,! (!H:' t<n.

'1'l Ol/ls first P:;;)/iCfJtion, :

ft:$ .tu:clwd :mln tne (tUt! cr axtWtnal}' :t/;(Ja :csdmg if' (t":dnVfields" \,
nel'eiiO !Je .Jo!/r..'wed with by
mte!lrgont ratJder. T:mfl$ .-
'Much of [ha book con carwin!v bu t$3o" '.Nid; pin<:,w((1 f-!rofit
D/at.flct .:'eofFE>phy fiwei an d t!:e infitt$
of num,m spm.: en, al/t'n wnntn compusuvv!,! mst7r;(<':tf trea$.
M(js( p<>,';;tnhl(' a
f
,!! "Ji;; Ch2pt7rs ?A, ;;'em.::t/!t!,; C!tango. <:luj 25
C
'3 '.,. . . . ..
wtursl Of!0\1'fng.1!1 w.'"::cn vr ngm tJ:t.-'3y1tom tne I7?Re1!itnic$
o! /dllgUf!(j;; :;,7'-"/ Cnii tf;.; pIn)' of hU.'I'&1I .F,F:nu .. _A
11-'otk cf 31lJ1..'.zillY fxud!dcr., bV 'J'llitcr who soems cq:mli1' ilf home-
!if prim!:/;lR c:"hinASI7t ft
IV/l/ De JUfjJ:tdsd iln ;ndist)tJffSiJuif., 3idta <il:' studems of :;r;mp81-
flt/;/(-; fihi1oJ''):l'/,' i'JAW St.-.tfiSmf'11
'{"h;; hi.1rr;r;r.<:i _;;;{"J(1}' nf rrmg::;;g.,s '.-vith whi,! tf;e t'0,;.,A O{)tJl1S,
is d ur cUllJpl!:fs:iL'.': !:t:d rel'!:!!:!'::: !IJt: rr:3pnitude of the
losi; and the. s:Jth:)t::; .f' nt' hi::; .1I.1hiet:r ft fS fi con.
Ui1IUl/!J!.' lu !i.','r;'w$/ic j!udl'.' EducaUan
:'P.::;t
[275
i55N 0 O'f -
J
l
!
, '
'r"
l
--- '1
.L
'1
L
l'
1
1
1
II!
Il
f
,j l:
)
1 .
1
1
1
,
"
i
, r-
LANGUAGE
Book! of SifJilar Inlemt
LOOM OF LANGUAGE
I!Y FREDERICK BODMER
Editcd and arranged by LANCELOT HOGllEN
Primers for tbe Age of Pknty, No. ~
By MARIO A. PEI
WORLD'S CHIEF LANGUAGES
THE STORY OF LANGUAGE
By OTTO ]f.S\'F.RSEN
LANGUAGE
Its Nature, Deve!opment and Origin
PHILOSOPHY OF GRAMMAR
ESSENTIALS OF ENGLISH GRAMMAR
A MODERN ENGLISH GRAMMAR ON
HISTORICAL PRINCIPLES Sevm Volwnes
HOW TO TEACH A FOREIGN LANGUAGE
LANGUAGE
BY
LEONARD BLOOMFIELD
PROFESSOR OF GERMANie PHILOLOGY
IN THE U.....IVERSITY OF CHICAGO
LOKDON
GEORGE ALLE", & U\'\VIN LTD
:\IUSEU.1>I STREET
COPYRIGHT IN U.S.A. 19H
REVISEO AND
FIRST PUBLISHED IN GREAT BRIT.AIN.19U
Reprinted HJ50, '955,1957, '958, '961 , 'g62, '965,1967,1959, 1970 and '973
This book is copyright the BerMe Convention. Ali rights
rcser1Jed. Apart from ony fair dcalint for the of privale
,tudy, resenrch, criridsm or review, a5 permirted undu the
Copyrighr Act, 1956, no part of this p"blicalion may he
reprodrlced, <torcd in a retrieval sysrem, or tronsmit/ed, in ony
form or by any means, electronic, declricoI, chcmical, mechanical
op/ieal, photowpying, ,"cording or orhcrwise, rhe prior
permission of tl,e copyrighr owncr. be addres,ed
10 the Pub/i,hm.
ISB" 0 04 400016 2
PRINTIlD IN GREAT li RITAIN
llY COMPTON LTD.
LONDON AND AYLESBURY
TO
A. S. B.
PREFACE
This book is a revised version of the author's Introduction io the
Siudy of Language, whieh appeared in 1914 (New York, Henry
Hoit and Company). The ncw version is much larger than the old,
because the science of language has in the interval made progrcss,
and bccause both men of science and the educated public now at-
tribute greater value ta an understanding of human speech.
Like its predecessor, this book is intended for the general reader
and for the student who is entcring upon linguistic work. Without
sucb an introduction, specialized treatises are uninteIHgible. For
the general reader an orderly survey is probably more i n t e r e s t i n ~
than a discussion of selected topics, for these, after ail, cannot be
understood without their background. No one will ask for an
anecdotal treatment who has once opened his eyes to the strange-
ness, beauty, and import of human speech.
The deep-rooted things about language, whieh mean most ta
all of us, are usually ignored in ail but very advanced studies; this
book tries to tell about them in simple terms and to show their
bearing on human affairs. In 1914 1 bascd this phase of the ex-
position on the psychologie system of Wilhelm Wundt, which was
then widely accepted. 8ince that time there has becn much up-
beaval in psychology; we have learned, at any rate, what one of
our masters suspeeted thirty years ago, namely, that we ean pursuc
the study of language without reference to any one psychologieal
doctrine, and that to do 80 safeguards our results and makes
them more signifieant to workers in related fields. In the present
book 1 have tried to avoid sllch dependenee; only by way of elu-
cidation l have told, at a few points, how the two main present-
day trends of psychology differ in their interpretation. The men-
talists would supplement the faets of language by a version in
terms of mind, - a version whieh will differ in the various schools
of mentalistic psychology. The meehanists demand that the
faets he presented without any assumption of such auxiliary
factors. 1 have tried to meet this demand not merely because 1
beIieve tbat mechanism ie the necessary form of scientific dis-
course, but also hecause an exposition whieh stands on its OWD
v
viii
PREFACE
feet is more solid and more easily surveyed than one which is
propped at various points hy another and c h a n g e a ~ l e doctrine.
I have tried everywherc to prcsent the accepted v!Cws, not cven
avoiding well-used standard examplos; on disputed matters 1 have
tried to state the point at issue; and in bath cases 1 have given
references, in the Notes and Bibliagraphy, which will enable the
reader to look inta things, and, if he chooses, ta arrive at an opinion
of his own.
Thanks are due to many scholars who contributed help and in-
formation and ta the publisher, the printer, and the very able
typesette:, ail of whom devoted great eare to the making of this
book.
L. B..
Chicago, January 1938.
PREFACE '1'0 THE BRITISH EDITION
This .. dition differs from the American form of this book (NewYork,
1933) in two respeets: the phonetic symbols conform to the usage
of thp. International Phonetic Association, and the transcriptions
of Engl..h forms represent a poli te type of British (' Received' or
. Public Rehool') prOlHlllcution. l\foreover, a few corrections havc
heen embodied in the text. AU thcsc changes were subject to li.
limitatiun imposcd by the method of manufacturing the book: the
paging and alignment of the American edition had to Le kept.
Aceol"(}ingly, the reader will.find some American features (snch as
the speUing -or for -our) and some passages where the point of view
(e.g., as tu topography) is American. However, in aIl cases where
corrections or additions seeme ta have material bearing, theso
have been either incorporated into the text, or, where this could not
he done, added in a list at the end of the book. For most of thesc
improvements l am indebted to Professars R. G. Kent and D.
Jones; t,he criticism and the publisbcd works of Professor Joncs
have aided me especiaUy as to British pronunciation.
L. B.
Chicago, A1l9ust, 1934.
CONTENTS
lLl.PTIl:R
1. THE STUDY OF LANGUAGE
2. THE USE OF LANGUAGE.
3. SPEECH-COMMUNITIES
4. THE LANGUAGES OF THE WORLD
5. THE PHONEME
6. TYPES OF PHONEME8
7. MODIFICATIONS .
8. PHONETIC STRUCTURE
9, MEANING .
10. GRAMMATICAL FORMS
11. SENTENCE-TYPES
12. SYNTAX
13. MORPHaLOGY
14. MORPHOLOGIC TYPES
15. SUBSTITUTION
16. FORM-CLASSES AND LEXICON
17. WRI'M'EN RECORDS .
18. THE COMPARATIVE METHOD
19. DIALECT GEOGRAPHY
20. PHONETIC CHANGE
21. TYPES OF PHONETIC CHANGE
22. FLUCTUATION IN THE F'REQUENCY OF FORMB .
23. ANALOGIC CHANGE .
24. SEMANTIC CHANGE
25. CULTURAL BORROWING
-26. INTIMATE BORROWING
27. DIALECT BORROWING
28. ApPLICATIONS AND UTLOOK
NOTES.
BtBLIOGRAPHY
TABLE OF PHONETIC SYMBOLS
ADDITIONS AND COlECTIONS
INDEX.
PAOlI
3
21
42
57
74
93
109
127
139
158
170
184
207
227
247
264
281
297
321
346
369
392
404
425
444
461
476
496
511
525
547
551
553
LANGUAGE
CHAPTER 1
THE STUDY OF LANGUAGE
1. 1. Language plays a groat part in our life. Perhaps because of
its familiarity, we rarely observe it, taking it rather for granted, as
we do breathing or walking. The effects of language are remarkable,
and include much of what distinguishes man from the animaIs, but
language has no place in our educational program or in the specula-
tions of our philosophers.
There are sorne cl'cumstances, however, in which the conven-
tionallyeducated person discusses linguistic matters. Occasionally
he debates questions of "correctness" - whether it is "better,"
for instance, to say it's Ior it's me. His discussion of such things
follows a fairly rigid pattern. If possible, he looks ta the
tians of writing for an answer - as, say, for the question whether
a t is ta be pronounced in words like often or soften. Otherwise he
appeals to authority: one way of speaking, he believes, is in-
herently right, the other inherently wrong, and certain learned
men, especially the authors of,grarnmars and dictionaries, can tell
us which is which. Mostly, however, he neglects ta consult these
authorities, and tries, instead, to settle the matter by a kind of
philosophical reasoning, which operates with terms such as
ject," "abject," "predicate," and so on. This ia the commonnse
way of dealing with linguistic matters. Like much else that mas--
querades as comrnon sense, it is in fact highly sophisticated, and
derives, at no great distance, from the speculations of ancient and
mcdicval philosophers.
It is only within the last century or so that language has been
studied in a scientific way, by careful and comprehensive
tion; the few exceptions will occupY us in a moment. Linguistics,
the study of language, is only in its beginnings. The knowledge it
has gained has not yet become part of our traditional education;
the "grammar" and other linguistic instruction in our schools
confines itself ta handing on the traditional notions. Many people
have difficulty at the beginning of language study, Dot in grasping
the methods or results (which are simple enough), but in stripping
3
4 THE STUDY OF LANGUAGE THE STUDY OF LANGUAGE
5
off the preconceptions which are forced on us by our popular-
scholastic doctrine.
1. 2. The ancient Greeks had the gut of wondering at things
that other people take for granted. They speculated boldly and
persistently about the origin, history, and structure of language.
Our traditionallore about language is duc largely to them.
Herodotus, writing in the fifth century B.C., tells us that King
Psammetichus of Egypt, in order ta find out which was the oldest
nation of mankind (whatever this may mean), isolated two new-
born infants in a park; whenthey began t speak, they uttcred
the word bekos, which turned out to be Phrygian for 'bread.'
In his dialogue Cratylus, Plato (427-347 B.C.) discusses the
origin of words, and particularly the question whether the relation
betwccn things and the words which name them is a natural and
necessary relation or merely the result of a human convention.
This dialogue gives us a tirst glimpse into a century-Iong contra-
versy between the Analogisls, who believed that language was
natural and therefore at bottom regular and logical, and the
Anomalists, who denied these things and pointed out the irregular
ities of linguistic structure.
The Analogists believed that the origin and the true meaning of
words could he traced in their shape; the investigation of this they
called etymology. 'Ve may illustratc thcir theory by English ex-
amples. The word blackbird obviously consists of black and bird:
the species was named for its color, and, indeed, blackbirds are
birds and are black. In the same way, the Greeks would have con-
cluded that there was sorne dcep-seated connection between a
gooseberry and a goose: it was the etymologist's task to fioq this
conoection. The word mushroom would have presented ft more
difficult problem. The components are oftcn altered; thus, break-
fast, in spite of the difference in sound, is evidently the meal by
which we break our fast, and manly a shortcr form of man-like.
In Greek, as in English, however, most words resist this kind of
analysis. Thus, early ends like manly, but the rest of the word is
obscure; woman resembles man, but what is the first syllable?
Then there is a residue of short, simple words that do not resemble
others - words such as man, boy, good, bad, eat, run. In such cases
// the Greeks and their pupils, the Romans, resorted to guesswork.
For instance, they explained the Greek word lithos 'stone' as
derived from the phrase Lian theein 'to run too much,' because this
is what a stone does not do. A Latin example of this sort has be-
come proverbial: lucus a non lucendo 'a grove (lucus) is so named
on account of its Dot being light (lucendo).'
These etymologies show us, at any rate, that the Greeks realized
that speech-forms change in the course of time. In the systematic
study of this change modern students have found the key to most
linguistic probtems. The ancients never scttled down to any
careful study of linguistic change.
The ancient Greeks studied no language but their own; they
took it for granted that the structure of their language embodied
the universal forms of human thought or, perhaps, of the cosmic
order. Accordingly, they made grammatical observations, but
confined these to one language and stated them in philosophical
form. They discovered the parts of speech of their language, its
syntactic constructions, such as, especially, that of subject and
predicate, and its chief infiectional categories: genders, numbers,
cases, persons, tenses, and modes. They defined these not in
tcrms of recognizable linguistic forms, but in abstract terms which
were to kil the meaning of the linguistic class. These teachings
appear most fully in the grammars of Dionysius Thrax (second
century B.C.) and of Apollonius Dyscolus (second century A.D.).
The Greeks made also sorne observations of detaU, but this
phase of their work, unfortunately, had less effect upon posterity.
Their great epic poems, the Iliad and the Odyssey, which they
viewed somewhat as sacred scriptures, were composed in an
ancient and otherwise unknown kind of Grcck. In order to un
derstand these texts and t make correct copies, one had to study
their language. Most famous in this work was Aristarchus (about
216-144 B.C.). Other works of Greek literature were composed in
conventionali7..cd forms of various regional dialeets: the Greeks
had the opportunity of comparing $Cvera] divergent forms of
their language. When the language of the great Athenian writers
of the fourth century had become antiquated, it was made a
special subject of study, sinee it represenkd the ideal form of
written discourse. Ail this work demanded careful observation
of detaUs. Sorne of the later grammarians, notably Herodian, the
son of Apollonius Dyscolus, assembled valuable information on
8uch topies as the inflection and accent of ancient Greck.
1. 3. The Greek generalizations about language were Dot im-
proved upon unti! the eighteenth century, when scholars ceased
6 THE STUDY OF LANGUAGE THE STUDY OF LANGUAGE 7
to view language as a direct gift of God, and put forth various
theories as to its origin. Language was an invention of ancient
hernes, or else the product of a mystical Spirit of the Folk. It
began in man's attempts to imitate noises (the "bow-wow"
theory), or in hisnatural sound-producing responses (the
dong" theory), or in violent outcries and exclamations (the" pooh-
pooh" theory).
In the etymological explanation of there was no
improvement. Voltaire is reported to have said that etymology
is a science in which the vowels count for nothing and the con-
sonants for very Iittle.
The Romans constructed Latin gramrnars on the Greek model;
the rnost famous of these, the work of Donatus (fourth century
A.D.) and of Priscian (sixth century A.D.), remained in use as
text-books through the Middle Ages. In the Middle Ages, when
Latin was changing from its ancient shape into the forms which
we know today as the Romance languages (French, Italian,
Spanish, and 50 on), the convention rernained of writing, as weIl
as one could, in the ancient classical fonn of Latin. The rnedieval
scholar, accordingly, in both the Latin countries and othera,
studied only classical Latin. The acholastic philosophers dis-
covered sorne features of Latin grammar, such as the distinc-
tion between nouns and adjectives and the differences between
concord, government, and apposition. They contributed much
less than the ancients, who had, at any rate, a first-hand knowl-
edge of the languages they studied. The medieval scholar saw in
classical Latin the logically normal form of human speech. In
more modern times this doctrine led ta the writing of general gram-
mars, which were to demonstrate that the structure of various
languages, and especially of Latin, embodies universally vaUd
canons of logic. The most famous of these treatises is the Gram-
maire gnrale et raisonne of the Convent of which
appeared in 166f}, This doctrine persisted into the nineteenth
century; it appears, for instance, in the classical acholar, Gott-
fried Hennann's work De emendanda ratione Graecae grammaticae
(1801). It is still embodied in our school tradition, which seeks
to apply logical standards ta language. Philosophera, ta this
day, sometimes look for truths about the universe in what are
really nothing but formaI features of one or another language.
An unfortunate outgrowth of the general-grammar idea was
the helief that the grammarian or lexicographer, fortified by his
powers of reasoning, can ascertain the logical basis of language
and prescribe how people ought to speak. In the eighteenth
century, the spread of education led many dialect-speakers to
learn the upper-class forms of speech. This gave the authoritari-
ans their chance: they wrote rwrmative grammars, in which they
often ignored actual usage in favor of speculative notions. Both
the belief in "authority" and sorne of the fanciful rules (as, for
instance, about the use of shaU and unIr; still prevail in our
schools.
For the rnedieval scholar, language meant classical Latin, as
it appears in books; we find few traces of interest in any other form
of speech. The horizon widened at the time of the Renaissance.
At the end of the Middle Ages, the study of Greek came back
into fashion; soon afterward, Hebrew and Arabie were added.
What was more important, sorne scholars in various countries
began to take an interest in the language of their own time.
The era of exploration brought a superficial knowledge of many
languages. Travelers brought back vocabularies, and mission-
&ries translated reIigious books into the tongues of newly-discovered
countries. Sorne even compiled grammars and dictionaries of ex-
otic languages. Spanish pricsts began this work as early as in the
sixteenth centuryj to them we owe a number of treatises on Ameri-
can and Philippine languages. These works can be used only
with caution, for the authora, untrained in the recognition of
foreign speech-sounds, could make no accurate record, and,
ing only the tcrminology of Latin grammar, distorted their ex
position by fitting it into this frame. Down ta our own tirne, per-
sons without linguistic training have produced work of this sort;
aside from the waste of labor, much information has in this way
been lost.
The increase of commerce and travel led also to the compila-
tion of grarnmars and dictionaries for languages closer at hand.
The linguistic horizon at the end of the eighteenth century can be
surveyed in the glossary of 285 words in two hundred languages
of Europe and Asia which P. S. Pallas (1741-1811) edited at the
behest of Empress Catharine of Russia in 1786. A second
tion of this, in 1791, added eighty more languages, including sorne
African and American. In the years 1806 to 1817 there appeared
a four-volume treatise under the tiUe Mithridates, by J. C. Adelung
8 THE STUDY OF LANGUAGE THE STUDY OF LANGUAGE 9
guages, and, above aIl, the documents which show us successive
stages of one and the same language, as for instance of Anglo-Saxon
(Old English) and modern Englsh, or of Latin and the modern
Romance languages. One kncw that sorne languages resembled
cach other, but the doctrine of linguistic decay diseouraged sys-
tematic study of this relation, since the changes which led, say,
from Latin ta modern French, wcrc vicwed as haphazard corrup-
tions.
The illusion that Latin had lived on, unchangcd, besidc the
Romance languages, led scholars ta derive contemporary languages
one from the other. they took Hcbrew ta be the language
from which ail others had sprung, but somc thought otherwise,
as, for example, Goropius Bceanus of Antwcrp, who patriotically
dcrived aIl languages from Dutch.
H was plain that the more farniIiar languages of Europe fell
into thrce groups by virtue of close resemblances within each
group, resemblances such as appear in the following words:
and J. S. Vater, which containcd the Lrd's Prayer in nearly live
hundred languages.
The Renaissance turned the intcrest of a few scholars ta the
aider records of their own languages. Franciscus Junius (1589-
1677) accomplished an enormous a.mount of work in the study of
the ancient documents of English and of the closely related lan-
guages, Frisian, Dutch, German, Scandinavian, and Gothie. This
last - a language no longer spoken today - Junius knew from
the famous Silver Codex, then reeently discovered, a manuscript
of the sixth century A.D. eontaining fragments of a Gospel transla-
tion; Junius published its text, together with that of the Anglo
Saxon Gospels. George Hickcs (1642-1715) continued this work,
publishng a Gothie and Anglo-Saxon grammar and a Thesaurus
of miseellancous information about the older stages of English
and the sister tangues.
1. 4. The development so far outlined shows us what eighteenth-
century scholars knew about language. They stated the gram-
matical features of in philosophieal terrns and took no
account of the structural difference between languages, but ob-
scured t by forcing their descriptions into the scheme of Latin
grammar. They had not observed the sounds of speech, and con-
fused them with the written symbols of the alphabet. This failure
to distinguish betwcen actual speech and the use of writing dis-
torted also their notions about the history of language. They saw
that in medieval and modern times highly cultivated persans
wrote (and even spoke) good Lat.in, while less educated or careless
scribes made many mistakcs: failing to see that this Latin-writing
was an artificial and academic exercise, they concluded that lan-
guages arc preserved by the usage of educated and careful people
and changed by the corruptions of the vulgar. ln the case of
modern languages like Englsh, they believed, accordingly, that
the speech-forms of books and of upper-class conversation repre-
scnted an aIder and purer levd, from whieh the "vulgarisms" of
the common people had branched off as "corruptions" by a process
of "linguistic decay." The grammarians feIt free, therefore, to
prescribe fanciful ruIes which they derived from considerations of
lo/!;ic.
These misconceptions prevented scholars from making use of
the data that were at hand: the modern languages and dialects,
the records of ancient languages, the reports about exotic
GERMANIe GROUP
'hand'
English hand
Duteh hand
German Hand
Danish haand
Swedish hand
'foot'
English foot
Duteh voet
German FU8Z
Danish fod
Swedish fot
'winter'
English winter
Dutch winter
German Winter
Danish vinter
Swedish vinter
ROMAl'iCE GROUP
French main
Italian mano
Spanish mano
French pied
l talian piede
Spanish pie
French hiver
Italian inverno
Spanish invierno
SLAVIC GROUP
Russian ruka
Polish rka
Bohemian ruka
Serbian ruka
Russian noga
Polish noga
Bohemian noha
Serbian noga
Russian zima
Pollsh zima
Bohemian zima
Serbian zima
10
THE STUDY OF LANGUAGE THE STUDY OF LANGUAGE 11
There was apparent also a less striking resemblance bctween
these groups; this wider resemblance extended to some other lan-
guages, such as, notably, Greek:
'mother': Greek miMr, Latin mater (with its modern forms in
the Romance languages), Russian mat' (genitive case materi-
with sirnilar forms in the other Slavic languages), English mother
(w}th s:rnilar forrns in other Germanie languages);
two : Greek duo, Latm duo, Russian dva English two' , ,
'three': Greek treis, Latin Ires, Russian tri English Ihree-
'is': Greek esti, Latin est, Russianjest', is t).
1. 5. Outside the tradition of Europe, several nations had de-
veloped Iinguistic doctrines, chiefly on an antiquarian basis. The
Arabs had workcd out a grammar of the classical form of their
as it appears in the Koran; on the model of this, the
Jews III Moharnmedan countrics constructed a Hebrew grammar.
At the Renaissance, European scholars beearne acquainted with
this tradition; the term root, for instance, as a designation for the
central part of a word, cornes from Rebrew grammar. In the Par
East, the Chinese had gained a great deal of antiquarian linguistic
knowledge, especially in the way of lexicography. A Japanese
grammar secms to have grown up independently.
It was in India, howevcr, that thcre arose a body of knowledge
which Will! destined t rcvolutionizc European ideas about lan-
gua?e. cfhe Brahmin religion guardcd, as sacred texts, some very
collections of hymns; the oldest of these collections, the
Rlg-Veda, dates in part, at a conservative estimate, from about
1200 B.C. As the language of these tcxts grew antiquated, the
proper way of pronouncing them, and their correct intcrpretation,
?ecame t.he task of a special class of lcarned men. The antiquarian
III language which arose in this way, was carried over
a more practical sphere. Among the Hindus, as among us,
dlfferent classes of society differed in speech. Apparently there
GERMAI,UC OROUP
'drink'
English drink
Dutch drinken
German trinken
Danish drikke
Swedish dricka
ROMANCE GROUP
French boire
rtalian bere
Spanish beber
SLAvre GROUP
Russian pit'
Polish pic'
Bohemian piti
Serbian piti
were forces at work which led upper-class speakers to adopt lower-
class forms of speech. We find the Hindu grammarians extending
their interest from the Scriptures to the upper-caste language,
and making rules and lists of forms descriptive of the correct type
of speech, which they called Sanskrit. In time they worked out
a systematic arrangement of grammar and lexicon. Generations
of such labor must have preceded the writing of the oldest treatise
that has come down to us, the grammar of Pt).ini. This grammar,
which dates from somewhere round 350 to 250 B.e., is one of the
greatest monuments of human intelligence. It describes, with the
mnutest detail, every inflection, derivation, and composition, and
every syntactic usage of its author's speech. No other language,
to this day, has been so perfectly described. It may have been due,
in part, to this excellent codification that Sanskrit became, in
time, the official and literary language of aU of Brahmin India.
Long after it had ceased to be spoken as anyone's native language,
it remained (as classical Latin remained in Europe) the artificial
medium for ail writing on lcarned or religious topics.
Some knowledge of Sanskrit and of the Hindu grammar had
reached Europe, through missionaries, in the sixteenth and seven-
teenth centuries. In the eighteenth century, Englishmen in India
transmitted more exact reports; round the beginning of the nine-
teenth century, the knowledge of Sanskrit became part of the
equipment of European scholars.
1. 6. The Indian grammar presented to European eyes, for the
first time, a complete and accurate description of a language,
based not upon theory but upon observation. Moreover, the dis-
covery of Sanskrit disclosed the possibility of a comparative study
of languages.
'1'0 begin with, the concept of related languages was strikingly
confirmed by the existence, in far-off India, of a sister of the
familiar languages of Europe; witness, for example, the Sanskrit
equivalents of the words above cited:
mata 'mother,' accusative case mataram;
dvau 'two';
trayah 'three';
asti 'he s.'
Even more important was the insight into linguistic structure
which one got from the accurate and systernatic Hindu grammar.
Datit now, one had been able to sec only vague and fluid similar-
12
THE STUDY OF LANGUAGE
THE STUDY OF LANGUAGE 13
ities, for the current grammars, buUt on the Greek model, did not
clearly set off the features of cach language. The Hindu grammar
taught Europeans to analyze speech-forms; when one oompared
the constituent parts, the resernblanccs, which hitherto had been
vaguely recognized, could he set forth with certainty and precision.
The old confused notions of linguistic relationship lived on for
a brief time in the opinion that the European languages were
derived from Sanskrit, but this opinion soon gave way ta the ob-
viously correct explanation, namely, that Sanskrit Latin Greek
, , J
and sa on, ",ere divergent later fOTIns of sorne one prehistoric
guage. This explanation seems ta have been first stated by Sir
William Joncs (1746-1794), the first great European Sanskrit
scholar, in an address delivercd in 1786: Sanskrit bears a resem-
blance to Greek and Latin which is tao close to he due 1{> chance
,
but shows, rather, that all three "have sprung from sorne common
source which, perhaps, no longer exists," and Gothic (that s,
Germanie) and Celtic probably had the same origin.
In arder ta work out the comparison of these languages, one
needed, of course, descriptive data for each one of them. The
prospect of comparison, however, with ail that it revcalcd about
ancient speech-forms and tribal migrations and the origin of
peoples and customs, proved so alluring that no one undert()ok
the humdrum task of analyzing the other languages on the model
of Sanskrit. European scholars had a sound knowledge of Latin
and Greek; most of them spoke sorne Germanie language as their
Confronting a precise statement of Sanskrit
grammar or a carcfully analyzed lexical form, they eould usually
recall a similar feature from some of the more familiar languages.
In reality, of course, this was a makeshift; often enough the com-
parer had ta make a preliminary investigation to establish the
faets, and sometimes he went astray for lack of methodical1y
arranged data. If European scholars had possessed descriptions
of thc sister languages comparable to the Hindus' description of
Sanskrit, the comparative study of the Indo-European languages
(as they are now called) would have progressed far more speedily
and accurately. Yet, in spite of pOOl' equipment, and thanks to
the energy of its workers, the historical and comparative study of
lndo-European languages beeame one of the principal enter-
pnses, and one of the most sueeessful, of European science in the
nineteenth ccntury.
The languages of Persia (the sa-called lranian languages) so
closely resembled Sanskrit that their kinship was certain from the
start. A similar l'dation, though less close, was found ta exist
between the BalUe languages (Lithuanian, Lettish, and Old
Prussian) and the SIavic. Jones' surmise that the Germanie lan-
guages ",cre related to Latin, Groek, and Sanskrit, at once proved
truc, as did later his surmise about Celtic (Irish, Welsh, Cornish,
Breton, and the aneient language of Gaul). Later, Armenian and
Albanese, and a few aneicnt languagcs known to us only from
seant written records, proved also ta bclong to the lndo-European
family.
Although there was some dispute as ta details, the general
suppositions of historical and comparativc language-study saon
became dear. Lanb'1lagcs change in the course of bme. Apparent
exceptions, such as the medieval and modern usc of Latin (or, in
India, of Sanskrit), amount only ta this, that by long schooling
people can he trained ta imitate the language of ancient writings.
This antiquarian feat is utterly different from the normal trans-
mission of speech from parents to children. Ali writing, in fact,
is a relatively rccent invention, and has remained, almost ta our
day, the property of only a ehosen few: the effect of writing upon
the forms and the development of actnal speech is very sIight.
If a language is spoken over a large area, or thanks to migration,
in several separate areas, then it will change differently in different
places, and the result will he a set of related languages, like Italian,
French, Spanish, Portuguese, Roumanian, and the other Romance
dialects. We infer that other groups of related languages, such
as the Germanie (or the Slavic or the Celtic), which show a similar
resemblance, have ariscn in the same way; it is only an accident
of history that for thesc groups wc have no writ.ten records of the
carlier state of the languagc, as it ",as spoken bcfore the differen-
tiation set in. To these unrecorded parent languages we give
names like Primive Germanie (Primitt've Slavie, Primitive Celtie,
and sa on).l In the same way, finding that aU these languages
and groups (Sanskrit, Iranian, Armenian, Greek, Albancse, Latin,
Celtie, Germanie, BalUe, Slavie) resemble eaeh other beyond the
possibility of mere chance, we cali them the l ndo-European family
1 The word primite i8 here poorly chosen, sinee it is intended to mean ol;ly that
we happcn to have no written of the language. German scholars have '"
bette,. dcvice in their prefix ur- 'prime",,]" with which they form, for this purposc,
names like urGer manisch. urslavisch, urkeltisch.
14 THE STUDY OF LANGUAGE THE STUDY OF LANGUAGE 15
of languages, and conclude, with Jones, that they are divergent
forms of a single prehistorie language, ta which we give the name
Primitive Ind(}-European.
The method of comparison, too, was clear from the start, In
general, any feature that is common to aIl or ta several of the
related languages, must have been present in their common ant-
cedent stage, in the "parent language." Thus, from the above
cited forms of the word for 'mother,' it is cIear that in Primitive
Indo--European this word must have begun with the sound whieh
we indicate in writing by means of the lettcr m. "Where the related
languages do not agree, some or aU of them must have made sorne
change. Thus, it is clear that the second consonant in the word
for 'mother' was in Primitive Indo-European a t-sound, and that
the th-sound in English (as weIl as the eariier d-sound in the Old
English form, modor) must be due to change.
1.7. The beginning of a systematic comparison of the Indo-
European languages was a treatise on the inHectional cndings of
verbs in Sanskrit, Grcek, Latin, Persian, and Germanic, published
in 1816 by Franz Bopp (1791-1867), In 1818 Rasmus Kristian
Rask (1787-1832) showed that the words of the Germanic
guagcs bear a regular formaI relation in matters of sound, to the
words of the other Indo-European languages, For instance,
where the others have p, the Germanic languages have f, as in
fathe:r: Latin pate:r, foot: Latin ps, flve: Greek pente, few: Latin
paud. In 1819 Jakob Grimm (1787-1863) pubIished the tirst
volume of his Deutsche Grammatz'k, which was not, as the title now-
adays would indieate, a German grammar, but a comparative
grammar of the Germanie languages (Gothic, Scandinavian,
English, Frisian, Dutch, and German). In the second edition,
in 1822, of this volume, Grimm prcsented a systematic
tion of the correspondences of consonants between Germanic and
the other Indo-European languages; sinee then, these correspond-
ences have been known to scholars as Grimm's
Law. These corrcspondences are a matter of historical detail,
but their significance was overwhelming, since they showed
that human action, in the mass, is not aItogether haphazard, but
may proceed with regularity even in so unimportant a matter as
the manner of pronouncing the individual sounds within the How
of speech. Grimm's comparison of the Germanie languages re-
mains to this day unrivaled; three more volumes appeared in 1826,
1831, and 1837; a fifth volume, which was to complete the syntax,
never appeared,
In 1833 Bopp began the publication of a comprehensive treatise,
a comparative grammar of the Indo--European languages. In
the years 1833 to 1836 there appeared the tirst edition of the
Etymological Investigations of August Friedrich Pott (1802-1887).
The term etymology, here as in aU modern discussions, has taken
on a precise meaning: the etymology of a is simply its
history, and is obtainrloy the sam-
-raiguage ana-tne'oI111s-tl1' relat'id--languages whichare" divergent
variants of the same parent form'; , Thus, to state the etymology
word mother is to say that this forrn is the modern
version of the ninth-century Old EngIish modor; that this is te-
lated ta Old Norse mo'Oer, Old Frisian moder, Old Saxon mdar,
Old High German muoter (these are the forms in our oldest
ords of the respective languages), in the sense that aIl these are
divergent variants of a single Primitive Germanie word, which
we symbolize as *moder; and that these Germanie forms are in
turn related to ("cognate with") Sanskrit mtilii, Avestan (Old
Iranian) rniitil, Old Armenian mair, ancient Greek rntr, Albanese
motre (which, however, means 'sister'), Latin mater, Old Irish
mtithir, Lithuanian mote (which means 'wife'), Old Bulgarian
(Slavic) mati, and with the other corresponding forms in eaeh of
the groups of languages here illustrated, in the sense that ail
these are divergent later forms of a single Primitive Indo-European
word, which we symbolize as *mtitr. As this example shows, ety-
mologies, in the modern sense, do not necessarily show us an older,
more transparent meaning of words. Our modern etymologies in
the Indo-European languages are due largely to the researches
of Pott.
During the foIIowing decades progress was so rapid that bath
smaller treatises and the great handbooks rapidly became
quated. Of the latter, Bopp's, in spite of new editions, was
scded in 1861 by the Compendium of the Comparative Grammar of
the Ind(}-European Languages of August Sehleicher (1823-1868).
In 1886 Karl Brugmann (1849-1919) and Berthold Delbrck
(1842-1922) began the publication of their Outline of the Compara-
tive Grammar of the Indo--European Languages; the standard work
of reference today is the second edition of this, which appeared
from 1897 ta 1916.
16 THE STUDY OF LANGU AGE THE STUDY OF LANGUAGE 17
As the work went on, other, more detailed treatises were
yoted ta the separate branches of the family, in
the manner of Grimm's great treatise on Germanie. Friedrich
Diez (1794-1876) began the serioue study of the Romance
guages in his Grammar of the Romance Languages (1836-1844);
Johann Kaspar Zeuss (1806-1856) opened the field of the Celtic
languages in his Grammai2'ca Celliea (1853) j Franz yon
sich (1813-1891) wrote a Comparative Grammar of the Slavic
guages (1852-1875).
1. 8, These studies could not fail to throw light upon many an
aspect of history and archaeology, but their immediatc interest
lay in whaf. they told about human speech. Although the various
Indo-European languages had a common origin, their lat.er earcers
werc indcpcndent: the student had now a vast collection of de-
tails concerning the changes in human speech, which enabled him
to generalize on the manner of this changc.
To draw the conclusions as to the way in which languages change,
was to replace the speculation of earler tirnes by the rcsults of
scientific induction. William Dwight Whitney (1827-1894), an
American scholar, wrote Language and the Study of Language
(1867) and The Life and Growth of Language (1874). These books
were translated into scveraI European languages; today they
seem incomplete, but scarcely antiquated, and still serve as an
exccllent introduction ta languap;e study. In 1880 t.here ppeared
the Principles of Linguistic History by Hermann Paul (1846-1921),
which, in its successive editions (the fifth appeared in 1920),
came the standard work on the methods of historical linguistics.
Paul's book of Principles iIlust.rates, with a wealth of examples,
the process of linp;uistic change whieh had been revealed by
European studies. Not sa welt ,vrittcn as \Vhitney's, but more
detailed and methodical, this book excrcised a great influence on
linguistic studies; studcnts of a more recent generation are neg-
lecting it, to their disadvantage. Aside from its very dry style,
Paul's Principles suffers from faults that scem obvious today,
because they are significanf. of the limitations of
century linguistics.
One of these fauIts is Paul's neglect of descriptive language
study. He admitted that descriptions of languages were
sary, but confined his aetual discusson to matters of linguistic
change. This shortcoming he shures with his cpoch. We can study
linguistic change only by comparing related languages or dif-
ferent historieal stages of the same language. For instance, by
noting the similarities and differences of English, Frisian, Dutch,
German, Scandinavian, and Gothie, we can gct a notion of the
older language (" Primitive Germanie") from which thcy haye
differentiated in the course of time, and we can then study the
changes which have occurrcd in cach of thcse later languages.
Or else, by comparing our records of Ol English (say, in the
ings of King Alfre) with modern English, we can see how
lish has changcd in t.he bst thousand years. Evidently our power
of making this eomparison depends upon our knowledge of the
things to be compared. For example, our knowledgc about the
compounding of words (as in blackbird or footsore) in the several
Germanie languages is decidedly incompIete; therefore wc can-
nat go very far with a comparative study of this matter, which
would tell us how words were cOIllIJOunded in Primit.ive C.:rermanic,
and how thcse habit.s have changed in t.he subsequent history of
eaeh Germanie language. The historieal language students of
the nineteenth century suffered under these limitations, but they
seem not to have grasped the nat.ure of the diffieulty.
The other great weakness of Paul's Principles is his insistcnce
upon "psyehologcal" intcrpretation. He accompanies his
ments about language with a paraphrase in tenus of mental
processes which t.he speakers are supposed ta have undergone. The
only evidence for these mental processes is the linguistic proceSSj
they add nothing t.o t.he discussion, but onl:y obscure it.. In Paul's
book and largely to the present day, linguistics betrays its descent
from the philosophical speculations of t.he ancient. Greeks. Paul
and most of his contemporaries dealt only with
languages and, what with their neglect of descriptive problems,
refused to work with languages whose history was unknown. This
limitation eut them off from a knowledge of foreign types of
grammatical structure, which would have opened t.heir eyes to the
fact t.hat even the fundamental features of
mar, such as, espeeially, t.he system, are by no
means universal in human speech. Belicving these feat.ures to be
universal, they resort.ed, whenever they dealt with fundamentals,
ta philosophical and psychological pscudo-cxplanations.
1. 9. Alongside the grcat stream of historical research, there ran,
however, a small but accelerating current of general linguistic
18 THE STUDY OF LANGUAGE THE STUDY OF LANGUAGE 19
study. The Hindu grammar of Sanskrit was never quite for-
gotten; while many pupils used Hs results without knowing of its
existence, the masters, who knew the antecedents of their science,
appreciated its value. For the less-known Indo-European lan-
guages descriptive studies could not be avoided. It is surely no
accident that the best of these, in the field of the Slavic and
Baltic languages, were furnished by August Lskien (1840-1916),
a scholar who took a leading part in laying the foundations of
historical methods of research.
For the most part, however, descriptive studies did not merge
with the main stream of historical work. Some students were
attracted by the structural peculiarities of languages outside the
Indo-European group, even though the history of these languages
was unknown. Other students examined a variety of languages in
order to get a philosophical SUl'"Vey of human speech j in fact, much
of the oIder descriptive work is almost unintelligible today because
it is pel'"Vaded by philosophical notions that are no longer familiar
ta us.
The first great book on generallinguistics was a treatise on the
varieties of human speech by Wilhelm von Humboldt (1767-
1835), which appeared in 1836. H. Steinthal (1823-1899), beside
more general writings on the fundamentals of language, published
in 1861 a treatisc on the principal types of language structure.
G. von der Gabelentz' (1840--1893) work on the science of language
(1891) is much less philosophical. This direction of study cul-
minated in a great work on language by the philosopher and
psychologist, Wilhelm Wundt (1832-1920), which appeared in
1900 as the first part of a treatise on social psychology. Wundt
based his psychology of speech upon anyand ail accessible descrip-
tions of languages. It is interesting today to read the Indo-
Europeanist Delbrck's critique and Wundt's rejoinder, both of
which appeared in the following year. Delbrck objects to Wundt's
use of languages whose history is unknown; for him the only aspect
of language worth studying is its change in the course of time.
Wundt, on the other hand, insists upon the importance of psycho-
logical interpretation in terms of his system, while Delbrck says
that it does not matter what particular system of psychology a
linguist may choose.
Meanwhile sorne students saw more and more clearly the natural
relation between descriptive and historical studies. Otto Bohtlingk
(1815-1904), who made the modern European edition of PQini,
applied the descriptive technique ta a language of totally different
structure, the Yakut of Asiatie Russia (1851). Friedrich Mller
(1834-1898) published an outline of linguistic science (1876-
1888) which contained brief sketches of the languages of the world,
regardless of whether a historieal treatment was possible. Franz
Nikolaus Finek (1867-1910), bath in a theoretical essay (1905) and
in a Iittle volume (1910) in whieh he analyzed descriptively eight
unrelated languages, insisted upon descriptive study as a basis for
both historical research and philosophieal generalization. Ferdi-
nand de Saussure (1857-1913) had for years expounded this matter
in his university lectures; after his death, they wcre published in
book form (1915).
Most convincing in this respect was the historieal treatment of
language famUies other than the Indo-European. On the one hand,
the need of descriptive data as a prerequisite for comparative work
was here self-evidcnt; on the other hand, the results showed that
the processes of linguistic change were the same in aIl languages,
regardless of their grammatical structure. The comparative study
of the Finno-Ugrian languages (Finnish, Lappish, Hungarian, and
their kin) began as early as 1799, and has been greatly elaborated.
The second volume of Humboldt's great treatise founded the
comparative grammar of the Malayo-Polynesian language family.
Today we have comparative studies of other families, such as the
Semitic family and the Bantu family in Africa. Students of
American languages could indulge in no self-deception as to the
need of descriptive data: north of Mexico alone there are dozens
of totally unrelated groups of languages, presenting the most varied
types of structure. In the stress of recording utterly strange forms
of speech one soon learned that philosophical prepossessions were
only a hindrance.
The mcrging of these two streams of study, the historical-
comparative and the philosophical-descriptive, has made cIear
sorne principles that were not apparent ta the great Indo-European-
ists of the nineteenth century, as represented, say, by Hermann
Paul. AIl historical study of language is based upon the comparison
of two or more sets of descriptive data. It can he only as accurate
and only as complete as these data permit it to he. In order to
describe a language one needs no historical knowledge whateverj
in faet, the obsel'"Vcr who allows such knowledge to affect bis
20 THE STUDY OF LANGUAGE
description, is bound to distort his data. Our descriptions must
be unprejudiced, if they are to give a sound basis for comparative
work.
The only usefuI gencralizations about language are inductive
generalizations. Features whic1 we think ought ta be universal
may be absent from the very next language that becomes accessible.
Sorne features, such as, fO: instance, the distinction of vcrb-like and
noun-like words as separate parts of speech, are cornmon ta many
languages, but laeking in others. The fact that sorne features are,
at any rate, widespread, is worthy of notice and caUs for an ex-
planation; when we have adequate data about rnany languages, we
shall have to return to the problem of general grammar and ta
explain these similarities and divergences, but this study, when it
cornes, will be not speculative but inductive.
As ta change in language, we have enough data ta show that the
general processes of change are the sarne in aIl languages and tend
in the same direction. Even very specifie types of change oceur in
much the same way, but independently, in the most diverse
languages. These things, tao, ,vill sorne clay, when our knowledge is
wider, lend thernselves ta a systematic survey and ta fruitful
generalizatian.
CHAPTER 2
THE USE OF LANGUAGE
2. 1. The most difficult step in the study of language is t,he
fust step. Again and again, schol;:;'shp has approached the study
of language without actually entering upon it. Linguistic science
arose from relatively practical preoccupations, such as the use of
writing, the study of literature and especially of aIder records, and
the prescription of elegant speech, but people can spend any
amount of time on these things without actually entering upon
linguistic study. As the individual student is likely to repeat the
delays of history, we may do well ta speak of these matters, so as ta
distinguish them from the subject of our study.
Writing is not language, but merely a way of recording language
by means of visible marks. In sorne countries, such as China,
Egypt, and Mesopotama, writing was practised thousands of
years aga, but to most of the languages that are spoken today it has
been applied either in relatively recent times or not at aU. More-
over, until the days of printing, literacy was confined to a very
few people. AH languages were spoken through nearly aH of their
history by people who did not read or write; the languages of such
peoples are just as stable, regular, and rich as the languages of
literat{l natious. A language is the same no matter what system of
writing may be used ta record it, just as a person s the same no
matter how you take his picture. The Japanese have three systems
of writing and are developing a fourth. When the Turks, in 1928,
adopted the Latin alphabet in place of the Arabic, they went on
talking in just the same way as before. In arder t study writing,
we must know something about language, but the reverse is not
true. Ta be sure, we get our information about the speech of past
times largely from written records - and for this reaSn we shaH,
in another conneetion, study the history of writing - but we find
this ta he a handicap. We have tu use great care in interpreting the
written symbols into terme of actual speech; often wc fail in this,
and always we should prefeT ta have the audible word.
Literature, whcthcr prcsented in spoken form Of, as is now our
21
22 THE USE OF LANGUAGE THE USE OF LANG UAGE 23
custom, in writing, consists of bcautiful or otherwisc notable ut
terances. ThfJ stur!ent of Iiterature observes the utterances of cer-
tain persons (say, of a Shakspere) and conccrns himself with the
content and with the unusual features of form. The interest of the
philologist is even broader, for he is conccrned with the cultural s i g ~
nificance and background of what he reads. The Hnguist, on the
other hand, studies t.he language of ail persons alike; the individual
features in which the language of a great writer differs from the
ordinary speech of his time and place, interest the linguist no more
than do the individual features of any other person's speech, and
much less than do the features that are common ta ail speakers.
The discrimination of elcgant or ., correct" speech is a by-product
of certain social conditions. The linguist has to observe it as he
observes other linguistic phenomena. The fact that speakers
label a speech-form as "good" or" correct," or cIse as "bad" or
"incorrect," is merely a part of the linguist's data concerning this
speech-form. Necdless ta say, it does not permit hirn ta ignore
part of his matcrial or ta falsify his records: he observes ail speech-
forms impartially. It is part of his task ta find out under ,,,hat
circumstances the speakers label a form in one \Vay or the other,
and, in the case of each particular form, why they label it as they
do: why, for example, rnany people say that ain't is "bad" and
am not is "good." This is only one of the problems of linguistics,
and since it is not a funclamental one, it can be attacked only
after many other things are known. Strangely enough, people
without linguistic training devote a great deal of effort to futile
discussions of this topie without pl'ogressing ta the study of lan-
guage, whieh alone could /!;ive them the key.
A student of writin/!;, of litorature or philology, or of correct
speech, if he were persistent and methodical enough, might realze,
after sorne waste of effort, that he had better first study language
and then return to theso problems. We can save ourselves this
detour by turning at once to the observat.ion of nonnal speech.
We begin by observing an act of speech-utterance under very
simple circumstances.
2.2. Suppose that Jack and Jill are walking down a lane. Jill
is hungry. She sees an apple in ft tree. She makes a noise with her
larynx, tangue, and lips. Jack vaults the fence, cUmbs the tree,
takes the apple, brings it to Jill, and places it in her hand. ,Jill
eats the apple.
This succession of events could be studied in many ways, but
we, who are studying language, will naturaHy distinguish between
the act of speech and the other occurrences, which we shaH caU
practical events. Viewed in this way, the incident eonsists of three
parts, in order of time:
A. Practical events preceding the act of speech.
E. Speech.
C. Practical events following the act of speech.
We shaH examine first the practical events, A and C. The events
in A coneern mainly the speaker, Jill. She was hungry; that is,
sorne of her muscles wero contracting, and sorne fluids were being
secreted, especially in her stomach. Perhaps she was also thirsty:
her tangue and throat were dry. The lightwaves refiected from
the red apple struck her cyes. She saw Jack by her side. Hel'
past dealings with Jack should now enter into the picture; let
us suppose that they consisted in sorne ordinary relation, like
that of brother and sister or that of husband and wife. Ali these
events, which precede JilI's speech and concern her, we calI the
speaker' s stimulus.
We turn now ta C, the practical events which came aHer Jill's
speech. These concern mainly the hearer, Jack, and consist of
his fetching the apple and giving it to Jill. The practical events
which follow the speech and concern the hearer, we cali the hear-
er's response. The events which follow the speech concern also
Jill, and t-his in a very important way: she gets the apple into her
grasp and eals it.
It is evident at once that, our whole story depends upon sorne
of the more remote conditions connected with A and C. Not every
Jack and Jill would behave like these. If Jill were bashful or if
she had !lad bad experiences of Jack, she might be hungry and
see the apple and still say nothing; if Jack were ill disposed to-
ward her, he might not fotch her the apple, even though she asked
for it. The occurrence of a speech (and, as we sha1l see, the ward.
ing of it) and the whole course of practical events before and after
it, depend upon the entire life-history of the speaker and of the
hearer. 'Ve shaH assume in the present case, that ail these pre-
disposing factors were such as ta produce the story as wc have
told t. Supposing this, wc want ta know what part the s p e e c h ~
utterance (B) played in this story.
If Jill had been alone, s h ~ might have been just as hungry and
24
THE USE OF LANGUAGE THE USE OF LANGUAGE 25
thirsty and might have secn the same apple. If she had sufficient
strength and skill ta get over the fence and c1imb the tree, she
could get hold of the apple and eat it; if not, she would have to
stay hungry. The lone Jill is in much the same position as the
speechless animal. If the animal is hungry and sees or smells
food, it moves toward the food; whether the animal succeeds
in getting the food, depcnds upon its strcngth and skill. Thc state
of bunger and the sight or smeH of the food are the stimulus (which
we symbolize by S) and the movements toward the food are the
readion (which we symbolize by R). The lone Jill and the speech
less animal act in only one way, namely
S1$ loR.
If this works, they get the food; if it does not work - if they are
not strong or skilful enough ta get the food by the actions R - they
must stay hungry.
Of course, it is important for Jill's weHare that she get the apple.
In most instances it is not a matter of life and death, though sorne-
times it is; in the long run, however, the Jill (or the animal) that
gets the food has far better chances of surviving and populating
the earth. Therefore, any arrangement which adds ta .JiIl's chances
of getting the appIe, is enormously valuable for her. The speak-
ing Jill in our story availed herseH of just such an arrangement.
She had, to hegin with, the same chance of getting the apple as
had the lone Jill or the speechlcss anima!. In addition ta this,
however, the speaking Jill had'a further chance whic1 the others
did not share. Instead of strugglng with the fence and the tree,
she made a few small movements in her throat and mouth, which
produced a little noise. At once, Jack began ta make the reactions
for her; he performed actions that were beyond Jill's strength,
and in the end Jill got tbe apple. Language enables one person to
make a readion (R) when another persan has the stimulus (S).
In the ideal case, witbin a group of people who speak to each
other, each person bas at his disposaI the strength and skiU of
every person in the group. Tbe more these persons differ as ta
special skills, the wider a range of power does each one person
controL Only one person needs to he a good climber, since he
can get fruit for aIl the rest i only one needs to be a good fisherman,
since he can supply the others with fish. The div'ion of Ibor,
and, with il, the whole working of human society, is due ta Inguage.
2.3. We have yet to examine B, the speech-evcnt in our story.
This, of course, is the part of the story with which we, as students
of language, are chiefly concerned. In ail of our work we are ob-
serving Bi A and C concern us only because of their eonnection
with B. Thanks to the sciences of physiology and physics, we know
enough about the speech-event ta sec that it consists of three parts:
(BI) The speaker, Jill, moved her vocal chords (two little
muscles inside the adam'g.apple), ber lower jaw, her tongue, and
so on, in a way which forced the air into the form of sound-waves.
These movements of the speaker are a reaction to the stimulus S.
Instead of performing the practical (or handling) reaction R-
namely, starting realstically off to get hold of the apple - she
performed these vocal movements, a speech (or substitute) reaction,
which wc shaH symbolize by a small letter r. In sum, then,
Jill, as a speaking person, has not one but two ways of reacting
ta a stimulus:
81$ ) R (practical reaction)
81$ ) r (1nguistic substitute reaction).
In the present case she performed the latter.
(B2) The sound-waves in the air in Jill's mouth set the sur-
rounding air into a similar wave-motion.
(B3) These sound-waves in the air struck Jack's ear-drums
and set them vibrating, with an effoct on Jack's nerves: Jack
heard the speech. This hearing acted as a stimulus on Jack: we
saw him running and fetching the apple and placing it in Jill's
grasp, much as if Jill's hunger-and-apple stimulus had been act-
ing on him. An observer from another planet, who did not know
that there \vas such a thing as human speech, would have ta con-
clnde that somewherc in Jack's body there was a sense-organ
which toid him, "Jill is hungry and sees an apple up there." In
short, Jack, as a speaking person, reacts to two kinds of stimuli:
practical stimuli of the type S (such as hunger and the sight of
food) and speech (or substitute) stimuli, certain vibrations in his
ear-drums, whieh wc shaIl symbolize by a small letter s. When
we seek Jack doing anything (fetching an apple, say), his action
rnay be due not only, as are an animal's actions, to a practical
stimulus (such as hunger in his stomach, or the sight of an apple),
but, just as often, ta a speech-stimulus. His actions, R, may he
prompted not by one, but by two kinds of proddings:
(practical stimulus) S1$ ) R
(linguistic substitute stimulus) sJt ) R.
26 THE USE OF LANGUAGE THE USE OF LANGUAGE 27
It is evident that the connection hetween JiIl's vocal move-
ments (BI) and Jack's hearing (B3) is subject to very little un-
certainty or variation, since it is merely a matter of sound-waves
passing through the air (B2). If we represent this connection by
a dotted line, then we can symbolize the two human ways of
responding to a stimulus by these two diagrams:
speechless reaction: S19 ) R
reaction mediated by speech: SB )or sil >R.
The difference between the two types is evident. The speechless
reaction occurs always in the sume person as does the stimulus;
the person who gets the stimulus is the only one who can make
the response. The response, aecordingly, is limited to whatever
actions the recciver of the stimulus can make. In contrast with
this, the reacHon mediated by speech rnay oceur in a person who
did not gct the practical stimulus; the persan who gets a stimulus
can prompt another person ta make a response, and this other
person may he able ta do things which the speaker cannot. The
arrows in our diagrams rcpresent the sequencc of events within
one person's body - a sequence of events which we think is due
to some property of the nervous system. Therefore the speechlcss
reaction can take place only in the body whch received the stim-
ulus. In the reaction mediated by speech, on the other hand, there
is the lnk, by a dotted line, which consists of sound-
waves in the air: the reaction mediatcd by speech can take place
in the body of any person who hea-rs the speech; the possibilities
of reaction are enormously increased, since different hearers may
be capable of a tremendolls variety of acts. The gap between the
bodies of the speaker and the hearer - the discontinuily of the two
nervous systems - is bridged by the sound-waves.
The important things, biologically, are the same in both the
speechless and the speaking occurrence, namely S (the hunger
and sight of the food) and R (movements which Il;et the food or
fail to get it). These are the practical phase of the affair. The
speech-occurrence, s r, is merely a means by which Sand
R may occur in different individuals. The normal hurnan being: s
interested only in Sand R; though he llses speech, and thrives by
it, he pays no attention to it. Saying the ward apple or hearng
it said, appeases no one's hungcr. H, along with the l'est of speech,
is only a way of gctting one's fellow-men tD help. As students of
language, however, we are concerned prcciflely with the speech
event (s r), worthless in itself, but a means to great ends.
We distinguish between language, the subject of our study, and
real or practical evcnts, stimuli and reactions. When anything
apparently unimportant turns out to be closely connected with
more important things, we say that i t has, after aIl, a "meaning";
namely, it "means" these more important things. Accordingly,
we say that speech-utterance, trivial and unimportant in itself,
is important because it has a meaning: the meaning consists of
the important things with which the speech-utterance (B) is
eonnected, namely the practical events (A and C).
2.4. Up to a certain point, sorne animaIs respond ta each
others' stimuli. Evidently the marvelous co-ordination in a group
of ants or bees must be due to some form of interaction. Sounds
as a mcans for this are common enough: crickets, for instance,
cali other crickets by stridulation, noisily rubbing the leg against
the body. Sorne animais, like man, use vocal noises. Birds produce
sound-waves by means of the syrinx, a pair of organs at
the hend of the lungs. The higher mammals have a larynx, a box
of cartilage (in man called the adam's-applc) at the top of the
wind-pipe. Inside the larynx, at the right and left, two shelf-like
muscles l'un along the walls; when these muscles, the vocal chfYfds,
are stretched taut, the outgoing breath sets them inta a regular
vibration which produces sound. This sound we calI the voice.
Ruman speech differs from the signal-like actions of animaIs,
even of those which use the voice, by Hs great di fferentiation.
Dogs, for instance, make only two or three kinds of noise - say,
barking, growling, and whining: a dog can set another dog acting
by means of only these few different signaIs. Parrots can make
a great many kinds of noise, but apparently do not make different
responses ta different sounds. Man utters many kinds of vocal
noise and makes use of the variety: under certain types of stimuli
he produces certain vocal sounds, and his feIlows, hearing these
saille sounds, make the appropriate response. 1'0 put it briefly,
in human speech, different sounds have different meanings. 1'0
study this co-ordination of certain sounds with certain meanings
is ta study language.
This co-ordination makes it possible for man to interact with
great precision. When wc tell someone, for instance, the address
of a house he has never seen, we are doing something which no
animal can do. Not only has each person at his service the abilities
28 THE USE OF LANGUAGE THE USE OF LANGUAGE 29
of many other persons, but this co-operation is very precise. The
extent and accuracy of this is the measure of
success of our social organization. The term society or social or-
ganism is not a metaphor. A human social group is really a unit
of a highcr order than a single animal, just as a animal
is a unit of a higher arder than a single cell. The single ecUs in the
many-ceUed animal co-operate by means of such arrangements as
the nervous system; the individuals in a human society co-operate
by ITIeans of sound-waves.
The different ways in which we profit by language are so obvious
that we need mention only a few. 'Ve can relay communication.
\Vhen sorne farmers or traders say TVe want a bridge over this stream,
this ne",'s may pass through a town meeting, astate legislature, a
bureau of roads, an engineering staff, and a contractor's office,
running through many speakers and many rclays of speech, until
at last, in response ta the farmers' original stimulus, a corps of
workmen make the actual (practieal) rcsponse Illovernents of
putting up a bridge. Closely connected with the relay chamcter
of speech is its abstraction. The reluys of speech, between the
practical stimulus and the pmctical response, have no immediate
practical effect. Therefore they can be put into aU kinds of forms,
provided only one changes them back correetly before procecding
ta the final, practieal response. The engineer who plans the bridge
does not have to handle the actual beams and girders; he works
mercly with speech-forms (sueh as numbers in calculation); if he
makes a mistake, he does not destroy any materials; he need only
replace the ill-chosen speech-forrn (MY, a wrong figure) by a suit-
able one bcforc he the aetual building. ln this lies the value
of talking to oneself or Ihinking. As children, we talk to ourselves
aloud, but, under the correct.ion of our eIders, we SOn learn ta
suppress the sound-producing movements and replace them by
very slight inaudible ones: we "think in words." The usefulness
of thinking can he illustrated by the process of counting. Our
ability to estimate nurnbers without using speech, is extrcmcly
lmited, as anyone may sec by glancing, say, at a row of books on
a shelf. To say that two sets of abjects "have the same number"
means that if wc take one object from the first set and place it
next to one objeet of the second set, and keep on doing ths without
usng any abject more than once, we shaU have no unpaired ob-
jects left aver. Now, wc cannat always do this. The objects may
he too heavy ta move, or thcy may be in different parts of the warld,
or they may exist at differcnt times (as, say, a f1.ock of sheep before
and after a storm). Here language steps in. The numerals one,
two, three, four, and sa on, arc simply a series of words which we
have learned ta say in a fixed arder, as substitutes for the abave-
described process. Using thern, we can "count" any set of abjects
by plaeing them into correspondence (as mathemati-
cians caU it) with the saying one for one of the
abjects, two for another, three for the next, and so on, taking care
ta use each abject only once, until the abjects of the set are ex-
hausted. Suppose that when wc had said nineieen, there were no
more abjects left. Thereafter, at any time or place, we ean decide
whether any set of objects has the same nmnber as this first set,
by merely repeating the counting proeess with the new set. Matha-
maties, the ideal use of language, consists merely of elaborations
of this proeess. The use of numbers is the simplest and clearest
case of the usefulness of talking to oneself, but there are many
others. We think before we act.
2. O. The particular speech-sounds which people utter under
particular stimuli, differ among diffcrent groups of men; mankind
speaks many languages. A group of people who use the same
system of speech-signaIs is a speech-communily. Obviously, the
value of language depends upon people's using it in the same way.
Every member of the social group must upon suitable occasion
utter the proper speech-sounds and, when he hears another utter
these speech-sounds, must make the proper response. He must
speak intelligibly and must understand what others say. This
holds good for even the least cvilized communities; wherever we
tind man, he speaks.
Every ehild that is barn into a group acquires these habits
of speech and response in the first years of his life. This is doubt-
less the greatest inteUcctual feat any one of us is cver required to
perform. Exactly how children learn ta speak is not known; the
proc<.)ss seems ta be somcthing like this:
(1) Under various stimuli the child utters and repeats vocal
sounds. This seems ta be an inhcrited trait. Suppose he makcs a
noise which we may represent as da, although, of course, the
aetuaI movements and the resultant sounds differ from any that
are used in conventional English speech. The sound-vibrations ,-/
strike the child's car-drums whilc he keeps rcpeating the mave-
30 THE USE OF LANG UAGE THE USE OF LANGUAGE 31
ments. This results in a habit: whenever a similar sound strikes
his ear, he is likely to rnake these same re-
peating the sound da. This babbling trains him to reproduce vocal
sounds which strike his ear.
(2) Sorne person, say the mother, utters in the child's presence a
sound which resembles one of the child's babbling sylIables. For
instance, she says doU. When these sounds strike the child's car,
his habit (1) cornes into play and he utters his nearest babbling
syllable, da. We say that he is beginning to "imitate." Grown-ups
seern ta have observed this everywhere, for every language seems
ta contain certain nurscry-words which resemble a chld's babbling
- words like marna, dada: doubtless these got their vogue because
children easily learn ta repeat them.
(3) The mother, of course, USes her words when the appropriate
stimulus is present. She says doll when she is actualIy showing
or giving the infant his dolI. The sight and handling of the doU
and the hearing and saying of the ward doll (that is, da) occur
repeatedly together, until the child forms a new habit: the sight
and feel of the dolI suffice to make him say da. He has now the
use of a ward. Ta the adults it may not sound like any of their
words, but this is due merely to its imperfection. It is not likely
that children ever invent a word.
(4) The habit of saying da at sight of the doU gives rifle to further
habits. Suppose, for instance, that day after day the ehild is given
his doU (and says da, da, da) irnmediately after his bath. He has
now a habit of saying da, da after his bath; that is, if one day the
mother forgets ta give him the dolI, he may nevertheless cry da, da
after his bath. "He is asking for his doU," says the mother, and
she is right, since doubtless an adult's "asking for" or "wanting"
things is only a more eomplicated type of the same situation.
The child has now embarked upon abstract or displaced speech:
he names a thing even when that thing is not present.
(5) The child's speech is perfected by its results. If he says da,
da weil enough, his eIders understand him; that is, they give him
his dolI. When this happens, the sight and feel of the doU aet as
an addiUonal stimulus, and the child repeats and practiscs his
successful version of the word. On the other hand, if he says his
da, da imperfectly, - that is, at lVCat variance from the adults'
conventional form doU, - then his elders are not stimulated to
give him the doU. Instead of getting the added stimnlus of seeing
and handIing the dolI, the child is now subject ta other distraeting
stimuli, or perhaps, in the unaceustomed situation of having no
doli after his bath, he goes into a tantrum whieh disorders his
rceent impressions. In short, his more perfect attempts at speech
are likely ta be fortified by repetition, and his faitures to he wiped
out in confusion. This process never stops. At a much later stage,
if he says Daddy bringed il, he merely gets a disappoinbng answer
such as Nol You must say " Daddy broughl il",. but if he says Daddy
broughl il, he is likely ta hear the form over again: Yes, Daddy
broughl il, and to get a favorable practical response.
At the same Ume and by the same proeess, the child learns
also ta act the part of a hearer. While he is handIing the doli he
hears himself say da, da and his mother say doll. After a time,
hearing the sound may suffice ta make him handle the doU. The
mother will say Wave your hand lo Daddy, wheu the child is doing
this of his own accord or while shc is holding up the chld's arm
and waving it for him. The child forms habits of acting in eon-
ventional ways when he hcar;; speech.
This twofold character of the speech-habits hecomes more and
more unificd, since the two phases always oceur together. In
each case where the ehild learns the connection SB )or
(for instance, ta say doll when he sees his doll), he learns also the
connection sH ) R (for instance, t() reach for his doli
or handle it whcn he hears the word doll). After he has learned
a numher of such twofold sets, he develops a habit by which one
type always involves the other: as soon as he leams ta spcak
a new ward, he is also able ta respond ta it when he hears others
speak it, and, vice versa, as soon as he learns how to respond ta
sorne new word, he is usually able, ahm, to spcak it OII proper
occasion. The latter transferencc scems to he the more difficult of
the two; in Iater life, we find that a speaker understands many
speech-forms which he seldom or ncver employs in his own speech.
2.6. The happenings which in our diagram are represented by
a dotted line, are fairly weU understood. The speaker's vocal
chords, tangue, lips, and so on, interfere with the stream of his
outgoing breath, in such a way as ta prouce these
waves are propagated through the air and strike the hearer's
car-drums, which then vibrate in unison. The happenings, how-
ever, which we have representcd by arrows, are very obscure.
Wc do uot undcrstand the mechanism which makes people say
32
THE USE OF LANGUAGE
THE USE OF LANGUAGE 33
certain things in certain situations, or the mechanism which makes
them respond appropriatc1y when these speech-sounds strike their
ear-drums. Evidently these mechanisms are a phase of our gen-
eraI equipment for responding ta stimuli, he they speech-sounds
or others. These mechanisms are studied in physiology and,
especially, in psychology. To study them in their special bearing
on language, is ta study the psychology of speech, linguistic
psychology. In the division of scientific labor, the linguist deals
only with the speech-signal (r s); he is not competent to
deal with problems of physiology or psychology. The findings of
the linguist, who studies the speech-signal, will he aIl the more
valuable for the psychologist if they are not distoried by any pre-
possessions about psychology. We have seen that many of the
aIder linguists ignored this; they vitiated or skimped their reports
by trying ta state everything in terms of sorne psychological.
ory. We shaU all the more surely avoid this fault, however, If we
survey a few of the more obviaus phases of the psychology of
language.
The mechanism which governs speech must be very complex
and delicate. Even if we know a great deal about a speaker and
about the immediate stimuli which are acting upon him, we usu-
ally cannat predict whether he will speak or what he will say.
We took our story of Jack and Jill as something known to us,
after the facto Had we been present, we could not have foretold
whether Jill would say anything when she saw the apple, or, in
case she did speak"what words she would utter. Even supposing
she asked for the apple, we could not foretel1 whether she would
preface her reque8t by saying Tm hungry or whether she would
say please or whether she would say l want that apple or Get me
that apple or l was just wishing l had an apple, and 80 on: the
possibilities are almost infinite. This enormous variability has led
to two theorie8 about human conduct, including speech.
The mentalistic theory, which is by far the aider, and still pre-
vails bath in the popular view and among men of science, supposes
that the variability of human conduct i8 due to the interferencc of
sorne non-physical factor, a spirit or will or mind (Greek psyche,
hence the terni psychology) that is present in every human heing.
This spirit, according ta the mentalistic view, is entirely different
from material things and accordingly follows some other kind of
causation or perhaps nonc at aU. Whether Jill will speak or what
words she will use, depends, then, upon sorne act of her mind 01
will, and, as this mind or will does not follow the patterns of suc-
cession (causfl-and-effect sequences) of the material world, we
cannat foretell her actions.
The materialislic (or, better, mechanistic) theory supposes that
the variability of human conduct, including spech, is due only
to the fact that the human body is a very complex system. Human
actions, ta the materialistic view, are part of cause-and-
effect sequences l:'xactly like those which we observe, say in the
study of physics or chemistry. However, the human body is so
complex a structure that even a relatively simple change, such
as, say, the on the retina of light-waves from a red
apple, may set off some very eomplicated chain of consequences,
and a very slight difTerence in the state of the body may result in
a great difference in its response ta the light-waves. Wc could
foretell a person's actions (for instance, whether a certain stimulus
will lead him ta speak, and, if sa, the exact words he will utter) ,
only if we knew the exact. structure of his body at the moment, or,
what cornes to the same thing, if we knew the exact make-up of
his organism at some early stage sayat birth or before - and
then had a record of every change in that organism, including
every stimulus that had cver affeeted the organism.
The part of thc human body responsible for this delicate and
variable adjustment, is the nervOlls system. The nervous system
is a very complex condllcting mechanism, \vhich makes it possible
for a change in one part of the body, (a stimulus, say, in the eye)
ta result a change in sorne other part (a response, say, of reach-
ing with the arm, or of moving the vocal chords and tongue).
Further, it is deaT that the nervous system is changed, for a time
or even permanently, by this very process of conduction: our
responses depend very largely upon our earlier dealings with the
same or similar stimuli. Whether Jill will speak depends largely
on her liking for apples and on her past experience of Jack. Wc
remember and acquire habits and learn. The nervous system i8
evidently a trigg-er-mcehanism: a very slight change may set the
match ta a large store of explosive materia1. To take the case that
interests us, only 80 can we explain the fact that large-scale move-
rnents like J ack's fetching the apple, arc set off by very slight
changes, such aS the minute thrumming of s,ir-waves on his ear
drum.
34 THE USE OF LANG U AG E
THE USE OF LANGUAGE 35
The working of t.he ncrvous system is not accessible to observa
M
Hon from without, and the person himself has no sense-organs
(such as he has, for instance, for the working of the muscles in his
hand) with which he hirnself couId observe what goes on in his
nerves. Therefore the psychologist must resort to indirect methods
of approach.
2.7. One such met-hod lS expcriment. The psychologist submits
Ilumbers of people ta carefully prearranged stimuli under the
simplest conditions, and records their responses. Usually he also
asks these persons ta "introspect," - that is, to descri be as mucb.
as possible of what goes on insidc thcm when they get the stimulus.
At this point psychologists oftell go astray for want of linguistio
knowledge. It is a mistakc, for instance, to suppose that language
cnables a person ta observe things for whioh he has no sense-
organs, such as the workings of his own nervous syst-em. An ob-
server's only advantage in reporting what goes on inside him is that
he can report stimulations which an outsider cannot dctcct - say,
a pain in his eye or a tickling in his throat. Even here, we must
not forget that language lS a matter of training and habit; a
person may be unablc ta report sorne stimulat.ions, simpIy because
his stock of speech-habits provides no formula; this is the case
with many of our less UReful adventures, such as smaller goings-on
in our internai organs. Often t.he very structure of our body leads
to a false report.; we show the physician exactly the spot ,yhere we
feel a pain, and he finds the injury SOIlle distance away, ai a
point whioh his experience Illay teach him ta locate at once from
our false description. In this respect many psycholof';ists go astray
by actually training their observers to U8e a set of technical terms
for obscure stimuli and then attachinf'; significance to the observer's
use of these terms.
Abnormal conditions in which speech is disturbed, seem ta
refiect genel'al maladjustments or Icsions and to thro,y no light on
the particular mechanism of languagc. 8tulkring i8 probably due 1,0
imperfcct specialization of the two cerebral hemisphcl'es: in the
normal speaker the 1eft hemisphere (or, if he is left-handed, the
right hemisphel'e) dominates more delicate actions, such as those of
speech; in thc stutterer this one-sided specialization is incomplete.
Imperfect production of specifie sounds (stammering), wherc it iB
not due 1,0 anatomical defects in the organs-of speech, seem8 to
rcsult from similar maladjustments. Hcad-,younds and diseases
" !
which injure the brain often result in aphasia, disturbances in the
manner of making speech-responses and in responding 10 speech.
Dr. Henry Head, who had unusually good opportunities for the
study of aphasia in wounded soldiers, recognizes four types.
Type 1 reacts weil 1,0 other people's speech, and in milder cases,
uses words for the proper objects, but mispronounces or confuses
his words; in extreme cases, the sufferer can say Httle more than
yes and no. A patient reports, with sorne difficulty: "1 know it's
uot the correct pronunciatian l don't a1-
ways correl it because 1 shouldn't get it right
........ in five or six times unless someone says it for
me." In a more serious case, the patient, when asked his name,
answers IIonus instead of 'Thomas,' and says ersl for 'fi1'st' and
hend for' second.'
Type 2 reacts fairly weil 1.0 simple speech, and pronounces
appropriate words and short phrases, but not in the convcntional
constructions; he may ialk an unintelligi bIe jargon, although eaeh
ward is correct enough. Ta the question "Have you played any
games?" a patient answers: "Played games, yes, played one, day-
time, garden." He says, "Get out, lay down, go to sleep, sorne-
timcs gs away. If 8it in kitchen, moving about working, makes
me getting worse on it." He comment8, "Funny thing, this worse,
that sort of thing," and by way of explanation, writes down the
words as and at. "\Ve shaH see later that the structure of normal
language forces us to distinguish between lexical and grammatical
habits of speech; the latter arc disturbcd in thesc patients.
Type 3 reacts with difficulty to the names of objects, and has
trouble in finding the right wards, especially names of things. His
pronunciation and arrangement are good, but he has to use i n ~
genious circumlocutions for the words he cannot find. For' scissors'
a patient says "what you eut with"; for 'black' he says: "people
who are dead, - the other people who are not dead, have this
eolor." He may use the wrong word, as bulton for 'scissors.' The
words lost are chiefly the names of cancrete abjects. This state
seems like an exaggcration of many normal persans' difficulty in
recalling people's narnes and the designations of abjects, especially
under preoccupation, excitement, or fatigue.
Type 4 often does not respond correctly ta the speech of others;
he has no trouble in uttering single words, but he cannat finish a
connected speech. It s signficant that these patients suffer from
36 THE USE OF LANGUAGE THE USE OF LANGUAGE 37
apraxia; they cannat find their way about and are confused by
being set, say, on the opposite side of the street. One patient
reports: "1 don't seem to understand ail you say, and then 1 forget
what l've got to do." Another patient says: "When at table, 1
am very slow in picking out the objeet, say the milk-jug, which 1
want. 1 don't spot it at once ... 1 sec them aIl, but 1 don't
spot them. When 1 want the salt or the pepper or a Hpoon, 1
suddcnly tumble to its presence." The disturbance of speech
appears in this answer of a patient: "Oh, yes! 1 know the differ-
ence betwecIl the Nurse and the Sister by the dress: Sister bine;
l\'urse - oh! 1 get muddled, just ordinary nurse's c10thes, white,
blue ..."
Ever since 1861, when Broca showed that damage ta the third
frontal convolutian in the left hemispherc of the brain \VaR accom-
panied DY aphasia, there has been dispute as to whether "Broea's
center" and other regions of the cortex act aS specific centers for
the activity of Hpecch. Head finds some correlation between
diffcrent points of leson and each of his four types of aphasia. The
demOllstrable functional identifications of cortical arcas always
concern sorne specific organ: an injury in one area of the brain is
accompanied by paralysis of the right foot, an injury in another
area by failure ta respond to stimulation in the left-hand side of the
retina, and so on. Now, Hpeech is a very complex activity, in which
stimulation of every kind leads to highly Hpecific movcrnents of
the throat and mouth; these !ast, moreover, are not, in a physi-
ologic sense, "organs of speech," for they serve biologically earlier
uses in man and in spcechless animuls. Many injuries to the
ous system, accordingly, will interfere with speech, and different
injurics will result in different. kinds of difficulty, but the points of
the cortex are surely not correlated with specific socially significant
feat ures of spcech, such as words or syntax; this appcars plainly
from the fluctuating and contradictory results of the search for
various kinds of" speech centers." We lIlay expect the physiologist
to get better results when he looks for correlations octween points
of the cortex and specifie physiologic activities concerned in
speech, such as the movement of special muscles or the transmission
of kinesthetic stimuli from the larynx and tongue. The error of
seeking correlations bctween anatomically defined parts of the
nervoua systrn and socially defined activites appoars clearly when
wc sec sorne physiologists looking for a "visual word*center" which
is ro control reading and writing: one might as weIl look for a
specific brain-center for telegraphy or automobile-driving or the
use of any modern invention. Physiologieally, language is not a
unit of function, but consists of a great many activities, whose
union into a single farAreaching complex of habits results from
repeated stimulations during; the individual's ear!y life.
2.8. Anothcr way of studying human responses is to observe
them in the mass. Some actions are highly variablc in cach per-
son, but fairly constant in large groups of persons. Wc cannot
predict whether any particulaI' unmarried adult will marry dur-
ing the next twelve months, or w11ich particular persons will
commit suicide, or which ones will bret into prison, but, given a
large enough comrnunity, and the figures fol' past years (and per-
haps certain other data, such as those which concern economic
conditions), statisticians can foretell the number of marriagea,
suicides, convictions for crime, and so on, whch will take place.
If we found it possible and wort.h while ta regist.er every
utterance in a large community, wc should doubtless be ablc t.o
foreteU how many times any given utterance such as Good-morning
or 1 love you or Il ow much are oranges torlay? would be spoken
within a fixed number of days. A detaled study of this kind
would tell us a great deuI, cspccially about the chang;es that are
constant!y going on in every lang;uage.
However, there is another and simpler way of studying human
action in the rnass: the study of conventional actions. 'Vhen we
go to a strange country, we saon learn many established modes of
action, such as the system of currency and of weig;hts and meas-
ures, the mIes of the road (docs one keep to the right, as in Amer-
ica and Germany, or to the left, as in England and Swedcn?),
good manners, hours for meals, and so on. The traveler docs Ilot
gathcr statistics: a very few observations put him on the track,
and these are confirmed or corrected by further experience. Here
the linguist is in a fortunate position; in no other respect are the
activities of a group as rigidly standardizcd as in the forms of
language. Large groups of people make up aIl their utteranccs
out of the same stock of lexical forrrlS and grammatical construc-
tions. A linguistic observer thercfore can describe the speech-
habits of a community without resorting to statistics. Need!ess
to say, he must work conscientiously and, in particular, he must
record every form he can find and not try ro excuse hirnself from
38 THE USE OF LANG UAGE
'f
THE USE OF LANGUAGE 39
this task by appealing to the reader's cornmon sense or to the
structure of sorne other language or ta sorne psychological thcory,
and, above ail, he must not select or distort the facts according
to his views of what the speakers ought to he saying. Aside from
its intrinsic value for the study of languag;c, Il, relevant and Un-
prejudiced description of this kind, serves as Il, document of major
importance for psychology. The danger here lies in mentalistic
views of psychology, which may tempt the observer ta appeal to
purely spiritual standards instead of reporting the facts. To say,
for instance, that combinations of words which are "feIt to be"
compounds have only a single high stress (e.g. blw:kbird as opposcd
to bkuk bird), is t tell exactly nothing, since wc have no way of
determining what the speakers may "feel": the observer's task
was to tdl us, by sorne tangible criterion, or, if he found none, by
a list, which combinations of words are pronounced with a single
high stress. A worker who accept.s the materialistic hypothesis in
psychology is under no such temptation; it may he stated as a
principle that in al! sciences like linguistics, which observe sorne
specific type of hurnan activity, the worker must proceed exactly
as if he held the materialistic view. This practical effectiveness is
one of the strongest considerations in favor of scientific materialism.
The observer who, by this mass-observation, gives us astate-
ment of the speech-habits of a community, can tell us nothing
about the changes whch are going on in the language of this as
of every community. These changes could be observed only by
means of genuinely statistical observation through a considerable
length of time; fol' want of this, we are ignorant. of many matters
concernin/!; linguistic change. In this respect, tao, the science of
language is fortunate, howewr, because comparative and geograph-
ical methods of study, again through mass-observation, supply a
good deul of what we should hope t.o get from statistics. The
fortunate position of our flcience in these mattNs is due to t.he
fact that lang;uage is the sirnplest and most fundamental of our
social (that is, poculiarly hurnan) activities. In another direction,
however, the study of linguistic change profits by a mere accident,
namely by the exist.ence of writien records of speech of the past.
2.9. The stimulus which caUfI forth speech, leads a1so to sorne
other reactions. Sorne of thcflc are not visible from the outsidc;
these are muscular and glandular actions which are of no imme-
diat importance to the speakor's Others arc impor-
tant handIing responsos, such as locomotion or the displacement
of abjects. Still other rcsponses are visible, but not directly im-
portant; they do not change the lay-out of things, but they do,
along with speech, serve as stimuli to the hearer. These actions
are facial expression, mimicry, tone of voiee (in so far as it is not
prescribed by the conventions of the language), insignificant
handling of objects .(such as fiddling with a rubber band), and,
above ail, gesture.
Gesture accornpanies al! speech; in kind and in amount, it
differs with the individual speaker, but ta a large extent it is
governed by social convention. Italians use more gest.ure than
English-speaking people; in our civilization people of the privileged
class gesticulate least. '1'0 sorne extent, individual g;estures are con-
ventional and differ for different, communities. In saying
wc wave the hand with palm outward; Neapolitans wave it with
the back outward.
Most gestures scareely go beyond an obvious painting and pic-
turing. American Indians of plains or woodland tribes will ac-
cvmpany a story by unobtrusve gcstures, foreign ta us, but quitc
intellig;ible: the hand, pa]rn in, thumb up, is hcld just undor the
eyes to represent spying; a fist is slapped into a palm for a shot;
two fingers imitate a man walking, and four t,he running of a horse.
Even \vhere gestures are symbolic, they go Jttle beyond the ob-
vious, as when one points back over one's shoulder to indicate
past timc.
Sorne communities have a ges/ure language which upon occasion
they use instead of speech. Such gcst.ure languages have bcen
observed among t.he 10wer-c1ass Neapolit.ans, among; TmppiHt
rnonks (who have made a vow of silence), among the Indians of
our weRtorn plains (where tribes of different. language met in
commerce and war), and among groups of deaf-mutcs.
It secms ccrt,ain t.hat these gest.ure languages are merely de-
velopments of ordinary gest.ures and that any and aIl complicated
or not imrncdiately intelligible gestures are based on t,he conven-
tions of ordinalJ' speech. Even such an obvious transference as
pointing backward to indicate past time, is probably due to a lin-
guist.ie habit of using the same word for 'in the rear' and 'in the
past.' "\Vhatever may be t.he origins of the two, gesture has sa long
played a secondary rle under the dominance of langua?;c th11.t it
has lost ail t.races of independcnt chamcter. Tales about peoples
40 THE USE OF LANGUAGE
THE USE OF LANGUAGE
41
whose language is sa defective that it has ta be eked out by gesture,
are pure myths. Doubtless the production of vocal sound by ani-
mais, out of which language has grown, originated as a response-
movement (say, contraction of the diaphragm and constriction
of the throat) which happened ta produce noise. It seems certain,
however, that in the further development, language always ran
ahead of gcsture.
If onc gcstures by moving sorne object sa as to leave a trace on
another abject, one has entered upon marking and drawing. This
kincl of reaction has the value of leaving a permanent mark, which
may serve as a stimulus repeatcdly and even after intervals of
time and can be transported ta stimulate persans far away. For
this reason, doubtless, many peoples attribute magic power ta
drawings, apart from thcir esthetic value, which is still "with us.
In sorne parts of the world drawing has developed into writing.
The details of this process will concern us latcr; the point of in-
tcrcst here is that the action of tracing an outline becomes suh-
ordinate ta language: drawing a particular set of lines becomes
attached, as an accompaniment or substitute, to the utterance of
a particular linguistic form.
IThe art of symbolizing particular forms of speech by means of
particular visible marks adds a great deal to the effective uses of
language. A speaker can be heard only a short ways and only for
an instant or two. A written record can be carried to any place
and preserved for any length of time.! \Ve can see more things at
one time than we can hear, and we can deal better with visible
thin!-;s: charts, diagrams, written calculations, and similar clevices,
enable us to deal with vcry complex matters. The speech-stimuli
of distant people, and espccially of persons in the past, arc available
to us through writing. This makes possible an accumulation of
knowledge. The man of science (but Dot always the amateur)
surveys the results of earlier students and applies his energics at
the point where they left off. Instead of always starting over
again from the bCl';inning, science progresses cumulatively and
with acceleration. It has been said that, as we preserve more and
more records of more and more speech-reactions of highly gifted
and highly specialized individuals, we approach, as an idcallimit,
a condition where aIl the events in the universe, past, present, and
future, are reduced (in a symbolic fonu to whieh any reader may
react) to the dimensions of a large library. It is no wonder that
the discove;ry of printing, which manifolds a written record ta any
desired numbcr of copies, brought about, in aH our manner of
living, a revolution which has been under way for sorne centuries
and is still in full swing.
There is no need of dilating upon the significance of other means
for recording, transmitting, and multiplying speech, such as the
telegraph, telcphone, phonograph, and radio. Their
for the simpler uses of language is obvious, as in the use of wrreless
telegraphy in cases of shipwrcck.
In the long run, anything whieh adds to the viability of lan-
guage has also an indirect but more pervasive effect. Even acts
of speech that do not prompt any particular immediate response,
may change the predisposition of the hearer for further responses:
a beautiful poem, for instance, may make the hearer more
tive ta larer stimuli. This general refinement and intensification
of human response requres a grcat deal of linguiste interaction.
Education or culture, or whatever narne we cho08e ta give it, de-
pends upon the repetition and publication of a vast amount of
speech.
l havc said nothing about biological grouping, because
does not, like the other groupings, depcnd upon language for lts
existencc. Most matings, of course, take place between
of like speech, so that a speech-cornmunity is always somethmg
of an inbred group; the exceptions, however, are very many, both
in the mat,ing of persons of different speech, one of whom usually
acquircs the other's language, and, what is more important, in
the assimilation into a of whole groups of for-
eigncrs, such as immigrants, conquered people, or captives. These
deviations are so many that, if we had records, we should doubUess
tind very fcw persons whose ancestors of a few generations aga aIl
spoke the same language. What concerns us most, however, is
the fact that the features of a language me not inherit.ed in the
biologie sense. ,A child crics out at birth and would doubtless in
any aft.er a time take to gurgling and babbling, the par- ,
ticular language he learns is entirely a matter of envlronrnent.
,'\ An infant that gets into a group as a foundling or by adoption,
learns the of the group exactly as does a child of native
parentage; as he learns to speak, his language shows no trace of
whatever language his parents may have spoken. Vlhatever
hercditary differences there may be in the structure of the larynx,
mouth, lips, and so on, of normal human beings, it is certain that
thcsc differcnces are not such as to affect the action" which rnake
up language. The child learns ta speak like the persons round him.
The rust language a hurnan being learns to speak is his native
wnguage; he is a native speaker of this language.
3. 2. Speech-communitics differ greatly in size. More than one
American Indian tribe of only a few hundred persans spoke a
language of its own. On the other hand, even before the
of modern communication and travel, sorne speech-commumhes
were very large: in the first centuries of the Christian Era, Lg,tin
and Greek were each spoken by millions of people over large areas
round the :?I.lediterranean. Under modern conditions, sorne speech-
cornmunities have grown to enormoUS size. Jespersen estimates
the number of speakers of the principal European languages, in
millions, for the years 1600 and 1912 as follows:
GERMAN R lJSS1AN FRENCF\ SPANISH ITAL!AN
SPEECH-COMM UNITlES
CHAPTER 3
SPEECH-COMM UNlTlES
3. 1. A speech-community is a group of people who interact by
means of speech ( 2.5). AlI the so-called higher activities of man
- our specificalIy human activities - spring from the close ad-
justment mong individuals whieh we calI society, and this ad-
justment, in turn, is based upon language; the speech-community,
therefore, is the most important kind of social group. Other
phases of social cohesion, such as economic, political, or cultural
groupings, bef1r some relation to the grouping by speech-commu-
nities, but do not usually coincide with it; cultural features,
cially, are almost always more widespread than any one language.
Before the coming of the white man, an independent lndian tribe
wm.ch spoke a language of its own, formed both a speech-cammu-
nity and a political and economic unit; as to religion and general
culture, however, it resembled neighboring tribes. Under more
complex conditions there is less correltion between language and
the other groupings. The specch-community which consists of
ail English-speaking people is divided inta two political
nities: the United States and the British Empire, and each or'
these is in turn subdivided;'econornically, the United States and
Canada are more closely united than politically; culturally, we
are part, of a great area which radiates from western Europe. On
the other hand, even the narrowest of these groups, the political
United States, includes persons who do not speak English: Amer-
ican Indians, Spansh-speakers in the Southwest, and linguistically
linassimilated immigrants. Colonial occupation, as in the Philip-
pines or India, puts a specch-community into political and eco-
nomic dependence upon a foreign spcech-comrnunity. In sorne
countries the population is divided iuto several specch-communities
that exist togcther without local division: a town in Poland con-
sists of Polish-spcaking and German-spcaking people; by religion,
the former are Catholics, the latter Jews, and, Imtil qui; recently,
vcry few persons in eit.her group troubled themselves to under-
stand the other group's language.
42
,
i
1
1
i
1
!
[.
1600
1912
ENGLISlI
6
150 -
la
90
3
106
14
47
8-}
52
43
94
37
44
SPEECH-CO::\1.:vI UNlTIES SPEECH-CO M M UN ITIES 45
Figures such as thesc have only a very indefinite value, because
one cannot always tell which local groups forrn a single
community. Tesnirc, estimating the numbers round the year
names Chinese as the largest with 400
mIlhon speakers, but the term Chinese denoks a family of mutually
unintelligible languages. Doubtless one of these, North Chinese
has today more native speakers than any other lanKllage, but i
know no estimate of their numher. Another of this
group, probably ranks among the largest speech-
commumhes. In any case, English (to continue with Tcsnire's
figures) rank" second, with 170 million native speakers. Russian
cornes third; divides the figures between Great Russian
(Sa millions), Little Russian (Ukrainian, 34 millions) and "Vhite
Russian milliom;), but these are illutually intelligible varieties
as as British and American English. Similarly;
Tesmere sphts the language, German, into
man (SO. millions) and Judeo-German (70} millions), although the
rest of h18 figures do not considcr dialectal differenees; Jespersen's
figure of 90 millions is probably right. Tesnire's
ing figures omit Javanese, which has at reast 20 millions of
speakers. Wi th these modifications his figures are: Spanish 65,
Japanese 55, Bengali) 50, French 45, l talian 41, Tartur
3?, !lind 1 38, Arabic 37, Bihari 1 36, Portugucse 36,
l',astern Hmdl 1 25, Telugu 224, Polish 23, Javancse 20, IVlarathi 1
19, Ta:Tlii 2 19: Rorcan 17, Panjabi 1 16, Annamite 14, Roumanian
14,. H.aJastham ) 13, Dutch 13, Bohemian-Slovak 12, Canarese 2 10,
OrIya 1 10, Hungarian 10.
Another element of uncertainty in figures like these arises from
the differences within Duteh and German
actually form only one in the sense that there
is no break local but the extremc types are
mutual1y unmtelhp;lble, and the political groups (on the one side
Flemish Belgiun:- and the NetherJands, and on the ot.hcr side,
Germany, AustrIa, and German Switzerland) have adopted two
mutually unintelligible spcech-forms, Standard and
Standard German, as their official languages. On the other hand
and sorne of the languages of India in our list
languages spoken in India; We should perhaps add GuJerati
Wlt SOrne 10 mJ1hon speakers. '
Dravidian languages spokcn in Indis.
';
,
,
L
ably include equally great differences, although the extremes may
he connected by local gradations. A final and insurmountable
culty lies in people's acquisition of foreign languages. If we could
determine a degree of proficiency which makes a student a member
of a foreign English, studied ail over the world,
would receive a much larger figure. Tesnire estimates that Malay
is native to sorne three million people, but is spoken as a foreign
language, espeeia11y in commerce, by sorne thirty millions.
3.3. The difficulty or impossibility of dctermining in each case
exactly what people bclong to the same is
not accidentaI, but arises from the very nature of
communities. If wc observed closelyenough, we should find that
, no two persons - or rather, perhaps, no one person at different
,\ -'. tmes - spoke exactly alike. 1'0 bc sure, within a re1atively homo-
geneous set of speakers - say, the native speakers of English in
the Middle 'Vestern part of the Vnited States - the habits of
speech are far more uniforrn than the nceds of communication
would demand. \Vc see the proof of this when an outsider - say,
a Southerner or an Englishman or a foreigner who has mastered
English - comes into our midst: his speech may be so much like
ours as to cause not the slightest difficulty in communication, and
yet strikingly noticeable on aeeount of inesscntial differences,
such as "accent" and "idiom." Keverthcless there are great
differences even among the native memhcrs of such a relativcly
uniform group as Middle Western Amcrican, and, as we have
just sccn, cven greater differcnces within a speech-community
(e.g. English) as a whole. These differences play a very important
part in the history of languages; the linguist is forced to consider
them very carcfully, even though in sorne of his work he is forced
provisiona11y to ignore them. When he does this, he is merdy
employing the method of abstraction, a mcthod esscntial to
scientific investigation, but the results so obtained have to be
correetcd bcfore they can he used in most kinds of further
work.
Thc difference hetween speakers is partly a matter of bodily
makc-up and perhaps of purely personal habit; we recognize our
fricnds by their voiees from the next room and over the tdephone.
Sorne people are more talentcd for speech than others: they remem-
ber more words and turns of phrase, apply them better w the
situation, and combine them in more pleasing style; the extreme
46
SPEECH-COM MUNITIES SPEECH-COM M UN l'fI ES 47
case is the literary genius. Sometimes convention assigna certain
speech-forms ta certain speakers, as when the soldier, the well-
trained servant, and the chiId in certain sch001s, learn ta say sir
or ma'm ta certain persans, who do not reciprocate. Sorne
clamations, such as Goodness grooous! or Dear me! are largely
served for the use of women. In sorne communities very differM
ent speech-forms are conventional for the sexes. The classical
instance is that of the Carib Indians; a recentY--lluthenticated one
is the language of the Yana Indians in northern California. Ex-
amples of Yana wards are:
The difIerences between the two sets of Yana forms can he stated
by means of a fairly complex set of rules.
3.4. 'The most important differences of speech within a com-
munity are due ta difIeronces in densily of communication. ; The \
learns to speak like the people round him, but we must not
pieture this learning as coming ta any particular end: there is
no hour or day when we can say that a persan has finished learnM
ing ta speak, but, rather, ta the end of his life,\the speaker keeps
on doing the very things whieh rnakc up infantile language-Iearning.\
Our description of the latter ( 2. 5) might he taken, in many"
respects, as a slowMmotion picture of the ordinary processes of
speech. Every speaker's language, exccpt for personal factors
which we must here ignore, is a composite result of what he has
heard other people say.
Imagine a huge chart with a dot for every speaker in the com-
munity, and imagine that every time any speaker uttercd a
tence, an arrow were drawn into the chart pointing from his dot
to the dot representing each one of his hcarers. At the cnd of a
given period of time, say seventy years, this chart would show us
of communication within the community. Sorne speak-
ers would -eurn out to have heen in close communication: thcre
would he many arrows from one ta the other, and there would be
many series of arrows connecting them by way of one two or
three intermediate speakers. At the other extrem there he
MEN'S LANGUAGE ,,"'OM01N'S LA:N"GUAGE;
'tire'
'my tire'
'deer'
'grizzly-bear'
'auna
'allnija
bana
t'en'na
'allh
'all'nich'
ba'
t'et'
widely separated speakers who had never heard each other speak
and were connccted only by long chains of arrows through many
intermeJiate speakers. If wc wanted to explain the likcness and
unlikeness between various speakers in the community, or, what
cornes ta the same thing, to predict the degree of likene8s for any
two given speaker8, our tirst step v-lOuld be ta count and evaluate
the arrows and series of arrows connecting their dots. We shaH
sec in a moment that this wouJd bc only the first step; the reader
of this book, for instance, is more likely ta repeat a speech-form
which he has heard, say, from a lecturer of great fame, than onc
which he has heard from a street-sweeper.
The chart we have imagined lS impossible of construction. An
insurrnountable diffieulty, and the most important one, would he
thc factor of timc: 8tarting with persans now alive, we should be
compelled ta put in a dot for every speaker whose voice had ever
reached anyonc now living, and thcn a dot for every speaker whom
these speakers had ever hcard, and so on, back bcyond the days
of King Alfrcd the Great, and beyond earlicst history, back in
M
definitdy into the prmeval dawn of mankind: our speech depcnds
cntrely upon the speech of the pasto
Since we cannat construct our chart., we depend instcad upon
the study of indirect results and are forced ta resort. ta hypoth-
esis. Wc believe thah;the differences in density of
tian within a, are not only personal and in
M
dividual, but that the cornrnunity is divdcd into various systems
of sub-groups such that the persans within a sub-group speak
much more to each other than to personsoutside their
Viewing the system of arrows as a network, we may say that
these are separated by lines of U'eakness in this net of
oral communication. The lines of wcakness and, accordingly, the
difference8 of speech vl'ithin a specch-eommunity are wcal-- due
ta mere g;cographic and non-local, or as we usually
say, social. In countries over which a speech-community. has
recently spread and settlcd, the Ilcal .. differences arc relatIvely
small, as, say, in the United States (especially the western part)
or Russia - in countries that have been lonK settled by the
, .. --- .. - ._-_.-
speech-community the local differcnce.s are as, say,
in England, where English has becn spokcnJi:- sorne 1500 years,
or in France whcre Llltin (nmv callcd French) has bcen spoken for
two-thousand years.
48 SPEECH-C0 M MUN1T 1ES
S PEEC H - C 0 l\1 :\1 UNI T 1E 8
49
3.5. We shaH examine tirst the simpler case, as it appears in
the United States. The most striking !ine of clcavage in our speech
is one of social class. Children who are born into homes of priv-
ilege, in the way of wealth, tradition, or education, become native
speakers of what is popularly known as "good" Enp;lish; the
linguist prefers to give it the non-committal name of standard
English. Less fortunate children become native speakers of "bad "
or "vulgar" or, as the linguist prefers to cali it, non-standard
English. For instance, 1 have none, 1 haven't any, 1 haven't got any
are standard (" good ") English, but 1 ain't got none is non-standard
(" bad") English. --
These two main types of American Eng-lish are by no means
treated alike. The standard forms are used!n schoo1, in .9purch,
and in aIl discciurse that officially concerns the wh01e community,
as in raw-courts and legislative assemblies. Ail our writing (except
hy way of jest) is based on the standard forms, and theoo forms
are registered in grammars and dictionaries and presented in
text-books to foreigners who want to learn our language, Both
groups of speakers, standard and non-standard, agree in cal1ing
the standard forms "good" or "correct' , and non-standard forms
"bad," "incorrect," dy.u.lgltr," or even, "not English." The
'\. speaker of standard English does not trouble himself to learn t,he
non-standard forms, but very many speakerl" of non-standard Eng-
lish try to use the standard forms. Anative ofthe less favored group
who aequires prestige, say, in the way of wealth or politieal emi-
nence, is almost sure to learn, as well as may be, the standard form:;::
of :;::peechj in fact, noticeab1e lapses in this respect -- even a single
1 seen it or l done i/- may endanp;er his newly acquired position.
Within the standard language there are minor diffcrcnces. In
this case again, the divergent forms are estimated as higher and
lower. A Chicagoan, for instance, who uses the ah-vowcl of father
instead of the more cornmon a-vowel of man in words likc laugh,
halJ, bath, dance, can't, is said ta he speaking a "higher-class"
kind of English. In cases like these, however, people's attitudes
differ: many Chicagoans find thesc ah-forms silly and affected,
Speakers of standard English often disput.e as to which of two
forms is "better"; it's 1 or il' s me, forehead or "forrid." 8ince the
disputants do not trouble themselves to agrcc on a definition of
"better," these disputes never' Teach any conclusion. This is fi,
matter which will occupY us again.
Within the standard Jang:uage, further, thcre arc differences
that obviously dcpend upon density of communication: differcnt
economic classes, --' say, the very rich and the so-called "middle
class ,. in its various g:radations, - differ in speech. Then thcre
are differcnces of education, in the way both of family tradition
and of schooling. These differcnces arc crossed by less important
divisions of technical occupation: different kinds of craftsmen,
merchants, engineers, lawycrs, physicians, scientists, artists, and
so on differ somewhat. in speech. Sports and hobbies have at
least own vocabulary. The factor of age-groups will concern
us laterj it is a tremcndous force, but work<! almost unseen, and
scarccly appe[l,rs on the levcl that now conccrns us, except perhaps
in young peoplc's fOIHiness for slang. .
The most stable and striking diffcrenccs, even in the Umted'
States and evcn in our standard lang:uage, are geographic. ln the
United 8latcs we have thrce great geographic types of standard
English: New England, Central-Western and Southcrn. "\Vithin
these types t hcre are sml111er 10cl11 clifferenees: speakers of standard
EIl"lish from older-settled P:1TtS of the country can oftcn tell a
home within faid}' narroW limits, ln matters of
pronunciation, especially, the rangc of <ltandard Enp;lish in America
is \vide: greatly different prolluneiations, such as those, say, of
North Carolina and are aceepted equally as standard.
On1y from the stage do wc 11 unifol'ltl pronunciation, and
here our aetol'S use a British type mthe. than an American. In
England there arc <limilar regional types, but. they are not granted
equ'al value. The highest social l'ecogni'li on is given to the "puhlic
school" English of the <lOUHL The innurnerable gradations from
this t;oward the decidedly provincial types of standard, enjoy less
prestige as they depart from the most favored type, The social
recognition of a speaker of st:lIldard Ellglish from Sco'l1and or
Yorkshire or Lancashire, depends in part upon hm,,- closely his
pronunciation approaches the upper-class southern type. In
Eng1and, but scarce1y in the United 8tat,es, provincial colorings of
standard English are tied up with dfferences of soc.ial leve!.
',' Non-standard speech shows grcater variety than standard.
The higher the social position of the non-standard speaker, the
more ncarly does ho approach the standard top
/ are the transitional speakers who use an almost standard form of
speech, v,ith only a sprinkling of non,,:st.andard forms, and pcrhaps
50
SPEECH-COMM UNITIES SPEECH 0 MMUNI TI ES 51
a pronunciation with too provincial a twang, At the hottom are
the unmistakably rustic or proletarian speakers who make no
pretense at using standard forms.
Apart from this continuous gradation, various groups of nOn-
standard speakers have their own speech-forms.
groups, such as fishermen, dairy workers, bakers, brewers, and so
have, at any raie, their own tcchnical language. Especially,
mmor groups who are in any way cut off from the g;reat mass use
clearly-marked varieties of speech. Thus, sea-faring men use'd ta
speak their own type of non-standard English. and sorne
of law-breakers have many speech-forms of their own; sa
do Clrcus people and other wandering entertainers, Among nOn-
standard speakers of German, Christians and Jews and in sorne
places Catholics and Protestants, differ in many of ;heir linguistie
forms, If the special group is at oclds with the rest of the commu-
nity, it may Use its peeuliarities of speech as a secret dialect as do
the English-spcaking Gipsies. Criminals in various have
deyeloped such secret dialects.
---The greatest diversity in non-standard speech however is
geographic, The geographic differenccs, which we even in'the
s.tandard English of the United States, are more audible when we
hsten ta non-standard speakers, In remote districts within the
older-settled parts of the country these local characteristics are
very pronounccd, ta the point where wc may describe thern as
local dialects.
In older-settled speech-communities, the type exemplified by
F,rance, or by the British part of the English-speaking group, local
dlaleets play a much greater part. In such communities the/nOD-.
standard language can he dividcd, roughly, ta be sure, and without
a sharp demarcation, into sub-standard speech, intelligible at least
"\ though not uniform, throughout the country and local dz'alec/
differs, from place to place ta such an that
hvmg sorne dIstance apart may fail ta understand each other.
standard speech, in such countries, belongs ta the "lower mfd:ie-
class," - ta the more ambitious ci-r .--
city workmen, - and the local dialects are by the
and the poorest people of the towns.
Yrhe local dialects are, of paramount importance ta the linguist,
not merely because thclr great variety givcs him work ta do but
because the origin and history of the standard and sUb-standard
types of speech can be understood only in the Iight of the local
dialects. Especially during the 1ast decades, linguists have come
ta sec that dialect geography furnishes the key to many problems.
In a country like France, Italy, or Germany - better studied in
this respect than England - every village or, at most, evcry
group of two or threc villages, has its own local dialect. The differ-
eoces between neighboring local dialects are usually sman, but
recognizable. The villagers are ready ta tell in what way their
neighbors' speech differs from theirs, and often tease their neighbors
about these peculiarities. The difference from place to place is
small, but, as one travcls in any one direction, the differences
accumulate, until speakers, say from opposite ends of the country,
calmot understand each other, although there is no sharp line of
linguistic demarcation between the places whero they live, Any
such geographic area of graduai transitions i8 called a dialect area.
Within a dialect area, we can draw lines bctween places which
differ as ta any feature of languag;e. Such Hnes are called isoglosses,
If a village has sorne unique peculiarity of speech, the isogloss based
on this peculiarity will he simply a Hne round this village. On the
other hand, if sorne peculiarity extends over a large part of the
dialect aroa, the isogloss of this feature will appear as a long line,
dividing the dialect area into two sections. In Germany, for in-
stance, the northern dialects pronounce the word bite with ai-sound,
as we do in English, but the southern dialeets pronounce it with an
s-sound (as in standard German the isoglm,s which sepa-
rates these two forms is a long and very irregular line, running cast
and west across the whole German speech area. In the north and
northeast of England one can mark off an area where the past tense
of bring has the form brang. Dialect collections of maps of
a speech area with isoglosses ddwn in, arc an important tool for the
linguist.
The speakers' attitude toward local dialects differs somewhat in
different countries. In England the local dialects have little pres-
tige; the upper-class speaker does Dot bother with them and the
native speaker of a local dialect who rises sociaUy will try to cast
it off, even if only in exchange for SOIlle form of sub-standard
speech. The Germans, on the other hand, have developed, within
the last century, a kind of romantic fondness for local dialects.
While the middle-class speaker, who is not quite sure of his social
position, will 8hy away from them, sorne upper-class Gerrnans make
52 SPEECH-COMMUNITIES
SPEECH-COMM UNI'fIES
53
it a point ta speak the local dialect of their home. In German
Switzerland this goes farthest: even the upper-class Swiss, who is
familial' with standard German, uses local dialect as the normal
medium of communication in his family and with his neighbors.
3. 7. The main types of speech in a complex speech-community
can he roughly classed as follows:
(1) lilerary standard, used in the most formaI discourse and in
writing (example: 1 have none);
(2) colloquial standard, the speeeh of the privileged class (ex-
ample: 1 haven't any or 1 haven't got any - in England only if
spoken wi th the southern "public school" sounds and intonation) ;
(3) provincial standard, in the United States probably not to he
differentiated from (2), spoken by the" middle" class, very close
to (2), but differing slightly from province t province (example:
1 haven't any or 1 haven't gol any, spoken, in England, with sounds
or intonations that doviate from the "public school" standard);
. (4) sub-standard, clearly diffcrent from (1), (2), and (3), spoken
III European countries by the "Iower mddle" class, in the United
States by almost aIl but the speakers of type (2-3), and differing
topographically, without intense local diffel'ence (example: 1 ain'l
gol none);
(5) local dialect, spoken by the lcast privileged class; only slightly
developed in the United States; in Switzel'land used also, as a
domestic language, by the other classes; differs almost from village
to village; the varieties sa great as often to he ineomprehensible ta
each other and ta speakers of (2-3-4) (Example: a hae nane).
3.8. Our survey of differences within a speech-community has
shawn us that the rnemhers of a 8peech-community may speak
so much alike that anyone can understand anyone eIse, or may
differ sa much that person8 who live sorne distance apart may
fail ta understand each other. The former case is illustrated by
an Indian trihe of a few hundred persans, the latter by a far-
ftung speech community like English, where an American and a
dialeet-speaking X,orkshireman, for instance, do not
each other's speech.''.Attnrt1y, however, wc can draw no line
between the two cases, because there are aIl kinds of gradations
between understanding and failing ta understand. Whcther the
American and the Yorkshireman undel'stand each other, may
dep.end on the intelligence of the two individuals concerned, upon
thelf generai experience with foreign dialects or languages, upon
their disposition at the moment, upon the extent ta which the
. t' clarifies the value of the speech-uttcrance, and sa on.
SItua Ion - cl d
A
. there are endless gradations between local and stan al'
gam, . h' h 'd
h
. e'ther or both persons may make concesslOnS w le al
speec , l . Il . th d'
undel'standing, and these concessions wlll usua y run m - e !reC-
tion of the standard language.
AIl this prevents our drawing a plain line round the bordel's
of many a speech eommunity. The clear cases are those where two
mutually uninteIligible languages abut on each othel', as ,do,
English and Spanish in our Southwest. each persan s natlve
l
guag
e - if for simplicity sake, we Ignore the languages of
an, 'h S 'h
Indians and recent immigrants - is either Enghs or ,
d we can draw an imaginary line, a language buundary, whlCh
:ll separate the English-speakcrs from the
This language boundary will of course not appear aS a
and fixed line between two topographically solid commu.mtles.
There will he English-speaking settlements thrown out, m the
shape of speech-islands, into totally Spanish
vice versa Spanish speeeh-islands surroundcd by Enghsh-speakmg
communities. Families and individuals of either group will be
found living among the other and will have to be enclosed in a
separate little circle of our language boundar!. Our
boundary, then, consists not only of a great Irregular lme, but
also of many little closed curves around speech-slands, sorne of
wmeh eontain only a single family or a single person. In spite of
its geometrical complexity and of its instability
this language boundary at any rate represents a plam dlstmctlon.
It is true that linguistie scholars have found enough rcsemblance
between English and Spanish to prove beyond a doubt that these
languages are related, but the resemblance and relationship are
too distant ta affect the question with which we are here concerned.
The same might he said, for instance, of German and Danish:
across the Jutland peninsula, just north of the city of Flensburg,
we could draw a boundary betwcen the two languages, and this
boundary would show, on a smaller seale, the same features as
the English-Spanish boundary in our Southwest. In this case,
however the resemblanee between the two languages is suffi-
ciently close ta warD us of further possibilities. The two languages
are mutually unintelligible, but resemble each other SO closely
tbat it takes no linguistic l'esearch to see the relationship. If one
54 SPEECH-COMM UNITIES
S PEECH - CO::\1 :\'1 UKITlES
55
can compare such things at aIl, the differenoo is no greater than
the difference between, say, a German local dialcct spoken in
SJeswick and one spoken in Switzerland. German and Danish,
where they abut on each other, show a difference no greater than
the differences which may exist within a single 10eal1y differenti-
ated speech-community - only that in the latter case the inter-
mediate gradations intervene, while between German and Danish
we lind no interrnediate dialects.
The purelyrelative nature of this distinction appears more
plainly in other cases. We spcak of French and Italian, of Swedish
and Norwegian, of Polish and Bohemian as languages,
because thesc communities are politically separate and use differ-
ent standard languages, but the differences of local speech-forms
at the border are in aIl these cases rclatively slight and no greater
than the differences which we tind within oach of these speech-
communities. The question cornes down to this: what dcgree of
difference bctwecn adjoining speech-forms justifies the name of a
language border? Evidently, wc cannat weigh differences as
accurately as ail this. In SOrne cases, certainly, our habits of nomen-
clature will not apply to linguistic conditions. The local dialcctR
justify no linc hetween what wc cali German and what we caU
the Dutch-Gernmn speech area is lnguistically a
unit, and the clcavage is primarily political; it is linguistic ouly
in the sense that the political units use diffcrent standard languages.
In sum, the term speech-community has_only a relative value.'':The
possibility of communication botwt:en groups, or evon betwecn
individuals, ranges aIl the way from zero up to the most delicate
adjustment. It is evident that the intermediate degrces contribute
very much to human welfare and progress.
3. 9. The possibilities of communication are cnhanccd and the
boundaries of the spccch-eomrnunity are further obscured by
another very important factor, namely, people's use of forcign
languages. This is by no moans a moern accomp!ishmcnt; among
peoples of simpler civilization, such as sOrne tribes of American
Indians, wcll-bred persons often speak more than ono of the
languages of neighboring tribes. The factor of foreign-language
speaking does not lend itself to measurernent, since proficiency
ranges aIl the way down ta a srnattcring sa slig;ht as to he of al-
most no actual use. 1'0 the extent that the learner can
cato, he may he ranked as aforeign speaker of a language. We have
already seen that the l.s':Ofulness of SOHlC lanj;uagcs, as Eng-
lisb or IVIalay, is partly due ta the adhercncc of. speake.rs.
Often cnough, as arnong the ebssc8 III In:ba, Enghsh
, tlle of commUnlcatlOll bpf,\vccn forclgn speakers
serves as ' '" . .
who do not understl1nd each otber's natwc
Sorne people entirely give up the use of thelr natwe
. f of a forei"n onc This har)pens frequently among lmml-
ID I1vor , .' <o' t .
grants in the United States. If thc immigrant does not. sta
y
.III a
settlement of others from his own country, and espeClally If .he
marries outside his original nationality, he may have no
at al! to use his naJ.ive hUlguage. Especially, would ln
case of less eucatcd persons, this may result, after a yme,. III
wholesale forgetting: people of this kind understand thelr natlvC
language when they chance to hear it spoken, but can no !o.nger
s eaI< it freely or even intelligibly. They have made a shi!: of
,J:nguag
e
; thcr only medium of communication is now English,
and it is for them not a native but an adu]Jted language. Some-
times these persons have neverthclcss acquired English very im-
perfectly and therefore are in the position of speaking no language
weil.
Another, more common case of shift of language in
children of immigrants. Very often the parents speak tbelr natIve
language at homc, an make the native language of thoir
dren but the children as soon as they bcgin ta play out of doOts
" .
or to attend school, refuse to speak the home language, and III
time sllcceed in forp:etting aIl but a srmttteriIlj; of it, and speak
ouly English. Fol' them, English has become what wc may cdl
their adult language. In general, they spcak it perfectly - that i8,
in a manner indistinguishable from tbat of the surrounding native
speakers - but in sorne they carry over foreign peculiarites
from their native language. This latter thoy speak very imperfectly
or not at ttll, but their passive undcrstanding, wheu they hear it,
is somewhat better. A study of sirnilar cases in 'Yales, where the
children of Welsh-speaking parents shift to English, seern8 to show
that this process retards the child's development.
3. 10. In the extreme case of foreign-language learning the
speaker becames sa profrcient as ta be indistinguishable from the
native speakers round hilil. This happens occasionally in adult
shifts of language and frcquently in the childhood shift just
describcd. In the cuses whcrc this perfect forei!!;n-bnV;l1uge Icarn-
56 SPEECH-CO:VI.:vr UNIT1ES
ing is not accompanied by 10ss of the native language, il. results in
bilingualism, native-like contra! of two languages. After early
childhood few people have enol1gh muscular and nervous freedom
or .enough opportunity and leisure ta reach perfection in a forcign
language; yet bilingualism of this kind lS eommoner than one might
suppose, both in cases like those of our immigrants and as a result
of trave1, foreign stl1dy, or similar association. Of course, one
cannot define a degree of perfection al. which a good forcign
speaker becomes a bilingual: the distinction is relative.
d l'dore commonly the bilngual acquires his second language in
carly childhood. This happens frequently in communities near a
language border, or where a family lives as a speech-island, or
where the parents are of different speech. :Many well-to-do Euro-
pean families rnake their children bilingual by employing foreign
nurses or governesses. The educated Swiss-German is bilingua!
in the sense that he speaks both the local dialect and the highly
divergent st.andard German. In the United States, better-educated
immigrants oHen succeed in making their children bilingual; this
development contrasts with the shifting of language among less
privileged groups. In ail these cases, apparently, the two languages
play sornewhat. differcn1. parts in the life of the bilingual. Ordina-
rily one language is the home language, while the other serves a
wider range, but other dispositions also occur. The apparent
frequency with whch one rneets bilinguals among artists and men
of science may indicate a favorable effect of bilinp;ualism on the
general deve1opment. of the child; on the other hand, it may mean
merely that bilingualisrn results from generally favorable ebild.
hood surroundings.
CHAPTER 4
THE LANGUAGES OF THE WORLD
4. 1. Among the languages that arc spoken today, only few
are even tolerably weil known to science. Of many we have in
adequate information, of others none at aIl. The oIder stages of
sorne present-day languages, and sorne languages no longer spoken
are known to us from written records; 1.hese records, however,
acquaint us with only an infinitesimal part of the
of the past. Sorne extinct languages are known from the scantwst
of records, such as a few proper names, many morC only by the
narne of the people who spoke them, and doubtlcss a vastly greater
number has disappeared without a trace. l\'1ore than one language
now spoken, especially in Africa and in South America, will pass
out of existence without being recorded.
The inadequacy of our knowIedge makes it impossible to deter-
mine the rclationships that may exist between many languages.
In general, studcnts who deal with slightly-known languages, have
a weakness for setting up relationships on insufficicnt evidence.
By relationship of languages we mean, of course, resemblances
that can he expIaincd only on the assumption that the languages
are divergent forms of a single older language. Such resemblances
show themsclves in phonetic correspondences like those cited in
Chapter 1, correspondences which can be determined only on the
hasis of extensive and accurate data. The less known the
guages and the less expert the student, the greater is the danger
of his making false assumptions of kinship. Even the most positive
announcements often tum out, upon examination, to be based
Upon insufficient evidence.
4.2. English is spoken by more native speakers than any other
language exccpt, presumably, North Chinesc; if we count. the
important factor of foreign speakers, English s the most ,vide-
spread of languages. The number of native speakers of English
Was estimatcd for 1920 at about. 170 millions ( 3.2). Almost ail of
these speakers use standard or sub-st.andard English; local dialccts
are of sman extent and for the most part mutual1y intelligible.
57
THE LAKGUAGLS OF THE WORLD THE LANGUAGES OF THE \VORLD
59
En!!;1ish is unmistakr1.bly tu the other Gennanic lan-
guages, but at the same lime differs plain!y from ail of them.
History tdls us 1 it came to Hritain as the languag;e of
the Saxom,. and Jutes, who conquered the islaml in the
fifth (:entury of our era. The marked Jifferencc of English from
the Germanie speech along the continental of the North Sea
is exphtrwd by the millcnnium and a half of separation. The oldest
written records of English, daling from the eighth and ninth
cent urif's, confirm ihis, fol' t heir language closdy resemblcs t hat
of (,he oldest records of continental Germi1nic spch, \dlich date
from about the samc tinw. The splittin,e: off of English is a elassical
exarnple of the \Vay in which a dialect aTea is divided by migmtion.
The resemblance is e]osest bet \VCn Lnglish the dialect.s
of the PrisJn aTt'U, spoken by some 350,000 persons on the coast
[Lnd coust,,! islands rl.1ong;! he ?\orth Sea. This rcscmblance ap-
pears strikingly in the oldesl Frisian texts, which from the
second half of the thirtcenth centnry. We conclude chat
is [1.n offshoot of an An(flo-Fi"isiun (or lngu;[onic) dtlecf, area.,
which must have been fairly extensive bcfore the migmtion to
Britain.
Outside of Frisian, the Gennanic-speaking area of the European
mainland (excluding Scandirmvia) shows no sharp deavages. The
nearest. thing ta a is a buncHe of isogJosses running
east and west across German)': uorth of the bunclle olle speaks
p, f, k in words like !Iope, hile, muke, south of il, sounds likef, 8, kh,
ns in standard German /io]Tw, be'iszen, mucher!. The speech of the
Ilorthern type is known as Low German, that of the southern as
Hi,e:h German; since the vaTious isoglosses do not coinc:ick, the
distinction can be sharply drawn only if one resorts ta an arbi-
t rary definition. This (Efference already in our o!dest.
records, \Ihich date from about the saille time as those of English.
\arious kinds of ,'.vidence show us that the dinTgenee of the
soutlwrn (y-pe is due ta ch:w(l,"es whidl took place in the sout.h
during; the fifth :md sixth {'('.ntmies of our era. The Conlinenlal
iYes! Oennrnu'c di:tlccts, as they are in contmst \Vith Anglo-
Fri"bn; made a vigol'oUS expansion during the
Al!:es; lo the cast and southeast, of t.he main urea there arc mflIlY
spch-islands, especially of the lIi?;h German type, sueh a" Yiddish
in Patand and Russia. Continental 1,\'est Germanie is spoken
today by over 100 millions of persons. Tt has clevelowd two great.
cl rd la
nguages Dutch-Plemish, which is used in J3elgium and
stan a "' .
the Netherlands and is based on western coastal dlalects of the
Low-German type, and New High German, based on eastern
t 1dialects of the distri0t that was gained by medICval expanSlO
n
.
raAnglo_Frisian and 'Vest eaeh
ther closcly enough to be vlewed as a TVesl Germamc umt, lB con-
with the smaller Scandinauian (01' North Gcrman) group.
Within this group, Iceland: difTers markedly from the rest,
with the thousltnd years of separation sinee Icdand was colomzed
from western :Norway. Icelandic is spoken today by sorne 100,000
speakers. The language of the Faroese Islands, with about
speakers, is close ta 1celandic., The rest of the area,
Denmark, Norway, Sweden, Gotland, and patt of the Fmmsh
coast, shows no marked eleavages; the speakers Humber sorne
15 millions. Our oldest records of :North Germanie speech are in-
scriptions, sorne of which may date as early as the fourth eentury
A.D.; the oldest manuserip(.s date from the t\ye!fth century, but
the wording of the texts, espeClially in the case of sorne Icelandic
litemture, may be several centuries older. The stand-
ard languages are Icelandi0, Dunish, Dano-Norwegian, Korwegian
Landsmaal, and Swedish.
Vle have sorne information about Germanie languages that are
no longer spoken, sneh as the languages of th'l Goths, Vandals,
Burgundians, and Lombards. Parts of ft Bible translation in the
Gothie language of the Visigot.hs, made by Bishop Ulfil:1 in the
fourth ntury, are presel'ved to us in sixth-eentury manuscripts,
notably the Silver Codex. While the language of the Lombards
seems to have been of the West Germanie type, the others, in-
cluding Gothie, were doser to Seandinavul and are usually set
apart as an Balil Oermanic group. Easl Gel'nli1nic settlers seern
1.0 have kept their lang;uage in the Crimea and elsewhere on the
Blaek Sea untU the eighleenth eentury.
Ali the languages so far named resemble each other cloRely in
eontrast with a1l others, and aecording;ly eonst:itule the Germanie
family of languages; lhey are divergent modern fonus of a single
prehistoric language ta whieh we give the name Primitive Ger-
manie ( 1.6).
4.3. The kinship of the Germanie family, as a wholc, with
certain other languages and hwgmtge fumilies of Europc and
Asia, is not supf'rficially apparent, hut has been fully cstabli"hed
60
THE LANGUAGES OF THE WORLD
THE LANG UAGES OF THE \VORLD 61
by the researches of the last centurYi together, aIl these languages
make up the Indo-European family ( 1.6).
Ta the west of the Germanie languages we find today the
nants of the Celtic family. Irish is knowil ta us from a manuseript
literature sinee the eighth contury of our erai a fcw inscriptions
on stone are perhaps much earlier. Irish is spokcn by some 400,000
people, and its offshoot, Scotch Gaelic, by sorne 150 000'
, ,. fl[.-W}
as a home language, alongside English, by a few hundrcd. Anothcr
branch of the Celtic family consists of Welsh and Breton, each
with about a million speakers and known through written records
since the eighth century. The latter, spoken on the northwestern
coast of France, was brought there from Britain, perhaps as carly
as the fourth eentury. Another language of this braneh, Cornish,
whose earliest records date from the ninth century, died out round
the ycar 1800. History and the evidence of plaee-Damcs show that
Ccltie was in earlier times spoken over a large part of Europe,
including what is DOW Bohemia, Austria, southern Germany,
northflrn Italy, and France. It was superseded in these region,; by
Latin, as a result of Roman conquests, and by Gennanic languages,
as a result of the great migrations in the carly centuries of our era.
Wc have a few scant inscriptions, dating from round 100 B.C. in
the aneent Celtic language of Gaul.
Northeast of the Germanic languages lies the Baltic family.
The two surviving languages of this family, Lithuanian, spoken by
sorne 2}. million people, and Lettish, spoken by sorne 1} millions,
have wntten records dating from the sixteenth eentury; thanks
ta the politieal independenee of Lithuania and Latvia, both of
these dialect-groups arc now developing vigorous standard lan-
guages. A third language of this group, Old Prus8ian, is known
to us from a few written documents of the fifteenth and sixteenth
centuries; it ceased to he spoken in the seventeenth century.
South of the Baltie language,;, and east and southeast of the
Germanie, we find the great Slavic family. The eastward expansion
of German in the Middle Ages overlaid various languages of the
West Slavic braneh. One of these, Lusatian (Wendish, Borman),
survives as.a speech-island of some 30,000 persons in Upper
SaxonYi another, Polabian, survived inta the eighteenth century
and has Ieft a few written texts; the rest have died out, leaving a
trace only in Germanized place-names. As a result of the struggle,
the two great. surviving West. Slavie dialect areas show a peculiar
graphie configuration: a narrow streak of speeeh+islands trails
north\vard from the main Polish area along- the Vistula toward
and Bohemian juts out wcstward as a kind of peninsula
nto the domain of German. PoUsh, recorded sinee the
century, is spoken by more than 20 million people. The Bohemlan
area, divided on the basis of standard languaj';es, ioto Czeeh and
Slovak, comprises perhaps 12 millions of speakers; the oldest
records date frorn the thirteenth cent ury. Ea8t Slavic consists of
but one enormous dialeet area, Rusan, with at least 110 million
speakers, and written records datinj'; back to the twelfth
The South 81avic branch is separated from the ot.hcrs by the mter-
vention of IIungarian, an unrelated intruder. H consists of Bul-
garian, with sorne 5 million speakers, Serbo-Croatian, with sorne
10 millions, and Sluvene, with about 1} millions. Our aIdest written
records of Slavic speech are Old Bulgarian records from the ninth
century, preserved in manuseripts written at least a eentury later,
and a seant tenth-century text in Old Slovene. Sorne students
find a relatively close resemblance between the Baltic and Slavie
groups, and include thorn together as a Balto-Slavie sub-group
within the Indo-European family.
To the south of the Germanie languages, Romance languages
are spoken: the Portuguese-Spanish-Catalan area (with three
standard languages indieated by these names) comprising in ail
over 100 million speakers, the French area with 45 millions, the
Italian with over 40 millions, and Ladin (RJweto-Romanic) in
Switzerland, spoken by sorne 16,000 persons. A further gfOUP,
the Dalmatian, s extinct: one of the dialects, Ragusan, died out
in the fifteenth eenturYi another, Veliote, survived into the nine-
OOenth. Ta the cast, on the Black Sea, cut off from the western
areas by the intrusion of South Slavic, lies the Roumanian area,
estimated as having 14 millions of speakers. Ali the Romance
languages, of course, arc modern forms of Latin, the aneient dialect
of the city of Rome. OUr oldcst records of Latin date from some-
where round 300 Re. In medieval and modern time, Latin has
been used as an artificial medium for writing and leamed diseourse.
Ancient inscriptions show us, in !taly, sorne sister languages of
Latin, notably Oscan and U11Jrian; these and others, which in
the course of Roman expansion were superseded by Latin, belong,
together with Latin, iuto the Italie family. Some seholars believe
that Italie and Celtic are conneeted by special resemblances, so
62 THE L A GUA G ES O.F THE \V 0 R L D THE LANGUAGES OF THE WORLD 63
as Lo form an Ita..lo-Celtc sub-p;roup within the Indo-European
family.
East of ihe Adriatic, south of 8el'bo-Croatian, is the Albanese
area-. Albanese, known from records only 'lince Lhe
ccntury, s by a population of millions. Al thollg;h
Albanese is full of Ioan-words from the surrollnding bnp;uages, the
native nudcus of its forms shows it to be a separaie of the
lndo-European st ock.
is 'lpoken today by some 7 millions of speakers, in many
local dl:J..lects and in n widesprrad 'ltandarcJ ianglHtge. The modern
dialects are almosl enrery descended l'rom the standard l::mguagc
(the so-called Koint) which prevailed in t,he first cent uril'S of
the Christian 1<:ra, having superseded Hw local and provincial
dialecLs of anent times. These Ancit,])! Grp(,k dialects arc known
to us frolll many inscriptions, be!/:inning in the seventh century
B.C., from frap;ments of writing on p:tpyrus, beginning in the fourth
century B.C., and from a copio\ls literature (tl'anslnitkd, to b" sure,
in mueh later rnanuscripts), whose oldest. compusitions, the Ho-
meric poems, are at least as old as 800 H,C.
In Asia .i.\'linor we find one brandi of the Indo-Europnan stock,
Armenian, spoken today by 3 or 4 million our oldest
wri tten records of ArrrH'nian date from the flfth century A. D.
The great A'liatic offshoot of the Indo-European fnmily is the
Indo-Iranian group, This of t,wo sub-groups, Iranian and
Indic (or Indo-Aruan), very diITerent today, but in the forms of
our carliest record'l so simihtr that wc can with certainty view them
as descendants of a Primitive Indo-Imnial1 parent langu:tge.
The prineipa! dia1cct areas of modern Iranian are Persian
(with a sta!Idard language of high prestige, spoken b.y perhaps
7 or 8 millions of people), the Caspian g-roup, and Kunlish; then,
eastward, the Pamir dialects, Afrlhan (Pushto), with some 4 million
speakers, and nalw:hi; an iso1tLtcd offshoot, far to the west is
Ossele, in the Caucasus, spokpn by SOIlle 225,000 persans. Our
oldest records of lranian arc t.he rock inscription'l, in Old Peran,
of King Darius the Great. and his successors (from t.he sixrh to the
fourth centuries H.C.), and the sacred texts, in Avestan, of the
Zoroastrian (Parsi) religion, whose aIdest portions may have been
composed as early as 600 B.C., though our manuseripts arc quite
modern und contain a tcxt which has undewone serious ortho-
graphie revisioll. Intermediate stages, except Persian (Pehlevi),
less weil knovm, but. carly in the prcsent cenf.ury discoveries of
ar:nuscript fragment.s in Chinese Turkestan gave us knowledge of
medieval Iranian languages, whieh have been identified as
Parlhian, 8ogdian, and Sakian.
The other sub-branch of Indo-Iranian, Indic, comprises a tobl
of more than 230 millions of speakers, distributed amonl!; a nllmbpr
of dialect areas which cover larger parl of Ind1 nml indude
such great Innguages as Mamthi (19 millions), CHjerati (10 mil-
lions), Panjabi (16 millioIlH), Haja8lhani (13 millions), H'estern
Hindi (38 millions), Baslan Hindi (25 millions), Oriya (10 mil-
lions), Bihari (36 millions), Bengali (50 millions). The bnguage of
the Gipsies (Romani) is an cmigrant offshoot. of the Pa;achi area in
northwestern Ind1. Our oldest written n'eord'l of Indic sprceh, the
inscript.ions of King Aoka, dating from the t.hirci Cl'ntllry D.C.,
show us a number of Indie dialects in whal is callec! the Prakrit
(or Indic) st.age; Indic languages in the Prukrit. stage are
known to us a1so from !aLer inscriptions and from manuscript
teX; among these last is Pali, the language of the Buddhist
scriptures. An even older stage of Indic speech, the Sans!crilic
(or Old Indic) stage, is known to us, strangely enough, from some-
what luter documents. Our oidest texts in this st.age are the Vedic
collections of hymns; the original composition of the oldest parts
of t.he oldest collection, the Rig:-Veda, is pi:J.ced consel'vativeiy at
1200 B.e. These hymns form the basic part of the scriptures of the
Brahmin religion. A second, 8/ightly divergent type of Old Indic
speech is known to us from the Brahmana.'s, the prose texts of
the Brahmin religion, and from the grammar of PU1.lini 1.5) and
its ancillary works. This language, known as was spoken
round the fourth century RC. by t.he upper class somewhere in
northwestern India. As a stando..rd dialcet and Iatef as a lit.erary
and scholastic language, it gradua.Ily came into officiai m;c ail over
Brahmin India; in t.he inscriptions it appears lirst round IBO B.e.
and 11 few centuries later entirely supersees the diaJects of the
Prakrit type; From f.hat Lime to the pl'eSnlt, writtcn according ta
the rules of PaQ-ini's grammar, it has sprved as the medium of an
enormous body of artist.ic and schola]"ly literatme.
Beside the branches so far named, al! of whieh are represented
languages spoken today, there must have existe at diffcrent
tlmes many other olTshoots of Primtivc Indo-European, sollie
closely related to surviving branches, others interrnediltk between
64 THE LANGUAGES OF THE \VORLD THE LANGUAGES OF THE WORLD
65
them, and perhaps still others quite apart. Of sorne sueh languages
we have a slight knowledge. Round the Adriatic, the lllyrian
languages were spoken in ancient timcs; lllyrian, in which we have
only a few proper names, Vendic, known from inscriptions that
date from the fourth to the second centuries B.C., and Jfessapian
in southern Italy, with inscriptions dating from 450 ta 150 D.C.
Of Thracian, in the western part of the Balkan peninsuia, we have
only a few names and words and a single inscription (round 400
B.C.); it scems ta have been closely rclated ta Phrygian, in Asia
Minor, which is known ta us from a set of inscriptions dating as
early as the eighth eentury B.C. and another set from the first
centuries of our era. Jfacedonian seems 1.0 have been closely related
ta Greek. Ligurian (round the present Riviera) and Sicilian in
Sicily, may have been close to Italie. Tocharian, in Central Asia,
is known to us from manuscript fragments of the sixth century A.D.,
found in Chinese Turkestan.
Primitive Indo-European, in its turn, must have been related
ta other languages; with one exception, however, these have cither
died out or eIse chang-ed so much as to obscure the kinship. The
one exception is Hittite, an ancient language of Asia Minor, known
ta us from cuneiform inscriptions that begin round 1400 B.C. This
relationship, though distant, enables us ta reconstruct sorne of the
pre-history of Primitive Indo-European and sorne features of a
presumable Primitive Indo-Hittite parent language.
4.4. As the various bnguages of the Indo-European stock
sprcad over their present vast territory, they must have obliterated
many unrelat.ed forms of speech. A remnant of such a language is
Basque, spoken today by sorne half-million people in the western
Pyrences. Our oldest texts in Basque date from the sixteenth
century. It is the only surviving form of ancient Iberian, once
spoken over southcrn France and Spain, and known ta us from
inscriptions and place-names.
Of other such languages, now extinct, we have only seant in-
formation. In Italy, Eiruscan, a totally unrelated neighbor that
exerted a powcrful influence on the Latin people, has left us
COpiOliS inscriptions, which begin as carly as the sixth century B.C.
They are in the Greek alphabet and can be rcad, but not
stood. The inscriptions in ancient Rhaetian show this language
to have becn an offshoot of Etruscan. An inscription of about
600 B.C. on the island of Lemnos and a series of inscriptions of the
rth and third centuries B.C., mostly from Sardis in Asia Minor,
:u that Etruscan was related to Lemnian and Lydian; the texts
(o:ly the last-named have been interpreted.
o ancient Crete we have several inscriptions in the Greek
alphabet but in an unknown language, two from the fourth century
and one (from the town of Praisos) somewhat aider. From a
earlier period, round 1500 B.C. we have Cretan inscriptions
:.nly in picture-writing and partly in a simplified system derived
(romtbis. .
From Asia Minor we have copious inscriptions in Lycwn, from
tbe fifth and fourth centuries B.C., and less extensive ones in
Carian, from the seventh century B.C. The former are in a Greek
alphabet and have been partly interpreted; the writing of the
latter may be of the same provenience, but is undeciphered. In
Syria and the adjacent part of Asia Minor copious inscriptions in
pictute-writing from about 1000 B.C. to about 550 B.C. have been
attributed to the Hittites, but there is no reason for believing that
these undeciphered inscriptions were made by the same people as
our Hittite cuneiform records ( 4.3).
Cuneiform inscriptions on rock and clay from the Near East
acquaint us with extinct languages of an older time: Sumerian in
Mesopotamia, from 4000 B.C., Elamitic, in Persia, from 2000 B.C.;
seant records of Cossean, cast of Mesopotamia, from 1600 B.C.,
Mitanni, east of Mesopotamia, from round 1400 B.C.; the language
of Van (near Lake Van) from the ninth and eighth centuries B.C.;
and several uninterpreted languages ,vithin the Hittite empire in
Asia Minor. Of the other languages represented in records of this
type, we have aIready mcntioned Old Persian and Hittite ( 4.3),
and shaH immediately spcak of Babylonian-Assyrian, a 8emitic
language.
4.6. Of the presentwday families which border upon Indo-
European, one or more may be distantly akin; the Semitic-Hamitic
and the Finno-Ugrian families seem ta show sorne resemblance
to but, in spite of mueh effort, no conclusive evi
w
dence has been found.
The Semitic-Hamitic family consists of four branches which
tesembIc each other but distantly; Semitic, Egyptian, Berber, and
Cushite.
The Semitic branch appears in two offshoots. The eastern, now
extinct, consists of Babylonian-Assyrian, known to us from in
w
66 THE LANGUAGES OF THE \VORLD
THE LA)fGUAGES OF THE WORLD 67
scripti;)ns on stone and clay in euneiform writing, from about
2500 B. c. onward; this bnguag-c was superseded by Aramaic bdore
the beg-illIling- of the Christian Em. The western branch of Sernitie
is divided, again, into two main offshoots, a northern and a south-
ern. The former appeal's in the Canaanile glosses in euneiform
tablet.s found al. Tel-el-Amarna, datng rouIHi 1400 B.C., and in
the .JIvalide of the famous inscription of King ?vf esha, Ilinth
eentury B.C. Phoenician, kIlown fi l'st, from imwriptions of the
ninth ccntury n.C, \Vas spoken not only in where it
died out bcfore the Christian Em, but also in the Phocnieian
colony of whcl'c it lived somc centuries longer. Ilebrew
is known from ins(:riptions of equal age and from the manuscript
tradition of the Old Testament. whosc eal'Iicsl portion may have
been eomposed by 1000 fl.C. Tt. \H1::; superseded by Ararnaic in the
second century B.C., but remllined in writtcn use through the
"Middle Ages; of lale, t Iwre have been attempts ta restore it,
artincially, to the sj:1tm of Il spoken language. AruJilaic, finally,
eonsists of a group of dialeet::;, fir::;t. known from inscriptions of the
eighth century n,Co In a ircmendous wave of expansion, Aramaic,
in the centuries just before the Christtn Eta, spread ove:' Syria
and large tracts of Asia" \'ying \Vith Greek, and replacing many
languages, amollg thcm Hebrew and Assyrian. For a rnillennium
(from round 300 B.C. to round 650 A.D,) it served as t.he leading
official and written langnage of the car East; in the latter capac-
ity it exerciscd a great. effect upon Asiatic systems of writing.
It was superscded, in its turn, by the spread of Arabie, and is
spoken taday in isolated patches by sornc 200,000 people. The
Routhern bnmch of West Semitic is reprcsented by several still
f10urishing languages. Smdh Amhic, known from inseriptions
ranging frorn about 800 n,co ta the sixth century A,D., is still
spoken, in several dialeds, along the sO\Jt,hern coast of Arabia and
on the isla.nd of Sokotra. Arahie, whose ear!iest record is an in-
scription from 328 A.D., owes its expansion, since the seventh
century of our era, to the conquests of the Molmmmedan Arabs.
lt is spoken today by sorne a7 millions of people and, beyond this,
has served for centuries as the sacred, literary, and oITtciallanguage
of Islam. Ethiopian, on the cast coast of Africa (Abyssinia), if!
first known ta us froUl inscriptions bel?;inning with the fourth cen-
tury A.D.; the pre5ent-day languages of t.his group are Tigre,
Tigrifia, and A mharic.
The Egyptian, Berher, and Cushite branches of Semitic-llamitic
are usually induded under the name of IIamihc languages.
Egyptian is recorded for us in hieroglyphic inscriptions from
4000 B.C.; the later form of the language, known as Cuptic, appears
in a manuscript literature of Christian times. Egyptian died out,
superseded by Arabie, in the sevcnteenth century.
The BeI'ber bmnch of Semitic-Hamit.ic, is known from ancient
Limes through inscriptions in the Lihyan language, from the fourth
century B.C.; il. is represented today by various languages, such
as Tuarcg and Kabyle, which have maintained theillseives against
Arabie in northern Africa and are sal to total SOine fi or 7 million
speakers.
The fourth brancll of Sernitic-Hamitic i::; Cushile, south of
Egypt; it includes a numbcr of languages, among them Somali
and Galla, the latter with some 8 million ;;peakers.
4.6. South of the Arab and Berber areas of northern Africa, a
broad belt of ma,ny langnages stretches across the continent from
the Ethiopian Cu::;hite areas in the cast to the Gulf of Guinea
in the west. The of thi;; vast belt, spoken by a popula-
tion of presumably sorne 50 millions, are known. Some
schobrs, upon vpry scant evidence, bclieve them aH ta be related;
others connect of these languages with Hamitic, or soille
wit.h Bantu. Among the b,nguages of this region that arc more
often named, we nmy mention lVolof and FuI in Senegal ; Grcho,
Evx!, and YOi'1Jna :tlong the Guinea coast; HI1lISSa in Hw central
region; and in the past, Nuba in li large terri t.ory round l'':hartoum,
south of this, J)inka, and still further south, Musai.
South of this Guinean and SOlldanese belt we come upon the
vast Bantn family of hnguagcs, which before the Europcan in-
va::\ion covercd ail the rest of Africa except only a southwestern
district. The bngmtges of the Ihntu family, totalinp; sorne 50 mil-
lions of speakers, are very nUlllCfolls; amonp; the bctt.r known
I1re Luganda, 8Wllhdi, Ka.{fil', Z1/I1I, 'l'elJc/c, Subiya, llerero.
The portion of sout.hwestern Africa that was not Bantu-speaking,
belon?;cc\, before the coming of the European, to two unrclated
linguistic areas: the Hushnwn, with some 50,000 speakers, and
the Il ot/cntot, \Vith sorne 2.50,000.
4. 7. Returninp; to the continent of EUTasia, we find, to the
east of the Indo-European lanp;uages and in topographie alter-
nation with Lhem, the great Finno-Ugrian famil)'. This family
68 THE LANGUAGES OF THE WORLD THE LA::-fGUAGES OF THE WORLD 69
consists of six major branches. The lirst is the Finnish-Lapponic.
In the northerly parts of Norway, Sweden, and Finland, sorne 30,-
000 people speak Lappish. The other languages of the Finnish-
Lapponic branch form a closer group, the Finnish (or Baltic-
Finnish). The largest language of this type is Finnish, recorded
in a fragmentary way as early as the thirteenth century and in
printed books since 1544; Finnish is native ta sorne 3 million
speakers. Esthonian, with earlest records of about the same dates
,
is spoken by about a million people. Bath Finnish and Esthonian
have standard languages which are official in the republics of
Finland and Esthonia. The other languages of the Baltic branch,
Carelian, Olonetsian, Ludian, Vepsian, Livonian, Ingrian, and
Votian, are far srnaller, and sorne of them are near extinction.
Four further branches of the Finno-Ugrian stock lie in patches
across thc extent of European and Asiatic Russia; thcy are Mord-
vine (a million speakers); Cheremiss (375,000); Permian, consist-
ing of Votyak (420,000) and Zyrian (2,58,000), the latter with
written records from the fourteenth century; Ob-Ugrian,
ing of Ostyak (18,000) and Vogule (5000). The sixth branch of
Finno-Ugrian is Hungarian, brought by invaders at the end of
the ninth century inta central Europe. Aside from scattered
words in Latin documents, the oldest written record of Hungarian
dates from the thirteenth century. In a flourishing standard
language and in a number of local dialects Hungarian is spoken
by sorne 10 million p()rsons.
1'0 the east of the Ostyak area, along the Yenisei River, sorne
180,000 persans speak languages of the Samoyede family. These
languages are dispersed over a wide area and show great local
diversity. Sorne investigators believe that Sarnoyede and Finna-
Ugrian are related.
4. 8. The Turkish (Turco-Tartar or Altaic) farnily of languages
covers a vast main area, from Asia Minor, conquered, at the end
of the Middle Ages, by the Ottoman Turks, al! the way to the
Upper reaches of the Yenisei. These languages, with little d-
ferentiation, are spoken by some 39 millions of people; Turkish,
Tartar, Kirgiz, Uzbeg, Azerbmjani are the more familiar language-
names. Our oldest texts are sorne Siberian inscriptions, dating
from the eighth century A.D., a Turkish-Arabic vocabulary from
the elcventh century, and a Latin-Pcrsian-Turkish vocabulary
from the fourteenth. Separated from the other languages of the
group, but not very different from thorn, is Yakut, spoken by
over 200,000 people in northernmost Sibcria. Some studcnts
believe that Turco-Tartar is related to the l\longol and Manchu
families; others, on cvcn slighter grounds, daim a relationship
of aIl these with Finno-Ugrian and Samoyedc (in what they caU
a Urai-Altaic family).
The Mongol languag('s lie for the most part cast of the TUrco-
Tartar, in ::YIongolia, but, in consequence of the former wander-
ing and predatory habits of these tribes, scattcrcd comrnunities
are found in various parts of Asia, and eVen in European Russia.
The total number of 'lpeakers is estimated at 3 millions. The old-
est known written record is an inscription from the time of Gengis
Khan, in the thirteenth eentury.
The Tungusc-It,[anchu fmnily lies ta the north of the MoniSol,
dividing Yakut from the rest of the Turco-Tartar area. Tunguse
is spoken by some 70,000 persons dwelling oyer a relative1y larp;e
tract in Siberia. The number of aetual speakers of Manchu is
uncertain, since most of the so-called 1'vfanchus in China speak only
Chinese; Deny estimates it at weil under a million. As li literary
and official language, l\Janchu has been printcd since 1647; the
manuscript tradition p;oes bfwk to an even earlier date.
The great Inrlo-Chinese (or 8ino-Tibetan) family consists of
three branches. One of these is Chinese, spoken by sorne 400
lions of people; it fonns really a vast dialect area containing many,
in part mutuaIly unintelligible, dialects or languages. TheBe
have been classified into four main groups: the Mandarin group
(North Chinese, including the language of Peking; Middle Chine8e,
including Nanking; West Chinese, in Szechuen), the Central Coas/al
group (Shanghai, Ningpo, Hangkow), t.he Kiangsi group, and the
South Chinese group (Foochow; A moy-Swatow; akka).
Our oldest t.exts are inscriptions, sorne of which may date as far
back as 2000 B.e., but sinee Chinese wl'iting uses a separate sym-
bol for each word, with litHe indication of sounds, oyen an in-
telligible document may tell us liule or nothing of thc language:
our knowledge of Chinese speech, t.herefore, does not set in beR
fore about 600 A.D. The second branch of Indo-Chinese is the
Tai family, which includes 8iarnese, spoken by sorne 7 millions
of people; the oldest record i8 an inscription from 1293 A.D.
The third bmnch is l'ibeto-BJlrman, consisting of four groups:
in the Tibetan group, the langull.!!;c of the same name, with rec-
70
THE LANGUAGES OF THE 'VORLD
THE LANGUAGES OF TIlE WORLD
71
ards reaching back to the ninte ecntury A.D., is the most impor-
tant; in the Bllrrnesc group, B1trmese, with sorne 8 million speakers,
holds a position; the other two groups,
and La-Io, consist of lesser diaJccts.
The H7!perborean famil)', in the extreme northeastcrn eorner
of Asia, consists of ChiJkchee, spoken by :some 10,000 persons,
Karyak, with almost f1S man)' speakers, nnd Karnciwdal, with
1000.
Along the Yenisei Hiver, y cnisci-08fyak, with Rome 1000
speakers, and Cottian, probably by this time extinct, forln an
independent family.
No relationship has heen found for seveml other languages of
castern Asia. Gilyak is ;;pokpn in the nortlwrn part of 8akhalin
Island and round tlw Illouth of the Arnur Rivpr. Ji in1l is
by some 20,000 persons in ,Japan. Japancse has :36 million speak-
ers; the written records begin in the eighth ccntury. Korean has
17 mllions of speakers.
4.9. Turning southeastward from Europe, wc find in the Cau-
casus region a great. varidy of languages. Apart from Ossete,
an Iranian language '1.:3), these are grmerally claHspd inlo two
,-Vorth Cauwsian and Soulh Cmrwan, with bctwecn 1
and 2 million speakers in each. The be,J, known of these hnguagcs
Georgian, hclDngs to the latter group; the \vritten
as carly as the tenth e(,ntmy A.D.
In India, south of t.he Indo-.Arya,n lanl!;uages, lies the great
Dravidian family, including, beside many lesser languages, the
great spcech-an,as (and st.andard li Lemry languages) of 'l'amil
(18 millions), Jialayalarr' (6 million::;), CarWTfSC (10 millions; () ldest
inscriptions from the fifth ceutury A.D.), Telugu (24 millions).
A single Dravidian langwlge, Bmhni (wi th 17<],000 speakers) is
spoken, far on from the l'est, in the mountains of Baluchistan;
it seems to be a relie of a time wlH'n Dravidian oeeupied a rnuch
wider territory, before the invasion of Indo-Aryan and Iranian
speech.
The languages of the ]\Iunda family are spoken by 3 millions
of persons in two separate parts of India, namely, on the southern
slope of tho Himalayas and round the plateau of Chota Nagpur
in central India.
The Jfon-Khmer fami!y lif's in patelles over southcflstcrn Asia,
including the Islands and some districts in the :Malay
Peninsllia. Our oldest records are inscriptions in Cambogian,
dating from the <leventh century A. D. This family includes at
present one grcat cultural language, Annamite, spoken by 14
millions of people. Sorne scholars beheve both t.he l'Tunda and the
l'don-Khmer fmnilies to he rdated to the :Vlalayo-Polyncsian
falllily (forming the Austric family of l:mgullgc::;).
The il[alayo-l)olyncsian (01' A w;lroncsm) f:mlily ext.enc!s from
the l\.Jalav Peninsula across the Pacifie 1,0 Easter Island. lt
consists of four bn1i1ches. The Malayan (01' Indoncsian) branch
incllldes Malay, \vith some 3 million m1rive speakers and wide use
as a language of commerce and dviliz:ttion; further, il, embmces
the languages of the great, h,Jands of the Vast, such aS Fomwsan,
J avaneSf (20 millions), 8ltndanese millions), AIaduran (3 mil-
lions), Bahncse (1 million), and the Illany Phihppw !a.nguages,
among them Bisaya (2'1 millions) and Tagalog (t millions); a
tant. offshoot is Alalagasy, the language of Madagascar, spoken
by some 3 million people. The second, }'i danesian, bmnch of
lVlabNo-Polyncsian indudes many languages of smaller isbnd
groups, snch as the l:mglUlgcs of the 8olOJnon 1slands und Fijian.
The ]lIicroncsm brunch contains t.he languages of ft, srnaller Trad,
the Gilbert, Alarsha!l, Caroline, Q,ud JIarianne f1rchipehgos and
the 1::;land of Yap. The fourth, Folynesian brandi inclucles 111 aori,
the native language of Npw Ze:tland, and the languages of t.he
more casterly Pacific isla,nds, such as Samoan, Tahilian, IJawaiian,
and the language of Easter Lland.
The ot.her families of ihis part of the e:lrth have been little
studied; the PaplllHI falni.!y, on Xew Guinea :tIld adjacent islands,
and the A'U.'jlmliaTt languages.
4. 10. There retll:1ins the Amcrican continent.
His estimatcd the territory north of :\'lcxico \Vas inhabited,
before the eoming of the white by nearly 1,500,UOO Indians;
in this same territory the number of speaker;,; of American lan-
guages today cannot be much over ft of a million, with
En"lish making ever more rapiel encroac!lment. As the languages
been insufficiently studied, they C:W be but, tentatively
grouped into farnilies: cstimates vary between twenty-five and
fifty entirely ulll'elated families of language::; for the reg-ion north
of l\lost of this rcgion is covered by great linguistic
stocks, but sorne areas, not,ably the region round Puget Sound
and the eoastal district of \Vere dosdy packed with
72 THE LANGUAGES OF THE WORLD THE LANGUAGES OF THE WORLD 73
smalt unrelated At least half a dozen ln-
guistic stocks are known ta have died out. Of those that still
exist, we may name a few of the largest. In the far north, the
Eskimo family, ranging from Greenland over Baffinland and Alaska
ta the Aleutian Islands, forms a fairly close-knit dialeet-group.
The A 19onquian family covers the northeastern part of the con-
tinent and includes the languages of eastern and central Canada
(Miemac, Montagnais, Cree), of New England (Penobscot, },Iassa-
chuseUs, Natick, Narraganset, 21,Iohican, and so on, wi th Delaware
to the south), and of the Great Lakes region (Ojibwa, Polawatomi,
Menomini, Sauk, Fox, Kickapoo, Peoria, Illinois, .llIiami, and sa
on), as weil as a few detached languages in the west: Blackfoot,
Cheyenne, and Arapaho. The Athabascan family covers ail but the
coastal friuge of northwestern Canada (Chipewyan, Beaver, Dogrib,
Sarsi, etc.), a number of isolated groups in California (such as
Hupa and Matok), and a third, large area in the south, the Apache
and Navajo languages. The Iroquoian family was spoken in a dis-
trict surrounded by Algonquian; it includes, among others, the
Huron (or Wyandot) language, and the languages of the Iroquois
type CMohawk, Oneida, Onondaga, Cayuga, Seneca, Tuscarora); in
a detached region to the south Cherokee was spoken. The Musko-
gean family includes, among other languages, Choctaw, Chickasaw,
Creek, and Seminale. The Siouan family includes many languages,
such as Dakota, Teton, Oglala, Assiniboine, Kansa, Omaha, Osage,
Iowa, Missouri, Winnebago, lV"andan, CroU!. A Uto-Aztecan family
has been proposed, on the basis of a probable relationship, to
inc1ude, as three branches, the Piman family (east of the Gulf of
California), the Shoshonean family (in southern California and
castward, including (/Ie, Paiulc, Shoshone, Comanche, and Hopi),
and the great Nahuaaan family in Mexico, including Aztec, the
language of an ancient civilization.
The number of speakers of Amcrican languages in the rest of
America is uncertain: a reeent estimate places the figure for
Mexico alone at millions and for Peru and Brazil at over 3
millions each, with a total of over 6 millions for Mexico and Central
America and of over 8-} millions for South America. The number
of languages and their relationships are quite unknown; sorne
twenty or sa independent familics have been set up for Mexico
and Central America, and round eighty for South America. In the
former rcgion, beside Nahuatlan, we may mention the Mayan
family in Yucatan as the bearer of an ancient civilization. In
South America, we note, in the northwest, the Arawak and Carib
families, which once prevailed in the West Indies; the Tupi-
Guarani, stretched along the coast of Brazil, the Araucanian in
ChUe, and Kechuan, the language of the Inca civilization. Bath
the Aztec and the Maya had developed systems of writing; as
bath the systems were largely hieroglyphic and have been only in
part deciphered, these records do not give us information about
the oIder fonus of speech.
THE PHOKE3JE
75
CHAPTER 5
THE PHO)J E:vIE
O. 1. In Chapter 2 we distinguished three succe;;sive events in
an act of speech: A, the speaker's sitlmtion; il, his ntterance of
speech-sound and its impingement on the lH'arer's car-drums: and
C, the hearer's response. Of these three types of events, A a;l.d C
include ail the situations that Illay prompt a person 1.0 spea,k and
ail the actions which a hcarer may perfonll in response; in SUIn,
A and C makc up the world in which we live. On the otlwr kmd,
D, the speech-sound, is merely a rueans whieh enables us to rp-
spond to situations that would otherwise leave us unaffeoted, or
io respond more accurately to situations that otherwise Illight
prompt [ess usefulresponses. In prineip!e, the sturIont of language
is eoncerned only with t.he actual speech (H); the stlldy of speakers'
situat.ions and hearers' responses (A and C) is eqllivalent to the
sum total of human knowledge. If we had an accurate knowlcdge
of every speaker's situation nnd of every hearer's response - and
this would rnake us lttle short of omniscient -, we could simply
register these two facts as the meaning (A-C) of any given speech-
utterance (B), and neatly scpamte our study from al! other rIo-
mains of knowledge. The rad that speech-utt.emnces themselves
orteIl play a part in the situation of a speaker and in the response
of a hearer, rnight complcate things, but this diffieulty would
not. be serious. Linguistics, on this ideai plane, would consist. of
two main invesli!!;ations: p!wnelics, in \vhich we studied the spceeh-
event withoui, refercnce 1,0 its meaning, investigatng only the
sound-producing I!lovements of the speaker, the sOllnd-waves, Hnd
the aeton of the hearer's ear-drum, and sernan{'8. in which we
;;tudied the relation of thesc features to the featun:s of meaning-,
showing that a certain type of speech-sound was uttered in certain
types of situations and led the hearer to perfonn certain types of
response.
Actualiy, however, our knowledg-e of the world in which wc live
is so imperfect that. we can rarely Hmke accnrate statcments about
the rneaning of a speech-form. The situations (A) which !cad to
74
an uttemnce, and the hcarer's responses (C), includc many things
that have Hot been mnstered by science. Even we knew mueh
more than wc do about the external world, wc should still have ta
reckon with the predispositions of the ;;peaker and the hearer.
\Ve cannot foretell whether, in li given situation, a person \vill
or if so, what words he will use, and we cannot forete1l
how he will respond lo a given speech.
Tt is truc that we a.re Ilot so much with each individual
as wit.h the wholc comrnunity. \'le do not inquire into t,he minute
nervous processes of a persoll who utters, say, the word apple,
but content. oursdves rather with determining that, by and larg:e,
for ail the members of the community, the w-ord apple means a
certain kind of fruit. However, ilS 800n as we try to dealaccurately
with this llmller, we find that the a?;recrnent of the community is
far from perfect, and ihat every person uses speech-forms in a
unique way. .. .
6.2. The stucly of language can be conducted wlthout specml
assumptions only so 10ll?; as we pay no attention to rneaning
of wh,ct is spoken. This phase of language study IS known as
phonctics (cxpern entaI phone/icI'!, la/Joratory phone/ics). The pho-
neticil1n can swdy either the sound-producing mm"ements of the
speaker (physiololjicol phonetiul) or the resuHing sound-waves
(physical or acoustic p!wnctics); '.,ye have as yet no means for study-
ing the action of the hearer's ear-drum.
l)hysio!o?;ica] phonetics begins with inspection. The laryngoscope,
for instance, is il mirror-device which cnables an observer ta see
nnother person's (or his own) vocal chords. Like other devices of
the sort, it interferes with normal speech and can serve only
very lmited ph:lses of observation. The x-ray does good serVlCe
where its limitations can he overcome; tongue-positions can be
photographed, for instance, if one lays a thin or chain
along the upper surbee of the tongue. Other dev\ces a
ferred record. For instanee, a false pa1ate covered wIth colorm?;-
matter is put into Hw mouth j after the speaker utters a
the places where the tangue has touched the palate Rl"C
able by the removal of the celoring-matter. In most devlCes of
this 80rt a bulb is attaehed io sorne part of the speaker's vocal
organs, say to the adam's-applei the mechanism
illovement into up-and-down movemenis of a pen-pomt WhlCh
touches a strip of paper. The strip of paper is kept moving at an
76
THE PHONEME
THE PHONEME 77
even rate of speed, so that the up-and-down movement of the
appears on the paper as a wavy Hne. This recording
devLCe lS cal1ed a kymograph. In acoustic phonetics One seCUres
imprints of the sound-wavcs. Records of this kind are familiar ta
us in the form of phonograph-disks; phoneticians have not yet
succceded in analyzing most features of such records.
A considerable part of our information about speech-sounds is
due to the methods we have just outlincd. Howevcr, laboratory
phonetics does not enable us ta connect speech-sounds with mean-
ings; t studies speech-sounds only as muscular movements or as
disturbances in the air, without regard to their use in
tion. On this plane we find that speech-souncls are infinitely
complex and infinitely varied.
Even a short speech is continuous: it consists of an unhroken
succession of movements and sound-waves. No matter into how
successive parts we break up our record for purposes of
mmute study, an even finer analysis is al ways conceivablc. A
speech-utterance is what mathematicians call a rontinuum; it
can he viewed as consisting of any desired number of successive
parts.
are infinitely varied. Everyday experiencc
us that different persans speak differently, for we can recog-
mze people by their voices. The phonetician finds that no two
utterances are exactly alike.
Evidently the working of language is due ta a resemblance be-
tween su?cessive Utterances which in ordinary life
we as conslsting of "the same" speech-forms _ say,
utterances of the sentence l'm hungry - evidently
contam some constant features of soundwave, cornmOn ta aIl
utterances of this "same" speech-form. Only on this assumption
can we account for our ordinary use of language. The phonetician,
ho,,:ever, cannat make sure of these constant featurcs, as long as
he ignores the meaning of what is said. Suppose, for instance,
that he had records of an utterance which we could identify as
representing the syllable man, spoken on two different pitch-
schernes. If the language of these utterances were English we
should say that contaioed the sarne speech-forrn,
the ward man, but if the language were Chinese the two records
lliight represent two different speech-forms, in Chinese dif-
ferenees of pitch-scheme are connected with different meanings:
the ward man with a high rising pitch, for instance means 'deceive,'
and the ward man with a fal1ng pitch means 'slow.' As long as
we pay no attention ta meanings, we cannat deeide whether two
uttered forms are "the same" or "different." The phonetician
cannat tell us which feutures are significant for communication
and which fcatures are immaterial. A fcature which is significant
in sorne languages or dialects, may be indiITerent in others.
5.3. The fact that two utterances of the syl!able man with
different pitch-schemcs are "the same" speech-form in English,
but" different" speech-forms in Chinese, shows us that the workw
ing of language depends upon our habitually and conventional1y
discriminating sorne features of sound and ignoring al! others.
The features of sound in any utterance, as they might he recorded
in the laboratory, are the groBs aeoush"c fea/ures of this utterance.
Part of the gross acoustic fentures are indifferent (non-distinctive),
and only a part are connected with mcanings and essential to
communication (distinctive). The difference between distinctive
and non-distinctive features of sound lies entirely in the habit of
the speakers. A feat.ure that is distinctive in one language, may
be non-distinctive in another language.
Since we can recognize the distinctive features of an utterance
only when we know the meaning, we cannot identify them on the
plane of pure phonetics. 'Ve know that the difference bctween
the English forms man and men is distinctive, because we know
from ordinary life that these two forms are used under different
circumstances. Tt is possible that sorne science other than lin-
guistics may define this diffcrence in accurate terms, providing
even for the case where we use man for more than one individual
(man wants but litile kae below). In any case, however, this dif-
ference cannot he recognized by purely phonetic observation: the
difference bctween the vowel sounds of man and men is in some
languages non-distinctive.
To recognize the distinctive featuI'Cs of a language, we must
leave the ground of pure phonetics and act as though science had
progressed far eoough to identify ail the situations and responses
that make up the meaning of speech-forms. In the case of our
own language, we trust ta our everyday knowledge ta tell us
whether speech-forms arc "the same" or "different." Thus, we
flod that the word man spoken on various pitch-schemes is in
English still" the sarne" ward, with one and the same meaning,
78 THE PHONE.ME
THE PHONEME
79
but that man and men (or pan and pen) arc "different J! words,
with different meanings. In the case of a strangc we
have to learn snch things by trial and error, or to obtain the mcan-
ings from someone that knows the language.
The study of signijicani speech-sounds is phonology or practical
phonetics. Phonology illvolves the consideration of meanings.
The mcanings of speech-forms could be scienilfieal1y dcfined only
if ail branches of seiencc, including, espceially, psychology and
physiology, were close ta perfection. L'nill that time, phollology
and, with it, aIl the semantic phase of language study, rests upon
an assumption, the fundamental assumption of linguistics: wc must
assume that in every speech-community some utterances are alike
in form and meaning.
6.4. A moderate amount of experimenting wiII show that the
significant features of a speech-form arc limited in nurnber. In
this respect, the significant features contrast with the gross acoustic
features, which, as we have seen, form a continuous whole and
can be subdivided into any desired number of parts. In order to
recognize the distinctive features of forms in our own language,
we need only determine which features of sound are" diITerent"
for purposes of communication. Suppose, for instance, that we
start with the ward pin: a few experiments in saying words out
Ioud saon reveaI the following resemblances and differences;
(1) pin ends with the same sound as fin, sin, tin, but begins
differently; this kind of resemblance is familiar to us because of
our tradition of using end-rime in verse;
(2) pin contains the sound of in, but adds something at the
beginning;
(3) pin cnds with the same sound as man, sun, hen, but the
resemblance is smaller than in (1) and (2);
(4) pin begins with the same sound es pig, piIl, pit, but ends
differcntly;
(5) pin bcgins with the same sound as pai, push, peg, but the
resemblance is srnaller than in (4);
(6) pin begins and ends like pen, pan, pun, but the rniddlc part
is differcnt;
(7) pin begins and ends differcntly from dig, fish, mill, but the
middle part is the sarne.
In this way, we can find forms which partially resemble pin,
by altering any one of three parts of the ward. We can alter first
one and thPll a second of the threc parts :tnd still have a partial
rcscmblanee: if we aller the first part and then the second, we get
. l'k 'n t;rt tait' l'f \"e alter the first part and then the
a scres 1 c pl -, - " , '
third, y.m get a series lib; pin-tin-tick; if wc alter the second. part
and then the thil'cl, we get 11 series like and if we
alter al! thl'ce parts, no rcsembbnee is left, as lU pw-tm-irw-tack.
Furthcl' experilllent fails to rcveul any rcplaceabk
in t.he worel pin: we condude that the distinctrve f:atures of thls
word are thrcc indivisible units. Each of these \lIut.s oeeurs al.so
in other eombinalions, buL cannot be further
resemblances: cach of the thrcc is a minimnrtl nmt af
sound-feature, a phonem!";. Thu;] we say thaL the .word pm consls,ts
of thrcc phonemcs: the first of thcse alm.
m
pack,
and Ulany other words; the second a];;o in fig, lnt, mISS, and many
other words; the thinl also in tan, om, hen, and many other words.
l th
. of p;n our alphabetic writing represents the three
n ,e case , . f
phonemes by thrce leUers, p, i, and 71: but conventions .0
't' ca poor "uI'dc' in the \Vord thIck, for mstancc, our wnt-
wnmg al' . ,..,. , i
ing represellts the first nhoneme by the iwo-letter group th am
the thini by the two-Ietter group ck. . .
A little prucce will enable the observer ta a phoneme
even when it appears in different parts of words, as pm, apple, 'lnOp.
Sometimes our stock of words docs not rellciily bring out. t.h.e
d
'if F 'n'tance the ,vord then eVl-
rcsemblances and l erences. 'or l s, .
dent.ly consist:; of thrce phonemes, buL under the lll-
f1uencc of our way of writing) wc might whcther
initial phonerne was or was Ilot the same as in Ouck; once we hIt
h
. tl' l nd thy or upon mouth and mouthe, ,ve see
upon l e palr ng ta,
that they are difTercnt..
6.6. Among the gross acoliStic features of any .utterance, the
n
,
t
. h" are 'lncll've recurrinlT in rccoglllzflble and
cel' a.tn on"" ."' ',''''' ., .
tively ooiistant shllpe in sucecssive utter::mccs. dlstmctrve
features occur in lumps or hundles, each one of wluch we cal! a
phoneme. The speaker has been trained ta make sound-prod.uCln
g
t
. such a way that thc phoneme-features Will be
movemen s 10 "
prescnt in the sound-waves, and he 1ms been tramed ta
only to these fentures and to ip;nore the rcst af the gross acoustlC
mass that reaches his cars. ..'
It wou1d be usdess to try ta praduce the istmchve features ma
pure state, frec from non-distinctive a<"lcornpaniments. For
80 THE PHONEI\IE THE PHONEME 81
ample, an English word, as such, has no distinctive pitch-scheme
- the features of ptch which appcar in any utterance of it are
non-distinctive - but of course \VC cannot speak a word likc man
without any features of pilch: in any one Iltterancc of it thcrc will
he sorne pitch-scheme - even, rising, falling, high, middle, low,
and SO 00. The phonernes of a language arc Dot souDds, but mcrely
features of sound which the speakers have been trained to produce
and recognzc in t.he current of actwl1 speech-sound - just as
mot.orists arc trained to stop bdorc a l'cd signal, be it an c!ectric
signal-light, a larnp, a flag, or what not, although there is no
disernbodied redness apart from thesc actual signais.
In fact, when wc observe closeiy, especially in a lan@;ua@;e foreign
to us, we orten notice the wide range of non-distinctive featuI'f?S
and the relatively slight consistency of the distinctive features.
The Menomini Indian, in a word like that for 'water,' which l
shall here render as nipew, secms to us ta be speaking the middle
consonant sometimes as a p and somdirnes as a b. Fol' his language,
the phonemic (that is, essential) feature is merely a clasure of the
lips without. escape of bre:tth t hrough the nose. Everything cise,
including the features by which English dist.inguishes between p
and b, is non-distinctive. On the other hand, a slight puff of
breath before the consonant, or eIse a slight catch in the throat-
either of which will probably escape the ear of an English hearer-
would produce in the Menomini language two entirely different
phonemes, each of whieh contrflsts with the plain p-IJ phoncrne.
In the same way, a Chinesc observer who had not been fore-
wamed, would probably have sorne troublp before he realized that
English words have the same mcaning (are" the same ") regardless
of their pitch-schellle.
In part, the non-distinctive featurcs receive a fairly conventional
treatment. \Vhen a foreign speaker reproduces the phonemic
values of our language so as to make himself understood, but dops
not distribute the non-distinctive features in accorclance with our
habit, we say that he spcaks our language weIl enough, but with :1
foreign "accent." In English, fol' instance, we produce the initial
phonemcs of words likc pzn, Nn, kick with a slight puff of breath
(aspiration) artel' the opening of the cloBure, but when an s p r e ~
cedes, as in spin, stick, 8kin, we usually lcave off this puff of breath.
As this difference is not distinctive, a forcign speaker who fails to
l'cproduce il., is still intelligible, but his speech will seem queer to
us. Frenchmen are likely to fail in this matter, because in French
the phonemes which resemble our p, l, k arc spoken always without
aspiration. On the other hand, an Englishman or American who
speaks French weil enough to he understood, is likely still ta dis-
please his hearers by using the aspiration after p, l, k.
Non-distinctive features OCCUl' in ail manner of distributions. In
most types of Arnerican English, the t-phoneme in words like
water or buller is often reduced to an instantaneous touch of the
tongue-tip against the ridge hehind the upper gums: in our habit,
the sound so prodllced suffices to represent the phoncme. In
England this variant is unknown, and is likely to be interpreted as
a variant of the phoneme d, - so that the American may flnd that
he is not underslood when he asks for water.
In the ordinary case, there s a limit lo the variability of the non-
distinctive features: the phoneme is kept distinct from al! ether
phonemes of its language. Thus, wc speak the vowel of a word like
pen in a gl'eat many ways, but not in any way that belongs to the
vowel of pin, and not in any way that belongs to the vowel of
pan: the thrce types arc kept rigdly apart.
5. 6. The faet that distinctions which are phonemic in one
language or dialect arc indifferent in others, and the faet that the
borders between different phonemes differ in dfferent languages
and dialects, appears most clearly when wc hear or try to speak a
foreign language or dalect. \Ye have just seen an instance of how
Amcrican English may be misunderstood in England. The vowc1
of words like fob, bomb, hot is in American Englsh much closer
than in British En!!;lish to t.he vowel of words like far, balrn, pa;
n sorne kinds of American English the t.wo sets of words have in
fact the sarne vowel. The Englishrnan of the south, moreover, has
lost the r-sound in words like far. A London cabman did not
undersumd me whcn l asked to be driven to the Comedy Theatre:
l had forgotten myself and spoken the American form of the first
vowel in comedy, and this the EnglishrmlO could take only as a
representative of the vowel phoneme in a word like car - so that
l was really asking for a Carmody Theatre, which does not exist.
Whell we try to speak a foreign language or dialect, we are lkely
to replace its phonemcs by the most similar phonemes of our own
language or dialect. Sometnes our native phoncme and the foreign
one overlap, so that part of the time our reproduction is correct,
but part of thc time it falls outside the range of the foreign sound.
82 THE PHONEME
THE PHONE:1IE 83
Thus, an American who pronounces the French word
(' same ') with the vowel of the English '.'lord ma'm, will only part
of the time produce a sound which meets the conventional require-
ments of the French phoneme; most of the tirne he will be produe--
ing a sound which differs decidcdly from the vowel which the
Frenchman is accllstomed to hear.
What saves the situation in wch cases is the native's complemen-
tary inaccuracy. Whcn we heur foreign speech-,;ounds we Tespond
ta them as if they eontained the eharacteristics of sorne aeoustically
similar phoneme of our native languag"c. The discrepancy disturbs
us, and wc say that thc foreigner speaks indistinctly or with a
strange "accent," but wc do not know where the diffcrence lies.
In our example, according"ly, the Frenehman will mostly under-
stand the Ameriean's pronunciation of mame, even when if con-
tains a vowel sound that wouJd never oeeur in the Frenchman's
own pronunciation. However, if our rendition deviates tao far
from the foreign phoneme, and especially if it cornes close ta some
other phoneme of the foreign language, we shall be misunderstood;
thus, sorne varieties of the American's ma'm which he uses for
French mme, will be unintelligible bccause the Frenchman accepts
them as renditions of a different phoneme whieh oeeurs, for
stance, in words lke lame (' blnde ').
The confusion is more serious when two or three of the foreign
phonemes resemble sorne one native phoneme of ours. Our infantile
trains us t ignore differences that are not
phonemic in our la'Jguage. The English-speaker will not hear any
difference hetween the l\lenomini forms a' kah 'yes, indeed,' and
ahkdh 'kettle,' and the first part of the ward akdhsemen 'plum.'
In the first of these frms, the phoneme which resembles our k
is preceded by a slight catch in the throat (a glottal stop) which
l have designated here by an apostrophe; in the second, the k
is preceded bya puff of breath (aspiration), which l have desig-
nated by h; in the third forIll these features are absent. The
English-speakcr was trained in childhood Ilot to respond to a
catch in the throat or a slight huskiness befOTe a consonant sound:
if a fellow-speaker occasionally produces such a noise, we pay no
attention ta iL
The Menomini, for his part, cannat distinguish differcnces like
thai of our t and d. Words like bad und bat sound alike ta hirn.
This appears, for instance, in the fact that the NIcnomini have
translated the ward Swede into their language as if il. wero
by the tenn sayelI1erwt 'one who is swed.' There is a ::'I.IenOlllini
pho
neme
which bath our t and d, t:
he
lVlcnomini speaker orten uHtrs variants of thls phoneme whlCh
fall within the range of our t-phoneme, and occasionally variants
which fa.l] wilhin the range of our d-phoneme, but his infantile
training taught him to ignore these differCIlccs of sound.
\Vhen wc try to speak a forcign languag"e, '',le reproduce, in such
cases, several foreig"n phonemes by one single phoneme of our own.
The native speaker, in tum, responds to our phoneme as if it
wcre Olle of his. Thus, the German hettrs no difference betwccn
the initi::l.1 phonemc of tl:n ttnd th: of thin, since bath of them
rcsemblc one of his native phonemes. When he English, he
uses th;.s German phonerne. him, wc respoml ta it as
though it wPre our t-phoneme; we are right, al any rate, in con-
cluding thal he does flot distinguish bet,ween lin and thin. In quite
the same way, when the English-speaker hears German, he will
respond Hl t.wo different. phonemcs of that lang"uage as though
they were identieal with t.he English phoneme that is initial in
words like cat, and he will fail, in consequence, ta distinguish
tween fiOlne words that are quit.e differcnt in the habit.s of the
German.
In other cases, the one phoneme which wc substitule for scv-
em] phom'mes of the foreign lang"uage, is acoustica.Ily intcrmediat.e,
and ta the n:,live spetlker wc secm t.o be intercha,nging" the sounds.
For instance, many Genllans (such as Alsatians) h:we only one
phoneme, of intel'Illediate acoustic qualty, in the spherc of our
p and b, and in speaking our language they use this for both of
our phonernes. \ylten lhey do this in a Wnllike pie, '.'le are struck
by t.he devialion in the direetion of band respond as t.hollg"h to
the word buy; on the ot.her hf1nd, when they use their interf"lCdiate
phoneme in 11 ward like buy, wc are struck by the deviation in t.he
direction of p, and respond as though we had heard pie. Hence
it secms t.o us (01' tu a Frcnehman) that the German can pl'onounce
both p and li, but perversely keeps int.erchrmg"ing t.he two.
The gl'eat.esl cliffieulty arises where a laIlg"uage makcs signifi-
cant use of features thal: play no such part ill our Ia.nguage. An
English-speaker who hears Chinesc (or any of quite a fcw other
languages), will fail lo undcrstand or to speak intelligibly, \lntH
he discovers and trains himself t.o hear and w reproduee the dis-
84
THE PHONEME THE PHONEME 85
tincHons of relative pitch which are significant in every syllable.
He docs not respond to thern at first, because as an infant he
was trained not to notice the different pitchschemes which occur
in successive utteranccs of a word like man; the Chinese infant,
on the other hand, was trained to respond to several types of snch
pi tch-schernes.
When the forcign languagc has only one phoneme in a general
acoustic type where our language has more than one, it often
secms to us as if the foreigner were using very different sounds
without a rcasonable distinction. Thus, the Menomini's or the
Alsatian's one p-b phoneme will strike our ears now as p and
now as b.
Sorne persons have an aptitude for hearing and reproducing
foreign speech-sounds; wc say that such persans arc good imitators
or havc a "good ear." 1\'1ost other people, if they hear enough of
a foreign langua@, or if they are carcfully instructed, will in time
learn to understand and make themselves understood. Practical
phoneticiuns sometimes acquire great virtuosity in discriminating
and reproducing ail manner of strange sounds. In this, tD he
sure, there lies sorne danger for linguistic work. IIaving learned
ta discriminatc many kinds of sounds, the phonetician may turn
ta sorne language, new or familiar, and insist upon rccording ail
che distinctions he has learned ta discriminate, cven when in this
language they are non-distinctive and have no bearing whatever.
Thus, having learned, say in the study of Chinese, ta hear the
difference between an aspiratcd p, t, k, (as wc usually have it
in words like pin, tin, kick) and a similar sound without aspiration
(as a Frenchman forms it, and as we usually have it in words like
spin, stick, skin), the phonetician may c1utter up his record of
English by marking the aspiration wherever he hears H, while
in reality its presence or absence has nothing to do with the rnean-
ing of what is said. Thc chief objection ta this procedure is its
inconsist.ency. The phonotician's equipment is personal and ac-
cidental; he heam those acoustic features which are discriminated
in the languages he has 0 bserved. Even his most "exact" record
is bound to ignore innurnerable non-distinctive features of sound'
the ones that appear in it are solectcd by accidentaI and persona!
There is no objection to a linJ.,'Uist's describing aIl the
acoustlC features that he cau hear, provided he does not confuse
these with the phonemic features. He should remernber that his
hearing of non-distinctive fcatures depends upon the accident of
bis personal equipment, and that his most elaborate account can-
not remotely approach the value of a mechanical record.
Only two kinds of linguistic records are scicntifically relevant.
One is a mechanical record of the gross acoustic features, such as
is produced in the phonetcs laboratory. Tho other is.a
in terms of phonemes, ignoring an features that are not dlstmctlve
in the language. Until our knowledgo of acoustics has progressed
far bcyond its present state, only the latter kind of record can
he uscd for any study that takes into consideration the meaning of
what is spoken.
In fact, the laboratory phonetician usually knows, from other
sources, the phonemic character of the speech-sounds he is
ing; he usually formulates his problems not in purely acoustlc
terms, but rather in terms which he has borrowed from practical
phonctics.
5.7. In arder to make a record of our observations, we need
a system of writton symbols which provides one sign for each
phoneme of the language we are recording. Such a set of symbols
is a phonetic alphabet, and a record of speech in the shape of these
symbols is a phonetic transcription (or, simply, a transcription).
The principle of a symbol for each phoncme is approached by
our traditional alphabetic writing, but our traditional writing does
Dot carry it out sllfficiently for the purposes of Hnguistic study.
We write sun and son differently, althollgh the phonemes are the
same, but lead (noun) and lead (verb) alike, though the phonernes
are different. The words oh, owe, so, sew, sow, hue, beau, Ihough al!
end with the same phoneme, variously represented in writing; the
words though, bough, through, cough, tough, hiccough end with du-
ferent phonemes but are ail written with the letters -ough. Our
lettN x is superfluous because it reprcsents the sarne phonernes
as ks (as in tax) or gz (as in examine); our letter c is superfluous
because it represents the same phoneme as k (in cat) or as s (in
cent). Although we have the letter j for the initial phoneme in
jam, we also use the leUer 9 (as in gem) for this same
Standard English, as spoken in Chicago, has thirty-two Simple
primary phonemes: the twenty-six lelters of our alphabet are tao
few for a phonetic record. For some phonernes we use combina-
tions of two letters (digraphs) , as th for the initial phoneme in
thin, ch for that in chin, sh for that in shin, and ng for the final
86 THE PHONE:VIE
THE PRONEME 87
phoneme in sing. This leetds to furtlH'r incoI};;isteneies: in then
we use th for a different phoneme, and in hot!wusc for tbe two
phonemes which are normally represenLcd by the separate let-
ters i and h; in Thomas the th has the value of the phoneme 01'-
dinarily represented by t. In singer we use ng for 11 sinp;le phoneme,
as in sing, but in finger the leti.ers ng repl'cscnt this phoneme 1)lus
the phoneme ordinal'ily rE'presented by the Ict,r (J, as if; go.
Tradilional aIphetbetic writ ing is etccuraLe only in the case of a
few languages, such as Spanish, TIohemian, Polish, and Finnish,
whe1'e it has been shetped or revised b,Y persom who had worked
out. the phonemic system of thei1' bngu[tge.
5.8. On account of the imperfections of tmditional writing and
t.he Jaek of a sufficient number of characters in our (so-ca!led
"L:ttin") alphabd, sehobrs have dcviscd lllany phonetic al-
phabets.
SOIlle of t.hese sehemcs depart enlirclv from our traditional
habils of writing. BeWs" Visible Speecl;" is the b('st-known of
these, chieRy because Henry Swcet (\S45 1912) liScd it. The sym-
bols of this alphabet are sirnplified and conventionalizcd diagmms
of the vocal org,lIls in position for the utterancc of the variolls
phonemcs. Visible Speech is bard to write and very costly 1.0
print.
Another system which departs from tlw historical tradition is
.Jespersen's "Analphabetic Notation." Hcre ('very phoneme is
l'eprescnted by li whole sct of symbobi which consist of Greek
letters and Arabie mimerais, \Vith Latin letters as exponents.
Each Greek letter indicatE's an organ and e:wh a degree
of opening; thus, a indieates t he li ps and 0 iwlieatcs cJosure, so
that aO will appcar in t.he lormula for any phoneme during
whose ult.er:\Ilee the lips are closed, sw:h as our p, b, and m
phonemes. The formula f0r the En?;lish m phoneme, as in man
is aO 02 fi, ,vher b2 means the !Jack of the paIate is lowered:
and fI means that the vocal cbords I1r<> ir.. vibration. The advan-
tages of this not.ation are evident, but of course it is not intended
for the recording of whole utterances.
l\Tost phonetic alphabets are modifications of the traditional
alphabet. They supplement the orclinary letters by ,mch devices
as small cupitals, letters of the Greek alphabet, distorted forms of
conventioua.1 letters, and lett.ers with li t. tlc marks, diacritical signs,
attached ta them (e.g. and a). Thel'e arc man)' alphabets of this
type, sueh us that. of Lcpsius, used for Afl'ican languages; of
Lundell, uscd for Swcdish dialects; of Bremer, used fol' German
dialects; of the American Ant.hropologieal Association, uscd for
Arnerican Indian languages. In this book wc shall use the alphabet
of the Int.ernational Phonctic Associatian; this alphabet l'las
veloped by Ellis, Sweet, Passy, und Daniel Jones. A crude form
of phonetie alphabet appears in the "keys to pronunciation" of
rnost. dictionaries. Similar dcvices have grown up in the
tional writing of sorne 1I1n?;u,.ges, devices such as the two dot.s over
vowel1et.ters in German writing (d, 0, ) or the diacritical marks in
Bohernian writing ( for our ch, s for our sh); the Hussian and
Serbi:tn alphabets supplement the Greek alphabet wit.h a number
of extra letters.
In principle, one phonetie alphabet is about as good a8 another,
sinee aIl we ncccl is a fcw dozen syrnbols, ,"nough to supply one for
cach phoneme of what.ever language we are recording. In their
application, howevcr, ail phonetic alphabets suffer frorn serious
drawbacks. 1,Vhen they were invented, t.he principle of the phoneme
had not been elcarly recognized. The inventors meant their alpha-
bet.s to he rich and flexible enouf!;h ta offer a symbol for every
acoustic varicty that could be heard in any languagc. It is evident,
today, t.hat a record of ihis kind would amount to llothing less t.hao
a rnechanical recording of t.he sound-waves, which would be the
same for no two utterances. In practiee, the phonemie principle
sornehow slipped in: usually one wrote a symbol for eaeh phoneme,
but these symbols l'lere highly differentiated and cluUered up with
diacritical marks, for the purpose of indicating "exact" acoustic
values. The val'ieties that l'lere in this ,vay distinguished, were
merely those which phoncticians happened t.o have noticed.
Henry Sweet devised a relatively simple system, bascd on the
Latin alphabet, which he ealled Rurnic, for use alongside of Visible
Speech. When the phonemic principle became clear ta him, he
reaIized that his Romic notation would still he sufficient if one
greatly simplified it. Accordingly he used a simplified form, with
a symbol for cach phoneme, and callcd it Broad Romic; he still
believed, however, that the more cornplex fonn, Narrow Ramie,
was somehow "more accurate" and better suited to scientific
purposes.
Out of Sweet's Romic thcro has grown the alphabet of the
ternatinal Phonetic Association, which consist.s, accordingly, of
88 THE PHONEME
THE PHONEME
89
the Latin symbols, supplemented by a number of artificialletters,
and a few diacritical marks. In a modified form, we shaH use it in
this book, pladng between square brackets, as is customary, every-
thing that is printed in phonetic symbols.
6.9. The principle on which the International Alphabet ig
based, is to empIoy ordinary letters in values approximating the
values they have in sorne of the chief European languages, and to
supplement these letters by artificial signs or by the use of dia-
critical marks whenever the numbcr of phonemes of a type ex
ceeds the numher of ordinary letters. Thus, if a language has one
phoneme of the general type of our tsound, we symbolize this
phoneme by the ordinary leUer ft], regardless of whether this
phoneme is acoustically quite like the English or the French t-
sound, but if the language has two phonemcs of this general type,
wc can symbolize only one of them by [t], and for the second one
we must resort to the use of a capital [T], or an italic ft], or sorne
other similar device. If a language has two phonemes of the gcneral
type of our e-sound as in pen, wc use the letter [el for one of them,
and the supplementary symbol [el for the other, as in pan [pen].
These principles, which the International Phonetic Association
formulated as early as 1912, have heen neglccted even by its
members; most studcnts have failed to break away from the tradi-
tion of the time when the phonemic principlc had not yet been
recognized. Thus, we find most writers using queer symbols for
EngIish phonemes because it has been recognized that English
phonemes differ from the most similar types of French phonemes.
For instance, having pre-empted the symbol [0] for the phoneme of
French eau [0] ('water'), these authors do not use this letter for
recording the English vowel in son, hecause this English phoneme is
unlike the French phoneme. In this and sorne other respects, 1
shall depart in this book from the usage (but not from the prin-
eiples) of the International Phonetic Association.
Where several languages or dialects are under discussion, each
one must he recorded in terms of its own phonemes ; the differ-
ences, sa far as we are able to state them, may deserve a verbal
description, but must not he allowed ta interfere with our symbols.
Thus, even a phonetician who thinks he can describe in accurate
terms the differences hetween the phonemes of standard English as
spoken in Chicago and as spoken in London, will add nothing to
the value of his statements by using queer symbols for one or the
other of these two sets of phonemes, and he will only make things
still harder if he uses outlandish symbols for bath of them, hecause
he happens ta know that the ordinary letters have been used for
recording the somewhat different phonemes of sorne other language.
The principle of a single symbol for a single phoneme may he
modificd without harm only where no ambiguity can result. It
rnay he advisable, where no ambiguity can result, ta depart from
the strict principle when this saves the use of extra symbols tbat
rnight be disturbing to the reader or costly ta print. In sorne lan-
guages, sounds like our [p, t, k] with a slight puff of breath after
them, are distinct from sounds like the French [p, t, k] without this
aspiration; if the language has no phoneme designated by [hl, or if
it has such a phoneme but this phoneme never occurs after [p, t,
k], then it is safe and economical ta use the compound symbols
[ph, th, khI for the former type.
6. 10. The matter of recording languages is eomplicated not
only by the existence of severaI phonctic alphabets and by in-
consistencies in their application, but also by the frequent use of
two other devices alongside phonetic transcription.
One of these devices is the citation of forms in their traditional
orthography. This is often donc where the language in question
uses the Latin alphabet. The author either supposes that his
reader knows the pronunciation, or else, in the case of ancient
languages, he rnay not care to guess at the pronunciation. Citation
is often helpful ta readers who are familiar with the ordinary
orthography; it is only fair, however, ta add a transcription, e.g.
French eau [0] 'water.' Even in the case of andent languages it is
often useful ta add a guess at the pronunciation, e.g. Old English
geoc [jok] 'yoke.' Only in the case of languages like Bohemian or
Finnish, whose traditional orthography is entirely phonetic, can
one dispense with a transcription. In the case of Latin, a citation
with a macron over long vowels is sufficient (e.g. amii.re 'ta love'),
since, sa far as we know, Latin orthography was phonetic except
that it failed to indicate the distinction hetween long and short
vowels.
For languages which use alphabets other than the Latin, citation
is less often employed. It is customary in the case of Greek, less
often of Russian, but is in every way to be deplorcd. Sorne luxurious
publications indulge even in Hebrew, Arabie, and Sanskrit type for
citing these languages. The only reasonable exceptions here are
90 THE PHONEME
THE PHONEME 91
forms of writing like the Chinese and the ancent Egyptian, whose
syrnbols, as we shaH see, have meaning-values that cannot he
represented in phonetic terms.
For languages wbich use writing of sorne form other tban the
Latin alphabet, iransliteration is often employed instead of tran-
scription. Transliteration consists in assigning sorne letter of the
Latin alphabet (or sorne group of letters or sorne artificia1 symbol)
ta each character of the original alphabet, and thus reproducing
the traclitional orthography in terms of Latin 1etters. Unfortu-
nate1y, differcnt traditions have grown up for transliterating differ-
cnt languages. Thus, in transliterating Sanskrit, the Latin letter
c is used ta represcnt a Sanskrit letter which seems to have des-
ignated a phoneme mueh like our initial phoneme in words like
chin, but in transliterating the Slavic alphabet, the letter c is
used to represent a letter whieh designatcs a phoneme resembling
our ts combination in hats. For most linguistie purposcs it would
he better to use a phonetie transcription.
6. 11. It is not difficult (even aside from the help that is af-
forded by our alphahetie writing) to make up a list of the pho-
nemes of one's language. One nced only proceed with a moderate
number of words as we did I1bove with the word pin, to find that
one has identified every phonerne. The number of simple primary
phonemes in different languages mns from about fifteen to about
fifty. Standard English, as spoken in Chicago, has thirty-two.
Compound phonemes are combinations of simple phonemes Wh1Ch
aet as units sa far as meaning and word-structure are concerned.
Thus, the diphthong in a word like buy can be viewed as a com-
bination of the vowel n far with the phoneme that is initial in
yeso Standard English has cight such combinations.
It is somewhat harder ta identify the secondary phonemes. These
are not part of any simple mcaningful speech-form taken by itself,
but appear only when two or more are combined into a larger
form, or else when speech-forms are used in certain ways - espe-
cially as sentences. Thus, in English, when wc combine severa1
simple elements of speech into a word of two or more syllables,
we a1ways use a secondary phoneme of stress which consists in
speaking one of these syllables louder than the other or others:
in the word foretell wc speak the tellloudcr than the fore, but in
foresighi the fore is louder than the sight. The noun contest has
the stress on the first syllable, the veIb contest on the second. Fea-
tures of pi/ch appear in Englsh as secondary phonemes chiefly
at the end of sentences, as in the contrast between a question (at
four o'clock?) and an answer (atfour o'clock). It is worth noticing
that Chinese, as well as many other languages, uses eatures of
piteh as primary phoncmes. The secondary phonemes are harder
to observe than the primary phonemes, because they occur only
in combinations or in particular uses of simple forms (e.g. John?
in contrast with John).
The principles we have outlined would probably enable anyone
familiar with the use of writing ta work out a system of tran-
scribing his language. In this book the English examples will he
transcribed, unless otherwisc indieated, according to the pronun-
ciation of standard English that prevails in Chicago. This re-
quires thirty-two symbols for simple primary phonemes and nine
for secondary phonemes.
PRIMARY l'aONEME8
[ a ] alms [amz] [ i ] pin [pin] [ r ] rod [rad ]
[ a 1 odd [ad
1 [ i 1
yes [jes
1
[ s 1 S(ld [sud 1
[ b] big [big 1 [ j ] gem [iem 1 [ il 1 shove [Sv 1
[ c 1 chin [Cin 1 [ k 1 cai [kEt 1 [ t 1 tin [tin
1
[ d 1 dig [dig 1 [ 1 ] lamb [lem 1 [ e J thin [Gin
1
j
[ ~ ] t h e n [tien] [ml mtS8 [mis ] [ u 1 put [put 1
( e ] egg [eg ] [ n ] knot [nut 1 [ V 1 van [ven ]
[ e ] add (ed
1
[ 1) ] smg [sil)
1
[ wl wag [weg]
[ f ] fan [fen 1 [ 0 ] up [op ] [ z ] zip [zip ]
[ g ] give [giv ] [ 0 ] ought rot l
[ z ] rouge [ruwz]
[h]hand [hend] [ p ] pm [pin]
COMPOUND PRIMARY PliONEME8
[ai] buy [bai ] [ij ] bee [bij ] 1oj ] boy [boj
1
[aw] bough [baw] [juw] few [fjuw] [uw] do [duw 1
[ ej] bay [bej ] [ow] go [gow]
SECONDARY l'HONEME8
[" }, placed before primary symbols, loudest stress: That's mine!
[tiet s "majn!].
[ , l, placed before primary symbols, ordinary stress: forgiving
[for'giviIJJ; l've seen it [aj v 'sijn it].
92
THE PHONEME
[ 1 J, placed before primary symbols, less loud stress: dinil/{J-room
['dajniIJ lruwm]; Keep it up [lkijp it 'op].
[ 1 J, under one of the primary symbols [l, m, n, rl, a sIight
stress which makes this primary phoneme louder than what
precedes and what follows: coral ['karl], alum ['dm] apron
('ejprI;IJ. pattern ['petrn].1 , ,
[ . J, placed after primary symbols, the falling pitch at the end of
a statement: l've seen it [aj v 'sijn it.].
[l ], placed after primary symbols, the rising-falling pitch at the
end of a question to he answered by speech-forms other than
yes or no: Who's seen il? ['huw z 'sijn itl].
[ ? J, placed after primary symbols, the rising pitch at the end of
a yes-or-no question: Have you seen it? [hev juw 'sijn it?J.
[ J], plaeed primary symbols, the distortion of the pitch-
scheme In exclamations: It's on fixe! lit s an 'fajd], Seven
o'clockf! ['seVI). 0 "klak?!].
[ , j, placed hetween primary symbols, the pause, orten preceded
by rising pitch, that promises continuation of the sentence:
John, the oider boy, is awayat school ['Jan {siJ 'owldr 'b:)J' iz
1 "
e weJ et 'skuwl.J.
1 Contrast above, forms like [karl], elm [elm]. It is eustomary and
ereatea no ambigu, ty to put the sign 1,] under the symbo!s [l, m, n, r] whenever
these phonemes are louder thall what precedes and what follows (and, aecordingly.
as we say, f?rm a even though no inerease of "trcas is required to make
them BO, as ln bOUle 1batll, boUom [' hat':ll, buUon (" bot.], bird [br
d
]; see 7.10.
CHAPTER 6
TYPES OF PHONEMES
6. 1. While the general principles which wc surveyed in the last
chapter will enabIe an observer to analyze the phonetic structure of
his own speech, they yield very little help, at the start, for the
understanding of a strange language. The observer who hears a
strange language, notices those of the gross acoustic features which
represent phonemes in his own language or in other languages he
has st.udied, but he has no way of knowing whether these features
are significant in the Jan!!;uage he is observing. Moreover, he fails
to notice acoustic features which are not significant in his own
language and in the other languages he has studied, but are signifi-
cant in the new language. His first attempts at recording contain
irrelevant distinctions, but. fail to show essential ones. Even a
mechanical record will not help at this stage, since it would register
the gross acoustic features, but would not tell which ones were
significant. Gnly by finding out which utteranccs are alike in
meaning, and which ones are different, can the observer lcarn to
recognize the phonemic distinctions. So long as the analysis of
meaning remains outside the powers of science, the analysis and
recording of languages will remain an art or a practical skill.
Experience shows that one acquires this skll more casily if one
is forewarned as to the kinds of speech-sounds that are distinctive
in various languages - although it is true that any new language
may show some entirely unforeseen distinction. This information
is most easily acquired if it is put into the form of a rough descrip-
tion of the actions of the vocal organs. This rough description is
what we mean by the term practical phonetics. After the observer
has found out which of the gross acoustic features are significant
in a language, his description of the significant features can he
ilIustrated by a mechanical record.
6.2. We have no special organs for speech; speech-sounds are
produced by the organs that are used in breathing and eating.
Most speech-sounds are produced by interferenee with the out-
going breath. Exceptions to this are suction-sounds or clicks. As
93
94 TYPES OF PHoNEMES TYPES OF PHONEMES 95
a non-linguistic sign of surprised commiseration (and a]so as a
signal ta urge horses), we sometimes make a click- the novelist
represents if. by tut, tut! - wiih the tangue against the ridge just
back of the upper teeth. As speech-soumIs, various clicks, formed
in different parts of the mouth, are used in some African languages.
6. 3. The first interference which the out!!;oing brcath rnay mect,
is in the larynx. The larynx is a box of cartilage at the head of the
wind-pipe, visible from the outside as the adam's-apple. \Vithin
the larynx, at the right and left, are two shelf-like muscu]ar pro-
tuberances, the vocal chords. The opening between them, through
whieh the breath passes, is called the y/oUis. In ordinary breathinp;
the vocal chords are relaxed and the breath passes freely through
the glottis. At the rear of the larynx, the vocal chords are attached
to two movable cartilaginous hinp;es, the arytenoids. Thanks io
delicate museu]ar adjusiments, both the vocal chords and the
arytenoids can he set into a number of positions. The extrernc
positions are the wide-opcn position ot' ordinary breaihing and the
firmly closed position which occurs when onc holds one's breath
with the mouth wide open. Varions lan!!;uages make use of various
intermediate positions of the glotts.
One of these positions is the position for voicing. In voicing, the
vocal chords are drawn rather tightly together, so that the breath
can get through only frorn instant t.o instant. In getting ihrough,
the breath-stream sets the vocal chords into vibration; the fre-
quency ranges from around eip;ht.y to around one-thousand vibra-
tions per second. These vibrations, communicated to t.he outer air,
strike our ears as a musical sound, which we call the voice. The
voicc does noi play a part in ail speech-soumIs: we distinguish
betwcen voiced and unvmd (or breathed) If one
places a finger on the or, better, if Oiie presses one's
palms ti!!;htly over one's cars, and then utters a voiced sound, such
as [v] or [z], the voice will be felt. as a trernbling or vibration, while
unvoiced sounds, such as [f] or [s] will lack this buzzing accompani-
ment. It seems thai in every language at least a few phonemes have
lack of voicing among their fixcd characteristics. During thc
production of most unvoiced sounds the glottis is wide open, as in
ordinary brcathing.
Variaus adjustments cnable us t alter thc loudness 'lnd the
pitch of the voice-sound as weil as its quality of resonance. These
last variatians, such as the "hcad regisr," "chest register, "
<lmufHed sound," "metallic sound," and the like, have not Dcen
physiologically analyzcd.
Among the positions intcrmediate between breathing and voicing,
several deserve mention. If the vocal chords are sa far separated
that the voice no longer sounds pure, but is accompanicd by the
friction-sound of the breath passing ihrough the glottis, wc get a
murmur. In English, the unstrcssed vowels are orten spoken with
murrnur instead of voice. As a phoneme, the murmur occurs in
Bohemian, where it may he transcribed by the symbol rh], which
is used in the conventional orthography of this langua!!;e. If the
glottis is still farther opened, the VOlee ceases and only a friction-
sound remains; this friction-sound characterizes our phoneme [hl,
as in hand [ht:nd]. Another intermediatc position is the whper,
in which only the - that is, the spaec between the
- is open, but the vocal chords are in contact. In
what we ordinarily call "whispering," the whispcr is substituted
for the voice and the unvoiced sounds are produced aS in ordinary
speech.
The sound-waves produced by the vibration of the vocal chords
in voicing, are modified by the shape and by the elasiicity of the
channel through which they pass before they reach ihe outer air.
If we compare the vocal chords ta the rceds of 11 wind-instrument,
we may view the mouth, or rather, the whole cavi ty {rom the vocal
chords ta the lips, including, in sorne cases the nasal caviiy, as a
By setting the rnouth into various positions,
by cutting off the exit eiiher through the mouth or through the
nose, and by tightening or loosening the muscles of this region, we
vary the configuration of the outgoing sound-waves.
In contrast with musical sound, noises, which cOiisist of irregular
combinations of sound-waves, ean be produeed by means of the
glottis, the tangue, and the lips. Sorne voiced sounds, sucb. as
[a, m, 1], are purely musical, that is, relatively free from noise,
while others, such as [v, z], consist of u. noise plus the musical sound
of voicing. Unvoiced sounds eonsist merely of noises; examplcs
are [p, f, 81.
6.4. When the breath leaves the larynx, ii passes, in normal
breathing, through the nose. During most speech, however, we
eut off this exit by raising the velum. The velum is the soft, mov-
able back part of the palate; at the rear it ends in the ulfUla, the
Httle lobe that can be seen hanging down in the cent.cr of the mouth.
96 TYPES OF PHONEMES TYPES OF PHONEMES 97
If one stands before a mirror, breathing qllietly through nase and
mouth, and thenspeaks a clear [a], one can see the raising of the
velum, especially if one watches the uvula. When the velum i.s
raised, its edge lies against the rear wall of the breath-passage,
cutting off the exit of the breath through the nase. Most sounds of
speech are purely oral; the velum is completely raised and no
breath escapes through the nase. If the velum is not completely
raised, sorne of the breath escapes through the nase and the
speech-sounds have a peculiar resonance; such sounds are called
nasalized sounds. In English the difIerence betwccn puroly oral
and nasalized sounds is not distinctive; we often nasalize our
vowels hefore and aSter the phonemes [m, n, Il], and we nasalize
more than usuaI when we are tired or relaxed. In sorne languages,
however, nasalized sounds, most commonly voweh'l, are separate
phonemes, distinct from similar sounds without nasalization. The
usual symbols for nasalization are a small hook under a letter (this
is used in the traditional orthography of Polish), or a tilde over a
letter (Portuguese orthography and International Phonetic As-
sociation), or an exponent [n] after a letter (used in this book,
because easier t' print). French has four nasalized vowels as
phonemes, distinct from the corresponding purely oral vowels:
bas [bal 'stocking,' but banc [ban] 'hench.'
If thc velum is not raised and the exit of the breath through the
mouth is in any way cut off, then, as in ordinary breathing, aIl the
breath escapes through the nosc. Phonemes where this is the case
are nasal. In English we have three nasals: lm], in which the Iips
are closed; ln], in which the tongue is pressed against the gums; and
[Il], as in sing [silJ], in which the back of the tongue is pressed
against the palate. These are purely musical sounds, characterized
by the resonances which the different shapes of the oral-nasal
cavity give ta the musical sound of the voice. Sorne languages,
however, have unvoiced nasals as phonemes; these are audible not
so much by the very slight friction-noise of the breath-stream, as
by the contrast with preceding or following sounds and by the
intervening non-distinctive glide-sollnds that are produced while
the vocal organs change their position.
A good test of nasalization is to hold a card horizontally with
one edge pressed against the upper lip and the opposite edge against
a cold pane of glass' if one DOW produces a purely oral sound, such
as [a], the pane will he misty only under the card; one produces
a nasalized sound, such as [an], the moisture will appear both above
and below the card; and if one produces a purely nasal sound, such
as [ml, the moisture on the pane appears only above the cardo
6.5. We change the shape of the oral cavity by placing the lower
jaw, the tongue, and the lips into various positions, and we affect
the resonance also by tightening or loosening the muscles of the
throat and mouth. By these means every language produces, as
phonemes, a number of musical sounds, such as our [al. in palm
[pam], our li] in pin [pin], our [u] in put [put], our [r] ID rubber
l'robr], and so on. In some of these the tongue actually touches
the roof of the mouth, but leaves enough room at one or both sides
for the breath to escape without serious friction-noise; auch sounds
are lalerals, of the type of our [1], as in littk ['litt]. In unvoiced
lateraIs, whieh occur in Welsh and in many American languages,
the friction-noise of the breath-stream is more audible tban in
unvoiced nasals.
We make noises in the mouth by movements of the tongue and
lips. If we place theae organs (or the glottis) so as ta leave a very
narrow passage, the outgoing breath produces a friction-noise:
phonemes characterized by thia noise are spirants (fricatives). They
may be unvoiced, as are our [f] and [aJ, or voiced, like our [v] and
[z]. Since the amount of friction can he varied ta any degree, there
is no real boundary between spirants and musical sounda such as
li] or [1]; especially the voiced varieties occur in different languages
with many degrees of closure.
If we place the tangue or the lips (or the glottis) sa as ta leave
no exit, and allow the breath to accumulate behind the closure, and
then open the cloaure, the breath will come out with a
slight pop or explosion; sounds formed in this way are stops
(plosives, explosives), like our unvoiced [p, t, k] and our voiced
lb, d, g]. The characteristic feature of a stop is usually the explo-
sion, but the making of the closure (the implosion) or even
brief period of time during closure, may suffiee to characterlZC
the phoneme; thus, in English we sometimes leave off the explosion
of a final [p, t, k]. These varieties are audible by contrast with
what precedes or follows (as a sudden stoppage of sound or as a
moment of silence), or else through the transitional sounds during
the movement of tangue or lips; al8O, during the closure of a voiced
stop one can hear the mufHed sound of the voice.
Sinee lipa, tangue, and uvula arc elastic, they can he placed sa
98 TYPES OF PHONEMES TYPES OF PHONEMES 99
that the hreath sets them ioto vibration, with alternate moments of
contact and opening. Snch trills OCCUT in many languages; an
example is the British English "rolled T," as in Ted or hOTTid.
'Ve shall take up the chief types of phonernes in the following
order:
noise-sounds:
stops,
trills,
spirants;
musical sounds;
nasals,
laterals,
vowels.
6. 6. Stops occur as phonemes in perhaps every language.
English distinguishes three types as to position: labial (more
exactly, babial) , in which the two lips form the closure [p, b];
dental (more exactly, a/veolar, or better gingival), in which the
tip of the tongue makes closure against the ridge just back of
the upper gums [t, d]; and velar (in older wri tings mis-called
guttural), in whieh the back of the tongue is pressed against the
velum [k, g].
Thesc last two types occur in many varieties, thanks to the
mohility of the tongue. Contact can he made by the tip of the
tangue (apical articulation) or by a larger area, the blade, round
the tip (coronal articulation); it can he made against the edgcs of
the upper teeth (interdental position), against the backs of the
upper teeth (pas/dental position), against the ridge back of the
upper teeth (gingival position), or against points still hi!J;her up on
the palate (cerebral or caC1Iminal or, better, inverted or domal
position). Thus, apical articulation in the damaI position (the
tip of the tongue touching almost the highest point in the roof of
the mouth) occurs as a non-distinctive variant alongside the
gingival [t, d] in American English. In French the ncarest
sounds to our [t, d] are pronounced not gingivally but as post-
dentals (the tip or blade tQuching the back of the teeth). In
Sanskrit and in many modern languages of India, postdentals [t, d]
and domals (usually transcribed by a letter with a dot under it,
or by italics, or, as in this book, by small capitals [T, Dl) are dis-
tinct phonernes.
Similarly, diffcrent parts of the back of the tongue (dorsal
articulation) may be raised so as ta touch different parts of the
palatc; one distinguishes, usually, between anterior or pakUal
position and pos/mor or velar position, and, still farther back,
uvular position. In English the velars [k, g] are closed farthcr
forward before sorne sounds, as in kin, {[ive, and farther backward
beforo others, as in cook, good - both types in contrast with, say,
calm, guard - but these variants are not distinctive. In sorne
languages, such as Hungarian, thcre arc separate phonemes of the
palatal and velar types, which wc distinguish in transcription by
such devices as [cl for the palatal and [k] for the velar unvoiced
stop. In Arabie a velar unvoiccd stop [k] and a uvular unvoiced
stop [q] are distinct phonemes.
A glottal or laryngal stop is produced by bringing the vocal
chords tightly togcthcr and then letting them spring apart under
the pressure of the breath. We sometimes produce this sound
before an initial stressed vowel when speaking under a strain, and
in German this is the normal usage; as a phoneme, the glottal stop
oecurs in rnany languages, as, for instance, in Danish, whcre there
is a distinctive difference, for example, between hun [hun] 'she'
and hund 'dog.'
As to the manner of forming the closme, aside from the differ-
ence of unvoiced and voiced, the amount of brcath-pressme and the
vigor of action in the lips or tongue may be varionsly graded:
pressure and acHon are gentle in lenes, vigorous in in $olu-
tion-lenes the opening-up is relativcly slow, so as to wcaken the
explosion. The unvoiced stops may be followed by a puff of
unvoiced breath (aspiration) or preceded byone (pre-aspiration);
the voiccd stops, similarly, may be preceded or fol1owed by un-
voiced breath or by a murrnur. The closure may be made simul-
taneously in two positions, as in the [gb] stops of sorne African
languages; many languages have glottaled oral stops, with a
glottal stop occurring simultaneously, or just bcforc, or just after
the opening of the [p, t, k]. In English the unvoiced stops are
aspirated fortes, but other types occur as non-distinctive variants,
notably the unaspirated lenis type after [s], as in Bpin, s/one, skin.
Our voiced stQPS are lenes; at the beginning or at the end of a word
they are not voiced through their whole duration. In French the
unvoiced stops [p, t, k] are fortes and, as a non-distinctive variant,
may be accompanied by li simultaneous glottal stop, but are never
aspiratcd; the voiced lb, d, g] are lenes, more fully voiced than in
100 TYPES OF PHONEMES TYPES OF PHONEMES 101
English. In North Chinese, aspirated and uuaspirated unvoiced
stops are different phonemes, e.g. [phal versus [pa], and voiced
stops occur only as non-distinctive variants of the latter. Many
SouthwGerman dialccts distinguish unvoiced unaspirated fortes
and 1enes, which wc may transcribe by [p, t, k] and [b, d, gJ; voiced
variants are not distinctive. Sanskrit had four such types of
stops: unvoiced unaspirated [pl, aspirated [ph], and voiced lln-
aspirated [b], aspirated [bh].
6. 7. The commonest trill is the apical or tongue-tip trill, in
which the tangue-tip vibrates in a few rapid strokes against the
gums; this is the "rolled" r of British English, Italian, Russian,
and many other languages. Bohemian distinguishes two phonemes
of this type, the one accompanied by a strong friction sound. The
uvular tri Il, in which the uvula vibrates against the uplifted back
of the tongue, occurs in Danish, in the commoner pronunciation of
French, German, and Dutch, and in varieties of English (the
"Northumbrian burr"); in these languages, as weIl as in Nor-
wegian and Swedish, the uvular and the tongue-tip trill arc gea-
graphic variants of t.he same phooeme. The phonetic symbol for a
trill is [r]; if a language has more thao one trill phoneme, [R] is a
handy character.
If the tongue-tip is allowcd to make only a single swing, with one
rapid contact against the gums or palate, we havc a tangue-ftip.
In the Central-"\Vcstern type of American English, a voiced gingival
tongue-Bip occurs as a non-distinctive variant of ft] in forms like
water, butter, ai ail; differcnt types of tongue-Bip occur in Norwegian
and Swedish dialects.
6.8. The positions in which spirants are formed in English
differ from thosc of the stops. In one pair, the labiodentals [f, v],
the breath-stream is forced to pass between the upper teeth and the
lower Hp. In the dentals [0, ti], as in thin [Oin], then [tien], the
blade of the tongue touches the upper teeth. Our gingival spi-
rants [s, z] are hisses or sibilants: that is, the tonguc is constricted,
50 as ta bulge up at thc sides and leave only a narrow channel
along the center, through which the breath is forced sharply
against the gums and teeth, giving a sonorous hiss or buzz. If
we draw the tangue a Httle ways out of this position - in English
we draw it back - the breath is directed less sharply against the
gurus and teeth, and seems to eddy round bcfore finding an exit:
in English these hushes or abnormal sibilanis are separate phonemes
[8, z], as in shin [sin], vision ['viz!].].! In each of these positions wc
have a pair, voieed and unvoiced. Many other varieties occur,
such as bilabial spirants, in whieh the narrowing is made between
the two lips (an unvoiced variety in Japancse, a voiced in Spanish).
In French the hisses are formed postdentally; to our ears the
Frenchman seems to have a slight lisp. German, which has no
[z], protrudes the lips for [8], so as to accentuate the eddying sound.
Swedish has a [8] with very wide opening, which sounds gueer to
English ears.
English has no dorsal spirants, but they occur in many languages,
in a great variety of positions, including lateral types. German
has an unvoiccd palatal spirant, in which the middle of the tongue
is raised against the highest part of the palate; as a non-dis-
tinctive variant of this, it uses a velar type, an unvolccd spirant
in the position of our [k, g, lJJ. The customary transcription of
German uses two symbols, [] for the palatal variety, as in ich
[i] 'l', and [x] for the velar variety, as in ach [ax] 'oh,' but only
one symbol is needed, sinee the varietics depend upon the pre-
ceding phoneme. Voiced spirants ['Y] of the same position occur
in sorne types of German pronunciation as variants of the stop [g];
in Dutch and in modern Greek they orcur as separate phonernes.
Uvular spirants occur in Danish as variants of the uvular trill,
in other languages as distinct phonemes.
In English we have an unvoieed glottal spirant, [hl as in kit [hit],
when [hwen], Mw [hjuw], in which friction is produced by the pas-
sage of the brcath through the slightly opencd p;lottis; Bohernian
has a similar sound in which the friction is accompanied by voice
vibrations (murmur). A further pair of glottal spirants, unvoieed
("hoarse h") and voiced (" ayin"), occurs in Arabic; their char-
acteristic feature is said ta he a tightcning of the throat-muse1es.
As to manner, spirants show perhaps less variety than stops.
Among languages which distinguish two varietics of manner,
French voices its [v, z, z] more completely than does English. Sorne
languages have glottalizcd spirants (preccded, accompanied, or
followed by a glottal stop).
6.9. ThE' positions of nasals are mueh like those of stops; in
English [m, n, lJ] are spoken in the same three positions as the
1 The IPA uses other characters, " "long 8" and "long z"; the characters ",bave
are use<! h the traditional orthography of Bohemian and in many schemcs of
tranaliteratian.
102
TYPES OF PHONEMES
TYPES OF PHONEMES 103
stops. On the same principle, French 8peaks its [n] in postdental
position, like its [t, dl On the other hand, French has no velar
nasal, but has a palatal nasal, in which the closure is made by
raising the middle of the tongue against the highest part of the
palate, as in Bigne [sill] 'sign.' As in the stops, Sanskrit and mod.
eru Indian languages distinguish betwccn a dental [n] and a
damai [N].
6.10. In English the lateral [IJ is apical, in gingival position;
at the end of words we use a non-distinctive variety in which the
middle of the tongue is excessively lowered; contrast wss with
well. In German and French the [1] is spoken with the surface of
the tangue more raiscd; the acoustic impression is quit.e differentj
in French, moreovcr, the contact is postdental. Halian has a
palatallateral, distinct from the dental, with the back of the tangue
touching the highest point of the pa.late but leaving free passage
for the breath at one or bath sidcs:figlio ['fi\o] 'son.' Some Amer-
ican languages have a whole series of laterals, with differences of
position, glottalizaton, or nasalization. Unvaiced laterals es-
pecially if the contact lS extensive, take on a spirant
laterals, especially if the point of contact is minute,
wlt.h vowPls; thus, one of the two laierai phonemes of Polish
strikes our oar almost as a [w]. On the other hand, the Central-
\Vestern Arnerican English vowc1 [r], as in red [rrdl, fur [fr], far
[far], is C!osely akin ta a lateraI: the tip of the tongue is raised to
damaI (inverted) position, but does not quite make a contact. In
transcription we use the same syrnbol [r] as for the trill of oUler
languages; this is convenient, hecause our sound and the British
English trill in red are geographic variants of the same phoneme.
6. 11. Vowels are modifications of the voice-sound that involve
no closure, friction, or contact of the tongue or lips. They are
ordinarily voiced; sorne languages, however, distinguish different
voice-qualiiies, such as mnifled vowels, murrnured vowcls, with
slow vibration of the vocal chords, or whpered vowels, in which
friction between the arytenoids replaces vibration of the vocal
chords.
l
1 In .....,th vowels, the other sounds (stops. trilIs. spirants. nasals, Iaterals)
are "ometJmes c"Ued consono.nts. Our school grammar uses the terms "vowel"
and in .an t way, referring to lettcrs rather than
t.he desenptlOn of mdl vldual languages. il. is often convenient ta use these terms
Hlother ways nnd to supplement them by such as or whos" ap-
plicano[l we "hall see in the next chapter.
Every language distinguishes at least several different vowel
phonemes. The differences hetween these phonemes seem ta he
largely differences of tangue-position, and ta consist, acousbcally,
of differences in the distribution of overtones. Even these principles
are disputedj in what follows 1 shall state the tongue-positions
according to the generally accepted seheme, which has this merit,
that it agrees with the relations of the vowels that are exhibited in
the phonetic and grammatical systems of many languages. Other
factors that enter into the distinction of vowel phonemes, arc the
tenseness and looseness of the tangue and other muscles, and
different positions of the lips, such as protrusion and retraction.
The Central-\Vestern type of American English distinguishes
nine vowel phonemes. One of these, [r], which we have aIready
discusscd, is peculiar in its inverted tangue-position. The other
eight form what we may calI a two-four system. As to position,
they occur in pairs; each pair consists of a front voweI, formed by
raising the middle of the tonguc toward the highest part of the
palate, and a back vowel, formcd by raising the back of the tangue
toward the velum. The four pairs differ as to nearness of the
tangue ta the palate; thus we have four degrees of raising; high,
higher mid, lower mid, and low. Instead of the terms high and
low, sorne writers use close and open. This gives us the following
scheme:
FRONT BACK
high i u
higher mid e a
lower mid e l
low a 0
Examples: in, inn [in], egg [cg], add [ed], alm8 [amz], put [put],
up [op], ought [;Jt], odd [ad]. These phonemes are subject ta a good
deal of nondistinctive variation, sorne of which depends upon the
surrounding phonemes and will intereflt us later.
Southern British Enp;lish has much the same system, but the
distribution of the back-vowel phonernes is different, in thl1t the
degrees of closurc of the vowels in words like up nnd odd are the
reverse of ours: higher mid in odd [ad], low in up [ap]. However,
thcre has arisen a convention of transcribinp; British English, Dot
by the symbols here indicated in accord with the principles of the
IPA alphabet, but by mcans of qucer symbols which are intended
104
TYPES OF PHONEMES
TYPES OF PHONEMES 105
to remind reader, irrevelantly enough, of the difference he...
tween Enghsh and French vowel phonemes:
CHICA.GO PRONUNCIA- BRITISH l'110NUNCIA- BRITISH PRONUNClA_
'l'ION ACCQRDING TO 'l'ION ACCORDING TO TlON, ACTUAL PRAe-
IPA PRINClPL>JS IPA PRINClPLE8 TIeZ
u
BACE:
a
INDtETER>JNT
INDIFFERENT
FRONT
FRONT
higb
low
The fewer the phonemes in a vowel-system, the more room is
there for non-distinctive variation of each phoneme. In Spanish
the mid vowels, for instance, vary, ta our ear, between higher and
lower positions, with much the same acoustic qualities as in
Italian, where these differences represent different phonemes. The
Russian vowels are subject to wide variation, which depends
chiefly on the preceding and following phonemes j especially one
variant of the high front vowel, as in [sin] 'son,' strikes our ear
very strangely, because in this variant the tongue is drawn back
much farther than in any variant of the English high front vowel.
The three-vowel system of Tagalog, finally, al10ws each phoneme
a range that seems enormous to our hearing; the variants of the
Tagalog phonemes symbolized above by the characters [il and
lu]' range ail the way from positions like those of our high vowels
to positions like those of our lower mid vowels.
6, 13. Different positions of the lips play no part in American
English vowels, except for one minor fact which we shaH take up
later. In many languages, however, lip-positions accentuate the
quality of different vowels: the front vowels are supported by
retraction of the lips (drawing back the corners of the mouth), and
the back vowels by protrusion or rounding of the lips. In general,
the higher the vowel, the more pronounced is the action of the
lips. These features appear in most European languages and con-
tribute to the difference between their and our vowels. Even
here we find decided differences; the Scandinavian languages,
high i u
mid e 0
low a
Spanish examples: si [si] 'yes,' pesca ['peska] 'fisbing,' tu [tuj
'thou,' pomo ['porno] 'apple,' ama ['ama] 'loves.'
Even simpler is the three-vowel system which appears in sorne
languages, such as Tagalog:
Sorne languages have simpler systems, such as the five--vowel
system of Spanish or Russian:
m
u
o
;)
in
in
inn
egg eg eg eg
add ed cd red
alms amz amz u:mz
put put put put
odd ad od od
ought :lt ot o:t

