Curt Sachs, Jaap Kunst History of Music

THE WELLSPRINGS OF MUSIC
Photo by Joseph Breitenbach
CURT SACHS
CURT SACHS
THE WELLSPRINGS
OF MUSIC
Edited by
JAAP KUNST
THE HAGUE
MARTINUS NIJHOFF
I9 62
ISBN 978-94-015-0427-0
DOI 10.1007/978-94-015-1059-2
ISBN 978-94-015-1059-2 (eBook)
Copyright I96I by Martinus Nijhott. The Hague. Netherlands

All rights reserved. including the right to translate or to
reproduce this book or parts thereof in any form
Softcover reprint of the hardcover 1st edition 1962
To
IRENE
alter lilty years
01 t.narried lile
and collaboration
EDITOR'S PREF ACE
When Irene Sachs asked me to take charge of the practically

completed manuscript left by her late husband, and to prepare
it for the press, I feU moved and honored to have been entrusted
with the task to bring Curt's last work safely into the harbour.
In favoring me with the editing of the book, Irene may have
thought of the long discussions we had together, Cutt Sachs and
I, during two successive years. It was in his study when, with
the little gramophone on the piano, the manuscript sheets piled
up before us, we listened intently to the records, merging our
opinions or deviating, and thus consolidating a bond of common
interest and warm~ personal friendship.
Be it as it may and although at the moment I am still recuperating from a serious operation, I have tried to carry out this
task in the shortest possible time and to the best of my knowledge
and ability.
I have avoided making important changes in the text, though
I have been tempted occasionally to put a question mark in the
margin, when our views differed somewhat in degree or emphasis.
But in Sachs' reasoning one thing follows from the other in such
a way as to make it unfeasable to attempt revisions without
compromising the logic of the whole. Only once - in the field of
Indonesian music, where I feel particularly at home - have I
omitted some facts I knew to be interpreted less correctly.
Futther, insofar as possible I have supplemented the bibliographical data in the footnotes.
The sum and substance of the book is a truly monumental
discussion of the development of the melodie element in all its
aspects (including rhythm). This development has been traced
and analysedstep by step, silhouetted against a general cultural
historical backdrop, in a calm, clear manner, seasoned with
VIII
EDITOR'S PREFACE
typical mild "Sachsonian" humor and illustrated with a wealth

of examples and evidence. I doubt whether any other of us
ethnomusicologists would have been capable of writing such a
comprehensive study: S~chs had at his fingertips a phenomenal
amount of factual information. Sometimes, under his wise
guidance, complicated propositions become surprisingly simple.
The reader will be as fascinated as I was by the originality of
some of his conelusions, captivated by his penetrating comparisons, and charmed by his rich and expressive language.
After having lived with this book for aperiod of some months,
I realize a1l themore what the death of Curt Sachs means to the
musicological world.
"The Wellsprings of Music" forms a worthy elose to that long
series of publications his boundless energy and unsurpassed
knowledge have given uso
I feel greatly indebted to Dr. Eric Werner, New York, for his
decisive encouragement towards the publication of this work
and for the active interest he took in it. I want also to express
my gratitude to the musicologist Miss Marijke Charbon, The
Hague, for having made an Index of N ames and for some useful
suggestions.
Amsterdam, 15th February 1960
JAAP KUNST
My dear friend Jaap Kunst did not live to see this book in
print. I wish to express my deep gratitude to him, whose dedicated help and unfailing and devoted interest made its publication possible.
After Jaap Kunst's death, his assistants, Messrs. Ernst Heins
and Felix van Lamsweerde, took over with selfless loyalty. They
read the proofs and saw the manuscript through its final stages.
They were aided by Miss M. van Walraven. The Royal Tropical
Institute, Amsterdam, kindly granted time for their work.
IRENE SACHS
The letter group EFw in the footnotes stands for "Ethnic Folkways Library,"
a vast collection of primitive, oriental and folk music from every part of the world
on L.P. recordings, published in New York under Moses Asch as producer and Harold
Courlander as editor-in-chief.
NOTE -
TABLE OF CONTENTS
Editor' s Prejace
Chapter One /
THOUGHTS AND METHODS
1. Preliminaries
II. The advent of the ethnomusicologist
IV
1
1
Theology and humanism as fundaments of earlier education - early music histories : Martini, Hawkins, Burney,
Laborde, Forkel, Krause, Fetis, HuIlah, Ambros - early
non-historieal sources : Rousseau and the romanties, Villotteau, J ones, doctoral theses - beginnings of modern
method with EIlis, Edison and Fewkes-Gilman-Stumpf phonographie archives - change of name from comparative musieology to ethnomusieology.
IH. The ethnomusicologist's workshop
16
Field work - phonograph, gramophone, tape - tone

measurements - acoustical deviees: forks, Tonmesser, tonometer, monochord - visual devices: oscilloscope, photophonography, stroboscope, electronie counter - editingtranscribing - logarithms and cents - savarts - Reiner's
music rule - notation.
IV. The question of origin
33
Theories - tone languages - mouth musie - drum languages

etc. - absence of singing - prehistory - radio carbon method
of dating - fluorine method - stone and metal ages anthropology - culture changes - the "savages" - primitive and non-literate man - scriptless society and folk
music - monogenism and polygenism - Kulturkreislehre (tracest'.
Chapter Two /
EARLY MUSIe
V. The oldest music: tumbling strains

Major and minor - tumbling strains or stair melodies high and low - domestieation - triadie and tetradie octaves.
49
49
TABLE OF CONTENTS
VI. The oldest musie: one-step melodies

Terminology - sharping the pitch - affixes and infixes tetrachords - bare fifths and sixths - pathogenie and
logogenie - meaningless syllables - burdens - glossoWia repetitive style - flexa - ontological remarks - imitation
and dramatization - a.ni1naJ. dirges or epikkedia.
59
VII. Conservatism and magie

Palaeolithie survivals - the fight of lIlusie and poetry- striet
tradition - magie - charms against siekness, death and
dearth - mana - unnatural singing - kazoos
VIII. Voeal mannerisms
8S
Von Hornbostel's and Merriam's descriptions - European,

Asiatie, Australian, Polynesian, Melanesian, African
styles - quilisma, pressus, pUca - tessitura - proposed
terminology - tempo and intensity - sense and value aesthetical aspects
IX. Instruments
Literature - extensions - other rudiments - male and female
- eosmology - connQtations of the flute - of stinged instruments and jaw's harp - conflicting connotations - bull
roarer - gourd rattle - on the way to melody - finger holes
- foot and inches - Blasquintentheorie - tuning of strlngs isotony - tuning of xylophones - East African and Indonesian paralleis - ehimes - the lithophone of the Musee de
I' H omme - panoramic view - Europe and Africa - early
instrumental music - zanza
X. Rhythm and form
9I
III
General notions - rhythmie patterns - quintary rhythm downbeat and upbeat - metrie alteration - meter and
time - additive meters - "fascinating rhythm" - art and
folk music - eycles - strictness and freedom - overall
patterns - isorhythm - panoramic view - form - repetition
- stimulus and narcotic medieval paralleis : lai, estampie,
lied, rondeau - antiphony and response - variation - zanza
cyclie forms
Chapter Three /
ON THE WAY
Growth, personality, art

Culture ehanges - musical criticism - musical contests individualism and art - copyright
I3 2
I3 2
XI
T ABLE OF CONTENTS
Chapter Four I THE
FATE OF SECONDAL AND TERTIAL
PATTERNS
143
The widening nucleus - "triadic and fanfare melodies" tonal anatomy and physiology - chains - double, tripie,
quadruple, quintuple and sextuple thirds - skeletons and
infixes - pentatonism and heptatonism - German chorale
dialect - Landino third - four-line notation - six-five
patterns - triadic octaves
Chapter Five /
THE FATE OF QUARTAL AND QUINTAL
159
PATTERNS
Disjunction and conjunction - quartal chains - chains of

fifths - hybrids - modulation - culture chance once more
Chapter Six !CENTRIC
168
MELODIES
Gregorian tenor melodies - the concept of
mes~
Chapter Seven / POLYPHONY
175
Vertical and horizontal - distribution - isochronous c0nductus - paralleis - drones - grounds - alternation and
canon - Stimmtausch- heterophony
Chapter Eight /
CROSS- OR POLYRHYTHM
Independent accompaniment - hemiola - "syncopation" Afro-Indian concordances
Chapter Nine /
PROFESSIONAL MUSIC AND MUSICAL SYSTEMS
200
Semi-professionals and professionals - bards - minstrelsy segregation and discrimination - professional wallers musical vocabularies - pitch syllables - concomitant
method of learning
Chapter Ten /
'PROGRESS'?
210
The fading theory of evolution - culture graft - progress evidences against progress - our gain is our loss
A note on Bibliography
223
Index 01 N ames
224
ONE
I. PRELIMIN ARIES
Western music, the pride of our culture, is no longer what

it should be and was: the highlight of our day, edifying and
blissful. Our modern lives are ad nauseam saturated with music
and wouldbe music. I do not speak of the dizzying quantity of
concerts and recitals - we attend them or stay away as we please.
But we cannot have our coffee break without the blaring interference of a non-stop loudspeaker on the wall or a jukebox in
the corner; the savings-bank pours music over our head while
we pass acheck across the counter; railroad cars and buses feed us
catchy or sentimental tunes instead of improving the service; and
the neighbors force us to share in their radio and television orgies.
Men of high civilization have become voracious hearers but
do hardly listen. Using organized sound as a kind of opiate, we
have forgotten to ask for sense and value in what we hear.
In primitive music, on the contrary, sense and value are
paramount qualities. Not only is singing indispensable for
special events, like wedding and childbirth, puberty rites and
death, and whenever luck must be forced on adverse powers in
hunting, harvest, and sickness. It also acts when regular work,
as rowing aboat, or rocking a child, or grinding edible roots,
demands and gives a rhythmical impulse. In this interweaving
with motions and emotions, music is not a reflex, remote and
pale, but an integral part of life. As Frer-Haimendorf puts it
exquisitely in words, this music "resounds in the darkness,
gripping the singers and blending them one and all, till they
finally merge in the unity of the dance. This rhythm is more
than art, it is the voice of man's primeval instinct, the revelation
of the all-embracing rhythm of growth and decay, of love,
baUle and death." 1
1
Christoph von Frer-Haimendorf, The naked Nagas, London, 1939. p.
208.
These words could not have been written half a century earlier.
Older travelogues - up to I900 - present us with queer and often
foolish descriptions of primitive music and often ridicule what
to the natives is their most sacred expression. One of the oldest
comprehensive books in our field, Wailaschek's Primitive Music,2
teems with quotations of this kind.
Matters have changed, luckily. In modern works by educated
anthropologists one may read about the extraordinary charm
of haunting tunes or of "the delightful transitions from one
musical phrase to the other which retain an element of surprise
however often one may hear them." 3 R. F. Fortune teils us
that "most truly Dobuan mourning is as fine as finest bird
song," 4 and Marius Barbeau says of the mountain songs of the
Tsimshian Indians in the Northwest that they are "swaying
and ethereal. The voices of the singers, especiaily the women,
are beautifully lofty and lyrical in character. Like the songs of
Mongolia and Siberia, they are imbued with color; color and
expressiveness of the voice are an essential feature of the native
singing on both sides of the Pacific Ocean." 5
Many a modern composer has been fascinated by the charm
of some primitive or oriental melody or a special feature like
the Javanese isotonic scale. But when they borrow a native
theme - of course in western dressing - such contact belongs
in the history of western music and does not justify our
endeavors to find, understand, and interpret the musical idioms
of non-western peoples. Such mistake would amount to seeing
in the awkward Romanesque-Gothic stew of Harvard's Memorial
Hall a justification for studying the history of medieval architecture.
Rather should our search bring horne to everybody the
inimitable uniqueness of all art and discourage artificial exotism
and pseudo-archaism, unless the spirit of modern art makes
legitimate an opening of the gates. Debussy, Bart6k, and Stravinsky are cases in point.
2 Richard Wallaschek, Primitive music, London, I893; German translation as
Anfnge der Tonkunst, Leipzig, I903.
3 Von Frer-Haimendorf, The Rai Gonds of Adilabad, London, 1948, vol. I,
pp. 406 ff.

4 R. F. Fortune, Sorcerers of Dobu, London, 1932, p. 255.
5 Viola E. Garfield, Paul S. Wingert and Marius Barbeau, The Tsimshian: their
arts and music, New York, 1952, p. 98.
Such music cannot be bought in stores, but comes from

faithful tradition or from personal contributions of tribesmen.
It is never soulless or thoughtless, never passive, but always
vital, organic, and functional; indeed, it is always dignified.
This is more than we can say of music in the West.
As an indispensable and precious part of culture, it commands
respect. And respect implies the duty to help in preserving it.
In pursuing our goal, we are particularly rushed, as the
venerable heritage of archaic cultures is threatened with imminent extermination. The melodies of primitive man, an organic,
essential part of his spirituallife, fall victim to Christian missionaries, Soviet agents, European colonizers, and American oil
drillers. And primitive as well as folk melodies are hopelessly
succumbing to a technical age with military service and factory
work, with rapid buses, planes, and cars, with phonographs,
radios, and television sets. "One of the most tragic things about
contact of the aboriginal with civilization is the destruction of
art and culture that so frequently follows." 6
A moving instance of such extermination is the fate of music
in the steppes of Central Asia. Not long ago, the autobiography
of the great Kazakhian rhapsodist Dzhambul called it "a mighty
power ... irresistible as the tempest." 7 Today, we read in
Haslund-Christensen's report "On the Trail of Ancient Mongoi
Tunes";8
"When in 1923 I arrived in Central Asia for the first time I
f<>und the whole of this isolated world, 'with its sparse nomad
population, like a living reminder of a remote past ... On my
return to Central Asia in 1936 I found that the few years during
which I had been absent in the civilized world had brought
about greater and more radical changes than the whole of the
previous five hundred years put together ... The occupying
powers had begun to enrol Mongoi youth in their respective
schools and armies, and the whole population was now being
exposed to the effective batteries of modern propagandistic
6 Verrier Elwin & Shamrao Hivale, Folksongs 01 the Maikal Hills, London etc.,
1944, p. xv.
? E. Emsheimer, Singing contests in Central Asia, in Journal 01 the International
Folk Music Council, vol. 8 (1956), p. 28.
8
Reports Irom the scienti/ic expedition to the Norlhwestern provinces 01 China,
VIII, Ethnography 4, Stockholm, 1943, pp.
20 ff.
technique
Troubadours who had themselves spent their
youth at the feet of the old masters to leam the chants and
folk-songs of the past were now without a single disciple to
whom they could pass on their precious inheritance ... Singers
and muslcians ... were incarcerated in the prison of the town
for having been too deeply rooted in the past .... "
Sad, too, is what we hear from Assam, on the Gulf of Bengal: 9
"A bell rang, feebly and 'tinnily,' and the sound of singing
rose in our ears. Were these N aga voices? I t sounded like a
hymn or a chapel-chant - or perhaps not quite like either. This
singing was entirely different from Naga music, the melodies
were not in harmony with N aga expression - as discordant as
the ugly tin-roofed chapel amidst the palm thatches of the
village houses. People with sullen faces came out of the chapel;
they seemed to me mere shadows of Nagas, or, even worse,
caricatures of Europeans .... "
In I949, a former army doctor in the service of China described
to me how "recruits from inner China ... where there are neither
radios nor movies" were .forced to learn "the, modern - and
fully western - national anthem of China. Some officer sang it
once to them, without accompaniment, and after that the men,
somewhere between one and three hundred, were forced to sing
it every morning on hoisting the flag ... " 10
Thus indigenous music is killed in the interest of poor and
second-hand western trash.
The 10ss is greater than most of us might realize. It means
irresponsible annihilation of an organically grown and vital
part of culture and a substitution of 'popular,' rootless, inane,
and often commercial hits. It means the increasing vulgarization
of our globe.
11 Von Frer-Haimendorf, l.c., p. 55.

10 W. Lurje, in a letter to, and translated and communicated by, Curt Sachs, in
Journal 01 the American Musicological Society, vol. 3 (1950), pp. 292 ff.
II. THE ADVENT OF THE ETHNOMUSICOLOGIST*
The two pillars of European education during the Middle

Ages and the Renaissance were theology and the classics. Constantly dealing with the Holy Scriptures and the books of
Antiquity, the musical writers could not eschew at least a
literary contact with the musical wodds of the Bible and of
the Greeks. Much as they were engrossed in the musical problems
of their days, in tuning, intervals, and counterpoint rules, they
kept a loose connection with the half primitive half oriental
sphere of either one. Even early treatises on music opened with
a reverent if noncommittal bow to the Hebrews and Hellenes,
to Miriam and David, to Pythagoras and the Roman Boethius.
True, the music of the Bible was irretrievably lost, and the dozen
relics of Greek composition that we own today had not yet been
rediscovered. Although the scholars had not the slightest conception of the music itself, they had been trained in the intricate
art of interpreting, however questionably, the many passages
on music in the Holy Scriptures and in the writings of ancient
Greece.
So powerful was this education even in later times that the
earliest of the great musical histories, Padre Martini's threevolume Storia della musica (1757-1781) covered nothing but
Antiquity. But far from being exclusively Graeco-Roman, it
included chapters on the Hebrews from "the creation of Adam,"
indeed, on the Egyptians, the Chaldeans, "and other oriental
peoples."
Martini's British successors were less generous. Chades
Burney's General history 01 music (1776) did not bother with
"Chaldeans and other oriental peoples"; but at least it opened
its discussions with a good-sized chapter on Egyptian music.
Sir John Hawkins, in his General history 01 the science and
practice 01 music of the same year, went not even that far. In
true cavalier fashion he did away with all the "barbarians."
* Much of this seetion has been printed in the author's contribution to 50me
aspects 0/ musicology, New York, The Liberal Arts Press, 1957. Reprinted by permission of the American Council of Learned Societies and of the publishers. Cf. also:
the author's The Orient and western music, in The Asian Legacy and A me1'ican Li/e,
New York, John Day, 1942, pp. 56-69.
Their best music - as he wrote in the Preface - His said to be

hideous and astonishing sounds. Of what importance then can
it be to enquire into a practice that has not its foundation in
science or system, or to know what are the sounds that most
delight a Hottentot, a wild American, or even a more refined
Chinese?"
This is, no doubt, achallenging statement. After all, the "more
refined Chinese" based his music very definitelyon science,
and all the ancient world from Egypt eastward to the coast of
the Pacific had its musical Hfoundation in science or system."
This, of course, Sir John could hardly have known. But as a
student of Hellenic sources he should at least have realized that
the chroai or shades in tuning Grecian scales with their manifold
and often very unfamiliar, irrational steps from note to note
were thoroughly oriental and hence no less inferior and "hideous"
than all the music excluded from his gigantic work. Worse: he
scorned oriental music not because it was hideous, but because
it was said to be hideous. He did not even take pains to know
that music himself and feIt no urge to test the hearsay. As to
the Greek and Hebrew chapters, on the contrary, there was no
danger of unfavorable reports: conceivably, no earwitness was
at hand. The two sections of the book were easily filled with
learned quotations from literary sources well known to all the
erudite contemporaries. The music itself was absent, to be sure;
but being Biblical or Greek, it must have been perfect by definition.
Four years after Burney and Hawkins, Laborde stressed the
universalistic viewpoint of the French. His Essai sur la Musique
of 1780 discusses again, as Martini's Storia had done, "Chaldean
and other oriental peoples," Hebrews, Egyptians, and Greeks.
But he also included Chinese and J apanese, Siamese, Arabs,
Persians and Turks, Negroes, and many others.
The Germans kept out of the picture. Johann Nicolaus Forkel's Allgemeine Geschichte der Musik of 1788, in two volumes,
limited Antiquity once more to Egypt, the Bible, Greece, and
Rome; and after him, the recent theory of evolution gave an
even sharper edge to the concept of history. The Orient and
the Primitives were mentioned, although in a negative light,
since any claim to our own perfection required the production
THOUGHTS AND METH ODS
of humbler predecessors. In K. C. F. Krause's certainly not topranking Darstellungen aus der Geschichte der Musik (1827), we
read with astonishment: "In Antiquity, which was the childhood
of music (!), only simple, unadomed melody was known, as is
the case today with such peoples as the Hindus, Chinese, Persians,
and Arabs, who have not yet progressed beyond the childhood
age (I)." This is a truly Hegelian progressivism: how far have
we come in our mature age (or is it senility, if not worse?) 1 Not
to mention the profound ignorance behind the notions of Hindus,
Persians, and Arabs singing in "simple, unadomed melody" they who are unrivalled masters in the art of highly adomed
singing and leave simplicity to the lower forms of children's
songs - and to the West.
It means modulation to a different key to open the H istoire
generale de la musique of 1869 by the Franco-Belgian Fran'toisJoseph Fetis. He offers sections, not only on India, China, and
Japan, but also on Kalmucks, Kirghizes, Kamtchadals, and
others. For to Fetis' imaginative insight, l'histoire de la musique
embrasse celle du genre humain - the history of music is the
history of mankind.
These words we should write upon the posts of our house.
John F. Hullah's History 01 modern music, written only two
years after Fetis, was not exact1y a landmark and could be
passed over in tactful silence, were it not for opinions that still
haunt quite a number of musicological minds. "Much as the
Orientals have or have had" a "music of their own," this music
"as at present practised ... has no charm, nor indeed meaning,
for us ... The European system, though the exigencies of
practice prevent its being absolutely true, is nearer the truth
than any other." Oriental music, then, has no charm, no meaning,
and is more remote from truth - thls is indeed an unmistakable
answer to the appropriative question : 'What does that music
mean to me? Can I enjoy it?'
The opposite urge, to widen the field, was not always successful
either. Theological and c1assic ideals were too strong and blinded
the greatest authors. The reader who reaches for the monumental
Geschichte der Musik by August Wilhelm Ambros (1861) finds a
whole Buch on the Kulturvlker des Orients, indeed on the
Primitives. But on these pages he also finds the most bewUdering
pronunciamentos such as: "Assyrian music seems (!) never to

have risen above the level of a mere sensual stimulus"; or: the
music of Babyion "was quite certainly (!) voluptuous, noisy,
and far from simple beauty and noble form"; and Phoenician
music was mainly meant to drown "the cries of the victims who
burned in the glowing arms of Moloch." What horrible brutes!
Even the hope that they may have mended their ways before
old Ambros was reprinted in 1887 is futile: the third, "gnzlich
umgearbeitete" edition has left these monstrous lines intact. l l
If the re-editor, engrossed in Westphal's then recent theories on
music in Greece, knew nothing of the fabulous excavations between the Two Rivers, nothing of the impressive art of Sumer,
Babylonia, Assyria, should he not have known from the books
of Kings and Chronicles that the Phoenician King Hiram helped
his friend Solomon with skilled artists and workmen to build
the Temple in Jerusalem? And did not this very Temple import
a number of musical instruments from Phoenicia, among which
the soft-sounding harps and zithers (nvalim and asorot) were
quite unable to drown the desperate screams of Phoenicia's
burning victims?
THE GENERAL FAlL URE of universal music histories is the
more surprising as an interest for non-western music had unmistakably appeared even before Sir John's and Burney's time.
Travel diaries of the eighteenth century had begun to add a
specimen or two of native songs in modern notation. These,
it is true, were rather doubtful sources, since untrained, unsuited
men had squeezed what they heard into familiar scales and
rhythms of Europe - by hook or by crook. Did not in the same
volumes the illustrations, too, falsify and 'idealize' the characteristic types of foreign races in the likeness of ancient statues
and render their garb in the drapery of ancient gods 12 that
art schools were teaching?
11 Prof. Sachs could have added one more appalling quotation from Ambros,
occurring, however, in the first edition only, as being apparently no longer palatable
when the later editions were prepared: "Auf Java geht die Tonkunst in barbarische
Trbung ber, so dass sich hier ihre letzte Spur verliert" - Le. "In Java music is
degenerating into such a barbaric mess that there its last vestiges are being lost."
(J.K.).
12 Cf. also Richard Thurnwald, Die menschliche Gesellschaft, vol. 2 of Reprsentative
Lebensbilder von Naturvlkern, Berlin, 1931, pp. 6 ff.
After these early attempts, it was an epochal moment when

Rousseau printed quite a few exotic melodies
in his Dictionaire de Musique (1768) - from Finland, from les
sauvages du Canada, and from China. The Chinese example later
found a place of honor in providing local color for Weber's
Turandot overture.
Oriental musicology was born eleven years after Rousseau' s
work, when in 1779 F ather J osephAmiot' s painstaking Memoire sur
la musique des Chinois, tant anciens que modernes was posthumously
printed. We still consult this work with respect and profit.
It might be permissible to link the advent of ethnomusicological interest with Romanticism, which in all the arts consisted
"in a high development of poetic sensibility towards the remote,
as such ... In the middle of the eighteenth century, that
romantic attitude (heralded by J ean-J acques Rousseau) can
already be recognized ... in a tendency to take interest in
remote kinds of art." Indeed, Romanticism "allows the poetic
interest of distant civilization to supplant the aesthetic interest
of form ... Under the influence of this poetry, Nature's unconsidered variety became the very type and criterion of beauty,
and men were led by an inevitable consequence to value what is
various, irregular, or wild, and to value it wherever it might be
found." 13
Less than twenty years after Amiot, in 1798, Napoleon
invaded Egypt and caused an impressive number of French
scholars in various fields of knowledge to study the country in
all its aspects, past and present. After years of additional
research in the Bibliotheque Nationale, the various works were
gathered together in a gigantic Description de l'Egypte, with
almost three of the twenty-five volumes being devoted to music:
(I) Memoire sur la musique de l'antique Egypte, (2) De l'etat
actuel de l' art musical en Egypte, and (3) Description historique,
technique et lifteraire, des instruments de musique des orientaux.
The author of these three musical essays was Guillaume-Andre
Villoteau.
One cannot easily exaggerate their merits.
Based on keen observation during three and a half years of
intensive field work, they are uncompromisingly correct and
J ean-Jacques
13
Geoffrey Scott, The architecture oj humanism, Garden City, 1954, pp. 41,43,60.
10
free of the preconceived ideas of a European. Villoteau, it must

be said, had his prejudice when he first set foot on Egypt's shore.
But, while studying with a native music teacher, he soon realized
that correct intonation was not a monopoly of western man.
Indeed, the European system was by no means, as I just quoted
from Hullah, "nearer the truth than any other." After all,
western man had changed his ideas about tuning in a long
evolution and was in Villoteau's time replacing the artificial
mean tone temperament by the no less 'untrue' equal temperament. In Egypt, Villoteau came to understand that oriental
music, though basically different from ours, was neither less
near to truth nor inferior; it had its own scientific foundation
and must therefore be judged according to laws of its own.
Meanwhile, an English High Court judge in Calcutta, William
lones, had in I784 written On the musical modes 01 the Hindoos,
of which a German translation by F. H. v. Dalberg came out in
I802 under the more general title Ueber die Musik der Indier.
Dalberg added to the original text and appended a number of
music examples beyond the one melody that lones had given.
But when he quotes his main collector, an English musician in
India, as saying that he had had no small trouble in setting
the native tunes in a regular (I) "tempo" (he means rhythm),
we cannot accept them without reserve and distrust. How the
'regularization' was done can indeed be seen from the neat
notation of a children's prayer heard in Malabar in the Southwest
of the subcontinent. The same piece, or at least a elose variant,
was later recorded on an Edison cylinder, found its way to the
Phonographie Archives of Berlin, and was published in Igo4. 14
The recorded original is sung, as many children's songs of India,
in the typical, charming triputa rhythm of seven beats or 4 3
in the measure.
Ex.
I Cf U r EJ I Cl U r I EI
J=IS8
aFCI ~Et U F:1

I
Bqt this was too strong a medicine for the good musician from
England.
14 Cf. Otto Abraham und Erich M. von Hornbostel, Phonographierle indische
Melodien, in Sammelbnde der Internationalen Musikgesellschatt, vol. 5 (1904), p. 361,
reprinted in Sammelbnde fr vergleichende Musikwissenschaft, vol. 1 (1922), p. 26x.
11
And so it eame to pass that his "regularized" version proeeeded

in the most insipid, hobbling, and banal six-eight.
Ex.
WE NEED NOT follow up the subsequent attempts of the

nineteenth eentury; this book is no bibliography. Suffiee it to
say that exotie musicology in its earlier stage eovered almost
all oriental high civilizations, and that a good many of the
studies published in those deeades have beeome classics.
Exeellent or poor, the eharaeter in all of these studies was quite
similar: the elusive pitehes of China or Japan, the melodie and
rhythmieal patterns of India or the Muhammedan lands, the
musical philosophies and musico-eosmologieal systems of the
East in general, in short, everything theoretical that eould be
gathered from native writings was presented to western readers
as a proof that oriental music was a very serious and scientifically founded art worth our eareful study. The musical instruments as weIl were deseribed, depicted, and interpreted in
numberless papers and books.
Music, the subject proper, was eonspicuously absent.
The decisive turn eame in the eighteen eighties. Baker's
doctoral thesis on the music of the N orth Ameriean Indians
(1882) 15 made the lore of primitive music an aeademie subjeet;
and two years later, in an important survey of the musical
scienees, Guido Adler in Vienna already aeknowledged M usikologie - our modern ethnomusicology - as a new and "very
meritorious" sideline.16 In the following year, Carl Stumpf, then
in Leipzig, published an exeellent study of the songs of Bellaeoola Indians. 17 Early in the nineties, Hagen wrote another
doetoral thesis, this time on Pacifie natives ;18 and in the foIlowTheodor Baker, Ueber die Musik der nordamerikanischen Wilden, Leipzig, 1882.
Guido Adler, Umfang, Methode und Ziel der Musikwissenschaft, in Vierteliahrsschrift f<ur Musikwissenschaft, vol. I (1885), p. 14.
17 earl Stumpf, Lieder der Bellakula-Indianer, ib., vol. 2 (1886) pp. 405-426.
Reprinted in Sammelbnde f<ur vergleichende Musikwissenschaft, vol. I (1922) pp.
87-104.
18 KarlHagen, Ueberdie Musik einiger Naturvlker (Australier, Melanesier, Polynesier), Hamburg, 1892.
15
16
12
ing year, Richard Wallaschek brought out the first comprehensive book on primitive music as a whole. 19
WHILE GERMANS raised the new branch of musicology to an
academic level, men in England and the States gave it a degree
of exactness still unknown on the Continent.
The first in time was an Englishman, Alexander John Eilis
(1814-1890). This prolific, versatile scholar, far from being
a musician, focused his attention mainly on phonetics and
spelling reform. In the course of his research, he made himself
familiar with acoustics and the psychology of hearing, and, in
doing so, met face to face the thrilling problems of exotic scales
and their unusual steps. Fortunately, we might say, he had no
musical ear and thus avoided the naive reliance on so unreliable
asense. Instead, he took refuge in mathematics and devised an
ingenious computing system of cents or hundredths of an equaltempered semitone. The details of this system will be expounded
in the following section; its gist is this: the clumsy ratios, so far
in almost exclusive use to express the interval between two
notes, are logarithmically transformed into handy, clear, and
graphic single numbers. The autosuggestive mishearing of the
western musician was hereby eliminated; our arbitrary evaluation of intervals yielded to mathematical exactness.
Eilis published this epoch-making method in his Tonometrical
observations on some existing non-harmonie scales, a paper first
printed in the Proceedings 01 the Royal Society (1884) and reprinted in the following year in the Journal 01 the Society 01 Arts
under the simplified title On the musical scales 01 various nations. 20
"The final conclusion is," in his own words, "that the Musical Scale
is not one, not 'natural,' nor even founded necessarily on the laws of
the constitution of musical sound so beautifully worked out by
Heimholtz, butverydiverse, very artificial, and very capricious."
This was a new realization and the decisive step away from the
standards that the West takes for granted.
19 Richard Wallaschek, Primitive music, London, I893; in German: Anfnge de1'
Tonkunst, Leipzig, I903.
20 E. von Hornbostel published a German translation, Ueber Tonleite1'n verschiedener Vlker, in Sammelbnde fr vergleichende Musikwissenschaft, vol. I, Mnchen,
1922, pp. I-76. -Cf. also ]aap Kunst, Alexander lohn Ellis, in Die Musik in Geschichte
und Gegenwart, vol. 3 (I954) col. I284 ff.
13
While Ellis was giving musicology a mathematical foundation and a theoretical tool of highest precision, another man,
though unwittingly, presented us with a practical tool, whose
boundless value is too weIl known to require renewed emphasis.
Thomas Edison had begun, at the age of thirty (in 1877), to
transform the vibrations of human voices into curves engraved
in the wax coating of rotating cylinders. At first, the sounds
were blurred, harsh, and creaking. But the revolutionary invention had been made. Sound, with its characteristic inflections
and individual timbres, could be recorded, reproduced, and
preserved for future generations; the aural counterpart of
photography had become a reality.
The paths of Edison's phonograph and the lore of exotic
music converged in 1889 when Dr. Jesse Walter Fewkes (18501930) - eminent anthropologist and, eventuaIly, Chief of the
V.S. Bureau of Ethnology - took such a machine along while
studying the Passamaquoddy in their Maine reservations and,
one year later, the Zufii in New Mexico. His output was delivered to Benjamin Ives Gilman (1852-1933), of Harvard and
shortly afterwards secretary of the Museum of Fine Art in
Boston, who transcribed the Zufii cylinders and printed the
melodies in western notation in the first volume of the Journal
of American Archaeology and Ethnology (1891). Two years later,
the German philosoph er and psychologist Carl Stumpf reported
on Gilman's paper in the Vierteljahrsschrift fr Musikwissenschaft and reproduced his transcriptions. 21 The new branch of
learning was thus established both in the V.S.A. and in Europe,
in the ethnological as weIl as the musicological literature.
IncidentaIly, Gilman's transcriptions bypassed the subtleties
of Ellis' cents, which had indeed been devised for the scaleconscious music of orient al high civilizations, but not for the
inconsistent intonations of primitive melodies. All that he did to
show deviations from the western system was to add a dash above
orbelowwhenanotewassensibly higher or lower than on his test
harmonium. And Stumpf agreed with a quotation from Heimholtz' s
Sensations of Tone: "Who would split firewood with a razor?"
21 earl Stumpf, Phonographierte Indianermelodien, in Vierteljahrsschrift fr
Musikwissenschaft, vol. 8 (1892), pp. 127-144; reprinted in Sammelbnde fr vergleichende M~usikwissenschatt, vol. 1 (1922), pp. II3-126.
14
IT TOOK LESS than half a generation after these events to

establish phonographic archives in a number of cities on either
side of the Atlantic. (The chronological order of these archives
is not safe, since some sources indicate the date when an actual
archive with a curator, an assistent, a technician, and a budget
was organized, while others are satisfied to mark the year in
which the earliest recording found its way into some science
museum, only to mildewand disappear in the basement).
These archives were meant to collect and conserve exotic
cylinders (and, later, discs and tapes), to have them ready for
serious studies, and to encourage the use of recording machines
on scientific expeditions. I myself witnessed for many years how,
with the Berlin Phonogramm-Archiv as their headquarters,
earl Stumpf and his disciple Erich M. von Hornbostel persuaded
explorers to take an Edison along, to record songs they might
hear in native villages, and to bring them back for conserving
treatment, dubbing, storage, transcription, and analysis. The
result was beyond all expectation. Ethnologists became eager,
able phonographers and delivered an ample harvest to be taken
care of in the Archiv. Dozens of German ethnological works of
that time include special chapters on music, based on this
harvest and unselfishly contributed by Von Hornbostel.
Similar activities and similar archives can be found today in
many countries, and first of all in the United States. 22
THE NEW BRANCH of learning - what was its name to be?
The Germans called it originally Vergleichende M usikwissenschaft in analogy to Vergleichende Sprachwissenschaft; the
Americans and English followed with Comparative musicology,
and the French with Musicologie comparee.
These names were acceptable fifty or sixty years ago when
haphazard bits of information trickled in: a box of Edison
22 Erich M. von Hornbostel, Geschichte des Phonogramm-Archivs der Staatlichen
Hochschule fr Musik in Berlin, Berlin, 1927, p. 4 ff. - id., Das Berliner Phonogrammarchiv, in Zeitschrift fr vergleichende Musikwissenschaft, vol. 1 (1933), pp. 40 ff.
George Herzog, The collections of phonographie records in N orth A merica and
Hawaii, in Zeitschrift fr vergleichende Musikwissenschaft, vol. 1 (1933) pp. 58 ff. id., Research in primitive and folk music in the United States, Bulletin 24 of the
American Council of Learned Societies, Washington, 1936.
Frances Densmore, The study of Indian music, in Smithsonian Report for I94I,
Washington, 1942, pp. 527-550.
]aap Kunst, Ethnomusicology, The Hague, 1959, pp. 16 ff.
15
cylinders from some North American tribe, a few doubtfully

rendered melodies from Mongolia, or the music of a touring
theatrical company from Siam. With this handful of unrelated
sampies out of an enormous stock still unknown, and unable
to draw connecting lines at so early a date, the seholars eould
not but compare their findings, weighing likeness against
differenee.
But today 'eomparative musicology' has lost its usefulness.
For at the bottom every braneh of knowledge is comparative;
all our deseriptions, in the humanities no less than in the sciences,
state similarities and divergenees. Even in the history of music
we eannot diseuss Palestrina's Masses without eomparing them
with Lasso's or Victoria's or with his own motets. Indeed, all
our thinking is a form of eomparison: to speak of a blue sky is
comparing it with a grey or a purpie one. Walter Wiora is
certainly right when he emphasizes that eomparison ean denote
only a method, not a braneh of learning. 23
After the name of Comparative Musieology had fallen into
disrepute, leading men in the field began to speak of Ethnomusicology or, in German, of musikalische Vlkerkunde, and, in
Freneh, of ethnologie musicale. 24 The term seems to have been
used for the first time in 1950.25 According to aresolution of
the Ameriean Soeiety for Ethnomusicology during the International Congress of Anthropological and Ethnologieal Scienees
in Philadelphia, September 1956, the title Ethnomusicology is
to be written without a hyphen. With or without the hyphen,
this word is somewhat unwieldy and its meaning obseure to
people who have not even a clear idea of musieology without a
prefix. The musikalische Vlkerkunde of the Germans and the
ethnographie musicale of the Freneh seem to put an exaggerated
stress on the ethnologieal part of the aggregate.
For an aggregate it is, not only in name. The man who works
in this field sits on the fenee between musieology and ethnology.
This must not be held against hirn - one needs only run over
the bulletin of some modern university to find quite similar
23 Walter Wiora, On the method of comparative melodie research, in: Journal of the
International Folk Music Council, vol. 9 (1958) pp. 55 ff.
24 Cf. also: Willard Rhodes, Toward adefinition 01 ethnomusicology, in A merican
Anthropologist, vol. 58 (1956) pp. 457-463.
25 In ]aap Kunst, Musicologica (Amsterdam, 1950).
16
aggregates: astrophysics, psycho-physiology, palaeo-botany,

and dozens of similar fusions. Connecting ethnology and music,
the ethnomusicologist, like the astrophysicist and other composites tries to break the ties of departmental sec1usion and
aims at a merger on a higher level.
ur.
THE
ETHNOMUSICOLOGIST'S
WORKSHOP
Activities of the ethnomusicologist are divided into two

successive stages: field work and desk work.
Fieldwork is done while the explorer stays in a native village.
And the word 'stays' must be stressed. A tourist can never hope
to achieve what the seientific explorer accomplishes in weeks,
in months, or even in years.
Inversely, it will not do bringing the 'natives' to our eities
and record their music in commereial studios or academic
laboratories. They are as a rule very ill at ease in an allen environment and rarely yield the quallties available at horne in
familiar surroundings in the midst of their fellow tribesmen and
on the very occasions for which their songs are intended. They
often refuse, quite rightly, to sing a. wedding song without an
actual marriage or a funeral dirge when there is no death. 26
To them it would be the same kind of frivolous profanation as
it would be in the West (excepting the stage) to simulate nuptials,
a deathbed, or a funeral for the benefit of press photographers.
Even a civilization as literate as that of India shows this deeprooted oneness of music and life: certain melody patterns are
assigned to certain definite hours of the day and cannot easily
be transferred to other hours in the interest of some concert or
recording session. L' art pour l' art, or music for the sake of music,
does not exist in the primitive world; nor is it imaginable in a
considerable part of the highly civilized East. A song is an
essential, inseparable element in primitive life and cannot be
isolated from the conditions that are its cause, its sense, and
its reason of being.
Be Cf. also Bishop Redly in earl Stumpf, Lieder der Bellakula-Indianer, in Vierteljahrsschrift tr Musikwissenschaft, vol. 2 (1886), p. 4II, reprinted in Sammelbnde
tr vergleichende Musikwissenschaft, vol. 1 (1922), p. 93.
17
It is hardly necessary to emphasize that the singer who has

been, euphemistically speaking, 'acculturated' through military
service, trade with white hawkers, or employment in port towns,
mines, and factories is of little value unless the researcher takes
greater interest in the dynamic change and decay of native
music than in the most traditional forms available.
Reception, adventures, and success in the village are impossible
to anticipate, and much depends on the knowledge and good will
of the guide and interpreter. Generally speaking, the difficulties
are the same in musicology and in anthropology, on which the
interested reader may consult the report of a seasoned explorer:
Chapter 20 on The Ethnographers Laboratory in M. J. Herskovits' Cultural Anthropology (New York, 1955, 4/1960).27
I also recommend to look into Frances Densmore's musical
experiences with American Indians. 28 But in all admiration for
her incessant, diligent work, I must take exception to one of
her principles, which I deern rather dangerous and inadmissible:
she told her singers that they "must sing in a steady tone and
not introduce the yells and other sounds that are customary to
Indian singers. The recording," she said, "is not intended to be
realistic, but to preserve the actual melody." The sentence is a
challenge. What is an "actual melody?" Certainly not an abstract sequence of notes shom of all its characteristic traits.
Everyone versed in European folksong as well as in primitive
music knows the importance, the indispensable role, the expressive and sometimes eerie power of those interrupting shouts,
be they magie incantations addressed to ademon or simply
diseharges of exeess vitality. What is the sense of improving
the teehniques of reeording and at the same time neglecting a
faithful, realistic, unpolished, and undistorted rendition?
The musicologieal explorer might find any shade of behavior
between the extremes of an unlimited willingness to show
27 Also in Hoebel, Jennings, Smith, Readings in anthropology, New York, 1955;
Cornelius Osgood, The selection 01 a community lor anthropological study and Ethnological Held techniques (pp. 6-I1); Philip Phillips, James A. Ford, James B. Griffin,
Archaeological Held work (pp. I7-2I). - Cf. also: Hugh Tracey, Recording Alrican
music in the Held, in Alrican Music, vol. I (I955) no. 2, pp. 6-u; J. Manchip White,
Anthropology, New York, 1955, pp. 170 ff.; Jaap Kunst, Ethnomusicology (The Hague,
1959), 3rd. ed. pp. 14-16.
28 Frances Densmore, The study 01 Indian music, in Smithsonian Report lor I94I,
Washington, 1942, p. 536.
18
musical prowess with gusto and pride, and an invincible shyness

to exhibit sacred songs or a deadly fear to lose one's seH, as
with that Eskimo who in honest simplicity said: "My songs are
part of my soul, and if the demon in the white man's magie box
steals my soul, why, I must die." 29
The explorer might find one village familiar with the phonograph from preceding visits, and another one to which the
phonograph is novel and the incomprehensible intention of the
white man and his whole complicated procedure utterly intimidating. When Mrs. M. Selenka visited the Vedda in inner Ceylon
in 1907, she had to tell them that "the box hid a little birdie
eager to hear the Vedda sing. But he could hear them only if
they came quite near. With-Iong intermissions and the gentle
persuasion on the part of the interpreter, I succeeded to seat
three of them elose enough to the horn." 30
Things are essentially easier today, not only because phonography is more widely known and needs no longer a tale of inquisitive birdies, but also because the phonograph itseH has
basically changed. Modem phonography facilitates the process
both in technique and in its psychological possibilities 31 in a
similar way as thephotographer snaps you unawares when you
are relaxed instead of demanding rigid, unnatural poses in front
of a slow-working black-elothed camera.
The old forbidding horn or receiving funnel from which you
were not allowed to swerve by a handbreath has disappeared.
The easily portable Edison phonograph of 1877 with its crank
and spring action and with wax cylinders fragile and hard to
preserve yielded to Emil Berliner's heavier 'gramophone' with
shellac disks, patended in 1887; to electric manipulation with a
tiny, often freely movable microphone that we can carry about
to any vantage point and even hide from the singer (1925); to
Peter C. Goldmark's long-playing micro-groove machines (1948) ;
and ultimately, so far, to magnetic tape recorders. Created by
the Danish inventor Valdemar Poulsen as early as 1898, tape
29 Christian Leden, Ueber die Musik der Smith Sund Eskimos, in Meddelelser om
Grllnland, vol. 152 (1952), p. 22.
80 Max Wertheimer, Musik der Wedda, in Sammelbnde der Internationalen Musikgesellschaft, vol. II (1909/10), p. 300.
81 Latest monograph: Roland Gelatt, The fabulous phonograph, Philadelphia &
New York, 1955.
19
has after a rather erratic career come into its own and offers to
workers in the field the advantages of easy transportation and
handling, excellent rendition, a capacity of practically indefinite
recording without interruption, and an equally indefinite number
of playbacks without a measurable deterioration.32 In I955, a
New York distributor presented a "Grooved Tape-Disk,"
developed in Cologne and said to play eight hours of uninterrupted music on three miles of a sound track one-thousandth
of an inch in diameter.3s
Whether on cylinders, disks, tapes, or tape-disks - phonography has grown indispensable in musical field work. The times
when musicians (and non-musicians, too) proudly referred to
their more or less well trained ears are gone. We know from
bitter experience how unreliable and deadly prejudiced man's
senses are, how easily we project into a totally foreign style of
music the tempered melody steps and even-stressed rhythms
of western tradition, and, hence, how small the documentary
value of such unverified impressions iso As a leading anthropologist has expressed it: "We hear music, no less than we
produce it, in terms of very subtle conditionings that make up
our musical enculturation. "S4
Still, even under such favorable conditions the ear is not always
up to expectation. Much as a listener may be trained in analyzing
unfamiliar types of music, he has at times desperate difficulties
with melody steps others than the ones of our habitual welltempered system or with confusing rhythms and counterrhythms.
The first fact that I touched upon - "steps others than the
ones of our habitual well-tempered system" - confronts the
ethnomusicologist with two essential problems: how to measure
and how to express them more correctly than by an embarrassed
"a bit larger (or shorter, smaller) than a wholetone." Of necessity, we look around for mechanical help. But, at least for
the purpose of field work, the helping device should, in J aap
Kunst's words, be "easily transportable, of a simple manipulated construction, and able to stand a certain amount of
knocking about. It should further," he adds, "not be too ex88
88
84
S. J. Begun, Magnetic l'ecording, New York, 1949.

New York Times, 1955, April 13.
Melville J. Herskovits, Cultul'al anth1'opology, New York, x955, p. 282.
20
pensive to buy and maintain; musicologists do not, as a rule,

excel as possessors of earthly riches, and only a very few are
privileged to receive adequate finandal assistance from sdentific
institutions or interested private persons." 35
TONE-MEASUREMENTS can be obtained either acoustically or
visually.
The older acoustical methods might in general suffice for the
less subtle needs of primitive music, while they are hardly fine
enough for the purposes of oriental music. They compare the
pitches heard with asounding standard of satisfactory pitch
persistence and measuring gradation. Helmholtz used a set of
spherical brass resonators; Eilis availed himself of quite a
battery of tuning forks. Stumpf and Von Hornbostel followed
Appun's Tonmesser, a kind of harmonium with many reeds at
short tonal distances but without actual keys; and Von Hornbostel also devised a tiny Reise-Tonometer for field work: a
single reed with a tuning wire manipulated by a dial, which
indicated the frequency when the spinning hand, in shifting
the wire up and down along the reed, had tuned it to the pitch
of the note to be measured. J aap Kunst sticks to his monastic
monochord, a narrow soundbox with graded surface and a
movable bridge under a metal string. After adjusting the open
string to the pitch of a tuning fork or pipe, any tone heard in the
phonograph (or its octave) can be reproduced on the monochord
by moving the bridge to a position where the ear is reasonably
certain that the two pitches, on the shortened string and in the
phonograph, are identical. The grading on the surfaee allows
for an approximate and in most cases suffident measurement
(except when the bridge draws too elose to the motionless end
of the string). The plucked sound, breaking off at onee, is another drawback.
All these acoustical methods suffer from multiple approximation. (r) It is the ear that deddes on the equality of the piteh
to be measured and the piteh to be the standard. ( 2) The distanees
from standard to standard are as a rule so wide on the test
instrument that the pitch to be tested lies between two of them
85 Jaap Kunst, Musicologica, Amsterdam, 1950; 3rd edition (as Etknomusicology),
The Hague, 1959, p. 10.
21
at a point left to our estimate. (3) In many cases the acoustical

method provides only one octave of standards and hardly ever
the whole range of men's and women's voices and of instruments
that we might want to test. The solution is transposition by an
octave or even two, once more incumbent on our unreliable ears.
The visual methods are essentially more exact, notably those
based on modern electronics. As the principal apparatuses, we
use the oscilloscope, the stroboscope, and the electronic counter.
The oscilloscope, ignoring sound as an audible phenomenon,
makes the sound waves or oscillations visible and measurable
and allows for a photographic reproduction of single notes as
well as of melodies. In the case of such a photographic recording,
we speak of photophonography,36
Oscilloscopic writing has of late been greatly improved,
above all by Olav Gurvin at the University of Oslo.37 In the
U.S.A., Chades Seeger of Santa Barbara (California), in collaboration with V. C. Anderson of San Diego, is developing an
oscillograph which needs neither processing nor mathematical
calculations. A heated stylus traces the waves by melting the
coat of a special plastic graph paper 63.5 mm wide. The design
that the stylus leaves permits the instantaneous reading of all
the pitches of an unaccompanied melody, including exact
rhythm, the subtle changes that a vibrato entails, and, additionally, the finest shades of speed,3s
The stroboscope, unlike the oscilloscope, centers around a
disk of alternately black and white sectors. Under the control
of a dial, the disk reacts to a light fed from the sound source.
When the sectors or 'spokes' appear to stand still, the dial is
indicating exactly the frequency of the outside sound. When
the spokes seem to move clockwise, the outside pitch is higher than
thestandardinside; when counter-clockwise, the pitch is 10wer.39
36 Milton E. Metfessel, Phonophotography in folk music, Chapel Hili, 1928. M. Grtzmacher und W. Lottermoser, Ueber ein Verfahren zur trgheits/reien Aufzeichnung von Melodiekurven, in Akustische Zeitschrift, vol. 2 (1937).
37 Olav Gurvin, Photography as an aid in folkmusic research, in Norveg, vol. 3
(1955). - Kar! Dahlback, New methods in vocal folkmusic research, Oslo, 1958.
38 Char!es Seeger, Toward a universal music sound-writing for musicology, in
journal of the International Folk Music Council, vol. 9 (1957), pp. 63-66.
39 Milton E. Metfessel, The strobophotography, in journal of General Psychology,
vol. 2 (1929), pp. 135 ff. - R. W. Young and A. Loomis, in journal of the Acoustic
Society of America, vol. 10 (1938), pp. II2 ff. - Fritz A. Kuttner, Der stroboskopische
Frequenzmesser, in Die Musikforschung, vol. 6 (1953), pp. 235-248.
22
Electronic counters are probably handiest for measuring

single pitches. They convert acoustical into electrical phenomena
and answer a tone by the appearance of the exact frequency
number in current figures - way beyond the range of audibility,
from one cyde per second up to 120,000 cydes. 40
A RELIABLE EAR is indispensable during the second phase:
when the ethnomusicologist sits down at home, studies the
yield of his field work, and makes it accessible to readers who
have not heard the originals. While the machine plays back the
recorded pieces, he eagerly tries to catch the tonal picture. This
time, the ear is not engaged in the weird and hopeless multiactivity of the non-phonographer: to follow in an impossible
haste lest a note should get lost, to evaluate the passage just
heard, to write it spasmodically down longhand while the singer
is already far ahead, and in all this to fail inevitably. We find
it a tremendous help to turn back the pick-up as often as necessary and repeat the piece or an elusive phrase until we are ready
to transcribe it into our western notation and thus enjoy the
benefit of leisurely study as weIl as of careful control. But also
he who studies a piece after the transcriber has a new and badly
needed control. For transcriptions are too subjective to be
trusted uncritically and should, wherever possible, be compared
with the phonographie original.
A BIG HOWEVER stands behind these gadgets. Much as they
are perfected, we1come, and indispensable, they serve one
purpose only: the faithful measurement of the constituents
in the melodies we hear from the phonograph. But the most
precise of pitches and metrical values have only a limited merit.
Even the unsophisticated songs of the earliest among the
primitives are much more than lifeless sequences of incoherent
pitches and lengths. Indeed, the less sophisticated a song is,
the less we can rely on individual pitches and lengths, which,
when consciously handled, belong to systematic music. Our
own 'blue' notes in jazz and similar types of music are a case
in point.
The editor's essential task is organization, analysis, interpre40
Hewlett-Packard Journal, vol. 6, no.
IX
(July 1955).
THOUGHTS AND METH ODS
23
tation of the raw materials that the phonograph provides; it

demands grouping, phrasing, and punctuating. This is, beyond
the gadget' S power of precision, a musician' s chore - though
not a musician who arrogantly relies on a purely western education or, much worse, on his "unfailing instinct" (which is
the prerogative of animals, not an asset of men). Only he who
knows the pitfalls of western habits and has learned to escape
them is up to such intricate, delicate work and can hope to do
justice to eastern and primitive music.
And even such a man is far from being infallible. When we
compare an original phonogram with a transcription made by
another person, we will more often than not disagree. This is
not necessarily a question of keener ears, but rather of the
analytical apparatus in the brain - just as two painters of equal
ability (and even photographers) might be at variance in seizing
the likeness of a modeL Indeed, our own transcriptions will often
be unsatisfaetory when we resurne and revise the work of
yesterday. One cannot too earnestly warn the student against
accepting printed transcriptions as gospel truth. 41
THE SECOND PROBLEM - how to express steps other than the
ones of our well-tempered system - pivots around numbers,
fraetions, divisions, and logarithms. The pitches that we hear
differ in only one faetual, physical quality: the frequency of
vibrations or, in other words (as we understand it today) the
number of to's plus fro's in which astring (or any other sounding
objeet) moves during a second. Thus, our tuning forks, in
rendering the standard a', vibrate 440 times in a second - 440
times back and 440 times forth. This so-called frequency is in all
cases the unique, unavoidable measure of a single note.
It is more complicated to describe a step or distance between,
or interval of, two notes (step or distance and interval are
psychologically not identical; but we may here be allowed to
negleet the diffetence). To express this interval in numbers, we
do not subtraet, but must divide the two frequencies. When one
of the two notes vibrates 440 times per second, and the other
one 330 times, their interval is 440 : 330, or 4 : 3.
But ratios are not always that simple. Let the two not es have
41
See also: ]aap Kunst, Ethnomusicology (3rd. edition, The Hague, 1959), pp. 37 ff.
24
THOUGH TS AND METHODS
435 and 3911 vibrations, and we get a fraction ungraphic,

unwieldy, irredueible. If we compare this fraction with another
one that the nearest neighboring step provides - 39I ! : 348 we would not, in fact we could not, from the forbidding looks
of these two ratios become aware that eitherone expresses
exactly the selfsame distance of a wholetone. They differ in
their constituent numbers, because, quite individualistically,
they tell us about the aceidental relations of a', g', and f', but
fall to indicate the general sizes of two wholetones, which after
all are the same whatever note in the whole range of hearing
you take as the starting point. The situation is as abnormal as
if we would express the distances from Boston to New York and
from New York to Washington in different, complicated ratios
instead of reading simply the mileage from road maps and
time-tables.
In order to attain such a musical mileage, two needs are
obvious: (I) to express similar distances by identical numbers,
regardless of pitches and frequeneies; (2) to express them so
that the numbers, firmly lodged within a frame-work of fixed
standards (as on graph paper) give instantly a c1ear picture of
the size in question.
The only way to meet these needs is to transform frequency
numbers into logarithms. This transformation implies automatically a transformation of the original awkward division into an
easy subtraction : log (x: y) = log x-log y. By the simplest
of operations this subtraction can be expressed by a single number
which, answering the first of our needs - "to express similar
distances by identical numbers, regardless of pitches and frequeneies" - is the log of the interval in question, or log i.
Need (2) - "to express distances so that the numbers, firmly
lodged within a frame-work of fixed standards, give instantly a
c1ear picture of the size in question" - was met by the Frenchman Felix 'Savart (1791-1841) and his system of logarithmic
savarts. Outside France, the savarts have been supplanted by
Alexander J. Eilis' system of cents (1884). One savart equals
3.99 or, rounded off, 4 cents, and the semitone is expressed by
25 savarts and 100 cents.
The following is Eilis' principle. All possible shades of pitch
are contained in one octave, which has the frequency ratio of
25
2 : I or, short, of 2. An octave contains in our 'well-tempered'

system twelve equal semitones, each of which, mathematically
speaking, amounts to \12. In EIlis' system, each semitone is
divided into 100 cents, and the whole octave, consequently, into
1200 cents. Thus,
a semitone has
a wholetone,
a minor third,
a major third,
a fourth,
a tritone,
a fifth,
a minor sixth,
a major sixth,
a minor seventh,
a major seventh,
an octave,
100
200
300
400
500
600
700
800
900
1000
1100
1200
cents,
cents,
cents,
cents,
cents,
cents,
cents,
cents,
cents,
cents,
cents,
cents,
To abbreviate our excursion into the dreaded mathematical

field, let us leave the theoretical aspects and try the practical
application. To find the cents of a distance or interval, you take
the following steps.
(I) Find the frequencies of your two notes by one of the
methods already mentioned in this section - monochord, tonometry, stroboscopy, phonophotography, oscillography, electronic
counting.
(2) Find the two logarithms and their difference, or log i.
(3) This log i can easily be transformed into cents with the
help of the following table.
cents
log
cents
log
cents
log
1.
.00025
10
.0025
100
.025
2.
.00050
20
.0050
200
.050
.00075
30
.0075
3 00
.075
.00100
.00125
.001 5 1
60
.01 5 1
4 00
5 00
600
.100
5
6.
40
50
.0100
. 001 7 6
70
. 01 7 6
7 00
.176
.0125
.125
.151
26
cents
log
cents
log
cents
log
8.
.00201
80
.0201
800
.201
.00226
90
.0226
9 00
1000
.226
.251
IIOO
.276
1200
.3 01
To give an example: suppose your log i is .12461, you find

yourself elose to .125 and hence to 500 cents, which are the
equiv?lent of an equal-tempered fourth. If you are interested
in absolute correctness, you add the two columns of logs and
cents:
.100
.0226
.00201
400
90
8
which reproduces the .12461 of log i and the resulting cent

number 498, numerical symbol of the perfect fourth.
Other shortcuts were made by Erich M. von Hornbostel,42
R. W. Young,43 and Heinrich Husmann. 44
When no table of logs is at hand, adetour can be made with
the help of an auxiliary number and four arithmetic operations,
as subtraction, multiplication, addition, and division. These
cumbersome procedures have been described in earlier books 45
and need no resumption - the less so as they have in the meantime
been superseded by the mechanical device of the music rule.
The music rule, devised in 1949 by Marcus Reiner, professor
at the Technion of Haifa, Israel, with the musical collaboration
of Edith Gerson-Kiwi in Jerusalem, is constructed like an
ordinary slide rule, which shifts two twelve-centimeter scales
along one another. The left scale pictures the equal-tempered
octave c'-c" from 264 to 528 cyeles (double vibrations or frequeneies); full dividing lines mark every ten cyeles, and four broken
lines in between, every two cyc1es. The distances up from line
42 Erich M. v. Hornbostel, Eine Tafel zur logarithmischen Darstellung von Zahlenverhltnissen, in Zeitschrift fr Physik, vol. 6 (1921), pp. 29 ff.
48 R. W. Young, A table relating trequency to cents, Elkhart, Indiana, 1939.
44 Heinrich Husmann, Fnf- und siebenstellige Centstafeln zur Berechnung musikalischer Intervalle, Leyden, 1951.
45 Curt Sachs, The rise of music in the ancient world, New York, 1943, p. 28.
27
THO UGHTS AND METHODS
to line, however, are not equal as on our thermometers, but

become steadily shorter. The reason is obvious. The differences
between the two frequency numbers of any interval transposed
upwards grow steadily larger; if you compare the wholetone
distances c'-d' and, an octave above, c"-d", the result will be:
d' has 297 vibr.
c' has 264 vibr.
diff. :
33 vibr.
d" has 594 vibr.

c" has 528 vibr.
diff. :
66 vibr.
In frequency numbers, the distance is in the upper octave

twice as wide; but to the musical ear and to the cent system
it is exactly the same. The only way out is to shrink the spaces
from line to line in the same ratio in which the vibration numbers
grow wider apart; in our example: the space between c" and d"
is on the music rule just half as wide as the space between c'
and d'. (The d" is actuaIly not on the rule). All the spaces narrow
on their way up.
The right-hand scale matches the left one in its length of
twelve centimeters. But since it represents the unchangeable
octave of I200 cents (or ellis, as the two Israeli scholars caIl
them), this length is divided by fulllines into centimeters, each
of which comprises the one hundred cents of a semitone, and, by
broken lines, each centimeter is divided into ten degrees of ten
cents. All that the measuring person has to do in order to
transform a distance into cents is: (I) shift the cent scale so far
that its base line stops exactly opposite the frequency number
of the lower note; and (2) find the cents opposite the frequency
number of the upper note. In case one or both notes exceed the
range of the c'-c" octave, there is an additional move: (3) for
the octave above, the cycIes on the left side must be multiplied
by 2, and the number of cents increased by I200; for the octave
below, the cycles must be divided by two, and the cents diminished
by I200.
For practical purposes this approximation will suffice. For
theoretical preciseness it is easy to give the music rule a larger
size, maybe 24 centimeters, and augment the number of Reiner's
subdivisions on either scale.
28
Three years after Reiner, Fritz Bose came out with an almost
identieal 'nomogramm.' 46 Instead of a slide rule, it necessitates
aseparate rule in millimeter division to be laid against the
logarithmic vibration scale (which amounts to the same). Moreover, its vibrations are marked from five to five cyc1es, as against
the more accurate division from two to two cyc1es on Reiner's
model.
IN SPITE of all these refinements, we are still very far from
a faithful transcription of the fleeting impression on our ear
into a lasting impression on our eye. The reasons are evident.
At first, there is in primitive or oriental music no silent composing with paper and pencil. Dreamily humming and strumming,
composers create their melodies, and even after polishing rugged
passages do they not pen a definite version. On playing in public,
they are not bound to an authentie printed form, to an Urtext.
There is none. Producing and reproducing fuse into a delightful
unit; the weIl-wrought, mentally definite form and the indefinite,
momentary impulse reach a perfect balance. Any notation would
spoil this equilibrium in the undue interest of finality; it would
destroy the potentialities of a free-flowing melody in favor of
stagnant impersonality. Although there are a few native notations in countries of high civilization - in India as weIl as the
Far East and the Muhammedan world, - they are poor and vague
and assist recollection rather than performance or study. Be
their symbols borrowed from the alphabet or from descriptive
gestures like the 'neumes' of the Middle Ages, be they fingering
guides like the instrumental 'tablatures,' or conventional signs
for certain groups of notes, they render at best a lifeless skeleton.
What makes it worse is that our notes on four or five staff
lines and in the spaces between them are intended to serve a
diatonie gender with its alternation of steps and halfsteps.
More accurately: the original concept was meant for ancient
Europe's tertial chains with infixes, such as C tl, E I GaB or
D e F gAb C, to be discussed later in this book.
Reading oriental musie from western staff lines is just as
deceptive as reading oriental poetry in a twenty-six letter
46 Fritz Bose, Ein Hilfsmittel zur Bestimmung der Schrittgrsse beliebiger Intervalle, in Die Musikforschung, vol. 5 (I952), pp. 205-208.
29
transliteration without array of 'diacritical' dashes, tildes, dots,

and hooks. The staff lines and spaces entice the reader into a
fatal misconception. The notes, let us say, of a Siamese me1ody,
which has neither whole nor semitones, have no proper place
anywhere on the staff. Forced upon and between the five lines,
they deceive the reader with perfect fourths, thirds major and
minor, and seconds major and minor where there are no such
steps at all. They also deceive the reader by suggesting, in the
conflict between the familiar lines we see and the unfamiliar
steps we hear, that the exotic melodies in question are out of
tune, in other words, that the West is right and the East is
wrong. Additionally, our script is not prepared to show the
nasaling, yodeling, vibrating, gliding, falsetto shades, which,
almost more than structure, are characteristic traits of the various
musical idioms outside the West. The unexperienced reader
deciphers such notations as a student wou1d decipher a Koran
text after studying Arabic from a grammar: the Arab wou1d not
be able to identify the sura, nor wou1d the Westerner recognize
the same text when read by a native.
Out of this want, Otto Abraham and Erich M. von Hornbostel published in 1909 "Suggestions for the transcription of
exotic melodies." 47
Most of their suggestions adapt the well-known symbols, like
notes, flags, keys, ties, slurs, dots, to the special needs of nonwestern music. Among the newly invented signs I want to
mention only the inverted fermata which occurs quite often in
transcriptions:
\.:,/ means the shortening of a note, and
means a lesser shortening of a note.
(\:.I)
These symbols are objectionable. The dotted arc is a familiar

symbol of lengthening, both in our current notation and in that
of the ethnomusicologist. I t is again used in ethnomusicological
transliteration when we want to indicate in a symbolic scale at
the end of a melody what relative degrees of weight and frequency its individual pitches have. In such symbolic scale, we write
47 Otto Abraham und Erich M. von Hornbostel, Vorschlge tar die Transkription
exotischer Melodien, in Sammelbnde der Internationalen M usikgesellschajt, vol. I I
(1909), pp. 1-25
30
the most important pitch (according to its frequency, stress, or

end position) as a wholenote, the next important one as a halfnote, and so forth; and we introduce the fermata sign to mark an
intermediate shade of rank. Inverting the unmistakable symbol
of lengthening to indicate the opposite meaning would be just
as artificially rationalistic as turning the Cross upside down to
symbolize non-Christian faiths: the form of a symbol impresses
our eyes much more than does its position. Incidentally, our
current notation inverts the fermata anyhow when its place is
below the note to be lengthened, without changing therewith or
even inverting its purport. Are we allowed to use a symbol now
for one meaning now for its opposite?
Parentheses, on the other hand, are generally accepted as
symbols of an aside, but not of a diminution. Both signs are
ungraphic.
Although the Abraham-Hornbostel suggestions are as a whole
still valid, we follow today the recommendations of a number of
experts convened at Geneva in July I949 by the Archives
Internationales de Musique populaire, which were resumed in
C.I.A.P. (Commission Internationale des Ans populaires) Information, nov.-dec. I949 (nos. I5-I6) and in Document du Conseil
International de la Musique (Unesco), I950. Among the symbols
suggested at that time, there are little arrows, aimed up or
downward to raise or lower the note by a microtone; an are
opening downward ('eonvex') marks a slight lengthening of a
note, and an are opening upward ('concave'), a slight
shortening.
Renouncing the current staff notation, a number of scholars
have replaced the symbols on lines by round or angular curves,
where the ordinate stands for pitch, and the abscissa for duration. 48 This form of writing is meant to depict the general
trend and flow of a melody and has as such advantages in certain
cases. But it is neither accurate nor graphic enough to be exclusively accepted.
Weighing the merits and defects against each other, we find
that even the most painstaking method will not give us ultimate
48 Examples: B. I. Gilman, HOPi songs, in Journal 0/ American Ethnology and
Archaeology, vol. 5 (I908). - E. Lineff, The peasant songs 0/ G1'eat Russia, London,
1905; id., Velikorusskiia piesni, vol. 2, St. Petersburg, 19II.
31
satisfaction. No musical script can ever be a faithful mirror of

music, just as no spelling, phonetic or otherwise, can give an
adequate idea of the sound of English, French, or any one
language to those who do not master it. Even for western music
our western notation is quite unsatisfactory from many viewpoints. Leaving this latter problem to the many untiring reformers around us, we might, for our present concern, take heart
in the rapid growth of phonography, which makes a good deal of
our writing symbols superlluous by restoring music to the ear where it belongs.
But as long as we still are forced to notate what we hear, three
details should be observed.
(1) It is often confusing and perfect1y useless to transcribe a
piece exact1y in the original key when this tessitura requires
many sharps and flats. Native singing does not follow tuning
forks, nor does it depend on the 'ethical' character of a certain
tonality. Another singer would very probably perform the same
piece in a different key. Why must we impose on the reader the
fortuitous pitch of a primitive melody when a simple transposition by a semitone rids him of the dead ballast of superlluous
accidentals? How much more readable would the six-sharp
songs in B. I. Gilman's Zufii collection be if a shift by a semitone
up or down gave them instead a single sharp or one flat! Our
duty to scholarly accuracy can be amply fulfilled by adding the
short remark: "original a semitone higher (lower)."
Such transposition, however, should never exceed one semitone.
Otherwise, characteristic tessituras might get lost in a fictional
range. He who knows the eerie, solemn prayers of Tibetan lamas,
which so often dwell down by (great) C,49 or even the songs of
North American Indians, which easily descend to E or, in
women's parts, to e, will realize that a transposition by more
than a semitone would distort the very character of that darkcolored music. 50 And just as notation must keep the lower ranges
intact, it should in other cases preserve the opposite impact of
extreme height - as in East Transcaucasia, where an Azerbaijani
tenor freely and without falsetto enters on eil and ascends to a
EFw FE 4504 (P 504) IV I3.
Cf. Frances Densmore, The study 0/ Indian music in Annual Rep01't, Smithsonian
Inst. for I94I, Washington, I942, p. 542.
49
50
32
woman's e",51 wbile, next door, Uzbekistani women go quite