The ninth vowel phoneme, which we transcribe for Central-
Western American English by [r], as in bird [brd], has no uniform
correspondent in Southern British English or in New-England or
Am.erican English. Before vowels, British English has a
tnU, which we transcribe by [r], as in red [redJ; where
Ameriean has [r] after vowels, British has merely
.1 catIOn (m sorne cases, a lengthening) of the vowel, whieh
IS mdlcated by a colon [:], as in part [pa:t], form [fo:m]j where in
Central-Western American the [r] is neither preceded nor followed
by a vowel, British English uses a mixed vowel, intcrmediate he-
tWeen front and back positions, whieh is transeribed by [;rJ
[a], as in bird [ba:dJ or bitter ['bita]. ' or
.6: 12,. Sorne Central-Western types of American English lack the
dlStmctIOn of [a] and [a]. The low vowel of such speakers strikes
my ear as .an [a], bath in alms and in odd; in their phonemic system
however, lts position is neither "front" nOr "baek " b t' d'fI '
. . " u ln 1 er-
smce thlS has only one low-vowel phoneme. A
system, wtthout the eccentric [r] vowel, OCcurs also in
Itahan. We may caU this a seven-vowel system:
FRONT INDlFF>JRENT
high
higher mid e
lower mid e
low a
Italian examples are; si [si] 'yes' pesca ['peska] 'fi h' ,
[
'pe k J ' h ' [J' s mg, pesca
s a peac, tu tu 'thou,' pollo [' polloJ 'chicken ' olla [' olla]
'pot,' ama ['ama] 'loves.' '
especially Swedish, round their back vowcls more than do the
other European languages: Il. Swedish [0], as in bo [bo:) 'to dwelI,'
has about the tongue-position of a German or French [0], as in
German so [zo:J 'thus' or French beau [bol 'beautiful,' but it has
the extreme lip-rounding of a German or French high vowel ru], as
in German du [du:] 'thou' or French bout [bu] 'end'; it strikes Us
as a kind of interrrwdiatc sound between an [oJ and an [u],
The languages just named rnake use of lip-positions also for the
distinction of phoncmes. The comrnonest distinction of this kind
is that bctween the ordinary front vowels (with retracted lip-
position) and rounded front vowe/s, with the Hp-position of the
corresponding back vowels, Thus, French, besidc eight vowel
phonernes in a distribution likc that of American English, has
three rounded front vowels:
Examples:
fini [fini] 'done,' t [ete] 'summer,' lait fit] 'milk,' bat [bal
'beats,'
rue [l'y] 'street,' feu [f.0] 'fi.re,' peuple [pcepl] 'people,'
rOue [ru] 'whcel,' eau [0] 'water,' homme rom] 'man,' bas [bal
'low.'
To these arc added four nasalized vowels (see above, 6.4), as
distinct phonemes: pain [pen] 'bread,' bon [bo") 'gond,' un [ce"]
'one,' banc [bo"] 'bench.' Furthermore, French has Il. shorter
variety of [], which is transcribed [a], as in cheval [sJvalJ 'horse.'
The symbols [y, .0J are taken from the traditional orthography
of Danish; that of German (and of Finnish) uSCs the symbols
and 0,
One can learn to produee rounded front vowels by practising
Hp-positions before a mirror: after learning to produce front vow-
els of the types li, e, eJ with the corners of the mouth drawn baek
,
and back vowels of the types lu, O,J] with the lps protrudcd and
rounded, one speaks an fi] and then trics to keep the tongue-position
unehanged while rounding the lips as for an [uJ; the result is an
TYPES OF PHONEMES
FRONT
UNROUNDED ROUNDED
107
u:u
0:0
a:a
INDlFFERENT BACK
(ROUNDED)
y:y
P:0
j: i
e:c
FRONT
UNROUNDED ROUNDED
T YP Ji; S 0 F PH0 N E ES
high
mid
low
Germanexamples: , cl'
'h [,,] 'hl'm 'in [in] 'in' Reet [be:t] 'flowcr-bed,' Bett [bet] he ,
t n Ln" , .. ' ['k . 'k] 'k' ,
Tr [tYT] 'door,' hbsch [hyps] 'pretty, Komg p.lll mg,
zw6lf [tsvplf] 'twelve,' ,. ,
Fusz [fu:s] 'foot,' Flusz [flus] 'river,' hoch ,[ho:x
1
,h1gh, Loch
[loxJ 'hole,' kam [ka:m] 'came,' Kamm [kam] comb..
The differences bctween the vowel phonemes of dlfferent lan-
1 Dutch lacks the short liS].
[y], In the same way one passes from [eJ ta [pl and from [El
ta [J. d d b k
A fmther distinction is created by the use of unroun e ac
vowcls, in contrast with rounded. This additional factor produces
in Turkish a three-dimensional vowel system: each vowcl phoneme
is either front or back, high or low, rounded or unrounded:
}"RONT B"'CK
UNROUNDED R01JNDED UNROUNDED ROUNDED
1 Y li high
low e p a a
G 14 Another factor in vowel-production is the or 100S6
:r of the muscles: to our cars, vowels of the former
pOSl 10 . the Enghsh
sound clearer and perhaps excessively preClSe, sillee ,
vowels arc allloose, Some authors use the terms narrow a.nd. wtde
instcad of tense and [oose, The most striking to
our car, of the French vO\vels is their tense It lS rela-
tive tenscness, too, which in addition ta hp-achon, makes the
Italian vowels very differcnt from those of although the
two languages make t.he same mnnber of
Tenscness and looseness are utlized for dlstmctlO11s of
. G d Du'ch In German and, to a lesser extent, m
m erman an ". , . f
D t h the tense vowels are also of longer duratlOn (a actor
U'C , f 'd' t te
which will concern us later) than the Ioose. l wc lU lca e ,nse-
ness, combined with greater lcngth, by a colon after th.e symbo,l,
we obtain for these languages the following system, wlth a palr
of phonemes in each position 1:
B..l.CK
(HOUNDED)
U
o
:>
a
y
.0
ce
e
e
a
high
higher mid
lower mid
low
106
lOB
TYPES OF PHONEMES
guages are not sufficiently understood It lk 1
th t . IS I e y, moreover
a one and the same phoneme OIay often he produced . ' th'
same language, by very different actions of the vocal ,ID e
and for the native hearer identicaI,
e ects: It Is supposed that in such cases the deviation f
gan (say dif 0 one
. ,a 1 erent tongue-position) is compensated by differen
actIon of SOrne other organ (such as a different action of the
CHAPTER 7
MODIFICATIONS
7. 1. The typical actions of the vocal organs described in the
la,st chapter may be viewed as a kind of basis, which may be modi-
fied in various ways. Such moditications are: the length of time
through which a sound is continuedj the loudness with which it
is produced; the musical pitch of the voice during its production;
the position of organs not immediatcly concerned in the charac-
teristic action; the manncr of moving the vocal organs from one
characteristic position to another. This distinction between basic
speech-sounds and modifications is convenient for our exposition,
but it is not always recognized in the phonetic system of languages j
many languages place sorne of the latter features quite on a par
with phonemes of the former 80rt. We have seen, for instance,
that features of pitch are utilizcd as primary phonemes in Chincse,
and featurcs of duration distinguish primary phonemes in German.
On the other hand, most languages do recognize the distinction
to this extent, that they use sorne of the modifying features as
secondary phonemes - phonemes which are not part of the
simplest linguistc forms, but merely mark combinations or
ticular uses of such forms.
7. 2. Duration (or quantity) is the relative length of time through
which the vocal organs are kept in a position. Sorne languages
distinguish between two or more durations of speech-sounds. Thus,
we have seen ( 6.14) that in German the tense vowels are longer
than the loose; this differenee of length is more striking than that
of tenseness. The sign for a long phoneme is a colon after the
symbol for the sound, as German Beet [be:t) 'flowcr-bed,' in con-
trast with Bett [betJ 'hed.' If more degrees of length are to be in-
dieated, a single dot or other signs can he used. Another method
of indieating long quantity is to write the symbol twice; this is
done in Fnnish orthography, e.g. kaappi 'cupboard' with long
[a] and long [pl.
In Ameriean English, vowel-quantity is not distinctive. The low
and lower mid vowels, as in pan, palm, pod, pawn, are longer than
109
the other vowels, as in pt'n, pen, PU"l" pull, Ail Our vowels, more-
over, are longer before voiced sounds than bofore unvoiced; thus,
the [cl in pan, poo is longer than in pat, pack and the [il in pin,
bid longer than in pit, bit. These diffcrences arc, of course, not
distinctive, since they depend upon the height of the vowel and
Upon the following phonemes.
In dealing with Inattcrs of quantity, it is often convenient ta
set up an arbitrary unit of relative duration, the mara. Thus, if we
say that a short vowel lasts one mora, we may describc the long
vowels of the same language as lasting, say, one and one-half morae
or two morae.
In French, the distinction between long and short vowels works
in a peeuliar way. Long vowels Occur only bofore the last con-
sonant or consonant-group of a ward: the mere presence of a long
vowel in French thus indicates that the next consonant or COn-
8<mant-group ends a ward. In this position, moreover, the length
of a vowel is for the most part determined entircly by the nature
of the phonemes themselves. The nasalized vowcls [011, en, on, llJ
and the vowels fo, arc in this position always long: tante [tant]
aunt,' faute [fo;tJ 'fauIt,' The remaining vowels are always long if
the final consonant is fi, r, v, vr, z, 2J, as in caVe [ka:vJ 'cellar,'
vert [ve:rJ 'green.' Only in the cases not covercd by these two rules,
is the vowel-quantity ever distinctive, as in bte [be:tJ 'beast'
versus bette [betJ 'et.'
Long consonants OCCur in English in phrases and compound
words, sueh as pen-kmje ['pen InajfJ or eat lwo ['ijt 'tuwJ; within a
single word (nnJ occurs in a variant pronunciation of forms like
meanness ['mijnnesJ beside ['mijnesJ. A distinction of two con-
s<>nant-quantities within simple words is normal in Italian, as in
fatto ['fatwJ 'donc,' butfato ['fato] 'fate,' in Finnish, and in many
other languages. In Swedish and Norwegian a consonant is long
always and only after a stressed short vowel; the difference of
consonant-quantities, accordingly, is not distinctive. In Dutch
there are no long consonants; cven when Iike consonants meet in a
phrase, only one consonant Inora is spaken, sa that the phrase
consisting of dat [datj 'that' and lal (talJ 'number' is pronounced
{'da 'taI].
7.3. Stress - that is, intensity or loudness - consists in greater
amplitude of sound-waves, and is produced by mcans of more
energetic movements, such as pumping Inore breath, bringing the
111
110
MODIFICATIONS
MODIFICATIONS
.. and uing the muscles
vocal chords closer Fngli;h we have three
. 1 foraI arcu a lOns..
more vlgorous y or 'h' h . t of increascd stress, in contrast
d w lC conS1S 0
secon ary p " unstresscd passages of phonemes, ur
with what wc may call h. f s usually in contrast or
t ["J marks emp atlC orm, ,
ht'ghest s ress d' 'tress ['] appears normally
d' t' ht'gh slress or or mary Il ]
eontra lC JOn; our d' our low stress or secondary stress [1
on one syllatle of each wor.
U
: bics of compound words and long
pp
e'HS on one or more s}' ,1 d' 1 d by a
a , h h' h stress of certain wor s IS rep ace words, In phrases, t e tg
, d EpIes'
low stress or entirely omte. xam .' k' l' lJ
Th' , m,y parking-place ['tiis iz "maJ 'par Il) IP eJs ..
IS lS [' ". t" aJ 'bIt en it
It n't my fault and it is your fault lt lZ l;t m
"iz "juwr 'fllt,1 ,.
insert, verb [in'sft]; noun l msrt]
l'm going oui [aj m IgowlJ 'awt,]
Let' s {Jo back ['let s Igow 'bek.]
business-man [' bimes 1men]
gentleman ['jent\mT,I] ..
domt'nating ['dam1lncJtllJ]
domination lldami'negl}-] . lan ages and in
This svstem is paralleled in aIl the Germamc, 19u , Chi-
,) l' S anish the SlavLC anguagcs,
many otlwrs, such, as Ha e;like tllese the st.ress characterlzes
nese. In stress-llsmg langl g " l 'th use of one
f r . L forms' the tYPICl1 case IS e
combinations 0 lllgms 'h 'hrase with certain unstressed or
high stress on each wor e p sorne languages of this
low-stressed words as h :'s unanalyzable words)
, . 1 rngmshc orms SIlC a,
type contam Slmp 'hich may be diffcrentiated, accordingIy,
of more than one sy a c, wh R i ln [' gorotJ 'city' and [ma' ras1
by the place of the stress; t us uSS! , fix ffix' here
'f t' re bath simple words, containing no pre or su h' ,
ros:l h lace of siress has the value of a primary
accordmgly, tep d f loudness as non-dlsncve
Other languages use egrees 0 ds as t(l ups
'1""1 'n' language a sentence soun ,
features, In tue ,;V enoml l , but these ups
and downs of stress, quite like Enghsh hooemes and
and downs determined of stress
bear no relatIOn ta the meanlllg, d' 'ly th'" end of a phrase
k' d of gesture' or man v
serves only as a III .' . phatic speech some other
' 1 cl th n the rest sometlmes, III em "
IS ou er a 'f h one hears a long succession
syllable is cspecially loud; 0 ten enoug
of syllables with very little fluctuation of stress.
112 MO D IFICATIONS
MODIFICATIONS
113
7. 4. Among languages there arc sorne differences in
the manner of applying stress. In English therc is a non-distinctive
variation by which the vowcls of unstrcssed words and syllables
appear in a "weakened" form: they arc shortcr and formed with
looser muscles, the voice is sometimcs redueed to a murmur, and
the tend toward a uniform placing, somewhere
near higher mid position. The degree of wcakening varies from
utterance to utterance, and differs a great deal in different geo-
graphie and social types of standard English, Phoneticians often
use special symbols for the weakencd vowels, but this is unneees-
sary, sinee the diiIercnces are not distinctive, but depend merely
upon the unstressed position. The unstresscd vowel is a shorter,
looscr, less cxtremcly formcd variant of the stressed vowel. Com-
pare the full [el of test ['test] with the wcakened [el of contest
['kantest]; this wcakcned [el appears also, in American English, in
forms like glasses ['glsczJ, landed ['knded]; in ail thesc cases
British English seoms to use a somewhat higher vowel. Similarly,
we may compare the full [oJ in Been and unseen ["on'sijn] with its
variant in undo [on'duw); this weakened [0] appears also
10 forms like eautiouB ['kJsos], parrot ['perot].
In other cases the weakcned syllables actually show a loss of
phooemes, .or substitution of one vowel phoneme for another;
usually varIOUS grades of weakening exist side by side:
concert ['kansrt] concerted [kon'srted]
address, noun ['edres1 addl'ess, verb [e'dres]
relay ['rijlej] return [re'trn]
vacate vacation [vej'kejsl)., ve'kDjSI)-J
protest, noun [prowtest] protest, verb [prow'test, pro'test]
rebel, ooun ['rebIJ rebel, verb [re'bel]
atom ['Btrp] a/omie [e'tomik]
maintenance ['mejnt!).!).sJ maintain [rncjn'tejn, m!).'tejnJ
. In cases like these, various grades of weakening exist side by
sldc, are used according to the spced and the mood (formaI,
famlhar, and so on) of utterance. There are also local and social
differenccs. American English says dietionary ['dikSn,ejrij]
tary . ['sekre,tejrij] (compare secretarial [,sekre'tcjrm; British
Enghsh uses weaker forIOs, saying ['diksI)-ri, 'sekritriJ. On the
other hand, in forms likc Latin ['lct!).], Mar/in this degree
of weakcning is dccidedly sub-standard in England, where the
standard forms are ['Jetin, 'ma:tinJ.
Not aIl languages that use stress as a distinctive feature weaken
their unstressed vowels. The Germanie languages other than Eng-
iish produce the vowels of unstrcssed syllables quite lke those of
strcS
sed
syllables. The unstressed vowels in German MQnat
['mo:
nat
] 'month,' Kleinod ['klajno:t] 'gem,'. Armut
'poverty,' are quite like the stressed vowels m hat [hat] has,
Not [no:tj 'distress,' Mut [mu:t] 'courage.' In these
one vowel, the short le], appears in a weakencd vanant when lt
is unstrcssed. Thus, in German haUe ['hate] 'had' or
[ge'ba:det] 'bathed,' the [e]-vowel is. spoken sh.orter and
tangue less raised and fronted than III a form hke BeU [OOt] OOd,
and in a fonn like baden [' ba:den] 'to bathe,' the second syllable is
acoustieally quite like the second syllable of an English form like
sodden ['sadI)-], and very different from a German [den]
'then.' Phoneticians orten indicatc this wcakening by usmg the
character [al for the unstrcssed form of le], transcribing
['hata], baden ['ba:dan] or ['ba:d!).], but this is unnecDssary, Slllce
the accent-mark suffices to indicatc the weakening.
Other stress-using languages, such aS Italian, Spanish, Bohe-
mian Polish do not use special variants for aoy of the unstressed
vowels; for instance, our restitution [,resti'tuwsI).] with ,an
Italian res/ituzione [restitu'tsjone]. In a Bohemian word hke
kozel ['kozclJ 'goat,' the [el is just as full y formed as in zelenec
['zelenets] 'evergrcen.' .
7. 5. Another difference between stress-usmg languages eon-
cems the point at which the increase of loudness se,ts in. In
Engiish, if the first syllable of a ward has a stress, the of
loudness OOgins cxactly at the beginning of the word. Accordmgly,
there is a difference between parrs like the following:
a name [e 'nejm] an aim [en 'ejm]
/hat sod ["od 'sadI that' s odd ["od s'ad]
that stuff ['?id 'stof] that's tough ['tict s 'tof] .
The same habit 'prevails in German and Scandinavian; German,
in faet marks the onset of stress so vigorously that it often takes
the shape of a (non-distinctive) glottal stop bofore initial
vowel of a stressed word or clement, as in ein Arm [aJn 'arm]
'an arm,' or in Verein 'association,' where the ver- is an
unstressed prefix. .
In many stress-using languages, on the other hand, the pomt
of onset of a stress is regulated entirely by the character of the
primary phonemes. In Dutch, for instance, when there is a single
consonant the vowcl of a stressed syllable, this consonant
always shares III the loudness, regardless of word-division or other
factors of meaning: een aam 'an aarn' (measure of forty gallons)
and een naam 'a name' are bath [e'na:m], and a phrase like het
'the eye' is [e'tande'ro:x]. The same habit prevails
lU Itahan, Spamsh, and the Siavic languages.
7..6. Differences of pitch, that is, frequency of vibration in the
muslcal sound of the voice, are used in English, and perhaps in
most languages, as secondary nhonemes. The actual acoustic
forms are. highly variable; there is also sorne geographic variation.
The Enghshman's rising pitch in Thank you! is striking to Ameri-
can ears, and his rising pitch in sorne statements oflen makes them
to us like a yes-or-no question. Ivloreover, we use features
of pltch very Iargely in the manner of gestures, as when we talk
harshly,. sneeringly, vetulantly, caressingly, cheerfully, and so on.
In and in the languages of Europe generaIly, pitch is the
aco.ustIe feature where gcsture-like variations, non-distinctive but
effective, border most ciosely upon genuine linguistic
dlstmctlOns. The investigation of socially effective but nOD-
patterns speech, an investigation scarcely bcgun,
concerns ltself, accordmgly, to a large extent with pitch. For the
it is Dot easy to define the cases where features of
pIteh.have III our language a genuine status as seeondary phonemes.
It lS clear that the end of a sentence (a term wc shaH have to
define later) is always marked by sorne special distribution of
piteh. Wc can speak the words Ifs ten o'clock l have to go home
. l "
as a slllg e sentence, with a final-pitch only at the end, or as two
sentences, with a final-piteh on clock and another at the end
It's ten o'clock. l have to go home. After a final-pitch we ma;
pause for any length of time, or stop talking.
Within the domain of final-pitch we can distinguish several
differences. II' sten o'clock, as a staterncnt, differs from
ft sten 0 clock? as a question; the latter cnds with arise, nstead
of a fall. Among questions, there is a differencc of
betwcen a yes-or-no question, such as l t' sten 0' clock? or Did you
see the show? and a supplement-question, which is to be answered
by sorne special word or phrase, as What time is il? or Who saw the
with a lesser risc at the end. In transcription we may in-
dICatc the latter type by placing the question-mark upsidc down
[;,]. The distinction appears plainly in thc eontrast betwecn a
supplement-qucstion and a yes-or-no question which asks whcther
this supplement-question if! to bc answered: Who saw the show?
['huw (Se asks for the person, but ['huw (Se
mcans 'Is this what you were asking about?'
ThcBe three types of finalpitch appear side by side in the follow-
ing exarnple. If someone said l'm the man who - who - , his
intcrlocutor mip;ht help him out by sayinp;, with the final-pitch
of a statement, Who took the money [huw 'tuk 'e monij.]. This
contrasts with the supplement-question Who took the moneyl
['huw 'tuk 'Oc 'moniji,], to which an interlocutor who wantcd to
make sure that this was the question, or to use it aS a formaI
starting-point, might answer by a yes-or-no question, Who took
the maney? ['huw 'tuk oc 'monij?] (l'U tell YOll who took it. .. ).
It appears, furt.her, that sentences of an three of thesc types
may he distorted as to pit.ch, and also as to stress, when the speaker
is ref!ponding to a strong stimulus. We are doubtless justified
in scttinr; up a sinp;lc secondary phoneme of exclamatory pitch,
symbol [1], for this type, and in supposing that the varicties within
this type, such as the intonations of anger, surprise, calI, sneer,
and thc likc, arc non-distinctive, gesturc-like variations. The
cxclamatory phoncme appcars in conjunction with aIl thrce of
the final-ptch phonemes. Contrast John ['jan.] as an answer to
a question, with John! ['jan 1] as a call for the hcarer's (John's)
presence or attent.ion; similarly John? ['jan?] as a simple question
('ls that John? ') cont.rasts with the same question aeeompanied
by exclamatory pitch: John?! ['jan?!] ('lt isn't John, 1 hope! ');
finally, Who was watching the door [;,] contrasts wit.h thc exclama-
tory Who was watching the door in an emergency or a calamity.
As a fifth sceondary phoneme of pitch in English wc must ree-
ognize pause-pitch or suspension-pitch ['1, which consists of a risc
of pitch berore a pause wit.hin a sentence. It lS used, in eontrast
with the final-pitches, to show that thc sentence is not ending at
a point whcrc otherwise the phrasaI form would make the end of
a sentence possible: l was waiting there L] when in came the man.
John [,] the idiot [,J missed us. (Contrast: John the Bapiist was
preaching.) The man l,] who was carrying a bug L] came up to our
door. Only one man is in the story; contrast: The man who was
carrying a bug came up to our door, which implies that severaI men
are in the story.
114 MODIFICATIONS
MODIFICATIONS
us
7. 7. In EngIish both stress and pitch, then, are used only as
secon?ary phonemes, ?ut there are sorne differences hetween the
functlOns of the two. The stress phonemes step in only when two
more elements of speech are joined into one form: a simple word
hke John, contains no distinctive fcature of stress' to hear a
tinctive feature of stress we must take a phrase a compound
word or, at lcast, a word containing two or more parts such as con-
test. The pit.ch phonemes, on the other hand, occur every utter-
ance, appearmg even when a single word is uttered as in John'
J ohn? ohn.. the other hand, the pitch phonem'es in English
are not ln attached to any particular words or phrases,
but vary, wlth dlfferenccs of meaning, in otherwise identical forms.
languages differ from English in using secondary phonemes
o! pltch as we use those of stress, in \Vords and phrases that con-
SlSt of more than one element. In Swedish and Norwegian a
word of two syllables, for instancc, has an ordinary high
on. one of them, quite as it would in English, but, in addition to
thIS, the stressed syllables are distinguished by two different
pitch. The stress may he accompanied by a rising
pltch, grvmg much the same acoustic impression as an English
high stress, as in Norwegian ['b.0ner] 'peasants' or ['akscl] 'shoul-
der,' or, with a distinctive difference, it may he accompanied by
a falling pitch, as in ['b.0ner] 'beans' or ['akseI] 'axle.' This dis-
tinctive is aIl the more remarkable because in aIl other
Swedish and Norwegian ciosely resemble English in
their use of secondary phonemes of pitch and stress.
The language is said to distinguish two relative pitches,
normal and hIgher; thus, [hana] 'nose' has normal pitch on both
syllables, ['hana] 'bcginning' has higher pitch on the first syllable
and [ha'na] 'flower' on the second; there seem to he no
phonemes of word-stress.
In still other languages features of pitch are used as primary
phonemes. North Chinese distinguishes four of thesc which wc
may symbolize by numbers: '
[1] high leve!: [mal] 'mother'
p] high rising: [ma
2
J 'hemp'
low rising: [ma
3
J 'horse'
[4] low falling: [ma
4
J'scold.'
. Can.toncse iB said to have nine such tones. Primary phonemes of
pltch, III fact, appear in very many languages, eitiler in a few simple
types, as in Lithuanian, Scrbian, and ancient or in what
seerns ta us a bewildering variety, as in sorne Afncan languages.
It is worth noticing that we have in American English a non-
distinctive variation of pitch on our stresscd vowels: before an
unvoiced sound, as in map or mat, the lS simple, but
before a voiced sound, as in mad or man, we have ordinarily, and
under loud stress quite clearly, a rising-falling pitch.
7. 8. Once we have obtained sorne notion of how a phoneme is
forrned, wc may observe various modifications in the way it is
produced. The English phonemes [k, .g], for instance, .are made by
closure of the back of the tongue agamst the velum: If we observe
carefully, we fmd that the dosure is made forward
the next phoneme is a front vowd, as in kin [km], keen [klJn},
give [giv], gear [gijr], and farther backward bdore a back vowel,
as in cook [kuk], coop [kuwp], good [gud], goose [guws], in contrast
with what we may call the normal position, as in car [kar], cry
[kraj] , guard [gard], gray [grej]. The English phonerne [hl is formed
with the oral position of the fol1owing vowd. These variants are
not distinctive, since they depend entirely upon the following
phoneme. In languages where differences of this sort are
tive we have really no right to caH them "modifications," for in
the:e languages thcy are essential foutures of the phoneme. 'Ve
might just as weil use the term "modification" of the action or
inaction of the voice during the production of a noise-sound, or of
the presence or absence of nasalizat.ion, or of the rounding or
tion of the lips dnring the production of a vowel. Nevertheless, It
is convcnient to view in this way some less familiar features which
are phonemic in certain languages.
The IOost important of these is palatalatio
n
: during the produc-
tion of a consonant the tongue and lips takc up, so far as is com-
patible with the main features of the phoneme, the a
front vowel, such as li] or [eJ. Thus, we may say that III Enghsh
[k] and [g} are subject to a non-distinctive palatalizatio
n
before a
front vowel. Palatalization occurs as a distinctive feature notably
in sorne of the Siavie languages. In Russian, for instance, most
consonant phonernes occur in pairs, with the distinctive difference
of plain versUS pala/alized. For the transcription of the latt.er,
various devices have been used, such as a dot, curvc, or
over the symbol, or an exponent i or an accent-mark after. it, or
the use of italic letters. We shall adopt the last-namcd devlCe, aS
116 MODIFICATIONS
MO D IFICATIONS
117
118 MODIFICATIONS
MODIFICATIONS
119
the most convenient for printing. In a Russian word like [pal]
'five' the corners of the mouth are retracted and the tangue is
raised ioto front-vowel position during the formation of bath
consonants. In the case of the [t] this means, of course, that while
the tip and edge of the tangue are making closure against the
backs of the upper teeth, the blade of the tongue is raised toward
the palate; similarly in words like ['dada] 'uncle' or ['nana] 'nurse.'
The distinctive character of the difference appears in cases like
[bit] 'way of heing,' [bit] 'to be,' [bil] 'ta beat.'
Sorne languages distinguish velarized consonants, in which the
tongue is retracted as for a back vowel. If the lips are rounded
during the production of a consonant, it is said ta he labialized.
These two modifications appear together in labiovelarized con-
sonants.
7.9. The manner in which the vocal organs pass from inactivity
ta the formation of a phoneme, or from the formation of one
phoneme ta that of the next, or from the formation of a phoneme
ta inactivity, will often show varieties which we label as lransitions.
This tenu is fair enough when the differences are Dot distinctive,
but when they are distinctive, we have really no right ta describe
some of the essential features of the phonemes as basic and others
as transitional.
In passing from silence to a voiced stop, as in bay, doy, gay,
we begin the voicing gradually, and in passing from these sounds
ta silence, as in ebb, add, egg, we gradually lessen the voicing. This
contrasts with the French manner, where the stops in these posi-
tions are fully voiced, from the very beginning to the very end.
In passing from silence ta a stressed vowel, we usually make a
graduaI onsct of the voice, wile the North German first closes the
glottis and then suddenly begins full voicing, so as ta produce a
(non-distinctive) glottal stop. Occasionally, as a non-distinctive
variant, we start in the German style and the German in ours. In
French and in sub-standard southern English a third variety of
onset is non-distinctive, in which the glottis passes through the
[hl-position. In standard English and in German this variety is
distinctive, as in English hearl [hart] versus art [art]. In passing
from a vowel to sUenee, the languages so far named use a gentle
off-glide, but athees pass through the fhJ-position or end sharply
with a glottal stop, and in still others these differences are pho-
nemie. In passing from an unvoiced stop to a voiced sound,
espeeially a vowcl, one may begin the at th:
of explosion, or the voeing may lag for an lUstant; 10. elther case lt
may bgin gently or with a glottal stop; these. dlfferences are
phonemic in SOllle languages, and were dlscussed ln 6.6.
or after palatalized consonants there mllY he a glide a
front vowel; velarized consonants, similarly, may he accompamed
by a back-vowel glide. .'.
In successions of consonants the ehwf transltlOnal fcature
ta be the difference between close and open transition. ln Enghsh
7ie use close t.ransition. Vlhen we pass from one to
we form the second closure before opening the first: lU a word hkc
ador ['ektr], for instance, the tip of the tongue touches the gums
for the [t] bcfore the back of the t.ongue is the
to release t.he [k]. French uses open transltlOn: III a ward
acteur [aktce:r} 'actor,' the [k] is opcned bdore the tongue-tlp
touches the teeth for the [t]. Similarly, combinations of stop plus
spirant in English have close transition, as in Betsy, cupful, it shall:
bofore the stop is opened, the organs are already placed, as far as
possible, into the position of the follo:ving spirant, ,sa that the
explosion of the stop is incomplete. ThiS contrasts wlth the open
transition of French, where the stop is fully exploded before
spirant begins, as in celle scne 'th,is ;appe fuee
[etap fasil] 'easy stage,' cette chawe [set se:
z
] thm chair..
difference appears in so-callcd double consonants,. In
hich the same consonant phoneme appears tWlCe lU SUCCCSSlOn.
English, forms like flrab-bag ['greb ,bqJ;], hot time ['hot 'tajm],
pen-knife ['pen lnajf] show only one closure for the groups tt,
nn]; this closure merely lasts longer than the elosure of .a smgle
consonant. The double consonant is marked also by the dlfference
of stress uetwecn the implosion (in our examples, weuk) and the
explosion (in om examples, strong). In French, similar as
in cette table [sd tabl] 'this table,' normally show two opemngs,
an implosion and an explosion for euch of the two umts.
li both types of transition occur in a language, .the dlfference
may be utilized as a phonemie distinction. Thus, Pohsh has mostIy
open transition, like that of French, as in tny [tsi] 'thre?< but the
combination of ft] and [] occurs also with close as
separate phoneme, which we may dcsignate by [cl, as Ln cz
y
. tl]
'whethcr.' Thcre is also, again as li separate phonemc, a palatahzed
variety of this, [], as in ci [i] 'ta thee.'
120 MODIFICATIONS 1-1 OD 11'1 CATlNS 121
This last cxampic shows us compound phonemes - that is, sounds
resembling a succession of two or morc phoncrnes of thc sarrle
language, but in way distinguished from sueh a succession,
and utilized as separatc phonemes. -:\1any compound phonemes
consist, like those in our examplc, of a stop plus a spirant or other
open consonant; phcnemes of this sort are called affricates. In
English, where ail consonant groups have close transition, this
could not be used as a phonemic feature. Nevertheless, English
has two affricate phonemes, [cl as in church [6p':], and [J] as in
Judge [joj]o These affricates are always palatalizccl, and it is this
feature which distinguishes them from combinations of [tJ plus
[8], as in beel-sugar ['bijt Isugr], it shall lit '!!d], and of [dl plus [i],
as in did Jeanne [did 'zan].!
7. 10. The treatrnent of successions of vowels and predomi-
nantiy musical sounds shows great variety, and many types of
transition are distinctive in one or anothcr language.
In uny succession of sound8, sorne strike the car more forcibly
ihan others: diffcrences of sonority play a great part in the transi-
tion effects of vmvds and vowel-like sounds. Thus, other things
(especially, the stress) being equal, a low vowel, such as [a], is
more than a high vowel, such as li]; any vmvcI is more
sonorous than a consonant; a nasal, trill, or lateral more than a
stop or spirant; a 8ibilant [s, zJ, with its concentratioIl of the
stream into a narrow channel, more than another spirant; a spirant
more than a stop; a voiccd sound more than an unvoiced. In any
succession of phonemes there will thus be an up-and-down of
sonority. In a series like [tatatata], the [ars will be more sonorous
than the [trs. In the following example four degrecs of sonority
arc distinguished by means of numbers:
.Jack cu11ght a red bd
[jrk kt e red brd]
314 414 1 213 323.
Evidently sorne of the phonemes are more sonorous than the
phonernes (or the silence) which immediately precede or follow.
This is truc of the phoncmes marked 1 in our example and, in one
case, of a phoneme marked 2, namely the [r] in bird, but not of the
[1'] in red. Any sueh phoneme is a crest of sonority or a Sil liabic; the
other phonemes arc non-syllabic. Thus the le] in l'cd and the [l'] in
1 Phonetciam often sylUbolize th" English by l 'Uld long 3, and by d
and long z, tUll dosl' togdhcr Or conncctcd by a "mail can'c.
bird arc syllabics, but j,he [r] in red ,lnd the [dl in red and bd are
non-syllabics. An utterance is (0 have as many syllables (or
natural syllables) a.s it. has syllabics. The ups and downs of
tian play an important part in the phonclic structure of ail lan-
guages.
, every language, only certain ones of the phonerncs ever occur
as syllabies, but in prim:iple any sound may be more sonorous than
ils surrollndings. The interjccl ions pst! [psl !] and sM [i5!] with which
we demand silence, differ from ordimtT)' English words in using
[s] and [8] as sylhbi(:s. Actually, most of the phonemes in any
language are used only as non-syllabics, as, in English, [p, t, k];
\ve calI these consonants. Othn phonemes, fewcr in number, oceur
only as syllabics, us, in English, [e, 0, a]; we cal! these vowe/s. In
rnost languap;eR there is a third, intermediate group of wnantB,
phonemes which oeeur in bath syll:tbic and non-syllabic positions;
thus, in Ameriean English, of the Centl'fl,l- \Vestern type, [1' J is
syllabic in bird [brdl, but non-syllabic in l'cd [red].
\Vhether a sonant in any word i.s syllabic or non-syllabc, 1S
dcterrnined in differcnt ,vays in different languages. If the
labic or non-syllabic clmractel' of a sonant depends entirely upon
the surrounding phonernes (as in birrl versus rcd) , then the liffer-
cnce is not distinctive, and, sa hr as transcription is concerncd,
we do not more than one s}'mbol. In many cases, hmvcver,
the syllabic or non-syllabic charaeter of the som' Dt is determined
arbitrarily, und constitutes a phonemic difference. ThUR, in
Btirring ['striQ] the [rJ is syllabic, but in stn:na [strilJ] it is non-
syllabic; in the sc:c:ond syllable of patient ['petrn] the [l'] is 8yllabic
and the [nJ is non-syllabic, but in the second sylIablc of patron
['pejtrr,IJ the [r] is non-syllabic and the [n] is syHabie. In snch cases
we need separate symbols for the (.wo phonemes. Cnfortunately,
our habits of transcription in this regard arc neither uniform nor
consistent.. In a few cases wc \Ise difIerent symbols: [i, u, y] arc gen-
eraUy used for syllabic values, and [j, w, q], respectively, for the
corresponding non-syllabics; many transcribers, however, U8e the
former also for certa,in non-syllabic occurrences. Another
device is ta pl::wc alittle curve above or below symbols like [i, u, y,
C, 0, a] ta indicate non-syllabic function. On the othcr hand, the
symbols [l', l, m, nJ uswdly have Il dot, eircic, or verticalline placed
under them t denote syllabic function.
'Vhen the sylIabic or non-sylIabic function of a sonant is deter
122
MODIFI CATI ONS MODIFICATIONS 123
mined by the surroundinp; phonemes (or silence), the distribution
is natural. Thus, in standard German, the phonemes [i u] are
non-syllabic when they precede or follow a vowcl, and in other
}Xlsitions they are syllabic. Non-syllabic [u] occurs only aftel'
[a],. as i? 'bouse'; non-syllabic [il occurs aftel' [a],
as 10 Et [aJ] egg, aftel' [0] (or [0]), as in mu [noj, nojJ 'new,'
and vowels and ru], as in ia [ia;] 'yes,' iung [jUIJ] 'young.'
The vanants after a vowcl are dccidedly lowered, and the nOn-
syllabic ri] before syllabics is spoken with close contact, so as to
give a decided friction-sound, but these differences are not dis-
tinctive; traditionally, transcribers use the symbols li, u] for the
former type, but [jl for the latter.
"Where the syllabic or non-syllabic funetion of sonants is not
by the surrounding phonemes, the difference is pho-
nemIe. Sorne languages use a slight incrcase of stress to make
a sonant syllabic. In English thi3 8lfllabic-stress acts as a sccondary
phoneme. In Central-Western American Englsh, syllabic-stress
makes an [r] syllabic in strel1s<,d syllables in cases like stir-
ring in contrast with stn"ng [striIJ1, or erring ['riIJ] in con-
trast wlth nng [riIJ]. In unstressed syllables, the sonants [l', l, m, n]
are syllabic by natural distribution, as in buller ['botr],
boule [batll, bot/am ['batrrJ, bullon ['bob)], but in other cases
theu syllabic value is determined by the use of syllabic-stress.
Thus, syllabic-stress marks off an Ir] from a preceding [rJ, as in
e,rTor ['err] ['bejrrL or from a preceding [rI, as in stirrer
[strr], and lt often determines which of successive sonants is
syllabic:
apron ['ejprJ;1] paltern ['pdrnl
pickerel ['pikrlJ mins/rel ['minstrj]
coral ['karj] Carl [karl]
char 'em ['car :rp.J charrn [I":arm]
mainterumce ['mejntJ;1J;1s1 penance ['penJ;1sj.
The syllabic-stress lilay even make [l', l, m, n] syllabic before a
more open phoneme:
baUery ['betrij] pantry ['pmtrij]
hastily ['hejstlij] chas/dy ['ccjst.Jij]
anatomy [e'netrpij] met me ['met mij]
botany ['batl,j] chutney ['cotnij].
The syllabic-stress, then, has in this type of English the value
of a secondary phoneme. If we omitted the little vertical strokc
under syl1abic Ir, l, m, n], as wc properly should, in ail cases where
the syllabic value is due mercly to the character of the surrounding
phonemes, this st.roke would serve in the remaining cases as a
consistent fiign of syllabic-stress.
By the use of syllabic-strcss sorne languages reverse the rela-
tions of natural sonoritYi thUil, South German dialects have the
li, u, y] syllabic and the [a] non-syllabic in forms like [Hab] 'deat,'
[guat] 'good,' [gryan] 'green.'
Anoiher type of distribution is the use of ar/iculatory differences
to set off the syllabic and non-syllabic functions of the sonants.
Usually this consists in forming the non-syUabic variety with more
dosure than the syllabic variety. In English, the sommts [il and
[u] occur as non-syllabics before and after vowels; symbolizing
these non-syllabic occurrences by [j] and [w], we have [j] in yes
[jes], say [sej]' buy [baj], boy [b:lj] and [w] in well [wei], go [gow] ,
nOUl [naw]. In thesc exarnples the non-syllabic function of li, w]
is suffieiently deterrnined by natural sonorit.y, since a more open
vowel precedes or followil. Therefore the actual variations in t.he
manner of forming the sounds are here non-distinctive: the [j, w]
after vowcls, especially in the types [aj, :Jj, aw1are very open, and
the [a] also is quite different from an ordinary [a]; before a vowel,
as in weil, the [j] has a higher and more fronted tongue-
position than a syllabic [il, and the [wJ has a higher tongue-
position than a syllabic lu] and is formed with a slight contraction
of the lips. Now, these latter diiIerences are utilized, in English,
as phonemic differences: even where the function is not determined
by nat.ural sonority, we distiIlguish the closer nOll-syllabic [j, w]
as separate phonemes, from t.he more open syllabie li, u]. Thus,
we dist.inguish between [uw] in ooze [uwz] and [wu] in wood [wud],
and betwcen [ij] in ease [ijz] and a rare [ji], as in slang Vip [iip]
'to squeal,' and we have evcn groups like [jij, wuwJ, as in yeast
[j ij st] , woo [wuw]. When two different membcrs of t.he set,
li, Il, l'] come together in a st.ressed syllable, the first is
syllabic: you [juw], yearn lirn], win [win], work [wrk], rid [rid],
roof [ruf, ruwf]. In unstressed syllables wc have, however, dis-
tinctions like he [hajr] versus higher ['haiT], pair [pejr] versus
payer ['pejr], sore [sowr] versus sower l'sowr]. A non-syllabic
sonant which, thanks to some modification, is phonemically
distinct from the corresponding syUabic sonant is called a semi-
vOUlel.
124 MODIFICATIONS MODIFICATIONS 125
We shall sec in the next chapter that in the phonetic structure
of our speech-forms, these groups play the same part as simple
vowel phoncmes. The pcculiar non-distinctive modifications of
the componcnts, espccially of [a, j, w], which we notccd above,
often appear in diphthongs, but this is of secondary importance;
the essential feature is the peeular strudural trcatment. Another
non-distinctive pcculiarity of our diphthongs is their divergent
sound, in most American types of prollunciation, befare [r]: in
this position they approach the character of a single long and
rather tense vowcl;
In the same way, French produces iis high vowels li, H, y] with
greater closure and tensity when they are non-syllabic, as in
hier [jE;r] 'yesterday,' oie [wa] 'goose,' ail [a:j] 'garlie,' huiie
[qil] 'oil,' and these t:nws as separate sernivowd phonernes,
distinguishing, for instance, between omO [wi] 'yes' and houille
[u;j] 'anthracite,' and eillploying the sequence lij], as in fille
[fi;j] 'daughter.'
7. 11. Vowds and sonants combine iIlto compound phonemes,
which are known as diphthongs, or, if therc are three components,
as triphthongs. \Vhether a succession of phonemes is to be viewed
as a compound phoneme, depends entirdy upon the phonetc
structure of t.he language. In English, successions like [je] in yes
or [wc] in well are ireated as two phonernes, like a,ny sequence of
consonant plus vowe1, but combinations of vowc! plus 8emi-
vowc! are treated as compound phonemes \Ve have seven
8uch combinatons, as well as one triphthong of semivowel-vowel-
semivowel:
In sorne pronunciations these modified varicties differ from
any simple vowel, wtness:
::\Iany t,ypes of pronunciation, howevcr, lack some or aIl of these
differences; in these types eit her SOille of the cliphthongs or some of
the simple vowels do not (weur bdore [rJ.
Diphthongs occur also in languages that do not treat syllabic
and non-syllabic vowds as sl'pamte phonemes. In German the
combinations [ajJ as in His [ajs] 'iee,' [oi] as in ncu [noj] 'new,'
and [aw], as in Jfans [haws] 'house,' are !reatecl, structllrally, as
unit phoncmcs. As in Englsh, the constiiuents differ greatly from
their ordinary forrn: the non-syllabics have mid-vowcl quaHty
rather than high, and the [oj], especially, exists in several varieties,
rescmbling, in some pronunciations, rather a combination of
munder! front vowels, say [Oq].
Dipht.hongs like the English and German, ,vhere the syllabic
part precedes, are called falling diphthongs, in contrast. with rising
diphthongs, in which the non-syllabii: part precedes. Thus, in
French, combinaons lke [jE], as in fier [fjer] 'proud,' and [wa],
as in moi [mwaJ 'l,' are t.reakd st.ructurally as unit phonemes; in
Italian, the combinations [jF, w:J] arc trcaled as diphthongs; the
same is true of [je, weI in Spa,nish.
Some languages have compound phonemes of syllabic vowels and
non-syllabic consonants. In Lithuanian the phonemes [1, l', m, n]
are never syllabic, but combinations like [al, ar, am, an] are treated
structurally and accelltually as diphthongs, quitc on a par with
[aj] or [aw].
7. 12. Since syllabication is a matter of the relative Ioudness of
phonemes, it can be re-enforced or opposcd by adjustments of
stress. The re-enforcing habit prcvails probably in most languages.
In French, where stress is not distinctive, cvcry syllable is re-
enforced by a slight increase of stress on its syllabic; if there is
only one brfore the syllabic, the tise rn.'gins on this
non-syllabie; if there arc t.wo, different groups are treated differ
cntI)': pertinac [pfT-ti-na-si-te] 'pcrt.inacity,' patronnesse
;:>-nrs] 'patroness.' This distribution of minute rises and falls
of stress is non-di stinctive, since it is detcrmined entircly by the
worc [wowr], hoarse [howrs]
horse [hors]
war [,v;:>r].
Mary ['Illejrij]
merTJj [' lllcrij]
maNY ['mrrij]
seeing [',lijiJ)J
saying ['sejilJ]
buying [' bajiJ)]
boy'ish ['b::Jjis]
doing [' duwiJ)]
going ['gowlJ]
bowing ['bawiI)]
fewer ['fjuwl-'].
sure [suwrJ
oar [owrJ
hour [awr].
gear [gijr]
air [ejr]
fire [fajrJ
see [sij]
my [sej]
buy [bajJ
boy [b;:>j]
do [duw]
go [gow]
bow [baw]
few [fjuw]
126
MODIFICATIONS
character of the primary phonemes. It gives the language, to our
ears, rapid,. pattering or drumming sound. The same habit
prevaIls also ln many stress-using hnguages, sueh as Italian
Po:ish, Bohemian, and oYen in Hussian, whieh not onl;
has dlstmctwe stress, but also weakens the unslressed vowcls
Thus, in Italian pertinacia 'stubbornness' or
tronessa [pa-tro-'nes-sa] 'patroness,' the syllables arc dividcd by
ups and downs of stress, whieh arc well-marked in the acccnted
syllables, and slight in the others.
English and the other Germanie languages do Dot mark off the
unstressed syllables by ups and downs of stress. In a word lik
dimity or ['pcjtrones], the stress merely drop:
off aCter lts hlgh pomt on the tirst syllable. Evidently there arc
syllables, there arc t.hrcc crests of natural sonority,
but lt would be Impossible to say where one syllable encls and
the next begins. In forms like pertinacity [prti'nesitiiJ or pm.
cra.stination [pro1krrsti'nejsll-l , the heginnings of the stressed sylla-
bIcs arc piainly marked by the onset of stress, but no other syllable-
boundaries are in any way marked off.
distribution of st.ress may create crests of sonority which
are llldependent of the natural sonority of the phonemes. We
have seen that in English the phonemes [r, l, ID, n] may be louder
than the surrounding phonemes, and therefore syllabie thanks to a
slight increase of stress. '
The distribution of stress may even overcome relations of
natural sonority. In a combination like [dzd], the [z] is more
sonorous than the [d] 's, and in [kst] the [s] is more sonorous than the
stops, but in English our single high stress on forms like adzed
[edzd], text [tekst], slep [step] is sa loud that it drowns out these
smaIl differences of sonority. Sorne st.ress-using languages in this
way drown out even the sonority of predominantly musical sounds:
thus, Russian speaks the fol1owing, thanks ta stress as one-syllable
words: .'of the [rta] 'of the mouth'; Polish, similarly
trwa [trvaJ lt lasts, msza [msa] 'mass.'
CHAPTER 8
PHONETIC STRUCTURE
C. 1. Descriptions of speech-sounds like those in t.he last two
chapters, are due merely to chance observation. Thesc descrip-
tions are made in terms of a speakcr's movements: more refined
physiological observation may show that SOrne of them are wrong.
What is more serious, the differences and variehes t.hat are ob-
served, such as, say, the differcnce bet.wcen French and English
unvoiced stops [p, t, k], are not selected by any fixed principlcs
(sueh as acoustic phonetics may sorne day give us), but owe their
currency ta the chance that sorne observer with a good ear had
heard both of the languages eoncerncd. Just as observation of
South German dialects or of certain American Indian languages
adds ta the varieties of unvoiced stops that could he gathered from
standard English and standard French, so the stndy of aIrnost any
new dialect will inerease the repertoire of differences which fi.
phonetician can hear. The extent of observation is haphazard, its
accuracy doubtful, and the terms in which it is reported are vague.
Practicnl phoneties is a skill, for the student of languages often a
very useful skill, but. it has littlc scientific value.
For this teason it is beyond oq.r power to analy7,c the general
acoustic effect of a language. We can explain certain superficial
effects: the "pattering" mn of Halian (ta English cars) is due ta
the syllable-division; the "guttural" sound of DuLch (ta our
sense), ta the use of a uvular trill ( 6.7) and of velar tlpirants
( 6.8). In goneraI, however, such observations of the "basis of
articulation" are bound t be vague. English (in contrast, say,
with French or German) retracts the jaw; t.he Central and Western
type of American English adds a tendeney ta raise the tip of the
tangue. German and French (in contrast with English) advance the
jaw and use the muscles more vigorously - German in large,
sweeping movements, French in smaller and more precise ones,
especially in the front of the mouth. Danish draws the muscles in
toward the median line. Such observations are often helpful
toward understanding or imitating a pronunciation, but they are
127
129
lJ
Ci
o
w
u
a
C
8
i
r
Il
rn
"
El>Gr.1SH (ClllCA(;")
PHOKETIC RTllUCTCRE
nasals
lateral
inverted
aemivowcls
vowels, high
higher mid
lower mid
low
secondaI'}' phonemes:
stress
syllabic-stress
pitch Z, l? !
'Tables Iike thesc, even when they cxclude non-distinctive
tures, are nevcrthdcss irrclcvant to the structure of the language,
stops, ullvoiced P
t
k
voiced
b
cl
g
aifricate, unvoced

voiced
j
spirants, unvoiced
f 0 s S
h
voiced
v i :Il Z
. erent phonemes, while in Language A ft:] lu] mi?ht he non-
va.rianls of la], and [e, 0] of ll, Il]. not.h
might se?rtl. to show two
la. ht he phonNmc III Language III C?er
llldll
) , \\ Iule. m
Illig B thev milTht be non-distmeve vanants. Both lmght
Language '". h ." ", "
show plain a.nd aspIned as (:hiIerc.nt 1Il
Language A and as rn(:fe non-dl:llIletn:e vanants Ul
Both might have a series of vOl:cd .spirants, hut these ,llllgl.l.l hl,
distinctive in Language n, whlle III Language A they eXlsted
mercly as v!lriants of stops hel.wcen vowels. .
Only the phoneHles of a language are l'devant to Its
_ that is, to the work it does. A descript iOIl of tllP non-(!lstmct!ve
features mighl he of gret1t intcrcst, but for tNs il would have to be
more complele and more copious than any thnt have su far been
macle.
S. 2. A list or lable of the phonertles of a. language should
fore ignore ail non-distiner ive fcatUrE'S. Such or t.fI arc
usuall)' made on the basis of practical-phoncllc clai3S1fieatlOns,
thus:
STRUCTURE 128
hazy and inuecurate. We must wail for Jaboratory phonet'
. . d 1 - grve us preCIse an trustwort Iy statements.
. The important thing about language, however, s not the wa'
It sounds. The speaker's movement, the disturbance in the
and hearer's vibrations B of *2.2) are,a:,.
themselve.s, .of ver:.: lIt.tle The lmportant thing abon
IS Its servICe III connecttnp; the speaker's stimulus (A .
2.2) wIth the hrarer's r?sponse (C in 2.2). This cOllnectio
depends, as wc have scen ( 5.4), upon only a relatively few kat
of the acoustic form, upon t.he features which wc calI phonem '
For the working of language, ail that, is necessary is that ea . '
phone,me. be different from aIl the others. ExcepL
for thls dlffcrent.latlOn, ILS range of variet.y and its acoustie
tcr are irrelevant.. Any language ean be replaced, for ail its
values, by any system of sharply distinct signais, providcd that on"ii1i
signal s made to rep1a(:{) each phonenw of the language. Such
replacement is made in a correct phonetie tmnscription - omi::n-;
which satisfies the demands of accuraey and rekvaney by
one and only one symbol for each phonerne. Impcrfectly and ye :; '.
sufficiently weil for praetical purposes, such ft replacement is made """
in tmditional alphllbet.le writing. The imporbmce of u phoneme;<
then, lies not in the actuul configuration of its sound-waves, but.',
rncrely in the difference between this configuration and the con .
figurations of all the other phonemes of the 'lame language.
For this reason even a perfeetNj kIlowledge of acoustics will not,
by itself, give us the phonetic structure of a language. Wc shaIl
always have to know which of the gross ncousc features are, by

virtue of rneanings, "the saIne," and which "different" for the
speakers. The only guide to this is the speaker's situation and the
hearer's response. Any description whieh fails t.o discrminate the
distinctive features from the non-distinctive, ean tell us little or
nothing about the struct.ure of a In this respect, a ,
mechanical record has at ieast the vi l'tue of not distorting the
acoustie facts. The" exact" freehand records of zeftIous phonetc .,'
experts are likely to insist upon differences that
owe their notation merely to the circumstance that the observer
has leal'ned to respond ta ()n this basis, it is possible ta find , ..
the same set of "sounds" in languages of entirely differcnt pho-
nernie structure. For instance, both bnguages might show seven
similar vowel "sound;;," but in Language B these might. be sev
e1l
130 PHONETIC STRUCTURE
PHONETIC 8TRUCTURE 131
because thoy group the phonemes according to the linguist's notion
of their physiologie character, and not according ta the parts which
the sevcral phonemes play in the working of the language. Our
table does not show, for instance, that two of the nasals, [ml and
[nJ, sometimes serve as syllabics in unstressed syllabIcs as in
boUom ['batrp], but/on ['botl;J, while the third one, [lJ], does not.
It fails to show that [1] serves as a syllabic in unstressed syllables
only, as in boUle ['botU, while Ir] may serve as a syllabic regardless
of stress, as in learner ['lrnrJ. It does not show which vowels and
semivowels combine into compound phonemes. 1'0 show these
structural faets, wc should Dccd a supplementary table somcthing
like this:
1. Primary phonemes:
A. Consonants, always or sometimes non-syllabic:
1. Mutes, always nOD-syllabic: [p t k b dg cJ f () s sh v
ti Z ZIJl
2. Sonants, sometimes syllabic:
a. Semi-consonants, syllabicity determined by sur-
roundings and by syllabic-stress:
(1) Consonantoids, syllabic only in unstrcssed syl-
lables: [m n 1]
(2) V ocaloid, syllabic also in stressed syllables: [rJ
b. Semivowels, syllabicity detcrmined also by manner
of articulation; diphthong-forming:
(1) Non-syllabic: [j w]
(2) Syllabic: [i u]
B. Vowels, always syllabic:
1. Diphlhongs and triphthong, compound phonemes:
[ij uw ej ow aj aw oj juw]
2. Simple vowels: [e 0 e Cl a a]
II. Secondary phonemes:
A. Syllabic-stress, applied to semi-consonants: [11
B. Porm-stress, applied to meaningful forms: [" 1 IJ
C. Pitch, relating ta end of utterance:
L Mcdial: L]
2. Final: [. ~ ? 1]
8. 3. The parts which our phonemes play in the structure of our
language arc in reality much more diverse than this; in fact, we call
easil)' show that no two of thern play exactly the same part.
Since every uttcrance contains, by dcfinition, at Icast one syllabic
phonemc, the simplest way ta descrbe the phonetc structure of a
language iB to state which non-syllabic phoncmes or groups of
non-syllabie phoncmes (clusters) appear in the throe possible posi-
tians: initial, beforc the first syllabic of an uttcrance; final, after
the last syllabic of an utteranee; and medial, betwecn syllabics.
In this respect the diphthongs and triphthong play in English
the same part as do the simple vowels; it is precisc1y t.his fact that
compels us to class them as compound phonemes and not as mere
successions of phonernes.
For convenience, 1 shall place a number befoN each phoneme
or group of phonemes that shows any peculiarity in its structural
behavior.
Taking first the initial non-syllabics, wc find at the outset that
two phoncmes never begiD an utterance; they arc (1) [IJ, li]. We
ignore foreign forms, such as the French name Jeanne [Zan].
Further, six of the non-syllabics that. occur in initial position
never appear as rnembers of an initial cluster: (2) [v, ti, z, C, J, il.
The initial clusters ail begin with one of the fol1owing non-
syllabies: (3) [p, t, k, b, d, g, f, e, s, s, hl. Herc we find an accord
between the structural grouping and our physiologie description,
Binee our structural group (3) embraces exactly the physiologie
groups of stops and unvoiced spirants.
If the first consonant of the c1uster is (4) [s], it may be followed
by one of the set (5) [p, t, k, f, m, n], as in spin, stay, sky, sphere,
small, snail.
Ail the initiais of group (3) and the eombinations of (4) [s]
with (6) [p, t, k] may be followed by one of the set (7) [w, r, 1],
with the following restrictions;
(8) [w] never cornes after (9) [p, b, f, s], and never aftcr the
combination of (4) [13] with (10) ft]. The actual c1usters, then,
are illustrated by the words twin, quick, dwell, Gwynne, thwart,
Mm, when [hwen], squall.
(11) [r] never cornes after (12) [s, hl. The clusters, therefore,
are those whieh begin the words pray, tray, craw, bray, dray, gray,
fray, three, shrink, spray, stray, scratch.
(13) [1] never cornes after (14) [t, d, 0, s, hl, and never aftcr
the combination of (4) [s] with (15) [k]. The clusters, aecordingly,
are those which appear in play, clay, blue, glue, flew, slew, split.
8. 4. We come DOW ta the final c1usters. These are subject to
132
PHONwrrc STRUCTURE
PHONETIC STRUCTURE 133
The which OCCUf as post-finals arc (17) [t, d, s, z].
In a forrn like test or texl wc calI the Hl a main fimd, because there
f?rms lke tests, texls, in which a further consonant (a post-
final) IS added, but in a form like wished [wEt.] we call the Hl
a the cluHtcl' [-St] is not paralleled by any cluster
wlth the adclitlOn of a further consonant: we have no such final
clustor as, say, [-Ms].
?ccurrence of the post-finals is limited by threc important
restnctlOns. The post.-finals (l.S) [t, s] are the only ones thai occur
after the main finals (19) [p, t, k, C, f, , s, :'i]; these same post-
finals never oceur aHer any other sounds; and the post-finals
(20} [t, d] are the only ones that occur after the main finals (21)
[03, J, s, z, s, il. It is worth noticng that set (19) agrees, cxcept for
the absence of [hl, with the physiological cluss of unvoiced sounds,
.general. r.ulc that the same phoneuH; never OCCurs in two ad.
pOSltlons: there are no such final groups as [ss] or [tt].
ThiS :ul: hnlds good also for initial clustcrs and is irnplied by our
deScriptIOn of thcrn, but it docs not hold good, as we shall see
for medial clusters. '
'Ve have undertakeo to vicw combnations of vowel plus [j]
Or [w] as compound phonemes (diphthongs) and aCCOl'dingly can-
nat eouut the sernivowels in tbese combinations as final oon-
syllabics or parts of clusters. If, accordingly, we eiirninate these
cases (e.g. say [sejJ, go [gow]J, find thl11 (10) [h, j, w] do not
occur as final non-sylhbics or members of final clusters. Ali the
remaining non-syllabics OCeur in both of thesc functions.
English final c1usters consist of t\Vo, three, or four non-syllabics.
One can describe the combinations most simply b:r saying that
eaeh c1uster consists of a main final consonant, which may be
preCl'ded by a pre-final, which in tum may be pl'eceded by a
second :urthcr, .the main fmal may be followed by a
post-final. TIns grves us SIX possibilities:
main final alone:
pre-final plus main
final:
second pl'e-final plus
pre-final plus main
finaL
\Vl TJWU1' POS1'-I'INAL
bet Hl
lest [-st]
text [-kst]
\VI1'lI POS'l'-FINAL
bels [-ts]
tests [-stsJ
lexts [-ksts].
and that set (21) embraces the physiological c1aHses of affricatos
and sibilants. These restrictions group the main finals int.o six
classes:
Those in (19) but not in (21) may he followed by [t, s], as [p]
in help, helped, helps;
those in neHhor (19) nor (21) may be followed by [d, z], as [b]
in grab, grabbed, grabs;
those in (19) and (21) may be followed only by ft], as l] in
reach, reached;
those in (21) but not in (19) may he followed only by [d], as
Ul in urge, urged;
[t] in (19) but. not in (21), owin?; to the rule of no doubling, may
be followed only by [s], as in wa, wails;
[dl in ncither (19) nor (21), owing to the 'lame rule, may he
followed only by [z], as in fold, folds.
vVe turn now to the pro-fmals. The main consonant.s (22) [g, '('i, f;,
Il, r] are never accompanied by a pre-final, and the consonants (23)
[b, g, , J, v, fi] never occur us pre-finals. The combinations that
romain are subject to the following; further restrictions:
The pre-finals (24) [l, rJ do not occur before t.he main final (25)
[z1. Their combinations, accordingly, aTe those which appeaT in the
following examplcs: harp, barb, heart, hard, hark, march, barge,
scarf, caroe, hearlh, farce, har8h, arm, barn; help, bulb, beU, he/d,
milk, filch, bi/ge, pelf, delve, wealth, else, Welsh, elm, kiln.
The pre-final (25) [n] occurs only hefore the main finals (27)
[t, d, C, j, li, s, zJ, as in ant, mnd, pinch, range, monlh, ome, bronze.
The pre-final (28) [mJ occurs only before the main finals (29)
[p, t, f, li], as in camp, dreamt, nymph; the combination with
(30) [el occurs with the second pre-final (11) [1'1: warmth.
The pre-final (31) [I)J occurs only before (32) [k, OJ, as in link,
length.
The pre-final (4) [s] occurs only before (fi) [p, t, k], as in 1j)asp,
test, ask. Before (10) [t] it may he preceded by the second pre-
final (15) [k], as in text.
The pre-finals (33) ['5, z] occur only beforc the main final (28)
lm], as in rhythm, chasm.
The pre-final (10) ft] occurs only hcfore the main finals (34)
[6, s], as in eighth [ejt6], Ritz (compare, with post.-final [t] added,
the slang rilzed [rit.st] 'snubbed'). The combnation with the main
final (4) [sI occurs also with second prc-final (11) Ir] in quartz.
134
scarce, course; bcfore the other combinaHons of pre-final [r] the
only permitted diphthongs are [ow], as in cord, fork, torn, and, in
anly a few dialeetal-sounding words IejJ: laird, cairn. Before pre
final Il] the only perrnitted diphthongs are [ij, aj, ow], and the first
two oeeur only when [dl follows, as in field, mild, old, coli. Before
pre-final [n] only [aj, awJ oecur with any freedom, as in pint,
mount, bind, bound; [oj, ej] occur when [t] follows, as in paint,
point. The diphthongs do not oeeUT before [1)].
The triphthong differs from ordinary eombinations of
[j] plus vowel or diphthong (yank, year, Yale) in that if. oceurs
after initial consonants; it oceurs after [p, k, b, g, f, h, v, m, n]
as in pew, cue, beauty, gules, few, view, muse, new, and after the
c1usters [sp, sk], as in spew, skew; after [n] there is a less elegant
variant with [uw] instead of [juw], but, on the other hand, [juwJ
occurs in an pronunciation after [t, d, Il, s, l, st], where
[uw] is the commoner variant, as in tune, dew, thews, sue, lute,
stew.
We shaIl find that the grammatical structure of a language
implies groupings of the phonemes which supplement the groups
definable on the basis of succession ( 13.6).
8. 6. The structural pattern differs grcatly in different languages,
and leads us to reeognize different types of compound phonemes.
German, for instance, has, on the whole, a structural scheme
mueh like that of English, but with sorne striking differences.
The voieed stops and spirants lb, d, g, v, z] never oeeur in final
position. The initial groups can he simply described only if one
takes the affrieate combinations [pf, ts] as compound phonemes,
as in Pfund [pfunt] 'pound,' zehn [tse:n] 'ten,' zwei [tsvaj] 'two.'
The only diphthongs are [aj, aw, oj]; the simplicity of structure in
this respect, leads phoneticians to transcribc them rather by [ai,
au, oi], sinee no ambiguity ean arise. The French system dffers
not only as ta the partieular clusters, but also in more general
respects. The diphthongs are rising, such as [je, wa]. The greatest
differenee ie in the use of the vowcl phoneme [;01], whose occurrence
is governed largely by the phonetie pattern, so that it may be said
ta play the part of a sceondary rather than of a primary phoneme.
The phoneme [;01] oceurs wherever without it there would arise
an unpermitted c1uster of consonants. Thus, it oceurs in le chat
[1;01 sa] 'the eat,' because [lsJ is llot permitted as an initial eluster,
but not in l'homme [1 JmJ 'the man,' where no cIuster arises. It ap-
PHONETIC STRUCTURE
The prefinal (35) [dJ OCCurs only before (36) [Il, z], as in width
adze. ,
The pre-finals (37) [p, kJ occur only before the main finals (18)
[t, sJ, as in lapse, ad, tax. Of these two, the pre-final (5) [k]
before the mam final (4) [s] oecurs also with the second pre-final
(31) as minx (compare, with a post-final ft] added, the
slang [Jll)kstj 'gave bad luck '); the other, [p], oeeurs with
the second pre-final (28) [m]: glimpse, tempt.
The pre-final (38) [f] oecurs only before (10) It], as in lift.
The medial non-syllabies of English consist of aIl the combina.
tions of final plus initial, ranging from complete laek of a
as in saw it ['SJ it], to sueh clusters as in glimpsed
str-], including repetitions of the same phoneme, as
rn that [-t t-] or ten nights [-n n-J.
8. 6. A survey of the 38 functional sets of non-syllabics will
show that this classification mffices to detine evcry non-syllahc
phoneme in our language. In the same way, most or possibly ail
?f our syllabie phonemes could he defined by the parts they play
In the structure of our language. Since different types of stand-
ard English differ in the distributions of the syllabic phonemes, 1
shall mention only a few of the pattern features.
Of the semi-consonants, only Ir] occurs in stressed syllables;
if. never Occurs before Ir]. The syllabie semivowel lu] is distin-
by the fact that it does not OCcur initiaIly, and oceurs
med1aIly only before [t, k, d, s, li, 1], as in put, look, wood, plUS,
push, pull; it oceurs also before [f, ml, as in roof, room, but here
always beside a more elcgant variant with [uwJ. Neither [il nor
[u] OCcurs in final position.
Of the vowels, [1::, a] do not occur before semivowels (in diph-
thong combinations) and [J] does not occur before [w]. Only
[a, aJ OCCur in final position, as in saw, ma. The vowcl [aJ occurs
only before [JI, m, r], as in garage, calm, far, and before medial
[Ci], as in father. The phonemes [i, e, e, a] oceur before [r] only if
another vowel follows, as in spirit, herring, nwrry, sorry; [oJ oc-
?urs before [rJ only when the [rI is a pre-final, as in horn, horse, north;
III many types of pronuneiation the combination [or] is entirely
laeking. The vowel [0] oceurs before [r] only if [wJ precedes, as in
war, dwarf. The vowel [a] oceUTS before [g] only as a less Common
variant of [J], as in log, fog.
Of the diphthongs, only [ij, cj, ow] oecur before [rs], as in fierce,
PHONETIC STRUCTURE 135
136
PHONETIC STRUCTURE PHOKETIC STHUCTURE 137
pears in cheval [s<Jval] 'horse,' since the duster [sv] is not permitted
initially, but since this duster is permitted in medial position,
one says un cheval [
n
s'laI] 'a horse.' The medial cl usters are
limited, for t.he most part tD two consonants; thus, [rt.] is p e r ~
mitted as a final clust.er, as in porte [p0rt] 'carries,' but if an ini.
tial consonant follows, [a] is inserted, as in porte bien [pDrta bjenJ
'carries weIl.' An entirely different system appears in a language
like Plains Cree. The structure groups the phonemes into five
sets: (1) the vowels [a, a: , e: , i, i: , u, o;J; these arc the only syl-
labic phonemes; (2) consonants of four types: st.ops [p, t, k],
induding the afTricate r]; spirants [s, hl; nasals [m, n]; semivowels
ri, w]. The initial possibilities are: no consonant; any one con.
sonantj stop, spirant, or nasal plus semivowel. The medial pos-
si bilities are: any one consonant; stop, spirant, or nasal plus semi-
vowelj spirant plus stop; spirant plus stop plus semivowel. The
only final possibility is one consonant. The Fox language, with a
somewhat similar patterni ng, permits of no final eonsonant: every
utterance ends in a short vowel.
\Vhile English is especially rich in consonant clusters, it is easy
to find others, such as initial [pf-, pfi-, pfr-, ts-, tsv-, sV-, kn-, gn-]
in German, e.g. Pflaume ['pflawme] 'plum,' schwer [sve:r] 'heavy,'
Knie [kni:] 'knee,' or the clusters in Russian [tku] '1 weave,' [mnu]
'1 squeeze,' UOiiJ 'cabbage-soup,' [lsu] '1 flatter.' Final dusters
foreign to English appear, for example, in German Herbst [herpst]
, autumn' and Russian [bors] 'beet-sup.'
8. 7. Once we have defined the phonemes as the smallest units
which make a diffcrence n meaning, wc can usually define each
individual phoneme according to t.he part it plays in the struc-
turai pattern of the speech-forms. We observe, especially, that
the structural pattern leads us to recognize alRo compollnd p h ~
nemes, whi ch reseroble successions of ot.her phoncmes, but play
the part of a simple phoneme, and that very slight acoustic d i f ~
ferenees, such as, in English, the syllabic-stress on [r, l, m, n],
or the greater tensity of [j, w] compared to syllabic li, u], may give
fise to separate phonemes.
The phonemes so defincd arc the unit.s of signaling; the mean-
ingful forms of a language can be described aR arrangements of
primary and secondary phonemes. If we takc a large body of
speech, we can cOllnt out the relative frequencies of phonemes
and of combinations of phonemes. This task has been neglected
by linguists and very imperfectly performed by amateurs, who
confuse phonemes with printed leUers. Taking the total number
of phonernes in the text used as 100 per cent., a recent couot for
English shows the following percentage frequencics for consonant
phonemes:
n 7.24 'l5 3.43 p 2.04 g .74
t 7.13 z 2.97 f 1.84 j .60
r 6.88 m 2.78 b 1.81 .52
s 4.55 k 2.71 h 1.81 j .44
d 4.31 v 2.28 1] .96 0 .37
l 3.74 w 2.08 H .82 Z .05.
The figures for [r, l, m, n] indude the occurrences in syllabic
function; those for [j] and [w] do not include the occurrences of
these phonemes as parts of dipht.hongs or triphthong. The count
of vowel phoncrncs i8 tao confused to allow of plain reading. Ap-
parently, [cl is the most-usrd, with a frcquency of over 8 per cent;
next cornes [ij], with over 6 per cent.; then Ir], with 3.5 per cent.
The figures for groups of phonemes ure unusable. From this and
similar counts it is cvident that. the phonemes of a language per-
form very difTerent. rles as t.o frequency. IVloreover, thcre seems
to be sorne resemblunce bct.wcen languagcR; thus, in languages
which use two types of st.ops, sueh as our [p, t, k] versus [b, d, g],
the stop of the unvoiced type in each pair is more frequent than
its voiced mate, - for inst.ance, [t] more frequent than [dl. A
seriaus study of t.his matter is much to be desired.
8. 8. We have seen three ways of studying the soumIs of speech.
Phonetics in the strict sense - that is, laboratory phonetics-
gives us a purcJy acoustic or physiological description. It reveals
only t.he gross acoustic fent.ures. In practice, the laboratory
phonet.ician usnally singles ont for study some feature which his
lay knowledgc recognizes as eharacteristic of a phoneme. Practical
phonetics is an art or skill, not a science; the practical phonetician
frankly accepts his everyday recognition of phonernic unit.s and
trics to tell how the speaker produces t.hem. The terrn phonology
is sometimes placed in contmst with the 1>','0 forms of phonetics:
phonolo!-,'Y pays no heed to the acoustic nature of the phonemes,
but mercly accept.s thcm as distinct uoHs. It defines each phoneme
by Hs rle in the structure of speech-forms. It i8 important t
remember that. practicul phondies and phonology presuppose a
138
PHONETIC STRUCTURE
knowle?ge of meanings: without this knowledge we could oot
ascertam the phooemic features.
The description of a language, then, begins with phonology, which
phoneme and states what combinations Occur. Any
of phonemes that occurs in a language, is pronounce-
able ID thIS. and is a phonetic Jorm. The combination
[mnu], for mstance lS unpronounceable in Engll'sh but th
b
. [ . , e com-
matIon men) IS pronounceable and is a phonetic form.
the phonology o.f a language has been establishcd, there
romams the of tdhng what meanings are attachod ta the
phonetIC forms. This phase of the description is semantic
It lS ordinarily divided into two parts, grammar and lexicon. 8.
A phonetic form which has a meaning is a linguislz'c .f
orm
Th
E l'sb 'J' us,
ng 1. sentence, phrase, Or word is a linguistic form, and sa
a syll.able, such as, say, [mEl] in maltreat, or [mon]
ID M onday, a meanmgful form may even consist of a single pho-
such as the [s] which means 'more than one' in
Iike caps, books. In the following chapters we shan see how
meanmgs are connected with linguistic forms.
CHAPTER 9
MEANING
9. 1. The study of speech-sounds without regard ta meanings is
an abstraction: in actual use, specch-sounds arc uttered as signaIs.
We have defined the meaning of a linguistic form as the situation
in which the speaker' utters it and the response which it caIls forth
in the hearcr. The speaker's situation and the hearer's response
are closely co-ordinated, thanks t<> the circumstance that every one
of us learns ta act indifferently as a speaker or as a hearer. In the
causal sequence
spcaker's situation 1& ) speech III ;.. hearcr's rcsponse,
the speaker's situation, as the earlier term, will usually present a
simpler aspect than the hearer's response; thereforc we usually
discuss and define meanings in terms of a speaker's stimulus.
The situations which prompt people ta utter speech, include
evcry abject and happening in their universe. In arder to give a
scicntifically accurate definition of meaninp; for cvery form of a
language, we should have to have a scientifically accurate knowl-
edge of everything in the speakers' world. The actual extent of
human knowledge is very small, compared to this. We cao define
the meaning of a speech-form accurately when this mcaniog has to
do with some matter of which wc posscss scientific knowledge.
We can define the names of mineraIs, for examplc, in terms of
chemistry and mineralogy, as when wc say that the ordinary
meaning of the English ward BaU is 'sodium chloride (NaCI),' and
we can define the narnes of iJlants or animaIs by means of the
technical terms of botany or zoology, but we have no precise way
of defining words like love or hate, which concern situations that
have not been accuratcly classified - and these latter are in the
great majority.
Moreover, eVen where we have sorne scientific (that is, u'Qi-
versally rccognizcd and accurate) classification, wc often find
that the meanings of a language do Dot agree with this
tion. The whale il) in German called a 'fish': Walfisch [' vahfisl
139
140
:.\IEANIKG
MEANING 141
and the bat a 'mouse': Fledermau8 ['fle:der-,maws]. Physicist,s
view the color-spectrum as a continuous scale of light-waves of
different lengths, ranging from 40 to 72 hundred-thousandths of 11
mllimetre, but languages mark off difforent parts of this seale quite
arhitrarily and without precise lirnit.s, in the meanings of sueh
color-names as violet, blue, green, yellow, orange, red, and the color-
narnes of different languages do not embrace the same gradations.
The kinship of persans seems a simple matter, but the terminol-
ogies of kinship that are usod in various languages are extrcmely
bard ta analyze.
The statement of meanings is tberefore the weak point in
language-study, and will remain so until human knowlodge ad-
vanees very far beyond its present state. In praet.ice, we the
meaning of a linguist.ic form, wherever wc can, in terms of sorne
ot.her science. ",Vhere this is impossible, we resort ta makeshift
devices. One is If someone did not know the mean-
ing of the \vorcl apple, we could instruct hill) by handing him an 1'-
apple or painting at an apple, and continuing, as long as he made
mistakes, ta handle apples and point. at them, until he used the
wO;d in t?e conventional way. This is essentially the process by
WhlCh chlldren learn the use of speech-forms. If a questioner
understood enough of our language, we could define the wonl
apple for hirn by circumloclIlion - that is, in the manner of our
dictiomuies, Ily 11 l'Oundabout 'lpeech which fiUed the same situa-
tions as does the ward apple, saying, for instance: "The well-known,
firm-fleshed, smooth-skinned, round or oblong pome fruit of the
trees of the genus 1\.-1alu8, varying greatly in size, shape, color, and
degree of acidity." Or else, if we knew enough of the questioner's
we could answer him by translalJ:on - fhat is, by utter-
mg a roug-hly equivalent form of his language; if he were a French-
man, for inst.ance, we could give pomme [pJm] as the meaning of
apple. This method of defil1tion appears in our bilingual dietion-
arics.
9. 2. The situations which prompt us ta utter any one linguistic
form, are quite varied; philosophers tell us, in faet, that no two
situations are ever alike. Each one of us uSCs the ward apple, in
the course of a few months, of many individual pieces of fruit.
which diffcr in size, shape, COlOT, odor, tast.e, and sa On. In a
favorable case, such as that of the ward apple, aIl the members of
the spccch-community have been trained, from childhood, ta use
the speech-form whenever the situation (in this casc, the abject.)
presents certain relatively definable characteristics. Even in
cases like this, our usag-e is never quit.e uniform, and most speech-
+- forms have less dear-cut meanings. Nevertheless, it. is clear that
we must discriminate bet.ween non-distinctive features of the situa-
tion, sueh as the size, shape, color, and so on of any one particular
apple, and the distinctive, or linguistic meaning (the semantic
features) which are cornIllon ta ail the situations that call forth
the utterance of the linguistic fonn, such as the features which are
common ta ail the objects of whieh English-speaking people use
the ward apple.
___ Since our study ordinarily concerns only the distinctive fentures
of form and meaning, 1 shall hcneeforth usuaIly omit the qualifica-
tion lt'nguist'ic or and spcak simply of forms and meaJ'l.-'
rgs, ignoring the exist.ence of non-distinctive feat,ures. A form is
often said ta express its rneaning.
9. 3. Even if we had an accurate definiLion of the meaning that
is attached ta every one of the forms of a Iang-uage, we should still
face a difficulty of another sort. A very important part of every
situation is the state of the speaker's body. This includes, of
course, the predisposition of his nervous system, whieh results
from ail of his experiences, linguistic and other, up to this very
moment - not to speak of hereditary and pre-nat.al factors. If we
could keep an external 'lituation ideally uniform, and put different
speakers into H, we should still be unable ta measure the equipment
each speaker brought with him, and unable, therefore, ta predict
what speeeh-forms he would utter, or, for t.hat matter, whether he
would utter any speech at aIl.
If we had perfect definitions, we should still discover that during
X-many utterances the speaker was not at aU in the situation which
we had defined. People very often li Uer a ward like apple when
no apple at aIl is present. We may caU this displaced sPSlch. The
frequency and importance of displaced speech is obvious. We re-
caU the infant" asking for" his doll ( 2.5). Relayed speech em-
bodies a very important use of language: speaker A sees sorne
apples and mentions them to speaker B, who has not seen them;
speaker B relays this news t.o C, C ta D, D to E, and sa on, and it
may be that none of these persans has seen thern, when finally
speaker X goes and Cats sorne. In other ways, too, we utter linguis-
tic forms when the typical stimulus is absent. A starving beggar
142
MEANING
1\1 EANING 143
at the door says Tm hungry, and the housewifc gives him food:
this incident, we say, embodies the primary or diclionary meaning
of the speech-form Tm hungry. A petulant child, at bed-time,
says Tm hungry, and his mother, who is up to his tricks, answers
by packing him off tD bcd. This is an example of displaced speech.
It is a remarkable fact that if a foreign observer askcd for the
meaning of the form Tm hungry, bath mother and child would still
,
in most define it for him in terms of the dictionary mean-
ing:i' Lyinj;!;, irny, jesting, poetry, narrative fiction, and the like,
are probably as old and certainly as widespread a;; language. As
Son as we know the dictionary meaning of a form, we are fully
able to use it in displaced speech; OUr dictionaries and handbooks
of foreign languages nced tell us only the dictionary meaning. The
displaccd uses of speech are derived in fairly uniform ways from its
primary value, and require no special discussion; nevertheless, they
add to Our uncertainty as to the forms that a given speaker will
utter (if he speaks at aIl) in a given situation.
l 9.4.. Adhere.nts of mcntals.tic psychology belicve that they
can aVOId the dlffieulty of defimng meanings, because they helieve
that, pror ta the utterance of a linguistic form, there occurs within
the speaker a non-physical process, a thought, concept, image, feelr
/{/, act of will, or the like, and that the hearer, likewise, upon Te-
ceiving the sound-waves, goes through an equivalent or correlated
mental process. The mentalist, therefore, can define the meaning
of a Iinguistic form as the characteristic mental event which
occurs in every speaker and hearer in connection with the utterance
or hearing of the linguistic fonn. The speaker who utters the ward
apple has had a mental image of an apple, and this word evokes a
similar image in a hearer's mind. For thc mentalist, language is
t0..
he
expre.%ion of ideas, feelings, or volitions.
The mechanist does not accept this solution. Ho helieves that
mental images, feelings, and the like are merely popular terms for
various bodily movements, which, so far as they eoncern language,
can be roughly divided into three types:
(1) large-scale proeesses which are much the same in different
people, and, having SOrne social importance, are represented by
conventional speech-forms, such as l'm hungry (angry, frightened,
sorry, glad; my head aches, and sa on);
(2) obscure and highly variable smaU-scale rnuscular contrac-
tions and glandular secretions, which dmer from persan to persan,
and, having no immediat.e social importance, are not represent.ed
by convent ional speech-forms;
(3) soundless movcments of the vocal organs, taking the place
of specch-movements, but not perceptible tD other people (" think-
ing in words," 2.4).
The mechanist views the processes in (1) simply as events which
the speaker can observe bettcr than anyone cIse; the various
problems of meaning, such as that of displaced speech (the naughty
child saying l'm hungry), exist hore no less than elsewherc. The
mechanist. believes that the proeessef\ in (2) are private habits left
over, as traces, from the vieisf\itudes of education and other ex-
perience; the speaker reports them as images, feelings, and sa on,
and they differ not only for every speaker, but for every occasion of
speech. The speaker who says, "1 nad the mental image of an
apple," is reully saying, "1 was rcsponding to SOille obscure internai
stimuli of a type which was a;;;;ociated at some time in my past
with the stimuli of an apple." The sub-vocal speeeh in (3) scems
ta the mechanist mercly a derivative of the habit of actual speech-
utterance; when we are assured that a speaker has inaudibly
performed the f\peech-movements of a certain utterance (" thought
it in words "), we face exacUy the same problem as when he has
audibly uttered the same speech-form. In sum, then, the" mental
processes " seem to the rnechanist to he merciy t.raditional names
for bodily processes which either (1) come within the definition of
meaning as speaker's f\ituation, or (2) are so distantly correlated
with speech-uttcrance as tD he negligible factors in the speaker's
situation, or (3) are mere reproductions of the speech-utterance.
Although this difference of opinion plays a decisive part in our
views the fundamentals of language, as of other human
activities, and although mentalists lean heavily upal! thcir termi-
nology in ail discussion of meaning, the dispute has rcally very liUle
to do wit.h problems of lingustic meaning. The cvents which the
mentalist designatps as mental processes and the mechanist clas-
sifies otherwise, affect in cvery case only one person: every one of
us responds to them when they occor within him, but has no way
of responing ta them when they occur in anyone else. The
mental processes or internaI bodily processes of other people are
known ta each one of us only from speech-utterances and other
observable actions. Since thcse arc ail we have to work with, the
mcntalist in practice defines mcanings exactly as docs the mecha-
nist, in t.erms of actual situations,' he defines apple not. "'h .
f 1
. ,IS ,e nu-
age 0 t le well-known, firm-fieshed etc frUI't " 1llt ]'k tl
. ' '.... ,},' 1 e IC
mechall1st, ollats t.he first three of thesc words, and, in fact, for
ail excppt himself, merely infers tha1. the imagc was pres--
ent, eher from the thaL the speakcr used t.he word apple, or
sorne more defimt.c utterance of thc speaker's (" 1had a mental
of an apple"). In practice, then, al1linguists, bot.h ment.aJ-
mechanis1.s, dcfine rneanings in terms of the speaker's
s1t.uatlOn and, whenever this scems t.o add anything, of the hearer's
response.
9. 5. meanings arc more specifie t.han the meanings
of ads. A great. deal of hurnan co-operat.ion is
cff:ct.?d wlthout by Ruch roeans as gestures (for instance,
at the. handling of abjects (placing an abject
mto s?meone s dashmg object to the ground), contact
caressmg), non-linguisric soumis, both non-vocal (snap-
pmgthe fingers,. applause) vocal (laughing, crying), and so on.
':e mention especial1y, in this last. connection, the non-
(non-distinctive) features of speech-sound, such as
plallltlve, angry, .commanding, drawling "tones of voice"; t.he
manner of speech, 10 fact, is, next to sppech itspif, our most effective
mct.hod of. signaling. Linguistic forrns, howevcr, result, for the
most part, 10 far more aecurate, specific, and delicat.e co-ordination
A than conld be reaehed by non-linguistic means; to see this, one
only listen to 11 chance speeches: FOla feet Ihree and a hall
tnch:s. -; If you dord hear frorn rne by eight o'clock, go without me.
- fi here s the small boUle of ammonia? Apparent exceptions, such
,elaborate systems deaf-and-dumb language, signal-
:ng-cod.es, t.he use of wnlmg-, telegraphy, and so on, turn out., upon
to be merely derivatives of language.
we no way of defining most rneanings and of demon-
strang thelr constancy, we havc to take the specifie and stable
as a presupposition of linguistic study, just
as lt m our everyday dealings with people. We
thlS presupposition as the fundamenlal assumplion of
lmgUlslws ( 5.3), namely:
In certain: communilies (speech-eommunities) some
anees are altke as 10 farm and meaning.
This is bought at the cast of ra1.ionality.
The non-Imgmsc modes of communication are based directly
upon our bodily make-up, or else arise directly simp.le social
situations, but the connection of linguisti
e
forms wlt.h theu
ings is wholly arbitrary. What we cali horse, the German
Fferd [pfe:rtL the Frenchman cheval [s;;val] , the Cree IudJ.'J.n
[misatim], and SO on; one set of sounds is as unreasonable as any
other.
ur fundamental assumption implies that each linguistic fonn
has a collstant and specifie meaning. If the forms are phonemically
different, we suppose that their meaning;s also are differe.nt - f?r
instancc, t hat eadi one of a set of forllls like quick, fasl, 8Wijt, rapld,
speedy, differs from ail the others in some cOIJ::itant and conven-
tional feature of meaning. We suppose, in short, that there arc no
actual 8ynonyms. On the other hand, our tlSSumption implics also
that if the forIlls arc semantically diffcrent (that is, different as ta
lingustic meaning), they are not. "the same,' even though lhe.y
may he alikc as to phonetic fonn. Thus, in English, the phonec
form [bejr] occurs with three different meanings: bear 'ta carry;
ta give bil'th to,' bear 'ursus,' and are 'uncovered.' Similarly,
[pejr] rcpre::ients two nouns (pear and pair) und a verb .an.d
many other examples vl'ill occur to the reader. Different hngUlstlC
forms which have the same phonetic form (and differ, therefore,
only as t.o lIlcaning) are known us homonyms. Since we cannot with
ccrt.ainty dcfine meanings, wc cannot always decide wbethcr Il
given phonetic form in its various uses has always the same
ing or represents a set of hOIllonyms. For instance, Enghsh
verb bear in bear a burden, bear troubles, ear fruit, bear ojJsJlTlng,
can be viewed as a single form or as tt set of two or perhaps even
more homonyms. Similarly, charge, in charge the cannon with
grapeshot, ehrge the man with larceny, chargc the gloves to me,
him a slijJ priee, can he viewed iu several \vnys; the infantry 'Will
charge Ihe fort SCCIllS to he different. The quality doth and the
_animal oth probably represent a pair of homonyms to sorne speak-
ers and a single meaning to others. Ali this shows, of course, th.at
our basic assumption is true only wit.hin limits, evcn though lts
ge}leral truth is presupposed not only in linguistic study, but by
ail our actual use of language.
G. Although the lnguist CtlllnoL define meanings, but must
appeal for this to studcnts of other sciences or tD common knowl-
edb'C, yet" in many cases, having obt.ained for sorne
forlUs, he can define the meanings of other forms m terms of
144 l\JEAKI:.rC
MEANli'irG
145
146 MEANING
MEANING
147
these first ones. The mathematician, for instance, who is here
acting as a linguist, cannat define such terms as one and add
. '
but if we give hirn a definition of these, he can define two (' one
added to one'), three ('one.addcd to two'), and sa on, without end.
What we see plainly in mathematicallanguage, where the denota-
tians are very precise, appears also in many ordinary speech-fonns.
If the meanings of the English past tense and of the word go
defined, the linguist can define went as 'the past of [fa.' If the
differencc male: female' is defincd for the linguist, he can assure
us that this is the diffcrence betwcen he : she, lion: lioness, gan-
der: goose, mm : ewe. The linguist has this assurance in very many
cases, where a language, by sorne recognizable phonetie or gram-
matical feature, groups a number of its forms into
in any one form-class, every form contains an element, the class-
meani'fl{J, which is the same for aIl fOrIns of this form-class. Thus,
aIl English substantives belong ta a form-class, and each Englisb
substantive, accordingly, has a meaning, which, once it is defined
for us (say, as 'object 'j, we can attribute ta every substantive
form in the language. English substantives, further, are subdivided
into the two classes of singular and plural; granted a definition of
the meanings of these two classes, we attribute one of these mean-
ings ta every substantive.
In every language we find certain forms, substitutes, whose
meaning consists largely or entirely of class-meanings. In English,
the pronouns are the largest group of substitutes. The pronouns
show us a very interesting combination of rneanings. The principal
features are class-meanings; thus, somebody, someone haVB the class-
meanings of substantives, singulars, personals; he has the class-
meanings of substantives, sngulars, personals, males; it has the
class-meanings of substantives, singulars, non-personals; they has
the class-meanings of substantives and plurals. In the second place,
a pronoun may contain an element of meaning which makes the
pronoun represent sorne particular substantive form of the lan-
guage. Thus, the pronouns sorne and none tell us that the particular
substantive is one which has been recently rnentioncd (Here are
apples : Lake sorne); in contrast with this, something, somebody,
someone, nothing, nobody, no one tell nothing about the species.
Thirdly, sorne pronouns contain an element of meaning which tells
us which particular abjects in a species are concerned. Thus, he,
she, it, they irnply that not only the species (say, policeman) has
been mcntioned, but also that the particlllar abject of this spccies
(say, Officer Smith, or the one at this corner) becn
This feature of mcaninl-';, once defined, will be found 10 vanous
other forms of our language; it occurs, apparently without admix-
ture as the meaning of the article the, for this litHe ward tells us
onl; tbat the following substantive denotes an identified individllal
of a species.
In sum, then, we may say that certain meanings, once tbey are
defined can be recognized as recurring in wholc series of fonns.
In partlcular, the last-named type, which has ta do with the
fication of indjvidual objects of a speciBs, in tbe way of selectlOn,
inclusion exclusion, or numbering, elicits very uniform rcsponses
from diffcrent persons, and recurs with relative uniformity in
diffcrent languages; these types of meaning, accordingly, give
to the specially accurate form of speech which we caH
d-
9
. 7. Vocal gestures, serving an inferi?r of c.ommUnlCatlOn,
occur Dot only olltside of spcech, as ln an
but a1so in combination with in the dISpOSltlO
n
of
non-distinctive features of speech-sound, such as the "tone of
voice." Sorne conventional speech-forms, in fact, seem to lie on
the border-line; thus, we have seen that, in English, the exclama-
tions pst [pst] and sk [8], with which wc demand silence, vi?late
the phonehc pattern by the use as syllabics of the relatlvely
un-sonorous phonernes [s, li]. Less striking deviations from the
phonctic pattern sornetirnes occur in words whos: .
resembles that of a pointing gcsture. In English the lmtml
neme [t1 occnrs only in words of demonstrative an.d
ings, such as this, tkat, the, then, there, thougk; RusSlan, the
phonerne (e] occurs initially in none but demonstratlve words, such
as ['eto] 'this.' .
L Non-phoncmic, gesture-like becarne
In Plains Cree the word le:] 'yes' 1S ordlllanly spoken wh a dlph-
thongal glide"in the vowc1 and a final glottal .somewhat
as [ :(], although neither of these features is phonemlC III lan-
guage. In our slang fashions, peculiar pitch-schemes occaslOnalIy
become fixed for certain values; in the last years, Yeah? and 1s that
sol with a peculiar modification of the question-pitch, have been
used as facetious vulgarisms, expressing dishclief.
The latter expression has also a fonn ls zat so? which illustrates
another phase of unusual linguistic features, facetious mispronun-
148
MEANING
l\lEANING 149
riation. To say oxcuse me, for instance, is a form of tired
wit. Thcse distortions get their value from a resernblance to other
linguistic forms (as in Our example, the ward ox) or to the speech.
forrns of fareigners, sub-standard speakers, and children, as in
the facctiaus use of [nj] for [l'] in words like bird (irnitating the sub-
standard speech of New York City), or in the Use of baby-talk
(Alla boy! Alta dirl!).
Certain expressions have slurred and shortened br-forms in
which the phone tic pattern is lost; thcse are cornrnon formulas of
social intercourse, such as greetings and terms of address. Thus,
!low do you do? is shortened in ail manner of ways inta farms which
cannat be reeorded in terms of Engli.'lh phonemes, but only sug-
gcsted by such sketches as [j 1duw or [d' duwl.J; FI ow are you?
is something liko [hwajl haj;,]; madam appears as [Ip] in Yes'm.
These by-forros oecur only in the formula; in asking How do Vou do
it? ['haw Ju 'duw iti.J for example, we do Dot use the over-slurred
forrn. These shortened forms occur in various langllag()s; their
relation to normal speech is obscure, but evidently they represent
a kind of sub-lingui8tic communication, in which the ordinary
meaning of the forrns plays no part.
'Ve can mention any sound by means of Il rough imitation in
tenDS of vocal sound, as when we tell the calls of animaIs, or when
we report the noise of an engine. In this way we can also
tian talking about a persan who lisps, for instance,
someone may say, "1 am tired of his etcrnal yelh, yelh." The eom-
monest case is hypostasis, the mention of a phonetically normal
speech-forro, as when wc say, "That is only an if," or "Thcre is
always a but," or when wc talk about "the ward normalcy" or
"the name Smith." One may even speak of parts of words, as l
shall speak in this book of "the suffix -h in boyish." Hypostasis
.. i8 closely related ta quo/atian, the rcpetition of a speech.
"\ 9. 8. The peculiarities of the forros discussed in the last para-
graph consist in deviations from the ordinary tie-up of phonetic
form with dictionary rneaning. When there is no such dcviation,
and only a normal phonetic form with a dictionary meaning is
to be considcrcd, the latter will still exhibit great complexity.
'Ve have alreudy seen that present-day knowledge does not suffice
ta unravel aIl the entanglernents of roeaning, but there are two
main features of the dictionary meaning of speech-forms which
dcmand such comment as we are able ta make.
Very many linguistic forms are used for more than one typical
"situation. In English, wc speak of t.he head of an army, of a pro-
cession, of a hOllschold, or of a river, and of a head of cabbage;
of the mou/h of a botUe, cannon, or river; of the eye of a needle,
and of hooks and eye8 on a dress; of the teeth of a SI1w; of the
tongue of a shoe or of fi wagon; of the neck of fi bottle and of a
neck of the woods; of the arms, legs, and buek of a chair; of the
foot of a rnount.ain; of hear/s of celery. A mltn may be a fox, an ass,
or f1 dirty dog; a woman, fi peach, Zeman, eat, or goose; people are
shal'p and keen or duIl, or else bright or foggy, as ta their wits;
warm or cohl in ternperament; crooked or straight in eonduct; a
persan may be up ! the a, at sea, off the handle, off his base, or
cven bes2'de himself, without actually rnoving from thc spot. The
reader will be able ta add examples practically wit.hout limit;
t.here is no greuter bore than the enurneration and classification of
these "metaphors."
The remarkable thing about thesc variant meanings is our as-
surance and our agreement in viewing one of the meanings as
normal (or central) and the others as marginal (metaphoric or
lransferred). The central meaning is favored in the sense that wc
understand a forrn (t.hat is, respond ta it) in the central meaning
unless sorne feature of the practic[l,1 situat.ion forces us ta look
ta a transferred meaning. If we hear someone say There gaes a
fox! we look for a real fox, and if t.his is out of the question, we are
likely to take the uttel'ance as displaced speech (say, as make-
believe or as part of a fairy-tale). Only if sorne situational feature
forces us - say, if the speaker is pointing at a man - do we take
the forrn in the transferred sense. Even if we heard someone say,
The fox prornised to help her, we should think of a fairy-tale rather
than of fox 'unscrupulolls and clever persan.' Sometimes the
practical feaiure that forces \1S to take a form in t.ransferred mean-
ing, has been given by speech: Gld il,{r. Smith is a fox is bound ta
be taken in transferred meaning, becausc wc do not caH l'cal foxes
"Mr." or give them family-names. Ile married a lemon forces us
ta the trnnsferred meaning only beeause wc know that men do
not go through a marriage cerernony with a pieee of fruit. On
the other hand, special practical situations may change ail this.
People who lived close ta the Fox Indians rnight, \vithout special
constraint, take fox in our examples in the transferrcd scnsc 'mem-
ber of the Fox nation.'
150 MEANING MEANING 151
In sorne cases a transferrc:d rneaning is linguistieally determined
by an accompanying fonn. The word cat al ways has a transferred
meaning when it is aecompanied by the suffix -kin (catkin) , and
.word PUIJ6'Y when it is compounded with willoU! (pussy-willow);
slmllarly, the ward eye wheu it has the suffix -let (cycle/). The words
dog, monkey, beard when they appear with the marks of verb
vatian (say, with a preceding to), alwa.ys have transferred lllean-
ing (to dog someone's foot6'feps; don'f monkey wh that; to beard a
lion in his den). These linguistic features may be purely negative:
give out, used without an Chis money gave out; our horses
gave out), always has a transferred meaning (' become exhausted ').
In these cases the structure of the language rccognizes the tran&-
ferred meaning. Even a linguist who made no attempt ta deline
meanin!!:s would have ta specify that give out, intransitive, meant
something different (was a different form) from give out, transi-
ti"le (he gave out tickets).
In manycases we hesitate whet.her ta view the fonn as a single
form ,vith several meanings or as a set of hornonyms. Examples
of this are air 'atmosphere; tune, rnelody; manner' (this last
including airs 'haughty manners '), !cry 'instrument for locking
and unlocking; set of tones in music,' charge 'att.ack; load; ac-
cuse; debit,' sloth 'name of an animal; laziness.'
We are likely to make the mistakc of thinking that the trans-
ferred meanings of our arc natura! and even inevitable
in human speech - the more so, as ther appear also in other Euro-
pean languages. This bst, however, is mely a result of our com-
mon cultural traditions; while transferred meanings occur in ail
languages, the particular ones in any givcn language are by no
means to be taken for granted. :r.,Teither in French nor in German
can one speak of the eye of a necdIe or of an car of grain. To speak
of the foot of a mountain seems natural to any European, but it
wo'.Ild be nonsense in Menomini and doubtless in many other
languages. On the other hand, in Menornini [una:?m:w] 'he places
hirn in position' has also t.he transferred rneaning 'he picks lice
from hirn.' In Russian, [no'gal 'leg' is not used of the leg of a chair
or table; this transferred rncaning apppars only in the diminutive
['noska] 'litHe leg; log of a chair or table.' Accordingly, when the
linguist tries ta statc meanings, ho safely ignores the Ilses of dis-
placed speech, but does his bost ta register al! cases of transferred
meaning.
AIl this appUes also ta another type of deviant meaning, the
narrowed rneaning, with this difference, that we are far more
ready ta acccpt a form in a narrowed rneaning. The practical
situation guides us at once ta take car in different narrowed senses
in The diner is the second car forward (' railroad-carriage '); Does
the car stop ai this corner? (' street-car '); Bring the car close to the
curb ('motor-car'). When we hear the command ta caU a doctor,
we take it at once ta mean a doctor of medieine. A burner is primarily
a persan or instrument that bums things, but usually, in a nar-
rowed sense, a gas-tap arranged ta give a certain kind of flame.
A bulb among gardeners is one thing and among electricians an-
other. A glass is usually a drinking-glass or a looking-glass;
glasses are usually eye-glasses. Narrowed meanings are hard ta
define, because, after aIl, every occurrence of a form is prompted
by some one practical situation which need not contain aIl the
possibilities of meaning: apple is used now of a green one, now of a
red one, and sa on.
The language itself, by formaI characteristics, rccognizes nar-
rowed meanings in certain combinations. For instance, blaekbird
is not merely any 'black bird ': in this cornbination the meaning
of b7m:k is greatly narrowed; sirnilarly blueberry, whitefish, and the
like.
Widened meanings are less cornmon. In general, cat is the do-
mestic animal, but now and then wc use the ward ta include lions,
tigers, and sa on; the word dog, however, is Dot similarly used to
include wolves and faxes. On the other llimd, hound is used
cally and facetiously of any kind of dog. Often, the widened
ing is recognized in the structure of the language, and appears
only when certain accompanying forms are present. Thus meai
is edible flesh, but in meat and drink and in sweetmeais it is food
in generaI; fowl is an edible bird, but in jish, flesh, or fowl or the
f owl of the air it is any bird.
Often enough the speakers of a language do not distinguish a
central and a marginal meaning in cases where an outsider might
sec two situationally different values; thus, day in English means
a period of hours (Swedish dygn [ciYlln]) or the light
part of this period (in contrast with night; Swedish dag [da:gJ).
9. 9. The second important way in which meanings show in-
stability, is the presence of supplementary values which we caU
connotations. The meaning of a form for any one speaker is nothing
152 MEANING 153
hindranee, in the premises, heirs and assigns; criminals' terms crass
but to the poi nt; a stickup, a shot (of 11)hiskey), gel pinehed.
The connotation of learned forms is vaguer but. more frequent;
almost any colloquial form has a parallel fmm with learned con-
notation.
As these examplcs show, the learned, e1cgant, and arehaic types
of connotation merge in rmlllY a fonn. ln formal speech and in
writing-, we custornarily prefer learned forms, up to a certain de-
grec: he who uses too many learned forms is a stihed or a
tircsome writer.
Foreign speech-forms bear connotations of their own, which
reRect our attitude toward foreign peoples. The forcign fcatures
of fonn may consist in peculia.ri tics of sound or of phondic pattern:
garage, mirage, rouge, a je ne sais quoi; alla podrida, chile con carne;
dolce far nierile, fortissimo; Zeitgeist, TVanderlld; intelligentsia. In
ot.her instances, the forcign fealure lies in the construction, in the
French types marriage of convenienee and Ihal goes wilhout saying.
This flavor is turncc! to face tious use in mock-foreign forms, sueh as
nix come erouse (mock-Gcrman), ish gabibble (' it.';; noIle of my co!l-
cern,' snpposcdly .Judeo-German). Schoolboys use mock-Latin-
isms, such as the nonsense-form qul sirli lJuidit, or macaronic verse;
Boyiblls priti girloTllm, girlibus likibus, 1J)anti somorum.
Some languages, and rnost notably, pcrhaps, English, contain a
great mass of or fvreign-learned forms - a class of
forms with a separatc styk of pattern and dcrivation. Our
books of rhetoric distinguish thcsc fonns, as the" Latin-French"
part of our vocabulary, from the "native" or "An?;]o-Saxon"
forms. The connotation, however, docs not c!cpcnd. direetly upon
the actual provenience of the forms. The word chair, for inst.ance,
more than a result of the situations in which he has heard this
form. If he has not heard it very many times, or if he has heard it
under very unusual circumstances, his use of the form may deviate
from the conventionaL \Ve combat such personal deviations by
giving explicit definitions of meaning; this is a chief use of our
dictionaries. In the case of scientific tenus, we manage to keep the
meaning nearly free from connotative factors, though cven here
we may be unsuccessful; the number thirteen, for instance, has
for many people a strong connotation.
Themost important connotations arise from the social standing
of the speakers who use a form. A form which is uscd by a less
privileged class of speakers often strikes us as coarse, ugly, and
ain't gol none, 1 seen ii, 1 done sound nasty to the
speaker of standard English. This may be offset by sorne special
factor; the speech-forms of tramps or criminals may bear a con
notation of devil-may-care wit, and those of a rustic type may
strike us as homely but poetic. A form uscd by a more privileged
class of speakers may strike us as ovor-formal or prettified and
affected. 110st speakers of Central-Western American English
find this connotation in the use of [a] instcad of [j,:] in forms like
laugh, bath, can't and of [juw] instead of [uw] in forms like tune,
sue, stupid.
Connotations of local provenience are closely akin to thcse; a
Scotch or an Irish locution has hs own tang; so have, in America,
certain rcal or supposed Anglicisms, such as luggage (for baggage)
or old chap, old dear as tenns of address.
Even in communities that have no writing, sOrne forms are
recognized (rightly or wrongly) as archaisrfts; in communities
that have written records, these serve as addiLional sources of
archaic forms. Examples arc, in English, the old second-person
singular forms (thou hast), the third-person forms in -th (he hath),
the old present subjunctive (if this be treason) , the pronoun ye,
and many forms like eve, e'en, e'er, morn, anent, and so on. Some-
times fully current locutions may preserve sorne special aphoris/ic
form; thus, an old sentence-construction survives in a few proverbs,
such as First come, firsl served or Old saint,
l-The connotation of technical forms gets its flavor from the stand-
ing of the trade or ctaft from which thcy are taken. Sea-terms
sound ready, honest, and devil-may-care: abafl, aloft, the wl of ht"s
jib, stand by; legul terms precise and a bit tricky: withoul let or
:-<ORMAL
IIe came too soon.
Ifs too bad.
-Where're you gmmJ?
now
if he comes
so (that) you don't lose il.
LIOAHNBD
l-le prematurely.
It is regrettable.
TVhat ; your destination?
at present
in case (in case Ihat, 't'n the event
that, in Ihe conlingency that) he
cornes; should he come, ...
in vrder that YOll may not lose it,
lest you lose it.
154 MEANING
MEANING 155
is Latin-French in origin, but does not helong to the foreign-
learned part of our vocabulary. The chief formaI characteristics
of our foreignlearned forms is perhaps the use of certain accented
suffixes and combinations of suffixes, such as Htij] ability; [-'ejsQ-]
education. Another feature is the use of certain phonetic alterna-
tions, such as [sijv] in receive, but [sep] in reception and [sij] in
receipt, or [vajd] in provide, but [vid] in provident, [viz] in visible,
and [vii] in provision. These peculiarities suffice to mark certain
words and constituents of words as foreign-learned, especially
certain prefixes (ab-, ad-, con-, de-, dis-, ex-, in-, per-, pre-, pr(}-,
re-, trans-); theso prefixes themselves in part show pcculiar phonetic
alternations, as in con-tain but collect, correct, and ab-jure but
abs-tain. Semantically, our foreign-Iearned forms are peculiar in
the capricious and highly specialized meanings of the combinations;
it seems impossible. for instance, to set up any consistent meaning
for elements like [sijv] in conceive, deceit'e, perceive, receive or [tend]
in attend, contend, distend, pretend, or [d(j)uws] in adduee, conduce,
deduce, induee, produce, reduce. The connotative flavor of these
forms lies in the learned direction: a spcaker's ability to use these
forms measures his education. Errars in their use (malaproprms)
mark the semi--educated speaker. The less cducated speaker fails
to understand many of these forms, and is to this extent shut out
from sorne types of communication; he may take vengeance by
using mock-learned forms, such as absquatulate, discombolulate,
rambundious, scrumptious. Many languages contain a foreign-
learned layer of this kind: the Romance languages have a Latin
type, Iargely identical with ours; Russian, heside a fair sprinkling
of this type, has learned forms from Old Bulgarian; Turkish has a
stratum of Persian and Arabic words, and Persian of Arabie; the
languages of lndia similarly use Sanskrit forms.
Opposed to the foreign-Icarned connotation, the slangy con-
notation is facetious and unrestrained: the users of slang forms are
young persons, sportsmen, gamblers, vagrants, criminals, and,
for that matter, roost other speakers in thcir rclaxed and unpre-
moods. Examplcs are familiar, such as guy, gink, gazebo,
gazook, bloke, bird for 'man,' rod or gat for 'pistol,' and so on; the
slang form may at the same time bc foreign, as loco 'erazy,' sabby
'understand,' vamoose 'go away,' from Spanish. The value is
largely faeetious; when the slang form has bcen in use too long,
it is likely to be replaced by sorne new witticism.
tf- 9. 10. The varieties of connotation are countless and indefinahle
and, as a whole, cannot he clcarly distinguished from denotative
meaning. In the last analysis, every speech-form has its own
connotative Havor for the entire speech-community and this, in
turn, is modified or even offset, in the case of each speaker, by the
connotation which the form has acquired for him through his
special experience. It may he wen, however, to speak briefly of
two more types of connotation which stand out with at least
clearness.
V'" In many speech-communities certain improper speech-forms are
uttered only under restricted circumstanccs; a speaker who utters
thern outside the restriction is shamed or punished. The strictness
of the prohibition ranges from a rnild rule of propriety to a severe
tabu. The improper forms helong for the most part to certain
spheres of meaning, but often enough there exist by their side
forms with the same denotation but without the improper con-
notation, as prostitute by the side of the improper form whare.
Sorne improper forms denote objects or persans that are not to
he named in a casual way, or pcrhaps not to be named at aIl.
In English, various terms of religion, such as God, devil, heaven,
heU, Christ, Jesus, damn are proper only in serious speech. Viola-
tion of the rule exposes the speaker ta reproof or avoidance; on the
other hand, in certain groups or under certain conditions, the
violation connotes vigor and freedom. In many communities the
names of persons are tabu under sorne circumstances or to sorne
people. The male Cree Indian, for example, does not speak the
names of his sisters and of sorne other female relatives; he explains
the avoidance by saying, "1 respect her tao much."
Another direction of impropriety is the tabu on so-called obscene
forms. In English there is a severe tabu on sorne speech-forms
whose rneaning is connected with excretory functions, and on sorne
that deal with reproduction.
A third type of improper connotation is less universal among us;
the avoidance of ominous which name something
painful or dangerous. One avoids the words die and death (if
anything should happen w me) and the names of sorne diseases.
Other peoples avoid mention of the left hand, or of thunder-
storms.
In sorne communities onc avoids the names of game animaIs,
either during the hunt or more generally. Under special conditions
156 l\lEANING MEANING
157
(as, on the war-path), many speech-forms may be avoided, or
inverled speech, saying the opposite of what one means, may be in
order.
11. The second more specialized type of connotation that
here deserves to he pointcd out, is inlensity. The most character-
istic intense forms are exclamat1,ons. For these we have in English
not only a special secondary phoneme [!], but also certain special
speech-forms, inlery"ections, such as oh! ah! ouch! These forms aU
retlect. a violent. stimulus, but. differ in connotation from an ordinary
statement in which the speaker merely says that he is undergoing
strong stimulus.
Certain spcech-forms have an animaled flavor, aldn ta the ex-
clamatory, as, for instance, the placing first of cert.ain adverbs:
Away Tan John; Away he Tan. In connect.ed narrative a similar
flavor appears inless violent transpositions: Yesterday he came (and
sal ...) is Illore lively than JIe came yesterday ... rn Enp;lish
the historieal present, in narratin?; past l'vents, is either c1egant,
as in t.he summary of a play or story, or, in ordinary speech, slightly
vulgar: Then he cornes bock and says 10 me . . .
English is especially rich in another type of intense forms, the
symbolic forms. Symbolic fonDS have a connotation of somehow
illustratin?; the meaning more immediately t.h::m do ordinary
speech-forms. The explanation is a maLter of grammatical st.ruc-
turc and will eoneern us later; t.o the speaker it sms as if the
soumis were espeeially suited ta the meaning. Examples are
flip, jlap, flop, fliUer, flimmer, flicker, fluller, flash, flush, }lare,
g/are, aliller, glvw, a/oat, glimmel', bang, bump, lump, Ihmnp, Ihwack,
whack, sniff, sni,[fle, sizzle, wheeze. Lan?;uages that. have
symbolic forrns show sorne agreement, but prohably more dis-
ngrcement as to the types of sounds and meanings which are as-
sociatcd. A special type of symbolic form, whidl is quite widely
distributed, is the repctition of the form with sorne phonetic varia-
tion, as in snip-snap, zig-zag, riff-raff, J'im-y"ams, jiddle-faddle,
teeny-tiny, ship-shape, hodge-podge, hugger-mugger, hon!ry-/rmk.
Closely akin ta t,bcse are imitaiive or onomalopoet; intense (imns,
which denote a sound or an abject which givcs out a sound: the
imitative speech-form resembles this sound: coc/;-a-door/le-doo,
meeow, moo, baa. Many bird names are of this sort cuckoo, bob-
white, Doubled forms are common: bow-wow,
ding-dong, pee-wcc, choo-choo, chug-chug. These forms diffcr from
language ta language: the French dog says gnaf-gnaf [paf paf];
the German ben says bimbam.
Among the forms just cited, sorne have an infantile connotation;
they are nursery-forms. The most familial' are papa and marna.
In English almost any doubled syllable may be used, in almost
any meanng, as a nursery-word; each family develops its pri-
vate supply of the type ['dijdij, 'dajdaj, 'dajdij, 'mijmij, 'wawa].
This custom provides spcech-forms which the infant can repro-
duce with relative ease, and it hclps adults to turn the infant's
utterances into conventional signais.
The pet-name or hypochoristic connotation largely merges with
that of the nursery. In English, relatively fcw pet-names like Lulu,
have the doubled nursery form; in French this type is cornrnon:
Mimi Nana and sa on. English petMnames are less uniform:
" .
Tom, Will, Ed, Pal, Dan, Mike can be described structurally as
shortenings of the full narne; this is not the case in Bob for Robert,
Ned for Edward, Bill for William, Dick for Richard, Jack for John.
Sorne have the diminutive suffix [-ij], as Peggy, Maggie for Mar-
garet, Fanny for Frances, Johnny, Willie, Billy.
There is sorne intensity also in the connotation of nonsense-
forms. Sorne of these, though conventional , have no denotation
at aIl, as triI"'la-Ia, hey-diddle-diddle, tarara-boom-de-ay; others
have an explieitly vague denotation, as folrde-rol, gadget, con
M
niption jUs. Any speaker is free to invent nonsense-forms; in fact,
any form he invents is a nonsense-form, unless he succeeds in
the almost hopeless task of getting his fellowMspeakers to accept
it as a signal for sorne mcaning.
GRAMMATICAL FORMS 159
CHAPTER 10
GRAMMATICAL FORMS
10.1. Our discussion so far has shown us that every language
consists of a number of signaIs, linguisiic forms. Each linguistic
form is a fixed combination of signaling-units, the phonemes. In
every language the number of phonemes and the numher of ae-
tuaUy occurring combinations of phonemes, is strictly limited.
By uttering a linguistic form, a speaker prompts his hearers ta
respond ta a situation; this situation and the responses to H, are
the linguistic meaning of the form. We assume that each linguis-
tic form has a constant and dctinite meaning, dillerent from the
meaning of any other linguistic form in the samc language. Thus,
hearing several utterances of sorne one linguistic form, such as
l'm hungry, wc assume (1) that the differences in sound are i r ~
relevant (unphonetic), (2) that the situations of the several speak-
ers contain sorne common features and that the differences 00-
tween these situations are irrelevant (unscmantic), and (3) that
this linguistic meaning is dillercnt from that of any other form in
the language. We have seen that this assumption cannot be veriw
fied, since the speaker's situations and the hearer's responses
may involve almost anything in the whole world, and, in particu-
lar, depend largely upon the momentary state of their nervous
systems. Moreover, whon we deal with the historieal change of
language, we shaU he eoncerned with faets for which our assump-
tion does not hold good. In the rough, however, our assumption
lS justitied by the mere fact that speakers co-operatc in a very
refined way by means of language-signaIs. In describin/!; a lan-
guage, we are eoncerned primarily with the working of -this cOw
operation at any one time in any one community, and not with
its occasional failures or with its changes in the course of history.
Accordingly, the descriptive phase of linguistics consists in a
somewhat rigid analysis of spcech-forms, on the assumption that
these speech-forms have constant and definable meanings ( 9.5).
Our basic assumption does have to be modified, howcver, right
at the outset, in a different way. When we have recorded a fair
158
numher of forms in a language, we always discover a feature
which we have so far ignored in our discussion: the partial resem-
blance of linguistic forms. Suppose we hear a speaker say
John Tan,
and a little later hear hOO or some other speaker say
Johnfell.
We recognize at once that these two forms, John Tan and John
fell, are in part phoneticaUy alike, since bath of them contain an
element John [jan], and our practical knowledge tells us that the
meanings show a corresponding resemblance: whenever a form
cantains the phonetic clement [jan], the meaning involves a cer-
tain man or boy in the eommunity. In faet, if wc are lucky, we
may hear somoone utter the forro
John!
aIl by itself, without any aecompanOOent.
After observing a number of such cases, wc shaU he constrained
ta modify the basic assumption of linguistics to read: In a speech
w
community some utterances are alike or paray alike in sound and
meaning.
The corrunon part of partly like utterances (in our example,
John) consists of a phonetic form with a constant meaning: it
answers, therefore, to the definition of a linguistic form. The parts
which are not common ta the partly-like utterances (in our e x ~
ample, Tan in the one utterance, and fell in the other) may, in
the same way, turn out ta he linguistic forms. Having hcard the
form John Tan, we may later hear the form Bill Tan, and perhaps
even (say, in answer to a question) an isolated Han. The saroe will
happen with the eomponent fell in John feU: we may hear a form
like Danfell or even an isolated Fel!.
In other cases, we may wait in vain for the isolated form. Know-
mg the forms John, Bill, and Dan, we may hear the forms, Johnny,
Billy, and Danny and hope to hear now an isolated -y [-ijJ with
sorne such meaning as 'Httle,' but in this instance we sha11 he dis-
appointed. In the same way, familiar with the forms play and
dance, we may hoar the forms playing and dancing, and thon hope,
in vain, to hear an isolated -ing [-ilJ], which might reassure us as to
the somewhat vague meaning of this syllable. In spite of the fact
that some components do not occur alone, but only as parts of
larger forms. wc nevertheless calI these components linguistic
160 GRAMMATICAL FORMS GRAMl\lATICAL FORMS 161
forms, sinee they are phonetic forms, such as [ij] or [i!]], with COn-
stant meanings. A linguistic form which is never spoken aIone is
a bound form; al! others (as, for instance, John Tan or John or TUn
or Tunning) are free forms.
In other cases we wait in vain for the occurrence of a form even
as part of sorne other form. For instance, having heard the form
cranberry, we saon reeognize the component berry in other forms,
such as blackberry, and may even hear it spoken alone, but with
the other component of cranberry we shal! have no such luck.
Not only do we wait in vain ta hear an isolated *cran, but, listen
as we may, we never hear this element outside the one combina,.
tian cranberry, and we cannot dicit from the speakers any other
fonn whch will contain this clement cran-. As a practical matter,
observing languages in the field, we soon learn that it is unwise to
try to elicit such forms; our questions confuse the speakers, and
they may get rid of us by sorne faise admission, such as, "Oh, yes,
l guess cran means red." If we avoid this pitfall, ,ve shall come
to the conclusion that the element cran- occurs only in the com-
binatian cranberry. However, since it has a constant phonetic
form, and since its meaning is constant, in so far as a cranberry
is a definite kind of berry, different from al! other kinds, we say
that cran-, tao, is a linguistc form. Experience shows that we
da wel! to generalize this instance: unique elements, which oecur
only in a single combination, arc linguistic forms.
Sometimes we may be unable to decide whether phonetically
like forms are identical in meaning. The straw- in strawberry is
phonetically the same as the siraw- in strawflower and as the isolated
straw, but whether the rneanings are "the same," we cannat say.
H wc ask the speakers, they will answer sometnes one way,
sometimes another; they are no more able ta tell than we. This
difficul ty is part of the universaI difficulty of semantics: the
practical world is not a world of clear-cut distinctions.
10. 2, We sec, then, that. some linguist.ic forms bear partial
tic resemblances t.o ot.her forms; examples are,
John Tan, John feU, Bill ran, Bill fell; Johnny, Billy; playing,
dancing; blackberry, cranberry; sirawberry, strawjlower. A linguistic
form whieh bears a partial phanctic-sernantic resemblance to sorne
other linguistc form, lS a campIez form.
The cornmon part of any (two or more) cornplex forms i5 a
linguistic farm; it is a constituent (or component) of thcse complex
forms. The constituent is said ta be contained t:n (or ta bo included
in or to enier inio) the complex forms. If a complex form, beside
the cornmon part, contains a remainder, such as the cran- in
cranberry, which does not occur in any other complex form, this
remainder a1so is a lnguistic form; it is a unique constituent of the
complex fonn. The constituent forms in our examples aboye are:
John, ran, Bill, feU, play, dance, black, berry, siraw, jlower, cran-
(unique constituent in cranberry) , -y (bound-form constituent
in Johnny, Billy), -ing (bonnd-form constituent in playing, danc-
ing). In any complcx forro, each constituent is said ta accompany
the other constituents.
A linguistic fonn which bears no parlial phonetc-semanlic
resemblance to any other form, is a simple form or morpheme.
Thus, bird, play, dance, -!l, -ing are morphemes. :\Jorphemes
may show partial phonctic rcsernblanees, as do, for instance, bird
and burr, or even homonymy, as do pear, pair, pare, but this
resemblance lS purely phonetic and is not paralleled by the mean-
ngs.
From aIl this it appcars that every complex form is entrely
made up, sa far as its phonetically definable constituents are con-
eerned, of morphemes. The nurnbcr of t.hese ultimale constuenls
may run very high. The farm POOl' John ran away contains five
morphemes: poor, John, ran, (l- (a bound form recurring, for
instance, in aground, ashore, alofl, around), and way. However, the
structure of complex forms is by no means as simple as this; wc
could not understand the forms of a language if we merely redueed
al! the complex forms ta their nltimate constituents. Any
speaking person who conccrns himself with this matter, is sure to
tell us that the immediale constituents of Poor John ran away are
the two forms poor John and mn away; that each of these is, in
turn, a complex form; t.hat the irnrnediate constituents of ran away
are ran, a morpheme, and aWGY, f1 complex form, whose constitu-
ents are the marphemes 0- and way; and that the constituents
of paal' John are t.he morphemes paal' and John. Only in this
way will a proper am11ysis (that is, one which takes account of the
meanings) lead to the ultimately constituent morphemes. The
reasons far this will occupY us later.
10. 3, A rnorphernc can he describcd phonctically, since it
consists of onc or mre phonemes, but hs meaning cannat be
analyzcd within the scope of our science. For instance, we have
162 GRAMMATICAL FORMS GRAMMATICAL FORMS 163
seen that the morpheme pin bears a phonetie resemblance to other
morphemes, such as pig, pen, tin, len, and, on the basis of these
resemblances, can be unalyzed and described in terms of three
phonemes ( 5.4), but, since these resemblances are not connected
wit.h resemblances of meaning, we cannat attribute any meaning
to the phonemes and cannat, within the scope of our science,
analyze the meaning of the morpheme. The meaning of a mor-
phemc is a 8ememe. The linguist assumes tbat each sememc is a
constant and definite unit of meaning, different from aIl otber
meaniogs, including aIl otber sememes, in the language, but he
cannot go beyond this. There is nothing in the structure of mor-
phemes like wolf, fox, and dog ta tell us the relation hetween theu
meanings; this is a problem for the zoologist. The zoologist's
definition of these m e a n i n ~ s is welcome to us as a practicaI heIp,
but it cannat be confirmed or rejected on the basis of our science.
A workable system of signaIs, such as a language, can contain
onIy a small number of signaling-units, but the things signaled
about - in our case, the entire content of the practical world
- may be infinitely varied. Accordingly, the signaIs (linguistic
forms, with morphemcs as the smallest signals) consist of different
combinations of the signaling-units (phonemes), and each such
combination is arbitrarily assigned ta sorne feature of the practical
world (sememe). The signaIs can be analyzed, but not the thiogs
signaled about.
This re-enforces the principle that linguistic study must always
start from the phonetic form and not from the meaning. Phonetic
forms -let us say, for instance, the entire stock of morphemes in
a language - can be described in terms of phonemes and their
succession, and, on this basis, can be classified or listed in sorne
convenient arder, as, for example, alphabetically; the meanings
- in our example, the scmemes of a lanp;uage - could he analyzed
or systematically listed only by a well-nigh omniscient observer.
10. 4. Since evcry complex form is made up entirely of mor-
phemes, a complete list of morphemes would account for an the
phonetic forms of a language. The total stock of morphemes in a
language is its lexicon. Howevcr, if we knew the lexicon of a
language, and had a reasonably accurate knowledge of each se-
meme, we ffiight still fail to understand the forms of this language.
Every uttcrance contains sorne significant features that are not
accounted for by the lexicon. We saw, for instance, that the five
morphemes, John, poor, ran, way, Cl- which make up the form
Poor John ran away, do not fully account for the meaning of this
utterance. Part of this meaning depends upon the arrangement
- for example, upon the arder of succession - in which these
morphemes appear in the complex form. Every language shows
part of its meanings by the arrangement of its forms. Thus, in
English, John hit Bill and Bill hit John differ in rneaning by virtue
of the two different ordeTS in which the morphemes are uttered.
The mcaningful arrangements of forms in a language constitute
its grammar. In general, there seem to he four ways of arranging
linguistic forms.
(1) Order is the succession in which the constituents of a com
w
plex form are spoken. The significance of order appears strikingly
in contrasts such as John hit Bill versus Bill hit John. On the other
hand "'Bill John hit is not an English form, because our language
, -
does not arrange these constituents in this order; similarly, play-ing
is a form, but *ing-play is not. Sometimes differcnces of order have
connotative values; thus, Away Tan John is livelier than John ran
away.
(2) Modulation is the use of secondary phonemes. Secondary
phonemes, we recaH ( 5.11), are phoncmes which do not appear in
any morpheme, but only in grammatical arrangements of mor-
phemes. A morpheme like John [Jan] or run [ron] is really an
abstraction, because in any actual utterance the morpheme is
accompanied by sorne secondary phoneme which conveys a gram-
matical meaning. In English, if the morpheme is spoken alone, it
is accompanied by sorne secondary phoneme of pitch ( 7.6): it is
eitber John! or John? or John [.]- this last with falling final-pitch,
as, in answer to a question - and there is na indifferent or abstract
form in which the morpheme is not accompanied by any final-
pitch. In English complex forms, sorne of the constitucnts are
always accompanied by secondary phonemes of stress ( 7.3); thus,
the difference in the place of stress distinguishcs the noun convict
from the verb convict.
(3) Phonetic modification is a change in the primary phonemes
of a form. For instance, when the forms do [duw] and not [nat]
are combined ioto a complex form, the [uw] of do is ordinarily
replaced by [ow] , and, whenever this happons, the not loses its
vowel, 80 that the combined form is don't [dow nt]. In this cxample
the modification is optional, and we have al80 the unmodified
164 GRAMMATICAL FORMS
G RAl\IlVr AT l CAL FOR::\lS 165
forms in do not, with a difference of connotation. In other cases
we have no choice. Thus, the suffix -ess with the meaning 'female,'
as in count-ess, is added also ta duke [d(j)uwk], but in this combi-
nation the form duke is modified ta duch- [do-J, for the ward is
duchess ['does].
Strictly speaking, we should say that the morpheme in such cases
has two (or, sometimes, more) different phonctic forms, such as
not [nat] and [nt], do [duw] and [dow], duke and duch-, and that
each of these alternants appears under certain conditions. In our
examples, however, one of the alternants has a much wid"r range
than the other and, accordingly, is a basic alternant. In other cases,
the alternants are more on a par. ln run and ran, for instance,
neither alternant is tied ta the presence of any accompanying form,
and we might hesitate as ta the choice of a basic alternant. We
flnd, howcver, that in cases like keep : kep-t the past-tense form
contains an alternant (kep-) which occurs only ,vith a certain
accompanying form (-t); accordingly, ta obtain as uniform as
possible a statement, wc take the infinitive form (keep, run) as
basic, and describe thc alternant which appears in the past tense
(kep-, ran) as a phonetically modified form. 'Ve shaH sec other
instances where the choice is more difficult;\ve try, of course, to
make the selection of a basic alternant so as to get, in the long
l'un the simplest description of the facts.
(4) Selection of forms contrihutes a factor of meaning because
different forms in what is otherwise the same grammatical ar-
rangement, will result in difTerent meanings. For instance, sorne
morphemes spoken with exclamatory final-pitch, are caUs for a
person's presence or attention (John! Boy!), while others, spoken
in the same way, are commands (Hun! Jump!), and this differ-
ence extends a1so ta certain complex forms (lrfr. Smith! Teachcr!
versus Hun away! Backwalel'.l). The forms which, when spoken
with exclamatory final-pitch, have the meaning of a cali, may he
said, by virtue of this faet, to make up aforrn-class of the English
language; we may cali it the form-class of "personal substantive
expressions." Similarly, the forms which, when spoken with ex-
clamatory final-pit.ch, have the meaning of a eommand, make up,
by virtue of this fact, the English form-class of "infinitive
sions." Whether an exclamation is a call or a command, depcnds
upon the selection of the form from the one or the other of these
two classes.
The meaning of a complcx form depends in part upon the selec-
tion of the constituent forms. Thus, drink milk and walch John
name actions, and, as we have just seen, are infinitive expressions,
but fresh milk and pOOl' John name abjects and are substantive
expressions. The second constituents, milk, and John, are the
same; the difference depends upon the selection of the first con-
stituent. By virtue of this difference, the forms drink and walch
belong to one English form-elass (that of "transitive verbs "),
and the forms fresh and poor ta another (that of "adjectives").
The features of selection are usually quite complicated, with
form-classes divided int.a sub-classes. In English, if we combine
a form like John or the boys (form-class of "nominative substan-
tive expressions ") with a form like ran or 11)enl home (form-class
of "finite verb expressions "), the resul tant cornplcx forrn means
t.hat t.his abject 'pcrforms' this action (John ran, the boys ran,
John wenl home, lhe boys went home). These feat.ures of selection,
however, are supplernented by a furtlwr 1mbit: wc say John runs
fast but the boys run fast, and wc oevel' rnake the reverse combina-
tians of John with l'un fast, or of the boys with T'uns fasl. The form-
class of nominative expressions is divided into two sub-classes
(" singular" and "plural JI) and the form-class of finite verb ex-
pressions] ikewise, inta t wo su (" singular " and "plurai"),
such that in the complex forms which mean that an abject per-
forms an action, the two constituents agree as t.o the" singular"
or "plural" sub-class. In Latin, the form pater filium amai (or
filium pater amat) means 'the father loves the son,' and the form
patrem fllius amal (or fflius patrem amal) means 'the son loves the
father'; the forms pater' father' and filius 'son' belon!'; ta a form-
class (" nominave case") whase forms, in cambinatian with a
verb likc amal 'he lovcs,' denote the 'performer' of the action;
the forms patrem 'father' and filium 'son' belong to a diffcrent
fonn-class (" accusative case "), whose forms, in combination with
a verb like amat, denotc t.he 'undergoer' ('abject' or 'goal')
of the action.
The features of selection are often highly arbitrary and whim-
sical. We combine prince, author, sculptor with the suffix -ess in
princess, auihoress, sculptress (in this last case with phonetic modi-
fication of [T] to [l'D, but not king, singer, paintcr. By virtue of
this habit, the former words belong to a form-class from which
the latter words are excluded.
166
GRAMMATICAL FRMS GRAMMATICAL FRMS 167
10.6. The features of grammatical arrangement appear in
various combinations, but can usually he singled out and sep-
arately descrihed. A simple feature of grammatical arrangement
is a grammatical fealure or taxeme. A taxeme is in grammar what
a phoneme is in the lexicon - namely, the smallest unit of form.
Like a phoneme, a taxeme, taken by itself, in the abstract, is
meaningless. Just as combinations of phonemes, or, less commonly,
single phonemes, occur as actual lexical signaIs (phonetic fonns),
so combinations of taxemes, or, quite frequently, single taxemes,
occur as conventional grammatical arrangements, tadic fO'l'Jn8.
A phonetic form with its meaning is a linguistic form; a tactie
form with its meaning is a grammatical farm. When we have oc-
casion to contrast the purely lexical character of a linguistic form
with the habits of arrangement to which it is subject, we shall
speak of it as a lexicalform. In the case of lexical forms, we have
defined the smallest meaningful nnits as morphemes, and their
meanings as sememes; in the same way, the smallest meaningful
units of grammatical form may he spoken of as tagmemes, and
their meanings as episememes.
The utterance Run!, for example, contains two grammatical
features (taxemes), namely, the modulation of cxclamatory final-
pitch, and the selective feature which consists in the use of an
infinitive verb (as opposed, for instance, to the use of a noun, as
in John!). Eaeh of these two taxemes happens to he, in English,
a tactie form, sinee each is currently used as a unit of signaling.
Taking eaeh of thern with its meaning, we descrihe them as units
of grammatical form (tagmemes). The tagmeme of exclamatory
final-pit ch occurs with any lexical form and gives it a grammatical
meaning (an episememc) which wc may roughly descrihe, per-
haps, as 'st.rong stimulus.' The tagmeme of selection by which
infinitive forms arc rnarked off as a form-class, has a grammatical
meaning (an episerncmc) which wc may call a class-meaning and
roughly definc as 'action.'
A tagmemc may consist of more than one taxeme. For instance,
in forms Iikc John ran; poor John ran away; the boys are here; l
know, wc find several taxemes. One constituent belongs to the
form-class of nominative expressions (John, poor John, the boys, 1).
Thc other constituent belongs to the form-class of flnite verb ex-
pressions (ran, ran away, are here, know). A further taxeme of se-
lection assigns certain finite verb expressions to certain nomina-
tive expressions; thus, the constituents arc not interchangeable
in the three examples l am, John is, you are. A taxeme of arder
places the nominative expression before the finite verb expression:
we do not say *ran John. Furthcr taxemes of order, in part r e ~
versing the basic one, appear in special cases like did John run?
away ran John; will John'! A taxeme of modulation appears only
in special cases, when the nominative expression is unstrcssed,
as in l know [aj 'now]. Taxemes of phonetic modification appear
also in certain special cases, such as John's here, with [z] for is,
or l'd (JO, with [dl for wouId. Now, none of these taxemcs, taken
by itself, has any meaning, but, taken all togcther, they make up
a grammatical form, a tagmeme, whose meaning is this, that the
one constituent (the nominative expression) 'performs' the other
constituent (the finite verb expression).
If we say John ran! wit.h exclamatory piteh, we have a complex
grammatical form, with three tagmemes. One of these is 'strong
stimulus,' the second is '(object) performs (action),' and the third
has the episememe of 'complete and novel' utterance, and con-
sists, formally, in the selective feat,ure of using an actor-action
phrase as a sentence.
10. 6. Any utterance can he fully descrihed in terms of lexical
and grammatical forms; we must remember only that the mean-
ings cannot he dcfined in terms of our science.
Any morpheme can be fully descrihed (apart from its meaning)
as a set of one or more phonemes in a certain arrangement. Thus,
the morpheme duke consists of the phonemes, simple and com-
pound, [dl, [juw], [k], in this order; and the morpheme -ess con-
sists of the phonemes le], [s], in this order. Any complex form can
he fully describcd (apart from its meaning) in terms of the im-
mediatc constituent forms and the grammatical fcatures (taxeroes)
by which these constituent forms are arranged. Thus, the com-
plex form duchess ['daces] consists of the immediatc constituents
duke [djuwk] and -ess [es], arranged in thc following way:
Selection. The constituent duke bclongs to a special class of
English forms which combine with the form -ess. This form-class
includes, for instance, the forms count, prince, lion, tiger, author,
waiter, but Dot the forros man, boy, dog, singer; it is a sub-class of
a larger form-class of male personal nouns. The form -ess c o n ~
stitutcs a little form-class of its own, by virtuc of the fact that it
(and it alone) combines with precisely the forms in the class just
168 FOHMS
GRA:\l AT1C.:\ L FOR:\1 S
169
described. Ali these fads, taken togethcr, may be viewed as a
single taxeme of selection.
Grder. The form -ess is spoken after the accompanying form.
1"!odula/ion. The form -css is spoken unstressed; the accompany-
ing form has a high stress.
Phone/ic modification. The [juw] of duke is replaced by [oJ, and
the [kJ by l].
Givcn the forms dukc and -css, the statement of these four
grammatical features fully describes the complex form duchess.
Any actual utterance can be fully described in terms of the
lexical forrn and the accompanying grammatical features. Thus,
the utteranee Duchess! consists of the lexical form duchess and the
two taxemes of excbmatory tinaI-pitch and selection of a sub-
stantive expression.
If sorne science furnished us with definHions of the meanings of
the units here concerned, defining for us the meanings (scmemes)
of the two morphemes (duke and -ess) and t.he meanings (cpi-
sernernes) of the thl"Ce tagmemes (arrangcment of duke and -ess;
use of exe!amatory final-pitch; selection of a substantive expres-
sion), then the rneaning of the utterance Duchess! would be fully
ana.lyzed and defined.
10, 7. The grammatical forms are no exception to the necessary
prineiple - strielly speaking, we should cali it an assumpiion -
that a language ean convey only sueh meanings as arc attached to
sorne formai feature: the speakprs can signal unly by means of
signaIs. lVIany students of language have been misled in this matter
by the faet that the fOl'mal feat.ures of grammar are not phonemes
or combinations of phonemes whieh wc can prollounce or tran-
scribe, but. merely arrangements of phonetic forrns. For this our
scholastie t.radition may be largely t.o blame; if il. were not for this
tradition, there would perhaps he nothing difficult about the faet,
for instance, that in English, John hil Bill and Bill Ail John signal
t"wo different situations, or that convict stressed on the first syllable
differs in Olcanng from convict stressed on the second syllable, or
that there is Il difference of rneaning between John! and John?
and John.
A forill likc John or /Un, mentioned in the abstract, without, for
instance, any specification as to final-piteh, is, properly speaking,
not a l'cal lnguistic forOl, but only a lexical form; Il lingustic
fOlm, as actually uttercd, always cont.ains a grammatical form.
No matter how simple a form we take and ho\\' we uUer it, we have
alnmdy made sorne selection by virtue of wltich the utlerance
conveys a grammatical rneaning in addition ta its lexical content,
and we have used sorne pitch-scheme which, in English at allY
rate, lends it a p;rammatical meaning such as 'stalement' 'yes-
. , '1 1.." 1 t ,
or-no questIOn, oupp ement-ques IOn, or exc ama .1On.
The grlunmatical forrno of a can he groupecl into three
great classes:
(1) \Vhen a form is spoken alone (that is, not as a constituent
of a larger fonn) , it appears in sorne sentence-lype. Thus, in English,
the use of the sccondary phoncrne [!] gives us the sentence-type of
exclamation, and the use of a substantive exprpssion givps us the
type of a call (John!).
(2) Whenever two (or, rarcIy, more) forms arc spoken together,
as eonstiluents of a complex fonn, the gramrm11ical features by
which they are combined, make up a construdion. Thus, the
grammatical features by which duke and -css combine in the forrn
duchess, or the grammatical features by which poot John and mn
away combine in the [orm poor John Tan away, up li con-
struction,
(3) A third great class of grammatical forrns must probably be
set up for the cases where lt form is spoken as the convenlional
substitut,c for any one of a whole class of other fonns. Thus, the
b'Blective fcaturc by which the fonn he in English is a conventiomtl
substitute for a whole cbss of other forms, such as John, poor John,
a policeman, the man l yesterday, whoever dl lh, and sa on
(which forms, by virtue of thio habit, constlIte form-class of
"singula
r
male substantive expressions"), musl cloubtless be
viewed as an example of a third cllios of grammatical forms, to
which wC may give the name of substitutions.
SENTENCE-TYPES
171
CHAPTER 11
SENTENCE-TYPES
11.1. In any utterance, a linguistic form appears either as a
constituent of sorne larger form, as does John in the utterance
John ran away, or else as an independent form, not included in any
larger (c?mplex) linguistic form, as, for instance, John in the
exclamatlOn John! When a linguistic form occurs as part of a larger
form, it is said to he in included position; otherwse it is said ta he in
absolute position and to constitute a sentence.
A form which in one utterance figures as a sentence, may in
another utterance appear in included position. In the exclamation
just citcd, John is a sentence, but in the exclamation Poor John! the
form John is in included position. In this latter exclamation
! is a but in the uttcrance POOT John ran away;
lt lS lU lllciuded posItion. Or again, in the utterance just cited
poor John ran away is a sentence, but in the utterance When th;
dog barked, poor John ran away, it is in included position.
An utterance may consist of more than one sentence. This is the
case wheu the uttcrance contains several linguistic forms which
are no.t by any meaningful, conventional grammatical arrangement
(that IS, by any construction) united into a larger form, e. g. : How
are you1 It's afine day. Are you going to play tennis this afternoon't
Whatever practical connection therc may be hetween these three
forms, there is no grammatical arrangement uniting them one
larger form: the utterancc consists of three sentences.
It is evident that the sentences in any utteranee are marked off
by the fact that cach sentence is an independent linguistic
form, not mcluded by virtue of any grammatical construction in
any linguistic form. In most, or possibly aIl languages, how-
ever, varIOUS taxemes mark off the sentence and further dis-
tinguish different types of sentence. " 1
In English and many other languages, sentences are marked off
by modulation, the use of secondary phonemes. In English,
secondary phonemes of pitch mark the cnd of sentences and
distinguish three main sentence-types: John ran away [.] 'John
170
ran away [?] Who ran away [il To each of these, further, we may
add the distortion of exclamatory scntence-pitch, so that we get
in aIl, six types, as deseribed in 7.6.
This use of secondary phonemes to mark the end of sentences
makes possible a construction known as parataxis, in which two
forms united by no other construction arc united by the use of
onlyone sentence-ptch. Thus, if we say D's ten o'clock [.l 1 have
to go home [.] with the final falling pitch of a statement on o'clock,
we have spoken two sentences, but if we omit this final-pitch
(substituting for it a pause-pitch), the two forms are united, by
the construction of parataxis, into a single sentence: It's ten
o'clock L] [ have to go hom,,-; [l
Another feature of sentenee-modulation in English and many
other languages, is the use of a secondary phoneme to mark
emphatic parts of a sentence. In Englsh wc use highest stress
for this ("Now it's my tum," 7.3). The cmphatic clement in
English may he marked also by the use of special constructions
(It was John who did that) and by word-order (Away he ran);
in languages where stress is not significant, such methods prevail,
as in French C'est Jean qui l'afait [s e zun ki la fe] 'It is John who
did it.' Sorne languages use special words before or after an em-
phatic element, as Tagalog [ikaw 'l)u? al) nag'sa:bi nijan] 'you
(emphatic particle) the one-who-said that./ i.e. 'Yon yourself said
so'; Menomini ['jo:hpeh 'niw, kan 'wenah 'wa:pah] 'Today (e:o-
phabc particle), not (emphatic particle) tomorrow.' Our h1gh
stress ean even strike forms that are normally unstressed: of, for,
and by the people; immigration and emigration.
11. 2. Beside features of modulation, fentures of selection may
serve to mark off different sentence-types. This is the case in sorne
of the examples just given, where 11 special construction, or the use
of a special particle, marks an emphatic clement. In English,
supplement-questions are distinguished not ouly by their special
pitch-phoneme [i,], but also by a selective taxeme: t.he form used as
a supplement-question either consists of a special type of word or
phrase, whieh we may calI an interrogative substitute, or else con-
tains such a word or phrase; Who? With whom? Who ran away?
With whom was he talking?
Perhaps aIl languages distinguish two great scntence-types
which wc may caU full sentences and minOT sentences. The differ-
ence consists in a taxeme of selection: certain forms are fav01it6
172
SENT F,NCE-TYP ES SENTENCE-TYPES 173
sentence-forms; when a favorite scntence-form is used as a sen-
tence, t.his is a full sentence, and when any other form is uscd as
a sentence, this is a minor sentence. In English we have two fa-
vorite sentencc-forms. One consists of phrases-
phrases whosc struet.ure is that. of t.he actor-action const.ruct.ion'
John ran an'ay, Who ran away? Did John run away? The
eonsists of a command - an infinitive verb with or without modi-
fiers; Come! Be good! This second type is ahvays spoken with
exeIamatory sentencc-pit.ch; the infinitive may be accompanied
by the word JjOU as an actor; You be good! As these examples
show, the meaning of the full sentence-type is somct.hing like
'complete and novel uttcrance' - that is, the speaker implies
that .what he says is a full-sized occurrence or instruction, and
that lt somehow altcl's the hearer's ;;it.uation. The more deliber-
ate the speech, the more likcly arc the sentences ta be of the full
t!pe. The nat.ure of the episcmeme of full sentences has given
nse to much philosophie di;;pute; to define this (or any other)'
meaning exact.ly, lies beyond the domain of linguistics. It is
a serious mistake t.o try ta use this meaning (or any rneanings),
rather than formaI features, as a starting-point for linguistic
discussion.
.Quit.e a .fe"': of the prcsent-day Indo-European languages agree
wlth Enghsh lU using an actor-action form as a favorite sentcnce-
type. Sorne, such as t.he other Germanie languages and French,
agree also in that the uctor-action form is always a phrase, with
the actor and the action as separat.e words or phrases. In sorne
of these languages, howcver - for instance, in Italian and Spansh
and in the Slavic languages - the actor and the action are Dound
forms which make up a single word; Italian canto ['kant-o] '1 sing'
canti [' kant-il 'thou singest,' can/-a [' kant-a] 'he (she, il) sings:'
and sa on. A word which cantaing a favorite sentence-form of
its language lS a sen/ence-word.
. Some languages hrlvc difTerent favorite sentence-types.
Slan has an actor-action type of sentence-ward tinite verbs lke
those of Italian; [po'ju] '1 sing-,' [po'jos] 'thou singest,'
'he (she, it) sings,' and so on. In addition ta t.his, it has another
type offul! sentence: [j'van du'rak]' John Os) a fool,' [sol'dat 'xrabr]
'the soldier (is) brave,' [o'/ets 'dama] 'Father (is) at home.'
In this second type, one component, which IS spoken fir;;t, i5
a substantive; the other farm is a substantive to whieh the first
is equated, or an adjective (adjectives have a special form for
this use), or an adverbial form.
When a language has more than one type of full sentence, these
types may agree in showing constructions of two parts. The com-
mon name for sueh bipartite favorite sentence-forms is predica-
tions. In a predication, the more objeet.-like component is caUed
the subject, the other part the predicate. Of the two Russian types,
the former is callod a narrative predication, the latter an equational
predication. For a language like English or Italian, which has
only one type of bipartite sentence, these terms are superfluous,
but often employed: John ran is said to be a predication, in
which the actor (John) is the subject and the action (ran) the
predicate.
Latin had the same types of full sentence as Russian, Dut the
narrative type existed in two varieties: one with an actor-action
construction: cantat 'he (she, it) sings,' amat 'he (she, it) loves,'
and one with a goal-action construction: cantti/ur 'it is being sung,'
amatur 'he (she, it) is loved.' ':'he equational type was less com-
mon than in Russian: beiitus ille 'happy (is) he.'
Tagalog has five types of predication, wit.h this cornmon fea-
ture: eit.her the subject precedes and a particle [aj] (artel' vowels,
[iD intervenes, or the reverse arder is used without the particle.
There is, first, an equational type: [alJ ' ba:ta j maba' it] 'the
child is good,' or, with inverse order, [maba'it alJ 'ba:tar] 'good
(is) the child.' Then therc are four narrative types, in which the
predicates are transient words, which denole things in four
ferent relations to an action. The four types of transient words
are:
aet.or: [pu'mu:tul] 'one who eut'
goal: [pi' nu:tul] 'something eut'
instrument.: [ipi'nu:tul] 'something eut wit.h'
place: [pinu'tu:lan] 'somcthing cut on or from.'
These transient words arc by no moans confincd, like our verbs,
ta predicative position; they can figure equally well, for instance,
in equational sentences, as: [alJ pu' mu; tul aj si 'hwau] 'the one
who did the cutting was John,' but in the predicate position they
produce four types of narrative predication:
actor-action: [sja j pu'mu:tul nal) 'kaJlUj] 'he eut sorne wood'
goal-action: [pi 'nu:t.ul nja alJ 'ka:huj] 'was-cut by-him the
wood,' Le. 'he eut the wood'
174 SENTENCE-TYPES SENTENCE-TYPES
175
instrument-action: [ipi'nu:tuI nja aI] 'gu:luk] 'was-cut-with by.
him the bolo-kne,' Le. 'he eut with the
0010'
placeaction: [pinu' tu:Ian nja aI] 'ka:huj] 'wascutfrom byhim
the wood,' i.e. 'he eut (a piece) off the wood.'
Gorgian distinguishes betwecn an action-type, as ['v-ts?er]
'Iwrite' and a sensation-type, as ['m-esmis] 'mesound-is,' i.e.
'1 hear.' Such distinctions are never carried out with scientific
consistency; Georgian classifies sight in the actiontype: ['v.naxavJ
, I-see.'
Not aIl favorite sentenceforms have bipartite structure: the
command in Englsh consists of merely an infinitive form (come;
be good) and only occasionally contains an actor (you be good).
In German, beside a favorite sentence.typc of actoraction which
c10sely resembles ours, there is an impersonal variety, which dif-
fers by not containing any actor: mir ist kali [mi:r ist 'kaIt] 'to-
me is cold,' that is, ' 1 feel cold;' hier toird getanzt [' hi;r virt ge' tantst]
'here gets danced,' that is, 'there is dancing here.' ln Russian,
there is an impersonal type which differs from the equational
predication by the absence of a subject: ['nuzno] 'it is necessary.'
11. 3. English has a subtypc of full sentences which we may
cali the explicitaction type; in this type the action centers round
the verb do, does, did. This taxeme of selection appears in the
contrast between, say, 1 heard him and 1 did hear him. The explicit-
action type has several uses. When the verb is an emphatic eIe.
ment (spoken with highest stress), the normal type emphasizes
the lexical content (the sememe) of the verb, as in "1 heard him"
(but did not see him), or in "Run home!" (don't walk); the explicit
action type emphasizes the occurrence (as opposed to non-
occurrence) or the time (present or pasi) of the action, as in "1
did hear him," or "Do run home!" Secondly, we use the
action type wherever the verb is modified by not, as in 1 didn't
hear him or Don't run away; thus, English, by a taxeme of selection,
distinguishes a negalive type of full sentence.
Further, within our explcitaction type, we distinguish a suD-
type in which the verb do, does, did precedes the actor. This in-
ver/ed type occurs in frmnal yes-orno questions, along with
question-pitch; Did John run away? Didn't John run awayf in
contrast with the uninverted (informal) type: John ran away?
John didn't run away?
The features just discussed are not so widely paralleled among
languages as the more gcneral characteristics of English full
sentences. In German, for instance, the negative adverb is not
tied up with a spccial-sentcncetype: Er kommt nicht [e:r 'komt
'nixt} 'he cornes not' is like Er kommt bald [e:r 'komt 'baIt] 'he
cornes soon.' Other languages, however, resemblc English in
using special sentence types with negative value. In Finnish,
negative sentences have a special construction: thc verb (whicb,
as in Italian, includes actor and action in one scntenceword)
is a special negative verb, which may he modified by an infinitive-
like form of another verb:
luen '1 read' en lue 'Idon't read'
lue! 'thou readest' ct lue 'thoudostnot read'
lukec 'he reads' ci lue 'hedoesn't read.'
ln Menomini there arc three main types of full sentence, equa-
tionaI, narrative, and ncgative:
narrative: [pi:w]
equational: [enu? pajiatJ 'he - the one who cornes,' that is,
'It's he that's coming'
ncgative: [kan upianan] 'not he-cornes (negative),' that s, 'He
does not come.'
ln the negative type the two parts arc, on the one side, the nega
tive word [kanJ in its various inflections and, on the other, the
rest of the scntence, markcd by the use of special verb-forms.
Special types of full sentences for formaI questions are more
widespread. German useS actoraction fonns in which the verb
precedes the actor: Kommt cr? ['komt e:r?] 'cornes he?' in
with Er kommt [e:r 'kornt] 'he cornes.' French also uses speclal
interrogative constructions; 'Is Johu coming?' is cither Jean
vient.il? [zan vj ent i?] 'John cornes he?' or Est-ce que Jcan vient?
[e s k;) zan vjen?J 'Is it that John cornes?' In Menomini the three
main types of full sentence have each an interrogative sub-type:
narrative: [pU?J 'Is he coming?'
equation: [enut pajiat?] 'he (interrogative) the one who cornes?'
that is, 'Is it he that is coming?'
negative: [kane:? upianan?] 'uot (interrogative) hecomes (nega-
tive)?' that is, 'Isn't he comng?'
Other languages lack a special sentence type for formaI yes-orno
questions, but sorne of them use special interrogative words, as
Latin vcnitne7 [we'nit nc?] 'Is he coming?' and num venit? 'You
176 SENTENCE-TYPES

177
don't mean to say he is coming?' (expectation of ncgative reply),
in contrast with venin 'He is coming?' This use of special liHie
words (particles) to mark a formai yes-or-no question, appears in
many languages, such as Russian, Chinese, Tagalog, Cree.
l\'To;:;t languages agree with English in supplement-
questions by the presence of special words, but the details differ:
in Tagalog and in Menomini, for instance, the supplement-question
is always an equational sentence, e.g., :Ylenomini [awF:?
'who t.he-one-who-comes?' that is, 'Who is coming?'
The English command is an example of a speci: sentence-t.ype
used in exclamations. Other languages also have special types of
full sentence for some kinds of exclarnationH. In Menomini there
are two such, one of surprise, where the occurrence is new or
unforeseen, and one of disappoiniment at the nonoccurrence of
something expected:
SURPRISE
narrative: [piasah!] 'and so he' s corninp;!'
equational: [enusar pajiat 1] 'and so it's he that's coming!'
negative: [kasa? upianan!] 'and so he isn't coming!'

narrative: [piapah!] 'but he was coming!'
equational: [enupa? pajiat!] 'but he wa;:; the one who was
coming!'
negative: [kapa? upianan!] 'but he wasn't cominp;!'
11, 4, A sentence which does not consist of a favorit.e sentence-
form is a minor sentenee. Some forms occur predominantly as
minor sentences, entering into few or no constructions other than
parataxis; sucb forms are Interjections arc cither
special words, such as ouch, oh, sh, gosh, hello, sir, ma'm, yes, or
eise phrases (secondary interJections), often of pcculiar construction,
such as dear me, me, gracious, goodness sakes alive,
oh dear, by gol/y, you angel, please, thank you, good-bye.
In general, minor sentences seem to be eher completive or
exc/amalory. The compleUvl> type consists of a forill which merely
supplements ft, situation - that is, an earlier speech, a gesture, or
the mere presence of an object: This one. Tomorrow morning.
Gladly, if 1 cano Whenever you're ready. Here. When? With whom?
Mr. Brown: .1.111'. Smith (in introducing people). Drugs. State
Street. They occur especially as answcrs t.o quest.ions; for this use
we have the special completive interjections, yes and no. Even in
this regard languages differ: French says si 'yes' in answer to
negative 'luestions, such as 'Isn't he corning'?' but oui [wi] 'yes'
in answer to othcrs, snch as 'Is hc coming?' SOlIle hnguages have
no such interjections. Polish answers with ordinary advcrbs,
aflirmatively with iak 'thus, so' and ncgalively with n [ne]
'not.' Finn"h answers affirrnalive1y by an ordinary form, e.g.
Tulette-ka kaup1tngista? - Tulernme. 'Arc yOU coming from town?'
_ '\Ve are corning,' and negatively by its negative verb: Tunneite-
ko herra T-,ehdon? - En (or En tunne) 'Do you know l'l'Ir. Lehto?'
- '1 clon't' (or '1 don't know').
Exciamatory minor s('nlenCN; occur unclcr a violent stimulus.
They eonsist of interjections or of normal fonns that do not bc10ng
to favorite ;:;entence-types, and oftcn show parataxis: Ouch, damn
il! This 1l.my, please.! A substantive form naming a hcarer is \lsed
in English as a demand for his presence or attention: John! Little
boy! Yon with the g/asses! \Vith parataxis: Hello, John! Come Itere,
Little boy! The int.erjections sir and ma'am are especially devoted to
this use; in the same way Russian us!'s an interjection [sJ, as
[da-s] 'yes, sir; yes, ma'am,' without distinetion of sex.
languages have special vocative fonns for this use, as Latin Balbus
(man's-narne), vocative Balbe, or Fox ',volllan,' voca-
tive [iskwe], and [islnve:wak] 'wornen,' vocahve [iiikwe:tike]. In
l'vIcnomni the terms of relationship have special, llighly irreg;ular
vocative forms: [m:r'm:hl 'illY older brother,' vocative [mUle:?] or
[neki:jah] 'my lllothcr,' vocative [nerF:h]. Other words arc spoken
as 'lOcatives with short vowels inslead of long: [mt:i2:muh] 'wo-
man,' vocative [mdf:muh]. In Sanskrit, vocative forrns were
unstressed.
Occasionally we find minor scntenees of aphol'istic type ( 9.9)
used with mueh the saille value as full sentences; English examples
arc The more you have, the more you want. The more, the mf:TTier.
First eome, jirst served. Old saint, young sinner.
11. 5, In most languages the sentence is characterizcd also bya
selective fcature more general t.han ail those we have been dis-
cussing: some linguistc forms, which we cali bound forms ( 10.1),
are never used as sentences. English examples are the -ess [es] in
countess, lioness, duchess, ete., or t.he lis] in boyish, childish,
greenish, etc., or the -s [8] in hats, books, cups, etc. Thesc are genuine
linguistic forms and couvey a rneaning, but they OCCLU only in
178 SENTENCE-TYPES SENTENCE-TYPES 179
construction, as part of a larger form. Forms which occur as
sentences are free forms. Not every language uses bound forms:
modern Chinese, for instance, seems ta have none,
A free form which consists entirely of two or more lesser free
forms, as, for instance, poor John or John ran away or yes, sir, is
a phrase. Afree form which is not a phrase, iB a word. A ward, then,
is a free form which does not consist entirely of (two or more) lesser
free forms; in brief, a word is a minimum free form.
Since only free forms can he isolated in actuai speech, the ward,
as the minimum of free form, plays a very important part in our
attitude toward language. For the purposes of ordinary lue, the
ward is the smal1est unit of speech. Our dictionaries list the words
of a language; for aU purposes except the systematic study of
language, this procedure is doubtless more useful than would he a
list of morphemes. The analysis of linguistic forms into words is
familiar ta us because we have the custom of leaving spaces he-
tween words in our writing and printing. People who have not
1earned ta read and write, have some difficulty when, by any
chance, they are caUed upon to make word"divisions. This diffi-
culty is 1ess in English than in some other languages, such as
French. Thc fact that the spacing of words has become part of our
tradition of writing, goes to show, however, that recognition of the
ward as a unit of speech is not unnatural ta speakers; indeed,
exeept for certain doubtful cases, people easily learn ta make this
analysis.
In our school tradition we sometimes speak of forms like book,
books, or do, does, did, done as "different forms of the same ward."
Of course, this is inaccurate, sinee thore are differences of form
and meaning between the memhers of these sets: the forms just
cited are different linguistic forms and, accordingly, different
words.
In other cases, inconsistencies in our habits of writing may make
us uneertain. We write John's in John's ready, where it is two
words (John and [z], an alternant of is) and in John's hal, where it
is one word (consisting of John and thc bound form [-z], posses-
sive). We write the boy' s as though it were two or three words, but,
strictly speaking, it is only one word, since the immediate eon-
stituents are the boy and [-z] possessive, and the latter is a bound
form; thjs appears clearly in cases like the king of England' s or
the man 1 SGW yesterday's, where the meaning shows that the [-z]
if; in construction with the entire preeeding phrase, sa that the
two are united inta a single long word.
11. 6, In the case of many languages, however, it is impossible ta
distinguish consistently, on the one hand, between phrases and
words and, on the other hand, between words and bound forms.
The linguist cannot wait indefinitely for the chance of hearing a
given form used as a sentence - that is, spoken alone. Sorne forms
are rarely sa used. Inquiry or experiment may caH forth very
different responses from hearers. Are English forms Iike the, a, is,
and ever spoken alone? One can imagine a dialogue: 1s? - No;
was. The word because is said to he a woman's answcr. An impa"
tient listener says And? We can imagine a hesitant speaker who
says The , . . and is understood by his hearers. Aside from such
far-fetchcd situations, the gcncral structure of a language may
make one classification more convenient than another for our
purpose. The form the, though rarely spoken alone, plays much
the same part in our language as the forms this and that, which
freely occur as sentences; this paraUelism leads us ta class the
as a ward:
this thing .' that thing : the thing
this : that : (the).
In other cases, the difficulty is due ta features of phonetic
fication. The forms [z] in ready, lm] in l'm hungry, or [nt]
in Don'tf are unpronouneeable in English, but we have to class
them as words, for they are merely alternants of the pronounceable
forms is, am, not. In French we have even the case of a single
phoneme representing two words: au [0] in a phrase like au Toi
[0 rwa] 'ta the king,' arises by phonetic modification of the two
words [a] 'to' and le [10] 'the'; this [0] is homonymous with the
words eau 'water' and haut 'high.'
In other cases the doubtfu1 forms are units of grammatical
selection rather than of modification, and yet, in view of the total
structure of their language, may he best classified as words. French,
again, has severaI forms of this sort. Absolute forms like moi
[mw:;,] 'l, me' and lui [lqi] 'he, him' are replaced in certain con-
structions by shorter forms that do not ordinarily appear in
absolute use, such as je [2:a] 'l,' me [ma] 'me,' il [il] 'he,' le [la]
'him 'j for instance: je le connais [Z0 1 bnE] '1 know him,' il me
connait [i m k::JllE] 'he knows me.' The replacement of the abso1ute
forms by these conjunct forms is ta be described as a feature of
180 SENTENCE-TYPES SENTENCE-TYPES 181
selection rather than of moditieatian; nevertheless, the conjunct
forms, iargcly because of their paralieJism with the absolute forms,
have the status of words.
A less important border-Jine case is the use of bound forms in
hypostasis ( 9.7), as when we speak of a girl in her teens, taking
up aIl kinds of isms and ologies.
At the other extreme we tind forms which lie on the border be-
tween words and phrases. A form like blackbird rcscmbles a twa-
ward phrase (black bird) , but we shall flnd that a consistent
description of English is bound ta class this form as a single (com-
pound) ward. In this case there is a clear-cut difference, since in
blnckbird the sccond ward (bird), has a weaker stress instead of a
normal high stress, a difference which in English is phonemic, and
this formaI differenec correlates with the semantic difference he-
tween blackbird and black bird. The distinction s not always sa
clear: ice-cream ['ajs ,krijm], spoken with only one high stress, will
he classed as a (compound) ward, hut the variant pronunciation
ice cream ['ajs 'krijm], with two high stresses, will he classed as a
two-word phrase. Similar variants exist in types like messenger
boy, lady friend.
This criterion of stress fails us in forms like devil-may-care (as
in a deuil-may-care manner) or jack-in-the-pulpit (as the name of a
plant). If the former were devil-may-care-h, we should Dot hesi-
tate ta class it as a ward, since here one of the immediate con-
stituents s the bound form -ish. The forms of the type devil-may-
care are classed as words (phrase-words) because of certain other
features whieh, within the system of the English language, place
them on a level wHh other words. One of these is their peculiar
function; as a phrase devil-may-care would be an actor-action form,
but as a phrase-word it fills the position of an adjective. Another is
their indivisibility: the plant-name jack-in-the-pulpit cannat be
modified by putting the ward /iule in front of pulpit, but the cor-
responding phrase permits of this and other expansions.
This latter prnciple, namely that a word cannat be interrupted
by other forms, holds good almost universally. Thus, one can
say black - 1 should say, - birds, but one cannat
similarly interrupt. the compound ward blackbirds. The exceptions
ta this principlc are sa rare as ta seem almost pathologicaL Gothie
had a bound form [ga-] which was prefixed especial1y to verbs:
['se:hwi] 'he should sec,' [ga'se:hwi] 'he should be able t.o see.'
Yet occasionally we tind words included between this [ga-] and
the main body of the verb, as in the translation of Mark 8, 23:
['frah ina ga- u hwa 'se:hw] 'he asked hirn whether [u] he saw
anything [hwa].'
None of these criteria can he strictly applied: many forms lie
on the border-line betwecn bound forms and words, or between
words and phrases; it is impossible ta makc a rigid distinction
between forms that may and forms that may not be spokcn in
absolute position.
11. 7. The ward is not prirnarily a phonetic unit: we do not,
by pauses or other phonetic features, mark off those segments
of our speech which could be spoken alone. In various ways, how-
ever, different languages givc phonetic recognition ta the word-
unit: sorne, like French, very li ttle, and others, like English, very
much.
As a free form, the word is capable of being spoken in absolute
position; accordingly, it is subject ta the phonetic patterning of
its language. It is sure ta contain at lcast one of the phonemes
which normally serve as syllabics; interjections, sueh as our sh
[Ii] and pst [pst], occasionally violate this principle. The initial
and final consonant.s and clusters in the ward arc necessarily such
as can occur at the beginning and at the end of speech; thus, no
English ward begins with [IJ] or [mb] and none ends with rh] or
[mb].
Beyond this, many languagcs place further restrictions on the
phonetic structure of the ward. We may find that sorne of the
permitted medial clusters do not occur wHhin the body of a single
word; in English, permitted clusters like [M, vt, tsv, ststr], as
in rash child, give ten, it' s very cold, least strong, and double con-
sonants, like [nn, tt, bb], as in ten nights, that time, nab Bill, do
Dot occur within simple words. On the other hand, French, with
its insertion of Fa], and languages like Fox or Samoun, whch use
no final consonants, tolerate no more clusters within a phrase
than withn a ward.
Sorne languages have the peculiar restriction, known as vowel-
harmony, of tolerating only certain combinations of vowels in the
successive syllables of a ward. Thus, in Turkish, the vowcls of fi
word are either ail front vowcls [i, y, e, pl, as in [sevildirerncmek]
'Dot to be able ta cause to be loved,' or al! back vowels [, u, a, 0],
as in [jazldramamak] 'Dot ta be able to cause ta be written'.
182 SENTENCE-TYPES SENTENCE-TYPES 183
In Chinese we have the extreme of structural word-marking;
each ward consists of one syllable and of two or threc primary
phonemes: a non-syllabic simple or compound phonemc as initial,
a syllabic simple or compound phoneme as final; and one of the
pitch-schemes ( 7.7); the initial non-syllabic may he lacking;
the language has no bound forms.
In English and many other languages, cach ward' is marked by
containing one and only one high stress (forgiving; convict, verb;
convict, noun). In sorne of these languages the ward-unit is even
more plainly marked, in that the position of a ward-stress bears
a definite relation ta the beginning or ta the end of the ward: in
Bohemian and in Icelandic the first syllable is stressed, in Cree
the third-last (the antepenult) , in PoUsh the next-to-Iast (the
penult). In Latin the penult was stresscd, as in amamus [a'ma:-
mus] 'we love,' unless this syllable had a short vowel followed by
no more than one consonant, in which case the antepenult was
atressed, as in capimus ['kapimus] 'we take.' In languages like
these, the stress is a ward-marker, which indicatcs the beginnings
or ends of words, but, since its position is fixed, it cannot distin-
guish between different words. In Italian, Spanish, and modern
Greek, the stress cornes always on one of the last thrce syllables
of a ward. In ancient Greek a ward had either a simple accent
on one of the last three syllables or a compound accent on one of
the last two, with sorne further restrictions based on the nature of
the primary phonemes in these syllablcs.
Among stress-using languages, sorne, Uke English, start the
stress at the beginning of a ward whose stress cames on the first
syllable; witness contrasts like a name versus an aim or that scold
versus thal's cold ( 7.5); others, sueh as Dutch, Italian, Spanish,
and the Slavic langua/!;es, regulate the onset of stress by purely
phonetie habits, starting the stress on a consonant which precedes
a stressed vowel, even though this consonant belongs ta another
ward, as in Italian un aUro [u'n aItro] 'another.' A language like
French, which uses no stress-phonemes, cannat in this way mark
its word-units.
Phonetic recognition of the ward-unit, in cases like the above,
is disturbed chiefly by two factors. 'Vords which contain, among
their ultimate constituents, two or more free forms, generally
have the phonetic character of phrases. In English, compound
words have the same medial clusters as phrases: stovewtop [vt],
chesl-strap [ststrJ, pen-knife [nn], grab-bag [bb]; phrase-dervatives
may eyen have more than one high stress: old-maidish ['owld
'mejdis], jack-in-the-pulpit ['Jek in 'l5e 'pulpit].
On the other hand, words in included position are subject to
modulations and phonetic modifications which may remove the
phonetic characteristics of word-marking. Thus not in the phrase
don't [Ldow nt]loscs both Hs high stress and its syllabic; the phrase
can't is homonymous with the word cant; compare, similarly, lock it,
with locket, feed her ['fijd rJ with feeder, and sa on. In the normal
pronunciation at aU [c't Dl] the stress begins on the [t] of at. These
included variants, in which a ward loses the phonetic features
that characterize words in absolute position, will concern us in
the next chapter. In the present connection it is worth noticing,
however, that in a smaH way these modified phrases may never-
theless involve phonetic recognition of the word-unit, bccause they
contain phonetic sequences that do not occur in single words.
Thus, the final sequence [ownt] is permitted in English, but oc-
curs only in the phrases don't and won't, and not in any one ward.
In South German dialects sorne initial clusters, such as [tn, tH.] oc-
cur in phrases, thanks ta phonetic modification of the first ward,
as in [t naxt] 'the night,' [t sta:st] 'thou standest,' but not in any
one word. In North Chinese a phrase may end in syllabic plus
[rI, as in [jaw
3
'ma r
S
] 'little horse,' but only as a result of phonetic
modification of two words, - in our example, [maS] 'horse' and
[r
2
] 'son, child, smaIl.'
In the few languages which use no bound forms, the word has
a double importance, since it is the smallest unit not only of free
form but also of linguistic form in generaI. In languages which
use bound forms, the word has great structural importance he-
cause the constructions in which free forms appear in phrases
differ very decidedly from the constructions in which free or
bound forms appear in words. Accordingly, the grammar of these
languages consists of two parts, called syntax, and morphology.
However, the constructions of compound words and, ta sorne ex-
tout, of phrase-derivatives, occupy an intermediate position.
SYNTAX
185
CHAPTER 12
SYNTAX
12. 1. Traditionally, the grammar of most languages is dis-
cussed under two heads, syntax and morphology. The sentence- -
types, which we surveyed in the last chapter, arc placed under
the former heading, and so arc the types of substitution (which
we shal! consider in Chapter 15), but grammatical constructions,
which wc shaH now examine, are dealt with partly under the head-
in!!; of morphology. There has been considerable debate as ta
the usefulness of this division, and as to the scope of the two head-
ings. In languages that havo bound forms, the constructions in
which bound forms play a part differ radically from the const.ruc-
tions in which ail the irnmediate constituents arc free forms. Ac-
cordingly, we place the former under the separatc heading of mor-
phology. The difficulty is this, that certain formai relations, such
as the relation between he and him, consist in the use of bound
forms, while the semantic difference bctween these forms can be
defined in terms of syntactic construction; he serves, for instance,
as an actor (he ran) and him as an undergoer (hit him). Neverthe-
less, the traditional division is justified: it merely happens that in
these cases the meanings involvcd in the morphologie construction
are definable in terms of syntax instead of being definable merely
in terms of practical life. Syntactic constructions, then, are con-
structions in which none of the immediate constituents is a bound
form. Border-line cases between morphology and syntax occur
chieRy in the sphere of compound words and phrase-words.
12. 2. The free forms (\Vords and phrases) of a language appear
in larger free forms (phrases), arranged by taxemes of modula-
tion, phonetic modification, selection, and arder. Any meaningful,
recurrent set of such taxemes i8 a syntactic construction. For
instance, the English actor-action construction appears in phrases
like these:
John ran
John feU
Bill ran
Bill feU
Our horses ran away.
184
In these examples we see taxemes of selection. The one con-
stituent (John, Bill, our horses) is a form of a large class, which we
calI nominalive expressions; a form like ran or very good could not
be used in this way. The other constituent (ran, fell, ran away) is
a fonu of another large class, which we calI fint'te verb expressions;
a fOrIn like John or very good could not he used in this way. Sec-
ondly, we see a taxeme of order; the nominative expression precedes
the finite verb expression. We need not stop here to examine the
various other types and sub-types of this construction, whieh show
different or additional taxemes. The meaning of the construction
is roughly this, that whatever is named by the substantive expres-
sion is an actor that performs the action named by the finite verb
expression. The two immediate constituents of the English actor-
action construction are not intcrchangeable: we say_that the
construction has two positions, which we may calI the positions of
actor and of action. (',ertain English words and phrases can appear
in the actor position, certain others in the action position. The
positions in which a form can appear are its functions or,
tively, its function. AlI the forms which can fi.!l a given position
thereby constitute a form-class. Thus, aIl the Englsh words and
phrases which can fill the aetor position in the actor-action con-
struction, constitutc a great form-class, and we calI them nomina-
tive expressions; similarly, ail the English words and phrases
which can fill the action position on the actor-action construction,
constitute a second grcat form-clas&, and we calI them finite verb
expressions.
12. 3. Since the constituents. of phrases are free forms, the
speaker may scparate them by means of pauses. Pauses arc mostly
non-distinctive; they occur chiefly when the constituenta are long
phrases j in English they are usually preceded by a pause-pitch.
We have scen ( 11.1) that free forms which are united by no
other construction may he united by parataxis, the mere absence
of a phonetic as in It's ten o'clock l,] l have to go
home [l In ordinary English parataxis a pause-pitch appears be-
tween the constituents, but wc have also a varlety of close para-
taxis without a pausc-pitch, as in please come or yes sir.
A special variety of parataxis is the use of semi-absolute forms,
which grammatically and in meaning duplicate sorne part of the
form with which they arc joined in parataxis, as in John, he ran
away. In French this type is regularly used in sorne kinds of
186
Sy:t\TAX SYNTAX 187
of another word. Various languages use sandhi-fonus of t.bis sort;
they are known as atonie forms. This term is not altogether ap
propriate, since the peculiarity is not always a lack of st.ress. In
the French phrase l'homme [1 'the man,' the article le [1;)] is
atonie, because its sandhi-form [1] could not be spoken aJone on
account of the phonetic pattern (Jack of a vowel). In the Polish
phrase ['do nuk] 'to the feet,' the preposition do 'ta' is at.onic
precsely because it has the stress, for the st.ress in this language is
placed on the next-to-last syllable of each word, and falls on do
only because this word is treated as part of the following word.
An atonie form which is trcated as part of the following word -
this is the case in our examples so far'- is a proclitic. An atonie
form which is treateci as if it were part of the preceding word is an
enditic; thus, in 1 saw him [aj im], the [aj] is proclitic, but the
[im] onclitic.
The sandhi which substitutes an for a, and the sandhi by which
this and other words are unstressed in phrasai combinations, are
examplcs of compulsory sandhi. Other Bnglish sandhi habits are
optional, bocause parallcled by unaltered variants, which have
usually a formai or e1evated connotation; for instance, the dropping
of [hl in hirn does Ilot take place in t.he more ekvat.ed variant 1
saw him [aj 's;) him]. Beside the sandhi-forrns in dlyou! ['dijuw?],
won't you ['wowncuw?], at ail [e'Lll] (in Arneriean with the
voiced tongue-Bip variant of [t]), we have the Illore elegant variants
['did juw? \vownt juw? et 'JI].
Sandhi-forms may be unpronounceablc when taken by them-
selves; this is the case in a numbcr of English cxamples:
questions, as Jean quand est-il venu? [zall kant et i vny?] 'John, when
did he come?'
Parenthesis is a variety of parataxis in which one form interrupts
the other; in English the parenthctic form is ordinarily precedcd
and followed by a pause-pitch: 1 saw the boy [,] 1 mean Smilh's boy
[,] running across the street [.] In a form likc Won't you please come?
the please is a close parenthesis, without pause-pitch.
The term apposition is used when paratactically joined fonns arc
gramrnatically, but not in mcaning, cquivalent, e.g. John [,] the
poor boy. When the appositional group appears in included posi-
tion one of its Illembcrs is equivalent ta a parenthesis: John [,]
the ;oor boy [,] ran awuy [.1 In English \vc have also close apposition,
without a pause-pitch, as in King John, John Brown, John the
Saptis!, Jlr. Broum, .Mount Everest.
Ofton enough non-linguistic factors interfere with const.ruction;
what. the speaker has said is nevert.heless meaningful, provided he
has already uttcrcd a fr('C form. In aposiopesis the speaker breaks
off or is l thought he -. In anacolouthon he starts
o'ler again: 11' s high time we - oh, weil, l guess if won't matier.
When a speaker h,'sitales, English and sorne other languages offer
special parenthetie hesitation-jorms, as [r] or [f:] in Afr. - ah-
Sniffen or 1I-1r. -what Y011 may cali him - Sm:iJen or that - thing-
amajig _..- transmdter.
12. 4. Features of modulation and of phonetie modification play
a great part in many syntactic constructions; they arc known as
sandhi. 1 The form of a word or phrase as it. is spoken aJone is Hs
absolute form; the forrris whieh appear in included positions are
its sandhi-jorms. Thus, in English, the absolllte form of the in
definitc article is a ['ej]. This form appears in included position
only when the article is an emphatie element and the next ward
begins with a consonant., as in "not a house, but the house." If the
next wonl begins with a 'lowel, we have instead a sandhi-form, an
['en], as in "not an uncle, but. her uncle."
A feature of modulation appears in the fact t.hat when a, an
is not an cmphatic it is spoken as an unstressed syllablc,
as in a house [e 'haws], an arm [en 'arm]. In Enp;lish, a ward in
absolute form has one high stress; hC11ce wc may say that. in a
sandhi-form without. high a word is spoken as if it \Vere part
l This term, like many tcchnical terms of lingllistics, cornes from the ancient
Hindli grammaria.ns. Li terally, it means puttin.-: togclher.
ABBOLt:"TE FORM
is ['iz]
has ['hez]
am ['em]
are ['ar]
have ['hev]
had ['hed]
would ['wud]
will ['will
them ["oem]

[z ] John's ready.
[s ] Dick's ready.
[z ] John's got il.
[ml l'm ready.
[r ] TVe're waiting.
fv] l've got it.
[d ] lIe'd seen it.
[d l TIe'd see it.
[1 ] l'Il go.
n ] That'l[ do.
[111] Watch 'em.
188
SYNTAX
SYNTAX 189
12. 5, Our examples so far illustrate special or irregular cases
of sandhi, peculiar to certain fonns and constructions. General
The French language has a great deal of sandhi. Thus, the
article la [la] 'the' (feminine) loses the [a] before a vowel or diph-
thong: la femme [la fam] 'the woman,' but l'encre [1 onkr] 'the ink,'
l'oie [1 wa] 'the goose.' The adjective ce [S;)] 'this' (masculine) adds
ft] before the sume sounds: ce couteau kuto] 'this knife,' but
cet homme [sat om] , this man.' A plural pronoun adds [z] before the
initial vowel of a verb: vous faites [vu fet] 'you make,' but tes
[vuz f::tJ 'you are.' A plural nouD-modifier behaves similarly:
lesfemmes [le fam] 'the wOlllen,' bulles hommes [lez ;:lm] 'the men.'
A or verb adds [z], f1 third-perso
n
verb
ft], bofore certain initial vowels: va [va] 'go thou,' but t'as-y [vaz iJ
'go thou there'; elle est [ri r J 'she is,' but [f t rI?J 'is she?'
A few masculine adjectives add sandhi-consonants before a vowe1:
un grand garon [
n
gra" garso
n
] 'a big boy,' but un gmnd homme
[" grunt :lm] 'a grcat man.'
In languages with distinctions of pitch in the wm'd, modifications
of piteh may play a part in sandhi. Thus, in Chinese, bcside the
absolute form ['il] 'one,' there arc the sandhi-forms in [,i
4
phF
'ma3J 'one horse' and Wko '.hm2] 'one man.'
Sandhi-modification of initial phonemes is less cornmon than
that of the end of a word; it ODCurS in the Celtie languages, as, in
modern Irish:
AlISOLUTll FOIl'"
not ['nat]
and ['end]
AnSOLU'I'I' FOIl'"
['bo:] 'cow'
['uv] 'egg'
['ba:n] 'white'
['bog] 'soft'
['bris] 'break'
SA"'DHI-FOIl'''
[1ft] It isn't.
[nt] 1 won't.
lt ] 1 can't.
[If ] bread and butter,
SA"'Dlll-FORM
[an 'vo:] 'the cow'
far 'mo:] 'our cow'
[an 'tuv] 'the egg'
[na 'nuv] 'of the eggs'
la 'huv] 'her cgg'
['bo: 'va:o] 'whitc cow'
['ro: 'vogJ 'very soft'
[do 'vriS] 'did break.'
or regular sandhi appHes to any and aIl words in a short (close-
knit) phrase. In sorne forms of English, such as New England
and southern British, words which in absolutc position have a
final vowel, add [r] before an initial vowel: water ['W0t::l] but the
water is ['l5c 'w0tar iZ]j idea [aj'dija] but the idea is ['oij aj'dij;)r
izJ. When three consonants come together in French, the word-
final adds [a]; thus, porte [pJrtJ 'carries' and bien [bjl;n] 'weIl'
appear in the phrase as porte bien [p0rta bjen] 'carries weil.' A
word Wh08C first syllable in absolute form cantains [;)], either be-
cause the word has no other syllabic or because otherwise it
would begin with an unpermitted c1uster ( 8.6), loses this [a]
in the phrase whenever no unpermitted group would ref:mlt: le
[laJ 'the' but l' hornme [1 <lm] 'the man'; cheval [8;)val] 'horse,'
but un cheval [
n
sval] 'a horse'; ie [za] '1,' ne [na] 'not,' le [la]
'it,' demande [d;)mand] 'ask,' but je ne le demande pas [b n 1;)
dmand pa] '1 don't ask it' and si ie ne le demande pas [si z n'" 1
demand pa] 'if l don't ask it.'
In Sanskrit there is a great deal of general sandhi; for instance,
final [ah] of the absolutc form appears in the following sandhi-
variants: absolute [de:'vahJ 'a god,' sandhi-forms: lde:'vas 'tatra]
'the god there,' [de:'va arati] 'the god wanders,' [de:'va e:ti]
'the god goes,' [de:'vo: dada:ti] 'the god !!;ives,' and, with change
also of a following initial, before ['atra] 'here,' [de:'vo: traJ 'the
god here.' Certain words, however, behave differently; thus,
['punah] 'again' gives ['punar dada:ti] 'again he gives,' ['punar
'atra] 'again here.' The divergent words may be marked off by
some structural feature. Thus, in sorne Dutch pronunciations thc
ahsolutc forms heb ['hep] 'have' and stop [stop] 'stop' behave differ-
ently in sandhi: heb ik? ['heb ek?] 'have I?' but stop ik? ['stop ek?]
'do 1stop?' The forms which have the voiced consonant in sandhi
have it also whenever it is not at the end of the word, as hebben
['hebe] 'to have,' in contrast with stoppen ['stope] 'to stop.'
Sandhi-distinctions based on morphologic features like this, may
he called reminiscent sandhi.
Sandhi may go so far as to restrict the word-final in a phrase
bcyond the ordinary medial restrictions of a language. Thus, the
sequence [ta] is permitted medially in Sanskrit, as in ['patati] 'he
falls,' but [t] at the end of the ward is in c1ose-knt phrases rc-
placed by [dl before a vowel: absolute ['tat] 'that,' but ['tud asti]
'that is.'
12. 6. Taxemes of selection play a large part in the syntax of
most languages; syntax consists largely in defining them - in
stating, for instance, under what circumstances (with what ac-
compan:ying forms or, if the accompanying forms are the same,
with what difference of meaning) various form-classes (as, say,
indicative and subjunctive verbs, or dative and accusative nouns,
and so on) appear in syntactic constructions. 'Ve have seen that
the selective taxemes delimit form-classes. These classes are
most numerous in the languages that use most taxemes of
selection. The syntact.ic constructions of a language mark off
large classes of free forms, such as, in English, the nomina
R
tive expression or the finite verb expression. Since different lan-
guages havc different constructions, their form-classes also are
different. 'Ve shaH see that the great form-classes of a language
are most easily describcd in terms of ward-classes (such as the
traditional "parts of speech "), because the form-class of a phrase
is usually determined by one or more of the words which ap-
pear in it.
In languages which rnake a wide use of selective taxemes, the
large form-classes are subdivided int.o smaner ones. For instance,
the English aet.or-action construction, in addition to the general
selective taxemes, shows some more specializcd taxemes of the
same sort. \Vith the nominative expressions John or thai horse
we can join the finite verb expression runs but not the finite
verb expression run fast; with the nominative expressions John
and Bill or horses the reverse selection is made. Accordingly,
we recognize in eaeh of these two form-classes a division into t.wo
sub-classes, which we call singular and plural, sueh that. Il singular
nominative expression is joined only with a singular finit.e verb
expression, and a plural nominative expression only ,vith a plural
finite verb expression. It would not do to define these sub-elasses
by meaning - wit.ness cases like 11)hcat grou:s but oats grow. Fur-
ther examination shows us several varieties of selection: (1) many
finite verb expressions, sueh as can, had, weni, appear with any
actor; (2) many, sueh as run : show the twofold selection
just described; (3) one, was : were, shows a twofold selection that
does not agree with the preccding; (4) one, fina!ly, am : is : are,
shows a thrcefold selection, with a special fonu that accompanies
the actor l, precisely the actor form as to which (2) and (3) dis-
agree:
190 SYNTAX
SYNTAX
191
(1) (2) (3) (4)
A l can [ run [ was [am
B the boy can the boy runs the boy was the boy is
C the boys can the boys run the boys were the boys are
A=B=C A=C A=B
Thus we find among nominative expressions and among finite
vel'b expressions a thrcefold subdivision, due to taxemes of se-
lection; among nominative expressions sub-class A cantains only
the form [; sub-class B contains those which are joined with finite
verb expressions sueh as runs, was, is, and sub-class C contains
those which are joined with finite verb cxpressions such as run,
were, are. In fact, we can base our definition of the thrcc suh-
classes on the selection of the three finite verb forills am : : are.
Conversely, we define the suh-classes of fini te verb expressions by
telling with which nominative expressions (say, [ : the boy : the
boys) they oceur.
The narrower type of selection in cases like this one is in prn-
ciple no different from the more inclusive type by which our
language distinguishes great form-classes like nominative expres-
sions and finite verb expressions, but there are SOrne differences
of detai!. The narrower type of selection, by whieh great form-
classes are subdivided into selective types, is caned agreement.
In a rough way, without real boundaries, we can distinguish three
general types of agreement.
12.7. In our example, the agreement is of the simplest kind,
which is usually caned comord or congrueme: if the actor is a form
of A, the action must he a form of sub-class A, and so
on. Sometimes one of the subdivisions is otherwise also rccog.
nized in the structure of the language; thus, in our exampIe, classes
Band C of nominative expressions are otherwise also definable
in our language; nameIy, by the use of the modifiers this, lhat with
class B, but these, those with class C: wc say this boy, this wheat,
but these boys, these oats. Accordingly, we view the subdivision of
nominative expressions into singulars and plurals as more funda-
mental than that of finite verb expressions, and say that the lat-
ter agree with or stand in congruence with the former. For the
same reason, wc say that the forms this, thal, these, those stand
in congruence with the accompanying substantive form. Congru-
ence plays a great part in many languages; witness for example
192
SYNTAX
SYNTAX 193
the infiertion of the adjectives in most lndo-European languages
in congruen with various sub-classes (number, gender, case)
of the noun: German der Knabe [der 'kna:be] 'the boy,' ich sehe
den Knaben [0. 'ze:e den 'kna:ben] '1 see the boy,' die Knaben
[di: 'kna:benJ 'the boys,' where the selection of der, den, die
all;rcs with the sub-classes of the noun (singular and plural,
nominative and accusative); in das Haus [das 'haws] 'the house,'
the form das, as opposed ta der, is selected in agreement with the
so-callcd gender-classes into which German nouns are divided.
These genders arc arbitrary classes, each of which demands
differcnt congruence-forms in certain kinds of accompanying
words. German has three ll;ender-classes; for each of these 1 ll;ive
phrases showing the congrucnce of the definite article and of the
adjective kalt 'cold':
"masculine gender": der Hut [dcr 'hu :tJ 'the hat,' kalter Wein
[,kalter 'vajnJ 'cold winc'
"femininc gender": die Uhr [di: 'u:r] 'the clock' kalte Milch
[,kalte 'milx] 'cold milk'
"neuter gender"; das Haus [das 'hawsJ 'the house,' kaltes Wasser
[Ikaltes 'vaser] 'cold water.'
French has twa genders, "masculine," le couleau [18 kutoJ
'the knife,' and "feminine," la fourchette [la fursetJ 'the fork.'
Some languages of the Bantu family distinguish as many as twenty
gender-classes of nouns.
12. 8. In other cases the subsidiary taxeme of selection has to
do with the syntactic position of the form. For instance, we say
l know but walch me, beside me. The ch01ee between the forms l
(he, she, they, wc) and me (him, her, them, us) depends upon the
position of the form: the 1-class appears in the position of actor,
the me-class in the position of goal in the action-goal construction
(walch me) and in the position of axis in the relation-axis construc-
tion (beside me). This type of selection is called government; the
accompanying form (know, walch, beside) is said ta govern (or tu
demand or ta take) the selected form (1 or me). Government, like
congruence, plays a great part in many lanll;uages, including many
of the lndo-European family. Thus, in Latin, different verbs
govern different case-forms in the substantive goal: videt bovem
'he sees the ox,' nocct bovI 'he harms the ux,' uti/ur bove 'he uses
the ox,' meminit bovis 'he remembers the ox.' Similarly, different
main clauses may govern diffcrcnt forms of subordinate verbs,
as in French ie pense qu'il vient [:7.8 puns k vjen] '1 tbink he is com-
ng,' but ie ne pense pas qu-il vienne [za n puns pu k vjen] '1 don't
tbink he s coming.'
Identity and non-identity of objects are in many languages
distinguished by selective features akin ta government. In English
we say he washed him when actor and goal are not identical, but
he washed himself (a reflexive form) when they are the same person.
Swedish thus distinguishes betwecn identical and non-identical
actor and possessor: han log sin haU [han 'to:g si:n 'hatJ 'he took
his (owo) hat' and han log hans hait [hans 'hat] 'his (someone else's)
hat.' The Algonquian languages use different forms for non-
identical animate third persons in a context. In Cree, if wc speak
of a man and then, secondarily, of another man, wc ~ e n t o n the
first one as ['na:pe:w] 'man,' and the second one, in the socalled
obvialive form, as ['na:pe:wa]. Thus, the language distinb'1lshes
between the fol1owing cases, where we designate the principal
persan as A and the other (the obviative) as B;
['utinam u'tastutin] 'he (A) took his (A's) hat'
['utinam utastu 'tinijiw] 'he (A) took his (B'il) hat'
[utina'mijiwa u'tastutin] 'he (B) took his (A's) hat'
[utina'mijiwa utastu'tinijiw] 'he (B) took his (B's) hat.'
12. 9. In the third type of agreement, cross-reference, the sub-
classes contain an aetual mention of the forms with whch they are
joincd. This mention is in the shape of a substitute-form, resem-
bling our pronouns. In non-standard English this occurs in sucb
forms as John his knife or John he ran away; here the form his knife
actually mentions a male possessor, who is more explieit.ly men-
tioned in the aecompanying semi-absolut.c form John; similarIy,
thc he in he ran away mentions the actar John - contrast .jlfary
lier knife and Mary she ran away. In French, cross-reference occurs
in the standard language especially in certain types of questions,
such as Jean o est-il? [zun u et i?J 'John where is he?' that is,
',"Vhere is John?' CS 12.3). A Latin tinite verb, such as cantat
'he (she, it) sings,' includes substitutive mention of an actor. It
is joined in cross-referencc with a substantive expression that
makcs specifie mention of t.he act.Qr, as in puella canlal '(the) girl
she-sings.' In IDfiny languages verb-forms include substitutive
(pronominal) mention of both an actor and an undergoer, as, in
Cree ['wa:pame:w] 'he saw him or her'; accordingly, more specifie
194 SYNTAX
8YNTAX 195
mention of both actor and undergoer is in cross-reference ['wa:-
pame:w 'atimwa a'wa na:pc:w] 'he-saw-nim (obviative) a-dog
(obviative) that man'; that is, 'the man saw a dog.' Similarly, in
many languages, a possessed noun includes pronominal mention of
a as, in Cree, ['astutinJ 'hat,' but [ni'tastutin] 'my hat,'
[ki'tastutin] 'thy hat,' [u'tastutin] 'his, her, Hs hat'; hence, when
the possessor is mentioned in another word or phrase, we have
cross-reference, as in ['ca:n u'tastutin] 'John his-hat,' i.e. 'John's
hat.'
12. 10. Every syntactic construction shows us two (or some-
times more) free forms eombined in a phrase, which we may call
the resullant phrase. The resultant phrase may belong to a form-
class other than that of any constituent. For instance, John ran
is neither a nominative expression (like John) nor a finitc verb
expression (like ran). Therefore we say that the English actor-
action construction is exocentric: the resultant phrase belongs ta the
form-class of no immediate constituent. On the other hand, the
resultant phrase may bclong ta the same form-class as onc (or more)
of the constituents. For instance, poor John is a proper-noun
expression, and so is the constituent John; the forms John and
poor John have, on the whole, the same functions. Accordingly,
we say that the Eng:1ish character-substance construction (as in
poor John, fresh milk, and the like) is an endocentric construction.
The exocentric constructions in any language arc few. In
English we have, beside the actor-aetion construction, also that of
relation-axis, as beside John, with me, in the house, by running
away; the constituents are a prepositional expression and an ac-
cusative expression, but the resultant phrase has a function differ-
ent from either of these, appearing in entirely different syntactic
positions (e.g. as a modifier of verbs: sit beside John, or of nouns:
the boy beside John). Another exoccntrie construction of English is
that of subordination. The const.ituents in one type (clause-
subordination) are a subordinating expression and an actor-action
phrase, as in if John ran away; the result.ant phrase has the function
of neither constituent, but serves as a modifier (subordinate clause).
In the other type (phrase-subordination) the constituent.s are a
subordinating expression and any other form, especially a sub-
stantive: as l, than John, and the resultant phrase has the function
of a modifier (as big as l, bigger than John). Although the resultant
phrase in an exocent.ric construction has a function different from
the function of any constituent, yet one of theoo constituents is
usually peeuliar to the construction and serves t.Q characterize the
resultant phrase; thus, in English, fini te verbs, prepositions, and
subordinating conjunctions regularly appear in the exocentric
constructions just illustrated, and suffice ta characterize them.
Endocentric constructions are of two kinds, co-ordinative (or
serial) and subordinalive (or attributive). In the former type the
resultant phrase bclongs to the same forrn-class as two or more of
the constituents. Thus, the phrase boys and girls belongs t.o the
same form-class as the constit.uents, boys, girls; these eonstit.uents
are the members of the co-ordination, and the other constit.uent is
the co-ordinator. Sometimes there is no co-ordinator: books, papers,
pens, pencils, blotters (were alllying ... ); sometimes there is one
for each member, as in both Bill and John, either Bill or John.
There may be minor differences of form-class between the resultant
phrase and the members; thus Bill and John is plural, while the
members are each singular.
In subordinative endoccnt.ric constructions, the resultant phr<'lse
belongs to the same form-class as one of t.he constit.uents, which
we caH t.he head: thus, poor John belongs to the same form-class as
John, which wc accordingly call the head; the other member, in
our example poor, is the attribute. The attribute may in turn be a
subordinative phrase: in very fresh milk the irnmediate constituent.s
are the head milk, and the attribute very fresh, and this phrase, in
turn, consists of the headfresh and the attribute very. In this way
there ean be several ranks of subordinative position; in very fresh
milk there are three: (1) milk, (2) fresh, (3) very. In the same way,
the head al80 may show an attributive construction: the phrase
this fresh mille consist.s of t.he at t.ri bute this and the head fresh milk,
and this, in t urn, of the attri bute fresh and the head milk.
12. 11. If ail the synt.actic constructions which go tQ make up a
phrase are cndoccntric, then the phrase will contain among it.s
ultirnate constituents sorne word (or several words, rnembers of a
co-ordination) whose form-class is the same as that of the phrase.
This ward is the center of the phrase. In the phrase ail this fresh
milk, the word milk is the center, and in the phrase ail this fresh
bread and sweet butter, the words bread and builer are the centers.
Since most of the constructions in allY language are endocentric,
most phrases have a center: the form-cluss of a phrase is usually
the same as that of sorne word tho.t is contained in the phrase.
196 SYNTAX SYNTAX 197
The exceptions are phrases of exocentric construction, and these,
too, we have seen, are definable in terros of word-classes. The
syntactic of phrases, therefore, can he derived From
the syntactic of words: the form-classes of syntax are
most easily describcd in terms of ward-classes. Thus, in English, a
substantive expression is either a word (such as John) which
belongs to this form-class (a substantive), or else a phrase (such as
pOOT John) whose center is a substantive; and an English fioite
verb expression is either a word (such as Tan) which belongs to this
form-class (afinite veTb), or eIse a phrase (sueh as Tan away) whose
center is a finite verb. An English aetor-action phrase (sueh as
John Tan or pOOT John ran away) does not share the farm-class of
any word, since its construction is exocentric, but the form-class of
actor-action phrases is defined by their construction: they consist
of a nominative expresf\ion and a finite verb expression (arranged
in li certain way), and this, in the end, again reduces the matter to
terrnf\ of ward-classes.
The term parts of speech is traditionally applied ta the most
inclusive and fundamental of a language, and then,
in accordance with the principle juilt stated, the syntactic form-
cIasses are def\crbcd in terms of the parts of speech that appear in
them. However, it s impossible ta set up a fully consistent scheme
of parts of speech, because the overlap and cross each
othcr.
In speaking of we use the term expresion to include
bath words and phrascs: thus John is a substantive, poor John a
subslantive phrase, and hoth forms are subslantive expressions.
Within the great far rn-classes which contain both words and
(thunks to endocentric constructions) a vast number of phrasaI
combinations, there may he duc to small differcnces of
phrasai construction. For instance, whcn an attribute lke fresh,
good, or sweet is joincd t' the head milk, as in fresh milk, tbis
resultant phrase is still capable of joining with other attributes, as/
in good, sweet,fresh milk: the phrase has entirely the same functions
as its center (and head), namely the word milk. If, however, we
join a form like mk or fTesh milk with the attribute this, the
resultant phrase, IMs milk or this fTesh milk has not quite the same
function as the head or center, since the resultant phrase cannot he
joined with attributes like good, sweet: the construction in this
milk, this fresh mk is partially closed. The possibilities in this
direction, in fact, are limited ta adding the attribute all, as in all
this milk or aU this fresh milk. When the attribute aU has been
added, the construction i8 cwsed: no more attributes of this type
(adjectives) can he added.
12. 12. An example of a taxeme of order is the arrang' .ment by
which the actor form precedes the action form in the normal type
of the Engli sh actor-action construction: John Tan. In lang'Jages
which use highly complex taxemes of selection, order is largely
non-distinctive and connotative; in a Latin phrase such as pater
amal jlium 'the father loves the son,' the syntacti c relations are ail
selective (cross-reference and government) and the words appear
in al! possible orders (paler f'ilium amal, faium paler amat, and so
on), with differenees only of cmph81ls and liveliness. In English,
taxemes of order appear in the diffcrcnce hctween actor-action and
action-goal, as in John Tan and catch John; the difference hetween
John hit Bill and Bill hit John rcsts entirely upon order. In general,
however, taxemes of arder in English occur along with taxemes of
selection. Languages which in this respect and in the general
configuration of their syntax resemble English, may still show great
differences as to taxemes of order. Thus, standard German differs
from English in allowing only one attribute (word or phrase) of the
verh to precede a finite verb: heute spielen wir Bail ['hojte 'spi:len
vi:r 'bal] 'today play we bail.' Further, it places several elements
last in the sentence: certain adverbs, as ich stehe um sieben UhT
auf [ix 'ste:e um 'zi:hen 'u:r 'awf] 'I get at seven o'clock up';
participles, as ich habe ihn heute gesehen [ix Iha:be i:n 'hojte
ge'ze:nJ 'I have him today sccn'; infinitives, as ich werde ihn heule
sehen [ix Iverde i:n 'hojte 'zc:n] '1 shaH him today see'; the verb
of a dependent clause: wenn ich ihn heute sehe [ven ix i:n 'hojte
'ze:e] 'if l him today see.'
French has a complicated and rigid system of ordering certain
suhstitute (" conjunct") accompaniments of its verbs. In the
ordinary (non-interrogative) sentence-type, it distinguishes seven
positions of thesc clements, which precede the finite verb:
(1) actors, such as je 'l,' il [il] 'he, it,' [il] 'they,' on
[on] 'one,' ce [sa] 'it, that'
(2) the negative adverb ne [na] 'not'
(3) farther goals of first and second persons, such as me [ma]
'to me,' vous [vu] 't' you,' and of the rcficxivc se [sa] 'to himself,
herself, themselves'
198 SYNTAX SYNTAX 199
(4) nearer goals, such as me [ma] 'me,' vous {vu] 'you,' se [sa]
'hirnself, herself, themselves,' le [la] 'him, it,' ws [1e] 'them'
(5) farther goals of the third person: lui [hJi] 'to him, to her,'
kur [l:r] 'to thern'
(6) the adverb y [il 'there, thi ther, to i t, to them'
(7) the adverb en [aD] 'from there, of it, of them.'
For example: (1-2-3-4) il ne me le donne pas [i n ma l dJD pa]
'he does Dot give it to me'
(1-3--6--7) il m'yen donne [i m j aD dJn] 'he gives me sorne of it
there'
(1-4-5) on le lui donne [aD IJ lqi dJn] 'one gives it ta him'
(1-2--6--7) il n'yen a pas li n j ann a pa] 'there aren't any,'
titerally 'it has not of them there.'
Occasionally arder serves finer distinctions. In French most
adjectives follow their nouns: une maison blanche [yn mezo
D
biaDs]
'a white house'; a certain few precede; une beUe maison [yn bel
mezo
D
] 'a pretty house'; others precede anly with transferred
meanings or with emphatic or intense connotations: une barbe
noire [yn barba nwa:r] 'a black beard': une noire trahison [yn
llwa:r t.raizo
D
] 'a black betrayal'; un livre excellent [
D
li:vr
eksda
D
] 'an excellent book'; un excellent livre 'a .splendid book!'
A few show greater differences of meaning: un livre cher [
n
li:vra
se:r] 'a costly book': un cher ami [
n
se:r ami] 'a dear friend,'
sa propre main [sa prJprJ meD] 'his own hand': une main propre
[yn meD prJpr] 'a clean hand.'
Viewed from the standpoint of economy, taxemes of order are
a gain, since the forms are bound t.o be spoken in sorne succession;
nevertheless, few languages allow feat.ures of order to work alone:
almost always they merely supplement taxemes of selection.
12. 13. The languages of the Indo-European famUy are peculiar
in having many part.s of speech; no matter upon what construc-
tions we base our scheme, a language like English will show at
least half a dozen parts of speech, such as substantive, verb, ad-
jective, adverb, preposition, co-ordinating conjunction, and subor-
dinating conjunction, in addition to interjections. Most languages
show a smaller number. A distribution into thr types is quite
frequent (Semitic, Algonquian); usually one resembles our substan-
tives and one our verbs. It is a mistake to suppose that our part,-.of-
speech system reprcsents universal features of human expression.
If such classes as object.s, actions, and qualities exist apart frorn
our language, as realities either of physics or of human psychology,
then, of course, they exist aIl over the world, but it would still
he true that many languages lack corresponding parts of speech.
In languages wit.h few parts of speech, the syntactic fonu-dasses
appear rather in phrases. Often the class of a phrase is indicated
by sorne special word, a marker; strictly speaking, the marker and
the form which it accompanies are joined in an exocentric con-
struction which determines the class of the phrase. Aside from
this selective feature, the constructions are likely to be distin-
guished by word-order.
The classical instance is Chinese. The parts of speech are full
words and particles (that is, markers). The principal constructions
are three.
(1) The favorite sentence-construction is one of subject and
predicate, mnch like the English actor-action construction; the
subject precedes the predicat.e: [tha
1
'xaw
3
] 'he is good,' [tha
l
'laj 2] 'he came.' In certain cases, depending on differences of form-
class, the predicate is marked by the particle [sa
4
] at its begin-
ning: [tha
1
S
4
'xaw
3
Izan
2
] 'he (p.) good man,' that is, 'he is a good
man.'
(2) There is an endocentric construction in which the attribute
precedes the head; in meaning this resembles the similar English
constructions: ['xaw
3
Izan
2
] 'good man,' ('man
4
,chy4] 'slowly go,'
that is, 'go slowly.' The attribute is in certain cases marked by
the particle (tP] at its end: ['tilJ3 Ixaw
3
tF 'zan
2
] 'very good man';
[IW0
8
t.F 'fu
4
chin1] '1 (p.) father,' that is, 'my father'; ['tso
4
coZ
tP 1zan2] 'sit (p.) person,' that is, 'a sitting persan'; [' wo
3
'je
8
Itsa
4
ti 'pP] '1 write (p.) brush,' that is, 'the brush l write with'
~ in this example t.he attribute is a phrase of subject-predicate
construction; ['maF ti 'SUl] 'buy (p.) book,' that is, 'the purchased
book.'
(3) A second endocentric construction, in which the attribute
follows the head, rosembles rather the English action-goal and
relation-axis constructions: Lkwan
l
'man
2
] 'shut the door,' [ltsaj4
'CUlJI kwo] 'in China.' We may caU this, somewhat inexactly,
the action-goal construction, t.o distinguish it from (2).
Taxemes of selection consist largely in the marking off of a
formclass which serves as subject in (1), as head in (2), and as
goal in (3), resembling t.he Englsh substantive expression. To
this form-cluBs (we may call it the abject expression) anly a few
200 SYNTAX SYNTAX 201
words may he Baid to helong in their own right; theoo are substitute-
words of the type [tha1] 'he, she' or [w0
3
] 'l.' The other objeet
expressions are phrases with various markers. 'l'he commonest
of these markers are certain particles which precede as attributes
of type (2), such as [ca
4
] 'this,' [na
4
] 'that,' [na
3
] 'which?' Thus,
['ca
4
k0
4
] 'this piece,' that is, 'this (thing).' In most instances
these markers do not immediately join with a full word; but only
with certain ones, like the [k0
4
] 'piece' in the last example, which
hereby constitute a form-class of numeralives; the phrl!OO of mar-
ker plus numerative joins the ordinary full word in const.ruction
(2), as: [ca ko 'zan2] 'tbis (individual) man'; [1 wu3 Ilj al] 4 'cha1]
'live (individual) cart;' that is, 'live carts.' Another kiod of object
expression is charaeterized by the particle [ti
l
] at its end: [lmaj4
'SUl ti] 'sell book (p.),' that is 'bookseller.'
In this way complex phrases are built up: [tha
l
'taw
4
'thjen
2
li
3
'Chy4] 'he eoter field interior go,' that is, 'he goes into the field';
here the first word is the subject, the rest of the phrase the predi-
cate; in this predicate the last word is the head and the other three
are an aHribute; this attribute consists of the action [taw4J 'enter'
and the goal ['thjen
2
Ii"] 'field interior,' in which the first ward is an
attribute of the second. In the sentence [nP 'meF pa
3
'maj3
'mep ti Ichjen
3
'kep WO"] 'you not take buy coal (p.) money give 1,'
the first ward is the subject, the rest the predieate; this predicate
consists of an attribute, [mep] 'not' and a head; within this head,
the first five words are again an attribute and the last two ['kcj3
w0
3
] 'give l'a head, whose construction is action and goal. In the
five-word at tri bute [pa 3 1maF 'meF ti 1chjenZj 'take buy coal
(p.) money,' the first word is an action and the rest a goal; this
goal consists of the head [hjen
3
] 'money' and the attribute [Imap
'mep ti] which is marked as such by the particle [tP] appended
to the phrase [Imaj" 'mep] 'buy coal,' whose construction is action-
goal. Thus the sentence means 'you not taking buy-coal-money
give me,' that is 'you haven't given me money to buy coa1.'
In Tagalog, the parts of speech arc, again, full ward and particle,
but here the full words are subdivided into two classes which we
may calI static and transient. The latter resemble our verbs in
forming a special kind of predicate (the narrative type, with four
sub-types, 11.2) and in showing morphologie distinctions of
tense and mode, but they differ from our verbs becausc, on the one
hand, they are not restricted to the function of predcate and, on
the othcr hand, there exist non-narrative predicates. The chief
constructions are subject and predicate, marked optionally by
arder (predicate precedes subject) or by the particle [aj] and arder
(subject precedes predicate marked by initial [am, as illustrated
in 11.2. The subject and the equational predicate are selectively
marked: the class of forms which fill thcse positions resembles the
English substantive expression and, even more, the Chinese abject
expression. A few substitute-words, such as [a'ku] 'l'and [si'jaJ
'he, she,' helong to this class by their own right; all other abject
expressions are phrases, characterized by the presence of certain
attributes, as tisa 1) 'ba:tan 'one child,' or by certain particles,
chielly [si] before names, as [si 'hwan] 'John,' an far)] bofore other
forms, as [alJ 'ba:tw'] 'the child, a child,' [al] pu'la] 'the red,' that
is, 'the redness,' [al] 'pu:tul] 'the eut,' or, to illu8trate traosient
forms, [al) pu'mu:tul] 'the one who eut,' [alJ pi'nu:tul] 'that which
wascut,' [alJ ipi'nu:tul] 'that which was eut with,' [al) pinu'tuJan]
'that which was cut from.' There are four attributive construc-
tions. In ono, a particle [na], after vowels [l]], int.ervencs bet.wccn
head and attributc, in either order, as [al) 'ba:ta 1) sumu: 'su :lat]
or [ar] surnu:'suJat na 'ba:ta?] 'the writing child'; [al] pu'la l]
pun'ju]' tho rcd handkcrchief,' [aIJ pan'ju l] i'tu] 'this handkerchief.'
Anothcr, morc restricted attributive construction lacks the particle,
us [hin'di: a'ku] 'not l,' [hin'di: maba'it] foot good.' In the third
attributive construction the attributc s an abject expression in a
special form: thus, [a'ku] 'l' i8 rcplaced by [ku], and [si'ja] 'he,
she' by [ni 'ja], and the particle [si] by [ni], the particle [alJ] by
[nalJ]: [al) pu' la nal) pan' ju aj mati 1)' kad] 'the rod of the handker-
chief is bright'; [al] 'ba: ta j ku' rna:in nal] ,ka:nin] 'the chHd ate
(sorne) rice,' (actor-action); [ki 'na :in nal] 'ba :ta? al] 'ka:nin]
'the rice was eaten by the child' (goal-action); see a1so the examples
in 11.2. In the fourth attributive construction, too, the attri-
bute is an abject expression: [si] is replaced by [kaj] and [al)] by
[sa]; the attribute tells of a place: [al] 'ba:ta j na'na:ug sa 'ba:haj]
'the child came out of the house, out of a house.'
12. 14. The details of syntax are often complicated and hard to
describe. On this point, any fairly complete grammar of a language
like English, German, Latin, or French, will prove more enlighten-
ing than would an abstract discussion. Syntax is obscured, how-
ever, in most treatises, by the use of philosophical instad of
formaI definitioos of constructions and fornH:lasscs. As a single
202
SYNTAX
SYNTAX 203
illustration of the more complex syntactic habits, we shall survey
the main features of one construction in present-day (colloquial
standard) English - the construction which we may calI eharacter-
substance, as in fresh mille.
This construction is attributive, and the head is always a noun-
expression - that is, a noun or an endocentric phrase with a noun
as center. The noun is a word-class; like ail form-classcs, it is ta he
defined in tcrms of grammatical features, some of which, in fact,
appear in what follows. When it has been defined, it shows a class-
meaning which can be roughly stated as 'abject of sueh and such
a species'; examples are boy, si one, water, kindne88. The attribute
in our construction is a!ways an adjective expression - that is an
adjective or an endoccntric phrase with an adjective as center. The
adjective is in English a word-class (part of speech), definable
preciscly by its function in the character-substance construction
which wc are now to discuss; its class-meaning will emcrge from
our discussion as something like 'character of specimens of a species
of abjects' ; examples are big, red, this, some. Reside these features
of selection, the character-substanee construction contains a fea-
ture of arder: the adjective expression precedes the noun expres-
sion: poor John, fresh mille.
The adjectives are divided inta two classes, descriptive and
limiling, by the circumstance that when adjectives of bath these
classes Occur in a phrase, the limiting; adjective precedes and
modifies the group of descriptive adjcctive plus noun. Thus, in a
form like this fresh milk, the immediate constituents are the Umi ting
adjective this, and the noun phrase fresh milk, which consists, in
turn, of the descriptive adjective fresh and the noun milk. This
difference subdivides Our character-substance construction into
two sub-types, the qualily-subslance construction, where the at-
tribute is a descriptive adjective expression, and the limitation-
substance construction, where the attribute is a limiting; adjective.
The quality-substance construction and the form-class of de-
scriptive adjectives are both dvided iuto severnl types by features
of order. For instance, wc say big black sheep and never *black big
sheep, kind old man and never *old kind man, and sa on. We shall
not stop ta examine thcse sub-types. The meaning of the form-
class of descriptive adjectives is roughly 'qualitative character of
specimens. '
The form-class of limiting adjectives ie much smaJler than that
of descriptive adjectives, and constitutes, in fact, what wc shall
later define as an irreguLar form-class ~ that is, a form-class \"hich
has ta he described in the shape of a list of the forms; however, the
boundary between lirnting and descriptive adjectives is not c o m ~
pletely definable. The class-meauing of lirniting adjectives will
appear from the following discussion as something like 'variable
character of specimens.'
Our limitinp; adjecti.ves fail into t wo sub-classes of determiners
and numeratives. These two classes have severaI subdivisions and
are crossed, ffiOfCover, by several other lines of classification.
The determiners arc defined by the fact that certain types of
noun expressions (such as house or big houBe) are always accom-
panied by a dcterminer (as, thiB house, a big house). The class-
meaning is, roughly, 'identificational charactcr of specimens.' This
habit of using <.:ertain noun expressions always with a determiner,
is peculiar ta some lanll;uages, such as the modern Germanie and
Romance. Many languages have not this habit; in Latin, for
instance, domUB 'house' requires no attrihute and is used indiffer-
ently where wc say the house or a house.
Anumber of features subdivides the detcrminers into two classes,
definite and indefinite. Of these features, we shall mention only
One: a definitc determiner can be preceded by the numerative aU
(as n aU the water) but an indefinite determiner (as, some in some
water) cannot.
The definite detcrminers are: any possessive adjective (John's
book, my house) and the words this (these) , that (those), the. The
class of possessive adjectives is definable in terms of morphology.
lt is worth observing that ltalian, which has a eharacter-substance
construction much like ours, does not use possessive adjectives as
determiners: il mio amico [il mio a'miko] 'the my friend' (that is,
'my friend ') contrasts with un [unI min arnica' a my frieod' (that
lS, 'a friend of mine'). The class-meaning of dcfinite detcrminers
is 'identiji.ed specimens.' A precise statemcnt of how the specimens
are identified, is a practical matter outsidc the linguist's control;
the identification consists in possession by some persan (John's
book), spatial relation ta the speaker (this hause), description by
sorne accompanying linguistic form (the house 1 saw) , or purely
situational features (Ihe sky, the chairman) , among which earlier
mention by speech is ta be reckoned (" l saw a man, but the man
did not sec me"), Among the dcfinite dcterminers, this: these
204
SYNTAX
SYNTAX 205
and that : those are peculiar in showing congruence with the nUffi-
ber-class of the noun (this house : houses).
The indefinite determiners are a (an), any, each, either, every,
neither, no, one, sorne, what, whatever, which, wMchever, and the
phrasaI combinations many a, such a, what a. The class-meaning
is 'unidentified specimens.'
The word a is peculiar in its sandhi-form an, used before vowels.
The word one OCCUrs not only as an indefinite determiner (one
man), but also in SOme entirely different functions (as in a big one
if one only knew); this phenomenon may he designated as
cleavage. The meanings of the varions indefinite determiners are
in part linguistically definable in terms of grammatical features
of wider hearing than our present subject. For instance, what and
which are interrogative, introducing supplement-questions, which
prompt the hearer to supply a speech-form (what man? which man?)
Whatever and whichever are relative, marking their noun as part
of a subordinate clause (whatever book you take, ...). No and
neither are ncgative, ruling out ail specimens. Each, which, and
whl'chever imply a limited field of selection: that is, the specimens
concerned helong tD an identified part (or to the identified whole)
of the species (which book? which parent?); eher and neither go
farther in limiling the field ta two specimens.
Some of the determiners are atonic (barring, of course, the
case where they are emphatic elements): my, our, your, his, her,
its, their, the, a; others are sometimes atonie or spoken with sec-
ondary stress.
The types of noun expressions which always have a determiner,
are preceded, when no more specifie determiner is present, by
the articles, definite the and indefinite a, whose meaning is merely
the class-meaning of their respective form-classes. A grammatical
classification, such as definite and indefinite, which always ac-
companics sorne grammatical feature (here the types of noun ex-
pression which demand a determiner), is said to he categoric. The
definite and inddinite categories may be said in faet to embrace
the cotire cluss of English noun expressions,' even thase
types of noun expression which do Dot always takc a determiner,
can he classed as definite or indefinite: John, for instance, as defi-
nite, kindness as indefinite.
According to the use and non-use of determioers, English noun
expressions fall into a number of interesting sub-classes:
1. Names (proper nouns) occur only in the singular number,
take no determiner, and are always definitc: John, Chicago.
The class meaning; is 'spccies of abject containing only one
specimen.' Here and in what fol1ows, space forbids our en-
tering into details, such as the class-cleavage by which a
name occurs also as a cornmon noun, in cases like homonymy
(two Johns, this John); nor can we take up such
as that of river-names, which are always preceded by the
(the Mississippi).
II. Common nouns occur in both categories, definite and in-
definite. The class-meaning is 'species of object occurring in
more than one specimen.' In the plural number they require
a determiner for the definite eategory (the houses), but not
for the indefinite (houses, corresponding to the singular form
a house).
A. Bounded nouns in the singular number require a determiner
(the house, a house). The class mcaning is 'species of objeet
oecurring in mOre than one specimen, such that the specimens
cannot be subd,ided or merged.'
B. Unbounded nouns require a determiner for the definite cate-
gory only (the milk : milk). The class-meaning is 'specics
of objeet occurring in more than one specimen, such that
the specimens can be subdivided or merged.'
1. Mass nouns never take a and have no plural (the mk : milk).
The class-rneaning is that of B with the added proviso that
the specimens' exist independently.'
2. Abstract nouns in the indefinite singular without a determiner
include aU the specimens (life is short); with a dcterminer and
in the plural, the specimens are separate (a useful life; nine
lives). The class-meaning is that of TI with the proviso
that the specimens' exist only as the demeanor (quality, ac-
tion, relation) of other abjects.'
Among the subdivisions of II, class-cleavage is frequent and
teresting, as, an egg, eggs (A), but "he gat egy on his necktie"
(BI); coffee (BI), but an expensive coffee (A).
The limiting adjectives of the other class, numeratives, fall
into various sub-classes, of which we shall merely mention a few.
Two of them, aU and both precede a dcterminer (ail the apples); the
rest follow (the other apples). Two, however, precede a in phrases
whieh arc determiners: many a, sueh a. The numeratives few,
206 SYNTAX
hundred, thousand, and those fonned with the suffix -ion (million
and sa on), arc prcceded by a in phrases which serve as numcra-
tives with plural nouns (a hundred years). The numeratives same,
very, one - this Just differs by class-cleavage from the determiner
one - arc uscd only with definite nouns (lhis same book, the very
day, my one hope); the numeratives much, more, less are uscd only
with indcfinitc nouns (much waler); the numerative all is uscd with
bath kinds of nouns but only with definite determincrs (all the
milk; aU milk). Some, such as bath, few, many, and the higher num-
bers, are used anly with plural nouns; others, such as one, mw;h,
liUle, only with singular nouns. Sorne numerativcs are used also
in other syntactic positions, as, many and few as predicate adjec-
tives (they were many), and aU, bath as semi-predicative attributes
(the boys were bath there). Some other interesting !ines of classi-
fication among the English numeratives will appear when we take
up the substitutive replacement of Doun expressions in Chapter 15.
CHAPTER 13
MORPHOLOGY
13.1. By the morphology of a language we mean the construc-
tions in which bounli forms appear among the constituents. By
definition, the resultant forms are either bound forrns or wards,
but never phrases. Accordingly, we may say that morphology
includes the constructions of words and parts of words, while
syntax includes the constructions of phrascs. As a border region
we have phrasc-words (jack-in-the-pulpit) and some compound
,yards (blackbird) , which contain no bound forms among their
irnmediate constituents, and yet in some ,vays exhibit morphologie
rather than syntactic types of construction.
In general, morpholo!!;ic constructions are more elaboratc than
those of syntax. The features of modification and modulation
are more numerous and often irregular - that is, confined to
particular constituent.s or combinations. The order of the con-
stituents is almost always rigidly fixed, permitting of no such
connotative variants as-John ran away : Away ran John. Features
of selection minutcly and often whimsically limit the constituents
that may hc united into a complex form.
Accordingly, lan!!;uages difTer more in morphology than in s y n ~
tax. The variety s so great that no simple schemc will classify
languages as to their morphology. One sueh scheme distinguishes
ana-lytic languages, which use few boumi forms, from synlhetic,
which usc many. At one extremc is a completcly analyhc lan-
guage, like modcm Chinesc, where each word is a one-syllable
morpheme or a compound ward or phrase-word; at the other, a
highly synthetie language like Eskimo, whieh unites long strings
of bound forms into single words, snch as [a:wlisa-ut-iss?ar-si-
niarpU-I]a] '1 am looking for something suitable for a fish-Ene.'
This distinction, however, except for cases at the former extrcme,
is relative; any onc language may bc in some respects more ana-
Iytic, but in other respects more synthetic, than sorne other lan-
guage. Another scheme of this sort dividcd lan!!;uages into four
morphologie types, isolating, agglulinalive, polysynthetic, and in-
207
208 MORPHOLOGY
MORPHOLOGY
209
fleeting. Isolating languages were those which, like Chinese, used
no bound fonns; in agglutinative languages the bound fonns were
supposed merely to follow one another, Turkish being the stock
example j polysynthetc languages expressed semantically impor-
tant elements, such as verbal goals, by means of bound forms, as
cloes Eskimoj inflectional languages showed a merging of seman-
tically distinct features eithcr in a single bound fonn or in closely
united bOl.lnd forms, as whcn the suffix -0 in a Latin form Iike
ami) '1 love' expresses the meanings 'speaker as actor,' 'only one
actor,' 'action in present time,' 'real (not merely possible or hy-
pothetical) action.' These distinctions are not co-ordinate, and
the last three classes were nevcr clearly defined.
13. 2. Since the speaker cannot isolate bound fonns by speak-
ing them aione, he is usually unable to describe the structure of
words. The statement of morphology requires systematic study.
The ancient Greeks made some progress in this direction, but,
in the main, our technique was devcloped by the Hindu grammari-
ans. No matter how refined our method, the elusive nature of
meanings will always cause difficulty, especially when doubtful
relations of meaning are accompanied by formaI irregularities.
In the series goose, gosling, gooseberry, gander, we shall probably
agree that the tirst two fonus are morphologically related, in the
sense that [gaz-] in gosling is a phonetic modification of goose,
but the [guwz-] in gooseberry docs Dot fit the and, on the
other hand, the formai resemblance [g-J of goose and gander is
sa slight that one may question whether it really puts the practical
relation of meaning into Iinguistic form. This last difficulty appears
also in the pair duck : drake, with their common [d ... k]. One
saon learns that one cannat look ta the speakers for an answer,
since they do not practise morphologie analysis; if one bothers
them with such questions, they give inconsistent or silly answers.
If the history of a language is known, one often tinds that the
ambiguity was absent in some older state of the language it
appears, for instance, that sorne centuries aga' gooseberry' was
*grose-berry and had nothing ta do with a goose but facts of this
sort evidently do not tell us how things work in the present state
of the language.
In describing the modulations and modifications which accut in
syntax, we naturally take the absolute form of a word or phrase
as our starting-point, but a bound form which occurs in several
shapes will lead to several entirely different fonus of description,
according to our choice of a basic aUernant. For instance, the
plural-suffix of English nouns appears ordinarily in three shapes:
[-el'l] glasses, [-z] cards, [-s] books; by taking each of these thrcc, in
turn, as one's starting-point, one can arrive at three entirely
different statements of the faets.
Very often therc are further difficulties. Sometimes a grammati-
cal featurc, such as a modification, appeaTs ta express a
meaning which is usually expressed by li linguistic form, as in
man: men, where modification of the vowcl takes the place of the
plural-suffix. In other cases there is not e'len a grammatical
feature: a single phonctic farm, in the mmlllet of homonymy, rep-
resents two meanings which are usually distinguished by means
of a linguistic form, as, singular and plural noun in the sheep
(grazes) : the sheep (graze). Here the Hindus hit upon the apparently
artificial but in practice eminently serviceable dcvice of speaking
of a zero element: in sheep : sheep the plural-suffix is replaeed by
zero - that is, by nothing at aIl.
13. 3. \Vhat with these and other difficulties, any incomstency
of procedure is likely to create confusion in a descriptive statement
of morpholog:.y. One must observe, above ail, the principle of
immediate constituents ( 10.2). This principle leads us, at the
outset, ta distinguish certain classes of words, according ta the
immediate
A. Secondary words, containing free forms:
1. Compound words, containing more than one free form:
door-knob uild-animal-tamer. The inc1uded free forms are
the memb'ers of the compound word; in our examples, the
rnembers are the words door, knob, tamer, and the phrase
wild animal.
2. Der'ed secondary words, containing one free fonu: boyish,
old-maidish. The ineluded free form is callcd the under
lying form; in our eXl1rnples the underlying forms are the
ward boy and the phrase old maid.
B. Primary words, not containing a free fonn:
1. Derived primary words, containing more than one bound
form: re-eeive, de-ceive, con-ceive, re-tain, delain, con-
tain.
2. consisting of a single (free) morpheme:
man, boy, cul, run, red, big.
210
MORPHOLOGY MORPHOLOGY 211
The principle of immediate constituents willlead us, for example,
ta class a fonn Iike gentlemanly not as a compound ward, but as a
derived secondary word, since the immediate constituents are the
bound fonn -ly and the underlying word gentleman; the word
gentlemanly is a secondary derivative (a so-ealled de-compound)
whose underlying form happons to be a compound word.' Similarly,
door-knobs is not a compound ward, but a de-compound, consisting
of the bound form [-z] and the underlying word door-knob.
The principle of immediate constituents leads us te observe the
structural order of the constituents, which may differ from their
actual sequence i thus, ungentlemanly consists of un- and gentle-
manly, with the bound form added at the beginning, but gentle-
m.anly consists of gentleman and -ly with the bound form added at
the end.
13. 4. As examples of relatively simple morphologie arrange-
ments. we may take the constructions of secondary derivation that
appear in English plural pouns (glass-es) and past-tense verbs
(land-ed). .
As ta selection, the bound fonns arc in bath cases unique, but
the undcrlying fonns bclong to two great fonn-classes; the plural
nouns are derived from singular nouns (as, glasses from glass) and
the past-tense verbs from infinitive verbs (as, landed from land).
Other, subsidiary taxemes of selection will COncern us later.
As to order, the bound fonn, in both cases, is spoken after the
underlying fonn.
By a feature of modulation common to ncarly ail constructions of
English morphology, the underlying form keeps Hs stress, and the
bound fonn is unstressed.
The taxemes of phonetic modification are more elaborate, and
will show us sorne pecuIiarites that appear in the morphology of
many languages.
To begin with, the bound form appenrs in several alternants
diffcrent shapes which imply, in this case, features of
modification:
glass: glasses [-cz]
pen : pens [-z]
book: books [-8].
If we collect examples, we SOon find that the shape of the bound
form ie dctermined by the Just phoneme of the accompanying for'1l:
[-oz] appears after sibilants and affricates (glaSSe8, roses, dishes,
garages, churches, bridges) i [-z] appears after aIl other voiced
phonemes (saws, boys, ribs, sleeves, pem, hills, cars); and [-s] after
ail other unvoiced phonemes (books, cliffs). Since the differences
bctween the thrce alternants [-cz, -z, -s] can he described in terms
of phonetic modification, we say that they are phonetic alternants.
Since thc distribution of the thrce alternants is regulated acoonHng
ta a linguistically recognizable characteristic of the acoompanying
forms, wc say that t.he alternation is regular. Finally, since the
deciding characteristic of the accompanying forms is phonemic
(namely, the identity of the last phonernc), we say that the alt<:rna-
tian is automatic.
Regular altcrnations play a great part in the morphology of most
languages. Not, aIl regular alternations are phonetic or automatic.
In German, for instance, the singular nouns are divided, by certain
syntactic features, into thrce forrn-classes which are known as
gonders ( 12.7); now, German plural nouns are derived from
singulars by the addition of bound forms which differ acoording t'
the gender of the underlying singular:
masculine nouns add [-el, with certain vowel-changcs: der Hut
[hu:t] 'hat': Rilte ['hy:tc] 'hats'; der Sohn [zo;n] 'son': S6hne
'sons'; der Baum [bawm] 'tree': Baume ('bai me] 'trocs'
Deuter nouns add [-cl without vowel-change: das Jahr [ja:r]
'year': Jahre ['ja:re] 'years'; das Boat [bo:t] 'boat': Boote ['bo;te]
'boats '; das Tier [tLr] 'animal': Tiere [' ti:rel'animaIs'
feminine nouns add [-en]; dz'e Uhr [u;r] 'dock, watch': Uhren
['u:renJ 'clocks, watches'; die Last [last] 'burden': Lasten ['lasten]
'burdens' i die Frau [frawl 'woman': Frauen ['frawen] 'women.'
This alternation (aside from special features which wc need not
oonsider) is regular, but it is not phonetic, sinee, of the threc
alternants, [-cl with vowel change, [-cl, and [-cn], the last is not,
in the system of the language, phonetically akin to the first two;
and t.he alternation is not automatic, but grammatical, sinee it
depcnds not upon phonctic, but upon grammatical (in this instancc,
syntactic) peculiarities of the underlying forms.
13. 5. \Ve have not yet described in tenns of phonetc modifica-
tion, the kinship of the three alternants [-ez, -z, -sl of the bound
farm that appears in English plural nouns. It is evidcnt that three
entirely different statements are possible, according to our choicc
of one or another of the three forms as our starting-point. Our
212 MORPHOLOGY
MORPHOLOGY
213
1 The tyPes of English pronunciation which diBtinguieh between [el and [il in
unetrfsscd position, use [il in both the bound form (1]/a..l8e<l) and the word (Be<l8'8).
aim is to get, in the long ron, the simplest possible set of statements
that will describe the facts of the English language. To try out the
different possible formulae with this aim in view, often involves
great labor. In the present instance our trouble is small, because
our alternation has an exact parallel in English syntax: the enclitic
word whose absolute form is i:s ['iz], alternates quite like our plural
suffix:
land : landed [-edJ
live : lived [-dJ
dame : dand HJ,
and wc nccd not hesitate, now, to take [-cd] as the basic fOnD for
our de8cription and to say that this form loses Hs vowel after al!
phonemcs except dental stops, and then replaces [dJ by [t] after aH
unvoiced sounds.
13. 6. A survey of English plural nouns will soon show that the
statement we have made holds good for an indcfinitely large
number of forms, but not for a certain limited number of exceptions.
In some instances the constituent form in the plural differs
phonetically from the underlying singular noun:
Bess's ready riz, ez] 1
John's ready [zJ
Dick's ready [sJ.
Sinee in this case the absolute fOnD is necessarily serves as the
starting-point of description, wc reach the simplest fOnDula if we
take [wez] as the basic aernant also of the bound form. We can
say, then, that in English any morpheme of the fOnD riz, ezJ, un-
stressed, loses its vowel after al! phonemcs exeept sibilants and
affricates, and then replaces [z] by [s] after unvoiced sounds. This
covers also the alternation of the third-person present-tense verb
suffix in misses: runs : breaks and of the possessive-adjective
suffix in Bess's, John's, Dick's. Moreover, it leads us ta use a
parallel formlila in the case of the past-tense suffix of verbs. This
8uffix appears in thrce similar alternants:
We can describe the peculiarity of these pIurals by saying that
the final [f 6, s} of the underlying singular is replaced by [v, '5, zJ
before the' bound form is added. The word "before" in this
statement means that the alternant of the bound form is the one
appropriate to the substituted sound; thus, the plural of kmfe
adds not [-s], but [-z]: "first" the [-f] is replaced by [-v], and" then"
the appropriate alternant [-zJ is added. The terms "before,
first, then," and so on, in such statements, tell the descnptwe
order. The aetual sequence of constituents, and their structural
order ( 13.3) arc a part of the language, but the descriptive order
of grammatical features is a fiction and results simply from our
method of describing the forms; it goes without saying, for instance,
that the speaker who says knives, does not "first" replace [f] by
[vJ and" then" add [-zJ, but merely utters a form (kni.ves) which
in certain features resembles and in certain features dlffers from
a certain other form (narncly, kmfe).
If the English plural nouns which cxhibit this voicing of a
spirant in the underlying form, showed any common phonetlc
or grammatical feature that distinguished them from other
we could describe this peculiarity aS a regular alternant. IhlS,
however, 880mS not to be the case; we have also plurals like cliffs,
myths, creases, where [f, 6, s] of the underlying form appcars un-
changed. We can make our general statcrnent coyer one group,
but will then have to furnish a list of the cases that do not fall
under the general statcrnent. A set of forms that is not covered
by a general statement, but has to be presented in the shape of
a list, is said ta be irregular. We try, of course, to arrange our
description so that as many forms as possible will be included
in general statements. The choice is often decided for us by the
circurnstance that one group of forms is of indefinite extent and
therefore amenable to a general statement, but not to a list. In
the case of English nouns in [-s], we obviously face this condition,
for house : houses is the only instance where [-sJ is replaced by [zJ
in thc plural, while an indefinite number of plural nouns retains
the [-s] of the underlying form (ylasses, creases, CUrses, dames,
and so on). Our list, in this case, inc1udes only one form, houses,
a unique irregularity. The list of plurais which substitute ['51
for the [-6] of the underlying form is not large, embracing ooly the
forms baths, paths, cloths, mouths (and for sorne speakers also
laths, oaths, truths, youths); on the other side we find a number
: knives [najv-z]
: mouths [maw'5-zJ
: houses ['hawz-ez].
knife [najf]
mouth [maw6]
house [hawsJ
,'of current. forms, sueh as months, widths, droulhs, myths, hearths,
and, what is more decisive, the habit of kccping [-6] in the fOTIna-
tian of plurals that arC not traditional and may be formed by a
speaker who has not heard them: the llfcGraths, napropaths, mo?W-
liths. In the case of the list is largcr: knives, wives, lives, calves,
halves, lhieves, leaves, sheaves, beeves, loaves, elves, shelves (and for
sorne speakers also hooves, rooves, scarves, dwarves, wharves); we
decide to calI thesc irrogular on the strength not only of counter-
instances, such as clijJs, toughs, reefs, oafs, but also of less common
or occasional forms, such as (some good) laughs, (general) staffs,
monographs.
,Yhere the two trcatments occur side by side, as in laths [le6s]
or [letiz], roofs or there is usually sorne slight difference of
connotation between the variants. The noun becf, as a mass-noun
( 12.14), has no ordinary plural by its side; the plural beeves is
a specialized derivative, since it deviates in its meaning of 'oxen,
cattle,' with archaic-poetic connotation.
We may note in passing that the grammatical fcatures we have
discussed, determine features of the phone tic pattern ( 8.5),
by defining groups like sibilant-ajJricate, dental stop, voiced, un-
voiced, and establishing the relation [f, 6, sl versus [v, 5, z], and
[t] versuS [dl.
We may describe "voicing of final spirant plus suffix [-{Oz, -Z,
(-s)] " as an irregular alternant of the regular plural-suffix [-cz, -z, -s];
the irregularity consists in a phone tic modification of the under-
lying form. The same modification is accompanied by modifica-
tion of the syllabic in the uniquely irregular staff: slaves. In
cloth [kkIO] : clothes [klowz] we have a uniquely irregular plural
with specialized meaning (' garment.s, clothing '), beside the ir-
rcgular plural elolhs [kl;fiSz] with normal meaning.
The homonymous third-person present-tensc suffix of verbs is
accompanied by phonetic modification of the underlying form in
do [duw] : does [doz], say [sej] : says [sez], have [hEV] : has [hez].
The suffix [-cd, -t] is accompanied by phonetic mod-
ifieatian in the irregular forms say: said, flee : fled, hear [hij r] : heard
[hrd], keep : kept (and, similarly, crepi, s/ept, swept, wept; leaped
and leapt are variants), do : did, sell : sold (and, similarly, told) ,
make:m.ade, have : had.
13. 7. Ins0
rt1
e cases the bound form appears in an unusual
shape. In die: dice the alternant [-s] appears against the general
habit; in penny: pence the same featuro is accompanied by modi-
fication (loss of [-ij]) in the underlying form, togethcr with special-
ization of meaning, in contrast with the normal variant pennies.
In the past tense, we find H.J instead of [-dl in the archaic-f1u.vored
variants burnt, learnt. If we say that in English thcunperrnitted
final cluster [-dt] is replaced by we can class here, with [-tl
instead of [-ed], the forms benl, lent, sent, spent, huilt.
Bath constituents show irregular phonetic modification in
feel : feU and similarly in dealt, knelt, dreamt, meant. If wc say
that thc unpermitted final clusters [-vt, -zt] arc replaced by
we can class here also leave : left and lose: lost. The bound form ap-
pears in the alternant Hl instead of [-d], and the underlying form
replaces the syllabic and ail that follows by [:J] in seek [sijk] : sought
[sot] and, similarly, in bought, brought, caught, taught, thought.
In the extreme case, an alternant bears no resemblance to the
other alternants. In ox : oxen the bound form added in the plural
is [-I,t] instead of [-cz, -z, -s]. If the language does not show parallel
cases which warrant our describing the dcviant form in tenus of
phonetic modification, an alternant of this sort is said ta be supple-
tive; thus, [-l,l] in oxen is a suppletive alternant of [-ez, -z, -s],
becuuse English grammar shows no phoneiic modification of [-ez]
to [-Q]. In other instances it is the underlying form which suf-
fers suppletion. Beside the ordinary derivation of kind : kinder,
warm : warmer, and sa on, we have goorl : better, where ,the
lying word good is replaced by an entircly difTerent form bet-,
which we describe, accordingly, as a suppletive alternant of good.
In the same way, the infinitive be suffers supplction, by [i-], in
the third-person present-tense farm is riz]. In child: children, a
suppletive alternant [-rQ-] of the bounel form is accompanied by
phonetic modification of the underlying ward.
Another extreme case is that of zero-alternants ( 13.2), in which
a constituent is entircly lacking, as in the plumls sheep, deer, moose,
fish, and so on. Thcse plurals arc irregular, for although some of
them (for instance, spccies of fish, like perch, bass, pickerel, large
enough ta be catcn in separat.e specimens, and not narned after
other objects) ean be classified by purely practical fpatures of
meaning, thcy have no formaI charaeteristic by whioh we could
define them. The past-tensc suffix of verbs shows a zero-alternant
in bet, let, set, wet, hit, slit, split, cul, shul, put, bcat, cost, bUTst,
shed, spread, wed. The third-person pres{'nt-tcnsp suffix has 11
214 MORPHOLOGY