naturally up to b".52
An exception must certainly be made when a number of
melodies are written beJow one another and transposed to the
same pitch for easier comparison.
(2) Avoid aIl key signatures reminiscent of those which we
write in western music. Created for tonality in our harmonical
sense - such as A major or B minor - they are nonsensical and
dangerously misleading in the transcriptions of early melodies
which have no such tonality. Give a transcription just the
signature of the few sharps or flats that you do not like to repeat
aIl the time within your notation, but not more. Do not argue
that, while you have only a sharpened C in your piece, you
should liberally add an F sharp in the signature since in our
western music no C sharp occurs without an F sharp. All that
you achieve with this illogical, illicit addition is to suggest one
of our major or minor tonalities, from which primitive melodies
are aeons away. In a similar way, it is inadmissible to print an
F sharp signature at the beginning of each staff, thus implying a
G major or an E minor, although F sharp does nowhere occur in
the melody at aIl. 53 Such procedure becomes suicidal when the
sharp not only is completely absent from the piece but must be
re-naturalized wherever there is an F: for aIl that the melody is
not G major but, if the transcriber thinks he cannot do without
an accepted scale name, at best Mixolydian. 54
The worst examples that come to my mind are a couple of
Pawnee songs, each of a one-step pattern. One of them, just a
zigzag of d and e, appears in the masquerade of D major, with
two sharps as the signature which raise the two non-existent
notes fand C. 55 If the author thinks he cannot do without the
make-believeof a western scale, why not just as weIl D minor?
EFw FE 4505 (P 505) IV l6; FE 4416 (P 4l6) 13, II I. Also: FE 450l (P 50l)
(from Syria).
52 Other examples of high tessitura in: Richard Wallaschek, Primitive music,
London, l893, pp. 75 ff., and Anfnge der Tonkunst, Leipzig, 1903, pp. 77 ff.
53 Cf. No. 43 in Frances Densmore, Teton Sioux music, Washington, 19l8, with
an F sharp in the signature without a a single F in the pentachordal melody. How
does the author know that, if there were one, the singer would sharp it?
54 Cf. the example transcribed by D. Arakchiev, in Proceedings of the MusicoEthnographical Commission, vol. I, p. 293, Music Supplement.
55 Transcribed by Edwin S. Tracy in Alice C. Fleteher. The Hako, a Pawnee
ceremony, Washington, 1904, p. 70.
51
33
The other one, a skiriki or coyote warrior song,56 boasts of three

sharps, but consistsin the second part only of the two unsharpened
notes a and e. The transcriber, unable to think of any two notes
outside the twenty-four tonalities that the tidy equal temperament of Europe provides, decided for A major. Why did he not
choose A minor, which would have allowed him to dispense with
all the three ghost accidentals? There is no answer. The obsession
of tonality where there is no tonality, with all its high-handed
decisions, is beyond reason and justification.
Even the method of Mieczyslaw Kolinski, otherwise an excellent, leading ethnomusicologist, falls. He squeezes the primitive
melodies into a system of church modes and imaginary cydes
of fifths and succumbs, in my opinion, to an ill-fated attempt at
over-dassification. 57
(3) Take good care to write the necessary accidentals in the
front signature on the line or in the space exact1y where they
belong. An F sharp that occurs exdusively in the one-lined
octave must appear in the first space and not, as in modern
notations, on the fifth line. A wider melody range might easily
leave this two-lined F unsharpened.
I also suggest that the bass def so often used in transcriptions
of men's songs should be replaced by the modern G or violin def
over an 8 as a transposing symbol in the interest of a unified
script (although I do not doubt in the least that the readers of
ethnomusicological books and papers are perfect1y able to
decipher the F def).
IV. THE QUESTION OF ORIGIN
Questions of origin fascinate the laymen no less than historians. To see the Mississippi, the Nile, or the Danube in all
their quiet majesty is an unforgettable experience; but it is
still more exiting to find their sources and watch the new-born
streamlets trickle from under the rocks.
Natalie eurtis, The Indians' book, New York, 1906.
M. Kolinski, SUl'iname music, Part 111 of Melville J. and Frances S. Herskovits,
SUl'iname 10lk-lo1'e, New York, 1936. - Kolinski's system has been adopted, for
example, by Alan P. Merriam; cf. his Songs 01 a Rada Community in Trinidad, in
Anth1'opos, vol. 51 (1956).
68
57
34
The currents of history, alas, delude our curiosity: while the

ageof man spans millions of years, historical evidences hardly cover
five thousand ofthem ora tinyfraction of onepercent. All that which
precedes these evidences is wide open to guess and conjecture.
The origin of music has likewise been a favorite object of guess
and conjecture. Primitives trace it occasionally to animals of
their world: the Luisefio in southern California relate that music
was invented in the earliest times, and Lion, Eagle, Raven, Frog,
Deer and others were the first musicians. 58 The ancient high
civilizations appointed one of their gods to be the inventor or at
least a patron of music - Thot in Egypt, Narada in India, Apollo
in Greece. An exception is Judaism, which as a monotheistic
religion had no special divinity available; Genesis 4: 2r just mentions Jubal as "the father of all such as handle the harp and the
pipe,"withouttakingtroubletotracetheroots of so innate an expression assinging. The ancient Chinese, quite modem in this
assumption, already believed that music had had a slow development.5 9
The later West draped its guesswork with the flowing gown
of the scholar. It pretended to find the origins of music, now in
the imitation of singing birds, now in shouts to signal "over the
hills and far away," or in the natural rhythmic alleviation of
organized toll,60 indeed, in human speech. 61
Not one of all the allegedly scientific theories is borne out by
facts. At best, some of their connections are in the realm of
possibility; and even the possibilities are depressingly weak.
A shout or a bit of regular work may doubtless have led to a
melodicstrain or a rhythmic pattern; but can they have begotten
music as a whole in all its ramifications?
Lately, the derivation from speech has been modified via the
so-called tone languages. 62
58 Constap.ce Goddard DuBois, The religion 01 the Luiseno Indians, in University
01 Calilornia Publications in American Archaeology and Ethnology, val. 8 (1908),
p.166.
59 L Pu-We, Shi Ch'un Ts'iu [3rd. century B. C.], ed. Richard Wilhelm, p. 66.
60 Kar! Bcher, Arbeit und Rhythmus, Leipzig, 1896 (3rd ed. 1903, 5th ed. 1918). Erich M. von Hornbostel, Arbeit und Musik, in Zeitschrift der Internationalen Musikgesellschaft, val. 13 (191Z), pp. 341-35.
61 Latest llesumes: Robert Lach, in Handbuch der Musikgeschichte, ed. Guido
Adler, znd ed., Berlin, 1930, val. I, pp. 3 ff. - Geza Revesz, Der Ursprung der Musik,
in Internationales Archiv lr Ethnographie, val. I I (1941), pp. 65 ff.
62 Fritz A; Kuttner, Die verborgenen Beziehungen zwischen Sprache und Musik,
in Musica, val. 5 (1951), pp. 13 ff. - Marius Schneider, Primitive Music in The New
OXford History of Music, 3rd ed. (1957), val. I, pp. 6-7.
35
In a number of languages past and present, sounds are not

only long and short or stressed and unaccented, but also, in a
musical sense, ascending, level, descending, or medium, high,
and low. Tone languages, as they are called, may appear in two
forms. Only one of them justifies the name.
We do not speak of tone languages when inflections follow
local traditions and habit rather than need, as the dropping off
at the end of a word in northem Germany and, instead, the rise
in southem Germany. Such inflections do not change the meaning
of words, but are at the very best, oratorical shades, beyond any
regional idiom: even in languages as unmelodic as English and
German, words like no and nein express as a rule a question as
they glide upwards; descending, they stand for a statement or
an interdiction. No language proceeds in an absolute monotone.
But in other languages, say Chinese or Siamese, the tone is
absolutely essential to identify the meaning of a syllable which
might have a quite different sense according to whether it is
high or low in pitch, or rising or descending. 63 The same can be
said of languages around the GuU of Guinea. In Y oruba, the word
oko, when middle-pitched, means 'husband,' when low-pitched,
'spear,' when ascending, 'hoe,' when descending, 'canoe.' "Tones
are more significant in Yoruba than vowels and consonants." 64
The language of the Jabo in eastem Liberia assigns to the
syllable ba four totally unrelated meanings, depending on its
pitch: spoken on high pitch, it means 'to be broad,' in the middle
register, 'namesake,' on a somewhat lower one, 'tail,' and spoken
quite low, it is a particle expressing command. 65 The language
of Uganda has, like ancient Greek, three 'accents' : a lift of the
voice or 'acute,' a half-lift to a medium pitch, and a lift and
following drop below the unaccented level or 'circumflex.' In speaking, these inflections are carefully observed, since, for example, a
word like musala has seven meanings unless the inflection serves
as a determinative. But they also must be observed in melody. 66
68 On tone languages in general, see Kenneth L. Pike, Tone languages (Ann Arbor,
1948).
64 Ulli Beier, The talking drums 0/ the Yoruba, in A/riean Music, vol. 1 (1954)
No. I, p. 29.
65 George Herzog, Speeeh-melody and primitive musie, in The Musical Quarlerly,
vol. 20 (1934), p. 453. - Also A. M. Jones, A/riean musie, Livingstone, 1949, pp. 11 ff.
66 Joseph Kyagambiddwa, A/riean musie /rom the souree 0/ the Nile, New York,
1955, pp. 22 ff.
36
Speech, as a consequence, can often be understood without

words. In northern Transvaal, the Venda, who are not allowed
to speak during initiation ceremonies, communicate satisfactorily by whistling; and a similar expedient is reported from the
Isle of Gomera in the Canaries. 67
There is an almost exact analogy on European ground.
Scandinavian herdspeople substitute voices for horns and horns
for voices, and even the vocal melody "did not need a text to
be quite understood by people and cattle as a signal of command
and of recognition. 68
An inverse substitution is the so-called mouth music on the
Hebrides. 69 Bagpipes, as a symbol and stimulant of Scotch
nationalism, were prohibited for some time after the Scots had
risen in 1745 to throw off the English yoke. Their lilts and dances,
however, were imitated vocally and are sung to this day on the
Hebrides alongside with the redeemed pipes.
With all these substitutions we are in the midst of the important horn-, drum-, whistle-, and slit-drum languages in
Melanesia, Africa, and South America, all based on the realistic
imitation of speech-pitches and rhythms, not on conventional
Morse codes with symbols long and short. Of greatest consequence
are the slit-drum languages. Their renditions are particularly
faithful as the drum lips alongside the slit are bevelled off at
changing angles to render different pitches - up to six 70 - so
that the striking sticks can easily reproduce all the intricacies of
speech melodies. George Herzog has given us a detailed analysis
of a four-note horn-Ianguage in Liberia and asserts that on the
whole it renders actual speech tones; 71 and so does Herbert
67 Percival R. Kirby, Primitive music, in George Grove, Dictionary 01 music and
musicians, New York, 1955, vol. 6, pp. 921 ff. - Cf., as an oider source, Richard
Wallaschek, Primitive music, London 1893, and in German, as Anfnge der Musik,
Leipzig 1903, p. 33 in either ed. - For Gomera see: A. Classe, The Si/bo Gomero, in
Scientific American, 1956, and id., Phonetics of the Silbo Gomero in Archivum Linguisticum, vol. 9 (Glasgow, 1957), pp. 44 ff.
68 Carl-Allan Moberg, Gm valltar, in Svensk Tidskritt tr Musikforskning, vol. 37
(1955), pp. 28 ff., 94. - EFw FM 4008 (P 1008) II 6 (Norway). - Liv Greni, Ueber
Vok altradition in norwegischer Volksmusik, in Les Colloques de Wegimont, vol. 1
(1954-5) pp. 154 ff.
69 EFw FE 4430 (P 430) I 3.
70 Dorothy R. Gilbert, The lukumbi, a six-toned slit drum 01 the Batetela, in Alrican
Music, vol. 1 (1955) no. 2 pp. 21-23.
71 George Herzog, Speech melody and primitive music, in The Musical Quarterly,
vol. 20 (1934), pp. 454 ff.
37
Pepper for Central Africa. 72 Only Henri Labouret, who studied

the whistle languages of tribes to the north of the GuU of Guinea,
teils us how one of them comprised merely two pitches; the
signals, he thinks, are purely conventional. 73
This is in a way confirmed when Andre Schaeffner reports
from the same region that its drum language consists in formules
rhythmo-melodiques equivalent of words without any phonetic
connection with spoken language. 74 This wou1d bring at least
one kind of drum language into the neighborhood of the vocabulary of our military bugle languages.
The exchange between words and instrumental sounds can
even assume three phases: vocal - instrumental - vocal. In
Africa, it wou1d occur that at first the words of a message are
translated into slit-drum beats, that, however, in the momentary
absence of such an instrument its message is shouted - not in
the original, pre-instrumental words, but in syIlables, like ke, ki,
le, or li, which render the beats of the translation into the drum
message. 75
The Gulf of Guinea extends language imitation even to the
so-called musical bow - a simple bow like that of an archer,
whose string can be stopped at various places with a tiny stick
and struck with another stick, while an end of the wood is held
in the player's mouth to get audibility, resonance, and changing
pitches. One such bow piece, from the Bassa in Liberia, has been
recorded; 76 the story that the seemingly so barren single string
relates is no less than eighteen lines long.
A bow language existed in Hawaii as weIl. 77
To the mutual assimilations of vocal and instrumental melodies, of speech and of music, we could easily add the case of
Vedic cantiIlation in the liturgy of India, where "the transition
from speaking to singing, and back to language is so continuous
72 Herbert Pepper, Musique Centre-alricaine, in Encyclopedie Coloniale et Maritime,
vol. Alrique Equatoriale Franfaise (1950).
73 Henri Labouret, Les tribus du Rameau Lobt, Paris, 1931, pp. 195 ff.
74 Andre Schaeffner, Les Kissi, Paris, 1951, pp. 26-47.
75 For a fairly exhaustive bibliography of drum languages (up to Aug. 31St. 1958)
see ]aap Kunst, Ethnomusicology 3rd. ed., The Hague 1959, and its Supplement,
ibid. 1960, s.v. talking drums.
76 EFw FE 4465 (P 465) I 4. - Cf. also Schaeffner, l.c., p. 64.
77 Helen H. Roberts, Ancient Hawaiian music, no. 29 of the Bernice P. Biskop
Museum Bulletin, Honolulu, 1926, pp. 18 ff. - Edwin G. Burrows, Polynesian music
and dancing, in The Journal 01 the Polynesian Society, vol. 49 (1940), pp. 334 ff.
38
that adefinite line of demarcation cannot be drawn." The

quoted authority goes on: "The Hindu [Inder] have in their
speech enunciation a distinct trend toward musical intonation,
indeed a mdimentary recitative [Sprechgesang] - an impression
confirmed in personal dealings with Indians." 78
There are counterexamples, to be sure. In Amerindian melodies, the accents "do not always correspond to the accents in
the words of a song when spoken." 79 And modem Chinese
music can be so diametrically opposed to spoken language and
so little concemed with the tones of speech that theaters distribute the texts of their operas which otherwise wou1d be incomprehensible in their melodie disguise.
But even if there were no deviations from speech, what do
these speech inflections prove or, for that matter, Monteverdi's
stile recitativo e rappresentativo and Wagner's Sprechgesang?
Is there a proof that speech inflections preceded singing? And
if they did? Post hoc, ergo propter hoc, 'after this and hence on
account of it,' is the worst of arguing fallacies. Musical imitation
of language at any time does not evince a linguistic origin of music.
Even an original identity of music and language cannot be
accepted. Had it existed, the separation of the two could only
have happened in order to build up two very different forms of
utterance: an emotional idiom (singing) as distinct from a
communicative idiom (speaking). But, as far as we can judge,
a good deal of primitive music remains totally unemotional,
while language is often emotional, excited, and passionate.
Furthermore : birds sing, too, and therewith testify to the
existence of melody in the realm of animals ; but no animal has
an articu1ate language of vowels and consonants or words and
sentences beyond impetuous sounds like gmnting, barking, or
roaring. The sounds of mamals are quite unmelodious; and the
calls and twitters of birds have in their well-pattemed uniformity
hardly a linguistic quality: "language proper is cu1ture, and out
of the reach of the birds." 80 In the world antedating the age
78 Erwin Fe1ber, Die indische Musik der vedischen und der klassischen Zeit, in
Sitzungsberichte der Kais. Akademie der Wissenschaften in Wien, Phil.-Hist. Kl.,
vol. X52 (X9U), pp. 9, xo.

79 Frances Densmore, The study of Indian music, in Smithsonian Report for X94X,
Washington, x942, p. 544.
80 William Howells, Back of history, Garden City, x954, p. 56.
39
of man, the precursors of language and the precursors of music

are distinct1y separate. Why should man have unified the two,
only to sever them anew?
In the face of so many chance opinions without any proven
facts we are resigned to leave the puzzling question alone,
instead of presenting mere suggestions with the air of authority
- just as the linguists, too, have shelved the once so urgent
question of the origin of languages and abandoned the speculative reconstruction of presumptive ancestors, like Ur-Indoeuropean or aboriginal Semitic.
A chapter on origins must not overlook that early travel
literature occasionally reports the total absence of singing and
dancing from the lives of certain tribes. 81 Such statements might
all too readily be accepted as a proof, or at least as an indication,
that music and the dance developed only in later stages of culture.
Such accounts must however be read with a good amount of
scepticism. They are nowhere supported by facts, just as "early
reports of peoples lacking language or fire, morals or religion,
marriage or government, have been proved erroneous in every
instance." 82 At best, such statements could be due to cultural
shrinkage under adverse circumstances; but it is more probable
that they derive from the explorer's failure to hear any music or
see any dance while visiting the tribes. For unless the white man
from abroad has been accepted as a reliable friend, the primitives
often shy from exhibiting their sacred songs and dances ; rather
they pretend that they have none.
cannot be solved. The beginnings
of music are lost in the days of yore, as are the rudiments of
speech, religion, and the dance. All we can achieve is to follow
these manifestations back to the time when the curtain slowly
rose over the earliest act of mankind' s history .
Even here, the difficulties of arriving at an answer are obvious
and insurmountable. Music, fleeting and evanescent, has left
THE QUESTION OF ORIGIN
81 e.g. the Charriia in Uruguay and the Guarani in Argentina. Cf. Don Fe1ix de
Azara, Reisen in dem sdlichen Amel'ika, I78I-I80I, in ]oumal fr die neuesten
Land und Seereisen, vol. 6 (1813), pp. 77 ff., III. - Also two Brazilian tdbes, recently
discovered by Hans Becher (New York Times Aug. 7, 1958).
82 George P. Murdock, Universals of culture, 1946, reprinted in, and quoted from,
Hoebel, Jennings, Smith, Readings in anthl'opology, New York, 1955, p. 4.
40
no trace but those surviving in the tenacious traditions of living

people. To detect these traces and to exploit the hints they might
give, we must of necessity join the non-musical scholars who
wrestle with the thorny problems of culture in times that antedate history proper: the prehistorians and the anthropologists
or ethnologists.
works in a vertical way: he excavates the
tombs and fireplaces of ages long past - the deeper they lie the
older they are. By and !arge they divide the older parts of prehistory into three Stone Ages. The Older Stone Age or Palaeolithic
Era and the Newer Stone Age or Neolithic Era, with a connecting
Middle Stone Age or Mesolithic Era. The Older Stone Age may
have begun more than half a million years ago; the Mesolithic,
tentatively, twenty thousand years ago; and the Neolithic about
seventy-five hundred years before the present. But the dates
differ from region to region, and the whole stratification applies
in the first place to western Europe and not necessarily to other
countries and continents.B S Thus the prehistorian's chronology,
connecting the excavated strata in terms of time, is more or less
relative: he operates with the concepts 'before' and 'after'
without providing accurate years or centuries.
Recently, it is true, the so-called radio-carbon or (14 method,
developed in the United States by Willard F. Libby, Hans Suess,
Meyer Rubin, and others, in the Netherlands by Hessel de Vries,
has begun to provide absolute dates if the fossils are organic
matter and not older than 30,000 years.B 4 Since half of the
radioactive carbon 14 in a fossil decays at a rate of about 5,570
years, the amount still found provide the age of the fossil. The
unqualified reliability of the method has just been challenged
by Frederick E. Zeuner at the University of London: alkali
might wash out some of the carbon 14 and therewith cause the
illusion of greater age.Bs
Another modem method of approximate dating consists in
THE PREHISTORIAN
83 This chronology follows Garrod, Zeuner, and Braidwood, as quoted in Melville

J. Herskovits, Cuttural anth1'opology, New York, 1955, p. 38.
84 Willard F. Libby, Radiocarbon dating, Chicago, 1952. - Also: Robert F. Heizer,
in A. L. Kroeber, Anth1'opology today, Chicago, 1953, pp. 14-17, and in Hoebel,
Jennings, Smith, Readings in Antkropology, New York, 1955, pp. 54-58, with
bibliography.
,85 The New York Tirnes, Sunday, February 3, 1957, p. 15.
41
testing the amount of fluorine, a non-metallic element that

bones absorb during their underground stay.
The enormous spans before the use of metal are called the
Stone Age, as man, still ignorant of melting and molding metals
or else residing in regions without ores, used stone to shape his
weapons, tools and implements. In the Palaeolithic and Mesolithic, such stones were simply though carefully chipped; in the
Neolithic, they were elaborately polished. But this distinction
in which we once believed is no longer quite valid. Often, techniques overlapped; polished stone might still belong to an age
otherwise palaeolithic, and chipped stone survived in the
Neolithic. N or must it be forgotten that a stone age pattern of
culture could easily exist without much stone in evidence: in
tropical countries, other substances, like gourds, bamboo, and
coconut shells and, generally, bones and wood permitted to live
without the cumbersome chipping of stones. Bones excepted,
excavators will hardly find their traces; for objects so perishable
do not resist the rot of thousands and ten-thousands of years.
For this reason, archery, with its wood-and-fiber bows and slight
and slender shafts, had completely disappeared from palaeolithic
layers and was attributed to neolithic cultures exclusively, until
prehistorians found Old Stone Age arrowheads and saw the bows
unmistakably depicted on cave murals of the Upper Palaeolithic
and the Mesolithic.
The Neolithic revolution added houses, probably basketry,
agriculture, and the domestication of animals other than the
dog, who had already been tamed in the Palaeolithic. 86 Pottery,
at least in Africa, can no longer be credited to the Neolithic
Age; it is common in the Mesolithic times of that continent and
seems to have been invented in the Upper Palaeolithic.8 7
In due time, the potentialities of ore were discovered, and such
metals as copper, copper-tin bronze (about 3500 years ago), and
iron (about 2500 years aga) superseded primeval stone, bamboo,
and wood, and material civilization assumed historical forms.
But even in metal times, two classes often continued using stone:
the humbler layers of society, for reason of cheapness, and the
performers of religious rituals, for reasons of tradition.
86 But cf. Richard Thurnwald, Rep1'sentative Lebensbilde1' von Natu1'vlJlkem, vol.
of Die menschliche Gesellschaft, Berlin, I93I, p. 25.
87 Sonia Cole, The p1'ehist01'Y 01 East AI1'ica, New York, I954, pp. 26, I9I.
42
ANTHROPOLOGY ENTERS THE PICTURE at this point. For in the

coexistence of Stone and Metal Age in the same society, we face
the principal topic of cultural anthropology: the culturallag,
which even in our present time has caused essential parts of
humanity to live in conditions and forms that high civilizations
have left innumerably long ago. Thus the anthropologist gains
a synopsis of the various successive civilizations living alongside
on the present surfaee of the earth. What to the prehistorian is
a dim progression in hundred-thousands of years, is to the
anthropologist a lucid, live existenee side by side.
Indeed, in defiance of our gigantic advanee to the Iron, the
Electrie, and the Atomie Ages, tribal groups, retired to refuge
areas in impenetrable virgin forests, have to this day persisted
in the Stone Age' existenee that their aneestors might have lived
some ten or a hundred thousands years ago.
Thus we find palaeolithie survivals an over the world, in single
traits as wen as in almost the whole of eulture. Asian and
Afriean pygmies, the Botoeudos in eastern Brazil, most Fuegians,
and many other peoples ean be termed palaeolithic. They an
are food gatherers and nomadie hunters with blowguns or
4lITOWS; they use ehipped stone implements and, instead of huts,
proteetive roofs or wind-sereens; they go stark naked or nearly
so and do not praetiee farming or pottery.
Mesolithie and neolithic survivals are by far more frequent
than palaeolithie remainders.
It might be wise to make the present tense in an these statements tentative. Many of the authoritative monographs from
whieh we derive our knowledge of these peoples were written
half a eentury ago. In the meantime, mueh of the seclusion and
eultural retard found on the threshold of this eentury may have
eome to a partial or total end. This handy present tense is after
an nothing else but what Chapple and C. S. Coon onee ealled the
"ethnographie present." Indeed, "it is unlikely that a tribe will
behave in 1953 in the way in which an observer deseribed it in
1883 or even 192 3." 88
While writing these lines, I reeeived a letter from Georges
Condominas, diseoverer of the sounding stones in the Musee
de l'Homme in Paris. As to Madagasear, he says, I would be
88
J.
Manchip White, AnthropoIDgy, New York, 19S5, p. 171.
43
disappointed - "I have seen only few instruments you have

described; acculturation on the High Plateaux is very strong." 89
Rigid, immutable cultures do not exist; all civilizations have
been open to change and renovation, in the past as weIl as today.
1ntertribal contact was unavoidable; palaeolithic men were
nomads who under the pressure of changing climates or depleted
stocks of game would cover enormous territories and encroach
on grounds that other people claimed; many tribes demanded
exogamy or marriage outside of the group; interpenetration,
neighborly barter, warfare did the rest. Weapons, tools, and
implements as weIl as habits, thoughts, and melodies 90 were
freely exchanged, and most civilizations are so thoroughly and
differently mixed that every human group and even every single
man represents an individual mixture.
CULTURES HAVE CHANGED in the last half century, and with
the cultures, the two sciences in charge of them, anthropology
and prehistory. Much of the rigid dogmatism in their beginnings
is no longer valid, and the whole picture of development and
coexistence has grown more flexible. It is the normal trend that
a new generation of scholars oppose the views of their predecessors. For "in science as in life, it is a good practice to attach
from time to time a question mark to the facts one takes most
for granted, to question the fundamental postulates or facts
which require no demonstration; for a fact as a postulate is
largely the opinion of those who should know - and those who
should know are but human, and therefore liable to err." 91
This reaction seems to climax in the apodictic statement that
contemporary primitive societies "were preceded by other forms
of society of which we can learn nothing, even indirectly .. . we
can know only certain aspects of a vanished civilization." 92
This ultra-sceptical attitude betrays an assumption that the
elements of a civilization are unrelated and therefore not indicative of the whole; and such an assumption makes of culture
Paris, Dec. 2I, I956.
Cf., e.g. Jaap Kunst, A Study on Papuan Music (Weltevreden, I93I), pp. 3738; A. M. Jones, Ajl'ican music, Livingstone, I949, p. I6.
91 M. F. Ashley Montagu, Man's most dangerous myth, 3rd ed., New York, I952,
P33
92 Claude Levi-Strauss. Race and history, Paris, I952, pp. 8, I7.
89
90
44
history a mere kaleidoscope in which a number of colored bits

of glass unite in ever new, entirely haphazard combinations.
This would be regrettable. Less sceptical scholars will rest
satisfied with the conviction that certain forms of civilization
-like the lack of clothing, of huts, and of planting - are coherent
elements of a weIl defined cultural pattern and exclude potential
components of a contradictory quality. We do not need the total
number of bones to reconstruct the skeleton of an extinct mammal or, for that matter, of a culture.
And we should be builders, not wreckers.
In a wiser, up-to-date discussion of these problems, Melville
J. Herskovits makes the concession, so valuable from the pen
of a critical master: "Inference is a workable and useful tool if
held within its limitations." 93 And beyond the Atlantic, Richard
Thurnwald has concluded in complete agreement that, though
the facts at the bottom of comparisons are only similar and never
entirely identical, careful conclusions are possible and advisable. 94
ARCHAIC CULTURE AND MAN HlMSELF, the man whom prehistory
and anthropology tried to portray, have changed their valuation.
The naive and arrogant underestimation and misunderstanding
of these cultures and men created in the sixteenth century the
beginning colonial exploitation, which needed an alibi for its
bestial crimes by proving that the "natives" were at best
subhuman. It made me fee! ashamed for our time when I found
that as early as 1758 a French colonial officer who saw and
described the impressive burial customs of the N atchez Indians
in Louisiana protested against the brutal word savage. 95
'Primitive' men - as we usually call them today - live infinitely closer to unharnessed nature than do we; 96 and they live
without the help of organized, scientific thinking. Though firmly
guided by experience and oral tradition, they have no insight
93 Melville J. Herskovits, Cultural anthropology, New York, 1955, chapter 3
and p. 32.
94 Richard Thurnwald, o.c., vol. I, Reprsentative Lebensbilder von Naturvlkern,
Berlin, 1931, p. 88.
95 Le Page du Pratz, Histoire de la Louisiane, Paris, 1758, in John R. Swanton's
English translation reprinted in Hoebel, Jennings, Smith, Readings in Anthropology,
New York, 1955, p. 254.
96 Cf. the significant title of Richard Thurnwald's book Der Mensch geringer
Naturbeherrschung, Berlin, 1950.
45
into the complex connection of cause and effect. Still, no human

being can lead the life of a beast. No human being is devoid of
civilization; for it is just civilization that separates men from
animals. Even on his lowest perceptible level man speaks rich,
involved languages, devises elaborate weapons, tools, and
implements, has proto-religious ideas, organizes tightly knit
societies, and follows ethical mIes. For lack of a bett er name,
we may best call ab original man a primitive, and his civilization,
low. But the epithet low is definitely an undesirable discrimination. Even 'primitive' implies too much of an order of precedence to be to the taste of modern anthropology, and many
writers prefix it with an embarrassed, face-saving 'so-called.'
The term archaie migllt be preferable.
Some recent scholars have taken refuge in the appellations
scriptless and non-literate (as distinct from illiterates within a
literate society). But the prefix non- and the suffix -less, of
themselves correct, give even to these qualities a negative and
hence disparaging tint. A glance at the Middle Ages confirms our
hesitation. Charlemagne, resplendent king of the Franks and
emperor of the Holy Roman Empire, tried in vain to learn the
difficult art of writing. Indeed, all through the earlier Middle
Ages hardly anyone except a few monks and cleric-scribes was
able to read or to write: as late as the thirteenth century, the
knightly minnesinger Walter von der Vogelweide was expressly
praised for writing his verses in his own hand. Not to speak of
the appalling majorities of illiterat es in allegedly literate civilizations both in the East and the West. 97
With all these reservations, the importance of script as a
criterion of culture cannot easily be overrated. Originally devised
to carry the memory of kings and their feats beyond the lifespans
of men and dynasties, although to a certain degree also to keep
account ledgers in trade, it has more and more become a surrogate
of memory. We write, not only what we want to communicate
to other people for lack of oral contact, but also that which we
ourselves fear will be forgotten. Ancient nations chiseled monumental inscriptions in stone; we modems have notebooks and
calendars, as the complex character of modern civilization has
97 Cf. the terminology in Mieczyslaw Kolinski, The determinants 01 tonal construetion
in tribai musie, in The Musical Quarterly, vol. 43 (1957), pp. 50 ff.
46
forced us to disburden our memories. Writing has allowed us

to remember as weIl as to forget.
It is evident that the growing transfer from memory to reading
means objectivation, indeed, mechanization. As long as no
wrltten records interfere, the reproduction of something remembered involves production: the Homeric bard precedes the
prose reports of Herodotos; history is still a poet's epos, and Clio,
its Muse, stands on the Parnassos next to Erato, the patron of
lyrics. Reproduction from memory is rarely mechanical; it has
the prerogative of poetical license, reflects the personality of
the relator, is often imaginative, and sometimes presents us with
pieces of immortal creation.
If we justly replace 'relator' by 'singer,' we carry the contrast
of scriptless and literate into the musical field.
Scriptless societies have, as a matter of course, no way of
notating music; the oldest notations of which we have knowledge
- Babylonian, early Jewish, Vedic, Greek, and others - draw
their signs from the symbols of current alphabets. But the
sentence cannot be inverted: music without notation is not
limited to scriptless societies. Many ancient notations were
merely devised by priests for priests and cantors, and some
were even kept seeret: "The initiated may show it to the uninitiated." 98 While in religious musie notation had adefinite plaee
in order to prevent the present and future generations from
breaking sacred traditions, seeular music relied on free invention
and memory, in the high civilization of the West as well as in
those of the East. Notation beeame indispensable only under the
pressure of worked-out polyphony. Still, in and outside the
polyphonie style the seript was sketchy even in Europe; the
chord-improvising 'realizers' of figured basses during the Baroque
Age were just given a fundamental bass-part to build upon to
the best of their skill and imagination; and the melody singers
and players of the same era performed from a skeletal seript
whieh left to everyone the right and the duty to embellish and
vary the seript at his diseretion or, in the words of the Flemish
musician Adriaen Petit Coelieus (1552), to spice the meat with
salt and mustard. Improvisation on the spur of the moment
98 Curt Sachs, The mystery 01 the Babylonian notation, in The Musical Quarterly,
vol. 27 (1941), pp. 62-9.
47
reached into the nineteenth century, and even today a scriptless

performance is, at least in principle, expected in the cadenza of
a concerto. But the sparse survivals of improvisation practice
do not change the vital fact that western art music is literate:
extant originals bind the performer and force him into obedience.
An actually scriptless tradition lives only in folkmusic. No
composer writes it down to have it played; but, inversely, a
notation that we may find, is taken from live performance; the
melodies follow oral tradition and are at the same time open to
wide individual and regional changes.
Beyond its musical scriptlessness, folk music on the whole
cannot be easily reconciled with generalliteracy. With exceptions,
it is most authentie, best, and richest in times and regions where
illiteracy prevails. Evidence are the South of the States and the
South and the East of Europe. This is easy enough to explain:
the backbone of a scriptless culture are tradition and memory;
both vanish under the impact of generalliteracy, and with them
fade the imagination and creativeness of uneducated performers.
Literacy and folk art bloom in inverse ratio.
THE FUNDAMENTAL PROBLEMS of human early history, it
appears, are equally less in the foreground of our thinking and
hence less rigidly answered. The basic alternative of monogenism
and polygenism, that is, the opposite theories that all mankind
has descended from one single ancestral type or else from several
of them, has for the moment not the importance it was given in
the nineteenth century. The question in how far cultural elements
spread in 'diffusion' from one people to other through marriage,
war, and trade, or developed by 'evolution' or innate disposition,
finds individual answers rather than dogmatic decisions and
sweeping generalizations. And a worldwide organization of
culture patterns, as in the German Kulturkreis-Lehre, is met with
admiration, but also.With scepticism. Most suspicious are we of
the slippery and often criminally exploited concept of 'race,' in
which biologically transmitted genes are all the time confused
with environmental culture. 99 "The word 'race' has been so
99 The venerable sire of a none too brilliant family of musical racists is Fran<;lois
]oseph Fetis, Sur un nouveau mode de classi/ication des races kumaines d'apres leurs
systemes musicaux, in Bulletins de la SocietB d'Antkropologie, n.s., vol. 2, Paris, 1867.
48
abused that it has become at onee meaningless and too full of

meaning." 100
All these coneepts will be rarely found in this book - not from
want of daring on the author's part, but from a hardwon insight
into the unbelievably eomplicated and treacherous web of
ever-ehanging cultures, tribal nomadism, and unidentifiable
mergers of patterns. Besides, musicologists, as spokesmen of a
small, however significant component of eulture, can at best be
expert witnesses in this lingering lawsuit but hardly judges or
jurors.
100
J.
E. Manchip White, Anthropology, New York, 1955, p. 73.
TWO
EARLY MUSIC
v.
THE
OLDEST
MUSIC:
TUMBLING
STRAINS
Turning to music proper, we face the old trouble of all musical

writing: descriptions of musical works and styles are hard to
put in words, even harder to read and absorb, and almost
impossible to translate into actual images. Yet, the issue cannot
be avoided and the attempt must be made. But the reader
should recur to good recordings as often as possible.
In describing non-western music, be it oriental or primitive,
one must strictly refrain from misusing incongruous coneepts of
western music. The terminology that has been learned in music
school applies to a harmonie structure of music and is inappropriate, indeed misleading and distorting in deseriptions of nonharmonic, non-western music.
Most unjustifiable is placing the current labels 'major' and
'minor' on non-western melodies, according to whether the note
the deseriber thinks more or less rightfully to be the third is at
a major or minor distance from the alleged 'tonic.'! How does
he act when the third, as so often, is larger than the minor third
of our pianos or else sensibly shorter than the major third, or,
as a so-ealled neutral third, just in the middle between them?
Is that Dirge of the Eagles in northern British Columbia 2
'major' or 'minor'? The note c, in this case the third without a
doubt, is in Marius Barbeau's transcription now natural, now
slightly heightened by a little + sign, now even more sharped by
a bracketed sharp over a plus, now fully sharped. In other words,
the third is diseoncertingly vague. And this is not the only
example of thirds that refuse to be crammed into our major1 As a shocking example, I recommend the reading of the Minor and Major chapter
in Richard Wallaschek, Primitive music (London 1893) and Anfnge der Musik
(Leipzig 1903).
2 Marius Barbeau, Asiatic survivals in Indian song, in The Musical Quarterly,
vol. 20 (1934), p. III.
50
EARLY MUSIC
minor dualism. Our 'blue' notes pose a similar problem. Moreover, major and minor are almost inseparable from a harmonie
system foreign to the primitives and apt to misrepresent their
musicallanguage. To understand this idiom, follow the advice
of Wagner's Hans Sachs:
Der eignen Spur vergessend,

Such davon erst die Regeln auf.
(Away from course and custom,
Do find the standards of its own).
Still worse than simply dubbing primitive melodies 'major'
or 'minor' is the amateurish habit of conferring ready-made
moods upon them in line with the unfounded belief of the West
that major must be gay and minor sad. Small wonder that on
reading such abuses of Affektenlehre, pseudo-aesthetics, and
Music Appreciation, one is under the impression that nine tenths
or more of all the world's humanity are deeply melancholy.
As a rule, such lamentable sadness is ascribed to a depressing
environment. But I am afraid that this environmental depression
applies to the visitor from Europe much more than to the wellacclimatized natives of the endless steppes and llanos. Religion,
too, is not entirely absent from the causes of sadness. Many
decades ago, I read in the dictionary of some East African language, compiled by a French missionary - the author's name and
book unfortunately long forgotten - that all the native tunes
were in minor because the joyousness of major is denied to everybody who does not believe in the Christian God. I refrain from
commenting and leave the answer to those familiar with the
Gregorian chant.
Incidentally, not even in the West is minor a compulsory
mode of mourning, and vice-versa. Many definitely mournful
pieces are written in the major mode - at random: Lasso's
Tragico tecti syrmate and Tristis est anima mea, Hans Leo
Hassler's later hymnalized Innsbruck ich muss dich lassen, Marin
Marais' beautiful Tombeau pour Lully for viol and harpsichord,
a Plainte by Fran<;ois Couperin le Grand, the dead march in
Handel's oratorio Saul, and, in the nineteenth century, Silcher's
popular melody for Heinrich Heine's Lorelei.
EARLY MUSIC
51
THE OLDEST PIECES of music, it must be emphasized, are

purely vocal and, hence, pure melodies.
Since the concept of melody is far from being unambiguous,
it might here be defined as the audible movement of a singing
voice (or, much later, of an instrument) from the beginning of a
piece through all successive steps to its end. But beyond the
lifeless sum total of its individual steps, we take for granted that
such a movement be an organic, living whole with breath and
flow, with tension and relaxation. It is not necessarily 'melodious'
or sweet-sounding, as the opera fans demand and the dictionaries
set forth. (Whether or not the metrical values of the notes in a
melody should enter the definition is irrelevant as long as we
limit our discussion to primitive music).
But even on the earliest level, a melody is never anarchie or
arbitrary. It folIows certain, almost unbreakable rules.
As a consequence of this lawfulness, the surviving tribes of
palaeolithie culture use two dear-cut styles side by side; and
they do it without geographie divide, in whatever part of the
world they might have settled.
The most fascinating of the oldest melody patterns may be
described as a 'tumbling strain. ' I ts character is wild and violent:
after a leap up to the highest avallable note in screaming fortissimo, the voiee rattles down by jumps or steps or glides to a
pianissimo respite on a couple of the lowest, almost inaudible
notes; then, in a mighty leap, it resumes the highest note to
repeat this cascade as often as necessary. In their most emotional
and least 'melodious' form, such strains recall nearly inhuman,
savage shouts of joy or walls of rage and may derive from such
unbridled outbursts.
The crudest style of this kind of melody seems to be preserved
in Australia. Midway between howling and singing, it is described
by an ear witness as performed with "frenzy" and "spasm" in a
"steadily growing excitement" and with a "good deal of
passion." 3
Today, we possess an easily accessible North Australian
8 E. Harold Davies, Aboriginal songs 0/ Gentral and Southern Austrialia, in Oceania,
vol. 2 (1932), pp. 454-467. - Cf. also Trevor A. Jones, Arnhem IAnd music in Oceania,
vol. 26 (1956), p. 336, republished in A. P. Elkin and Trevor A. Jones, Arnhemland
Music (Oceania Monograph No. 9), Sydney, 1957.
52
EARLY MUSIC
recording,4 in which the soloist starts his cascade in a fortissimo

falsetto on elf while several voices throw in the lower third e".
After a long sostenuto, he reaches up for g" and tumbles in an
audacious curve to a final c', as shownin the following illustration:
Ex. 3*
"
This music carries meaningless syllabies, but no words; it is

purely emotional and keeps completely independent of the
definiteness of words.
While such fierce haphazard cascades can hardly be transcribed in the neat notation of the West, others, in almost every
part of the globe, are less turbulent and at a pinch accessible
to our five-line staffs. N owhere are these tamer 'stair melodies'
more impressive than among the N orth American Indians - in
the Northwest as well as in the Southwest. 5 The most beautiful
of them, full of noble pathos and powerful passion and yet
restrained and solemn, is of particular historical interest, as it
belongs to the earliest set of melodies ever recorded by a phonograph. 6 It was sung by the Zum in New Mexico and spans the
amazing range of more than two octaves in an essentially tertial
organization.
Ex. 4
It also was very probably a tumbling strain when in the

eighteenthcentury women of the now extinct Abip6n in Argen*
Transer. C. S.
EFw FE 4439 (P 439) II 3.
5 Cf. e.g. Garfield, Wingert, Barbeau, The Tsimshian, New York, 1952, example
no. 64. - Also David P. McAllester, Enemy Way Music, Cambridge (l\fass.), 1954, no. 14.
6 Benjamin Ives Gilman, Zuni melodies, in ]oumal of American Arckaeologyand
Ethnology, vol. I (1891). - Carl Stumpf, Phonographi(e)rte Indianermelodien, in
Vierteljahrsschrift {Ur Musikwissenschaft, vol. 8 (1892), pp. 134 ff., and in Sammelbnde far vergleichende Musikwissenschaft, v.ol. I (1922), p. 120.
4
EARLY MUSle
53
tina, they too lamenting their dead, toppled down from the
highest to the lowest note and did it with whistling hisses (ein
pfeifendes Gezische).7
The Indians may have obtained their tumbling strains from
Asia via the Bering Strait: Marius Barbeau found a J apanese
tune closely resembling some of the Yukon and of northern
British Columbia. At the beginning, he says, it "scaled a high
curve, touched a top note, then dropped over wide intervals to
the bottom, where it droned leisurely just as do the tunes of a
number of typical Indian songs." Unfortunately, Barbeau gives
no notation of the J apanese melody; 8 but his description is
unmistakable and would be fully convincing if it were supported
by more and better evidences.
A similar example, from a Japanese no play,9 looks more
convincing. Anyway, the c10sely related folksong of Korea has
definitely tumbling strains. 10
The instrumental music of Japan as we know it today has
occasionally a remarkable kinship to vocal tumbling strains.
The following piece of theatrical music, played on the popular
three-stringed lute or samisen,ll descends in all its phrases and
from the depth leaps back to the upper octave or even the
double octave eH.
Ex. 5
,J=r I cCa ffl

p
Altogether, the majestic strains of the Indians remind us of

the melodies we hear when Polynesians wall for their dead.
Excitedly, the voice would leap up no less than two octaves and
reach the three-lined c sharp.12
Martin Dobrizhoffer, Geschichte der Abiponer, vol. 2, Wien, 1783, p. 367.
Marius Baxbeau, Asiatic survivals in Indian song, in The Musical Quarterly,
vol. 20 (1934), p. 108.
e Constantin Brailoiu, Sur une melodie russe in Pierre Souvtchinsky (ed.), Musique
Russe, 11, Paris, 1953, p. 380, no. 176.
10 EFw FE 4424 (P 424) I 6.
11 From: O. Abraham and E. von Hornbostel, Tonsystem und Musik der Japaner,
in Sammelbnde der Internatwnalen Musikgesellschatt, vol. 4 (1903), p. 349 no. 2,
and in Sammelbnde "Ur vergleichende Musikwissenschaft, vol. 1 (1922), p. 222 no. 2.
lD E. G. Burrows, Songs 01 U vea and Futuna, in Bernice P. BishoP Museum Bulletin,
vol. 183, Honolulu, 1945, passim.
7
54
EARLY MUSIC
Edwin G. Burrows heard quite similar songs in 1947 on

Ifaluk Atoll in the Caroline Islands. Mourners were sitting by
a dying man. The wailing, while not regular, followed a general
pattern. "Starting on a high note (how high varies from a shrill
shriek to a medium tone), the voice descends in a whine, portamento, to a low pitch, repeated with slow, equal time value. Then
up a minor third for oue tone, down again for one or more. This
rise and fall of a minor third sometimes repeated. Sometimes
all sang in unison. Again one group sang low, another
higher.
Several times, when I tested the interval between them, it was a
perfect fourth, or nearly enough to sound so to me. The interval
from the first high note to the low monotone varied, but was
often an octave or very near it. The last tone often broke into a
shuddering sob. At the end of some phrases, a short wall, one or
a few notes. " 13
On another occasion (January 15, 1948): "Each line of the
song - tliey are known songs, not improvisations, for all the
people sang the same words - was sung mainly on a monotone,
with the occasional rise of a minor third noted earlier. Sometimes
the rise amounted to a major third. Between each two lines was
a long cry, without definite pitch. There was great variation in
these. Some could be called shouts, other howls, still other
shrieks. Then the voice would drag down, portamento, to
the pitch on which the lines were being intoned. There
was some singing in parallel parts. Noted major third and
fourth."14
A PERIODIC, AUDACIOUS RECAPTURE of the higher octave is
the essential, identifying trait of tumbling strains, even on a
level as primitive as that of Australia and Torres Straits, not to
enumerate later habitats. 15 This is the more remarkable, as
outside the tumbling strains the octave is by no means a selfevident interval in primitive music.
13
id., M usic on I faluk Atoll in the Caroline I slands, in Ethnomusicology, vol. 2, (1958),
p.IO.
ib., p. 13.
SampIes in EFw FE 4428 (P 428) I 3, 6 (Watutsi); FE 4500 (P 500) I I (Zulu);
FE 4500 (P 500) B I 15, 11 '2, FE 4432 (P 432) I 6 (Haiti and Trinidad); FE 4401
(P 401) I I, 3 b (Sioux); FE 4401 (P 401) 11 2 (Navaho); FE 4445 (P 445) I 2, 7
(Flathead).
14
15
EARLY MUSIC
55
Ex. 6*
From the octave as the primary concept, tumbling strains

developed by inner consolidation and not by accretion and
expansion, as did the horizontal melodies to be discussed in the
following section. The beginnings of this process were fierce and
unruly, with the stepping stones inside the octave left to chance
without an idea of scales and steps. Little by little the picturesque
wildness yielded to asolid organization in recurrent distances,
with a preference to either fifths or fourths or thirds.
TUMBLING STRAINS, with the forceful intonation of the highest
notes and the subsequent flagging on a melodie descent, give
an impressive example of physiologically conditioned changes
in intensity. Indeed, everyone of our own singers and wind
players knows the difficulties of intoning a high note softly and
a low note with force.
It is tempting to attribute to such correlation of pitch and
intensity the double meaning of the words high and low in quite
a number of languages. Whoever leams French is told to read
a haute voix, that is, aloud; and in a similar way our Italian
teacher might 'recite his Dante a voce alta.1 6 In late medieval
France, les instruments hauts, as trumpets, drums, and shawms,
were penetrating and served outdoors; les instruments bas, as
strings and flutes, were soft enough for indoor use. And to find
our way back to the archaie level: the Bellacoola Indians in
British Columbia have likewise one word for strong and high,
and one for soft and 10w. 17 In the Pacific, the Hawaiians have
Watutsi, after Rose Brandel.

High and low seem to be concepts of the West. The Bantu speak of small and
great (Hugh Tracey, The state of music in Bantu Af1'ica, in African Music, vol. I
(1954), p. 9), as weIl as the Javanese (Jaap Kunst, Music in Java, The Hague 1949,
vol. I p. 101). On the inversion of high and low in ancient Greece and the Semitic
languages, cf. Curt Sachs, The 1'ise of music in the ancient world, New York, 1943,
pp. 69 ff.
17 C. Stumpf, Lieder der Bellakula-Indianer, in Vierteljahrsschrift fr Musikwissenschaft, vol. 2 (1886), p. 410, reprinted in Sammelbnde fr vergleichende Musikwissenschaft, vol. I (1922), p. 92.
16
56
EARLY MUSIC
exactly the same ambiguity. But when their voiees "go high and
low," they sing crescendo and diminuendo without any change
of pitch within their one-note cantillation (cf. example 23 on p.
70).18 Their shade of in~ensity is emotionally conditioned, not
physiologica11y.
Otherwise, changes in intensity do not often occur in primitive
musie.
THE EVIDENTLY RAPID DOMESTICATION of tumbling strains has
forced us to discuss in the same part on Early Musie types of greater
and of lesser age. In the interest of coherence, even more recent
forms of tumbling strains should be outlined in this chapter, though
they no longer are part of the lowest observable level of musie.
As a rule, they cling to the old characteristlcs: that the octave
stands out as the back,bone of the skeleton; that again and again the
melody leaps up to recapture the higher octave; and that within the
octave certain descending intervals become established landings.
This establishment developed in two directions: a triadic or
tertial and a tetrachordic or quartal pattern. The triadic pattern
shall be our first topic.
A chronology of triadic octaves in the primitive world is not
yet possible. Such melodies occur in every possible shade. Some
appear in root and some in six-four position (to borrow terms
from harmony). This amounts, in the terminology of scale
structures, to 'authentie' and 'plagal' or, as I suggested in an
earlier book, to fourth-over-fifth and to fifth-over-fourth melodies. 19 In either case we meet, now empty skeletons, now fully
or partly heptatonic and diatonic structures.
N eat examples of tumbling strains organically transformed
into triadic octaves are found in a recorded song of the Mandingo
in Liberia. 20 and the following song of the Vogul, a Finnish
people on the northern end of the divide between Europe and Asia.
Ex. 7*
After Visnen.
Helen H. Roberts, Ancient Hawaiian music, Honolulu, 1925.
19 Curt Sachs, The rise 0/ music in the ancient world, New York, 1943, p. 65.
20 EFw FE 4465 (P 465) I I.
18
EARLY MUSIC
57
The skeleton is empty but for an occasional 'infix' or filler 21

in the thirds. One would hardly suspect this 'civilized,' western
melody to descend from tumbling strains - as it stands on paper,
it could easily introduce some symphonie movement of Brahms.
And yet it reminds us remotely of wild and passionate walls.
For it has the unmistakable leap up to the top of the range, the
cascade down to the lowest note and, again and again, the
recapture of the upper octave and arepetition of the whole pattern.
The same traits occur, among many others, in melodies of
the Navaho 22 and also of the Teton Sioux 23 which however
have infixes and plagal structures, with the triad above and the
tetrachord below.
It seems that tumbling strains developed even into formations
of a mere sixth instead of the habitual octave, as in the following
example from Ruanda in East Africa: 24
Ex. 8
Indeed, on seeing Junod's old notation of a Baronga tune as

he heard it, we are irresistibly tempted to call it a tumbling strain,
although it has only four descending notes in the narrow ambitus
of a fifth. 25 Beagle Bay Australians 26 and the Wanyamwezi in
Central East Africa,27 on the other hand, leap to the upper ninth
or tenth:
Ex. 9
21
22
1954
Cf. Sachs, l.c., p. 37.

Musical examples in David P. McAllester, Enemy Way music, Cambridge (Mass.),
23 Frances Densmore, Teton Sioux music, in Bureau of American Ethnology

Bulletin 61. Washington 1918 no. 43.
24 Erich M. von Hornbostel, Gesnge aus Ruanda, in Wissenschaftliche Ergebnisse
der deutschen ZentralAfrika-Expedition I907-08, vol. 6 I (1917), ex. 8.
25 Henri-A. Junod, Les Ba-Ronga, Neuchtel, 1898, p. 146.
26 Carl Stumpf, Die Anfnge der Musik, Leipzig, 19II, p. 122.
27 Erich M. von Hornbostel, Wanyamwezi-Gesnge, in Anthropos, vol. 4 (1909),
eX5
58
EARLY MUSIC
At the end of this long migration, we meet a familiar melody

from ancient Greece, the skolion or drinking song of the GraecoSicilian composer known as Seikilos : 28
Ex.
10
"., J (' F"
mr
1.1' r
mr J r J I
The charming piece has the well-stressed range of an octave

in fifth-over-fourth arrangement. In Greek terminology, its mode
is Phrygian; counterpoint nomenclature would call it Hypomixolydian. The lower fourth is pentatonic, the upper fifth,
heptatonic. Like a genuine tumbling strain, the melody leaps from
the mese or tonal center up to the highest note and cascades
down to the final note an octave below. Eleven of the twelve
measures move within the upper fifth; and only the last measure
descends to the lower tetrachord. Strophic repetition, very
probable in so short a melody though, as a matter of course, not
carved in the tomb stete that has preserved the tune, would provide the recapture of the highest note by way of the mese.
Around and behind this delightful song stand the untold
thousands of fifth-over-fourth melodies created between India
andEurope.
OPPOSED TO TRIADIC OCTAVES are the octaves formed by two
conjunct or disjunct tetrachords. These terms as well as a conjunct example of the Iroquois Indians with an appended octave
at the bottom will be discussed in the second-part section on
The Fate of Quartal and Quintal Patterns. Similar melodies are
sung by the Pawnee Indians. 29
The following, disjunct example from Croatia betrays the
derivation from pristine tumbling strains in its definite descent
and its recapture of the higher octave:
Printed in Curt Sachs, The rise 01 music in the ancient world, 1943, p. 245.
Tracy's transcription in Allce Fletcher, The Hako ceremony, Washington,
1904, p. 103
S8
S9
59
EARLY MUSIC
Ex. u*
For once, we meet with the rare case of a development from

the most ab original types to modem forms through a steadily
growing organization from within and through interpenetration
with other structural patterns.
VI. THE OLDEST MUSIC: ONE-STEP MELODIES
Side by side with tumbling strains, we meet on the lowest

cultural level an. apparently less emotional type of melody,
which in its most rudimentary form consists of only two pitches
sung in alternation. The voice moves up and down and more or
less describes a horizontal zigzag line. 30
The words alternation and zigzag might easily suggest a
regular pendulum progression to and fro. Such shuttle motion
does not exist, unless we think of a trill at moderate speed.
Sometimes the upper and sometimes the lower note is struck
more often and assurnes the role of a principal, starter or final.
This makes the melody either 'hanging' or 'standing' :
Ex. I2**
'a
8
V tr
I Er CJ I F
Ex. I3***
, ; J VJ J
*
**
n ; J J ) J n ;;
<J
After Wiora.
East Flores, after Jaap Kunst.
*** Eskimo, after Zygmunt Estreicher.
30 Cf. Zygmunt Estreicher, Teoria dwytonowych melodii, in Kwartalnik Muzycky
VI (Warschau, 1948). - General on form: Mieczyslaw Kolinski, The structure 0/
melodie movement, a new method 01 analysis, in Misceldnea de Estudios dedicados al
Dr. Fernando OrUz, La Habana, 1956, pp. 881-918.
60
EARLY MUSIC
Ex. 14*
One note - as in example I4 - might be so predominant in

importance and frequency, and the other note so subordinate,
that we are tempted to speak of an appogiattura in the modem
sense. Alternation, after all, demamis a certain balance of the
two notes, both in weight and in quantity.
A simple zigzag - up and down or down and up - might readily
be labeled monotonous. In doing so, we use 'monotonous' in
its current figurative sense but risk having it understood in the
literal sense as a composite of monos, 'one' and tonos, 'tone.'
With this ambiguity, the reader is caught in the cobweb of
terminology and cannot be spared a few minutes of semantics
for the sake of clarity and consistency.
SOME RECENT ETHNOMUSICOLOGICAL WRlTINGS name the to and
fro of two notes, not monotonie, but ditonic, which, de-Grecized,
means 'of two tones.' Accordingly, they call the aggregates of
three and more notes tritonic, tetratonic, and pentatonie. This
is inadmissible or at least impractical and dangerously misleading.
The word ditonic is illicit for two reasons. Phonetically,
ditonie differs from the everyday term diatonic merely by the
absence of an a, which is in English pronunciation almost
inaudible after i. Moreover, not many readers might realize
that the first word derives from di, 'two,' and the second, from
dia, 'through.'
The component 'tonic' is even worse, since, instead of one,
it has no less than six different and even contradictory meanings.
(I) In the form tonic, an adjective used as a noun, it stands for
the main, gravitational pole of a harmonized or harmonizable
melody. The original Greek noun, tonos (and hence, via Latin
and Old French, our 'tone') is related to 'tension' and means
(2) acoustically speaking, any regular sound as opposed to
A women's dance from Ifaluk, Carolines, after George Herzog.
EARLY MUSIC
61
irregular noises ; (3) the pitch, vibration number, or frequency

of such asound, say C or C sharp; (4) its color or timbre, warm or
cool; (5) a me10dy pattern (like 'psalm-tone'); and (6) the distance
or interval of a major second. There is a degree of danger in
handling a word with six different meanings. In composition
with di, it may become outright destructive; for in its Grecian
homeland, as well as in the western Middle Ages, ditonos meant
logically twice the distance of a major second and was the accepted term for the major third. Thus nobody should confuse
the issue by re-introducing a well established term for something
entirely different and even contradictory.
'Tritonic,' again, must not be used for a melody of three notes.
In musical terminology, it has a place of long standing as the
uncontested name for a succession of three wholetone steps,
such as F GAB or, without the intermediaries, the stretch
F-B, both in succession or as a dissonance.
The word pentatonic, on the contrary, is universally accepted
to denote a scale (or me1ody) of only five steps in the octave mostly, though by no means always, three seconds and two thirds,
as in the Far East or in the Celtic countries.
We will studiously avoid such 'ambiguous' words, to make
an understatement. Terminology is meant to c1arify, not to
confuse.
The concept that we need is 'step.' Its meaning is quite unequivocal; and, strictly speaking, no melody has 'tones' or
'notes' ; as a form of musical movement, it is aseries of steps (or
even of leaps) while the notes are simply stations and terminals.
All considered, we had much better call the primitive patterns
one-step, two-step, three-step melodies, even if such terms, not
being Greek, are somewhat less impressive.
To use at least a Latin-derived word, we could hardly denote
the approximate size of the step as a secondal one-step melody,
when the two altern~tting notes are a major or minor second
apart; in a tertial one-step melody, a major, neutral, or minor
third; in a quartal melody, a, usually descending, fourth; in a
quintal melody, a fifth. Later on, after having met more complicated melodies, it will be easier to speak of single seconds, thirds,
fourths, and fifths, as opposed to double, tripie, quadrupie, or
quintuple such intervals.
62
EARLY MUSIC
WE DO NOT KNOW what factor determines the size of a tribal

group's characteristic step. It is certainly not the degree of
narrow-mindedness that early scholars ascribed to primitives:
the narrower the mind, the narrower the step. This was an
illicit equation. People do not stick to shallow secondal patterns
because they stand on the lowest rung of the culturalladder.
Nor are the fourths and fifths reserved for bolder climbers. Often,
a person's sex appears to be the shaping power; women seem to
prefer a smaller step, just as they do in dancing, while men
proceed in larger strides and leaps. Here and there, the customary motor habits of a tribe, so often conditioned by its way of
life, might play a decisive role; or it might be the geologie
formation of the habitat, as flat or mountainous, wide-open or
enc1osed; and, occasionally, even its climate. But whenever a
satisfactory quantity of evidence seems to speak for one of these
possibilities, there are always discrediting counterexamples which
cannot be identified as justifiable exceptions. Thus, if !ittle
Eskimo girls sing empty fourths in unison, 31 this very archaie,
female, and, in fact, infantile music proves that wide steps are
not necessarily dependent on advanced mentality or on sex:
Ex. 15*
f~J"I28
fit! I D,) Inn n InJDJIDA In}
Indeed, adult Eskimo women might sing in fifths and at times
overdo them to form a kind of minor sixth. 32
The opposite pole of the globe presents another disconcerting
example of empty fourths: Fuegian medicinemen sing them in
their rituals: 33
.. Transcr, C. S .
After von Hornbostel.
81 EFw FE 4444 (P 444) 11 I.
32 Z. Estreicher, Cinq chants des Esquimaux Ahearmiut, in Geert van den Steenhoven, Research-reporl on Caribou Eskimo law, La Haye, 195688 Erich M. von Hornbostel, The music 01 the Fuegians, in Ethnos (1948, posth.),
p. 102, first staff_
EARLY MUSIC
63
while the most characteristic and most frequent pattern of the

Patagonians is strait and secondal. Add western Polynesia to
these antipodes.34 Here, too, we have no convincing explanation
to offer.
A strict non liquet applies to the realm of secondal melodies.
Their distribution is truly universal, and special conditions
- as sex, climate, and others - are not apparent. At best, the
seconds are so frequent in the very lowest civilizations that we
must assign them to the earliest known societies of mankind,
without siding in the dispute of the monogeneticists and the
polygenetidsts for or against ~ one-home origin of man.
The situation is not much clearer in the realm of the third.
Tertial patterns, too, are most universal; but the map of distribution shows a particular denseness in Negro Africa, which in
turn is nearly devoid of fourths. Indeed, the third appears to
have almost the monopoly in Black Africa. Is this a case of
radal condition? I doubt it. The fourth is equally absent from
Europe, excepting much of the East and the Southeast, while
the third is all dominating. Again, as far as our knowledge goes,
the Asiatic world has a tertial N orth and a preponderantly quartal
South, from Tibet all the way down to Indonesia. Thus we must
face the possibility that an originally tertial region, the whole
Old World of Asia, Europe, and Africa, was in a remote past cut
into two by an East-bom wedge of quartal melody.3 5
In Indian America, tertial and quartal melodies live alongside
without betraying any clue to a convincing chronology or
stratification.
This section deals with single steps only, leaving multiple
steps to the Second Part.
SINGLE STEPS from note to note remain very often unchanged
within a melody or even in all the melodies of a tribe. But, as a
matter of course, the listener should not expect that the steps
we call seconds, thirds or fourths be identical with those of our_
pianos. They have a considerable latitude 36 and might even run
34 E. G. Burrows, Native music of the Tuamotus, in Bernice P. Bishop Museum
Bulletin 109, Honolulu, 1933, p. 74.
35 Curt Sachs, The road to major, in The Musical Quarterly, vol. 29 (I943), pp. 38I404. - Also id., Towards aprehistory of music, in The Musical Quarterly, vol. 24
(I938), p. I47 ff.

36 Cf. e.g. Hugh Tracey, The state
vol. I (I954) p. 9.
0/
folk music in Bantu Africa, in African Music,
64
EARLY MUSIC
into one another so that the transcriber is uncertain whether to

rate some step as a second or as a third, as a third or a fourth. 37
A good transcription will indicate the cent number of the
distance and herewith remove any doubt.
The situation worsens when for the sake of expressiveness
good singers (much more than poor ones) expand or shrink the
steps on purpose. This is particularly true of the Amerindians,
who have no scale-wise tuned instruments and hence no standardized interval. At times, they rise up to a major third above the
original pitch; and they also would, in 'vertical augmentation,'
increase the step as such. 38 We know similar techniques from
Japan, where a flutist is expected to drive the pitch up after
attacking correctly. Yemenite Jews as weil as Persian Jews
sharpen the notes the more their frenzy increases. 39
EMPTY STRUCTURES might be the right term for the one-step
patterns discussed so far. There are just the two structural notes
a second, a third, a fourth, or a fifth apart. But the emptiness
of these patterns can be, and often is, disturbed by additional,
auxiliary notes that some richer imagination or a different
motor habit impose.
As long as the nuc1eus is c1early recognizable as such, any
additional note in the melody is an affix if it joins the nuc1eus
outside 40 and, when the specification is necessary, more distinctly a supra/ix added above or an in/ra/ix added below.
Secondal melodies can be adorned with affixes only. Tertial,
quartal, and quintal patterns, on the contrary, have had a
growing tendency towards resolution into smailer steps. The
filling notes within a third, a fourth, or a fifth are called infixes.
Ex. 17*
&HCU I r ~ I CO I r ~ I m
*
rrr d
nl.lc/evs
111"1fxra.
F
In.
~x
Lithuanian daina, after Wiora.

Cf. EFw FE 4446 (P 446) (Mato Grosso) II I.
38 Cf. David McAllester in Journal of the American Musicological Society, vol. IO
(I957), p. 45
39 Curt Sachs, The rise of music in the ancient world,t New York, I943, p. 81,
after Idelsohn.
40 Cf. the characteristic f, g, and bp in a dream song of the Temiar in inner Malaya
(rec. in EFw FE 4460 (P 460) I I).
37
EARLY MUSle
65
Such fillings must not be mistaken for melodies of several

consecutive seconds, which might look rather similar - C D E.
The third, fourth, fifth would still act as the structural frame,
and the fillers are merely non-structural stepping stones. Of
thirds, usually only the major ones are eleft. Minor thirds, too
elose to the size of a wholetone, do not need a stepping stone and
use it rare1y. A fine example is the archaic me10dy from Lithuania on EFw FM 4009 (P 1Oog) I e, whose seventeen double verses
use exelusively the notes FE'p D.
A structural fourth or 'tetrachord' with or without infixes
often keeps the minor third intact, so that one infix suffices to
divide the quartal frame into a minor third and a wholetone,
usually with the third above and the second below. This is the
kernel of the so-called pentatonic gender, which will be discussed
in the second part of this book. When tetrachords are eleft into
three steps (as a rule two wholetones and one semitone) we speak
of two infixes, an upper and a lower one, and call the fourth
diatonic or heptatonic. Three basic arrangements are possible
(here taken in the usual descending order): the semitone at the
lower end, tone, tone, semitone or T T s (Greek Dorian); the
semitone in the center, T s T (Greek Phrygian); and the semitone
above, sT T (Greek Lydian) All these arrangements can be
traced to primitive civilizations.
The two infixes within a tetrachord are rarely of equal weight
and importance. In the following Rumanian lullaby the tetrachord reads d c b a; but far from being peers, the note b occupies
only 3 time units, where d takes 15, and c and a, either one 14.
There is only one possible interpretation: the infix c has almost
become skeletal, while b is still a passing note. The melody as
such is nearly pentatonic in the form second-above-third:
Ex. 18*
&B ({j I GI ttJ I tilF I tCt I Clljl ~ J I ) j

It happens in folksongs - Rumanian and others - that towards
the end the structuraI fourth, by addition of an inferior note,
becomes a fifth. When such is the case,41 the added note must
* After Riegler-Dinu.
41
As on EFw FE 4419 (P 419) II 7 from Rumania.
66
EARLY MUSIC
not be called a tonic (as has indeed been done); it is simply an

infrafix. The title tonic - misleading as it is in a tetrachordal
melody - can only apply to one of the tetrachordal terminals.
Almost an oddity - when seen from a modern western viewpoint - are pentatonic tetrachords with the infix so dose to
one end of the skeleton that the five-semitone span of the fourth
appears in the forms 4
I or I
4 and not, as usually, in the
form 2
3 or 3
2. Yet this strangely unbalanced tetrachord
is neither artificial nor unnatural; it must be very old as it can
be found in many places all over the world. 42 To all these
evidences we ought to add the famous Kol nidrei sung on the
eve of the day of Atonement:
Ex.
20
, J. J I {) J Il 1 11) J I t;l
n IbJ
In all these cases the semitone is situated above the third.