1
1
"
,1
1
i
MORPHOLOGY 215
216 MORPHOLOGY
MORPHOLOGY
217
In cases likc have [hr,v] : had [h-::l] or make [mejk] : made [mej-dJ,
one of the constituents is modificd by the loss of a phoneme. This
loss may he described as a minus-feature; like zero-features or
substitution-fcatures, minus-features may occur independently.
For instance, in a French adjective, the regular type has only
one form, regardless of whct.her the adjective accompanies a
masculine or a femininc noun, c.g. rouge [ru:z] 'rcd' : un livre rouge
[ce" li:vr;) ru:z] 'a red book,' masculine, and une plume rouge
[yn plym ru:z] 'a red feather or pen,' feminine. In a fairly large
irregular type, howcver, the masculine and feminine forms diffcr:
un livre vert [ve:r] 'a green book,' but une plume verte [vert] 'a
green feather or pen.' Thus:
It is evident that two forms of description are here possible.
We could take the masculllf; forms as a basis and tell what con-
sonant is added in each case in the feminine form, and this would,
of course, result in a rairly complicated staternent. On the other
hand, if we take the feminine form as our bass, wc can describe
this irregubr type by the simple st.atement that the masculine
form is derived from the feminine by means of a minus-feature,
namely, 108s of the final consonant and of the c1uster [-kt]. If wc
take the latter course, we find, moreover, that ail the other
cnces betwecD the two forms, as to vowel quantity and as ta
nasalization (as in our last example), re-appear in other phases of
French morphology and can in large part be attributed ta the
phonetie pattern.
The last part of our discussion has shawn us that a word may
have the chameler of a secondary derivative and yet consst of
zero-alternant in can, shall, will, must, may, and, in certain con-
structions (for instance, with the modifier not), in need, dare; this
is a regular grammatical alternation, since these verbs are defin-
able by their syntactic function of taking an infinitive modifier
without the preposition ta. Our possessive-adjective suffix [-ez, -z,
has a 1.ero-alternant in one instance, namely, after an under-
lying form which ends in the plural-suffix [-ez, -z, -s,] as the-boys'.
A zero-alternant may go with modification of the accompany-
ing form. Thus, the plural nouns geese, teeth, feet, mice, lice, men,
women ['wimJ}.] add no bound form to the singular, but contain a
difIerent syllabic. In these plurals a grammatical feature, phonetic
modification, expresses a meaning (namely, the sememe 'more
than one abject ') which is normally expressed by a linguistic form
(namely, the morpheme [-ez, -z, -s]). Wc may say that "substitu-
tion of [ii]" (for the stressed syllabie of the underlying farm) in
geese, teeth, ft, "substitution of [aj]" in mice, lice, "substitution
of [e]" in men, and "substitution of [i]" in women, are alternants
of the normal plural-suffix - substitution-aliernants or substitution-
forms. In our past-tense verbs we filld substitution of various
syllabics taking the place of [-ed, -d, -t], as:
[ a ] got, shot, trad
[ e ] drank, sank, shrank, rang, sang, sprang, began, ran, swam,
sat, spat
[ e ] bled, fed, led, read, met, held, fell
[ i ] bit, lit, hid, slid
[ 1saw, fought
[ 0 J clung, flung, hung, slung, swung, spun, won, dug, stuck,
struck
[li J shook, took
[ej ] ale, gave, came, Zay
[ aw J bound, found, ground, wound
[ ow] clave, drave, wove, bore, swvre, tore, wore, broke, spoke,
woke, chose, froze, rose, smote, wrote, rode, sto/e, shane; with dove
as a variant besidc regular dived
[(j)uw] knew, blew, flew, slew, drew, grew, threw.
In stand: stood we have a more complex case with an alternant
describablc as "substitution of [u] and 10ss of [n]."
A zero-alternant replaces the bound form, and a suppletive al-
ternant the underlying form, in cases like be: wus, go: went,
1 : my, we : our, she : her, bad : worse.
MASOULINE
plat [pla] 'fiat'
laid [le] 'ugly'
distinct [dis;"] 'distinct'
long [la"] 'long'
[bal 'low'
gris [gril 'gray'
frais [frE] 'fresh'
gentil [zaDti] 'gentle'
lger [leie] 'light'
soul [su] 'drunk'
lJln [pieD] 'full'
FEMININJ1I
platte [plat]
laz'de [led]
distincte [disteDkt]
longue [long]
basse [ba:s]
grise [gri:z]
frache [fn:::s]
oentille [zaDti:j]
lgre [lezer]
soule [sul]
pleine [plen].
218
MORPHOLOGY
MORPHOLOGY 219
only one morpheme, accompanicd by a zero-feature (sheep, as
a plural; eut as Il past) , by a substitutioll-feature (men, sang),
by suppletioll (went, worM) , or by a minus-feature (French vert,
masculine). 'Ve class t.hesc words as secondary dcrivatives and rec-
ognize thcir peculiarity by cl1lling them secondary morpheme-words.
13.8. The bound forms which in secondary derivation arc added
ta the underlying form, are ca.l1ed ajJixes. Affixes which precede
the underlying form are prefixes, as be- in be-head; those which
follow the undcrlying fonn are called suffixes, as [-ez] in glasses
or -ish in boyish. Affixes added within the undcrlying form are
ea11ed infixes; thus, Tagalog uses several infixes which are added
befare the first vowel of thc I1nderlying form: from ['su:lat] 'a
writing' are derived [sl1'mu:lat] 'one who wrote,' with the infix
[-um-], and [si'nu:1at] 'that which was writtcn,' with infix [-in-Jo
Reduplication is an affix that consists of rcpeating part of the
underlying form, as Tagalog [su:-'su:lat] 'one who will write,'
['ga:mit] 'thing of use': [ga:-'ga:mit] 'one who will use.' Redu-
plication may be of various extent: Fox [wa:pame:wa] 'he looks at
him': [wa:-wa:pamewa] 'he examines him,' [wa:pa-wa:pame:wa]
'he keeps looking at him.' l t may differ phonctically in sorne con-
ventional way from the underlying ward: ancient Greek ['phajncj]
, it shines, it appears ': [pam- 'phajnej J 'it shines brightly'; San-
skrit ['bharti] 'he bears': ['bi-bharti] 'he bcars up,' ['bhari-bharti]
'he bears off violently.'
13.9. 'Ve have seen that whcn forms are partially similar, there
may be a question as ta which one we had better take as the under-
Iying form, and that the structure of the language may decide this
question for us, sinee, takinp; it one way, we get an unduly com-
plicated description, and, taking it the other way, a relatively
simple one. This same consideration often leads us 1.0 set up an
artificial underlying form. For instam:e, in German the voiced
mutes lb, d, g, v, z] are Dot permitted finals, and are in final
position replaced by the correspondinr; unvoiced phonemes. Ac-
cordingly we gct sets like the following:
[tani] 'weep' [tanisia] 'wept'
[inu] 'drink' [inumia] 'drunk'
lulu] 'enter' [ulufia] 'cntcred.'
It is clear that a useful description will here set up the basic forms
in theoretical shape, as [tanis-, inum-, uluf-].
'VITH SUFFIX [-lA] WlTHOUT BUFFIX
SINGULAR (RUFFIX ZERO)
[nen8:h] 'my hand'
[mete:h] 'aheart'
[wi :ki:h1'birch-bark'
It is evident that if we taok the underlying words in their actual
shape as OUI' basic forms, we should have ta give a long list ta tell
which ones appcared in derivativcs with [z] instead of [s]. On the
other hand, if wc start from an artificial underlying form with
[-z], as hawz-], in contrast with [spa:s, aws], we need give
no list and can account for the uniform final [-sJ which actually
appears in the independent forms, by the rule of permitted finals.
Similarly for the othcr voiced mutes, as in
rond [runt] 'round' runde [' rund.-e] 'round ones'
bunt [bunt1'motley' bunte [' bunt-e1'motley ones,'
where we set up a theoretical basic form [rund-] in contrast with
[bunt]. Wc have seen that in sorne languages these theoretical
forms appear a1so in the phrase, by reminiscent sundhi ( 12.5).
Similarly, some languages permit no final c1usters and yet show
included free forms with clusters. Compare the following noun-
forms in Menomini:
PLURAl. [-AN])
[nene:hkan] 'my hands'
[mete:hjan] 'hearts'
[wi:ki: hsan] , pieces of birch-
bark' "-
[nekdceneh] 'my thumb' [nekdcene:hcjan] 'my
[pe:hcekuna:h] 'medicine- [pe:hcekuna.:htjan] ,
bundle ' bundles. '
It is evident that a description which took the singular forms as
a basis would have ta show by elaborate lists what consonants, as,
[k, j, s, cj, tj], arc added before IL suffix; the simple and natural
description is ta take as a starting-point the free forms Ilot in their
absolute shape, but in the form which appears before suffixes, as
[wi:ki:hs-] and the likc.
Another example is furnished by Samoan, which permits no
final consonants at a11, and therefore has sets like the following:
DER1VF.n WORD
grasen ['gra:z-cn] 'ta graze'
hausen ['hawz.-en] 'ta keep house,
ta earry on'
spaszen ['spa:s-enl 'ta jest'
auszen ['aws-en] 'on the outside.'
Spasz [spa:s] 'jest'
aus [awsJ 'out'
UNDlLYlNG W()RD
Gras [gra:s] 'grass'
HaW3 [haws] 'house'
220 MORPHOLOGY MORPHOLOGY
221
13. 10. Modulation of secondary phonemes often plays a part
in morphologie constructions. In English, affixes arc normally
unstressed, as in be-wail-ing, friend-li-ness and the like. In our
fOl'eign-Iearned vocabulary, shift of stress to an affix is a taxemc
in many secondary dcrivatives. Thus, some suffixes have pre-suf-
fixal slress: the accent is on the syllable befol'C the suffix, regardless
of the nature of this syllablc; thus, -ity in able: ability, formal:
formality, major: majority; [-jQ-] in music: musician, audit: audi-
tion, edw;ate : edw;alion; [-ik] in demon : demonic, anarchist: an-
archistic, angel : angdic. In the derivation of some of our foreign-
lcarned nouns and adjectives from vcrbs, the stress is put on the
prefix: from the verb insert [in'sft] wc dcrive the noun insert
['insrt]; similarly, contract, conviet, convert, converse, discourse, pro-
lesl, project, rebel, transfe!". In other cases this modulation appears
along with a suffix: conct:ive : concept, perceive: percept, portend .. por-
tent; in some, the underlying verb has to be theoretically set up, as
in precept.
In sorne languages modulation has greater scope. In Sanskrit,
with sorne suffixes the derivative form keeps the accent of the
underlying farm:
['ke:a-] 'hair' : [ke:a-vant-] 'having long hair'
[pu'tra-] 'son' : [pu'tra-vant-] 'having a son.'
Others are accompanied by shift of accent to the first syllable:
['purusa-] 'man' : 'coming from man'
[va'sti-] 'bladdcr' : 'of the bladder.'
Others have presuffixal accent:
'man' : [puru'sa-ta:-] 'human nature'
[dc:'va-] 'god' : [de:'va-ta:-] 'divinity.'
Other affixes are themsclvcs acccnted:
[' fsi-] 'sage' ; [:1:rs-e:' jar] 'descendant of a sage'
[sa'rama:-] (proper noun) : :'ja-] 'dcsccnded from
Sarama.'
Others require an accentuation opposite to that of the underlying
ward:
['atithi-) 'guest' : [a:ti'th-ja-] 'hospitality'
[pali'ta-] 'gray' : ['pa:lit-ja-] 'v;rayness.'
Tagalog uses bath stress and vowel-Iengthening as auxiliary
phonemes; thrce suffixes of the fOTIn [-an] differ in the treatment of
these modulations.
Suffix [-an]
1
is characterized by presuffixal stress and by long
vowel in the first syllable of the underlying form:
[' i: big] 'love' : li: 1bi;gan] 'love-affair'
[j'num] 'drink' ; [i:'nu:manJ 'drinking-party.'
The meaning is 'action (often reciprocal or collective) by more
than one actor.'
Suffix [-anF is stressed when the underlying word has stress on
the first syllable; othel'wise it is treated like [-anJ!:
['tu:lug] '8lcep' : [tulu'gan] 'sleeping-place'
[ku'lul)] 'enclose' : [ku:'lu:lJanJ 'place of imprisonment.'
The meaning is 'place of action, usually by more than one actor,
or repeated.'
Suffix [-anJ3 has presuffixal stress when the underlying word is
stressed on the first syllablej it is stresscd when the underlying
ward is strf'ssed on the last syllable; there is no vowel-lengthening
beyond what is demandcd by the phonetic pattern;
(a) ['sa:gil)] 'banana' : [sa'gi;l)an] 'banana-grove'
[ku'lu!J] 'enclose' : [kulu'IJan] 'cage, crate'
(h) ['pu:tul]' eut' ; [pu'tu:lan] 'that which may be eut from'
[Ia'kas] 'strength' : [laka'san] 'that upon which strength may
he expended.'
The meaning is (a) 'an object which serves as locality of the
underlying object, action, etc.,' and (h) 'that which may be acted
upon.'
In languages with auxiliary phoncmes of pitcll, these may play
a part in morphology. Thus, in Swcdish, the suffix -er of agent-
nouns shows the normal compound word-pitch of polysyllables
( 7.7) in the resultant form: the verb-stem [Ie:s-] 'read' forms
liiser [vle:ser] 'reader'; but the of the present tense demands
simple word-pitch in the resultant form: (han) liiser ['Ie:ser] '(he)
reads.'
13. 11. In ail observation of word-structure it is very important
to observe the principle of immediate constitucnts. In Tagalog, the
underlying form ['ta:wa] 'a laugh' appears reduplicated in the
derivativc [ta:'ta:wa] 'one who will laugh'; this form, in turn,
222
MORPHOLOGY
MORPHOLOGY 223
underlies a derivative with the infix [-um-], namely [tuma:'ta:wa]
, one who is laughing.' On the other hand, the form ['pi:litl 'effort'
jirst takes the infix [-um-], givinp; [pu'mUit] 'one who compellcd,'
and is tken reduplicated, gvinp; [-pu:pu'mUitJ, which underlies
[nap;-pu:pu'mi:lit] 'one who makes an extreme effort.' Close
observation of this prindple is ail the more neccssary because now
and then we meet forms which compromise as to immediate
stituents. Tagnlog has a prefix [paIJ-], as in [a'tip] 'roofing' :
[paIJ-a1ti p] 'that used for roofing-; shingle.' The [IJ] of this prefix
and certain initial consonants of an accompanyinp; form are subject
ta a phonctie modification - we may cali it morphologie sandhi-
by which, for instance, our prefix joins with ['pu:tul] 'a eut' in the
derivative [pa-'mu:tul] 'that Ilsed for cuttinv;,' with substitution
of [ml for the combination of [-IJ] plus [p-]. In sorne forms, howevcr,
we find an inconsistency as ta the structural arder; thus, the fOrIn
[pa-mu-'mu:tul] 'a cutting in quantty' implies, by the actual
sequen of the parls, that the redupli cation is made "beforc" the
prefix is added, but at the same time implies, by tbe presence of
[m-] for [p-J in bath reduplication and main form, tbat the prefix is
addcd "before" the reduplication s made. A carelessl}' ordered de-
scription would fail ta bring out the pecllliarity of a form like ihis.
13. 12. In languages of complcx morphology wc can thus ob-
serve a ranking of constructions: a complex ward can be described
only as thoup;h the various compoundings, affixation>!, modifica-
tions, and sa on, were addcd in a certain order ta the basic form.
Thus, in English, the ward acfresses consists, in the first place,
of adress and [-ez], just as laMes consists of lass and [-ez];
in tum consists of actor and just as consists of couni
and -css; actor, finally, consists of act and [-rJ. There would Le
no parallel for a division of actresses, say int.o actor and -esses. In
languages of this type, then, we can distinguish severai ranks
of morphologie structure.
In many languages these ranks fall into classes: the structure
of a complex word reveals first., as ta thc more Immediate oon-
stituents, an outer layer of inflectional constructions, and then
an inner layer of constructions of word-formation. In our 1ast
cxample, the outer, inflect.ional layer is represented by the
struction of ac/ress with [-ez], and the inner,
layer by the remaining const.ructions, of actor with -css and of
aet with [-rl.
This distinction cannat always be carried out. Tt is based on
severaI features. The constructions of inftcction usually cause
closure or partial c!osure ( 12.11), sa that a ward which contains
an inftectional construction (an t'nflected ward) can figure as a
constituent in no morphologie constructions or else only in certain
inflectional constructions. The English form adresses, for in"tance,
ean enter into only onc morphologie construction, namely the
derivation of the possessive adjective adresses' (with the zero-
alternant of [-ez, -z, -s], 13.7). This latter form, in turn, cannat
enter into any morphologie construction; it has complete C!osure.
Anoiher peculiarity of inflection, in contrast with word-
formation, is the rigid parallelism of underlying and re>!ult.ant
forms. Thus, nearly al! English singular nouns underlie a derived
plural noun, and, vice versa, neal'ly aIl Enp;lish plural nouns are
derivcd from a singular nonn. Accordingly, English nouns occur,
for the most part in paraUel sets of two: a singular noun (hal) and
a plural noun derived from the former (hats). Given one of these,
the speaker is usually capable of producing the other. Each such
set of forms is called a paradigmatic set or paradigm, and each
form in the set is called an inflected form or infieetion. Sorne lan-
guages have large paradigms, which contain many inflections.
In Latin, for instance, the verb appears in sorne 125 inflectional
fonns, such as amiire 'ta love,' am 'I love' amas 'thou lovest' , ,
amat 'he loves,' amamus 'we love,' amem '1 may love,' amor
'I am loved,' and so aD; the occurrence of one form usually guar.
antees the occurrence of ail the others. It is this parallelism of
the infiections which forces us ta treat a single phonetic form,
like sheep as a set of homonyms, a singular nOlm sheep (eorrespond-
ing ta lamb) and a plural noun sheep (corresponding ta lambs).
It is tbis paraIlelism also, which leads us to view entirely differ-
ent phonetic forllls, like go: went, as morphologically related
(by suppletion): go as an infinitive (parallel, say, with show) and
went as a past-tense form (parallel, then, with showed).
The parallelism, ta he sure, is sometimes imperfect. Defective
paradigms lack sorne of the infiections; thus, can, may, shall, will,
must have no infinitive, must has no past tense, scissors no singular.
If, as in these cases, the lacking form happens ta underlie the
tually existing ones, we do best to Het up a theoretical underlying
form, Hlleh as a non-cxistent infinitive *ean or singular *scissor-.
On the other hand, sorne irregular paradigms are over-differenti-
224 MORPHOLOGY
MORPHOLOGY
225
nominative ['hipp-osJ ['hipp-oj]
vocative ['hipp-e] ['hipp-oj]
accusative ['hipp-on] ['hipp-ows]
dative ['bipp-o;jJ {'bipp-ojs]
genitive ['hipp-ow] ['hipp-o:n]
In seeondary dcrivation, however, tbis paradigm is represented
not by the common clement [hipp-], but by a special deriving-
form [hipp-o-] as in 'horseman,' or with loss of the [oJ
by phonetic modification, in [hipp-i'kos] 'pertaining to horses.'
Similarly, as a cornpound-mcmber, the paradigm is represented
by a special compounding-form, homonymous with the preceding:
[hippo-'kantharos] 'horse-beetle.' Thus, wc distinguish bctween
the kernel [hipp-], which actually (subjcct, however, in principle,
a whole, is represented by the underlying form; the EngHsh lan-
guage, accordingly, may he said to have word-inftection, word-
derivation, an word-composition.
In many languaJ.,'"Cs, c8pecially in those which have a more com-
plex morphology, none of the forms in a paradigm can convenicntly
be viewed as underlying the others. Thus, the rcgular paradigms
of the German verb contain a cornmon element which is not equal
to any of the inflectional forms. For instance, the pamdigm
represented by the forms !.achen ['Iax-en] '(to) laugh,' (ich)
!.ache ['lax-e] '(1) laugh, , (er) lacht [lax-t] '(he) laughs,' (er) lachte
'(he) laughed,' ge/acht [ge-'lax-t] 'laughed' (participle) ,
and sa on, shows a common clement lach- [lax-] in aU the inflec
tional forms, but none of thesc inflectional forms consists simply
of the clcment lach- without an affix. In secondary dcrivation and
composition the paradigm is reprcscnted by this same fotm, as in
Lacher ['lax-er] 'laugher' and Lachkrampf [' lax-Ikrampf]
spasm.' This lach-, strictly speaking, is a bound form; it is called
the kernel or stem of the paradigrn. The German verb is an example
of stem-inftection, stem-derivation, and stem-composition. In our
description, wc usually treat the stem as if it wcre a free form.
In sorne languages of this type, the common clement of the
paradigm differs from the stem which represents the paradigm
in derivatives and compounds. Thus, an ancient Greek noun-
paradigm has stcm-inflection. It contains a common element, a
kernel, mueh likc the German verb-stem, e.g. [hipp-] 'horse':
a/ed. Thus, corresponding to a single form of an ordinary
digm like play (to play, 1 play, we play), the paradigm of be has
three forms (to be, 1 am, we are), and, corresponding ta the single
form played, it has the forms (1) Wa8, (we) were, been. The exist-
ence of even a single over-differentiated paradigm implics homon-
ymy in the regular paradigms.
The parallelism of inflected forms goes hand in hand with a
further charaeteristic: the differcnt inflections dilIer in syntactic
function. If wc say the boys chauffe, our syntactlc habit of con-
gruence ( 12.7) requires us, when the boy is the actor, to supply also
the form chauffes. In the case of the present and past inflections
of the English verb this is not true : the parallelism of plays : played
is not required by any habits of our syntax, but is carried out
none the less rigidly.
If there are several tanks of inflcction, wc get compound para-
digms; the inflections of the Fnglish noun, for instance, consist of
an outer construction, the dcrivation of the possessive adjective,
and an inner one, the derivation of the plural:
SnlGULAR PLURAL
nominative-accusative man men
possessive adjective man's men's
In the Latin verb wc find a very complicated compound
digm: an outer layer for different adors or undergoers, distin-
guished as to persan (speaker, hearer, thir person), number (sin-
gular, plural), and voice (actor, undcrgoer), an inner layer for
differenccs of tense (present, past, future) and mode (real, hypo-
thctical, umeal), and an innermost layer for a dilIercnce as ta
completion of the act (imperfectic, perfeetic).
13. 13. Wc come, finally, to an important charactcristie of
inflection, akin to those we have mcntioncd, the derivational unity
of paradigmf' The inflectional forms of a paradigm do not each
enter into nuposition and dcrivation, but the paradigm as a
whole is represented by sorne one forrri. ln English, the forms of
a noun-paradigm are represented by the singular, as in man-
sla1<ghler, mannish, and those of the verb-paradigm by the in-
finitive, as in playground, player. An English paradigm consists
of an underlying word (itself a member of the paradigm) and sorne
secondary derivatives containing this underlying word; as a con-
stituent in further derivation and composition, the paradigm, as
SINGULAR PLURAL
226 MORPHOLOGY
t.o phonetic modification) appears in aIl the forms, and the stem
[hippo-], which underlies the further derivatvs.
Sorne exceptions to the principle of paradigmatic unity are only
apparent. The possessive-adjective form in the English compounds
like bull's-eye or the plural form in longlegs are due, as we shaH see,
to the phrasaI structure of these compounds. Real exceptions do,
however, occur. German has a suffix chen [-xen] 'small,' which
forms secondary derivatives from nouos, as; Tisch [tis] 'table' :
Tischchen ['tis-xeo] 'ltt.1e table.' In the system of German mor-
phology, this is a construction of word-formation, but in a certain
few instances the suffix [-xen] is added tQ nouns which aIready have
plural inficction: bcside Kind [kint] 'child': Kindchen ['kint-xcnJ
'little child,' the plural infiection K inder [' kinder] 'children' under-
lies the detivative Kinderchen ['kinder-xenJ 'littlc children.' If a
language contained too many cases of this sort, wc should simply
say that it did not distinguish such morphologie layers as are
denoted by the terms inficction and word-formation.
)
CHAPTER 14
MORPHOLOGIe TYPES
14. 1. Of the three types of morphologie constructions which can
he distinguished according to the nature of the constituents ~
namely, composition, secondary detivation, and primary dcriva-
tion ( 13.3) ~ the constructions of compound words are most
similar to the constructions of syntax.
Compound wards have two (or more) frce forms among their
immediate constituents (door-knob). Undcr the principle of im-
mcdiate constituents, languages usually distinguish compound
words from phrasc-derivatives (as, old-maidh, a secondary dcriva-
tive wit.h the undcrlying phrase old maid), and from de-compounds
(as, genllemanly, a secondary derivative with the underlyinf!; c o m ~
poun word gentleman). \Vithin the sphere of compound words,
the same princi ple usual1y involves a defini te structural order; thus,
the compound wild-animal-ho1lEW does not consist, say, of three
membets 11!ild, animal, and house, and not of the members wild
and animal-house, but of the members wild ammal (a phrase) and
house; and, similarly, t.he compound doorknob-wiper consists, un-
mistakably, of the members door-knob and wiper, and not, for in-
stance, of door and knob-w/per.
The grammatical features which lead us to recognize compound
words, differ in difforent languages, and some languages, doubtless,
have no such cluss of forms. The gradations between a word and a
phrase muy he many; often enough no tigid distinction ean be
made. The forms which we cluss as compound words exhibit. sorne
feature which, in their language, characLerizes single words in
contradistinction to phrases.
In meaning, compound words are usually more specialized than
phrases; for instance, blackbird, denoting a bird of a particular
species, is more specialized than the phrase black bird, which de
not.es any bird of this color. lt. is a very common mist.ake to try
to use this difference as a criterion. \Ve cannot gauge meanngs
accurately enollgh; moreover, many a phrase is as specializcd in
rneuuing as any compound: in t.he phrases a queer bird and meat
227
228 MORPHOLOGIC TYPES MORPHOLOGIC TYPES 229
and drink, the words bitd, meal are fully as speciaIized as they are
in the compounds jailbird and sweetmeats.
14. 2. In languages which use a single bigh stress on each
word, this feature distinguishes compound words from phrases.
In English the high stress is usuaHy on the first member; on the
other member there is a lesser stress, as in door-knob
upkeep ['op-1kijp]. Certain compounds have the irre@larity of
leaving the second member unstressed, as in gentleman ['Jent\mQ],
Frenchman ['frenmQ]; contrast milkman Certain
types of compounds, chiefiy sorne whose members are adverbs and
prepositions, stress the second member: without, upon. Aecord-
ingly, wherever we hear lesser or least stress upon a ward which
would always show high stress in a phrase, we describc it as a
compound-member: ice-cream ['ajs-1krijm] is a compound, but
ice cream ['ajs 'krijm] is a phrase, although there is no denotative
difIerence of meaning. However, a phrase as priar rnember in a
compound keeps aH its high stresses: in ['wajld-
'nimhhaws] the stress assures us only that house is a compound-
member; the rest of the structure is shawn by other criteria.
As to the phonetic pattern, compound words are generaHy
treated like phrases: in English, clusters like [vt] in shrovetide or
[nn] in pen-knife do not occur wit.hin simple words. Sandhi-like
phonetic modifications mark a compound as a single ward only
when they differ from t.he sandhi of syntax in the same language.
Thus gooseberry ['guwzbrij] is marked as a compound becausc the
substitution of [z] for [s] is not made in English syntax, but only in
morphology, as in gosling ['gazlilJ]. Similarly, in French, pied--Ie:n-e'
[pjet-a-tc:r] 't.emporary lodging' (literaHy 'foot-on-ground ') be-
side pied [pje] 'foot,' or pot-au-feu [p::Jt-o-fp] 'broth' (literally
'pot-on-the-fire ') beside pot [po] 'pot,' or vinaigre [vin-egr] 'vinegar'
(li teraHy 'sour-wine ') beside vin [ven] 'wine,' are marked as com-
pounds, because French nouns do not exhibit these types of sandhi
in t.he phrase, but only in word-constructions, such as pieter [pjete]
'toe the mark,' potage [pJta:z] 'thick soup,' vinaire [vine:r] 'per-
taining ta wine' i cont.rast, for instance, the phrase vin aigre [VEn egr]
'sour wne.'
More st.riking phonetic modifications may mark a compound;
thus, in the following examplcs the priar member suffers greater
modification than it does in any phrase of its lauguag: holy
['howlij] : holiday ['hulidcj], moon: Monday, two [tuw] : twopence
{'topl).S]; Old English ['fe:ower] 'four' : ['fitler-,fe:te] 'four-footed';
the second member, in Sanskrit [na:wh] 'ship' : [ati-'nuh] 'gone
from the ship' ; ancient Greek [parte:r] 'father' : [ew- 'pato:r] 'well-
fathered 'i Gothie dags 'day' : fidur-dogs ' four days old'; hoth
members, in English breakfast ['brekfest], blackguard ['blegrd],
boatswain ['bowsQ], forecastle ['fowkslli in sorne cases there is also
a variant form without modification, as in forehead ['fared], waist-
coat ['wesketJ. In extreme cases, of course, the form may be sa
unlike the independent word that we may hesit.ate botween calling
it a compound-,ember or an affix: a form likefortnight ['forhnajt,
'fortnet] lies on the border bctween compound and simple ward.
The order of the membcrs in a compound word may bo fixed,
while that of the phrase is free, as in bread-and-buUer botr]
'slices of bread spread with butter,' contrasting with the phrase,
as in she bought bread and buUer, she bought butter and bread. This
criterion is likely to break down, however, because the order in a
phrase, tao, may be fixed: wc have a1so a specialized phrase ['bred
fi. 'hotr] with the same order and the same meaning as the com-
pound. Contrasting order is a surer mark: French blanc-bec
[blan-bek] 'callow young persan' (litcrally 'whi tc-beak ') is char-
acterized as a compound, because adjectives like blanc in the phrase
always foHow thcir noun: bec blanc 'whi te beak.' EngIish exampIes
are to housekeep, to barkBlide, to undergo, since in a phrase a noun
goal like house and adverbs of the type back, under would follow
the verb (keep house, slide back).
14.3. The commonest, but also the most varied and most
difficult ta observe, of the features which lead us to distinguish
compound words from phrases, are grammatical features of se-
lection.
The plainest contrast appoars in languages with stem-eomposition
( 13.13). A stem like German lach-, which represents a whole verb
paradigm in a German compound like Larhkrampf
'laughing-spasm,' but does not actually occur as an independent
word, makes the compound unmistakably different from any
phrase. Even more plainly, a compounding-stem, such as ancient
Greek [hippo-] 'harse,' may differ formally from aIl the inflections
of its paradigm, and, in any case, characterizes a compound by it.s
invariability; t.hus, [hippo-] joins some other st.em, such as ['kan-
tharo-] 'beetle,' ta form a compound st.em,
'horse-beetle,' but remuins unchanged in aIl t:le inflcctional forms
230 1IORPHOLOGIC TYPES MORPHOLOGIC TYPES 231
of this compound: nominative [hippo'kantharo-s], accusative
[hippo'kantharo-n], and so on.
Evrn when the compound-mernber is formal1y equal ta sorne
ward, it rnay characterize the compound. In aocient Grcck a
nono-stern is infiect ('(1 by means of suffixes. Accordingly, the
first mernber of a compound noun-stem will remain the same in all
forms of the paradigm. Thus, the phrase 'new city' will show
various inflectional forms of two paradigms:
nominative [ne'a: 'polis]
accusative [ne'a:n 'polin]
p;cnitive [ne'a:s 'poleo:s],
and so on, but the compound stem [ne'a:-poli-] 'Naples,' whose
first member is in nominative singular form, will show this first
member unchanged in ail the inflcctioo!\:
nominative [oc'a:polis]
accusative [ne'a:polin]
genitive [nea:'poleo:s).
In German, the adj ecti ve has word-inflcction; the unrierlying
form is uscd as a complement of verbs: Das isl roi [das ist 'ro:t]
'that is red,' and the derived infiections appear as modifiers of
nouns: roter Wein ['ro:ter 'vajn] 'red wine.' The absence of i n ~
fleclional suffixes thcreforc characterizes the compound-member in
a fonn like Rotwein ['ro:t-1vajn] 'red-wine.'
The use of prefixes and suffixes may decide for us what is the
be!/;inning and what the end of a ward or stem. In German, the
past particip1e of verhs is formed by t.he addition ta the stem of a
prefix [ge-] and a suffix [-t], as in gelacht [ge-'lax-t] 'laughcd.' The
position of these affixes, accordingly, shows us that a form like
geliebkost [ge-'li:p,ko:s-t] 'care&;l:-d' s one '1'1 0 I"(! , derived from a
compound stern, but that a form like liebgehabt ['li:p ge-,hap-t]
'liked' is a two-word phra.-"C. This gives us a standard for the
classification of other inflectional forms, such as t.he infinitives
liebkoscn ['li:p-,ko:zen] 'ta caress' and liebhaben ['li:p Jha:ben}
'ta like.'
Sometimcs the compound-member resembles an infiectionai
fonn, but one which would he impossible in the phrase. The
[-z, -sJ on the prior mernbers of bondsman, kinsman, landsman,
marksman resembles the possessive-adjective suffix, but possessive
adjectives Iike bond' s, land's and 30 on, would not be sa used in the
phrase. In French, the adjective grande [I;rand] 'great,' as in une
grand,e maison [yn grand mezo
n
] 'a big house,' drops the final con-
sonant ( 13.7) ta rnake the inflectional fonn used with masculine
nouns: un grand garon [
n
gra
n
garson] 'a big boy'; but, as a
compound-member, the latter form appears a1so with certain
Feminine nouns: grand'mre [grun-me:r] 'grandmother,' grand'porte
[gran-pJrt] 'main entry.' Compound-members of this type are
especially common in German: Sonnenschein ['zonen-Isa.jn] 'sun-
shine' has the pror member Sonne in a farm which, as a separate
ward in a phrase, could only be plural; in Geburistag [ge'burts-1ta:k]
'birthday,' the [-s] is a genitive--casc ending, but would not he
added, in an independent word, ta a ferninine noun like die Geburt
'birth.'
A compound-member may be characterized by some feature of
ward-formation which differs from what would appear in an
independent ward. In ancient Grcck therEil was a highly irregular
verb-paradigm, containing such forms as [da'mao:] '1 tame,'
[e'dme:the:] 'he was tamed,' and so on, which grammarians con-
veniently describe on the hasis of a stern-foml [dame:-]. From this
paradigm there is derived, on the one hand, the ndcpendent agent-
noun [dme:'te:r] 'tarner,' and, on the other hand, with a different
suffix, an agent-noun [-damo-J, which is used only as a second
member of compound words, as in [hip'po-damo-s] 'horsc-tamer.'
Compounds with special features of ward-formation arc known as
synthetic compounds. Synthetic compounds occurrcd especially in
the aider stap;es of the lndo-European language'l, but the habit is
by no means extinct. In English, the verb to black underlics the
independent ag-ent-noun blacker (as in a blacker of boots), but forms
a1so, with a zero-elcmcnt, the agent-nOllll -black which appears in
the compound boot-black; similarly, to sweep forms sweeper and the
second member of chimney-sweep. Even forms likc long-tlIilcd or
red-bearded are not aptly describcd as containing- the words iilcd,
bearded (as in tai:d monkeys, bearded lady); the natural starting-
point is rather a phrase like long iail or red beard, from which they
differ by the presence of the suffix ~ e d . This is the same thing as
saying that wc use compounds of the type long-tailed, red-berded
regardless of the existence of words like tailed, bearded: witness
forms like blue-eyed, four-footed, snub-nosed. Anothcr modern
English synthetic type is that of three-masler, thousnd-legger.
In English, we freely form compounds likc met-eater and meat-
232 MORPHOLOGIe TYPES MORPHOLOGIe TYPES 233
eating, but not verb-compounds like *to meat--eat; thesc exist only
in a few irregular cases, such as 10 housekeep, to bootlick. Now, ta
be sure, words like ealer and eating exist alongside the compounds;
the synthetic feature consists merely in the restriction that a
phrase like eat meat is paralleled hy compounds only when -e:r or
-ing is at the same time added. We may designate the types meat-
eating and meat-eater as semi-synlhetic compounds.
14. 4. Among the word-like features of the forms which we
c1ass as compound words, indivisibility ( 11.6) is fairly frequcnt:
we can say black -1 8hould say, bluish-black - birds, but we
do not use the compound word blackbird with a similar interrup-
tion. In sorne instances, however, other features may lead us ta
c1ass a form as a compound word, even though it is subject to
interruption. In Fox, a form like [ne-pje;i-wa:pam-a:-pena] 'wc
have come to see him (her, thern) , has to he classed as a compound
word, because the infiectional prefix [ne-] '1 (but not thou)' and
the inBectional suffixes [-a:-] 'him, her, them' and [-penal (plural
of first person' unmistakably mark the beginning and end of a
word ( 14.3). 'l'he members of the compound are the particle
[pje:ci] 'hither' and the verb-stem [wa:pam-] 'sec (an animate
abject).' Nevertheless, the Fox language sometimes inseris words
and even short phrases between the rnembers of such compounds,
as in [ne-pje:ci-keta:nesa-wa:pam-a:-pena] 'we have come to see
her, thy daughter.' In German, compound-members can be c o m ~
bined serially; Singvogel ['zIhfl'5:gel] 'songbirds,' Raubvogel
['rawpv,fp:gel] 'birds of prey,' Sing- oder Raubvagel ['zilJ-o:der-
'rawp-,fl'5:gel] 'songbirds or birds of prey.'
GenArally, a compound-member cannot, like a ward in a
phrase, serve as a constituent in a syntactic construction. The
word black in the phrase black birds can be modified by very (v!3ry
black birds), but not sa the compound-member black in blackbirds.
This feature serves ta class certain French forms as compound
words: thus, sage-femme [sa:z-fam] 'midwife' is to he c1assed as a
compound, in contrast with a homonymous phrase meaning 'wise
woman,' because only in the latter can the constituent sage' wise'
he accompanicd by a modifier: tres sage femme [tre sa:z fam]
'very wise woman.' This restriction, like the prcceding, is oc-
casionally absent in forms which by other featurcs are marked as
compound words. In Sanskrit, where stem-composition plainly
marks the prim member of compound words, this member is
nevertheless occasionally accompanied by a modifying word, as
in [Citta-prama'thini; de;'va;na:m 'api] 'mind-disturbing o f ~ g o d s
even,' that is 'disturbing ta the minds even of gods,' where the
genitive plural nOun (' of gods ') is a syntactic modifier of the
compound-member [Citta-] 'mlnd.'
14.6. The description and classification of the forms which
the structure of a language leads us ta describe as compound words,
will depend upon the characteristic features of tbis language.
Linguists often make the mistake of taking for granted the uni-
versai existence of whatever types of compound words are current
in their own language. It is true that the main types of compound
words in various languages are somewhat similar, but this Sm-
ilarity Is worthy of notice; moreover, the details, and especially
the restrictions, vary in different languages, The differences are
great enough to prevent our setting up any scheme of classifica-
tion that would fit aIl languages, but two lines of classification are
often useful.
One of these two lines of classification concerns the relation of
the members. On the one hand, we have syntactic compounds,
whose members stand ta each other in the same grammatical
relation as words in a phrase; thus, in English, the rnembers of
the compounds blackbird and whitecap (the difference between
these two examples will concern us later) show the same construc-
tion of adjective plus noun as do the words in the phrases black
bird and white cap. On the other hand, we have asynlaclic com-
pounds like door-knob, whose members stand ta each other in a
construction that is not paralleled in the syntax of their language
- for English has no such phrasai type as *door knob.
The syntactic compound differs from a phrase only in the es-
sential features which (in its language) distinguish compound
words from phrases - in English, then, chieBy by the use of
only one high stress. It may differ lexically from the correspond-
ing phrase, as does dreadnaught; the corresponding phrase, dread
naughi, has an archaic connotation, and the normal phrase would
befear nothing. Wc can set up sub-classes of syntactic compounds
according to the syntactic constructions which are paralleled by
the members, as, in English, adjective with noun (blackbird, white-
cap, bull's-eye) , verb with goal noun (lickspiUle, dreadnaught) ,
verb with adverb (gadabout) , past participle with adverb (cast-
away), and so on.
234 IHORPHOLOGIC TYPES
MORPHOLOGIe TYPES 235
Many compounds are intermediatc between the syntactic and
asyntactic extremes: the relation of the members parallels sorne
syntactic construction, but the compound shows more than the
minimum deviation from the phrase. For instance, the compound
verb ta housekeep diffcrs from the phrase keep house by the simple
feature of word-order. In such cases we may gpeak of various
kinds of semi-syntadic compounds. The difference of arder appears
also in upkeep versus keep up, and in the French versus
bec blanc ( 14.2). In turnkey versus turn the key or turn keys, the
difference lies in the use of the article or of the number-category.
Even types like blue-eyed, three-master, meat-eater, viewed as syn-
thetic compounds, can be said to correspond ta blue eyes, three
masts, eat meai, and ta difIer from these phrases by simple formaI
characteristics, including the addition of the bound forms -ed, -er
ta the second member. In French, [bwo:t-a-Ietr],
literaUy 'box-for-letters,' and bote-aux-lettres [lJ\vo:t-o-Ietr], lit
eraUy both meaning 'mail-box, post-box,'
differ in the choico of preposition and in the use of the article
from the normal phrasaI type, which would givc bote pour des
lettres [bwo:t pu:r de ldr] 'box for letters'; the use of and certain
other prepositions in place of more specific ones, and difforences
of article (especially of zero in place of the phrasaI article ropre-
sented by the form are in French well-marked feature8 which
enable us to set up EL class of semi-syntactic compounds.
Where semi-syntactie eompounds are definable, they can be
further classified in the same manner as syntaciic compounds:
thus, in the semi-syntactic the members have the same
construction as in the syntactic bladcbird, in three-master the same
as in three-day, in housekeep, turnkey the same as in lickspittle, in
upkeep the same as in gadabout.
Asyntactic compounds have members which do oot combine
in syntaetic constructions of their langUf1ge. Thus, in door-knob,
horsefly, bedroom, salt-cellar, tomcal we sec two nouns in a construc-
tion that does not occur in English syntax. Other asyntactic types
of English compounds are illustrated by fly-blown, frost-biiien-
crestfallen, foot80re, fireproof, foolhardy - by-law, by-path,
glade - dining-room, mvimming-hole - bindweed, cry-baby, drive-
way, playground, blowpipe - broadcast, dry-clean, foretell- some-
where, everywhere, nowhere. Compounds with obscure members,
Ruch as smokestack, mushroom, or with unique rncmbers, such as
cranberry, huckleberry, zigzag, choo-choo, are, of course, to be classed
as asyntactic.
Although the relation between the members of asyntactic com-
pounds is nccessarily vague, yot wc can sornetimes extcnd the
main divisions of syntaciic and semi-syntactic compounds to
caver also the asynt.actic class. In English, for instance, the co-
ordinative or copulative relation which we sec in a semi-syntactic
compound like biitersweet (compare the phrase biUer and sweet) , can
be discerned also in asyntactic compounds like zigzag, fuzzy-
wuzzy, choo-choo. 110st asyntactic compounds secm ta have a
kind of attribut.e-and-head construction: door-knob, bulldog, cran-
berry. '1'0 the extent that one can carry out this comparieon, one
can thereforc distinguish between copulative compounds (San-
skrit dvandva) and determinative (attributive or subordinative)
pounds (Sanskrit tatpuTUsha); these divisions will cross t.hosc
of syntactic, and asyntactic compounds. One
may even be able t mark off smal1er divisions. The Hindu
grammarians distinguished among copulative compounds a special
sub-group of repetitive (amredita) compounds, with idcntical mem
bers, as in choo-choo, bye-bye, goody-goody. In English, we can
mark off also a class in WhlCh the members show only some elc-
mentary phonetic differcncc, aS zigzag, flimflam, pell-mell, fuzzy-
wuzzy. The Hindus found it convenient to set off, among the
detcrminaves, a special class of syntactic attribute-and-hcad
pounds (karmadharalla), such as blackbirrl.
14. 6. The othor frequently usable Hne of classification
cerns the relation of the compound as a whole to its members.
One can often apply to compounds the distinction between en-
docentric and exoceniric constructions which wc met in syntax
( 12.10). Since a blackbird is a kind of a bird, and a door-knob
a kind of a knob, wc may say that these compounds have the
same function as their head members; they arc ondocentric. On
thc other hand, in gadabout and turnkey the head member is an
infinitive verb, but the compound is a noun ; theso compounds
are exocentric (Sanskrit bahuvrihi). To take a copulativc type as
an cxample, the adjective biUersweet ('bitter and sweet at th.e
same time ') is endocentric, since the compound, like its co-ordl-
nated mcmbers bit/er and mveet, has the function of an adjective,
, .
but the bitiermveet is exocentric, since, as a Doun, lt
differs in grammatical function from the two adjective members.
236 MORPHOLOGIe TYPES MORPHOLOGIC TYPES 237
Another type of English exocentric compounds consists of adjeo-
tives with noun head: two-pound, half-mile, (in) app[e..
pie (order).
The difference of form-class may he less radical, but still recog-
rnzable in the system of the language. In English, the noum
longlegs, bright-eyes, bUUerfingers are exocentric, because thev occnr
both as singulars, and, with a zero-affix, as plurals (that longlegs,
thase longlegs). In French, the noun rouge-gorge [ru: 'robin'
(literally 'red-throat ') is exocentric, because it helongs to the
masculine gender-class (le rouge-gorge 'the robin '), while the head
IDember belongs ta the feminine gender (la gorge 'the throat').
In the English type sure-footed, blue-eyed, straight-backed the syn-
thetic suffix -d, -t] gocs hand in hand with the exoccntric value
(adjective with noun head); however, one rnight perhaps hesitate
as to the classification, since -eyed, -backed might he viewed
as adjectives (compare horned, bearded). Types like clambake, up-
keep are better described as endocentric, in English grammar, he-
cause the head members -bake and -keep can he viewed as nouns of
action derived, with a zero-featurc, from the verbs; if English did
not use many zero-features in derivation and did not fOTIn many
types of action nouns, we should have ta class these compounds as
exocentric. Similarly, our description will probably work out best
if we class bootblack, chimney-sweep as endoccntric, with -black and
-sweep as agent-nouns.
On the other hand, the large class of English compounds tbat is
cxemplified by whitecap, longnose, SWallO'll)-tail, blue-coat, blue-
stocking, red-head, short-horn has noun function and a noun as head
member, and yet is ta be classcd as exocentric, because the con-
struction implies precisely that the abject does not bclong to the
same species as the head member: these compounds mean 'abject
possessing an object (second member) of such-and-
such quality (first member).' This appears in the faet that the
number-categories (longlegs) and the persanal-impersonal cate-
gories (nose ... it; longnose ... he, she) do not always agree.
In three-master, thousand-legger the synthetic suffix gocs hand in
hand with this exocentric relation. Nevertheless, there are
line cases which may prevent a clear-cut distinction. The com-
pound is endocentric if we view the inseet as 'like a
bottle,' but exocentric if we insist that the 'bottle' is only part of
the insect.
The Hindus distinguished two special sub-classes among exo-
centric cornpounds, namely numeratives (dvigu) , nouns with a
numbcr as prior membcr, such as, in English, sixpence, twelvemonth,
f ortnight, and adverbials (avyayibhava) , adverbs wi th noun head,
such as bareback, barefoot, hotfoot, or with noun subordinate, such as
uphill, downstream, indoors, overseas.
11/ 14. 7. In secondary derivative words we find one free form, a
'-1 phrase (as in old-rnaidish) or a word (as in mannish), as an im-
mediate constituent; in the latte:- case, the underlying word may
be a compound word (as in gentlemanly) or, in its own tum, a
derived word (as in adresses, whcre the underlying word aclress
is itself a secondary derivative from the underlying word ac/or).
We have seen, howevcr, that for the description of SOIlle languages,
we do well to set up theoretical underlying forms, namely stems,
which enable us to class certain forms as secondary derivatives
although, strictly spcaking, they do not eontain a free form
( 13.13). A similar dcvice is caUed for in the description of forms
like English scissors, oats, where we set up a thcoretical scwr-,
oat- as underlying forms, just as we class cranberry, oatmeal,
as compound words.
The underlying frce form, aetual or thcoretical, is accompanied
either by an affix, or, as we saw, in Chapter 13, by a grammatical
feature.
In many languages, secondary derivatives are divided, first of
aIl, into inftcctional forms and word-formational forms ( 13.12),
but we may do well to recaB that languages of this sort neverthcless
often contain border-line forms, such as, in English, beeves or
clothes, which predominantly resemblc infiectionul types, but show
a deviation. In the same wuy, learned ['lrned],
drunken, laden, sodden, molten, and the slang broke 'out of funds'
deviate from the strictly inflectional past participles learned
[lrnd], drunk, loaded, seeihed, meUed, broken.
The inflcctional forms ure relatively easy to describe, since they
occur in parallcl paradigmatic sets; the traditional grammar of
familiar lun/!;uages gi ves us a picture of their inflcctional systems.
It may be worth noticing, however, t.1at our traditional grammars
fall short of scientific compactness by dealing with an identical
feature over and over again as it oocurs in difTerent paradigmatic
types. Thus, in a Latin grammar, we tind the nominative-singular
sigu -s noted separatcly for each of the types am"icus 'frieud,' lapis
238 MORPHOLOGIe TYPES MORPHOLOGIe TYPES 239
'stone,' dux 'leader,' tussis 'cough,' manus 'hand,' facies 'face,'
when, of course, it should be noted only once, with a full statement
as ta where it is and whcre it is not used.
Ward-formation offcTS far more difficulty, and is largely neg
lected in our traditional grammars. The chief difficulty lies in
detennining which combinations exist. In very many cases we
have to resign ourselves ta calling a construction irregular and
making a list of the forms. Only a list, for instance, can tell us
from ,,,hich English male liouns we derive a femalc noun by means
of the suffix -ess, as in caunless, lianess, and it will probably require
a subsidiary list to tell in which of these derivatives a final [rJ is
replaced by non-syllabie [rL as in waiter : waitress, t1:ger : tigress-
for the type without this change, as in author : authoress is proh-
ably regular. Special cases, such as duke : duchess, master : mis
tress, thief : thievess demand separate mention.
Once we have established a construction of this kind, we may he
able ta set up a typical meaning and then, as in the case of inflec
tian, to look for parallels. Our suffix -ess, for instance, has a
definable linguistic meaning, Dot only because of the parallel
character of aU the sets like caunt : lion: lioness, but
also because English grammar, by the distinction of he : she,
recognizcs the meaning of the -ess derivatives. Accordingly, we
arc able to decidc, much as wc are in the case of inflection, whether
a given pair of forms, such as man: woman, does or does not show
the same relation. This cnablcs us to draw up supplcmentary
statements, resembling our descriptions of paradigms, ,vhich show
the various fonnal aspects of sorne grammatically deterrnincd
semantic unit. Thus, wc find the sememe 'female of such-and-such
male' exprcssed not only by the suffix -ess, but also by camposition,
as in elephant-cow, she-elephant, nanny-goat, and by suppletion, as
in ram : ewe, boar : sow; sorne such pairs show inverse derivation,
the male derived from the female, as goose : gander, duck : drake.
Similarly, we should probably need a complete list ta tell which
English adjectives underlie comparative forms in -er of the type
kinder, shorter, longer, and, having this list, wc could recognize
scmantically equivalent pairs, such as good : bet/er, much : more,
liUle : less, bad : worse.
In other groups the semantic relations are not Krammatical1y
definable. Thus, we dcrive a many verbs from nouns by
means of various changes, including a zero-clement, but the
ings of these derived verbs in relation to the underlying noun are
manifold: ta man, ta dog, to beard, to nose, ta milk, to tree, ta table, ta
skin, to boUle, la father, ta fish, to clown, and so on. Or, again, wc
derive verbs from adjectives in several varieties of the meanings
'ta become so-and-so' and 'ta make (a goal) so-and-so,' with
variollS formai devices:
zero: to smoo/he
zero, from comparative: ta lower
zero, from quality-noun: old : to age
modification of vowel: full : to fill
suppletion (?) : dearl : to kil/.
prefixes: enable, embitter, refresh, assure, insure, belittle
suffix -en: brigh/en
suffix -en, from quality-noun: long : lengthen.
1'0 this list wc must add a large number of foreign-Iearned types,
such as equal : equalize, archaic : archaize, English : anglicize, sim-
ple : simplify, vile: vilify, liquid: liquefy, valid: validate, long:
elongate, different : differentiate, debile : debililate, public: publish.
'Vhen derivation is made by means of grammatical features,
such as phonetic modification (man: men; mauth : to mouthe) or
modulation (convid verb : convid noun) or suppletion (go: went)
or zero-elements (eut infinitive: cut past tense; sheep singular :
sheep plural; man noun : to man verb), we may have a hard time
deciding which form of a set wc had better describe as the under-
Iying farm. In English, wc gct a simpler description if wc take
irregula,r paradigms (such as man: men or run : ran) as undcrly
ing, and regular paradigms (such a.."l ta man or a run) as derived.
In most cases this criterion is lacking; thus, we shaU find it hard ta
decide, in cases like play, push, jump, dance, whether ta take the
noun or the verb a-s the underlying form. Whatever our decision,
the derivative ward (e.g. ta man derived from the noun man, or
a run dcrived from the verb ta run) will often contain no affixes, and
will be described (for reasons that will shortly appcar) as a second-
ary
In the same way, phrase-derivatives, such as old-maidish, derived
from the phrase old maid, offer no special difficulty so long as they
contain a derivational affix, such as but when the phrase is
accompanied only by a zero-fenture, as in jack-in-the-pulptl or
devil-may-care, we have the diffirult type of phrase-words. These
240 MORPHOLOGIC TYPES MORPHOLOGIC TYPES 241
differ from phrases in their uninterrupted and syntactically m-
expansible character, and often in their exocentric value.
r;{. 14. 8. Primary words contain no free forms among their im-
mediate constituents. They may he complex, consisting of two
or more bound forms, as per-eeive, per-tain, de-eeive, de-tain, or they
may he Bimple, as boy, Tun, Ted, and, in, ouch.
The bound forms which make up complex primary words,
are determined, of course, by features of partial resemblance, as
in the examples just cited. In many languages the primary words
show a structural rcsemblance to secondary words. Thus, in
English, the primary words hammer, ruddcr, spider rcsemble scc-
ondary words Iike dance-T, lead-er, ride-T. The part of the primary
word which resembles the derivational affix of the secondary ward
(in our examples, -er) can he descrihed as a primary affix. Thus,
the primary words hammer, rudder, spider are said to contain
a primary suffix -er.* The remaining part of the primary ward-
in our examples, the syllable [hem-] in hammer, [rod-] in rudder,
[spajd-] in spider - is called the TOOt. The root plays the same part
in primary words as the underlying form (e.g. dance, lead, ride)
in secondary words (dancer, leader, rider).
This distinction between primary affixes and roots is justified
by the faet that the primary affixes are relatively few and vague
in meaning, while the roots are very numerous and therefore rela-
tively clear-cut as to denotation.
1
In aecordanee with this terminology, primary words that do
not contain any affix-like constituents (e.g. boy, run, Ted) are
classed as primary roo!rwoTds. The roots which occur in primary
roo"t--words are free roots, in contrast with bound roots which
occuronly with a primary affix, such as the root [spajd-] in spider.
Primary affixes may he extremely vague in meaning and aet
merely as an obligatory accompaniment (a determinative) of the
root. In English, the commonest primary suffixes do not even
tell the part of speech; thus, we have, with -cr, spider, bitter,
linger, etier, under; with -le, boille, li/tle, hustle; with -ow, furrow,
1 Early students of language, who confused description with the entirely differ-
ent. (and much harder) problem of ascertaining historical origina, somehow got the
notIon that possessed mysterious qualities, especially in the way of age. Now
and then One hears the c1aim that the roots which wc set up must once upon a
time have been spoken as independcnt worde. The reader need be told
that. this ;8 utterly unjustified; the roots, like ail bound forms, are merely units of
par bal resemblance between word. Our analysis guarantees nothing about earlier
stages of the language which wc are analyng.
yellow, borrow. In other cases the meaning is more palpable; thus,
-ock, i!l hummock, mattock, hassock, and so on, forms nouns de-
noting a lumpy object of moderate size, and this is confirmed by
its use as a secondary suffix (class-cleavage) in words like hillock,
bullock. Our foreign-Iearned prefixes get a vague but recogniz-
able meaning from contrasts like con-tain, de-tain, peT-tain, re-tain.
In sorne languages, however, primary affixes bear rclatively con-
crete meanings. The Algonquian languages use primary suffixes
that denote states of matter (wood-like solid, stone-like solid,
liquid, string-like thing, round thing), tools, parts of the body,
animaIs, ,,>'Oman, child (but not, apparently, adult males). Thus,
in Menomini, the verb-form [kepa:hkwaham] 'he puts a cover on
it,' has a stem [kepa:hkwah-], which consists of the root [ker-l
'obstruction of opening,' and the primary suffixes [-a:hkw-] 'wood
or other solid of similar consistcncy,' and [-ah-] 'act on inanimate
object by tool.' Similarly, in Mcnomini, [akuapi:nam] 'he takes
it from the water,' the verb-stcm consists of the root [akua-J
'removal from a medium,' and the suffixes [-epi:-] 'liquid' and
[-en-] 'act on abject by hand'; [ni:sunak] 'two canoes' is a par-
ticle consisting of the root [ni:sw-] 'two' and the primary suffix
[-unak] 'canoe.' Thesc affixes are uscd also in secondary deriva-
tion. Sorne of them arc derived from indepcndent words or stems;
thus, in Fox, [pje:tehkwe:wc:wa] 'he brings a woman or women'
is an intransitive verb (that is, cannot be used with a goal--object,
- much as if we could say *he woman-brings) containing the
primary suffix [-ehkwe:we:-] 'woman,' whieh is derived from the
noun [ihkwe:wa] 'woman.' In Menomini, the cOl'{Ilate [-ehkiwe:-],
as in [pi:tehkiwe:w] (same meaning), does not stand in this rela-
tion to any noun, because the old noun for 'woman' is here ob-
solete, and the aetual word is [mete:muh] 'woman.' In sorne lan-
guages the use of primary affixes derived from nouns covers mueh
the same semantic ground as our syntactic construction of verb
with goal-object. This habit is known as incorporation; the clas-
sical instance is Nahuatl, the language of the Aztecs, where a
noun like [naka-tl] 'meat' is represented by a prefix in a verb-
form like [ni-naka-kwa] 'I-meat-eat,' that is, '1 eat meat.'
A root may appear in only one primary ward, as is the case
with rnost ordinary English roots, such as man, boy, cut, red, nast-
(in nasty), ham- (in hammer), or it may appear in a whole series
of primary words, as i8 the case with many of our foreign-Iearned
242 MORPHOLOGIC TYPES
MORPHOLOGIC TYPES 243
roots, like in deceive, conceive, perceive, receive. In either
primary ward may underlie a whole series of secondary
denvatlves; thus, man underlies men, man's, men' s, mannish, manly,
(to) man (mans, manned, manning); deceive underlies deceiver
deceil, deception, deceptive; conceive underlies conceivable, conce:
concept, conception, conceptual; perceive underlies perceiver, per-
cept, perceptive, perception, perceptible, percsptual; and receive
underlies 'l'eceiver, 'l'eceipt, 'l'eception, receptive, 'l'eceptacle. More-
over, derivatives like these may ex8t where the primary
lS lackmg; thus, we have no such primary ward as *pre-
cewe, but wc have the words precept, preceptor, whieh are best
described as secondary derivatives of a theoretical underlying
form *pre-ceive.
The roots of a language make up its most numerous class of
morphologieal forms and bear its most varied and
specifie mcanngs. This is cleare8t in languages whieh have roots
as free forms, as, in English, boy, man, eut, 'l'un, red, blue, green,
brown, white, black. The ciear-eut meaning will be found alsa
in bound roots, sueh as yell- in yellow, pu'l'p- in purple, nast- in
nasty, and so on. In most languages, however, thcrc arc also
roots of very vague meaning, such as, in English, the forci/!;n-
learned roots of the type -ceive, -tain, -fer (conceive, contain, confer,
and so on). This is partieularly the case in languages \yhose pri-
mary affixes are relativcly varied and specifie in meaning.
Once we have set up il fOot, we face the possibility of its modi-
fication. This possibilit.y i8 obyious when the root occurs as an
ultimate constituent in il secondary derivative; thus, in the sec-
ondary derivative duchess the modification of the underlying word
duke is at the same time a modification of the root duke, and in
the secondar:r dcrivatives sang, sung, song, t.he modifications of
the undcrlying sing, are necessarily modifications of the root sing.
The alternant shapes of roots are in sorne languages sa varied
that the describer may weIl hesitate as to the choice of a basic
form. In ancient Greek we find the alternants [dame:-, dme:-,
dmo:-, dama-, dam-] in the forillS [e-'dame;] 'he tamed,' [e-'dme;-
the;] 'he ,;"as tamed,' ['dmo:s] 'slave,' '1 tame,' [hip'po-
Our whole description of Greek morphol-
ogy, mcludmg even the distribution of derivatives into primary
and sccondary types, will depend upon our initial choice of a
basic form for roots of this sort. In the Germanie languages,
modification of the root, with or without affix-like determinatives,
occurs in words of symbolic connotation, as flap, flip, flop. If
wc take flap as the basic form of this root, we shaH descrihe jlip,
flop as dcrivatiyes, formed by substitution of li] 'smaller, neater'
and by substitution of [a] 'larger, dulIer.' Similar cases are, with
substit.ution of [i]: snap : snip, snatch: sniteh, snuff: miff,
bang: bing, yap: Vip; of [ij]: squall: squeal, squawk: squeak,
crack: creak, gloom: gleam, iiny: teeny, of [oJ: mash: mush,
jlash : flush, crash: crush. At first glancc, wc should describe
thcsc forms as secondary derivatives, sinee the word jlap can be
said t() undcrlic the words jlip, flop, but it is possible that a de-
tailed description of English morpholo/!;Y would work out better
if we viewed words like jlip, jlop as primary modifications of
"the root instead of deriving them from the actual
ward jlap.
The root.s of a language are usually quite uniform in structure.
In English they are onc-syl1ablc clements, such as man, eut, red;
many of thern are free forms, oeeurring as root-words, but many,
such as [spajd-] in spider, [hr-m.] in hammer, and, especially,
forci/!;n-Iearned root.s like [-sijv] in conceive, perceive, are bound
forms. Sorne of thcse bound roots end in clusters that do not oc-
cur in ward-final, as [10mb-] in lumber or in linger. In
sian, the root.s arc rnonosyllabic, with the exception of sorne that
have [I] or [r] between vowels of the set [e, oJ, as in ['golod-J 'hun.
ger,' ['gorod-] 'city.' vVe have seen an example of the variability
of a root in ancient Greek; for this language, as weIl as, apparcntly,
for Primitive Indo-European, wc probably have to set up roots
of several different shapes, rnonosyllabic, such as [do:-] 'give,'
and disyllabic, such as [darnc:-] 'tame.' In North Chinese, ail
the roots are monosyllabic free forms consisting, phonetically,
of an initial consonant or cluster (which may be lacking), a final
syllabic (including diphthongal types with non-syllabic [j, w, n, 1)]),
and a pitch-scheme. The Malayan languages have two-syllable
roots, with stress on one or the other syllable, as in the Tagalog
root-words ['ba:haj] 'house' and [ka'maj] 'hand.' In the Semitie
languages the roots consist of an unpronounceable skeleton of
three consonants; aecordingly, every primary ward adds to the
root a morphologie e1ment which consists of a vowel-scheme.
Thus, in modern Egyptian Arabie, a root likc [k-t-b] 'writc' ap-
pears in words like [katabJ 'he wrote,' [ka:tib] 'writing (person),'
244 MORPHOLOGIC TYPES
MORPHOLOGIC TYPES
245
[kita:b] 'book,' and, with prefixes, [ma-ka:tib] 'places for writing,
studies,' [ma-Hab] 'place for writing, study,' (je-ktub] 'he is
writing;' similarly, the root [g-l-sJ 'sit' appears in [galas] 'he sat,'
[ga:lis] 'sitting persan,' [ma--gaJis] 'councils,' [ma-glas] 'council.'
In a few languages, such as Chinese, the structure of the roots is
absolutely uniform; in others, we find sornc roots that are shorter
than the normal type. It is a remarkable fact that these shorter
TOots belong almost always to a grammatical or a semantic sphere
which can be described, in tenns of English grammar, as the sphcre
of pronoun, conjunction, and preposition. In Gennan, which has
much the same root structure as English, the detinite article con-
tains a TOot [d-], for in the forms der, dem, den, and sa on, the rest
of the word (-er, -em, -en, and sa on) is in each case a normal in-
fiectionai ending, appearing also in the infiectional fonns of an
adjective like 'red ': rot-e:r, rotem, rot-en. The same applies to the
interrogative pronoun 'who?' with forms like wer, wem, wen. In
Malayan and in Semitic, many words in this semantic sphere have
only one syllable, as, in Tagalog, rat] 'and,' or the syntactic par-
ticles [alJ] 'sig-n of object-expression,' [aj] 'sign of predication,'
[na} 'sign of attribution.' This semantic sphere is roughly the same
as that in which English uses atonie words.
14. 9. Perhaps in most languages, most of the roots are mor-
phemes. Even in cases like English sing : sang : sung : song or
flap : flip : flop, a relevant description will view one of the forms
as basic and the others as sccondary derivahves or as primary
derivatives with phonetic modification of the root. In other cases,
however, wc tind clearly-marked phonetic-semantic rcscmblances
between elements which we vicw as different roots. The pronominal
words of English are probably best described as containing mono-
syllabic roots that resemble each other, especially as to the initial
consonants:
[ti-]: Ihe, Ihis, tkat, then, there, Ihilh-er, thus.
[hw-]: what, when, where, whith-er, which, why,. modiflcd ta [h]
in who, how.
[s-]: su, suck.
[n-]: no, nol, none, 7W7', nev-er, neith-er.
Complex morphologie structure of the root is much plainer in
the case of English symbolic words; in these we can distinguish,
with varying degrees of clearness, and with doubtful cases on the
border-line, a system of initial and final root-forming
of vague signification. It is plain that the intense, symbohc con-
notation is associated with this structure. Thus, we find recurrent
initiaIs:
[fl-l 'moving light' : flash, flare, ftame, flick-er, jlimm,..e:r.
[6-] 'movement in air': fly, ftap, flil (flutt-er). .
[gl-] 'unmoving light': glow, glare, gloal, gloom (gleam, gloam,..mg,
glimm-er) , glint. . .
[sl-] 'smoothly wet': slime, slush, slop, slobb-er, slzp, slule.
[kr-] 'noisy impact': crash, crack (creak), crunch.
[skr-] 'grating impact or sound': scratch, scrape, scream.
[sn-] 'breath-noise': sniff (snuff), snore, srwrt, snot. .
[sn-] 'quick separation or movement': snap (smp), snatch
(sni/ch).
[sn.] 'creep': snake, snail, sneak, srwop. ....
[H 'up-and-down movement': jump, jounce, Jlg (Jog, Jugg-Ie),
jangle (jingle).
[b-] 'dull impact': bang, bash, bounce, bijJ, bump, bat.
In the same vague way, we can distinguish finals:
[-es] 'violent movement': bash, clash, crash, dash, flash, gash,
mash, gnash, slash, splash.
[-ejr] 'big light or noise': blare, flare, glare, stare.
[-awns] 'quiek movement': bounce, jounce, pounce, trounce..
[-imJ, mostly with determinative [-Tl. 'small light or nOise':
dim, flimmer, glimmer, simmer, shimmer.
[-omp] 'clumsy': bump, clump, chump, dump, frump, hump,
lump, rump, stump, slump, thump.
[-et), with determinative [-T], 'particled movement': balter, clat-
ter, chalier, spalter, matler, scatter, ratlle, prattle.
In this last instancc we sec a formaI peculiarity which confirms
our classification. In English morphology there is no general
restriction to the occurrence of [-T] or H] aS suffixes, and, in par-
ticular, they are not ruled out by the presence of [r, 1] in the
of the word: forms like brother, rather, river, reader, reaper or llttle,
ladle label are common enough. The symbolic roots, however, that
an [r], are never followed by the deterrninative
but take an H] instead, and, conversely, a symbolic root contamlllg
(I] is nevcr followed by [-Il, but only by [-Tl: brabble and blabber
are possible as English syrnbolic types, but not *brabber or *blabble.
246
MORPHOLOGIe TYPES
The of minute features, such as the mor-
to be uncertain and incomplcte, because a
phonetlC slmllarlty, such as say the [b-J in box b t b
sent 1" . t f " , ea, ang, repre-
. .s a. lllgUlS le onn only when it is accompanied by a sernantic
slmllanty, and for this 2ast, which belongs ta the practicaI world
we have no standard of measurement. '
CHA PTJ<JR 15
SUBSTITUTION
15. 1. IIaving surveyed sentence-types (Chapter 11) and con-
structions (Chapters 12, 13, 14), we tum now ta the third type of
meaningful grammati cal arrangement, substitution ( 10.7).
A substitu/e is a linguistic fonn or grammatical feature whieh,
under certain conventional circumstances, replaces any one of a
class of linguistic forms. Thus, in English, the suostitute 1 replaces
any singular-number subst,antive expression, providcd that this
substantive expression denotes the speaker ')f the uUerance in
which the substitute is used.
The grammatical peculiarity of substitution consists in selective
features: the substitute replaces only forms of a certain class, which
wc may call the domain of the subst.itute; thus, the domain 'of the
suostitute 1 is the English forrn-class of substantive expressions.
The 8uhstitute differs from an ordinary linguistie form, sueh as
tking, person, object, by the fact that its domain is grammatically
definable, \Vhether an ordinary fOTIn, even of the most inclusive
meaning, such as tking, can be used of this or that practical situa-
tion, is a practica.l question of meaning; the equivalence of a sub-
stitute, on the other hand, is grammatically det.ermined. For
instance, no matter whom or what we address, we may menton
this real or pretene hearer in the foml of a substantive expre8sion
by means of the substitute you - and for this l'le nccd no practical
knowledge of the persan, animal, thing, or abstraction that wc are
treating as a hearer.
In very many Ci18es, substitutcs arc marked a180 by other pecul-
iarities: they arc ofkn short words and in many languages atonie;
they often h"ve irregular inf!ection and dcrivation (I : me : my) and
special syntactic constructions. In many languages they appear
as bound fonns and may then be characterized by morphologie
features, such as their position n structural arder.
16.2. One clement in the meaninl'; of every substitut is the
class-meaning of the form-class which serves as the domain of the
substitute. The class-meaning of the substitutc you, for example,
247
248 SUBSTITUTION
SUBSTITUTION 249
lS the class-meaning of English substantive expressions; the class
meaning of J is that of singular substantive expressions, and the
class-meaning of the substitutes they and we is that of plural sub-
stantive expressions.
Sorne substitutes add a more specific meaning whieh does not
appear in the form-class, but even in these cases a set of several
substitutes systematically represents the whole dornain. Thus,
who and what together caver the class-meaning of English sub-
stantive expressions. In the same way, he, she, and it together
cover the c!ass-meaning of singu!ar substantive expressions; within
the set, he and she cover the same sub-domain as who, and il the
same sub-domain as whai, but the distinction betwcen he and she
implics a further and indcpendent subdivision. Our selection of
substitutes, then, divides Englsh substantive expressions into the
sub-classes of personal (replaced by who and he-she) and non-
personal (replaced by whai and ii), and it subdivides the personal
sin!!;ulars into the sub-c!asses of male (replaced by he) and female
(replaced by she).
In addition ta the class-meaning, every substitute has another
element of meaning, the subslituiion-Iype, which consists of the
conventional circumstances undcr which the substitution is made.
Thus, J replaces any singular substantive expression (this domain
gives us the dass-meaning of 1), provided that this substantive
expression denotcs the speaker of the very utterance in which the
J is produced: this is the substitution-typo of J. The eireumstances
under which a substitution is made aro praetical circumstances,
which the linguist, for his part, caIlnot accurately define. In de-
tail, they differ grcatly in different languages; in speaking a foreign
langtlal:,'1O, wc have great difficulty in using the propel' substitute-
fOrIns.
15. 3. Nevertheless, it will he worth our while to leave, for a
moment, the ground of linguistics, and to examine the problems
which here confront the student of sociology or psychology. Wc
find, at once, that the varions types of substitution represent
elementary cireumstances of the aet of speech-utterance. The
substitution-types in J, we, und you are based upon the speaker-
heurer relation. The types of this, here, now and that, there, then
rcprcsent relations of distance from the speaker or from the speaker
and the hearer. The interrogative type of who, what, where, when
stimulates the hearer to supply a The negative type
of nobody, no/king, nowhere, never excludes the possibility of a
speech-form. These types are remarkably widesprcad and Ilniform
(except for details) in the languages of the world; among them wc
find the practical relations to which human beings respond more
uniformly than to any others - numerative and identificational
relations, such as positivc-negative, aU, sorne, any, same, olher,
and, ubave aIl, the numbers, one, two, three, and so on. These
are the relations upon which the language of science is baBCd;
the speech-forms which express them make up the vocabulary of
mathematics. Many of these substitution-types have to do with
species and individuals: they select or identify individuals (ail,
sorne, any, ea<:h, every, none, and sa on) out of a species. Perhaps
every language has a form-cluss of object-expressions, with a
class-meaning of the type 'species occurring in individuaI speci-
mens.' Accordingly, the substitutes for object-expressions, pro-
will usually show the most varied substitution-types.
In En!!;lish, where object-expressions are a special part. of speech,
the noun, the substitutes for the noun make up a part of speech,
the pronoun; together, these two constitute a greater part of
speech, the substantive. The pronouns differ from nouns, for one
thing, in not bciqg accompanied by adjective modifiers ( 12.14).
To a large cxtent, sorne substitution-types are characterized,
further, by the circumstance that the form for which substitu-
tion is made, has occurred in reeent speech. Thus, when we say
Ask that policeman, and he will tell you, the substitute he means,
among other things, that the singular male substantive expression
which is replaced by he, has been recently uttered. A substituto
which impljes this, is an anaphoric or dependent substitute, and
the recently-uttercd replaced form is the antecedent. This dis-
tinction, howevcr, sccms nowhere ta be fully carried out: wc
usually find sorne independent uses of substitutcs that are ordinarily
dependent, as, for instance, the independent use of it in it's
ing. Independent substitutes have no antecedent: they tell the
forrn-class, and thcy may even have an elaborate identificational
or numerativc substitution-type - as, for instance, somebody, nO-
body - but they do not tell which form of the class (for instance,
which particulaI' noun) has been replaced.
On the whole, then, substitution-types consist of elementary
features of the situation in which speech is uttered. These fea-
turcs are so simple that, for the most part, they could he indicated
250 SUBSTITUTION SUBSTITCTION 251
by gestures: l, yon, this, that, none, one, two, aU, and so on. Es-
pecially the substitutes of the 'this' and 'that' types rescmble
interjections in their semantic closeness ta non-linguistic forms
of response; li ke interjections, they occasionally deviate from the
phonetic pattern of their language ( 9.7). Since, aside from the
the substitution-type represents the whole mcan-.
ing of a substitute, wc can safely say that the meanings of sub-
stitutes arc, on the one hand, more inclusive and abstract, and,
on the other hand, simpler and more constant, than the meanngs
of ordinary linguistic fonns. In their class-meaning, substitutes
are one step farther removed than ordinary forms from practical
reality, since they designate not l'eal abjects but grammatical
form-classes; substitutes are, sa ta speak, lingui:\tic forms of the
second degr8e. In their substitution-type, on the other hand, sub-
stitutes are more primitive than ordinary linguistic forms, for
they dcsignate simple features of the immediate situation in
which the speech is being utt.ered.
The practical usefulncs8 of substitution is casy ta see. The sub-
stitute is used more often than any one of the forms in its domain;
consequently, it is easier ta spcak and to recognize. Moreover,
substitutes are often short forms and often, as in English, atonie,
or, as in French, otherwise adapted ta quick and casy utteranee.
In spite of this economy, substitutes often work more safely and
accurately than specific forms. In answcr to the question Wou/d
you like some fine, fresh cantaloupes? The answer How much are
cantaloupes? is perhaps more likely to be fol1owed by a delay or
aberration of response (" misunderstanding") than thc answer
How much are they? This is especial1y truc of certain substitutes,
such as l, whose meaning is unmistakable, while the aetual men-
tion of the speahr's name would ruean nothing ta many a
hearer.
15.4. Returning ta the ground of lnguistics, wc may be some-
what bolder, in view of what wc have seen in our practical excur-
sion, about stating the mcanings of substitutes. \Ve observe,
also, that in many languages, t.he meanings of substitutes recur
in other forms, such as the English limit.ing adjectives ( 12.14).
The meaning of the substitute yon may be stated thus:
A. Class-meaning: the same as that of the form-class of sub-
stantive expressions, say' object or abjects';
B. Substitution-type: 'the hearer.'
The meaning of t.he 8ubstitute he may be stated thus:
A. Class-meanings:
1. Definable in terms of fonn-classes:
(a) the same as that. of the form-class of singular
stantive expressions, say 'one object' ;
(b) the same as thai of t.he form-claEs defined by the
substitut.es who, somcone, say' personal';
2. Crcating an otherwisc llllestablished form-class: he is used
only of cerhtin singular pf'rsonal abjects (the rest are re-
placed, instead, by she), whieh, accordingly, constitute
a sub-class wi th a class-meaning, say 'male';
B. Substitution-types:
1. Anaphora: he implies, in nearly aU iLs uses, that a
substantive designating a specics of male personal ob-
jects has recently been uttered and tha-t he means one
individual of this spedes; say' recently mcntioned';
2. Limitation: he implies that the individual is identifiable
from among ail the individuals of the species mentioned;
this element of meaning is the same as tll!1t of the syn-
tacUc category of definite nouns ( 12.14) and can he
sta say, as 'ident.i fied.'
15.5. Subst.itules whose substitution-typc consists of nothing
but anl1phora, are (simple) anaphoric substitutes: apart from
their class-mcanings (which differ, of course, according cO the
grammutieal form-classes of differcnt languages), they say only
that the particular form which is being replaced (the antecedent)
has just been mentioned. ln English, finite verb expressions are
anaphoricl111y replaced by forms of do, does, did, as in Bill will
misbchave just as John did. The antecedent here is misbehave; ac-
cordingly, the replaced form is misbehaced. A few English verb-
paradigms, such as be, have, will, shall, can, may, must, lie outside
the domain of this substitution: Bill will be bad just as John was
(not did). Nouns, in English are anaphorical1y replaced by one,
plural ones, provided they are accompanied by an adjective at-
tribute: 1 prefer a hard pencil to a soft one, hard pencils to soft ones.
This use of one as an anaphorc pronoun differs by class-cleavage
from the several attributive uses of the ward one ( 12.14), espe-
cially in forming a. plural, ones. The details of this anaphoric sub-
stitution will conccrn us later ( 15.8-10).
In subordinute clauses introduced by or than, we have in Eng-
252 SUBSTITUTION SUBSTITUTION 253
lish a second kind of anaphora for a finite verb expression: wc say
not only Mary dances belter Ihan Jane does, but also Mary dances
better than Jane. We can describe this latter type by saying that
(after as and than) an actor (Jane) serves as an anaphoric sub-
stitute for an actor-action expression (Jane dans) , or we can say
that (after as and than) a zero-feature serves as an anaphoric Ruhsti-
tute for a finite verb expression accompanying an actor expression.
Another case of an anaphoric zero-featurc in English is the replace
w
ment of infinitive expressions after the preposition to (as in 1
haven't seen it, but hope 10) and after the finite verbs which take
an infinitive attribute without la (as in l'll come if 1 can). Simi-
larly, wc have zero-anaphora for participles after forms of be
and have, as in You were running faster than 1 was; 1 haven't seen if,
but Bill has. Zero-anaphora for nouns with an accompanying
adjective occurs freely in English only for mass nouns, as in 1 like
sour milk bei/er than fresh. For other nouns we use the anaphoric
one, ones, except after certain limiting adjectives.
While sorne forms of simple anaphoric substitution seem to
occur in every language, there are grcat differences of detai!. The
use of one, anes, is peculiar to English; related languages of similar
structure use zero-anaphora quite freely for nouns after adjectives,
as, German grosze Hunde und kleine ['gro:se 'hunde unt 'klajne]
'big dogs and Httle ones'; French des grandes pommes et des pdites
Ide grand pDm e de ptit] 'big apples and smaU ones.' In sorne lan-
guages the subject in the fuU sentence-types can he replaced by
zero-anaphora; tlms, in Chinese, to a statement like [wo
3
'jUI)4 i2
khwaj 'pu
4
] '1 need one pieee (of) cloth,' the response may he
['j Ull 4 i
4
phi 1 mo?] , Nced one roll (interrogative particle)?' In
Tagalog this happens in subordinate clauses, as in the sentence
[al) aj halJ'gaIJ sa mag
J
bu:l)a] 'the tree (predi-
cative particle) grew until (attributive particle) bore-fruit.'
16. 6. Perhaps ail languages use pronominal substitutes which
combine anaphora with definite identification: the replaced form
is an identified specimen of the species named by the anteeedent.
This, we have seen, is the value of the English pronoun he, as in
Ask a policeman, and he will tell you. Substitutes of this kind are
often, but misleadingly, called "anaphoric"; a better oame would
he definite. In most languages, including English, the definite
substitutes are not used when the antecedent is the speaker or the
hearer or includes these persons; for this reason, the definite
ij
:1
substitutes arc often spoken of as substitutes. They
usually share various peculiarities with the substitutes that refer
ta the hearer and to the speaker.
The English definite or thrd-person ptonouns, he, she, it, they,
differ for sngular and plural replaced forms, and, in the singular,
for personal and non-perMnal anteeedents: personal he, she, versus
non-personal it. Wc have seen that the difference of singular and
plural is otherwise also recognized by the language (as, for instance,
in the inflection of nouns: boy, boys), and we shall see that the same
is true of the dfference of personal and noo-persooal. Within the
personal class, however, the distinction between he used with a
male antecedent, and she, with a female antecedent, is otherwise
imperfectly recognized in our language (as, in the use of the suffix
-ess, 14.7). The distinction, then, between the pronoun-forms
he and she, createR a classification of our persanal nouns into male
(defined as those for which the definite substitute is he) and female
(similarly defined by the use of the substitute she). ScmantieaUy,
this classification agrees fairly weIl with the zoological division
into sexes.
In languages with ( 12.7), the third-person pro-
nouns usually differ according to the gender of the antecedent.
Thus, in masculine nouns, RUch as der Afann [der 'man]
'the man,' der HuI [hu:t] 'the hat,' have the sub-
stitute er [e:r], as when e'r ist grosz [e:r ist 'gro:sJ 'he, it is big,' is
said of either a man or a hat, or of any other antecedent that
belongs to the "masculine" congruence-class;
feminine nouns, such as die Frau [di: 'fraw] 'the woman,' die
Uhr [u:rJ 'the clock,' have the third-person substitute sie [zi:],
as in sie ist grosz, 'she, it is big';
neuter nonns, such as das JJaus [das 'hawsJ 'the house,' or das
Weib [vaj p] 'the wornan, ' have the third-pcrson substitute es
[es], as in es ist grosz.
This distinction, unlike that of he and she in Enp;lish, accords
with a distinction in the fotm of noun-modifiers (sueb. as der : die:
das 'the').
The meaning of dcfinite identification - that is, the way in
which the individual specimen is identified from among the species
named by the antccedent - varies for dilIerent languages and
would probably be very hard to define. It is important ta notice,
however, that in lanp;uages which have a category of "definite"
2M SUBSTITUTION SUBSTITUTION
255
noun-modifiers (such as, in Eng"lish the, this, that, my, John's,
etc., 12.14), the definite pronoun identifies the individual in the
same fashion as a definite modifier identifies it.s head oouo; thus,
a he after the antecedent policeman is equivalent in denotation,
except for the peculiar value that lies in the use of a substi tu te, to
the phrase the policeman. We need mention only a few widespread
pcculiarities, such as the case, not very common in English, t.hat
the definite pronoun is spoken before its ant.eccdent: He is foolish
who says so. If the antecedent is a predicate complement after a
form of the verb to be, the definit.e pronoun is normally it, regardless
of number, personality, or sex: it was a two-storey house; it's he;
it's me (l), it's the boys. Instead of an infinitive phrase as an actor
(to seold the boys was foolish), wc more cornmonly use il, with the
infinitive phrase following in close parataxis ( 12.2): il was foolish
to seold the boys. An act.or-action phrase, such as you can't come,
does not serve as an actor; but. docs appear in close parataxis with
it as an actor: it's too bad 1I0U can't come. This antieipatory use of
the definit.e pronoun extends, in German, to almost any act.or, with
the restriction that the pronoun cornes first; thus, beside ein Mann
kam in den Garten [ajn 'man 'ka:m in den 'gart.en] 'a man came
into the garden,' there is the fonn es kam ein Mann in den GaTten,
where the use of es resembles the English use of the adverb there.
If the noun in parataxis is plural, this German es accornpanies a
plural verb: beside zwei Mdnrwr kamen in den Garten [t.svaj 'mener
'ka:mcn] 'two men came inta the gardcn,' thcre is the form es
kamen zwei dnner in den Garten.
ln French, the definite pronoun replaces an adjective: les-vous
heureux? - je le suis [e:t vu r0? - fa 1 sqi,] 'Arc you happy?
- 1 am.' A step beyond this, we find dcfinite pronouns in marginal
uses without any antecedent, as in English slang beat il 'run away,'
cheese it 'look out,' he hot-Jooted it home 'he mn home,' let 'cr go.
Wc use they as an act,or for people in gcneral: they Bay Smith is
doing very well. The commonest use of this sort is the pseudo-
impersonal use of a definite pronoun as a merely formai actor, in
languages that have a favorite actor-action construction: it's min-
it's a shame. This may occur a.Iongside a genuine impcrsonal
construction ( 11.2). Thus, in German, beside the genuinc im
personal mir war kali [mi:r va:r 'kalt] 'to-me was cold; 1 felt
cold,' hier wird getanzt ['hi:r virt ge'tantst] 'here gets danced;
there is dancing here,' the dcfinite pronoun es may appear as an
actor, provided it COrnes first in the phrase: eswar mir kalt; es wird
hier getanzt. In Finnish, the impersonal and the pseudo-mpersonal
are used for different meanings: puhutaan 'there is talking' is a
genuine impersonal, but sadaa 'it's mining' contains a definite sub-
stitute actor 'he, she, it,' just as does puhuu 'he, she, it is t.alking.'
15. 7. The dcfinite substitutes in mosl languages arc not used
whcn the replaccd form designates the speaker or the hearer or
groups that include these persons; in this case a different type, the
personal substitute is used. The just-person substitute 1 replaces
mention of the speaker, and the sccond-person substtute thou,
of the hearcr. These are independent substitutes, requiring no
antecedent utterance of the replaeed form.
In addition ta the land thou substitut.es, most languages use
aIso forms for groups of people that include the speaker or the
heurer or both. Thus, in English, for a group of people whieh
includes the speaker, the substitute is wc; if the speaker is not
inc1uded, but. the hearer is, the substit ute is !Je. l'vI any languages
distinguish a11 t.hrce of these possibilites, as, Tagalog, which,
side [a'kuJ 'l'and [i'kaw] 'thou,' has the plural-liko forros:
speakcr only included (exclusive first persan plural): [ka'mi] 'we'
speaker an hearer inc1uded (inclusive first person plural):
['ta:ju] 'we'
hearer only inc1uded (second person plural): [ka'ju] 'ye.'
Similarly, languages which distinguish a dual nurnber, allow of
five combinations, as in Samoan: 'I-and-he,' 'I-and-thou,' 'ye-two,'
'l-and-thou-and-he (-or-they),' A
few languages distinguish also a trial number (' threc persons') in
their personal pronouns.
The English forms thou, ye are, of course, archaic; modern
English is peculiar in using the saIlle form, you, bath for thc hearer
and for a group of persons that includes the hcarer.
Many languages use different sccond-person substitutes
cording ta different social relations bctwccn speaker and hearer.
Thus, French uses vous [vu] 'you' much likc English, for bath
singular and plural, but if the hearer is a near relative, an intimate
friend, a young child, or a non-human bcing (such as a god), there
is a special intimate toi [twa]. German uses the third-
person plural pronoun 'they' for both singular and plural second
person: Sie spaszen [zi; 'spa;scn] is both 'they are jesting' and' you
(singular or plural) are jesting,' but the intimate forms, used much
256
SUBSTITUTION SGBSTITUTION 257
like those of French, distinguish singular and plural: du spaszest
[du: spa:scst] 'thou art jesting,' ihr spaszt [i;r spa:st] 'ye are jesting.'
The meaning of second-person substitutes is limited in sorne
languages by the eircumstance that they are not used in defer
cntial speech; instead, the hearer is designat.ed by sorne honorific
term (your Honor, your Excellency, your Majesty). In Swedish or
in Polish, one says, for instance, 'How is 111other feeling?' or 'Will
the gentleman come to-morrow?' where the terrns here italicized
denote the hearer. Sorne languages, sueh as Japanese and Malay,
distinguish several substitutes for both first and second persons,
according to deferential relations between speaker and hearer.
The personal substituteR and the definite (" third-person ") sub-
stitutes in many languages group themselves, by virtue of common
features, into a kind of closed system of personal-definite substitutes.
In English, both sets he, she, it, they and 1, we, you (thou, ye), are
atonic in the phrase; most of them have a special accusative case-
form (me, us, him, her, them, thee); most of them derive their pos-
sessive adjectives irregularly (my, our, your, his, her, tMir, thy)"
and sorne of these adjectives have a special form for zcro anaph-
ora (mine, etc., 15.5). ln French, the personal-definite pronouns
have special (conjunct) forms when they serve as actors or
goals of verbs ( 12.12); these have case-infl.ection for different
positions, which is otherwise foreign to French substantives; more-
over, they underlie possessive adjectives, as moi [mwa] 'l,' mon
chapeau [mon sapa] 'my hat,' while other substantives do not;
le chapeau de Jean [10 sapo d zan] 'the hat of John; John's hat.'
Very commanly the personal-definite substitutes have special
syntactic constructions. Thus, in English, German, and French,
the finite verb has special eongruence-forms for different persons
as actors; 1 am: thou art: he is; French mus savons [nu savon] 'we
know,' vous savez [vu save] 'you know,' elles savent [el sa:v] 'they
(feminine) know,' ils savent [i sa:v] 'they know.'
The personal-definite pronouns may even have a fairly sys-
tematie structure. Thus, in the Algonquian languages, an initial
element [ke-] appears in the forms that include the hearer; if
the hearer is Dot included, [ne-J denotes the speaker; if neither
is included, the initial is [we-], as, in Menornini:
[kenahJ 'thou' 'we' (inclusive) 'ye'
[nenah] 'l' 'wc' (exclusive)
[wenah] 'he' [wenua?] 'they.'
Samoan, with a distinction of dual and plural numbers, has:
[a?u] 'l' [ima;uaJ 'we t,vo' (excI.) [ima:tou] 'wc' (excl.)
[ita:ua] 'we two' (inel.) [ita;tou] 'wc' (ine!.)
[?oo] 'thou' [?ouluaJ 'ye two' [?outou] 'ye'
lia] 'he' [ila:ua] 'they two' [ila:tou] 'they.'
The dual-trial-plural distinction appears in the language of
Annatom Island (l\lelanesian):
[ainjak] 'l,' [aijurnrau] 'we two' (excI.), [aijumtai] 'we three'
(excl.), [aijama] 'wc' (exc!.),
[akaijau] 'we two' (ine!.), [akataijJ 'we three' (ine!.) , [akaija]
'we' (inc!.),
[aiek] 'thou,' [aijaurau] 'ye two,' [aijautaij] 'ye thrce, , [aijaua]
'ye,'
[aienl 'he,' [araul 'they two,' [ahtaijJ 'they three,' [ara] 'they.'
In many languages, persond-cfinite substitutes appear as
bound forms. Thus, Lat.in had definit.e-personal actors or goals
in the finitc verb-farms:
amo '1 love,' amas 'thou lovest,' amal 'he (she, it) loves,'
amtimus 'wc love;' am.dtis 'ye love,' amant 'they love,'
amor '1 am loved,' amiiris 'thou art loved,' amtitur 'he (she, it)
is loved,' amamur 'wc are loved,' amtimini 'ye are loved,' amantur
'they are 10ve,(1..'
Sorne languages, in the same way, include both actor and
goal, as Cree; [nisa:kiha;w] '1 love him,' [nisa:kiha;wak] '1 love
them, , [kisa:kiha;,,,] 'thou lovest hirn,' [nisa:kihik] 'he loves me,'
[niRa:kihikuna:nl 'he loves us (excl.),' [kisa:kihitina:n] 'we love
thee,' [kisa;kihitin] '1 love thee,' and so on, through a large
paradigm.
Likewisc, in Cree, the possessor of an objeet. appears in a bound
farm: [nitast.utinj 'my hat,' [kitastutinJ 'thy hat,' [utastutin]
'his hat,' and so on. In ail these cases, the third-person bound form
may stand in cross-reference with a nouo anteccdent; Latin paier
anw.t 'father he-loves; the father loves' ( 12.9).
The personal-definite system may be elaborated by distinctions
of identity and non-identity, such as thc difference of me and my-
self, where the latter forto implics identity with the actor (1
washed myself, 12.8), or the Scandinavian 'his' and sin
'his (own).' These differcnccs appear also in bound forms, as in
the obvin.tive forms of Algonquian ( 12.8); similar1y, ancient
Groek, bcsidc an ordnary bound actor, as in ['clowscJ 'he washcd,'
258 SUBSTITUTION SUBSTITUTION 259
Thus: ceite plume-ci [sd plym si] 'this pen,' dc ces deux plumes,
je prfre celle-ci celle-l [da se d0 plym, :1.;) prefe:r sd si a sd la]
'of these two pens, I prefer this one to that one'; but, of unnamed
things, je prfre ceci cela [sa si a so la] '1 prefer this to that.'
The pronouns without ci and l are confined to certain construc-
tions: de ces deux plumes, je prfre celle que vous avez [sd k;) vuz ave]
'of these two pens, l prefer the one you have'; independent: c'est
asscz [s Et ase] 'that's enough.'
Demonstrative substitution-types are not always fully distinct
from dcfinite, and, similarly, demonstrative limiting modifiers
may merge wit.h mere definite of the type 'the.' In
German, more than one dialect has only a single paradigrn whose
forms are used proclitically as a definite article, der Mann [der
'man] 'the man,' and with accent as a dcmonstrative limiting
adjective, der llfann ['de:r 'man] 'that man,' and as a pronoun,
dRr ['de:r] 'that one.' This last use, in German, is but slightly
distinguished from t.hat of the definite pronoun er [e:r] 'he'; the
chief diffel'ence, perhaps, is the use of der (not er) in the second of
two paratactic full sentences: es war einmal ein },[ann, der haile
drei Shne [es 'va:r ajnlma:l ajn 'man, de:r [hate Idraj 'z0:ne]
'there was once a man, he (literally, 'that-one ') had three sons.'
Many languages distinguish more types of demonstrative sub.
stitution; thus, sorne English dialects add yon, for things farthest
away, to the distinction of this and that. Latin had hic for things
nearcst the speaker, iste for those nearest the hearer; and ille
for those farthest away. The Kwakiutllanguagc makes the saillC
distinctions, but doubles the number by distinguishing a1so betwecn
'in sight' and 'out of sight.' Cree has [awa] 'this,' [ana] 'that,'
and [o:ja] 'that recently present but now out of sight.' Eskimo
has a whole series: [manna] 'this one,' [annal 'that one in the uorth,'
[qanna] 'that one in the south,' [panna] 'that one in the east,'
[kanna] 'that one clown there,' [sanna] 'that one down in the sea,'
[iI)I)u] 'that one,' and so on.
Outside of pronouns, we have the adverbial forms here: there,
hither : thither, hence : thence, now : then; the th-forms, however,
had a middle-voice form, where the actor is al. the ,'lame time affected
by the action: [e'lowsato] 'he washed himself' or 'he washed for
himself.'
Gther specializations are less common; thus, Cree, btside a
verb with actor and goal, such as [ninituma:w] '1 ask for him,
call him,' [ninitut:n] '1 ask for it,' and a form with actor and two
goals, [ninitutamawa:w] '1 ask him for it,' has also a form with
actor, goal, and interested person [ninitutamwa:n] '1 ask for il.
with refcrence to him,' that is, 'for his use' or 'at his hehest.'
15.8. Demonstrative or deictic substitution-types are based on
relative ncarness to the speaker or hearer. In English '',le have two
sueh types, for nearer and for farther away; they eoincide with
the values of the limiting adjectives this and that ( 12.14). De-
monstrative substitutes may he dependent (that is, they may refer
anaphorically ta an antecedent speech-form that names the
species), or independent. In cither eaoo, however, they identify
the individual objeet within the (named or unnamed) species.
Demonstrative pronoun substit.ution, in Eng-lish, is made by the
pronouns this, (these) , that (those) , which differ, by class-cleavag-e,
from the limiting adjectives, or by phra8es consisting of thesc
limiting adjectives plus the anaphoric one ( 15.5). These forms
are not ordinarily uscd to replace personal nouns - for the an-
t.icipatory use in Th is my brother; thcse are my brothers cannot
he viewed as persona!. The dependent substitutes in the singular
are this one, that one, and the independent this, that; hence we have
the distinction hetween, say, of these books, 1 like this one better
than that one, but, of unnarned objects, llike this belter ihan ihat.
In the plural, however, thCS8 and tho8e arc in either case used with-
out the anaphorlc ones.
In French we can see a more differentiated systcm. There are
three types of demonstrative limitation and substitution: a gen-
eraI type from which two special types arc differentiated by the
addition of the adverbs ci [si] for ncarer position and l [la] for
farther away. The forms of the limiting adjective, the depndent
pronoun, and the independent pronoun, lue distinct:
Al)JEC'l'IVE DEPENDENT INDEPENDE:N'l'
PRONO"N PRONOON
singular ce [sa]
masculine ce [sa] celui [sall1i]
feminine ceite [sd] [sd]
plural
masculine
feminine
ADJECTIVE
ces [se]
ces [se]
DEPENDE"T
PRONom'!
ceux [S0]
celles [sd]
INDEPENDE1'IT
1'1lONOUN
260 SUBSTITUTION S UB STITUT l ON 261
merge with simple anaphoric use, as in Going to Ihe circus? l'm
going there too. Similarly, so (and archaically also thus) is both
demonstrative and, more usually, anaphoric (l hope to do so).
Forms like (do il) this way, lht's sori (of Ihing) , this kind (of thing)
are on the border between substitutes and ordinary linguistic
forms.
16. 9. Inlerrogative substitutes prompt the hearer ta supply
either the species or the identification of the individual; in English,
accordingly, interrogative substitutes occur only in
questions. Of pronouns, we have the indepcndent who? (accusative
whom?) for personais and whal? for these ask for
both species and individual. For only we have also
the independent which? askng for identification of the individual
object from a limited field, but not for the species. The dependent
substitutes, asking for the identification of the individual from a
limited field, are which one? which ones?
Outside the pronouns, we have the interrogative substitutes
where? whith&T? whence? when? how? why? Interrogative
stitutes oceur in sorne languages, as in Menomini [wdsc:kcWll
'what sort is he? '
The limitation of interrogative forms to certain syntactic posi-
tions is quite common. Frequently \VC tind them restricted to
positions in the predicate of a binary The word-
order and the plural in who are they? what are those
things? arc features of this kind. In French, the
personal quoi? [kwalJ 'what?' is searceiy ever used as aetor or goal,
but instead, figures as a predieate complement, appearing in the
eonjunet form que [b], as in qu'eske que c'est? [k es ka SEi,] 'what
is it that this is? what's this?' and qu' qu'il a vu? [k E s k il
a vyII 'what is it that he has seen? what did he see?' In sorne
languages the interrogative substi tutes are always predicates of
equational sentences, as, in Tagalog, ['si:nu al) nagbi'gaj sa i'jull
'who the to you? who gave it to you?' or, in Me-
nomini [awe] pe:muhnetll 'who who is walk
ing there?'
16. 10. The various possibiIitics of sclecting individual abjects
from a species are rcprcscnted by al! manner of
especially of pronouns. In English, nearly aIl forms of this sort
consist of limiting adjectives with the anaphoric one, ones ( 15.5)
or of substantive uses, by of the same words. There
are many distinctions, not always rigidly carried out, bctween
dependent and indepcndcnt substitution, and in the latter, between
personal and classes. The various limitng adjectives
differ in treatment; those differenees add another line of classifica-
tion among them ( 12.14).
(1) Sorne lmiting adjectives are, like ordinary adjectives, fol-
lowed by one, ones to form anaphoric substitutes. 'Ve have seen
that this is the case of the sing:ular Ihis, Ihat and, undcr certain
conditions, of which? what? It is true also of each, every, whatever,
whichever, and of the phrasai expressions many a, such a, what a.
Thus, we say he was pleased wiih the chi/dren and gave each one a
penny. As independent substitutes wc use this, thal, which, what,
whichever, whalever of only; corresponding to every,
we have personal everybody, evcryone and everylhing;
each has no independent form.
(2) We have bath simple pronoun use or combination with the
anaphoric ones, one, in the case of eilher, former, latter, lasi, neilher,
other, such, and the ordinals, firsl, second, etc. The variants differ
chiefly in connotation. Thus, wc say Here are the books; lake eilher
(one). The word other forms a special in tilat it has a
plural form, others: Vou keep this book and TU take the others (the
other ones). In independent use these words serve chiefly as
personals.
(3) The remaining limiting adjectives arc pcculiar in not taking
the anaphoric one, ones. Thus, we say: Ilere are the books; take one
(two, three, any, both, all, afew, some, and so on). The independent
substitutes show great variety. 1'hus, aU is used as a
Ali is not That's all. On the other hand, one, as an atonie, is
personal: One hardly knows whai to say. Several form compounds
for indepcndent use, such as the personal somebody, someone, any-
body, anyone and the sornething, anyihing.
(4) Several limiting adjectives show an eecentric trcatment.
The article the with the anaphorie one, ones forms a depcndent
substitute, provided sorne other modifier follows: the one(s) on the
table; othcrwise it does not appear in pronominal use, and the
definite pronoun serves instead. The article a in combination
with another adjective does not influence the treatment of the
latter: many a one; anather (one). Otherwise, the article a is
accompanied by the anaphoric one only in the cmphatic farm
not a one. An other pronominal uses show us one repln.cing a: to
262 SUBSTITUTION
St'BSTITCTION 263
take an apple there corresponds the pronominal lake one. The deter-
miner no is paralleled by the dependent substitute none, but ordi-
narily we use instead the combination of not with any (I didn'/ see
any); the independent substitutes are the compounds nobody, no
one, no/hing (archaic naugh/).
Among thesc subst.itution-types, the negative is, of course,
represented in aU langUflges, and often shows special pecularitics;
to t belong alsu the non-pronominal nowherc, never, and sub-stand-
ard nohow. In many languages, as in most forms of sub-standard
English, these substit.utes arc accornpanied by the general negative
adverb: l ean't see nolhing. The nurllerative types (ail, one, two,
/hree, and so on) Beem also t be universal. As to the selective
types, however, t.here is great rOOIIl for varie(y; other languages
have substitution-types that are not exactly matched in English.
Thus, Itussian ['ne-xto] 'someone' implies ihat the speaker can
(but does not) identify the individual (' someone told Ille the other
day that ... '), whilc [xto-ni-'but] does not imply this ability
(' therc's someone at the doar '). Stll another type, ['koj-xto]
implies t.hat a different individual is 'iclected on differeni occasions
(' now and t.hen someone trics ').
16. 11. Substitutes frequently are tied up with special syntactic
functions; thus, we have seen that interrogative substitutes in
English and many other languages are confinud to certain positions
in the sentence. Some languages have special pronouns for predi-
cative use. Thus, in Menomini, besidc such fOrIlls as [nenah]
'l,' [enuh] 'that one' (animate), [eneh] 'that' (inanimat.e), there
are paraUd forms which occur only as prodicates; the normal sub-
stit ute appears in [kPhke;nam oneh] 'he-knows-it that (thing);
he knows that,' but the predicative farm in [ener ke;hkenah]
'that (thing) that is what he knows,' or in
[enui' ke;hkenah] 'that (persan) that one
is the one who knows it.' These predicative forms vary infiection-
ally for the same categories as a verb, such as interrogative [enet
ke:hkenah?] 'is it that which he knows? is that the thing he
knows?' or surprised present [cnesa? ke;hkenah!] 'and so that is
what he knows!' and sa on.
OUr relative substitutes belong ta a fairly widesprcad, but. by no
mcans universal type: the substitute indicates that the phrase in
which it figures is an included (or completive) form. In English,
the phrase has the favorite structure (actor-action
construction), and is marked by the relative substitute as not con-
stituting a full sentence. Our relatives who (whom), which, where,
when, that differ from other substitutes by class-clel1vage. They,
or their immediate phrase, come first in the clause. We have,
firstly, the anaphoric type, that, and personal who, non-personal
whieh: lhe boy who (that) ran away, the book which (tha/) he read;
the house in which we lived. If the relative 'iubstitut.e fills in its
clause the position of verbal goal, prepositional axis, or predicate
complement, wc have hore also a the man l saw,
the house we lived in, the hero he was. In ordinary speech, Englsh
relative clauses identify the individual antecedent; in more formaI
style we have also non-identifying relative clauses wit.h paratactic
sentence-modulation: the man, who was carrying a big bag, came up
ta lhe gale.
In languages with the infiection of the relative pro-
noun is normally determined by the forms in its clause: 1 saw the
bay who ran away; the boy whom l saw ran away. In Latin, a normal
form would be l:n hile vta quam nunc ego dego 'in this lite which l
now lead,' where the anteccdent, vila happens to he in the ablative
case (as axis of the preposition in), and the relative pronoun, quam
'which,' in the accusative case, as goal of the verb degJ. However,
lar:.guages with complicated inflechon now and then show attrac-
tion of t.he relative pronoun into an infiectional form that belongs
properly to the antecedent; the Latin form vHa in Mc qua nunc
ego dego, with the same denoiation as the above normal form, has
the relative pronoun qua in the ablative case, concording with the
antecedent, instead of the accusative case demanded by Hs position
in the clause.
Independent relative substitutes, having; no ant.ecedent, allow
the clause to replace an indication of species: lake what(ever) you
want; ask whom(ever) yon like; whoever says so is mistaken. In
English such clauses are used alsa as paratactic modifiers of a full
sentence: wha/ever he says, l don't believe him. The same difference
between dependeni and independent use appears in our adverbial
substitutes; dependent the lime (when) he dl it; the hause where we
lived; independent we'll see him when he gels here; we visit them
whenever we ean; we take them where(ver) we find them.
CHAPTER 16
AND LEXICON
16. 1. The mcaningful fcatures of linguistic signaling arc of two
kinds: lexical forms, which consist of phonerncs, and grammatical
forms, which consist of tagmemes (featurcs of arrangement,
10.5). If we extend the term lexical to cover aU forms that can he
stated in terms of phonemes, including even such forms as already
contain some grammatical features (e.g. poor John or duchess or
ran), then the parallelism of lexical and grammatical fcatures can
he exhibited in a set of tcrms like the followin/Z;:
(1) Smallest and meaningless unit of -linguistic signaling:
phememe;
(a) lexical; phoneme;
(b) grammatical: taxeme;
(2) Smal1est meaningful unit of linguistic signaling: glosseme;
the meaning of a glosseme is a noerne;
(a) lexical: morpheme,' the meaning of a rnorpheme is a
sememe;
(b) grammaticaL lagmeme; the meaning of a tagmeme is an
episememe;
(3) JVleaningful unit of linguistic signaling, smallest or complex:
linguistic form; the meaning of a linguistic form is a lin-
guislic meaning;
(a) lexical: lexical form; the meaning of a lexical form is a
lexical meaning;
(b) grammatical: grammalicalform; the meaning of a gram-
matical form is a grammatical meaning.
Every lexical form is connected in two directions with gram-
matical forms. On the one side, the lexical form, even when taken
by Uself, in the abstract, exhibits a meaningful grammatical
structure. If it is a complex form, it shows sorne morphologie or
syntactie construction (duchess, poor John), and if it is a rnorpheme,
it may still exhibit morphologie features (a modified morphcme,
e.g. men ?r ran, 13.7); in an unmodified mQrpheme (man, run)
we may VleW the absence of grammatical construction as a positive
264
FORl\l-GLASSES AN D LEXICON 265
characteristic. On the other side, the lexical fonu in any actual
utterance, as a coucrete linguistic form, is always accompanied by
sorne f!:rammatical fonn: it appears in sorne function, and these
privilcges of occurrence make up, col1cctively, the grammatical
functm of the lexical form. The lexical form appears in certain
sentence-types or, if it is a bound form, in none at aIl; it appears in
certain positions of (,'ertain constructions or, if it is an interjection,
in few or none; it appears as replaced farm in certain substitutions,
or, if it be a substitute, as substitute in certain substitutions. The
functions of lexical forms are created by the taxemes of selection
which help to make up grammatical forms. Lexical forms which
have any function in cOIllIllon, belong to a commonform-class.
The functions of lexical fonus appear as a very complcx system.
Some functions lire common to a great number of fonus nnd define
a large fonu-class; for instance, the functions which define the
English fonu-duss of substantive expressions (serving in the
sentence-type of caB, filling the positions of actor with a verb, of
goal ,vith 11 verb, of axis with a preposition; undcrlying a possessive
adjective, and 130 on), arc cornIllon to an almost unlimited number
of words and phrases. Different functions may create overlapping
form-classes; thus, the function of fi lling the actor position is
common ta substantive expressions and to marked infinitive
phrases (to scold the boyl! would be foolish). Other functions may be
limited to a very few lexical forms or to only a single one; thus,
phrases with the nounway as center stem to be the only substantive
expressions which fUIlction as advcrbs of manner, with the in-
terrogative substitute how? (this way, the way 1 do, and so on).
Particular lexical forms may, by clasil-cleavage ( 12.14) cxhibit
unusual combinations of function. Thus, eyy is in English a
bounded noun, (the eyy, an eyy) but oceurs also as a mass ooun (he
spilied eyy on his necktie). SaU is a mass noun and aeeordingly
underlies a plural only in the specialized mcaning 'kinds of,' but,
by class-cleavage, there is also a plural salts (as in Epsom salts)
with the meaning 'consisting of particles,' in a class with vats,
grils, and the like. Man is a (bounded, personal) male noun (a
man, the man, ... he), but by class-cleavage is treated also as a
proper noun, parallel in this with God, as in man wanls but little,
man il! a mammal. The ward one by a complieated class-eleavage
belongs to five as a detenuiner ( 12.14) it fulfils
the rcquircment that boundcd singular nouns be preceded by a
266
FRM-CLASSES AND LEXICON
FOR:''1-CLASSES AND LEXICON 267
modifier of this cIass (one house, one mile); as an ordimlry nurnera-
tive it oecurs with the definite determiners (the one man, this one
book, my one friend); it replaces a with anaphora of the noun
( 15.10) no other modifier is present (Here are sorne apples;
take one); It occurs as an indepcndent pronoun for' any person in
goneral' and in this use is always atonic and underlies the deriva-
tives one's and oneself (one can't help rmeself); tinall}', it is the
anaphoric substitute for nauns after an adjective, and in this US()
forms a plural, (the big box and the small one, these boxes and the
ones in the kitchen, ]5.5).
16. 2. gratnmar of a language includes, then, a very complex
set of habIts (taxemes of selection) by which every lexical fonn i8
used only in certain conventional functions; every lexical form is
assigned always to the customary form-classes. To dcscribe the
grammar of a language, we have to state the forrn-classes of each
lexical fonn, and to determine what characteristics make the
speakers assign it to these form-classes.
The traditional answer ta this quc:;tion appenrs in our schoo1
gramr.nars, which try to deline the form-classes by the c/ass-
meanmg- by the feature of meaning that is Cornrnon to aIl the
forms in the form-class. The school gramtnar tells us, for
Il noun is "the name of Il person, place, or thing."
rhlS dcfimtlOn presupposes more philosophical and scientific knowl-
edge than the human race can command, and impliCfi, fUl'ther, thl1t
the forrn-classes of a languagc agree with the classifications that
bc made bya philosophcr or scientist. ls fire, for instance,
a thmg? For ove1' a century physicists have believed it to he an
or proccss rather than a thing: \Inder this view, the verb
bum IS tnore appropriate than the nounfire. Our language supplies
the adjective hot, the noun heaf, and the verb to heat for what
physicsts belicve to be a movement of pal'ticles in a
bod!. school grammar defines the class of plural nouns
by ItS meamng "more than one" (person, placc, or thing), but
who could gather from this that oats s a plural ,vhile wheat is a
Class-meanings, like aIl other meanings, eludc the lin-
gUlst's power of dcfinition, and in general do not coincfdc with the
meanings of strictly-defincd techncal terms. To accept definitions
of mcaning, wbieh at best are makeshifts, in place of an identifica-
tion in formaI tcrms, is to abandon scientific discourse.
Class-meanings are merely composites, or, one might say, great-
est comTllon factors, of the grammatical meanings which accom-
pany the forms. '1'0 state a clnss-meaning is to find sorne formula
that includes the grammatical meanings in which the forms occur.
An English finite verb expression (runs, ran away, is very kind,
scolded the boys, and so on) occurs only in one position of one con-
struction, namely as action in the actor-action construction (John
ran away). Even when it is used a10ne, it appears only as a com-
pletive senknce which, accordingly, presupposes an actor. Now,
we can state the meaning of the actor-action construction very
roughly as 'A performs B,' where A is the nominative expression
(John) and TI the finite verb expresilion (ran away). This statc-
ment defines for us the meanings of the two positions; the mean-
ing of the actor-position is 'perforrner of n,' and that of the action-
position is 'performed by A.' Therefore, since English finite verb
expressions occur onIy and always in 'this latter position, their
class-meaning is the same as that of their one pORition, namely,
'performed by an object.' If we definc the class-meaning of the
larger forrn-class of verbs as 'action,' then the class-rneaning of
English finite verb expressions is '(action) performed by an actor.'
When a form-class has more than one function, its class-meaning
is harder to state, but iB still merely a derivativc of the grammatical
meaningfi in which the forms occur. English fiubstnntive expres-
sions occur, for instance, in the position of actor in the actor-action
construction (John Tan), with the positional meanng 'performer
of an action.' They occur in the position of goal in the action-goal
construetion (hit John), with a positional meaning something like
'underf,!;oer of an action.' They occur in the position of axis in
the relation-axis construction (beside John), 'with a pOilitional
meaninf,!; of, say, 'center from which a relation holds good.' They
Occur in morphologic construction with the possessive suffix
(John's), with the positional meaninf,!; of 'possessor.' "'WTithout list-
ing ail the other functions of :'::nglish substantive cxpressions, we
can say that the class-meaning cornIllon to aIl the lexical forms
in this form-class is 'that which can bc the perfo1'mer of an action,
the undergoer of an action, the center from which a relation holds
good, the possessor of objects,' and so on. Whether we can sum
this up in a shorler formula, depends upon our resources of
minology; for instance, we can sum up the class-meaning just
given, under the term 'object.'
These instances suffice to show that c1ass-meanings are not
268
FORM-CLASSES AND LEXICON FORM-CLASSES AND LEXICON 269
clearly-efinable unitH which could Hcrve as a basis for our work
but only vague situutonal features, undefinable in terms of ou;
science. The people who speak English and keep their substantive
expressions within the accepted functions, do not guide themselves
by deciding whether each lexical form denotes an object. Form-
classes, like other linguistic phenomena, can be defined, not in
ferms of meaning, but only in terms of linguistic (that is, lexical
or grammatical) features.
16. 3. The form-class of a lexical form is determined for the
speakers (and consequently for the relevant description of a
language) by the structure and constituents of the form by the
inclusion of a special constituent (a marker), or by the of
the form itsclf.
(1) A complex form is usually assigned to a by its
structure and An endocentric phrase, for instance,
such as fresh milk, belongs ta the same form-class as its head or
center ( 12.10). An exocentric phrase, such as in the house, Con-
tains .characteristic constituent (as, in our example, the
preposItlOn tn) wl",ich determnes its form-class. Thus, the forrn-
class of a phrase is usually determined, at bottom, by the form-
class of one Or more of the ncluded words. For this reason the
speaker (and the grammarian) need not deal separately with
each phrase; the form-class of almost any phrase s kno,vn if we
know the syntactie constructions and the forrn-classes of words.
The form-classes of words are therefore fundamental for syntax.
Our school grammar rccognizes this: it tries, by a mistaken method,
to be sure, to deterrnne the form-classes of words, partcularly
the most inclusive of these form-classes (parts of speech), and then
shows how phrases arc constructed.
(2) Somctimes the function of a phrase is determined by sorne
constituent, a marker. For inst.ance, in English, a phrase
consJStmg of the preposition to and an infinitive expression helongs
to the special forrn-class of marked infinitive phrases func-
tion differs from t.hat. of unmarked infinitive since
they serve as actors (toscold the boys was foolish) and as attributes
of nouns, verbs, and adjectives (a chance to go; he hopes ta go; glad
to go). The determining adjectives form noun phrases which am
distinguished by closure : this fresh milk cannot take adjecti ve modi-
fiers as Cao fresh milk or milk ( 12.10). 1Vhcnever a form-class of
small extent determines a peculiar function in phrases, we may
regard its forms as markers. Thus, our dctermining adjectives,
our prepositions, our co-ordinating conjunctions, and our subor-
dinatillg conjunctions, may be viewed as markers; they are smalt
form-classes, and the presence of any of their forms in a phrase de-
termines something about the form-class of this phrase. Other
amples of markers are the particles of Chinese or Tagalog ( 12.13).
(3) Finally, lexical forrns may helong arbitrarily or irregularly
to a form-class that is indicated neithcr by their st.ructure nor
by a marker. For instance, the phrase in case has the structure
of preposition plus substantive and yet serves as a subordinating
conjunction: In case he isn't there, don't wail for him. The phrases
this way, tkat way, the other way, the same way have substantive
structure, but are used as verb-modifiers of the special sub-class
(manner) that has the interrogative substit.ute how? Simlarly,
quit.e a few English nouns and noun phrases serve as verb-modifiers
in the when? class, either alone or in phrases: Sunday, last winter,
tomorrow morning. The form-classcs of English words are largely
arbitrary: there is nothing to tell us that man, boy, lad, son, father
are male nouns, that run, bother are verbs, that sad, red, green
are adjectives, and so on. In particular, of course, the formclass
of every morpheme is arbitrarly det.ermined. A complete descrip-
tion of a language will list every form whose function is not
termined either by structure or by a marker; it will include,
cordingly, a lexicon, or list of morphemes, which indicat.es the
form-class of each morpheme, as weIl as !ist.s of ail complex forms
whose function is in any way irregular.
16.4. Form-classes are not mutually exclusive, but cross each
other and overlap and are included one within the ot.her, and so
00. Thus, in English, the nominative expressions (which serve as
actors) include both substantives and marked infinitives (/0 scold
the boys would be foolish). On the other hand, among the substan-
tives are sorne pronounforms which, by do
not serve as actors: me, us, him, her, them, whom. One group of
substantives, the gerunds (scalding) , belongs ta a form-class with
infinitives and with ot.her verh-forms, in serving as head for cer-
tain types of modifiers, such as a goal (scolding the boys). For
this reason a system of parts of speech in a language like English
cannat he set up in any fully satisfact.ory way: our list of parts
of speech will depend upon which functions we take ta he the most
important.
270 FORM-CLASSES AND LEXICON FORl\J-CLASSES AN D LEXICON
271
One can often distinguish, however, between great form-classes
like the above, and petty form-clases like that of foot, goose, tooth
or of ox (with irregular plural-forms). Large form-clusses which
completely subdivide either the whole lexicon or sorne important
form-class into form-classes of approximately equal size, are called
categories. Thus, the English parts of speech (substantive, verb,
adjective, and so on) are categories of our language. So are
singular and plural substanlives, since these t wo form-classes, of
approximately equal size, complet.cly subdivide the form-class of
substantives. In general, inflectionalOforms, what with the parallcl
occurrence in every paradigm, represent categories - for instance,
the various forms of the verb-paradigm, including the congruence-
forms of finite verbs (am: is: are or was : were) and, crossing
thesc, t.he tenses and modes of verbs (he is: hewas : he were).
Not aIl categories, however, are inftectional. The selection of the
pronouns he versus she divides our personal nouns int.o the cate-
gories of male and female; yet. there is no inflcction or regular
derivation to distinguish these, but only n sporadic use of markers
(count : countess, Paul: Pauline, Albert: Alberta) or of entirely
irregular derivation (duck : drake, goose : gander) or of composition
(he-goal, billy-goat, bull-buffalo) or suppletion (son: daughter, l'am:
ewe) or merely elass-cleavage (a teacher ... he; a teacher ... she;
Francis: Frances).
Again, sorne catel;ories are syntactic, and appcar not in inflec-
tion, but in phrases. Such are the categories of indefinite and
definite substantives (a book: the book), or, in our verbs, the as-
pects (wrote: was writing) , completion (wrote: had written) , or
voice (wroie : was written).
The categories of a language, cspecially those which affect
morphology (book: books, he : she), f1re so pervasive that anyone
who reficcts upon his language at ail, is sure to notice thern. In
the ordinary case, this person, knowing only his native language, or
perhaps some others closely akin to it, may mistake his categores
for universal forms of speech, or of "human thought," or of the
universe it.sclf. This is why a good deal of what passes for "Iogic"
or "metaphysics" is merely an incompetent restating of t.he chief
categories of the philosopher's language. A task for linguists of the
future will be ta compare the categores of different languages and
see what features are universal or at Icast widespread. Thus, a
form-class comparable to Our substantive expressions, with a
class-meaning something like 'object,' scems to cxist everywhere,
though in many languages it is not an arbitrary class, like our
substantive part of speech, but dcpends largely upon the presence
of markers, as in l\hbyan or Chinese ( 12.13).
16. 5. Our knowledge of the practical world may show that sorne
Iinguistic categories agree ,vith classes of real things. It ma:v be,
for instanc'e, that our non-linguistic world consists of obJects,
action8, qualities, manners, and relations, comparable wHh the
substantives, verbs, adjectives, adverbs, and prepositions of our
language. In this case it would still be truc, how,cver, many
othcr languages do not recognizc these classes III thmr part--of-
speech system. ::\'Ioreover, wc should still have to the
English parts of speech not by their correspondence wlih dtfferent
aspects of the pradical worId, but mercly by their functions in
English syntax. .
This appears plainly in w circuffiatance that languages wIth an
elaborate part-of-speech system always contain abstract forms;
they have parallel forms with the same Icxical meaning for.use. in
different synt.actic positions. Thus, a verb like l'un or an adJectivc
like smooth cannot scrve as an actor, but we have for this function
the abstract noun forms l'un (as in the l'un will warm you up) and
smooihness. It is an error to suppose that abstract forms like these
occur only in the languages of literate peoples; they occur in ail
languages that lmit different form-classes ta differcnt syntactic
positions.
Linguistic categories, thcn, cannat bc defined in phlosophical
terms; having defined them in formaI terms, we may have great
di ffi culty in describing thcir meaning. Ta show this, we need only
glance at sorne of the more familial' categories.
Numbcr, as it appears in our singulars and plurals, seems to. be
close to some universal trait of hurnan responsc; yet, cases hke
oals versus whcat, or Epwm salts versus table salt, seem to have
little non-linguistic justification.
The categorics of gender in English arc close to our non-linguistic
recognition of personality and sex, but even here some animais
(the bull . . . he or it) and other things (the good ship . . . she or
it) are variously treated. The gender-catcgories of most Inda-
European languages, such as the two of French or the three of
German ( 12.7), do not agree with anything in the practical world,
and this is truc of most Bueh classes. In the Algonquian languages,
272
FOR.M-CLASSES AND LEXICON FRM-CLASSES AND LEXICON 273
aIl persons and animaIs belong t() one category, an 'animate ,
gender, but so do sorne other objects, such as 'raspberry,' 'kettle,'
and 'knee'; aIl other objects (including, for instance, 'strawberry,'
'how1,' 'elhow ') beIong to the other, 'inanimate' gcnder. Sorne
of the Bantu languages mn up to as high as twenty such classes;
distinctions of number, however, are merged with the gendcr-
classification.
Case-categories, ranging from t-wo, aS in English (he: him) ,
up to twenty or sa, as in Finnish, resemble varions situations of the
practical world, but never with any consistency. Thus, in German,
the goal of a verb is in the accusative case, as in er bal mich [e:r
'ba:t mix] 'he asked me (for something),' but certain verbs have
it in the dative case, as er dankle mir [e:r 'daIJktc mi:r] 'he thanked
me'; compare the Latin examples in 12.8,
The categories of [ense have a surface rationality, espccially in a
language like Latin, which distinguishcs present (cantat 'he sings'),
past (canivit 'he sang'), and future (canUibit 'he will sing'), but
even here one soon finds that these categories disagree with our
non-linguistic analysis: the "historical present" i8 used in Latin,
as in English, of past events, and the meanings of the Latin mse-
forms are mixed up with considerations other than relative time.
The English categories of aspect distinguish betwecn 'punctual'
action (sorne grammarians cali it 'perfective '), envisagcd as a unit
(he wrote the letter), and 'durative' action (some calI it 'imperfec-
tive '), which extends over a segment of time during \"hich other
things can happen (he was writing the letter) , This distinction is at
best hard to define for the practical world, and in English suffers
marked dislocations; some verbs, for instance, appear persistently
in punetual fonn (1 think he is there; he funny) and are durative
only in special constructions or rneanings (I am Ihinking of him;
he is being funny). In Russian, which has much the same aspects
as English, certain verbs, such as 'eat' and 'drink,' appear per-
sistently in durative form.
A cornmon verb-category that is lackin!l; in En<J"lish is itemlion
h' ,
which distinguishes between an action occurring once and a
peated action, as, in Russian [011 be'zal do'rnoj]' he was running
home' (on one particular occasion) and [on 'bcgal do'moj] 'he
ran home; he was running home' (repeatedly, e.g. every day).l
1 In EngIish, iteration plays no part in the verb-form; he played tenn';" eecry <!ay
(punrtual) and wu. playi"1J tennis every <!al/ (durative) are like he played a 8.t of
Perfection contrasts contemporary, 'imperfectic' action with
1 perfcctic' action, whosc effect is contemporary: versus
he has wH'Uen; he writing versus he has been wntlng; he wrote
versus he had written; he waswring versus he had been writing,
The diffcrencc is scarcely definable in terms of practical situation,
and different languages show different di8tri butions.
English has IIlany modes, distinguishing various approaches of
an action to its aclual occurrence. l\lorphologically, English dis-
tinguishes between 'real' (he is here) and' unreal' (if he were here);
syntactically, English recognizes a whole series by the pcculiarity
of certain irrcgular ('auxiliary') verbs which are followed by an
infinitive without to: he will write, shall write, can wrile, must write,
may write. We may observe that in these combinationH the in-
finitive is rather persistently punctual, and only now and then
durative (1 shall be writing); in Russian, the futUre tense, which
corresponds fairly weIl ta our shall and will phrases, distinguishes
aspect just as exactly as do the present and past tenses. The uses
of different modes are tied up in many languages with differences
of syntactic position and congruence. In Engli8h, for instance, the
unreal appears only in clauses introduced by or though, or in
combination with the phrasaI mode-forms (he would help llnreal
of he will he/p us). Similar complication8 appear in the UReS of the
various modes of other languages, as, in French, je pense qu'il
vient [lI8 puDs k i vj!"] '1 think he is coming,' with the verb of the
clause in the' indicative' (actual) mode, but je ne pense pas qu'il
vienne [2:8 n pUllS pa k i vjen] '1 don't think he is coming,' with the
verb of the clause in the 'subjunctive' (possible) mode.
16. 6. We saw in 16,3 that the function of sorne forms is
det,errnined by their constituents or their construction. Any
tion that is so determined is said to be regular, and a function
which is not 80 determined is said to be irregular. Thus, if we know
that the words fox and ox are sinf';ular common nouns, wavering
betwoen non-personal and male personal gender, then we can say
that fox hU8 the regular function of combining with the plural-
suffix [-oz] in the form foxes (since this function is shared by an
unlimited number of singular nouns), but that ox has the irregular
tennis (punctual) and he was playill{] a ed of lennis (durative). In French,
and modern Greek, rcpcated action and durative action are merged III one clasa:
French il criMit [il ekrivtl is both 'he writ.ing' and' he (repeatcdly!;
he use<! to writc.' In RU8sian, rcpcu.tcd "ct.iolls are cias,cd M duratIve, but, Wl thm
the durative nee for certain from single actions.
274
FORM-CLASSES AND LEXICON
FORM-CLASSES AND LEXICON 275
function of combining with the plural-suffix [-V.]. Linguists usually
apply the terms regular and irregular to the form itself, saying, for
instance, that the noun fox is regular and the noun ox irrogular;
we must speeHy, of course, the function with respect to which
these tNms hold good, since in their other functions the nOUTIS fox
and ox are quitc alike. By another extension of these tenIls, lin-
guists apply thorn also t the resultant forms in which the functions
appear, saying, for instance, that the plural nouo foxes is regular
and the plural noun oxen irregular.
The speaker can use a forrn in a regular function even when he
has never heard the resultant form: he may uUer a form like faxes,
for instance, even when he has never heard this particular plural.
He can use a form in an irregular function only if he has heard it
used in this function: the form oxen is uttered only by speakers
who have hcard it from other speakers. In the description of a
language, accordingly, regular functions are stated for whole form-
classes, in the mass: wc can state the regnlar plural-formation of
English nouns without attempting to list aH the nonns in the
language. Irregular functions, on the other hand, force us to list
al! the forms of the class: we have ta mention the noun ox as
taking -en in the plural, and the nouns foot, tooth, goose as taking
substitution of [ijj in the plural, and so on.
lf we insist on this distinction, we may say that any form which
a speaker can uUer without having heard it, is regular in its irn-
mediate constitution and embodies regular funchons of its con-
stituents, and any form which a speaker can uUer only after he
has heard it from other speakers, is irregular. St.rict!y speaking,
then, every morpheme of a language is an irregularity, since the
speaker can use it only after heuring it used, and the reader of a
linguistic description can know of Hs existence only if it is listed
for him. The lexicon is really an appendix of the grammar, a list
of basic irregularities. This is all the more evident if meanings
are taken int consideration, sinco the meaning of each morpheme
belongs ta it by an arbitrary tradition. In a language like English,
where each morpheme ls arbitrarily assigned ta sorne grammatical
class, this feature also is an irregularity: the speaker must learn
from experience and the describer must list the fact that pin is
a noun, spin a verb, thin an adjective, in a preposition, and so on.
This task also is customarily assigncd t the lexiconj the grammar
!ists only the kinds of irregularity that are not present in aU
the morphernes of ft language, and the terms regular and irregular
are used ouly of features that appear in the gramrnar.
lf we rnake this restriction, it is obvious that most speech-forms
are regular, in the sense that the speaker who knows the
ents and the grammatical pattern, can utter them without ever
having heard thern; moreovcr, the observer cannat hope to list
thern, sinoo the possibilities of combination are practically infinite.
For instance, the classes of nominative expressions and finite
verb expressions in English are sa large that many possible actor-
action forms - say, a red-headed plumber bought five oranges-
may never before have been uttered; by the same token, however,
we cannot he sure that this is true of any particular combination
which we may chance > hear. A grammatical pattern (sentence-
type, construction, or substitution) is often called an analogy.
A regular analogy pcrmits a speaker to utter which
he has not heard; we say that he utters them on the analogy of
similar forms which he has heard.
An irregular analogy, on the other hand, may caver a number
of forms, but a speaker will rarely utter a new form on the analogy
of those which he has heard. For instance, the phrases at /.ell8t, at
most, at best, at worst, al first, at last are built up on the same pat-
tern (at plus adjective in -st), but the analogy is limited w a very
few forms. In al ail (wherc the adjective dacs not end in -st and
the sandhi is irregular) or in don't we have a unique analogy.
When the automobile came into use, one speaker was as well able
as another to form the compound automobile-driver, on the analogy
of cab-driver, truck-driver, and sa on; a compound like cranberry,
on the other hand, with its unique first member, is uttered only
by speakers who have heard it. lf we take meanings into considera-
tion, we can say the same of a speaker who uses the term blackbird
of the species of bird t which it cuswmarily applies, for the com-
pound bears this meaning by an arbitrary tradition. A form like
charlestoner 'one who performs the dance called charleston' is
fonned on the regular analogy of dancer, waltzer, two-stepper, and
so on; a form like duchess ( 10.6) is unique. On the border-line
we have cases like the feminines in -css, which on the whole are
limited to traditional forms: we say poetess, sculptress, but not
*paintress; occasionally, however, a speaker will extend this an-
alogy, uttering such forms as, say, profiteeress, 8Wnd/.eress. Even
our root-forming morphemes ( 14.9) have sorne fiexibility; hear-
276 FORM-CLASSES AND LEXICON
FORM-CLASSES AND LEXICON
277
ing a form like squunch in a.meanng 'step with suctionnoise on
wet ground,' we cannot tell whether the speaker has heard it or
is using the analogy of as in squirt, squash, and [-on],
as in crunch.
The regular analogies of a language are habits of substitution.
Suppose, for instance, that a speaker had never heard the form
give Annie the orange, but tbat he had heard or spoken a set of
forms like the following:
Baby t's hungry. Poor Baby! Baby's orange. Cive Baby the orange!
Papa is hungry. Poor Papa! Papa' s orange. Cive Papa the orange!
Bill is hungry. Poor Bill! Bill's orange. Give Bill the orange!
Annie is hungry. Poor A nnie! Annie' s orange. . ..' .
He has the habit, now, - the analogy, - of usng Annie in
the same positions as Baby, Papa, Bill, and aecordingly, in the
proper situation, will utter the new form Cive Annie the orange!
When a speaker utters a complex farm, wc are in most cases unable
to tell whether he has heard it before or has created it on the an-
alogy of other forms. The utterance of a farm on the analogy of
other forms is Iike the solving of a proportional equation with an
indefinitely large set of ratios on the left-hand side:
Baby is hungry : Annie is hUngry )
Poor Baby: Poor Annie = Cive Baby the orange: x
Baby' s orange : Annie's orange
or
dog : dogs 1 -
pickle : -pickles .
=radw'x
: .
pwno: pwnos
16.7. The power or wealth of a language consists of the mor-
phemes and the tagrnemes (sentence-typcs, constructions, and
substitutions). The number of rnorphemes and tagmemes in any
language runs weil into the thousands. In cvery language,
over, many complex forms carry specialized meanings which
cannot figure in a purely linguistic description but are practically
of grcat importance. The linguist can determine for instance , ,
that English compounds of the type blackbird, bluebird, whitefish,
or phrases of the type give out, fall out, throw up, bear specialized
meanings, but he cannot evaluat these meanings, although in
practica! life they are fully as llseful as aDY sememe.
popularly, the wealth of a language is supposed ta depend upon
the number of different words which it uses, but this number is
indcterminate, since words are freely formed according ta the
analogies of morphologie construction. For instance, having
counted play, player, and dance, shaH we count dancer as a fourth
word, even though it contains no additional glosseme? If so,
then the number of words in any language is practically infinite.
When wc are toId that Shakspere used 20,000 different words in
his writings, and Milton in his poems sorne 8,000, wc mistakenly
conclllde that less cloquent speakers use far fewer. It is an indica-
tion of Shakspcre's genius that he uscd sa many different words
in sa srnaII a volume of speech as is contained in his works, but
this volume of speech is srn:lll compared to the amount which
even a taciturn person will utter in the course of a year. The
myths about peasants, workingmen, or savages who use only
a few hundred words have no foundation in fact; in so far as one
can count words (ignoring, for instance, the inflected fomls of a
language Iike ours), every adult speaker uses at least somewhere
round 20,000 to 30,000 words; if he is educated - that is, if he
knows technical and learned words - he uses many more. Every-
one, moreover, Ilnderstands more words than he uses.
The relative frequency of the various lexical and grammatical
units (morphcmes and tagmemes) in a language can be studied
wherever \ve have copious records of normal utterances. In the
next chapters we shaH sec that our lack of such records is one of
the impediments to the historical study of language - for fluc-
tuations in the frequency of glossemes play an important part in
the changes that occur in every language.
The frequency of most lexical forms is doubtless subject ta a
great deal of superficial fluctuation, according to the practical
circumstanccs. A word Iike thimble, say, or stove, might not occur
at ail in lonf!; stretches of flpeech; yet such forms as these are used
by everyone- wheu the occasion presents itself. The most frequent
forms, on the other hand, both lexical and, especiaIly, grammatical,
are constantly demanded by the structure of the language. Such
counting as has becn done has been confined to words. 1t is found
that the commonest words (the, to, is, etc.) make up a consistently
high pcrcentage of what is spokcn.
16. 8. The practical question as to what things can be said in
different languages, is often confused with questions of word-
278 FORl\l-CLASSES AND LEXICON FRM-CLASSES AND LEXICON 279
meanings and of categories. One language will use a phrase where
another uses a single word and still another a bound fonn. A mean-
ing that is categoric in one language (as, for instance, plurality
of abjects in English) may appear only under particuhtr practical
stimuli in another Ill.nguage. As to denotation, whatever can he
saie! in one language caTI doubtless be s:d in an)' other: the differ-
ence will concern only the structure of the forms, and their con-
notation. 'Vhat one language expresses by a single morpherne
will in another language require perhaps a long phrase; what
one language says in a ward may appear in another language as
a phrase or as an affix. Elements of meaning that appear in one
language !.lecaliSe they belong ta some category, even thougb they
are irrc!evant to the praceal situation, will be absent in another
language. In English we say Fike's Peak is high with a
tense verb; in ChiIlDf1C or in Rllssinn there would be no
tiJnse element in a sirnilnr message.
Tt is a striking fact that the smallest units of signaling, the
semes, of different differ vastly in practical value. This
is true even of closely re!ated languages. 'Vhere we Sfty ride,
German says reiten ['rajten] for riding on an animal, but jahren
['fa:ren] for other kinds of riding, as in a vchicle. \Vhere wc say
on, German says auj when the force of gra:vity helps the contact,
as in 'on the table,' but othenvise an, as in 'on the wall.' Our
morning matches the French matin [matr
n
], cxcept when the
moming is viewed as a segment of time during which something
else can happen, as in ' l 81ept ail morning' or ' during the morning' ;
in this case French uses a derivative matine [matine]. Even things
which are easily defincd and classified, receive the most diverse
treatment in differcnt languages. Nothinp; could be more dcfinite
than terms for simple biologica! relatioIlship between persons.
Yet, beside wocds corresponding ta our bro/her and sister, German
has a plural Geschwister [ge'svister] that includes both sexes, as in
Wieviele Gesehwister haben Sie? [vi: 'fi:l ge'vister 'ha:ben zi:?]
'fIow many brothers and (or) sisters have you?' Sorne languages
have here one word, regardless of gendcr, as Tagalog [kapa'tid];
our brother corresponds to a Tagalog phrase [kapa'tid na la'la:ki],
where the lust ward means 'male,' and our sister ta [kapa'tid na
ba'ba:ji], with the attribute 'fomale.' On the other hand, sorne
languages insist upon relative age: Chinese ['ko
1
ka!] 'eider
brather,, ['cjU!)l ti
4
] 'younger brather,' ['cje
3
jc
3
J 'eider 'lister,'
['mej 4mej 4] 'younger sister.' An even more complicated
ogy appears in Menomini, which wo can best elucidate if wc use
the tenn sibling ta mean 'brother or 'lister.' In Menomini the
terms arc [ndnrh] 'rny eIder brother,' [neme:h] 'my eIder sister,'
[m:hse:h] 'ruy younger sibling,' [neko:?semaw] 'my sibling of
opposite sex' (i.e. 'my brother' when a woman says it, 'my 'lister'
when a man says it), [nc:hkah] 'my brother (man speaking),'
[ne:tckEh] 'my 'lister (woman speaking).' The general term
[ni:tcsj anak] 'InY si blings' is used in the plural when the siblings
are of both sexes and not ail younger than the possessor.
Terrns of relationship not only vary as in the above examples,
but also are used in situations that one cannat define. The Me-
nomini terms for 'brother' and 'sister' are used also for cousins,
provided the related parents arc of the same sex: a man says
[ne:hkah] of his father"; brother's son, and so on. Moreover, these
and some other terms are inherited: my father's brother's son's son
is also [ne:hkah]. Conseqllently, the rneaning rcally hinges on the
consistency with which these relationships are rcmembered and
recognized.
In the same way, plant-names, for example, are perhaps nowhere
used in a way that would be consistent with a botanist's
fieation - even aside from such vague terms as tree, shrub, bush,
herb, reed, grass.
Even in such a sphere as that of the numbers, languages show
many deviations. Our system of decimal numbers (tU'enty-two,
thirty-five, etc.) shows traces of a duodecimal or twelves system
(eleven, twelve instead of *one-teen, *tU'o-teen). Other irregularities
are fonnal, as tU'o : tU'enty : second: halj, or threc : thirtecn, thirty,
third. Furthermore, the connotation of certain numbers like three,
seven, thirteen, and of additional terms like dozen, score, gross, can-
not be stated mathematically. In Danish there is an admixture of a
vigesimal or twenties system. In French one counts from 'sixty ,
ta 'seventy-nine' without a special ward for the intervening
multiple of ten: 'seventy' lS soixante-dix [swasant-dis] 'sixty-ten';
'seventy-one' is soixante et onze [swasant e OUz] 'sixtY and eleven,'
and sa on; 'eighty' is quatre-vingt [katro ven] 'four-twenties,'
and then one counts up twenty more to rcach one-hundrcd; thus,
'ninety-two' is quatre-vingt douze [kati'd ven du:z] 'four-twentics-
twelve.' Peoples who have Ettle use for higher numbers may
Use very few: the Kham Bushmen are said to count by simple
280 FORM-CLASSES AND LEXICON
numbers only to 'three,' and to use 'two and two' for 'four,'
and sa on.
In other spheres which are subject to scientific analysis, this
may still provide no gauge for the linguistic classification. Color,
for instance, is a matter of frequency of refracted or reflected light-
waves. The visible spectrum is an unbroken scale of frequencies.
Different languages use diffcrcnt color-names (such as our red,
orange, yellow, green, blue, violet, 9.1) for different parts of this
scale. We should have a hard time deciding at what points on the
actual scale the domain of each English color-name begins and
ends. If we showed people colors in minute grades of wc
should flnd that betwecn the frequencies which were namcd con
sistently, say, as yellow and as green, there would be a harder-zone,
whcre the naming wavercd. If we went outside the European
culture-sphere, wc should tind entirely different distributions.
For most of our meanings we have not even this approach ta an
external standard. Terms which relate ta social behavior, such as
love, friend, kind, haie could be defincd in terrus of ethnology, folk-
lore, and sociology, provided studics had reached a perfect,ion
and accuracy undrcamed of today. Terms which relate ta states
of the speaker's body that arc perceptible only to him, such as
queasy, qualmish, ,md, gay, glad, happy, could be defined only if
we had a minute knowledh'" of what goes on inside a living person's
body. Even ail this would not sulfice for linguistic meanings that
have less practical bearing, such as categories of noun-gender or
verbal aspect. There seems ta be no practical criterion by which
the gender of a noun in German, French, or Latin could be
mined: ta define the meaning of the episememe 'masculine' in such
a language would be simply ta list the markers of masculine nouns
and the nouns that belong arbitrarily ta the class, and ta say that
whatever is common, in the practical world, to aIl these objects,
is the "meaning" of the masculine gender-catcgory. The same is
true of the verbal aspects of English: the difference between wroie
and was wriiing is fla elusive and differs sa much for different verbs
and in different phrases, that the dcfiner, after stating the main
prnciples, cannot do hettcr than to resort to a demoDstration by
means of examples.
CHAPTER 17
WRITTEN RECORDS
17. 1. The language of any appears ta an
observer as a complicated signaling-system, of the kind that has
occupied us in the prcccding chapters of this book. A language
presents itself ta us, at any one moment, as a stable structure of
lexical and grammatical habits.
This, however, is an illusion. Every language is undergoing, at
ail times, a slow but unceasing proccss of linguistic change. 'Ve
have direct evidence of this change in the case of communities
which possess written records of theii' earlier speech. The English
of the King James Bible or of Shakspere is unlike the English of
today. The fourteenth-century English of Chaucer is intelligible
ta us only if wc use a glossary. The ninth-century English of King
Alfred the Great, of whieh ,,"'e have contemporary manuseript
records, seems to us like a forcign language; if we could meet
English-speakers of that time, wc f'hould not understand their
speech, or they ours.
The flpced of linguistic change cannot be statcd in absolutc
tcrms. A flpeaker has no dilficulty, in youth, in conversing with his
grandparents, or, in age, in conversing with his grandchildrcn, yet
a thousand years - say, thirty ta forty generations - have
suffieed to change the English language to the extent we have just
indicated. During these generations, it must havc seemed ta each
London-English mother that her children were learning ta speak
the same kind of English as she had lcarncd in her infancy. Lin-
guistie chanp;e is far more rapid than biological change, but prob-
ably slower thun the changes in other human infltitntions.
Linguistic change interests us especially because it offerfl the
anly possibility of explaining the phenomena of language. Speak-
ers acquire their habits from earlier speakers; the only explanation
of their habits lies in the habits of these earlier speakers. If we
ask, for instance, why present-day speakers use the form dog for
the animal 'canis domesticus,' or, let us say, why they add the
suffix [-ez, z, -s] to derive plural from singular nouns, the obvious
281
282 WRITTEN RECORDS WRITTEN RECORDS 283
answer is that they acquired these habits, in infancy, from the
aIder people round them; if we then ask the same questions about
the habits of these aIder people, we are referred ta the habits of
still aIder people, and sa on, back into time, without limit. If wc
could realize our diagram of density of communication ( 3.4), in
which every speaker was represented by a dot and every utteranc
byan arrow from the dot that represented the speaker ta the dot
or dots that represented the hearer or hearers, we should find that
the network reached indefinitely back into time.
In the normal casc, then, the explanation for a speech-habit is
simply the existence of the Same habit at an earlier time. 'Where
linguistic change has bccn at work, however, the explanation will
he the existence of sorne other habit at an earlier time, plus the
occurrence of the change. Our lexical habit, for instance, of using
the word meat 'edible fiesh,' is not very old; a few centuries aga,
the ward flesh was used in this meaning, and the ward meat meant
'food.' The explanation of our present-day habit, in this case,
consists in (1) the earlier habit, and (2) the intervening change.
Since linguistic change never stops, it sooner or later affects every
habit in a language; if we know enough of the speech of the past,
the second type of explanation will apply to evcry present-day
speech-form.
Since written records give us direct information about the
of the past, the first step in the study of linguistic
change, wherever we have written records, is the study of these
records.
We today are sa used t,o reading and writing that we oHen
fuse these activi ties with language itseif ( 2.1). Wri Ling is a rel-
atively recent invention. It has been in use for any considerable
length of time in only a few speech-communities, and even in these
its use has been confined, until quite recently, ta a very few persons.
A speech-utterance is the same, whether it receives a written
record or not, and, in principle, a language is the same, regardless
of the extent to which specch-utf,erances of this language are
recorded in writing. For the linguist, writing is, except for certain
matters of detail, merely an cxtcrnal device, like the use of the
phonograph, which happens to preserve for our observation sorne
features of thc speech of past times.
17.2. Writing is an outgrowth of drawing. Probably ail peoples
make pictures by painting, drawing, scratching, or carving. These
i
i.
r
,
l'
l'
i
pictures, aside from other uses ( 2.9), sometimes serve as mes-
sages or reminders - that iB, they modify the conduct of the
beholder - and they may he persistently used in this way. The
Indians of North America are skilful draftsmen, and in aIder
times made extensive practical use of pictures. Thus, we are told
of an jibwa Indian who owned a long strip of birch-bark with a
series of pictures, which he used 1.0 remind himself of the succrossion
of verses in a sacred song. The third picture, for instance,
sents a fox, because the third verse of the song says sornething
about a fox, and the sixth picture represents an owl, becausc the
sixth verse says, "It is an ill omen." A Mandan Indian sent the
following picture to a fur-trader: in the center are two crossed
Iines; at one side of these lines are outline drawings of a gun and of
a beaver, with twenty-nine parallel 8trokes above the picture of the
beaver; at the other side of the crossed lines arc drawings of a
fisher, an oUer, and a buffalo. This means: "1 am ready to trade
a fisher-skin, an and a buffalo-hide for a gun and thirty
beaver-pelts."
Records and messages of this sort arc u8ually spoken of as
"picture-writing," but this term is mislcading. The records and
messages, like writing, have the advantagc of being permanent
and transportable, but they fall short of writing in accuracy, since
they bear no fixed relation to linguistic forms and accordingly do
not share in the delicate adjustment of the latter.
We have no record of any people's progress from thi8 use of
pictures ta the use of real writing, and can only guess at the
steps. In the use of pictures wc can often see the bcginnings of
the transition, and traces of it remain in the actual systems of
writing.
Real writing uses a limited number of conventional symbols.
We must suppose, therefore, that in the transition the pictures
became conventionalized. The way of outlining cach animal, for
instance, becomes so fixed that even a very imperfect sketch leaves
no doubt as ta the species of animal. To sorne degree this is true
of the pictures of Arnerican Indians. In actual systems of writing
we orten find symbols which still betray this origin. In the so-called
hieroglyphic writing of ancient Egypt, most of the symbols are
OOnvcntional but realistic picturcs, and many of them actually
denote the name of the object which they represent; thus, the
picture of a goose (drawn always in the same way) denotes the
284
WRITTEN RECORDS WRITTEN RECORDS 285
word [S?}l whieh means 'goose.' In Chinese writing, sorne of the
symbols, such as, for instance, the syrnbol for the word [ma
3
J
'horse,' still resemble a picture of the meaning of the word, and
this is sometimes true of the older shapes of characters whose
modern form shows no such resemblanCB.
When the pieturc has become rigidly conventionalized, we may
calI it a characler. A character is a uniform mark or set of marks
which people produCB under certain conditions and ta which,
accordingly, they respond in a certain way. Once this habit is
established, the resemblance of the charaetcr to any particular
abject is of secondary importance, and rnay be obliterated by
changes in the convention of forming the chameter. These changes
are often due ta the nature of the writing-rnaterials. Sorne of
the charaeters of the cuneiJarm writing of the aneient l\lesopota-
mian peoples still betray thcir origin in pictures, but for the most
part this is not the case: the characters consist of longer and shorter
wedge-shaped strokes in various arrangements, and evidently
got this shape because they were scratched into tough clay. In the
hieroglyphic writing of ancient Egypt the characters were carefully
paintecl, but for rapid writing with a reecl brush on papyrus the
Egyptians developed a simplified and roundecl version (known as
hieratic writing) whose charaeters have lost ail resemblanee to
pictures. Our own writing is ultimately derived from the ancient
Egyptian, but no one could recognize pictures in our lctters; as a
matter of fact, our letter F still has the two horns of the snaH
whichwas pietured in the hieroglyphie ancestor of thisletter.
The other, more important phase of the transition frorn the use
of pictures ta real writing, is the association of the characters with
linguistic fomlS. Most situations contain features that do not lend
themselvcs to pJcturing; the to aH sorts of
devices that proper response.Thus, we saw the
Indian drawing twenty-nine strokes above his beaver to represent
the number of beaver-pelts. Instead of depicting the process of
exehange by a series of pietures, he represented it by two crossed
lines with the sets of traded abjects at either side. The Ojibwa
represcnted "ill omen" by an owl, in accordance, no doubt, with
sorne tribal belief.
When the picture-user was confronted by a problem of this kind,
we may suppose that he actually spoke to himself, and tried out
l "'e do not know the vowcl sounds of anoient Egyptian.
varions wordings of the troublesome message. Language, after aIl,
is our one way of communicating the kind of things that do not
lend thernselves to drawing. If we make this supposition, we can
understand that the picture-users might, in time, arrange the
characters in the order of the spoken words of their language, and
that they rnight develop a convention of representing every part
- say, every word - of the spoken utterance by sorne characrer.
We can only guess at the steps of this transition: real writing
presupposes it.
In real writinp;, some characters have a twofold value, for they
represent both a pieturable abject and a phonetic or linguistic
fotm; other charaeters, having lost their pictorial value, represent
only a phonetic or linguistic form; purely pictorial characrers that
are not associated with speech-forms sink into subsidiary use. The
linguistic value predorninates more and more, especial1y as the
characters become conventionalizcd in shape, losing their resem-
blance ta pictured abjects. The characters become symbols - that
is marks or groups of marks that conventionally represent sorne
linguistic form. A syrnbol "represcnts" a linguistie form in the
sense that people write the symbol in situations where they utter
the linguistic form, and respond to the symbol as they respond to
the hearing of the linguistic form. Actually, the writer utters the
speech-fonn before or during the act of writing and the hearer
ntters it in the aet of reading; only after considerable practice do
wc succeed in making these speech-movements inaudible and
inconspicuous.
17.3. Apparently, words are th linguistic nnits that are tirst
symbolizcd in writing. Systems of writing which use a symbol for
each word of the spoken utterance, are known by the misleading
name of ideographic writing. The important thing about writing
is preeisely this, that the characters reprcsent not features of the
practical world (" ideas "), but features of the writers' language;
a better name, accordingly, would he or logographic
writing.
The main difficulty about logographic writing is the providing
of symbols for words whose meaninp; dacs not lend itsclf to pic-
torial representation. Thus, the Egyptians used a characti:r that
rcpref:>'Bntcd a tadpole, t symbolize a word that meant 'one
hundred thousand,' presumably becausc tadpolcs were very nu-
merous in the swamps. The Chinese symbol for the word
286 WRITTEN RECORDS
WRITTEN RECORDS
287
'l'';ood' is a cornbination of the symbols for 'woman' and for
'child.'
The most important device of this sort is to use the symbol of
sorne phonetieally similar word whose meaning is picturable.
Thus, the ancient Egyptians used the character that depicted a
goose, not only for the ward [s?] 'goose,' but also for the ward
[s?] 'son,' and they used the eharacter that depicted a convention-
alized eheekerboard, not only for [mn] 'eheckers,' but also for
[mn] 'remain.' Chinese writing used the conventionalizcd charae-
ter depieting a wheat-plant not only for a word that meant 'wheat,'
but also for the homonymous ward that meant 'come' - in
present-day North Chinese, [laj2]. The ambiguity that arises in
this way, leads to a further development: one adds flome eharae-
ter that shows which of the similar words lS to he read; these ad-
ditional eharaeters are ealled classifiers or determinant8. In Chinese
writing, whieh earries the logographic system to perfection, the
phonetic (as the basic symbol is callcd) and the classifier are united
into a single compound character. Thus, the symbol for [ma
3
]
'horse' and the symbol for [ny"] 'woman' are united into a com-
pound character, which serves as the symbol for the word [mal]
'mother.' The symbol for [farJl] 'square' combines with the sym
bol for [thu
2
] 'earth' into a compound symbol for [fa!)!] 'district ';
wit.h t.he symbol for [sr
l
] 'silk,' it forms a compound symbol rep-
resenting the ward [faIJ3] 'spin.' The phonetic part of the corn
pound symbol, as these examples show, docs not always accuratcly
represent the sound of the word; we have to suppose, however,
that at the time and in the dialect where this dcveJopment Look
place, the compound symbols (that is, such as wcre there and then
created) were phonetically accurate.
The logographic system, as we see H in Chinese writing, has the
disadvantage that one has to learn a symbol for every word of
the language. The compound symbols of Chincsc writing can
al! be analyzed into 214 constituents ("radicals"), but, even so,
the labor of learning to read and write is enormous. On the other
hand, this system has a great advantage in that the symbols arc
non-committal as ta the phonetic shape of the '.'lords, The Chinese
speak a number of mutually unintclligible dialects, but in writing
and printing- they adhere,l to certain conventions of lcxicon and
ward-arder and are thus able L rcad each others' writings and,
with sorne training, also the writings of their ancient literaturo.
Our numerals (derived from ancient India) arc exarnplcs of
logographic writing. A like 4 is intelligible to many na-
tions, although we read li as [fowr], the Germans as [fi;r] , the
French as [katr], and 50 on, Moreover, since we arrange the nu-
merals according to a fixed convention, we can rcad each others'
numeral phrases even thmlgh our languages differ as to the struc-
ture of these phrases: 91, for instance, is everywhore intelligible,
although we say not ['najn 'won] but ['najntij 'won], and the Ger-
mans say, in opposite arder, ['ajn unt 'nojntsik] 'one and ninety,:
and the French [katra w;n onz] 'four twenties eloven,' and the
Danes ['e?n [l hal 'fem?sJ 'one and haU five-times.'
17. 4. In the device of representing unpicturable words by
phonetically similar picturable words, we see the emergencc of
the phonetic factor in writing. Once a symbol is associated with
a particular word, the phonetic features of this word may suffice
ta bring about the writing of the syrnbol. In Chinese, where the
words arc of uniform structure, this transfcrence has been made
only from word to word, and the compound characters, in accord-
ance with this structure, are written as units ana held down to
uniform size. In the writing of other languages, where '.'lords are
a! lengths, wc find uscd for phonetically
simllar parts of longer words. Thus, the Egyptians wrote the
symbol for [mn] 'checkerboard' twice over to reprcsent the word
lmnmn] 'move.' Bya succession of the symbols for [m] 'duster'
and [Dr] 'basket,' tobey wrotc the ward [mnr] 'car.' In accord-
ance with the structural variety, they represented words not always
by one. symbol, but also by various arrangements of logograms,
phonetlCs, and classifiers. Similarly, in Aztec writing, the place-
Dame Teocaltz'tlan, literally 'god-house-people,' '.'la:; represented by
the Symbols for tentli 'Jips,' otli 'path,' calli 'house ' and /lantli
'teeth' th . h " .. . . '
. ' IS IS t e more mtelhglble as the -th lU these words is
an lllfiectional suffix.
The s,rrnb l . th
J'" a s 10 lS way may take on a more and more constant
phonQgraphic value: they bccome phonograms - that is symbols
not for r . t' f '
mgUls IC onns, but for phonetic fonns. The commonest
result 8eCm t b .
s 0 e a sot of syrnbols, each one of which do-
one syllabic sound with (or without) preceding and follow-
mg non-syllab' 1'h 'f ..
ICS. e cunel orm wntmg of the ancient Meso-
Pt[aInlam; reached this stage; it had charaeters for such syllablcs
as ma' .
, ml, mu, am, lm, um, muk, mut, nam, tim]. Throughout
288
WRITTEN RECORDS
WRITTEN RECORDS 289
its use, as it passed from nation ta nation, it carried aiong logo-
graphic features. For instance, the ancient Sumerian ward for
'god' was [an]; when the Babylonians learned the use of writing,
they took over the Sumerian symbol as a logogram for the Baby-
10nian word [iluJ 'god,' and as a classifier which they placed before
the names of gods. This kind of retention often occurs when a
system of writing is adapted ta a new language; thus, we retain
Latin abbreviations, such as & (Latin et) for and; etc. (Latin et
cetera 'and other things ') for and 80 forth; i.e. (Latin id est) for
that is; e.g. (Latin exempli gratia 'for the sake of an example')
for for instance; lb. (Latin libra) for pound, and so on.
In Babylonian writing the syllabic principle was never fully
carried out; thus, a single symbol (a vertical wedge with two small
wcdges aslant at the left) represented the syllables [ud, ut, UT, tarn,
par, pir, lax, xi;';] and, 1ogographically, the words [u:mu] 'day,'
[samsu] 'sun,' and [piu] 'white.' In its Old Persian form, cunei-
form writing had developed into a genuine syllabary, with a rela-
tively small number of symbols, eaeh representative of sorne one
syllable. In general, syllabic systems of writing are widespread
and seem ta be easily devised. The ancient Greeks on the island
of Cyprus used a syllabary of sorne sixty-five symbols. The Jap-
anese largely use Chinese logographs, but supplement them with
two syllabaries, both of which are derived from Chinese charae-
ters. The Vai, in Guinea, are said to have a system of 226 syllabic
signs. '''hen persons acquainted with modern writing devise a
system for an illiterate people, they sometimes find it easiest w
teach syllabic writing. Thus, Sikwaya, a Cherokee, devised a set
of eighty-five syllabic symbols for his language; the Fox Indians
have several syl1abaries, ail based on English script forms; and
the Cree have a syllabary consisting of simple geometrical
acters.
17.5. It seems that only once in the history of writing there
bas been any advance beyond the syllabic principle. Sorne of
the Egyptian hieroglyphic and hieratic symbols were used for
syllables containing only one consonant; in the use of these, differ-
ences of the accompanying vowel were disregarded, and the r&
sultant arnbiguities were removed by the use of classifiers and
logograms. In ail, there \Vere twenty-four of these symbols for one-
consonant syllabics. At an early date - certainly before 1500 B.C.
- Semitic-speaking people became acquainted with Egyptian
riUng, and hit upon the idea of lietting down words of their lan-
;uage by means of the twenty-four simplcst .symbols.
This was feasible because the structure of ScmltlC Identifies each
root by its consonant-scheme ( 14.8); the non-indication of vowels
could leave a reader in doubt only as ta sorne features of
derivation which he might, in most instances, guess from the con-
text.
Our oldest examples of this Semitic writing are the Sinai Inscrip-
tions, which date from somewhere round 1800 ta 1500 B.C. One
later style of writing these characters is known as the South Sem-
itic; it is represented by old inscriptions and, in modern times, by
the Ethiopian alphabet. The other, North Semitic, style, was
used by the Phoenicians, the Hebrews, and the Arameans. The
Aramaic varicties include the style which we see in the modem
l'Hebrew'' type, the Syrian style, and the writing of modern
Arabic. It ie the North Semitic character, in its Phoenician and
its Aramaic varieties, that has spread, with many changes, over
Asia and Europe.
The syllabaries Ilsed in India seem ta be derived in part from
Aramaic, and mostly from Phoenician writing. For the languages
of India, indication of the vowel phonemes was necessary. The
Indians used each Semitic character for the syllable of consonant
plus [a] and then devised additional marks (diacriiical signs) which
they added ta the symbol ta designate the combination of the
consonant with sorne other vowel. Thus, a simple sign means [bal,
and the same sign \Vith various marks means [ba:, bi, bi:, bu, bu:)
and sa on. Further, the Indians devised a mark which meant that
the consonant was fol1owed by no vowel at ail, and a set of sym-
bols for vowcls without any consonant. At the same time, they
increased the number of basic symbols until they had one for
each consonant phoneme. In this way they arrived at a system
which recorded their specch-forms with entire phonetic accuracy.
17.6. Of aIl the offshoots, immediate and other, of Semitic
writing, we need trace onlv the one which includes our own system
of writing. The ancient Greeks took over the Phoenician system
and made a decisive change. Sorne of the Phoenician symbols
represented syllables containing consonants that were Foreign ta
Greek; thus, A represcntcd glottal stop plus vowel, 0 11 laryngal
spirant plus vowcl, and l the consonant [il plus vowel. The Greeks
Used these superfluous symbols ta indicatc vowel values, combining
290 RECORDS
WRITTEK RECORDS 291
bvo symbols, such as TA or TO or TI, to fepresent a single syllable. '
In this way they arrived at the principle of phonemic or alphabetic .
writing - the principle of using a symbol for each phoneme. They'
feIl short of complete aecuracy only because they faUed to invent :
cnough symbols for vowels: they never distinguished betwccn the
long and short quantities, distinctive in their language, of the'
vowels [a, i, u]. They did later devise diacriticai marks ta indicate
the position and the two qualities of their ward-accent, and sorne
signs of punctuation ta indicate sentence-modulation.
From the Creeks the alphabet spread ta other Mediterranean
peoples. The Romans reccived it apparently through the mediation
of the Etruscans. In the Middle Ages it passed from the Greeks
to the Bulgarians, Serbians, and Russians, and from the Romans,
dircctly or indireetly, to the other nations of Europe.
The transfer of writing to a new langnage oeCUTIJ, apparently,
in this way, that sorne bi!ingual persan who knows writing in one
language, hits upon the notion of usinl'; the alphabet also for his
other langual';e. He may retain whakver defects the alphabet had
in the first language and he may retain letters that are necessary in
the first language but superfluous in the new one, and he may fail
ta devise new letters for additional phonemes of the new language.
On the other hand, he or his successors may be C!ever enough to
mend these defects, either by inventing new characters or by
putting superfluous eharacters ta good use, Of by
devices, such as using comhinations of letters for a sIDgle
phoneme.
The phonetic pattern of Latin was such that the Greek alphabet,
as the Romans got it (probably from the Etruseans), was almost
sufficient. One defect, the use of the symbol C for both [k] and
[g], they mended by inventing the modified symbol G for tg]. A
more serious matter was the laek of syrnbols ta distinguish long
and short vowels; the practice of plaeing a stroke over the letter or
of writing the letter twiee ta indieate length, never gained much
ground. There \Vas no need for indieating the word-accent, since
this in Latin was automatically reguIa.ted aecording to the primary
phonemes. .
The Germanic-speaking peoples took ove:- the Cracco-Roman !
alphabet, we do not know whcn Of where, ID a shape somewhat [
dit1ercnt from the ordinary Greek or Latin styles. This. 0: the 1:,.,
alphabet, known as the runes, was used for short mscnptlOns,
r
cbiefly of magic Of religious charact:r,. such as epitaphs. The
ere not used skilfully, but they dld mclude letters for sorne
:ny Germanic phonemes, [&, ,v, j]. The customary arder of the
alphabet, too, was different from of the Craeco-Roman pro-
totype; it ran: [f u & a r k g w h Il 1 J P E Z s t b c mil] 0 d]. For
this reason the runic alphabet is sometimes c1-111ed the futhark.
The oldest runic inscriptions date from round 300 A.D. Later, as
the Germanie-speaking pcoples were christianized by Homance and
Irish missionaries, they l';ave up the runes in fayor of the Latin
alphabet. However, the Gothic bishop Ulfila, who in the fourth
century deviscd an alphabet for his Bible-translation, retained
severa1 runic letters, and the Old English priests, in the eighth
century, when they took to writing English, retained the runic
cl1aracters for [0] and [w], sinee the Latin alphabet providcd none.
It was only after the :-Jorrnan Conquest that English writers gave
up thesB leUers in ffl,vor of the combinations th and VI' (whenee our
w). The live Latin yowelletters have never sufficed for English;
on the other hand, wc l'etain the superfluous 1etters c, q, and x.
The writing of prcsent-day English lacks symbols for the phonemes
[a, E, ;:l, &, 5, S, :il, , 1]] and for the stress-accent. This lack is only
partially repaired by the use of digraphs, such as th, sh, ch, ng.
Occasionally wc find our alphabet fully adapted ta the phonetic
system of some lanl';uage. In the ninth ccntury, the apostles Cyril
and Method added enough extra letters ta the Greek alphabet ta
make it cover the primary phonemes of the Old Bulgarian language.
This Slavie alphabet, in its modern form, is weIl suited to the
Slavic languages; for Serbian, sOlDe extra eharacters have bcen
added. Several modern hlllguages have adequate forrns of the
Latin alphabet; in the case of Bohemi:lll and of Finnish, this result
has been reaehed by the use of di::teritical marks, and in the case of
PoIish by the use also of digraphs, such as cz for l] and IIZ for [s].
17.7. The prineiple of nJphabetie writing - one symbol for
each phoneme - is applicable, of course, ta any language. The
inadequacy of the actual systems i" due largely to the conservatism
of the people who wrte. The writer does not analyze the phonetie
system of his speech, but mercly writes eaeh ward as he has seen
it in the writings of his Rredccessors. When the art of writing be-
COrnes weil established in a community, oot ooly the spellings of
wordE, but even lexical and grammatical forms becomc conven-
tioual for written records. In Lhis way, a literary dialed may bccome
292 WRITTEN RECORDS WRITTEN RECORDS 293
established and obligatory for written records, regardless of the
writer's actual dblect.
This conservatism, as time goes on, works also in another way:
the conventions of writing remain unaltered even though the
speech-forms have undergone linguistic change. For instance, in
Latin writing the letter C represented the phoneme [k]. 'Vhen the
Irish and the English took over the Latin alphabet, they used this
letter for their [kJ-phonemes; in Old English, cu spelled [ku:] 'cow,'
cinn spelled [kinn] 'chin,' and scip spelled [skip] 'ship.' Later on,
the phoneme [k] underwent certain ehanges in the various dialects
of Latin. In Haly, [k] before front vowels became [6]; Latin ['ken-
tum] 'hundred,' for instance, became Iia.lian ['csnto]. The Romans
wrote their ward as centum; the Italians still write cento. In France,
the Latin [k] before front vowels has become [s], as in [sun] 'hun-
dred,' but the French still write this word as cent. In English, we
have taken our foreign-Iearned words from French, with the [s]
pronunciation, but also with the traditional spelling ,vith C, as in
the word cent [sent]. In Latin, the letters A, E, l, 0, U were used
for the phonemic types [a, C, i, 0, u], and they were taken into
English wrHing in these values. Thus, in medieval English writing,
a graph like name rcpresented Il forrn like ['na;me] 'name.' In the
fifteenth century, English spelling becarne eonventionally fixed in
much Hs present shape. Since that tir;le, howevcr, our vowel
phonemes have undergone a great deal of change. The result has
heen that we use the Latin vowel-letters not only in entircly new
values - this, after all, would do no harm - but in inconsistent
ways. We have kept on using the letter A in graphs like name, hal,
aU, far, although these words havc now entirely different syllabic
phonemes. SouneIs which existed when our spelling becarne habit-
uai, but have since been lost by linguifltic change, arc still repre-
sented in our writing by silent letters, as in name, know, gnat,
bought, would.
Once a system of spcllng; has bccomc antiquated in its relation
ta the spoken sounds, learned scribes are likely to invent pseudo-
archaic spellings. The words debt, doubt, subtle contained no lb]
sound in Old French, whence English received them, and were
written both in French and in English as dette, doute, sutil; the
present-day spellings with b were invented hy scribes who knew
the far-off Latin antecedents of the French wards, debitum, dubilo,
sublilis. The let ter s in isle reflccts the Old French spclling isle
(from Latin insula); at the time when the ward was taken inta
English t no longer had an [s] (compare modern French ile [i:l])
and was appropriately spelled ile. The scribes not only favored the
spelling with s, but even introduced the letter s into two similar
words which had never contained any namely the native
English island (from Old English iglond) and the French loan-ward
aisle (French aile, from Latin ala). People who saw the runic Ietter
pin ancient English writing-s but did know Hs va:ue [0], took
it ta be a form of the letter y and arrlvcd at the notIon that the
article the was in eider English ye.
17. 8. It is evident, from all this, that written records give us
ouly an imperfect and often distorted picture of past speech, which
bas ta he deciphered and interpreted, ofton at the cost of great
labor. 1'0 begin with, the values, Iogographic or phonographic, of
the written signs may be unknown. In this case, the problem of
deciphcrment is sometimcs n.esperate. The best help is fi bilingual
inscription, in which by the side of the undeciphercd text there is a
version in some known language; other aids arc some knowledge
of the lanbruage or of the contents of the inscription. In 1802
Georg Friedrich Grot.efend succeeded in deciphering cuneiform
inscriptions in Old Persian, and round the midle of the nineteenth
century a succession of wor!rs (E. Hincb" Rawlinson, Oppert.)
deciphered thos8 in Babylonian-Assyrian; in bath instances the
decipherers made ingenious use of their knowledge of relatcd
languages. The cuneiform texts in other languages (Sumerian, the
language of Van, and Hittite) werc deciphercd thanks to bilingual
texts, such as dictionary-like tablets of word-list.s in SUlllerian,
Assyrian, and Hittite. In 1821 Jean Franois Champollion began
the deeipherment of ancient. Egyptian \\'fitings by using the famous
Rosetta Stone (found by t.he French in 1799; now in the British
Museum), which bears parallel inscriptions in hieroglyphics, in a
laLer form of Egyptian writing, and in Greck. In 1893 Vilhelm
Thomsen deciphered thfl Old Turkish Orkhon inscriptions; Thom-
sen saw that the writing was alphabetical and the language of the
Turk family. Thc hieroglyph-like inscriptions of t.he Hittites and
those of the ancient Cretans have never been deciphered; of t.he
. Maya pcture-writing in Central America only sorne charactcrs.
denoting manths, days, numbers, and colors, have been interpreted.
If the system of writing is known, but the language is not, the
situation is litt.lc better. The most famous instance of this is the
294 WRITTEN RECORDS
WRITTEN RECORDS 295
Etruscan language in ancent Italy; we have extensive texts in
a form of the Greek alphabet, but cannat interprct them, beyond
reading personal names and a few other words. We have dice
with the first six numbers written on the faces, but cannat deter-
mine the arder of these numbers. The Lydian inscriptions in Asia
Minor arc intelligible, thanks ta a bilingual tcxt in Lydian and
Aramaic; the alphabet is Greek, and thc language apparently
related ta Etruscan.
17. 9. 'Vhen both the syst,em of writing and the language are
intelligible, we am, of course, ta learn from the texts ail we can
get as to phonetics, grammar, and lexicon. The phonetic values
of the chamet.ers in ancient writings ean never he surely known;
thus, the aetual sounds rcpresented even by t.he alphabetic sym-
bols of languages like Ancient Greek, Latin, Gothie, or Old Eng-
lish, are in part. uncertain. When the writ.ng has become conven-
tional and unphonetic, the lapscs of scribes or the way they write
uncornmon words, may bctray the real phonetic valucs. Our
Old English manuscripts show t.he same inflectional syst.em from
t.he ninth century until weIl int.o t,he eleventh century, distin-
guishing t.he vowels of unstressed syIlables and the presence of
final m and n; but oecasionallapses of the scribes betray the fact
that already in the tcnth century most of these vowels had ehanged
to le] and the final [ml and [n] had becn 10st; such lapses are, for
instance, spellings like worde for tiSual wurda 'of words,' fremme
for normal fremman 't.o make,' gode for godum 'to good ones.'
When an English writ.er in the fifteenth century spells behalf
without an l, wc infer that he no longer pronounced the [I] in this
word, although the tradition of writing insists upon t.he symbol
ta this day. So-called inverse spellings tell t.he same story. Old
English had a sO'.llld [x] in words like light, bought, eight, which
is still refiected in our spelling with gh. When wc find the word
deleite (a loan from Old French delei/cr), which never contained
the sound [x], spelled delight , then we may be sure that the [x]
was no 10nfSer spoken in words like light: for t.he writers, the gh
wa-s now a mre sHent graph, indicative only of yowel-quantity.
A serious factor in the lingustc int.erpretation of writtcn docu-
ments is their t.ransmission. Inscriptions, chiefly on stone or metal
or, as in the cuneiform text.s, on clay, are generally original nota-
tions; we need rcckon only with one scribe's errors of speHing or
dictation. 110st writing, however, is made on perishable material,
and has come ta our time through successive copyings. Our manu-
scripts of Grcek and Latin writings date from t.he Middle Ages,
often from the later :Jiddle Ages or from the early modern period;
only fragment.s have been prescrved on papyrus in the sands of
Egypt. His rare good fortune when wc have a contemporary man-
uscript of an ancient text, like the RaHon manuscript of Alfred
the Great's translation of Pope Gregory's Pastoral Care. The
scribes not only made mistakes in copying, especially wherc they
did not understand the text, but they even tampered with it, by
way of improving the language or falsifying t.he content. The
studyof ancient writing, pawography, and the technique of rccon-
structing ancient texts from one or more imperfect copies, lex-
tual criticism, have developed into separate branches of science.
Unfortunat.cly, textual critics have sornetimes lacked linguistie
knowledge; our printed editions of ancient lexts may fail to re-
port linguistically valuable forms that appear in t.he manuseripts.
Sometimes t.he t.ext. which appears in our written records has
undergone re-spelling into a nc,\' alphabet or a new system of
orthography. This is the case with our t.ext. of the ancient Greek
Romeric poems, and with our texts of the Avesta. We try, in
such cases, t.o reconstruct t.he original spollings and ta detect
misleading or orroneous features in t.he traditional t.ext.
17. 10. Thore are a few side-issues which somctimes help us
in the linguistic interpretation of writtcn records. In the forms
of composition which we group t.ogether under the name of verse,
the author binds himsclf ta observe certain phonetic patterns.
In modern English verse, for instance, the allthor shapes his word-
ing so that st.ress-phonemcs come at ccrbtin intervals, and t.hat
words of like ending, from the stresscd syllabic to the end, occur
in pairs or larg-er sets, again at certa.in intervals. Thus, if we know
that a poet composed ulider a convent.ion of exact rimes, we can
gather from his rimc-words a groat deal of information that may
liot appear in t.he s p e 1 1 i n ~ s . Chaucer rimed ~ ta quote the words
in their present-day spellings ~ mean with clean, but not wit.h
keen, queen, green: he evidently spoke different vowels in these
two sets of words. On the other hand, inconsstencies are equally
Hluminating. When the Aisatian poet Brant, at the end of the
fiftecnth century, rimes the word for 'not.' both in the Alsatian
form [nit], as, for instance, with Ritt [bit] 'rcqucst,' and in the
present-day standard German form [nixt.], as, for instance, with
296 WRITTEN RECORDS
Geschicht [ge'sixt] 'story,' we know that in his day the modern
standard form, nicht [nixt] 'not' had already gained curreney
alongside the provincial form of the ward. Even whcn rimes are
uscd traditionally after they cease ta he phonetically true, as,
in modern English poetry, rimes like move : love or seant: want,
a study of the tradition may he of interest.
Other types of verse lead ta similar deductions. In old Germanie
poetry, high-stressed words occurred in alliterative sets with the
saille initial consonant, as in hOUS8 and home, kith and kin. Ac-
cordingly when in ancient Ieelandic verses of the Eddie poems wc
tind [',vega, 'vega] 'strike' alliterating with [rejt5rJ 'wroth,' wc
conclude that the men who eoined this alliteration still pronounced
the latter ward with an initial [wr-], although the spdling of our
manuscripts, in aecordance with the later language, no longer
shows the [w]. In Grcek and Latin verse the succession of long
and short syllables was regulated; a syllable containing a long
vowel or u. diphthong, or any vowe1 followed by more than one
consonant, counted as long; the position of words in verse thus
often informs us as ta vowel-quantities, which are only in part
shown by Greek orthography and not at aIl by Latin.
Another oeeasional help toward the interpretation of written
records is the transcription of speeeh-forms from one language
into another. At the begirming of the Christian era we tilld the
name of Caesar written in Greek texts as kaisar: sinee the Greek
language has not undergone a change of [k] ta [cl or the like, and
the Greek k, accordingly, represented always a phoneme of the [k]
type, this transcription makes it likely that Latin at that time
still prescrved the [k-]. The old Chinese transcriptions of Indo-
Aryan Mmes in Buddhist texts give information about the sOunds
which wero attaehed ta Chinese 10gographic symbols.
Finally, written records may contain statements of a linguistic
nature, as in the case of Sanskrit grammar and lexicon ( 1.6);
the Hindus, moreover, were excellent phoneticians and interpretcd
the written symbols in physiologie terms. Often enough, however,
we have ta distrust the information in our texts. The Latin /;ram-
marians give us little help as ta the English phone-
tieians of the early modern period, likewise, confused sounds with
spellings and give very poor guidance as ta the actual pronuncia-
tian of their time.
CHA PTER 18
THE COMPARATIVE METHOD
18.1. 'Ve saw in Chapter 1 that some languages resemble each
other ta a degrec that can be explained only by historieal connec-
tian. Sorne resemblanee, ta be sure, may result from universa.I
factors. Such fcatures as phonemes, morphemes, words, sentences,
constructions, and substitutian-types, appear in every language;
theyare Inherent in the nature of human speech. Other features,
sueh as noun-like and verb-like form-classes, categ;ories of nllmber,
persan, case, and tcnse, or grammatical positions of actor, verbal
goal, and possessor, ure not universal, but still sa widesprt,ad
that better knowlcdge will doubtless sorne day connect them with
univcrsaJ charact.eristics of mankind. features that are
Dot widesprcad - mnong thern some very specifie and cven mi-
nute oncs - arc found in distant and wholly unrelated languages;
these features, tao, may be expccted sorne day ta throw light on
human psychology.
Other resemblances bctwn languages bear no significance
whatevcr. lvlodern Greck ['mati] means 'eye,' and so does the
Malay ward [mata]. If wc knew nothing of the history of thesB
languages, we should have to work through their lexicons and
grammars in search of other rcscmblances, and then weigh the
probabilities of historieal conncction, taling into aeeount both
the number of rescmblanccs and their stnlctural position. Ac-
tually, our knowledg;e of the past forms bath of Grk and of
Malay shows us that the resemblance of the t\\'o words for 'eye'
is accidentaI. 1.fodern Grk ['mati] is a relativc1y recent develop-
ment from an ancient Greek [om'mationj 'Ettle eye,' and this
word \Vas in ancient Greck connccted, as a secondary derivative,
with an underlying word ['omma] 'eye.' The Malay word [mata],
on the other hand, had in ancient tirnes much the saIlle phonetic
shape as today. Even if, a,gainst ail present seeming, it should turn
out, Some day, that these two languages are related, the relation-
ship would lie far back of Primitive Tudo-European and Primitive
Malayo-Polyncsian time, and the l'esernblancc of the modern
ward:> for' eye' would have nothing ta do with. this relationship.
297
Still other resemblances are duc to the borrowing of speech-
forms. In modern Finnish there arc many words like abslraklinen
'abstraet,' almanakka 'almanac,' arkkitehti 'architect,' ballaadi
'ballad,' and so on through the dictionary - cultural words of
general European distribution, which have been borrowed, in the
last centuries, from one European language into the other, and
evidence nothing about kinship. To be sure, we cannot always
distinguish between this sort of transmission and the normal hand-
ing on of linguistic habits within a speech-community, but for the
most part the two processes arc very different. If the Finno-Dgrian
languages should he related to the Indo-European, then the kin-
ship dates from a time when the words abslract, almanac, etc., were
not yet in use.
lB. 2. When wc say, in contrast with these eascs, that a Te-
semblance betwCn languages is due to relationship, we mean that
these languages arc later forms of a single earlier language. In the
case of the Romance languages, we have written records of this
parent language, namcly, Latin. After the Latin language had
sprcad over a large area, it underwcnt different linguistic changes
in different parts of this area, so that today these differcnt parts
differ greatly in speech, and ''Ile call the divergent spcech-forms
"Italian," "French," "Spanish," and so on. If we could follow
the speech, say of Italy, through the last two-thousand years, we
could not pick out any hour or year or century when "Latin"
gave way to "Italian" i these names are entirely arbitrary. Dy
and large, any foature that is common to aIl the modern territorial
forms of Latin, was present in the Latin of two-thousand years ago;
on the other hand, when the modern forms of Latin disagree as to
any feature, thon sorne or al! of them have, in this feature, under-
gone sorne change during the 1ast two-thousand years. The re-
semblances appear especially in features that are common in
everyday speech - in the commonest constructions and form-
classes and in the intimate basic vocabulary. The features of
difference, moreover, appcar in systematic groups, with each terri-
torial form diverging in its own characteristic way.
In most cases we are less favorably situated, in that wc possess no
written records of the uniform parent speech. The Germanie lan-
guages, for instance, resemble cach other much as do the Romance,
but we have no records from a time when the differences had not
yet arisen. The comparalive melhod, however, lllakcs the same
'house' haws
hpys
haws