In reverse, with the major third above the semitone, this
tetrachord is a characteristic trait of Tunisian Berber music,43
Examples:
America: Uitoto, Colombia (Fritz Bose, Die Musik der Uitoto, in Zeitschrift fr
Vergleichende Musikwissenschaft, val. 2 (1934), musical supplement, pp. 10, 12,
13, 15, 16, 19); Pawnee, U.S. (Alice C. Fletcher, The Hako: a Pawnee ceremony,
Washington, 1904).
Europe: Faeroer, due north of Scotland (Walter Wiora, Europischer Volksgesang,
Heemstede, 1952, p. 19).
Africa: Ethiopia (EFw FE 4405 (P 405) II 2), ex. 19:
42
Ex. 19
(transcr. C.S.)
Asia: Korean folksongs (EFw FE 4424 (P 424 12)); Buriat Mongols (Carl Stumpf,
Mongolische Gesnge, in Vierteljahrsschrift fUr Musikwissenschaft, val. 3 (1887),
p. 33); India, rgas bihg and tilanga.
Pacifk: West Carolinas (George Herzog, Die Musik der KarolinenInseln, Hamburg,
1936, pp. 315, 349 ff.); Marind-anim, New Guinea (]aap Kunst, De inheemse muziek
in westelijk Nieuw-Guinea, Koninklijke Vereniging Indisch Instituut, Mededeling
93 (1950). pp. 51 ff.).
43 Erich M. von Hornbostel, Phonographierte tunesische Melodien, in Sammelbnde
der Internationp,len Musikgesellschaft, vol. 8 (1906/07), p. 31, reprinted in Sammelbnde fr vergleichende Musikwissenschaft, vol. I, (1922) p. 3I! ff. - Id., and Robert
Lachmann, Asiatische Parallelen zur Berbermusik, in Zeitschrift fr vergleichende
Musikwissenschaft, vol. I (1933) p. 4.
EARLY MUSIC
67
but also occurs in the archaic Javanese two-step orchestras

munggang and kodok ngorek. 44
We shall meet such major-third pentatonics again in a later
examination of double and tripie fourths.
ALL TETRACHORDAL ORGANIZATIONS must be treated with the
greatest care. To be more explicit, the reader is referred to a weIl
recorded folksong from Anatolia (Asia Minor). 45 The listener
faces a confusing fact: the two infixes are uncertain and variable
enough to make the tetrachord d'-a now Lydian (in the Greek
sense of the word) , now Phrygian, now Dorian, noweven 'gypsy'
with one of the wholetones augmented at the cost of the other
wholetone, as in our so-called harmonic minor scale. In fact,
here as elsewhere, such terms express mere approximations;
infixes, free from rule or consistency, are mobile and subject to
the singer's discretion. In this context I weIl remember Von
Hornbostel's bewilderment when, in 1932, the two of us attended
a Coptic service in Cairo: all the tetrachords so neatly engraved
in an edition of Coptic chant he had studied were due to the
transcriber's imagination and, to say the very least, debatable.
Even in a system as thoroughly established as that of ancient
Greece the infixes were officially granted an incredible liberty.
The chroai or 'shades' within the otherwise normal scales, of
which we read in Archytas, Aristoxenos, Ptolemy, and other
theorists, were hardly more than futile attempts to mathematize,
legalize, and codify what in fact was entirely left to the player's
taste and whim. Hence Ptolemy's tolerance when, instead of
insisting on lawful correctness, he openly admitted that "players
of the lyra favored (!) two normal forms of intonation ... Those
who performed on the kithara preferred (!) ... " 46 The bordernotes of the tetrachord were hestotes or 'immovably' fixed; the
infixes were freely movable.
BARE FIFTHS are frequent in the music of Bushmen and
Patagonians, Lapps and Eskimo:
44
45
48
Curt Sachs, The 1'ise 0/ music in the ancient w01'Zd, New York, 1943. p. 129.
EFw FE 4404 (P 404) I 7 (Asia Minor).
Curt Sachs, Z.c., pp. 2II-2IS.
68
EARLY MUSle
Ex.
21*
, pIrr r I p;. p Irr r I p7 PI F' I ~ .,

Such quintal patterns occur in songs of the Menominee Indians47
with two infixes but without a semitone:
Ex.
22
These melodies cannot possibly be mistaken for double thirds

as not one of the infixes forms a third.
Sixths, transcribed now as minor now as major, have been
found in Polynesia. 48
FOR TUMBLlNG STRAlNS I once coined the term 'pathogenie'
or 'passion-born.' The name is still to the point and has generally
been . understood. But I hesitate to continue calling the usual
horizontal melody 'logogenie' or 'word-born.' Experience has
shown that the term suggests something elose to the stile recitativo of the early seventeenth century or to Richard Wagner's
more or less faithful rendition of natural speech inflection rather
than a neutral, indifferent vehiele of words. Like the tunes of our
hymnals, primitive melodies are often used for entirely different
texts and hence cannot be considered text-born. At best, they
are insignificantly adapted to deviations in the number of
syllables. 49
Indeed, the words themselves are often immaterial and little
apt to shape a melody in their own image. California Indians,
says a Spanish mission report from r8II, do not make any
statement in their songs, "but only use fluent words, naming
birds, places of their country, and so on." 50 Among many
* After Lucy Lloyd and A. Weisbecker.
Frances Densmore, Menominee music, Washington, I932, p. 7I, ex. 39.
E. G. Burrows, Native music 01 the Tuamotus, in Bernice P. Bishop Museum
Bulletin I09, Honolulu, I933.
49 Example: Walter Kaufmann, Folk-songs 01 the Gond and Baiga, in The Musical
Qual'tel'ly, vol. 27 (1941), p. 280.
50 A. L. Kroeber, A mission reporl on the Calilornian Indians in Univ8'Tsity 01
Calilornia Publications in American Archaeology and Ethnology, vol. 8 (1908), p. I9.
47
48
EARLY MUSIC
69
paralleIs, the Sakai of Malacca recite long series of river and

mountain names. A song of the South Californian Luisefio ends:
"I have mentioned all the names of the seasons and stars ... " 51;
and Polynesians carry these lists of incoherent items so far that
E. G. Burrows dedicates a whole section of his Tuamotu book
to "Enumeration Songs." 52
Emancipation from meaningful words also are the non sense
burdens of our kindergarten ditties, the numberless jiggety jigs
and tooUe tooUe toots. Their time-honored predecessor is the
omnipresent ta la la at the stanza end of Italian baUetto songs
around I600. Actually, such refrains can be traced to archaic
music. The Caucasus Georgians have their delivodelivodeli; 53
the almost untouched Caribou Eskimo sing their ja-ja-ha-ja-ja
refrains, 54 the Copper Eskimo their i ya i ya,55 and the Smith
Sund Eskimo their nalegak-nalegaksoak. 56 Beyond the matter
of refrains, J ohn Ross, who discovered this last tribe in I8I8,
mentions that the tribesmen sang for ten minutes merely on the
syllables amnaj sjah and hejaw heja. 57 Indeed, in the music of
the Flathead Indians, "very few of the songs have texts; rather
the majority employ nonsense syllables usually beginning with
the consonants hand y." 58
Textless are also the rapt nigunim of Chassidic Jews on ay or
bay, on oi or doi 59 - syllables probably no more "non-sensical"
than the glossolalia of religious ecstasy which the English Bible
calls the "speaking with tongues" (Corinthians I I4). In a similar
way the Cheremisses in east central Russia sing their sacred songs
without texts. 60 Indeed, the East African Watutsi sometimes
51 Constance Goddard DuBois, The religion 0/ the Luiseno Indians, in University
0/ Cali/ornia Publications in American Archaeology and Ethnology, vol. 8 (1908), p. 19.
52 E. G. Burrows, Native music 0/ the Tuamotus, in Bernice P. Bishop Museum
Bulletin 109, Honolulu, 1933, pp. 42-45.

53 Victor Belaiev, The lolk-music 01 Georgia, in The Musical Quarterly, vol. 19
(1933), p. 422.
54 Zygmunt Estreicher, La musique des Esquimaux-Caribous, in Bulletin de la
Societe Neuchdteloise de Geographie, vol. 54 (1948), p. 4.
55 He1en H. Roberts and D. Jenness, Eskimo songs, Ottawa, 1925, p. 23.
Gr6nland, vol. 152, no. 3 (1952), p. 9.
57 ibid. p. 9.
58 Alan P. Merriam introducing EFw FE 4445 (P 445).
59 Good examples in Chemjo Vinaver, Anthology 01 ]ewish music, New York,
1955, pp. 259-281. - Cf. also Frances Densmore, The use 01 meaningless syllables in
Indian music, in American Anthropologist, vol. 45 (1943), pp. 160 ff.
60 Communication by Dr. Yury Arbatsky (New York, 1956).
70
EARLY MUSle
sing whole seetions of their songs without any words or even

syllabies. 61
In the face of all this evidence, we must realize that many
melodies in this group are not word-bom at all. Thus it might
be better to drop the term 'logogenie' and to speak of horizontal
and of tumbling melodies.
words that matter might arrest the
melodie flow and find a musically unassuming expression without
the help of interesting rhythms or melody proper. Such is the
case in the liturgie recitations on one unchanging pitch, as in
Gregorian psalms, prayers, and lessons. There is, as a rule, a
tlexa or lower cadential note as a mark and remainder of lessening
voice force, and there might be initial and final melody formulas.
But the steps that they form are not essential; whether the voice,
after a score of even-spaced eighthnotes, sinks with the Ilexa by
a semitone or a wholetone or even a third is irrelevant except
from the viewpoint of modal correctness (cf. ex. 68 on p. I72).
All that counts (in principle) is the monotonous seriation of
a single note, which in humility avoids drawing the hearer's
attention from the sacred words to a melodie cloak.
One-tone recitation is not limited to any particular Church.
A classic example of this repetitive or stationary style are in the
archaic world the solemn ancient songs of Hawaii, in which
often dozens of measures repeat the upper note before the singer
gives way to the needs of structure and vocal effort and marks
the end of a phrase by touching the lower note. 62
QUITE TO THE CONTRARY,
Ex. 23
~)U
CCI l;r ce" I
P& Fa I SI rer r I ru
Also the Maori of New Zealand delight in one-note recitations,

from which they swerve only seldom to some auxiliary note a
second or a minor third apart. 63 The islands in between perform
EFw FE 4428 (P 428) I 5.
Helen H. Roberts, Ancient Hawaiian music, in Bernice P. Bishop Museum
Bulletin 29. Honolulu, 1926.
63 EFw FE 4433 (P 433).
61
8S
EARLY MUSIC
71
their stationary recitations without dropping to auxiliary pitches;

after a shout, the second part might follow a semitone higher and
end on a still higher final note from which a cadential glissando
trails down. 64
The Polynesian pieture would not be complete without the
graceful dances of sitting women, to whose quiet, sleepily moving
arm and torso movements the stationary litany style appears
particularly adequate. 65
The distribution chart of one-note melodies records furthermore the (musieally primitive) Caroline Islands; 66 dark-skinned
Jahai Semang pygmies, who live in Malaccan jungles on the
lowest cultural level; 67 Buddhist priests 68 and South Arabian
warsingers; 69 the mestwirebi bards of Caucasian Georgia; 70
and Serbs in magie incantations whose occasional auxiliary
not es are sigh-like and often microtonic. 71 But in most examples
the one-note recitations are nothing but episodes or repetitions
within an otherwise alternating style. As an example of a
intermediary position, I quote a Pawnee song of eleven measures
with only two and a half measure adding band a to the otherwise
exclusive e,72 and a Cheyenne melody, where the whole second
part consists in nineteen measures on d and e. 73
Because in the current of culture and time - as later seetions
will show - alternation patterns grew often from one step to two
steps, from two to three steps, and so forth, the conclusion
would seem logical that litany-like melodies on one note (or
nearly so) must be older even than two-note melodies. But
counting is a poor help in cultural history, the more so as repetition and alternation seem to differ beyond comparison, one
64 E. G. Burrows, Songs of Uvea and Futu11a, in Bernice P. Bishop Museum
Bulletin 183, Honolulu, 1945, p. 64. - Cf. also: id., Native music of the Tuamotus, in
Bulletin I09 (1933).
65 Cf. Curt Sachs, World history of the dance, New York, 1937, pp. 37, 175, 188.
66 George Herzog, Die Musik der Karolinen-Inseln, in Ergebnisse der SdseeExpedition I908-I9IO, H B, Band 9, H. Halbband, Hamburg, I936.
67 Mieczyslaw Kolinski, Die Musik der Primitivstmme auf Malacca, in Anthropos,
val. 25 (1930), pp. 588ff.
68 EFw FE 4449 (P 449) I 2.
69 EFw FE 4421 (P 421) I 4.
70 Robert Lach, Gesnge russischer Kriegsgefangener, in Akademie der Wissenschaften in Wien, Phil.-Hist. Klasse, Sitzungsber., val. 205, Vienna, 193I.
71 EFw FE 4434 (P 434) I 2.
72 Transcibed by Edwin S. Tracy in Alice C. Fletcher, The Hako: a Pawnee ceremony, Washington, 1904, p. rr6.
73 Natalie Curtis, The Indian's book, New York, 197, no page.
72
EARL Y MUSIC
exhibiting the need for either rest or for tension, and the other,
for leisurely motion. The only statement that we dare propose is:
one-note, as well as two-note formations belong in the earliest
age ofman.
Those who believe in a parallel evolution of the human
individual and the human kind will be satisfied to find both
the one-note and the two-note formations among the very first
babble songs of small children. The psychologist Heinz Werner
recorded such songs a few decades ago. 74
BOTH HORIZONTAL AND TUMBLING STRAINS are often found
within the same tribe 75 and indeed within the same piece. This
peaceful coexistence in so many cases forbids aseparate attribution of either style to certain races or minor groups. N or does it
allow us to think of different layers of the Palaeolithic. The way
in which the two species mingle rather bares two different roots
of singing, one derived from the violent howl, and the other,
from recitation. There are at least two roots of mere melodical
urge, not even counting the motor impulse of rhythm - how can
one possibly search for "the" origin of music?
A CONSPECTUS OF A MORE general, stylistic nature may end
this rather technical section.
All art is dual in its concept and trend; it satisfies opposite
urges, and often at the same time and in the same civilization.
It might be imitative to the border of illusion and suggest the
reality of an object, a situation, or an experience of the senses.
r, on the contrary, it can be designedly abstract, non-imitative,
totally unconcerned with objective reality. In between, we have
all shades of transition from one extreme to another; harsh
reality idealized or weakened to mere abstraction and, in the
other direction, ornamental, playful, meaningless patterns
enlivened and transformed to a suggestion of well-known objects
or symbols. This happens in painting, engraving, and sculp74 Heinz Werner, Die melodische Erfindung im Irhen Kindesalter, in K.K. Akademie der Wissenschaften zu Wien, Phil. -Bist. Klasse, Sitzungsber., vol. 182 (1907),
no. 4. - See also Bruno Nettl, Inlant musical development and primitive music, in
Southwestern Journal 01 Anthropology, vol. 12 (1956), pp. 87-9I.
75 Cf., e.g., Hall and Nettl, illusical style 01 the Modoc, in Southwestern Journal
of Anthropology, vol. I I (1955), no. 1 and 2.
73
EARLY MUSle
ture; and it also happens in poetry, music, and the dance.

The two basic forms of early melody discussed so far seem
to fit very weIl in this general picture. Tumbling strains, I said,
"recall savage shouts of joy or wails of rage." They are not
passionate but highly naturalistic; they first repeat, then merely
imitate, and at last only suggest the unbridled outbursts of
strong emotions familiar both to the singers and to the listeners.
Horizontal melodies, on the other hand, being sometimes a
steady repetition of the same note, sometimes a ceaseless repetition of two notes at different pitches, remind us much of those
unassuming, primitive ornaments which iterate the same little
dot or dash around the rim of a bamboo quiver or zigzag across
the osiers of a basket, without any intention of imitating or
expressing the inner and outer experiences of man. Either type
develops away from its own to the opposite type: tumbling
strains move from natural, genuine walls to tame and neat
stylizations; horizontal tunes adjust themselves to the correct
infleetion, meter, and meaning of the words they carry and beeome
realistic speech melodies, i.e., imitations, as in the following
"Harken Israel" of the Babylonian liturgy:
Ex. 24*
,~ j
J I n J I J7J J I n
She-m..
Is-ra-el
J I 11 ; I; ,I
A-do nai E-Io-he-nu
A-do-nai
e-had.
Oceasionally, the profiles of melodies, partly taken from prephonographie publieations,76 have been interpreted as actual
images, say, of longing and gentleness, of mountains and rivers,
of the sun, the wind, and the earth. It is not only possible but
quite probable that here and there the text ereates automatically
a melodie symbol. But unless the tribesmen themselves point
out and explain this symbolism, it is utterly dangerous to read
it into a foreign, primitive me10dy whieh one has never heard.
It is even more dangerous than the notorious abuses of the
'hermeneutic' method applied to the music of Europe. 77
After A. Z. Idelsohn, Hebrisch-01'ientalischer Melodienschatz, vol. 2, Wien, 1922.
76 Cf., e.g. Werner Danckert, Tonmalerei und Tonsymbolik in der Musik der Lappen,
in Die Musikforschung, vol. 9 (1956), pp. 286-296.
77 The latest publication on the subject: Nils-Eric Ringbom, Ueber die Deutbarkeit der Tonkunst, Helsinki, 1955.
74
EARLY MUSIC
While we must be wary where the symbol is meant to be an

automatie, almost unconscious part of the creative act, we are
safer in the face of a conscious, intentional art of illustration.
Most of the latter belongs in post-primitive times. The dramatizations of scenes of war and quarrel in southern Arabia 78
are, no doubt, on a post-primitive level; and this is also true when
the Iru in Uganda describe their hunting parties in long cantatas
from the first alarm on the trumpets to the kill and the noisy
distribution of the meat in a dialogue, climaxing in a final chorus
of all the hunter's horns. 79 The semi-dramatic, gesticulating
recitations of professional and almost professional minstrels also
belong in a later stratum and will be touched upon in the penultimate section.
On a lower level, certain musical imitations are reminiscent of
the realistic cave paintings in southern France, the north of
Spain, and Bushmansland with their striking movements of men
and beasts. 80 One of the most characteristic examples is provided
by Bushmen, who skilfully render the typical gaits of their
many animals by softly beating their bows in adequate rhythms.
With just as much gusto, the Kayabi in the Mato Grosso imitate
the voices of birds, otters, monkeys, and jaguars; 81 and with
exquisite art the Eskimo render the cries of geese, swans, and
walruses - not for entertainment, not for magical purposes, but
simply to bring these animals within the hunter's shooting
range.8 2 In mythical accounts, the Apache act the songs of
smaller birds "in falsetto, while that of the slowwitted lumbering
giant is intoned in a gruff voice." 83 New Guinean Papua weave
into their songs and flute melodies the voices of birds and other
animals, which perhaps "playa certain part in the(ir) religious
ideas." 84 Finally, North Australians, famous also for rock
paintings and naturalistic wood carving, imitate on awkward
wooden trumpets the dancelike movements of the 'Native
EFw FE 4421 (P 421) I 6, 7.
K. P. Wachsmann, Folk musicians in Uganda, Kampala, 1956, p. 7.
80 Cf., e.g., Herbert Khn, Primitive Kunst, in Max Ebert, Reallexikon der Vorgeschickte, vol. 10, Berlin, 1927-1928, pp. 264-292.
81 EFw FE 4446 (P 446) I 10.
82 EFw FE 4444 (P 444) n 2, 3. Also FE 4446 (P 446) II 7.
83 Morris Edward Opler, An Apacke life-way, Chicago, 1941, p. 455.
84 Van der Sande in Jaap Kunst, A study on Papuan music, Weltevreden, 1931,
pp. 7 ff.
.
78
79
EARLY MUSIC
75
Companion,' a graceful blue-gray bird with long thin legs, and

of other birds. 85
We also have an Australian recording of a singer describing,
and perhaps with humor, the drilling of soldiers, commands,
marking time, marching, halting, and marching again.8 6 A. P.
Elkin gives a good idea of descriptive me10dies in Australia,
with the song of the anvil imitating "vividly the darting of
sparks and the 'cry' of resounding noise when the heated iron
is struck with the hammer. The 'boat song' pictures a boat with
mast and sail and an auxiliary engine and the singer travelling
in it. And the 'washing song' portrays the c1othes, washed, hanging
on theline, blowingin the breeze, and then beingironed." 87
The strange doings of the white man are suggestive indeed:
the Temiar in inner Malaya ce1ebrate Bah motoh, the spirit of
the motor boat, in an overwhelming blend of naturalism and
stylization.88
But in general, expression outside of what we somewhat
embarrassedly call by the undefined name of 'art,' avoid naturalism and even realism - mueh the same as small children do not
eopy nature. They draw with their eyes on the paper, not on
objeets, and merely from a eomposite of memory and imagination; their simplified, suggestive peneil strokes are not realistic,
And this is true of all folk art.
The songs in animal stories, an arehaiepart of Amerindian
music, are imitations of animal eries, "although not neeessarily
very naturalistic." 89 Our folksongs are so neutral and uneharacteristic that the same, unehanged melody can be sung for
all the stanzas, indeed, for texts in different, contrasting moods,
and even for unrelated poems. Only by exeeption does the true
folksinger abandon his noneommittal melody to beeome a
dramatie actor. There is an easily aeeessible instanee in the
Sicilian tragi-eomie eloge for a departed donkey with the marvelous rendition of his bray.90 But this might be a survival of
85 EFw FE 4439 (P 439) I 7; cf. also A. P. Elkin, Arnhemland music, in Oceania,
vol. 25 (1956), reprinted 1957 (Oceania Monograph no. 9).
86 EFw FE 4439 (P 439) I 2.
87 Elkin, l.c. vol. 24 (I953), p. 9I; reprinted 1957 (Oceania Monograph no. 9).
88 EFw FE 4469 (P 460) II 2.
89 George Herzog, Special song types in Norlh Ame1'ican Indian music, in Zeitsch1'itt tr vergleichende Musikwissenschaft, vol. 3 (I935), p. 9
90 EFw FE 4520 (P 520) II 2.
76
EARLY MUSIC
the ancient semi-parodical dirges or epikedia for animals, to be

expected on the classical soll of Magna Graecia. 91 More to the
point is that Mississippi folksong of the Old Woman, who once
saw a corpse in her church, asked the preacher: "Will I look just
so when I am dead? ," was told: "Just so," and to the preacher
said - but there is no final word, no final note, only, naturalistically, "a very loud scream." 92
VII. CONSERVATISM AND MAGIC
One-:step melodies have a universal distribution all over the

globe. The majority occurs - and often to the exclusion of other
styles - among extremely archaic peoples: Californian as well
as N orthwest Indians and, in South America, East Brazilian
Botocudos and Fuegians of Patagonia; Bushmen and other
pygmies and pygmoids, both in Africa and Asia; and Solomon
Islanders in Melanesia.
But one-step melodies are by no means limited to Stone Age
civilizations; they are found on essentially higher levels: in the
oldest form of Indian Veda cantillation, in the modern Pacific,
in Muhammedan countries: 93
Ex. 25*
and in the folksong of eastern Europe, as in Bulgaria, Rumania,

Lithuania.
The European habitat is, once more, an evidence of palaeolithic survival in the midst of European civilization, which has a
striking parallel in "the occasional appearance of Palaeolithic
physical types in the present European population."94
Such survivals are by no means paradoxicalor even confusing.
* Transer. C. S.
91
193
About anima! dirges: G. Herrlinger, Totenklage um Tiere in der antiken Dichtung,
92 Published in Arthur Palmer Hudson and George Herzog, Folk tunes Irom
MississiPPi, New York, 1937, no. 43.
93 From EFw FE 4469 (P 469) II I (Kurds).
94 Ra!ph Linton, The tree 01 culture, New York, 1955, p. 27.
EARLY MUSle
77
Nor do they lure us into misdating cultural traits; for the

method of prehistory has taught that Hit is by the appearance
of new forms that a stage is recognized, not by the survival of
old types." 95
Thus, we need not wonder at finding here and there a chronologieal disharmony between musie and other elements of a
civilization: Hthe integration within cultures is loose enough so
that some parts of a culture may be highly evolved while others
have remained simple." 96
This fact has a bearing upon one much contested problem in
the situation of Polynesia: the utterly simple style of Hawaiian
musie. Helen H. Roberts attributed it to influences from the
cantillation ofthe old-Indian Veda texts. 97 With all due respect
it must be said that her suggestion is not acceptable. The
supposed migration and diffusion from India through half of
Asia and half of the Pacific is hardly convincing, on account
not only of the enormous geographie distance, but also of an
enormous difference in religion, culture, and musical styles.
Farther to the west, "the extremely primitive" character
of Carolinian melodies surprised George Herzog as Ha fact unexpected within the partly high civilization of the Caroline
Islanders." 98 Tentatively, he suggested 99 that the luxuriant
poetry of Micronesia and Polynesia might have enslaved their
melodies and killed their musical development. This idea, if
not exactly conclusive, has a certain merit. It is a weIl known
fact that poetry and music, though mated by nature and often
inseparable, have their latent frictions and open conflicts. Up
to the twentieth century, the histories of western song and of
opera provide examples on every page: musical drama, as
Gluck's and Wagner's, stands against common opera, one proclaiming the mastery of drama, and the other, the dominant
rights of music; and, in a similar way, the song inspired by a
poet's accents, thoughts, and words, as Hugo Wolfs Lieder,
avoids and opposes the neutrality of melodies that fit the
Sonia eole, The prehistory 01 East Alrica, London, 1954, p. II3.
Ralph Linton, o.c., p. 56.
97 Helen H. Roberts, Ancient Hawaiian music, in Bernice P. Bishap Museum
Bulletin 29, Honolulu, 1925, pp. 378-380.
98 George Herzog, Die Musik der Km'aUnen-Inseln, Hamburg 1936, p. 263.
99 ibid., p. 279.
95
96
78
EARLY MUSIC
different stanzas of the same poem or even different poems as

happens so often in anthems and hymns. It is characteristic that
Goethe, as a poet, liked the simple and neutral settings of his
friend Kar! Friedrich Zelter better than the illustrative, meddlesome interpretations of Schubert.
How strict this domination of poetry over music can be - and
here we return to our primitive ground - appears in the unbelievably simple recitation of national epics. We think of the Georgian
mestwirebi and their recitations of myth and of history:
Ex. 26
of the Finnish kalevala, 100 the Balkan epics of the guslari,101 or the
Arabian Abu Said romances, with their weansome, hardly varied
melody patterns as a vehic1e for many thousands of lines; of the
archaie cantillation of Persian and Yemenite Jews; 102 and of
the medieval French chansons de geste whose seetions or lais,
up to fifty verses long, obeyed each the same melody before the
last broke loose to form the cadence.103 And we may assume that
the Homeric epics were no exception. Such melodie scraps can
be repeated for hours and hours without annoying a listener
exc1usively interested in the narrative.
A SECOND REASON for inertia is the kind of majority archaism
that we can best observe in Europe. While the leading composers,
often in a breathtaking tempo, hurry from style to style, the
idiom of our anthems, hymns, and many of our hits has basically
not changed since the eighteenth century. We take development,
evolution, and revolution for granted in the concerthall and on
the operatic stage, but resent the simplest ninth chord in the
settings of popular music. The ages of the machine, of electricity,
100 A. Launis, Ueber die Art, Entstehung und Verbreitung der estnisch-finnischen
Runenmelodien, Helsingfors, 1910.
101 Gustav Becking, Der musikalische Bau des montenegrinischen Volksepos, in
Proceedings of the International Congress of Phonetic Sciences I932 (1933). - Walther
Wnsch, Die Geigentechnik der sdslawischen Guslaren, Diss., Prag, 1934.
102 Curt Sachs, The rise of music in the ancient W01'ld, New York, 1943, p. 83.
108 Curt Sachs, Our musical heritage, 2nd edition, New York, 1955, pp. 71 ff.
EARLY MUSIC
79
and of the atom may have remodeled our thought and our vision,
but have not touched the ears of our majorities. If this is true
of our own civilization, how much more does it apply to the
primitives whose balance of change and perseverance is almost
always in favor of tradition.
for the tenacity of archaic music applies in
particular to primitive civilization. There, only material equipment is in need of change and advance. Weapons and tools must
never lag too much behind the ones of hostile neighbors; and a
change of climate and habitat can force a tribe to readapt its
whole material existence.
Immaterial traits are exempt from such competitive adaptation. The help that music might give in battle, hunt, and
personal crises of life, to be sure, is of a magic and therewith
practical, almost material nature. But magics as weil as advanced
religion depend essentially on faithful accuracy in the sense of
unadulterated tradition rather than on change, experiment, or
so-cailed progress. Should the songs of the Apache shaman not
be sung in the right way, "the masked dancer would fall down
like a man knocked out by a blow." 104 (Which brings to mind
the fatal consequences that Hindu mythology ascribes to the
unskilfulsingingof hailowed rgas).105 "As a matter of fact, the
high standard of accuracy facilitates continued belief in the
powers of the medicineman and his ritual; the stock explanation
for failure to cure is that amistake must have been committed
in the ceremony or the singing." 106
To avoid this, accuracy was often enforced - as strictly as
possible. "On the island of Gaua in the New Hebrides it is said
that formerly the old men used to stand by with bows and
arrows and shoot at every dancer who made amistake." 107 And
this formidable discipline applies to music, too: "N ewly initiated
Wasuto (South Africa) boys are beaten with specially plaited
grass switches for mistakes in singing." 108 Which is certainly a
A THIRD REASON
Morris Edward Opler, An Apache lile-way, Chicago, 1941, p. 107.

Curt Sachs, The rise 01 music in the ancient world, New York, 1943, pp. 173 ff.
106 George Herzog, Speech-melody and primitive music, in The Musical Quarterly,
vol. 20 (1934), p. 462.
107 Curt Sachs, World history 01 the dance, New York, 1937, p. 219.
108 Hugh Ashton, The Basuto, London, 1952, p. SI.
104
105
80
EARLY MUSIC
gentle rebuke when compared with the methods of Polynesia,

where carelessness in performing a ritual "was usually discouraged by executing the one who made the mistake." 109
It is less serious and yet within the same ideology when in
French Upper Guinea a visitor trying to enter the retreat of
newly circumcised boys must first intone a circumcision song
which the boys in the hut take up. If the guest or the boys make
amistake, the culprit must pay a moderate fine. 110
The solicitous regard for pure tradition is not just 'pagan.'
The sacred cantillations of Brahmans, Buddhists, oriental J ews,
and Muhammedans are all archaic; and the Catholic Church
has been anxious to preserve the Gregorian chant and to restore
it to its pristine austerity whenever practice swerved from
authenticity. The same is true of the Eastern Orthodox Churches.
The oneness of material and immaterial culture, which we so
easily take for granted, often falls a victim to the extraordinary
inertia of music. Civilization as a whole would change its contents
and its forms, but musical instruments, much as they are
connected with ideas and customs, change at a lesser speed,111
and music as a whole might stagnate.
Where this is fully intentional, inertia has magic connotations.
conjures up an all-important complex of
mental functions overwhelmingly active in the primitive world
and still alive within and without religious rites of the highest
ciVilizations, however weakened they may be by scientific
knowledge and reasoning.
Primitive man is exposed to countless occurrences whose
causes he fails to see; all his life he suffers destinies whose
reasons of being are hid from his eyes. Conception and childbirth, sickness and death, abundance and want, rainstorms and
dearth - who makes, who sends them? What can man do to
avert misfortune? How may he summon health and copious
food?
Primitive answers to these questions are not rational; being
unscientific, they ignore deduction, test, and hence the lawful
THE WORD MAGIe
Ralph Linton, O.C., p. 192.

Andre Schaeffner, Les rites de circoncision en pays Kissi, in Etudes Guineennes,
no. 12 (1953), p. 44.
111 Cf. Andre Schaeffner, Les Kissi, Paris, 1951, p. 6.
109
110
EARLY MUSIC
81
processes at the bottom of nature and life. Instead, all that which
befalls a man, his livestock, and his crops, must be the intentional
doings of benign or malevolent beings, be they demons or mortals; and, in the ever-present dangers of existence, he tries to
call, appease, or combat these powerful shapers of fate.
In doing so, his strongest and often exclusive weapon is magie,
"a supematural technique," in Malinowski's short definition,
"by which man can, in his conceit, bring about all that which
his rational technique falls to accomplish." 112
The ideas and practice of magics take up an incredible space
in the lives of men and of peoples - far beyond the primitive
level; and special libraries have been founded to house the
gigantic literature devoted to this form of approach to the
secrets of our existence. Here, we cannot give a comprehensive
survey of this field; but a few principal forms connected with
sound and with music must be mentioned.
Charms against sickness are supreme among them. In many
civilizations, certain rites are performed by an intelligent,
excitable, and often psychic 113 spokesman of the tribe whom
we know as the medicine man, witch doctor, or shaman. In astate
of self-inflicted trance he acquires the superhuman power to foresee
the future, to heal, and to chase the evil forces. He dances with
clattering anklets, shakes the seed-filled, rattling gourd, strikes
his shallow, circular drum, and sings or bawls his hectic incantations for hours and hours until complete exhaustion silences him.
The strength of believing in these nerve-wrecking rituals even
in the midst of a western environment and in the face of American
doctors, can be judged from arecent case of seventy frightened
Navaho Indians, tuberculosis patients in an Arizona sanitarium,
who during a thunderstorm had to be quieted by the magie herbs
and the 'sing' of a native medicine man flown to the hospital. 114
An Amerindian healing song, breathless and breathtaking, is
reprinted in my 'Rise f Music' on p. 22. 115
Bronislaw Malinowski, Tlf,e dynamics 01 culture change, New Haven, 1945, p. 49.
But cf. A. P. Elkin, Aboriginal men 01 high degree, Sydney, 1944, pp. 22 ff.:
"Medicine-men are normaL" - Psychoneurotic: Paul Radin, The world 01 primitive
man, New York, 1953, chapter 4.
114 Associated Press report, quoted from The Boston Herald, Aug. 17, 1955, p. 16.
115 After Erlch M. von Hornbostel. -Cf. also Frances Densmore, Music in the treatment 01 the sick by American Indians, in Hygeia, April 1923, pp. 29 ff.; The Musical
Quarterly, vol. 13 (1927), pp. 555 ff.; Scientilic Monthly, vol. 79 (1954), pp. 10g-U2.
Also: EFw FE 4525 (P 525) A 14 (Eskimo).
112
118
82
EARLY MUSIC
After sickness - death.