hu:s
'mouse'
m0Ys
mawS
mu:S
maws
'louse' laws
Ipys laws
lu:?s
lu:s
'out' awt
1yt
aws u:?ti
u:t
'brown' brawn
brpyn
brawn
bru:?n
bru:n.
299
SWlOPl8R
SWEPLBll
DANlBH
THE COMPARATIVE METHD
an
man
'man' mEn man ID
h t h t
hand
, hand ' hlOnd an an
f
fo:t6 fo:t
'foot' fut vu:t u:s
, . 'fiIJer 'feIJ?ar 'fiIJer
'finger 'fiIJgT VIl)er
h
h hu.?s hu:s
'house' haws 0YS aws
, . t 'venMar 'vinter
'winter' 'wintr 'winter VlO cr .
,
'summer' 'somT 'zo:mer zomer s;) v "
'd . k 't . ken 'drega vdrika
'drink' dri1)k rll) C rll)
b
'b 'brl'I)en 'breIJa v brirJa
'bring' ril) rel)e
'lived' livd 'le: vde 'le:pte 'lc:vda v le:vde
This list could be extended almost indefinitely; the .resemblances
d the
Y
S
o thoroughly pervade the baslc vocabulary
are sa many an ' . . .
and grammar, that neither accident nor borrowmg WIU expIal.o
them. Wc need only tum to languages outside the
n

group ta see the contrast, as in 'hand': [me ].' Russmn
[
'k] F' . h ka"si' or 'house" French [mezo J, RUSSlan [dom],
ru a, mOlS ., ". .
Finnish talo. Another remarkable feature lS the systematIC
ing of the differences within the Germani? famil}'. Where
has the compound intonation, there Danlsh lacks glottal stop,
where the others have initial [f], thcre Dutch has imttai [vli ,:here
the others have [d], therc German has [t]. In fact, senes of
forms show the same divergences from one Germamc Janguage to
the other. Thus, the divergent syllabic phonemes JO the word
house arc parallclcd in 11 whale sct of forms:
ENGLlBB: D(;TCR GERMAN
h
In the latter case wc merely rack the
ferences in bot cases. h . t t
t
of the written record. We assume t e eXls ence, a
confirma lOn . .. . l
. . the past of a Pnmtltve Germamc parent anguage,
some tlme m , 1 b
but the speech-forms of this arc known to. us on y . y
. f wc writc thern clown, wc indicatc thls by placmg
10 erence. n
an asterisk bcfore them. .'
lB. 3. Compare, for instance, the followmg words present-day
standard English, Dutch, German, Danish, and Swedlsh:
DUTCIl GER.\!AN DANlBIl
THE COMPARATIVE METHD 298
301
ROU!llANIAN
'fioarJ
nod
ROUMAN!AN
SPAN18H
LADIN FREI<l.'H SPANlSH
ITALIA::>/
ITALlA"I LADIN FRE:>Il.'H
THE COMPARATIVE METHOD
'flower' 'fjore flur fi:r
'knot' 'nodo nuf np
, vow' ''loto vud v0 bodas
1
'tail' 'koda 'kua k0 'kola 2 'koada
The Latin prototypes appear in the first. three of these words, as
weIl aS in a number of smilar cases, with a syUabic 0, which we
interpret aS [0:]: florem, nodum, uotNm. In our fourth ward,
cordingly, we infer that the Latin prototype contained this same
vowel and had the form *['ko:damj. An illference of this kimi is a
reconstruction; we mark the reconstructed form, *['ko:dllml or
*codam, with an asterisk. Now, in the written records of Latin,
the ward for 'tail' appellrs in a different shapc, namely as caudam
(accusative singularj the nominative s cauda). Ths dsagrces
with our reconstruction, for ordinarly Latin au (presumably
[aw1) is reflected in the Romance languages bya different type of
vowel-correspondenee. Thus, Latin aurum 'gold' and causam
1thing, affair' appear as:
'gold' 'oro aur
'thing' 'koze oo:z 'kosa
It is truc that our Latin manuscripts, written in the l\1iddlc Ages,
occasionally spell the word for (tail' as coda, but this may he duc
merely to the errors of eopyists; the older manuscrpts from whch
1 Pluml furm, mCall ;ng
Re-shaped from Old Spanisb coa, prcsumably [koaJ.
Here we follow the same procedure as with the Germanie cor-
respondences, observing the local types in each area, and the
spelling
s
of the older records. The difference is only
written notations of the form of the parent language, Latlll, are m
wast inst.ances available. The Romance words in our examplc are
modern forms of the Latin wors whch appear in our records as
nasu
m
, eaput, capram,Jabam.
After we have learned ta draw inferences from the Romance
forms l'le may find diserepanees bctween the resuIt of our n
4
and the writtcn records of Latin. These discrepancies are
especially interesting because of the light they throw on the value
of our nferences in cases where no record of the parent language
is available. Take, for instance, the syllabic in the followng types:
nas
kap
'kapI';}
'fawa
1
,
naso
'kabo
'kabra
'aba
'nose' 'naso nas ne
'hoad' 'kapo kaf sef
'goat' 'kapra 'kavra 5e:vr
'bean' 'fava 'fave fcv
l Macedonian
300 THE CO::VIPARATI VE M CTHOD
fact that the differences themRdves foltow a system, - that
the say,?f EngliRh and German [aw] and Dutch [py] ap-
pears III a sencs of forms - confirms our surmisc that these
forms are connected. The divergence, we suppose, is due
ta characte".r1stlc changeR undergone by some or ail of the related lan-
If we cxtend our observation to cover more of the dialccts
III each we find many other varicties, with a similar parailelism.
In partJcular, we find, in our example, that forms with the vowel
such as [hu:s, mu:sJ etc., occur also in local dialects of the Eng-
.Ish, Dutch, and ?erman aroas - as, for instance, in Scotch English.
Further, avaiimg ourselves of the written records of these
languages, we find that the oldest. records from the English and
Dutch-German. areas, datin/!; round the eighth and ninth centuries
of our eru, wnte the forms in our example uniformly with the
u, as hus, mus, lus, ul (southern German uz), brun. Since the
wrItmg of these peoples was based on Latin, where t.he letter u
:cpresented of t.he type lu]' wc conclude t.hat the divergences
III the syllablc of our. fo::ms had not. yet arisen in the ninth century,
and that the syllablC m tho8e days was lu] in ail the Germanie
languages; other evidence lead8 us ta belicve that. the vowe! was
long [u:]. Accordingly, wc conclude that t.he Primitive Germanic
language spoke t.hcse forms with [u:J as the syllabic. His
lmpor:,ant t.o observe, however, thai. this description of the pho-
neme IS only a supplementary detai!; cven if wc made no surmise as
to the chamcter of the Primitive Germanic phonernc, the
regulartty of the corrcspondcnces, in the way of agrcement. and in
the wa,): paralle! disagreemenl, could still be explaincd onl von the
that so:ne phonemc of the parent ap-
III of the forma hou,:e, mONse, and so on.
. 8. 4. Il 1S mtcrestlng ta compare these Illferences with the
mferences that. made in the more favorahle casc, where the
parent. language 18 known to UR from written records. The
blance betwecn the Romance languages is much Iike thnt bctween
the Germanic languages. -
ITALlA;'f LADlN FnE:l'IGH SPA/i"'SH ROt:"MANIA!;
ours wero copied may have had the lisuaI Latin form cauda. This
error would he natural for copyists whose school pronunciation of
ancient Latin did not distinguish between Latin 0 and au, and
would he almost inevitable for eopyists who spoke a form of
Latin in which our ward already had, as in the present-day lan-
guages, the vowcl of fiorem, nodum, volum and not that of aurum,
causam. That sorne people were in this latter position appears
from the glass, preserved to us in ninth-century manuscripts,
which explains the ward cauda by saying that it means coda:
apparently, the former seemed antique and difficult, while the
latter was intelligible. The conclusive support for our reconstruc-
tion appears in this, that inscriptions of carly date show occasional
spel1ings of 0 in words that ordinarily have au, as POLA for the
name Paulla in an inscription dating from the year 184 B.C.
Further, we learn that this o-pronunciation for au-forms was a
vulgarism. Suetonius (who died about 160 A.O.) tells us that the
r1Jetorician Florus corrccted the Emperor Vespasian (died 79 A.O.)
for saying plostra instead of the more elegant plausira 'wagons';
the next day, the emporor gat back at him by calling him Flaurus
instead of Florus. As ta our ward, a grammarian of the fourth
century A.O. speaks of cauda and coda as variant pronunciations.
:Moreover, we occasionally tind over-elcgant forms, likc Vespasi-
an's Flaurus for Plorus; an inscription dating from before the
beginning of the Christian Era has the spdIing AVSTIA for ostia
[o:stiaJ 'doors.' In sum, wc conclude that our reconstructed
*coda *[ko:da] is by no means illusory, but represents a less elegant
pronunciation which rcally existed in ancient Lime.
Cases like this givc us confidE'nce in the reconstructcd forms.
Latin writiug did uot indicatc vowel-quantities; a graph like se-
cale 'rye' could represcnt several phonetic types. As this ward
does not occur in verse, wherc its position would show us the vowel-
quantities ( 17.10), wc should be \lnable ta determine its form,
had we not the evidence of the comparative method: forms likc
Italian segola ['segola], French seigle [se:gl] show us that the Latin
graph represents the form ['se:kaleJ. Students of the Romance
languages reconstruct a Primitive Romance ("Vulgar Latin ")
form before they turn to the written records of Latin, and they
interpret those records in the light of the reconstructed farm.
18.5. A reconstructed form, then, is a formula that tclls us
which idontities or systematic corn:spondonces of phoocrncs
303
302 THE COMPARATIVE METHOD
THE COMPARATIVE METHOD
in a set of related languages; moreove
r
, since these identities
pear ondences refieet features that were already present
and eorresp t language the reconstruetcd form is also a kind of
in the paren ,
ic diagram of the ancestral form.
oldest records of the Germanie languages we find the
f llowing forms of the wordfather: d
oGothie, text composcd in the fourth ccntury A.D.,'
. th century manuseript fadar, prcsumably [fadar], the
in a SlX
prese
nted by d may have bCn a splrant.
P
honeme re . f th t
Id N se
in thirteenth-century manuscnpts 0 texts a
or, "'. hl
'0 part composed much earlier: fauzr, presuma y
were, 1 ,
['fa'6erj. bl
Old EngIish, ninth-century manuscripts: fder, presuma Y
['feder] 1
Old Fnsian, thirteenth-century manuscripts texts that were
composed somewhat earlier: feder, presumably [feder].
Old Saxon (that is, northerly parts of the area),
ninth-century manuscripts: fader, presumably [fader].
ld High German (southerly parts of the area),
ninth-century manuscripts: fater, presumably .
We sum up thcse facts by putting down the Pnm.
1ve

prototype as *['fader]; moreover, we daim that. thlS summanzmg
formula at the same Ume shows us the phonemiC structure of the
prehistorie form. .'
Our formula embodies the followmp; observatIons. .
(1) AH the Germanie languages stress tho first syllable of th1S
ward as of most others. 'Ve indicato this in our formula by an
or since accent" on the first syllable is normal in
Germanie, no accent-mark at aH. This means, at
same time that in the Primitive Germanie parent language thlS
word with most other words a phonemic feature (cali it. x)
which appears in aH the actual Germanie languages as a h1gh
stress on the first syHable of the ward. Of course, it is almost a
certaint.y that this featuro x in the parent speech was the sa:n
e
as appcars in aIl the aetual Germanie languages, a h1gh
stress on the first syHablo, but this additional surmlse ln no way
affects the validity of the main conclusion.
(2) Ail the old Germanie languages begin the ward wit.h [f].
1 Thp. Otd Engli8h syllahlc [-der] ha. in mod?rn .EnJdiBh changed to [-or]: honcll
We say fa/her. I/wthcr. (Jallwr, etc., where Old Engh8h had [-der].
304 THE COMPARATIVE METHOD THE COMPARATIVE METHOD
,305
If we had not the older records, we shouId have to consider the
fact that sorne present-day dialects of the English and of the
German areas have here a voiced spirant of the type [v], but
the geographic distribution would even then show us that [f] wa,a
the older type. In any case, the structural value of the symbol
[f] in our formula is merely this, that the word fa/her in the Ger.
manic languages begins, and in Primitive Germanie began, with
the same phoneme as the words foot, five, fee, free, fare, and 80
on, aIl of which wc symbolize by formulas with initial If].
(3) The [a] in our forrnula says that we have here the same
respondence as in words like the following:
wateT: Gothie ['wato:], Id Korse [vatn], Old English ['weter],
ld Frisian ['weter], Old Saxon ['watar], Old High German ['>vas-
sar], Primitive Germanie formulas *['water, \vat':];
acre: Gothie ['akrs], Old Norse [ah], Old English ['eker], Old
Frisian ['ekker], Old Saxon ['akkar], Old High German ['akxar],
Primitive Germanie formula *['akraz];
day: Gothic [dags], Old Norse [dagr], Old English [dej], Old
Frisian [dej], Old Saxon [dag], Old High German [tag}, Primitive
Germanic formula *['dagaz].
In this case the dcviations, namely Old English [el and OId
Frisian [el beside the [a] of the other languages, do not oecur in
aIl formR; ail the dialecte have [a], for instance, in cases like the
following:
fare: Gothie, Old English, ld Saxon, Old High German ['faran],
Old Norse, Old Frisian ['fara], Primitive Germanie formula *['far-
anan].
In fact, the English [el and the Frisian [el occur under fixed
phonetic conditions - namely, in monosyllables, like day, and
before an [el of the next syllable, as in fatheT, water, acre. This
deviation, we infer, is due to a latcr change, perhaps in a cornrnon
intermediate Anglo-Frisian parent language. We are safe, in
any case, in setting up, for ail these words, a single structural
phonemie unit [a] in the Primitive Germanie parent language.
(4) The aeoustie value of the Gothie letter whieh wc have
literated as d is douhtful; it may have bcen a stop of the type
[dl or a spirant of the type [\"il, or it may have fiuetuated, in whieh
case [dl und ['5J were variants of one phoneme. The old Scandi-
nuviun p;raph speaks for [t'il in this arca. The West Germanie
have an unmistakable [d], which, in this as in other
s appears in South German as ft]. In our Primitive Germanie
case 'la we indieate ail this by the symbol [dl or ['{i]; the former
forn
lU
.' 0 f l'cl t'fi th
. ferable because eaSler to prmt. ur ormu a 1 en 1 es e
lShPr:m
e
with thatwhich appenrs in cases like the following:
p onther: Old Narse ['mo:'{ier], ld English ['mo:dor], Old Frisian
f
' Old Saxon ['mo:dar], Old High German ['muotar],
mo. ,
Primitive Germanie formul::t *[ 'mo:der];
mead: Old Norse [mjair], Old English ['meodo], Old Frisian
['mede], Old High German ['metu], Primitive Germanie formula
*['meduz]; ['<:]0111' l'h[''d ["d]
ride: Old Norse 'ri: oa , ( <,ng fi: an, 1 nSlan rJ: a,
ld High German [;ri:t,anj, Primitive Germanic formula *['ri:danan].
(5) The next phonerne shows us li divergence in Gothie, whieh
is obviously due to later change: Gothie always has nT for the un-
stressed er of the other languages, e.g.: Gothie ['hwa6ar], Old
English ['hwr'iSer] 'which of the two.'
(6) The dialects agree as ta the lnst phoneme, [r].
18.6. Whilc wc have no written records to eonfirrn our reeon-
struetior.<;; of Primitive Germ:mie, wc oecasionally get almost this
from the very aneient Scandinuvian runie inscriptions ( 17.6).
Take, fol' instance, the fol1owinll; reconstructions:
guest: Gothie [gastsJ, Old Norse [gestr], Old English, Old Frisian
[jest], ld Saxon, ld High German [gast], Primitive Germanie
formula *['gastiz];
horn: aU the old dialects [horo], Primitive Germanie formula
*['hornan].
Here our Primitive Germanie reconstructions are longer than
the actually attested forms. Space forbids our cntcring into the
reasons that lead UR to .'let up the additional phonemcs; suffice it to
say that in most case,;, as in guest, thcse additional phonernes are
made entirely definitc by the formOl in the actual dialeCts, while in
others, sueh as hOTn, the presence of additiona! phonemeR in
Primitive Germanie is certain from the comparison of the Germanie
languages, although the nature of these phonemes is decided only
by the considerations which ,ye now approaeh. l have chosen the
words guest and horn as cxamples becauRe they oeeur in a runie
insCliption on a golden horo, dating probably round 400 A.D.,
found near Gallehus in Denmark. Transliterated, the inscription
rends;
ek hlewagastiz holtilJ!1z horna tawido
306 THE COMPARATIVE METHOD
THE COMP ARATIVE METHOD
307
'l, Fame-Guest, the Holting (man of the family of Holt), made the
horn.' The same '.'lords in our Primitive Germanie formulas, would
appear as *['ek 'hlowa-Igastiz 'holtingaz 'hornan 'tawido:nl, and
the inscription confirms the final syllable of our reconstruction of
guest, and the vowel, at any rate, of the final syllable in our recon-
struction of horn.
The Finnish, Esthonian, and Lappish languages, belonging ta
the Finno-Ugrian family ( 4.7) and therefore unrelated to ours
contain many WOl'ds which they must have borrowed from
Germanie language at an ancient time - aIl evidence points to the
beginning of the Christian Era. As these languages have since
that time gone through entirely different changes than have the
Germanie languages, these borrowcd forms give us indepcndent
evidcnce as ta the ancient form of Germanie '.'lords. Our recon-
structions of Primitive Germanie f01'fis, like ring, Old English
[hring], Old Norse [hringr], as *['hringaz], or king, Old English
('kyning], as *['kuningaz], or gold, Old English [gold] as *['goI6an],
or yoke, Old English [jok], as *['jokun], are confirmed by such
Finnish loan-words as rengas 'ring,' kuningas 'king,' kuUa 'gold,'
jukko 'yoke.'
18.7. The comparative method gives us an even more powerful
check upon our Primitive Germanie reconstructions, Since the
Germanie languages are a brunch of the Indo-European family
our Primitive Germanic forms enter as units into
with forms of the other Indo-European languages, The recon-
structed forms of Primitive Indo-European give us a scheme of a
stin earlier structure, out of which the Primitive Germanic struc-
ture has grown,
Among our last examples therc are two good instances. Our
reconstruction of Primitive Germanie *['gastiz] 'guest' matches the
Latin form hoslis 'stranger.' From the comparison of the Siavic
forms, Old Bulgarian (gOStI], Russian [gost], and soon, we recon-
struct a Primitive Siavic *['gostIj; this, however, s under strong
suspicion of having been borrowed from a Germanie dialect and
must thercfore stay out of account. The comparison of the Latin
fonn, however, leads uS to set up a Primitive Indo-European
fonnula *[ghostis], which tells us, in shorthand fashion, that the
Latin second syllable confirms the final phonemes of our Primitive
Germanie formula.
Similarly, on the basis of Gothie [ga'juk] 'pair' and the other
otd Germanie forms of the '.'lord yoke; namely, Old Norse
Englis
h
[jok], Old Righ German [Jax], we set up a PrImltlve
Germanie formula *['jokan], confinned by the loan-form
.ukko. The phonemes n the second syllable of thls reconstrueted
would be in some respects indeterminl'.te, were it not that
this formula enters in tum into comparison with other formg of the
lndo-European group. Sanskrit [ju'gam] leads us to set up a
Primitive Indo-lranian *[ju'gam]. Further, we have Greek [zu'gon]
and Latin ['jugum]. The Slavc fonns, such as ld Bulgarian
[igo], Russian ['igo], lead us ,ta set ,up a Primi.ti,:e, Slavic
*['igo]. Cornish iou, Welsh taU, pomt ta a PrImitive Ce,ltlC
gom]. Even languages which have reshaped our word, LIthuaOlan
['jungas] and Armenian lue, give sorne evidence as to the structure
of the '.'lord in Primitive lndo-European. AIl of this evidence we
subsume in the formula, Primitive lndo-European *[ju'gom].
The case of the word father shows us an Inference of a more
complex character. Sanskrit [pi'ta:], Greck [pa't
e
:
r
], Latin
('pater], Old Irish ['a''irJ, Primitive Germanie *['faer], are the
principal forms which lead us ta set up the Primitive Indo-Euro-
pean formula as *[pa'te:r], The initial phoneme here illustrates
the simplest case, a constant and normal set of corresponences:
initial [pl of the lndo-European languages in general is matched
by [f] in Germanic, and by zero in Celtic; Latin ['porkus] 'pig,'
Lithuanian ['parsas], corresponds to Primitive Germanie *['farhaz],
ld English [fcarh] (modern jarrow), and Old Irish [ork], and the
Primitive Indo-European formula is *['porkos].
The second phonemc in our formula shows a more complex case.
In our Primitive Indo-European formulas wc distinguish three
phonemes, [a, 0, a], although no Iudo-European lan-
guage has this threefold distinction. Wc do this because the
correspondences betwecn the languages show three diffcrcnt com-
binations. We use the symbol [a] in those cases whore Indo-Iranian,
Greek, Latin, and Germanic agree in having [a], as in
acre: Sanskrit ['ajrah], Greck [a'gros], Lat.in ['ager], Primitive
Germanie *['akraz]: Primitive Indo-European formula *[agros].
We use the symbol [0] for the many cases where Indo-Iranian
and Germanie have [a], but Greek, Latin, and Celtie have [0], as in
eight: Sanskrit [as'Ta;w}, Greek [ok'to:], Latin ['okto:], Primi-
tive Germanie *['ahtaw], Gothie ['ahtaw], Old German ['ahto]:
Primitive Indo-European formula *[ok'to:
w
].
308 THE COMPARATIVE METHOD THE COMPARATIVE METHD 309
'Ve use the symbol [a]or the cases where Indo-lranian has [il,
while the other languages have the same phoneme as in the forms
of the first set:
slead: Sanskrit ['sthitih] 'a standing,' Grk [' stasis], Primitive
Germanie *['stadiz], Gothie [sta6s], Old High German [stat]:
Primitive Indo-European formula *[sthJtis].
Evidently the forms of the word father show this last type of
eorrespondenee; hcnce we use [a] in our formula. The morphologie
structure of Primitive Indo-European, as it appears in the totality
of our formulas, confirms our thrcefold distinction [a, 0, a], in that
thcse three units take part in thrce different types of morphologie
alternation.
The third syrnbol in our formula, whieh is the last we sha11 con-
sider, il!ustrates a vcry interc:sting type of infcrcnce. Ordinarily
when the other lndo-European languages have a ft], the Germanie
langunges have a [fl]. 'l'hus,
brolher: Sanskrit ['bhra:ta:], Greek ['phra:te:r] ('member of a
phratry'), Latin ['fra:tcl'], Old Bulgarian [bratre], Primitive
Germanc *['bro:Ocr], Gothie ['bro:flar], Old Norse ['bro:t5er],
Old English ['bro:tor], ola High German ['bruodcr]: Primitive
Indo-European formula *['bhra:te:rJ;
Ihree: Sanskrit ['trajah], Greek ['trc:js], Latin [trc::s], Old Bul-
garian [tnje], Primitive Germanie *[flri:z], Old Norse [Ori:r],
Old Righ German [dri:]: Primitive Indo-European formula
*['trejcsJ.
The word father, together with some others, is anomalous in
Primitive Germanie in contrning ld] instead of [OJ. One might,
of course, assume that two distinct Primitive lndo-European
phoncmes were horo involved, which had coincidod as [t] in ail
the Indo-European languages exccpt GC'rmanic, which nlone dis-
tinguished them as [01 versus [dl. In 1876, however, Karl Verner
(1846-1896), a Danish lingust, shmved that in a number of the
cases where Germanie has the troublesome [d], this consonant
follows upon a vowel or diphthong which is unstresscd in Sanskrit
and Greek; this correlation occurs in onough instances, and, in the
morphologie structure, systematieally enough, t.o cxclude the factor
of accident.. The contrast of the words brather andfalher illustrates
this correlation. Since the place of the ward-accent is determined
by the primary phonerncs in Italie, Ccltic, and Germanie, wc can
casily believc that its position in cach of thesc languages is due ta
ter change. Sanskrit and Grcck, moreove1', agree sa ofton, al-
the place of the accent in both is highly irregular, that wc
d
t h
esitate to attribute this fcature ta the parent language.
o no . .
We tbus face a dcfinite succession of events 10 the penod between
primitive Indo-European and Germanie - a period to
wbicb wc give the name pre-Germamc:
Primitive Indo-European: [t] a unit phoneme; word-accent on
different syllables in different words:
*['bhra:te:r] 'brother' *[pa'te:r] 'father'
Pre-Germanie period:
first change: [tJ becomes [6]:
*[' bra:ac:r1*[fa' Oe:r]
second change: [a] after unstressed syllabic becomes [dl, pre-
sumably a voced spirant:
*[' bra:OO:r] *[fa' de:r]
third change: the accent is shifted ta the first syllable of each
ward; this brings us ta
Primitive Gernwnic *['bro:6er] *['fader].
In a similar way, the corrcspondences reveal the pre-history of
each branch of the Indo-European family. Thus, in the case of
Latin cauda and coda' taU,' the Lithuanian ward ['kuodas] 'tuft'
probably represPnts the saroe form of t.he parent speech; if so, then,
in the light of otIler correspondences, in which Lithuanian [uo]
and Latin [0:] appear side by side, we may take coda to he the
older of the two Latin forrns, and eauda to be a hyper-urban
(over-elegant) variant ( 18.4).
Our Primitive lndo-European reconstructions are Dot subject
to any check by means of carlier recorded or reconstructed forms.
In the last decades, 1.0 he sure, it has been ascertained that the
Hittite language, known to us from records in cuneiform writing
from 1400 H.C. onward, is distantly relatcd to Indo-European.
Accordingly, it has becn possible to uncoycr a few features of a
Primitive lndo-Hittite parent language - that is, to trace the
earlier history of a few of the featurcs of Primitive Indo-European.
18. 8. The comparative method tells us, in principle, nothing
about the acoustic shape of reconstructed forms; it identifies the
phonemcs in rcconstruckd forms mercIy as recurrent units. The
Indonesian languages show us a striking cxample of this. Each
language has only a fcw phonemes of the types [d, g, 1, r], but the
varlety of the correspondences assures us of a largcr numher of
310 THE COMPARATIVE METHOD THE COMPARATIVE METHOD 311
phonemes in the parent language. The acoustic character of these
phonemes can only bc guessed at; the symbols by which we rep..
resent them are mercly labels for correspondences. It is worth
noticing that we have older written records for nonc of thesc
guages except Javanesej tbis in no way affects the application of
the comparative mcthod. The eight normal types of
ence will appear sufficiently if we consider three languages:
log (on the island of Luzon in the Philippines), Javanese, and
Batak (on the island of Sumatra). In the following examples the
consonant under discussion appears in the rniddle of the word.
T.\.G.\.LOG JAVANIlBE BATAK PF\IMITIVE
INDu"EHIA:S
(1) l 1 l 1
'ohoose' 'pi:m pilik pili *pilik
(2) 1 r r L
'lack' 'ku:larJ kural) bural) *kuLaI)
(3) l r g g
'nose' i'lul) iruIJ igul) *iguI)
(4) l D d D
'desiro' 'bi:lam iDam idam *hiDam\
(5) r d d d
'point out' 'tu:rul tuduk tudu *tuduk
(6) r d d d
'spur' 'ta:ril tadi tadi *tadi
(7) g g g g
'saga' 'sa:gu sagu sagu *tagu 2
(8) g zero r
l'
'addled' bu'guk buruk *bu1'uk
18.9. The comparatiVe method assumes that each branch or
language bears independent witness to the forms of the parent
language, and that identities or correspondences among the re-
lated languages reveal features of tbe parent speech.
This is the
Baille thing as assurning, firstly, that the parent community was
completely uniform as to language, and, seeondly, that this parent
eommunity split suddenly and sharply into two or more daughter
communities, wbieh lost aU contact with eaeh other.
1 JavancS\) [DI i8 a domal stop, distinct from the dental {dl. The Tagalog word
means 'pain, smart.' The Bata.k Iorm here givcn is Il ot listed for the Toba dialect
from whieh ou r other ""amples are ta.kcn, but it oceurs in the Dairi dialect. '
2 The Tagalog f OrIll mp'-"H 'exuda.tion'; in poetic use, a1so 'sap.'
Orien enough, the comparative method assumes successive
splittings of this sort in the history of a language, It assumes that
Germanie split off neatly from Primitive Mter
this split, any change in Germanie was independent of changes in
the sister languages, and any resemblance between Germanie and
the sister languages betokens a common inheritance. The differ-
enees between Primitive and Primitive Germanie
are duc to changes which occurred during the period.
In exactly the same way, the comparative method interprets
the special similarities among the \Vest Germanie languages
(in contrast with Scandinavian and Gothie) by saying that a
West Germanie community split off, neatly and suddenly, from
the uniform Primitive Germanie parent community. After this
splitting off cornes a pre-West-Germanie period, during which
there arOse the diffcrences that Primitive West
Germanie. Again, on the basis of peculiarities common to English
and Frisian (sueh as, espeeially, the lE, el for Primitive West
Germanie [a], which we noticed above), we may speak of a
period, during whieh therc occurred the changes
which led to Primitive Upon this there followed
a period, whieh leads to the forms that appear in our
earliest records of English. Thus, the comparative method re-
constructs uniform parent languages existing at points in time,
and deduces the changes which took place after eacb such parent
language split, up to the next following parent language or recorded
language. The comparative rnethod thus shows us the aneestry
of languages in the form of a family-tree, with successive braneh-
ings: the points at whieb branches separate are designated by
the word primitive; the branches between the points are
ignated by the prefix pre-, and represent periods of linguistic change
(Figure 1).
18. 10. The earlier students of did not realize
that the family-tree diagram was merely a statement of their
method; they accepted the uniform parent languages and their
sudden and clear-cut splitting, as historical realities.
In actual observation, bowcver, no speeeh-community is ever
quite uniforrn ( 3.3). When we describe a language, wc may
nore the lack of unfonnity by confining o\lrselvcs ta some
trarily chosen type of speech and leaving the other varieties for
later discussion, but in studying linguistic ehang;e we c:mnot do
312
THE COMPARATIVE METHOD
THE COMPARATIVE METHOD 313
this, because ail changes are sure to appear at first in the shape
of variant features.
Primitive Indo-European
FIGURE 1. (Above) Farnily-tree diagram of the relatiouship of the Indo-
European (Bclow) Part of a family-tree diagram, showiug the
epochs lU the hlstory of Euglish.
times, ta he sure, history shows us a suddcn cleavagc, such
as IS assumed by the comparative method. A eleavage of this
sort occurs when part of a community cmigratcs. Aftcr the Angles,
Saxons, and Jutes settled in Britain, they were fairly weil eut
off from their fellows who remained on the Continent; from that

IS{i:(P
. on the English language dcveloped independently, and any
between English and the continental dialects of
;:t Germanie can he taken, in the ordinary case, as evidence
for a feature that existed ?cfore the emigration of the English.
When the Gipsies, in the Middle Ages, started from northwestern
India on their endless migration, the changes in their language,
from that time on, must have bcen independent of whatever lin-
guistic changes occurred in their former home.
A less common case of elear-eut division of a spcech-community,
F.lGURE 2. Eastern Europe: the splitting. of speech-areas br inv.asion.
Once a unit, was split, in the earl}: MIddle Ages, \:>y the mtruslOn. of
SlaVlC. In the ninth cent ury this area, lU turn, wa.s split by the mtrusloll
of Hungarian.
preGermanie period
Primitive Indo.European
: English (aetual records)
preEngHsh period
Primitive AngroFrisian
preAngloFrisian period
Primitive West Germanie
preWest Germanie period
Primitive Germanie
'.
314 THE COIHPARATIVE METHOD
THE COMPARATIVE METHOD 315
is splitting by the intrusion of a foreign eommunity. the
Roman Empire, Latin was spoken over a solid area from Italy
ta the Black SM. In the carly Middle Ages, Slavs came in from
the north and settled sa as to cut this area complct.cly in two:
since that time, the dcvelopment of Roumanian, in the east, has
gone on independently of the development of the other Romance
languages, and a feature cammon ta bath Roumanian and the
western Romance lanl';uagcs is presumably guarantced as Latin.
ln the ninth century, the great Siavic area in turn suffered a
similar split, for t,he Ma/.,"yars (Hungarians), coming from the
east, settled sa as ta cut the Slavic area into a northern and
a southern part (see Figure 2). Since that time, accordingly, the
changes in South Siavic (Slovenc, Serbian, Bulgarian) have been in-
dependent of those in the northcrn area of Slavic, and any common
features of the two areas presumably date from beforc the split.
Such clear-eut splitting, however, is not usuaI. The differ-
ences among the Romance languages of the western area are evi-
dently not due to [1;Nll';raphic separation or ta the intrusion of for-
eign specch-communities. Aside from English and from Ieelandic,
the same holds good of the Germanic languages, including the
sharply defined difference bctwccn 'Vest Germanic and Scandi-
navian, whieh border on each other in the Jutland pcninsula. Evi-
dently sorne other historical factor or factors beside sudden sep-
aration may creatB several speech-eommunitics out of one, and in
this case we have no guarantee that ail changes after a certain
moment are independent, and therefore no guarantec that fea-
tures eommon to the daughter languages were present in the par-
ent language. A feature common, let us say, to French and Italian,
or to Dutch-German an Danish, may be duc to a common change
which occurred after SOIlle of the difference::; were aIready in
existence.
18, 11. Since the comparative method does not allow for varie-
ties within the parent language or for common changes in re-
lated laIll';lll1ges, it will carry us only a certain distance. Suppose,
for instance, that within the parent language thcre was sorne dialec-
tal difference: this dialectal difference will be reflected as an ir-
reconcilable difference in the related languages. Thus, certain of
the inflectional suffixes of nOuns contain an [ru] in Germanie and
Balto-Slavic, but a [bh] in the other Indo-Europcan languages, and
there is no parallel for any such phonetie corrcspondence.
(a) Primitive Indo-European *[-mis], instrumental plural:
Gothie ['wulfam] 'ta, by wolves,'
Primitive Indo-European *[-mi:s], instrumental plural: Lithu-
anian [nakti'mis] 'by nights,' Old Bulgarian [nostrmi],
Primitive Indo-European *[-mos], dati ve-ablati ve plural: Lith-
uanian [vil'kams] 'to wolves,' Old Bulgarian [v\komu],
(b) Primitive Indo-European *[-bhisJ, instrumental plural: San-
skrit [pad' bhih] 'by ft,' 0 Id Irish ['fenw] ' by men,'
Primitive Indo-European *[-bhjos], dati ve-ablati ve plural: San-
skrit [pad'bhjllh] 'to, from the feet,'
Prirnitivo Indo-European *[-bhos], dati ve-ablati ve plural; Latin
['pedibus] 'to, from the fcet,' Old Celtic [rua:trebo] 'to the moth-
ers.'
In cases like these, the comparative method does not show us
the form of the parent speech (whieh is cfined as a uniform lan-
guage), but shows us irreconeilably different forms, whose relation,
as alternants or as dialectal variants, it does not reveaI. Yet these
cases are very many.
On the other hand, if, like the older scholars, we insist that the
diserepancy is due to a eommon change in the history of Germanie
and Balto-Slavie, then, under the assumptions of the comparative
method, we must say that these two branches had 11 period of
cornmon development: we must postulate a Primitive BaIto-
Slavo-Germanic speech-community, which split off from Primitive
Indo-European, and in turn split into Germanie and Baito-Slavic.
If wc do this, however, wc are at once illvolved in contradictions,
beeause of other, discordant but over!apping, resemblances. Thus,
Balto-Slavic agrees ",ith Indo-Iranian, Armenian, and Albanesc,
in showing sibilants in certain forms where the other languages
have velars, as in the ward for' hundred':
Sanskrit [a'tam], Avestan [satDm], Lithuanian [Vsimtasl, but
Greek [he-ka'ton], Latin ['kentumJ, Old Irish [ke:(i], Primitive
Indo-European *[kJTl'tom]. Wc suppose that the parent language
in such cases had palatalized vc!ar stops.
Likewise, where the four branehcs just named have vclar stops,
there the others, in many forms, have combinations of velars with
a labial e1ement, or apparent modifications of thesc; \ve
that the parent language had labialized vclar stops, as in the in-
terrogative substitute stem:
Sanskrit [kah] 'who'!' Lithuanian [kas], Old Bulgarian [ku-to],
316 THE COMPARATIVE METHOD
THE COMPARATIVE METHOD 317
FIGURE 3. Sorne overlapping fcatures of special resemblanco among the
Indo-European languages, conflieting with the family-tree diagram.-
Adapted from Schrader.
1. Hibilants for velars incertain forms.
2. with [ml for [!lh].
3. Passive-voiee ndings with [ri.
4. Prcfix [t c-} in past tenses.
5. Feminine nouns wi t h masculine suffixes.
6. l'erfect lense used as general past tenso.
Again, we find special resemblances between Germanie and
Italie, as, for instance, in the formation and use of the past-tense
verb, or in sorne features of voeabulary (goal: Latin haedus;
Gothic gamains : Latin commnis 'eommon'). These, too, confliet
with the special rcsemblances between Germanie and Balto-
Slavie. In the same way, Italie on the one side shares peeuliarities
with Ccltic and on the other side with Grock (Figure 3).
18.12. As more and more of these rcsemblances were revealed,
the older scholars, who insistcd upon the family-trec diagram, faced
an insoluble problem. Whichcver special resemblances one took as
evidence for closer relationships, there rernuincd others, incon-
sistent with thcse, whch could be explained only by an entirely
dferent diagram. The decision, moreover, was too important to
be evaded, since in each case it profoundly altcred the value of
resemblances. If Germanie and Balto-Slavic, for instance, have
passed through a pcriod of eommon development, then any agree-
ment hetween them guarantccs nothing about Primitive
ropean, but if they have not passed through a period of common de-
velopment, then such an agreement, on the family-tree principle, is
practically certain evidence for a trait of Primitive Indo-European.
The reason for these contradictions was point.ed out in 1872
by Johannes Schmidt (1843-1901), in a famous essay on the
relationship of the Indo-European languages. Schmidt showed
that special rescmblances can be found for any two branches of
Indo-European, and that these special resemblances are most
numerous in the case of branches which lie geographically nearest
each other. Johannes Schmidt accounted for this by the so-called
wave-hypothesis. Different linguistic changes may spread, like
waves, over a speech-area, and each change may he carried out
over a part of the area that does not eoincide with the part covered
by an earlier change. The rcsult of successive waves wll he a net-
work of isoglosses ( 3.6). Adjacent districts will resemble eaeh
other most; in whatever direction onc travels, differences will
increase with distance, as one crosses more and more isogloss-lines.
This, indeed, is t.he picture presented by the local dialects in the
areas we can observe. Now, let us suppose that among a series of
adjacent dialects, which, to consider only one dimension, wc shaH
designate as A, B, C, D, E, F, G, ... X, one dialect, say F, gains
a political, commercial, or other predominance of sorne sort, so
that ts neighbors in either direction, first E and G, then D and H,
and then even C and l, J, K, give up their peculiaritics and in tirne
COme to spcak only the central dialect F. When this has happened,
F borders on B and L, dialects from which it differs sharply enough
ta produce elear-cut language boundaries; yet the resemblance
between F and B will be greater than that between F and A, and,
similarly, among L, M, N, ... X, tbe dialects nearest to F will
show a grcater resemblance to F, in spite of the clearly marked
__ -2- __
1
...
.- .......
"'
/ .
Balto-Siavic /
Ge"!"f) '.
,
al)' }
,
le ,
,
" -.,.- ....
/
, ... ....
--
... /'... Indolranian \
/'
1
...
1
Arme _
'l'"J
1q
/
1) /
/'
- -
...
/
'%
/
al)eSe
....
/

Cellie
-3_
, ....
/
1
1
\
\
\
\
"
"
but Greek ['po-then] 'from where?' Latin [kwo:] 'by whom, by
what?' Gothie [hwasJ 'who?' Primitive Indo-European *[kWos]
'who?' and derivati ves.
Only in a limited number of cases do the two sets of languages
agree in having plain velar stops. Accordingly, many scholars
suppose that the earliest traceable division of the Primitive Iudo-
European unity was into a western group of so-ealled "cenlum-
languages" and an eastem group of "salem-languages," although,
ta be sure, Tocharian, in Central Asia, belonged to the former
group. This division, it will be seen, clashes with any explanation
that supposes Balto-Slavic and Germanie to have had a common
period of special development.
318 THE COMPARATIVE METHOD
THE COMPARATIVE METHOD 319
boundary, than will the more distant dialects. The presentation of
these factors became known as the wave-theory, in contradistinction
to the aider family-Iree theory of linp;uistic relatiollship. T'oday wc
vicw the wave process and the splitting process merely as two
types - perhaps the principal types - of historical processes that
lead ta linguistic differentiation.
18. 13. The comparative rncthod, then, - our only method for
the reconstruction of prehistoric language, - would work accu-
rately for absolutcJy uniforrn speech-communities and sudden,
sharp cleavages. Sin,,' these presuppositions arc never fully
ized, the comparative method cannat c1aim ta picture the historical
process. 'Vhere the reconstruction works smoothly, as in the Tndo-
European word for father, or in observations of less ambitious
scope (such as, say, reconstructions of Primitive Romance or
Primitive Germanie), there wc are assured of the structural
turcs of a spccch-form in the parent language. 'Vherever the corn-
parison is at aIl ambitious as ta the reach of time or the breadth of
the area, it will revcal incommensurable forms and partial similar-
ities that cannat bc reconciled with the family-trcc diagram. The
comparative method Can work only on the assumption of a uniform
parent language, but the incommensurable forms (such as *[-mis]
and *[-bhis] as instrumental plural case endings in Primitive Indo-
European) show us that this assumption is not justified. The com-
parative method presupposes elcar-cut splitting off of successive
branches, but the inconsistent partial similarities show us that
later changes may spread across the isoglosses left by earlier
changes; that resernblance hetween neighboring languages may he
due ta the disappearance of intermediate dialeats (wave-theory);
and that languages already in sorne respects differcntiated may
make like changes.
Sometimes additional facts help us ta a decision. Thus, the
adjective Sanskrit ['pi:va:] 'fat,' Greek ['pi:o:n] occnrs only in
Indo-Iranian and Greck, but its existence in Primitive Indo-
European is guaranteed by the irregular formation of the feminine
fOrIn, Sanskrit ['pi:vari:], Greek ['pi:ejra]; neither language
formed new feminines in this way. On the other hand, the
manie ward hemp, Old English ['henep], l\'1iddle Dutch ['hannep],
and so on, corresponds to Greek ['kannabis]; nevcrtheless, learn
from Herodotus (fifth century B.e.) that hemp was known to the
Greeks only as a Foreign plant, in Thrace and Seythia; the word
caille into Greek (and thencc int.o Latin) and into Gcrmanie (and
thence, presumably, into Slavie) from some other language - very
likely from a dialcct. - at sorne time before the
Germanic changes of [k] to [hl and of [b] ta [pl. But for this pieee
of chance information, the corrospondence of the Groek and
Germanic forms would have led us ta attributo this ward to
Primitive Indo-Europcan.
18. 14. The reconstruction of ancient speech-forms throws sorne
light upon non-linguistic conditions of early tillles. If we consider,
for instance, that the composition of our earliest Indic records can
scarcely be placed later th/tn 1200 B.e., or that of the Homeric
poems later than 800 n.e., wC arc bound to place our reeonstruct.ed
Primitive Indo-European forrns at. least a thousand years earlier
than thesc dates. We can thus trace the history of language, often
in minute detail, much farthcr baek than that of any other of a
people's institutions. Unfortunat.c1y, wc cannat transfer our
knowledge to the latter field, espeeially as the meanings of speeeh-
forms are largely uncertain. 'Ve do not knw where Primitive
Indo-European was spoken, or by what manner of people; we
cannat link the Primitive Iudo-European to any
particular type of prehistoric abjects.
The noun and the verb snow appear 50 generally in the Indo-
European languages that wc ean exc1ude India from the range of
possible dwellings of the Primitive Tndo-European community.
The names of plants, even where thcre is phonetic agreement,
differ as ta meaning; thus, Latin ['fa:gus], Old English [bo:k] mean
'beech-tree,' but Greek [phe:'gos] means a kind of oak. Similar
divergences of meaning appear in other plant-names, such as our
words tree, birch, withe (German Weide 'willow'), oak, corn, and
the types of Latin saUx 'willow,' quercus 'oak,' horrleum 'barley'
(cognate with German Gerste) , Sanskrit ['javah] 'barky.' The
type of Latin glans 'acorn' occurs with the same meaning in Greck,
Armenian, and Balto-Slavic.
Among animal-names, cow, Sanskrit [ga:wh], Grcck [Ybows],
Latin [bo:s], Old Irish [bo:], is uniformly attested and guaranteed
by irregularities of form. Other designations of animaIs appear in
par: of the territory; thus, (joat, as we have seen, is confined ta
ermamc and Italie; the type Latin caper: 01d hafr 'goat'
?ccurs also in Celtic; the type Sanskrit [a'jah], Lithuanian [o'zi:s]
18 confined to thesc two and the type of Greek ['ajks]
320 THE COMPARATIVE METHon
appears aIso in Armenian and perhaps in Iranian. Other animaIs
for whieh we have one or more equations covering part of the
Indo-European territory, are horse, dog, sheep (the ward wool
is eertainly of Primitive Indo-European age), pig, wolf, bear, stag,
otter, beaver, goosc, duck, thrush, crane, eagle, fiy, bee (with
mead, which originally meant 'honey'), snake, worm, 6sh. The
types of our milk and of Latin lac' milk' are fairly widespread, as
are the ward lfoke and the types of our wheel and German Rad
'wheel,' and ofaxle. We may conclude that cattle were domesti-
cated and the wagon in use, but the other animal-names do not
guarantcc domestication.
Verbs for weaving, sewing, and other processes of work are
widespread, but vague or variable in meaning. The numbers
apparently included 'hundred' but not 'thousand.' Among terms
of relationship, those for a woman's relatives by marriage ('hus-
band's brother,' 'husband's sister,' and sa on) show widespread
agreement, but not those for a man's relatives by marriagc; Olle
concludes that the wife became part of the husband's family,
which lived in a large patriarehal group. The various languages
furnish severai equations fol' names of tools and for the metals gold,
silver, and bronze (or copper). Sevcral of these, however, are loan-
words of the typc of hemp; sa ccrtanIy Greek ['pelekus] 'axe,'
Sanskrit [para'uh] is connccted with Assyrian [pilakku], and our
axe and silver are ancient loan-words. Accordingly, scholars place
the Primitive Indo-European community into the Late Stone Age.
CHAPTER 19
DIALECT GEOGRAPHY
19. 1. The comparative method, with Hs assumption of unifonn
parent languages and sudden, de6nitive eleavagc, has the virtue
of showing up a residue of forms that cannat he cxplained on this
assumption. The conflicting large-seale isoglosses in the Indo-
European area, for instance, show us thaL the branches of the
Indo-European family did not arise by the sudden brcaking up of
an absolutely uniform parent community ( 18.11, Fip;ure 3).
We may say that the parent community was dialectally differ-
entiatcd before the break-up, or that aHer the break-up various
sets of the daughter communities remained in communication;
both statements amount to saying that arcas or parts of areas
which already differ in sorne respects may still make changes in
common. The result of successive changes, therefore, is a network
of isoglosses over the total area. Accol'dingly, the study of local
differentiaaons in a speech-area, dialect geography, supplements
the use of the comparative method.
Local differcnces of speech within an al'Ca have never escaped
notice, but thcr significance has anly of late been appreciated.
The eighteenth-century gramrnarians belicved that the literary
and upper-class standard language was oider and more true ta a
standard of reason than the local speech-fonns, which were due
ta the ignorance and carelessness of cornmon people. Nevertheless,
one noticed, in time, that local dialects preserved one or another
ancient feature which no longer exsted in the standard language.
Toward the end of the eighteenth centnI'".Y thcrc began to appear
dialect dictionarieB, which set forth the lexical pcculiarities of non-
standard speech.
The progress of historical linWlistics showed that the standard
language was by no means the type, but had arisen, under
historical conditions, from local dialects. Standard
ghsh, for instance, is the modern form not of literary Old
English, but of the old local dialect of London which had become
first a provincial and then a national standard language, absorbing,
321
322 DIALECT GEOGRAPHY DIALECT GEOGRAPHY 323
meanwhile, a good many fOnTIS from other local and provincial
dia1ccts. Opinion now turned ta the other extreme. Because a.
local dialect preserved sorne forms that were extinct in the stand.
ard language, it ,vas viewed as a survival, unchanged, of some
ancient type; thus, we still hear it said that the speech of some
remote locality is "pure Elizahethan English." Recause the
mixture of fOnTIS from other dialeds had been obscrved only in the
standard language, one jumped at the conclusion that local dialects
were free from this admixture and, therefore, in a historical sense,
more regular. At this stage, accordin!!;ly, we find dialect grammars,
which show the relation of the soumis and inflectiom, of a local
dialed to those of sorne aider stap;e of the language.
Investigation showed that every language had in many of its
fOnTIS suffered dispiacements of structure, which ,vere due to the
admixturf) of fomlS from other dialects. Old English [f], for in-
stance, normally appears as [f] in standard En?;lish, as in father,
foot, fill, five, and so on, but, in the '.'lords vat and viren, from Old .
English [fd] and ['fyksen] 'female fox,' it appears as [v], evidently .
because these fOnTIS are admixtures from a dialect whieh had
changed initial [f] to [v]; and, indeed, this initial [vl appears
regularlv in some southern English dialects (\Viltshire, Dorset,
Somers;t, Devon), in form3 like vut, vil, Sorne
students hoped, therdore, ta find in local dialects the phonemic
regularity (that is, adherence ta older patterns) that was broken
in the standard language. In 1876 a German scholar, Georg
Wenker, began, with this end in view, ta survcy the local dialecU
in the Rhine country round Dsseldorf; hter he extended his
survey ta coyer a wider area, and published, in 1881, six maps as
a first instalment of a dialed atlas of northern and centra.1 Ger-
many. He then gave up this plan in favor of a survey which '.'las to
caver the whole Gennan Empire. With government aid, Wenker
got fort,y test-sentences translatcd, largely by schoolmasters, into
more than forty-thousand German local dialects. Thus it was
possible ta mark the different local varictics of any one feature aD
a map, which would then show the geographc distribution. Since
1926 thcse maps, on a reduced scale, have becn appcaring in print,
under the editorship of F. "\Vrede.
The result, apparent from the very start, of 'Yenker's study, was
a surprise: thc local dialects were no more consistent than the
standard language in their relation to aIder speech-fonns. Dialect
eography only confirmed the conclusion of comparative study,
that different linguistic changes cover different porlions
of an area. The new approach yielded, however, a close-range view
of the network of isoglosses.
19.2. At present, then, we have three principal forms of dialect
study. The oldest is lexical. At first, the dia.1ect dictionaries
included only the forms and meanings whch differed from standard
usage. This critcrion, of course, is irrelevant. Today '.'le expect a
dictionary of a local dialect to give an the words that are current
in non-standard speech, with phonetic aceuracy and with reason-
able esre in the definition of meanings. A dialect dictionary for a
whole province or area is a mueh bigger undertaking. It should
give a phonemic scheme for each. local type of speech, and thcre-
fore can hardly be separated from a phonologic study. Wc ex-
pect a statement of the geographic area in which every form is
current, but this statemcnt can be given far better in the form
ofamap.
Grammars of local dialects largely confine themselves to stating
the correspondence of the phonemes and of the inflectional fonus
with those of an older stage of the language. The modern demand
would be rather for a description sueh as one might make of any
language: phonology, syntax, and morphology, together with
copious texts. The history of the forms can he told only in con-
nection with that of the afCa as a whole, since cvery feature hru;
been changed or spared only in so far as some '.'lave of change has
reached or failed ta fCach the speakers of the local dialect. The
grammar of a who1e area represents, again, a large undertaking.
The first work of this kind, the single-handed perfonnance of a
man of the people, was the Davarian grammar, published in 1821,
of Johann Andreas Schmeller (1785-1852); it is still unsurpassed.
For English, '.'le have the phonology of the English dialects in the
fifth volume of Ellis's Early English Pronunciation, and Joseph
Wright's grammar, published in connection with his Englh
Dialect Dictionary. Here tao, of course, '.'le demand a statement of
the topographie ext.ent of each feature, and this, again, can bc
m.ore clearly given on a mup.
Exoopt for the complete and organized description of a single
local dialect, then, the map of distribution is the c1Carcst und most
compact form of statement. The dialect atlas, li sct of such maps,
allows Us to compare the distributions of different foatures by
324 DIALECT GEOGRAPHY DIALECT GEOGRAPHY 325
comparing the different maps; as a practical help for this
parison, the German atlas provides with each map a loose trans..
parent sheet reproducing the principal isoglosses or other marks
of the map. Aside from the self-understood demands of accuracy
and consistency, the value of ft map depends very largely on the
completeness with which the local dialects are registered: the
finer the network, the more complete is the tale. In arder to record
and estimate a local form, however, we need to know its structural
pattern in terms of the phonemic system of the local dialect.
Furthermore, several variant pronunciations or grammatical or
lexical types may be current, with or without a difference of
denotation, in a local dialect, and the"e variants may be decidedly
relevant to the history of the change which procluced thorn. Finally,
to reproduce thc ,,,hole grammar and lexicon would rcquire so
vast a number of maps that even a very large atlas can only give
samples of distribution; we ask for as many maps as possible. In
view of aIl this, a dialcct atlas is a tremendous undertaking, and in
practicc is likely ta fall short in one or another respect. The
sentenccs on which the German atlas is based, were written clown
in ordinary German orthography by schoolmastenl and other
linguistically untrained pcrsons; the material does not extend to
great parts of the Dutch-German area, such as the Netherlands
and Belgium, Switzerland, Austria, Baltic German, Yiddish, Tran-
sylvanian, and the other speech-islands. The data are largely
phonologic, since the informant, except for striking lexical or
grammatical differences, would mcrcJy transcribe the forms into
a spellng that represented the local pronunciation; yet the pho-
nologie aspect is precisely what will he least clear in such a tran-
scription. The data for the French atlas were collected bya tmined
phonetician, Edmond Edmont; one man, of course, courd visit
only a limited numbcr of localities and stay but a short time in
each. Accordingly, the maps register on!y something over
hundred points in the French area (France and adjoining strips
in Belgium, Switzerland, and Italy), and the forms were collected
in each case from 11 single informant by means of a questionnaire
of sorne two-thousand words and phrases. Hmyever fine his car,
Edmont could not know the phonologie pattern of each local
dialect. The results for both phonetics and lexicon are more
copious than those of the Germllll atlas, but the looseness of the
nctwork and the bck of who!e sentences are drawbacks. The atlas
'tself was planned and worked out by Jules Gilliron (1854-1926),
bas appeared in full (1896-1908), together with a supplement
for Corsica. An Italian atlas, by K. Jaberg and J. Jud, bas
been appearing since 1928; it tries for great accuracy and pays
close attention ta meanings. Smaller atlases exist for Swabia
(by H. Fischer, 28 maps, published, in connection with a careful
treatise, in 1895), for Denmark (by V. Bennicke and M. Kristensen,
1898-1912), for Roumania (by G. Weigand, 1909), for Catalonia
(by A. Griera, 1923 ff.), and for Brittany (by P. Le Roux, 1924 ff.).
Other atlases are in preparation, including a survey of New England
under the direction of H. Kurath. A single-handed obscrver can
cover a smaU part of an area, as did Karl Haag in his stndy of a
district in Southern Swabia (1898); or else, he may restrict himsclf
to one or two features but follow them over a larger district, as
did G. G. Kloeke in his study of the vowel phonemes of the words
mouse and house in the Netherlands and Belgium (1927).
Needlcss to say, the map or atlas may he accompanied by a
treatise that interprets the faets or accounts for their origin, as
in the publications of Fischer, Haag, and Kloeke. The great
atlases have given rise to many studics, such as, notably, Gil-
liron's various books and essays, based on the French atlas, and
a whole series of studies, under the editorship of F. Wrede, by
workers on the German maps.
19.3, Our knowledge is confined, so far, ta the conditions that
preval in long-settlcd areas. In these, there is no question of
uniformity over any sizable district. Evcry village, or, at most,
every c1uster of two or three village,;, has its local peculiarities of
speech. In general, it presents a unique combination of forms, each
of which also appears, in other combinations, in sorne of the neigh-
boring localities. On the map, accordingly, each seUlement or
amall cluster of settlements will be eut off from each of its neigh-
bors by one or more isoglosses. As an example, Figure 4, reproduc-
ing a sman portion of Haag's map, shows the Swabian village of
Bubsheim (about ten miles east by southeast of Rottweil). The
nearest neighbors, within a distance of less than five miles, are ail
separated from Bubsheim by isoglosses; only two of these
hors agree with each othcr as to ail of the features that were
studied by Haag. The appended tab1c (Figure 5) shows under the
name of cach locality, the forms in which its dialect differs from
the forms of Bubsheim, which are given in the first column; where
326 DIALECT GEOGRAPHY DIALECT GEOGRAPHY 327
...--
z
"

"
'0$
<> <>
III
.Cl El b.o
"
2l
'l.
'"
.Cl

'0$ ....
<>
8
"
,..::l
0
,.
'" "
" ..Cl
;il

";.; if
""
.=;

z
.Cl 'Oi' oS d
" A

'" "
'l.
..,
'0$
E
<>
" .Cl
'0$
8
'0
(ll
,
,.
"
'l.
..,

<>
" :'1 .Cl
'0$
III

i!!
.. -
"
...,
i:l
'<]
'1.
<>
'"
.c
'"
...,
-0

'3
...,
z'" .r;; '<]
"=
III " .:::
'"
..,

.:0:
:..
1
0 -",
0 ....
-",
;" 8 0
.:::
b.o
i!!
>
-;::..
a a'J
'0
;" ::l
.3 .::: 0

.g
c
"
:::1 :;; .c
"
'-'
:- ;g
dl
. . :"
"
'" "'
"
.:2- -'"
"
p:;
{;
,::;>

Ifl
:
...,
w ..,
:::1
'"
'l. 01
"
00
.,
s .s 'Cl
...
bOl ;:l +-' .-0 w ...,
,....;
C' M ..;< >ri <0 00 ci
:=:
--'---
1
4

Mahlslelten

Bottingen

6
FIGURE 4. Isoglosses around the German villa[';e of Bnbsheim (SwlI.bia),
after Haag. The village of Denkin[';en has beeu added, with a fcw of its iso-
glosses, in order ta show the rceurrt'nce of Lille 6.
no form is given, the dialect ap;rces with Bubsheim. The number
before each form is the same as the number attached ta the cor.
responding isogloss in Figure 4.
If we followed the further course of tbese isogIOf\SeS, we should
find them running in varioui:\ directions and dividing the territory
iuto portions of differing size. The isoglosses numbcrcd 1, 2, and
3 in our Figures, cut boIdly across the German arca; Bubsheim
agrees, a" to these features, with the south and southwest. In
contrast with these important lines, others, sueh as our number 9,
surrollnd only a small district: the form ['tru:"ke] 'drunk,' ",hich
is listed for Denkingen, is spokcn only in a small patch of settle-
ments. The isogloss we have nurnbered as 6 appears on our map
as t wo li nes; theRe are really parts of an irrep;ularIy windinp; line:
Denkingen agrecs with Bubsheim as to the vowel of the verb
mow, although the intermediate village" speak diITercntly. We
find cven isoglosses whieh divide a town inta two parts; thus,
alonp; the lower Rhinc, just southwcst of Duisburg, the town of
KaldenhauRen is cut through by a bundle of isoglosses: the eastern
and western portions of the town speak different dialects.
The reason for this intense local diffcrentiation iR cvidently to
be sought. in the principIc of denRity ( 3.4). Evcry speaker is
constantly adnpting his speech-habits 1.0 those of his inter!ocutors;
328 DIALECT GEOGRAPHY
DIALECT GEOGRAPHY
329
he gives up forms he has been using, adopts new ones, and, perhaps
oftenest of aIl, changes the frequency of speech-fonns without
cntirely abandoning any oid ones or accepting any that are really
new to him. The inhabitants of a settlement, village, or town,
howevcr, talk much more to each other than to persons who live
elsewhere. When any innovation in the way of spcaking spreads
over a district, the limit of this spread is sure to he along sorne line
of weakness in the network of oral communication, and these lines
of weakness, in 130 far as they are topographical lines, are the
boundaries betwoon towns, villages, and settlements.
THE NETHERLANDS
50
flm

UIUIl [m,h. n;:! J
(mm,hhsl
F1QUnJ.i; 6. Distribution of syllabie souods in the words mouse and /wuse l
the Netherlnnds. -- After Klockc.
19. 4. Isoglosses for different forms rarely eoincide along their
whole extent. Almost cvcry feature of phonetics, lexicon, or
grammar has its own area of prevalence - is bounded by its own
isogloss. The obvious conclusion has becn weIl stated in the form
of a maxim: Every word has ils own hislory.
The words mouse and house had in early Germanie the same
vowel phoneme, a long [u :J. Sorne modern dialects - for instance,
sorne Scotch dialects of English - preserve this sound apparently
unchang
ed
. Others have changed it, but keep the ancient structure,
in the sense that these two \Vords still have the same syllabie
pho
nemc
; this is the case in standard English and in standard
German, where both words have [awJ, and in standard Dutch,
where both have [l'ill]. In the study above referred ta, Kloeke
traces the syllubies of these two words through the present-day
local dialcets of Bclgium and the Netherlands. Our Figure 6 shows
Kloeke's map on a reduced seale.
An easlern area, as the map shows, has prcserved the Primitive
Germanie vowel ru:] in both words: [mu:s, hu:sJ.
Several patehes, of various size, spcak [y:] in bath words:
[my:s, hy:sJ.
A district in the extreme west speaks [0:J in bath words: [myi:s,
h0:S].
Agreat central lirca speaks a diphthong of the type [Oq] in bath
words: [mpqs, hl'illS]. Since this is the standard Duteh-Flemish
pronunciation, it prevails in the usage of standard speakers also
in the other districts, but this fact is not indieated on the map.
In these last three districts, then, the sound is no longer that
of Primitive Germanie and medieval Dutch, but the structure of
our two words is unehanged, in 130 far as they still agree in their
syllabic phoneme.
Our map shows, however, three fair-sized districts which speak
lu:] in the word mouse, but [y:] in the word house; hence,
sistently, [fllU:S, hy:sJ. In these districts the structural relatlon of
the two words has undergone a change: they no long;er agree as t.o
their syllabie phoneme.
We sec, then, that the isogloss which separates [mu:s] from
[my:s] does nol coincidc with the isogloss which scparates [hu:s]
froIn [hy:sJ. Of the two \Vords, mouse has preserved the ancicnt
vowel over a larger territory than house. Doubtlcss a study of other
words whieh contained [u:] in medieval times, would show us still
other distributions of lu:] and the thher sounds, distributions
which would agree only in part \Vith those of mouse and house.
At some time in the 1\1iddle Ages, the habit. of pronouncing [y:]
instead of the hitherto prevalent lu:] must have originated in sorne
cultural center - perhaps in FJanders - and sprcad from there
330 DIALECT GEOGRAPHY DIALECT GEOGRAPHY
331
over a large part of the area on our map, ncluding the central
district which today speaks a diphthong. On the coast at the
north of the Frisian area there is a Dutch-speaking district known
as he! Bilt, which was diked in and settled undcr the leadership of
Hollanders at the bcginning of the sixteenth century, and, as the
map shows, uses the [y:]-proll1111ciation. It is [y:], moreover, and
not the old ru:], that appears in the loan-words which in the carly
modem period passed from Dutch into the more east.erly (Law
German) dialeets of the Dutch-German area, and into forcign
languages, such as Russian and Javanese. The Dutch that was
carried ta the colonies, such as the Creole Dutch of the Virgin
Islands, spoke [y:]. The spelling-s in written documents and the
evidenee of poets' rimes confirm this: the [y:]-pronunciation spread
abroad with the cultural prestige of the great coastal cities of
Holland in the sixteenth and seventeenth cent.uries.
This wave of cultural expansion was checked in the castem part
of our district, where it conflicted with the expansion of another
and similar cultural area, that of the North German Hanseatic
cities. Our isoglosses of mouse and and doubtless many
others, are results of the varying balance of these two cultural
forces. Whoever was impresscd by the Hollandish official or
merchant, Icarned to speak [y:]; whoever saw his superiors in the
Hanseatic upper class, retained the old ru:]. The part of the popu-
lation which made no pretensions to clegance, must also have long
retained the [u:], but in the course of time the [y:] filtercd
cven to this class. This process is still going on: in parts of the area
where ru:] still prcvails - both in the district of [mu:s, hu:s] and
in the district of [mu:s, hy:s] - the peasant, when he is on his g-ood
bchavior, speaks [y:] in words where his everyday speech has
ru:]. This Bavor of the [y:]-variants appears strikingly in the shape
of hyper-urbanisms: in using the elegant [y:], the spe3,ker sorne-
times substitutes it where it is entirely out of place, saying. for
instance, [vy:t] for [yu:t] 'foot,' a ward in whieh neither nor
present-day upper-class Dutch ever spoke an [y:].
The word house will occur much oftener than the ward mouse
in official speech and in conversation with persons who represent
the cultural center; mouse is more confined ta homcly and familial'
situations. Accordingly, we find that the word house in the upper-
class and central form with [y:] spread into districts where the
word m011se hl1:1 persisted in the old-fashioned form with [u;]. This
shows us also that the Holland influence, and not the Hanseatic,
was the nnovator and aggressor; if the reverse had bcen the casc,
wc shou!d find districts wherc house had [u:] mouse had.[y:].
In the sixtecnth and seventeenth centurICs, even whlle the
[y:]_pronunciation was making its arose; it wOllld
seern in Antwerp, a st.ill newer pronUnelatlOn wIth [04J mstcad of
the hitherto elegant [y;]. This new style spread to the
dties and with this its fortune "las madc. The [0\.l]-pronunClat.lOn,
as in' standard Duteh hllis [h011S], m1s [mpt 1s], is today the only
truty urbane farm. On our map, the area of this [pq] looks as if it
had been bd on top of a former solid area of [y:], leaving: only
disconnected patehes uncovered along the edge. This picture of
disconneet.ed patches at the periphery is charact.oristie of aider
styles, in language or in ather activities, that have boeo superseded
by sorne new cent.ral fashion. It is characteristic, too, that the
more remote local dialects are taking up a feature, the [y:]-pro-
Dunciation, which in more central districts and in the more priv-
leged class of speakers, has long ago OOen superseded by a still
newer fashion.
19. 6. The map in our !ast exarnple could not show the occurrence
of the present.-day standard Dutch-Flcmish pronunciation with
[04] in the dist.ricts where it has not conquered the local dialect.s.
Ta show this would bo to cover our whole map with a dense and
minute of [0lJ]-forms, for the educated or social1y
placed persons in t.he whole area speak standard Dutch-Flemish.
The pcrsifltence of old fcat.ures is easier to trace than the occur-
rence of new. The oost dat.a of dialect. geography are furnished by
relie forms, which attest some older feature of speech. In 1876,
J. Winteler published ",hat was pcrhflps the first adequat.e study
of a single local dialect, a monograph on his nat,ive Swiss-German
dialect of the settlement Kerenzen in the Canton of Glarus. In this
study, 'Vinteler mentions an archaic impcrative form, [lux] 'let,'
irreguia.rly derived from the stem [Ias-J, and says that he is not
certain that anvane still uscd it at the time of publication; most
speakers, at rate, already used the widespread and more
regular form [las] 'let.' A 1ater observer, C. Streiff, writing in
1915, has not heard the old form; it has veen totally replaced by
[los].
In the same way, 'Vinteler quotes a yerse in which the Glarus
people are mocked for their use of the prcscnt-tense plural verb-
332 DIALECT GEOGRAPHY
DIALECT CEOGRAPHY
333
.FIGURE 7. The of G1arus, SwitzerIand. - In 1915 the shade<! a
81.111, the provincial [haJd, wajd] as pluraJs of ,. have" and" want 1.0" .rs::
c arca use<! the general Swiss--German forms [band, wand]: - fter
forI1lS [hajd] '(we, ye, they) have' and [wajdJ '(we, ye, they)
wa
nt
to,' forms which sounded offensively rustic ta thcir neighbors,
who used the more generally Swss provincial forms [hand, wand].
FortY years later, Streiff reports a similar verse, in which the
people of the central region of the canton (including the largest
FIGURE 8. The French speeeh-area. - A discontinuolls isoglos8 encloses the
two nlarginal shaded areas in whinh reflexes of Latin rnulturn "mueh, very"
are litill i u use. - After GamilIacheg.
community and scat of government, the town of Clarus) mock
the inhabitants of the outlying vaUeys for their use of these same
forms, [hajd, wajd]. Our Figure 7, based on Streiff's statements,
shows the distribution in 1915: the more urbane and widespread
[hand, waud] prevail in the central district along the river LintQI
334 DIALECT GEOGRAPHY
nIALECT GEOGRAPHY 335
Our figure shows us six disconnected and, as to commerce, remote
districts which still speak fonns without the addcd vowel, such
as [kwc: 11 'bo\\' l,' in one or more of these four words. These
, ery' a modern form of Latin trans 'through, beyond, exceeding,'
;d beaucoup [boku] 'very,' whieh represents a Latin *bonum
lpum 'a p;ood blow or stroke.' Figure 8 shows the two detached
:arginal areas in which modern forms of Latin multum are still
in use.
In Latin, the word fal/it rneant 'he, she, it deceivcs.' By way
of a mcaning 'i t fails,' this word came ta mean, in medieval French,
'it is lacking,' and from this there has developed the modern
French use of ilf1<l [i fol 'it is necessary; one must.' This highly
specialized development of meaning can hardIy have occurred
indepcnd!mtly in more than one place; the prevalcnce of the mod-
ern locution in the greater part of the French area must be due
to spread from a nter, presumably Paris. Figure 9 shows us,
in the unshaded district, the prevalence of phonetic equivalents
of standard French il faut in local dialeets. The shaded districts
use other forms, princpally reflexes of Latin calet 'it's hot.' It
is evident t.hat the modern form spread southward along the
Rhne, which is a great hig-hway of commercc. We sec here how
an isogloss running at right angles ta a highway of communication,
will not cross it with unchang-ed direction, but will swerve off,
mn paral1el with thc-highway for a stretch, and thcn cit.her cross
it or, as in our example, reappear on t.he other side, and then run
back before resuming its former direction. The bend or promon-
tory of the isogloss shows us which of the two speech-forms has
been spreading at the cast of the other.
19. 6. If we observe a set of relie forms that exhibil. sorne one
ancient feature, we gel. a striking illustration of the principle
that cach word has its own history. The Latin initial eluster
[sk-] has taken on, in the French area, an initial [e-], a so-called
prothetic i;OWel, as, for example, in the following four words with
which our Figure 10 is coneerned:
which includes the capital, Glarus, and communicates freely with
the city of Zurich (toward the northwest); th" old rusUc forms
are uscd in the three more romote valleys, including the settlement
of Korenzen.
The relie form, as this example shows, has the best chance of
survival in remote places, and therefore is likely ta appear in
FIGURE 9. The French speech-area. - The unshaded district uses reflexes
of Latin Jal/il in t.he rneaning "it is neccssary." The shaded areas use other
forms. - After Jaberg.
small, detached areas. The Latin farm mullum 'much,' surviv-
ing, for instance, in Italian mollo ['molto] and Spanish mucha
['muco] 'much,' muy [mui] 'very,' has been rcplacod in nearly
ail of the French area by words likc standard French trs [tre]
'ladder'
'bowl'
'write'
'schol'
scala
swlella
scribere
schola
['ska:laJ
[sku'tellaJ
['skri:bereJ
['skola]
)'IODER" ST.\."D.l.RD FRE"CH
chell,e [e!isl]
cuelle [ekljd]
crire [ekri:rJ
cole [ebI]
336 DIALECT GEOGRAPHY DIALECT GEOGRAPHY 337
districts include 55 of the 638 places that were obscrved by Ed-
mont ( 19.2). The districts are:
A. A fairly large area in Belgium, overlapping the political
border of the French Republic at one point (Haybes, Dcpartment
of the Ardennes), and covering 23 points of the Atlas.
FiGURE JO. The French spech-area. - The shaded districts speak re-
ftexes of Latin [sk-] wthout an added initial vowel. - After Jaberg.
B. A somcwhat smallcr area in the Departmcnts of the Vosges
and of Meurthe-t-Moselle, overlapping into Lorraine, 14 points.
C. The village of Bobi in Switzerland, 1 point.
D. Mentone and two other villages in the Dcpartment of Alpes--
Maritimes on the Itallan border, 3 points.
E. A fair-sized district along the Spanish border, in the Depart-
ment of Hautes-Pyrnes, and overlapping into the neighboring
Departments,.11 .. .
F. A smallmtenor dlstrlCt ln the hIll-country of the Auvergne,
Departments of Haute-wire and Puy-de-Dme, 3 points.
Number of places where forms without
Words in which forms
added vowel are still spoken
without added vowel
are still spoken
BV DISTRICTS
TOTAL
A 1 B C D E F
- .. -
ladder, bowl. write, school 2 2
1
ladder, bowl. write 11 1 12
-" .- f------
_..
ladder, bowl. schoo! 1 3 4
..-
bowl, write. schocl
4+
1
--
-_...
ladder, bowl 5 12
ladder, write -' 1 1

ladder, schocl 5 5
--
bowl, write

1

- 1-
3 13
ladder 2 8
.-
bowl 1 1
write 1 1
TOTAL 23 14 1 3 11 3

FIGURE Il. Prothet.ic vowel in French. - Occurrence of the forms in the
shaded areas of Figure 10, by communities.
What interests us is the fact that most of the settlements in
these baekward districts have adopted the prothetic vowel in one,
tw, or three of our words. Thns, in district B, the village of Sainte-
Marguerite (Vosges) says [co:l] 'ladder' and [kwd] 'bowl,' but,
in th' modern style, [ekrir] 'write' and [eko:l] 'school.' Moreover,
the dialects do not agree as ta the words in which the innovation i8
338 DIALECT GEOGRAPHY
DIALECT GEOGRAPHY 339
old [ew] bofore labial or velar:
19,7. The final resuIt. of the procoss of spread is the complete
submergence of the old forms. Wherc wc find a great area in which
som
e
linguistic change has bcen uniforrnly carried out, we may be
sure that the greater part. of the uniforrnity is duo to geographic
leveling. Sometirnes place-names show us the only trace of the
strugg
le
. In the German area gencrally, two anciont diphthongs,
whieh wc represent as [ew] and [iw] are still distinct., as in standard
New High German, with li:] for ancient [ew], Flge 'fly' (noun),
Knie 'knee,' Stiefvater 'step-father,' lief 'deep,' but, with [oj] for
ancient [i w], scheu 'shy,' teuer 'dear,' neun 'nino.' The dialect of
Glarus has apparently lost the distinction, as have adjoining
dialects, wherever a labial or velar consonant followed the diph-
thong:
old [iw]:
shy *['skiwhiz] [sy:x]
dear *['diwrjaz] lty:r]
nine *['niwni] [ny:n]
Apparontly, thon, these two old types are bath represented in
Glarus by modern [y:], in accordance with the general South-
German dovelopment. A single form suggests that the [y:] for old
[ew] is rcally an import.ation, namely, the word deep, Primitive
Germanie type *['dowpazl, which appcars in Glarus as [t.jf]. Our
suspicion that the diphthong [il is the aIder representative of
[ew] beforo labials and velars in this rcgion, is eonfirmed by a
place-name; [' xnj -grn: t], li teral1y 'Knee-Ridge.'
The southwestern corner of German-speaking Switzerland has
changed the old Germanie [k] of words like drink to a spirant [x]
and has lost. the precedinj1; nasnJ, as in ['tri:xJ] 'to drink.' This is
today a crass localism, for most of Switzerland, alonp; \vith the
rest of t.he Dutch-German area, speaks [k]. Thus, Olams says
['trrl)k<:ll 'to drink,' in accord wit.h standard German trinken.
Place-names, however, show us that the deviant pronunciation
once extcndcd over a much larger part of Switzerland. GIarus,
made; thus, in contrast with the preceding case, the village of
Gavarnie (Hautes-Pyrnes), in our district E, says ['ska:lo]
'ladder' and ['sko:lo] 'schooI,' but [esku'de:lo] 'bow1' and
[eskri' beJ 'write.' Only t wo paints, bath in district A, ha'le pre.
served the old initial type in ail four of our words; the othcrs show
various cornbinations of oId and ncw fonus. Figure 11 givcs, in the
first column, the combinations of words in which the old form is
still in use, thcn the numbcr of points (by districts and in total)
where each combination has survived. In spite of the great variety
Words in which
Number of places where forms without
forms without
added vowel are still spoken
added vowel
8Y DISTRICTS -J
are still spoken

f--
A(23) 8(14) C(1)
0
0(3) E(lI) F(3) f--
'iadder' 21 14 2 11 1 49
'bowl' 20* 6 1 3 3 2 35*
'write' 16 1 3 20
'school' 2 1 1 8 12
*O f l ~ $lOJllt ,5 dOlJbdvl
FIGURE 12. Pr<;lthetic vowel in French. - Occurrence uf the forms in the
shaded arcas of Figure 10, by words.
that appears in this table, the survey by individuaI words, in
Figure 12, shows t.hat t.he hOHlely terms 'ladder' and 'bowl' appear
morc often in the old form than do 'writc' and' schoo1 ' which are
. '
assocutted with official instit.utions and wit.h a wider cultural out,..
look. Ta be sure, at Bobi (dist.rict C) it is prccisely 'ladder' which
has the new farm, but wherever the field of observation is larger,
as in districts A, B, and E, or in the tot.al, the terms for' ladder'
and 'bowl' knd to lead in the nurnber of conservative forms.
jly
knee
step-
PRBlI"I"IVE
GERYlII-KI TYPE
*[' f1ewgo:nJ
*['knewan]
*[' StBWpa-J
GLII-RU8
['f1y:gaJ
[xny:]
['sty:f-fatar]
340 DIALECT GEOGRAPHY
DIALECT GEOGRAPHY
341
weU ta the east, alongside the cornrnon noun ['wIlJk;}l] 'angle
corner,' has the place-name of a mountain pasture ['wIXl;}]
ers,' and alongside [xrulJk] 'sick' (formerly, 'crooked ') the name
of another pasture 'Crank-Dale,' that is,
Valley.'
19.8. Dialect geography thus gives evidence as to the former
extension of linguistic features that now persist only as relie forms.
Especially when a feature appears in detaehed districts that are
separated by a compact area in which a competing feature is
spoken, the map can usually be interpreted to mean that the de-
taehed districts were once part of a solid area. In this way, dialect
geography may show us the stratification of linguistic features;
thus, our Figure 6, without any direct historical supplementation,
would tell us that the [u:]-forms were the oldest, that they were
superseded by the [y;]-fol'ms, and these, in turn, by the diphthongal
forms.
Since an isogloss presumably marks a line of weakness in the
densy of communication, ''le may expect the dia.Iect map to show
us the communicative conditions of successive times. The
habitants of countries like England, Germany, or France, have
always applied provincial names to rough dialectal divisions, and
spoken of snch things as "the Yorkshire dialcct," "the 8wabian
dialcct," or "the l'i'orman dialect." Earlier seholars accepted these
classifications without attempting ta define them exactly; it waS
hoped, later, that dialect geography would lead ta exact definitions.
The question gained intel'est from the wa'le-theory ( 18.12), sinee
the provincial types were examples of the differentiation of a
without sudden cleavage. 1'vloreover, the question took
on a sentimental intcrest, since t.hc provincial divisions largely
represent old tribrtl groupings: if the extension of a dialect, such
as, say, the "Swabian dialect" in Germany, eould be shawn ta
coincide with the area of habitation of an ancient tribc, then
guage would ag"ain he throwing light on the conditions of a bygonc
time.
In this respect, howevcr, dialect geography proved to be dis-
appointng. It showed that almost every village had its own dialee--
tal features, so that the whole area was by a network of
isoglosses. Tf one bcgan by setting up a list of characteristic pro
vincial peculiariiies, one found thcrn prevailing in a solid core,
but shading off at the edges, in the sense that each characteris-
tic was bordered by a set of isoglosses its pres-
ence in different words - Just as the house and mouse lsoglosses
for {y:] and ru:] do not coincide in the eastern Netherlands (Fig-
ure 6). A local dialect from the center of Yorkshire or 8wabia
or Normandy couId be systcmatically classed in terms of Hs prov-
ince, but at the outskirts of sueh a division there lie whole bands
of dialects which share only part of the provincial characteristics.
In this situation, moreover, there is no warrant for the initial
list of characteristics. If these were differently selected - say,
without regard ta the popularly current provincial classification
_ we should obtain entirely different cores and entirely different
zones of transition.
Aecordingly, sorne students now despaired of aIl classification
and announced that within a dialect area there are no real bound-
aries. Even in a domain such as that of the western Romance lan-
guages (Italian, Ladin, French, Spanish, Portuguesc) it was urged
that there were no real boundaries, but only graduaI transitions:
the difference between any two neighboring points was no more
and no less important than the difference between any two other
neighboring points. Opposing this view, sorne scholars held fast
ta the national and provinciaL classifications, insisting, pcrhaps
with sorne mysticai fervor 1 . on a terminology of cores and
zones.
It is true that the isoglosses in a long-settled area arc sa many
as ta make possible almost any desired classification of dialects
and ta justify almost any claim concerning former densities of
communication. It is easy ta see, however, that, without
dice of any kind, we must attribute more significance to sorne iso
glosses than ta others. An isogloss which cuts boldly acrOS8 a
whole area, dividing it into two nearly equal parts, or even an
isogloss which neatly marks off sorne block of the total area, is
more significant than a petty line enclosing a localism of a few
villages. In our Figures 4 and 5, isoglosses 1, 2, 3, which mark
off southwestern German from the rost of the German area, are
evidently more significant than, say, isogloss 9, which encloses
only a few villages. The great isogloss shows a feature which has
spread over a large domain; this spreading is a large event, simply
as a fact in the history of language, and, may refiect, moreovcr,
SOme non-linguistic cultural rnovement of comparable strength.
As a criterion of description, too, the largc division is, of course,
342
DIALECT GEOGRAPHY DIALECT GEOGRAPHY 343
make ['ma:ke] ['maxen]
l [ik] [ix]
sleep ['sla:pe] ['sla:fen]
thorp 'village' [dorp] [dorf]
pound fpunt] [pfunt]
bite ['bejte] ['bajsen]
t"al [dut] [das]
to [tu:] [tsu:]
The isoglosS3S of these and other forms that contain Primitive
Germanie [p, t, k] run in a great bundle, sometimes coinciding,
but at other times diverging, and evcn crossing cach other. Thu5,
round Berlin, the isogloss of make, t{)gether with a good many
ot:hers, a northward bend, so that there one says [ik] '1'
with unshtfted [k], but ['maxen] 'make' with [k] shifted ta [x];
on the other hand, in the west the isogloss of l swerves off in a
northwestrly direction, so that round Dsseldorf One says [ix}
'1' with the shifted sound, but ['ma:kenJ 'make' with the old
[k] preservcd.
more signifieant than small ones; in faet, the popular classifica.-
tion of dialects is evidently based upon the prevalence of certain
peculiaritics over large parts of an area.
Furthcrmore, a set of isoglosses running close togethcr in much
the same direction - 11 so-called bundle of isoglosses - evidences
a larger historieal proeess and offers a more suitable basis of classi.
fication than does 11 single isogloss that represents, perhaps, saIne
unimportant featurc. It appears, that these two char.
acteristics, topographie importance and bundling, often go hand
in hand. Thus, Franee is divided by a grcat bundle of isoglosses
running cast and west aeross the area. This division refleets the
medieval division of France into the two cultural and linguistic
domains of Fn)Qeh and Provenal.
The most famous bundle of this kiwI, perhaps, is the east-and-
west bundIe which runs aeross the Dutch-German area, separat-
ing Low German from High German. The difference is in the treat-
ment of the Primitive Germanie unvoiced stops [p, t, k], which in
the south have been shifled to spirants and affricates. If wc take
standard Dutch and standard German as representativcs of the
two types, our isoglosses separate forms like these:
NORTHERN SOUTllBllN
In this way we flnd t.hat the topographie distribution of lin-
guistic features within a dialect area is not Indifferent, and
exhibits decided clcavagcs. 'Ve must make only two obvious
reservations: we cannat guarantee to preserve the popular termi-
nology by provinces, but, if we retain provincial names, must
den
ne
them; and we Can bound our divisions either imperfectIy,
by wnes, or arbitrarily, by selecting sorne one isogloss as the rep-
resentative of a whole bundle.
19.9. Having found the linguistic divisions of an area, we may
compare thern with other lines of cleavage. The comparison
shows that the important lines of dialectal division run close to
political lines. Apparently, cornrnon governrncnt and religion,
and especially the custorn of intermarriago within the political
unit, lead to relative uniformity of speech. Tt is estimated that,
under older conditions, a new political boundary led in less than
fifty years to some linguistic difference, and that the isoglosses
slong a political boundary of long standing would persist, with
Httle shifting, for sorne two-hundred years after the boundl1ry
ha<! been abolished. This soems to be the primary correlation.
If the important isoglosses ag;ree with other lines of cultural di-
vision - as, in northern Germany, with a difference in the con-
struction of farm-houses if they agree with geographic
barriers, such as rivcrs or mountain-ranges, then the agreement
is due merely to the fact that these features also happen t con-
cord with politieal divisions.
This has been shown most plainly in the distribution of the
important German isoglosses along the Hhine. Sorne forty kilo-
meters east of the Rhine the of thc great bundle that
separates Low German and IIigh German bcgin to separate and
spread out northwestward and southwestward, so as to form what
bas been called the "Rhenish fan" (Figure 13). The isogloss of
northern [k] versus southern [x] in the ward make, which has been
arbitrarily, as the criticalline of division, crosses the Rhine
Just north of the town of llcnrath and, accordingly, i.e; called the
"B
enrath line." It is found, now, that this line corresponds
l'Oughly to an ancient northern boundary of the territorial do-
mains of Berg (cast of the Rhine) and Jlieh (west of the Rhine).
The isogloss of northern [k] versus southcrn [x] in thc ward l
8werves off northwestward, crossing the Hhine just north of the
\':illage of rdingcn, and ia known aecordingly, as the "rdingen
344 DIALECT GEOGRAPHY DIALECT GEOGRAPHY 345
Hne;" sorne studcnts take this, rather than the Hne of make, as
the arbitrary boundary between Low and High German. The
rdingen line corresponds closely to the northern boundaries of
the pre-Napoleonic Duchies, abolished in 1189, of Jlich and
Berg - the states whose earlier limit is reftected in the Benrath
]ine - and of the Electorate of Cologne. Just north of rdingen
., 1
the town of KalJenhausen is split by the Urdingen ]jne into a
western section which says [ex] and an eastern which says [ek]i

S L'" VIC
j-LL
FIGURE 13. The Duteh-German speech-area, showing the isogloss of [kl
versus [x] in the word Jnake, a.nd, in the western part, the divergence of three
other i8og1osse8 wruch in the East run fairly close to that of make. - After
Behaghel.
we learn that up ta 1789 the western part of the town belonged
to the (Catholic) Electorate of Cologne, and the eastern part ta
the (Protestant) County of Mrs. Our map shows also two iso-
glosses branching southwestward. One is the !ine between
ern [pl and southern [fI in the word [dorp dorl] 'village'; tbiB
line agrees roughly with the southern boundaries in 1789 of Jlicb,
Cologne, and Berg, as against the Electorate of Treves. In a still
more southcrly direction there branches off the isogloss between
northern [t] and southern [s] in the word [dat - das] 'that,' and tbis
liDe, again, coincides approximately with the old southern bound-
arY of the Electorate and Archbishopric of Treves.
Ail this shows that the spread of fe.atures depends
upon social conditions. The factors in this respect are doubtless
the density of communication and the relative prestige of differ-
eut social groups. Important social boundaries will in time at-
tract isogloss-lines. Yet it is evident that the peculiaritics of the
severallinguistic forms themselves play a part, since each is likely
ta show an isogloss of its own. In the Netherlands we saw a now
form of the word house spreading farther than a new fonu of the
homely word mouse ( 19.4). We can hope for no scientifically
usable analysis, such as would enable us to predict the course of
every isogloss: the factors of prestige in the speakers and of mean-
ing (including connotation) in the forms eut off our hope of this.
Nevertheless, dialeet geography not only contributes to our under-
standing of the extra-linguistie factors that affect the prevalence
of linguistic forms, but also, through the evidence of relie fonus
and stratifications, supplies a great many details concerning the
history of individual fonus.
PHONETlC CHANGE 347
CHAPTER 20
PHOKETlC CHANGE
20. 1. '\Vritten records of earlier speech, rescmblancc betwcen
languages, and the varieties of local dialects, ail shO\v that lan-
guages change in the course of time. In our Old English records
wc find a word stan 'stone,' which we interprd phondically as
[sta:n); if we believe that the present-day English ,vord stone
[stown] is the modern form, by unbroken tradition, of this Old
English ward, then we must suppose tlmt OId English [a:] has
here changed ta modern [ow]. If wc bdicve that the rcscmblanccs
are due not ta accident, but to the tradition of speech-habits,
then we must infer that the differences between the rescmblant
forms are duc ta changes in these speeeh-habits. Earlier students
recognized this; they collectcd sets of resemblant forrns (dymol-
ogies) and infcrrcd that the diffcrenees between the forms of a
set wcre duc ta linguistie ehange, hut, \lntil the beginning of the
ninctcenth century, no one sllceeeded in classifying these differ-
enees. The rescmblances and difTerences varied from set to set.
An Old English bat, which we interpret phonetically as [ba:t],
is in one meaning paralleled by modern English boat [bowt], but
in another meaning by modern English bait [bcjt]. The initial con-
sonants are the same in Latin dies and English day, but dillerent
in Latin duo and English two. The rcsults of linguistic change
prcsented themsclves as a hodp;e-podge of resemblances and dif-
flOrences. One could suspect that sorne of the resemblances were
merdy accidentai (" false etymologies -"), hut t.here was no test.
One could reuch no clel1f formulation of linp;uistic relationship-
the l ~ s s so, since the persistence of Latin documents through the
Middle Ap;cs alongside of documents in the Romance languages
distorted one's ,vhole view of linguistic chronology.
It is not. useless to look back at those tirnes. Kow t,hat we have
a method which brings order into the confusion of linj'l;Uist.ic re--
scmblances and throws somc light on the nature of linguistic
relationship, wc arc lkely ta forget how chaotic are the results
of linguistic change when one has no key to their classification.
346
Since the beginning of the nineteenth cent.ury we have learned ta
!assify the differences between related forrns, attributing them
~ several kinds of linguistic change. The data, who$O variety
bewildered ear1ier students, lend them$Olves with facility t.o this
classification. Resemblanees which do not fit into our classes of
change, arc relatively few and can often he sufely ruled out as
accidentai; tbis is the case, for instance, with Latin dies: English
day, which wc now know ta be a false otyrnology.
The process of linguistic change has never been directly ob-
served; we shall see that such observation, with our present fa-
cilities, is inconceivable. We are assuming that our method of
classification, which works well (thouf';h not by any means per-
fectly), refiects the actual factors of change that produced our
data. The assumption that the simplest classification of observed
facts is the true one, is eornmon ta ail science; in our case, it is
weil ta remember that the observed facts (namely, the results of
linguistic change as they show themselves in etymologies) r e ~
sisted aIl comprehension until our method came upon the scene.
The first step in the development Qf method in historical lin
guistics was the seeking out of uniform phonetic corresponrlences;
we take these correspondencel'.' ta be the rcsults of a factor of
change which we cali phonetic change.
20.2. At the heginning of the ninetccnth eentury we find a
few scholars systematically picking out certain types of resem-
blance, chiefiy cases of phonetic agreement or correspondence.
The first notable step was Rask's and Grimm's observation ( 1.7)
of correspondences bctween Germanie and other ludo-European
languages. From among the ehaotic rnassof resemblant forms, they
selected certain ones which exhibited uniform phonetic correlations.
Stated in prcsent-day terms, these correlations appear as follows:
(1) Unvoieed stops of the other languages are paralleled in
Germanie by unvoiced spirants:
[p - f] Latin ps : English foot; Latin piscis : English fish; Latin
paler: English fathe/';
[t - 0] Latin tres : English three; Latin lenuis : English thin;
Latin lactre 'ta he silent' : Gothic ['Bahan];
fk - h] Latin centum : English hundred; Latin caput : English
head; Latin corn : Englisb homo
(2) Voiced stops of the other languages are paralleled in Ger-
manie by unvoiccd stops:
348 PHONETIC CHANGE PIIONETIC CHA.NGE 349
[h - p] Greek ['kannabis] : English hemp;
[d - tl Latin duo: English two; Latin dens : English tooth; Latin
edere : English eaf;
[g - k] Latin granum: English corn; Latin genus: English kin'
Latin ager : English acre. '
(3) Certain aspirates and spirants of the other languages
(which we denote today as "refiexes of Primitive Ino-European
voiced aspirates") are paralleled in Germanie by voiced stops and
spirants:
Sanskrit [bh], Groek [ph], Latin [f], Germanie lb, \T]: Sanskrit
['bhara:mi] '1 bear, , Grcek ['phero:], Latin fero : English bear;
Sanskrit ['bhra:ta:], Greek ['phra:te:r], Latin frater : English
brother; Latin frangere : English break;
Sanskrit [dh], Greek [th], Latin [f], Germanie [d, ti]: Sanskrit
['a-dha:t] 'he put,' Greek ['the:so:] '1 shall put,' Latin fci '1
made, did' : English do; Sanskrit ['madhu] 'honey, mead,' Greek
['methu] 'wine' : English mead; Sanskrit ['madhjah], Latin medius:
English mid;
Sanskrit [h], Greek [kh], Latin rh], Germanie tg, 'Y]: Sanskrit
[han'sah] : English goose; Sanskrit ['vahati] 'he carries on a vehicle,'
Latin t'ehi! : Old English wegan 'to carry, move, transport'; Latin
hostis 'stranger, eoemy' : Old English giest 'guest.'
The only reason for assembHng cases like these is the helief
that the correlations are too froquent or in sorne other way too
peculiar to he due to chance.
20. 3. Students of language have accepted these correlations
(calling them, bya dangerous metaphor, Grimm's "law"), because
the classification they introducc is confirmed by further study:
new data show the same correspondences, and cases which do not
show these correspondenees lend themselves to other classifi
cations.
For instance, from among the cases which do not show Grirnm's
correspondences, it is possible to sort out a fair-sized group iD
which unvoiced stops [p, t, k] of the other languages appear also in
Germanic; thus, the [t] of the other languages is paralleled by
Germanie ft] in cases like the following:
Sanskrit ['asti] 'he is,' Greek ['esti], Latin est: Gothie [ist] 'is';
Latin captus 'taken, caught' : Gothie [hafts] 'restrained';
Sanskrit [as'Ta:w] 'eight,' Greek [ok'to:] Latin ocW : Gothie
['ahtaw].
Now, in aIl these cases the [p, t, k] in Germanie is immediatcly
rec
cded
by an unvoiced spirant [s, f, h], and a survey of the cascs
conform to Grimm's correspondences shows that in them
the Germanie consonant is never precedcd by these sounds.
Grimm's correlations have thus, by leaving a residue, led us to
tind another correlation: aHer [s, f, h] Germanie [p, t, k] parallel
the [p, t, k] of the other Indo-Europcan languages.
Among the te;;idual forms, again, we tind a number in which
initial voiced stops lb, d, g] of Germanie arc paralleled in Sanskrit
not by [bh, db, gh], as Grimm would have it, but by lb, d, g], and in
Greek not by the expccted [ph, th, kh], but by [p, t, k]. An example
is Sanskrit ['bo:dba:mi] '1 observe,' Grcck ['pewthomaj] '1
experience' : Gothie [ana-'biwdanJ 'tD cornmand,' Old English
['be:odan] 'tD order, announce, offer, , English bid. In 1862,
Hermann Grassmann (1809-1877) showed that this type of
relation appears wherever the next consonant (the consonant after
the intervening vowcl or diphthong) belongs to Grimm's third
type of corrcsponences. That is, Sanskrit and Greek do not have
aspirate stops at the beginning of two successive syllables, but,
wherever the related languages show this pattern, have the first
of the two stops unaspirated: corresponding to Germanie *[bewda-J,
we find in Sanskrit not *[bho:dha-] but [bo:dha-], and in Greek
not *[phewtho-] but [pewtho-]. Here tao, then, the residual data
which are marked Dff by Grimm's reveal a
correlation.
ln this case, moreover, we get a confirmation in the structure of
the languages. In Grcek, certain fonns have a reduplication ( 13.8)
in which the first consonant of the underlying stem, followed by a
vowel, is prefixed: ['do:so:] '1 shaH give, , ['di-do:mi] '1 give.'
Wc find, now, that for stems with an initial aspirat<l stop the
reduplication is made with a plain stop: [' the: so:] '1 shaH put,'
['ti.the:mij '1 put.' The same habit appears elsewhere in Greek
morphology; thus, there is a noun-paradigm with nominative
singular ['thriks] 'hair,' but other case-forms like the accusative
['trikha]: when the consonant after the vowel is aspirated, the
initial consonant is [1.] instcad of [th]. Similarly, in Sanskrit, the
normal reduplication repeats the first consonant: ['a-da:t] 'he
gave,' ['da-da:miJ '1 give,' but for an initial aspirate the redupIica-
tion has a plain stop: ['a-dha:t] 'he put,' ['da-dha:miJ '1 put,'
and similar alternations appear elsewherc in Sanskrit morphology.
350 PI-IOKETIC CHANGE PHONETIC CHANGE 351
*['bhewdhomaj] *['phewthomajJ ['pewthomaj]
*[' dhidhe:mi] *[' thithe:mi] [' tithe:mi]
*['dhrighrr] *['thrikha] ['trikha].
On the other hand, in the nominative siugular of the word for
'hair,' we suppose that there never was an aspirate aiter the vowe1:
Thcsc altcrnations are obviously rcsults of the dis.
covercd by Grassmann.
20. 4. If our correspondcnees arc not due t.o chance, they must
result from sorne historieal conncction, and this connection the
comparative method reconst.ructs, as we have seen, by t.he assump-
tian of cornmon descent. from a parent language. 'Vhere the related
languages agree, they are prcserving features of the parent
guage, such as, say, the fr] in the ward brother, the [m] in the words
mead and mid ( 20.2), or the [s] in the verb-forms for 'he is'
( 20.3). \Vhere the corrcspondence connect.s markcdly different
phoncmes, we suppose that one or more of the bnguages have
changed. Thus we st.ate Grimm's correspondences by saying:
(1) Primitive Indo-European unvoiced stops [p, t, k] changed
in pre-Germanc ta uovoiced spirants [f, G, h];
(2) Primitive Iodo-Europcan voiced stops lb, cl, g] chaoged in
pre-Germanie ta unvoiced stops [p, t, k];
(3) Primitive voiced aspirate stops [bh, dh, gh]
ehanged in pre-Germanie ta voiced st.ops or spirants [b, d, g], in
pre-Greek to unvoiced aspirate stops [ph, th, kh], in pre-Italie
and pre-Latin to [f, f, hl. In this case the acoustie shape of the
Primitive Indo-European phonemes is by no means certain, and
sorne scholars prefer to speak of unvoiced spirants [f, e, xl; sim
Harly, wc do not know whet.her the Primitive Germanie reflexes
were stops or spirants, but thesc doubts do not affect our con ,
clusions as to the phonetic patkrn.
The correspondenccs where [p, t, k] appear also in Germanie "
demand a restriction for case (1): immediately after a consonant
(those which aetually occur are [s, p, k]), the Primitive
pean unvoiced stQPS [p, t, k] were not changed in pre-Germanic.
Grassmann's correspondenees we state hist.orieally by saying
that at a certain stage in the history of pre-Greek, forms which'
contained two successive syllables with aspirate stops, lost the
aspiration of the first stop. Thus, wc recollstruct:

INDo-EUF\OPFJAN
>
PRE-GREEK
>
Primitive Indo-European *[dhriks] as Greek. We
infer a similar change for pre.-Ind?- a Ptlllllve Indo-
European *[bhewdho-] III Mnsknt as [bo:dha-], a
}'rilniti
ve
Indo-European "'[dhedhe:-] [dadha:-], on.
A further step in the reconstructlOll of the hlStoflcal events
roceeds from the fact that the loss of aspiration results in San-
in lb, d, g], but in Greek in [p, t, k]. This implies that the
Primitive Indo-European [bh, dh, gh] had already become un-
voiced [ph, th, kh] in pre-Greek when the loss of aspiration took
place. Since this unvoicing does not occur in Indo-lranian, we
oonelude that the de-aspiration in pre-Greek and the de-aspiration
in took place independently.
The interprctlttioo, then, of the phonetic correspondences that
appear in our rcsemblant forms, assumes that the phonemes of a
language are subjccl tu historical change. This change may be limited
to certain phonetic conditions; thus, in pre-Germanic, [p, t, k]
did not change to [f, 0, h] when another unvoiced consonant
immediately preceded, as in *[kaptos] > Gothie [hafts]; in pre-
Greek, [ph, th, kh] became [p, t, k] only when the next syllable
began with an aspirate. This type of linguistic change is known as
phonetic change (or sound change). In modern terminology, the
8ssumption of sound-change can be stated in the sentence:
nemes change.
20.6. When we have gat.hered t.he roscmblant forms which show
the recognized correlations, the remainders will offer two
evident possibilities. 'Ve may have stated a correlation too
narrowly or too widely: a more careful survey or the arrivaI of now
data rnay show the correction. A notable instance of this was
Grassmann's discovery. The fact that residues have again and
again revealed new correlations, is a strong confirmation of our
method. Sccondly, the resemblant forms may not be divergent
pronunciations of the same earIler form. Grimm, for instance,
mentioned Latin dies: English rlay as an etymology which did not
faU within his correlations, and sincc his time no amount of re-
Beareb has revealcd any possibiIity of modifying the otherwise
correlation-classes sa that they may include this set.
ilarl
y
, Latin habrc 'to have' : Gothie haban, Old High German
hab
en
, in spite of the striking resemblance, confiicts with types of
COrrelation that otherwise hold good. In such cases, we may
attribute the rescmblance t accident, mcaning by this that it is
352 PHONETIC CHANGE PHONETIC CHANGE 353
not due to any historieal connection; thus, Latin dies: EngIish CJ.y
is now regarded by everyone as a "false etymology." Or cise, the
resemblance may be due to grammatical resemblance of forIlls in
the parent language; thus, Latin habre 'to have' and Old High
German habn 'to have' may be descendants, respcctivcly, of two
stems, *[gha'bhe:-] and *[ka'bhe:-] which were morphologieally
parallel in Primitive Indo-European. Finally, our resemblant
forrns may owe their likcncss to a historieal connection other than
deRcent from a common prototype. ThuR, Latin dentlis 'pertain.
ing tD a tooth' and English dental reRemble each other, but do not '
show the correlations (e.g. Latin d : English t) which appenr in
Latin and English reflexes of a common Primitive Indo-European.
prototype. The reason is that dental is merely the English-speaker's ,
reproduction of the Latin word.
To sum up, then, the residual forms which do not fit into
nized types of phonetic correlation may be:
(1) descendants of a common ancestral form, deviant only he-
cause we have not correct.ly ascertained the phonetic
relation, e.g. Sanskrit ['bo:dha:mi] and English bid, before
Grassmann's discovery;
(2) not descendants of a common ancestral form, in which case
the rcscmblanee may be duc to
(a) accident, e.g. Latin dies: English day;
(b) morphologie partial resemblance in the parent language,
e.g. Latin habre : English have;
(c) other historical relations, e.g. Latin denialis : EngIish
dental.
If this is correct, then the study of residual resemblant forms
will lead us to discovcr new types of phonetic correlation (1), to
weed out false ctymologies (2a), to uncaver the morphologie struc-
ture of the parent speech (2b), or to recognizc types of linguistio
change other than sound-change (2c). If the study of rcsidual
forms does not lead ta these results, then our scheme is in
correct.
20. 6. During the first three quarters of the nineteenth century
no one, sa far as we know, ventured to limit the possibilities in the
sense of our scheme. If a set of resernblant forms did not fit into
the recognized correlations, scholars feIt free to assume that these
forrns were neverthelcss related in exactly the same way as the
normal forms - narnely, by way of descent from a common
tral form. They phrased this historically by saying that a
might change in one way in sorne forms, but might
change in another way (or fail to change) in other forms. A
Primitive Indo-European [dl might change to [t] in pre-Gennanic
. mos
t
forms, such as two (: Latin duo), ten (: Latin decem), tooth
f: Latin dens), eat (: Latin but romain unchanged in sorne
otber forms, such as day (: Latm dtes).
On the whole, there was nothing to be said against this view -
in fact, it embodied a commendable caution - unless and until
an extended study of residual fonus showed that possibilities
(1) and (2a, b, c) were realized in so grcat a number of cases as to
rule out the probability of sporadic sound-change. In the seventies
of the nineteenth century, several scholars, most notably, in the
year 1876, August Leskien ( 1.9), concluded that exactly this had
taken place: that the sifting of residual forms had resulted so
often in the discovery of non-contradictory facts (1, 2b, 2c) or in
the weeding out of false ctymologies (2a), as ta warrant linguists
in supposing that the change of phonemes is absolutely regular.
This meant, in terms of our mcthod, that aH rescmblances between
fm'IDS which do not fall intD the recognized correspondence-classes
are due ta features of sound-change which we have failed ta
recognize (1), or clse are not divergent forms of a single prototype,
either because the etyrnology is falf'>'C (2a), or bccause sorne factor
other than sound-change has led to the existence of resemblant
fonus (2b, c). Hstorically interpreted, the statement means that
sound-change is merely a change in the speakers' manner of
producing phoncmes and accordingly affects a phoneme at every
occurrence, regardless of the nature of any particular linguistic
form in which the phoneme happens to occur. The change may
COncern sorne habit of articulation which is common to severaI
phonemes, as in the unvoicing of voiced stops lb, d, gJ in pre
Germanie. On the other hand, the change may concern sorne habit
of articulating successions of phoncmes, and thcrefore take place
only unJer particular phonetic conditions, as when [p, t, k] in
Pre.Gennanic became [f, 6, h] whcn not preceded by another
IlOUnd of the same group or by [s]; similarly, [ph, th, kh] in pre-
Greek became [p, t, k] only when the next syllable began with an
aspirate. The limitations of these conditioned sound-changes arc,
of course, purely phonetic, since the change conccrns only a habit
of articulatory movement; phonetic change is independent of
PHO)J"ETIC CHANGE PHONETIC CHANGE 355
non-phonetic factors, such as the meaning, frcquency, homonymy,
or what not., of any particular linguistic form. In present-da;'
terminology the whole assumption can he briefiy put into the
words: phonemes change, sinee the term phoneme designates l'I
meaningless minimum unit of signaling.
The new principlc was adopted by a number of linguists, who
received the nickname of "neo-grammarians." On the other hand
,
DOt only seholars of the older gCDcratioD, snch as Georg Curtius
(1820-188;:', but also some younger men, most notably Hugo
Schuehardt (1842-1927), rcjccted the new hypothesis. The dis-'
Cllfifiion of pro's and con's has never ccased; linguists are as
much divided on thifi point today as in the 1870's.
A great part of this dispute was duc mercly to bad terminology.
In the 1870's, when technical terms were less precise than today,
the assumption of uniform sound-change. the obscure and
metaphorical wording, "PhoneLic laws have no exceptions." It
is evident that the t enn "law" has here no precise me(1ning, for
a sound-ehangc is not in any sense a law, but only a historicai
OCCUITen. The phrase "have no exceptions" is a very inexad
way of saying that non-phonet.ic factors, such as the frequency or
meaning of partic\J1ar linguistic forms, do not interferc with the
change of phonernes.
The real point at issue is the scope of the phonetic correspond-
ence-classes and the significance of the residues. The neo-gram-
marians claimed that the results of study justificd us in making
the correspondence-classes non-cont.radict.ory and in seeking a
complete analysis of the residues. 1f we fiay tilat Primitive Indo-
European [dl appears in Germanic as ft], then, according to the
neo-grammarians, the resemblancc of Latin dies and English MY
or of Latin dentlilis and English den/al, cannot be classed simply
as "an exception" - that is, historically, as due to the
manic speakers' faUure t make the l1sual change of habit - but
presents a problem. The solution of t!:lis problem is either the
abandonment of the etymology as due to accidentaI resemblanCEl
(Latin dies: English day) , or a more exact formulation of the
phonctic corrcspondence (Grassmann' s discovery) , or the recog-
nition of sorne other factors that produce resernblant forms (Latin'
dent ulis borrowed in English dental). The insiste,
part.icularly, that his hypothesis is fruitful in this last direction:
it sorts out t.he rescmblances that are due to factors other tha
O
phonetic change, and accordingly lcads us to an understanding of
tbese factors.
The adual dispute, then, conccrns the weeding-out of faIse
tymologies, the revision of our statements of phonetic corre-
and the recognition of lingllistic changes other than
soundchange.
20. 7. The opponcnts of the neo-grammarian hypothesis claim
that resemblances which do DOt fit int recognized types of pha-
Detic correspondence may he due mcrcly to sporadic occurrence or
deviation or non-occurrence of sound-change. Now, the very
foundation of modern historical Iinguistics consisted in the seUing
up of phonctic correspondenco-classes: in this way alone di.d Rask
and Grimm bring order into the chaos of rcsemblances WhlCh had
bewildered aIl earlier students. The advocates of sporadic sound-
change, accordin!!;ly, agree with the neo-grammarians in discard-
mg such etymologies as Latin dies: En!!;lish day, and retain only a
few, where the resemblance is striking, such as Latin habere : Old
High German haben, or Sanskrit [ko:kilah], Greek ['kokkuks],
Latin cuculus : English cuckoo. They admit that this lcuves us
no crit.erion of decision, but insist that our inability to draw a
tine does not prove anything: cxceptional sound-changes oc-
curred, even though \ve have no certain way of recognizing
them.
The neo-grammarian sees in this a serious viobtion of scien-
tific method. The beginning of our science was mnde by a proce-
dure which implied regularity of phonctic change, and further
advances, like Grassmann's discovery, were based on the same
implicit assllmption. It may Le, of course, that some otller as-
8umption wOllld lead t.o an even better correlation of facts, but
the advocates of sporadic sound-change orrer nothing of the kind;
they accept t.he results of the actual method and yet cluim to
plain some facts by a contradiclory mcthod (or lack of method)
which was triel! and found wanting through aIl the centuries that
preceded Rask and Grimm.
In the historical interpretation, the theory of sporadic
change faces a very serious difficulty. If we supposc that a form
ke r:uckoo resisted the pre-Germanic shift of [kJ to [hl and still
preserves fi Primitive Indo-European [kJ, thon we must a1so
Pose that during many generations, when the pre-Germanic
People had changed their way of Primitive Tndo-
356 PHONETIC CHANGE PHONETIC CHANGE 357
European [k] in most words, and were working on through SUc-
cessive acoustie types such as, say, [kh - kx - x - hl, they Were
still in the ward cuckoo pronouncing an unehanged Primitive Iudo-
European [k]. If such things happened, then every language would .
be spotted over with all sorts of queer, deviant sounds, in fonus .
which had resisted sound-change or deviated from ordinary .
changes. Aetually, however, a language moves within a limited '
set of phonernes. The modern English [k] in cuckoo is no different
from the [kJ in words like cow, calf, kin, which has developed
normally from the Primitive Indo-European [g]-type. We should
have ta suppose, thcrdore, that some later change brought the
prcserved Primitive Imio-European [k] in cuckoo into complete
equality with the Germanie [k] that refiects a Primitive Indo-
European [g], and, since every language moves within a limited
phonetic system, we should have ta suppose that in every case
of sporadic sound-change or resistance to sound-change, the
discrepant sound has been reduc'Cd ta some ordinary phonemic
type in Lime to escape the ear of t.he observer. Otherwi,,-e we should
find, say, in present-day standard English, a sprinkling of forms '
which preserved sounds from eightcenth-century English, carly
modern English, Middle English, ld English, Primitive G e r ~
manie, and sa on - not to speak of deviant sounds resulting from
sporadie changes in sorne positive direction.
Actually, the forms which do not cxhibit ordinary phonetio
correlations, conform to the phonemic system of their language
and are peculiar only in t.hcir correlation with other forms. For
instance, the modern standard English correspondents of Old
English [0:] show sorne decided irregularitics, but these consist
simply in the presence of unexpected phonemcs, and never in :
dcviation from the phanetic system. The normal representation
seCInS to be:
[a] befare [s, z] plus consonant other than [t]: goshawk, gosling,
blossom;
[J] before Old English consonant plus [t]: f oster, soft, sought
(Old English sahle), brought, Ihought;
[u] before [k] book, brook (noun), cook, crook, hook, look, rook,'
shook, took;
[0] before [n] plus consonant other than (tJ and bcfore consonant.
plus [r]: Monday, monlh; brother, mother, olher, rudder;
[aw] bcfore [nt] and [rl and frolll the cambination of Old Englisb.
[o:wl: don't; floor, ore, swore, toward, whore; blow ('bloom '), floUl,
glPw, groW, l o ~ (verb), row, stow;
[uw] otherwlse: do, drew, shoc, slew, tao, to, woo, brood, food, mood,
Mof, roof, woof, cool, pool, school, slool, tool, bloom, broom, doom,
gloom, loom, boon, moon, noon, soon, spoon, SU'oon, whoop, goose,
wosc, boot, moot, root, sool, booth, sooth, toolh, smooth, soothe, be-
hoove, prove, ooze.
If we take the correlation of Old English [a:] with these sounds
as normal under the phonetic conditions of eaeh case, then wc have
the following resdue of contradictory farms:
[ a] shod, fodder;
[aw] bough, slough;
[ e J Wednesday;
ra] blood, flood, enough, tough, gum, donc, must, doth, glove;
[ow] woke;
[ u] good, hood, slood, bosom, JOOI, and optionally hoof, roof,
broom, sool;
[uw] moor, roost.
Ail of these seven deviant types contain sorne ordinary English
phoneme; the [0], for instance, in blood, etc., i8 the ordinary [ o ] ~
phoneme, which represents Old English lu] in words like love,
tongue, son, sun, come. In every case, the dscrepant forms show
not queer sounds, but merely normal phonemes in a distribution
tbat runs couoter ta the cxpect.f1tions of the hstorian.
20. 8. As ta the correction of our correspondenee-groups by a
oareful survey of the rcsidual cases, the neo-grammarians soon
got a remarkable confirmation of their hypothesis in Verner's
treatment of Germanie forms with diserepant [b, d, gJ in place
of [f, e, hl ( 18.7). Verner collectBd the cases like Latin pater .-
Gothie ['fadar], Old English ['feder], where Primitive Indo-
European [t] appears in Germanie as [d, Ci], instead of [eJ. Now,
the Voicing of spirants between vowcls is ft very common fonn of
SOund-change, and has actually oceurred at various times in the
history of several Germanie languages. Primitive Germanie [el
appears as a voiced spirant, coinciding with the reflex of Primitive
Germanie [dl, in Old Norse, whieh says, for instance, ['bro:Cier],
'Vith the same consonant as ['faCier]. In Old English, too, t.he
Primitive Germanie [Il] had doubtless become voiced between
vowels, as in ['bro:tior], although it. did not coincide with rd],
the :reflex of Primitive Germanie [d], as in ['federJ. In bath Old
358 PHONETIC CHANGE PHONETIC CHANGE 359
Narse and Old Primitive Germanie [f] had beeome voiced
[v] between vowels, as in OldCnglish ofen ['aven] 'oven' (Old
Righ German ofan ['ofan]), coinciding with the [v] that repre.
scnted Primitive Germanie lb], as in Old English yfel ['YVel]
'evil' (Old Righ German ubil ['ybilJ). Nothing eould be more
natural, t.herefore, if one admitted the possibility of irrcgular
sound-change, than ta suppose that the voicing of intcrvoealie
spirants had begun sporadieally in some words already in pre-
Germanie time, and Hw.,t a Primitive Germanie *['frlder] alongside
*['bro:Scr] reprrsented merely the Leginning of a proeess ihat was
ta End ils eompletion in the Old Norse, Old English, and Olti Saxon
of our aetual records. Yet in 1876 Verner's study of the deviant
forms showed an unmistakable correlation: in a fair numwr of
cases and in eonvineing systematic positions, the deviant [b, d, g]
of Germanie appeared where Sanskrit and Greek (and therefore,
presumably, Primitive Indo-European) had an unaceented vowcl
or diphthong before the [p, t, k], as in Sanskrit [pi'ta:], Greek [pa-
'te:r] : Primitive Germanie *['fader], contrasting with Sanskrit
['bhra:ta:], Grcek ['phra:te:r] : Primitive Germanie *['bro:6er].
Similarly, Sanskrit [' vaurah] ,father-in-law,' reilec ting, pre-
sumably a Primitive Indo-European *['swekuros], shows in
manic the normal reflex of [hl for [k], as in Old Righ German
['swehar], but Sanskrit [va'ru:h] 'mother-in-law,' reflecting a
Primitive Indo-European *[swe'kru:s] appears in Germanie with
tg], as in Old High German ['swigar], representing the Primitive
Indo-European [k] after the unstressed vowel.
A confirmation of this rsuH was the facto t.hat the unvoiced
spirant [s] of Primitive Indo-European suffered the same change
under the same conditions: it appears in Germanie as [s], except
when the precedinp; syllabic was unuceented in Primitive Indo-
European; in this case, it was voieed in pre-Germanie, and appcars
as Primitive Germanie [z], which hter became [1'] in Norse and in
West Germanie. In a number of irregular verb-paradigms the
G<;rmanic languages have medial if, 0, h, s] in the present
tense and in the singular indicative-mode forms of the past
tense, but lb, d, g, z] in the plural and subjunctive forms of
the past t.cnse and in the past partieiple, as, for inst.ance, in Old
English:
('weor6an] 't bccome,' [he: 'wear6] 'he bccame,' but [we:
'wurdonJ 'we Lecarne';
['ke:os
a
n] 'ta choose,' [he: 'ke:as] 'he chose,' but [wc: 'kuron]
'wc chose';
['wesan] 'to he,' [he: 'wes] 'he was,' but [we: 'weron] 'wc were.'
This alternation, Verner showed, corresponds t.o the alternation
in the position of the word-accent. in similar Sanskrit paradigms,
as in the verb-rorms cognate with the abo\'e:
'['vartate:] 'he turns, bccomes,' [va-'varta] 'he turned, , but
[va--vrti'ma] 'wc turned';
*['jo:sati] 'he enjoys,' [ju-'Jo:sa] 'he enjoyed,' but [Ju-Jusi'maI
'we enjoyed';
['vasatiI 'he dwells,' [u-'va:sa] 'he dwclt,' but [u:si'rnu] 'we
dwelt.'
This was so siriking a confirmat.ion of the hypothesis of regular
sound-change, that the burden of proof now fell upon the op-
ponents of the hypothess: if the residual forrns can show such a
correlation as this, we may well ask for very good reasons before
we give up our separation of forms iIlt.n rceognized correspondences
and rcmainclers, and our prineiple of seanning residual forms for
new correspondences. '\le may doubt whether an observer who
was satisficcl wit.h a verdict of "sporaclic sound-change" could
evcr have diseo\'cred these correlations.
In a small way, the accidents of obso'Vation sometimes furnish
similar confirmations of our method. In the Central Algonquian
languages - for which we have no older records - wc find t.he
following normal correspol1dcnccs, which we may syrnbolizc by
"Primitive Central Aigonquian" reeonstructed fonns:
PUlM1TlVE
PL-UNS C!;;:>ITRAr,
Fox OJ1DWA ME:<IO)1I GR"!;; AL(10NQvIAN
(1)
hk sk k sk k
(2)
sk sk sk sk sk
(3)
hk hk hk sk xk
(4)
hk hk hk hk hk
(5)
k ng hk hk nk
Examples:
(1) Fox [kehkje:wa] 'he is old,' Menomini [krki:wJ, PCA
*[keckj c:waJ.
(2) Fox 'fire,' Ojibwa [iskudr.:], IVlenomini
[esko:k:w], Cree [iskute:w], PCA *[iskutr:wi].
(3) Fox [mahkese:hi] 'moccasin,' jibwa [mahkizin], Menomini
[mahkE:scn], Cree [maskisin], PCA *[maxkcsini].
(4) Fox [no:hkumcsa] 'my grandmother,' Ojibwa [no:hkumie],
Menomini [no:hkumeh], Cree [no:hkum], PCA *[no:bkuma].
(5) Fox [takeskawe:wa] 'he kicks him,' Ojibwa [tangiSkawa:d],
Menomini [tahl:skawe:vd, Cree [tahkiskawe:wJ, PCA
*[tankeska\"8:wa].
Now, there is a residual morpheme in which none of tbese
respondences holds good, namely the element which means 'red':
(6) Fox [meskusiwaJ 'he is red,' jibwa [miSkuzi], Menomini
[mehko:nJ, Cree [mihkusiw], PCA *[mekusiwa].
Under an assumption of sporadic sound-change, this would
have no significance. After the sixth correspondence had been set
up, however, it was found that in a remote dialect of Cree, which
agrees in groups (1) to (5) with the Plains Cree scheme, the mor-
pheme for 'red' has the peculiar c1uster [htkJ, as in [mihtkusiw]
'he is red.' In this case, then, the residual form showed a special
phonctic unit of the parent speech.
The assumption of rcgular (that is, purely phonemic) sound
change is justified by the correlations which it uncovers; it is n-
consistent to accept the results which it yields and to rcject it
whenever one wants a contradictory assumption (" sporadic
sound-change") to "explain" difficult cases.
20. 9. The relation of our residual forms to factors of linguistic
history other than sound-change, is the crucial point in the dis-
pute about the regularity of sound-change. The nco-grarnmariaos
could not elaim, of course, that linguistic resemblances ever run
in regular sets. The actual data with which we work arc extremely
irreguiar, - so irregular that centuries of study before the daye
of Rask and Grimm had found no uscful correlations. The nec-
grammarians did elaim, however, that factors of linp;uistic change
other than sound-change will appear in thc rcsidual forms ufter
wc have ruled out the correlations that resuh from sound-change.
Thus, Old English [a:] in stressed syllablcs appears in modern
English normally as [ow], as in boat (from id English [ba:tJ),
8ore, whole, oa/h, mow, :>tom, bone, home, dough, goat, and many
other forms. In the residue, we find forms like id English [ba:t] :
bait, Old English [ha:l] : hale, Old English [swa:n] 'herdsman' :
swain. Having found that Old English [a:] appears in modern
standard English as [owJ, we assign the forms with the discrepant
xood
ern
English [ejJ to a residue. The forms in this residue are
t the results of a deviant, sporadic sound-change of Old English
fl to modern English [ej]; their deviation is due not to sound-
but to another factor of linguietic change. The forms
Uke bail, hale, swain are not the modern continuants of Old English
form
s
with [a:], but borrowings from Scandinavian. Old Scandi-
navian had [ej] in forms where Old English had [a:]; ld Scandi-
navian (Old Norse) said [stejnn, bejta, hejn, swejnn] where Old
English said [sta:n, ba:t, ha:l, swa:n]. The regularity of corre-
spondence is due, of course, to the common tradition from Primi-
tive Germanie. After the Norse invasion of England, the English
language took over these Scandinavian words, and it is the Old
Norse diphthong [ej] which appears in the deviant forms with
modern English [ej].
In cases like these, or in cases like Latin dentalis : English dental,
the opponents of the neo-grammarian hypothesis raise no objcc
tion, and agree that linguistic bOTTOWing accounts for the resem-
blance. ln many othcr cases, however, they prefer to say that
irregular sound-change was at work, and, strangely enough, they
do this in cases where only the neo-grammarian hypothesis yields
a significant result.
8tudents of dialect geography are especially given ta this con-
fusion. In aoy onc dialect we usually find an ancient unit phoneme
represented by scveral phonemes - as in the case of Old English
[0:] in modern English food, good, blood, and so on ( 20.7). Often
one of these is like the old phoneme and the others appcar to em-
body one or more phonetic changes. Thus, in Central-Western
American English, wc say gathd with [12], rather with or with
[a], and father always with [a]. Sorne speakers have (juw] in words
like tune, dew, stew, new; sorne have [uw] in the first three types,
but keep [juw] ordinarily afrer [n-]; others spcak [uw] in aIl of them.
Or, again, if wc examine adjacent dialects in an area, we find a gra-
dation: sorne have apparently carried out a sound-change, as when,
Bay, in Dutch, sorne districts in our Figure 6 have [y:] for ancient
ru:] in the words mouse and house; next ta these we may find
dialects which have apparently carried out the change in sorne of
the forms, but not in others, as when sorne districts in our Figure 6
say [hy:s] with the changed vowel, but [mu:s] with the unchanged;
finally, we rcach a district where the changcd forms arc lacking,
&uch as, in Figure 6, the area where the old forms [mu:s, hu:s] arc
360 PHNETIC CHANGE PHNETIC CHANGE 361
362
PHONETIC CHANGE
PIlON ETIC CHANGE 363
still being spoken. Under a hypotho3is of spomdic sound-change
no dcfinite conclusions could he drawn, but under the
of rcgular sound-change, distributions of this sort can at once he
interpreted: an irregular distribution shows that the new forms
in a part or in aB of the area, are due Ilot ta sound-change, but
borrowing. The sound-chrlllge took place in sorne one cent.er and
after this, forros which had undergone the change spread from
center by linguistic borrowing. In other cases, a community may
have made a sound-change, but the changed forms may in part he
superseded by unchangcd forms which spread from a cent.er which
has not made the change. Student.s of dialect geography make this
inference and base on it their reconstruction of linguistic and
cultural movements, but many of these studcnts at the same tirne
profcss to reject the assumption of regular phonetic change. If
thcy stopped ta examine the implications of this, they would saon
sec that their work is basod on tho supposition that sound-chan!,'B
is regular, for, if we admit the possibility of irregular sound-change,
then the use of [hy:sl beside [mu:sJ in a Dutch dialect, or of ['ratir]
rather beside ['ge'5rl gather in standard English, woule! justify no
deduetions about linguistic borrowing.
20. 10. Another phase of the dispute about the regularity of
concerns residual forms whose deviation is connectee!
with features of meaning. Often enough, the forms that deviate
from ordinary phoneti c correlation belong ta sorne c1early marked
semantic group.
ln ancient Greek, Primitive Indo-European [s] betwecn vowels
had boen lost by sound-change. Thus, Primitive Indo-European
*[' gewso:] '1 taste' (Gothie [' ki wsal '1 choose ') appears in Greek
as ['gewo:] '1 give a tastc'; Primitive Indo-European *['genesos]
'of the kin' (Sanskrit [' j alll1sah]) appenrs as Greek j' geneosj, later
['genows]; Primitive Indo-European *['e:srp] '1 was' (Sanskrit
['a:sam]) appcars in Greek as [Ve:a], bter ['e:].
Over ap;ainst cases like t.bese, there is a considerable residue of
forms in which an olti intervocalic [s] secms ta ho preservcd in
ancient Greek. The principal type of this residue consists of aorist-
tense (that is, past punctual) verb-forms, in which the suffix [-s-J of
this tcnse occurs after the final vowe! of a root or verb-stem. Thus,
the Greek root [plow-] 'sail' (present tense ['plewo:] '1 sail,'
kled by Sani'Jkrit !'pinvate:] 'he sails') has the a01'i8t form ['eplewsaJ
'1 sailed'; Hw Greek aorist ['etejsa] '1 paid a penn,1ty' pt,raIlcls
Sanskrit ['ara:jsam] '1 collected'; the Greek root [stc>] 'stand'
( resent tense ['histe:mi] '1 callse to stand') has the aorst fonu
'1 t? stand,' with O:d, Bulgarian
'1 stood up,' Prnmtlve 1ndo-l', mopean type -[ esta:sTfl]; a Pnm-
itive Ioda-European aorist type *['cbhu:sJTl] (Old
[byxu] '1 becamo') is apparently represented by Greek ['ephu:sa]
'1 caused ta Opponents of the neo-grammarian rnethod
suppose that when inkrvocalic [s] was weakened and finally lost
during the pre-Greek period, the [sJ of thcse fonns resisted the
change, becausc it expressed an important meaning, namcly that
of the aor:st tense. A sound-chanp;e, they daim, can be chccked in
fOTIns v/here it threatcn::; to remove sorne semantically important
feature.
The neo-grammarian hypothesis implies that sound-change is
unaffectcd by Remantic fcaLures and concerns merely the habits of
articubting spcech-sounds. If residual fonus are characterized
by some semaIltic feature, then their deviation must be due not
ta sound-change, but to sorne other factor of Iinguistic change-
ta sorne factor which is connectcd with rneanings. In our exarnple,
the sound-change which lce! to the loss of intervocalic [s] destroyed
l'very intervocalic [s]; forrns like Greek ['este:saJ cannat be
tinuants of forms that existed before that sound-change. They
were created after the sound-change "l'aS past, as new combinations
of morpherncs in a complex form, by a process which we caU
analogie new combina/ion or analogie change. In many forms l',..here
the aorist-suffix was not betwecn vowcls, it had come unscathed
through the sound-change. Thus, a Primitive Indo-European
aorist *['ele:jkwsJTll '1 left' (Sanskrit ['ara:jksnmJ) appearl in
Greek, by normal phonetic development, as ['clejpsa]; Primitive
Indo-European *[eje:wksJTl] '1 joined' (Sanskrit ['aja:wksam])
appears as Grcek ['ezewksaJ; the Primitive Indo-European root
*[gews-] 'taste' (Greek present [' gewo :J, cited above), cornbining
with the aorist-suffix, would give a stem as double
lss] was not lost in pre-Greek, but merely at a Iater date simplified
t [s], the Greek aorist ['egewsa] '1 gave a taste' is the nonnal
phonetic type. Accordingly, the Greek language possessed the
aorist suffix [-s-]; at aU times this snffix was doubtless combined
with ail manner of verbal stems, and our aorists with the [+] be-
twecm vowels are merely combinations which were made after the
Sound-change which affected [-s-] had ceased to work. On models
364
PHONETIC CHANGE PHONETIC CHANGE 365
like the inherited present-tense ['gewo:] with aorist ['egewsa], one
formed, for the present-tense ['plewo:), a new aorist ['eplewsa]. In
sum, the residual forms are not due ta deflections of the process of
sound-change, but reveal ta us, rather, a diffcrent factor of lin-
guistic change - namely, analogie change.
In much the same way, sorne students bclieve that sounds whieh
bear no important meaning are subjeet ta excess weakening and
ta loss by irregular sound-change. In this way they explain, for
instance, the weakening of will ta {I] in forms like l'U go. The nea-
grammarian wauld attribute the weakenng rather ta the fact that
the verb-form in phrases Iike these is atonie: in English, unstressed
phonemcs have been subjeeted ta a series of weakenings and lasses.
20.11. The neo-grammarians define sound-change as a purely
phonetic process; it affects a phoneme or a type of phonemes either
unversally or under certain strictly phonctic conditions, and is
neither favored nor impeded by the semantic character of the forms
which happen ta contain the phoneme. Thc effeet of sound-change,
then, as it presents itself ta the comparatist, will he a set of regular
phonemic correspondences, such as Old English [sta:n, ba:n,
ba:t, ga:t, ra:d, ha:!]: modern English [stown, bown, bowt, gowt,
rowd, howl] stone, bone, boat, goat, road (rode), whole. However,
these correspondences ,,,Hl almost always be opposed by sets or
scatterings of deviant forms, such as Old English [ba:t, swa:n,
ha:1] versus modern English [hejt, swejn, hejl] bait, swain, hale,
because phonehc change is only one of several factors of linguistic
change. 'Ve must suppose that, no matter how minute and accurate
our observation, we should always tind deviant forms, hecause,
from the very outset of a sound-change, and during its entire
course, and after it is over, the fOnTIS of the language are subject to
the incessant working of other factors of change, such as, especially,
borrowing and analogie combination of new complex forms. The
occurrence of as defined by the neo-grammarians,
is Dot a fact of direct observation, but an assumption. The neo-
grammarians beleve that this assumphon is correct, because it
alone has enabled linguists to find order in the factual data, and
hecause it alone has Ied to a plausible formulation of other factors
of linguistic change.
Theoretically, we can understand the regular change of pho
nemes, if we suppose that language consists of two laycrs vf habit.
One layer i8 phonemic: the speakers have certain habits of voie-
'og tongue-movement, and so on. These habits make up the pho-
system of the language. The other layer consists of formal-
semantic habits: the speakers habi tually uUer certain cornbinations
of phonemes in response to certain types of stimuli, and respond
a,ppropriately when they hear these same eombinations. These
habits make up the grammar and lexicon of the language.
One may conceivably aequire the phonetic habits of a language
without using any of its Rignifieant forms; this may be the case of
a singer who has been taught to render a French song in correct
pronunciation, or of a mirnic who, knowing no French, can yet
imitat-e a Frenchman's English. On the other hand, if the pho-
nemes of a foreign language are Dot completcly incommensurable
with ours, wc may utter significant forms in this language without
acquiring its phonetic habits; this is the case of sorne speakers of
French and English, who converse freely in eaeh others' languages,
but, as we say, with an abominable pronuncttion.
Historically, we picture phonetic change as a graduaI favoring
of sorne non-distinctive variants and a disfavoring of others. It
could be observed only by mcans of an onormous mass of mechan-
ieul records, reaching through several generations of speakers. The
bypothcsis supposes that such a collection - provided that wo
could rule out the effects of borrowing and analogie change-
would show a progressive favoring of variants in some ono direc-
tion, coupled with the obsolescence of variants at the other ex-
treme. Thus, Old English and Middle Englsh spoke a long mid
Yowel in fonns like gos 'goose' and ges 'gocse.' \Ve suppose that
during a long period of timo, higher variants were favored and
lower variants went out of uso, until, in the eighteenth century, the
range of survivng variants could be describcd as a high-vowel
type [u:, i:]; since then, the more diphthongal varirwts have Leon
favored, and the simple-vowel types have gone out of use.
The non-distinctive acoustic features of a language are at al!
times highly variable. Even the most accurate phonctic record
of a language at any one time could not tell us which phonemes
were changing. Moreover, it is certain thflt these non-distinctive,
Bub-phonemc variant.s are subjcct to linguistic borrowing (imi-
tation) and to analogic change (systematization). This appears
from the fact that whcnever the linguist deals with a sound-change
- and certainly in sorne cases his documents or his observations
must date f,'Um a time very shortly after the occurrence of the
366 CHANGE
PHONETIC CHANGE
367
change - he finds the results of the sound-change disturbed by
thesc other factors. Ineed, when wc observe sub-phonemie Vari.
ants, we. somet,imes find them difltributed among speakers or
among forms, quite in the manner of linguistic
and,of change. In the Central-Western type
of AmerIcan hnghsh, vowcl-quantitics are not distinctive, but
some speakers habitually (though perhaps Ilot invariably) use
a s?ortor variant of the phoneme [a] bcfore the ciusters [rk, rp],
as. III dark, Bharp, and before the ciustcrs [rd, rt] followed by a
prImary suffix [-f, J;I-], as in barter, Carter, garden, marten (Jlartin).
Before a seeondary suffix, [-r, -l,], however, the longer variant is
used, as in starter, carter (' one who carts '), harden; here the exist-
ence of the simple words (start, cart, hard), whose [a] is not suh-
ject. ta shortening, has led to the favoring of the normal, longer
variant. The word larder (not part of the colloquial vocabulary)
could bc read with the shorter variant, but the agent-noun larder
('one who lards') could he formed only with the longer type of
the [aJ-phoneme. This distribution of the variants
is quite like the results of analogie change, and, whatever its
origin, the distribution of this habit among speakers is doubtless
e.ffeet.ed. by a of imitation whieh we could identify with
hngmstle borrowmg. If the diffcrence between the two variants
should become distinctive, then thc comparatist would say that
a sound-change had occurred, but he would find the results of
this sound-change overlaid, from the very start, by the effccts
of borrowing and of analogie change.
Wc can often observe that a non-distinctive variant has hecome
obsolete. In ightecnth-century English, forms like geese,
goat had long vowcls of the typtS ri:, c:, u:, 0:],
WhlCh smce then have changed ta the diphthongal t.ypes [ij, cj,
uw, ow]. This displacemcnt has had no bearing on the structure
of language; a transcription of prcsent-day standard English
whlCh usod the symbols [i:, e:, U:, 0:] would be perfectly ade-
quate. It is only the phonetician or acoustician who tells us that
there has been a displacement in the absolute physiologie and
acoustic configuration of thcse phonemes. Nevcrtheless we cao
sec the non-diphthongal variants, which at first the
predommant ones, are today obsolete. The speaker of present-
day standard English who tries to spcak a language like German
or French which has undiphthongizcd long vowcls, bas a hard
. learning ta produce these types. It is as hard for him to artie-
these acoustic types (which cxisted in English not so many
li nerations ago) as it is for the Frenchman or the German ta
the English diphthongal types. The speaker learns
Pth difficlllty to produce tipecch-sounds that do not occur lIt
:s native language, eyen though the irrolcvantly, may
assure him that an carlier stage of hls language potisessed these
very sounds.
We can speak of sound-change only when the displaeement of
habit has leri ta sorne alteration in thc structure of the
Most types of American English speak a low vowel [a] lU forms
like gai, rod, noi, where British English has kept an
type [J]. In sorne types of Ameriean standard Enghtih, thm [a]
is distinct from the [a] of forms likc calm, far, pa - so that
bother does not rime ",-ith fa/her, and bomb, is not homonymous
with balm: thcrc has been no displaccment of the phoncmie
system. In other types of Arnerican standard English, however,
the two phoncmcs havc eoincided: gOl, rod, boiher, bo-mb, calm,
far, pa, falher, balm ail have one and the .'lame low yowel [a], and
we say, accordingly, that a sound-change has taken place. Sorne
speakers of this (as wcll as .'lame of the other) typc pronounce
bomb as [bom]: this fOTm is due t.o sorne sort of linguistic borrowing
and accordingly cannat exhibit the normal correlation.
Thc init.ial ciusters [kn-, gn-j, ati in knee, gnai, lost. thcir stop
sound carly in the cighteenth eentury: hereby knot and noi, knight
and night, gnash and Nash became homonymous. English-tipeakers
of today learn only with diffieulty to produce initial c1ustcrs like
these, as, say, in German Knie [kni:] 'knce.'
In Dutch-German area, the Primitive Germanie phoneme [8]
changed toward [Z] and then toward [d]; by the end of the l\liddle
Ages this [dl coincided, in the llorthern part of the area, with
Primitive Germanic rd]. Henee modern standard Dutch has ini-
tial [dl uniformly, both in words like dag [dax] 'day,' doen [du:nJ
'do,' droom [dro:ml 'drcam,' whcre English has [d], and in words
like dik [dik] 'thiek,' doorn [do:rn] 'thorn,' drie [dri:] 'throe,' where
English has [el. The distinction has bccn entirely obliterated,
and eould he re-introduced only by borrowing from a language
in which it has been preservcd. Ncedless to say, the Dutchman
or North German has aS hard a Lime learning to utter an Eng-
Hah [6] as though this sound had nCver existed in his language.
TYPES OF PHONETIC CHANGE
CHAPTER 21
21.1. Phonctic change, as defined in the last chapter, is a change
. the habits of performing movements. Strictly
:eaking, a change of this kind has no importance so as it
does Dot affect the phonemic system of the language; m fact,
even with perfect records at our command, wc
he unable to determine the exact point where a favormg of certam
variants began to deserve the name of a historicn.l change. At
the time when speakers of English bcgan to favor the variants
with higher of the vowcls in words like gas 'goose'
and gs 'geese,' the was entircly without significance.
The speakers had no way of the acoustic qualities of
their vowels with the acoustic qualitics of the vowds which their
a few generations back, had spoken in the same
guistic forms. When they heard a dialect which had not made
the change, they may Imve noliced a difference, but they could
have had no assurance as to how this difference had arisen. l'ho-
netic change acquin's significance only if it results in a change of
the phonemic pattern. For instance, in the carly modern period,
the Middle English vowel [r-:], as in 8ed [sr-:d] 'seed,' was raised
until it eoincided with the [e:J in (les [ge:s] 'geese,' and this
cidencc for ail Ume changed the distribution of phonemes in the
forms of the language. Aguin, the Middle English short [el in a
so-called "open" syllable - that is, before a single conson:lllt
lowed by another vowcl, as in ete ['eteJ 'eat' - was lengthened
and ultimately coincided with the long vowels just mentioned.
Accordingly, the phoncmie structure of modern Enp;lish is differ-
ent From that of medicval English. Our phonemc [ij] continues,
among others, these three oIder phonemes; we may note, espe-
cially, that this coincidence has given rise to a number of
homonyms.
Old and Middle English le:] has changcd to modern [ij] in heel,
8teel, (Jeese, queen, green, meei (verb), need, keep.
Old and Middle English [:] has changed to modern [ij] in heal,
369
PHONETIC CHANGE 368
The favoring of variants which leads to is a
historical occurrence; once it is past, we have no guarantee of
its again. A later may end by favoring the very ,
same aeoustIc types as were ellfIllnated by an earlier change.
The Old and Middle English long vowels ri:, u:], as in [wi:n, hu:sJ
were climinated, in the carly modern period, by change toward
the diph thongal types of the wine, house. At about
the same Ume, however, the Old and l'diddle Enp;1ish long mid
vowels, as in [gc:s, go:s], were being mised, so that eighteenth-
century English again had the types [i:, u:] in words like geese,
goose. The new [i:, u:] arrived too late to suffer the change 10
[aj, aw] which had overtaken the Middle English high voweIs.
Similarly, we must suppose that the speakers of the
generations that were weakening the phoneme [s] between vowels,
could learn only with difficulty to utter such a thing as a distinct
simple [8] in intervocalc position, but, after the change was over,
the simplification of long [ss] re-introduced this phonotie type, and
(doubtless indcpendently of this) new combinations of the type ,
['este:sa] ( 20.10) were again fully pronounccable. In this way, '
we can often dotermine the succession (relative chronology) of '
changes. Thus, it is dear that in time, the Primi-
tive ludo-Europcan lb, d, gJ can have reaehed the types of Prim-
itive Germanie [p, t, k] only after Primitive [p, t, k]
had already beeo changed somewhat in the direction of the types
of Primitive Germanie [f, 6, h] - for the actual Germanie forms
show that these two series of phonemes did not coincide ( 20.2).
371 TYPES OF PHOKETIC CHANGE
t
's ta predict its occurrence. The greater simplicity of the
tba l,. f' ff 'b'l
red
variants 1S a permanent actor; It cnn 0 er no pOSSI 1-
favo
ities of correlation. .
Simplification of final consonant-cImters lS oven more common.
A Primitive 'f?ot: (nominative
pears in Sanskrit as [pa:t] and 10 Latm as [pe:sJ; a 1 rIml tl ve
*['bheronts] 'llCaring' (nominative singular mas-
euline) appears in Sf1nskrit as ['bharan], and in Latin as ferens
['ferens], later ['fere:s]. It is this type of change ,."hich ta
habits of pcrmitted final ( RA) and to alternatIons
of the type describcd in 13.9. Thus, a Pnnlltlve Central AI-
gonquian *[axkehkwa] 'kettle,' plural *[axkehkwaki], refiected in
Fox [ahko;hkwa, ahko:hko:ki], loses its final vowel and part of
the consonant-cluster in Cree [askihk, askihkwak] and in Men-
omini [ahkf:h, ahkehkuk], 100 thM, the plllral-form in these lan-
guages cont-ains a consonant-cluster that cannat be determined
by inspection of the sin!!;ular form. In English, final [lJg] and
{mb] have lost their stop; hence the contrast of long: longer [blJ -
'WIJgr], climb : clamber [klajm - 'klembr]
Sometimes even sin!!;lc final consonants are weakened or dis-
appear. In pre-Grk, final [t, d] were lost, as in Primitive Indo-
European "'[tod] 'that,' Sanskrit [tat]: Greck lto]; final lm] bccame
ln], as in Primitive Indo-European "'[jll'gom] 'yoke,' Sanskrit
[ju'gam]: Greek [zu'!-';on]. The same changes ,-,eern to have occllrred
in pre-Germanic. Sometimes ail final consonants are lost and
there results a phonetic pattern in which every word ends in a
yowei. This happ(,ned in pre-Slavic, \\'itness forms like Old Bul-
garian [to] 'that,' [i!!;o] 'yake.' It is a chan!-';e of this sort that
a.ccounts for morphologie situations like that of Samoan ( 13.9);
a Samoan form like [inu] 'drink' is the descendant of an aIder
'l'[inum], whose final l:onsonant has becn kept in Tagalog [i'num].
When changes of this sort appear at the beginnin?; or, mOre often,
at the end of words, \ve have tn suppose that the languages in which
they took place had, at the time, some phonetic marking of the
word-unit. If there were any fonns in which the beginning or the
end of a word had not the characteristic initial or final pronuncia-
tian, these forms would not suITer the change, and would survive
as Thus, in Middle English, final ln] was lost, as in
eten> ete 'eat,' but the article an beforc vowels must have been
Pronounced as if it wcre part of the following word - that is,
TYPES OF CHANGE
370
meal (' taking of food'), cheese, leave, clean, lean (adjective) st .
mead ('meadow'), meel (adjective). ' teetj
Old English [e] bas ehanp;ed to modern [ijJ in steal
meal ( flour ), weave, lean (vcrb), quean, speak, meat, mete, ,.
mead ('fermenter! drink').
. the the restriction of this last change to a
lImited phonetlC posltlOn, has produccd diffe:ent phonemes in
forms. that.used t the same phoncrne: the olcl [el was length.
ened m MIddle Enghsh weve < weave, but not in .:YIiddle English
weft < we!l. In the same way, a phonetic change which consisted
of shortenmg long vowels bcfom certain consonant-cInsters bas
produced the diiIerenco of vowel between meadow ( < Old English
['me:dweJ) and mead, or between kept ( < Old English ['ke:pte])
and keep.
A few hundred years ago, initial [k] was lost beiore [n]: the
rcsult was a change in the phonemic system, whieh included such.
{caturcs as the of knol and not, or of knight and m'ght,
and the alternabon of [n-] and [-kn-] in know, knowledge : 00;:
knowledge.
. 21. 2. The gencral direction of a great deal of sound-change
IS toward a simplification of the movements which make up the
utterance of any given linguistic form. Thus, consonant-groups
are often simplified. The ld English initial clusters [hr, hl, hn,
kn: gn, have lost their initial consonants, as in ld English
hrzng > nng, hliJapan > leap, hnecca > neck, cnow > knee, gna-'
yan> gnaw, wl"ingan > wring. The loss of the [hl in these groups
occurred in the later Middle Ages, that of the other consonants
in early modern time; wc do not know what ne\\' factor intervenoo
at these times to destroy the clusters which for many centuries
had bccn spoken without change. The [hl-elusters are still spoken.
in Icelandic; initial [kn] remains not only in the other Germanie
languages (as, Dutch knie [kni:], German Knie [kni:], Danish
[km::?], Swedish [kne;]), but also in thc English dialects of the
Shetland and Orkney Islands and northeastern Scotland. The
persists almost as widdy - in English, more widely; [wr-j,
lU the shape of [vr-] , remains in Scandinavian, the northern part
of the Dutch-German arca, including standard Dutch, and in:
several scattered dialects of English. As long as wc do Dot know
what factors led ta thcsc changes at one time and place but Dot
ut another, wc cannat cIaim ta know the causes of the change-
373 TYPES OF PHONETIC CHANGE
Bulgarian, which. did D?t make the changes in qucstion.
*['svatrba] 'marnage' glves Russian ['svadba]; compare Russmn
svat] 'arranger of a marriage.' Old Bulgarian [ot'Gbe:zati] 'to
appeurs in Russian as [odbe'zat]; compare the simple Old
Bulg
arian
[otu] 'from, away from' : Russian lot]. On the other
band, ld Bulgarian [podukopati] 'to undermine' appears in
Russian as [potko'pat]; contrast Old Bulgarian [podu igo] 'under
theyoke' : Russian ['pod igo].
The assimilation may affect the action of the velum, tongue, or
lips. If sorne difference between the consonants is kept, the assim-
ilation is partial; thus in pre-Latin [pn] was assimilated ta [mn], as
in Primitive Indo-European *['swepnos] 'sleep,' Sanskrit ['svap-
Bah] : Latin sumnus. If the difference entire!y disappears, the
assimilation is total, and the result is a long consonant, as in
Italian sonna ['s<mno]. Similarly, Latin oeta 'eight' > Italian oUo
['()tto]; Latin ruptum 'broken' > Italian roflo ['rotto].
In progressive assimilation the latter consonant is aitered. Thus,
pre-Latin *[kolnis] 'hill' gives Latin collis; compare Lithuanian
['ka:lnas] 'mountain.' Our ward kill underwcnt the same change
[lnJ > [lI] in witness Primitive Indo-European
*{pl: 'nos] 'full,' Sanskrit [pu:r' Nuh], Lithuanian [' pilnas] : Prim
itive Germanic *['foIlaz], Gothicfulls, Old Englishfull, or Primitive
Indo-European *['w\:na:] 'wool,' Sanskrit ['u:rNa:J, Lithuanian
('vilna} : Primitive Germanic *['wollo:], Gothie wu/la, Old English
wull.
21. 4. A great many other changes of consonants can be viewed
as assimilative in character. Thus, the unvoicing of final con-
SOnants, which has occurred iR the history of various languages,
can he viewed as a sort of regressive assimilation: the open position
of the vocal chords which fol!ows upon the end of speech, is antic-
ipated during the utterance of the final consonant. Thus, many
diaIects of the Dutch-German area, including the standard lan-
guages, have unvoiced al! final stops and spirants; the result is an
alternation of unvoiced finals with voiced medials ( 13.9):
Old Righ German tag 'day' > New I-ligh German Tag [ta:kJ,
but, plural, taga 'days' > Tage ['ta:ge], with unchanged [g];
buOid High German bad 'bath' > New High German Bad [ba:t],
t, genitive case, bades > Bades ['ba:desJ;
Righ German gab '(he) gave' >New High German gab
.p], but, plural, gabun' (they) gave' > gaben ['ga:benl.
TYPES OF PI-IONETIC CHANGE 372
.the phonetic of final position - 130 that the
ln] thls case was not lost (hke a final ln]), but preserved (Iike a-
medlal [n]): a kouse but an arm. Latin vos' ye' gives French VOUa
[vu], but Latin phrase-types like vos amatis 'ye love' are reBected
in the French of saying vous aimez [vuz eme]. Latin.
est 'he is' gave French est [e] 'is,' but the phrase-type of Latin
est t'lle? 'is that one?' appears in the French sandhi-form in esf.il:l
[et i?] 'is he?' In the same way, a Primitive
*['bheronts] is refiected not only in Sanskrit ['bharan], above cited'
but also in the Sanskrit habit of adding a sandhi [13] when the nexi:
word began with ft], as in ['bharans 'tatra] 'carrying there.' '
21. 3. Simplification of consonant-ciusters is a frequent
of sound-change. Thus, a pre-Latin *['fulgrnen] 'flash (of Iight-'
ning) , gives a Latin fulmen. Here the group [lgm] was simplifiedi
by the change ta [lm], but the group [Ig], as in fulgur 'flash,' was,
not changed, and neither was the group [gm], as in agmen 'army,'.
In describing such changes, we speak of the conditions as con,.;
ditioning factors (or causing factors) and say, for instance, that one
of these was absent in cases like fulgur and agmen, where the,
tg], accordingly, was preserved. This form of speech is inaccurate,'
since the change was really one of [lgmJ to [lm], and cases Iike
fulgur, agmen are irrcievant, but it is often convenient to use thes'
terms. The result of a conditioned change is often a morphologie,
alternation. Thus, in Latin, wc have the suffix -men in agere 'ta:
Icad': agmen 'army' but fulgere 'to flash': f ulmen 'flash (of light-:
ning).' Similarly, pre-Latin [rkn] became [m]; beside pater' father',
paternus 'paternal,' wc have querCU8 'oak' : quernus 'oaken.'
Quite commonly, ciusters change by way of assimilation: the,
position of the vocal organs for the production of one phoneme is,
altered to a position more like that of the other phoneme. Th
commoner case is regressive assimilation, change of the prior
phoneme.
Thus, the voicing or unvoicing of a consonant is often altered'
into agreement with that of a following consonant; the [s] of
goose and hause has been voiccd ta [z] in the cornbinations goslin{h
husband. This, aguin, muy give rise to morphologic aIternationB.
In the history of Russian the loss of two short vowels (1 shall
transcribe thern as [IJ and [uD produced consonant-clusters; in
these clusters a stop or spirant was then assimilated, as to voiciDg,
ta a following stop or spirant. The old forms can be seen in Old
375
TYPES OF PHONETIC CHANGE
bsolute form [ba:] 'cows.' This type of sandhi is preserved in
nurnber of cases, as, in our instance, after the pronoun
'biS.' In the same way, [s] between vowels was weaken:d to
and then lost: a Primitive 1ndo-Europcan *['sweso:r] 'slster,'
krit ['svasa:], giving first, presumably, *['13weho:rJ, and then
siur. Final [s] similarly was lost: a Gallic tarbos 'bull'
pears in Old Irish as tarb, We have to suppose, now, that the
ge [8 > h] between vowels took place a180 n close-knit phrases,
a:hat an *[esa:8 o:wjo] 'her egg' (compare Sanskrit [a'sja:h]
'h r' with [-hl from [-s]) resulted in a modern [a huv] 'her egg,'
e, l" . h bOt
in contrast with the independent [uv - agam, a a 1 pre-
served only in certain combinations, aS after the word for 'her.'
Siroilarly, [ml was first changed to [n] and then lost at the end of
words but between vowels was preserved; both treatments appear
in 'holy place,' Old Gallic [nf'me:ton], Old Irish
nemed, At the stage where [-ml had become [-n], an old *[sen-to:m
o:wjo:mJ 'of these eggs' (compare the Greek genitive plural
l'to:nJ) gave what is now [na nuv], in contrast with the absolu
[uv] 'egg,' 1'0 a similar, but more complicated development we owe
the sandhi-alternant wi th ni tial [tJ, as in [an tuv] 'the egg';
ultimately this is duc to the fact that the Primitive IndoEuropean
nominativc.aecusative sngular neuter pronoun-forrns ended in
[dl, as Sanskrit [tat] 'that,' Latin id 'it.'
We may interprct the pre-Germanie chanp;e discovered by
Verner ( 18.7; 20.8) as a weakening of unvoiced spirants [f, 6,
h, s] between musical sounds to voiced [v, ti, ")', z]; then the re-
striction of the change to cases where the preeeding vowel or
diphthong was unstressed is subject to a further interpretation
of the same sort: after a loudly stressed vowcl there is a great
amount of breath stored up behind the vocal chords, so that their
opening for an unvoiced spirant is easier than their closure for a
voiced. We cannot view these interi)retahons as correlating
(" causal ") explanations, however, for enough languages keep
unvoiced spirants intact botween vowels, while others change
thern to voiced regardlcss of high stress on a preccding voweL
lIere, too, the condiUoning factor \Vas afterwards removed by other
changes: in an early pre-Germanie *['werllonon] 'ta becorne' ver-
SUs *[wur'('iu'me] 'we becamc,' the altcrnation [6:tiJ depended on
the place of the stress; later, when the stress had changed to the
first syllable of al! words, the alt<;rnation in Primitive Germanie
TYPES OF PHONETIC CHANGE
The voiced consonant may be proserved in sandhi - that i13, in'
traditional phrase-types whero it did not come at the end of
speech. This does not happen in standard German; here the ,
form has bei.'ll carried out for every word-unit. In Hussian,
ever, wc have not only the final-form, by which an old [podu],
ufter loss of the vowe1, becarno [pot], but also phrasaI types like
['pod igo] 'undor the yoke.' There is a type of Dutch pronunciation ,
whore an old hebbe '(I) have' appears, after loss of the final vowel,
not only in the final-form with [-p], as in ik heb [ok 'hepJ, but alsa
in the phrasaI sandhi-type, hcb ck? ['hob ek'!] 'have 1?' This is the '
origin of remniscent sandhi ( 12.5).
A very common t.ype of change i13 the weakening of consonants
bctween vowols or other open sounds. This, too, is akin to assim-
ilation, since, when tho preceding and following sounds are open
and voiced, tho less rnnrked dosme or the voicing of a stop or
spirant represents an economy of movement. The change which
gave rise to the American English voiced tongue-flip variety of
ft], as in water, butter, at aU 6.7), was surcly of this sort. Latin
[p, t, k] bet,veen vowels are largel}" weakened in the Romance
languages: Latin TTpam 'bank, shofC,' setam 'silk,' focum 'hearth'
appear in Spanish as liba, seda, fucgo 'fire,' ",here the lb, d, gJ are
largely spirant in charader, and in French as rive, soie, feu [ri:v,
swa, f.0]. Some languages, such as pre-Grook, lose sounds like
[s, j, w] between vowels. The Polynesian languages and, to sorne
extent, the medieval Indo-Ar.yan languages, show a loss of the oIrl
structure of medial consonants, much liko that in the French forms
just cited. In the history of English, loss of [v] is notable, as in
Old English ['hEvde, 'havok, 'hla:vord, 'hla:vdije, 'he:avod,
'navoga:r] > modern had, hawk, lurd, lady, head, auger; this change
seems tD have occurred in the thirtoenth Cntury.
If the conditioning factors are removed by subsequent change,
the result is an irregular alt.ernation. In this way, arose, for ex
ample, the sandhi-alternation of initial consonants in Irish ( 12.4),
In the history of this language, stops between vowels woro ",eak
cned ta spirants, as in Primitive Indo-European *['pibo:mi] '1
drink,' Sanskrit ['piba:rni): Old Irish ebaim ['cvim], Apparently
the language at this stage gavc little phonetic recognition to the
word-unit, and ca,rried out. this change in close-knit phrases,
changing, for instance, an *[eso bowes] 'his cows' (compare
Sanskrit. [a'sja 'ga:vahJ) to what i" now [a va:], in contrast with
374
377
MODERN E1<GLISR
drench
stitch
singe
bridge
> Qu, EN<;LISH >
drencean ['drenkanl
stice ['stike]
sengan ['sengan]
brycg [bryggl
TYPES OF PHONETIC CHANGE
PRE-E"'GLISII
*['drenkjanJ
*['stiki]
*['sengjan]
*['bryggju]