There are funeral rites in the primitive world with the noble
pathos of deepest emotion; nothing could be more strongly
moving than the dirges of N orth American Indians. But oftener
the rites derive from terror and the fearful anticipation of
imminent danger, be it through the hostile speil which must
have caused the present death, or through mischievous acts of
the deceased himself. From such peril the living must be protected by magical practices. Inversely, the magical acts may
serve the interests of the dead, especially in cattle-driving
societies where the living try to safeguard life in the beyond.
In planter civilizations, everybody's existenc~ depends upon
the mysterious growth of the seed. Thus it is protection of life,
too, when tillers bridge with magical rites the critical time between the seasons of sowing and reaping. Processions cross the
fields or walk around them with sacred songs and often with a
sheil trumpet, which, coming from the water, must unfailingly
attract the needed rain. Or else the boys leap over the pastures
with shouts and tinkling beils to drive the evil spirits away.
Peasants in Europe still cling to the old tradition, and the
blessing of the crops is part of Catholic no less than of Hindu
and Malay rites.
Nowhere are such ceremonials more beautiful than in Indonesia. To sow the rice, planters advance in solemn dance steps
and open the ground with pounding poles whose rattle tops are
tuned each to one note of the scale; the women foilow and drop
the grains in the pits. On the evening of the harvest sacrifice,
the maidens hit the soll with seed-fi1led bamboos and to their
rhythm sing the charming ditty: "Stamp, ye friends, for we look
down, look down to the praying, the praying new rice."
The dull and hollow-sounding shell trumpets and the seedfilled bamboos (as many other instruments) belong in the province ofhomoeopathic sorcery, which, based on a presumed
affinity and attraction of phenomena equal or simllar, forces
nature to imitate what is done in its face. Pounding the soll with
seed-filled tubes must act on the seeds interred in the ground
and makes them thrive when their time has come. The charms
for wind and rain must be of an analogous kind of homoeopathy.
The rainmaker ties a narrow, thin-cut slab of wood to a cord
EARLY MUSIC
83
and whirls such a 'bull-roarer' round his head; size and speed
make it whistle like a squall or roar like a thunderstorm. Squall
and storm, irresistibly attracted by their own characteristic
noises, must come and bring the rain that the crops require.
Mere whistling with the mouth has the same effect. 116 Again,
should there be too much rain, the charm is simply reversed.
The Zulu shaman climbs a hill when an undesired thunderstorm
nears, blows a squeaking fife of bone, and shouts: "Ye sky, be
gone, I hold nothing against you, I will not fight!" 117
EVERYTHING THAT SOUNDS, be it in the cruder form of frightening noise or the organized patterns of music, bears the brunt of
mankind's eternal strife against the hostile forces that threaten
his life and welfare ; and, just as weH, nothing better than sound
can summon the powers of luck and prosperity. Songs, say the
Melanesians, can have mana, or a hidden supernatural energy,
and the Polynesians "chanted hymns which had mana because
of the things which they recounted and the rhythmical, forceful
way in which they were recited." 118 The songs of American
Indians, too, "were believed to come from a supernatural source
and their singing was connected with the exercise of natural
powers." 119 Sound, as we mentioned, affects our nervous system
more than other sense perceptions; and since the primitives
project their own emotions onto the invisible forces around them,
these too must succumb to the unique mysterious spell of timbre,
rhythrn, and tune.
Even language stresses the unity of singing and magics as the
Latin word incantatio, 'magie formula,' derived from cantare,
and the English charm, from carmen.
WHEREVER SINGING is an act of ecstasy and depersonalization,
it moves away from ordinary human expression. The voice is
often remote from being as 'natural' as we believe our own
execution to be. It is colored by pulsating, yodeling, ventriloquizing, or bleating. One screams, yells, squeaks, mumbles, and
Leslie Milne, The home of an eastern clan, Oxford, 1924, p. 23I.
Henri A. Junod, The life of a South African tribe, val. II, Neuchtel, 1913, p. 291.
118 William Howells, The heathens, Garden City, 1948, p. 234.
119 Frances Densmore, The study of Indian music, in Smithsonian report for I94I,
116
117
84
EARLY MUSIC
nasals. Some people set such mannerisms strict1y apart according

to the purpose of singing; the Sioux, for instance, reserve a
nasal intonation for love songs; 120 on the Orinoco in northem
South America the singing women pinch their noses when the
gruesome mask dancers make their appearance; 121 and Yemenite
Jews use a boy's high-pitched voice to chant the targumim or
Aramaic translations of the Bible. 122 If this last example is not
exactly a mannerism, it is a special coloring for a special purpose.
Scattered all over the world is the queer nasaling caused by
a tiny membrane which screens the voice and colors it in covibration. Its original form, a simple tree leaf stretched out in
front of the mouth, has survived in practice and in the otherwise
inexplicable German phrase: kein Blatt vor den Mund nehmen,
which means literally 'to hold no leaf against the mouth,' or,
in idiomatic English, 'to be plain-spoken.' The membrane,
whatever its substance, form, or frame, is indeed a concealing
agent; more speciflcally, it is used as asound mask to depersonalize the singer's voice just as visual face masks dehumanize
the ritual dancers.
Similar 'kazoos' are described in my earlier books.1 23 There
are pseudo-trumpets serving as megaphones, or drums held at a
certain angle in front of a shaman's mouth, or conch shells to
make a chieftain's voice suprahuman, hollow, and eerie. The
strangest device is the nystaranga of India: a singer presses
two trumpet-like tubes right and left against his throat, and
the vibrations of the vocal chords are caught by a membrane
in the 'mouthpiece' of the trumpet and reinforced by its tube.
Two further examples, not mentioned before, Inight be added:
according to Tacitus' Germania, c. IOO A.D., the Germans sang
their battle songs against the shield to make them more raucous
and frightening; and in a curious game little Eskimo girls bend
down and whisper into a kettle words in strict, strong-breathing
eighth an<! sixteenth notes (J = c. I08 MM) which irresistibly
Frances Densmore, Chippewa music, Washington, I9IO, p. 6L
Alain Gheerbrant, L'expedition Orcnoque Amazone, Paris, I952, p. I25.
122 J ohanna Spector's tape recordings and her paper On the trail 01 Oriental music,
in The Reconstructionist, vol. I8 (I952), p. u.
123 Curt Sachs, Geist und Werden der Musikinstrumente, Berlin, I929, pp. Io6 ff.id., The rise 01 music in the ancient world, New York, I943, p. 23. - Cf. also: Curt
Sachs, The history 01 musical instruments, New York, I940, pp. 462 ff. - EFw FE 445I
(P 45I) I 2 (Cameroons).
120
121
EARLY MUSle
85
remind of jolting railway imitations 124 (cf. ex. 15 on page 62).

Therewith we have entered the realm of voice mannerism.
VIII. VOCAL MANNERISMS
The impersonalization of the human voice in magical rites

does not explain the total of distinctive mannerisms in singing
styles all over the world. The neat notations of non-western
music that we are offered to read, withhold from us one salient
fact: that nowhere outside the modem West do people sing with
a voice for which we have coined the honorific title of 'natural,' 125
that, to the western ear, all oriental and primitive singing is
unnatural and seasoned with strange, unwonted mannerisms;
and that these regional differences in handling the voice are often
more significant than tonal ranges and structures. How people
sing is no less meaningful than what they sing. An American
Indian, a Japanese or Korean, an Arab, a Negro -let them sing
a couple of notes, and they cannot escape recognition.
VOCAL MANNERISM is the hardest and most neglected branch
of our studies. For expressing what we hear, we have no proper
classification or terminology beyond the arbitrary lingo of
western voice teachers.
The difficulties grow when we face a phonograph recording
instead of the voice itself. Although the most recent methods
of phonography with microphones and tapes diminish the danger
of distortion and mishearing, even a first-class modem loudspeaker still might leave the listener in doubt as to whether the
singer is a man or a woman.
There is a pioneering practical guide through this field in an
album of Ethnic Folkways Library,126 which Henry Cowell has
compiled and introduced under the title The World's Vocal Arts.
Unfortunately, for us, the stress is on the word Arts: all the
singing styles of American Indians, African N egroes, and AustraloPacific peoples, in short the Primitives, are excluded from this
collection.
184
125
188
EFw FE 4444 (P 444) I 8.

But cf. Douglas Stanley, YOU1' voice, New York, I945, pp. 295 ff.
EFw FE 45IO (P 5IO).
86
EARLY MUSIC
The weighty problem of voice mannerism entered our field

of vision at first when Erich M. von Hornbostel, going beyond
the previous exclusive interest in melodie and rhythmical aspects,
drew our attention towards the vocal style of the Amerindians
and described its emphasis and over-accentuation, its sharp
staccato, indeed, its puffing steam-engine quality, with the
characteristic break-up of longer notes in even pulsation.1 27
More recent is Merriam's description in a monograph on the
Flathead Indians: "Using a tight, though open throat but without employing full resonance possibilities of the upper nasal
caveties, a penetrating quality is produced in the style often
labeled as the 'clenched-teeth' or 'ventriloquistic' style. 128
T enseness and emphasis recur in Siberia, from which the Red
Man first migrated to the American continent from twelve to
fifteen thousand years ago. 129
One more connection between the musical styles of Indians
and East Asia (Japan) is the frequent upwards drive, especially
of higher notes. I t is particularly clear in the N avajo Creation
Songs recorded by Harvard's Peabody Museum. But McAllester
emphasizes rightfully that the N avaho style "is not as tense, as
throat-splitting, as Sioux singing." 130
Indeed, an (essentially smaller) part of the Indians, "the
Yuman tribes, the Pima and Papago, and the Serrano and
Cahuilla," have no share in the typically American technique.
The area of these dissenters "corresponds to those of the mythical,
dreamt song series, of a certain simplicity of the melodie development, and of the prominent use of the gourd rattle." 131
127 Erieh M. von Hornbostel, in Theodor Koch-Grnberg, Vom Roroima zum
Orinoco, vol. 3, Stuttgart, 1923, p. 408. - id., Ueber ein akustisches Kriterium fr
Kulturzusammenhnge, in Zeitschrift tr Ethnologie (19II), pp. 601 ff. - id., Fuegian
songs, in American Anthropologist, N.S., vol. 38 (1936), p. 363.
128 Alan P. and Barbara W. Merriam, The ethnography of Flathead Indian music,
in Western Anthropology no. 2 (I955), p. 13.
129 Ralph Linton, in the Scientific Monthly, vol. 72 (1951) and in Hoebel, ]ennings,
Smith, Readings in anthropology, New York, 1955, pp. 48 ff. - "The somatic and
cultural affiliations of the American Indian and the Siberian aborigine are today
taken for granted; archaeological research has made steady progress toward documenting the route of ab original migration to the Americas across the Bering Straits
and the Aleutian Islands." (Melville ]. Herskovits, Franz Boas, New York, 1953,
P.19).
David P. McAllester, Enemy Way music, Cambridge (Mass.), 1954, p. 55.

George Herzog, Musical styles in North America, in Proceedings ofthetwentythird International Congress of Americanists, 1928, p. 457. - id., The Yuman musical
style, in Journal 0/ American Folklore, vol. 48 (1928). - id., Maricopa music, in
Leslie Spier, Yuman tribes of the Gila River, Chicago, 1933, pp. 271-279.
130
131
EARLY MUSIC
87
Beside these two fundamental styles, the American Indians

perforrn their love charm songs in a completely different way.
Supposed to derive from flute melodies, these songs have indeed
a distinctly instrumental character. They are slow, unaccented,
nasaling, and often fluttering under the impact of a waving hand
before the mouth. 132 In a similar way, we leam from an eighteenth
century source that the (now extinct) Abip6n in South America
changed their voices in a manner consistent with the text. 133
Few regions have completely unified styles and mannerisms.
Polynesians, alike, adapt their voices to changing purpose of
singing: the Tuamotu in the West make their voices breathy,
low, mysterious for chants to exorcise the evil spirits.l 34 But they
also know a strait and steady singing and, alongside, a quavering
manner, which at times becomes a triil on two pitches. 135
In the South of Polynesia, the Maori of New Zealand performed their war songs prestissimo in a lashing rnicrotonic
falsetto, which to the unprepared listener may weil sound like
excited cackling in a bamyard (at least in the phonograph).
Incantations, although in a similar high-strung style, give this
impression much less.l 36
The excited treatment of the voice reaches across the Tasman
Sea to Australia. The Australians, so teIls us areport from the
r860es, sing every note with a kind of (strong) expiration. Ail
the time they gasp with a convulsive quaffing (Schlrfen), only
to waste their wind with an obvious effort. 137 But Australian,
as weil, is astrange abdominal gruntinghardlyheard elsewhere. 138
Even in a comparatively smail territory like Uganda, so we
leam from Wachsmann,139 voice production is very different;
women in the extreme North tap their throats with the finger
tips, while others ululate, yodel, or sing in glissando.
Classic singers in the Far East are trained in "seven different
132 George Herzog, Special song types in N orth A merican Indian music, in Zeitschrift
fr vergleichende Musikwissenschaft, vol. 3 (I935), pp. 5 ff.
133 Martin Dobrizhoffer, Geschichte der Abiponer, Wien, I783, vol. 2, p. 577.
134 E. G. Burrows, Native music of the Tuamotus, in Bernice P. Bishop Museum
Bulletin I09, Honolulu, I933, p. 57.
135 ibid., p. 58.
136 EFw FE 4433 (P 433).
137 Hermann Beckler, Die Corroberri, in Globus, vol. I3 (1868), p. 82.
138 A. P. Elkln, Arnhem Land music, in Oceania, vol. 25 (1955), pp. 322,326,329
(reprinted as Oceania Monograph No. 9, Sydney, 1957).
139 K. P. Wachsmann, Folk musicians in Uganda, Kampala, I956, p. 4.
88
EARLY MUSle
vocal styles; one for each of seven emotional situations, instead

of one voice placement as in the West." 140
This is correct as far as it goes. Actually, the modem West
has an official style as weIl as a crooning, a toreh, and a diseuse
or diction style. 141
From these motley piCtures we leam that many tribal groups
have several voice mannerisms according to sex, geographie
situation, and the singing occasion. No attempt must be made
to ascribe the individual styles to 'races,' whatever this word
may mean. They are not biologically conditioned and hence
not inherited but purely environmental, that is, cultural. An
attempt at classifying vocal styles in the framework of Kulturkreise has been made by Herbert Hbner. 142 But the field is
marked off too narrowly; it comprises nothing but the islands
of N ew Britain and N ew Ireland; a solution requires a broader
basis and a careful avoidance of the dangerous ground of
pre-conceived theories.
'A QUAVERING MANNER' appears in our description of Tuamotu
singing. This mannerism is still important on the high level of
India's culture: in the classification of Nrada's Sangitamakaranda, her melody patterns or rgas were divided into three
groups: "the first includes those sung with a quivering voice
throughout; the second, those with parti3J. quivering; the third
those without any quivering." 143 Indeed, two symbols in the
neumatic notation of Gregorian song, quilisma and pressus,
indicated in early times a singing "tremula voce"; and Magister
Lambertus, the 'Pseudo-Aristotle,' mentions a third neume, the
plica, as due to half-closing the epiglottis or glottal valve, combined with a repercussio of the throat itself.
There seems to be a connection with the bleating technique
that reaches from Mongolia to the East.
Henry Cowell, l.c., p. 2.
"Coon shouting, blues singing, torch singing come from an uncoordinated,
or mixed, lower register and a mouth resonance adjustment ... Crooning is a
throaty, mixed falsetto screaming". (Douglas Stanley, Your voice, New York, 1954,
p. 7). - "Diction singer (diseuse) is one who concentrates on the words with complete
disregard for the music or import of the composition. Each and every syllable is
carefully and deliberately articulated and one syllable is sung at a time. The conso
nants are over-articulated .... " (l.c., pp. 23 ff.).
142 Herbert Hbner, Die Musik im Bismarckarchipel, Berlin, 1938.
143 Sachs, The Rise 0/ music in the ancient world, New York, 1948, p. 182.
140
141
EARLY MUSle
89
Beyond all these mannerisms we find, as the most remarkable

fact, in many parts of the world an emphasis on the higher register of man. Real barytones and bassos are generally absent.
This seems to have been the case in the European Middle Ages,
too. Guillaume Dufay, in the fifteenth century, was commended
for having expanded the lower limit of the musical tessitura by a
fourth from G to D; in the Trent Codices, principal repository
of fifteenth century music, only six percent of all voice parts
follow the bass def; and bass instruments were not manufactured
before the end of the 1400s.
BEYOND SUCH GENERALITIES and single traits expressed in a
vague, inconsistent terminology, the method we need for future
research is strict1y physiological. There is essentially only one
question to answer : which musdes, between diaphragm and face,
are used in the cases to describe?
Such detailed analysis from continent to continent cannot
be carried through in this general study. It must be left to
a specialist, and even the specialist will need the pioneering work
of many monographs before he surveys the globe as a whole.
Being neither a physiologist nor a trained singer, I must step
back and leave the work to scholars better prepared. For obtaining at least a sampie of the method I visualize, I enjoyed
the competent help of Rose Brande1, who, as an ethnomusicologist and a singer, was infinitely better equipped to analyze
foreign singing styles. Here is the analysis of the J apanese no
style, which is essentially due to her empathy and knowledge.
The J apanese no style is to a high degree guttural, nasal, and
impure in its low register. Physiologically, the word guttural
means the extreme tension of the constrictor musdes of the
throat; and nasaling, the tension of facial musdes (though not
the resonance of the nasal cavity). Japan's low or chest register
is not (in our western -sense) coordinated with the upper, head,
or falsetto register; and the tension of the laryngeal muscles
has little balance. The back of the tongue blocks the pharynx
almost completely, so that the mouth provides the resonance.
Breathing is extremely shallow; and no vibrato enlivens the
voice. Once in a while we hear the short and sudden leap to and
from the upper register for which there is no name (since it does
90
EARLY MUSIC
not occur in western music) and which we might adequately call

a spasm or jerk.
Beyond the sample of J apanese singing, our discussions found
evidence of an enormous province, reaching at least from the
Far East to Negro Africa, with the common characteristics:
guttural obstruction, lack of vibrato, shallow breathing, closed
pharynx, and mouth resonance.
These scanty sentences are meant to show an acceptable
method of avoiding an unfounded terminology and unfounded
impressions. But the way is long, and the going rough.
THE ELUSIVE QUESTIONS of tempo and intensity should at
least be touched upon.
Shamans accelerate their songs very often in order to increase
the excited and exciting character of the healing act. But an
increase of speed might occur under other circumstances as weIl.
John Ross, discoverer of the Smith Sound Eskimo in '1818,
mentions that these people sang accelerando and with an ever
shriller voice. Strangely enough, R. Stein, wo accompanied Robert
W. Peary to Greenland in the I8goes, reports the contrary: these
same men sang in "a low tone, inaudible thirty feet away" and
refused to raisetheirvoices. 144 Neither do the Batak in Sumatra
tolerate loud voices; they" even subdue the tone of their gongs. "145
Here, too, we should take to heart the moral of that classic
example given by two South American Indian tribes of the same
race and living side by side on the same spot of the northeastern
Gran Chaco. One of them excels in wildest dance leaps, vigorous
rattle swinging, and violent singing. The dances of the other
tribe are "a hesitant, indolent weaving to and fro; their musical
instruments sound weak and timid; their drum, as though
muffled; and their flutes, never shrill; and even their war horn
makes less noise than a toy trum pet ... No effervescent enthusiasm appears in their melodies. Their singing is soft and listless. "146
A similar example has in later years been reported from two
tribes of the Bororo Fulah in the Sudan: one of them shows
Gronland, vol. 152, no. 3 (1952), p. 10.
145 Raden Suwanto, notes to EFw FE 4406 (P 406).
146 Curt Sachs, Rhythm and tempo, New York, 1953, p. 53 after Herbert Baldus,
Indianerstudien im nordstlichen Chaco, Leipzig, 193I, p. 106.
EARLY MUSIC
91
.more youthful vigor in its dances, more freshness and nerve

than the other one, and the tempo of its songs is markedly
faster.147 In Uganda, some tribes delight in aslowtempo; others
disdain it and prefer to hurry so much that "the sung words
are often unintelligible on account of the pace." 148
The ultimate cause of such a contrast within the same people
is unknown. But it cannot be racial, cultural, or climatic.
Physiologically conditioned intensity has been exemplified
earlier in this book: true tumbling strains start always in fortissimo and weaken until the bottom note is sung in pianissimo.
In the same paragraph I mentioned the psychological change
of loud and soft in the one-note litanies of the Maori.
IX. INSTRUMENTS
The knowledge of instruments, although important enough in

every chapter of music history, is vital in its primitive and
oriental branches: ten thousands of years have left us no other
musical relies.
This book ean give them but limited spaee. For a more
comprehensive presentation, the reader is referred to my History 0/ Musical Instruments 149 and to my German Geist und
Werden der Musikinstrumente,150 whieh specialize in primitive,
folk, and oriental instruments and their problems.
To supplement such studies by audible illustrations, the
author has compiled an album of two reeords with sixty-seven
selected examples from the non-western world (I9S6).151
that the beginnings of music
were purely voeal. Peoples representlng a partieularly pristine
phase of man's evolution, like the Vedda in inner Ceylon, the
East Afriean Wanege, the Siriono of eastern Bolivia, and
most Pat~gonians, have no instruments. When, in a later
LET US ONCE MORE REMEMBER
147 Zygmunt Estreicher, Chants et t'ythmes de la danse d'hommes Bororo, in Bulletin

de la Societe Neuchateloise de Geographie, vol. 51 (1954-'55), p. 68.
148 K. P. Wachsmann, Folk musicians in Uganda, Kampala, 1956, p. 3.
149 New York, 1940.
150 Berlin, Dietrich Reimer, 1929.
151 EFw FE 4525 (P 525).
92
EARLY MUSIC
phase, instruments begin to appear, they do so in rudimentary

forms without any attempt at melodie achievements. 152
The use of instruments goes back to late palaeolithic times.
A number of instruments - strung, beaded rattles, scrapers,
bullroarers, and bone :nutes - have been excavated at least from
Magdalenian strata, which represent the ultimate steps before
reaching the end of the Older Stone Age. 153
In early periods, the reasons for using instruments were totally
different from those of singing. Early vocal music had carried
words as a mere vehicle or else burst forth as a powerful discharge
of passionate feeling. But instrumental music, at first remote
from passion, began in general as a percussive act of the body:
slapping the buttocks, the belly, the thighs, or clapping the
hands, or stamping the ground. I ts cause was a muscular urge
concomitant to the nervous tension of those who sang or listened
to singing; its aim was audible order. Against the basically
melodie character of vocal music, the beginnings of instrumental
music were basically rhythmic. Singing is allowed to be free,
and the warbling bird does not obey a conductor's baton;
percussion, like any recurring motion of the body, invites,
indeed, demands a regular time: rhythmus in corporis motu est,
states Marcus Fabius Quintilianus, the orator, around the year
IOoA.D.
Percussion stands seldom alone; rhythm, as a regulator, becomes nonsensical where there is nothing to regulate, be it a
melody or at least the steps of dancers and marchers. Thus,
early percussion acts mostly as an accompaniment, although we
shall see that in quite a number of cases the drums keep a rhythm
and tempo totally different from that of the voices.
But our narrative is still rather far from times when drums
were used. Incidentally, the drum as a criterion of chronology
must be dealt with very carefully. We can more or less exactly
determine when the drum appeared; but the lack of drums in
152 For a c1assification of musical instruments cf. Erich M. von Hornbostel and Curt
Sachs, Systematik der Musikinstrumente, in Zeitschrift fr Ethnologie, vol. 46 (1914),
pp. 553-590. - Abridged in Curt Sachs, The hist01'Y of musical instruments, New York,
1940, pp. 454-467. - Contradictory: Andre Schaeffner, Origine des instruments de
musique, Paris, 1936, and Hans-Heinz Drger, Prinzip einer Systematik der Musikinstrumente, Kassel, 1948.
153 Otto Seewald, Beitrge zur Kenntnis der steinzeitlichen Musikinstrumente Europas, Wien, 1934.
EARLY MUSIC
93
some civilizations does not necessarily indicate an earlier stratum.

In Polynesia, for example, drums were once in use, but have in
many islands been abandoned. 154
In a long evolution, early man increased the effect of pristine
body percussion by replacing the struck or striking part of the
body with a harder object taken from nature or made by human
hands. Clapping the hands was superseded by the wooden click
of boomerangs ; stamping the ground was reinforced by rattling
nuts or hooves strapped around the ankles, indeed, by using some
resonant, hollow bamboo as a pestle. Such 'extensions,' as the anthropologists call them, led from the bare, uncovered body in a slow
evolution to the actual instruments of later times and the present.
Extensions of the limbs were not the only beginnings of
instrumental music, and intensified rhythm was by far not its
sole purpose and meaning. There were the many objects, mute
themselves, but able to make a section of the air vibrate and
thus to provide sounds. A hollow branch or reed or conch distorts the hum or howl of those who press it as a loudspeaker
against the lips; an eagle' s bone becomes a screaming whistle;
the bull-roarer - a flat slab of wood whirled around the head on
a twine - roars like a motor propeller. Eerie, non-human, terrifying were those sounds and roused emotions and the imagination of man beyond whatever slapping, clapping, or pounding
could do. Rather, this horrific, awe-inspiring din stands for the
voices of spirits and demons; as we behold them in the gruesome
masks of demon-dancers. HA cold sweat," confesses Gheerbrant,
"had broken out of all my pores and run down my body. The
hairs on my skin had stood upright. The blast of the trumpets
had shaken the walls, the roof of the hut, and the very ground
itself. The sounds had seemed full of sweat and smoke, and they
had moaned and howled as though they really were the voices of
spirits, simple animal spirits of the heaven and the earth. They
rose from the ground and echoed from the walls and the roof;
they went through my vitals like beasts roaring in the forests." 155
Once more we find ourselves in that proto-religious and prescientifie world in whieh both outlook and action obey the
paralogicallaws of magie.
154 E. G. Burrows, Native music 01 the Tuamotus, in Bernice P. Bishop Museum
Bulletin 109, Honolulu, 1933, p. 71.
155 Alain Gbeerbrant, Journey to the Far Amazon, New York, 1954, p. 104.
94
EARLY MUSIC
The role of instruments in magic-ridden cultures is confusingly manifold. Instruments, as the only objects, have not
only meaningful shapes and colors but also meaningful sounds
and even significant substances. The ensuing connotations are
often in agreement and sometimes contradictory. All of them can
be reduced to one of the two universal principles: through the
worlds of men and beasts, across all groups and families, a great
divide leaves, left and right, what Chinese cosmology has called
the yang and the yin, the male sex and the female sex as the
fundaments of all organic life. Not even divinities are exempt
from this elementary dualism - the skies and the underworld
have their gods and their goddesses: Zeus with Hera, Osiris with
Isis, Odin with Frigg. Differing and complementing each other in
their anatomical, physiological, psychical, social traits, the sexes
are in their polarity important, indeed, indispensable to ascertain
the survival of society and procreation. The impact of this
dualism has been so strong, essential and consistent on the human
mind that the universe in all its manifestations, as planets,
seasons, liquids, colors, numbers, pitches, seemed to be an
interplay of male and female qualities. If in these cosmological
juxtapositions the sun and daytime, blood, color red, odd numbers
stood on the masculine side, the moon, and nighttime, milk and color white as wen as even numbers stood across on the feminine side.
Musical instruments were vitally involved in this sex dualism.
Masculine, in unimpaired purity, is the trumpet - even in its
pristine, pre-metal form when, made of reed or wood, it served
solely as a megaphone or voice-disguising mask without requiring the resilient tension of the lip muscle characteristic of
actual trumpeting. With its aggressive, menacing sound, it
is clearly virile; as a tube, it has definitely phallic connotations;
often it is painted red, and even into the twentieth century it
has in military bands been adorned with tasseis in red or wrapped
in feit of the same color. In virtue of such masculine symbolism,
it became in later civilizations an instrument for war and princely
pomp; but in more archaic cuItures it was confined to rituals of a
male and solar nature. It acted as acharm for rebirth or resurrection after death or when in rites at dusk and at dawn the sun
must be forced to reappear in the morning. Women who happened
to see a trumpet were often put to death.
EARLY MUSIC
95
The role of the flute is originally as virile as that of the trumpet.

Owing to its tubular shape, it represents the penis; and the sound
is male enough; the flutes are either long and wide and yield a
subdued roar, or they are short and narrow and have the shrill
agressiveness of a whistle. Often, in this latter ease, they are
made of bone which, once more, has generally a phallie signifieance. Accordingly a flute must, in many archaie civilizations,
not be seen by uninitiated strangers, women, or ehildren. In
New Guinea, saered bamboo flutes, up to six meters long, are
aetually used as penes for ritual cohabitations with the chieftain's wife; they are played in the forest while a widow is eompulsory sleeping with a relative in the degree of kinship that the
triballaw requires; and girls are deflowered in front of them in
the eeremonial house of the village. 156
The flute seems to be a love charm everywhere. Emest
Hemingway, in A Farewell to Arms, makes one of the soldiers
tell how at horne in the Abruzzi mountains young men were not
supposed to play the flute in serenades "because it was bad for
the girls to hear the flute at night." Among Ameriean Indians,
the flute belongs greatly to lovers and love; but we do not know
whether ornot its effeet is still considered magie. Often, as
among the Sioux, young men in love leam the flute so they may
woo their girls in the proper way; playing the flute amounts to a
proposal; 157 and many love songs are derived from flute melodies.
Flute melodies are dangerous not only to girls. Herdsmen of
the Gikuyu in East Afriea are sometimes forbidden "to play
the flute while the flock is in the grazing field, for it is feared
that sweet melodies might invoke evil spirits and thus bring
defilement to the herds. It is also taboo to play a flute or to
whistle inside a hut, because it is believed that sweet music can
easily attraet the attention of wandering evil spirits and lure
them to come and cause misehief inside the homestead."158
Stringed instruments, rather scarce in primitive civilizations,
are on the feminine side. Originally struek with asiender stick
(before being plucked or bowed), they render a frail, subdued,
and quite unaggressive sound so weak that some resonance
156 Richard Thurnwald, Banaro society, in Memoirs 01 the American Anthropological
Association, vol. 3 (1916), pp. 261, 269.
157 EFw FE 4505 (P 505) II 6 (ed. Henry Cowell).
158 Jomo Kenyatta, Facing Mount Kenya, London, 1953, p. 95.
96
EARLY MUSIC
cavity is necessary to make it audible to others - a pit in the

ground, an earthen vesse1, a gourd, or even the pIayer's mouth;
and this resonator gives them a womb-like cavity. They are
allotted intimate, introvert roIes: the musical bow or a hunter's
bow whose string is gently struck with asIender stick has here
and there the reputation of inducing contact with benevoIent
spirits. 159 Among non-stringed instruments, a widespread importance in courtship is assigned to the jaw's harp, whose eIastic
tongue vibrates between the player's jaws andreceives pitch and
resonance from the changing form and size of his mouth. 160
SOME INSTRUMENTS have conflicting characteristics of either
sex and thus are ambiguous or contradictoryin their connotations.
A trumpet made by cutting off the apex of a conch sheil is
masculine in virtue of its aggressive, frightening sound; but as it
derives from a water animal and in its slit and lips reminds of
a woman's sex organ, it is feminine as weil. Thus being androgynous, it serves at one pIace for solar, and at another place
for lunar cults, here for boy's initiations and there for women's
mystic dances which a man is forbidden to see.
The change can also occur through a dash of form and playing
action: drums and wooden slit-drums, rounded and hoilow, are
feminine as forms, but masculine because the primitive sees in
the beating act a symbol of cohabitation. Either instrument
deve10ps from feminine to masculine connotations, from the
predominant consideration of form to that of the act. 161
But even without any changing or conflicting sex characteristics, all instruments become in their connotations increasingly weaker with the weakening or reversing of magic
beliefs. They then assurne new sodal rules, be it within their
indigenous environment or in foreign places to which they have
been exported.
Two impressive examples of revaluation are the bull-roarer
and the gourd rattle.
On the lowest culturalleve1 of Australia, the bull-roarer, as a
ghost's voice, frightens the women away when boys are being
Curt Sachs, I.c., pp. 56 ff.
e.g. A. L. Kroeber, Peoples 01 the Pkilippines, 2nd ed., New York, I943, p. 221.
- R. F. Fortune, Sorcerers 01 Dobu, London, I932, p. 29.
161 Sachs, I.c., eh. I.
159
160
EARLY MUSIC
97
circumcised in the bush; a woman who has happened to see the

instrument must be killed. Indeed, the bull-roarer is supposed
to hold the forces of ancestors and to fecondate the wombs of
women. But in New-Guinea, on a somewhat advanced level,
girls (!) having reached maturity whirl it around as a love charm
to seeure a young man's affection. In early planters' civilizations,
it serves as a weather charm to attract both wind and rain; in
Malaya, it frightens elephants away from damaging the plantation ; and in a village on the Lido near Venice I once saw a boy
with a simple school mIer spun round on a twine. Here is, in one
short paragraph, the gamut of meaning from utmost secrecy to
open magie, and over sheer utility to children's fun.
The gourd rattle runs through a shorter gamut. Where shamanism comes to an end or women no longer shake the rattle in
their original feminine rites, they leave it to the infants as an
irresistibly fascinating toy - its way from places of worship to
cribs in the nursery is but one step.
How quick the transition to a children's toy can be is c1early
shown by a custom in French Upper Guinea: newly circumcised
boys, dismissed from their retreat, perform a ceremonial dance
which they accompany with a kind of sistrum especially made
with little disks of gourd stringed on a shaken bow. Then they
drop their instruments to be picked up by the children. 162
THE ALL-DOMINATING CONTRAST of the sexes also lends a
hand in shifting certain instruments from the pristine concepts
of mere sound to the pre-melodic contrast of high and low, even
if these two qualities have not yet definite pitches or adefinite
distance from one another. On the Orinoco, Gheerbrant saw all
instruments in pairs of unequal lengths, "the male and the
female." 163 The dwarfish aborigines of Malacca have in either
hand a stamping bamboo tube; the right tube, higher in pitch,
would give one blow quickly followed by three beats of the left
and low-pitched tube. The higher tube, a little shorter, is called
father or man, the lower and taller one, mother or woman. This
is bewildering: we, today, would invert the names and give the
169 Andre Schaeffner, Les rites de circoncision en pays Kissi in Etudes Guin4ennes,
no. 12 (1953), p. 52.
163 Gheerbrant, l.c., p. 94.
98
EARLY MUSIC
male connotation to the taIler, low-pitched tube, since men are

higher in stature and lower in voice. The aboriginal ideas are
different: the primitives are, above aIl, aware of the aggressiveness of higher pitches and the passive, more subdued character
of a lower one. The same is tme of quite a number of regularly
coupled instmments: in the New Hebrides, the biggest slit-drum
is called the mother (which may also be due to matriarchal
concepts), and in the Islamic belt the larger of two kettledmms
is defined as female (this time in the absence of matriarehai
connotations). Only late did the sex connection reverse itself.
The two sexes lend their names even to the two lips of a slitdrum (although they are originally, seen from an anatomical
viewpoint, strict1y feminine); drum sticks and the very beats
follow the same dualism: the Japanese call the right stick 'male'
and the left stick 'female' ; 164 and American jazz percussionists
call the left-hand 'hard' beat mama, while the right-hand 'soft'
beat is the daddy.165 The Olombo of the Upper Congo dub the
open string of their two-stringed zithers nyang6 or mother,166
evidently without bestowing the name of father upon the other
string which is meant to be stopped.
Nor should we forget that in the Far East the twelve pitches
within an octave (l's in Chinese) are alternately male and
female and, with them, the stone slabs, bells, and pipes in which
they materialize; that the ryo mode on C is male, and the ritsu
modes on D are female; that the satendro gender of Java is male
and the peloggenderfemale; and that the melodie speech inflections
of China are subdivided into male and, a tone lower, female groups.
A generally accepted western remainder of sex dualism as a
metaphor are the 'masculine' and 'feminine' word endings in
rhymes of our poetry, according to whether the accent is on the
last or the penultimate syllable - probably because the accented
ending sounds stronger and more decisive.
THE MALE-FEMALE CONCEPT illustrates an important stage in
the history of instmmental music. While on an archaic level
164
F. T. Piggott, The music and musical instruments 01 Japan, znd ed., Yokohama,
I909, p. I6z.
Cf. Harold Courlander introducing EFw FE 450Z (P 50Z).

J ohn F. Carrington, A comparative study 01 some Central All'ican gong-languages,
Ph.D. diss. Univ. of London, Brussels, I949, pp. 37 ff. (Courtesy Rose BrandeI).
165
166
EARLY MUSIC
99
instruments serve basically to mark rhythm or to utter mere

sounds without vying with voices to produee melodies, we wateh
in the stage of sex eontrasts an enriehment of the instrumental
language by opposing high and low. The 'father' and the 'mother'
of a pair of stamping tubes might produee them; or the two lips
of a slit-drum; or the two sections that a stopping finger ereates
on the string of a musical bow.
lt will not always be easy to decide whether or not this premelodie distinction and enjoyment of high and low is a natural
evolution from one-tone eoneepts to melody or rather the retroeessive adaptation of higher, melodie forms from some advaneed
civilization down to the lesser needs and potentialities of primitives. The eonclusive eriterion will probably be provided by
some alien detail in the non-musical workmanship of the
instrument. Where every trait eonforms with the primitive level
of the home civilization, the musical feature is almost eertainly
genuine; where an ornamental earving betrays foreign influenee,
the indigenous eharaeter of the music is open to doubt.
Of aetually melodie instruments we meet, among the first,
the ubiquitous flutes. At that stage, they are not too long and
not too short, not roaring and not shrill, but soft-spoken and
of medium size; and one feature lifts them to a level where they
ean eompete with human voices and begin to sing: fingerholes,
opened and closed, vary the piteh-producing length of the flute
and allow it to play in melodies, brisk and longing, staeeato,
legato, by steps and by leaps, in tender pulsations and foamy
easeades.
The holes appear in strange arrays. lt is obvious - as every
violimst or guitarist knows - that two musical steps of equal
width require different strides on the instrument, with the
seeond smaller than the first one, and so forth. This applies to
finger holes as mueh as to frets: their progression should be, in
mathematieal language, 'harmonie' rather than 'arithmetic.'
But no array of holes proeeeds in a harmonic progression; the
holes are equidistant throughout or else arranged in two groups,
either one with equidistant holes and separated from one another
by a somewhat larger spaee. And often the holes begin only in
the exaet middle of the flute and leave the upper half intaet.
This is astrange abdieation of the aural to the visual sense. I t
100
EARLY MUSIC
would be incomprehensible were not the individual holes quite

often musicaIly corrected' by increasing or diminishing their
normal sizes. It is, to be true, not only the visual but
also the tactile sense to which the ear must yield: the holes
are always at an easy fingering distance, not excluding the
larger gap between the two groups of holes, which is exempt
from fingering concerns, since either hand is assigned to one
group.
The easiest fingering distance, not too long and not too short,
is about an inch - a fact that shifts the whole question from the
visual, tactile, and acoustical to the metrical field. All over the
ancient Orient, the measuring unit was the foot with its nine,
ten, or twelve inches as the subdivisions. Far from being a merely
utilitarian device, as it is in our own civilization, the ancient feet
were venerable symbols of order, rule, and cosmic unity important enough to be engraved as etala on the sodes of royal
statues in Mesepotamia; and, forcibly expressed in figures and
ratios, they reflected the mystic power of numbers, immanent in
the tenets of both religion and philosophy from the Pacific to
the Mediterranean. Philo, the Jewish-Hellenistic thinker, who
lived in Alexandria at the beginning of the first century A.D.,
has given this shorl formulation of numeric control: "Whatever
is not considered worlhy of being conceived by a number, is
profane and not sanctified; but whatever is based on a numerical
value is recognized as already proved." 167
But feet and inches were rarely consistent; the feet had
different length and a different number of inches. They changed
from country to country, from dynasty to dynasty, from ruler
to ruler. As a consequence, the flutes as weIl were substantially
different in length and fingerhole distances. In many cases, a
careful comparison with the various ancient feet and inches in
metrological records allows to identify the provenience and
approximate date of a flute. On the other hand, many flutes have
undergone a disturbing diffusion from high to lower civilizations
which themselves had no conception of measuring standards.
Where such was the case, the flutes lost exactness to thoughtless
I
167 Philo, Posterity 01 Cain, 28, quoted after Hanoch Avenary, Magie, symbolism
and allegory 01 the old-Hebrew sound-instruments, in Collectanea Historiae Musieae,
vol.
(1956), p. 23.
EARLY MUSIC
101
copying and to casual measuring by the maker' s digits and

span. 16S
Even outside the concem of hole-drilling, tuning seems to be
a visual and tacti1e procedure. Panpipes, those rafts or bundles
of stopped flutes without fingerholes, each rendering one different pitch, are adjusted without re1iance on the ear; the players
rather introduce a measuring rod to take their sounding (a
word particularly inadequate in this case) and rests satisfied
when the depth is exactly like the one of the model pipe. To
quote the best example: on the Solomon Islands in Melanesia
this is carried out in several days of ceremonies by reproducing
painstakingly the paragons in the possession of the grandchieftain. 169 As late, culturally speaking, as the European Middle
Ages, all the many treatises on organ building or the casting of
bell chimes simply teach to proceed from pipe to pipe or from
bell to bell at the ratio of 9 : 8 for the wholetone distance (or of
3 : 2 for the fifth, and so forth), measuring there the mere
lengths and here the mere weights. Actually, pitch depends on
more than just a single measuring unit, and musically such
unilateral way of tuning can never yield satisfactory results.
I t is important to stress these facts. Any attempt at measuring
a couple of ancient or exotic flutes for the purpose of finding the
'scale' used in a certain age or by a certain people must ultimately
fail. If there is such a scale, it is the player's, not the maker's
concem, even when the two are one person. He achieves it with
the strength and direction of his breath, with the tension of his
lips, with a movement of his tongue, and with a complicated
fingering that might cover the lowermost hole only in part or
open one hole between two c10sed holes. The musical imperfection
of a non-musically perforated instrument allows the flutist almost paradoxically - a welcome freedom of intonation which
168 An excellent description of such measuring and burning procedures in East
Africa can be found in (Margaret Trowell and) K. P. Wachsmann, Tribal crafts of
Uganda, London, x953, p. 339. - Cf. also, for the Provent;:al galoubet, Claudie MarcelDubois, Extensions du domaine d'observations directes en ethnographie musicale
tranyaise, in Les Colloques de Wegimont, vol. x (X954-'5S), Paris-Bruxelles, X956, p.
X09
169 Erich M. von Hornbostel, Die Musik auf den nordwestlichen Salomo-Inseln,
in R. Thurnwald, Salomo-Inseln und Bismarck-Archipel, vol. x, X9X2. - id., Die
Maassnorm als kulturgeschichtliches Forschungsmittel, in Festschrift fr Pater Wilhelm
Schmidt, X928, pp. 303 ff. Curt Sachs, Les instruments de musique de Matlagascar, in
Universite de Paris, Tl'avaux de l'Institut d'Ethnographie, vol. 28, X938, pp. X4-23.
102
EARLY MUSle
no player of a modern Boehm flute enjoys. He is independent

of the tyrannie authority of a preconceived, demanding system.
THE OLDEST FOOT of China that we know measures 230 mm; a
stopped pipe of this length produces a pitch very elose to I'
sharp. Erich M. von Hornbostel found this pitch so often on
panpipes and other wind instruments between East Asia all the
way across the Pacific to South America that he safely exeluded
mere coincidence. Indeed, he found that all over this enormous
territory the scales available on individual instruments were
different from ours but conspicuously similar among themselves.
He saw an explanation in the two assumptions that (I) to tune
the canes of a panpipe the easiest way is overblowing the largest
cane and thus producing a twelfth ~hich, transpos~d ~own by
an octave, provides the fifth; and that (2) this overblown fifth in
stopped pipes is slightly shorter than the perfect fifth, so that
any tuning due to a cycle of fifths and fourths yields of necessity
steps shorter than 'Pythagorean.' 170
This sensational Blasquintentheorie was contradicted by the
late Manfred Bukofzer, whose physical measurements showed
that overblown fifths have very different degrees of deviation
from perfect fifths according to their size. 171 Another weakness
of this ingenious theory is that the way of tuning, wherever
described, consists in comparing the lengths of tubes, not their
pitches.
N evertheless, even with methodical flaws, the agreement in
tuning over such enormous stretches is impressive enough to
forbid a mere shelving of Hornbostel's astounding results.
THE TUNING OF STRINGS and chimes follows other principles.
Stringed instruments are on the whole of two kinds: those with
open strings and those with stopped strings. Lyres, harps, and
zithers belong to the first group, and lutes to the second. There
are however lyres, harps, and zithers with strings occasionally
170
569 ff.
Von Hombostel's first communication in Anthropos, vols. I4/I5 (I9I9), pp.
171 Manfred Bukofzer, Przisionsmessungen an primitiven Musikinstrumenten, in

Zeitschrift tr Physik, vol. 99 (I936), pp. 643 ff. - id., Kann die Blasquintentheorie
zur Erklrung exotischer Tonsysteme beitragen?, in Anthropos, vol. 32 (I937), pp. 402 ff.
- id., Blasquinte, in Die Musik in Geschichte und Gegenwart, vol. I (I95I) col. I9I8 ff.
- Jaap Kunst, Around von Hornbostel's theory ot the cycle ot blown titths, in Koninklijke
Vereeniging Indisch Instituut, Mededeeling 76, Amsterdam, I948.
EARLY MUSIC
103
stopped, and lutes with open strings. Since most of all these
instruments are post-primitive, it might suffice here to say that
the first group is usually tuned by ear in a cyde of fifths and
fourths, and the second group is ruled by a 'divisive' system,
half the string yielding the octave of the 'open' note, a third
of its length the fifth, a quarter the fourth, a fifth part . the
major third, a sixth part the minor third. 172
But there are many more ways of tuning, and some of them
quite unexpected. In Uganda, harps are not only 'well-tempered'
but also 'isotonic' : one Ganda people divides theoctave into
five, in principle equal, steps of around 240 cents, which corresponds to the satendro gender of Java's and Bali's gamelan
orchestras. We might call this tuning 'iso-pentatonic.' Another
Ganda people divides the octave into four equal steps, of a minor
third each, which makes it 'iso-tetratonic.' This, of course, is
highly suspect of oriental influences, although we are not yet
able to trace their exact diffusion routes.
To achieve isotony, one could (r) start from the octave as a
given whole and divide its tonal space into equal parts, four or
five; or else (2) advance in equal steps until the octave is reached.
The first procedure is not feasible; the ear in all its frailty cannot
divide an octave into a given number of equal parts. The second
course is problematic, too; a unit which, multiplied by four or
five, leads to a perfect octave is not easily found. And yet this
second method is a fact. The performer proceeds from the unit
to the octave, which however yields approximations only and
requires subsequent corrections.1 73
In a special paper,174 K. P. Wachsmann, then curator of the
Uganda Museum in Kampala, describes the actual procedure to
the last detail. The harpist pulls the highest string to about g'.
Next, he tunes the adjacent string to a distance of about 240
cents without any standard other than his tonal memory and goes
this way down over all the eight strings. This method, it is true,
requires control and adjustment. To achieve it, the player
rapidly plucks the strings r, 2, 3, which in succession yield
Curt Sachs, The rise 01 music, New York, 1943, pp. 71 ff., 75 ff.
K. P. Wachsmann, l.c., pp. 4 ff.
174 K. P. Wachsmann, An equal-stepped tuning in a Ganda harp, in Nature, vol.
165 (1950), pp. 40 ff. - id., A study 01 norms in the tribai m1tsic 01 Uganda, in Ethnomusicology, Newsletter no. II, Sept. 1957, pp. II ff.
172
173
104
EARLY MUSIC
about 480 cents or almost the 498 cents of a perfect fourth. He

continues in a similar way with 2.3.4 and 3.4.5 down to 6.7.
8. all these groups being expected to affect the ear as nearly
identical fourths. An easy second control tests the octaves between 1 and 6. between 2 and 7. between 3 and 8. With all this
reliance on mere hearing accuracy. the maximum deviation from
normal that Wachsmann found with the help of a modem
stroboscope was. on any string. no more than fifteen cents or
the negligible 13.6th part of a tone.
THE ISOTONIC TUNINGS in Uganda have astonishing parallels
in African xylophones: two South African peoples. Bapende and
Chopi, give their xylophones an 'iso-heptatonic' arrangment
with seven equal steps of about 171 cents in an octave. which
amounts exact1y to the current genders of Siam and Burma, that
is, of another region in Southeast Asia. This striking coincidence
has of course attracted attention. 175
From a methodical viewpoint, it is instructive to hold the
Chopi xylophones, as they actually sound, against H. A. Junod's
very different notation of the eighteen nineties, which makes
believe that these instruments are tuned in a neat European
tonality, be it E ttat minor or G ttat major, however you want
to read it.l 76 Such comparison amounts to a good waming of
the treachery of our ears as weil as of western staff notation,
and hence, alas, of most of our printed music examples.
These South African xylophones lead us into a third domain
of melody-carrying instruments: percussive chimes, or sets of
ready-made elements, be they slabs, be1ls, gongs, bowls, and
here and there even drums, each element rendering one single
note and all together constituting a scale. The set is finished by
the instrument maker to the minutest detail; the player must
obey his tuning, although he is allowed to make a few minor
175 Cf. Jaap Kunst, A musicological argument /01' cuUural relationshiP between
Indonesia - probably the Isle 0/ java-and Central A/rica, in Proceedings 0/ the Musical
Association, Session 62 (I936), translated into German as Ein musikologischer Beweis
/r KuUurzusammenhnge zwischen Indonesien - vermutlich Java - und ZentralA/l'ika, in Anthropos, vol. 3I (I936), pp. I3I ff. - Hugh T. Tracey, Chopi Musicians.
Their music, poetry and instruments, Oxford, I948, pp. I2I ff. - Ernest Haddon,
Possible O1'igin 0/ the Chopi timbila xylophone, in the Ajrican Music Society NewsletteI',
vol. I No. 5 (I952), p. 6I. - In a morereserved attitude: K. P. Wachsmann, Musicology
in Uganda, in joul'nal 0/ the Royal Anthropological Institute, vol. 83 (I953), pp. 50-57.
176 Henri-A. Junod, Les Ba-Ronga, Neuchte1, I898, p. 264.
EARLY MUSIC
105
alterations during changes of temperature by adding appropriate

lumps of wax or by chiseling matter away.
The roots of chimes reach down to prehistoric, pre-melodic
levels, when two or three and exceptionally more rough-hewn
slabs were loosely laid on the stretched-out legs of a person
sitting on the ground and struck with two knobbed sticks. But
whether these primitive leg xylophones, in Melanesia and westward to Indonesia, are actually beginnings or rather retrogressive
forms is hard to decide. N or do we know from reports at hand
whether or not a genuine, well-tuned scale is planned. Where on
the contrary the slabs are shorter, more numerous, and solidly
assembled in some portable framework or on a cradle-shaped
resonance box, as in the Southeast of Asia, in Africa, and in
Negro-America, we can be sure of a careful tuning in the scalesystem of the country (usually done by cutting out the surface
of the slab, which is done in the middle for flatting the tone,
and near the end, for sharping it). This applies to the principal stations along the great migration route: from Indonesia to Southeast
Africa and from there with the black slaves across the Atlantic
to Central America, plus minor branches to Japan and Europe.
Everywhere the playing technique is highly developed; and
the often incredible virtuosity of the performers has created
a genuine, idiomatic instrumental music which we can best
compare to our western keyboard toccatas. In this respect, the
next of kin of the xylophone - the lithophones, metallophones,
and crystallophones -lag somewhat behind; and so do the second
cousins: the drum chimes, porcelain bowl chimes, 177 gong chimes,
and .bell chimes. The unmuffled resonance, particularly of the
metal, exc1udes the uninhibited, agile technique that the dry
and short-breathed wooden slabs invite.
As an aside: metallophones and gong chimes, chief instruments
of Indonesian music, do not occurin Madagascar,178 which was
an Indonesian colony. Since the last wave of Malayo-Polynesians
seems to have reached the island off Southeast Africa as late as
the fifth century A.D.,179 we must presume the non-existence
of bronze sets at that time.
177 EFw FE 4409 (P 409) 11 I (India).
178 Curt Sachs, Les instruments de musique de Madagascar, in Universite de Paris,
Travaux et Memoires de l'Institut d'Ethnologie, vol. 28, Paris, 1938.
179 Ralph Linton, The tree 01 culture, New York, 1955, p. 203.
106
EARLY MUSIC
THE SUBJECT OF LITHOPHONES cannot easily be examined

without a glance at the curious stones, excavated early in I949
in Indochina and shipped to their present quarters in the M usee
de I'Homme at Paris. 180 Rarely, musical instruments have been
given a publicity like' this one; musicological periodicals and
even dailies have view in conferring on these venerable stones a
"megalithic" age and "a complete Eastem or Javanese pentatonic scale."
After such magniloquence, it is almost a disappointment to
read the more cautious report by Andre Schaeffner, music
curator of the Museedel'Homme. 18l Schaeffner is careful enough
to say that (from his viewpoint) nothing allows to date the
stones; they may be a few centuries old or several millenniums.
In the discussion of these stones we face two apparently
contradictory qualities; they have been chipped and hence prepared in a technique officially known as palaeolithic; but tuned
chimes, to the best of our present knowledge, nowhere occur
before the Bronze Age.
The conflict might be alleviated by the modem loosening of
the time-honored equations: chipped stone = palaeolithic,
polished stone = neolithic. "This classification was one of the
first to break down." 182 We knowwell-polishedimplementsfrom
the Magdalenian (late Palaeolithic) and, inversily, chipped
artifacts from the Neolithic, and Linton mentions expressly that
Southeast Asia's Neolithic still knew chipped stone implements,
much as they were rare and crudely shaped. 183
As to the stones of the Musee de l'Homme, the archaeologists
leave no doubt that their findis neolithic "de facture bacsonienne."
The musical side of the question is still more complicated.
The trove itself consists in eleven disconnected, chipped slabs
of sonorous stone in various sizes, the largest being IOI.7 cm long.
As to the allegedly pentatonic scale, we must for the time being
rest satisfied with non-committal and contradictory measure180 G. Condominas, Le lithophone prehistorique de Ndut Lieng Krak, in Bulletin
de l'Ecole Franyaise d'Extdme Orient, vol. 45, 2 (1952), pp. 359-392. - Cf. also:
Bemard Fagg, The discovery 0/ multiple rock gongs in Nigeria, in Man, vol. 56 (1956),
pp. 17-18, and African Music, vol. I, 3 (1956), pp. 6-9.
181 Andre Schaeffner, Une importante decouvertearcheologique: le lithophone de Ndut
Lieng Krak (Vietnam), in Revue de Musicologie, vol. 33 (1951), N.S. pp. 1-19.
182 Me1ville J. Herskovits, Cultural anthropology, New York, 1955, p. 43.
188 Ralph Linton, The tree 01 culture, New York, 1955, p. 174.
EARLY MUSIC
107
ments taken at the museum, which in their divergences can

perhaps be explained by the quantity of irregular partials above
the actual, intended pitches. The two measurements taken with
undisc10sed apparatuses, but not with an electronic counter,
by three scholars and published by Schaeffner, vary at times by
no less than 134 cents or two thirds of a wholetone; and among
the steps from slab to slab one distance reaches only 18 (or, in
the other measurement, 24) cents, which amounts to more or
less a tenth of a tone and does not fit in any conceivable system
of tuning, pentatonic or non-pentatonic. Possibly, it has been
suggested, the stones do belong to two lithophones, not just to
one.
Moreover, we read in the archaeological report: 184 "It is
possible that originally the longest slab was even longer." and:
"It is hard to say whether one of the slabs (lame XI) is not the
remnant of a more important element or a defective, broken
slab, or whether it represents in itself a piece of which we ignore
the role." And on the following page: "Two slabs present
important breaks. One end of slab V is completely broken, and
the accident has taken away two other pieces on either edge of
the slab; one of them has left a cut 17 cm long, and the 9ther one
15.51ong - a break, it seems, posterior to the manufacttire of the
slab and causing a false sound. Slab II is damaged on either
end ... "
Therewith, any discussion of a scale becomes mere speculation.
As matters stand, I am at this stage inclined, not to doubt 185
whether these stones were a musical instrument, but whether it
was meant for actual musical performance. They were found in
an unsuitable, upright position against the wall; they have
apparently none of the forms known from other sounding stones
discovered between the Carolines in Micronesia, the Far East,
Chios in the Mediterranean, Ethiopia, and Venezuela; 186 unlike
anyone of them - the kawa bowls on Ponape (Carolines) excepted
- they have no hole for passing a suspension-cord. The makeshift
playing arrangement that the Musee de l'Homme has given them
- horizontal on two longitudinal supports like those ofaxylo184 Condominas, l.c., p. 6.
185 As does Fritz A. Kuttner, Nochmals: Die Steinzeit-Lithophone von Annam, in
Die Musikforschung, vol. 6 (1953), pp. 1-8.
186 Curt Sachs, Geist und Werden der Musikinstrumente, Berlin, Ig~g, pp. IIg ff.
108
EARLY MUSIe
phone - does nowhere occur where stones are involved. All

considered, the only musical datum is the fact that these like
many other stones yield definite, different tones when struck
with a hammer. Even if they were meant to be struck and to
sound in some ritual, there is no indication that they were
used in scale-like tuned chimes. 187
BEFORE DISCUSSING INSTRUMENTAL MUSIe, we should briefly
touch upon the problem of general distribution. A panoramic
view discloses a number of areas strangely distinct in their
instrumental possessions and predilections in spite of diffusion
from country to country and across the seas.
Indonesia shows the sharpest outline: tuned idiophonic sets,
like xylophones, metallophones, and gong chimes, have an
uncontested leading role and (together with one single drum),
often unite to form the famous gamelans or orchestras. Strings,
and wind instruments are by no means absent but mostly due
to Muhammedan incursions around the middle of the second
thousand years A.D. and hence of minor importance. In the Far
East and westward to Burma and India, idiophonic chimes
occur but have an essentially weaker position. Only the xylophones, hardly used in the Far East and unknown in India,
crossed the ocean, reached N egro Africa, and from there were
taken to the West Indies and Central America.
India gtves a leading musical role to the drum. This role is
slightly weaker in the Middle East and more or less accessory
in the rest of the world.
This statement excepts Black Africa, where the drum in
solos and ensembles seems to be almost indispensable. Yet there
is a strong possibility that in Africa the drum belongs in a later
187 Since then, however, there have been found in the same region (theremnants
of) two more of such lithophones. One of them, originally heptatonic (but one of the
slabs was totally crushed by the bulldozer that brought the instrument to light)
is now in the possession of Mr. Claire Omar Musser, Los Angeles. Mr. Musser who is
an expert in acoustics, sent me the measurements of the intervals, made with a
Strobokonn. The scale has nothing in common with the Parisian instrument:
1473.6
I241.09
II73
Io087
959.2
841.3
I 297 II 971 III 2611 IV 871 V 2271 VI

The other instrument, according to the population of the village where it was
found, originally had had six slabs, three of which had been lost in the course of the
years. The remaining three slabs were held in high honor and awe and were sounded
only at special occasions. (l.K.)
EARLY MUSIC
109
stratum; in several cases we hear that a certain tribe owes its

drums to an import from the north. "Indeed, among the Bashi
no particular emphasis seems to be placed upon drums and
drumming; these percussion instruments are re1atively rarely
used '" One seldom hears drums being played about the
countryside ... Singers, with some exceptions, do not call for
drums nor seem particularly pleased when a drummer happens
to be present." 188 Among the Gikuyu in Kenya "the use of drums
as musical instruments has been adopted only in recent years;
the idea was borrowed from the Wakamba, who are the eastern
neighbours of the Gikuyu." 189
Below this possibly later stratum of drums, we behold a
strange prevalence of stringed instruments, from the simplest
musical bows, ground harps, and ground zithers way up to lyres,
harps, and zithers in many forms and hybrids. A number of them
are c1early dependent upon old-Egyptian and Arabian diffusion;
but many types do not occur anywhere else and must be considered indigenous.
Once more, we see old Europe in a certain agreement with
N egro Africa: the earliest literary sources mention harps and
lyres in the north; prehistorians have found the strangest horns
of bronze-age character, as the Baltic lurer and the side-blown
instruments from ancient Ireland ; modern folk usage still knows
zithers and fiddles as weIl as rustic horns, trumpets, and flutes.
Incidentally: the side-blown Irish horns have their only paralleis
- in antler and ivory - all over N egro Africa. Oboes and c1arinets
are originally absent from either region.
The Pacific Islands, inc1uding Australia, are comparatively
poor in instruments. With the exc1usion of panpipes, their
instrumental music is not important enough to settle in definite
patterns. The same applies to American Indians with their strong
predilection for flutes and their apathy to stringed instruments
if we except the Spanish import of recent ages.
Altogether, the diffusion of instruments across the seas and
along the caravan routes has been more essential than the
diffusion of singing styles, as they are material goods. We can
188 Alan P. Merriam, Musical instruments and techniques oj perjormance among
the Bashi, in Zaire, vol. 9, 2 (I955), pp. I23, I31.
189 Jomo Kenyatta, Facing Mount Kenya, London, I953, p. 95.
110
EARLY MUSIC
follow their traffic from Prehistory up to the modern regrettable,

destructive trade in accordeons and mouth organs all over our
globe.
Furlher interpretation of this sketchy survey must be left to
future detail research.
of vocal and of instrumental music
are utterly different. The instrumental impulse is not melody
in a 'melodious' sense, but an agile movement of the hands which
seems to be under the control of a brain center totally
different from that which inspires vocal melody. Altogether,
instrumental music, with the exception of rudimentary rhythmic
percussion, is as a rule a florid, fast, and brillant display of
virtuosity. Unlike the distinct and clearcut melodies of voices,
it consists of quick figures and passages, which often remind the
listener of seventeenth century 'divisions.' Quick motion is not
merely a means to a musical end but almost an end in itself,
which always connects with the fingers, the wrists, and the whole
of the body: the inventive power of a drummer, for instance,
depends on the nature and particularly on the beating position
of his drum and therewith on the position of the hands and their
striking arc. A drummer who holds the drum like a hobby-horse
between his legs plays rhythms different from those which he
would perform on a drum he squeezes under his shoulder or
carries horizontally in front of his abdomen.
Consequently, the vocal and the instrumental expression of a
tribe are never one in style. They shape two separate arts.
N evertheless, the beginnings of instrumental music remind
in a way of the early vocal one-step melodies. We find this onestep pattern in the female-male contrast of two stamping tubes
or the two slabs of rudimentary xylophones in Melanesia. Strangely enough, the one-step music has been preserved all the way up
to the most sophisticated xylophones. The amadinda of Uganda,
for example, is played by three men, one of whom faces the other
two; and while two of them perform their complicated parts,
the third plays only on the two highest keys about a wholetone
apart. 190
THE ORIGINAL CONCEPTS
190 ]oseph Kyagambiddwa, Atrican music trom the sources ot the Nile, New York,
1955, p. II7 and examples 101-162.
EARLY MUSIC
111
The same applies to the Negro regions of Madagascar. The

xylophone of the Sakalava in the west consists in an unusual
number of disconnected slabs arranged across the stretched-out
legs of a woman. A second woman, sitting at right angles, beats
only two of the slabs, while the third and the first one strike the
melody proper. 191
Even zanzas (cf. later in this book) follow in Central Africa
the principle of a one-step upper voice part. In a piece of the
Batwa in Central Urundi, transcribed by Rose Brandei, the
first of three cooperating zanzas plays nothing but a-b (excepting
one single g). 192
Yet another elementary form of instrumental music is the
one-sided; scale-like advance from neighbor to neighbor, from
slab to slab, from fret to fret, from fingerhole to fingerhole.
X. RHYTHM AND FORM
Rhythm and form are the two organizing powers of melody.

In recurrent patterns they dam and divide its flow; they control
its tension and relaxation; they balance its law and its freedom.
The concepts of rhythm and form overlap and cannot be
strictly kept apart; rhythm is as much a form quality as form
is a rhythmical quality. But in current terminology, we speak
of form when we want to discuss the organization of an entire
piece or movement, and of rhythm, when we analyse a smaller
section which we easily perceive and remember.
Rhythm in its beginnings is extra-musical.
Melody does not of necessity depend upon rhythm; like the
songs of many birds, it can exist without recurrent, strictly
gauged patterns of organization. Solo singing on the oldest
levels confirms it; in many cases "we are at a loss to detect a
binding principle or pattern in any western sense except for
an emotional stress here and there or for the to and fro, however
irregular, of tension and relaxation." 193
The inroad of rhythm is due to ourselves, to our bodies and
191
192
19B
Curt Sachs, Les instruments de musique de Madagascar, Paris, 1938, pI. XIII.
Transcribed from Riverside: Voice of the Congo, RLP 4002, Bd. 9.
Curt Sachs, Rhythm and tempo, New York, 1953, p. 36.
112
EARLY MUSle
our minds. To the voices Plato assigned melodie structure or,

as the Greeks called it, harmonia; but rhythm he allotted
to the artful motion of bodies. We saw a similar antithesis in
a quotation from the Roman rhetor Marcus Fabius Quintilianus.
Almost two thousand years later, Karl Bcher, the sociologist
and economist (d. 1930), stated in his often reprinted Arbeit und
Rhythmus that rhythm was basically not a quality inherent in
music or poetry but only in bodily movement. Rhythmical
structure of verses and melodies, as a consequence, was nothing
but a transfusion from the moving body, although, we might
add, from a body consciously moving under the firm control
ofthemind.
Rhythm springs from man as a slowly developing psychophysiological urge. Its physiological part is the impulse to
equalize continuous, regular movements, such as walking,
dancing, running, or rowing. Psychological is the awareness of
greater ease and gusto through constant evenness in motion.
Hence we experience the often irresistible need to impose such
evenness on every succession of acts and movements, if they can
be automatized and removed from wilful change (which excludes
all ordinary speech and any doings that demand the worker's
permanent, watchful attention).
To those interested in rhythmical problems, poetry might be
the most familiar carrier of regular rhythm. But taken in its
precise, metronomical sense, poetical rhythm would drown the
verses in a soulless, meaningless, antipoetic scanning. All
versification depends on irrational respiration with a meaningful
lengthening and shortening of individual syllabies. As a spoken
art, it requires a freedom in which both the stresses and the
meters are !ittle more than merely suggested.
But the central field of rhythm is music. Singing and playing
are indeed successions of acts; they can to a certain degree be
called automatie; and they have a high degree of ease and gusto
both as a cause and as a result of rhythm. Consequently, even
the most primitive music has a trend towards regular time units
or chronoi protoi, albeit not always in the sense of a ticking
metronome. This irresistible regularity retroacts upon man, its
creator, and, by way of his mind, to the musc1es. The archaie
Vedda in inner Ceylon allow us to behold the beginnings of such
EARLY MUSIC
113
reciprocity; they accompany their songs with a slightly roclcing

trunk or a leisurely moving toe. 194 Needless to say, dances, too,
belong in the field of kinetic reaction.
Early, an urge develops to make the bodily reaction an influential part of music: the muscles no longer follow in silent comovements, but are used to produce a noise and, in another
retroaction, to increase, by this noise, the equalization and
effectiveness of melody. Out of this reciprocation, mankind
proceeded to a sonorous, audible movement of the body, be it
slapping the buttocks or the chest or the flanks, stamping the
ground, or clapping the hands - activities of which the last two
are still common in Western folksong and in the kindergarten.
On a higher level, the striking limbs or the beaten body yield
to extensions: natural or artificial replacements intensify the
result. The numberless varieties of rattles and clappers, of pestles
and stamping boards, of tapping sticks and beaten gourds are
extensions of the hands, the feet, and the torso.
Rhythm, in music (as in poetry and the dance) an organization of lapsing time, is based on the keen awareness of time
units, equally spaced and ever repeating like the monotonous
ticktack of a clock. Such uniform and mostly rapid ticking is
often found in percussion, in clapping, rattling, and drumming,
along with the song or in an independent tempo of its own as
"an inexorable and mathematical background to the song." 195
It may be open to discussion whether or not such rhythm as an
aesthetic experience demands a transformation of purely
mechanical pulses into characteristic, stimulating patterns. But
in our field it is unavoidable to accept the rhythmical quality
of any simple pulsation, albeit on the lowermost level.
RHYTHMIC ORGANIZATION into clear-cut patterns - in music
as in poetry - assumes essentia1ly three different, however
overlapping, forms.
One form of rhythm - not the earliest - is purely numerical.
It consists in counting out and repeating again a number of
time units (or several numbers in a characteristic arrangement),
194 Max W ertheimer, Musik der Wedda, in Sammelbnde der Internationalen M usikgesellschaft, vol. 9 (I909), p. 300.
196 A. M. Jones, Atrican rhythm, in Africa, vol. 24 (I954), p. 28.
114
EARLY MUSIC
without depending on accent or meter, although such qualities

might be present or at least be suggested. In the Orient - as this
section will show - such counted rhythm is common in the Far
East. The West knows the counting principle from the endecasillabo or eleven-synable verse of Italian poetry, as in Dante, and
musically, from the folksong of Russia. 196 In the primitive world,
numerical rhythm has hardly any importance.
Another form of rhythm relles on actual or merely suggested
stresses; it groups aseries of even-spaced pulses by accenting
the first of two or more of them. This is the common form of
later western 'time' with its two-four, three-four, four-four:
x.x.x.x.x.x.x
x .. x .. x .. x .. x
x ... x ... x ... x
Ties, syncopations, triplets, counteraccents may disturb the
basic time-pattern as added spices; there also may be conflicting
episodes in some different rhythm for the sake of contrast. But
the straight and simple 'time' is always clearly recognizable and
can be easily beaten with abaton - most western music throbs
in even pulses.
In primitive music, two-four or binary time prevails, unless
some surplus synable in the text expands it to ternary time. All
over East Asia two or four-beat rhythms are almost exclusive;
man's two-foot stride might account for it. They rule supreme
in the Far East; and only Korea uses three beats here and there.
Binary rhythms, with
as the nucleus, play an all-important
role in children's and women's music. Their normal length is
eight units in allne, most often in 4 + 4. 197
Quintary rhythms are exceptional. Among the K wakiutl on
Nootka Sound, the half-initiates of the Sparrow Society strike
the sounding boards in front of the dance-house "in five part
rhythm, which is called 'one beat between.''' 198 As Franz Boas,
Curt Sachs, Rhythm and tempo, New York, 1953, pp. 29 ff.
Cf. Constantin BraiIoiu, La rythmique enlantine, Paris-Bruxelles, 1956. - Robert
Lachmann, Jewish cantillation and song in the isle 01 Dierba, Jerusalem, 1940, pp.
72 ff. - id., Musik der aussereuropischen Natur- und Kulturvlker, in Ernst Bcken,
Handbuch der Musikwissenschaft, Wildpark-Potsdam, 1931, p. 8.
198 Franz Boas, Ethnology 01 the K wakiutl, in Bureau 01 A merican Ethnology, 35 th
Report (1913-4) part 2, Washingte>n, 1921, p. 1I68.
196
19?
EARLY MUSIC
115
the reporter, was familiar with music, we must accept this

statement as correct; evidently there are five beats in what
we would call a measure; and according to the native term in
single quotes, this quintary time seems to be an irregular rhythm
in the eyes of the Kwakiutl. Incidentally, their neighbors, the
N ootka 199 and the Bellakula, have also drum-accompanied songs
in strictest 5/8 time,2oo which might confirm the report on the
Kwakiutl.
Downbeat and upbeat - the stress on the first or on the second
beat of the melody - carry in primitive music by far not the
significance granted them in the yvest of today. An overwhelming
majority of pieces start on the:downbeat, and an analysis of
the civilizations in which the upbeat occurs is not yet possible.
The situation is surprisingly reminiscent of earlier music in
Europe, where the upbeat has no evidences before the fourteenth century.201
Binary beats are occasionally submitted to 'metric alteration' ;
the first of two presumably even notes is slightly lengthened at
the cost of the second, just as it was in Europe up to the eighteenth century.202
Metric alteration can be invefted, by shortening the 'good,'
accented notes and lengthening the 'bad,' unstressed ones.
This so-called Scotch snap, in which the second note so to speak
recoils from the first note, is co~on among American Indians
and Siberians in their 'rumblingl styles. It also is common in
Hungarian folk music. Alteration appears sometimes in the form
ofthe Greekamphibrachicmeter ot 'short on both sides,' as fJ1
Altogether, Indians and Siberi~s (as well as Norwegians and
other folk musicians) are particularly fond of dotted and doubledotted, craggy rhythms in whatever form.
THE WORD METRIC, in these last few paragraphs, leads to the
third type of rhythmical patterns.
The vocables metric and meter fall all too easily victims to a
199 Helen H. Roberts and Morris Swadesh, Songs of the Nootka Indians, in Transactions of the American Philosophical Society, N.S. vol. 45, part 3 (1955).
200 C. Stumpf, Lieder der Bellakula-Indianer, in Vierteljahrsschrift fr Musikwissenschaft, vol. 2 (1886), p. 413, reprinted in Sammelbnde fr Vergleichende Musikwissenschaft, vol. I (1922), pp. 94, 95.
201 Curt Sachs, l.c., pp. III ff., 124,261 ff.
202 id., l.c., pp. 296-301. - Cf. EFw FE 4419 (P 419) I 2.
116
EARLY MUSle
lax and irresponsible terminology which gives them without

the slightest justification to measures from barline to barline
(as '3/4 meter') and to 'time' or accentual rhythm. A four-four
or a three-four are 'times,' as they obey a 'time-signature';
meter, on the contrary, belongs exclusively to some characteristic
pattern of longer and shorter notes. The longer ones may extend
over one pulse of the ticktack or more; the short ones may
subdivide such pulse into halves, thirds, or otherwise. The weIl
known verse feet of ancient Greek poetry (and vocal music)
provide the standard examples with their trochees and iambs,
dactyls and anapaests, spondees and paeons. But while Greece,
at least in theory, gave every short syllable the length of one
time unit or pulse, and the long one two of them, other countries,
such as India ancient and modern, give the long note from two
to seven units and feel entitled to assign to the short note more
than one unit if necessary, as, for example, in the dactylic pattern
of 7 + 2 + I units (an ambiguity that later recurred in the
western Middle Ages as longa per/eeta and imper/eeta, and brevis
altera and reeta) , the former two measuring three and two breves,
and the latter ones, three and two semibreves.
Meter, though often divisible. by two or three like a divisive
stress rhythm - as in the cases of the Greek daetyl (long-shortshort) or the molossos (long-Iong-Iong) is in principle 'additive.'
It can be best, and often exclusively, described as the sum of
different amounts of time units, such as 2 + 3, 2 + 4, 3 + 4,
3 + 3 + 2, or 2 + 2 + I + I. Such metrical patterns are
found quite early in percussive accompaniments. On the Carolines in Micronesia, for example, percussion would sometimes
proceed in I + 2 or 1+3 halves, that is, in an iambic meter. 203
The finest Amerindian examples are found among the N ootka
in Vancouver.
Outside India and the Middle East, such additive meters are
most important and vital in Bantu Africa. The Zulu have a
9 + 7 meter,204 as
x .. x.x ...
+ x.x .. x.
More frequent and of greater historical interest is the common

203
204
George Herzog, Die Musik der KaroUnen-Inseln, Hamburg, I936, passim.