>
OLD ENGLl8H
>
MODERN E"'GLISll
*['ko:ni > cene ['ke:ne] keen
*['kunni> 'kynni] cynn [kyn] kin
*['go:si > ges [ge:s] geese
*['guldjan > 'gyldjan] gyldan ['gyldan] gild
Dnpalatalized velars, not followed by front vowel:
PUE;-ENGLl8H > OLD E"'GLISll > MODEn;>;
*['drinkan] drinean ['drinkan] drink
*['stikka] sticca ['stikka] stick
*['singan] singan ['singan] sing
*['frogga] frogga ['frogga] frog
PRE-ENGLlSll > OLD E"'GLR > MODER'"
*['h::si] ciese ['ki:ese] cheese
*[kinn] cinn [kin] chin
*['geldanJ gieldan ['jeldanJ yield
*[gern] gearn [jarn] yarn
Unpalatalized velars, before ncw front vowels:
A third factor of the same kind was the loss, by later sound-
change, of the conditioning featurc, - that is, of the front vowel
[e, i, il which had caused the palatalization:
Palatalized velars, followed, at the critical Ume, by a front
vowel:
An ther way in which the pre-English palataIization in time
aff the structure of the language, was by the obscuration of
tb:Cconditioning factor. The back vowels [0, u], ,:hich not
aff t a preceding velar, were changed, under certam conditIOns,
vowels y] and later to le, il, which coincided with old
rnt vowels that had effect.ed palatalization. Hence, in the
::'er stages of English, both palatalized and unpalatalized velars
oceur
red
before front vowels.
pa.latalized velars, before old front vowels:
TYPES OF PHONETIC CHANGE
NonTR
PnE-E"'GLISR
> OLD ENGL18R
>
MODER'"
GERMA'" Em)LlSR
gold *[gold] gold [gold] gold
god *[go:d] yod [go:d] good
geldan *['geldan] gieldan ['jeldanJ yield
Garn *[gtrn] gearn [jarn] yarn
jok *[jokJ geoe [jok] yoke
jar *[je:r] gear [je:ar] year
376
*['werllanan - 'wurdume], Old English ['weorllan - 'wurdon], WIlB'
an arbitrary irregu1arity, just as is the parallel was : were, frc
Primitive Germanie *['wase - 'we:zume], in modern English'.
A similar change occurred much later in the history of
lish; it accounts br such differences as luxury : luxurious ['lok8-'
fij -log'zuwrjosJ in a common type of pronunciation, and for the
two treatments of French [s] in forms like possessr:rr [po'zesrL
This ehange involved the voicing of oid [s] after an unstressed.
vowel in suffixes, as in glasses, misses, Bess's; a few forms like
dice (plural of die) and pence show the preservation of [s] after
a stressed vowel. Immediately after this change the stressed
forms must have been off [of], with [will], is lis], his [his], and the
atonie forms of [ov] and [witi, iz, hiz,] but this alternation has>
bcen dcstroyed: off and of have bccn redistributed by analogie
change, [will] survives as a variant of [witi], and the [s]-forms of
is and his have fallen into disuse.
21. 6. Consonants arc often assimilated to the tongue-position
of preceding or following vowels. The commoncst case is the as-
similation especially of dentals and velars to a fol1owing front
vowel; this is known as palatalization. A change of this kind which
did not cause phonemic alterations, must have occurred not too'
long ago in English, for phoneticians assure us that we make the'
tougue-contact of [k, g] farther for ward before a front vowel, as,
in kin, keep, kept, give, geese, gel, than before a back vowel, as in ,
cook, good. In pre-English there occurrcd a change of the same
sort which led to alteration of the phonemic structure. To begin
with, the palatalized form of [g] - presumably this phoneme had
a spirant character - coincided with another phoneme, [j]. The
change in phonemic distribution appears plainly when we compare
the cognate forms from North German (Old Saxon), where the
old phonemic distribution remained intact:
379 TYPES OF PHONETIC CHANGE
. that the diverse vowels of the European languages were ?ue
., ter change, made during a common perlOd.
la the [a] of the Indo-lranian languages, Pnmlve Indo-
ore an, velars [k, g] appeared somdnes unchanged and some-
pe [e V
J
] In the 1870'8 several students indcpendently saw
es as , . 1 l' t' d
t these latter reflexes are probably due to pa ata lza an ,
f t correlate fairly weil with the cases where the European
;,' have [e]. Thus we with vowels i,n the
:> s of Europe and velar stops 10 Indo-Iraman, correspondences
rgusge
Primitive Indo-European *[kWod], Latin quod [kwod] 'what':
Ss.nskrit kat- (as first mcmber compounds); . "
.}'.;, Primitive Indo-European *[g 0:W8], Old Enghsh cu [ku:] cow:
;:;/. ,Sanskrit [ga:wh]. .
, ',On the other hand, with the front vowel [e] III the languages of
Europe and affricates instead of velar stops in Indo-lranian, we
&d correspondences like
... , Primitive Ind<rEuropcan *[kWe], Latin que [kwe] 'and' : San-
skrit [ca];
. Primitive Indo-European *[gWc:nis], Gothie gens [kwe:ns] 'wife':
. Sanskrit [-ja:nih] (final membcr in compounds) .
. From cases likc these we conclude that the uniforrn [a] of Indo-
lranian is due to a later devcloprnent-: in pre-Indo-lranian there
must have been an [el distinct from the other vowels, and this
{el must have caused palatalization of preceding velar stops. Since
, ,this le], moreover, agrees with the [el of the European bngun.ges,
the distinction must have existed in Primitive Indo-European, and
esnnat be due t() a joint innovation by the languages of Europe.
This discovery put an end to the notion of a cornrnon parent speech
". intermediate between Primitive lndo-European and the European
.....""" (as 0Pposed to the Indo-lranian) languages.
:c 21. 6. The weakening or loss of consonants is sometirnes ac-
companied by eompensatory lenglhening of a preceding vowel. The
Oid English combination [ht] preserved ta this day in northern
dialects, has lost the [hJ and' longthened the preeeding vowel in
lI10st of the aroa. Thus, Old English niht [niht, nixt] 'night,' modern
Scotch [nixt, next], became [ni:t] , whence modern nighl [najt].
Los
s
of a sibilant before voiced non-syllabics with compensatory
lengthening of a vowcl is quite common, as in pre-Latin *['dis-lego:]
Il pick out, 1 like' > Latin dHigo (compare dis- in dispendo '1
SPAN1SH FREI\TR LATlN > ITALlAN
TYPES OF PHONETIC CHANGE 378
The sound-change which we caU palntalization changes 00
sonants at first to varieties which the phonetician calls palatali
the modern English forms in our prcceding examples, with t
l, j, j], show us that these palatalized types may undergo furt
changes, These, in fact, arc extremely cornmon, although t
direction varies. In the case of both velars and dentals, affrica.
types [c, j] and si bilant types, both abnormal [s, i] and normal [s, z
are fairly frequent. In modern English we have a developme
of [tj > , dj > 1, sj > S, zj > i], as in virtue, Indian, ses'
vision [' vreuw, 'in}l}, 'ses!}, 'viZ!}]; more formaI variants, such
['vrtjuw, 'indj!}], have arisen by later changes. The Roman
languages exhibit a great variety of development of palatali
velars:
'hundred' cenlum cenlo cent cienlo
['kentum] ['cento] [san] ['ajonto]
'natian' genlem gente gens genle
['gentem] ['jenteJ [zan] ['xente]
Part of the French area has a palatalization of [k] before [a]
in the Middle Ages, when English borrowed many French wor
this had roachod the stage of [c], 80 that a Latin type like ca
[kan'ta:re] 'ta sing' > Old French chanter [can'te:r] appears .
English as chanl; similarly, Latin cathedram ['katedram] app
as chair; Latin calenam [ka'te:nam] as chain; Latin cameram ['ka
eram] as chamber. In modern standard French, further chan
of this [l has led to [!J]: chanIer, ehaire, chane, chambre [sanUj
se:r, se:n, sanbr].
Palatalization has played a great part in the history of t
Slavic languages: it has occurred at diffcrent times with differen
results, and has affocted cvery type of consonant, including ev
labia18.
A case of palatalization whose causing factor was obscured b
later change, played an important part in the development of
Indo-European studies. In the Indo-lranian languages a single
vowel-typc [a] corresponds to the three t.ypes [a, e, 0] of the oth ,
Indo-European languages. Thus, Latin ager 'field,' equos 'horse,
oct6 'eight' are cognate with Sanskrit ['ajrah, 'avah, as'Ta:
W
],
For a long time students belioved that the Indo-lranian langua
had here preserved the Primitive Indo-European statc of affaira.
381 TYPES OF PHONETIC CHANGE
pre-English type *['Oanho:o] 'clay' appears tirst as thoh
fIlO:hcl, then a ~ [Ilo:]; Gothie ['ahwa] ';ive:' (cognate with Latin
ua 'water ') lS parallclod by 0Id hnghsh ea [e:a], from pre-
i'ng
lish
*['ahwu]; Gothie ['sehwan] 'to see' is matched by Old
English scon [se:on].
21. 1. Vowcls are often assimilated to vowcls that precede or
follaw in the next syllable. During the early Middle Ages, changes
of this kind occurrcd in several Germanie dialects. These changes
in the Germanie languages are known by the name of umlaut; sorne-
what confusin?;ly, this ierm is applied also ta the resultant gram-
matical alternations. The commonest type of umlaut is the partial
assimilation of a stressed back vowel ta a following li, j]. The
resulting altcrnations, after the loss of the conditioning li, jl, be-
came purely grammatical:
PRE-ENGLI8H > OLD E:<rGLIRH > MODERN E:<rGLI8H
*[gold] gold gold
*['guldjanJ l gy/dan gild
*[mu:s] mus [mu:s] mouse
*['mu:si] mys [my:s] mite
*[fo:t] fot [fo:t] foot
*['fo:ti] fct [fe:t] feet
*[gans] gos [go:s] goose
*['gansi] ges [ge:s] gcese
*[drank] dranc [drank] drank
*['drankjan] drencean ['dronkan] drench
Old Narse had also other types of umlaut, sueh as assimilation
of [a] toward the back-vowel quaHty of a following lu], as in
*['saku] 'accusation' (cornpare 0 Id English saeu ' dispute') > Old
Norse [sJk]. Similar chan?;es, supplemented, no doubt by regular-
izing new-fonnations, must have led to the vowel-harmony that
prevails in Turco-Tartar and some othor languages ( 11.7).
The effeet of simplification appears most plainly in shortening
and loss of vowe\s. In the final syllables of words, and especially
inflnal position, this occurs in ail manner of languages. Among the
Central A\gonquian languages, Fox alone has kept the final vowels:
Primitive Central Algonquan *[eloowa] 'man' > Fox [neniwa],
Oji bwa [inini], ),1cnomini [cm: :niw], Plains Cree [ij ini"l]. Certain
1 T h ~ lu] in thi. fonn i . ~ due tu an earlicr assimilation of [oJ to the high-vowe!
!JOlIitioll of the foliowiJlg Iil.
TYPES OF PHONETIC CHANGE 380
weigh out,' and lego '1 pick, gather'); early Latin cosmis 'kind'
> Latin comis; pre-Latin *['kaznos] 'gray-haired' > Latin CnU8
(compare, in Paelignian, a neighboring Italie dialeet, caSMr 'old
man'); Primitive Indo-European *[nisdos] 'nest' (compare English
nest) > Latin nidus.
If the lost consonant is a nasal, the preceding vowtJ1 is often -
nasalized, with or without eompensatory lengthening and other .
changes. This is the origin of the nasalized vowels of muny l a n ~ ,
guages, as of French: Latin canlare > Freneh chanter [suntel,'
Latin cenlum > French cent [san], and so on. The morphology of
Old Germanie shows parallel forms with and without nasal, snch '
as Gothie ['bringan - 'bra:hta] 'bring, brought,' ['6ankjan-
'6a:htaJ 'think, thought.' Tho forms without [n] ail have an [hl
immediately following a long vo,,,ol. Tho suspicion that in theoo
forms an [n] has been lost with componsatory lengthoning, is
confirmed by a few eomparisons with other Indo-Europoan lan-
guages, such as Latin vincere 't conquer' : Gothie ['wi:han] 'to
fight.' Furthcr, wc have a twelfth-century lcelandic grammarian's
statement that in his language forms like [6e:!] 'file' (from *['6in-
h10:]) had a nasalized vowol. In Old English, the [a:] of the o1.her
Germanie ianguugcs, in forms liko t.heso, is rcprescn1.cd by [0:],
as in ['bro:hteJ 'brought,' ['Oo:hteJ 'thought-.' \Ve have reuson 1.-0
bclieve that this divergent vowel quality is a reflex of older na-
sulization, because in other cases also, Old English shows us an
[0:] as a reflex of an earlicr nasalized [a]. The loss of ln] before [hJ
occurred in pre-Germanie; before t.he other unvoiced spirants
[f, s, 6l an [n] rcmained in most Germanie dialects, but "las lost,
with cornpensatory kngthoning, in English, Frisian, and some of
the adjacent dialects. In 1.hese cases, too, wo tind an [0:] in ld
English as the reflex of a lengthenod and mlsalizcd [a]. Thus, the
words flve, us, mouth, soft, Goose, other appoar in tho oldest German
documents as [finf, uns, mund, sunfto, gans, 'ander] (with rd] as
reflex of an old [Il]), but in Old English as [fi:f, u:s, mu:6, 'so:fte,
go:s, 'o:ter].
When a consonant has been lost between vowels, the resulting
succession of vowels often suffers contradion into a single vowel or
diphthongal combination. Our earliest English records still show
us an [hl between vowels, but very soon afterward this h disappears
from the texts, and single vowels are written. Thus, the word to6
appcars first as tah, presuillebly ['ta:he], but 800n as la [ta:];
383 TYPES OF PHONETIC CHANGE
[saP); since the time of these shortenings, French has lost
the strong word-stress and ceased shortemng lts forms.
If a language goes through this kind of change at a time when
roorphologically related forms stress different syllables, the re-
may be an extremely irreguhtr morphology. Wc can sec the
beginnings of this in our foreign-Iearned voeabulary, which stresses
different syllables in different derivatives: angel ['cjnj!J, but
angelic (en'jelik]. In Primitive Germanie the prefixes were un
sf,reSsed in verb-forms but stressed in most other words; the
weakenings that ensucd broke up sorne morphologie sets, sueh as
pre-English *[bi-'ha:tnn] 'to threaten' > ard English behatan
[be'ha:tan], but
pre-English *['bi-ha:t] 'a threat' > Old English beoi [be:ot].
A similar process rendercd the morpholof.,'Y and, as to sandhi,
the syntax of Old Irish extrcmely irregular:
pre-Irish *[' bereti] 'he bears' > Old Irish berid [' berW'J;
preIrish *[eks 'beret] 'he bears out, brings forth' > Old Irish
48beir [as' ber] 'he says';
preIrish *[ne esti 'eks beretJ 'not it-is that-hc-forth-brings'
(that is, 'he does Dot bring forth') > Old Irish nI epir [ni: 'epir]
'he does not say.'
21.8. Sorne changes which superficially do not seem like weaken-
ings or abbreviations of movement, may yet involve a simplifi-
cation. In a good many languages we flnd an intermediate con
sonant arising in a cluster. A Primitive Indo-European [sr] appears
:s [strl in Germanie and in Slavic; thus, Primitive Indo-European
fsrow.j (compare Sanskrit ['sravati] 'it flows') is refleeted in
Primitive Germanie *['strawmaz] 'stream,' Old Norse [strawmr],
Old English [stre:am], and in Old Bulgarian [struja] 'stream.'
English, at more than one time, has inserted a [d] in the groups
ni] and a [b] in the groups [mr, ml]: Old English ['6unrian]
ihunder; Old English ['aIre] (accusative case) > alder;
['timrjan] 'ta eonstruct' as weIl as ['tmbrjan], but
h Enghsh has only ['timbrian] and [je'timbre] 'carpcntry-work,'
;:nce timber; Old English ['6ymle] > thimble. These
Bhnnges lllvoive no additional movernent, but rnerely replace
for movements by successive. Ta pass from [n] to [r],
6rJ.d UlStance,. the speaker must simultaneously raise his velum
.... .H_ lllove hls tongue from the closure position ta the trill posi-
: 1Il\,l1l:
TYPES OF PHONETIC CHANGE 382
types of two-syllable words are exempt from this shortening'
*[ehkwaJ 'lause' > Fox [ehkwa], Ojibwa [ihkwa],
[ehkuah], Cree [ihkwaJ.
Languages with strong word-stress often weaken or lose their ..
unstressed vowels. The loss of final vowels, as in Old English
(ic) singe> (1) sing, is known as apocope; that of medird vowels
as in Old English stiinas > stones [stownz], as syncope. The
trast between the long fonus of Primitive Germanie, the shorter '
forms of OId English, and the greatly rcdueed words of modern ..
Englsh, is due ta a succession of sueh chf1Dges. Thus, a Primitive
Indo-European *['bheronom] 'act of bearing,' Sanskrit ['bhara-
Nam], Primitive Genuanic *['berananJ, gves Old English beran,
Middle English bere, and then modern (lu) bear. The habit of
treating certain words in the phrase as if they were part of the
preceding or following word, was inherited from Primitive Inda-
European; when, in pre-Gerrnanie time, a single hilSh stress was
placed on each wont, these atonie forms reeeived none; bter, the.
weakening of unstressed vowels led to sandhi-variants, strcsood
and unstrcssed, of sueh words. Weakenings of this kind have
occurred over and OVf'r again in the history of English, but the
resultant alternations have bcen largely removed by re-fonuations
which consisted either of using the full fonus in unstressed posi
tions, or of using the weakened fonus in stressed positions. Our
on, for instance, was in the medieval periad the unweakened form; .
the weakened fonu of this ward was a, as in away, from Old
BngUsh on wcg [on 'wejJ; this weakened form survives only in llo
limited number of combinations, sueh as away, ashore, agrouruJ,
aloft, and the unweakened on is now used in atonie position, as in
on ihe table, but has here been subjected to n new weakening, which
has resulted in unstressed [on] beside stressed [an], as in go on .
[gow 'an]. In contrast with this, our pronoun l, which we use in .
bath stressed and unstressed positions, refieets an old unstresse
d
.
fonu, in which the final consonant of Old English ic has been lost;'
the old stressed form survives in the [ie] 'l'of a few local dialects. .
These changes have left their mark in the unstressed sandhi
variants of many words, such as is, but [z] in he's here; will, but
[1] in TU go; not, but [I).t] in isn't; and in the weakencd forIllS of
sorne unstressed compound members: man, but [-ml).] in gent:ema.
n
;
8wain but [-SI).J in boaiswain. The same factor accounts for the
shortness of French words compared ta Latin; as in centum > cent
384
eoed in forms like weve ['weve] > weave, slele ['stele] > steal,
nos
e
['nose] > nose, but stayed short in forms like weft, steUh >
steaUh, nos(e)thirl > nostril. In sorne languages, sueh as Menornini,
wc find a very eomplicated regulation of long and short vowcls
according ta the preceding and following consonants and accord-
ing to the number of syllables intervcning after the last prcceding
long vowcl.
The complete loss of quantitative differenccs, whieh oecurred,
for instance, in medieval Greek and in sorne of the modern Slavic
languages, makes articulation more uniform. The same can be
said of the abandonment of distinctions of sy11able-pitch, \"hich
bas oceurred in these sarne languages; similarly, the removal of
ward-accent uniformly to sorne one position such as the first
syllable, in pre-Germanie and in Bohemian, or the next-to-last,
in Polish, probllbly involvcs a faeilitl1tion.
ln the same sense, the loss of a phonemic unit may he viewed
as a simplifieation. Except for English and lcelandic, the Germanie
languages have lost the phoneme [0] and its voiced development
[tJ; the reflcxes coincide in Frisian and in Scandinavian largely
with ft], as in Swedish lom [to:rn] : thorn, with the same initial
as tio ['ti:eJ : ten, and in the northern part of the Duteh-German
area with rd], as in Dnteh daorn [do:rnJ : thorn, with the same ini-
tial as doen [du:nJ : da. Old English [hJ before a consonant, as in
niht 'night,' or in final position, as in seah '(1) saw,' was acousti-
cally doubtless an unvoieed velar or palatal spirant; in most of
the English area this sound has been lost or has coincided with
other phonemes.
21. 9. Although many sound-changes shorten linguistic forms,
simpWy the phonetic system, or in sorne other way lessen the Ja-
bor of utterance, yet no student has succeeded in establish-
ing a correlation bet.wecn and any anteeedent
rhenomenon: the causes of sound-change are unknown. When
flnd a large-scale shortening and loss of vowels, we feel safe
1.D. assUming that the language had a strong word-stress, but many
languages with strong word-stress do not weaken t.he unstressed
examples arc ItaIian, Spanish, Bohemian, Polish. The
nghsh change of [kn-, gn-] ta [n-] seems natural, after it hBS
but. why did it not oeeur before the eighteenth cen-
and why has it not. occurred in the other Germanie
guages?
TYPES OF PHONETIC CHANGE
ln] [r}
velum lowered !B velum raiscd
dental dosure !B ) trill position
If, with a less delieate co-ordination, the velum is raised before
the change of tongue-position, there results a moment of Ull.w
nasalized dosure, equivalent to the phoneme [d]:
[n] [dl [rI
velum lowered *i ) velum raised
dental closure B ) trill position
The second of these performances is evidently easier than the,
firs(;.
In other cases, too, an apparent lengthening of a form may he
viewed as lcssening the difficulty of utterance. When a relatively
sonorous phoneme is non-syllabic, it oft-en acquires sy11abic fune-
tioo; this change is known by the Sanskrit name of samprasarana.
Thus, in sub-standard English, elm [elm] has ehanged to ['ellfl].
This is often fo11owed by another change, known as anaptyxi8!
(;he rise of a vowel beside the sonant, which becornes non-syllabie.
Primitive Iodo-European *[agros] 'field' gives pre-Latin *[agr]i
in this the [r] must have become sy11abie, and then an anaptyctic:
vowel must have arisen, for in the historical Latin form ager
['ager} the e represcnts a fully formnd vowel. SimilarIy, Primi
tive Germanie forms like *['akraz] 'field,' *['foglaz] 'bird,' *['tajk
nan] 'sign,' *['majllmaz] 'precious object' 10st their unstressed
vowels in a11 the old Germanie dialects. The Gothie forms [alas,
fugls, tajkn, majllms] may have been rnonosyllabie or may have
had syllabie sonants; anaptyxis has taken place in the Old E
lish forms ['eker, 'fugol, 'ta:ken, 'ma:tiom], though even here
spellings like fugl are not uncommon.
Another change whieh may he regarded as a simplificatio
occurs in the history of sorne stress-using languages: the quant":
ties of stressed vowels are regulated aceording to the charaetet
of the following phonernes. Generally, long vowels remain long
and short vowels are lengthened in "open" sy11ables, that is, b&
fore a single consonant that is fo11owed by another vowel; in oth '
positions, long vowels are shortened and short ones kept sho
Thus, Middle English long vowels remained long in forms . ,
clene ['kIE:ne] > clean, kepe ['ke:pe] > keep, mone ['mo:neJ '7
moon, but were shortened in forms like cleWJe > cleanse, ke
> kept, mon(en)dai > M onday: and short vowels were lengtb
TYPES OF PHONETIC CHANGE
385
387 TYPES OF PHONETIC CHANGE
MIDDLE ENOLlSH > EARLY MODERN > PR:F:SENT-DAY
['na:me] [ne:m] [nejrn] name
[de:dJ [di:d] [dijd] deed
[ge:s] [gi:s] [gijs] geese
[wi:n] [wejn] [wajn] wine
[st:J:n] [sto:n] [stown] stone
[go:s] [gu:s] [guws] goose
[hu:s] [hows} [hawsJ house
Another theory seeks the cause of sorne sound-changes in fonnal
oonditions of a language, supposing that forrns of weak meaning
are slurred in pronunciation and thereby permanently weakened
or lost. Wc have met this doctrine as one of those which deny
the Occurrence of purcly phonemic changes ( 20.10). Wc have
no gauge by whieh we could mark sorne fonnal features of a lan-
guage as semantically weak or superiluous. If we condemn ail
fea.tures of meaning except business-like denotations of the kind
that could figure in scientific discourse, we should have to expect,
On this thcory, the disappearance of a great many fonns in almost
El'l"ery language. For instance, the inflectional endings of adjec-
, btles
s
incapable of pronouncing the original sounds, just as
.'. U English-speaker of today is incapable of pronouncing the
:;:neh unaspirated [p, t, k]. At a later however,
lb, d, g] were 10 to un-
;'\toiced stops [p, t, k]. Thesc sounds dld not COlllC1de with those
of the first group: the of the first had no the
. [p t k] character, havmg changed to asplrates or affncates or
already to spirants; the sounds of the second group, on
the other hand, were uot subjected to the same change as those
of the first group, because, as wc say, the sound-change of [p, t, k]
te [f, 6, hl was pasto More accurately, we should say that the sound-
change of [p, t, k] was already under way: the new [p, t, k] consti-
tuied a different habit, which did not take part in the displace-
ment of the old habit. In bme, the new [p, t, k] became aspirated,
as they are in prcsent-day English; sa that, once more, we are in-
capable of pronouncing unaspiratcd unvoiced stops.
The English sound-chan!,'"s that are known under the narne of
"the great vowel-shift," are of a type that has Httle effect beyond
altering the acoustic shape of each phoneme; the long vowels
were progressively shifted upward and into diphthongal types:
TYPES OF PHONETIC CHANGE 386
Every conccivablc cause has been allcgcd: "race," elima
topographie conditions, diet, occupation and gencral mode
life, and sa on. \Vundt attributcd soundchange to increase in .
rapidity of speech, and this, in tum, to the community's adva
in culture and gcneral intelligence. It is safe to say that we sJ)e'
as rapidly and with as lWe effort as possible, approaching alw
the limit where our interlocutors ask us ta repeat our utteran
and that a grcat deal of sound-change is in some way connee
with this factor. No permanent factor, however, can account f
specifie changes which occur at one time and place and Dot'
anothcr. The same consideration holds good against the th
that sound-change arises from imperfections in children's le
ing of language. On the other hand, temporary operation of facto
like the above, such as change of habitat, occupation, or diet,
ruled out by the fact that sound-changes occur too oiten
exhibit tao great a variety.
The substratum theory attributes sound-change to transferen
of language: a community which adopts a new language
speak it imperfectly and with the phonetics of its mother-ton
The transference of language will concern us later; in the prese.
connection it is important ta see that the substratum theory
account for changes only during the Ume when the langua
is spoken by persons who have acquired it as a second langu
There is no sense in the mystical version of the substratum theo
which attributes changes, say, in modern Germanie langua
to a "Ccltic substratum" that is, to the fact that many ce
turies ago, sorne adult (',cltic-speakers acquired Germanic speee
Moreovcr, the Celtic speech which preceded Germanie in south
Germany, the Netherlands, and England, was itself an invadi
language: the theory directs us back into time, from "race"
"race," ta account for vague "t.endencies" that manifest the
selves in the actual historical occurrence of sound-change.
Aside from their failure to establish correlations, theories
this kind arc confutcd by the fact that when sound-change h
removed sorne phonetic feature, later sound-change may res
in the renewal of just this featUre. If we attribute sorne partie
character to the Primitive Indo-European unvoiced stops [p, t, .
- supposing, for the sake of illustration, that they were unas
rated fortes then the pre-Germanie speakers who had begtl
ta change these sounds in the direction of spirants [f, 6, hl,
388 TYPES OF PHONETIC CHANGE
TYPES OF PHONETIC CHANGE 389
tives in modern German are IOf!;ically superfluous; the use of ad.
jectives is quite like the English, and a text in which these end.
ngs are covered up is intelligible.
In fact, sound-changes often oblitemte features ,,,hose meaning
s highly important. No grammatical difference could be more
essential than is that of actor and verbal goal in an Indo-European .'
language. Yet the difference betwcen the Primitive Indo-European
nominative in *[-os], as in Sanskrit ['vrkah], Greek ['lukos], Latin-
lupus, Primitive Germanie *['wolfaz], Gothie wuljs, and the ae--
cusative in *[-om], as in Sanskrit ['vrkam], Greek ['lukon], Latin'
lupum, Primitive Germanie *[\volfan], Gothie wulj, had been_
obliterated by the weakening of the in pre-English,
sa that the two cases were merv;ed, even in our earliest records,
in the form wul! 'wolf'. In Old English a few noun-types, such
as nominative cam : accusative care 'care,' still had the distinction;
by the year 1000 these wcre probably merged in the form ['karel,
thanks to the ,veakening of unstressed vowcls. In the ;;:une way,
sound-change leads to al! manner of homonymies, sueh as met!:
meat; meed: mead (' meadow'); mead (' drink '), knighl: niyht.
The classical instance of this is Chinese, for it can be shown that:
the vast homonymy of the present-day languages, especiaJly of
Korth Chinese, is dlle to phonetic changes. Homonymy and,
syncrelism, the merging of inflectional categories, are normal te-
sults of sound-change.
The theory of semantic weakness does secm to apply, however,
to fixed fonnulas with excess slurring ( 9. i). Historical1y, these
formulas ean be explained only as far in excess of
normal sound-change. Thus, good-bye rcprcsents an older Gad be
with yc, ma'm an older madam, Spanish usled [u'sted] an aider
vueslra rnerced ['vwestra mer'6ed], and Russian [s], as in [da sI
'yes, sir,' an older ['sudar] 'lord.' In thcse cases, however, the
normal speech-farm exists by the side of the slurred form. The
exccss weakening in these forms has DOt bccn explained and doub
less is connected in sorne way ,vith what ,ve may cal! the sub"
linguistic status of these conventional formubo. In any e'Vent, their
excess weakening differs very much from ordinary phonctic change
Since a is a historical happening, with a bcginnill
and an end, lirnited to a definite time and to a definite body of
speakers, its eause cannot he round in universal consideratio
or by observing speakers al. other times and places. A pho
neti
clan tried to establish the cause of a change of the type [azna >
asn
a
], which occurred in the of the Avesta language,
by observing in t.he laboratory a number of persons who were
directed to pronounce the sequence [azna] many times in succes-
sion. l\Iost of t.he persons - they were Frenchmen - yielded
no result, but at last came one who ended by saying [asna]. The
phonetician's joy was not clouded by the fact that this last person
was a German, in whose native language [z] occurs only before
syllabics.
Il. has been suggestcd that if a phoneme occurs in a language
with more than a certain relative frequency ( 8.7), this phoneme
will he slurred in articulation and subjected to change. The IIpper
limit of tolerable frequency, it is suppospd, varies for different
types of phonemes; thus, [t] represents in English more than 7
pel' cent of the total of uttered phonemes, and in several other
languages (Russian, Hungarian, Swedish, Italian) the unvoiced
dental stop runs to a similar perccntagc, while the type [d], on the
other hand, with a lower relative frequency (in English it is less
than 5 pel' cent) would in any language suffer sound-change,
acoording to this theory, before iL reached a relative frequency
like that of Enf!;lish ft]. The relative frequency of a phoneme is
governed by the frequency of the significant forms that contain
it; thus, ['l5] in English s evidently favored by the high frequency
of the word the. The frequcncy of significant forms is subject, as
shaH sec, to unceasing fluetuation, in accordance with changes
ID practical life. This theory, therefore, has the merit of
ing sound-change with an ever present and yet highly variable
It could be tested if we could determine the absolute upper
limit for types of phonemes, and the actuai frequency of a phoneme
at a stage of a language just before this phoneme was changed -
; say, of [v] in just hefore the havok > hawk.
e should then still have ta account for the speClfic nature of the
since phonemes of any one general type have changed in
thifferent ways in the history of various languages. Against the
eory we must weigh the great phonetic difference hetween lan-
guages and the high frequency, in same languages, of what we may
?&li unusual phonetic types; [-Ci], which plays such a great part
was at one time eliminated (by a
le-: ge to [dD and has remained su in Dutch-German; later it was
lDtroduced inta English by a change from [61 to [<5].
391 TYPES OF PHONETIC CHANGE

B,. ilation, to *['pempc], for we have Primitive Germanie *['fimfe]
, and Old High German fimf, Old English fif, and so on.
\' has [ - l. in words whcre we expect [s -5]'.
.,,': Metathesis is the mtcrchange of two ph?nemes wlthm ward.
, Beside thc expect.ed 'ask,' O.Id Enghsh has also acenan. In
Tagalog sorne morphologlC alternat-lOns seern to be due ta changcs
of this kind; thus, the suffix [-an], as in [a:sin] 'sa.lt' : [as'nan] 'what
. to he salted,' is sometimes accompamed by mterchangc of two
that come together: [a'tip] 'roofing' : [ap'tan] 'what js
to he roofed'; [ta'nim] 'that p!anted' : [tam'nanl. 'what is ta
lants put into it.' In the languages of Europe distant metathesls
[r-I] is fairly cornmon. '1'0 Old English alor 'aider' there cor-
responds in Old High German not only elira but also erila ( > mod-
ern ErIe). For Gothie ['werilo:s] 'lips,' Old English has weleras.
. wtin parabola 'word' (a borrowing from Greek) appears in
Spanish as palabra.
When a phonemc or group of phonemes rccurs within a ward,
one occurrence, toget.her with the intervening sounds, may be
dropped: this change is known as haplology. Thus, from Latin
nutri '1 nourish' the regular feminine agentnoun would he
*nutri-Irix 'nurse,' but the form is actually nutrix. Simlarly, the
compound which would normally have the form *stipi*pendium
lwage-payrnent' appears actually as stipendium. Ancient Greek
(amphi-pho'rews] 'both-side-carrier' appears also as [arnpho'rews]
.. , lamphora.' Changes like these arc very different from those which
'- are covcred by tht; assumption of sound-change; it is possible that
they are akin rather t.o the t.ypes of linguistic change which we have
still ta considcr - analogic change and borrowing.
TYPES OF PHONETIC CHANGE 390
21. 10. Certain linguistic changes which are usually descri
as soundchange, do Dot come under the definition of phone'
change as a graduaI alteration of phonemic units. In various pa'
of Europe, for instance, the old tongue-tip trill [r] has been repla
in modern times, by a uvular tril!. This has happened in No
umbrian English, in Danish and southern Norwegian and Sw
ish, and in the more citified types of French (especially in P .
and DutchGerman. Aside from its spread by borrowing, the n'
habit, in whatever times and places it may first haw ariscn, co
have originated only as a sudden replacement of one trill by
other. A replacement of this sort is surely diffcrcnt frorn the
ual and imperceptible alterations of phonetic change.
Sorne changes consist in a redist.ribution of phonemes.
commonest of these seems ta be dissimilation: when a phoneme ,',
type of phoneme recurs within a form, one of the occurrences
somctncs rep!accd by te different sound. Thus, Latin peregrn
'foreigner, stranger' is replaced in the Romance
type *pelegrinus, as in Halian pellegrino, and in English pilg L'
borrowed from Romance; the first of the two [rJ's has been '
placed by [I]. In the languages of Europe, the sounds [r, l, n] a
cspecially subject ta this replacement; the replacing sound ,,'
usually one of the same group. Where the replacement occurs,
follows quite definite rules, but wc cannat predict its occurreD
The change, if carried out, would produee a st.ate of affairs w
recurrence of certain sounds, such as [r] and [1], was not allo
within a ward - the state of affairs which actually prcvails
the modern Englsh derivation of symbolie words, where we ha.
clatter, blubber, but raltle, crackle 14.9). Probably this type
change is entirely different from ordinary phonetic change.
There is also a type of dissimilation in ,,,hich one of the r
phonemes is dropped, as when Latin quinque ['kwi:nkwe] 'fi
i8 rcplaced, in !tomance, by a type *['ki:nkweJ, Italan ci '
['Cinkwc], French cinq {senk].
Thcre arc severai other kinds of phonetie replacement whf
cannot properly be put on a leve! with ordinary sound-ch
In distant assimilation a phoneme is replaeed by another of
lated acoustic type which occurs elscwhere in the same w
Thus, Primitive Indo-European *('penkWc] 'five,' Sanskrit ['pan
Greek ['pente] appears in Latin not as *[pinkwe], but as qui
In pre-Germanie this ward sccms to have suffered the reversa
PL UCTUATION IN FORMS
393
CHAPTER 22
FLUCTUATION IN THE FREQUENCY OF
FORMS
22. 1. The assumption of phonetic change divides linguistic
changes into two principal types. Phanetic change affects only the
phonemes, and alters linguistic forms only by altering their pho-
netic shape. The English farm wolf is the modern pronunciation of
Primitive Germanic nominative *['wolfaz], accusative *['wolfan],
and several other case-forms, and the merging of these (syncretism)
s merely the l'Csult of the phonetic change. English [mijd] meed,
mead is the modern pronunciation of 0 Id English [me:d] 'meadow,'
[me:d] 'reward,' and ['medu] 'honey-drink'; the homonymy results
simply from the change in habits of articulation. When wc have
listed the phonetic correlations, there l'Cmain a great many dis-
crepancies. Thus, having found that Old English [a:] appears in
modern standard English as [owJ, as in [ba:t] > boat, and sa on,
wc see a discrepancy in the parallelism of Oid English [ba: t] 'hait'
with the modern bait. Secing Old English initial [f] preserved in
father, five, foot, and so on, we find a discrepancy in the sets Old
English [fd] : modern vat and Old English ['fyksen] : modern vixen.
While the modern form cow stands in a normal phonetic correlation
with Old English [ku:], just as house, mouse, out correspond to
Old English [hu:s, mu:s, u:tJ, the plural cows cannot he the mod-
ern form of the Old English plural [ky:} 'cows,' in view of cases
like Old English [hwy:] > wh1f, [fy:r] > fire, [my:s] > mice. If we
adhere ta the assumption of regular phonetic change, wc cannot
class forms like bait, vat, vixen, cows as modern pronunciations of
Old English forms, but must view them as the products of factors
other than simple tradition. Our problem, therefore, is ta find
among these residual forms some uniformity or correlation; to the
extent that we succeed in this, we sha1l have confirmed the value
of the assumption of phonetic change and of the particular phonetic
correspondences we have set up. The neo-grammarians claim that
the assumption of phonetic change leaves residues which shoW'
striking correlations and allow us to understand the factors of
392
1
. guistc change other than sound-change. The opponents of the
ID dff'
hypothesis imply that a i erent assumptlOn
:oncerning sound-change willleave a more intelligible residue, but
they have never tested this by TIHlassifying the data.
If the residual fonns are not continuants of ancient forms with
ooly the alterations of soumlchange, then they must come
into the language as innovations. 'Ve shall see that two kmds of
innovation account for the residual forms namely, the adoption
of forms from other languages (bait from Old Narse) or other
dialccts (vat, vixen from southern-English local dialccts) and the
combining of new complex forms (cow-s on the pattern" singular
noun plus plural-suffix gives plural noun"). These two kinds of
innovation, borrowing and analogie change, will occupy us in the
following chapters; now wc are concerned merely with the claim
that the frills which arc not accountcd for by phonetic correlation,
got into the language at various points in Ume.
22.2. If a form which has becn into a language pre-
vails in gcneral usage as, for instance, prevails as the ordi-
nary plural of cow '''0 have ta suppose that it has gained in
popularity since its first introduction. Conversely, if an old form
- snch as t.he Old English plural [ky:], which, by phonctic develop-
ment, would today be pronounced *[kajJ - has disappeared, wc
must suppose that it went through a period of decline, during
which if. was uscd less and less as the years went by. Fluctuation in
thefrequency of speech-forms is a factor in aIl non-phonetic changes.
This fluctuation can be observcd, ta sorne extent, both at first
hand and in our wrilten records. For instance, since the introduc-
tion of the automobile, the word garage, borrowed from French,
has become very common. \Ve can actually name the speakers
who first used the words chorile, kodak, and blurb; since the moment
of that first use, each of these words has become common. The
disappearance of a form cannot be obscrved at first hand, sinee
we can have no assurance that it. will not he used again, but in
older written records wc find many specch-forros that are no
longer in use. In Old English, [' weorfJan] 't() become' wus one of
the commonest words: [he: 'wearll 'tom} 'he got angry,' [he:
je'wearfJ 'me:re] 'he became famous,' [hc: 'weare of'slejen] 'he
got killed, , [heo 'weare 'widuweJ 'she becaInc a widow.' In the
Dutch-Gcrman area this verb, Dutch worden ['wurdo], German
Werden ['verden], is still so used. The ordinary Old English ward
394
IN FOR:MS
FLUCTUATION IN FORMS 395
for myeel, survives in Scotch mickle, but has -
from standard English. In our fragments of the Gothie Rible...
translation, the word mother is entirely replaced by a krm ['ajOi:],
and the word fa/her occurs only once (Galatiam; 4, 6) and is in ail
other passa!!;es replaced by ['atta), a ward famliar ta us frorn th
Gothie nickname of the king of the Huns, Attila 'littlB father:
This, apparcntly in its original connotation a nursery-word, i's
perhaps somchow connected with the Slavie term for 'father'
Primitive Siavic *(Otl'tSI), Russian (o'lets], which in pre-Slavie
must have crowded out the reflex of Primitive Indo-European
*[p;)'te:r).
Most frequently we observe the complemcntary fluctuation of
tw.o forms; il's 1 and il's me or ralher with [d and \vith [a], are
eVldently rwalforms in present-day Arnerican English. The plural-
form kine beside eows is still very rarely used as a poetic archaism.
In Elizabethan \vritings we still tind the spelJingfat for val, evidenc-
ing a survival of Olcl English [ff:t] , \vhich has sinee been crowded
out by val. Where a speaker knows two rival forms, they differ
in connotation, since he has heard thern from different persans and
under different circumstances.
in the frequency of forms could be accurately
ohserved If we had a record of every utteranco that was made in a
speech-community during whatever period of time we wanted w
study. We eould then kcep a tally-sheet for every form (including
grammatical forms, Bueh as the type he ran away; he feU down in
contrast. with away he ran; down he fell); whenever an uttcrance
:vas r:n
ade
, we could score f1 point on the tal1y-sheet of every form
ln utterance. In this way we should obtain tables or graphs
WhlCh showed the ups and downs in frequency of every form during
the time covered by our records. Such a system of scoring will
doubtless remain beyond our powers, but this imagina.IY system
gives us a picture of what is act.ually going On at al! times in every
spcech-community. 'lole can observe the fluctuation with the naked
eye when it is especially rapid, as in the sudden rise and equally
sudden disuse of popular slangy witticisms. On a smaller seale,
but eontributing ta the total fluctuations in the eommunity, small
groups and individuals indulge in similar whims; everyone cao
recall aId favorite wards and phrases which he and perhaps his
associat.es Once used at every tum. :'Iast fiuctUltfion lS less rapid
and escapes direct observation, but reveals itself in its results _ in
the differences of vocabulary and grammar which when we
compare different historical St3{,TOS of f1 language, or dlaIects of an
area, or related languages.
Leaving aside the origination of new forms, which will concern
us in the following chapters, we must now consider t.he factors
which lead ta the rise or to the dedine in frequency of speech-
forros. Until rccently this topic was neglected, and our knowledgc
is still far from satisfactory.
22.3. We naturally a..'lk at once whether any linguistical!y
finable characteristics of a farm may favor or disfavor its use.
The stylist and the rhetorician tell us that sorne
sound better than othcrs. The only criforion of a phonetic sort
seems ta he this, that repetition of phonernes Or sequences is
often avoided: a phrase like the observation of the I!ystematizaiion
of education is disfavored. In ordinary speech, however, euphony
seems ta play no part; the stock examples of troublesome pha-
neties are far-fetched combinations like Peier Piper picked a peck
of pickled peppers or she seIls sea-shells. On the other hand, various
patternings of recurrcnt phoncmes, such as alliteration (hearth
and home, cabbages and kings) , assonance (a in lime saves
nine) , and rime, and rhythmic repetitions (firsl come, first served),
seem to favor many a speech-form.
In aIl ordinary cases, semantic rathcr than formaI factors con-
tribute to the favor or disfavor of a form. It is natura1 ta suppose,
however, that a fonn which differs strikingly from thc other forms
of comparable meaning, will be disfavorcd. Several students
have eonjectured that certain fell into disuse
cause they were shorter than ordinary speech-forms of similar
meaning. Gilliron believed that Latin apis 'bec' has died out. in
nearly aIl dialects of the French aroa because its modern pronun-
ciation would consist of only a single phonerne [el. It would he
no counter-argument to say t;.at French has grammatical and
lational words of this pattern, such as et [el 'and,' but a case like
eau [0] 'water' ( < aquam) does rnilitate against the theory. It
5ms that some in the oler stages of the Inda-Euro-
pean languages feli into disuse because they were shorter than
ordinary forms of the same kind. The Menomini language, like
French and English, secms ta tolerate words of ail sizes.
ini [o:s] 'canoc' is sharter than ordinary nouns, and [uah1'he
Uses it' shorter than ordinary verrrforms. These forms, which
are ancient inheritances, have bccn largely replad in the sister
languages: Primitive Central Aigonquian *[0:8iJ 'canoe' by longer
derivative nouns, such as Fox [anake;weni], Cree and Ojibwa,
[i :ma:n], - though Cree has also [o:siJ - and Primitive Central
Algonquian *[o:waJ 'he uses it' by a reduplicated form, Fox [ajo:.
wa] or by other words, such as Cree [a:paihta:w]. AlI this,
ever, is doubtful.
The sernantic factor is more apparent in the disfavoring of
speech-forms that are homonyrnous with The reader
will have no difficulty in finding speech-forms that he avoids for
this reason. In America, knocked up is a tabu-form for 'rendered
pregnant'; for this rcason, the phrase i3 not u3ed in thc British
sense 'tired, exhausted.' In older French and English there was
a ward, French connil, connin, English coney, for' rabbit' ; in both
languages this ward died out becauso it resernbled a word that was
under a tabu of indecency. For the same reason, r008/er and don,..
key are replacing cock and a88 in American English. In such cases
there is little real ambjguity, but SOrne hearers react ncverthcless
ta the powerfu1 stimulus of the tabu-word; having caned forth
ridicule or embarrassment, the speaker avoids the innocent homo-
nym. It is a remarkable faet that the tabu-word itself has a much
tougher life than the hannless homonym.
22.4. These cases suggest that homonymy in general may in-
jure the frequency of a form. Many homonyms are distinguished
by differences of grammatical function, as are leader (noun) and
lead'(';f' (infinitive phrase) or bear (noun), bear (verb), and bare
(adjective); in French, [sun] ig sang 'blood,' cent 'hundred,' SM
'without,' sent 'feels, smells,' and s'en 'oneself of it,' as in s'en
aller 'ta go away.' Even with largely similar grammatical fune-
tians, homonymies like pear, pair or pieee, pence or mead, meed
do not seem to lessen the frequency of forms.
Neverthelcss, there is sorne evidence that homonymy may lead
ta troubles of communication which result in disuse of a form.
The classkal instance is Gilliron's explanation of the disappear-
ance of Latin gallu8 'cock' in southwestern France (Figure 14).
ln southern France generally this word is still in use in its modern
forms, such as [gal] or [luI]. A arCll in the extreme south,
however, uses for 'cock' another Latin ward, pullus, modern
[pul], which originally meant 'chick.' NOW, the southwcstcrn
COrner of the French area has made a sound-change by which
E:::::::l modern forms of
l:::::::3 latin Rullus 'chick'
'pheasanf
'farmhelper'
397 FLUCTUATION IN FORMS
. li at the end of a ward has become [tl; thus, Latin bellus
Latin [ ,1 dern [bd]l appears in the southwestern corner as
'retty m, Il d' t . t
P The isogloss of this sound-change cuts the pu us- IS nc
eastern part, whero one says [pul]. a western part
lllto s [put] Outside the PUUUs-dlstnct wc should ae-
h Te one say ., . h
w e
d
. Iy expect to find a forrn *[gat] 'cock,' correspondmg to t e
cor mg
FIGURE 14. The southwest.erll part. of F:ench - South-
West of the heavy Ene - Latin lll] appears ln final poslt,l?n as ln" k "
The unshadcd part of the area uses fo;,nls of Latm g<JUus coc.
The shaded arellS use othcr words for cock. - After Dauzat.
[gall- of ordinary southcrn French, but actually this form nowhere
appears: the entire [-tl-area, in so far as it does not say
calls the cock by qucer and apparently slangy nnames,
by local forms of the word pheasant, such as [aza ], from Latm
PMsianus or by a ward [begej] which means handy-
man' and' is thought to rcpresent Latin vicarius 'deputy, proxy,
vicar.' ,. . .
Now Gilliron points out, the form *[gat] 'cock III thls diS-
trict be homonymous with the ward 'cat,' namely [gat],
1 Stand..rd French bd [hclJ Leforc vowcls, beau [bol before consonants.
FLUCTUATION IN FORMS
396
399
FLCCTUATION IN FORMS
, . 1 H Id himr After a few snch experienees he wauld
_ htm. or o .
; '8
WP
one
of the effective forms at the first trial.
use 5 We frequently fiud rep;ular, or at least more regnlar,
, 22.. by the of irrcglllar eomplex forms, as, roofs,
. mbinatlOns . d
, CO d fs bv the side of rouves, hooves, dwarves, or dreame ,
.' TwQ!::ea ;ide of dreamt, kami, or you ought by t.he side.of
had better. Jn some cases the irrcp;uIar form decldedly m-
g t as l'n cuws e1es shoes, brothers versuS kme, eyne, shoon,
frequen , ' . , h d]
th
Other cxamples are, regular forehead ['fowr-, e ,goose-
bre ren. . . l ['fa cd
berrY ['guws-1bcrijJ, seamstress ['sijmstres] agal.nst lfregu ar r . '
, wzbrij 'semstres]. IIistory shows us that ID such the lr-
go If' frequcntly dies out or survives only in speCIal senses,
rego ar orm " .'
88 when sodden, the oid participle of seeihe, surv.rvcs onl
y
. lU a
transfcrrd meaning. The pinrals of yoat, book, cow, If we contmued
using the Old English forms [gr,:t, be:k, woulel today
*[gijt, bij, kaj]. ,\Yhenever wc know the hlstory of a
through any considerable period, we finc! many of thls kmd,
but the operation of this factor is obscure, beeanse ID many cases
the reguIar form makes no hcadway at aU. The of a reg-
ular foots inst.eac! of feei, or bringed instead of brought IS so rare
88 to he classed as a childish "mistake" or, in aIder people., as a
_ "slip of the tangue." Languages seem ta differ in of
irregular forms, but in general it would seem that a regular rival,
given a good start, has much the better chance. Very common
forms, such aS in English the paradigm of the verb .and the
pronouns l, we, he, she, they, with their over-differentIatlOn, per-
mst in spite of great irregularity.
, 22.6. For the most part, fluctuation does not depend upon
format fcatures, but upon rneaning, and aceordingly escapes a
t>urely linguistie investigation. The changes whieh are always
'going on in the practical life of a community: ,are ?ound to. affect
'the relative frequences of speech-forrns. Ihe introductIOn of
_railways, street-cars, and motor-cars has lesscned the frequen.cy
"of many terms relating to horses, wagons, and harness, and m-
, 'llleased that of terms relating to machinery. Even in the most
lemote and conservative community t.bcre is a constant displace-
l'nent of things talked about i if nothing cIse should alter, there i8
a.t lell.st the change of birth and death.
Anew abject or practicc whieh gains in vogue, carries a speech-
;.fol'tn, Qld or Dew, iolo increased frequencYi examples are many
FLUCTUATION IN FORMS 398
from Latin galtus. This hornonymy must have caused trOuble
in practicallife; therefore *[gat] 'cock' was avoided and replaced
by makeshift words.
What lends weight ta this theory is the remarkable faet thll.t
the isogloss which separates the queer words [azan] and [he-
gej] from the ordinary [gal], coincides exactly ,vith the isoglo
ss
between [-t] and [-1]; this is highly signifieant, because isoglosses',
even isoglosses rcpn:senting closely related features - very'
rarely coineide for any considerable distance.
Adjoining this stretch, the isogloss betweell [-t] and H] coin.
cides for a ways with the isogloss between [put] and [gal]. This"
too is striking and seems to be explicable only if we suppose thll.t "
this part of the Hl-region formerly used gallus and, when the
change of [-Ill to H] had occurred, rep1accd the troublesome *[gat]
by borrowing [put] from the ncighboring pullus-district.
On the rest of its course, the isogloss between H] and H] cuts
through the pullus-district, and mereIy scparates western [putl .
from eastern [pul]; in the pullus-district the sound-change caused
no homonymy and left the Iexicon undisturbed.
One may ask why *[gat] 'cock' rather than [gat] 'cat' was al- ,
fected by the homonymy. Dauzat points out that the morpheme '
*[gatl 'cock' occurred only in this one word, since the derived
form, Latin gallina 'hen' was subject t a differeilt change, giving ,
[garina], while [gat] 'cat,' on the other hand, was backed by a
number of unambiguous derivatives, such as the equivalents of
standard French chaUe 'she-cat,' chaton 'kitten,' chatire 'cat;.:
hale.'
While few instances are as cogent as this, it is likely that
ymy pIays more than an occasional part in the obsolescence of
forms. A few centuries ago, English had not only our present-day
verb let (which represents the paradigm of Old English ['lr:tan]),
but also a homonymous verb whieh meant 'ta hinder' (represent-
ing Old English ['lettanJ); we still have the phrases wilhout let (Jf'
hindrance and a let ball, at tennis. When Shakspere has Haml
et
say l'U make a ghost of him that lets me, he means 'of him that,
hinders me.' After it had come homonymous with let 'permit:
this word must have been singularly ineffective. A speaker who
wanted his hearers ta stop someone - say, a child that was
ning ioto danger, or a thief and cried Let him! might fiud bill
hearers standing aside ta make way. Then he would have to ad
d
If l do prove her haggard,
Though that her jesses were my deaT heart-strings,
l' d whistle her off, and let her down the wind,
To prey at fortune.
401
FLVCTUATION IN FORMS
. spiritual rcpresentatives. The term for the 'left' side appears
118 bave boen replaced in various languages; the Indo-European
use many words, among whieh Ancient Greek [ew-'o:-
os], litcrally 'of good Dame,' is evidcntly euphemistic. One
nU1lloften observe people avoiding unpleasant. words, such as die,
_ thesc words in pre-Germanie replaccd the Primitive Indo-
European term reprcsented by Latin mort' to die.' - .or numes of
serious diseases. The t.erm undertakcr was, to begm wtth, vaguely
ev8Sve, but the undertakers are now trying ta replace it by
ciano In cases like these, wherc the unpleasantness inheres in
the practical situation, the speech-form becomcs undesirablc as
soon as it is tao spoeifically ticd up wit.h the painful meaning.
Tabus of indccency do not seem to load to obsolescence; t.he
tabu-forms are oxcludcd in many or most social situations, but.
by no means avoidcd in others. The substitutes lllay in time bL'-
come tao closely associated with the rneaning and in turn become
tabu. Our word whurc, cognat.e with Latin CaTUS 'dear,' must have
been at one time a polite substitute for \vord now lost to us.
On the whole, however, words of this type do noL seem espccially
given to obsolescence.
The practical situation works in favor of words t.hat cali fort-h
a good response. In commerce, the finds advant.age in label-
ing his goods attractively. This is probably why t.erms for t.he
young of animais sometimes replace the more general nalIle of
the species, as when we say chicken for 'hen.' French poule [pull
'hen' and dialectal [pull 'cook' continue a Latin ward for 'chick.'
The word home for 'house' has cloubtless bccn favored by specu-
lative builders. In Germany, an express t.rain has corne to meo,n
a slow train, as has Schnellwg ['snehtsu:k], literally 'fast-train';
a really fast train is Blitzzug ['blit.s-1tsn:k], literal1y 'lightning-
:trall' - just as in the Unit.ed St.atesfirsl class on a railroad means
the ordinary day-coach accommodation.
There is an advanto,ge, often, in applying well-favored terms to
one's hearer. The habit of usi ng the plural pronoun 'ye' instead
of the singular 'thou,' spread ovcr Europe during the 1Jiddle
In English, you (the of ye)
erowded thou into archatc use; III Dutch, JI} [JcJ] has led ta
t?e entire obsolescence of thou, and has in turn become the in-
form, under the cncroachment of aIl originaUy still more
Il.onfic u [y:], representinp; [/WiJ Edelheid ['Y:W8 'e:delhcjt] 'Your
FLUCTUATION IN FORMS 400
in modern life, such as the terms of motoring, flying, and
less. If the practical situation ceases ta exist, the forms whi
are used in this situation are bound ta beeome less common a
may die out. The tenns of falconry, for instance, have suffe
this fate. Though we still hear beauty in Othello's wards, we
not understand them:
The word haggard was used of a wild-caught, unreclaimed mat'
hawk; were leather stmps fastened to the legs of a haw'
and were not rernoved when the hawk was unleashe; if a haw
flew with the wind behind her, she scldom returned. '
In the early centuries of our em, sorne of the Germanie tri
contaioed a class of people called [la:t], South-German [la:tIl
who were intermediate in mnk between freemen and serfs.
English farm of this word, [le:t], oecurs only once in our recor
in the oldest English law-code, and even herc the ward is explain'
- incorrectly, at that - by the ward [6e;ow] 'serf' written abo
the hoe. The ncw social organization of the English-speaki
tribes in Britain contained no such class of people, and the wo'
went out of use along with the institution.
22. 7. Words that arc under a ritual or ill-omened tahu,
Iikely ta disappear. The Indo-European languages use the m
varied words for' moon'; it is notable that Russian has borrow
Latin ['lu:na} as [lu'na], t.hough otherwise it rnakes scarcely
but highly learned borrowings from Latin. It may be due to
ritual or hunt.ers' tabu t.hat t.he Primitive lndo-European w
for 'bear,' surviving in Sanskrit ['rksah], Gn.'ek ['arktos], La "
ursus, has disappeared in Germanie and in Balt.o-Slavic. In Sla .
it has been replaced by the type of Russian [med'vel], origin
a transparent compound meaning 'honey-eater.' The likfl of t
seems to have happened in 1-lenomini, where the old word f,
'bear,' preserved in Fox [mahkwaJ, Crce [maskwaJ, has been
placed by [awe:hseh], a diminutive formation that seems to ha:
mcant origirially 'Httle what-you-may-call-him.' Cree
'he goes hunting' originally meant simply 'he goes away' - P
sumably there was danger of being overheard by the game or b
402
FLUCTUATION IN FORMS

FLUCTUATION IN FORMS 403
Nobility.' Honorifics of this sort often replace the ordinary seco d
person substitutes Similarly, one speaks in
of :,,hat. to the hearer. In Ibtlian, 'my wife' ia
mw ",:oghe [mla moA:e], but for 'your wife' one says rather la,
sua slgnora [la sua si 'pora] 'your lady.' In French and in German
one 'Mr., l'vIrs., l\liss' to the mention of the hearer's
relatives, as, madame /Joire mre [madam v:Jtr me:r] 'your mother'.
in one likes ta use for the hearer's husband
or wlfe archalC terms of distinguished Bavor: meine Frau [majne
;frawJ 'IllY wife,' but lhre Frau Gemaldin ['i:re fraw ge'ma:lin]
your rli1rs. consort,' and mein },{ann [majn 'man] 'my husband'
but Ihr Herr Gemahl [i:r her ge'ma:l] 'your :VIr. consort.' In
Central Algonquian languages the literai tenns for both 'my wife'
and 'thy wife' are tabu - ogres use them in fairy-tales - and
one says rather 'the old woman' or 'the one l live wi th' or eVen
'my cook.'
In generai, honorific tcrms for persons spread at the cost of
plain ones; gentleman and lady are more genteel than man and
woman.
22.8. General effectiveness, in the shape of violence or wit is
a in fluctuation, which unfortunately quite
the imgUlst s controi. lt leads, for instance, to the sudden rise and
fall of slang expressions. lwund 1896 or so, a transferred use of
the word rubber in the sense of 'stare, pry' played a great. part in
slang; ten years later it was obsolescent, and only rubberneefv.
wagon 'sight-seeing omnibus' has now any great frequency. Then,
round 1905, an interjection skidoo 'be off' and, in the same mean-
ing, an interjectional use of Iweniy-three, came into fashion and as
sud.denly out.. The rise of such forms is due, apparently, to
theIr effectIveness III producing a rcsponse from the hearer. At
tirst they owe this to their novelty and apt .ret violent
ence of meaning; later, the hearer responds weil because he has
heard thern in favorable situations and from attractive people.
Ali these favorable factors disappear from sheer repetition; the
novelty weurs off, the violent metaphor lapses when the transferred
meaning more familiar than the central meaning; the
average of sltuatlOns and speakers associated with the form he-
cames indifferent. Thereupon the slang form dies out. In sorne
cases, however, the older form has moanwhile gone out of use or
becorne archaic or speeialzed; the witticism, having 108t Hs point,
remains in use as a normal form. Thus, Latin eapul 'head' sur-
vives in ItaBan and French in specialized and transfermd senses,
but in the central meaning has been displaced by reflexes of Latin
testa 'potsherd, pot,' Italian testa ['testa], French tte [hd.]. Sim-
Harly, in German, the cognate of our head, namely Haupl [hawpt],
survives in transferred uses and as a poetic archaism, but has
been replaced, in the sense of 'head' by Eopf, cognate \Vith Eng-
lish cup. The forceful or wittYterm, weakened through frequency,
may suffer encroachment by new rivais, as in the countless slang
words for' head' or 'man' or 'girl' or 'km,' or in a set alike awfully,
terribly, frighifully (glad to see you).
This factor is casily recognized in extremc cases, but figures
doubtless in many more which elude our grasp, especially when
the fluctuation is observable only from far-off time.
22. 9. The rnost powerful force of aU in fluctuation works quite
outside the Iinguist's reach: the speaker favors the forms which
he has heard from certain other speakers who, for some reason of
prestige, influence his habits of speech. This is what decides, in
countless instances, whether one says it's me or it's l, rather with
[el or with [a], either and neither with [ij] or with [aj], roofs or
rooves, you ought to or you'd better, and so on, through an endless
list of variants and nearly synonymous forms. Dialect geography
and the history of standard languages show us how the speech of
important eornmunities is constantly imitated, now in one feature
and now n another, by groups and persons of less prestige. The
more striking phases of this leveIing process will concern us in
connection wit.h linguistic borrowing. We may suppose that many
features of lexicon and grarnrnar, and sorne features of phonetics,
have a social connotation, different for different groups and even
for individual speakers. In the ideal diagram of density of com-
munication 3.4) wc should have to distinguish the arrows that
lead from each speaker to his hearers by gradations representing
the prestige of the speaker with reference ta each hearer. If we
had a diagram with the arrows thus weighted, we could doubtless
predict, to a large extent, the futUre frequencies of linguistic forms.
1 It is in childhood, of course, that the speaker is most affect.ed by
the authority of older speakers, but ail through life he goes on
adapting his speech t the speech of the persans whom he strives ta
resemble or ta please.
AN ALOGIC CHANGE 405
CHA PTER 23
ANALOGIC CHANGE
23. 1. Many speech-forms :lrc not continuants of forms that
existed in an older stage of the same language. This is obvious in
the case of borrowings: a ward Iike toboggan, taken over from an
Ammican Indian language, cannot have been used in English
before the colonization of America, and, of course, wc do not. find it
in documents of the English language \vhich date from before that
time. ln very many instances, however, the new fonn is not
borrowed from a foreign lanll;uage. Thus, the plural-form COW8
does not appritr in OId and l\1ddle English. The Old English
plural of [ku:] (whcnce modern row) is cy [ky:j, which surllives,
as [kaj], in a number of modern Englsh dialccts. Round the year
1300 thcre appears in our records a form kyn, which survives in the
modern archaic-poetic form kine. Only SOrne centuries later do we
meet the form cows; the New English Dictionary's first rcference,
from the year 1607, has it as an alternative of the aIder form:
KI'ne or Cows. Evidenlly cows s not thc continuant., with only
phondic change, of kine, any more than kine bears this relation ta
kye: in bath cases a new speech-form has come into t.he language.
The faet that the form cows is not the continuant, with only
alterations of sound-change, of the older forms; is self-vident.
Sirietly l';pel1king, howcver, this is only an inference which we
make from thc primary fact of phonctic discrepancy. '''le know
that Ole! English ly:] appe:us in modern standard Enp;lish as [aj],
e.p;. inwhy, mice, bride from Ole! English [hwy:, my:s, bry:d], and
that modern [awJ, as in cOlOS, represents an 0 Id English [u:], as in
colO, how, mme, out from Old English [ku:, hu:, mu:s, u:t]. Further,
wc know ihat modern [z], as in cows, is not. added by any sound.
change, but rcprcsents ld Enr.;lish ls]' as in stones from Ole!
English l'sta:nasJ. In many cases, however, the novelty of a specch-
farm is not so apparent and il'; rcvealed onl}' by a systcmatic
comparison of sounds. The form days superficially resembles t.he
Ole! English plural-form dagas, which we interpret. as ['dagas], pre-
sumably with a spirant 19], but the phonetic dcvelopmcnt of the
404
Ole! English sound-group [ag] appcars rather in forms like ['sap;e]
> saw (implem,mt.), ['sagu] > saw 'saying,' ['hagu-'60rn] > haw.
thorn, ['dragan] > draw. This is confirmee! by the fact that in
earlier "Yliddle English we find spellings likc daues, for the
plural of dei 'day,' and that. spel1ings which agree wit.h the modern
form days appcar only round the year 1200. If our st.atements of
phonetic correspondence are correct, the residues contain t.he
new forms. One of the strongest reasons for adoptmg t.he assump-
tian of regular phonetic change is the fact that the constitution of
the rcsidues (aside from linguistic borrowings, which we shaH con-
sider inlater chapters) throws a great deal of light. upon the origin
of new fomm. :l\Jost of the word-forrns which arise in the course of
time and reveal themselves by their deviation frOIll normal phonetic
correspondence, belong to a single well-defined t.ype. This cannot
he duc to accident.; it confirms t.he assumption of phonetic change,
and, on the other hane! allows us ta st.udy the proccss of new-
formation.
The great mass of ward-forms that arise in the course of histDrY
consists in new cornbinations of complex forms. The form cows,
arising by the sde of kye, kine, consists of the singular cow ( < Old
English [ku:]) plus t.he plural-suffix Old English [-as]);
similariy, days, arising by the side of aIder daws, consists of the
singular day ( < Old English [dd]) plus the .'lame suffix. A vast
number of such instances, from the history of the most diverse
languages, leads us ta believe that the analogie habits ( 16.6) are
subject to displacement - t.hat at a time when the plural of
cow was the irregubr form kine, the speakers might create a
regular form eows, which then entered into rivalr}' with t.he old
form. Accordingly, this type of innovation is called analogie
change. Ordinarily, linguists use this tenu to include both the
original creation of the new form und it.s subsequent rivalry with
t.he old form. Strictly speaking, we should distinguish betwccn
these two events. After a speaker has heard or uttered the new
form (say, co-ws), his subsequent utterance of this fonn Or of the
aider form (kine) is a matter of fluctuation, such as we considercd
in the last chapter; what we did not there consider and what. con-
cerns us now, is the utterance, by someone who hus never heard it,
of a new combination, such as co-w-s insteud of kine.
23. 2. In most cases - and t.hese are the ones we come nearest
to underst.anding - the process of uttering a new form is quite
406
AN ALOGIC CHANGE
ANALOGIC CHANGE 407
like that of ordinary grammatical analogy. The speaker who
without having hcard it, produced the form COW8, uttered thi;
form just as he uttercd any other rcgular plural noun, on the
scheme
sow : sows = cow : x.
The rnode1 set (sow : sows) in this diagrarn represents a series
of models (e.g. bough : boughs, heifer : heifers, stone: stones, etc.,
etc.), whieh, in our instance, includes aU the rcgular noun-para,..
digms in the language. 1'loreover, the sets at ither side of the
sign of equality arc not limited t two members. The independent
utterance of a form like dreamed instead of dreamt [dremt], could
he depicted by the diagram :
saeam : sereams : screaming : screamer : sercamed
= dream : dreams : dreaming : dreamcl' : x
Psychologists sometimes abject to this formula, on the ground
that the speaker is not capable of the reasoning which the pro-
portional pattern implies. If this objection hdd good, linguists
woul be debarred from making almost any grammatical state-
ment, since the normal speaker, who is not a linguist, does not
deseribe his speech-habits, and, if wc are fooHsh enough to ask him,
fails utterly to make a correct formulation. Edueated persons, who
have had training in school grammar, overestirnate their own
ability in the way of formulating speech-habits, and, what is
WC'lrse, forget that ther olYe this ability to a sophisticated philo-
sophical tradition. They vielY it, instead, as a natural gift which
they expect to tind in aIl people, and feel free to deny truth of
any linguistic statoment which the normal speaker is incapable of
making. We have to rernmber at ail times that the speaker,
short of a highly specialized training, is incapable of clescribing his
speech-habits. Our proportional formula of analogy and analogie
change, like ail other statements in linguistics, describes the action
of the speaker and cloes not imply that the speaker himself could
give a similar description.
In studying the records of past speech or in cornparing related
languages and dialeets, the linguist will recognize many differences
of word-form, such as the emergence of cows beside aIder kine.
The habits of morphology arc fair!y rigid; word-lists and tables of
infiection are relatively easy to prepare and help us ta dctect inno-
vations. It is otherwise with phrasaI forms. Aside from the
fection of our descriptive technique in syntax, rctarded, as it has
been, by philosophie habits of approach, the syntactic positions of
a language can be filled by so many different forms that a survey
is hard ta rnako. The linguist who suspocts that a certain phrase
departs from the older syntactic habits of its language, may yet
/ind it difficult or impossible to make sure that this older usage
really excluded the phrase, or ta detennine the exact boundary
between the aider an the newer usage. Nevertheless, we can
sometimes recogniz syntactic innovations on the proportional
pattern. From the sixtccnth century on, we find English sub--
ordinate clauses introduced by the word like. We can picture the
innovation in this way:
10 do better Ihan Judith: to do beUer than Judith did
= to do like Judith : x,
where the outcome is the construction to do like Judith did.
A phrasaI innovation which dors not disturb the syntactic habit
may involve a new lexical use. In this case, our lack of control over
mf:anings, especially, of course, whore the speech of past times is
concerned, acts as an almost insuperable hindrance. The practical
situations which make up the rneaning of a speech-form arc not
strictly definable: one could say that every utterance of a speech-
forrn involves a minute semantic innovation. In older English, as
in sorne modern dialects, the word meat had a meaning close to
that of food, and the word flesh was used froely in connection with
eating, as in this passage (from the year 1693): who fiesh of animals
refused to eat, nor held aU sorts of pulse for lawful meat. A campound
flesh-meal served, for a while, as a compromise. The prevalence of
food and foddcr where at an earlier time the word meat was
mon, and the prevalencc of flesh-meat and meut where at an earlier
time flesh would have been the normal term, must be attributed to
a graduaI shifting of usage. The difficulty of tracing this has lod
linguists to view the process as a kind of whimsical misapplication
of speech-forms. If we remember that the meaning of a speech-
form for any speaker is a product of the situations and contexts
in which he has hoard this form, wc can see that here too a dis-
placement must be merely an extension of sorne pattern:
wave the bones and bring the flesh : leave the banes and bring the meat
=give us bread and jlesh : x,
408 ANALOGIe CHANGE ANALOGIe CHANGE 409
resulting in give us bread and meat. Doubtless wc have ta do, in
both grammatical and lexical displacements, with one general type
of innovation; we may call t analogic-semanlic change. We shaH
leave the lexical phase of this, semantic change, for the next chapter,
and consider first the morc manageable phase which involves
gramnlatical habits.
23.3. Wecan distingush only in theory between the actual
innovation, in which a speaker uses a form he has not heard, and
the subsequent rivalry betwcen this nm.,.. form and sorne older fOTIn.
An observer who, a few years ago, heard the form radios, might
suspect that the speaker had never heard it and was creating it on
the analogy of ordinary noun-plurals; the observer could have no
assurance of this, however, since the form could be equally weil
uttered by speakers who had and by those who had Dot heard it
before. Bath kinds of speakers, knowing the singular radio,
would be capable of uttering the plural in the appropriatc situa-
tion.
It may be worth noticing that in a case like this, which involves
dear-cut grammatical categories, our inability ta define meanings
need give us no pause. A formula like
SINGULAR PLURAL
piano : pianos
= radio : x
will hold good even if our definitions of the meanings of these
categories (e.g. 'one' and 'more than one') should tum out to he
inexact.
The form radios did not conflict with any aIder form. The diffi-
culty about most cases of analogie change is the existence of an
older form. An observer round the year 1600 who heard, let us
suppose, the earliest utterances of the form cows, could probably
have made the same observations as we, a fev,' years aga, could
make about the form radios: doubtless many speakers uttered it
independently, and could not be distinguished frorn speakers who
had aiready heard iL However, the utterances of the form cows
must have been more thinly sown, sinee thero was a1so the tradi-
tional farm kinc. In the ensuing rivalry, the new fOTIO had the
advantage of regular formation. It is safe ta say that the factors
which lead ta the originaton of a form are the same as those which
favor the frequency of an existing form.
Wc do not know why speakers sometimes uUer ncw combina-
tians instead of traditional forills, and why the new combinations
sometimcs rise in frequcncy. A form like foots, instead of feet, is .. "
occasionally u by children; we call ita"childish eITor"
and expect the child soon to acquire the traditional habit. A
grown person may say foots when he is tired or illL"itered, but he
does not repeat the farm and no one adopts it; we caH it a "slip of
the tangue."
It seems that at any one stage of a language, certain features
are relatively stable and othcrs relatively unstable. We must
suppose that in the sixteenth century, owing ta ant,r-cedent de-
velopmr,nts, there were enough lternative plural-forms (say,
eyen: eyes, shoon: shoes, brethren : brothers) to make an innovation
like cows relativcly inconspiclloUS and acceptable. At present, an
innovation like foots seems tn have no chance of survival when
it is pl'Oduccd from time to time; we may suppose that inno-
vation and ftuctwltion are at work rather in the sphere of
plurals with spirant-voicing: hooves : hoofs, laths [klSz : IEeS],
and so on.
The creation of il form like is only an episode in the rise
in frequeney of the regular pluml-suffix [-z, -z, -s]. Analogie
scmantic is mere!y fluctuation in frequency, in sa far as
it displaces grammatical and lexical t.ypes. The extension of a
farm into a new combinaton with a new accomprmying- form is
probably h.vored by ils earlier occurrence with pbonetically or
semantically rcbted forllls. Thus, the use of [-z] with cow was
probably fr1vorcd by the existence of other plurals in [-aw-z], sueh
as sows, bi'ows. Simhrity of rneaning plays a part.: soU's, he/fers,
ewes will attra.ct cows. Frequent occurrence in context probably
increascs the attraction of 11 model. The Latin noun senatus
[sc'na:tus] 'scn[1to' had an il'l'egular infkction, including a geni-
tive seM/US [so'na:tu:s]; by the side of this there arose a new geni-
tive on the regular modc1, suw/i [se'na:ti:J. Tt has bcen suggested
that tho chief model for t.his innovation was the noun
populus ['populus] 'people,' gcnitive populi ['populi:], for t.he two
words \Vere habitually uSt'd together in the phrasc sena/us popu-
lusque [se'na:tus popu'lus kwe] 'the Senate and People.' The most
powerful factor is surely that of nurnbers and frequeney. On the
one hand, regular form-classcs increase at the cost of smaller
groups, and, on the other hand, irregular forms of very high fre-
410 ANALOGIC CHANGE ANALOGIe CHANGE 411
Among the Old English paradigms of ather types, that of 'foot'
shows us an interesting redistribution of forms;
SINGULAR PLUI\AL
nom.Macc. [fo;t] [fe:t]
dat. [fe:t] ['fo:tllm]
gen. ['fo:tes] ['fo:ta]
Here the form with [0:], modern foot, has been generalized in
the singular, crowding out the ald dative, and the form with [e :],
modern feel, in the plural, crowding out the old dative and geni-
tive forms.
In a few cases, two forms have survived with a lexical difference.
Our words shade and shadow are reflexes of different forms of a
single Old Eoglish paradigm:
MODEI\N
OLD ENGLIBH PEONETlC RESULT
singular
nom.-acc. [sko;h] *[sof]
clat. [sko;] [suw]
gen. [sko:s] *[SOi'l]
plural
nom.-acc. [sko:s] *[Ss]
dat. [sko;m] *[suwm,
gen. [sko:] [suw]
PLURAL
['dagasJ
['dagum]
['daga]
[d8j]
['deje]
['dejes]
SINGULAR
nom.-acc.
dat.
gen.
Later, there came a change of [g] to [w], whence the Middle Eng-
lish irregularity of dei, plural dawes; the latter form, as we have
seen, was superseded by the regular new combination of day plus

The early Old English 10ss of [hl between vowels with oontrac-
tion ( 21.6), led ta paradigms like that of 'shoe,' which were
regular in Old English, but by subsequent phonetic change, would
have led to highly irregular modern sets:
quency resist innovation. Irrcgular forms appear chiefly among
the commonest words and phrases of a language.
23. 4. The regularizing trend of analogie ehange appears plainly
in inflectional paradigms. The history of thc regular plural-for_
mation of English is a long series of extensions. The suffix [-cz,
-s] is the modern form of an Old English sllffix [-as], as in slan
[sta:n] 'stone,' plural stanas ['sta:nas] 'stoncs.' This sllffix in ld
English belonged only t the nominative and accusative cases of
the plural; the genitive plural stana ['sta:na] and the dative plural
stanum ['sta:num] would both be represented today by the form
stone. The replacement of this fonn by tbe nominative-accusative
form which is now used for the whole plural, regardless of
syntactic position, is part of a largcr process, the loss of case-
infleetion in the noun, which involved both phonetic and analogie
changes.
The Old English nominative-accusativc plural in -as occurred
with only one type (the largest, t be sure) of masculine nouns.
There were some classes of masculine nouns which formed the
plural differently, as, ['mou] 'son,' plural ['suna]; among these was
a large class of n-plurals, such as ['steorra] 'star,' plural ['steorran].
Some nouns fluetuated: [feld] 'field,' plural ['fe1da] or ['fcldas].
Wc do not know the origin of t,bis fluct.uation, but, once granted
its existence, we can see in it a favoring condition for the spread
of the [-as]-plural A ncologism likc ['suoas] inst.ead of older ['suna]
'sons' would perhaps have had no botter chance of success than
a modern foots, had it not bcen for the familiar fluctuation in cases
like the word 'field.'
Neuter and feminine oouns in ld English had not the s-plural.
Examples of neuter types are [word] '\vord,' with homonymous
plural, ['spere] 'spear,' plural ['sperul, ['e:ajej 'eye,' plural ['c:agan]j
feminine types, [' karuJ 'care,' plural [' kara], [' tungeJ 'tongue,'
plural ['tungan], [bo:k] 'book,' plural [be:k].
Even where the II-plural was iraditional, sound-change led to
divergent forms. Thus an carly voicing of spirants between
els led to the type kmfe : knivel!. Other irregularities of this sort
have been overlaid by In pre-English, [a] becarne
[8] in monosyllablcs and before [el of a following syllable; after
this change, [g] bccame [j] befare a front vowel and in final
sition after a front voweL The result was a set of alternations, as
in the paradigm of 'day':
412 ANALOGIC CHANGE ANALOGIC CHANGE 413
MOOERN
OLD ENOLlaa PHONETIC EQUIVALWNT
singular
nominative ['skaduJ [kjd]
other cases ['skadwe] ['sEdow] shadow
plural
dative ['skadwum] ['sedow] shadow
other cases ['skadwa] ['sedow] shadow
whence the modern singular form riddle. This creation of shorter
or underlying forms is caUed back-formation. Another example
is Old English ['pise] 'pea,' plural ['pisan]; aH tho forms of the
paradigm lead ta modern pease, peas [pijz], and the singular pea
is a back-formation. Similarly, Old French cherise 'cherry' was
borrowed in l\liddle English as cheris, whencc modern cherries;
the singular cherry is an analogie creation.
23. 6. In ward-formation, the most favorable ground for
logic forms is a derivativc type which bears sorne cIear-eut mean-
ing. Thus, we form aU manner of new agent-nouos in -er, on what
is at present a normal grammatical analogy. This suffix was bor-
rowed in pre-English time from Latin, and has replaeed a number
Both forms, shade and shadow, have been generalized for the
whole singular, and have served as underlying forms for new
regular plurals, shadeB, dwdows; the rivalry of the two resultng
paradigms has ended in a lexical differentiation. The \Vords mead
and meadow arob' in the same way, but in this case the fluctuation
seems to be ending in the obsolescence of the form mead.
The word 'gate' had in OId English the nominat\'e-accusative
singular geat [jat], plural galu ['gatu]. The old singuIar, ,vhieh would
give a modern *yat, has died out; the modern form gale repre-
sents the oid plural, and the new plural gates has been formed on
the regular moder.
Analogie creation is not limited ta complex forms. A simple
form may he created on the analogy of cases wherc a complex form
and a simple form exist side by side. The l\.Iiddle English noun
redels 'riddle,' with homonymous plural, was subjected ta ana-
logie change of the pattern
stones
= redels
SINGULAI\
stone
x,
of native types. In Old English, the agent of ['huntian] 'ta hunt'
was ['hunta], whieh has been replaced by hun/er. At a later time,
webs/er was replaeed by weaver, and survives only as a
name. In boat-black, chimney-sweep old forms survive as
members. 'Ve not only form new agent-nouns, such as
flager, debunker, charlestoner, but also make back-formations, such
as the verb chauffe [sowf] 'drive (someone) about in a motor-
car' from chauffeur ['sowfr]. An analogy that permits of new for-
mations is said ta be "living."
The old suffix in webs/er is an example of a type which
perhaps never could have been described as "regular" or "living"
and yet had its period of expansion. It seems to have denoted
(as is still the case in Dutch) a fcmale agent. The female meaning
survives in originally 'spinneress.' Apparcnt.ly, the fe-
male meaning was not obviolls in aIl the words: the suffix became
indifferent as ta sex and appcars in lapster, huckster, teamster,
maltster, webster 'weaver,' dunster 'dunner, bailiff.' The action was
not necessarily useful, witness sonyster, rimester, trickster,
ster, puns/er. A non-human agent appears in lobster, which prat.
ably rcprescnts Old English Zoppeslre, originally 'jumper.' An
inanimate object is roadster. An adjective, instcad of verb or
noun, underlies youngBter. After the restriction to females was
lost, words in -sler combined with -ess: huckress, songstress,
stress. This last, by the shortening of vowels before clusters, be-
came ['semstres]; the more regular rival form ['sijmstres] is ana-
logic, with the vowel of the underlying seam. In cases like -ster
we see a formation spreading from form to form without ever at-
taining to the free expansion of "Iiving" types.
Sorne formations beeome widely usable without pre-cmpting
a domain of meaning. In Englsh, the suffixes -y, -ish, -ly, whieh
derive adjectives, have an remained quite "alive" through the
historical periud, spreading from ward to word, and settling in
varions semantc patehes. Thus, with the suffix -y (from Old Eng-
lish -ig), sorne words appear in our Old English records (e.g.
mighty, misty, moody, bloody, speedy) , white others appear only
luter (e.g. earthy, wealthy, hasty, hear/y, fiery). When the suffix is
added to words of foregn origin, the date of the borrowing gives
us a limit of age ("terminus post quem") for the new combination:
8Ugary, fiowery, creamy. At present, this suffix ls expanding in
certain zones of mcaning, sllch as 'arch, affected': summery (e.g.
414 ANALOGIe CHANGE
ANALOGIC CHANGE 415
of clothes), sporty, wanky, arly (' pretendedly artistic'), booky
('pretendedly bookish'). In the same way, -ish, in sorne c o m b i ~
nations a mere adjective-former (boyish, girlish) , has staked a
claim in the zone of 'undesirably, inappropriately resembling,' as
in mannish, womanish (contmst manly, womanly) , childish (con-
trast childlike). The starting-point of semantie specialization is
to be sought in forms where the undcrlying ward has the special
value; thus, the unpleasant flavor of -ish cornes from words like
loutish, boorish, swinish, hoggish.
The shape of morphologie constituents is subjeet ta analogic
change, especially in the way of enlargcmcnt. In Latin, the set
argen1um [ar'gentum] 'silver' : argenlarius [argcn't:1:rius] 'silver-
smith' reprcsents a regular type of derivation. In the history of
French therc was repeated losses of final phonemes; the modern
forms are argenl [adon] : argentier [ariantje]. The formula of deri-
vation has become: add the suffix [-tje]. This suffix, aeeordingly,
appears in words which (as the historian, quite irrdevantly, re-
marks) never eontaincd a [t] in the critical position: French fer-
blanc [fEr-blo"] 'tin' (Latin type *ferrum blankum 'white iron,'
with the Germanie adjective blank) underliesferblantier [ferblontje]
'tinsmith'; bijou [bizu] 'jewel' (from Breton bizun) underlies
bijoutier [bizutje] 'jeweler,' and so on.
In time, an affix may consist entirely of aeeretive elements, with
no trace of its original shape. In Old EngHsh, verb-paradigms
were derived from nouns on the pattern [wund] 'a wound': ['wun-
dian] 'to wound,' and this is still the living type, as in wound :
to wound, radio : to radio. In a few instances, however, the under-
lying noun was itself derived, by means of a suffix [-en-], from an
adjective, as in the set [fest] 'firm, strong' : ['festen] 'strong place,
fortress' : ['festenian] 'ta make firm, to fortify,' Thanks to sorne
fluctuation in frequency or meaning - sueh, perhaps, as a decline
or specialization of the noun ['fEsten] - the pair [fest] 'firm' :
['festenian] 'to make firm' served as a model for new-formations
on the scheme
fast: fasten = hard : x,
with the result of forms like harden, sharpen, sweeten, faiten,
g1,ad(kn, in whieh a suffix -en derives verbs from adjectives.
Less often, a relatively ndependent form is redueed to affixal
status. Compound-members arc occasionally reduced, by sound
change, to suffixes; thus, the suffix -ly (manly) is a weakened form
of like, and the suffix -dom (kingdom) of the word doom. This hap-
pens especially when the independent word goes out of use, as
in the case of -hood (childhood), which is a relie of an Old English
wo
rd
[ha:d] 'person, rank! German jl,[esser ['meser] 'knife' is
the modern form, with analogic as weIl as phonetic shortening,
of Old High German ['messi.rahs] original1y 'food-knife,' in which
the second member, [sahs] 'knife,' had been disfigured by Verner's
change ( 20.8) and the subsequent change of ['1] to [r]. In German
8chuster ['su:ster] 'shoemaker' the unique suffix [-ster] reflects
an old compound-membcr [su'tE:re] 'eobbler.' Merging of two
words into ODe is excessively rare; the best-known instance is
the origin of the futUre tense-forms in the Romance languages
from phrases of infinitive plus' have': Latin amare habeo [a'ma:
re
'habeo:] '1 have ta, am to love' > French aimerai [Emre] '(1)
shalilove'; Latin amare habet [a'ma:re Jhabet] 'he has to, is to love'
> French aimera [ClOra] '(he) will love,' and sa on. This develop-
ment must have taken place under very unusual conditions;
above aIl, we must remember that Latin and Romance have a
complicated set of verb-infiections which scrved as a model for
oneword tense-forms.
Back-formations in word-structure are by no means uneommon,
though often hard to recognize. Many verbs in the foreign-
learned vocabulary of English resemble Latin past participles;
this is aU the more striking since English has borrowed these
words From French, and in French the Latin past participles have
been obscurc by sound-change or replaced by new-formations:
Latin agere ['agere] 'to tead, carry on, do,' past participle actus
['aktus] 'led, donc' : French agir [azi:r] 'ta act,' participle (ncw-
formation) agi [azi] 'acted' : English to act; Latin affligere [af-
'fli:gerc] '"ta strike clown, affiict,' participle afflictus [af'fliktus]
'stricken, affiicted' : French ajJliger [afiize], participle affiig
[aflize] : English 10 ajJlict; Latin separare [se:pa'ra:re] 'to separate,'
participle separatus [se:pa'ra:tus] : French sparer [separe], par-
ticiple spar [separe] : English ta separate. The starting-point
for this habit of English seems to have been back-formation from
nouns in -tion: English verbs like act, afflict, separate are based on
nouns like action, affliction, separation, from Latin actionem, af-
ftictionem, separationem [akti'o:nem, afflikti'o:nem, se:para:ti'o:-
nem] via French action, affliction, sparation, in modern pronun-
ciation [aksjoO, afliksjoD, separasjoD]. The immediate modela
416 ANALOGIC CHANGE
. .. '
ANALOGIC CHANGE
417
must have been cases like communion: to commune (Old French
communion: comuner); the general background "las the English
homonymy of adjective and verb in cases like warm : to warm ""
separale : 10 separale. This supposition is confirmed by the fact
that the nouns in -lion appear in our records at an earlier time
on the whole, than the verbs in -l. Of the 108 pairs with
Ain the New English Diciionary, the noun appears earlier thant':
the verb in 74 cases, as, action in 1330, but to ad in 1384; affliction
in 1303, but to afflici in 1393. Moreover, "le sometimes see
late rise of the verb with -t: in the case of aspiralion : to aspire
we have stuck ta the Latin-French scheme, but round 1700 there
appears the new-formation to aspiraie. Modern formations of this
sort are Volute, based on evolulion, as a rival of the older evolve,
and elocute based on elocution.
23. 6. The task of tracing analogy in "lord-composition has
scarcely been undertaken. The present-day habits of word-com
position in English produce the illusion that compounds arise
by a simple juxtaposition of words. The reader need scarcely he
told that the modern English pattern, in which the compound
word equals the independent forms of the members, with modi
fication only of ward-stress, is the product of a long series of
regularizing analogie changes. Thus, ['fowr-lhedJ forehead, as
a rival of ['fared], which has bcen irregularized by soundehange,
IS due to analogie re-formation:
fore, arm : fore-arm ['fowr-1arm]
= fore, head : x.
The relation of the compound ta independent words often suf
fers displacement. Primitive Indo-European did not use
stems as compound-members; to this day, English lacks a verbal
type, *to meal-eat, which would match the noun and adjective
types meal-ealer and meat-eating ( 14.3). Several Indo-European
languages, however, have developed compounds with verbal mem-
bers. In English we have a few irregular forms like housekeep,
dressmake, backbile. From a compound noun like whitewash WB
derive, with a zero-elerncnt, a verb to whitewash, and from tbis
an agent-noun whilewasher. The irregular type to housekeep is
probably a back-formation on this model:
whitewasher : 10 whitwash
= housekeeper : x.
In a now classical investigation, Hermann Osthoff showed how
fonDS of this kind arase in several of the Indo-Europcan languages.
In Dld High German, abstract nouns like ['beta] 'prayer' were
use
d
, in the normal inberited fashion, as priar of com-
pounds: ['beta-1hu:s] 'prayer-house, house for prayer. Tbe mor-
pbologically eonnected ver? ['beto:nJ. 'ta pray' had a
suffixal vowel and did not. mterfere wlth the compound. Durmg
the Middle Ages, however, unstressed vowels were weakened ta
a. uniform [el and in part lost; hence in Middle High German
(round the year 1200), in a set like ['boten] 'to pray' : ['bete]
'prayer' : ['bete-1hu:s] 'house for prayer,' the compound-member
resembled the verb as much as it resembled the noun. If the noun
lost in frequency or \Vas specialized in meaning, the compound-
member became equivalent to the verb-stem. Thus ['bete] 'praye
r
'
lost in frequency - the modern language uses a different deriva-
tive, Gebet [ge'be:t] 'prayer' - and, for the rest, was specialized
in a meaning of 'contribution, tax.' As a result of t.his, compounds
Iike Bethaus ['be:t.-1haws] 'house for praying,' Bellag ['be:t-,ta:k]
'day of prayer,' Betschwester, ['be:t-Isvester] 'praying-sister,' that
is 'nun' or 'over-pious woman,' can he described only as containing
the verb-stem [be:t-] of beten [be:ten] 'to pray.' Accordingly,
ever since the :\Iiddle Ages, new compounds of this sort have been
formed with verbal prior members, as Schreibtisch ['srajp-1ts]
'writing-table,' from schreiben 'ta write,' or Lesebuch ['Ie:ze-[bu:
x
]
'reading-book' from lesen 'ta read.'
The fluct.uation between irrcgnlar compounds, such as ['fared]
forehead, and analogically fOl1TIcd regular variants, sneh as ['fowr
Ihed] , serves as a model for new-formations which replace an ob.-
SCure form by a compound-mernber. Thus, inmosl, norlhmost, ut-
mast (and, with regularization of the first. member, oulmosO, with
the word most as second member, are analogie formations which
replace the Old English type ['innemest, 'norOmest, 'u:temest];
the [-mest] in these words was a special form (with aecretion)
of the superlative suffix [-cst]. Regularizing new-formations likc
this, which (as the historian finds) disagree with the earlier struc-
ture of the form, arc sometimes called popular etymologies.
23. 7. Analogie innovation in the phrase is most casily seen
when it affects the shape of single words. Conditioned sound-
changes may produce different forms of a ward according ta its
IJhonetic positions in the phrase. In the types of English which
[el : [di] = rem] : x
419
ANALOGIC CHANGE
0/.0 ENGLI811 >
EAHLY M,DDLE ENG/.I811
bcfore vowel
otherwise
singular
oxe
nominative
oxa
ox
other cases
oxan
oxen
oxe
plural
oxe
nom.-acc.
oxan
oxen
dat.
oxum
oxen
oxe
oxena
oxen
oxen
gen.
[ t 1'] 'does he
1
cl
in a modern sandhi-form aime-i-il em
res
ute
l
ove?' t cl l
the later Old Englsh period, final [n] after an uns ,resse, vowe,
lost, except in sandhi before a vowel. Thus, eten. to eat
was i hand became a hand but an arm remamed. In
J)ecame ee, an '. h . cl .
f the article a . an the resulting alternatlOn as survive ,
the case 0 .' . d One
. early modern English one still said my fnen : mIne enem
y
:
IIIust suppose that at the time of the loss of -n, the language d1d
word-boundaries in the manner of
h
dh ln] was generalized in a few cases as a word-1mtIal. Old
T e san l, ddl 1 h ' t and
En lish efeta '[eveta] 'lizard' appears in l'vIl e <,ng IS as ew e
whence modcrn newt. A phrase likc an ewte must have.
[a'newte] and (doubtlcss under sorne special
frequency or mcaning) sllbjected to the new-formatIOn
[a'na:me] 'a name' : ['na:me] 'name'
= [a'ncwte] 'a lizard' : x,
'th the result that one said newte. Similarly, eke-name .' supple-
name' gave rise ta a by-form with n-, modern
for ihen anes is now for the nome. On the other hand, an mIt.lal
ln] was in sorne forms trcated as a sandhi ln] .. Thus, Enghsh
nafogar ['navD-lga:r], literally 'nave-lance,' 1\llddle
gar has been l'cplaced by auger; Old English ['ne:drc] glves Middle
naddere and addere, whence modern adder; Old French
naperon borrowed as napron, has been replaced by apron.
After' this loss of final [nJ, another sound-change Ied to
of certain final vowels, through which many hitherto medlal [n] ,s
got into final position, as in oxena > oxen. These final ln] s
came into final position tao late ta suffer the droppmg; hencc the
language had now, beside the sandhi ln], which appeared be-
fore also a stable final ln]. This Icd ta sorne complIcated
relations:
ANALOGIC CHANGE
/Ost [l'] in final position and before consonants, but kept it
vowels, there resulted sandhi-alternants of words like water: in.
final position and bcfore consonants this became ['w<lta], but he.
fore a vowel in a close-knit phrase it kept its [r]: the water is ['W<ltar :
iz], the water of ['w<ltar ov]. The final vowel of water was now like
that of a ward like idea [aj'dija], whieh had never had final [r].
This led to a new-formation:
waler ['w<lta] : the VJGi,er i8 ['w<ltar iz]
=idea [aj'dija] : x,
which resulted in the sandhi-form the idea-r is [aj 'dijar
In a language Iike modern English, which gives special phonetic '(..,"
\"
treatment to thc beginning and end of a word, the phonemes in
these potions rarely fuIftl the terms of an ordinary conditioned
sound-ehanl'!:e, but are subject rather to eonditioned changes of their
own. Only phrases with atonie words parallel the conditions which
exist within a ward. Hen English sandhi-alternation is limited
largely to cases like the above (... of, ... is) or ta sueh as dem't,
at yrm did you ['dijuw]. Moreover, the plain phonetic mark
ing of most words, and in sorne positions even of ordinarily atonie
words, favors the survival or new-formation of variants that agroo
with the absolute form: do not, al you ['et juw], did you ['did juw].
In languages whieh give a less specialized treatment ta ward
boundaries, sandhi-alternants arise in great numbers and give rise
to irregularities \vhieh are in turn levelcd out by new-formations.
We saw in 21.4 the origin of the initial-sandhi of Irish. In French,
the noun is on the whole free from sandhi-altcrnation: words like
pot [po] 'pot' or pied [pjc] 'foot' are invariable in the phrase.
However, we need only look ta phrase-like compounds ( 14.2),
such as pot-au-feu [p:Jt 0 f!"l] 'pot-on-the-fire,' that is 'broth,'
or pied--terre [pjet a te:r] 'foot-on-ground,' that is 'Iodgings,'
ta see that the apparent stability js due ta analogie reglliarization.
Third-person singular verbs which were monosyllabic in the carly
Middle Ages, have, by reglllar phonetic development, a final
[t] in sandhi before a vowel: Latin est> French est [el 'is,' but
Latin est ille > French est-il [et iJ 'is he?' On the other hand,
verb-forms of more than one syllable had not this [t]; Latin amaf
'he loves' gives French aime rEm] 'loves' even before a vowel.
However, the pat.trn
418
420
ANALOGIC CHANGE AN ALOGIC CHANGE 421
This complicated habit was re-shaped into our present distribution
of singular ox, plural oxen.
In most cases, a phrasai innovation results not in a new word.
farm, but in a new syntactic or lexical usage, sllch as the use of
as a ( 23.2). In German we find such apposi.
tIonal groups as em Trunk Wasser [ajn 'truDk 'vaser] 'a drink of
water,' where the related languages would lead Ils ta expect the
second noun in genitive case-form, Wassers 'of water.' The geni-
tive case-ending in feminine and plural nouns has been reduced
to zero by phonetic change: the genitive of 1'1,hlch [milx] 'milk'
(feminine noun) is homonymous with the nominative and accusa,..
tive. The old locution n J'runk TVassers has been replaced by
the present one, which arose on the scheme
Jfih:h trinken 'to drink milk' ein Trunk Milch 'a drink
of mlk'
= l'Vasser trinken 'to drink water' x.
was fitvored, no doubt, by the existence of nouns whose geni.
trve wavered between zero and -es, and by the cireumstance that
the p;enitive case was declinin.e; in frequcney. It seems likely, in
spite of the obvious difficu!ties, that furiher research will find
many cxamples of analogie innovation in the phrase, bath syo-
tac tic and lexicaL Our philosophie prepossessions have led us too
often to seek the motives of change in the individual ward and in
the meaning of the individual word.
23. 8. For many new-formations wc are not able to give a pro-
portional model. \Ye believe thut this is not alwuys duc to our io-
to find the model sets, and that therc s rcally a type of
Imguistic chanp;e which resembles analogie change, but goes on
without modd sets. These adaptive new-formations an
old form with some change in the direction of semantieully rclated
forms. For instance, of the two slang forms aelorine 'actress' and
chorine 'chorus-p;irl,' only the former can be described as the rEr
suit of a proportional analogy (Paul: Pauhne = actor : x). Now,
chorine seeHlS to be based in sorne way on adorine, but the set
chorus: chorine is not parallcl with actor : actorine cither in fonn
0: in. meuning. The set Josephus : Josephine [jow'sijfos,
fiJn] lS uncommon, remote in meaning, and phonctically irregular.
We can say only that many nouns have a suffix [-ijn], e.g. chlo.
rine, colleen; that this sufiix derivcs sorne women's numcs and
especially the noun actorine; and that the -us of chorus is plainly
suffixal, in view of the adjective choral. This general background
lllust have sufficed to make someone uUer the form chorine, even
tbough there was no exact analogy for this form.
A new form (such as chorine), which is based on a traditional
farm (chorus, chorus-girl), but departs from it in the direction of
a. series of semantically related forms (ehlorine, colleen, Pauhne,
etc., including especially actorine) , is said ta originaLe by adap-
tation. Adaptation seems to he favored by morc than one factor,
but ail the factors taken together would not allow us ta predict
the new form. Gften, as in our example, the new form has a face-
tions connotation; this connotation & probably connected with
the nnpredictable, far-fetched shape of the new ward. This is true
of mock-Iearned words, lke scrumptious, rambunctious, absquat
ulale. It scems unlikely that more than one speaker hit upon
tbese forms: we suspect them of bcing individual creations, de-
termined by the linguistie and pradical peeuliarities of sorne one
speaker. They must have agreed ta sorne cxtent, however, with
the general habits of the community, since they were taken up
by other speakers.
Sorne adaptations arc less far-fetched and merely produce a
new form which agrees better with semantically related forms.
English has borrowcd many French words with a suffix -ure,
sl1ch as cen8ure, fracture. The Old French words plaisir,
loisir, tresor, whieh contain other suffixes, have in Enl';lish heen
adapted to the -ure type, for the [-zr] of pleasure, leisure, treasure
reflects an old [-zju:r]. Among our foreign-lcarned words, egoism
follows the French model, but egotism is an adaptive formation
in the direction of despotism, nepolism.
In the Romance languages, Latin reddere ['reddere] 'to give
back' has heen largely replaced by a type *rendere, as in l talian
rendere ['rmdere], French rendre [randr], ,,,hence English render.
This *rendere is an adaptation of reddere in the direction of the
series Latin prehendere [pre'hendere, 'prendereJ 'ta take' > Italian
prendere ['prendere], French prendre [prandrJ; Latin attendere
[at'tendere] 'ta pay attention' > Italian atlendere [at'tmdereJ
'to wait,' French attendre [atandrJ (and other compounds of Latin
tendere); Latin vendere ['we:ndere] 'ta sell' > Italian vendere
['vendere], French vendre [va"dr]; here the word for 'take,' with its
close kinship of was doubUess the main factor.
422
ANALOGIC CHANGE ANALOGIC CHANGE 423
Sometimes it is a single form which exereises the attraction.
Beside the old ward gravis' heavy,' later Latin has also a fonu
grevis, whose vowel seems ta be due ta the infiucnce of levis 'light
(in weight).' Formations of this sort are known as blendings or
contaminations. 'Ve cannat always be sure that the attraction was
exercised by only a sin!!;!c form; in our examplc, the word brevis
'short' may have helped taward the formation of grevis.
The para-digm of the word for 'foot,' Primitive Indo-European
*[po:ds], genitive *[po'dos], Sanskrit [pa:t], gcnitive [pa'dah], ap.
pears in one ancient Greek dialcet in the expected shape, ['po:s],
genitive [po'dos], but in the AtUc dialect has the unexpected
nominative form ['pows]; this has been explained as a contamina.
tian with the ward for' (Doth,' [o'dows], genitive ['dontos], whieh
is a normal reflex of a Primitive Indo-European type
*[o'donts].
In the earlier stages of the Germanie languages, the personal
prOlJouns must have been in a state of instability. The old form
for' ye' seems to have been a Primitive Germanie type *[ju:z, juz],
which appears in Gothie as jus [ju:s] or [jus]. The other Germanic
dialect.s l'efieet a Primitive Germanie type *[j iz]: Old Norse [e:r],
Old English [je:], Old High German [ir]. This form has been ex-
plained as a contamination of *[juz] 'ye' with the ward for 'we,'
Primitive Germanie *[wi:z, wiz], refiected in Gothic [wi:s], Old
Narse [ve:r], Old English [we:], Old High German [wir].
Similarly, in Gothic the accusative case of' thou' is [Huk] and the
dative case [llus]. These forrns disagrce with the other dialects,
which refieet the Primitive Germanie types accusat,ive *['Hik],
Old Narse [llik], Old English [Oek], Old High German [dih], and
dativc *[Oiz], Old Narse [He:r], Old [Ile:], Old Righ German
[dir]. The Gothie forms have been explained as contaminations
with t.he nominative *[Ou:], Gothic, Old Narse, Old English
[llu;], Old IIigh German [du:]. For this, the ward 'l,' wbich had
the same vowe! in aIl three forms, Gothie [ik, mik, mis], may have
sen'ed as a kind of model, but thC1'e is no exact analop;y covering the
two paradigrns, and we might cqually well expect [mik, mis] ta work
in favor of *[llik, !lis].
NumeraIs seem to have been contaminated in the history of
vunous languages. In Primitive Indo-European, 'four' was
*[kwc' two:res], and 'five' *[' penkWc]; wi tness Sanskrt [ca' tva.:rah,
'panca] or Lithuanian [ketu'ri, pcn'ki]. In the Germanic languages
both words begin with [f], whieh refiects a Primitive
[pl, as in English four, five; and fire, moreover, has an [f] for t.he
[k"'l of the second syllablc, as in Gothie [fimf]. In Latin, on the
other hand, both words begin with [kw]: quattuor, quinque ['kwat-
tuor, 'kwi:nkwe]. Ail of these deviant forms could he explained
as due ta "distant assimilation"; it seems more probable, how-
ever, that the changes deseribed under this and similar terms
( 21.10) arc in reality contaminative or adaptivc. Ancient
Greek [hep'ta] 'seven' and [ok'to:] 'eight' led in one dialect to
a contamnative [op'to:] 'eight,' and in others ta [hok'to:]. The
words 'nine' and 'ten,' Primitive Indo-European *['newJ;l,
'dekrp], as in Sanskrit ['nava, 'daa], Latin nouem, decem, bath
have initial [dl in Siavic and Baltic, as in Old Bulgarian [deventI,
desentrJ.
Psychologists have ascertained that under laboratory conditions,
the stimulus of hcaring a word like 'four' often leads to the ut ter-
ance of a ward like 'five' - but this, arter aH, does not account for
contamination. There is perhaps more relevance in the fact that
contaminative "slips of the tangue" arc not infrequent, e.g. "l'Il
just grun (go plus run) over and get it."
Innovations in syntax sometimes have a contaminative aspect.
The type 1 am with him has been explained as due ta
contamination of 1 am friendly wi/h him and we are friends.
rcgulari ties such as the" attraction" of relative pronouns ( 15.11)
scem ta be of this nature.
So-callcd popular etymologies ( 23.6) are largely adaptivc and
contaminative. An irregular or semantically obscure form is re-
placed by a new farm of more normal struct.ure and sorne seman-
tic content - though the latkr is often far-fetched. Thus, an old
sham-fast 'shame-fast,' that s, 'modest,' has given way to the
rcgular, but semantically quccr compound shame-faced. Old
English sam-blind, containing un otherwise obsolete first rnember
which meant 'haU,' was replaccd by the Elizabethan mnd-blind.
Old English bryd-guma 'bride-man' was rcplae<;d
by bride-groom, thanks ta the obsolescence of guma 'man.' Foreign
words are especiaIJy subject to this kind of adaptation. Old
French Middle English crevise has bn rcplaced by
fish, craw-fish: mandragora by man-drakc; asparagus in older sub-
standard speech by sparrow-grass. Our gooseberr!f secms to be
a replacement of an oldcl' *groze-berry, ta judge by dinlect forms
424 ANALOGIC CHANGE
such as grozet, groser; thcse forms refiect a borrowcd French fonu
akin to modern French groseille [gr;:lze:j] 'currant; gooseberry.'
Probably forms like our symbolic words, nuTSCry words, and
short-names are created on general formai patterns, rather than on
exact analogie mdels. It seems, however, that forms like Bob, Dick
existed as common nouns, perhaps with symbolic connotation,
before they were specialized as hypochoristic forms of Robert,
Richard. It is a great mistake to think that one can account for
the origin of forms like these by merely stating their connotation.
In sorne instances we know that a certain persan invented a
fOln. The most famous instance is gas, invented in the seventecnth
century by the Dutch chemist van Helmont. In the passage
where he introdl1ces the word, Van Helmont points out its rcsem.
blanee to the word chaos, which, in Dutch pronunciation, is not
far removed (thoul';h phonemically quite distinct) from gas. More-
over, van Helmont \lfled also a technical term blas, a regular deriva-
tivc, in Dutch, of the verb blazen 'ta b10w.'
It is evident that in such cases we cannot reconstruct the in-
ventor's private and personal world of connotations; wc can only
guess at the general linguistic background. Charles Dodgson
(" I.Rwis Carroll "') in his famous poem, "The Jabberwocky" (in
l'h
r
ough the Looking-Glass) , uses a numller of new-formations of
this sort and, later in the book, explains the connotative signiti-
canee they had for him. At least one of them, chortte, has come
into wide use, l\/lore rcccnt examples are the mercantile term
kadak, invented by George Eastman, and blurb, a creation of
Gelett Burgess.
i ,
CHAPTER 24
SEMANTIC CHANGE
24. 1. Innovations whieh change the lexical meaning rather
than the grammatical function of a form, are classed as change of
rneaning or semantic change.
The contexts and phrasai combinations of a form in our older
writtcn records often show that it once had a difTerent meaning.
The King James translation of the Bible (1611) says, of the herbs
and trees (Cenes 1,29) ta you they shall befor meat, Similarly, the
Old English translation in this passage used the ward mete. 'Vc
infer that the word meat used ta mean 'food,' and we may assure
ourselves of this b,)' looking inta the foreign texts from which these
English translations were made. Sometimes the ancients tell us
meanings outright, chiefly in the fonu of glosses; thus, an Old
English-glossary uses the won1 mde ta translate the Latin Gibus,
which wc kno\V ta mean 'food.'
In other instances the comparison of related languages shows
different lIleanings in forms that we feel justitied in viewing as
cognatc. Thus, chin agrees in meaning with German Kinn and
Duteh kin, but Gothic kinnus and the Scandinavian forms, from
oId Korse kinn to the present, mean 'cheek.' In other Inder
European languages we tind Greek ['genus] 'chin' agrceing with
West Germanic, but Latin (Jena 'cheek' agrceing \Vith Gothie and
Seandinavian, \Vhile Sanskrit ['hanuh] 'jaw' shows us a third
meaning. \\Te concluclc that the olel meaning, whatever it was, has
changed in some or ail of thesc languages.
A third, but much less certain indication of semantic change,
appears in the structural analysis of forms. Thus, understand
bad in Old English timc the same meaning as now, but since the
ward is a compound of stand and under, wc infer that at the time
the compound was first formed (as, an analof!;ic new-formation)
it must have meant 'stand under' j this gains in probability frorn
the fact that under once rneant also 'among,' for the cognates,
German un/er and Latin inter, have this meaning. Thus, I under-
stand these things may have meant, at first, '1 stand among these
425
things.' In other cases, a fonn whos structure in the present
state of the language ?oes not imply anything as ta meaning,
may have been semantIcally analyzable in an earlier stage. The
ward ready has the adjective-fonning suffix -y added ta a unique
root, but the Old English form [je're:de], which, but for an ana-
logie re-fonnation of the suffix, can be viewed as the ancestor of
Teady, meant 'swift, fmted, skilled' and was a derivative of the
!.'ri:dan] 'ta past tense [ra:d] 'rode,' derived DOUil [ra:d]
a ndmg, a road. We infer that wheD [je're:de] was first formed
it meant 'suitable or prcpared for riding.' '
Inferences like these are sometimes wrong, because the make-up
of a form may he of later date than its meaning. Thus, crawfish
and gooseberry,. adaptations of crevise and *groze-berry ( 23.8),
cau tell us nothlllg about any older meanings.
24. 2. We can easily see today that a change in the meaning
of a speech-form is merely the result of a change in the use of it
and other, semantieally related speech-fonns. Earlier students
however, went at this problem as if the speech-fonn were a rola:
tively permanent object to which the meaning was attached as
a kind of changeable satellite. They hoped by studying the suc-
cessive meanngs of a single fonn, such as meat 'food' > 'fleshM
food,' to find the reason for this change. This led them ta classify
semantic changes according ta the logical relations that connect
the successive meanings. They set up such classes as the fol-
lowing:
Narrowing:
Old English mete 'food' > meat 'edible flesh'
Old English deor 'beast' > deer 'wild ruminant of a particular
species'
Old English hund 'dog' > hound 'hunting-dog of a particular
breed'
Widening:
Middle English bridde 'young birdling' > biTd
Middle English dogge 'dog of a particular (ancient) breed' >
dog
Latin viTtus 'quality of a l;nan (vir) , manliness' > French vertu
(> English virlue) 'good quali ty'
Metaphor:
Primitive Gcnnanc *['bitraz] 'biting' (derivaUve of *['bi:to:]
'1 bite ') > biller 'harsh of taste'
Metcmymy - the meanings are near each other in space or
time:
Old English cew;e 'jaw' > cheek
Old French joue' check' > jaw
Synecdoche - the meanings are related as whole and part:
Primitive Gennanie *['tu:naz] 'fence' (80 still German Zaun) >
town
pre-English *[' stobo:] 'heated room' (compare German Stube,
formerly 'heated room,' now 'living-room') > stove
Hyperbole - from strongcr ta weaker rneaning:
pre-French *ex-Wndre 'ta strike with thunder' > French tonner
'ta astanish' (from Old French, English borrowed astound, MM
tonish)
Litotes - from weaker to stronger meaning:
pre-English *['kwalljanl 'ta tonnent' (so still Gennan qwlen) >
Old English ewellan 'to kill'
Degeneration:
Old English cnafa 'boy, servant' > knave
Elevation:
Old English cniht 'boy, servant' (compare Gennan Knecht
'servant') > knight.
Collections of examples arranged in classes like these are use-
fuI in showng us what changes are likely ta occur. The meanings
'jaw,' 'cheek,' and' chin,' which we found in the cognates of our
word chin, arc found ta fiuctuate in other cases, such as that of
check from 'j aw' (0Id Englsh l)leaning) to the present meaning;
jaw, from French joue' cheek,' has changed in the opposite direc-
tion. Latin maxilla 'j aw' has shifted ta 'cheek' in most modern
dialects, as in Italian mascella [ma'sella] 'cheek.' Wc suspect
that the word chin may have meant 'jaw' before it meant 'check'
and' chin.' In this case wc have the confirmation of a few Old
High German glosses which translate Latin molae and maxillae
(plural forms in the sense 'jaw' or 'jaws ') by the plural kinnc.
Old English ['weorllan] 'ta become' and its cognates in the other
Germanie languages (8Och as German werden, 22.2) agree in
forrn with Sanskrit ['vartate:] 'he turns,' Latin verta 'I tum,'
Old Bulgarian [vrte:ti] 'ta tum,' Lithuanian [ver' cu] 'I tum';
we accept this etymology because the Sanskrit word has a
ginal meaning to becorne, , and bccauge English furn shows a
parallel development, as in turn sour, turn traitor.
426 SEMANTIC CHANGE
SEMANTIC CHANGE
427
24, 3. Yiewed on this plane, a change of meaning may impl
a between practical things and thereby throw light y
the hfe of aider times. English fee is the modern form of th on
d' f Id E 1'. e para
19m a , ng l8h feoh, whtch meant 'live-stock, cattle, property,
money" Among the Germanie cagnates, only Gothie faihu ['fehu]
pr?perty'; aU the others, such as German Vieh [fi;] or
[fc;], have like '(head of) cattle, (head of)
hve-stock. The same IS true of the cognates in the other Iudo.
Eur?pcan languag:s, such as Sanskrit ['pau] or Latin pecu; but
has the dcnved words pecunia 'money' and peculium 'sav-
This confirms our holief that live-stock served
III anmcnt tunes as a medium of exchange.
Eng;ish hase corresponds forrnally ta Dutch hoos [ho;sJ, German
Hose [ho:ze], but these words, usually in plural form mcan not
'sto 'v 'b t 't ' T . ' . c mgs u rousers. . hc Scandmavian forms, such as
oId Narse hosa, mcan 'stocking' or 'legging.' An ancient form
presurnably West Gcrmanic, came into Lr;tin in the early
of our cra, doubt1ess through the mediatian of Roman saldicrs
thc languages have a type *hosa (as, It.aIian uos;
[w;)saj) III the scnse '1egging.' 'Ve eonclude that in old Germanie
our. ward me:mt a cmcring for the leg, either including the foot or
endmg at the ankle. Round his waist a man wore anothcr gar-
the br.eeches (Old English broc). The English l'tnd Scandi-
naVIan indicates no change, but the German dcvelop-
seems ta mdlCate that on the Continent the hose were later
Jomed at the top into a frouser-like garment,
In this way, a semant.ically pccuIiar etymology and cultural
traces may confirm each other. The German ward Wand [vant]
denotes the wall of a roorn, but not a thick masonry wall, t.he
latter is Mauer ['rnawer], a loan from Latin. The German
saunds like a derivative of the verb /0 wind, German winden (past
tense wand), but etymolo!!;ists were at loss as ta the of
thesc meanings, until .l\lcringer showed that the derivative noun
applied at first to wattled walls, which were made of
tWlsted wlthos covered with mud. In the same way Primitive
in Gothie waddjus, Old Narse veggr,
Old Enghsh wag, IS now taken to have originated as a derivativc
of a verb that. meant 'wind, twist.' Wc have seen that scholars
t.ry, by a of scmantic and archaeolop;ic data, to throw
hght on prehlstoric conditions, such as thosc of the Primitive Indo-
European parent community ( 18.14). The maxim "Words an.d
Things" has been used as the title of a journal devoted to thls
aspect of etymology. , . . .
Just as formaI features may arIse from h1ghly specIfie and :r
ar1
-
bl
factors ( 23.8) so the meaning of a form may be due to sltua.-
a e' l 'f h' t . 1
t
, s that wc cannat reconstruct and can know on Y l 1S OrIca
iOn . ['k' ]' '
tradition is kind to us. The German J(mse:r . aJzer
and the Russian [tsar] are offshoots, by borrowmg, of the Latm
caes
ar
['kajsar], which was generalized the n.ame. of a par-
ticular Roman, Gaius Julius Caesar. ThIS name lS ta be a
derivative of the verb caed '1 eut'; the man ta 1t w.as first
given was barn by the aid of the surgical operatI.o
n
whlCh,. on
account of this same tradition, is caUed the cacsanan operatIOn.
Asidc from this tradition, if wc had not the historical knowledge
about Caesar and the Roman Empire, wc could Dot gucss that
the ward for 'emperor' had begun as a family-name. The now ob-
solesccnt verb burke 'suppress' (as, to burke opposiiion) was
derived from the name of one Burke, a murderer in Edinburgh
who smothered his victims. Th" ward panda cames from the
name of Pandarus; in Chaucer's version of the ancient story of
Troilus and Cressida, Pandarus acts as a go-between. Buncombe
comes from the name of a county in South Carolina, thanks to
the aoUcs of a congrcssman. Tawdry cames from Si. Audrey;
at St. Audrey's fair one bought tawdry lace, Terms like landau
and sedan come from the original place of manufacture. The ward
dollar is borrowed ultimately from German Taler, short for Joa-
chimstaler derived from Joachimsial ('Joachim's Dale'), a place
in BohcU:ia where silver was mintod in the sixteenth centur:y.
The Roman mint was in the temple of Juno M onCia 'Juno the
Warner' . honcc the Romans uscd the ward moneia bath for' mint'
and for: coin, money.' English mint is a pre-Enf?lish
from this Latin ward, and Eng1ish money is a mediCval borrowIllg
from the Old French continuation of the Latin ward.
The surface study of semantic change indicates that refined and
abstract meanings largely grow out of more concrete meanings.
Meanings of the type 'rcspond accuratc1y to (thlngs or speech)'
develop again and again from meanings like 'he near to' or 'get
hold aL' Thu8, undersiand, as we saw, seems to have meant 'stane
close to' or 'stand among.' German versiehen [fer'ste:en] 'under
stand' seems to have meant 'stand round' or 'stand bcfore'; thti
428 SEMANTIC CHANGE
SEMANTIC CHANGE
429
430
SEMANTIC CHANGE
SEMANTIC CHANGE 431
Old English equivalent f orstandan appears both for 'understand'
and for 'proteet, defend.' Ancient Greek [e'pistamaj] '1 under.
stand' is literally '1 stand upon, , and Sanskrit [ava'gachatiJ is
both 'he goes down into' and 'he understands.' Italian capire
[ka'pire] 'ta understand' is an analogie new-formation based on
Latin capere 'to seize, grasp.' Latin comprehendere 'to under.
stand' means also 'to take hold oL' The Siavic word for 'under-
stand, as in Russian [po'nat], is a compound of an old verb that
t " t k ' A' ,
mean S81ze, a e. margmal mcaning of underst:md' appears in
Our words grasp, catch on, get (as in [ don't get that). :\Iost of our ab-
stract vocabulary consists of borrowings from Latin, through French
or in gallicized form; the Latin originals can largely he traccd to
eoncretc meanings. Thus Latin defin'ire 'tQ define' is literally 'to
set bounds ta' (finis 'end, boundary'). Our elimz'nate has in Latin
only the 'concretc meaning 'put out of the house,' in accordance
with its derivative character, since Latin i!/'iminare is structural1y a
synthetic compound of ex 'out of, out from' and limen 'threshold.'
24.4. Ali this, aside from its extra-linguistic interest, gives us
some m:asure of by which we can judge of etymologic
compansons, but lt does not tell us how the meaning of a linguistic
form can change in the course of time. When wc find a form uscd
at one time i.n a meaning A and at a later time in a meaning B,
what We s.ee lS evidently the result of at !cast two shifts, namely,
an expanSIOn of the form from use in situations of type A to use in
of a wider type A.B, and then a partial obsolescence by
WhlCh the form ceases ta be used in situations which approximate
the old type A, sa that finally the form is used only in situat.ions
of type B. In ordinary cases, the first process involves the obsoles-
cence Or restriction of some rival form that gets crowded out of
use in the and the second process involves the en-
some rival form into the A-situations. We ean sym-
bohze thlS dIagramrnatically as follows:
meaning;
'nourish-
'edible
'edible
'muscular
ment'
thing'
part of
part of
animal
animal
body' body'
tirst stage:
food
meat
flesh fiesh
second stage;
food
meat &-----+ meal flesh
third stage: food
'&---+ food
meat fiesh
In the normal case, therefore, we have ta deal here with fluc-
tuations of frequency likc those of analogie change; the diffcrence
is only that the fluctuations result in lexical instead of
cal displacements, and therefore largdy elude the grasp of the
Iinguist. The first student, probably, ta see that semantic change
consists of expansion and obsolescence, was Hermann Paul. Paul
saw that the meaning of a form in the habit of any speaker, is
merely the result of the utterances in whieh he has beard it. Sorne-
tirnes, to he sure, we use a form in situations that fairly well caver
its range of meaning, as in a definition (" a town is a large settle-
ment of people") or in a very general statement (" vertebrate
animais have a head"). In such cases a form appears in its general
meaning. Ordinarily, however, a form in any one utterance rep-
resents a far more specifie practieal feature. When we say that
John Smith bumped his head, the ward head is used of one
ular man's head. When a speaker in the neighborhood of a city
says l'm going to town, the ward town means this partieular city.
In such cases the form appears in an occasional meaning. In eat
an apple a day the word apple has its general meaning; in some one
utteranee of the phrase cat this apple, the ward apple has an oc
casional meaning: the apple, Ict us say, is a large baked apple.
Ali marginal mcanings are occasional, for - as Paul showed-
marginal meaninp;s differ from central meanings precisely by the
fact that we respond to a marginal meaning only when sorne spe-
.cial circumstance makes the central meaning impossible ( 9.8).
Central meanings are occasional whenever the situation diffcrs
from the ideal situation that matches the whole cxtent of a form's
meaning.
Aecordingly, if a speaker has heard a form only in an occasional
meaning or in a series of oceasional meanings, he will utter the
form only in similar situations; his habit may difler from that of
other speakers. The ward meat was used of a1l manner of dishes;
there must have come a time when, owing ta the encroaehment of
sorne other ward (say, food or dish), many speakers had heard the
word mea! only (or very predominantly) in situations where the
actual dish in question consisted of fiesh; in their own utterances
these speakers, aceordingly, used the word meat only when flesh-
food was involved. If a speaker has heard a form only in sorne
marginal meaning, he will use this forrn with this same meaning
a central meaning - that is, he will use the form for a meaning lU
432
SEMANTIC CHANGE SEMANTIC CHANGE
433
which other speakers use it only under very special conditions_
Iike the city child who eoncluded that pigs were very properl
i
.
caUed pigs, on account of their unclean habits. In the later
Ages, the German ward Kopf, cognate with English cup, had the
central meaning 'cup, bowl, pot' and the marginal meaning 'head"
there must have come a time when many speakers had heard
ward only in its marginal meaning, for in modern German Kopf
means only 'head.'
24. 6. Paul's explanation of semantic change takes for granted
the occurrence of marginal meanings and of obsolescence, and
views thcsc processes as adventurcs of individual speech-forms
without reference to the rival forms which, in the one case, yiek
ground ta the form under consideration and in the other ca=
" ...., ...... ,
encroach upon its domain. This view, neverthc1ess, represents a
great advance over the mere classification of diffcrences of mean-
ing. In particular, it enablcd Paul ta show in dctai! some of the
ways in which obsolescence breaks up a unitary dornain of meaning
- a process which he called isolation.
Thus, beside the present central meaning of the ward meat
w.e have today the strange marginal (apparently,
wldencd) uses III meat and drink and in sweetmeals; for dishes other
than 6esh, the ward meat went out of use, except in these two
expressions, which are detached from what is now the central
meaning of the ward: wc may say that thesc two expressions have
been isolaled by the invasion of the intermediate sernantic domain
which is DOW covered by food, dish. In the same way, knave
been shifted from 'boy, young man, servant' ta 'scoundrel ' but
the card-player's use of knave as a Dame for the lowest of the'three
picture-cards (' jack') is an isolated remnant of the older meaning.
The ward charge is a loan from Old French charger which meant
originally 'to load a wagon.' Hs present multiplicity of meanings
is evidently due ta expansion into marginal spheres followed by
obsolescence of intermcdiate rneanings. Thus, the agent-noun
charger is no longer used for 'load-bearer, beast of burden,' but
only in the special sense 'war-horse'; the mcaning charge 'make a
swift attack (on)' is a back-formation from charger 'war-horse.'
The ward board had in Old English apparently the same central
meaning as today, 'fiat piece of wood,' and, in addition ta this,
several specialized meanings. One of these, 'shield,' has died out
entirely. Another, 'side of a ship,' has Ied to sorne isolated forms,
snch as on board, aboard, 10 board (a ship), these have. bccn
tended ta use in connection with other vehICles, such as 1'away
:1's. A third marginal meaning, 'table,' survives, in ele-
vated turns of speech, such as feslive board. Bcfore s general
obsolescence, howevc1', board 'table' underwent further trans-
ference ta ' regular meals,' which is still current, as !fi bed and
board and lodging, la board (al a boarding-house), und so on. flus
use of board is sa widely isolated today from board' plank' that we
should perhaps spcak of the two as hornonymous wards.
In Old Germanie the adjective *['bajlaz] meant 'unharrned, weIl,
prosperous,' as heil still does in German; this meaning remains in
our verb ta heal. In modern Ellglish wc have only a transferred
meaning in whole. DCl'ived from *['hajlaz] therc was another
jective *['hajlagn,z] which meant 'conducive to we.lbre, or
prasperity.' This ward seems ta have been uscd III ft rellglous or
superstitions sense. It Gecurs in a Gothie inscription in runes,
as Bishop L1fila did not use it in his Bible, we may suspect that lt
had helLthen associations. In the other Gcrmanic lanp;uages it
appears, from the bcp;inning of our records, only as an equivalent
of Latin sanctus 'holy.' Thus, the semantic connection between
whole and holy has been completely wiped out in English; cven in
Grman heil 'unharmed, prospcrous' and heilig 'holy' lie on the
betwecn distant semantic connection and mere ho-
monymy of roo!s.
The Old English adjective lward 'hard' underlay two adverbs,
hearde and hea-rrihce; the fornwr survives in Hs olt! relation, as
hard but the latter hardl1 has been isobted in the remotely
, ,.,
transferrcd meuning of 'barcly, scarcely,' through loss of inter-
mediate meanings as' only \Vith difficulty.'
Isolation may be furlhcred by thc obsolescence of sorne con-
struction. 'Ve flnd it hard to connect the rneaning of undersland
with the meanings of under and Ilot only becausc the mean-
ing 'stand close to ' or ' st:md among,' which must have been central
at the time the compound was formed, has been obsolete since
prchistoric time, but also because the construction of the com-
pound, preposition plus verb, with stress on the latter, hn,s died out
except for traditional forms, which survive as irregularitics, such
as undertake, undergo,underlie, overlhrow, ottercome,
give, forget, forbid. The '.'lords slraw (Old English strerlw) and t?
strew (Old English sirewian) wcre in prehistoric time
cally :onneetedj the Primitive Germanie types are *['strawwan] ,
strewmg, that strewn,' and *['strawjo:] '1 strew.' At that tim:
(Old English streawberige) 'strewnberry' must have
descnbed the strawberry-plant as it lies along the ground'
straw became spocialized to 'dried stalk, dried stalks' and'
h 1
. . ' e
morp 0 ogle conn:ctIOn with strew disappeared, the prior member
of strawberry was lsolatcd, with a deviant meaning, as a homonym
of straw.
change may prompt or aid isolation. A clear case of
thls lS ready, whi:h has diverged tao far from ride and road; other
examples are hohday and holy, sorry and sore dear and dearlh and
especially, with old umlaut ( 21.7) whole heal, dole and'deal.
word lord (Old English hlaford) was at the time of its forma-
tIOn 'loaf-ward,' doubtless in a sense like 'bread-giver" lad
(Old hldfdige) seems to have been 'bread-shaper:' Th:
was formerly 'lack of case, un.ease'; in the present
meaning 'sickness' it is ail the better isolated from
dzs-- and ease through the deviant form of the prefix, with [z] for
[sI after unstressed vowel ( 21.4).
Another contributory factor is the intrusion of analogie new-
formations. Usually these overrun the central meaning and leave
only some marginal meanings to the old form. Thus slolh' laziness'
was originally the quality-noun of slow, just as trulh is still that of
tr:ue, but the deeline of the th derivation of quality-nouns and the
:lS0 of slowness, formed by the now regular -ness derivation, has
Isolated slolh. An Old English compound *hswf 'houscwife'
through various phonetic changes rcached a form which survives
today o?ly in a transferred meaning as hussy ['hozij] 'rude, pert
woman. central meaning it was replaced by an analogic
new composItion of has and wf. This, in its turn, through phonetic
change a form hussif [hozef] which survives, though now
obsolescent, the transferred meaning 'sewing-bag,' but has becn
crowdc? ID the c:ntral by a still newer compounding,
housewife [haw8w,waJfj. In medleval German, sorne adjectives with
an umlaut vowel had derivative adverbs without umlaut: schoene
['sJll:ne] 'beautiful,' but schone ['S:ne] 'beautifully" feste 'firm' but
faste 'firmly.' ln the modern period, these have becn
out by regularly formed adverbs, homonymous with the
adJective: today schOn [j:S:n] is both 'beautiful' and as an adverb
'beautifully,' andfcst both 'firm, vigorous' and 'firmy, vigorously,:
but. the old adverbs have survived in remotely marginal uses,
schan 'already' and 'never fear,' and fast 'almost.'
Finally, wc may he able to recogni.ze a change in the practical
wo
rld
aS a factor in isolaUon. Thus, the isolation of German Wand
'wall' from winde:n 'to wind' is duc to the disuse of wattled waHs.
Latin penna' feather' ( > Id French penne) was borrowed in
Dlltch and in English as a designation of the pen for writing. In
French plume [plym] and German Feder ['fe:der], the vernacular
word for' feat.her' is used a1so for' pen.' The disuse of the goose-
quiIl pen has isolated these meanings.
24.6. Paul's explanation of semantic change docs not account
for the rise of marginal meanings and for the obsolescence of
forms in a part of their semantic domain. The same is true of so-
called psychologicai explanations, such as Wundt's, which mercly
paraphrase the outcome of the change. Wundt defines the cen-
tral meaning as the dominant element of meaning, and shows how
the dominant clement may shift when a form occurs in new typ-
ical contexts. ThUR, when meal had been heard predominantly
in situations where flesh-food was concerned, the dominant element
becarne for more and more speakers, not 'food' but
This statement. leaves the matter exact.Iy where it was.
The obsolescence which plays a part in many semant.ic changes,
need not present aoy characteristics other than those of ordinary
loss of frequenc}'; what Uttle we know of fluctuations in this di-
rection (Chapter 22) will apply here. The expansion of a fOfln
into new meanings, however, is a special case of rise in frequency,
and a very difIicnlt one, since, strictly speaking, almost any utter-
ance of a form il" prompted by a novel situation, and the degree
of noveIty is Dot subject to precise measurcment. Older students
accepted the rise of marginal meanings wthout seeking specifie
factors. Probably they took for granted the particul
ar
trans-
ferences which had occurrcd in languages familiar to them (fool
of a mountain, neck of a bottte, and the like, 9.8). Actually,
languages difIer in thiR respect, and it s precisely the sprcad of
a farm into a new meaning that concerns us in the study of semantic
change.
The shift into a new meaning is intelligible when it merely
reproduces a shift in the practical world. A form like ship or hat
or hose dcsignates a shifting series of objects bccausc of changes
in the practical world. If cattle were used as a medium of exchange,
434 SEMANTIC CHANGE
SEMANTIC CHANGE
435
the ward fee 'cattJo' would naturally he Hsed in the mea .
, 'd'f' nIng
money, an 1 one,vrote wlth a goose-feather, the word for 'feather'
would naturally be used of this writing-implemcnt. At this point
however, has been no shift in the lexical structure of
languago. Thls cornes only whon a learned loan-word pen is d'
tinct from fcather, or when fee on the one hand is no longer
of cattlo and, on the other hand, loses ground in the domain f
'money' until it retaillA only the specialized value of 'suro of
money paid for a service or privilege! a
The only.typo of somllntic expansion that is rdatively weil
understood, lS what we may cali tho accidentai type: sorne formaI
change:- an[1logie re-flhaping, or borrowing-
results 10 a 10cutwn which coincides with sorne old form of Dot
tao rernotc meaninp;. Thus, Primitive Germanie *['awzo:] de
noted the 'car' of a persan or animal; it appears as Gothie ['awso'J
ld Old GC1:man ora ( > modem Dutch oor [o:r:
Enghsh. [ e:are], and \S cognate \vith Latin auris, Old
nan [uxo], III same meaning. Primitive Germanie *['ahuz]
denoted the gram of a plant with the husk on it; it appears in
Q.hs, Old Korse ax, ld Connan ah and, with an analogie
nommatlve form duo ta oblique case-forms, Old Genmm ahir
( > aQr [n:r]) , Old English ['fhher] and ['e:arJ,
and \s cogmtte wIth Latin acus 'hnsk of g-min, chaff.' The 10ss of
rh] and of unstressed yowels in English has made the two fm'fis
phonetically alike, and, since the mellllings hrtVe sorne resem-
blance, ear of grain has become a marginal (tmnsferrcd) meaning
of car of an animai. Since ld English lwe:od] 'weed' and
'garment' have coincided through ;:;ound-change, the sur-
vlvmg use of the latter, in widow' s weeds, is now a marginal meaning
?f the Of course, the degree of ncarness of the meanings
\S subJeet. ta precise measurcment; the lexicographer or
tonan who knows the origin;:; will insist on describing Ruch forms
as pairs of homonyms. Nevert.hcless, for many speakers, doubt-
less, co:n ,on the foot r?presents merely a marginal rneaning of
corn gram. The latter 18 a continuation of an old native ward;
a borrowing from Old French corn ( < Latin cornu
horn, cognam \Vith English horn). In French, allure is an ab-
straet nouu derivcd from aller' to wa.Ik, ta go,' and means 'manner
of carriage,' and in a specialized mcaning 'good manner
of walkmg, good carriagc.' In English we have borrowed this a[...
. sinee it coincidcs formally with the verb to allure (a loan from
lure, . 'h 'It b
old French aleurer), we use it in the m?am
ng
c arm. ,may e
that let in let or hindrance and a let ban IS for sorne a queer
marginal use of let 'permit,' and that even the Ellzabethan let
'hin
der
' ( 22.4) had this value; wc have nO standard for anSwer-
ing such questions.
phonetic discrepancies in such cases may be removed by new-
formation. Thus, the Scandinavian loan-word benn 'equipped,
dy' would give a modern English *[baw
n
]. This form was
and in meaning sO close to the reflex of Id English
bunden, past participle of bindan 'to bind,' (> bound
Ibawnd], past partcipk of bind) , that a new-formatIOn bound
[bawnd] replaced it; the of [-dl was favored by
a babit of sandhi. The resu1t IS that bound 1Il phrases. as
bound for England, bound to sec it figures as a margmal mealling
of the past participle bound. 130th the word law and its compound
are loan-words from Scandinavian. The first member of
the latter was Olel Narse lhy:r] 'manor, town' - wit,ness the
oldcr English forms bir-law, bur-law - but the r<rshaping
turned it into a marginal use of the preposition and adverb by.
Beside the central meaning please 'ta give pleasure or satis-
faction,' wc have the marginal meaning , be willing' in if you please.
This phrase meant in 1\liddle English 'if it pieuses you! The
solescence of the use of finite verbs without actors, and of the
ponement of the finite verb in clauses, the near-obsolescence of
the subjunctive (if it please you), and the analogie loss of
distinction (nominative ye : dativ<raccusativo you), have left if
you please as an I1ctor-acton clause with you as the actor a.nd
anomalous marginal use of please. The same factors, actmg III
phrases of the type if you like, seem to have led ta a complete
turn-about in the meaning of the verb like, which used to mean
'suit, pleas
e
,' e.g. Old English [he: me: 'wcl 'li:kaO] 'he pIeuses
me well, 1 like him.'
Partial obsolescence of a form may leave a queer marginal
iog. Ta the examples already given (c.g. meat, we may a?d
a few where this feature hus kd to further Shlfts. The
French loan-word favoT had formerly in English two woll-separated
meanings. The more original one, 'kindly attitude, inclination,'
with its offshoot 'kiudly action,' is still central; the other, 'cust
of countenanee/ is in general obsolete, but l:lurvives as a marginal
436 SEMANTIC CHANGE
SEMANTIC CHANGE
437
meaning in ill-favored 'ugly.' In the aphoristic sentence Ki8sing
goes by favor, our word had formerly this marginal value (that is
'one prefers ta kiss good-Iooking people'), but now has the central
value (' is a matter of inclination '), Similarly, prove, proof had
a central meaning 'test' which survives in the aphorism The
proof of the pudding is in the eating; this was the meaning also in
The exception proves the rule, but now that prove, proof have been
shifted to the meaning '(give) conclusive evidence (for),' the
latter phrase has become a paradox.
The old Indo-European and Germanie negative adverb '"[ne]
'not' has left a trace in words like no, not, never, which reflect
old phrasai combinations, but has been supplanted in indcpendent
use. Hs loss in the various Germanic languages was due partly
to sound-change and led to sorne peculiar semantic situations,
In Norse it left a trace in a form which, owing to its original phrasaI
make-up, was not negative: '"[ne 'wajt ek hwerr] 'not know l who,'
that is, 'I don't know who,' resulted, by phonetic change, in Old
Narse ['n0kurr, 'nckkwer] 'someone, anyone.' In other phonetic
surroundings, in prc-Norse, '"[oc] was entirely lost. Sorne fonns
which were habitually used with the negatioo must have got in
this way two opposite meanings: thus, an *['ajnan] 'once' and a
'"[ne 'ajnan] 'not once, not' must have led to the same phooctic
result. Actually, in Old Norse, various such expressions have
survived in the negative value: *[ne 'ajnan] givcs Old Narse a
'not'; '"[ne 'ajnato:n] 'not one thing' gives Old Norse at 'not';
*[ne 'ajnaz gel 'not even one' gives Old Norse einge 'no one';
*[ne 'ajnato:n gel 'not even one thing' gives etke, ekke 'nothing';
*[ne 'ajwan gel 'not at any time' gives eige 'not'; '"[ne 'mannz gel
'not even a man' gives mannge 'nobody.' In German, where ne
has been replaced by nicht [nixt], originally 'not a whit,' the double
meanings due to its 10ss in sorne phonetic surroundings, still appear
in our records. At the end of the Middle Ages we fiod clauses of
exception ('unless , .. ') with a subjunctive verb formed both
with and without the adverb ne, en, n in apparently the same
meaning:
with ne: ez en mac mih nieman troesten, si en tua z 'there may
no one console me, unless she do it'
without ne: nieman kan hie fraude finden, si zerge 'no one cao
find joy here, that does Dot vanish.'
The first example here is reasonablc; the second contains a
hirosical use of thc subjunctive that owes its existence only ta
phonetic disappeurance of ne in similar cont.exts. \Ve, observe
, our examples also a plus-or-minus of ne, en III the malll clause
Ill
I
ng \Vith nieman 'nobody.' This, too, left an ambiguous type:
aO dh' 't ' t
bath an old dehein 'any' and an old ne e ,no any mus,
have led, in certain phonetic contexts, to dehem any; not any,
Both these meanings of dehein appenr in. .older texts, ,weil
as a ne de/win' not any'; of the three posslblhtles, only dehem Dot
any' (> kein) survives in modern German:
In French, certain words that arc wldely used wlth a verb and
the ncgative adverb, have also a negative meaning when used
ithout a verb. Thus, pas [pa] 'step' Latin passum) has the
w , d' '(" Il '1
two uses in je ne vm:s pas [za TI ve pa] l on t go ongma y
go not a step') and in pas mal [pa mal] 'not badly, not 80 bad,';
personne [pfTs:m] 'person' Latin personam) appears, also
je ne vois personne [Z<l n vwu persJn] 'I don't see anyone, and lU
P
ersonne 'nobody" rien [rjen] Latin rem 'a thing') has lost or
, , nJ
dinary noun values, and occurs in je ne vois rien [za 0 vwa rJf
'1 don't see anything' and in rien 'nothing,' This development
has becn described as contagion or condensation, Tt can be bctter
understood if we suppose that, during the medieval period of
high stress and vowel-weakening, French ne Latin non) was
phonetically lost in certain eontexts. .'
The reverse of this process is a loss of content. Latm forms like
canto '1 cantas 'thou-singcst,' cantat 'he-shc-it-sings ' (ta
which more specifie mention of an actor was added by cross-
rcfcrenee, 12.9), appear in French as chante(s) [sant]
uscd only with an act.or, or, rarely, in completive speech, Just
like an English verb-form. This loss of the pronominal.actor.
meaning is evidently the result of an analogie change WhlCh re
placed the type cantat 'he-sings' by a type ille cantat 'that-one
sings' (> French il chante [i sant] 'he sings'). This latter change
has been explained, in the case of French, as a result of the ho-
monymy, due to flound-change, of the various. Lati?
however, in English and in German, forms like 8'lng, stn{lest, smgeth
have come to demand an actor, alihough there i3 no homonymy.
24.7. Special factors like these will account for only a
proportion of the walth of marginal mcanings that faces us III
every language. Tt remained for a modern scholar, H. Sperber,
t<> point out that of meaning are by no means ta he
438 SEMANTIC CHANGE SEMANTIC CHA?>rGE
439
440 SEMANTIC CHANGE
SEMANTIC CHANGE 441
taken for granted, and that the first step toward understanding
thorn must be to find, if we can, the context in which the new meao-
ing first appears. This will always be difficult, because it demands
that the student observe very closely the meanings of the fonn
in ail older occurrences; it is especially hard to mako sure of nega-
tive features, such as the absence, up to a cortain date, of a cer-
tain shade of meaning. In most cases, moreover, the attempt is
bound to fail because the records do not contain tho criticai
tions. Nevertheless, Sperber succeeded in finding the critical
text for the extension of older German kopf 'cup, howl, pot' to
the meaning 'head': the new value tirst appears in our texts at
thc end of the Middle Ages, in battle-scenes, where the matter
is one of srnnshing someone's head. An English example of the
same sort .; the extension of bede 'prayer' to the presont mean-
ing of bead: the extension is known to have occurred in conncction
with the use of tho rosary, where one counted one' s bedes (originally
'prayers,' then 'Iittlc sphcres On a string ').
In the ordinary case of scmantic extension we must look for a
context in which our form can be applied ta both the old and the
new meanings. The obsolescence of other contexts - in Our ex-
amples, of German kopf applied to oarthen vessels and of bead
'prayer' - will then leave the new value as an unambiguous
central meaning. The reason for the extension, however, is another
matter. We still ask why the medieval German poet should speak
of a warrior smashing his enemy's 'how\' or 'pot,' or the pious
Englishman of counting 'prayers' rathor than 'pearls.' Sperber
supposes that intense emotion (that is, a powerful stimulus) leads
ta such transfcrences. Strong stimuli lcnd to the favoring of novel
speech-forms at the cost of forms that .have been heard in indif-
ferent contexts ( 22.8), but this general tendency cannot account
for the rise of specific marginal meanings.
The methodical error which has held back this phase of our work,
is Our habit of putting the question in non-linguistic terms - in
terms of meaning and not of form. When we say that the word
meat has changed from the meaning 'food' to the meaning 'edible
flesh, , we are merely stating the practicai result of a linguistic
procoss. In situations where both words were applicable, the ward
meat was favored at the cost of the wordjlesh, and, on the model of
such cases, it came to he uscd alsa in situations where formerly the
word flesh aione would have boon applicable. In the same way,
words like food and dish encroached upon the word meat. This
second displacement may have resulted from the fir8t because the
ambiguity of meat 'food' and meat was troublesome in
practical kitchen life. We may some day tind out why jlesh was
disfavored in culinary situations.
Once we put the question into theoo terms, we see that a normal
extension of meaning i8 the same process as an extension of
matical function. "\Vhen meat, for whatever reason, was heing
favored, and flesh, for whatever reason, was on the decline, there
must have occurred proportional extensions of the pattern ( 23.2):
leave the bones and bring the jlesh : leave the b01WS and bring the meat
= give us bread and flesh : x,
resulting in a new phrase, give us bread and meat. The forms at the
left, containing the word jlesh, must have borne an unfavorable
connotation which was absent from the forms at the right, with the
word meat.
A semantic changc, then, is a complex process. It involves favor-
ings and disfavorin!;s, and, as its crucial point, the extension of a
favored form into practical applications which hitherto helonged to
the disfavored form. This crucial extension can he observed only
if we succeed in finding the locutions in which it was made, and in
finding or reconstructing the modellocutions in which both forros
were used alternatively. Our records give us only an infinitesimal
fraction of what was spoken, and this fraction consists nearly
always of elevated speech, which avoids new locutions. In Sperher's
example of German kopf 'pot' > 'head,' we know the context
(head-smashing in battle) where the innovation was made; there
remains the problem of finding the model. One might surmise, for
instance, that the innovation was made by Gcrmans who, from
warfare and chivalry, were familiar \Vith the Romance speaker's
use of the type of Latin testam, testum 'potsherd, pot' > 'head,'
which in French and Italian has crowded the type of Latin caput
'head' out of aIl but transferred meanings. We confront this
plex problem in aIl semantic changes except the fortuitous ones like
English let, bound, ear, which are due to sorne phonetic accident.
We can best understand the shift in modern cases, whcro the
connotative values and the practical background are known.
During the !ast gonerations the growth of citics has led to a lively
trado in city lots and houses, "development" of outlying land iuto
residenco districts, and speculative building. At the same time, the
prestige of the persans who live by these things has risen to the
point where styles pass from them ta the working man, who in
language is imitative but has the force of numbers, and to the
"educated" persan, who enjoys a fictitious leadership. Now, the
speculative builder has learned to appeal ta every weakness, in.
cluding the sentimentality, of the prospective buyerj he uses the
speech-fonus whose content will tum the hearer in the right direc-
tion. In many locutions house is the colorless, and home the
mental word:
Smith has a lovely house : Smith has a lovely home
= a lovdy new eight-room house : x.
Thus, the salesman cornes ta use the ward home of an empty
shell that has never been inhabited, and the rest of us copy his
style. It may be too, that, the ward house, especial1y in the sub-
standard sphere of the salesman, suffeTS from sorne ambiguity, on
account of meanings such as 'commercial establishment' (a
reliable house), 'hotel,' 'brothel,' 'audience' (a halfempty house).
The lcarned word transpire in its use, meant 'to
breathe or ooze (Latin spIrlire) through (Latin trans),' and thus, as
in French transpirer [transpire], 'ta exhale, exude, perspire, ooze
out,' and with a transfer of rneaning, 'ta becorne public (of news).'
The oId usage would be t() say of what real/y happened, very liUle
transpired. The ambiguous case is transpired that the president
was out of town. On the pattern
SEMANTIC CHANGE
443
SEMANTIC CHANGE
The (Jods approve
The depth and not the tumul/ of the soul,
he waS only continuing the metaphoric use current in such ex-
pressions as deep, ruffled, or stormy feelings. By making a new
transference on the model of theoo old ones, he revived the "pic-
turc." The picturesque saying that "language is a book of faded
metaphors" is the reverse of the truth, for poetry is rather a
blazoned book of language.
Dot ulways he able h) trace its origin. lt may have arisen under
.$Om
e
very special practical circumstances that are unknown ta us,
or what cornes to the same thing, it rnay be thc successful coinage
of' sorne one speaker and owe its shape to his individual circum
stances. One suspects that the queer slang use, a quarter of a
century ago, of twenty-three for' get out' arose in a chance situation
of sportsmanship, gambling, crime, or sorne other rakish environ
ment; within this sphere, it rnay have started as sorne one pcrson's
witticism. Sinee every practcal situation is in reality unprece-
dented, the apt response of a good speaker may always border on
semantic innovation. Both the wit and the poet oflen cross this
border, and their innovations may become popular. To a large
extent, however, these personal innovations are modeled on current
forills. Poetie metaphor is largely an outgrowth of the transferred
uses of ordinary speech. Ta quote a very weil chosen example, when
Wordsworth wrote
SENTrMENT.l.L,
PLE.l.8.1.NT CONNOT.l.TrON COLORLE88
442
COLORLEB8 ELEG.l.NT-LF,.I.Rl'<ED
happened that the president was it transpired that the
out of town president . . .
= what happened, rernains a secret x,
we now get the forrnerly impossible type what transpired, remains
a secret, where transpire figures as an elegant synonym of happen,
occur.
This parallelism of transference accounts for successive en-
croachrncnts in a semantic sphere. As saon as sorne form like
terribly, which means 'in a way that arouses fear,' has been ex-
tended into use as a stronger synonym of very, the road is clear for
a similar transference of words Iike awfully, frightfully, horribly.
Even whcu the birth of the marginal meaning is recent, we shall
CULTUH.AL BORROVnNG 445
CHAPTER 25
CULTURAL BORROWING
25.1. The child who is learning to speak may get most of his
habits from sorne one person - say, his mother - but he will aIso
hear other speakers and take sorne of his habit.s from them. Even
the basic vocabulary and the grammatical features which he ae-
quires at this time do not reproduee exactly the habits of any one
aIder person. Throughout his life, the speaker continues ta adopt
fentures from his feliows, and these adoptions, though Jess funda-
mental, are very eopious and come from al! manner of sources.
Sorne of them are incidents in large-scale levelings that affect the
whole community.
Accordingly, the comparatist or historian, if he could discount aIl
analogic-semantic changes, should still expcct to find the phonetic
correlations disturbed by the transfer of speech-forms from person
ta person or from group to group. The actuaI tradition, could we
trace it, of the various features in the language of any one speaker,
rons back through entirely diverse persons and communities. The
historian can recognize this in cases of formaI discrepancy. He sees,
for instance, that forms which in older English containcd a short
[a] in certain phonetic surroundings, appcar in Central-\Vestern
American English as [el in man, hat, bath, gather, lalher, etc. This
represents the basic tradition, even though the individual forms
may have had very different adventures. Accordingly, when the
speaker uses an [a] for the same old phoncme in the word father and
in the more elegant variant of the word ralher. the historian infers
that somewhere along the line of transmission thcS forms must
have come in from speakers of a differcnt habit. The adoption of
features which differ from those of the main tradition, is lir1{juistic
borrowir1{j.
Within the sphere of borrowing, we distinguish between dia,.-
lect borrowing, where the borrowed features come from within the
same speech-area (as, father, rather with [a] in an [e]-dialect), and
cultural borrowir1{j, where the borrowed features corno from a dif-
ferent language. This distinction cannot always he carried out,
444
since there is no absolute distinction ta be made diaJect
boundaries and language boundaries ( 3.8). In this chapter and
the next '',le shaH speak of borrowing from foreign languag<.ls, and
in Chapter 21 of borrowing betwccn the dialects of an area.
/25.2. Every speech-community learns from its neighboro/ Ob-
jects, bath natural and manufactured, pass from one community
ta thc other; and so do patterns of action, such as technical pro-
cedures, warlike practices, religious rites, or fashions of individ-
ual conduct. This spread o(things and habits tS studied by eth-
llolo?;ists, who call it cultural dijJuSiof. One can plot on a map
the diffusion of a cultural feature, such as, say, the growing of
maize in pre-Columbian North America. In general, the areas
of diffusion of different cultural features do not coincide. Along
with objects or practices, the speech-forms by which these are
named often pass from people to people. For inst.ance, an
lish-speaker, either bilingual or with sorne foreign knowledge of
French, introducing a French article to his countrymen, will
ignate it by Hs French name, as: rouge [ru:z], jabot [zabo], chauf-
feur [sof:r], garage [gara:z], camouflage [kamufla:z]. In most in-
stances we cannot ascertain the moment of actuaI innovation:
the speaker himself probably could not be sure whether he had
ever before heard or used the foreign form in bis native language;
Several speakers may independcntly, none having heard the
others, make the same introduction. In theory, of course, we must
distinguish between this actual introduction and the ensuing rep-
etitions by the same and other speakers; the n8W form embarks
upon a careet of fluctuation in frequency. The historian finds,
however, that sorne of the lateradventures of the borrowed form
are due to its foreign character.
IIf the original introducer or a later user has good command of
the foreign language, he may speak the foreign form in foreign
phonetics, even in its native context. More often, however, he
will save himself a twofold muscular adjustment, replacing sorne
of the foreign speech-movements by speech-movement.s of the
native language; for example, in an English sentence he will speak
his French rouge wit.h an English [r] in placc of the French uvular
trili, and an English [uw] in place of the French tense, non-diph-
thongal lu:]. This phonetic substitution will vary in degree for dif-
ferent speakers and 00 differeot occasions/speakers who have
Dot learned ta produce French phonemes arc certain ta make it.
446
CULTURAL BORROWING CULTURAL BORROWING 447
The historian will class it as a type of adaptation ( 23.8), in which
the foreign form is al,red ta meet the fundamental phonetic
habits of the language.
1In phonetic substitutian the speakers replace t.he foreign SOllnds
by the phonemes of t.heir language. In sa far as the phonetie sys-
tems paralIel, this involves only the ignoring of minaI' differ-
ences/ Thus, we replace the various [l'] and [1] t.ypes of European
languages by our [rJ and [IJ, t.he French unaspirated stops by our
aspirated, the French postdcntals by our gingivals (as, say, in
and long vowels by our diphthonp;al types [ij, uw, ej,
owJ.jWhen t.he phonetic systems are less alike, the substit.utions
may secm surprising ta members of the lending community. Thus,
the older l\lcnomini speakers, who knew no English, reproduced
automobile as [atamo:pen]: Menomini has only one, unvoiced
series of stops, and no lateral or trill. Tagalog, having no [f]-type,
replaced Spanish [f] by [pJ, as in [pi'jesta] from Spanish fiesta
[' fjesta] 'celebration.;/
In the case of ancient speech, phonetc substit.utions may in-
form us as ta the acoustie relation between the phonemes of t.wo
languages. The Latin name of the Greek nation, Oraeci ['grajki:],
later ['gre:ki:], was borrowed, early in the Chri8tian era, into the
Germanie languages, and appears here wit.h an initial [k], as in
Gothic krkos, Old English crcas, Old High German kriahha
'Greeks.' Evidently the Latin voiced stop [g] was acoustically
closer t the Germanie unvoiced stop [k] than ta the Germanie
phoneme which we tran8cribe as tg], say, in Old English grne
'green'; presumably, at the time the old ward for 'Greek' was bor-
rowed, this Germanie [g] was a spirant. Latin [w] at this carly
time was reproduced by Germanie [w], as in Latin vinum ['wi:-
num] 'wine' > Old English win [w:in], and similarly in Gothie
and in German. In the early Middle Ages, the Latin [w] changed
to a voiced spirant of the type [v]; accordingly, this Latin phoneme
in loan-words of the missionary period, from the seventh century
on, was no longer reproduced by Germanie [w], but by Germanie
[f]. Thus, Latin versus ['versus] 'verse,' from oIder ['wersus], appears
in Old English and in Old High German as fers. A third stage ap-
pcars in modern time: German, having changed its old [w] to a
spirant type, and English, having in another way aequired a pha-
neme of th;) [v]-typc, now give a fairly accurate reproduction of
Latin [v], as in French vision [vizjonJ (from Latin visionem [w:-
si'o:nem]) > German [vi'zjo:n], English ['ViZI,l].l In Bohemian,
where every word is stressed on the tirst syllable, this accentu-
ation is given to foreign words, such as ['akvarijumJ 'aquarium,'
['konstelatse] 'constellation,' [' sofe:r1'chauffeur.'
25. 3. If the borrowing people is relatively familial' with the
lending language, or if the bOITowed words are fairly numerous,
then foreign sounds which are acoustically remote from any na-
tive phoneme, may be preserved in a more or less accurate render-
mg that violates the native phonetic system. In this respect, there
are many local and social differences. Thus, the French nasal-
ized vowels are very widely kept in English, even by people who
do oot spcak French, as in French salon [salon] > English [sa'lo
n
,
'selon], French rendez-vous [rande-vuJ > English ['randevuwJ, French
enveloppe [aUv(a)bp] > English envelope ['aUvelowp]. Sorne speak-
ers, howcver, substitute vowel plus [D], as in ['faIJdevuwJ, and
others vowel plus ln], as in ['randevuwJ. The Germans do the like;
the Swedcs always replace French nasalized vowcls by vowel
plus [IJ]. In sorne forms English does not reproduce the nasal-
ized vowel, as in French chiffon [sifonJ > English ['sifan], and in
the more urbane variant ['envlowp] envelope.
This adoption of foreign sounds may bccome quite fixed. In
EngIish the cJuster [skJ is due ta Scandinavian loan-words; the
[skJ of Old English had changed in later Old English time ta [8],
as in Old English [sko:h] > modern shae. This Scandinavian
ciuster occurs not only in bOITowed words, such as sky, skin,
skirt (beside native shirt) , but also in new-formations, sueh as
scatter, scrawl, scream; it has become an integral part of the pho-
netic system. The initiaIs [v-, Z-, J-l came into English in French
words, such as very, zest, just; a11 threc are quite at home now, and
the last two occur in new-formations, such as zip, zoom, jab, jounce.
Thus, the phonetic system has been permanently altered by bor-
rowing.
/ Where phonetic substitution has oecurred, increased famili-
arity with the foreign language may lead to a newer, more cor-
rect version of a foreign form. Thus, the Menomini who knows
a little English no lnger says [atamo:pen] 'automobile,' but [ata-
mo:piJ], and the modern Tagalog speaker says [fi'jestaJ 'celebra-
tion.' /The old fonn of the borrowing may survive, however, in
1 The discrcpanciee in this and similar cxamples are due (.0 changes which the
vllrious languages have made sinee the time of borrowing.
448 CULTURAL BORROWING
CULTURAL BORROWING 449
special uses, sueh as derivatives: thus, even the modern Tagalog
speaker says [kapijes'ta:han] 'day of a festival,' where the prefix,
suffix, and accentuation are native, and in English the derived
verb is always envelope [en'velop], with vowel plus [n] in the first
syIlable.
A similar adjustment may take place, at a longer interval of
time, if the bOITowing language has developed a new phoneme
that does better justice to the foreign form. Thus, English Greek,
German Grieche ['gri:xc] embody corrections made after these
languages had devcloped a voiced stop [g]. Similarly, English
verse is a revision of the oldfers; German has stuck to the old form
Vers [fe:rs]. In revisions of this sort, especiaIly where literary
terrns arc concerned, learned persans may exert some influence:
thus, the replacement of the older form with [kr-] by the later
form Greek \Vas surely due ta educated people.
1For the most part, however, the influence of literate persons
works also against a faithful rendcring. In the first place, the lit-
erate persan who knows nothing of the foreign language but has
seen the writtcn notation of the forcign form, interprets the latter
in tcrms of native orthography./ Thus, French forms like pme,
ruche, menu, Vicior Hugo [pys, rys, rnany, ygo] would doubt-
less be reproduced in English with [ij] for French [y], were it not
for the spelling with the letter u, which leads the literate
speaker to pronounce [(j)uw], as in [pjuws, ruws ,'menjuw, 'viktr
'hjuwgow]. Spanish Mexico, older ['mesiko], modern ['mexiko],
has [ks] in English because of li terate people's interpretation of
the symbol x; similarly, the older English renderiog of Don Quix-
ole (Spanish [don ki'xote]) is [dan 'kwiksat]. The latter has been
revised, certainly under learned influence, to [dan ki'howtij], but
the older version has been retained in the English derivative
quixotic [kwik'satik]. We reproduce initial [ts] in tsar or tse-tse-fly,
but not in German forms like Zeitgeist ['tsajt-Igajst] > English
['zajtgajst], or Zwieback ['tsvi:bak] > English ['zvdjbak], where
the letter z sug-gests only [z]. Even where there is no phonetic diffi-
culty, as in German Daehshund ['daks-1hunt}, Wagner ['va:gner],
Wiener ['vi:ner], the spelling leads to such reproductions as ['des-
Ihawnd, 'wegnr, 'wijnr, 'wijnij].
This l'dation is further complcatcd by literate persons who
know sornething of the foreign pronunciation and orthography.
A speaker who knows the spelling jabot and the English form
['zebow] (for French [zabo]), may rcvise ['tej tel tejtJ
(from French [te:t a kt]) to a hyper-foreign ['tejtetej], without the
final ft]. The literate person who knows parlez-vous franais?
['parlej 'vuw 'fransej?l (for French [parle vu franse?]), may decide to
join the Alliance Franaise [ali'jaOs 'fransejl, although the French-
man here has a final [z]: [alj (lnS fraose:z].
25. 4. The borrowed word, u.side from foreign sounds, often
violates the phonetic pattern. Thus, a German initial [ts], even
aside From the orthography, may be troublesorne to many English-
speakers. Generally1adaptation of the phonetic pattern takes
place togethcr with fLdaptation of morphologie structure. Thus,
the final [il of garage, which violates the English pattern, is re-
placed by [j] and the accent shifted in the form ['garell, which
conforms to the suffixal type of cabbage, baggage, image. Like-
wise, beside chauffeur [sow'fejr] with normal phonetic substitution,
we have a more fully adapted ['sowfr). /
The description of a language will thus recognize a layer of
eign forms, such as salon [sa'10], rouge [ruwz], garage [ga'rai],
which deviate from the normal phonetics. In sorne languages a
descriptive anaIyss will recognizc, further, a layer of semi-foreign
forms, which have been adapted up to a conventional point, but
retain certain conventionally determined foreign characteristics.
The foreign-Iearncd vocabulary of English is of this type. Thus,
a French prciosit [presbsite] was anglicized only to the point
where it became preciosity [pre' sj ostij, pre' il (j) ositij]; the un-
stressed prefix, the suffix -ity (with presuffixal stress), and the for-
mally and semantically peculiar relation to precious ['presos], do not
lead to further adaptation. The English-speakers (a minority)
who use the word at aIl, include it in a set of habits that deviates
from the structure of our commonest words. This secondary layer
of speech-habit owes its existence, historically, to old waves of
borrowing, which will concern us in the sequet.
/When the adaptation is completed, as in ehair (anciently
rowed from Old French) or in ['sowfr] chau:tfeur, the foreign origin
of the form has disappearcd, and nether the speaker nor, conse-
quently, an honest description can distinguish it from native
fOrIlls. The historian, however, who is concerned with origins,
will class it as a loan-form. Thus, chair and ['sowfrl chauffeur" in
the present state of the language, arc ordinary English wordsi:JUt
the historian, taking the past into view, thcm as loan-words.
450 CULTURAL BORROWING
CULTURAL BORROWING
451
At aU stages, the assimilation of foreign words presents man
problems. The phenomena of t.he type of phonetic
( 21.10), as in French marbre> English marble, are fairly fre-
quent. \Ve probably have t.o reckon here with highly variabl
factors, including adaptations based on the habits of
speakers. Both during thc progress taward the status of a loan-
form, and after this status has been reached, the structure is
likely to he unintelligible. The languages and, within a language,
the groups of speakers that are familiar with foreign and semi-
foreign forms, will tolerate this state of affairs; in other cases
a further adaptation, in the sense of popurar etymology, ma;
render the form structurally or lcxically more intelligible, as in
> *groze-berry > gooseberry; asparagus> sparrow-grass;
cremse > crayfish > crawfish ( 23.8). The classical instance is the
replacement, in medieval German, of Old French arba/este 'cross-
bow' by an adaptive new-formation Armbnt ['arm-,brust], lit-
craIly 'arm-breast.'
i The borrowed form is subject to the phonetic changes that
occur aiter its adoption./ This factor is distinct from phonetic
substitution and other adaptive changes. Thus, wc must suppose
that an Old French form like vis2on [vi'zjo:n] (reflecting a Latin
[wi:si'o:nemJ) was taken int.o medieval English wit.h sorne slight
amount of no longer traceable phonetic substitution, and that
it rise ta a successful adaptive variant, with stress on the
first syllable. The further changes, however, which lcd ta the
modern English ['viZr,I] are merely the phonetic changes which
have occuITed in English since the time when this word was bor-
rowcd. These two factors, however, cannot. aIways be distin-
guished. After a number of borrowings, there arase a fairly regular
relation of adapted English forms t' French originals; a new
borrowing from French could he adapted on the mode! of the
aIder loans. Thus, the discrcpancy between French prciosit
[presiosite] and English precio/y [pre'sjasitij, pre'l'ijasitij] is not
due to that occuITed in English aHer the time of
borrowing, but merely reflects a usual relation between French
and English types - a relation which has set up in the English-
speakers who know French a habit. of adapting forms along cert.ain
lines.
26.5. Where we can allow for this adaptive fact.or, the phonetie
development of borrowed forms often shows us the phonetic form
at the time of borrowing accordingly t.he approxUn.ate date
f various sound-changesI The name of Caesar appears m Greck
a spelling (with the let.ters k, a, i) which for earlier .tirne we
jnterpret as ["kajsar] and for later as ['kc:sar], and II. a?pears lll.
a similar spelling in Gothie, where the value of dlgrap.h ut
is uncertain and the form may have been, accordmgly, elther
['kajsar] or ['ke:sar]. forms t.hat at the time of
thcs
e
borrowings, Latin still spoke an mlal [k] and had not yet
gone far in the direction of modern forms like Halian cesare ['e-
zare] ( 21.5). In West Germanie, the foreign ward appears as
Old High German keisur, Old Saxon kesur, Old English casere,
this last representing presumably somet.hing like ['ka:sc:re}. These
forms eonfirrn the Latin [k]-pronunciation; moreover, they guar-
anWe a Latin diphthong of the type [aj] for the first syllable, since
the correspondence of southern German ei, northern Ie:], and Eng-
lish [a:} is the ordinary reflex of a Primitive Germanie diphthong,
as in *['stajnaz] 'stone' > Old High German stein, Old Saxon
[ste:n], Old English [sta:n]. Thus, for the time of the early CO?-
tact of Rome with Germanic peoples, we are assured of [kaJ-]
as the value of the first syllable of Latin caesar. On t.he ot.her hand,
the West Germanic forms show us that the various changes of
the dipht.hong [aj], in Old Saxon to le:] and in Old English to [a:1
occurred after the early contact wit.h the Romans. The vowel of
the second syllable, and the addition of a third syllable in
English, are surely due to some kind of an adaptation; the Enghsh
form, especially, suggests that the Roman ward was .taken up
as though it were *[kaj'so:rius] > pre-English *['kajso:fJaz]. The
ward was bOITowed from a Germanie language, doubtless from
Gothie, by the Slavs; it appears in Old Bulgarian as [t.se:sanj.
Now, in pre-Slavic time, as we know from the correspondences
of native words, [aj] was monophthongized ta [e :l, and then a [k]
before such an le:] changed to [tsl. Thus, Primitive Indo-European
*[kWoj'na:] 'penalty,' Avestan [kacna:], Greek [polne:]
in Old Bulgarian as [tse:nal 'price.' The Slavic borrowmg, ac-
cordingly, in spit.e of its actual deviation, confirms our reconstruc-
tion of the old Germanie form, and, in addition ta t.his, enablcs
us ta date the pre-Slavic changes of [ka] ta [tse:] aft.er the time
of early borrowing from Germanie, which, history tells us,
from round 250 to round 450 A.D. 11oreovcr, the second and thlrd
syllab1es of the Slavic form show the SUffie adaptation as the Old
English, ta a Germanie type *['kajso:rjaz]; we may conclude that
this adapted form existed also among the Goths, although 0
Gothie Bible, representing a more learned stratum of speech hUr
the correetly Latin kaisar. ' as
. Latin (via) road' appears in Old Saxon as ['stra:ta],
10 High German as [stra:ssa], and in Old English as [stre:t].
We lOfer that this term, like caesar, was borrowed before the emi.
?,ation of the E?glish.. The correspondcnce of German [a:] Eng_
hsh [e]: refleets, III native words, a Primitive Germanie [e:J as in
*['de:dizJ 'deed,' Gothie [ga-'de:Os], Old Saxon [da:d], Old' High
German [ta:t], Old English [de:dJ; aceordingly we eonclude that
at the Ume when Latin slrata was borrowed, West Germanie
speaker.s had alrcady made the change from le:] ta [a:], since they
used thls vowcl-phoneme ta reproduce the Latin [a:J. On the other
hand, the Anglo-Frisian change of this [a:] taward a front vowel
Old English [e:], must be later thao the borrowing of the \vorcl
street; this is confirrned by the Old Frisian form (of much later
documentation, to be sure), narnely strele. The medial [t] of the
shows us that, at the time of borrowing, Latin
stIll sald [stra:ta] and not yet ['strada] (Italian strada). This
contrasts with later borrowings, such as Old High German ['si:dal
'silk,' ['kri:da] 'chalk,' which have [dJ in accordanee with later
Latin prom:nciation ['se:da, 'kre:daJ from earlier Latin ['se:ta,
'kre:ta] ( 21.4). Finally, the [ssJ of the Righ German form shows
us that the South-German shift of Germanie medial ft] to affricate
and sibilant types ( 19.8) occurred after the adoption of the Latin
strIta. In the same way, Latin ['te:gula] 'We' appears in Otd Eng-
Hsh as ['ti:gol] (whenee the modern tile), but in Old High German
as ['tsiagal] (whence modern German Ziegel ['tsi:geIJ): the borrow-
ing oeeurred before the South-German eonsonant-shift, and t.his
is the case with a whole series of borrowings in the sphere of use-
fuI abjects and techniques. In contrast with this, Latin words in the
literary and scientific domains, which were burrowed presumably
in the missionary pcriod, from the seventh ccntury onward, came
too late for the South-German Latin templum
'temple' appears in Old High German as ['tempal], Latin tinda
'colored stuff, ink' as [' tinktaJ, and Latin tegula was borrowed
over again as Otd High German ['tegal] 'pot, retort' (> modern
German Tiegel ['ti:gelJ). The Same re-borrowing of this 1ast
ward appears in Old English ['tijele]; but here we have no striking
to distinguish the two chronological layers of
borrowing.
The South-German change of [t] to affricaLe and sibilant types
showS us, in faet, a rcmarkable instance of dating by means of
borrowed forms. A Primitive Germanie type *['mo:to:J is rep-
rescnted by the Gothie ward ['mo:ta] which translates the Greek
words for 'tax' and for 'toU-station' (e.g. in Romans 13, 7 and
Matthew 9, 9-10); there is also a derivative ['mo:ta:ri:s] 'tax-
gatherer, publean.' The Old English engnute [mo:t] oc.curs on.ce,
in the meaning 'tribute money' (Matthew 22,19); the :\1iddle Hlgh
German ['muosse} 'miller's fee' shows us the regul:1r High German
shift of [tJ to a sibilant and an equally regular shift of [0:] to [uoJ.
Now, in the southeastern part of the German area we tind also
an Old High German [' lnu: ta] 'toB' (> modern },{aut) and the
place-name ['mu:ta:run] (litcrally, 'at the toll-takers") of a town
on the Danube (> modern Mautern). These forms not only laek
the shift of [t] but also have an altogether unparalleled [u:] in place
of Germanie [o:J. Wc have reason to believe t.hat Gothie [0:] was
close to [u:J and in later time perhaps coincided with it. History
tells us that in the tirst half of the sixt.h eentury, Theodoric the
Great, the Gothie emperor of Haly, extended his rule to the
Danube. 'Ye eonclude that the German word is a borrowing from
Got.hic, and, aeeordingly, that at the Ume of borrowing, Primitive
Germanie [t] in Bavarian German had already changed toward
a sibilant: the [t] of the Gothie word was reproduced by the Ger-
man reflex of Primitive Germanie [d], as in Old High German
[h1u:t] 'Ioud' (> modern /aut) from Primitive Germanie *['hlu:-
dazJ; compare Old English [hlu:d]. The spread of the Gothie
['mo:taJ or rathcr *['mu:ta] is confirmed by the borrowing into
Primitive Sluvic *['myto, 'mytarr), e.g. Old Bu1garian [mytoJ
'pay, gift,' [mytanJ 'publiean.'
25.6. Grammatically, the borrowed form is subjected to the
system of the borrowing language, both as to syntax (sorne rouge,
this rouge) and as to the indispensable infiections (garages) and
the fully eurrent, "living" constructions of eompositian (rouge-pot)
and word-formation (to rouge; she is rouging her face). ss often,
a simultaneous borrowing of several foreign form.'> saves this adap-
tation; thus, frorn Russian we get not only bolshevik but also
the Russian plural bol.shevih", which wc use alongside the English
plural-derivation bolsheviks. On the other hand, native gram-
452
CULTURAL BORROWING
CULTURAL nORROWING 453
454 CULTVRAL BORROWING
CULTURAL BORRO\VING
455
matieal constructions which occur, at the time of borrowing,
only in a few traditional formR, will scarcely he cxtended to caver
the foreign word. After complete adaptation, the is
subject to the same analogies as any similar native word. Thus,
from the completely nativized ['sowfrl chauffeur, we have the
ta chauffe [oowf], as in 1 had ta chauffe my mather
araund aU day.
'Vhen many forms are borrowed from one language, the forcign
forms may exhibit their own grammatical relations. Thus, the
vocabulary of English has its own
morphologie systm ( 9,9). The analogies of this system may lcad
to Thus, mutinous, mutiny, mutineer are derived,
in English, according ta Latin-French morphology, from an old
mutine, a loan from French mutin; French has Dot these derivatives.
Simibrly, due is a loan from French, but dut y, duteous, dutiable
(and, with a native English suffix, dutiful) probably had no French
source, but were formed, with French-borrowed suffixes, in English.
The back-formation of verbs in ( 23.5) is a case
in point.
'Vhen an affix oecurs in f'nough foreign words, it may be
tended ta with native materia1. Thus, the Latin-
French suffix as in agreeable, excusable, variable, has
been extended to forms like bearable, eatable, drinkable, where the
underlying verb is native. Other examples of French suffixes with
native Englsh underlying forms are breakage, hindrance, murderaus,
bakery. In Latin, nouns for' a man occupied with such-and-such
things' were derived from other nouns by means of a suffix
as monetrius 'coiner; from manela 'mint;
coin'; gemmrius 'jeweler' from gemma 'jewel'; telonarius
gatherer, publican' from telOnium 'j,oll-house.' :Many of these
were borrowed ioto the old Germanie htnguages; thus, in Old
English wc have myntere, tolnere, and in Old High German
gimmilri. Already in our earliest records, howevcr, we flnd
this Latin suffix extended to native Germanie underlying nouns.
Latin lZma 'waal' : lanarius is matched in Gothie by
wulla 'wool' ; wullreis ['wulla:ri:sJ similarly, bOka
'book' : bOkreis 'scribe,' mata 'toil' : motareis 'toll-gatherer' or , ,
in Old English, [wejn] 'wagon' : ['wejnere] 'wagoner.' Cases like
01d English [re:af] 'Rpoils, booty' ; ['re:avere] 'robber,' where there
was a morphologically rclated verb, ['re:avian] 'to despoil, rob,'
led ta new-formations on the modcl ['re:avian: 're:avere] even in
cases where there was no underlyng noun, such as ['re:danJ 'ta
read' : [':dere] 'reader' or ['wri:tan] 'ta write' : ['wri:tereJ 'writer.'
Thus arase our suffix -er 'agent,' \vhieh appcars in ail the Germanie
languages. Quite similarly, nt Il mueh later time, the same suffix
in Spanish pairs like banco ['banko] 'bank' : banquera [ban'keroJ
'banker,' was added ta native words in Tagalog, as ['si:pa?]
ball' : [si'pe:ro] 'football-pbyer,' beside the native derivation
[ma:ni' ni:par] 'football-piayer.'
If many loans have been made from sorne one language, the
foreign structure may even attraet native words in the way of
adaptation. In sorne German dialects, including the standard
language, we find native words assimilated to
centuation: Old High German ['forhana] ['holuntar]
'eIder, lilac, , l'wexxolter] 'juniper' are represcnted in modern
standard German by Farelle [fo'rele], /lalunder [ho'lunder], Wa--
cholder [va'xolder].
25.7. The speakers who introduce foreign things may calI thern
by the native name of sorne rc1ated abject. In adopting
ity, the Germanie peoples kept sorne of the heathen religious
terms: god, heaven, heU were merely transferred to the new religion.
Needle&s to say, the levcling to whieh these terms owe their uniform
selection in various Germanie languages, is only another instance of
bOITowing. The pagan term Easter is used in English and German;
Dutch Scandinavian adopted the term
pascha (Danish paaske, etc.).
If there is no closely equivalent native terro, one may yet
deseribe the foreip;n object in native words. Thus the Grcek-Latin
technicai term baptize was not borrowed but paraphrased in older
Germanie: Gothic said daupjan and (perhaps under Gothie
fiuence) German taufen (ta dip, ta duck'; 01d English said
jan], apparentIy from *['full-wi:hjan] 'ta make fully sacred'; Old
Norse said ['ski:rja] 'ta make bright or pure.' This involves a
semantic extension of the native term. American Indian languages
resort ta descriptive forros more oftcn than ta borrowing. Thus,
, '1 d u: ,
they render whiskey as 'fire-water, or rm roa as
1Ienomini uses [ri:tewew] 'he reads,' from English read, less often
than the native description [wa:pahtam], literally 'he looks at it.'
For electricity the 1Icnomini says ohis glanee' (mcaning the Thun-
dcrer's) and telephoning is rcndered as speech' rather
456 CULTURAL BORROWING
.'
"-1
CL'LTURAL BORROWING
457
than by [tdefo:newew] 'he telephones '; a compound
wagon' is commoner than the borrowed [atamo;pen]. Tools and
kitchcn-utcnsils arc designated by native descriptive terms.
If the foreign tcrm itself is descriptive, the borrower may re-
produce the description; this occurs espeeially in the abstract
domain. Many of our abstract technical terms are mercly transla-
tions of Latin and Greek descriptive terms. Thus, Groek [sun-
'ejde:sis] 'joint knowledge, conscioUfmess, conscience' is fi deriva-
tivc of the verb [ej'denaj] 'to know' with the preposition [sun]
'with.' The Romans translated this philosophical term by con--
scientia, a cornpound of Bcientia 'knowledge' and con- 'with.' The
Germanie languages, in turn, reproduced this. In Gothie ['mi6-
wissi: J 'conscience' the first member means 'with' and the second
is an abstract noun derived from the verb 'to know,' on the Greek
model. In Old English [je-'wt] and Old High German [gi-'wissidaJ
the prefix had the old meaning 'with'; in North-German and
Seandinavian forms, such as Old Norse ['sam-vt], the prefix is the
regular replacer of an oid [ga-Jo Finally, the Slavic languages
translate the term by 'with' and 'knowledge,' as in Russian
[' so-vest] 'conscience.' This proeess, ealled loan-translation, in-
volves a semantic change; the native terms or the components
which arc united ta create native terms, evidently undergo an
extension of meaning. The more Ii terate and elevated style in ail
the languages of Europe is full of semantic extensions of this sort,
chiefly on ancient Greek modcls, with Latin, and often also French
or German, as intermediaries. The Stoic philosophers viewed aIl
dceper emotion as morbid and applied to it the term ['pathos]
'suffering, disease,' abstract noun of the verb [' paskho:1'1 suffer'
(aorist tense ['cpathon] '1 suffered '). The Romans translated this
by passio 'suffcring,' abstract of patiar '1 suffer,' and it is in this
meaning that we ordinarily use the borrowed passion. German
writers, in the seventeenth century, imitated the Latin use, or that
of French passion, in Leidenschaft 'passion,' abstract of leiden 'to
suffer,' and the Slavic languages foIIowed the same model, as, for
instance, in Russian [strast] 'passion,' abstract of [stra'dat] 'to
suffer.' Ancicnt Grcck '1 throw (something) before
(someone)' had a1so a transferred use of the middle-voice forms,
[pro-'ballomaj] '1 accuse (somcone) of (somcthing).' The Latin
usage of a sirniIar compound may be a loan-translation: one said
not only canibu8 cibum ob-Jicere ' t throw food ta the dogs,' but also
alicuT probra obJicere 'to mproach someone for his bad actions.'
This was imitated in German: er wirfi den Fu.uer vor 'he
thro
ws
food bofore the dogs,' and er wirft mtr merne Mwset.aten vor
'he reproaches me for my misdeeds.' The use of hke
calling for' profcssional occupation,' detives frorn a
f Christian theology. Our terms imitate the late Latlll use III thlS
of voctio, abstract noun of vocare' ta cali'; similarly, German
Beruf 'calling, vocation, profession' is derived from rufen 'to calI,'
and Russian ['zvanijc] 'calling, vocation' is the abstract of [zvat]
'to cali.' A great deal of our grammatical has g.one
through this proccss. With a very peculiar extenslOn, the anCl(:nt
Greek grammarians used the tenn ['pto:sis] 'a fall' ,at for
f1cctional form' and then espccially for' casc-form. ThIS was Ull-
itated in Latin where, casus, literally 'a fall,' was used in the san:-
e
way (whenee our borrowed case); this, in tum, is
the German Fall 'fall; case,' and in Slavic, where Russmn [pa,des]
'case' is the learned-foreign (Old Bulgarian) variant of [pa'dos] 'a
fall.' In English the loan-translations have largcly
as in these examples, by Latin-French semi-lcarned borrowmgs;
thus the complex scmantic sphere of Latin commun, now covercd
by the borrowed common, was in ld English b!
ons of the native word [je-'me:ne], of paraUd formatIOn, Just aS It
still is in German by the native formE gemein and gemeinsam. In
Russian the loan-translations are often in Old Bulgarian form,
because 'this language served as the medium of theological writi.ng.
In a less elevated sphere, we have GalIicsms, such as a marnage
of convenience or il goes without saying, or l'ue told him l don't kn,ow
how many times, word-for-word imitations of French phras:-s. '1 he
ttm 8uperman is a translation of the German term comed by
Nietzsche. For' conventionaIized,' French und German use a
derivative of the noun style, as, French stylis [stiIize]; one oc-
casionally hears this imitated in English in the form slylized.
These transferences are sometimes sa clumsily made that we
may say they involve a misunderstanding of the imitated form.
The aneient Greek grammarians caUed the case of the verbal goal
(the "direct object") by the term [ajtia:ti'kc: 'the :ase
pertaining to what is effectcd,' employing an denved
from [ajtia:'tos] 'effected,' with an ultimately underlymg noun
[aj'tia:] 'cause.' This term was chosen, on of
constructions like 'he built a house,' where house lU Indo-
458
CULTURAL nORROWING
l
CULTURAL BORROWING 459
syntax has the position of a verbal goal. The ward
[al tm:], however, had a1so the transferred meaning 'fauIt, blarne'
and th; derived. verb [ajti'aomaj] had come to mean '1
accuse. Accordmgly, the Roman grammarians mistranslated the
Greek grammatical term by accus6tvus, derived from accusa '1
unintelligible term, accusative, was in tum translated
mt.o Russlan, where the name of the direct-abject case is [v'nitel-
nOJ], deriv,ed from ,[d'nit] 'to accuse.' The Menomini, having only
one (unvOlced) serles of stops, interpreted the English term Swede
as sweet, and, by mistaken dnsignate thn Swedish
by the term [saje:wenet] iiterally 'he who is sweet.'
HavlOg neliher the types [l, r] nor a voieed [z], they interpreted
the name of the town Phlox (\Visconsin) as frogs and translated it
as [uma:hkahkow-meni :ka: n] 'frog-town.'
25.8. Culturalloans show us what one nation has taught an-
The recent borrowings of BogUsh from French are largely
III the sphero of women's clothes, cosmetics, and luxuries. From
German we gct coarser articles of food (frankfurter, wner, ham-
sauerkraut, .pretzel, lager-beer) and some philosophical and
sClentlfic (zetlgeist, wanderlust, umlaut); from Italian, musi-
cal tenns (plana, sonala, scherzo, virtuoso) , From India we have
pundit, thug, curry, calico; from American Indian languages, lom-
ahawk, wampum, toboggan, moccasin. English has given roast
beef ta other languages, (as, French btfteck [biftek],
Russlan [blf steks]); also some terms of clcgant life such as club
high lIfe, five-o'clock (Iea) , smoking (for 'dinner-ja;ket'), fashio;"
able, and, above al!, terms of sport, sueh as match, golf, football,
base.ball, rugby. Culturalloans of this sort may spread over a vast
terntory, from language to language, aIon,!!; with articles of com-
merce. \Vords like sugar, pepper, camphor, coffee, Ica, tobacco have
spread al! over the world. The ultimate Source of sugar is proh-
a,bly Sanskrit ['arkara:] 'gritty substance; brown sugar'; the
TlOUS shapes of such words, sueh as French sucre [sykr] Italian
zucchero ['tsukkero] (whenee German Zucker ['tsukerj)', Greek
['sakkharon] (whence Russian ['saxar]), are due to substitutions
a,nd which took place under the most varied condi-
tlOUS III the borrowing and lending languages; Spanish aZUcar
[a'Oukar], for instance, is a borrowing from Un Arabie form with
the definite article, [as sokkar] 'the sugar' - just as algebra al-
cohol, alchemy contain the Arabie article [ail' the.' It is this
factor of widespread cultural borrowing which interferes with
our reconstruction of the Primitive Indo-European vocabulary,
in cases like that of the word hemp ( 18.14), Words like axe, sa<:k,
silver occur in various Indo-European languages, but with pho-
nctic discrepancies that mark them as ancient loans, preswnably
from the Orient. The word saddle oceurs in aIl the Germanie lan-
guages in a uniform type, Primitive Germanic *['sadulaz], but, as
it cantains the root of sit with Primitive Indo-European [dl (as
in Latin 'I sit') unshifted, we must suppose saddle to have
been borrowcd into pre-Germanie, too late for the shift Id > tl,
from sorne other Indo-European language - presumably from
sorne equestrian nation of the Southeast. The Slavie word for
'hundred,' Old Bulgarian [suto], phonetically marked as a loan-
word from a similar source, perhaps lranian, belongs to the same
geographic sphere. The early contact of the Germanic-speaking
peoples with the Romans appears in a layer of culturalloan-words
that antedates the emigration of the English: Latin vnum >
ld English [wi:n] > wine; Latin strnta (via) > Old English
[stret] > sireet; Latin caupo is reflected in Old Eng-
lish ['ke:apian] 'ta buy' (German kaufen) and in modern cheap,
chapman; Latin mango ' slave-dealcr, peddler' > Old English
['mangere] 'trader' (still in fishmonger) j Latin mane/a 'mint, coin'
> Old English mynet 'coin.' Other wards of this layer are pound,
inch, Old English [kirs] 'cherry,' ['persok] 'peach,' ['pise]
'pea.' On the other hand, the Roman soldiers and merchants
learned no less from the Germanie peoples, This is attested not
only by Roman writ.ers' occasional use of Germanie wards, but,
far more cogently, by the presence of very old Germanie loan-
words in the Romance languages. Thus, an old Germanie *l'werro:]
'confusion, turmoil' (Old Righ German ['werra]) appears, with
a usual substitution of [gw-] for Germa,nic [w-l, as Latin *['gwerra]
'war' in Italian guerra ['gwerra], French guerre (in Bng-
Iish war, wc have, as oiten, a borrowing back from French into
Euglish); Old Germanie *['wi:so:] 'wiso, manner' (Old English
, [wi:s]) appears as Latin *['gwi:sa] in Italian and Spanish guisa,
French guise [gi:z]; English guise is a 10an from French, alongside
the native wise, Germanic *['wantuz] 'mitten' (Dutch want,
dish vante) appears as Latin *['gwantusJ in Halian guanto 'glove,'
French gant (gunJ; English gaunilet is a loan from French,
Germanie words which passed iuto Latin in the early centunes
460 CULTURAL BORROWING
of our em are hose ( > Italian uosa 'leg!!;ing" cf b
soap ( > Latin .WipO) , *f'Owahljo'] 'towel' ('> F'a ovhe, 24.3),
when e . tE' . rene touaille
c ,m. urn, nghsh towel), roast ( >French r6tir wh . '
English roast), helmet ( > French heaume) .' ence, ln
creche), flask ( > Italian }iasca), harp ( > French > French
ample of a loantranslation is Latin companio 'co rp.)', An
thetic compound of con- 'with, along' and a syn_
mode! of Germanie *[ga-'hlajbo:], Gothie [ga'hla'ba] ,e: ' th;
a characteristically Germanie formation ct. . J h mpamon,
'along, with' and *['hlajbaz] 'bread' ( > *fga-]
CHAPTER 26
INTIMATE BORROWING
/' 26.1. Cultural borrowing of speech-forms is ordinarily mutual;
it is one-sided only to the extent that one nation has more to
give than the other. Thus, in the missionary period, hom the
seventh century onward, Old English borrowed Latin terms re-
lating ta Christianity, sueh as church, minister, angel, devil, apostle,
bishop, priest, monk, nun, shrine, cowl, mass/and imitated Latin
semantics in the way of loan-translation, but Old English gave
nothing, at this time, in returnlThe Scandinavian languages
contain a range of commercial ana nautical terms from Law Ger-
man, which date from the trading supremacy of the Hanseatic
eities in the late :Middle Ages; similarly, Russian contains many
nautical terms from Law German and Duteh.
In spite of cases like these, we ean usually distinguish between
ordinary cultural borrowing and the intimatt borrowing which
occurs when two languages are spoken in what is topographieally'
and politically a single community. This situation arises for the
most part by conquest; lcss often in the way of peaeeful mi-
gration. Intimate borrowing is one-sided: we distinguish betwcen
the upper or dominant langua;ge, spoken by the eonquering or
otherwise more privileged group, and the lower language, spoken
by the subject people, or, as in the United States, by humble
immigrants. The borrowing goes predominantly from the upper
language to the lower, and it very often extcnds to speech-forms
that are not connccted with cultural noveltics.
/We see an extreme type of intimate borrowing in the contact
of immigrants' languages with English in the United States.
English, the upper language, makes only the most obvious cul-
tural loans from the languages of immigrants, as spaghetti from
ltalian, delcalessen, hamburger, and 50 on/tor, by way of loan-
translation, liver-sausage) from German. /The immigrant, ta
begin with, rnakcs far more cultural laans. In speaking his native
language, he has occasion ta designate by their English names
any number of things which he has learned to know since coming
461
462 INTIMATE BORROWING INTIMATE BORROWING
463
to America: baseball, alderman, boss, ticket, and so on. At the
very least, he makes loan-translations, such as German ersie
PapieTe 'first papers' (for naturalization). The cultural reason
is less evident in cases Iike policeman, conductoT, street-car, depot,
road, fence, saloon;but we can sayat least that the American van-
eties of these things are somewhat different from the European.
In very many cases, however, not even this explanation will
hold. Saon after the German gets here, we find him using in his
German speech, a host of English forms, such as coat boUle kick
, , ,
change. He will say, for instance, ich hofJe, Sie werden's enioyen
[ix 'hofe, zi: 'verden s en'tSojen] '1 hope you'Il enjoy it,' or ich
hab' einen kalt gecatched [ix ha;p ajnen 'kalt ge'ketst] 'l've caught
a cold.' He makes loan-translations, such as ich gkich' das nicht
[ix 'glajx das 'nixt] '1 don't like that,' where, on the model of
English like, a verb with the meaning 'he fond of' is derived from
the adjective gleich 'equal, resemblant.' Sorne of these locutions
like this last, have hecome convcntionally established in
ean immigrant German. The phonetic, grammatical, and lexical
phases of these borrowings deserve far more study than they have
received. The assignment of genders ta English words in German
or Scandinavian has proved a fruitful topic of observation.
! The background of this process is evident. The up--
per language IS spoken by the dominant and privileged group;
many kinds of pressure drive the speaker of the lower language
ta use the upper ianguab'. Ridicule and serious disadvantages
punish his imperfections. In speaking the lower language to his
fel1ows, he may go sa far as ta take pride in garnishing it with
borrowings from the dominant speech.
In most instances of intimate contact, the lower language is
indigenous and the upper language is introduced by a body of
conquerors. The latter arc often in a minority; the bolTowing
rarely goes on at such headlong speed as in Our American instance.
Hs speed seems to dcpend upon a nurnber of factors. If the speak-
ers of the lower language stay in touch with specch-fellows in
an unconquered region, their language will change less rapidly.
The fewer the invaders, the slower the pace of borrowing. Another
retarding factor is cultural superiority, real or conventionally as-
serted, of the dominated people. Even among our immigrants,
educated families may keep their language for generations with
little admixture of
The same factors, apparently, but with sorne difference of
weight may finally lead to the disuse (extinction) of one or the
ther Numbers count for more here than in the matter
of borrowinp;. Among immigrants in America, extinctionJ like
gocs on at great specd. fIf the immigrant is
isolated, if his cultural level is low, and, above all, If he marnes
a persan of different speech, he may cease entirely ta use h1s na-
tive language and cven lose the power of speaking it intelligibly.
English becomes his oniy language, though he it vcry
imperfectly; it bccomes the native language of hls chl:dren. They
may speak it at first with foreign features, but outslde contacts
soon bring about a oompletc or nearly complete oorrection. In
other cases the immigrant continues ta speak his native language
in the home; it is the native language of his children, but at school
ag
e or evcn earlier they cease using it, and English becomes their
" f .
only adult language. Even if their English kccps sorne orelgn .
ooloring, they have little or no command of the parentallanguage;j
bilingualism is not frequent. In the situation of conquest the
process of extinction may he long delayed. One or more
tians of bilingual o;peakers may intervene; then, at sorne JXlmt,
there may corne a generation which dnes not use the lower
in adult life and transmits only the upper language ta Its chl1-
dren.
(rhe lower lanW1age may survive and the upper language die
out. If the conquerors are not numerous, or, especially, if they
do not bring their own women, this outcorne is likely. In less ex-
treme cases the conquerors continue, for generations, ta speak
their own language, but find it more and more necessary to use
also that of the conquered. Once they form merely a biIingual
upper class, the loss of the less useful upper can easily
take placo; this was the end of Norman-French ln
26. 2. The confl.ict of languages, then, rnay take many dlfferent
tums. The whole territory may end by speaking the upper lan-
guage: Latin, brought into Gaul round the beginning of the Chris-
tian em by the Roman conquerers, in a few centuries crowdcd out
the GelUc speech of the Gauls. The whole terri tory may end by
speaking the lower language: NormanFrench, brought into Eng-
land by the Conquest (1066), was crowded out by English in
three hundred years. Thcre may he a territorial distribution:
wheu English was brought iuto Britain in the fifth century of our
464
INTIMATE BORROWING INTIMATE BORROWING 465
era, it crowded the native Celtic speech into the remoter parts
of the sland. In such cases there follows a geographic struggle
along the border. In England, Cornish died out round the YCar
1800, and Welsh, until quitc recently, was losing ground.
1 In aIl cases, however, it is the lower language which borrows pre-.
dominantly from Ihe upper. Accordingly, if t,he upper language
survives, it remains as it was, cxcept for a few cultural laaus
such as it might take from any nf'ighbor. The Romance
contain only a few cultural loan-words from the languages that
were spoken in thcir terrtory before the Roman conquest j Eng-'
lish has only a few cultural loan-words from the Celtic languages
of Britain, and American English only a few from American
dian languages or from the languages of ninetcenth-century im-
migrants/ In the case of conquest, the culturalloans which remain
in the surviving upper language are chiefly place-names j witucss,
for example, American Indian place-names such as .IIIassachuseUs
,
Wisconsin, Michigan, Illinois, Chicago, Jfilwaukee, Oshkosh,
Sheboygan, Waukegan, Muskegon. It is intercsting to sec that
where English in North America has superseded Dutch French
, ,
or Spanish as a colonial language, the latter has left much the same
traces as any other lower language. Thus, from Duteh wc have
culturalloan-words like cold-slaw, cook, cruller, spree, scow, boss,
and, especially, place-names, such as Schuylkill Catskill Harlem
, , ,
the Bowery. Place-names give valuable testimony of extinct lan-
guages. Thus, a broad band of Celtc stretches across
Europe from Bohemia to England; Vienna, Paris, London are
CeItic names. Slavic place-nll.mes cover eastern Germany: Berlin,
Leipzig, Dresden, Breslau.
/ On the other hand, if t.he lower language survives, it bears the
marks of the struggle in the shape of copious borrowings. Englsh,
wit.h its loan-words from Norman-French and its enormous layer
of semi-Iearned (Latin-French) vocabulary, is the classical instance
of this/The Battle of Hastings, in 1066, mll.rks the beginning. The
first appearances of French \vords in written records of EngIish falI
predominantly into the period from 1250 ta 1400; this means
probably that the actual borrowng in each case occnrred sorne
decades earlier. Round 1300 the upper-cbss Englishman whatever
- ,
his descent, was either bilingual or had aL 1east a good
speaker's command of French. The mas:;: of the people spoke anly
English. In 1362 the use of English was prescribed for law-courts;
in the same year Parliament was opcned in English. The conflict
between the two languages, lasting, say, from 1100 to 1350, scems
not to have affected the phonetic or grammatical structure of Eng-
lish, except in the sense that a few phonemic features, such as the
initiais [v-, Z-, J-], and many features of the morphologic system of
French were kept in the borrowed forms. The lexical effect,
however, was tremendous. English borrowed terrns of government
(slate, crown, rign, power, counlry, people, prince, duke, duchess,
peer, courl), of law (Judge, jury, just, sue, plea, cause, accuse, crime,
marry, prove, false, heir), of warfare (war, baitle, arms, soldier,
officer, navy, siege, danger, enemy, march, force, guard), of religion
and morais (religion, virgin, angel, saint, preach, pray, l'Ide, save,
tempt, blame, order, nature, virtue, vice, science, gracc, crud, pity,
mercy) , of hunting and sport (leash, falcon, quarry, scenl, track,
sport, cards, diee, ace, suit, tnlmp, partner), many terms of general
culturll.1 import (honor, glory, fine, nuble, art, beauty, color, figure,
painl, arch, tower, column, palw:e, caslle), and tenus rclating to the
household, such as servants might lourn from master und mistress
(chair, table, furniture, serve, soup, fruit, jelly, boil, fry, mast, toast) j
in this last sphere wc find tho oft-eited contrast bet.ween the native
English names of animais on the hoof (ox, calf, Bwine, sheep), and
the French loan-word names for their flesh (beef, veal, pork, mutton).
It is worth noting that our personal names arc largely French, as
John, James, Frances, Helen, including even thosc which ulti-
mateIy are of Germanie origin, such as Richard, Roger, Henry.
26. 3. The presence of loan-words in a wider scmantic sphorc
than that of cultural novelties enables us to recognize a surviving
lower language, and this recognition throws Jight not only upon
historienl situations, but also, thanks to the evidence of the loan-
words themselves, upon the lnguistic features of an aneient time.
Mueh of our information about older stages of Germanie speech
cornes from loan-words in languages that once were under the
domination of tribes.
Finnish, Lappish, and Esthonian contain hundreds of words
that arc plainly Germanie in origin, snoh as, Finnish kuningas
, king,' lammas 'sheep,' rengas 'ring,' niekla 'needle,' napakaira
'auger,' pello 'field' 18.6). These loan-wards occur not only in
such semantic spheres as political institutions, wcapons, tooIs, and
garments, but also in such as animais, plants, p/1rts of the body,
minerais, abstract relations, and adjecLive qualities. Binee the
466
INTIMATE BORROWING
INTIMATE BORROWING 467
sou.nd-changes whieh have oeeurred in Finnish differ from those
WhlCh have oceurred in the Germanie languages, these
supplement the results of the comparative method, espeeiaIly as
the oldest of these borrowings must have been made round the
beginning of the Christian era, centuries before Our earIiest written
records of Germanie speech.
In aIl the Slavie languages we flnd a set of Germanie loan-words
that must have been taken, accordingly, into pre-Slavic. Therc is
an older layer whieh resembles the Germanie loan-words in
nish, Old Bulgarian [kunendZI] 'prince' < *['kuninga-], Old
Bulgarlan [xle:bu] 'grain, bread' < *['hlajba-] (Gothie hlaifs
'b:e
ad
,' English loaf), Old Bohemian [neboze:z] 'auger' < *[' naba-.
galza-J; A later stratum, wbieh ineludes cultural terms of Greeo-
Roman origin, sorne speci.!ieally Gothie traits; to this layer
belong hke Old Bulgarian [kotdu] 'kettle' < *('katila-.],
Old Bulganan [myto] 'taU' < *['mo:ta], Old Bulgarian [tse:sarI]
'emperor' < *['kajso:rja-] ( 25.5), Old Bulgarian [userendzl] 'ear-
ring'. < *['awsa-hringa-]. We infer that the earlier stratum is pre-
GOthlC and dates from the beginning of the Christian Era, and that
the later stratum cornes from the stage of Gothie that is repre-
sented in our written documents of the fourth ccntury.
In what is known as the Great Migrations, Germanie trihes
conquered various parts of the Roman Empire. At tbis tilne Latin
aIready contained a number of old cultural Ioan-words from Ger-
manie ( 25.8); the new loans of the Migration Period ean he
distinguisbed, in part, either by their gcographie distribution, or
by formai characteristics that point to the dialect of the conquerors.
Thus, the vowel of Italian elmo ['elmo] 'helmet' refiects an oid
(i], and the Germanie [eJ of a word like *['helmaz] (Old English
h.elm) appears as [il only in Gothic; the Goths ruled Italy in the
slXth century. On the other hand, a layer of Germanie words with
a consonantshift like that of South German, represents the Lom-
bard invasion and rule. Thus, Italian tallera ['tattera] 'trash' is
presumably a loan from Gothie, but zazzera ['tsattsera] 'long hair'
represents the Lombard form of the same Germanie word. Italian
n'ceo 'rieb,' elso 'hilt,' tujJare 'ta plunge' are similarly marked as
loans from Lombard.
The most extensive borrowing in Romance from Germanic
appears in Freneh. The French borrowings from the Frankish
rulers, heginning with the name of the country France, pervade
the voeabulary. Examples are Frankish *[helm] 'helmet' > Old
French helme (modern heaume [o:m]) j Frankish *['falda-,sto:li]
Ifolding-stMI' > Old French faldestoel (modern fauteuil [fot:j]);
Frankish *[bru:n] 'brown' > French brun; Frankish *[bla:w]
'blue' > French bleu; Frankish *['hatjan] 'to hate' > French
har; Frankish *['wajdano:n] 'to gain' > Old French gaagnier
(modern gagner; English gain from French). This last example
illustrates the fact that many of the French loan-words in English
are ultimately of Germanie origin. Thus, English ward is a native
form and represents Old English ['weardjan]; the eognate Frankish
*['wardo:n] appears in French as garder [garde], whence English has
borrowed guard.
It is not surprising that personal names in the Romance lan-
guages arc largely of Germanie origin, as French Louis, Charles,
Henri, Robert, Roger, Richard, or Spanish Alfonso (presumably
< Gothie *['hallu-funs] 'eager for fray'), Adolfo (presumably
< Gothie *[' allal-ulfs] 'wolf of the land'). The upper-class style
of name-giving survives even when the upper language is otherwise
extinet.
Repeated domination may swamp a language with loan-words.
Albanese is Baid to eontain a ground-stock of only a few hundred
native \Vords; ail the rest are dominance-Ioans from Latin, Ro-
mance, Greek, S1avic, and Turkish. The European Gipsies speak
an Indo-Aryan language: it seems that in their various abodes they
have been sufficiently segregated to keep their language, but that
this language figured always as a lower language and taker of
loan-words. Ali the Gipsy dialeets, in particular, contain loan-
words from Greek. F. N. Finck defines German Cipsy simply as
that dalect of the Gipsy language in which "any expression lacking
in the vocabulary" il' replaced by a German word, as ['flikerwa:waJ
'1 patch' from German flickcn 'to patch,' or ['stu:lo] 'chair' from
German Stuhl. The inflectional syst.em, however, is intact, and the
phonetics apparently differ from those of German.
The model of the upper language may affect even the gram-
matical forms of the lower. The anglicisms, say, in the Arnerican
German of immigrants, .!ind many a parallel in the languages of
dominated peoplesj thus, Ladin is said ta have largely t.he syntax
of the neighboring German, though the morphemes are Latin.
In English we have not only Latin
4
French affixes, as in eatable,
murderous, ( 25.6), but also a few foreign features of phonetic
468
INTIMATE BORROWING INTIMATE nORROWING 469
pattern, as in zoom, jounce. Non-distinctive traits of phonemes do
llot seem to he borrowed. When wc observe the Arnerican of
German parentage (whose English, at the same time, may show
sorne German traits) using an American-English [1] or [r] in his
German, we may account for this by saying that German is for
hr a foreign language.
:fWith a change of political or cultural conditions, the speakers
of the lower language may make an effort to cease and even to
undo the borrowing. Thus, th(; Germans have waged a long and
largely successful campaign aguinst Latin-French
and the Slavic nations against German. In Bohemian one
even loan-translations; thus, [zana:ska] 'entry (as, in a ledgcr),'
abstract of a verb meaning 'to carry in,' a loan-translation of Ger-
man Eintragung 'a carrying in, an entry,' is bcing replaced by a
genuinely native [za:pis] 'writng in, notation.'
26. 4. Beside the normal conflct, with the upper language, if it
survives, remaining intact and the lower language, if it survives
hearing off a mass of loan-words und loan-translations, or
syntactic habits, we find a numher of cases where something cise
must have occurred. Theoretically, therc would seem to be many
possibilities of an eccentric outcome. Aside from the mystic ver-
sion of the substratum theory ( 21.9), it seems possible that a
large population, having imperfectly acquircd an upper language,
might perpetuate hs version and even crowd out the more orig-
inal type spoken by the upper class. On the other hand, wc do not
know the limt ta which a lower language may he altered and yet
survive. Finally, it is conceivable that a conflict might end in the
survival of a mixture sa evenly balanced that the historian could
not dccide which phase t regard as the main stock of habit and
whieh as the borrowed n,dmixt.ure. Howcvcr, we do not know
whieh of these or of other imagin9.ble complications have ac-
tually ccurred, and no one, apparently, has succeeded in explain-
ing the concrete cases of aberrant mixture.
From the end of the cighth century on, Danish and Norwegian
Vikings raidcd and settlcd in England; from 1013 to 1042 England
ruled by Danish kings. The Scandinavian clements in
hsh, however, do not conform ta the type which an upper
guage leaves behind. They arc restrictcd ta the intimate part of
the vocabulary: egg, sky, oar, skin, gate, bull, bait, skirt, fel/ow,
husband, sister, law, wrong, loose, low, meek, weak, {live, take, caU,
ca8t, hit. The advcrb and conjunction though is Scandinavian,
and 50 are the pronoun forms they, thr, them; the native form
[1\1], as in 1 8aw 'em ( < Old English him, dative plural), is
treated as an unstressed variant of the loan-form them. Scandt-
navian abound in northern England. Wc do not know
what circumstances led to this peculiar resuIt. The languages at
the time of contact were in ail likelihood mutually intelligible.
Perhaps their reiation as ta number of speakers and as ta domi-
nance dffered in different localities and shiftcd variously in the
course of time.
/ Most instances of aberrant borrowing look as though an upper
language had been affccted by a lower. The clearest case is that
of Chilean Spanish. In Chile, the prowess of the natives Icd ta
an unusually great influx of Spanish soldiers, who settled in the
country and married native womcn. In contrast with the rest of
Latin America, Chile has lost its Indian languages and speaks
only Spanish, and this Spanish differs phonetically from the Span-
ish that is spoken (by the dominant upper class) in the rest of
Spanish America. The differenees run in the direction of the in-
digenous languages that were replaced by Spanish; it has bcen
surmised that the children of the tirst mixed marriages acquired
the phonetic imperfections of thcir mothers';
Some features of the normal type of the Romance languages
have been explained as rcfiections of the languages that were
superseded by Latin. It would have ta he shawn that the features
in question actually date from the time when speakers of the car-
lier languages, having impcrfectly acquired Latin, transmitted
it in this shape t their children. If this were granted, we should
have ta suppose that the official and colonizing elass of native
Latin-speakers was not large enough to provide an ever-present
model such as would have led ta the leveling out of these imper-
Actually, the pecuHar traits of the Romance languages
appear at sa late a date that this explanation seems improbable,
unless one resorts w the mystical (atavistic) version of the sub-
stratum theory ( 21.9).
Indo-Aryan speech must have bcen brought into India by a
relativcly sman group of invaders and imposed, in a long pro-
gression of dominance, by a ruling caste. Some, at loast, of the
languages which were superseded must have been kin ta the pres-
non-Aryan linguistic stocks of India. The principal one
470
INTIMATE BORROWING
INTIMATE nORROWING 471
of stocks, Dravidian, uses a damaI series of stops [T, D, NJ
alongslde the dental [t, d, nl; among the Indo-European
guagcs, only the Indo-Aryan have the two series, and in theil' hig.,
tory the domals have become more numerous in the course of time,
The Indo-Aryan languages exhibit also an ancient confusion of
[I] and [l'] whieh has been explained as due ta substrata that pas--
sessed only one or neUher of these sounds. The noun-deelension
of lat Indo-Aryan shows a re-formation, by which the same
case-endings are added to distinct stems for the singular and plural
as in Dravidian; t.his replaced the chameteristic
habit of differcnt sets of case-endings, as the sole distinction be-
tween singular and plural, added to one and the same stem,
In Slavic, cspeeially in Russian and Polish, the impersonaI
and partitive constructions closcly pamllel the Finnish habit.
The languages of the Balkan peninsula show various resernhlances,
although they represent four branches of Indo-European: Grock,
Albanesc, Slavie (Bulgarian, Serbian), and Latin (Roumanian).
Thus, Albanese, Bulgarian, and Houmanian, al! use a definite ar-
ticle that is placed after the nOlm; the Balkan languages generally
lack an infinitive, In other parts of the world, tao, we find pha-
netic or grammatical features prevailing in unrelated languages.
This is the case with SOrne phonetie features in the Caucasus,
which are COIrnnon bath to the several non-Indo-European stocks
and tQ Armenian and to the lranian Ossete. On the Northwest
Coast of North America, phonetic and morphologie peculiarities
appear in similar extensions. Thus, Quilleute, Kwakiutl, and
Tsimshian aIl have different articles for common nouns and for
names, and distinguish between visibility and invsibiIity in de-
monstrative pronouns; the latter peculiarity appears also in the
neighboring Chinook and Salish dialeets, but not in those of the
inOOrior. The suggestion has been made that different tribcs cap-
wornen from one another, who transmitted their speech,
wIth traces of their native idiom, ta the next generation.
Where wc can observe the historieal proeess, wc oceasional1y
tind phonctie and grammatical habits passing from language ta
language without actual dominance. In the modern period the
uvular-trill [l'] has spread over large part.s of western Europe as
a replacement of the tongue-tip [l']; t.oday, in France and in the
Dutch-Gerrnan arca the former is cit,ified and the latter rustie or
old-fashioncd. At the end of the Ages, large parts of the
English, Dutch, and German areas, including the socially favored
dialeets, diphthongized the long high vowels. The risc of the ar-
ticles and of phrasaI verb-forms consist.ing of 'have,' 'he,' or 'he-
come' plus past partieiple, in perfectic and passive values, took
place in both the Latin and the Germanie areas during the carly
Middle Ages.
26.5. Thcre remains a type of aberrant borrowing in which we
have at. lcast the assurance that an upper languag-e has been modi-
fied, though the dels of the process are no less obscure.
The English (now largely American) Gpsies have lost their
language nnd speak a phonetieally and grammatieally normal
variety of sub-standard English; among themselves, however,
they use anywherc from a fe",' dozen to several hundred words of
t.he olcl Gipsy lang-uage. These words are spoken with English pho
nemes and Eng-lish inflection and syntnx. They arc terrns for the
very commonest things, and include grammatical words, such as
pronouns. They are used int.erehangeably with the English equiv-
alents. aider reeordings shO\\' great. numbers of t.hese words;
apparently a long speech could be made almost entirely in Gipsy
words with Englsh phonetics and grammar. 1-1odern examples
are: ['mendij] 'l,' ['IEdij] 'you,' [sJl 'aJl,' [kejk] 'not,' [pon] 'say,'
['grajrl 'hor8e,' [aj 'dow nt 'kam tu 'dik e 'mus e-'i<;umrn e 'gruvl)]
'1 don't like t.o see a mltn a-kissin' n cow.' Occasional1y one hcars
a Gipsy infiecon, snch as ['rukjr], plural of [ruk] 'tree.' The
phonetics and grammar of Gipsy words mark them unmistak-
ably as borrowings by native speakers of English from a freign
language. PresuHlably they passee! from native speakers of the
Gipsy language, or from bilinguals, into the English of their chil-
dren or other persons for whom Gipsy was no longer a native
language. It is remarkablc, however, that speakers of t.he latter
sort should have interlarded their English with borrowings from
the senescent lower language. Under the general cireumstances
of segregation, these borrowings had perhaps a facetious value;
ccrtainly they had the merit of one's speech unintelli-
gible to outsiders. Amerieans of non-Englsh parentage who do
not speak thcir parents' language, sometimes, by way of jest,
use words of this language, speaking them with English sounds
and inficctions. Thus, will occasionaUy use
forms likc [swits] 'to sweat' (fram German schwilzen) , or [klac]
'to gossip' (from German klatschen). This trick 8eemB tQ be eom-
472 INTIMA'fE BO RRO \V ING INTIMATE BORROWING 473
IDonest among Jews, who live under a measure of segregation,
and the borrowings, moreover, are to a large extent the very words
which in German also are peeuliarly Jewish, namely, semi-Iearned
words of literary Hebrew origin, such as ['ganef] 'thief,' [g:>j]
'gentile,' [me'suga] 'crazy,' [me'zuma] 'money,' or dialect-forms
of Judeo-German, such as ['nebix] 'poor felIow' ( < Middle High
German ['n eb ix] 'may 1 not have the like'). It seems likely that
the Gipsy forms in English represent merely an extension of this
habit under conditions that made it especially useful.
Speakers of a lower language may make so Httle progress in
learning the dominant speech, that the masters, in communi-
cating with them resort to "baby-talk." This" baby-talk" is
the masters' imitation of the subjects' incorrect speech. There
is reason to believe that it is by no Hleans an exact imitation, and
that sorne of Hs features are buscd not upon the subjccts' mis-
takcs but upon grammatical relations that exist within the upper
language tsclf. The subjects, in turn, deprived of the correct
model, can do no better now than to acquire the simplfied " b a b y ~
talk" version of thc upper language. Thc result may be a con-
ventionalized jargon. During the colonization of the last few
centuries, Europeans have repeatedly given jllrgonized versions
of thcir language to slaves and tributary peoples. Portuguese
jargons arc found at various places in Africa, India, and the Far
East; French jargons exist in l\lauritius and in Annam; a Spanish
jargon was formerly spoken in the Philippines; English jargons
are spoken in the western lslands of the South Seas (here known
as Beach-Ia-Afar), in Chinese ports (Pidgin English), and in Si-
eITa Leone and Liberia. Unfortunately, these jargons have not
been weIl recorded. Examplcs from Beach-I2rMar are:
Whal for you put diss belonga master in fire? Him cost plenty
money and that fellow kai-kai him. '\Vhy did you put the master's
dishes into the fire? They cost a lot of money and it has destroyed
them' - spoken to a cook who had put silverware into the oven.
What far you wipe hands belonga you on clolhes belonga e s s e ~
poon? 'Why did you wipe your hands on the napkin?'
Kai-kai he finish? 'Is dinner ready?'
You not like soup? He plenly goad kai-kai. 'Don't you like the
soup? It's very good.'
What man you give him stick? 'Ta whom did you give the stick?'
Me savey go. '1 can go there.'
In spite of the poor rccording, wc may perhaps reconstruct the
creation of speech-forms like these. The basis is the foreigner's
desperatc attempt at English. Then cornes the English-speaker's
contemptuous imitation of this, which he tries in the hope of
making himself understood. This stage is represented, for instance,
by the linga which the American, in slumming or when traveling
abroad, substitutes for English, ta make the foreigner understand.
In our examples we notice, especially, that the English-speaker
introduces such foreign words as he has managed ta learn (kai-
kai 'eat' from sorne Polynesian language), and that he does not
discriminate between foreign languages (savey 'know,' from Span-
ish, figures in al! English jargons). The third layer of alteration
is due ta the foreigner's imperfect reproduction of the English-
speaker's simplified talk, and will differ according to the phonetic
and grammatical habit of the foreigner's language. Even the
poor orthography of our cxamples shows us substitution of [s]
for [fi] in dish and failure to use final [IJ], in belonga, and initial
[sp], in esseppoon for spoon.
A jargon may pass into general commercial use between persons
of various nationality; we then call it a linguafranca, using a term
which scems to have bcen applied ) an Italian jargon in the eastern
Meditcrranean rcgion in the early modern period. Pidgin English,
for inst.ance, is uscd quite generally in commerce betwcen Chinese
and Europeans of other than English speech. In Washington and
Oregon, Indians of various tribes,. as well as French and English-
spcaking trader8, formerly used a Hngua franea known as" Chinook
Jargon," which was based, strangcly enough, on a jargonized forro
of the Chinook language, with admixtures from other Indian
languages and from English.
It is important to keep in view the fact, often neglected, that a
jargon or a lingua franea is nobody's native language but only a
compromise bet\veen a forcign speaker's version of a language and
a native spcaker's version of the foreign speaker's version, and so
on, in which each party imperfectly reproduces the other's repro-
duction. In many cases the jargon or lingua franca dies out, like
Chinook Jargon, without ever becoming native to any group of
speakers.
In sorne cases, however, a subject group gives up its native
language in favor of a jargon. This happens especially when the
subjcct group is made up of persons from different specch-coro-
474 INTIMATE BORROWING INTIMATE BORRO\VING
475
munitics, who can comrnunicate among themselves only by means
of the jargon. This waR the case, presumably, among Negro slaves
in rnanyparts of America. 'Yhen the jargon has become the only
language of the subject group, it is a crcolized language. The creo-
lized language has the status of an inferior dialect of the mastNs'
speech. It is subject to constant leveling-out and improvement in
the direction of the latter. The various types of "Negro dialect"
which wc observe in the United States show us sorne of the last
stages of this lcveling. \Vith an improvement of social conditions,
this leveling iR aeederated; t.he result is a caste-dialeet whase
speakers, so far as linguistic faet.ors are eoncerned, have no more
diffi culty than other sub-standard speakers in acquiring the stand-
ard language.
It isa question whether during this process the dialect that is
being de-ereolizcd may not influence the speech of the community
- whether the ereolized English of the southern slaves, for in-
stance, may not have influenced local types of sub-standard or
even of standard English. The Duteh of South Africa, known as
Afrikaans, shows sorne features that remind one of creolized lan-
guages - Ruch, for instance, as extreme inflectional simplification.
Since it lS spoken by the whole communit.y, one would have ta
suppose that the Dutch settlers developed a jargonized fonn of
Dut.ch in communication with native Africans, and that this
jarp;on, through t.he medium of native servants (especially, of
nurses) then influenccd the language of the masters.
In the very unusual case where the subject group, after losing
hs native language or languages and spcaking only a creolized
language, is removed from t.he dominance of the model language,
the creolized language escapes aRsirnilation and embarks upon an
independent career. A few such cases have been observed. Thus,
the descendants of runaway slaves who scttled on the island of
San Thom off the coast of West Africa, spoke a creolized Port.u-
gucse. A creolized Dutch was long spoken on the Virgin Islands.
Two crcolized fOTIns of English arc spoken in Suriname (Dutch
Guiana). One of these, known as Ningre Tongo or taki-taki, is
spoken by the descendants of slaves along the coast.. The other,
more divergent. from ordinary types of English, is known as Jew-
Tongo; it is spoken by the Bush Negroes on the Saramakka River,
descendants of slaves who won t.heir liberty in the eighteenth
century by rebellion and flight. It owes Hs name ta the fact that
sorne of the slaves wcre owned by Portuguese Jews. The rernark-
able fcature of Bush-Negro English is its extreme adaptation t.o the
phonetics and struct.ure of \Vest African.languages, a n ~ the reten
tion of much 'Vest African vocabulary: If the slaves stiJl spoke an
African language, it is a puzzle why t.hey should have abandoned it
in favor of English jargon.
The following examples of Ningre-Tongo arc taken from texts
rccorded by 1'1. J. Herskovits:
['kom na 'ini:-sej. mi: se 'gi: ju wan 'sani: fo: ju: de 'njam.]
'Come inside. 1 shaH give you something to eat.'
[a 't.aki: , 'gran 'tal)gi: fo: 'ju:] 'He said, "Thank you very
much." ,
[mi: 'njam mi: 'bere 'fum.] '1 have eaton my belly full.'
In the first of the following Bush-Negro English proverbs, kindly
supplied by Professor Herskovits, t.he tones are indicated by
nurnbers: Irising, 2levd, ofalling, and by combinations of numbers,
such as 13rising t.hen fal1ing, nJevel then falling, and so on.
[fulo kri 21 ki 23 anl tal)13 hD1J 2 W]21J 'full creek not stand uproot
weeds,' that is, 'A full creck docsn't uproot Rny weeds' - said
when a person boast.s of what he is going t.o accomplish.
[di: ju: sei: ju: hede, t.e ju: baj hati:, pt ju: pati: el)] 'If you sell
your head, then you buy hat, whcre you put him?' t.hat. is, 'If you
sell your head t.o buy a hat, whcre will you put it.?'
[pi:ki: maaw faa ga
n
paw] 'SmaIl axe fell great stick,' that is,
'A small axc can cut down a large t.ree.'
DIA LECT BORROWING
477
CHAPTER 27
DIALECT BORROWING
1 27. 1. The infant begins by acquiring the of the
people who take care of him. He gets most of his habits from sorne
one person, usualIy from his mothor, but he does nat repraduce this
person's speech exactly, because he takes SOrne forms from othet---..
personsl It is a matter of dispute whether any permanent habits
. h '
m t e normal case, arise as mere inaccuracies of imitation. Later
on,/the child acquires spcech-forms from more people; children are
especially imitative in their first contacts outside the immediate
family circle. As time goes on, the range of imitated persons be-
cornes wider; throughout his lire, the speaker continues ta adopt
from his fellows. At any momfmt, his language is a
composite of habits acquired from various people. '
1 whole groups of speakers agree in adopting or favoring
or dlsfavormg a an age-group, an oecupational
group, or a neighborhood group, a turn of speech will pass from
person to person. The borrowing of speech-habits within a com-
munity is largely one-sided; the speaker adopts new forms and
favoritisms from sorne people more than from ot.hers/ In any
group, sorne persons receive more imitation t.han others; they are
leaders in power and prestige. Vagucly defined as they arc, the
dlf'ferent groups make similarly one-sided adoptions. Every person
belongs to more than one minor speech-group; a group is inftu-
enced by persons who, along sorne other line of division, belong
to a dommant class. Among his occupat.ional companions for
example, a speaker will imit.ate those whom he believes to t.he
highest "social" standing. To take the extreme case, when a
speaker cornes in contact with persons who enjoy much greater
prestige, he eagerly imitat.es not only their generaI conduet. but
also thoir speech. Here the direction of leveling is most
apparent. The humble person is not imitat.ed; the lord or leader is
a model to most of those who hcar him. In conversation with hitn
the common man avoids giving offense or- cause for ridicule'
suppresses such of his habits as might seem pcculiar, and to
476
ingratiate himself by t.alking as he hears. Having conversed with
the great, he himself may become a in his own group for
those who have not had that privilege: Every speaker is a mediator
botween various groups.
adjust.ment.s are largely minut.e and eonsist in the favoring
of speech-forms more often than in the adoption of wholly new ones.
A great deal of adjustment probably concerns non-distinctive
variants of sound. On the other hand, when rival forms enjoy
somet.hing Iike cquality, the choice may be actually discussed: a
speaker dclibemtes whet.her he will say it's 1 or ifs me, or speak
either, neither with [ij] or with [ail. In our community, with its
tradi t.ion abollt the "correctness" of speech-forms, the speaker
asks "'Nhich form is better?" instead of asking "With which
persans shan l agree in speech?" In t.he main, howcver, the process
does not risc to the level of discussion.
Every speaker, and, on a larger scale, every local or social group,
act.s as an imit.atr and as a rnodel- as an agent in the leveling
process. No pcrson and nogroup acts always in one or the other
capacity, but"the privileged castes and the central and dominating
communities aet more often as rnodels, and the humblest classes
and most N'mote localit.ies more often as imitators.
27. 2. The import.ant. historical process in this leveIing is the
growth of central speech-forms that spread over wider and wider
areas. Suppose, for instance, that in a loeally differentiated area,
sorne one town, thanks t.o personalities that live in it or thanks
ta a favorable topographie sit.uation, becomes the scat of a re-
current religious rite or political gat.hering or market. The in-
habitants of t.he villages round about now resort at intervals
to this central town. On these visits they learn ta avoid the strik-
ingly divergent forms of t.heir domestic speech, rcplacing them
by forms that do not caU forth misunderstanding or rnockery.
These favored speech-forms will he sueh as are current in aIl or
most. of the local groups; if no one form is predominant, the choice
will fall usually upon the form that is used in the central town.
\Vhen the villager goes home, he continues t.o use one or another
of these new locutions, and his neighbors will imitate it., both he-
cause they know its source and because the speaker who has visited
the central t.own has gained in prestige at home. At. second, third,
and lut.er hand, these locutions may pass to still more rcmote
persons and places. The central town bccomcs a speech-center,
478 DIALECT BORROWING
DIALECT BORRO\VING
479
whose forms of speech, when there is not tao much weight against
them, become the "better" forms for a whole area of the surround-
ing country.
As commerce and social organization improvc, this process
repeats itself on a larger and larger scale. Each ccnter is imitated
over a certain area. A new concentration of political power c1e-
vates sorne of these centers to a higher rank; the lesser centers
themselves now imitate this main center, and continue ta spread
both its forms and their own over their petty spheres. This d e ~
velopment took place in the Middle Ages in Europe. At the end
of the medieval period, countries like England, France, and
Gennany containcd a number of provincial speech-centers, though
even by that time, in England and in France, the capital city
was taking the rank of a supreme speech-center for the whole area.
These levelings, where they occurred on a large scale, are reflected
in the great isogloss-bundles that mark the conflict of cultural
systems, such as the bundles which separate Low German and
High German or Northern and Southern French. The lesser provin-
cial and parochiallevelings appear as minor isoglosses; thus, we saw
that the boundaries of the petty states along the lower Rhine
that were swamped by the French invasion of 1789 arc reflected
in lesser isogloss-bundles of today. AlI this would he plainer,
were it not for the frequent shifting both of political boundaries
and of the relative influence of eenrers. The most variable
factor, however, is the differcnce between the speech-forms
themselves, since sorne will spread more vigorously than others;
either for semantic reasons or, less often, for reasons of formal
structure.
1 A similarity of speech in a district of any size may date from
the time when the speech-community first spread over this dis-
trict. The word house, for example, spread over England with the
entrance of the English language, at the time of the Saxon con-
quest. It then had the form [hu:s], and in the northern dialects
which still speak sa, the modern form may he a direct continua-
tion of the old form. /
In very many instances, however, we know that a uniformity
does not date from the time of seUlement. Thus, wc know that
the diphthong [aw] in house, mouse, etc., arase from oider [u:llong
after the oottlement of England. In these cases, older students
took for granted a uniform Iinguistic change over a large aroa,
supposing, for instance, that a large part of the English area made
a phonetic chan!!;e of [u:] ta [aw]. At present, we believe rather
thatjthe aetual change occurred among a relatively smaU group
of speakers, and that after this, the new form spread by linguistic
borrowing over the large area'; \Ve are led to this opinion by the
fact that isoglosses for parallel forms do Dot coincide. A diver-
gence like that of the isoglosses of the vowels in rnouse and house
in the Netherlauds ( 19.4) fits into our classification of linguistic
borrowing, but not into our classification of phonetic change. Some
students sec in this a reason for giving up our classifications, and
insist thata"phonetic change" spreads in this irregular fashion.
This staternent, however, is inconsistent with the original applica-
tion of the term "phonetic change" to phonemic parallelism in
cognate speech-forms ( 20.4). Accordingly, we should have to
devise a new classification or cIse to find sorne way of reconciling
the two kinds of phenomena that are included in the ne"i' use of
the teml "phonetic change" - and no one has ven attempted ta
do either of these things. The method which distinguishes b e ~
tween a uniform phonetic change and the spread by bOTTowing of
resultant variants, is the only formula that has sa far been de-
vised to fit the facts.
Even when a uniform feature could represent the type that
was imported in the original settlement, we may find upon closer
investigation that this feature has merely overlaid an aIder di-
versity. This may be disclosed by isolated relie forms ( 19.5),
or by the characteristic phenomenon of hyper-forms. Of these,
Gamillscheg gives a beautiful examplc. In the Ladin of the Dolo-
mite Mountains, Latin [wi-] has become [u-]: a Latin [wi'ki:num]
'neighbor,' for instance, appears as [uzin]. ln one corner of this
district, however, the Rau Valley, this change apparently did
not take place: Latin [wi-] is reprcsented by [vi-], as in [vzin]
'neighbor.' However, there is a queer discrcpancy. The Latin
type [aw'kellum] 'bird,' which appears in Italian as [u'ello]
and in the Ladin of the Dolomites as [uel], and did not have ini-
tial [wH, has in the Rau valley the form [viel] 'bird.' If the Hau
valley had really preserved Latin [wi-] as [vi-], the form [viel]
'bird' would he inexplicable. It can be understood only if we sup-
pose that the Rau dialect, like the other Dolomite dialects, changed
[wi-] to [u-j, and aftcrwards took ta borrowing the more urbane
Italian [vi-] as a replacement for the native [u-j. In doing this,
480 DIALECT BORROWING DIALECT BORROWING
481
the Rau speakers went too far, and substituted [vi-J for [u-] eVen
in the word *[ucel] 'bird,' where Italan has [u-] and not [vH.
An isogloss tells us only that there has occurred somewhere and
at sorne time a sound-change, an change, or a
cultural loan, but the isogloss docs not tell us where or when this
change occurred. The form which resulted from the change was
spread abroad and perhaps pushed baek, we know not with what
vicissitudes, in a process of dialeet borrowing whose outeome is
represented by the isogloss. The present area of a form may CVen
fail to include the point at whieh this fonn originated. It is a very
nalve error to mistake isoglosses for the limits of simple Unguis-
tic changes. The results of dialect geography tell us of linguistic
bonowing.
27. 3. If the geographic domain of a linguistie form is due 1.0
borrowing, we face the problem of determining who made the
original change. A cultural loan or an analogie-semantic innova--
tion may be due to a single speaker; more often, doubtless, it is
made independently by more than one. Perhaps the same is
true of the non-distinctive deviations whieh ultimately lead ta
a sound-change, but this matter is more obscure, since the actual,
linguistieal1y observable change is here the result of a cumula-
tian of minute variants. The speaker who favors or exaggcrates
sorne aeoustic variant, as weIl as the speaker who adopts sueh a
variant, has merely altered a non-distinctive featUre. By the time
a succession of sueh favorings has resulted in a change of phone-
mie structure, the borrowing process has doubtless long been at
work. There must have bccn a timc, for instance, when some parts
of the American English speceh-community favored the lower and
less rounded variants of the vowel in words like hot, cod, bo/her.
It is useless to ask what person or set of persons first favored these
variants; wc must suppose only that he or they enjoyed prestige
within sorne group of speakers, and that this group, in tum, in-
f1ucnced other groups, and so on, in the manner of widening
cles: the new variants werc fortunate enough through sorne time
and in rcpeatcd situations, to helong ta the more dominant speak-
ers and groups. This favoring went on until, over a Inrge part of
the area, and doubtless Dot cvcrywherc at the same time, the vowel
of hot, cod, bother coincided with that of far, palm, fathm" Only
at this moment could an observer say that ft sound-change had
occurred; by this time, however, the dlstribution of the variants
llJllong speakers, groups, and localities, was a result of
The moment of the coincidence of the two former phonemes mto
one could not he determined; doubtless even one speaker might
at one time make a difference and at another time speak the two
alike. By the time a sound-change becomes observable, its effect
has been distributed by the leveling process that goes on within
each community.
The linguist's classification of changes into the three great
types of phonetic change, analogic-semantic change, and borrow-
ing, is a classification of facts which result from minute and eom-
plicated processCs. The processes themsclves largely escape our
observation; we have only the assurance that a simple statement
of thcir results will boar some relation to the factors that created
thesc results.
j8ince every speaker a?ts as an intermediary groups
ta which he belongs, dlfferences of speech wlthm a dIalect area
are duc merely t.o a lack of mediatory speakers. The influence of a
will cause a specch-form ta spread in any direction
until, at some line of weakness in the density of communication,
it ceases t.o find adopters. Different speech-forms, with differ-
ent semantic values, different formai qualifications, and different
rival forrns ta eonquer, will spread at differcnt speeds and ovcr
iffercnt distances. The advance of the new form may be stoppcd,
moreover, by the advance of a rival form from a neighboring
speech-center, or, perhaps, merely by the faet that a neighboring
speech-center uses an unchanged forml
One other possible source of diITerentiation must be reckoned
with: absorpt.ion of a foreign area, whose inhabitants speak their
new language with peeuliar traits. We have Been ( 26.4) that this
is entirely problematic, since no certain example has been found.
For the most part, then, differentiation within a dialect area is
merely a rcSl1lt of imperfect lcveling.
27. 4. Increases in the area and intensity of unification are due
to a number of factors which we sum up by saying that the eco-
nomic and political units grow larger and that the means of com-
munication improve. 'Ve know little about the details of this
process of centralization, because our evidencc eonsists almost
tircly of writteo documents, I1nd written documents are in this
matter espeeially misleading; to begin with, they are in Europe
mostly couched in Latin und not in the language of the country.
482 DIALECT BORROWING
DIALECT nORltO\VING
483
In the non-Latin (rernacular) records of the English und Dutch_
German areas, we flnd aL the outset, - that is, from the eighth
century on, - provincial dialects. InternaI evidence shows that
evon these have arisen through sorne degree of unification, but
we do not know how much of this unification existed in actual
speech. In the later l\Jiddle Ages wc find begiIlnings of gre:1tBr ccu-
tralization. In the Dutch-German area, cspecialiy, we find three
fairly uniform types of lang-uage: a Flemish (" Middle Dutch")
type, a decidedly uniform North German (" l'vI iddle Law German")
type in the Hanselttic area, and a South German High
German ") type in the aristocratie literature of the southern
states. The language of these documents is fairly uniform over
wide geographic areas. In some respects, wc ean sec how local
peculiarities are excluded. The Nort.h German type is based pre-
dominantly on the speech of the city of Lbeck. The southern
type strikes a kind of average between provincial diaIects, ex
cluding SOIlle of the localisms that appear in present-day dialect.
In old Germanic the personal pronouns had separate forms for
the dual ::md plural numbers; in general, the distinction was re
mO\'ed by an extension of the plural forms ta the case where only
two persans were involved, but in some regions the old dual forms
were extended to plural use. In most of the Gcrman area the old
plural forrns, Hig-h German ir 'ye' (dative iu; accusative
iuch), survived, but certain districts, not.ably Bavaria and Aus-
tria, t.ook t.he second alternative: the modern local dialects use
thc old dual fortn css 'ye' (dative and accusative enk). our
Middle High German document.s from the latter region scarcely
cver show us these provincial forms, but write only the gener-
ally German ir 'ye.' On the other hand, careful study of a text
will usually show in what part of southern Germany it originated,
because many details had not been standardized. Poets' rimes,
especialiy, conform, on the one hand, ta certain conventions,
but, on the other hand, betray each poet's provincial phonetics.
Tt is remarkable that at the beginning of the modern period, in
the fifteenth and early sixteenth centuries, this South German
convention had broken dcwn and our documents arc again de-
cidedly provincial, unti! the coming of the modern national
ard language.
The modern standard languages, which prevail within tho
bounds of an cntire nation, supersede the provincial types. These
standard languages become more and marc uniforrn as time goes
on. In most instances they have grown out of t.he provincial type
that prevailed in the upper class of the urban center that hecame
the capital of the unificd nation; modern standard English is based
on the London type, and modern standard French on that of
Paris. ln other instances eveu t.he center of orig-in is obscure.
l...fodern standard German is not hased on any one provincial dia-
lect, but seems to have crystallzed out of an official and
cial type of speech that developed in the eastern froutier region.
It. ,vas not created, but only helped toward supremaey, by Luther's
use in his Bible-translation. This origin is refiected in the fact
that the documents of standard German until well into the eight-
eenth century are far less uniforrn and show many more provin-
cial traits than do those of English or French; the same ean br
said of the standard language as it. i8 sflokrm Loday.
The modenl state, then, pos8esses a standard language, which
is used in aU official discourse, in churches and schoo1s, and in aU
written notation. As saon as ft speech-group attains or sceks
political independence, or even asserts its cultural peculiarit.y, it
works at setting up a standard language. Thus, the
tians, cmcrging {rom Turkish rule, possessed no sLandard language;
a scho1ar, Vuk Stefanovieh Karadjich (1787-1864) made one on
the basis of his local dialect, ,vriLiug- a gTalllmar and lexicon. Bo-
hemia, governed from German-spel1king- centers, had nevertheless
deveJoped somethinj.i like a standard language at the tune of t.he
Reformation. The great reformer, Jan Hus (1369 -1415), in par-
t.icull1r, had devised an excellent system of spelling. ln the seven-
teenth and eighteenth centuries this movement dicd down, hut,
with the national revival at the end of this period, a Ilew standard
based on the old, WilS created large1y by the efforts of a
philologil1n, Josef Dobrowsky (1753-1829). \Vithin the mernory of
persans now living, the Lithuanian standard language, today
official and fully current in the confines of its nation, arase from
out of a ,velter of local dialccts. Groups that have not gained
poltical independence, such as the Slovaks, the C:alans, and the
Frisians, have developed standard languages. The case of Norway
is cspecially int.cresting. For sorne centuries Norway belonged
politically to Denmark and used standard Danish as ils national
language. The latter was similar enough t Norwegian speech-
forms t< make this possible for persans who got school training.
484
DIALECT BORROWING
DIALECT BORROWING 48.)
The Norwegians modified their standard Danish in the direction of
Norwegian speech-forms. This Dano-Norwegian Riksmaal (' na,.
tionallanguage ') beeame the native speech of the educated upper
elass; for the unedueated majority, who spoke local dialects, it
was almost a foreign language, even though after the politieal
separation From Denmark in 1813, it "las more and more assim-
ilatcd to the general type of the native dialects. In the 1840's a
Ivar Aasen (1813-1896) constructed a standard
language on the basis of Norwegian local dialects and proposcd its
adoption in place of Dano-Norwegian. With many changes and
variations, this new standard language, known as Landsmaal
(' native language '), has been widely adopted, so that Norway has
today t"lo officially rccognized standard languages. The advocates
of the two arc often in earnest conflict; the two standard languages,
by concessions on either side, are growing more and more alike.
27. 5. The details of the rise of the great standard languages,
such as standard English, arc not known, becausc written sources
do not give us a close cnough picture. In its early stages, as a local
dialect and later as a provincial type, the speech which later
became a standard language, mity have borrowed widely. Even
aftr that, before its supremacy has been decilied, it is subject ta
infiltration of outside forms. The native London development of
Old English [y] i8 probably [i], as in fill, kiss, sin, hill, bridge; the
[0] which appears in bundle, thrush, seems to represent a 'Vest-of-
England type, and the [e] in knell, mcrry an eastern type. In bury
['bcrij] the spelling implies the western deve1opment, but the
actual pronunciation has the eastern [e]; in bllSY ['bizijJ the spelling
is western, but the actual spoken form indigenous. The foreign
[oJ and [el must have come at a very early time into the official
wndon speech. The change of old [er] into far], as in heart, parson,
far, dark, 'varsity, Or clerk in British pronunci:ion (contrastng
with the developmcnt in earth, lcarn, persan, unersity, or elerk
in American pronunciation) scems to have been provincial; the
far]. forms filtered into upper-class London speech from the four-
teenth century on. Chaucer uses -th as the singular
prescnt-tcnsc ending of verbs (hath, giveth, etc.); our [-ez, -z, -sJ
ending was provincial (northern) until well into the sixteenth
century. Espeeially the East l\Iidlands influenced London English
during the carly ccn turies of the latter's pre-eminence. In later
times, the standard lunguage borrmvs from other dialects only
tcchnical terms, such as vat, vixen ( 19.1), or laird, eairn (from
Scotch), or else facctiously, as in hass, euss as jesting-forms for
horse, curse; here bass (' ",pedes of fish ') for */ierse, (Old English
bears) represent.s a more serious borrowing of earUer date.
The standard language influences the surrounding dialects at
wider range and more pervasively as it gains in prestige. It
affects espccially provincial centers and, through them, their
satellite dialects. This action is relative1y slow. Wc have seen
that a feature of the standard language may reach out.lying dialccts
long aiter it has been superseded at home ( 19.4). In the imme-
diate surroundings of the capital, the standard language acts very
stronglYi t.he neighboring dialcct.s may be so permeated with
standard frms as to lose ail their individuality. vVe are tdd that
within thirty miles of London there is no speech-form that could
he described as local dialeet.
The standard language takes speakers from the provincial and
local dialects. The humblcst people make no pretensc at acquiring
it, but with the spread of prosperity and educ:lton, it becomes
familiar to a larger and larger stratum. In western Europe:m
countries today most people possess at 1east a good srnatterinp; of
the standard language. The person who rises in the worlel. speaks
it as his adult language and transmits only it to his children: it
comes to be the native dialect of a growng upper layer of the
population.
Both in the graduai assimilation of !esser dia1cct.s and in the
conversion of individuals and families to standard speech, the
result is usually imperfect and is to be describcd as sub-standard or,
in the favorable case, as provincially colored standard ( 3.5).
The evaluation of these types varies in different countries: in
England they are counted inferior and their speakers are drivcn
toward a more rigid standardization, but in the United Stat.es or
in Germany, where the standard language belongs to no one local
group, the st.andard is less rigid and a vagucly-defined mnge of
varieties enjoys cqual prestige. The English which the first settlers
brought to America consisted, apparently, of provincialized types
of the standard language and of sub-standard, rat.her than of local
dialects. The characteristic features of sub-standard American
English secm to be general features of dialectal and
British English, rather than importations from any special British
local dialects.
486 DIALECT BORROWING
DIALECT BORRO\VING
487
27. 6. The study of written records tells us little about the
ntralization of speech and the rise of standard languages, not
only because the conventions of wriiing develap ta a hrge extent
independently of actuaI specch, but aIso because they arc mare
rapidly standardized and then Itctually influence the standardizing
of speech. 'Ve have seen that cven the early written notations of a
language tend ta use uniform graphs which saon become traditional
( 17.7). The spellngs of medieval manuscripts seem very diverse
to the modern student, yet closer inspection shows that they are
laThrcly convontional At the end of the :YIiddle Ages, as the use of
writing increases, the provincial types of orthography become more
and more fixed. After the invention of prnting and with the
spread of literacy, the convention grows bath more unified and
more rigid; at last come grammars and dictionaries whose teachings
supplement the examp1c that everyone has before him in the shape
of printed books. Schooling becomes more common, and insists
upon conventional style.
This developmcnt caneeais from us the actuaI centralization of
the spoken language. The historian has ta deal constant1y with
two opposite possibilities. The written convention, at bottom,
reflects the forms that have prestige in actual speech; on the other
hand, it conventionalizes much more rapidly and affects the pres-
tige of rival spoken fOnTIS. The decisive events occur in the spoken
language, yet the written style, once it has seized upon a form,
retains it more exc1usivc1y, and may wei!!;ht the seales in its
fayor. Wc get a glimpse of the slate of affars in the spoken bn-
p;uage frorn occasional aberrant spellings or from rimes. Thus, occa-
sional spellings and rimes sho"... us a rivalry in standard En!!;lish
between pronunciations with [aj] and with [Jj] in words like ail,
bail, join; the decisive yictory, in the last two centuries, of the
latter type is doubtless duc to its agreement ,vith the spclln!!;; we
may contrast the still unsettled fluctuation in similar matters
where the spelling does not exert pressure, sueh as [a] versus [E] in
father, rather, ga/her, command, or [a] versus [J] in dog, log, fog, doll.
In syntax and vocabulary the message of the wriUen record
is unmistakablc, and it exerts a tremendolls effect upon the stand-
ard language. In Old English and t this day in sub*standard
English, certain nep;ative forms requirc a negative adverb with
a finite verb: 1 don't wan/ none; the habit of the standard
guage seems ta have arisen first in writing, as an imitation of
Latin syntax. Everyone has had the experience of starting ta
speak a word and then realizing that he does not know how ta
say it, because he has seen it only in wriling. Sorne words have
become obsalete in actual speech and have then been restored,
from written sources: thus, soo/h, g,tise, prowess, paramour,
hes/, caitijJ, meed, ajJray were revivcd by eighteenth-century
poets.
\Ve get a clearer notion of the influence of written notation in
cases where it leads t actual chan!!;es in the language. Now and
then a reviver of ancent. forms mmnderstands his text and pro-
duces aghas/-ward. Thus, anigh 'near' amI idlesse 'idkness' are
pseudo-antique formations made by ninetoenth-century poets. In
Hamlet's famous speech, bourne means 'lirnit,' but modems, mis-
understanding this passage, use bOllrne in the sense of 'realm.'
Chaueer's phrase in derring do that longeth ta a kniyht 'in daring
ta do what is propcr for a knight,' was misunderstood by Spenser,
who took derring-do ta be a compound meaning 'brave actions'
and succeeded in introducing this ghostAorm into our elevated
language. I\Hsinterpretation of an old !cHer has 1ed ta the ghost-
form ye for the 17.7).
It is not only archaic ,....ritings, however, that lead to change in
aefual speech. If thore is any rivalry betwcen speech-forros, the
chances are weighted in favor of the form that if; repl'esented by
the ,....riHcn convention; eonsequently, if the written convention
deviates from the spoken form, people are likeIy ta infer that there
exists a prekmble variant that matches the wriUen form. Espe
ciany, if, wouid scern, in the last centuries, with the sprend of
acy and the great influx of dialect-speakers and sub-standard
speakers into the ranks of standanl-spel\kers, the influence of
the written form has grown - for these speakers, unsure of
selves in what is, after ail, a forcign dialcct, look t the \\Titten
convention for htUidance. The school-teacher, coming usually
from a humble and unfamiliar with the actual upper-class
style, is forced ta the pretense of knowng it, and exerts authority
over a rising generation of new A great deal of
3pelling-pronunciation that has becorne prevalent in English and
in Freneh, is due to this source. In a standard language like .the
German, which belongs originally to no one class or district, this
factor i" even more the spoken standard is thcre
iargcly derivcd from the written.
488 DIALECT
DIALECT nORROWING
489
In standard English an old [sju:] developed to as wc see
in the words sure [suwr] and sugar ['sugr]. This change is reflectfld
in occasional spellings since about 1600, such as shuite 'suit'
shew/id 'suited.' John Joncs' Pradical Phonography in 170'1
prcscribes the pronunciation with [s] for assume, assure, censure,
ensue,insure, sue, suel, sugar. The modern [sI or
[sJ] III of these words is doubtless a result of spelling-
pronuncIatlOn. The same is probably truc of [t, d] or [tj, dj] in
words like tune, due, which replaces an authentic [, j]; witness
forrns like vir/ue ['vrcuw], soldier ['sowl}r]. The British standard
pronunciation ['inj,,] India is probably older than the American
['indja]. Since old final [mb, IJg], as in lamb, long have lost the
it may be. that the preservation of the stop in [nd], as in hand,
lS due to spelltng-pronunciation; in the fifteenth, sixtnth, and
seventeenth centuries wc fin d occasional spellings li ke blyne ' blind, '
thousan, pOlin. The old ft] in forms like often, Mften, fasten is
being constantly by the lower reaches of standard-
speakers.
. The most cogent evidence appears where purely graphie de-
VIceS !ead to novel speech-forms. \Vritten abbreviations like
prof., lab., ce. lead to spoken forms [prof, lcb, ek] in students' slang
for professor, laboratory, economics. These serve as models for
innovations, such as [kw:Jd] for quadrangle, [dorm] for
dorrmtory. The forms [ej em, pij em] come from the A.M. and P.M.
of railroad time-tables. Other examples are [juw es ej] for United
of Amedea, [aj sij] for IlUnois Central (Railroad) , and [ej
bll, eJ em, em dij, pij ejc dij] for academic degrees whose full
designations, Bache/or of Ar/s, Mas/er of Arts, Doc/or of kledine,
Doctor of Philosophy, are actualiy iess current; the abbreviations,
morcover, have thc of the original Latin terms. French
like [t.e es d] for tlgraphe sans fil 'wireless telegraphy,
radlO ; III Russla many new rcpublican institutions are knowD
by names read off from graphic abbreviations, such as [komso'rnol]
for [kommuni'sticeskoj so'jus molo'dozi] 'communistic union of
or for [fseros'sijskoj tscn'tralnoj ispol'nitelnoj
komi/et] all-Russian central executive committee.'
influence of written notation works through the standard
langua.ge, but featurcs tilat are thus introduced may in time seep
clown IUto levels of speech. Needlcss to say, this influence
can he descrlbed only in a superficial sense as conservativc or
regularizing: the loans from written notation cleviate from the
results of ordinary devclopment.
27. 7. Thc full effect of borrowing from written documents can
be seen in the cases where written notation is carried on in sorne
that deviates widely from the actuallanguage.
Among the Romans, the upper-class dialect of the first century
lI.C. - the Latin tilat we find in the writings of Caesar and Cicero
_ bccame established as the proper style for written notation and
for formai discourse. As the centuries passed, the real language
came to differ more and more from this convention, but, aS lit-
erate people were few, the convention was not hard to maintain:
whoever learned ta write, learned, as part of the discipline, to
use the forms of classicai Latin. By the fifth century A.D., an
ordinary speaker must have needed serious schooling bcfore he
could produce writings in the conventional fornl. In reading aloud
and in formal speech, the custom apparently was ta follow the
written for
m
, giving 8ach letter the phonetic value that was sug-
gested by the current forms of the language. Thus, a graph like
centum 'hundred,' which in the classical period represented the
fonu ['kcntum], was now pronounced successively as ['kcntum,
'erntum, 'tsrntum] ttnd the like, in accordance \Vith the phonetic
development of the actuallanguage, which spoke, in the
cases, say ['kentu, 'centu, 'tsentuJ. To this day, in readwg Latm,
the different nationalities follow this practicc: the Italian rcads
Latin centum as ['centurn] because in his own language he writes
cento and speaks [' cEnto]; thc Frenchman reads it as [sent<>m]
bccause in his own hwguage he writes cent and speaks [san]; the
German got his tradition of Latin-rcading from a Romance tradi-
tion thHJ used [ts] for c and accordingly reads Latin centum as
[' tsentum]; in England one can still hear an "English" pronuncia-
tion of Latin, which says centum ['sentom], becausc it derives from
a French tradition. These traditional pronunciations of Latin
are now being superseded by a system which attempts to recon-
struct the' pronunciation of c1assical times.
This custom of carrying on written and formaI or learned dis-
course in classicai Latin passed, with Christianity, to non-Latin
countries. Records in the actual Romance languages, or in Celtic
or Germanic, begin round the year 700; they arc scarce at first
and become copious only in the twclfth and thirteenth centuries;
until sorne (ime after the invention of printing, Latin books re-
main in the majoril}'. Since Latin is still the official language of
the Roman Catholic church, we may say that its use as 11 written
and formai language persists ta the present day.
As saon as classical Latin had begun ta antiquate, persons who
had not been sufficient.ly schooled, were sure ta make mistakes in
writing it. In the non-Latin countries this was truc, of course,
from the moment when Latin-writing was introduced. As ta tbe
thoroughness of the training, there wcre differenccs of time and
place. The Latin written in l\Ierovingian France, from the sixth
to the cighth centuries, is decidedly unclassical, and reveals many
characteristics of the authors' spoken language - the language
whose hter form we cali French. In the ninth century, undcr
Charle" the Great, therc came a revival of schooling: our lexts
to a far more conventional Latin. Kccdless ta say that
ln the Romance countries, and ta some ext.ent., pcrhaps, even in
the others, errors in Latin-writing give us informntion about the
aetual language spoken by the l111thors. 'Ve have already seen
scholars miseonstrued thifl situation, mistl1king changes
ln Latm-writing for linguistic change and drawing the moral thai
linguistic changes were duc ta ignorance :md carelessnesfl and rep-
resented a kind of decay ( 1.4). Another error has proved more
tenacious - namely, that of viewing the "nwdieval Latin" of
OUr documents as an ordinary language. "\Vhen ,vc tind a new form
in these doeuments, thore is only a -remote possibility that this
:orrn represents an aetuai tradition of a classical Latin form;
ln by far the most imhtnces, it is eithor a new-formation on the
basis of classieal Latin, or a l::ttinization of some spokcn form.
Thus, the form quidilas 'whatness, charactcristic quality' whieh
appcars in medieval Latin-,vriting, is roughly constructpd On the
analogies of elassical Latin, and doefl not reJ1cct any ilpokcn form
either of clasflical or of medieval Limes. The fonn mansionaticum
'place for a feudallord to stop over nip;ht; domestc establishmont'
does not evidcnce the use of this form in classieal Latin: it is
merely a latinization of an actually spoken Old Frencn masnage
(or of it.s pre-l,'rench antecedent), which appears in later French
mesnage, modern mnage [mena:zJ 'houschold'; English manage
bo:r
owed
from a dcrived verb, French mnager. The htiniza-
han IS correct, to be sure, in the seme that masnage ls a mor-
phologie Corn bination \\'hose clements, if wc put thorn back into
classical Latin form, would have combined ns *mansinatum:
the medieval scribe hit upon the historically correct Latin
1 ts although actually, classical Latin formed no sueh com
en " . 'h k' M
tian. When wc read a perfect tense form prestt e III 1era-
vingian documents, we should do wrong to calI thls the
of forms like Italian prese ['prose] 'he took,' or French [pn];
it is merely an error in Latin-writing, the of a scnbe wh?
was not familiar enough with the classlcal Lat.m form
'he took' and wrote instead a pseudo-Latin form based on hlS,
spoken This error tells us t.hat the
employed the new-formation of the type Latm prenstt, ,,hleh
underlies the Romance forms and probably dates from a very carly
t . but it ,vould be a grave methodic confusion to say that the
1me, " . IL. f "
Romance forms arc derived from the medleva atm o.rm.
Again, when we find in Latin documents of
a word muta' toH,' il. would he a nave error to l'lee m thlS mcdlCval
Latin" word the source of Old Righ German muta' toU: (
the writer mercly used the German teehnical term 1.0 Latm-
writing because he knew no exact equivalent; one wnter even
speaks 'of nullum teloneum neque quod lingua muta va-
catur 'no toU or what is in German caUed muta. Moreover, we
find the derivatives mutarius, mutnarius 'ton-taker' the latter
wit,h an analogic that is peculiar ta German
em Mautner). In sum, then, the medieval dcvla-
tions from classieal Latin usage may throw hght upon hls actual
speech, but d.'1re not be confused with the of the lat-
ter, even in cases where the scribe sueceeded III makmg a correct
latinization. . .
27.8. 'Ye find, now, that at aH times, and wlth. the
modern spread of education, the Romance peoples mto
their formaI speech and then into ordinary levels,
book-Latin in the phonetic form of the tradttlOnal readmg-
pronunciation. These borrowings from the wriW.',n language are
known as learned words, or, by the French term, as mots savants
[ma savu
U
]. Art.er a book-Latin word came into spoken
use, it was subject, of course, ta the normal changes whleh
after occurred in the language; however, these were sometlmes
followed by re-shaping in the direction of the bookish farm. Many
a Latin word appears in a Romance language bath in its
developed modern farm, as a so-called popular word, and lU a
half-modernized Latin (or pseudo-Latin) form, as a learned ward.
490
DIALECT BORROWING
DIALECT BORROWING 491
Latin redemplionem [redempti'o:nemj 'redemption' appeaTS, by
developm.ent, as modern French rancon [ranson] 'ransom'
(l'-,nghsh lS a loan from Old French), but, as a borrowing
the .wn,tten in modern French rdemption [redonpsjon]
rcdemptlOn. At the bme of bookish borrowings the Frenchma
h d' L . 'n,
w en roa mg atm, used a pronunciation (based, as we have seen
Upon the actual linguistic carrespondences) which renclerccl
graph like by a pronunciation, say, of fredemp'
nern]: the dlfferences between this and the present-clay F-..n h
f
d n . n] <'Olle
re a pSJo are due to subsequent changes in the French language.
Only sorne - pcrhaps anly a minority - of the learned words ac-
tually went through this deveJoprncnt, but on the model of thase
that did, one rc-shapcs any new ones that may be taken from
the books; thus, if an educated Frenchman wanted to take u
Latin procrasiinationem 'procrastination,' he wouJd
It, accordancc with thesc models, as procrasiination [pr;:.krasti-
nasJo
n
].
Other examples of twofold development are: Latin Jabricam
> French forge ff;nz] 'forge,' learned Jab-
nque [fabnk] factory'; Latin fragile ['fragile] 'fragile' > French
[frel] 'frail,' learned fragile ffraZil] 'fragile'; Latin securum
(se: 'secure' > French Sr [sy:r] 'sure,' Latin securitaiem
> French srei [syrte] 'sureness, guarantee,'
learned secunie [sekyrite] 'security.'
Sometimes the book-ward got into the language eariy enough
to undergo some sound-change which gives it a superficially nor-
:nallook. Thus Latin capiiulum [ka'pitulum] 'heading' was taken
mto F:ench .speech early cnough to share in the development
ca > sa], and appears in modern French as chapitre fsa-
chapter.' Tho Ir] for Latin flJ is due apparently to an adapta-
tion of the type usually classcd as aberrant sound-change ( 21.10) ;
doUbtle.ss quite a few such changes are really due 1,0 re-shapings
of boolush words that presented an unusual aspect. In other cases,
a word b.orrowed after a sound-change, is still, by way of
adaptatIOn, put mto a form that partly or wholl
y
imitates the
of Thus, fi Latin discipulum [dis'kipulum]
.\\ould. glve by nonnal developmcnt a modern
[de seppJo]; thm does not exist, but the Icarned loan in
Hah
a
? partly apes these vowel-changes; it is not *[di'sipulo]
but dlscepolo [di'sepoJo]. The number of learned and semi-learned
forms in the western Romance languages is very large, especially
aS the standard languages have extendcd the analogy t the r:oint
where almost any Latin or Greco-Latin word can be mOd?rlllzed.
Among the French fonns that Wcl'e borrowed by Enghsh dur-
ing the period after the Norman Conquest, th:re were many of
thesc lcarned French borrowings from the Latm of books. The
literate Englishman, famlliar with bath French and Latin, got
into the hubit of using Latin words in the form they hud as French
mois sa/lanls. Wo have seon how the Englishman made his own
adaptations ( 25.4). In later Ume, the English writer
to use Latin words. In making these loans, we alter the Latm
graph and pronounce it in accordance with a fairly well-.fixed set
of habits; these habits are composed of (1) the adaptatIOns and
phonetic renderings that were eonventional in the French use of
book-Latin words round the year 1200, (2) adaptations that have
becomc conventionai in the English usage of Latin-French forms,
and (3) phonetic renderings due to English that
have occurrcd since the Norman time. Thus, the LatIn procras-
iinationem which is not currcnt in French, is borrowed from Latin
books int; English as procrasiination [prolkrrsti'ncjsI}], in accord-
ance with the above set of analogics. Dndcr (1) we have the fact
that French borrows its Latin words not in nominative singular
fonn (Latin procraslinatio) , but in accusative or ablati-:e form,
with loss of ending: had the v,ord bcen used, as a bookIsh loan,
in the Old French of 1200 to 1300, it would have appcarcd as
*procrastination *[prokrastina'sjo:n], with phonetic. changes which,
like the selection of the case-form, arc due ultImately, to the
model of non-learned French words. The rernaining deviations of
the actual English form, namely le] for a in the second syllable,
(ei] for a in the third, [;J fol' It" before vowel, and the weakening of
the end of the word to [-!)], are modcled on the phonetic changes
which have been undcrgone by words of similar structure that
rcally were borrowed during the Norman pcriod, such as Latin
nationem > Old French [na'sjo:n] > English nation ('nejs!)].
Final1y, the shift of accent to position copies an adap-
tation which English made in its actual loans from French. In
the same way, when we borrow from Latin books the verb prr:
crMtinare we renclor it as procrasiinate, adding the suffix -ate III
with an adaptation that has become habituai in Eng-
lish ( 23.5).
492
DIALECT BORROWING
DIALECT BORRO\VING 493
494 DIALECT nORROWING
DIALECT BORROWING
495
Bath the Romance languages and English can borrow, in this
way, not only actual Latin wards, but evcn medieval seri bal eoin-
ages, such as English quiddity from scholastic quiditas. We cven
invent new words on the general mode! of Latin morphology:
eventual, immoral, fragmentary are exumples of lcarned words
whose models do not oecur in Latin. Since the Rrnans borrowed
words from Greek, wc can do the saIlle, altering the Greek word
in aeeordance with the Romun's habit of latinization, plus the
Frcnchman's habit of gallieizinp; Latin book-words, plus the
lish habit of anglieizinp; French learned words. Aneient Grcck
[philos'phia:] thus gives an Englsh [fi'lasofijJ IJhilosophy. As in
the case of Latin, we are free to coin Greck ''lords: telegraphy rep-
resents, with the same modifications, a non-existent ancicnt Greek
*[te:lcgra'phia:] 'distan-writing.'
Ncedless to say, wc sOlIletimes confuse the analogies. 'Vc ren-
der ancicnl Greek [th] in English, against the custom of the Ro-
mance languages, by [OJ, as in [mu:tholo'gia:] > myth%gy. It
is true that ::meient Groek [th] has changed t [0] in modern Grock,
but the English habit is probably indcpendent of this and due
mercly to the spdling. l\Ioreover, medicval scribes, knowing th
as an abstruse Greek graph and pronouncing it simply as t [t], oc-
easionally put it into words thut were not Greek ut aIl. Thus,
the name of the Goths, old Germanie *['goto:z], appears in medi
eval Latin-writng Ilot only as goti but also as gothi, and it is from
the latter graph that we get our pronunciation of Goth, Gothic
with [OJ; the use of [6J in Lithuanian is a modern instance of the
same pseudo-learned pedantry. The same thing has happened in
English to an ordinary Latin word, auclorem > French autor
(modern auteur [otanD > l\liddle English autor; in Englsh it
was spellcd author and finally got the spclling-pronunciation
with [OJ.
The habit of learncd borrowing from the classieal languages
has spread to the other languages of Europe; in each one, the
learncd borrowing is aecompanied by adaptations which rcReet
the circumstances of the contact, immediatc or mediate, with the
Rornanee-speaker's use of book-Latin. Thus, the German, who
says .Valton [na'tsjo:n], Station [sta'tsjo:n], eould eonceivably bol'
row a *Prokrastinalion *[prokrastina'tsjo:n], - and similar habits
exist in the other languages of Europe.
This wholc history finds its parullel, including cven the graphie
archaization of spoken forms (like the medieval scribc's mansio-
naticum, presit) , in the use of Sanskrit in In?ia. In the
guagcs of India, graphie Jouns from Sansknt arc known as
tatsama Likc t.he mots savants of Europe, thcse
fonnations show us writtcn notation cxercising an influence upon
language.
CHAPTER 28
APPLICATIONS AND OUT LOOK
28. 1. The normal speaker faces a lnguistic problem whenever
he knows variant. forms which differ only in connotation _ for in.
stance, it's 1 and it's me. He states this problern in the question
"How shaU l talk?" In most cases he has no di fficulty,
the social connotations are obvious, and the speaker knows that
Sorne of the variants, (e.g. 1 done il) have an undesirable con.
notation and lead people to deal unkindly with the User. We ex-
press this t.raditionally by saying that the undesirable variant is
"incorrect" or "bad English" or even "Dot English" at aIl.
These statements, of course, are untrue: the undesirable variants
are not foreigners' crrors, but perfect1y good English; only, they
are not used in the speech of socially more privileged groups, and
aeeordingly have failee! t.o get into the repert.ory of standard speech-
forms. Even in smaller and less stratified speech-communities
whieh have segregated no standard the speakc;
usually knows which variants will do him hetter service.
When there is no obvious differenee between the variant forms
tbere should be no problem at aIl, sinee it evidcntIy will make
difference wbich variant the speaker uses. A speaker who is in
doubt whether to say it's 1 or il's me, has heard these two variants
from approximately the same kinds of since other-
. ,
Wise they would bear clear-cut connot.ations of desirability and
Since his assoeiates, then, use both forms, his
standmg Will not he affected by his use of one or the other.
theless, people devote time and energy to such problems, and
suffer anxiety on account of them.
The background of our popular ideas about language is the
fanciful doct.rine of the eight.eenth-century "grammarians."
This doctrine, still prevalent in our schools, brands aIl manner of
forms as "incorrect," regardless of facto Having heard t.he tcrm
"incorrect" applicd to variants which bear no undesirabIc con-
not.ation, the speaker grows diffident and is ready to suspect al-
most any speech-form of "incorrect.ness."
496
APPLICATIONS AND OUTLOOK 497
It would not have becn possible for" grammarians" t.o bluff a
large part. of our and they would not have
undertaken ta do 00, if the public had not bccn ready for the decep-
tion. Almost aU people, including even most native speakers of
a standard language, know that someone else's type of language
has a higher prestige. At the t.op, of course, there should be a most
privileg(;d group, whose membcrs nre sure of themselves in speech
as in aIl other issues of mannerism; in the English-speaking com-
munity, this should bo the Brit;h upper class, \vhich speaks the
"public school" variety of southern English. One may suspect,
however, that even within this group, the modcl of printed books
and the minor variations of modish cliques, make many speakers
unsure. Snobbery, the performance of acts which bclong to a more
privileged group, often lkes the shape, thcreforc, of unnatural
speech; the speaker utters forms which are not. current among his
asoociatos, because he believes (very often, mistakenly) t.hat these
forms are favored by Sorne "better" class of speakers. He, of
course, falls an easy prey to the authoritarian.
It is no accident that the "grammarians" arase when they did.
During t.he eight.eonth and nineteent.h centuries our society went
through grcat changes: many persons and families rose jnt.o
tively privilegcd positions and had to change from non-standard
ta standard speech. The problem t1mt faces the speaker who makes
this change, ,vil! concern us later; we see now that the aut.horitari:Ln
doctrine battened on t.he diffidenee of speakers whose background
was non-standard - speakers who were afraid t.o trust. t.he spceeh-
forms they had heard from their parents and gr:llldparents. In the
United St.ates this is complicated by the faet that cven many
native sper..kers of standard English have a foreign background and
are easHy frightened into t.hinking that a speech-form which is
natural to them is actually "not English."
Indeed, diffidence as to one's speech is an almost universal trait.
The observer who sets out t.o study a atrange bnguage or a local
dialcct, often get.s data from his informants only to find them using
enUrely different fonns when t.hey speak among themsc1ves. They
count. these latter forIlls inferior and are ashamed to givc them to
the observer. An observer may thus record a language entirely
unrelat.ed ta the one ho is looking for.
The tendency to revisc one's speech is universal, but the rcvision
consists normally in adophng forms whieh one heurs from one's
498
APPLICATIONS AND OU'l'LOOK
APPLICATIONS AN D OUTL OOK 499
The doctrine of our grammarians has had very !ittle effect
10 the way of or establishing specifie speech-forms, but it
has set up among hterate people the notion that forms which one
has not heard may be "better" thun those which one actually
hears and speaks. The only danger that threatens the native
speaker of a standard language is artificiali ty: if he is snobbish
p:iggish, or tnid, he may fiIl his speech (at le"st, when he is
hls good behavior) with spelling-pronunciations and grotesque
" t" f 'rh
. ?rIllS. e speaker to whom the standard language
18 na.tlve, will hardly ever tind good rCason for replacing a form
that IS natural to him. Variants such as it's 1: it'I; me haye beeo
used for centuries in the upper levels of English speech; there is
no reason why anyonc should make himsclf uncomfortable about
them.
It .is not often that a speaker has to choose between genuine and
well-defined variants within the standard language. In
t.he Umted States, the speaker of Central-\Vestern standard Eng.
hsh, who uses the vowel le] indifferently in man, mad, mat and in
laugh, bath, can'I, is confronted by a higher-toned type of the
standard language, which uses a different vowel [a] in words of the
set. \Vhether he tries to aequire this more elegant feature,
wlll depend upon how highly he values conforrnity with the speak.
ers wh? Use it. If he is placed entircly among them, say, by resi-
dence 10 New England or in Great Britain, he may naturally faH
iuto the new habit. One does weI! to remember t.hat the change is
not easy to make, and that a novice is likely to put. the ncw feature
iuto places where it does not. belong, producing outlandish hyper-
forms, such as [man] for [men] man. Unless the speaker constantly
hcars the preferred type from his associats, he had better not
moddle with it. Unnatural spcech is not pleasing. In England
wh.cre provincially tingcd types of the st.andard language
ferlOr to the "public-school" type, this question may wear a
ent aspect.
As to non-distinctive features of speech, the sit.uat.ion is differcnt.
Although t.hey are habituai, they do not form part of the signaling-
system, and are subject t divergence and improvement. Just as
one may be considerate and agrCCl1ble in other mannerisms one
may speak in a pleasant "tone of voice" - that is with a
1
. ,
atIon of acoustic ft'atures. The Same may be
sald of the cornbmatlon of non-distinctive and semantic features
which we call style; here too, one may, without affectation, use apt
and agreeable forms. Unfortunately our handbooks of rhetoric
confuse this wit.h the silly issue of "corrcctness."
For the native speaker of sub-standard or dialectal English, the
acquisition of standard English is a roal problem, akin to that of
speaking a forcign language. To be told that one's habits are due
to "ignorance JJ or "carelessness" and are "not English," is by
no means helpfuL Our schools sin greatly in this regard. The non-
standard speaker has the task of replacing sorne of his forms (e.g.
rseen il) by others (I saw il) which are current among people who
enjoy greater privilege. An unrealistic attit.ude - say, of humility
- is bound to impede his progrcss. The unequal dist.ribution of
privilege which injured him in childhood, is a fault of the societ.y in
which he lives. Wit.hout embarrassment, he should try t.o substitute
standard forms which he knows from actual hearing, for those
which he knows to be non-standard. In the beginning he runs a risk
of using hyper-urbanisms; such as l have saw (arising from the
proportion 1 seen il : l saw it = 1 have seen it : x). At a later stage,
he is likely to climb into a region of stilted verbiage and over
inyolved syntax, in his effort to escape from plain dialect; he should
rather take pride in simplicity of speech and view it as an ad-
vantage that he gains from his non-standard background.
28. 2. Society deals with linguistic matters through the schoo1
system. "'Thocver is accustomed to distinguish betwecn linguist.ic
and non-linguistic bchaYior, will agree with the criticism that our
schools deal too much with theiormer, drilling the chiId in
response phases of arit.hmetic, geography, or history, and neglecting
to train him in behavior toward his actual environment. In the
simpler community of a few generations ago, matters of art. and
science were remote, and mechanical and social processes worked
on a scale which placed them (or seemed to place them) within
direct everyday observation: the child lcarned practical matters
without the help of the school, which nceded to train him only in
the three R's. The schools have clung to this pattern, in spite of
the complexities of modern Iife. Attempts at irnprovement have
not becn encouraging: practical (that is, non-linguistic) matters
have been introduced in the shape of ill-considered fads. In view
of our schools' concentration on verbal discipline, it is surprising to
see that they arc utterly benighted in linguistic mattcrs. How
training is best impartcd must be for the pedagogue to determine..
500 APPLICATIONS AND OUTLOOK
APPLICATIONS AND OUTLOOK
501
but it is evident that no pedagogie skill will help a teacher who does
Dot know the subject which is to he taught..
Our unfortunate attitude toward mattcrs of standard and non-
standard speech (U correct English ") is largely kept up by Our
schools. Their attitude is authoritarian; fanciful dogmas as ta
what is "good English" are handed down by educational author-
ities and individual teachers who are utterly ignorant of what is
involved - dogmas such as the shall-and-will mIes or the alleged
"incorrectness" of wcll-established locutions (l've gat it) or con-
structions (the hause he lived in). l\Jeanwhile the differences between
standard and prevalent non-standard forms (such as 1 saw il : 1
seen il) are made the subject not so much of rational drll as of
preachment about" ignorance," "carelcssnoss," and" had associa-
tions." Ali of this, moreover, is set in a background of pseudo-
grammatical doctrine, which detines the categories of the English
language as philosophical truths and in philosophical tenus ("a
naun is the name of a person, place, or thing," "the subject is that
talked about," and so on).
The chief aim, of coursc, is literacy. Although our writing is
alphabetic, it contains sa many deviations from the alphabetic
principle as ta present a real problern, whose solution has bcen in-
definitely postpaned by our edueators' ignorance of the relation of
writing to speech. Nothing courd he more discouraging than ta
read our ueducationalists' " treatises on methods of teaching
children ta read. The Hize of this book does not permit a discussion
of their varicties of confusion on this subject. The primers and
first reading books which embody those doctrines, present the
graphie fonns in a mere hodge-podge, with no rational progression.
At one extreme, there is the metaphysieal doctrine which sets out
to connect the graphic symbols direetly wi th "thougil ts" or
"idcas ,., - as though thesc symbols \Voro correlatcd with objects
and situations and not with speech-sounds. At the other extreme
arc the so-called "phonic" methods, which confuse lcarning to
read and '''Tite with learning; to speak, and set out t.o train the child
in the production of soumis - an undertaking cornplieated by the
crassest ignorance of elernentary phonctics.
Pedagogues must determine how rcading and writing are ta he
taught. Their study of eye-moverncnts is an instance of progress
in this direction. On the other hand, they cannot hope for suc-
cess until they inform themselves as to the nature of writing. The
persan who learns to read, acquires the habit of responding to the
sight of letters by the utterance of phonemcs. This does not menn
that he is learning ta utter phonemes; he can be taught 1.0 read
only after his phonemic habits arc thoroughly established. f
course, he cannot utter phonemes in isolation; to make him re-
spond, say, to the letter b by uttcring the phonemo lb], which in
the English phonetic pattcrn cannat be spoken alonc, is to oreate
a difficulty. The co-ordination betwecn letters and phonemes,
accordingly, has t.o be established as an analogic process by prae-
tice on graphs in which the symbols have a uniform value, sueh
aS bai, cal, fat, hat, mat, pat, rat, sai - can, Dan, fan, man, pan, ran,
tan, van - bib, fib, rib and so on. The real factor of difficulty is
the host of irregular spellings which will romain, no maLter what
values are assigned as rcgular. Two deviccs obviously dernand
to be tried. One is to teach children to rcad a phonctic
scription, and to turn ta traditional writing only after the
dal reading habit has been set up. The ot.her Is to begin with
graphs that contain only one phonemic value for eaeh lettr'r-
sds sueh as were illustratcd above and eUher to postpone other
bTfaphs until the elementary habit has been fixed, or else to intro-
duce thern, in sorne rationally planned way, at earlier points.
The irregular graphs should bc prescnt.ed systematically (e.g. si-
lent gh: fight, light, might, night, righl, sight, tight; a for [J] bofare
1: aU, ball, call,jall, gall, hall, laU, wall, hall, malt, salt, bald, fatse).
It may prove advantageous to use some distinguishing mark
(such as differcnt COIOTS) for silent letters and for lctters in irreg-
ular phonemic values. The methods of procedure, the order of
presentation, und the varions minor devices can be deterrnined
only by experiment; from the outset, however, one must know
what one is trying to do.
28. 3. The difficulty of onr spelling great1y delays elementary
education, and wastes even mueh time of adults. \Vhen one sees
the admirably consistent orthographies of Spanish, Bohemian,
or Finnish, one naturally wishes that a similar system might be
adopted for English. It is not true that to change our
phy would be to "change our languagc": our language is the same,
regardless of how we write iL In the long run, to be sure, the
thography docs cause sorne linguigtic alterations ( 27.6); esthetI-
cally - and this is here the only consideration - we should
gain by eliminating the factor of ugly spelling-pronunciations.
502 APPLICATIOKS AND OUTLOOK
APPLICATIONS AND OUTLOOK
503
It is an error, also, 1,0 suppose thal Englsh is somchow an
phonctic language," which eannot be consistently symbolized
by alphabetic writing; like ail languages, English maves within a
precisely dcfinable mnge of phoncmic units. IL would be nes-
sary only ta reach Sorne compromise between the regional types
of standard English pronunciation; thus, the [1'] of types like Cen
tral-\Vcstcm American wonld have ta be kept, bec!1nso it gives
the simplost phonomie analysis for forIlls liko British red [red],
far [fa:], bird [b<l:d], biller ['bit;;]. On the ather hand, the South-
em British distinction of [t:] as in bad and [a] as in bath would evi-
dently have 1,0 be maintained. IL is wrong to suppose that writing
would be unintdligible if homonyms (e.g. pear , pair, pare or
pl:ece, peace) were spelled alike; writing whch reproduces the
phonemcs of speech is as intelligible as speech. :\Toreover, our
present irrcgular writing sins exactly in this respect by using iden-
tical graphs for phonemically diffcrent forms, such as read [rijd,
l'cd], lcad [lijd, Icd], or tear [tijr, tejr]. Lit.emry entertain
the notion that graphie eecentricities, such as the spcllings of
ghost or rhyme, somehow contribute to the connotation of words;
for a small minority of over-literate personfl they undoubtedly
produce the sort of bookiflh connotation which good writers try
1,0 avoid. There would be no serious difficulty about dcvising a
simple, effective orthography for aIl types of standard Englsh;
the use of it would save an enormous amount of time and Jabor,
and, far frorn injuring our language, would mise the goneral level
of standard speech, both by rea;;suring native speakers of non
standard and by removing the tendency to spelling-pronunci-
ations.
The real difficulty is eeonomic and poltieal. A new orthography
would within fifty years or so turn our whole present stock of
printed texts into sornething difficult and antiquated; for our
grandchildren the printed forms of today would bear the saHle
quaint connotation that Chauccrian spellings bear for us. The
confusion and expense of reproducing al! the more useful texts
would be cnormous. l\lorcover, the change itself, cxtending to
every printer and every school-teacher (not to speak of the public
at large), would dcmand a uniformity of co-operation in changing
deep-seated habits that far transccnds our present political and
administrative powers. Same years ago there was a movement
to "rcform" our spcllng by a series of le8ser changes. Small
changes have worked weil for orthographies like the Spanish, Ger-
man, Dutch, Swedish, or Russian, where the irregularities were
few and could he removed or noticeably lessened by a few simple
adjustments. In our case, however, fragmentary changes can
only incrcase the trouble; for instance, the spelling of no English
word in the present orthography ends with the letter v; to omit a
final silent e after v in some words (writing, for instance, hav for
have), but not in others, is a douhtful expedient. As long as our
main habits are kept up, minor alterations only make things
harder. We may expeet that at sorne time in the future our so-
cial organism will reach a degree of co-ordination and flexibility
where a coneerted change becomes possible, or eise that
ieai devices for reproducing speech will supersede our present
habits of writing and printing.
28.4. At a later stage in schooling we eneounter the many-
sided problem of foreign-language teaching. For the sake of what
is called cultural tradition or continuity, sorne part of the popu-
lation ought to be famiIiar with ancient languages, especially
with Latin and Greek. For the sake of contact with other nations,
and, especially, to keep up with technologic and scientific prog-
ress, a fairly large body of persons must understand modern for-
eign languages. The large part of the work of high schools and
oolleges that has been devoted to foreign-language study, includes
an appalling waste of effort: not one pupil in a hundred learns to
speak and understand, or even ta read a foreign language. The
mere disciplinary or "transfer" value of learning the arbitrary
glossemes of a foreign language can he .safely estimated at almost
nil. The realization of ail this has led to much dispute, particu-
larly as to the methods of foreignMlanguage teaching. The various
"methods" which have been claborated differ greatly in the mere
exposition, but far less in actual classroom practice. The result
depfmds very little upon the theoretical basis of presentation, and
very much upon the conditions of teaching and on the competence
of the teacher; it is only necessary to avoid certain errors to which
our tradition inclines.
A minority of the population stf!.Ys in school long enough to
reach the stage where foreign-language instruction begins. In
the old days, this minority was condcmncd en bloc ta study Latin
and Greek. The bitter struggle against the abandonment of this
custom seerns unwarrantcd, in view of the fact that the pupUs
504 APPLICATIONS AND OUTLOOK APPLICATIONS AND OUTLOOK 505
learned ta read neither of these languages. There remains the
fairly widespread four years' Latin course of our high schools;
apart from other factors, its ineffectiveness is explained by the
fact that scarcely any of the teachers have a reading knowledge
of Latin. The modern foreign languages are better taught, 00-
cause sorne of the teachers know the subject; here too, however,
the results are seareely good cnough ta counter a movement for
abolishing the instruction. Even as it is, very few persons, even
of our middle-class population, have a useful command of any for-
eign language. Whether the nurnber of such persons should be in-
creased, and, if 80, how the selection is to be made, js a large-scale
educational problem. We are far from the point where this is de-.
termined by the pupil's aptitude rather than by his parents'
economic means, eombined with chance or whim. In particular,
we could gain by having children of forcign background study
the language they had heard at home.
Another question of generaI bearing is that of the student's
age. Our eight years' grammar-school course reprcsents a down-
right waste of something lke four years of every child's time. The
European, after four or fivc years of elementary schooling, cnters
upon an eight or nine years' course in a secondary school, in which
he obtains his general education; at the end of this, he is ready ta
take up professional studies. At about the same age, the American
has had only four years of high-school study, and, ta get a general
education, must still go through a four years' college course.
In all respects except formaI education, he is tao mature ta find
satisfaction in general and elementary studies; accordingly, he
turns, instead, tD the snobberies and imbeeilities which make a
by-word of the Ameriean college. The four years' delay whieh
appears plainly in the history of the students who go on into pro-
fessional study, is as serious, if less apparent, for the great majority
who do not, and works most advcrsoly upon the effectiveness of
foreign-languR!!;e study. The eight years' grammar-school course
has OOcome something of a vested interest of administrators and
educational experts; there secrns ta be little hope of beginning
secondary-school studies, and foreign languages in particular, in
the fifth or sixth year of schooling. Yet it is probably to this
eartier beginning that wc must attribute the vastly greater success
of foreign-language instruction in Europe. The formaI and repe
titious nature of this study, the necessarily simple content of
the reading-matter, and the need of make-beIieve, aIl work in favor
of young children. The pupil who takes up his first foreign lan-
guage at high-school age or later, is likely to substitute analysis
for mere repetition, and thus tD meet halfway the incompetent
teacher, who talks about the foreign language instead of using it.
Between the two, they have kept alive the eightcenth-century
scheme of pscudo-grammatical doctrine and puzzle-solving trans-
lation.
The goal to he sought in an ancient language, and, for many
students, in a modern, is the ability to read. This circumstance
serves tao often as an excuse for slovenly teaching. A student who
does not know the sound of a language, finds great difficulty in
learning to read t. He cannat remember the foreign forms sa
long as they figure for him as a mere jumble of letters. Asidc
from the esthetic factor, a clear-cut set of phonetic habits, whether
perfectly correct or not, is esscntial ta fluent and accurate reading.
For the students who are to speak the foreign language - and
they should be more numerous than they are - this question
requires no argument.
The matter that is ta he presented, the thousands of morphemes
and tagmemes of the foreign language, can he mastered only
by constant repetition. The lexical phase, being the more exten-
sive, presents the greater difficulty. Every form that is introduced
should he repeated many times. Many of our text-books are prof-
ligatc in their int.roduction of new words, and fail ta let them re-
eur in later lessons. Recent experience has shown the tremcndous
gain that results from control of the lexical matter: textbook-
writer e-nd teacher should know exact.ly when a new lexical unit
(in rnost instances, a new word) is introduced, and keep exact
track of its recurrences, which must he froquent. Word-formation,
the stepchild of traditional school grammar, must play an im-
portant part in the presentation of sorne languages, such as
Latin or German. The meaning of the foreign forms is hard to
convcy. Translation into the native language is bound t mislead
the learner, bccause the semantc units of differcnt languages do
not m a V ~ h , and OOcause the student, under the practiscd stimulus
of the native forrn, is almost certain to forget the foreign one.
The nucleus of the foreign language should he presented in con-
ncction with practical abjects and situations - say, of the class-
roorn or of picturcs. Mueh can be gathcred from the contexts of
506 APPLICATIONS AND OUTLOOK
APPLICATIONS AND OUTLOOK 507
reading, provided the native speeeh-forms are kept as remotely
as possible in the background.
Grammatical doctrine should he accepted only where it passes
a test of uscfulncss, and even there it should be re-shaped to suit
the actual nccd. In Latin or German the case-forms, and in Latin
or French the vcrb-forms, are essential to understanding, but the
traditional presentation is uneconomic and confusing. The mcmo-
rizing of paradigms, cspccially, produces collocations of forms that
bear so little relation to actual speech as to he nearly worthless.
It is essential, in al! linguistic phases of education, that the
practical hearing be kept in view. The content of what is read in a
foreign language should show the life and history of the foreign
nation. Above aIl, what is read or spoken should he weIl within the
competence of the learner; solving puzzles is not language-Iearning.
28.6. The application of linguistics to the recording and trans-
mission of speech, as in stenography or codes, depends largcly upon
the phonemic principle and requires no special discussion. Thcre
is one undertaking, howcver, which would seem to demand aIl the
rcsources of our knowledge, and more to boot, and that is the
setting up of a universal language. The advantages of an inter-
national medium of communication are self-cvident. An inter-
national language would not involve anyone's giving up his native
speech; it would mean only that in every nation there would he
many foreign speakers of the international language. Wc should
nccd to agree only upon some one language which would bc studied
in every country. It has bcen argued that actually existing lan-
guages are difficult and that the adoption of any one would give
rise to jealousy; accordingly, various artificiallanguages have been
devised. The only type that has met with any success is that of
simplified Latin or Romance, especially in the shape of Esperanto.
Languages of this sort arc semi-artificial. They retain the chief
grammatical categories of the languages of western Europe. They
are morphologically simpler than actuallanguages; the syntax and
the semantic pattern are taken quite navcly from the western
European type, with not cnough analysis to insure uniformity. In
the semantic sphcre, cspecially, wc can scarcely hope to set up a
rational or stable scheme; there arc no natives to whom we could go
for decisions. The political difficulty of getting any considerable
number of people aIl over the world to stndy, say, Esperanto, will
probably prove so grcat that sorne naturallanguage will outstrip it.
English is the most likely choice; it is handicapped chiefly by its
irregular written notation.
28.6. The movement for a universallanguage is an attempt to
make language more useful cxtensively. One might expect the
linguist to try aiso to increase the usefulness of language inten-
sively, by working out speech-forms that will lead to valuable
responses in practicallife. However, it scems that aIl languages are
flexible enough to provide sueh speech.forms without artificial aid.
We can coin and deline scientilic tcrms at will; mathematical
reasoning can be translated into any language. The problem is
not one of linguistic structure, but of practical application. The
logic and dialectics of ancient and medieval times represent a
mistaken effort to arrive at pregnantly useful formulae of discourse.
Meanwhile, a genuinc system of this kind has grown up, in the
shape of mathematics. If wc can state a situation in mathematical
terms, mathematics enables us to re-state it in various simplified
shapes, and these, in the end, lead to a useful practical response.
Theoo procedures, however, depend upon our understanding of the
p r a c t i ~ a l world. The tasks of stating a situation in mathematical
(usually, in numerical) terms, and of deciding what types of re-
statement are consistent (that is, lead to a correct response), are
independcnt of linguistic features. When we have defined two
as 'one plus one,' three as 'two plus one,' and Jour as 'three plus
one,' it is not the linguist who can tell us that we shaH get into
trouble if we now act on the statement that two plus two equals
three. AIl that Hnguistics can do is to reveal the verbal character of
mathematics and save us from mystical aberrations on this score.
H this is true of the relatively simple speech-forms that are
involved in mathematical discourse, it holds good aIl the more of
vaguer and more complicatcd forms of spcech. _Lexical and gram-
matical analysis cannot reveal the truth or falsity of a doctrine;
linguistics can merely make us critical of verbal response habits.
Linguistics cannot tell us whethcr it is helpful to subject one tenth
of the childrcn born into the community to dcsperate handicaps,
hecause their parents failed to go through a ceremony of marriage.
The linguist will mercly note that this matter is hardIy ever dis-
cusood and that until quite rccently its mention was under a tabu.
Assuming that certain practices are injurious, the linguist will
observe that failure to reaet to them by speech (evasion) is a
characteristic symptom. At a higher level, when such practices
508 APPLICATIONS AND OUTLOOK
APPLICATIONS AND OUTLOOK 509
(lome into discussion, we often observe a speech-response that
invokes sorne obviously valuable but irrelevant sanction, as when
the Cree Indian says that he does not speak his sister's name be-
cause he respects her too much. This appeal ta a higher sanction
merges, at a later stage, into rationalization, a habit of diseussing
the practice in apparently reasonable ("cornmon-sense " or "log-
icaI") terms.
Something more like a practieal application of linguistics can
be made in the analysis of popular (and philosophic-scholastic)
beliefs that account for phenornena which in reality are due ta
language. It is remarkablc that popular helief, the world over,
exaggerates the effect of language in superstitious ways (magic
formulae, charms, curses, name-tubu, and the like) , but at the
same time takes no aceount of its obvious and normal effects.
When one persan stimulates another by speech, popular belief
dcems the speech alone insufficient, and supposes that there is
also a transference of sorne non-physical entity, an idea or thought.
When a person describes an act by speech before performing it,
popular hclief is not satisfied with the obvions connection, but
views the speech as the more immediate manifestation of a meta-
physical will or purpose, which determines the subsequent act.
The analogy is then transferred to the conduct of inanimate ob-
jects in the guise of lelealogic expIanations : trees strive toward
the light; water seeks its own lovel; nature abhors a vacuum.
28. 7. Although the linguist eannot go far toward the explana-
tion of practical things, he has the task of c1assifying linguistic
forms wherever their meaning has been determined bv sorne other
science. Thus, we ean voueh for the existence, in language
that has been studied, of a set of cardinal numbers, and we can
investigatC the grammatical structure of these forms, finding, for
instance, that arrangements in groups of ten, decimal systems,
are decidedly widespread. The anthropologist tells us at once
that this is due to the habit of counting on one's fingers. Both
the restriction of our extra-linguistic knowledge and, what con-
cerns us more, our lack of accurate and complete information
about the languages of the world, have so far frustrated attempts
at general grammar and lexicology. Vntil we can carry on this
investigation and use its results, we cannot pretend to ::.ny sound
knowledgc of communal forms of Imman behavior.
Adequate descriptive information about languages is a pre-
requisite for historical understanding. It is apparent even now
that we can sec historical change in human affairs most intimately
in the change of language, but it is evident also, that we shaH
have ta know far more both of practical (that is, extra-linguistic)
events and of linguiste changes that have actually occurred, be-
fore wc can reach the level of scientific classification and predic-
tion. Even now it is clear that change in language tends toward
shorter and more rcgularly constructed words:
shortens the ward, and analogie change replaces irregular deriva-
tives by regular. The speed and the consistent direction of this
process differ in different times and places. Starting from a
mon parent language, wc find modern English with greatly short-
ened words and simple morphology, but Lithuanian with fairly
long words and a complex morphology. The result of this sim-
plification seems to be a greater number of words in response to
lkc practical situations; modifying and relational features and
substitute forms that were once expressed by affixes or other
morphologie features, appear later in the shape of separate words.
The ultimate outcome may he the state of affairs which we sec in
Chinese, wherc each word is a morpheme and every practical
feature that receives expression recCvcs it in the shape of a word
or phrase.
The methods and results of linguistics, in spite of their modest
scope, resemble those of natural science, the domain in which
science has been most successful. It is only a prospect, but not
hopelessly remote, that the study of language may help us toward
the understanding and control of human events.
NOTES
Full titles of books and journals will be round in the Bibliogrnphy at the
end of these Notes.
CHAPTER 1
History of linguistie studies: Pedersen, Li1lfJUwtie science. Older period:
Benrey. Indo-European studies: Delbruck, Einleitung; Streitberg, Geschiehte.
Germanie studies: Raumer; Paul, Grundrssl.9; W. Streitberg and V. Michels
in Streitberg, Ge8chichte 2.2. The history of a single scholastic tradition:
.Tellinek, Ge8chichte der deJdschen Grammatik. Sorne interesting details in the
first chapter of OerteL
1. 2. The ancients' philosophical views about language: Steinthal, Guchichte,
The anecdote about the children in the park: Herodotus 2.2.
The etymology of lilhos in Eymologiccm magnum (ed. T. Gaisford, Oxford,
1848) 565.50; that of lueu.s from Quintilian 1.6.34, and in Lactantius Placidus'
gloss on Statius, Achillei8 593 (ed. R. Jahnke, Leipzig, 1898, p. 5(2).
Greek grammariana: G. Uhlig, Grammalici Graeci, Leipzig, 1883 ff.; Herodian
edited by A. Lentz, Leipzig, 1867 ff.
1.3. Theories about the origin of language: Steinthal, Ursprung; Wundt,
Sprache 2.628.
The epigram about etymology is attributed ta Voltaire by Max Mller,
Lectures an the 8cience of langu.age: Second serie8 (London, 1864), p. 238; 1
have sought it in vain in Voltaire's writings.
Latin grammarians: H. Keil, Grammalici Latini, Leipzig, 1857 ff.; H. Fu.
naioli, Grammalicae Romanae fragmenta, Leipzig, 1907.
Medieval work in Latin grammar: Waekernagel, Varle8ungen 1.22;
Thurot.
The Port-Royal grammar was written by A. Arnauld and C. Lancelot; it
appeared in Paris in 1660, a second edition in 1664, another in Brussels in
1676; I have seen only this last (at the Newberry J..ibrary, Chicago); modern
reprints with additions appeared at Paris in 1803 and 1810.
Eighteenth-century normati ve grammar: Fries; Leonard (very full account).
The shall and will doctrine: C. C. Fries in PMLA 40.963 (1925).
Pallas, Peter Simon, Linguarum totius orms l'ocabularia comparaliva,
St. Petersburg, 1786-89, two volumes (Newberry Library, Chicago). l have
not seen the second edition. An alphabetieal index, anonymous (accorcling
ta the Newberry Library catalog, by Theodor Jankovic von Mirijevo) under
the title Sravnilel'nyj s/ovar' vsex fazykav ~ . nareif, in four volumes, appeared
in St. Petersburg, 1790--91. Vuk Stefanovich (Karadjich) published a sup-
plement (Dodalak) at Bec in 1822, correcting the Serbian and adding Bulgarian
forms (oopy in Newberry Library).
Adelung-Vater's Milhridmes was named after the first book of its kind, an
alphabetieallist of languages, with a very few specimens, by Konrad Gessner
511
512 NOTES
NOTES 513
(1516-65), which appeared in Zurich in 1555; a. new edition of this, with a
commentary by Kaspar Waser, Zurich, 1660 (both editiODS in Newberry
Library).
Junius, F., QuatUOT D. N. Jeau. Chmti Euangelim-um versi0ne8 perantiquae
duae, Dordrecht, 1665.
Hickes, G., lmslitldionea grammaticae Anglo-Saxonicae et Moesogothiroe,
Oxford, 1689; Antiquae lileraturae septentrionalis !ibn duo (ldnguarum veU.
theeaurm), Oxford, 1705.
1.5. On the philological.linguistic work of the Chinese, Karlgren, Philology.
On Hindu grammar, Belvalkar; bibliography in Lg 5.267 (1929).
1. 6. Jones' address appeared in Asiatick reaearches (Calcutta, 1788) 1.422;
this volume has been reprinted, repeatedly, as volume 1 of the Transacliorns of
the Royal Asialie society of Bengal.
1.7. Etymology: Thurneysen; Thomas 1.
On Brugmann: W. Streitberg in 1J 7.143 (1921). On Delbrck: Hermann.
1.8. The second edition (1886) of Paul's Prinzipien served as the baais for
the excellent English adaptation by Strong-Logeman-Wheeler. On Paul's life
and work: W. Streitberg in IJ 9.280 (1924).
1. 9. On Leskien: W. Streitberg in 1J 7.138 (1921); K. Brugmann in Berich/e
Leipzig 68.16 (1916). On Bhtlingk: B. Delbrck in 1F Anzeiger 17.131 (1905).
On de Saussure: A. Meillet in BSL 18.C!xv (1913).
CHAPTEa. 2
Psychologists generally treat language as a side,issue. General discussion:
Marett 130; Boa.s 1.5; Wundt, Sprache; Sapir; Allport; de Laguna; and,
especially, Weiss.
2.1. The term philology, in British and in older American usage, is applied
not only ta the study of culture (especially through literary documents), but
also to linguistics. lt is important to distinguish between philology (German
Philowgie, French philologie) and linguislice (German Sprachurissenschajt,
French linguistique), since the two etudies have little in common. On the
confusion in English usage: H. Pedersen in Liueri8 5.150 (1928); G. M. Bolling
in Lg 5.148 (1929).
2.4. The popular bclief seems to be that in thinking we finaJly suppress
the speech,movements altogether, Hke the horse in the story, that finally
learned to go without fodder.
The use of numbers js characteristic of speech-activity at i18 best. Who
would want to live in a world of pure mathematics? Mathematics is merely
the best that language can do.
2.5. The child's learning of language: Allport 132; \Veis.s 310. Almost
nothing is known uecause observers report what the child says, but not what
it has heard; sa Stern; Preyer; Bhler. Learning to speak is the greatest feat
in one's life: Jespersen, Language 103.
2.8. Disturbances of speech: Kussma.u]; Gutzmann, Sprachheilkunde;
Wilson; Head; Travis.
2. 9. Gesture: \Vundt, Sprache 1.143.
2.10. The universe symbolically reduced to liurary dimensions: A. P. WeillS
in Lg 1.52 (1925).
CHAPTER 3
3.2. The !argest speech-comrnunities: Jespersen, Growth 252; L. Tesnire
in Meillet, Langues. For the languages of lndia, Tesnire's figures deviate
slightly from those of Grierson (volume 1); both estimates are based on the
census of 1921.
3.3. Scx...<:Jifferences: Jespersen, Language 237; E. Sapir in Donum Schrij-
nen 79.
3.9. Saer discusses children's shift of language in Wales. Sacr uses the term
bilingucl of children who have shiftcd from Wclsh ta English - an
nate extension; thus, in spite of Saer's careful distinction (32 ff.), West,
lingwdiem confuses the situation of these children "With genuine bilingualism,
and both of these things "With the position of a child who hears an entircly
foreign language in school.
On rcal bilingualism: Ronjat; a realistic fictional account, based on the au,
thor's childhood, will he found in George Du Maurier's Peter lbbeteon, pub-
lished in Harper's new rrn:mthly magazine, volume 83 (1891) and in book form.
CHAPTER 4
F. Mller surveys the languages of the world, giving grammatical sketches
and bits of text. Finck, SprachsUimme gives a bare list. MeilIet,Cohen is a
collection of surveys by specialists; it contains maps and sorne bibliography.
W. Schmidt has excellent bibliographies and, in a separate atlas, several maps.
Useful charts also in Kroeber; for America in Wissler. India: Grierson. Af
rica: Meinhof, Modm-ne Sprachfwschung.
4. 3. Relation of Hittite to Indo,European: E. H. Sturtevant in Lg 2.25
(1926); TA PA 40.25 (1929); AJP 50.360 (1929); a different view: W. Petersen
in AJP 53.193 (1932).
4.4. Languages now extnct: Pedersen, ldnguistic science. A few legible
but unintelligiblc inscriptions represent the language of the Picts in Scotland;
it is uncertain whether Pictish was (CcIUc) or not; sec Hubert
247.
4.8. Deny in Meillet,Coben. Chinese dialccts: Arendt, Ilandbu.ch 258;
340; map.
4.9. Papuan: S. H. Ray in Festschrijt Meinhof 377.
4.10. On the grouping of the Algonquian languages (in the tert listed geo,
graphically) see T. Michelson in BAE Annual reporl 28.221 (1912).
CHAPTER 5
5. 1. Sernon/ice, from eemantic 'pertaininp; to meanin!/;.' These words are
less c1umsy than eemasiology, semalriological. Literally, then, semantics is the
study of mcaning. If one disregards the speech-fonns and tries to study mean-
ing or meanings in the abstract, one is reully trying to study the uni verse in
general; the term semantics is sometimes attacbed to such attempts. If one
studies speech-forms and their meanings, semantics is equivalent to the study
of grammar and lexicon; in this sense l have defined it in the text.
5.2. Laboratory phonetics: Rousselot, Principes; Scripture;
Calzia, Einfi.ihrung (excellent introductory survey); Phonetik
514
NOTES NOTES 515
(theoretical outline); Gutzmann, Physiologie; Russell; Fletcher (especiall
y
for analysis of sound-waves and on the ear); Paget (except Chapters 7,8 9
and Appendix 8, which denl inadequately with unrelated topies). '
6.3. The phoneme: Baudouin de Courtenay 9; rie Saucaure 55; 63; E. Sapir
in Lu 1.37 (1925); Bee also L(J 2.153 (1926); Modem philology 25.211 (1927)'
H. Pedersen in Lt"tteris 5.153 (1928). '
6. 8. The chief systems of phonetic transcription are assembled by Heepe.
Visible Speech: Sweet, Primer. Analphabetic Notation: Jespersen, Lehrbuch.
Other systems: Lepsius; Lundell; Bremer; Phonetic transcription.
International Phonetic Association Alphabet: Sweet, Handbook; CoUeded
papers 285; Pa.'lI:Iy-Jones; Jespersen-Pedersen. Discussion and texts in Maitre
phontique.
5.10. On transiiteration and the like: G. M. Bolling and L. Bloomfield in
Lu 3.123 (1927); Palmer, Ramanization.
CHAP'tER 6
1. phonetics; Passy, Phontique (the best introduction); Sweet,
Prtmer,o Rippmann; Soames; NotH-Armfield. Larger works: Sievers, Gruw:J,.
zige (the cla.'ll:lical text); Jespersen, Lehrbuch,o Vitor, Elemente.
English: Krapp; Kenyon; H. Kurath in SPE 30.279 (1928);
L. Strong lU RP 5.70 (1928); Maitre phontique 3.5.40 (1927); bibliography:
H. Kurath in Lg 5.155 (1929).
British English: Sweet, Sou7Ui,o Jones, Ou/line; Palmer, First cmM'se; Lloyd.
Phonetic dictionaries: Michaelis-Jones; Jones, English pronouncing dictmary,'
Palmer-Martin-B1andford (the American part is inadequate).
German: Hempl; Vitor, German pronu.ru:ialion, AU8Sprache; AU8Sprache-
wiirierbuch; Bremer; Bieha.
French: Passy, Sons,' Sou.7Ui,o Passy-Rambeau; G. G. Nicholson; Michae1is-
Passy; Passy-HempL
Dutch: Kruisinga, Grammar,o Scharp. Danish: Jespersen, FoneI.ik, Forch-
hammer. Swedish: Noreen VS. Spanish: Navarro Tomlls. Russian; Trofunov.
Jones. North Cbinese: Guernier.
6.2. African languages: Meinhof, Moderne Sprachforschung 57.
6.3. Voiced h: Broch 67; E. A. Meyer in JVS 8.261 (1900).
Resonance: Paget.
6.6. Domals: E. Sramek in RP 5.206 (1928); Nol-Armfield 99. Palatal
stops: 91. Glottal stop: Jespersen, Fonetik 297. Glottalized
stops: Boas 1.429; 565; 2.33. South-German stops: Winteler 20.
6.7. Trills: Jespersen, Fonelik 417; Lehrbuch 137; Bohemian: Chlumsky in
RP 1.33 (1911). Tongue-Hips: Lunde1l48; Noreen VS 1.451.
8. German spirants: MaUre ph<m/ique 3.8.27 (1930). Arabie glottal
spirants: Gairdner 27; W. H. WorreIl in Vox 24.82 (1914); G. Panconcelli-
CaiJ:iain Vox 26.45 (1916).
6.10. Laterals: Sweet, Collected papers 508; Boas 1.429; 565; Broch 45.
6.12. Vowe1s: RusseIl, Vowel,o Paget; C. E. Parmenter and S. N. Treviiio in
Quarlerly jour1'U 18.351 (1932). Vowel systems: N. Troubetzkoy in Travaux
1.39 (1929). For the English-apeaker, study of the French voweIs is especially
enlightening. H. Pernot in RP 5.108; 289; 337 (l928).
CHAl'TER 7
7.2. Mora: E. Sapr in Lg 7.33 (1931).
7. 4. For the oontl'll.8t between American and British treatment of unstressecl
vowels, ace the introductory remarks of Palmer-Martin-Blandford; thei:r
general outlook, however, will scarcely find acceptance.
7.5. A name: an aim: many examples are assembled by D. Jones in Matre
phontique 3.9.60 (1931).
7.6. Fitch in (British) English: Jones, Curves; Palmer, IntonatW1i; Arm-
strong-Ward. German: Barker; Klinghardt. French: Klinghardt-de Fourme-
straux.
Eduard Sievers (1850-1932) gave many years to the study of
speech-patterns; summary and bibliography: Sievers, Ziele; IpsenKarg.
7.7. Word-pitch in Swedsh and Norwegian; Noreen VS 2.201; E. Selmer in
Vox 32.124 (1922). In Japanese: K. Jimbo in 8SGS 3.659 (1925). North
Cbinese; Guernier; Karlgren, Rroder. Cantonese: Jones-Woo. Lithuanian:
R. Gautbiot in Parole 1900.143; Leskien, Lesebuch 128; in Serbian: R. Gauthiot
in MSL 11.336 (1900); Leskien, Grammatik 123; in African languages: E. Bapil
in Lg 7.30 (1931); in Athabascan: E. Bapir in Journal de la Socit 17.185
(1925).
7. 8. Palatalization: Broch 203; velarization: 224.
CaAP'tER 8
8.1. An example of two languages with similar 80undB in entirely different
phonemic distribution: E. Sapir in Lg 1.37 (1925).
8.7. Relative f.equency of pbonemes: Dewey; Travis 223; Zipf. The con-
clusions of Zipf do Dot acem warranted by his data; ace also bis. e8811.Y in Harvard
atudies 40.1 (1929).
CaAP'tER 9
Many of the examplea in the text are taken from the excellent popular
treatise of Greenough-Kittredge. Sec also Bral; Paul, Prinzipien 74; Me-
Knight; Nyrop Uv; Darmesteter, Vie; Hatzfeld. For individual Englsh words,
ace NED. Position of the study of meaning: L. Weisgerber in GRM 15.161
(1927). The mentalistic view of meaning: Ogden-Richards. Bibliography:
Collin; G. Stern.
9.1. Kinsbip terms; L. Spier in Um"versily of Wash.glon publicatioTl-S 1.69
(1925). Demonstration: Weiss 21. The definition of apple is taken from
Websler' MW inl.ernal.i<m.al dictionary, Springfield, 1931.
9.7. Facetious malformation: M. Reed in American speech 7.192 (1932).
Over-slurred formulas: Horn, Sprachkiirper 18.
9. 8, Sec especially Collin 35.
9.9. Examples of speech-leve1s: Noreen VS 1.21, with table on p. 30.
Slang: Farmer-Heuley; Mencken, The American Language.
9.10. Tabu: Meillet, Linguistique 281; G. S. Keller in Streitberg Featgabe
182.
9.11. Jespersen, Language 396; Hilmer; Wheatley. Hypochoristic forms:
Sundn; Rotzoll; L. Mller in Giessener Beitriige 1.33 (1923).
516 NOTES
NOTES
517
CHAPTER 10
On the structure of Sweet, Pradical study; de Saussure; Bapil';
Hjelmslev; see also Lg 2.153 (1926). The best example of descriptive analysis
is the Hindus' work on Sanskrit; see note on 1.6. English; Jespersen,
Grammar; Philosophy; Kruisinga, Handb(){)k; Poutsma, Grammar; German:
Cuime; French; Vllrious languages are analyzed in Boas and by
Finek, Haupttypen.
10. 1. The asterisk before a form (as, *cran) indicates that the writer has
Dot heard the form or found it attested by other observera or in written docu-
ments. It appears, aecordingly, befofl.l forms whose existcnce the writer is
denying (as, *ran John), and before theoretieally construeted fonna (sueh as
*cran, the theoretically posited independent word corresponding to the eom-
pound-membcr cran- in cran/>erry). Among the latter the most important are
ancient speech-forms not attested in our written records, but reconstrueted by
the lingllst.
CHAPTER Il
In this and the following chapters, cxamples from less familiar languages
have been taken from the fol!owing sources; Arabie, Finck, Hauptlypen; Bantu
(Subiya), samej Chinese, same, and Arendt, Einfhrung; Crcc in Alti 2.427;
Eskimo, Finck, Hauptlypen and Thalbitzer in Boas 1.967; Fnnish, Rosenqvistj
Fox, T. Michelson in various in IJAL 3.219 (1925); Geor-
gian, Finck, HaupUypen; Gothie, Streitberg, ElemenlaTbw:;h,' BoI"thwick;
Menomini, Proceedinys 21s1 1.336; Polish, Soerensenj Russian, Berueker,
Grammatik; Samoan, Finek, Haupttypen; Snnskrit, Whitney, Grammar;
Tagalog, Bloomfield; Turkish, FillCk, Hauptlypen.
11. 1. Traditiollally and in sehool gmmmar, the term senlnu;e is used in a
much narrower value, to the subject-and-predicllte sentence-type of
the Indo-European languages. If we adhered to use, we should have to
coin a new tcrm to designate the largest form in nn utterancc. The older
definitions are philosophieal rather than linguistic; they are assembled by Ries,
Salz. The defiuition in the text is due to MeiIJet, lntrodw:;tion 339; compare Lg
7.204 (1931).
11. 2. Impersonal are Ilsually with pseudo-imper-
selllai types, which contain a pronominal aetor it's raining, 15.6).
11. 5. Difficulty of making Passy, Phontique 21.
11. 7. The Freneh-speaker oecasionally uses stress to mark word--divisions
(Passy, Sons 61), Lut this nse is not it is comparable to our or
the Frenehman's occasional pause between words. The word-unit in South
German: Willteler 185; 187.
CHAPTER 12
On syntax: Morris; Wackcrnagel, Vorlesungen; Blmelj Jespersen, Phi-
wsophy. For English, Leside the books cited for Chapter tO, see Curme-Kurath;
for German, Paul. Grammalik.
12.1. Definition of Ries, Syntax.
12.4. Pitch and stress in Chinese sandhi: Karigren, Reader 23; examples
from Arendt, Einjhrung 14.
12.10. Ranke; Jespel'f!Cn, Philosophy 96. . ..
12.12. Bibliography of writings on word-order; E. Schwendtner lU Worter
und Sachen 8.179 (1923); 9.194 (1926).
CHAPTER 13
Description of a complex morphologie Greek):
13. 1. Classification of languages accordlUg to theIr. morphology. Steinthal,
Charakteristik; Finck, KlasBifikation; HaupUypen,' SapIT.
CHAPTER 14
14.1. COillPounds: Knzel; Darmesteter, Trait.
14.4. Inclusion of words between memLers of eompounds; T. Michelson
in /JAL 1.50 (1917).
14.6. Exocentric compounda; Uhrstr0m; Last; Fabian. .
14.7. Denominative verbs: Bladin. On drunken: drunk the
M. Deutschbein in Streitberg Festgabe 36. Male and female III Enghsh;
Knutson.
14.8. Conerete suffixes of Algonquian in F'est8chrif/ Meinhof 393. Incor-
poration; Steinthal, Charakteristik 113. English ftip: flap: flop, etc.; Warnke.
CB"APTER 15
IG. 6. Impersonal and pseudo-impersonal types, bibliography; Ljunggren.
!li. 7. Annatom Island: F. Mller 2.2.73.
CHAPTER 16
Some dietionaries;
NED; Bosworth-Toiler; Stratmannj Grimm, Warler?uch;
Benecke-Mller-Zarncke; Lexer; Graff; Dutch: VerwlJB--Verdam; de. Vnes-te
Winkel; Dauish: Ordbok; OId Norse: Cleasby-VIgfusson;
Fritznerj Rllssian; Biattller; Latin: Thesaurus; .French: Hatzfeld-Darmesteter-
Thomas; Sanskrit: Bhtlingk-Rothj Chinese: Giles.
16.6. English aspects: Poutsma, Character8; Jespersen, Grammar 4.164;
Kruisinga, Handbook 2.1.340.
'-6.7. Number of words used: Jespersen, Language 126; Growth 215.
Relative frequency of words; Zipf; Thorndike. .
16.8. Kham Bushman numerals: F. Mller, Grundnss 4.12; nUIlleraIs,
bibliography: A. R. Nykl in Lg 2.165 (1926).
CHAPTEIl. 17
Lingllstie change: Paul, Prinzipien; Sweet, His/ory of language,' Oertelj
Sturtevantj de Saussure.
History of various languages: . .., . .
The Indo-European family; the best lIltroductlOn IS Meillet, Introdw:;twn,
standard refcrence-book, with bibliography, Brugmann-DelbI"ck;
Brugmann, Kurze vergleichende Gram.matikj more speculative, Rlrt,
Indogermanische Gram17Ultik; etymologlcal ..
The Germanie brandi: Grimm, (stlllmdlspensablel; Streltberg,
518 NOTES NOTES
519
Grammmik; Rirt, Handbuck de.! UrgemULniscke:n; Kluge. Urgermanisck;
etymological dictionary, Torp, Worlchatz.
English: readable introduction, Jespersen, Growth; Sweet, Grammar; His-.
tory of souruh; Horn, Gramma/ik; Kaluza; Luickj Wyld, Hislorical sludy,'
Hi8WTy; Short history; Wright, Elementary; Jespersen, Progress; etymologica.l
dictionaries: NED; Skeat, Dictionary; Weekley, Didionary. Old English:
Sievers, Grammalik; Sweet, Primer; Reader.
German: readable summaries, Kluge, Sprcu;hguckichle; Behaghel, Sprcu;he;
larger works: Wilmannsj Paul, Grammalik; Stterlin; Behaghel, Geschichl.e;
Syn/az; etymologica! dictionary, Kluge, Wtirierbueh. Old High German,
Braunej Old Low German (Old Saxon), Holthalll'leIlj Middle High German:
Michels.
Dutch: Schonfeldj van der Meer; etymological dlctionary, Franck-van
Wijk.
OId Norse: Heuslerj Noreen, Grammalik. Danlsh, Dahlerup, Historie.
Dano-Norwegian: Seipj Torp-Falk, Lydhislorie; Falk-Torp, Syntax; etymologi-
cal dictionaries: Falk-Torp, Wrierbuch; Torp, ()rJ);ok. Swedish: Noreen VS;
etymological dictionary, Tamm; sec also Hellquist.
Gothic: Streitberg, Ekm(;'farbuch; Jellinek, Guchichte der gotischen SprCU;M;
etymological dictionary, Feist.
Latin: Llndsayj Sommerj StolzSchmalz; Kentj etymological dictionary,
Walde.
Romance: introductions, Zaunerj Bourciezj Meyer-Lbke. Einfhrung;
Iarger works: Groberj Meyer-Lbke, Grammatik; etymological dictionary,
Meyer-Lbke, Wr/erbuch. French: Nyrop, Grammaire; Dauzat, HisWire;
Meyer-Lbke, Historische Grammalik. Haliaa: d'Ovidio j Grandgent. Span-
ish: Hanssen j Men6ndez PidaL
Oscan and Umbrian: Buckj Conway.
Celtic: Pedersen, Grammalik. Old Irish: Thurneysen, Handbueh.
Slavic: Mikloaich, Grammalik; Vondrakj Meillet, Slave; etymological dic-
tionaries: Miklosich, Wiirterbuch; Berneker, Wiirierbuch. Russian: Meyer.
Old Bulgarian: Leskien.
Greek: Meillet Aperu,' Brugmann-Thumbj Hirt, HandJ:,uch; etymological
dictionary, Boisacqj ancient dialects: Buckj modern Greek: Thumb.
Sanskrit: WackernageI, Grammalik; etymoIogical dictionary, Uhlenbeck.
Marathi: Bloch.
Finno-Ugrian: Szinnyei. Semitic: Brockelmann. Bantu: Meinhof, Grund-
ziige; Grundriss.
On writing: Sturtevantj Jensen; Pedersen, Linguistic science; Sprengling.
17. 1. Picture messages: Wundt, Sprache 1.241; in America: G. Mallery in
BAE Annual reporl 4 (1886); 10 (1893); Ojibwa. song record in W. Jones,
Ojibwa lexis, Part 2, New York, 1919 (Publications of the American ethnolog-
ical society, 7.2), 591.
17.2. Egyptian writing: Erman. Chinese: Karlgren, Sound. Cuneiform:
Meissner. Runes: Wimmerj O. v. Friesen in Hoops, Realkxikon 4.5.
17.9. Conventional speIIings in Old English: S. Moore in Lg 4.239 (1928);
K. Malone in Curmevolume 110. OccasionaJ spellings as indications of sound:
Wyld, His/ory. Inscriptions: Kent. Re-spelling of Homeric poems: J. Wacker
nagel in Bei/rage zur Kunde 4.259 (1878) j R. Herzogj of Avesta: F. ?Andr:as
and J. WackernageI in Nachrichten Giittingen 1909.42j 1911.1 (especmlly thIS);
1913.363.
17.10. Rimes: Wyld, S/udie8,' theoretical discussion: Schauerhammer. Al-
literation as evidencc: Heusler 11. Inaccuracy of older English phoneticians:
Wyld, His/ory 115.
CHAPTER 18
Comparative method: Meillet, Linguistique 19; Mtlwde; K. Brugmann in
IZ 1.226 (1884).
18. 4. Latin cauda, cuda: Thesaurus under caudal Schuchardt, Vokalismus
2.302j Meyer-Lbke, Einfii.hrung 121. Latin secale; same 1a6. Suetonius:
Vespasian 22;
18.6. Ga.llehus horn: Noreen, Altislandische Grammatik 379. Germanie
loan.words in Finnish: see note on 26.3.
18.7. On K. Verner: H. Pedersen in IF Anzeiger 8.107 (1898). Verner's
d8covery ln ZvS 23.97j 131 (1877).
The acoustic value of the Primitive Indo-European vowcl phoneme which
in our formulae is represented by the inverted letter e, is unknown; linguists
sometimes speak of this phoneme by the name shwa, a term takcn from He-
brew grammar.
Primitive Indo-European form of Latin cauda: Walde under caudal K. Ettr
mayer in ZrP 30.528 (1906).
Hittite: sec note on 4.3.
18.8. The Indonesian example from O. Dempwolff in Zeilchrift fr Ein-
geborenensprachen 15.19 (1925), supplementcd by data which Professor Demp-
wolff has kindly communicated.
18. 11. DiaIect differences in Primitive Indo-European: J. Schmidt; Meillet,
Dialectes; Pedersen, Groupement. Figures 1 and 3 are modeled on those given
by Schrader, Sprcu;hvergleichung 1.59; 65.
18. 13. Hemp: Schrsder, Sprachvergleichung 2.192. Herodotus 4.74.
18. 14. Schrsder, Sprachvergleichung; Mcillct, ln/ror!uctm 364, Hirt, Jndo-
germanen,' Feist, Kultur; Hoops, Wald/Xiume; Hehn; Schrader, Reallexikon.
Germanie pre-history: Hoops, ReaUexkon. General: Ebert.
Terms of relationship: B. Delbrck in AbhanrIlungen. Leipzig 11.381 (1889).
CHAPTER 19
Dialect geography: Jabergj Dau7-at, Goqraphie,' PaWis;
Gamillschegj Millardetj Schuchardt, Klw;llifikation; E. C. Roedder III Ger
manic review 1.281 (1926). Questions of principle in special studies: 1,. Gauchat
in Archiv 111.365 (19(n)j Termcher; Haag; KIoeke; A. Horning in ZrP 17.
160 (1893), reprind in Meisterwerke 2.264.
Discussion of a single dialect: Wintelcrj of an arcn.: Schmcller, Mundarten;
Bcrtoni' J ut7-. Dictionaries: Schmeller, Wiirterbuch; Feilhcrg.
dilllccts: Ellis, volume 5; Wright., DidiQnory; Grammar; Skeat,
Dialeds: P"lJlicalioT1.8 "f !he rllled society; Dialecl nr.tes. On the Amer
iean atlas: H. Kumth in Duect wites ti.liS (19:l0); M. L. Hanley in Dia./.ect
notes (j.nI (t9Jl).
520
NOTES NOTES 521
21
'changed into' and the symbol < means 're.
19.2. With the fifth issue (1931), the German atlas takes up some of
the hitherto omitwd parts of the area. Studes based on the German atlas:
Deutsche Dialekl(}ecgraphie; Teuthonisla.
19.3. Kaldenhausen: J. Ramisch in Deu18che Dialeklgeographie 1.17 62
(1908). '
19.4. Every word has its own history: Jaberg 6.
19. 5. Latin mullum in France: Gamillscheg 51; fallil: Jaberg 8.
. 19.6. Latin sk- in French: Jaberg 5; my figures, taken directly from
lironEdmont's maps, differ slightly from Jaberg's.
19.8. French and Provenal: Tourtoulon-Bringuier. Low and High German:
W. Braune in Beitrti/Je zur Geschichle LI (1874); T. Frings in Beilrage zur
Geschichle 39.362 (1914); Behaghel, Geschichte 156 and map; see also map 3
of \Vrede and the map given by K. Wagner in De!dsche Dialeklgecgraphie 23
(1927).
19.9. Rhenish fan: J. Ramisch in Deulsche Dialektgeographie 1 (1908);
plawa 1 and 2 of Wagner' a study, cited in the preceding note; Frings.
20
20. 2. Germanie consonant-shift: Russel'.
20.3. H. Grassmann in ZvS 12.81 (1863).
20.6. The neo-grammarian hypothesis: E. Weehssler in Festgabe 8uehier
349; E. Herzog; Delbrck, EinleillLng 171; Leskien, Declinalion "xviii; 2;
Osthoff-Brugmann, preface of volume 1; Brugmann, Stand; Ziemer. Against
the hypothesis: Curtius; 8ehuehardt, Laulgesetze; Jespersen, Language; Horn,
8prachki5rper; Hermann, Laulgesetz.
20.7. Tabulations of OId English and modern English correspondenoos in
8weet, lIis!ory of sounds.
20.8. Algonquian forms: Lg 1.30 (1925); 4.99 (1928); E. Sapir in S. A.
Riec 292.
20. 9. English bail, etc.: Luick 387; Bjorkman 36.
20.10. Greek forms: Brugmann-Thumb 143; 362.
20.. 11..Observation of sub-phonemic variants: Passy fl:tude; Rousselot,
ModijicatlOns; L. Gauchat in F(J)jtschrift Morf 175; E. Hermann in Nachrichten
G6Uingen 1929.195. Relative chronology: O. Bremer in IF 4.8 (1893).
21.1. The symbol > means
sulting from.'
21.2. Simplification of finaI consonants: Gauthiot.
21. 3. Latin clusters: Sommer 215. Russian assimilations: Meyer 71.
21. 4. Origin of Irish sandhi : Thurneysen, Handbuch 138; Brugmann.
Delbrck 1.922. English voicing of spirants: Jespersen, Grammar 1.199;
Russer 97.
21. 5. Palatalization in Indo-lranian: Delbrck, Einleiwng 128; Bechtel 02;
Wackernagel, Grammalik 1.137.
21. 6. Nasa!ization in Old Norse; Noreen, Altislandische Grammalik 39.
21. 7. English away, etc.: Palmgren. Irish verb-forms: Thurneysen Rand--
buch 62. '
21.8. Insertion of stops: Jespersen, Lehrbuch 62. Anaptyxis, etc.: Brug
mann-Delbrck 1.819.
21.9. Causes of sound-change: Wundt, 8prache 1.376; 522. Relative fre-
quency: Zipf (see note on 8.7). Experiment misapplicd: J. Rousselot in
Parok 1901.64.1. Substratum theory: Jespersen, Language 191. Homonymy
in Chinese: Karlgren, tudes.
21. la. Types of r in Europe: Jespersen, Fonelik 417. Dissimilation: K. Brug-
mann in Abhandlungen Leipzig 27.139 (1909); Grammont; A. MeiUet in
M8L 12.14 (1903). Assimilation: J. Vendryes in M8L 16.53 (1910); M. Gram-
mont in BSL 24.1 (1923). Metathesis: Brugmann-Delbrck 1.863; M.
mont in MSL 13.73 (1905); in SITeitberg Festgaoe Ill; in Fesischrift
nagel72. Haplology: Brugmann-Delbrck 1.857.
CHAPTER 22
22.2. The Old English word for "become": F. Klaeber in JEGP 18.250
(1919). Obsolescence: 'reichert. .
22.3. Latin apis in France: Gilliron, G(;na1ogie; Meyer-Lbke, Etn-
fiihTungl3. Short verb-forms: A. Meillet in MSL 11.16 (1900); 13.359 (1905);
J. Waekernagel in Nachrichlen G6Uingen 1906.147. English concy NED under
coney; Jaberg 11.
22.4. Homonymy: E. Richter in F(J)jlschrift Kretschmer 167. Latingallus
in southern France: Gilliron-Roques 121; Dauzat, Gographie 05; Gamill-
scheg 40.
22.6. Othello's speech (Act 3, Seene 3) explained in H. H. Furness' New
variorum edilion, volume fi (Philadelphia, 1886).
22.7. Tabu: see note on 9.10.
CHAPTER 23
Analogie change: Wheeler; Paul, Prinzipien 106; 242; Strong-Logeman-
Wheeler 73' 217' de Saussure 221; Darmesteter, Cration; Goeders.
23. 1. vcrsus irregular combinations: Jespcrsen, Philosophy 18.
23.2, Objections to proportional diaf/;ram of analogy: Herman, Laulr;esetz 86.
23.3. English s-plural: Roedler. Latin senati: Hermann, Lautgesc/Z 70.
23. 5. Back-formation: N:icht.enhauser; O. Jespersen io Feslskrift Thomsen 1.
English verbs in -en: Raith. English verbs in -ale: Strong-Logeman-Wheelcr
220.
23. 6. Verbal compound-members: OsthofT; de Saussure 195; 311.
Popular etymology: A. S. Palmer; Andrescn; Hasse; W. v. Wartburg in
Homenaje Menndez Pidal 1.17; Klein 5;); H. Palander in Neuphilologisehe
Miaeilungen 7.125 (1905); J. Hoops in Englische Siudien 30.157 (1906).
23.7. Analogie in syntax: Ziemer; Middleton.
23.8. Adaptation and contamination: M. Bloomfield in AJP 12.1 (1891);
16.409 (1895); IF 4.66 (1894); Paul, Prinzipien 160; Strong-Logeman.
Wheeler 140' L. Pound in Modern language review 8.324 (1913); Pound, Blends;
Bergstrom; G. H. McKnight in JEGP 12.110 (1913); bibliography: K. F.
Johansson in ZdP 31.300 (1899). In pronouns: Brugmann-Dclbrck 3.386.
Psychological study: 'rhumb-Marbe; Esper; Oertel 183. Slips of the tangue:
Meringer-Meyer. Bob, Dick, etc.: Sundn.
522
NOTES
NOTES 523
CHAPrER 24
See the referenOO3 W Chapt& 9.
. 24.3. The wattled wall: R. Meringer in Festgabe Heinzel. .
ln Germanic revicw 1.40 (1926). Words and things: Wiirter 1J/tl../1
3
'hH. Collitz
24.4. Paul, Prinzipien 74. ac en.
24. 6. On hard : hardly, Uhler.
24.6. Marginal meanings in aphoristic forms: Taylor 78.
24.7. Sperber; S. Kroesch in Lg 2.35 (1926); 6.322 (1930)' Modern ph'
lology 26.433 (1929); Stud!1l Collitz 176' Studt"es K' - -b 50' L'
A Za . R . h ' ...., er . atm testa.
. uner ln omtlTlfIlC C Farllchungen 14355 (1903) P f .
worth: Greenough-Kittredge 9. . . a88age rom Word&-
CE;APTER 25
2. F!-rst phonetic adaptation of borrowed words' S Iehik . Gr
matical mMcellany 179. " awa III am
26. 3. sk- in Englsh: Bjorkman 10.
.. 6: Latin in. Stender-Petersen 350. German Mau! from
Go hic. F. ln Beltrage Oilur Geschichte 35.156 (1909).
26.6. Enghsh words with ioreign affix . GAN'
sen, Growth 106. Suffix -cr: Stterlin n.
es
. . . IChol/lOn; Gadde; Jesper-
26.7. Loan-translation: K. Sandfeld Jensen in Fest .
Grammatical terms' Waekernagel V l schrijt Thomsen 166.
25 . , or Cilungen.
. 8. Early Germanie loans iro L t' KI
Growth 31 1atl'n 1 f m am: uge, Urgermanisch 9; Jespersen
. oans rom earl G . . B .- ,
jhrung 43. y errnaruc. ruch; Meyer-Lbke, Ein-
eaAPTER 25
26.1. La.tin missionary \Vords in English: Jespersen Growth 41 10
man \Vords ln Scandinavian: Hellquist 561 10w G' d D . .w
sian' van der MI' 0 S h d" erman an uteh III
G cl f . eu en, . e ra er ln Wisllenllchajtliche Beihejle 499 (1903)
en er 0 Enghsh words in American German' A W A . C . .
Il; in Arnerican Norwegian' G T FI . " .. . ron ln UT17le volume
W t' . . . om ln LFt-U!Ct notes 2 257 (1002)
lan es s (Bilingualism 46) about the' iate of nt
article (whieh a
rernarks in a Iiterar y use m erpretahon) and on sorne
yessay.
2
11
6. 2
E
So.nlli?t of.languages, bibliography: Paul Prinzipien 390' Bee
ma y ...'mdlsch lU Berichte Lei . 1897' ,espe-
J. Wackernu el in N h .' .101; G. Hempl in TAPA 1898.31;
Welsh: Parry:WilIiam':. rtehlen Gottmgen, Geschajtliche Mitleilungen 1904.90.
Place-names: Mawer-Stenton' Mei r 145'
Meyer-Lbke, Einjiiltrung 2;54; Oroon. e , 322; Dauzat, Noms de lieux;
Dutch words in American E l' h
with much.older these are not to be
re ch words III Enghsh: Jespersen, Growth 84; lI5.
Personal Mmes: Barber' Ewen' WeekJey R S
D ',omance' urnamu' Bah' h
auzat, N oms personnes; Meyer-Lbke,' 244 ' mac ;
26.3. Germaruc wQrs in Finnsh: Thomsen' E N
g
F' . h
' . . ....a III
UgNsc/w Forschungen 13.345 (1913); later references will be found in W. Wiget
in Streitberg lt'estgabe 399; K. B. Wiklund in same, 418; Collinder.
Germanie words in Slavie: Stender-Petersen. In Romance: Meyer-Lbke,
Einjiihrung 43 with references.
Gipsy: Miklosich, Mundarten; bibliograpby: Black; German Gipsy: Finek,
Lehrbw;h.
Ladin: Meyer-Lbke, Einfhrung 55.
26.4. Scandinavian elements in EngIish: Bjorkman; Xandry: Flom; Lind
kvist; A. Mawer in Acta. phiJJgi;;a Scandiwwica 7.1 (1932); E. Ekwall in
Grammatical misceUany 17.
Chilean Spanish: R. Lenz in ZrP 17.188 (1893); M. L. Wagner in ZrP
40.286; 385 (1921), reprinted in Meisterwer/ 2.208. Substrata in Romance
languages: Meyer-Lbke, Einjhrung 225.
Dravidian traits in S. Konow in Grierson 4.278.
Balkan languages: Sandfeld. Northwest Coast languages: F. Boas in Lg
1.18 (1925); 5.1 (1929); American anthrQPologist 22.367 (1920).
26.6. English and American Gipsies: J. D. Prince in JAOS 28.271 (1907);
A. T. Sinclair in BuJJ.elin 19.727 (1915); archaic form: Sarnpson.
Jargons, trade languages, creolized languages: Jespersen, Language 216.
English: Kennedy 416; American Negro: J. A. Harrison in Anglia 7.322 (1884);
J. P. Fruit in Dir:dect noies 1.196 (1892); Smith; Johnson. West African:
P. Grade in Archiv 83.261 (1889); Anglia 14.362 (1892); E. Henrici in Anglia
20.397 (1898). Suriname: Schuchardt, Sprache; M. J. Herskovits in Proceedings
23d 713; West-Indische gids 12.35. Pidgin: F. P. H. Prick van Wely in En{/"'
lische Studien 44.298 (1912). H. Schuchardt in Sitzungsberichte
Wien 105.151 (1884); Englische Studien 13.158 (1889); Churchill. India:
H. Schuchardt in Englische Studien 15.286 (1890).
Dutch: H. Schuchardt in Tijdschrijt 33.123 (1914); Hesselingj de Josselin
de Jong; Afrikaans: van der Meer xxxiv; cxxvi.
For various Romance jargons, sec the studies of H. Schuchardt, listed in
Schuchardf.-.Brevier 22 ff.
Chinook jargon: M. Jacobs in Lg 8.27 (1932). Siavic German and Italian:
Schuchardt, Slawo-Deulsches. trade language: O. Broch
in Archiv jr Illwische Philologie 41.209 (1927).
CHAPTER 27
27.1. The child: Jespersen, Langua,ge 103; J. M. Manly in Grammatical
miscellany 287.
27.2. Garnillscheg 14.
27.4. Rise of standard languages: Morsbach; Flasdieck; Wyld, History;
L. Morsbach in Grammatical millcellany 123. German: Behaghel, Ge:;chichte
182; Kluge, Luther. Dutch: van der Meer. French: Brunot. Serbian: Leskien,
Grammatik xxxviii. Bohemian: Smetanka 8. Lithuanian: E. Hermann in
NachrichJen Gttingen 1929.25. Norwegan: Burgun; 8eipe
27. 6. Englsh busy, etc.: H. C. Wyld in Eng/ische Studien 47.1; 145 (1913).
Englsh er; ar, etc.: Wyl, History.
Obsolete worda revived: Jespersen, Growth 232; dmng-do'
Kittredge 118.
524
NOTES
Half.leam
e
?words Romance: Zauner 1.21; Meyer-Lbke Ei-.
27.7. Medieval Latin: Strccker Bonnet C CR"' f '. nfuhrung 30.
, ,.. lce, orms III Du Cange.
CH..... l'TER 28
of new speakers to the standard language; Wyld, HisWrica},
28.2. Reading: Passy, Enseignement' Erdmann-Dod e. F:
28.4, Foreign-language teachin . . g, echner.
ta teach; Vitor, M ethadik' Palmg. Pradtcal study; Jespersen, How
Bb ,er, '--'Unguage study Colema . M M
1 hOKraphy; Buchanan-MePhee. Vocabulary' 'V t' l . li, curry.
28 5 A 'fi , es .-earntng
. . rt! clllilanguages: R. M. Me er in IF 12: .
R. Jones in JEGP 31.315 (1932). bb/ h. .33, (1901); Gurard;
26 6 Gen 1 t cl .' 1 IOgrap y III Bulletm 12.644 (1908)
., era en ency of Iinguistic development: Jespersen, Progress..
BIBLIOGRAPHY
General bibliographie aids, including the periodic national bibliographies,
are described in H. B. Van Hoesen and F. K. Walter, Biblioqraphy; practical,
enumerative, hislorieal; an introductory manual (New York, 1928) and in Georg
Schneider, Handbw;h der Bibliographie (fourth edition, Leipzig, 1930).
Bibliography of various language familles: 'V. Schmidt; Meillet-Cohen.
Oriental languages, annually; Orientalische Bibliographie. Indo-European;
Brugmann-Delbrek; annually in IF Anzeiger, since 1913 in 1J; these annuals
list also sorne gcneral lnguistic publications. Greek: Brugmann-Thumb;
Latin: Stolz-Schmalz; both of these languages annually in Bursian, Romance:
Grober; annually in Vollmoller, then in ZrP Supplement. Gennanie, includ-
ing English, in Paul, Grundriss; annually in J ahreslicrichl; the latter, up to 1900,
is summarized in Bethge. English: Kennedy; eurrent in A ngNa Deibla/t.
SeriaIs dealing with Germanie languages arc listed in Dieseh, Germanie and
Romance, bimonthly in Literaturblatt.
Items 1 have not Seen are braeketed.
Abhandlungen Leipzig: Sach-sische Akademie der Wssensehaften, Leipl:ig;
Philologiseh-hJlorische Klasse; Alihandlungen. Leipzig, 1850-.
Acta philologica Scandinavica. Copenhagen, 1926-.
AJP: American j01tT1U of philology. Baltimore, 1880-.
Allport , F. H., Social psychology. Boston, 1924.
Ameriean spee<;h, Baltimore, 1925-.
Andresen, K. G., Ober deutsehe Volkselymologie. Sixth cdition, Leipzig,
1899.
A nglia. Halle, 1878-; bibliographie supplement: Beiblatt; Mitteilungen,
sinee 1890.
Archiv: Archiv jr das Studium der neucren Sprachen. Elberfeld (now Braun-
schweig) 1846---. Often rcferred to as lIA ("Herrigs Archiv").
Archiv jr slavische Philologie, Berlin, 1876---.
Arendt, C., Einjiihrung in die nordehinesisehe Umgangssprac!w. Stuttgart
and Berlin, 1894 ( = Lehrbcher des Seminars jr urientalische Sprachen zu
Berlin, 12).
Arendt, C., Handbuch der nordehinesischen Umgangssprache. Erster Theil.
Stuttgart, 18\H ( = same series as preeeding, 7).
Armstrong, L. E., and ''liard, 1. C., Handbook of English intonation. Cam-
bridge, 1926.
Alti dei XXII rxmgresso intcrnaziona!e deU Americanisli. Rome, 1928.
BAE: Bureau of American eOmology; Annual reJY!rts. Washington, 1881-.
Bahlscn, L., The teaching oj modern languages. Roston, 1905.
Biihniseh, A., Die deu/sehen Persone;rmamen. Third cdition, Leipzig, 1920
( = A us Nalur und 296).
Barber, H., British fumily names. Second edtioIl, London, 1903.
525
526 BIBLIOGRAPHY
BIBLIOGRAPHY
527
Barker, M. L., A handbook of German intonation. Cambridge, 1925.
Baudouin de Courtenay, J., Ver.mch einer Theorie der phonetischen AUer-
1llItiimen. Strassburg, 1895.
BechteI, F., Die Hauplprobletnc der indogermanischen Lautlehre seit Schlei-
cher. Gttingen, 1892.
Behaghel, O., Die deuische Sprache. Seventh edition, Vienna, 1923 ( = DWJ
Wi8sen der Gegenwart, 54).
Behaghel, O., Deutsche Syntax. Heidelberg, 1923-32 ( = Germanische Bib-
liothek, 1.10).
Behaghel, O., Geschichte der deu18chen Sprache. Firth edition, Berlin and
Leipzig, 1928 ( = Grundriss der germani8chen Philclogie, 3).
Bei/rage ZUT Geschichte der deutschen Sprache und Literatur. Halle, 1874-.
Orten refel'l'ed to as PBB ("Paul und Braunes Beitriige").
Beilrage zur Kunde der indogermanischen Sprachen. Gttingen, 1877-1906.
Often refel'l'ed to as BB (" Bezzenbergers Beitxiige").
Belvalkar, S. K, An account of the differenl e:cisting systems of Sanskrit
grammar. Poona, 1915.
Bcnecke, G. F., Mller, W., Zarncke, F., Mitlelhochdeutsches Worlerbuch.
Leipzig, 1854-61. Supplemented by Lexer, below.
Benfey, T., Geschichle der Sprachwi8sen8chaft und Philologie in Deutschland.
Munich, 1869 ( = Geschichle der Wissen8chaften in Deutschland; Neuere Zeit, 8).
Bennicke, V., and Kristensen, M., Kart over de damke folkemaal. Copen-
hagen, 1898-1912.
Bergstrm, G. A., On blendings of synony1lWU8 or cognate expressiom in
English. Dissertation, Lund, 1906.
Berichte ber die Verhandlungen der sachsischen Akademie der Wi.wm-
schaften zu Leipzig; Philologisch-hist0ri8che Kla8se. Leipzig, 1849-.
Berneker, E., Russische GramTlUllik. Second edltion, Leipg, 1911 (=
Sammlung Gchen, 68).
Berneker, E., Slamsches etymologisches W6rlerbuch. Heidelberg, 1908-
( = Indogermani8che Bibliothek, 1.2.2).
Bertoni, G., ltalia dialeuale. Milan, 1916 (Manuali Hoepli).
Bethge, R., Ergebnisse und FortschriUe der germanistischen Wissenschaft
im lelzten Viertel;'ahrhundert. Leipzig, 1902.
Beyer, F., and Passy, P., Eletncntarbuch des gesprochenen FranziilJi8ch.
Cthen, 1893.
Bj1.irkman, E., Scandinavian loan-words in Middle English. Halle,190Q-{)2
( = Siudien zur englischen Philologie, 7; Il). '
Black, G. F., A Gypsy bibliography. wndon, 1914 ( = Gypsy lote weiety
monographs, 1).
Bladin, V., Studies on denominal.ive verbs in English. Dissertation, 'Upp"-
sala, 1911.
B1attner, K, Taschenw6rlerbuch der russischen und deulschen Sprache.
Berlin, 1906.
Bloch, J., La !ormatwn de la langue marathe. Paris, 1920 ( = Biblioth-
que de l' gcole des hautes tudes; Sciences historiques el philologiques, 215).
Bloomfield, L., Tagalog texls. Urbanll., Illinois, 1917 ( = Unwersity ollUi-
nais sludies in language and literature, 3.2-4).
B1mei, R, Einfhrung in die Syntax. Heidelberg, 1914 ( = Indogerman-
ische Bibliothek, 2.6). .
Boas, F., Handbook of American lndian languages. 1911-
( = Smithsrmian institu.lirm; Bureau of American. e}hnoloU!J; .Bulle/l
n
40).
Bhtlingk, O., Panini's Grammatik. Second edltlon, :lpzlg, 1887.
Bhtlingk, O., Die Spradw der J akuten. St. Petersburg, 1851 ( =.
3 of A. T. von Middendorf, Reisc im aUilzers/en Norden und Osten
Bhtlingk, O., and Roth, R, St. 18,)5-75;
additional matter in O. Bhtlingk, Sansknt-W6rlerbuch ln krzerer Fassung,
St. Petersburg, 1879-89 and in R. Schmidt, Nachtrdge zum Sans/crit-W6rlerbueh,
Hannover, 1924.
Bohlacq, E., Dictionnaire tym,ologique de la langue grlXf[lW. Heidelberg and
Paris, 1916. .
Bonnet, M., Le latin de Grgoire de Tours. Pans, 1890.
Bopp, F., Ober das Kon;'ugationssys/em der Sanskritsprache. Frankfurt am
Main,1816. ....
Bopp, F., Vergleichende Grammalik ries Samkrit, Zc:w, Grw;hlSCh;m,
ischen, LiUhaui8chen, Golhischen und Berlm, 1833. Third edltlOn,
1868-71. d'ted .
Borthwiek, N., Ceachda beoga galuingi. Irish reaaing less01t8, e 1 ln
simplified spellng by O. Bergin. Dublin, 1911.
Bosworth, J., and Toller, T. N., An Anglo-S=rm dictionary. Oxford, 1898.
Supplemen/, by T. N. Toiler, 1921. . .
Bourciez E. lments de lin[/Uis/ique romane. Second edltlOn, Pans, 1923.
Braune, 'w.,' AUhochdeutsche Grammalik. Third-fourth edition, Halle, 1911
( = Sammlung kurzer Grammatiken 5). .
Bral, M., Essai de smantique. Fourth edltlon, PariS, 1908. An Englil:l
translation of the third (1897) edition, by Mrs. H. Cust, appeared under the
title Semantics in London, 1000.
Bremer, O., Dcub.che Phrme!ik. Leipzig, 1893 ( = Sammlung kurzer gram-
TIUIliken deu1scher Mundarten, 1).
Broch, O., SlatJische Phone/ik. Heidelberg, 1911 ( = Sammlung slaui8cher
Elenwntar- und Handbcher, 1.2).
Brockclmann, C., Semitische Sprachwi8senschaft. Leipzig, 1906 ( = Samm-
lung Goschen, 291).
Brflndum-Nielsen, J., Dialekier og diaklforskning. Copenhagen, 1927..
J., Der Einjl.uss der germanischen Sprachen aul das
Heidelberg, 1913 ( = Sammlung romani8cher Elernentar- und Handl:riicher,
5.1). . ken S
Brugmann, K., KurZC vergleichende Grammatik der indogermants
c
pra-
chen. Strassburg, 1902-{)4.
Brugmann, K., Zum heuligen Stand der Sprachwi8sen8chaft. Straashurg,
1885. .
Brugmann, K, and Delbrck, R, Grundri8s der vergleichenden
der indogermanischen Sprachen. Strassburg, 1886-1000. Second edltlOn,
1897-1911. ..
Brugmann, K., and Thumb, A., Griechische Fourth edltlon,
Munich, 1913 ( = llandbuch der klassischen AltertumswtssBnschaft, 2.I).
528 BIBLIOGRAPHY BIBLIOGRAPHY
529
Brunot, F., Huloire de /.a langu.efranaUe des O1'igines 1900. Paris, 1905-.
B8L: Bulletin de la 80cit de linflUistiq1MJ de Paris. Paris, 1869-.
B80B: Bullelin 01 the School 01 Oriental sludies, London inslitution. London
1917- . '
Buchanan, M. A., and McPhee, E. D., An annolaled bibliography of modern
lanflUage melhodology. Toronto, 1928 ( = Publications 01 the American and
Canadian commiUees on modem languages, 8); also in volume lof Modern la....
guage instruclion in Canada ( = same series, 6).
Buck, C. D., A grammar 01 Oscan and Umbrian. Boston, HlM.
Buck, C. D., Introduclion wthe sludy ollhe Greek dialeds. Second edition
Boston, 1928. '
Bhler, K., Die geish'ge Entwicklung des Kindes. Fifth edition, Jena, 1929.
Bulletin of lhe New Y01'k Public Library, New York, 1897-.
Burgun, A., Le dveloppement linguislique en Norvge depuis 1814. Chris-
tiania, 1919-21 (= Videnskapsselskapets sm11er; HislO1'isk-jologisk k/a.sse,
1917.1; 1921.5).
Bursian, K., Jahresberichl ber die ForlschriUe der klassuchen Altertums-
11!Jsenschafl. Berlin (now Leipzig), 1873-.
Churchilt W., Booch-la--mar. \Vashngton, 1911 ( = Carnegie inetilution o}
WIUlhinglon; Publications, 154).
Clea.sby, R., and Vgfusson, G., An Icelandic-English dictionary. Oxford
,
Coleman, A., The teaching of modem foreign, languages in lhe United 8ta1.es.
New York, 1929 ( = Publications 01 the American and Canadian commiUees
on modem lanflUG(Jes, 12).
Collin, C. S. R., Bibliographical guide 10 semalology. Lund, Hl14.
Collinder, B., Die urgermanischen Lehnwiirler in Finnischen. Uppsala,
1932 ( = 8krifter tUgivna av K. humanisliska vetenskaps-samfundet i Uppsala,
28.1).
Conway, R. S., The Ilalic dialeds. Cambridge, 1897.
Cunne, G. O., A grammar of the German langu,age. Second edition New York
1922. ' ,
Curme, G. O., and Kurath, H., A grammar 01 the English language; volume 3,
Syntax, by G. O. Curme. New York, 1931.
Curme volume ollinguislic studies. Baltimore, 1930. (= Language mmw-
graphs published by lhe Linguistic society of America, 7).
Curtius, G., Z,w Kritik der neueslen 8prachforschung. Leipzig, 1885.
Dahlcrup, V., Del danske sprogs histMie. Second edition, Copenhagen, 1921.
Dahlerup, V., Ordbog ouer dei danske sprog. Copenhagen, 1919-.
Dannesteter, A., De la cralion acluelle de mols nouveaux dans la langue
franaise. Paris, 1877.
Dan,nesteter, A., de la lormation des mols composs dans la laUflUC
franatse. Second edltlOn, Paris, 1894.
Darmesteter, A., La vie des mols. Twelfth edition, Paris, 1918.
Dauzat, A., La gographie linguulique. Paris, 1922.
Dauzat, A., Hislcnre de la langue franaise. Paris, 1930.
Dauzat, A., Les nmns de lieu.:::. Paris, 1926.
Dauzat, A., Les noms de personnes. Paris, 1925.
Dauzat, A., Les palois. Paris, 1927.
Debrunner, A., Griechische Worlbildungslehre. Heidelberg, 1927 ( = Indo-
germanische Bibliolhek, 2.8).
de Josselin de Jong, J. P. R, Hel huidige ,\'egerhoUandsch. Am.sterdam,
1926 ( = Verlw.ndelingen der K. akademie van welenschappen,' /.eUer-
kunde; Nieuwe reeks, 14.6).
dc Laguna, G. A., Speech; ils junclion and Haven, 1927.
Delbrck, B., Einleitung in das Stwiium der tndogermamschen Sz:'"W:hen.
Sixth edition, Leipzig, 1019 ( = Bibliolhek indogermanischer Grammattken, 4).
Delbrck K, Grundfragen der 8prachjorschung. Str:J.8sburg, 1901: . .
de Sau:;sure, F., Cours de lin,guistiqtte gnrale. Second edlhon, Paris,
1922.
Deulsche Dialeklgeographie. Marburg, 1908-.
de Vrics, M., and te Winkel, L. A., Woordenboek der Nederlandsche laal.
The 1882-.
Dialect noies. New Haven, 189(}-.
Dicsch, C., Bibliographie der germanistischcn Zeitschriften. Leipzig, 1927
( = Bibliographie pub/icatians, Germanistic section, ,'fIodern language associa-
tion of Amenca, 1).
Diez, F., Grammalik der romanischen Sprachen. Bonn, 1836-44. Fifth
tion, 1882.
Donum natalicum Schrijnen. Nijmegen and Utrecht, 1929.
d'Ovidio, F., Grammatica storica della lingua e dei dialelli italiani. Milan,
1901} (Manuali Hoepli). .. .
Du Cange, C. du Fresne, Glossarium mediae et infimae Second
edition, Niort, 1883-87.
Ebert, M., ReaUerikon der Vorgeschichte. Berlin, HI24-.
Ellis, A..J., On early EU{I!ish pronuneiation. London, 1809-89 ( = Early
English lext sociely; Extra series, 7; 14; 2:3; o).
En{llische Sludien. Heilbronn (no,," Lcipzig), 1877-.
Erdmann, B., and Dodge, R., Psyclwlogi-sche Uutersuchungen ber d=
Lesen. Haile, 1898.
Erman, A., Die Hieroglyphen. Leipzig, 1912 ( = Sammlung GiJschen,
Esper, E. A., A technique jor the experimenlal im'estigation af assocwtwe
inlerference in artificial lillgllisti-c mater. Philadelphia, 1925 ( = Language
rrumowaphs publishe,z by lhe Linguis/ic sacie/y of .1merica, 1):
Ewen, C. L'Estrange, A hisWry of Slirnames ()f the Bntlsh Isles. London,
1931.
Fabian, E., Dus exuzentrische Kornpositum irn Lei pr. g, 19;1l
( = Form l.nd Gdst, 20). . . .
Falk, H., and Torp, A., Daru;k-N()rskens syn/ax. 1900...
Falk, H., alld Torp, A., Norwegisch-diinisches elymologlsches Worterbuch.
Heidc1berg, 1910 ( = Germanische Bibliothek, 1.4.1).
Farmer, J. S., and Henley, V'l. E., Slang and ils analogues. London, 1890-
1904. Second editon 1903-. Abridge<l version: A dicti-onary of slang and col
loquial English. New York, 1921.
Fechner, H., Gnmdriss du Geschichte der u,"chtigsterl Leselehrarten. Berlin,
1884.
530
BIBLIOGRAPHY BIBLIOGRAPHY 531
Feilberg, H. F., Bidrag /il en ordJ;og OVer iyske almuesmaal. Copenha
1886-1910. gen,
Fe.ist, S., Etymo/.0gi8ches Wo'irrerbuch der gol8chen Sprache. Second editi
Halle, 1923. On,
Fe.ist, S., KuUUT, Ausbreilung und llerkunj.t der Irnfonermanen B lin
1913. . er,
FIJ8tu.abe Hrinzel: Abhandlungen ZUT germani8chen Philolgoie; FlJ8tgabe ft
R. Hemzel. Halle, 1898.
FlJ8tgabe Suchier: Forschungen ZUT Tomanischen PhiWlogie,' Futgabe ft
H. Suchier. Halle, 1900.
FIJ818chrijt fr . , . P. KTe18chmer. Berlin, 1926.
Feslschrift Meinhof. Hamburg, 1927.
FIJ8l8chnfl Morf: Aus romani8chen Sprachen und Literaturen; Fest8chrift
H. Morf. Halle, 1905.
FIJ8l8chrifl V. Thomsen. Leipzig, 1912.
Fesl8chrift Wackernagel: Antidoron; Fe<;18chrifl J. Wackernagel Gttin"",n,
1924. .
til V. Thomsen. Copenhagen, 1894.
F. N., Dre Au/gabe und Gliederung der Sprachll.:iss8n8chaft. Halle,
Finck, F. N., Die Hauptlypen de8 Sprachbaus. Leipzig, 1910 ( = A us Nalur
und Geisleswell, 268).
F. N., Die Klrl88ijikation der Sprachen. Marburg, 1901.
Finck, F. N., LehTbuch de<! I>iakkts der Mut8chen Zineuner Marburg,
1903. .
Finck, F. N., Die SpracluJUimme des Erdkrei8es. Leipzig, 1909 ( = Aus
Nal.ur und Geteswcll, 267).
Forschungen. Helsingfors, 1901-.
H.. Geographie der schwabU;chen Mundar/. Tbingen, 1895.
Flaadleck, H. M., Forschungen zur Frh.zeil der neuenglischen Schriftsprache.
Halle, 1922 ( = Studien zur englischen Philologie, 65; 66).
Fletcher, H., Speech and heating. New York, 1929.
F1om, G. T., Scandinavian influence lm soulhern Lowland Scokh. New York,
1901 ( = Columbia Univmn:ly Germanic studies, 1).
Forcbhammer, H., How wlcarn Danish. Heidelberg, 1906. 1 have seen
only the French version, Le danois parl, Heidelberg, 1911.
Franck, J., and van Wijk, N., Etymologisch woardenboek der Nederla1l(jche
taal. The Hague, 1912.
C. C., .leaching of the English language. New York, 1927.
T., Rhein1Sche Sprachgeschichte. Essen, 1924. Also as contribution
to GesChlChte des Rhrinlande<;, Essen, 1922.
Fritllner; J., Ord1Jog wer dei gamle rwrske sprog. Christiania, 1886-96.
Gabelentz, G. von der, Die Sprachwiss8n8chaft. Second edition Leipziu
1901. ' ..",
Ga.dde, F., On the his/ory and use of lhe suffixes -cry (-1'1/), -<lfle and -ment in
Engluh. Lund, 1910 (Svea English tTeal.fSCS).
.W. H. -r:., The phonetics of Arabie. London, 1925 (The Ameri-
can Unwersfty at Cafro; On'enlal 8tudies).
Gamillscheg, E., Die Sprachgeographie. Bielefeld and Leipzig, 1928 ( =
Neuphilologisehe Handbibliothek, 2).
Gauthiot, R., La fin de mot en indo-europen. Paris, 1913.
The Germanic rcview. New York, 1926-.
Giessener Beitrage zur ErforschUT!{! der Sprache und Kullur Englands und
Nordamerikas. Giessen, 1923-.
Giles, H. A., A Chinese-English dictitmary. London, 1892.
GiIliron, J., L'aire clavel/us. Neuveville, 1912.
Gilliron, J., Gnalogie dlJ8 mo18 gui dsignent l'abeille. Paris, 1918 ( =
Bibliothque de l'cole des haute<; tudes,' Science<; hisloriques et philologiqUC8,
225).
Gilliron, J., Pathologie et thrapeutique verbales. Paris, 1921 ( = CoUeetion
lingnistique publie par la Socit de linguistique de Paris, 11).
Gilliron, J., and Edmont, E., Atlas linguistique de la France. Paris, 1902-10.
Supplement, 1920; maps for Cornica, 1914-15.
Gi1ron, J., and Mongin, J., Scier daM la Ganle romane du sud et de l'est.
Paris, 1905.
Gilliron, J., and Roques, M., tude<; de gographie linguistique. Paris,
1912.
Goeders, C., Zur Analogiebildung im Mit!el- und Neuenglischen. Disserta-
tion, Kiel, 1884.
Graff, E. G., Althochdeutscher Sprachschatz. Berlin, 1834-42. Index by
H. F. Massmann, 1846. A supplement, Die allhochdeul.schen Prapositionen,
Konigsberg, 1824, appeared before the main work.
A grammalical misceUany oJJered wO. Jespersen. Copenhagen, 1930.
Grammont, M., La dissimilation consonantique dans /e.s langues indo-euro-
pennes et dans les lngues r(}1TW,ne8. Dijon, 1895.
Grandgent, C. H., From Lal.in wltalian. Cambridge, Mass., 1927.
Greenough, J. B., and Kittredgc, G. L., Words and their ways in English
speoch. New York, 1001.
(Gricra, A., Atlas /ingistic de Catalunya. Barcelona, 1923-.]
Grierson, G. A., Linguistic Sllrvey of India. Calcutta, 1903-22.
Grimm, J., Deulsche Grammatik. Gottingen, 1819-37. Second edition of
first volume, 1822. Index by K. G. Andresen, 1865. Reprint with additions
from Grimm's notes, Berlin, then Gtersloh, 1870-98.
Grimm, J. and \V., Demsches Wiirterbuch. Leipzig, 1854-.
GRJI: Germanischromanische Monotsschrift. Heidelberg, 1909-.
Grober, G., Grundriss der Tomanischen Philologie. Second edition, Strass-
burg, 1004--06.
Gurard, A. L., A short histor1j of the international language movement.
London, 1922.
Guernier, R. C., Notes sur la prorumciation de la lang1le mandarine de Pkin.
London, 1912 (Supplement to Motre phontique).
Gut.zmann, H., Physiologie der S/imme llnd Sprache. Second edition,
schweig, 1928 ( = Die Wissenschaft, 29).
Gutzmann, H.) Sprachhei!kunde. Third edition, Berlin, 1924.
Haag, C., Die Mundarten des oberen Nockar- und Donaulanrks. School pro-
gram, Reutlingen, 1898.
532
BIBLIOGRAPHY BIBLIOGRAPHY
533
Ransoon, F., Spanische GrammaWc auf hislorischer GrundlafJe. Halle, 1910
( = Sammlung kurzer Lehrlnicher der romanischen Sprachen, 6).
Han'ard studies in classical philolo(j1j. Boston, 189(}-.
Hasse, A., Sludien ber englische Volksetymologie. Dissertation,
burg, 1904.
Hatzfeld, A., Darmestet.er, A., Thomas, A., gnral de la
franaise. Sixth edition, Paris, 1920.
Hatzfeld, H., Leitfaden der vergleichenden Bedemungslehre. Second edition
Munich, 1928. '
Head, H., Aphasia and kindred disorders of speech. New York, 1926.
Heepe, M., Lautzeichen. Berlin, 1928.
Hehn, V., KuUurpjlanzen und Haustiere. Seventh edition Berlin 1902
Ill! . EDo " .
. e qmst, ., el svenska ordf6rrude/s alder oeh li,rsprung. Lund 1929-30
Hempl, G., German ortlwyraphy and phonology. Boston, 1897. ' .
Hermann, E., Berlhold Delbrck. Jena, 1923.
Hermann, E., Lautgesetz und Analogie. Berlin, 1931 ( = Abhandlungen der
Gesel!chaft di:T Wissenschaften zu GUingen; Philologisch-historische Klasse'
Neue Folge, 23.3). '
Herwg, E., Streitfragen der romanischen Philologie. Halle, 1904.
Herzog, R., Die Umschrifl der dUeren yriechischen Literalur in das wnische
Alphabet. University program, Basel, 1912.
HesseIing, D. C., Hel Negerhollantl.<; der Deense Antillen. Leiden, 1905.
Heualer, A., Aitislndisches Elementarbuch. Second cdition, Heidelberg
1921 ( = Germanische Bibliothek, 1.1.3). '
Hilmer, H., Schallnachahmung. Halle, 1914.
J:iirt, H., Hawlbuch der [Jriechischen Lau/- und Formenlehre. Second edition,
Heidelberg, 1912 ( = Indogermanische Ribliothek, 1.1.2).
Hirt, H., Handbuch des Urgi:Tmanischen. Heidelberg, 1931- ( = Indoger-
manische Bibliolhek, 1.1.21).
Rirt, H., Die Indogi:Tmanen. Strassburg, 1905.
Hirt, H., lndogermanische Grammatik. Heidelberg 1921- ( = Indogerman-
ische Bibliothek, Ll3). '
L., Principes de grammaire gnrale. Copenhagen, 1928 ( = Del
k. danske vrJenskabi:Tnes Histmisk-filologiske medJelelser, 16.1).
Holthausen, F., Allsttchsisches Blcmentarbuch. Second edition, Heidelberg
1921 ( = Gerrnanisehe Riblialhek, 1.5). '
IJomenaie ofreda il Menr.dez Pidal. Madrid, 1925.
Hoops, .7., Re(lllexikan der germanischen Aller/umskunde. Straashurg, r911-19.
HOOJJ8, .J., Waldboiume und 1\ullurpflanzen im germanischen AlIertml!. Strass.
ourg, 1005.
Horn, W., Hislorische Grammalik; Erster Teil, Lau/lehre.
Strassurg, 1908.
Horn, W., Sprachk6rper und Sprachfunktian. Second edition, Leipzig
1923 ( = Palaestra, 135). '
Hubert, H., Les Celles el l'expansion cellique. Paris, 1932 ( = L'lJolutwn de
l'humanit, 1.21).
Humboldt, W. von, Ober die KaU!isprru;he. Berlin, 1836-39 (in Abhantl-
lungen der Akademie der Wissenschajlen zu Bi:Tlin, as of 1832). Part 1 was
republished with an elaborate commentary, in two volumes and a supplement,
by A. F. Pott, Berlin, 1876--80.
IF: Int!germanische Forsehungen. Strassburg (now Berlin), 1892-. Sup-
plement: Anzeiger.
IJ: Indogermanisches Jahrbuch. Strassburg (now Berlin), 1914-.
]JAL; International Journal of American linguisties. New York, 1917-.
Ipsen, G., and Karg, F., Schallanalytische Versuche. Heidelberg, 1928
( = Gi:Tmanische Bibliothek, 2.24).
IZ: Internationale Zeil.8chrift fr allgerncine Sprru;hwiBsenschaft. Leipzig,
1884-90.
Jaberg, K., Sprachgeographie. Aarau, 1908. .
Jaberg, K., and Jud, J., Sprach- und Sachatlas Italiens und der Sd8chwelz.
Zofingen, 1928-.
Jahresbericht ber die Erscheinungm auf dem Gebiete der germanischen Phi-
lologie. Berlin, 1880--.
JAOS: Journal of the American Oriental society. New York (now New
Haven), 185G-.
JEGP; Journal of Euglish and Gi:Tmanic Philology. Bioomington, Indiana
(now Urbana, Illinois), 1897-.
JeUinek, M. H., Geschichte der deutschen Grammatik. Heidelberg, 1913-14
( = Germanische Biblio/hek, 2.7).
Jellinek, M. H., Geschichle der gotischen Sprache. Berlin and Leipzig, 1926
( = Grundriss der germanischm Philologie, 1.1).
Jensen, H., Geschichte der Sehrift. Hannover, 1925.
Jespersen, O., Fonelik. Copenhagen, 1897-99. . .
Jespersen, O., Growth and s/ructuTe of the Bnglish language. Fourth edltlOn,
New York, 1929.
Jespersen, O., How w tooch a foreign language. London, 1904.
" Jespersen, O., Language; it8 nature, developmenl, antl migin. London and
New York, 1923.
Jespersen, O., Lehrbuch der Phonetik. Second edition, 1913..
Jespersen, O., A modern Englwh grammar on hislorical princlples. Heldel
berg, 1909- ( = Germanische Bibliothek, 1.9).
Jespersen, O., The philosophy of gramma:r. London and New York, 1924.
Jespersen, O., Progress in language. London, 1894. .
(Jespersen, O., and Pedersen, H.,) Phorwtic transcription and translttera-
tion. Oxford, 1926 (Supplement ta MaUre phontique).
Johnson, G. B., Folk culture on St. Helma Island. Chapel Hill, 1930 (in
University of Nor/h Carolina social study series).
Jones, D., An English pr01wuncing diclionary. London, 1917.
Jones D. Inwnation cumes. Leipzig, 1909.
Jones: D.: Outline of English phonetics. Second edition, Leipzig and Berlin,
1922.
Jones, D., and Woo, K. T., A Canlonese phonetic reader. London, 1912.
Journal Je la Socit des americanistes de Paris, 1895-.
Jutz, L., Die alemannisehm Afu1I.darten. Halle, 1931.
Kaluza, M., Hislorische Grammatik der englischen Sprache. Second edition,
Berlin, \906--07.
534
BIBLIOGRAPHY BIBLIOGRAPHY 535
Karlgren, B., :2tudes sur la phonologie chinoise. Leiden and Stockholm
1915 ( = Archives d'tudes orientales, 15). '
Karlgren, B., A Mandarin phonetic reader. Uppsala, 1917 (= Archives
d'tudes ill'.mtaks, 13).
Karlgren, R, Philo/.ogy and ancient China. Oslo, 1926 ( = 1118tituttet fUT'
8ammenliil'/Wnde kullurfarskning; Serie A: 8).
Karlgren, R, Sound and symbol in Chinese. London, 1923 (Language and
literature series).
Kennedy, A. G., A bibliography of writings on the English language. CamM
bridge and New Haven, 1927.
Kent, R. G., The oounds of Latin. Baltimore, 1932 ( = Language mana-
graphs published by the Linguistic society of America., 12).
Kent, R. G., The tex/ual criticism of inscripti0118. Philadelphia, 1926 ( =
Language 7/W1U)graphs published by the Linguistic society of America, 2).
Kenyon, J. 8., American pronunciation. Ann Arbor, 1924.
Klein, E., Die verdunkeen Worlzusammensetzungen im Neuenglischen.
Dissertation, Knigsberg, 1911.
Klinghardt, H., Om.mgen im deutschen TonfaU. Leipzig, 1927.
Klinghardt, H., and de Fourmestraux, M., Franz6sische Intonati0118bungen.
Cthen, 1911. English translation by M. L. Barker; French intonation exer-
cises, Cambridge, 1923.
Kloeke, G. G., De HoUarulsche expamie. The Hague, 1927' ( = Noord-
en Zuid-Nederlandsche dialectbibliotheek, 2).
Kluge, F., Deutsche Sprachgeschichte. Second edition, Leipzig, 1925.
Kluge, :1<'., Etymologisches Wiirterbuch der deutschen Sprache. Tenth edition
Berlin and Leipzig, 1924; eleventh edition, 19::10--. '
Kluge, F., Urgermanisch. Third edition, Strassburg, 1913 ( = Grundriss
der germanischen Philologie, 2).
Kluge, F., Von LuJher bis Lessing. Firth edition, Leipzig, 1918.
A., The gender of words denoting living beings in English. Dis-
sertatIon, Lund, 1905.
Krapp, G. P., The English language in America. New York, 1925.
Krapp, G. P., The pronunciation of standard English in America. New
York, 1919.
Kroeber, A. L., Anthropology. New York, 1923.
Kruisinga, E., A grammar of modern Dutch. London, 1924.
Kruisinga, E., A handbook of presenl-day English. Fourth cdition, Utrecht
1925; Part 2 in fifth edition, Groningen, 19,'H-32. '
Knzel, G., Das zusammengesetZle Substamiv und Adjektiv der englischen
Sprache. Dissertation, Leipzig, 1910.
KussmauJ, A., Die Stiirungen der Sprache. Fourth edition Lf'ipg
1910. ' ",
Last, W., Das Bahuvrihi-Compos-ilum im Altenglischen, Millelenglischen und
Ne-uenglischen. Dissertation, Greifswald, 1925.
Le?nard, S. A., The doctrine of correclnei,s in English usage 1700-1800.
MadIson, 1929 ( = University of Wisconsin studics in language and literatnre
'
Lcpsius, C. R., Standard ulphabe'. Seeond edition, London, t8ti3.
Le Roux, P., Atlas linguistique de la Basse Bre-fugruJ. Rennes and Paris,
1924-.
Leskien, A., Die Decliwltion im Slavisch-Litauischen und Germanisehen.
Leipzig, 1876 ( = PrC8ilchriften der Jablonowski'schen GeseUschaft, 19).
Leakien, A., Grammalik der altbulgarischen Spr(l,(;he. Second edition, Heidel-
berg, 1919 ( = Sammlung slavischer und Handtfcher, 1.1).
Leskien, A., Grammalik der serbokroatischen Spr(l,(;he. 1. Te. Heidelberg,
1914 ( = Sammlung slavischer Lehr
w
und Hand.b"acher, lA).
Leskier:, A., Handbu.ch der altbulgarischen (altkirchenslavischen) Sprache.
(Fifth edition, Weimar, 1910.]
Leskien, A., Litauisches Lese1:rnch. Heidelberg, 1919 ( = Indogermanische
Bihlwthek, 1.2).
Lexer, M., MiUelhochdeutsches IIandwiirterbuch. Leipzig, 1872-78.
Lg: Language; J DUrnal of the Linguistic society of America. Baltimore, 1925-.
LindkviBt, H., Middle English p/.aclHUlme8 of Scandinavian origin. Part 1.
DiBSCrtation, UppBRia, 1912 (alBo in Uppsala universitets arsskrift, 1911.1).
LinQsay, W. M., The Latin language. Oxford, 1894.
Literaturblatt fr germanische und romanische Philologie. Heilbronn (now
Leipzig), 1880-.
Ljunggren, R., Om den opersonliga konstruktionen. Uppsala, 1926.
Lloyd, R. J., Northern English. Second edition, Leipzig, 1908 ( = Skizzen
lebender Sprachen, 1).
Luick, K., Historische Gramm<Itik der englischen Sprache. Leipzig, 1914-.
Lundell, J. A., Det svenska Stockholm, 1879 ( = Nyare
bidrag tiU kannedom om de SVC118ka land-s-nWlen, 1.2).
Le matre phontique. Bourg-la-Reine (now London), 1889-----.
Marett, R. R., Anthropology, New York, 1911 ( = Home university liltrary,
37).
Mawer, A., and Stenton, F. M., Introduction ta the survey of English
Wlme8. Cambridge, 1921 ( = English pla-ce--name society, LI).
McKnight, G. H., English words and tMir background. New York, 1923.
McMurry, R. E., MueHer, M., Alexander, T., Modemforeign languages in
France and Germany. New York, 1930 ( = Studies of the International instit1d6
of Toochers college, 9).
Meier, J., Deutsche Volkskunde. Berlin and Leipzig, 1926.
Meillet, A., Aperu d'une histoire de la langue grecque. Third edition, Paris,
1930.
Meillet, A., Les dialectes indo-europens. Second edition, Paris, 1922 ( = Col-
kdion linguistique publie par la Socit de linguistique de Paris, 2).
Meillet, A., Introduction a l'tude comparative des langues in.r1o-europen1Ui8.
Third edition, Paris, 1912.
Meillet, A., Les langues dans l'Europe nouveUe. Second edition, Paris, 1928.
Meillet, A., Linguistique historique et linguistique gMrale. Paris, 1921
( = CoUeetion linguistique publie par la Socit de linguistique de Paris, 8).
Meillet, A., La mthode comparative en linguistique historique. Oslo, 1925
( = Instituttet for sammenliil'/Wnde kuUurforskning; Serie A: Forelesninger, 2).
Meillet, A., Le slave commun. Paris, 1924 ( = CoUeclion de manum publie
par l' Institut d'ludes slavll8, 2).
536 BIBLIOGRAPHY BIBLIOGRAPHY 537
Meillet, A., and Cohen, M., Les langues du mcnde, Paris, 1924 ( = Collu-
tion linguistique par la Socit de linguislique de Paris, 16).
Meinhof, C., GrundrifJs der Lautlehre der Bamusprachen. Second edition,
Berlin, 1910.
Meinhof, C., Grundziige einer !JergkU:henden Grammal.ik der Banf:u.-.Sprachen.
Berlin, 1906.
Meinhof, C., Die moderne Sprachforschung in Afrika. Berlin, 1910.
MeiBSIler, B., Die Keilschrift. Leipzig, 1913 ( = Sammlung Goschen, 708).
Meis1erwerke der romanischen Sprachwissenschaft. Munich, 1929-30.
Menndez Pidal, R., Hl idioma espafWl en sus primieros tiempos. Madrid,
1927 ( = CoUecci6n de manuales Hispania, B2).
Menndez Pidal, R., .il{anual de gramcitica histwa espafiola. Fourth
tion, Madrid, 1918.
Menndez Pidal, R., Origenes del espaiid. Madrid, 1926 ( = RevUm de
filologia espafWla; Anejo 1).
Meringer, R., and Meyer, K., Versprechen und Verlesen. Stuttgart, 1895.
Meyer, K. H., Historische Grammatik der TussUchen Sprache. Bonn,)923.
W., Einfhrung in da8 Studium der romanischen Sprachen.
Third edition, Heidelberg, 1920 ( = Sammlung romanischer Elementar und
Handbcher, 1.1).
Meyer-Liibke, W., Grammatik der Tomanischen Sprachen. Leipzig, 1890-
1902. French translation, Grammaire des langues romanes. Paris, 1890-1906.
Meyer-Lbke, W., Historische Grammatik der franzosischrm Sprache; 1.
Teil. Second edition, Heidelberg, 1913; 2. Teil, 1921 ( = Sammlung romarv-
ischer Elemrmtar und Handbcher, 1.2).
Meyer-Lbke, W., Romanisches etymologisches Wiirterbuch. Heidelberg,
1911-19; third edition, 1930- ( = Sammlung romanischeT Elementar und
Handbcher, 3.3).
Michaelis, H., and Jones, D., A phonetic dietionary of the English language.
Hannover, 1913.
Michaelis, H., and Passy, P., Dictionnaire phomfique de L langue franaise.
Second edition, Hannover, 1914.
Michels, V., Mittelhochdeutsches Elementarbuch. Third edition, Heidelberg,
1921 ( = Germanische Bibliothek, 1.1.7).
Middleton, G., An essay on analogy in synfax. London, 1892.
Miklosich, F., Etymologisches W6r1erbuch der I!lavischen Sprachen. Vienna,
1886.
Miklosich, F., Ober die Murularten und die Wanderungrm der Zigeuner
Europa's. Viennn, 1872-81 (aIso in volumB 21-23, 25-27, 30, 31 of Denk
8chriften der Akademie der Wissen8chaften; philosophisch-historische Klasse).
Mikloaich, F., Vergleichende Grammatik der slalJischen SprlUhtm. Weimar,
1852-74; second edition of volume 1, 1879; of volume 3, 1876.
Millardet, G., Linguistique et dialecwlogie romanes. Montpellier and Paris,
1923 ( = Publications de L Socit des langues romanes, 28).
The modern Lnguage reWew. Cambridge, 1906-.
Modern philology. Chicago, 1903-.
Morris, E. P., On principles and melhods in Latin syntaz. New York,
1901.
Morsbach, L., Ober den Ursprung der neuenglischen Schriftsprache. HeiI.
bronn, 1888.
MSL: Mbnaires de la Socwt de linguistique de Paris. Paris, 1868-.
Mller, F., Grundriss der Sprachwissenschaft. Vienna, 1876-88.
Mller, M., Die Reim- und Ablamkomposita des Englischen. Dissertation,
Strassburg, 1909.
NachricMen von der Gesellschaft der Wissenschaften zu G6uingrm; Philologisch-
hiBtorische Klasse. Gottingen.
Navarro Toms, T., ManlUll de pronunciaci6n espafWla. Madrid, 1918.
[English adaptation by A. M. Espinosa, Primer of Spanish pronundatwn.
New York, 1926.)
]',,'ED: A new Bnglish dictionary on historical principles, editcd by J. A. H.
Murray. Oxford, 1888-1928.
Neuphilulogische MiUeilungen. Helsingfors, 1899-.
Nicholson, G. A., English words with native ruots and with Greek, Latin,
or Romance suffixes. Dissertation, Chicago, 1916 ( = Linguistic studies in
Germanie,3).
Nicholson, G. G., A practical introduction w Frtmeh phonetics. London, 1909.
Nicht.enhauser, D., Rckbildungen im Neuhochdeutschtm. Dissertation,
Frciburg, 1\)20.
Nol-Annficld, G., General phonetics. Third cdition, Cambridge, 1924.
Noreen, A., Grammatik: 1. AUislandische und altnorwegische
Grammalik. Fourth cdition, Halle, 1923. 2. Allschwedische Grammatik, Halle,
1904 ( = Sammlung kurzer Grammatikrm germanischer Dialek/e, 4; 8).
Noreen VS: A., Varl.lprak. Lund, 1903-18. Selections translated
into German by H. W. Pollak, Einfiihrung in die wissenschaftliche Betrack-
tung der Sprache. Halle, 1923.
NS: Die neueren Sprachen. Marburg, 1894-.
Nyrop, K, Grammaire historique de la langue franaise. Copenhagen, 1899-
1930.
Nyrop, K., Ordrmes liv. Second edition, Copenhagen, 1925-26. A German
translation by L. Vogt, Das Leben der W6rter, Leipzig, 1903. [Second edition,
19231
Oertel, H., Lectures on the study of language. New York, 1901.
Ogden, C. K., and Richards, 1. A., The meaning of meaning. London, 1923.
Olscn, M., Farm;; and fanes of ancient NorUJay. Oslo, 1926 ( = Institut-
tct for sammenlitlnrmrie kulturjorsknintl; Serie A: Fore/esninger, 9).
Ordbok iWer svenska ulgiven av Svenska akademien. Lund, 1898-.
Orienlalische Bibliographie. Berlin, 1888-.
Osthoff, H., Das Verbum in der Nominalcomposition. Jena, 1878.
Ostoff, H., and Brugmann, K., Morphologische Unlersuehungen. Leipzig,
1878-1910.
Paget, R, Human 8peech. London, 1930.
Palmer, A. S., Folk-elymology. London, 1882.
Palmer, H. E., English intonation. Cambridge, 1922.
Palmer, H. E., A course in English phonetics. Cambridge, 1922.
Palmer, H. E., A grammar of 8pOken English. Cambridge, 1924.
Palmer, H. E., The principles of Lnguage-study. London, 1921.
538 BIBLIOGRAPHY
BIBLIOGRAPHY 539
Palmer, H. E., The prindples of romanization. Tokyo, 1931.
Palmer, H. E., Martin, J. V., Blandford, M. A., A dictJnary of EnglJh
pronunciation with American variants. Cambridge, 1926.
Palmgren, C., A chronological list of English forms of the types alive, aloOO,
agZow. School program, Norrkoping, 1923.
Panconcelli-Cala, G., Einfhrung in die angewandte Plwnetik. Berlin, 1914.
Panconcelli-Cal zia, G., EzperimenteUe Phonetik. Berlin and Llipzg, 1921
( = Sammlung Gikichen, 844).
La parole. Paris, 1891-1904.
Parry-Williams, T. H., The English element in Welsh. London, 1923.
. Passy, J., and Rambeau, A., Chreslorrw.thiefranaise. Fourth edition, Llip-
ZIg, 1918.
Passy, P., L'enseignement de la lecture. London, 1916 (Supplement to Mat-
tre phontique).
Passy, P., 1!:tOOe $UT les changements phontiques. Paris, 1890.
Passy, P., Petite plwntique compare. Second edition, Llipzig, 1912.
P., Les sons du franais. Eighth edition, Paris, 1917 [English trans-.
latlOII, The sounds of the French language. Second editioD, Oxford, 1913.J
Passy, P., and Hempl, G., /nternmwnal and English-French
dictiorwry. New York, 1904.
(Passy, P., and Jones, D.,) Principles of the InterruLtional phonetic auoci-
ation. London, 1912 (Supplement to Matre phontique).
Paul, H., Deutsche Grammatik. Halle, 1916-20.
Paul, H., Grundriss der gerrrmnischen Philologie. Second edition, Straas-
burg, 1900--09. Sorne of the contributions have appeared in later editions
as separate volumes; see Behaghel, Geschichte; Kluge, Urgermanisch. '
Paul, H., Primipien der Sprachgeschichte. Halle, 1880; fifth edition, 1920.
The second (1886) edition was translated into English by H. A. Strong, Prin-
Giples of the hisWry of langUQ{le, London, 1889; an adaptation of the same
tion is Strong-Logeman-Whccler.
Pedersen, H., Le grouprmunt des dialectes i1U:HmTopens. Copenhagen,
1925 ( = Det k. danske videnskabcrnes selskaJJ; meddel-
elser, 11.3).
. Pederseu, H., Linguistic science in the nineteenth century; English
tlOu by J. Spargo. Cambridge, Mass., 1931.
PederBCn, H., Vergleihende Gramrrw.tik der keUischen Sprochen. Gttingen
19o9-13. '
T.he transcription of lndian langUQ{les. Washington, 1916 (=
Smtthsoman mtsceUanellus coUeclions, 66.6).
PMLA: Publications of lhe Modcm language association of America.
more (now Menasha, 'Vis.), 1886--.
Pott, A. F., Elymologische Forschungen auf dem Gcbiete der
8chen Sprachen. Llmgo, 1833. Second (entirely different) edition, 1859-76.
Pound, L., Blends; their relation J English ward forma/ion. Heidelberg
1914 ( = Anglislische Forschungen, 42). '
Poutsma, H., The characl6Ts of the English verb. Groningen, 1921.
Poutsma, H., A grammar 0/ lute modern English. Groningen, 1904-26.
Preyer, W., Die Seele des Kindes. Seventh cdition, Leipzig, 1908.
Procudings of tM 21st international congr688 of Am8ricanists. Part 1. The
Hague, 1924.
Proceedings of tM 28d interruLtional congrU8 of Americanists. New York,
1930.
PublicatWn8 of the English dialect 8ociety. London, 1873-.
The quarterly journal of speech. Chicago, 1915--.
Raith, J., Die englischen Nasawerben. Llipllig, 1931 ( = Bei/rage ZUT engli-
schen Philologie, 17).
Rask, R. K., Underspgelse cm det nordiske eUer islandske sprays oprindelss.
Copenhagen, 1818.
Raumer, R. von, Geshichte der gerrrw.nischen Phil<Jlogie. Munich, 1870
( = Geschichl.e der Wissensdwften in Deutschland; Neuere Zeit, 9).
Rf!1)uc des patois ga/W.-romans. Paris, 1887-93.
Riec, C. C., The PhonoloUy of GdJlic clerical Latin after the rizth century.
Dissertation, Cambridge, Mass., 1902.
Riee, S. A., MetMas in social science. Chicago, 1931.
Ries, J., Was ist ein SaUf Prague, 1931 ( = his Beitrage zur Grundlegung
der Syntax, 3).
Ries, J., Was ist Synlaxt Second edition, Prague, 1927 ( = his Brilrdge
zur Grunlegung der Syntax, 1).
Rippmann, W., Elements of phanetics. Second edition, New York, 1903.
.Roedler, E., Die Ausbreitung des s-Plura/8 im Englischen. Dissertation,
Kiel, 1911.
Romani. Paris, 1872-.
Romanische Forschungen. Erlangen, 1883-.
Ronjat, J., Le dveloppement du langage observ chez un enfant bilingue.
Paris, 1913.
Rosenqvist, A., und Lesebuch der finnischen Sprache. Llipllig, 1925
(Sammlung Jgel).
RotllOll, R, Die DeminWivbildungen im Neuenglischen. Heidelberg, 1910
( = Anglistische Forschungen, 31).
Rousselot, J., Les modifr.=tWn8 phontiques du langage. Paris, 1893 (also
in Revue des patois, volumes 4, 5, and 5 supplement).
Rousselot, J., Principes de phontique exptNmentale. Paris, 1897-1908.
RP: Revue de phontiqu.e. Paris, 1911-.
Russell, G. O., Speech and voice. New York, 1931.
Russell, G. O., The vowel. Columbus, 1928.
Russer, W. S., De Gerrrw.ansche klankverschuiving. Haarlem, 1931 ( = Ne-
derlandsche bijdragen op het gebied TIan Germaansche philologie en linguistiek, 1).
Baer, D. J., Smith, F., Hughes, J., The Ililingual problem. Aberystwyth,
1924.
Samp80n, J., The dialect of the Gypsies of Wales. Oxford, 1926.
Bandfeld, K., Linguistique balkanique. Paris, 1930 ( = Colledim linguis
tique publie par la Socit de linguistique de Paris. 31).
Sapir, K, LangUQ{le. New York, 1921.
Scharp6, L., Nederlandsche uitspraakleer. Lier, 1912.
Schauerhammer, A., Mundart und Heimat Kaspar Scheits. Halle, 1908
( = Herm&J(l., ).
540 BIBLIOGRAPHY BIBLIOGRAPHY 541
Schleicher, A., Compendium der verrjlcichende:n Grammatik der iruiogerm<1n-
ischen Sprachen. Weimar, 1861; fourth edition, 1876.
Schmellcr, J. A., Bayeri.sches Wiirterbuch. Second edition, Munich, 1872-
77.
SchmclIer, J. A.. Die Mundarten Bayerns. Munch,1821. A partial reprint,
with an index aB Il separate volume (Registerband), by O. Mau&;er, appeared
in 1929.
Schmidt., J., Die Verwandtschaftsverhaltnisse der indogermanisclum, Sprachen.
Weimar, 1872.
Schmidt, W., Die Sprachfamilien und Sprachenk:reise der Erde. Heidelberg,
1926 ( = KuUurgeschichUiche Hibliothek, 1.5).
Schonfeld, M., Historiese grammatika van het Nederlands. Second edition,
Zut.phen, 1924.
Schrader, O., Reallexikon der indogermanischen Allertumskunde. Second
cdition, Berlin and Ldpzig, 1917-29.
Schrader, O., Sprachverg/eichung und Urgeschichte, Third edition, Jena,
1906-07. English translation of the second (1890) edition, l'rehis/oric anti
quilies of Ihe Aryan peoples. L{)udon, 1890.
Schuchardt, H., Slawu-Dcutschcs und Slawo-ltalienisches. Graz, 1884.
Schuchardt, II., Die Spro.che der SaramakkanegCT in Surinam. Amsterdam,
191:4 ( = Verhandelingen der k. Akademie van wetenschappen; Afdeeling fet-
terkunde; Nieuwe reeks, 14.f.
Schuchardt, H., Ober die Klassifiko.tion der romanisehcn ,lilundartcn. [Graz,
1900.) Reprinted in Schuchardt-Brevier, lOG.
Schuchardt, H., Ober die Lmolgeselze. Berlin, 1885. Reprinted in Schu-
chardt-Breder, 51.
Schuchardt, IL, Der Vokalismus des VulgarlateiIIS. Leipzig, 186....f>7.
Schuchardt-Brevier: Hugo Schuc/wrdl-Brevier. Second edition, Halle, 1928.
Scripture, E. W., 'The elements of experimenlal phonetics. New York, 1902.
Seip, D. A., Norsk sproghistorie. Christiania, 1920.
Siebs, T., Deutsche BhnenauBsprache. .Fiftccnth cdition, Cologne, 1930.
Sievers, K, AngelBiichsische Grammalik. Third edition, Halle, 1898 ( =
Sammlung kurzer Grammatiken germanischer Dialekte, 3). English tr:msla.
tion by A. S. Cook, under the title An Old En.qlish grammar, Boston, 1903.
'Sievers, E., Grumlzge der Phonelik. Fifth edition, Leipzig, 1901 ( = Bib-
lialhek iudogermanischer Urammaliken, 1).
Sievers, E., Ziele und Weg6 der Schallanalyse. Heidelberg, 1921 ( = Ger-
manische Bibliothek, 2.14; also in Stand und Allfguben, 65).
Sitzungsberichte der philasophisch-historischeu Ktasse der Akademie der Wis-
Vienna, 1848-.
Skeat, ,\Y. W., An etymological diclionary of the ErUJlish language. Third
edition, Oxford, 1898.
Skeat, W. \V' j Euglish dialecls. Cambridge, 1911 (Cambridge manualB of
$cience and literature).
Smetanka, E., To'chechische Grammalik. Berlin and Leipzig, 1914 ( =
Sa.mmllmg Goschen, 721).
Smit.h, R., Gullah. Columbia, S. C., 1926 ( = Bulletin of the University of
Sauth Carolinu, 190).
Soames, L., Introduction ta Engluh, French, and German phonetics. Third
edition, London, 1913.
Soerensen, A., rolnische Grammalik. Berlin, 1900.
Sommer, F., Handbuch der lateiniBchen Laut- und Formenlehre. Second
edition, Heidelberg, 1914 ( = lndoqermanische Bibliothck, 1.1.3).
srE: Society far pure EngliBh; Tracfs. Oxford, 1919--.
Sperhcr, H., Einfhrunfl in die Bedeulungslehre. Bonn and Leipzig, 1923.
Sprengling, M., The alphabet. Chicago, 1931 ( = Oriental inslilule communi-
cations, 12).
Stand und Aufgaben der Spra'chtlJissert8chajt,' Festschrift /Ur W. Streitberg.
Heidelberg, 1924.
Steinthal, H., Charakteri.stik der hauptsachlichsten 'Typen des Sprachbaues.
Second edition, revised by F. Misteli. Berlin, 1893 ( = bis Abriss der Sprach-
W8sensehajt, 2).
Stcintha!, H., Geschichte der Sprachwissert8chaft bei den Griechen und Riimern.
Berlin, 1863.
Steinthal, H., Der Ursprung der Sprache. Fourth edition, Berlin, 1888.
Stender-Petersen, A., Slavisch-germanische Lehnw()I'tkunde. Gothcnburg,
1927 ( = Goteborgs k. och vitterhets-samhall.e8 handlingar, 4.31.4).
Stern, C. and W., Die Kindersprache. Leipzig, 1907.
Stern, G., Meaning and change of meaning. Gothenburg, 1932 ( = Goteburg
hogskolas lirsskrift, 38.1).
Stolz, F., and Schmalz, J. H., Lateinische Grammatik. Fifth edition, Munich,
1928 ( = Handbuch der klaBsischen Altertumswissenschaft , 2.2).
Stratmann, F. H., A l'vfidJle English dictionary. Oxford, 1891.
Strecker, K., Einfhrung in das MitteUatein. Berlin, 1928.
Strciff, C., Die Laute der Clarner Mundarten. Frauenfeld, 1915 ( = Beitrage
zur schweizerdeutschen Grammatik, 8).
Streitbcrg, W., Gotisches Bleme:ntarbuch. Fifth edition, Heidelberg, 1920
( = Germanische Bibliothek, 1.2).
Streitbf'.rg, W., Urgermanische Grammatik. Heidelberg, 1896 ( = Sammlung
von Blementarbchern der altgeTmanischen Dialekte, 1).
Streitberg, W., and others, Geschichte der indogermaniBchen Sprachwissen-
schajt. Stras.sburg (now Berlin), 191G- (part of Grundriss der indoger-
maniBchen Alterlumskunde, begrndet von K. Brugrnann und A. Thumb).
Streitberg Festgabe. J-Bipzig, 1924.
Strong, H. A., Logeman, W. S., Wh!er, B. 1., Introduction to the sludy of
the hislory of language. London, 1891 (see Paul, Prinzipien).
Studies in English philololJY; A miscellany in honor of F. Klaeber. Minuea
polis, 1929.
Studies in 1=of Hermann Collilz. Baltimore, 1930.
SturtevfLnt, E. H., Linguistie change. ChicfLgo, 1917.
Sundn, K., Contributions to the study oj elliptical words in modern English.
UppSfL\fl" 1904.
Stterlin, L., Geschichte der Nomina AgentiB im Germanischen. StraB8burg,
1887.
Stterln, L., Neuhochdeulsche Grammatik. Munich, 1924-.
Sweet, H., An Anglo-Saxon primer. Eighth edition, Oxford, 1905.
542 BIBLIOGRAPHY BIBLIOGRAPHY 543
Sweet, H., An Angfo.Saxcm reader. Eighth edition, Oxford, 1908.
Sweet, H., CoUeeted papera. Oxford, 1913.
Sweet, H., HandOOok of pM!<JJI:ica. Oxford, 1877.
Sweet, H., A of sounda. Oxford, 1888.
Sweet, H., TM of language. London, 1900.
Sweet, H., A Mw English grammar. Oxford, 1892-98.
Sweet, H., TM practical sludy of languages. New York, 1900.
Sweet, H., A primer of phonetica. Third edition, Oxford, 1906.
Sweet, H., TM sounda of Engluh. Second edition, Oxford, 1910.
Szinnyei, J.. Finnuch-ugrische Sprachurissenschaft. Leipzig, 1910 ( = Samm-
lung GOSCMn, 463).
Tamm, F., EtymOJgisk svensk orrIbok. Uppsala, 1890--.
T APA: Transactions of the American philologil associatm. Hartford,
Conn. (now Middletown, Conn.), 1871-.
Taylor, A., The proverb. Cambridge, Maas., 1931.
Teichert, F., Ober daa Auaslerben aller W6rler im Verlaufe der englischen
Sprachgeschichte. DiBBertation, Erlangen, 1912.
Terraeher, A. L., Les aires morphologiques. Paris, 1914.
Teulhonuta; Zeitschrifl fUr deu/.sche Dia1ekifarschung und Sprachgeschichte.
Bonn und Leipzig, 1924-.
Thesaurua lingua.e Latiruu edilua auctaritate et wnsilio ademiarum quinqtre
Germanicarum. Leipzig, 1904-.
Thomas, A., NOWJeaux essau de philolor;iefranaue. Paris, 1904.
Thomsen, W., Ober den Einfiuss der germanuchen Sprachen auf diefinnuch--
lappischrm. Halle, 1870.
Thorndike, E. L., A teacher's word book. New York, 1931.
Thumb, A., Grammatik der neugriechuchen Volkssprache. Berlin and Leip-
zig, 1915 ( = Sammlung Goschen, 756).
Thumb, A., Handbm:h der neugriechuchen Volkssprw;he. Strasaburg, 1910.
Thumb, A., and Marbe, K., Experimentelle Untersuchungen ber
gische Grundlagen der sprachlichen Analogiebildung. Leipzig, 1901.
Thurneysen, R., Die EtyrnologUJ. Freiburg i. R, 1905.
Thurneysen, R., Handbuch dea Altirischen. Heidelberg, 1909 ( =
manuche Bibliothek, 1.1.6).
Thurot, C., Notices et extraits de divers manuacrits latins prJUr servir l'histoire
des dcclTines grammaticales au moyen due. Paris, 1868 ( = Notices et extraits
dea manuscrits de la bibliothque impriale el auires IJibliothques, 22.2).
Tijdschrijl voor Nederlandsche taal- en leUerkunde. Leiden, 1881-.
Toll, J. M., Niederlnduches Lehngut im Mittelenglischen. Halle, 1926
( = Sludien zur englischen Philologie, 69).
Torp, A., Nyrwrsk etymalogisk ordbok. Christiania, 1919.
Torp, A., Wortscha der germanischen Spracheinheit. Gottingen, 1909 ( =
A. Fiek, Vergleichendes Wrlerbuch der indogermanischen Sprachen, volume 3,
fourth edition).
Torp, A., and Falk, H., lydhistarie. Christiania, 1898.
Tourtoulon, C. J. M., and Bringuier, M. O., tude sur la limite gographique
de la langue d'oc el de la langue d'D'il. Paris, 1876.
Travaux du Cercle linguuliqil de Prague. Prague, 1929-.
Travis, L. E., Speech pathology. New York, 1931.
Trofunov, M. V., and Jones, D., The prcmuncialion of RusBian. Cambridge,
1923.
Uhlenbeck, C. C., Kurzgefasstes elynwlogisches Worlerbueh der altinduchen
Sprache. Amsterdam, 1898-\19.
Uhler, K, Die Bed.euI,ungsgleichheil der altenglischen Adjekliva und Adverbia
mil und ohne .-1ic (-lice). Heidelberg, 1926 ( = Anglislische Forschungen, 62).
Uhrstrom, W., Pickpocket, turnkey, and similar formations in
English. Stockholm, 1918.
University of Washingtcm p1Jlicalions in amhropology. Seattle, 1920-.
van der Meer, M. J., Historische Grammatik der niederlanduchen Sprache.
Heidelberg, 1927 ( = Germanische Bibliothek, 1.1.16).
van der Meulen, R., De IloUandsche zee- en scheepslermen in het RusBisch.
Amsterdam, 1909 ( = Verhandelingen der K. akademie van welenschappen;
AJdeeling leUerkunde,' Nieuwe reeks, 10.2).
Verwij.s, E., and Verdam. J., Middelnederlandsch woordenlwek. The Hague,
1885-1930.
Vitor, W., Die AU8Sprache des SchTiftdeutschen. Tenth edition, Leipzig,
1921.
Vitor, W., Deutsches Aussprachew6rtsrbuch. Third edition, Leipzig, 1921.
Vitor, W., Elemente der Pfwnelik. Sixth edition, Leipzig, 1915.
Vitor, W., German pronunciatitm. Third edition, Leipzig, 1903.
Vitor, W., Die Methodik des neuaprachlichen Unterrichts. Leipzig, 1902.
Vollmoller, K. G., KTilischer Jahresbericht ber die Fortschritte der
ischen PhiloWgie. Munich and Leipzig (then Erlangen), 1890--1915.
Vondrak, V., VergleichendeslavischeGrammatik. Second edition, Gttingen,
1924-28.
Vox: [nternationales Zentralblau fr experimenteUe Phomtik; Vox. Berlin,
1891-1922.
Wackernagel, J., Altinduche Grammatik. Gttingen, 1896--.
Wackernagel, J., Vorlesungen ber Symax; Erste Heihe. Second edition,
Basel, 1926. Zweite Reihe, Basel, 1924.
Walde, A., Lateinisches etynwlogisches Worterbuch. Second edition,
herg, 1910; third edition, 1930- ( = lndogermanische Bibliothek, 1.2.1).
WaJde, A., and Pokorny, J., Vergleichendes WOrlerbuch der indogermanischen
Sprachen. Berlin and Leipzig, 1930.
Warnke, C., On thefarmalion of Engluh words Dy means of ab/auto DiBBerta-
tion, Halle, 1878.
Weekley, E., A concise etymological dictionary of rrwdern English. New
York, 1924.
Weekley, E., The r01m'>nce of names. Third edition, London, 1922.
Weekley, E., Surnames. New York, 1916.
Weigand, G., Linguistischer Allas des dacorum<inischeri. Sprachgebeites.
Leipzig, 1909.
Weiss, A. P., A theorelica/ basis of human biihavior. Second edition, Columbus,
1929.
West, M., BilingualUm. Calcutta, 1926 ( = Bureau of education, [ndia;
Occasional reports, 13).
544 B.IBLIOGRAPHY BIBLIOGRAPHY 545
Weet, M., Leaning to read aforeign language. London, 1926.
De West-Indische gi.<bJ. The Hague, 1919-.
Wheatiey, H. B., A diclionary of reduplicaled WQrds in the English language.
London, 1866 (Appendix to the Transactions of the PhiWlogiccl society for
1865).
Whee1er, B. 1., Analogy and the scope of ils applicalion to langua.ge. Ithaca,
1887 ( = CorneU univeTsity studies in classical phiWlogy, 2).
Whitney, W. D., Language and lhe sludy of language. New York, 1867.
Whitney, W. D., The life and growth of language. New York, 1874.
Whitney, W. D., A Sanskrit grammar. Third edition. Boston, 1896.
Wilmanns, W., Deutsche Grammatik; volume 1, third edition, StraB1lburg,
1911; volume 2, second edition, 1899; volume 3, 1906.
Wilson, S. A. Kinnier, Aphasia. London, 1926.
Wimmer, L., Die Run8nschl'ift. Berlin, 1887.
Winteler, J., Die KerenzeT Mundart. Leipzig, 1876.
Wiss8nschafllieh Beihefte ZUT Zeitschrift des AUgemeinen deldschen Sprach-.
vereins. Leipzig, 1891-.
Wissler, C., The Am.erican lndian. Second edition, New York, 1922.
WQrler und Sachen. Heidelberg, 1909-.
Wrede, F., Deutscher Sprachallas. Marburg, 1926--.
Wright, J., The English diateel dictwnary. London, 1898-1905.
Wright, J., The English dWlect grammar. Oxford, 1905 (also as part of bis
English dialect dictionary).
Wright, J., and E. M., An elemen.tary histrYrirol New English grammar.
London, 1924.
Wright, J. and E. M., An elemen.tary Middle English grammar. Second
edition, London, 1928.
Wundt, W., Sprachgeschichle und Sprachpsychologie. Leipzig, 1901.
Wundt, W., Volkerpsychologie; Ers/er Band; Die Sprache. Third edition,
Leipzig, 1911.
Wyld, H. C., Hisl.oricalstudy of the 'flWlheT longue. London and New York,
1906.
Wyld, H. o., A histl'lTy of nwdern colloquial English. London, 1920.
Wyld, H. o., A short history of English. Third edition, London and New
York, 1927.
Wyld, H. C., Studies in English rhymes from Surrey to Pope. London, 1923.
Xandry, G., Das skandinamsche Element in den neuenglischen Dioteklen.
Dissertation (Mnster University), Neu Isenburg, 1914.
Zauner, A., Romanische Sprach'UJUJS8nsclwjl; 1. Teil. Fourth edition, Berli'!
and Leipzig, 1921. e. Teil. ThW edition, 1914. (= Sammlung G&Jchen,
128; 250).
ZdP: Zeitschrift fur deul8ch PhiWlogie. Halle, 1869-. Olten referred to as
ZZ (" Zachers Zeitschrift ").
Zeitschrift fur Eingeborenen8prachen. Berlin, 1910-.
Zeuss, J. K., Grammatica Celtica. Berlin, 1853; second edition, by H. Ebat,
1871.
Ziemer, H., Junggrammatische Slreijzge im GelYiele der Syntax. Second
edition, Colberg, 1883.
Zipf, G. K., Beleeled studies of the principte of rdativefrequency in language.
Cambridge, Mass., 1932.
ZrP: Zeil$chrift fur romanische Philologie. Halle, 1887-; Supplement; Bib-
liographie.
ZvS: Zeil$chrifl fr VeTflleichende SprachfOTschung. Berlin, 1852-. Often
referred to as KZ ("Kuhns Zeitschrift").
TABLE OF PHONETIC SYMBOLS
The phonetic a I p h ~ b e t used in this book is a slightIy modified
fOTIn of the alphabet of the International Phonetic Association.
The main principle of this alphabet is the use of a single letter for
each phoneme (distinctive sound, see Chapter 5) of a language.
The symbols are used very fiexibly, and represent rather different
sounds in the transcription of different languages, but the use is
consistent within each language. Thus, [t] represents an EngIisb
5O,und in tin [tin] and a somewhat different French sound in tout
[tu] 'aIl.' Additional symbols are used only when a language dis-
tinguishes additional phonemes; symbols such as italic Ii] or capital
[T] are used in addition to [t] onIy for languages like Russian or
Sanskrit which distinguish more than one phoneme of the general
type of [t].
The following indications are ta be read: "The symbol ..
represents the general type of the sound in ..."
[a] palm [pam]
[a] hot [hot]; French bas [bal
[A] son, sun [sAn]
1
[b] big [big]
[cl chin [Cin]
[} Modern Greek ['ei] 'has'
[dl do [duw]
[t5] then [t5enJ
[el men [men]; French gai [geJ
[el French petit [peti]
[el man [men]; French deUe [det]
[f] Jew [fjuw]
[g] go [gow]
['Y] Dutch zeggen ['OO'Ye]
[hl how [haw]
[il tin [tin]; French fini [fini]
[il Turkish [kiz] 'girl'
[j] yes [jes]
[j] Jg [Jig]
1CUBtomarily lllled in transcribing British EngliBh; 10] would do jUBt as weil.
547
548 TABLE OF PHONETIC SYMBOLS
TABLE OF PHONETIC SYMBOLS 549
[k] cook [kukJ
[1] lip [lip]
[] Halian figlio ['fio]
[ml me [mij]
[n] no [now]
[IJ] sing [siIJ]
[Pl French signe [siJl]
[0] son, sun [son]; French beau [bol
fol saw [8J]; French homme [;lm]
[ ~ ] French peu [pp]
[] French peuple [ppl]
[pl pin [pin]
[r] red [red]; French riz [ri]
[s] say [sej]
[8] show [sow]
[t] tin [tin]; French tout [tu]
[6J thin [Gin]
ru] put [putJ; French tout [tu]
[v] veil [vejl]
[w] woo [wuw]
[x] German ach fax]
[y] French vu [vyJ
[4J French lui (I4i]
[zJ zoo [zuw]
[.l:J rouge [ruw.l:]
[C] Danish hus [hu?sJ
Additional signs:
'''hen a language distinguishes more than one phoneme within
any one of the above types, variant symbols are introducedj thus,
capitals dcnote the domal sounds of Sanskrit [T, D, N], which arc
distinct from dental [t, d, n], and capital [l, u] denote opener
varieties, distinct from ri, u], as in Id Bulgarian; italic leUers are
used for palatalized consonants, as in Russian [bit] 'ta beat,' dis--
tinct from [bit] 'way of being.'
A smalt vertical stroke under a 1etter means that the sound fonns
a syllablc, as in button ['botr,t].
A sman raised [n] after a letter means that the sound is nasalized,
as in French bon [bon]. A sman raised [W] means that the preceding
sound is labialized.
The mark l'] means that the next syl1able is accented, as b e ~
nighted [be'najtcd]. The signs [H , d are used in the same way,
wherever severa1 varicties of accent are distinguished. Numbers
[1 234] indicate distinctions of pitch.
The colon means that the prcceding sound is long, as in German
Kahn [ka:n], contrasting with kann [kan].
Other marks of punctuation [. , 1] denote modulations in the
sentencej [t] is used for the modulation in Who's there? ['huw z
'tiejrt], contrasting with Are you there? lar ju "i5ejr?]
ADDITIONS AND CORRECTIONS
Page 13. Alhanese, the form used throughout the book, should
perhaps be replaced by the more current Alhanian.
Page 14. On Rask, see the Introduction by H. Pedersen to
Rask, R. K., Ausgewahlte AbhmuiZungen, Copenhagen, 1932-33.
Page 53. In expressions like "our Southwest," the angle of
vision ia that of the Uniu-d States of North America.
Page 59. For FaTOO$e read Faroe.
Page 61. Ladin iB Bpoken also in the southem Tyrol and in
north-eaatern lto.ly.
Page 65. The term Aceadian is now preferred ta Babykmian-
Assyrian.
Page 70. On the basiB of an entirely new definition and theory
of the relationship of languages, the Russian scholars N. 1. Marr
and F. Braun view the CaucaBian languages as survivaIs of a once
widespread Japhetic family, soroe features of which appear also in
Basque, in Semitic, and even in Indo-European languages, notably
Arroenian. However, the stateroents of these relations and the
evidence for thero do Dot seero precS\'l enough ta warrant accept.
ance. See Materialy po jafetieskomu jazyJwz'fU1niiu, Leningrad,
1910; Marr, N. L, Der japhetitische Kaukasu&, Bellin, 1923
( = Japhetitische Btudien, 2); Marr, N. 1., Etapy rawitija jafetieslw,j
teorii, Leningrad, 1933 ( = Iz1:wannyje raboty, 1).
Page 143. Instead of "(2) are so distantly carrelated," etc., it
would be better ta say:'" (2) are 80 variably correlated with speech-
forma that these cannot guide us in determining the speaker's
situation."
Page 164. 'l'he exarople Backwaier! seelIlB ta be an unusual
speech.form; DismmJ,nt/ would be better.
Page 284. 'l'he cuneiform charactera were not "scratched;" but
pre58ed with a stylus ioto tough clay.
Page 323. On Joseph Wright (1866-193Cl). Bee Wright,. S. M., The
Life of Joseph Wnght, London, 1932.
Page 328. On the map, the dotted patch which represents the
Frisian area extends too far 6Outhward; the dots abould reach only
551
552
ADDITIONS AND
to the boundary line which can he secn on the map.
Page 332. On the map, for Kerensrn read Kerenzen.
Page 358. The Old English word for 'become' was doubtless
pronounced not with [ElJ, but with [l representing an aIder [el.
Page 363. Greek [lelejpsaJ '1 ldt' is probably a late formation;
a relevant example would be Primitive lndo-European *[lete:rp8lPJ
'1 satisfied ' (Sanskrit [lata:rpsam] '1 was pleased '), Greek [Ieterpsa.].
Pagc 372. Latin agmen is a new formation and does Dot preS/:;rve
old -g-m-; in this combination the 9 was lost, witness e:z:dmen 'swann.'
With fulmen we should contrast, rather, mnUMJ't 'rampart'
derived from munire 'ta fortify.'
Page 413. Jespersen, Linguistica, Copenhagen, 1933, page 420,
does not believe that the suffix -ster was ever restricted ta females.
Page 414. We should add the following example, because it gives
the historical explanation of a phenomenon described earlier in the
book.
The Latin adjective grandis (accusative grandem, etc.) leada
phonctically ta k French grand [gr], masculine and feminine;
actually a new feminine form grande [grd] has been created aDa-
logically, according ta the type of adjective that loses a final con
sonant in the masculine infiection ( 13.7); the oid feminine form
survives as a prior member in certain compound wards ( 14.3).
Page 423. To crayflsk, etc., add: French fMUsseron re-shaped in
English as muskroom.
Page 512. Note on Chapter II: see aIsa the lively and readable
survey of linguistics by J. R. Firth, Speech, London, 1930,
Page 514. See also Armstrong, L, K, The Pkonetics of French,
Lundon, 1932.
Page 515. On pitch in Ja.panese, see also 0, Pletner in BSOS
3.447 (1924).
Page 519. On Verner, Bee a1so Jespersen, Linguistica, 12. On
Primitive Indo-European formulre, see C. D. Buck in Lg 2.99
(1926).
PagB .520. On English voicing of spirants, see also Jespersen,
Linguislica, 346.
Page 324. On foreign-language teaching, see a1so Palmer, H. E.,
The ScientiJw Sindy and Teaching of Languages, London, 1911.
Page !J33. Athird cdition of D. Joues' Outline of English Plwnetics,
Lonon, 1932.
INDEX
4 100, 102-106, 109f., 112, 114, 117,
288,488 12H., 124f., 127, 129, 152,187,325,
ablative 263, 315 361, 366r., 374, 394, 396, 401, 444,
abnarrnal 100,378 464 471... 480, 484f., 488, 498, 502
abso!ute 170,186-189 American Indian 7, 19, 42f., 71-73,
7 429f 456f
87, 97, 102, 127, 283, 404, 455f.,
abstract 205, 2 1, ., . 473
accent 80, 82, 182, 308f., 358f., 385, 458, 464, 469,
. h t Arnharic 66
450, Bee pIte ,S re65 arruedita 235
accretion 414, 417 anacolouthan 186
accusative 165,272,388,392, 457f. 362 "66 376 391
Aoka 63 analagic change ....." , ,
acoustic 77-79, 93, 128 393, 404-424, 426, 436, 439, 509
action 172---175,267,271 anaiogy 275-277,4.54,501
actian-goal 192, 197,267 analphabetic natation 86
action noun 236 analytic 207
active, see actar-action anaphora 249-266
actar 172-175, 267, 297 anaptvxis 384
actor-acti on 165-167, 172---175, 184f., Ang1a:'Frisian 58, 304, 311f., 4.52
Of Anglo-Saxon, see Old English
1!lOf. , 194,19 . animal 27, 155 ..----
adam's-apple 27 272
adaptation 420--424, 426, 440, 449f., animate 193, 232, 262,
f
animatcd 156, 197
458, 492 . A 't 44 71
addrE'ss 148, 152, 255f., 40If. nnaml 13 ,
C 7 Annatom 257
J
6
, i05, 173, 188,192, 198, answer 91, 115, 159, 163, 176f., 179,
202-206,231, 201,271,387f 249-263
adult language 55,463, 485 cv
adverb 175,177, 197f., 237,258,260, antepenult 182
262f., 271, 433--435 anticipatory 254, 258
affix 218, 414, 454, 509 aorist 362-364, 456
atfricatc 120, 133, 214, 342, 378 Apache 72
Afghan 62 aphasia 35f.
Africa 7,56, 67,87,94, 99, 117,472 aphoristic 152, 177, 438
Afrikaans 474 apical 98, 100, 102
agent 221,366, 412f., 454f. 5
agglutination 207f. aposiopesis 186
165f., 190--194 apposition 6, 186, 420
Amu 70
Albanese 13, 15, 62, 312, 315f., 467, 10, 21, 44, 6t, 89, 99,
Ali72J King, 17, 47, 281, 295 101, 154, 243f., 289,458
Algonquian 72, 193, 198, 241, 256f., Aramaic 66,289,294
271f., 359f., 371, 381f., 396, 402 73
alliteratian 296, 395 Arawak 73
alphabet 79, 85f., 128,290-294,500- archaic 152f., 292, 331, 401-404, 487
503 G Aristarchus 5
Alss.tian, sec erman Armenian 13, 15, 62, 307, 312, 315f.,
Altaic 68f. 7 319f 470
altemation 164, 210-219, 37D-3 6, .,
f 418f arrangement 163-168
i
3S
Y, ., . article 147, 192 204, 259, 261, Z71f.,
English 44, 47-52, 81, 98, 419,458, 470f.
553
555
conflict 463-475
congruence 6, 19If., 204, 224, 253,
256, 263, 270, 273
eonjunct 179f., 197f., 256, 260
conjunction 195, 198, 244, 269, 420,
469
connotation 151-157, 163, 197, 214,
402f., 421, 424, 441f., 496-498, 502
conquest 42L' 57, 60L' 64, 66, 68-70,
313L, 361, 386, 461-470, 472
consonant 102, 117-121, 217, 219,
243-246, 370-381
constituent I60f.
construction 169, 183-246, 407, 433,
437, 453f.
contamination 422-426
context 409, 440f.
oontraction 380f., 411
co-ordination 195, 198, 232, 235, 269
Coptic 67
copulative 235
Cornish 13,60,307,464
coronal98
correctness 3, 2If., 48, 477,496
Coss('-sn 65
Cottian 70
counting 28f.
Cree 72, 136, 145, 147, 155, 176, 182,
193f., 257-259, 288, 359f., 371,
38lf., 396, 400,508
Creek 72
creolizcd 474f.
Crdan 65, 293
Croatian 61
cross-reference 193f., 197, 257, 439
Crow 72
cuneiform 21, 64f., 284, 287f., 293f.,
309
Curtius, G., 354
Cushite 65, 67
Cyprian 288
Czeeh 61
Dakota 72
Dalmatian 61
Danish 8-10, 53f., 59, 99-101, 106,
127, 279, 287, 299f., 314, 325, 370,
390,455,468,483f.
Dano-Norwegian 59,483f.
Darius 62
dative 272, 437
Dauzat, A., 398
deaf-mute 39,144
decay 8f.,490
dccipherment ML, 72, 293f.
de-compouud 210, 227
defcctive 223
definitc 251-261, 266, 270
definition 139-146,152,266---268,280,
408
INDEX
Chaucer 281, 295, 429, 484, 487
Cheremiss 68
Cherokee 72,288
Chevenne 72
Chickasaw 72
chld 28-31, 43, 46, 56f., 84, 140f.,
148, 157, 386, 399, 403, 409, 432,
444, s..'Ul, 485, 512
Chinese 10, 44, 57, 69, 76L, 80, 83L,
91, 100, 109, 111, 116, 176, 182f.,
188, 199-201, 207f., 243L, 252,
269, 271, 278f., 296, 388, 509
Chinese writing 21, 69, 90, 284-288
Chinook 470,473
Chinook jargon 473
Chipewyan 72
Choctaw 72
chronology 309, 340, 346, 368, 413,
416, 451-453
Chukchee 70
circumlocution 140
ci tation 89f.
class-clcavage 204-206, 241, 251, 258-
270
classification 207f.
classifier 286-288
class-meaning 146, 166,202-205,247-
251, 266-268, 271
clause 192-194, 197, 204, 25If., 263,
273, 407, 437f.
click 93f.
close transi tion 119f.
closed construction 196f., 223, 268
closed vowel 103
cl osure, stop
cluster 131-136, 183, 219, 228, 243,
335,367,370-373,383
collective 221
colloquial 52, 153
color 140, 280
Comanche 72
command 164,172, 174, 176
common noun 205, 273, 470
comparative 215, 238f.
comparative method 11-20, 38, 64,
297-321,346-364,466
compensatory 379f.
complement 230,254,200,263
completion 224, 270, 273
completive 176f., 262, 266, 439
complex 160-170, 240, 244-246,
268f.,276,405,412
compounding form 225
compound phoneme 9Of., 120, 124f.,
130-132, 135f., 167, 182
compound word 17, 38, 180-184,
209f., 224-237, 275, 382, 413-418
condensation 439
conditioned sound-change 353, 372-
385, 417-420
Bohemian 9f., 44, 54, 61, 86f., 89
95, loof., 113, 182 291 385 447'
466, 468, 483 501' , , ,
BhtIingk, O., 18f.
Bopp, F., 14f.
bOITowing 298, S06f., 320-345, 361-
367, 398, 412-416, 429, 444-495
bounded noun 205, 265
bound form 160 177-184 207-246
257 ' , ,
Brahmana 63
Brahui 70
brain 36f.
Brant 295f.
27, 31, 80, 93-102, 110, 120,
Bremer, O., 87
13, 60, 325, 414
BntIsh English 44 47 49-52 81 98
100, 102-104, I12, '114, Il8 '152'
367, 396, 484f., 488, 497f., 502 '
Broca, P., 36
Brugmann, K., 15
Bul/1;arian 15, 61, 154, 290f., 306---308
314f., 363, 371, 373, 383 423 427'
B
437, 451, 453, 457, 459, 466 470 '
urgess, G., 424 '
Burgundian 59
Burmcse 70
Bushman 67,279f.
cali 115, 164, 169 177
Cambogian 71 '
Canarcse 44 70
Cantonese 44, 69, 116
Carelian 68
Caran 65
Carib 46,73
Caroline 71
Carroll, L., 424
Carthage 66
Case 5, 165, 192, 256, 272, 297 388
392, 457, 506 ' ,
Caspian 62
Catalan 61,483
category 204,270-273 388 408
Catharine, Emprcss 7' ,
Caucasian 70 '
Cayuga 72
Celtic 12f., 16, 6Of., 188, 307f. 312
315f., 319, 386, 463L, 489 ' ,
center 174, 195f., 202 265
meaning 149,151, 402f., 431-
centum languages 316
Champollion, J. F., 293
change 5, 13-20, 38, 158 208 277
281-495, 5G9 '"
character 284-286 294
charaeter-substande 194, 202-206
INDEX
554
artificiailanguaF;e 506
arytenoids 94f., 102
.270, 272f., 280
aspiration 80-82 8-1 89 99f 129
348-351, 446' , , ., ,
assimilation 273-381 390 423
Assiniboine 72 ' ,
aSSOnance 395
Assyrian 65f., 288 293 320
asterisk 516 ' ,
asyntactic 233-235
Athabascan 72
atonic 187, 204, 244 247 250 256
261, .266, 364, 376', 31d, 418' ,
attractIOn 263 423
attribute 188, '191, 194-206 230-235
251-263, 266-269 ' ,
Australia 71
Austric 71
Austronesian 71
autly?rity 3, 7f., 496-500
auxl!lary, Bee seeondary phoneme
15, 62, 295, 315, 389, 451
avyaYlhhava 237
192, 194, 199, 263, 265 267
aym 101 '
Azerbaijan 68
Aztec 72f., 241, 287
Babylonian 65, 288, 293
baby-talk 148 472
baek 412-416, 432, 454
back vowcl 103-107 117-119 181
376-381 ' "
bahuvrihi 235
Bali 71
Baltic 13, 18, flOf., 312-319 400 423
Balllchi 62 ' ,
Ba1!-tu 19, 67, 192, 272
baSIC alternant 164 209 211f 217-
219, 222, 231, 242-244 .,
basis 127
Basque 64
Batak 310
Bavarian, see German
Beach la Mar 472f.
Bp.ll.ver 72
Bell, A. M., 86
Bengali 44, 63
Bennicke, V., 325
Benrath Line 343
Bcrber 65,67
Bihari 44 63
bilabial 98, 101
bilingual 56, 290 293f., 445, 463f.,
471,513 '
Bisaya 71
Blackfoot 72
blend 422-424
Bodo 70
INDEX
557
Germanie 57-59,298-301, passim.
fl;erund 269
Gessner, K., 511f.
gesture 39f., 111, 114f., 144, 147, 176,
250
ghost-form 293, 487
Gilbert Islands 71
Gilliron, J., 325, 395--397
Gilyak 70
gingival 98, 100, 102, 446
Gipsy 63, 313, 467, 471
Gipsy English 50, 471f.
G1arus 331
glide 96f., 118---120, 147
glosseme 264, 277f., 503
glottal 80, 82, 99, 101, 113, 118f.,
147,289,299 '
glottalizcd 99, 10lf.
glottis 94f., 97, 101, 118
goal 165, 173, 192, 197f., 229, 233,
241, 257f.,265,269, 272, 297, 457f.
goal.aetion 173, 201, 316, 471
Goropius 9
Gothlc 8,14,17,59,453,466, passim.
government 6, 192f., 197,273
gram
mar
3, 7, 135, 138, 183, 266,
274,322f.,365,408,506
grammatical feature 35,166-169,209,
214, 216, 239, 264f., 268, 275,277,
394, 467f.
grammatical termB 457
Grassmann, H., 349-355
Grebo 67
Greek 43,62, passim.
Groek alphabet 64f., 86f., 288-296
Greek gramma.r 4-7,12,208,457
greeting 148
Griera, A.,. 325
Grimm, J., 347-351, 355, 360
Grotefcnd, G. F., 293
Gujerati 44,63
gums 96,98, 100, 119
guttural 98, 127
Haag, K., 325
Hakka 69
Hamitic 65, 67
haplology 391
Haussa 67
Hawaiian 71
head 195f., 199-202,235--237
Head, H., 35f.
Hebrew 9f., 66, 89, 289, 455, 472,
519
Herero 67
Hermann, G., 6
Herodian 5
Herodotus 4, 318
Herskovits, M. J., 475
hesitation 186
INDEX
Finck, F. N., 19, 467
finite 165--167,172,185,190-197,
251f., 256f., 267, 270
Finnsh 19, 68, 86, 89, 106 109f.,
175,177,255,272,291, 298?, 306L,
465f., 470, 501
Fintlo-Ugrian 19, 65, 67f., 298, 306,
319
first person 247f., 255--258
Fischer, H., 325
Flemish, see Dutch
foreign form 131, 153f., 423f., 449,
454
foreign language 45, 54-56, 8-84,
93, 142, 148, 248, 365-367, 386,
445-475,481,497,499,503-506
foreign-learned 153f" 220, 239,241-
243, 292, 383, 415f., 421, 449, 454-
458, 464f.
form-class 146, 164-167, 185, 190,
194-196, 199-204, 21Of., 247-251,
265-276, 409
Form()1j8 71
fortis 99f., 386
Fox 72, 136, 177, 181, 218, 232, 241,
288, 359f.,371,396,400
Frankish 466f., see German
free form 160, 178, 181-206, 209,219,
243
French 43f., 61, passim.
frequeney of forms 277, 354, 389,
392-403, 405, 408f., 414, 420, 431,
435,445
frequency of phonemes 136f., 389
triction, see spirant
Frisian 8, 14f., 17,58,303-305,311,
330, 380, 385, 452, 483
front vowel 103-107, 117-119, 125,
181,376-381,410,452
Fui 67
full sentence 171-177,252,259, 262f.
full word 199f.
function 185, 194-196, 265-274
fundamental assumptioll 78, 144f.,
158f., 162
futhark 291
future 224, 272f., 415
von der Gabelentz, G., 18
Galla 67
Gallic 13, 60, 375, 463
Gamillscheg, E., 479
gender 5, 192, 211, 217, 236, 253f.,
271-273,278,280,462
general grammar 6, 20, 233, 270f.,
297,508f.
general meaning 431
genitive 231, 375, 409, 420
Georgian 70,174
German 43f., 58f., passim,
dvandva. 235
dvigu 237
eardrum 25, .31, 74f., 128, 514
East Germamc 59
Easter Island 71
Eastern Hindi 44 63
Eastman, G. 424'
Edda. 296 '
Edmont, E., 324
Egllfatian 21, 65, 67, 90, 283-289,
Elamitic 65
elevated 48, 152f., 156, 330,
Ellis, A. J., 87 323
emphatic Ill,' 171 174 186 197f
204, 261 ",.,
enclitic 187, 212
endocentric 194-196 199 202 235f
268 ",.,
EI!glish 43-45, 57f., passim
166-168,172 .
e<;I ul!'tlOnal 173-176, 201 260
Esklmo 72, 207f., 259 '
Esperanto 506
Eathonian 68, 306 465
Ethiopian 66f., 289
Etruscan 64,290,294
etymology 4, 6 15 346 351-355,
427-430 ' , ,
euphemism 401
euphony 395
Ewe 67
exclamation 6,92, lI5 147 156 164
166--172, 176f. ' , , ,
exclusive 232, 255-257
exocentric 194-196 199 235-237
240,268 " ,
4, 34, 75f. 389 423
expilclt 174 "
explosive 97
expression 196
extitlet languages 13 57 59-51, 63-
66,68,70,72, 463f.; 513
faeetions 147f., 151 153f 394 402f
421, 443, 471 ' ., , .,
false palate 75
familytrec 311f., 316 318
Faroese 59 '
favorite sentence-form 171-177 199
254,262 ' ,
146, 238, 248, 253, 270
fernllllllc 192, 211, 217, 253 410 420
field of selection 204 260' ,
Fiji 71 '
final 131-136, 181-183 218f. 245
371-374, 381f., 418f. ' , ,
finalpitch l14f., 163-171, 185
Delaware 72
Delbrck, B., 15, 18
demonstration 140
demonstrative 147 248 2"8-
260, 470 ,,"''''
Dempwolff, O., 519
denotation 146
density of communication 46f 282
326, 328, 340, 345, 403, 481" ,
dental 98, 100 lO2 214 376 378
384, 470 ' , , , ,
see anaphora, subordinate
enved 209-227, 237-246 412--416
453--458,491 ' ,
ieriving form 225
de F., 19
adjecti ve 202f,
descnptlve order 213
descriptive study lIf 16--20
274, 311 ." 158,
determinative 240-245
compound 235
203-206,262 265-269
86-88,289-291
dlalect 5, 47-52, 152 314-318 321-
.345,476--485,499 ' ,
d!alect ar 51, 477-481
atlas 51, 322-325
dJalect geography 51 321-345 361f
480 ".,
dictionary 3, 87 140 142 152 178
. 320-323, 486' , , ,
dl::tlOnary meaning 142 148
DICZ, }'., 16 '
79, 85f., 89, 291 451
150, 157 226 400
Dlilka 67 "
Thrax 5
90, 124f., 13lf.
dISf.i9f.ed speech 28, 30, 141-143,
d!ss!milation 349-351, 390 450
d!stlDctive 77-80, 141, 366
dlsturbance of speech 34-37
Dobrowsky, J., 483
Dodgson, C. 424
Dogrib 72 '
domain 247-251
dom!l'l 98, lO2, 470
dommant 435
Donatus 6
dorsal 98, 101
double consonant 110, 119 132-134
181., ?28, 363,368,373' ,
DraVidian 44, 70, 470
:Inal 255,257 482
Du G., 513
iurat!on, see qnantity
iuratlVe 272f
Dutch 44,59:328-331, passim.
556
558
INDEX INDEX 559
hiatus 134
Hickes, G" 8
hieroglyphs, see picture writing
high vowel lO3-J07, 120
Hmcks, E., 293
hisa 100
historical present 156, 272
Hittite 64f., 293,309
hoarse h 101
home language 56,60
Homeric poems 5, 62, 295, 319
homonym 145, 150, 161, 179, 183,
205, 209, 214, 223-225, 232, 286,
354, 367, 369L, 388, 392, 396-399,
410, 412, 416, 420, 433f., 436, 439,
502
Hopi 72
Hottentot 67
Humboldt, W. V., 18f.
Hungarian 19, 44, 61, 68, 99, 3t3f.
389 '
hunting 155,400
Hupa 72
Huron' 72
Hus, J., 483
hyperbole 426
Hyperborean 70
hyper-forms 302, 309, 330, 449, 479,
499
hypochoristie 157, 424
hypostasis 148, 180
Iberian 64
leelandie 59, 182, 296, 314, 370,380,
385, see Norse
idea 142, 508
identification 146f., 203L, 249-263
idcogram 285
Illinois 72
Illyrian 64
imitation 6,30, 127, 148, 156f., 36.5f.
403, 472, 476-478, 496-500 '
Immediate eonstitllents 161, 167,
209f., 221f.
immigrant 43, 55f., 461--463, 467
Imperative 331
impcrfeet 224, 273
!mpers.onal 174, 254f., 470, 516
ImplOSIOn 97, 119
inanimate 241,262,272
included 170, 183, 186 219 262
inclusive 255-257 ' ,
incorporation 241
indefinite 203-206 260-262 270
independent 249, 255-266 '
India 42, 55, 98, 102, 154, 289, 458
469,472,495 '
Indic 62L, 296, 312, 319, 374 467-
469 '
indicative 190, 208, 273, 358
individual 22, 3D, 45--47, 75f., 142f.
152, 155, 157, 393, 403, 421, 424'
431, 443, 450 '
indivisibility 180L, 232, 240, 252
Indo-Aryan, Indic
Indo-Chinese 69f.
Indo-European 12-19, 57--65, 306-
321, passim.
Indo-Iranian 62, 30n., 315-318, 351
378f. '
Indonesian 71, 243f., 271, 309L
infinitive 164-166, 197,210,
. 215L, 252, 254, 265, 268f., 273, 470
mfix 218, 222
inflect.ing languages 207f.
infiection 5, Il, 222-232, 237f., 256
263, 270, 294, 387L, 406, 410--412;
453, 470f.
Ingrian 68
Ingwconie 58
initial 99, 131, 134-136, 147, 181-
183, 188, 243-246, 296, 367, 370,
. 418, 447--449, 465, 473
IllSCrJ pilOn 60-66, G8t, 71f., 289-
294,302,305L,433
instrument 173f.
instrumental 315,318
intense 156t, 198, 245
intl'.rdental 98
interjection 121, 156, 176L, 181, 198,
250,265, 102
intermarriage 43, 343, 463, 469f.
International Phonetic Alphabet 87-
92, 96, 101, 103f.
interpretation 64f., 293-296
interrogatve 171, 204, 244, 248, 252,
260, 262, 265, 269, 315f.
intimate 255f., 401
intransitive 150, 241
invasion, .Iee conquest
inverse spelling 294
nverted 98, 102f.
inverted order 174f.
inverted speeeh 156
Iowa 72
Iranian 13, 15, 62f., 70, 312 320
459 470 ' ,
Irish 'ra, 15, 60, 188, 291f., 307, 315,
319, 374f., 383, 418
Iroquoian 72
irregular 177, 188, 203, 207t, 213-
217,223,228, 231L, 238f., 247, 256,
269-275, 279, 309, 318f., 331, 358,
374, 376, 383, 399, 405, 409-411,
416-420, 423, 433, 509
iBoglosa 51, 58, 317f., 321-345, 398,
478--480
iso]ating 207f.
isolation 432--435
Italian 43f., 61, passim.
Italic 61, 308, 312, 319, 350, 380
Iterative 221, 272f.
Jaberg, K, 325
Japanese 10, 21, 44, 70, 101, 116,
256,288
jargon 472--474
Javanese 44, 71, 310, 330
jaw 25, 97, 127
O., 43, 86
Jones, D., 87
Jones, W., 12L
Jud, J., 325
Junius, F., 8
Kabyle 67
Kachin 70
Kaffir 67
Kamehadal 70
Kansa 72
Karadjich, V. S., 483, 511
karmadharaya 235
Kechua 73
kernel 225
Kickapoo 72
King James Bible 281,425
Kirgiz 68
Kloekc, G. G., 325, 329
Koin 62
Korean 44, 70
Koryak 70
Kristensen, M., 325
Kurath, H., 325
Kurdish 62
Kwakiutl 259,470
kymograph 76
la.bial 98 339, 378
labialized 118, 315
labiodental 100
labiovelar 118, 315L
laboratory 75-77, 85, 128, 137, 389,
423
Ladin 61, 3oof., 341, 467, 479f.
Landsmaal 59, 484
language boundary 53f., 56, 314,
3l7f., 464
Lappish 19, 68, 306, 465
laryngal 99,289
laryngoscope 75
larynx 25,27,36,43,941., 108
lateral 97, 10lf" 120, 446
Latin 43, 47, 61L, passim.
Latin alphabet 21, 86-90, 237, 288,
290-292, 296, 300, 302
Latin grammal' 4-8, 237f., 296, 458
law 354
learned 153, 277, 400, 436, 442, 448,
452,472,491--495
Lcrnnian 65
length, sec quantity
Icnis 99f.
Lepsius, C. R, 87
Le ROllX, P., 325
Leskien, A., 18, 353
letter 79, 284, 290-294, 300, 304,
487, 489, SOI
Lettish 13, 60
levels 47-50, 52
lexical form 35, 166-168, 264-269,
277
lexical meaning 169, 174,271,425
Icxicon 21, 39L, 138, 162, 269, 274-
280, 297, 316, 319L, 365, 407f.,
431, 459, 465, 486
liaison, sec sandhi
Libyan 67
Ligurian 64
lirniting 202-206, 250, 252, 258-262
lingua franca 473
linguistie form 138, 141, 145, 158--.
162, 166, 168L, 208f., 265, 283-287,
353f., 389
linguistic meaning 141, 145, 15S, 280
lips 31,43, SO, 86, 97-107, 117f., 123,
373
lisp, sec stammering
list 38, 203, 213, 219, 238, 269, 280
literaey 21
literary 52, 291f.
literature 2If., 286
Lithuanian 13, 15, 60, 117, 125, 307,
309, 315, 319, 373, 422, 427, 483,
509
litotes 426
living analogy 413f., 453
Livonian 68
loan-translation 456-458, 460-462,
468
Joan-ward 449
local difference 112, 114
logogram 285-288,293,296
La-la 70
Lombard 59,466
loosc vowel 103, 107, 109, 112
low vowel 103-107, 109, 120, 367
lower language 461-475
lucus a non luccndo 4
Ludian 6H
Luganda 67
LUlldell, J. A., 87
Lusatian 60
Luther 483
Lycian 65
Lydian 65, 294
macaronic 153
Macedonian 64
Maduran 71 .
Malagasy 71
560 INDEX INDEX 561
malapropism 154
45, 55, 71, 256, 297
70
71
19, 71, 297
male 146, 238,248,251,253,270
69
72, 283f.
69
60
71
44,63
149-151, 254, 427, 430-437
Islands 71
marker 199f., 258, 265, 268-271, 280
Islands 71
67
masculine 192,211,217,253,280,410
Massachusetts 72
mass noun 205, 214, 252, 265
lIllI.9S observation 37r.
mathematics 29, 146f., 249, 507, 512
Matole 72
Maya 72r.,293
lneaning 27, 74-78, 84f., 93, 128,
138-159, 247-251, 264, 407l., 425--
443
mechanical record 76, 85,87, 93, 128,
365
mechanistic 33, 38, 142-144
medial 131f., 134, 136, 18U., 189,
373l., 382, 452
medieval use of Latin 6, 8, 13, 61,
301f., 316, 346, 481, 489--494
Melanesian 71, 257
member 195, 209, 227-237
Mencken, H. L., 515
Menomini 72, 80, 82-84, Ill, 150,
171, 175-177, 219, 244, 256, 260,
262, 279, 359f., 371, 38lf., 385,
395, 400, 446r., 455l., 458
mentalism 17, 32f., 38, 142-144
Meringer, R., 428
66
Mesopotamia 21, 65,284,287
Messapi:J.n 64
metals 320
metaphor 149, 426, 443
metathesis 391
metonymy 426
Miami 72
72
Micronesian 71
mid vowel 103-109, 112
English 365, 368--371, 382,
384l., 387, 404l., 41ll., 419, 423,
426,437
middle voice 258,456
migration 12f., 58, 60, 64, 69, 312i.,
461-475
F. v., 16
Milton 277
minor sentence 17lf., 176i.
minus feature 217f., 231
Missouri 72
Mitanni 65
Mithridates 7, 511
mixed vowe1 104
Moabite 66
IIIock forcign 153
mock learned 154, 421
mode 5, 193, 200, 224, 270, 273
modifiect phoneme 117f.
modifier, su attribute
modulation 163, 166-171, 183-186,
207-210, 220l., 239, 263, 290
Mohawk 72
Mohican 72
Mongol 69
70f.
72
mora 110
Mordvine 68
morpheme 161-168, 209, 244-246,
264,274-278,412,509
morpheme word 209, 218, 240, 412
morphology 183f., 189, 207-246, 308,
349, 352, 371, 380, 383, 391, 406,
449, 454, 465, 506, 509
mots savants 491-495
mouth 97
muffied 102
Mller, F., 19
Munda 70f.
murmur 95, 99, lOH., 1I2
musical 97, 120-126, 375
Muskogean 72
mute 130, 218f.
Naga 70
Nahuatl 72,241
name 57, 64, 131, 155, 157, 201, 205,
288, 294, 413, 420, 429, 451, 465,
467, 470
NarraJ;:anset 72
narratl ve 173 175l., 200f.
narrow vowcl 107
narrowed meaning 151, 426
nasal 96f., lOH., 120, 130, 136, 339,
380
nasalizoo 96l., 102, 106, IIO, Il7,
217,380,384,447
Natick 72
na.tive 43
na.tural syllable 122f., 126
Navajo 72
negatIve 174-177, 197, 204, 248f.,
262, 438f., 486
neo-grammarian 354--364, 392f.
nervous system 26, 33L, 36, 141, 158
192,211,253,375,410
new formation 214, 276, 363l., 368,
38lf., 393, 405-425, 430, 434, 437,
447, 454f., 490f.
Nietzsche, F., 457
noeme 264
nominative 165--167, 185, 190--196,
237f., 267,269,388,392,422
non-distinctive 77-85, 96-105, 110-
129, 141, 144, 147, 365--367, 468,
477, 480, 498l., 516
non-personaI146, 236, 248,253, 260l.,/
263,273
nonsense fonn 153, 157
non-standard 48-52
non-syllabic 120-125, 13lf., 134, 182,
238,243,287,379,384
Norman Conquest 291,463-465,493
Norse 15,303-308, passim.
Northumbrian bUIT 100,390
Norweg:ian 54, 59, 100, 1I0, 116,
390, 468, 483l., see Norse
nose 80, 95f.
noun 166, 190, 192 1940, 198, 202-
206, 21{}-216, 224i"., 228-231, 236f.,
249,251-25'*, 266, 269, 272, 297,
388,392,406,408-412,418,470
Nuba 67
number 5, 192, 204-206, 224, 234,
236,254-257, 27If., 297, 320
number of speakers 43-45,57-73
numeral 29, 147, 152, 206, 237, 249,
279f.,294,320, 422f.,508
numeralsymbol 86, 287
numerative 200,203, 205f., 249, 262,
266
nursery form 157, 394, 424
object 146, 165, 167, 173, 198, 202,
205,216,221,232,236, 25Of., 257f.,
260, 267l., 27lf., 278
199-201,244,249
obJect 0 verb, see goal, of preposi-
tion, see axis
obscene 155, 396, 401
ob5olescence 154, 241, 321, 331-340,
365-368, 376, 393-403, 412, 415,
423, 430-435, 437, 440, 487
Ob-Ugrian 68
ouviative 193f., 257
occasional meaning 431
occupation 50
Oglala 72
O/bwa 72, 283f., 359f., 38lf., 396
o d Englisn 8L, 15, 17,89,303-308,
paslrim.
Olonetllian 68
Omaha 72
ominous forro 155,4OOf.
Oneida. 72
onomatopocia 156f.
Onondags 72
onset of stress 1I3f., 126, 182
open syllable 369, 384
open transition 119
open vowe1 103
Oppert, J., 293
oral 96f.
order 163, 167f., 184f., 197, 201, 207,
210,213,222,227, 229f., 234, 236f.,
247,263,285
origin of language 6, 40--
Oriya 44,63
Orkhon inscriptions 293
Osage 72
Oscan 61
Ossete 62, 70, 470
Osthol, H., 417
Ostyak 68
outcry 6, 147
over-dilerentiation 223f., 269, 399
Paelignian 380
Psiachi 63
Paiute 72
palatal 99 101f., 385
palatalized 117-120, 315, 376---379
palate 86, 95--103, 118
paleography 295
PaU 63
Pallas, P. S., 7
Pamir 62
Panini Il, 19, 63
Panjabi 44,63
Papuan 71
papyrus 295
paradigm 223-226,229--231,237-239,
257,270, 349, 358f., 399,406, 41Q-
412, 422, 506
parataxis 171, 176f., 185f., 254, 259,
263
parent language 12, 14,298-321, 350,
352,360,379,509
parenthesis 186
Parthan 63
participle 197, 225, 230, 233, 237,
252,358,399,415,437,471
particle 171, 173, 176, 199-201, 232,
241,244, 252, 269
parts of speech 5, 17, 20z..190, 196,
198-202,240,249,268-211,274
passive, see goal-action
Passy, P., 87
past 164,174,210,212,214-216,224,
272f., 316,358
Paul, H., 16f.
l
19, 43lf., 435
pause 92,114.,171,181, 185l.
Pehlcvi 62
Penobscot 72
penult 182
562 INDEX
INDEX
563
Peoria 72
perfect 224,273,316,471,491
Permian 68
permitted, su phonetic pattern
Persian 13f., 62,65, 154,288,293
person 5, 224, 297
personal 146, 164, 167,236,248,251,
253, 258, 260f., 263, 265, 270, 273
personal substitute 255-258, 422,
482
pet-name 157
phememe 264
Philippine 7, 42, 71
philology 21, 512
philosophy 6, 17, 172, 201, 270, 456,
508
Phoenician 66, 289
phoneme 79-138, 158, 162, 166f., 179,
264, 289-292, 300, 302-305, 308-
3IO, 350-360, 389, 395, 465, SOIf.
phonemic, see alphabet, distinctiv.;
phonetic alphabet 85-92
phonetic alternant lM, 211
phonetic change 309, 329f., 335, 339,
342, 346--393, 404, 4lOf. , 415,
418-420, 434, 436, 438f., 450f.,
479--481 492, 509
phonetic (orm 138, 145, 148, 159f.,
162, 164, 166, 168, 209, 223, 285,
287
phonctic modification 156, 163-168,
179f., 183f., 207-218,222,226, 228f.
235, 238f., 242-244
phonetic pattern 103, 124f., 128-138,
147f., 153, 181L, 187, 214, 217-
219, 221, 228, 250, 290, 295, 324,
350, 369-371, 376f., 385, 395f.,
449, 457f.
phonetic substitution 81-84, 365,
445--449, 458f., 472
phonetic symbol 286f.
phonetics 74-138,294,328,365
phonie method 500
phonogram 287,293
phonograph 41, 76
phonology 78, 137f., 323
phrase 178-209, 372, 374f., 417--419,
passim.
phrase derivative 178f., 183, 227, 239
phrase word 180, 184,207, 239f.
Phrygian 4, 64
physiology 25, 32, 75f., 78, 127, 130-
133, 137, 296
Pictiah 513
picture writing 65, 73, 283-288, 293
Pidgin EngliBh 472f.
Piman 72 .
pitch 76f., 80, 84, 9If., 94, 109, 1I4-
Il7, 147, 163, 167, 169-172, 174,
182, 185, 188, 221, 243, 299, 385
place 173f.,201,221
place-name 60, 64, 339f., 453, 464,
469
place of stress III
Plato 4
plural 19or., 195, 205f., 209-216, 219,
224,226,236,255-261, 265f., 270f.,
358, 376, 392-394, 399, 401, 404-
406,408--412,453,470,482
Polabian 60
Polish 9f., 42, 44, 54, 61, 86, 96, 102,
1I3, Il9, 126, 177, 182, 187, 2.56,
291,385,470
Polynesian 7l, 374, 473
polysynthctic 207f.
popular etymology 417, 423f., 450
Port Royal 6
Portuguese 13, 44, 61, 96, 341, 472,
474
position 185, 192, 265, 267, 27l, 273,
297
possession 178, 193f., 203, 212, 216,
223f., 226, 230, 236, 256f., 267,
297
postdental 98, 102, 446
Potawatomi 72
Pott, A. F., 15
practical event 23-27
practcal phonetics 78, 8M., 93-127,
129, 137
Prakrit 63
pre-- 309, 31lf.
predicate 5, 173f., 199-201, 206, 244,
252,260,262
predisposition 23-34, 75, 141
prefix 154, 180L, 218, 220, 230, 232,
241, 383, 434
pre-history 12, 16, 319f., 428
preposition Hl4f., 198, 216, 228, 234,
244,252,265,268,271
present 156; 174,212,214,224, 272f.
278,358,364
pre-suffixal 220L, 449, 493
primary dcrivative 209, 227, 240-246,
366
prmary phoneme 85, 9OL, 109, Ill,
114, Il6, 126, 135f., 163, 182, 290f.,
308
primitive 13,299,302, 31lf.
printing21, 41, 286,486, 502f.
Priecian 6
proclitic 187, 259
pronoun 146f., 152, 188, 193f., 244,
249-263, 266, 269f., 375, 382, 399,
401, 422f.,439, 469f.,482
proper noun 194, 205, 265
proportion 276, 40&-420, 44lf.
propriety 155
prothetic 335-338
protrusion 101, 103, 105f.
proverb 152
provincial 49, 52, 62, 296, 340, 478,
482--485
Prussian 13, 60
Psammetichus 4
pseudo-impersonal 2Mf.
psychology 17f., 32-38, 78, 142, 199,
248,297,406,423,435
punetual 272f., 362
Pushto 62
quality 198,202, 205,236,239, 271,
434, 465
quantitv 89, 104, 107, 109r., 129,
177, 217, 221, 290, 294, 296, 302,
366,369,379-381, 384f.
qucstion 91f., Il4f., 147, 169,171,
174-177,186,193,204,250,260
Quilleute 470
quotation 148
race 43, 386
Ra"'U5all 61
Rajasthani 44,63
rank 195, 222, 224, 226
Rask, R. le., 14,317,355,360
Rawlinsoll, H. C., :wa
reading 37,21>2. 285f., 500
rcal, sec indicative
reciprocal 221
reconstructlOu 15, 300-310, 351, 451,
459,516
reduplication 218, 22H., 349, a96
rel1exive 193, 1iJ7
rcgister 94f.
reguhr 189, 211, 213, 216f., 224f.,
238f., 273-276, 399, 405f., 409-
413, 434, 50!)
rebtion-nxis 192, 194, 199, 263, 267,
27l
relatiollshi p 140, 177, 278f., 320
rcbtiouship of 9-la, 57,
Ml, 64, 68f., 71f., 293f., 2g7-318,
316 425
relu.;c substitllte 204, 2(j2f., 423
rdayed speeeh 28, 141
reli fonll 331-310, 479
relig,on 42, 50, 155, :343, 455, 461
rcminiscent. sandhi 189, 219, 374
Renaissance 7f., 10
repctition 156f., 235
resonanee 94-07, 102
re-spelling 62, 295
response 23-M, 7M., 128, 139, 142-
144 147 158.250,2851.,365
194-i96, 207, 221, 223, 274
retraction 103, I05f., 117f.
Rhactian 64
Rhaeto-Romanic 61
Rhenish Fan 343,478
rhythm 395
Rig-Veda 10,63
rime 78, 295f., 330, 395, 482, 486
ri tuai 400
rival, sec variant
Romance 6, 9f., 61, 300-302, 489-
494, passim.
root JO, 240-246, 289, 362f., 426,
433, 459
root forming 245f., 275f.
root word 239f., 243
Rosetta Stone 293
Roumanian 13, 44, 61, 300f., 314,
325,470
rounding 105-107, 117f., 125
runes 290f., 293, 305f., 433
Russian 43L, 47, 61, 457, pa88im.
rustic 152, 331-340
Sakian 63
Salish 470
Samoan 71,181,219,255,257,371
Samoyede 68
samprasarana 384
sandhi 110 B3f., 135f., 163f., 173,
178f., 181, 183, 186-189, 201,204,
219, 222, 228, 275, 371L, 3741.,
378, 3821., 41Sf., 137 .
Sanskrit Il-15, 53, 4H5, passun.
Sanskrit grammar 10-12, 18, 208f.,
235, 237, 2H5, 384
Sarsi 72
satem-language>; 31
Sauk 72
Saxon 303-305, 358, 376, 45lf.
Scandin:tviun M\f., pasm.
Schleicher, A., 15
Schmeller, J. A., 323
Schmidt, J., 317
scholastic 6
school grammar 6, 102, 178, 237f.,
266, 268, 406, 4!lO, 500, 505, 516
8chuchanlt, H., 36'1
Scotch English 152; 300, 329, 370,
394, 485
Scotch Gadic 60
second persll 152, 11)8, 197,224,247,
2,50, 255-258,400f.
secondary deri vuti \'c 20f., 217f.,
'220 224 237-242, 244, 297, M6
secondarv 'phoneme 90--92, W9, Ill,
Il4-116, 122, 134, 136, 156, 163,
109-171, 22f.
secret dialect 50, 471
selection 161-169, 171, 174, 177,
179f., 184f., 190-199, 20H., 207,
229-237, 247,265f.
semantic change 335, 407f., 414,425-
443,456
scmantics 74, 13S, 141, 160, 513
564 INDEX INDEX 565
sememe 162, 166, 168, 174,216,238,
264,276
semi-8baolute 185f., 193
Seminole 72
semi-predicative 206
Semtic 19,65-67, 198, 243f., 2881
semivowel 102, 123I., 130, 132, 134,
136
Seneca 72
seJlS.<Otion 174
sentence 90-92, 114I., 138, 167, 170-
177, 179, 185, 197, 200, 262, 297,
516
sentence-type 152, 169-177, 184, 197,
247, 260, 265, 275I.
sentence-word 172,175
Serbian 9f., 6If., 87, 1I7, 290f., 314,
470,483
seriai, see co-ordi D.lI.tion
se" 46
Shnkspere 22,277,281,398,400,487
shiIt of language 55, 463
Shoahone 72
shwa 519
Siamese 69
sibilant 100, 120, 133, 211f., 214,
315[., 378I., 452I.
Sinilian 64
Sievers, E., 515
signal 80, 128, 136, 139, 144, 157I.,
162, 166, 168,281
significant, see distinctive
Sikwaya 288
Silver Codex 8, 59
simple, see morpheme, ta"eme
SiD.ll.i inscriptions 289
singular 146, 165, 190I., 205f., 208-
213, 219, 223I., 236, 270I., 358,
371,401,405,408-412,470
Sino-Tibetan 69
Siouan 72
situation, see stimulus
slang 49, 133f., 147, 154, 254, 394,
397, 402f., 420, 443
Slavic 9f., 60L, 466, passim.
slip oI the tongue 399, 409, 423
Slovak 61,483
Slovene 61, 314
slurred form 148, 388
socialleveis 47-52, 112, 476I.
society 24-34, 42
Sogdian 63
Solomon Islands 71
Somali 67
SODant 102, 121-124, 384
sonority 100, 120-126, 147,384
Sorbian 60
soundwaves 25-28,31,75-80,87,95,
Ill, 128, 142
Spawsh 42-44,61,467, passim.
specialized meaning 150, 214I., 227-
229,265, 276, 402I., 414, 417,432,
434, 436
species 146I., 202, 204I., 236, 249-
253,258,260,263
speech 22-27, 74, 248
speech community 29, 42-56, 140,
155, 281, 298, 3U, 313L, 317, 319,
394,445
speech-isld.nd 53, 56, 58, 61
speliing pronullcuotion 487I.,494,
498, SOif.
spelling reform 501-503
Spenser 487
Sperber, H., 439I.
spirant 95-97,100-102, llM., passim.
sporadic sound-change 353-364
stage 49
stammering 34, 101, 148
standard language 48-52, 57, 59-63,
68, 296, 321-323, 329, 334, 339,
474, 482-487, 496-500
statement 92, 114, 156, 169, 171
static 200
Steinthal, H., 18
stem 221, 225f., 229-232, 237, 241,
315
i
331,349, 362f., 416I., 470
stimu us 23-34, 74, 114, 128, 139-
144, 151, 156, 158, 166I., 177,285,
365,435, 440
stop 80, 86, 97-102, 214, passim.
Streiff, C., 331, 333
stress 90-92, 11o-Il4, 120-126, 130,
154, 163, 268, 174, 180, 182I.,
186f., 220I., 228, 233, 259, 303,
375I., 382I., 385, 447, 450
stridulation 27
structural ordcr 210, 213,222,227,247
structure 135,264,268
stuttering 34
style 45, 153, 499
Subiya 67
subject 5, 173f., 199--201, 252
subjunctive 152, 190, 224, 273, 358,
437-439
subordinate 192-195, 197I., 204,235,
237, 25H., 269, 407
sub-standard 50-52
substantive 146, 100., 177, 185, 196,
198, 249, 267-271
substitute 146f., 169, 184,247-263,
509
substitution leature 112, 216-218,
222,228,243,274
substratum 386, 468--470, 481
sub-vocal 143
Suetonius 302
suffi" 154, 218-221, 23().-232, 240f.,
244I., 314f., 318, 366, 410-417,
454I.
Sumerial1 65, 288, 293
Sundanese 71
superlative 417
suppletion 215f., 218, 223, 238I., 270
Swaheli 67
Swedish 9f., 54, 59, 87, looI., 106,
IlO, Il6, 151, 193,221,256, 299I.,
370, 385, 389f., 428, 447, 459, 503
Sweet
1
H., 86f.
syllablc 12().-125, 130-137, 181,384
syllabic stress 122f., 136
syllabic writing 287I.
syllabie 120-126, 243I., 287-290, 349-
351
symbol 283-290
symholic 6, 156, 243-246, 390, 424
syncope 382
syncretism 388, 392
synecdoche 426
synonym 145, 442
syntactic compound 233-235
syntax 5, 11, 183-206, 212, 216, 224,
232-235, 247-264, 268, 270-273,
407, 417-420, 423, 453, 467f., 486f.
synthetic compound 231-234, 236,
430, 461l
synthctic languages 207
syrinx 27
tabu 155, 396, 4D0--402, 507f.
tactic Iorm 166
Tagalog 71, IIl5, 171, 173f., 176, 2ooI.
218, 22lf. , 243f., 252, 255, 260,
269, 278, 3I1l, 371, 391, 446--448,
455
taKmcme 166-168, 264, 276I., 505
T!l.hiti 71
Tai 69
Tamil 44,70
Tartm 68
tatpurusha 235
tatsama 495
taxcme 166-171, 174, 184I., 190-192,
197-109, 2I1l, 220, 264-266
Tebele 67
technical 49f., 152f., 277
teeth 98, 100, Il8f.
tclephone 41, 45
Telugu 44, 70
tense 5, 200, 224, 270, 272, 297
tcnse vowel 103, 107, 109, 124, 136,
445
Tesnire, L., 44f.
Teton 72
textual criticism 5, 295
theoretica! form 218-220, 223, 237,
242, 516
thinking 28f., 142f., 508
third person 152, 188, 193, 198, 212,
214L, 224, 253f., 256-258, 418I.
Thomsen, V., 293
Thracian 64
Tibetan 69
Tigre 66
Tocharian 64,316
tone of voce 39, 114I., 144, 147, 498
tones 116, 475
tongue 25,31,36,75,94-97,99,101-
105, 108, 112L, 117-119, 123, 127,
365, 373, 376, 383I., 390, 470
tongue-flip 81, 100, 187, 374
transcription 85-92, 96, 98-104, 109,
112-114, 117, 120-123, 128, 135,
168, 296, 366, 501
transferred meaning 39, 149f., 198,
402f., 425-443, 456, 458
transient 173f., 2oof.
transition 118-120
transitive 150, 165
translation 140
transliteration 90, 101
transmission 294f.
trial 255, 257
trill 98, lOO-102, 104, 120, 127,383f.,
390, 445, 470
triphthong 124, 131, 135,137
Tsimshian 470
Tuareg 67
Tune;use 69
Tupi-Guarani 73
Turco-Tartar 44, 68f., 381
Turkish 21, 68!., 107, 154, 181,208,
293, 467
Tusearora 72
Ukrainian 44
ultimate constituent 161, 182, 195,
242
Umbrian 61
umlaut 381, 434
unbounded 205
undergocr, sec goal
undcrlying Iorm 209--226, passim.
understandlng 31, 55, 80-82, 84, 93,
127, 149, 179, 250, 277, 281, 295,
386, 457I., 487
unique IOOL, 210, 213I., 234I., 275,
415 426
unreai 224,273
unrounded 107
upper l a n ~ u a g e 461-475
Ural-AltalC 69
rdingcn Line 343
Ute 72
uvula 95-97, 99-101, 127, 390, 445,
470
Uzbeg 68
Vai 288
Van 65,293
566 INDEX
Vanda! 59
van Helmont 424
variant 81, 83, 98-103, 105, 1I0--
1I4, Il7 ., JXl8sim.
Vater, J. S., 8
Verne 63
velar 98f., !OH., 127, 315f., 339, 376--
379,385
velarized 1I8f.
Veliote 61
velum 95f., 98, 103, 1I7, 119, 373,
383f.
Venetic 64
Vepsian 68
verb 20, 165-167, 172-175, 190--194,
197f., 210, 212, 214-216, 223-
225, 229-233, 238f., 251, 254, 256,
258, 260, 297, 358f., 362-364, 383,
395,414-417,439,471,506
vernacular 482
Verner, K., 308,357-359, 374,415
verse 78, 295f., 302
Visible Speech 86f.
vocabulary, see lexicon
voeal ehords 25, 27, 31, 75, 94f., 99,
102,111,373,375,505
vocative 177, 225
Vogule 68
voee 27, 94-97, 101f., 112, Il4,
11 7f., 120, 221, 258, 364
voiee of verb 173, 201, 224
voicing 94-97, 99-102, 118, 120
135, 137, 189, 218f., 357f., 372-:
376,389,458
Voltaire 6
Votian 68
Votyak 68
vowcl 8If., 102-126, 134f., 216, 243,
288-290, 292, 295, 300-302, 306f.,
329,356-358,376-387
vowel harmony 181,381
vowel-shift 387
vulgar 147, 152, 156, 302
Vulgar Latin 302
war 156
wave-thOry 317f., 340
Wcigand, G.,32.5
Welsh 13,55,60,97,307,464
Wendish 60
Wenker, G., 322
West Germanie 59,304,311-314,
389,425,428,451
Western Hindi 44, 63
whispcr 95, 102
Whitney, W. D., 16
wide voweI 107
wdencd meaning 151,426,432
Winnebago 72
Winteler, J., 331
Wolof 67
word 90, 99, 102, 110-114, 116, 138,
17H., 176, 178-189, 195f., 200,
207-247, 254, 265, 268, 277f.,
284-287, 291, 297, 303, 309, 328,
371, 374f., 38lf., 395f., 414f., 417-
420, 447, 509
word-class IVO, 196, 202
ward-formation 222f., 226, 231, 237-
240,412-416,453,505
ward order 156, 171-175, 197-201,
229, 234, 254, 260, 263, 286, 437,
470
Wordsworth 443
Wrcde, F., 322, 325
J., 323
writing 3, 8;13, 2Jf., 3-7, 40, 66, 73,
79, 85f., 144, 152f., 178,
282-296, 448f., 486-495, 500-503,
506
written records -10; 13, Wc; 38,
57-73,1-52, 277, 28]-296, 298-
305, 309-311, 319, 330, 346, 359,
380L, 393, 400, 404f., 416, 425,
438, 44Df., 455, 459, 464, 48If.,
484
\Vundt, W., 18, 386,435
Wyandot 72
x-ray 75
Yakut. 19,69
Yana 46
Yap 71
Yenisei-Ostyak 70
Yoruba 67
zero-featurc 209, 215-219, 223, 231,
236, 238f., 252, 256, 263, 416, 420
Zeuss, J. K., 16
Zulu 67
Zyrian 68
,
,
\
1