EFw FE 4506 (P 506) 111 45 b.
117
EARLY MUSIC
pattern of 7
+5=
12;
twe1ve time units are grouped this way:
X.X.x ..
+ x.x ..
The 7 + 5 recurs in the Arabian'poetical meters of seven and

five time units and in the strange caesura of Roman hexameters,
such as, in Virgil's Aeneid:
Arma virumque cano I Troiae qui primus ab oris
-8
..
-81.848-6.
Jf'JJrJJ
JJJJjJJJ
or, inversely,
Infandum regina iubes
Irenovare
.,
dolorem
l-e.-411
JJJJJf"JJ
nJrJJJ
And it again recurs in the Far East. Japan distinguishes poems

in the haiku form, with three lines of five, seven, and five
syllabies, and in the tanka form, with five, seven, five, seven,
seven syllables. 205 In Chinese music, the combination of five and
seven beats appears even where the text follows some other
organization:
Ex. 27*
, r Eil ci r ICi er I F'? I J n I J d n Im

12
45.7
12
lj.
Cl ViF sJI t1 crlrylciPIJ Er1tU z
~'
84
84
These parallels, reaching from Japan to Bantu Africa, are

startling. But it seems premature to try an explanation.
Keeping quite independent from the singer's melody, rhythm,
and tempo modifications, the c1apping hands or the drum repeat
such patterns over and over again, even if their beating falls
between two syllables of the text.
After Van Aalst.
205 Curt Sachs, l.c., p. 56.
118
EARLY MUSIC
Africa also favors the ubiquitous meter 3

x .. x .. x.
+ 3 + 2:
x .. x .. x.
(in which the members can be exchanged, as 3 + 2 + 3 or

+ 3 + 3).
This "Fascinating Rhythm" - title of one of Gershwin's
piano pieces - has an important key position in nearly every
chapter of music history, from the Orient way up to our twentieth
century with hot jazz and the Charleston.
Whatever the rhythm, percussion seldom starts together with
the voice; it enters a few notes after the singer, or it precedes
him.
All these patterns, though, are later developments; and if we
hesitate to derive them from the Orient, l>ecause they play an
eminent role also in Africa, we should realize that in this case
as in others the Sudan and Bantu Africa are parts of the oriental
rather than of the primitive world.
N ext to Bantu Africa, additive rhythm prevails in Muhammedan countries and in India - not only in art music, but also
in folk songs. In southwestern India even children sing in rapid
4 + 3 = 7 units (our example r). A majority of Turkish pieces
follow a nine-eight measure in the arrangement 3/4 + 3/8; the
a
three eights do not form a triplet, as they would in ; J J
but three exact half-quarters as in ; J J J.; other Turkish
pieces have seven eights: J J J. or five eights: J J.
To thispicture we must add the intricate meters of Southeast
Europe:
2
m,
Ex. 28*
.~)ni)=270
i J J 3.11
JI)i I' ~ 1?~ J J JI _~ .,~ I'
L
in Bulgaria, Macedonia,206 Rumania, and Hungary, with up to

twenty one units.
Here as elsewhere we face the weighty question : does this
folk music derive from art music or vice-versa? I discussed it in
an earlier book without arriving at a convincing solution. 207 The
*
After Christo Obreschkoff, Das bulgarische Volkslied.

Yury Arbatsky, Beating the tupan in the Central Balkans, Chicago, 1953.
207 Curt Sachs, l.c., pp. 93-95.
208
EARLY MUSIC
119
question itself might be due to the bias of the modern West.

In our civilization, the two worlds of art and folk music are
rigidly separated. Art music - the product of desk work by men
of inteilect and systematic instruction - has little in common
with illiterate 'folk' melodies. This gap is almost non-existent
in the Orient. Music in ail social layers is practically scriptless
and non-intellectual, as tradition mIes at court as weil as on the
village level. The roots are the same, even if there are shades of
refinement. The Muhammedan Orient is most characteristic in
this respect: the best trained and most highly regarded 'art'
singers perform regularly, not in concert halls, but in urban
coffee houses for a thoroughly mixed audience, effendis no less
than common people, and the music they offer is the highest
available. In many cases the music that we are inclined to dub
'folksong' is rather 'tribal' and rural or belonging to one of the
ethnic minorities which often preserve a more archaic style of
melody.
arises: is there a law, or at least a principIe, to determine the growth or change of rhythm? Does rhythm
tend to become simpler and more regular within a certain
civilization, or does it on the contrary move to an ever greater
complexity?
Let us test a few examples.
In certain Amerindian communities, 'modern' songs are
said to have a more regular rhythm (and a smailer range) than
the 'old' ones. 208 There is no reason for doubting the statement
as such. Still, the contrast of old and recent within a primitive
man's quite personal memory is too insignificant where we deal
with traditions of ten-thousands of years. We should here refrain
from generalizations, and the more so as modern songs might
easily be influenced by white-American tunes.
Growing rhythmic rigidity, however, also occurs elsewhere.
Rhythm was free in the archaic style of India's Veda cantillation; but in the more recent style of this chant, dating from
about400B.C., uniform meters and regular structures prevai1. 209
ONE MORE QUESTION
208 Frances Densmore, Chippewa music, Washington, 1910, p. 59. - id., The study
of Indian music, in Smithsonian Report 1941, Washington, 1942, pp. 568 ff.
209 Cf. Erwin FeIber und Bernhard Geiger, Die indische Musik der vedischen und
der klassischen Zeit, in Sitzungsberichte der Kais. Akademie der Wissenschaften in
Wien, Phil.-Hist. Klasse, vol. 170 (1912) no. 7.
120
EARLY MUSIC
Maybe we should not take our examples fromPrimitives or

Orientals, but rather from the West, where the historical development is to the last detail familiar in the sequence of its cyc1es.
In Graeco-Roman antiquity, Plutarch speaks of the modey
patchwork in the rhythms of older Hellenie music, and especially
in its instrumental parts. Later masters, he says, gave up such
rhythmic complication and turned once more to melodie emphasis.
Indeed, he contrasts the two groups of composers as the older
'rhythmophiles' and the subsequent 'melophiles.' The fact is
dearly stated; but an interpretation can hardly be given, since
we neither know the time that Plutarch had in mind nor the
musical circumstances.
In the Middle Ages, the best visible rhythmic event was the
change from the rational, meager modi of the thirteenth century
to the most incredible complication with its climax around 1400.
This luxuriant sunset of the Middle Ages was a typical end-ofa-cyde trait. That the nascent Renaissance discarded these
excesses in the interest of austere simplicity was a just as typical
start-of-a-cyde trait ...
In our own century, we behold the crisis of 1910, when the
uniform 4/4 and 3/4 of the eighteenth and nineteenth yielded
almost suddenly to the rhythmical wealth of jazz and of Stravinsky's and Bart6k's scores. This impressive revolution contradicted the plausible events of 1400, giving rhythmic simplicity
to the end of a cyde, and complication to the beginning of a
newone.
The way out of this confusion is indicated by the changing
role and character of rhythm. Around 1400, rhythm was the
prineipal carrier of an end-of-a-cyde sophistication. Around
19H>, this inevitable sophistication devolved upon the colors
of harmony and orchestration. Rhythm, starving under their
impact, had to begin the coming cyde in a primitive freedom,
which was not sophistication, but the elemental force of a startof-a-cycle. Hence, a general rule as to the direction in which the
form of rhythm moves does not, and cannot, e~st.210
STRlCTNESS AND FREEDOM of rhythm are nevertheless by no
means developmental stages. Often the same civilization uses
210
For all details: Sachs, l.c., passim.
EARLY MUSIC
121
the two types alongside, even in the same piece, with freedom
in the song and strictness in the instrumental accompaniment.
Sometimes, the freedom of rhythm depends on the singer's
sex. On Pur in the West Carolines, women's songs are rhapsodie,
but men's, strictlyrhythmical. 211 This is no general rule. On the
contrary, where such a difference in sex behavior has been
observed, men are usually freer, while neatness and strictness
belong to the women, who, within or near the house, do their
regular, rhythmical chores, as cradle rocking, kneading, grinding,
pestling, seeding, and weaving. 212 In the villages of the Northwest Siberian Vogul - to mention one example - the men, who
do almost all the singing, keep melodies free in rhythm and
structure, while the women arrange their tunes in simple and
regular verses of the kind already mentioned. 213
RHYTHMIC OVERALL PATTERNS, well known and highly thought
of in medieval Europe, are by no means absent from primitive
music. Such advanced form of rhythmic organization shows even
in 'isorhythm.' This we find in the repertory of the Menomini
and otherNorth American Indians (and elsewhere) as well as
in the Gothic age of France among Machaut's and Dufay's works.
There, in compliance with arhythmie drive more dominant than
the melodie urge, a characteristic metrical pattern, maybe four
measures long, is forced, not upon a percussive accompaniment,
but upon the melody itself. If this melody has a span of, sa,y,
twelve measures, the "patternization' divides it into three
groups with the same arrangement of halfnotes, quarters, and
eighths to whatever ups and downs the melodie line might turn.
"Measures 5 and 9, though melodically different, have the exact
metrical organization of measure I; and measures 6 and IO
follow just as strictly the metrical pattern of measure 2; and so
forth, with measures 7 and II after 3, and 8 and I2 after 4." 214
Along with this description an example from the Menominee
west of Lake Michigan is given in my bpok. Other examples are
provided by the Arapaho in the Great Plains. 215
George Herzog, l.c., p. 277.
Sachs, l.c., p. 50 ff.
218 A. O. Visnen, Wogulische und ostjakische Melodien, Helsinki, 1937, p. 3.
914 Sachs, l.c., pp. 47 ff.
215 Bruno Nettl, Musical culture 01 theA,.apaho,master's thesis, Indiana University,
Bloomington, 1951. Also in Tke Musical Quarlerly, vol. 41 (1955).
211
212
122
EARLY MUSIC
Isorhythm, though, is a dangerous term. If the example from

the Comanche Indians that Bruno N ettl caUs isorhythmic because every measure is built on the same short pattern, then
we must of necessity caU isorhythmic everyone of the numberless
recurrent patterns, such as the meters of Greece and the modi
of the Gothic age, the rgas of India and the Radetzki March.
Nobody would go that far. All the metrical patterns just
mentioned - the Greek term polypodies might be a good common
label- have a strong and immediate motor appeal. This is hardly
ever the case in medieval isorhythmy. Most of its patterns do
not strike our sense of rhythm; and in many cases, we must
actually parse the score to come to an isorhythmic diagnosis.
But without motor appeal and immediate perceptibility, such
a repeated pattern moves from rhythm in a narrower sense to a
unifying articulation, that is, to a structural device. I think it
would be preferable not to confuse the two types by a distorting
oversimplification.
BEFORE LEAVING the manifold questions of rhythm, we pause
for a last panoramic view.
The styles of rhythm, as do the families of instruments, have
settled in a few distinct areas. The first one, almost without
exception, is the whole stretch of East and Southeast Asia
with a regular, monotonous four-four time. The second, with
rich and complicated additive meters, is the enOrmOUS region
from India through the Middle and N ear East to Negro Africa.
Ancient Greece presents an attractive blend of these contrasting
types of rhythms: it has additive rhythms up to the epitritic
meters of 3 + 4 units as a gift from the Orient, whose music
determined to a great extent the music of Hellas; but it also
has a distinctly binary attitude in many meters and their combinations - probably due to the concept of Greek orchestics as the
common ground of music, poetry, and the dance.
Ancient Europe is strangely problematic. There is no trace
of percussion and hence a strong suspicion that rhythm played
a minor role. A confirmation of this belief can be seen in the
bewildering fact that the central field of medieval music, the
Gregorian chant of the Catholic Church, despite its rather stdct
supervision by ecclesiastic authorities, lost its rhythmic tradition
EARLY MUSIC
123
within a couple of centuries and became arhythmie already in

the eleventh century. Indeed, around 1300 johannes de Grocheo
stated explicitly that no monophonie melody - including the
Gregorian chorale as well as the songs of the courtly troubadours,
trouveres, and minnesinger - was precisely measured. 216
This intentional lack of precision must have kept the earlier
notations from indicating metrical values; both the neumes
and their offshoot, the plainsong script, bypassed them: and
even the 'mensural notation,' developed in the thirteenth
century for giving such values to the plainsong symbols, struggled
almost three hundred years before being able to express them
in a clear and unambiguous form. The fight of striet against free
rhythmization extended at least into the Renaissance: the ItaloFlemish polyphony, though strictly formalized, still shows two
characteristically conflicting trends: on the one hand, the
inexorable binding time unit in the form of the handbeaten
tactus and in its augmentations and diminutions under the telling
title proportiones (at least in principle) and, on the other hand,
a free-flowing melody in all the voice parts without evenly
spaced stresses or regular additive patterns. This confliet leaves
the modern editor all the time in doubt as to where to put the
barlines in. Only with the decline of polyphony after 1600 and
probably in the retinue of the ever more influential social
dancing in the seventeenth century did simple four-four and
three-four win out.
FORM, as the beginnings of this seetion define it, is the organized
flow of a musical piece as a whole. 217
A very general statement must head this examination:
practically all the forms that we meet in primitive, oriental,
and western folk music are surprisingly short-winded. Even an
African N egro, otherwise so outstanding in musical achievements
both vocal and instrumental, can play the same phrase for a
whole day or a week if he is fond of it, without any signs of
fatigue.
216
The Latin text and a German translation: ]ohannes Wolf, Die Musiklehre des
J ohannes de Grocheo, in Sammelbnde der Internationalen M usikgesellschatt, vol.
(1899), p-. 84.

217 Wilhelm Heinitz, Strukturprobleme in primitiver Musik, Hamburg, 1931,
covers much more ground than mere form.
124
EARLY MUSIC
Scholars have blamed this deficiency on a still undeveloped

mentality, and this reasoning has doubtless its merits; birds
go hardly beyond the one call of their species, and children
display an irritating endurance in repeating a tiny scrap of
melody again and again. But the argument can certainly not
apply to the sophisticated music of high oriental civilizations,
which all the eloquence of Hegelian evolutionism is unable to
relegate to mankind's childhood. The growth of musical forms
that we observe in Europe from the seventeenth century on
seems to be connected with the growing separation of music
from social life and extra-musical claims. Contents, sizes, and
forms underwent an obvious change when music became selfcontained, to be listened to in public concert halls and opera
houses by audiences wholly devoted to its artful elaboration.
It might be helpful to take into consideration Radin's ideas
on the contrast of the man of action and the thinker, of whom
the former category, homo laber, is among the primitives much
more frequent. "His mental rhythm is characterized by a
demand for endless repetition of the same event or, at best, of
events all of which are on the same general level. Change for
him means essentially some abrupt transformation. Monotony
holds no terrors for him." 218
Just as tools and implements of pristine manufacture are
often decorated with a simple, unassuming row of ever recurring
uniform dashes without the aesthetic need of a change, much
primitive music is but a mere seriation of tiny groups of notes. 219
It must be conceded, however, that there is a flaw in this comparison, inasmuch as visual ornaments are easily taken in as a
whole, while musical ornaments are singly heard after each other.
Examples abound. Pygmoids in Central New Guinea repeat
persistently two notes a fourth apart: 220
Ex. 29
&1 ~ )1;):11
~
8
(After Kunst)
218 Paul Radin, The world 01 primitive man, New York, 1953, p. 39.
819 Cf. also: Robert Lach, Das Konstruktionsprinzip der Wiederholung in Musik,
Sprache und Literatur, in Sitzungsber. Akademie der Wissenschaften in Wien, Phil.Hist. Kl., vol. 201, Wien, 1925.
sso ]aap Kunst, A study on Papuan music, Weltevreden, 1931, pI. 11.
EARLY MUSle
125
andsodolittIe Eskimo girls 221 (our example I5). The Botokudos,

a truly palaeolithic people in eastem Brazil, would sing seven
times, without variation, one group of four notes: g f g f, so
that actually the secondal step g-f repeats no less than fourteen
times;222 on a recording at the Musee de l'Homme in Paris, a
women's chorus on Lifou, one of the Loyalty Islands in southwest Polynesia, iterate twenty times a very small motif of four
notes; 223 H. A. Junod heard Baronga lilts of two or three
measures ten, twenty, fifty times in a stretch; 224 Lagu Muna
women in North New Guinea endlessly repeat a scanty five-beat
melody; and a medicine man of the Taulipang Indians.in northem
Brazil dweIls mercilessly on a tiny mot of a rapid triplet with
a sustained note a semitone higher, the whole in a frantic
accelerando. 225
This last example draws our attention to another probable
cause for at least some cases of primitive shortness and repetition. The enervating, ceaseless recurrence of an identical sense
perception no longer interesting in itself, affects, indeed intoxicates us in either one of two opposite ways, as a potent stimulus
or, inversely, as a narcotic. Such impairing or even deadening
removal of man's unfettered volition is appropriate in cultures
whose thoroughly magic orientation assigns to much of their
music the role of depersonalization and release from normalcy.
Widely known examples are the trance and ecstasy of Derwishes
and other Muhammedan brotherhoods under the impact of
musical monotone and Koran reading.
Unassuming repetition, whatever its cause and reason may
be, occurs in many parts of the world, inc1uding the folksong
of Europe, as the following ditty from Rumania shows:
Ex. 30*
After Riegler-Dinu.
EFw FE 4444 (P 444) II I.
B22 J. D. Strelnikov, La musica y la danza de las tribusindias Kaa-Ihwua (Guarani)
y Botocudo, in Proceedings 0/ the 23rd Congress 0/ Americanists I928, New York,
1930, p. 801.
223 Transcribed in Curt Sachs, The rise 0/ music in the ancient world, New York,
1943, p. 50.
224 Henri-A. Junod, Les chants et les contes des Ba-Ronga, Lausanne, 1897, p. 33.
225 Sachs, l.c., p. 22 in E. M. von Hornbostel's transcription.
221
126
EARLY MUSIC
Herewith, we face the remarkable paralleis between primitive

structuresand thesecular forms of the Middle Ages in Europe. 226
To begin with, the repetition form appears in medieval epics
of the so-called litany type: each line of the text was sung to the
same melody: AAA ... 227
A step further is taken in the rather recent Ghost Dances
of the Plains or Buffalo Indians like the Algonquin and the
Sioux: "every phrase is rendered twice." Such paired patterns
or paired progression, as George Herzog has called this arrangement,228 are equally weU known to thehistorianofwestern
music; they form the lai or sequence (as AA BB CC ... ) so often
used in the Gothic Middle Ages.
An advanced type of this medieval lai, the instrumental
court dance estampie, .repeated the musical phrase in a more
flexible way; instead of a strict AA BB CC, it answered the
overt, or half cadence, of the first line by a etos, or fuU cadence,
in the second line of each of its pairS. 229
This principle has been known in the very earliest layer of
music still preserved. The first line and its repetition were
twinned but provided with different endings: the first time,
the voice rested on a non-final level that kept the listener's
ear in suspense; the second time, it shifted to the final level to
give a satisfactory conc1usion. "To put it technically: the first
phrase ended on a semi-cadence, and the second, on a full
cadence ... By uniting two phrases with cadential distinction
to form what musical theory calls aperiod, primitive peoples
at a very low level of civilization had created the most fertile
of musical structure schemes, the lied form." 230
Early examples are as archaie as Vedda and Patagonian
music. 231 Later, we find them in the liturgies of Oriental lews,
for instance the Song of the Red Sea from Gen. 14 : 30 in the
Yemenite version. 232
Sa!)hs, Our musical heritage, 2nd edition, New York, 1955, p. 71.
Cf. Gustav Becking, Der musikalische Bau des montenegrinischen Volksepos,
in Archives neerlandaises de Phonetique experimentale, vol. 8/9 (1933); and Felix:
Hoerburger, Weststliche Entsprechungen im Volksepos, in Die Musiklorschung,
vol. 5 (1952), pp. 354-361.
228 George Herzog, Plains ghost dance and (deat Plains music, in American Anthropologist, vol. 37 (1935), p. 404.
229 Sachs, Our musical heritage, l.c., p. 72.
280 Sachs, The rise 01 music in the ancient W01'ld, pp. 34 ff.
281 Sachs, ibid., p. 34.
BaB Sachs, ibid., p. 83.
226
S2?
EARLY MUSle
127
Iteration can also be enriched by repeating a shorl motif

at a different pitch - a practice exemplified in the Rain Dance
of the Navaho Indians: 233
Ex. 31
Such shifted iteration, as we shaIl caIl it, would be a sequence

in western terminology. But for the sake of c1arity it seems
Wiser to avoid so ambiguous a word, as it also covers the lai
and a c10sely related form of Catholic Church music.
The section on medieval forms in Our musical H eritage discusses after the litany and the lai, as the last one, the rounddance, rondeau, or refrain type, characterized by the coperation
of asolist and a chorus: "a leader sang the stanzas, and the
chorus ofthedancers answered with the refrain." 234 In primitive
music, the regular alternation of a chorus in response to a solo
'verse' is quite general. It is natural enough to occur in aIl
continents and archipelagos but is most vital in Negro Africa.
Next comes the stanza form, in which the lines have different
melodies. On Dobu in the western Pacific, the lines follow in
one of the arrays:
ABCA
ADEFC
or
ABA
CDEA
or in some other arraJ1gement. 235 A, repeated at the beginnings

and often at the ends of half-stanzas, serves as a refrain and leads
us to the following type of structure.
The response principle can assurne a variety of forms. In the
ass George Herzog, Speech-melody and primitive music, in The Musical Quarlerly.
vol. 20 (1934), p. 458.
284 Sachs, Our musical Heritage, 2nd. edition, New York, 1955, p. 72.
285 Cf. R. F. Fortune, Sorcerers oj Dobu, London, 1932, ch. VII.
128
EARLY MUSIC
simplest, the same verse melody (v) alternates steadily with the
chorus refrain (R), like:
vRvRvR ....
Elsewhere, the verses have different melodies in alternation
with an unchanging refrain, as in the wel1-known form of the
western instrumental rondo:
RaRbRcR ....
to be repeated as often as necessary and desired. 236
The answering chorus often performs in dance steps; sometimes it sings meaningless syl1ables and occasionally in a faster
tempo. 237
H.-A. junod shows in a diagram the way of staging responses
in the DelagoaBayin eastern South Africa. 238 The whole array
is semicircular: silent dancers move in an outer semicircle; an
inner semicircle belongs to young girls, who sing the refrain and
clap the rhythm; the soloist, between the two semicircles, dances
and sings the verses; and in the common center of these three
semicircles, a drummer takes care of the leading rhythmical
patterns.
Sometimes, the responding chorus interjects two notes without affecting the line of the soloist, as it so often occurs in our
passions, cantatas, and oratorios. Indeed, in Africa the choral
responses would fol1ow so breathlessly that the soloist and the
chorus almost dash in a two-part polyphony, in which the short
quartal motifs of the chorus act as an ostinato (cf. later), just as
in certain French motets of the thirteenth century. Such a twopart polyphony is actual1y achieved in Uganda. 239
While response is the alternation of a song leader or cantor
and an answering chorus, we call antiphony the alternation
(and occasionally merger) of two half-choruses. 240 The two
principalforms of antiphony are: (1) line by line the second
chorus repeats the melody of the first one; (2) the lines are
236 Cf., e.g., Erich M. von Hornbostel, in Gnther Tessmann, Die Pangwe, Berlin,
1913/1914, vol. 2, no. 2, pp. 334 ff. Also: no. 3, pp. 336 ff.; no. 5, pp. 340 ff.
237 Cf. e.g., Cora Du Bois, The people of Alor, Minneapolis, 1944, p. 136.
238 Henri-A. ]unod, Les Ba-Ronga, Neuchtel, 1898, p. 271.
289 ]oseph Kyagambiddwa, African music from the sources of the Nile, New York,
1955, p. 48.
240 Nguyen Van Huyen, Les chants alternes des garfons et des filles en Annam.
Paris, 1934. - Sachs, The rise of Inusic in the ancient world, pp. 50, 59, 92-95, 101.
EARLY MUSIC
129
alternately sung by one chorus, the odd-numbered by the first,

and the even-numbered by the second chorus, although they
might join forces at times.
The answering forms, response and antiphony, are intimately
and often exclusively linked with group dancing. But here and
there this connection breaks. When such is the case, musicseems
to be the surviving partner, whereas the dancing crumbles away.
This happens in primitive societies 241 as well as it occurs in the
West - from the musical rondeaux of the Middle Ages to recent
times with their pavanes, sarabandes, courantes, or minuets
divorced from the ballroom.
often defies the patience of singers and
listeners even within alternating patterns. To break the wearisome sameness, the variation form comes up as the most natural
remedy: the melody or theme repeats in a modified version,
different enough to keep the interest awake and similar enough
to secure familiarity and oneness. Being at once a faithful
rendition and a free improvisation, the procedure of variation
is indeed one of the oldest and most persistent of formal principIes; it has been omnipresent without interruption from the
earliest strata up to western twentieth-century music.
Variations may follow the original in its course and size. But
they also are flexible enough to undergo expansion or contraction.
To show variation incipient and mature, it is best to juxtapose
the word-conditioned changes of the earliest strata and the free
elaborations of later instrumental music. The word-conditioned
changes oceur whenever two lines of the text have a different
number of syllabies. This primitive type of variation is hardly
eonsciously eoneeived and has aesthetical values at best as a
starting point for later, imaginative treatments of a given theme.
Variations of the very finest kind are played on an exclusively
Negro Afriean instrument, in general known as the zanza.
Pressed against the performer's ehest, is a small resonanee box,
usually of wood, to which a set of earefully tuned rattan or iron
lamellas are tightly strapped with their lower ends so that the
CONSTANT REPETITION
241 Verrier Elwin & Shamroo Hivale, Folk-songs 01 the MaikaIHills,London&c.,

1944, p. xv.
130
EARLY MUSIC
slightly deflected upper ends can freely vibrate when plucked

with the two thumbs. A. M. Jones is right when he feels "something soothing and restful in the constant reiteration of a short
piece of beautiful music, whereas any possibility of monotony
is removed by the delightfully light and airy nuances produced
by the subtle variations." 242
Incidentally: it is bewildering to find in the latest musicological literature of Germany this refined virtuoso instrument
somewhat discourteously described as a Klimper. This neologism
is completely inadequate, as it derives from the verb klimpern,
'to thrum,' denoting a very strong derogatory connotation of
poorest, incompetent bungling on the keyboard. But Mr. Jones,
who writes so enthousiastically about this little instrument,
is in his terminology no paragon either. The disdainful and often
cynical nicknames that old-time travelers or colonials bestowed
on instruments are by now discredited; the obsolete misnomer
Kafir Piano has long been the laughing stock of scholars. By no
stretch of the imagination does this delicate, pocket-size instrument resemble our pianos, either in sound or construction; and
the Kafirs are an arbitrary, tiny segment of the untold millions
of Bantu who play the zanza in the enormous span between
IO o N. to 25 S. all across Africa, which amounts to the expanse
of the N orth American continent.
BESIDES ALL serial or additive forms, although less often,
primitive music knows da capo or ABA structures. Polynesia
provides very fine examples with, in the middle of a piece, a
contrasting passage in a contrasting tempo.243
The rare occurrence of ABA is understandable. All the other
forms of primitive music are open, additive, repetitive, be they
the simple seriation of some tiny group of notes or, more elaborately, lais, estampies, lieder, rondeaux, responses, antiphonies,
variations. ABA is the only form that does not permit an endless
addition of similar groups. It is a symmetrical, closed, and selfsufficient configuration superior to repetitive forms.
242 A. M. J ones, A/rican music, Livingstone, 1949, p. 27. - To be heard on EFw I 3
(from Equatorial Africa) and on English Decca (collected by Hugh T. Tracey)
LF 1084 I I.
243 E. G. Burrows, Songs 0/ Uvea and Futuna, in Bernice P. Bishop Museum
Bulletin 183, Honolulu, 1945.
EARLY MUSIC
131
Our knowledge of primitive form is altogether more than

limited. The phonograms, our principal source, are unduly
short - be it for economic or technical reasons. We seldom know
that which precedes or what follows; and the side of arecord
or, worse, the band on a side represents at best a sample rather
than a complete structure. 244 That the almost ceaseless singing
during day- or night-long feasts takes often the shape of compound forms out of coherent, though contrasting movements
cannot be doubted. In an earlier section I mentioned a cantata describing a hunt in Uganda, which is adefinite compound; and the cycle of the Hako ceremony, as held by the
Pawnee Indians, was published in a valuable volume more than
fifty years ago. 245
But it seems doubtful whether these aggregates are forms
in the sense of this chapter. We certainly speakof a symphonie
form, embracing, as a superstructure, all the four movements
in their changes of tonality, rhythm, and mood. But this overall
pattern is a pale configuration when compared with the individual
types, as the sonata form of the first movement, the variations
of the adagio, the ABA of the scherzo, and the rondo form of the
last movement. The night-long span of a primitive ceremony
does hardly alter the short-winded forms of its components.
N or are the complex liturgies of Good Friday or the Day of
Atonement musical forms.
244 Cf. Gilbert Rouget, Apropos de la forme dans les musiques de tradition orale,
in Les Colloques de Wegimont, vol. 1, p. 132 ff. (1954-1955).
245 Tracy's transcription in Alice Fleteher, The Hako ceremony, Washington, 19 0
THREE
ON THE WAY
GROWTH, PERSONALITY, ART
The oldest music described in the preceding sections has no

claim to pristine purity. We know from modern anthropology
that in itself "every culture element is essentially a range of
variation." This fact contributes "very considerably to the
flexibility of cultures and their capacity for under-going numerous
changes and stresses without actual disruption." 1
Even beyond such innate capacity for change, the very
oldest music cannot be thought of without admixtures of later
or foreign elements. But two types of melodies (treated in Seetions
Five and Six) keep rather near to the ideal of unadulterated
purity: the tumbling strains in an earlier state of organization
and the one-step melodies out of shortwinded and endlessly
repeated motifs. Examples come from every continent and from
the islands in between, and though survivals occur in many
higher civilizations, it is no undue generalization to state that
all the oldest cultures partake in these, and no other, styles.
This cannot be a consequence of diffusion or influence from group
to group. Such diffusion would suppose an intercontinental
globe-girding cosmopolitanism in palaeolithic ages and a sub sequent isolation in mesolithic or neolithic times. This does not
make sense. Our oldest music must be a general, innate, compulsory quality of palaeolithic man.
There are not enough such compulsory qualities to permit
speaking of mesolithic or neolithic music. On the contrary: after
the Palaeolithic, we are confronted with a good many different
styles, which cannot be assigned to definite ages and therefore
have been attributed to 'races,' to geographie provinces, to
culture patterns, or to Kulturkreise. Most of these ascriptions
are doubtful or incorrect, and reluctantly we have to leave an
1
RaIph Linton, The tree oj culture, New York, 1955, p. 34.
ON THE WAY
133
acceptable solution to later generations of ethnomusicologists.

Instead, we face another question: what forces form a musical
culture and lead it from palaeolithic stage to post-palaeolithic
levels?
There are two such forces: influence from outside and change
from within.
Influence from outside or 'diffusion' is natural and does not
need explanation or justification. Preceding sections of this
book referred repeatedly to the inevitable contact with neighbors
in trade and intermarriage and to c1ashes with foreigners during
the ceaseless migrations of hunters and cattle drivers. That in
spite of continual influences native styles have to an amazing
degree preserved their distinctive qualities, derives from the
fact that every group opens easily up to influences homogeneous
enough to be assimilated, but is hermetically c10sed to indigestible, 'outlandish' traits. "Borrowing is always selective." 2
Borrowing could not have occurred if mankind everywhere
had strict1y, exc1usively stuck to its palaeolithic heritage. Most
societies must have had a latent urge and the versatility to
change traditional patterns. We do not know the causes; just
as we do not know why almost of a sudden the polyphonic ideal
of the European Renaissance yielded to the stile recitativo e
rappresentativo. Very little in music can be fully traced to social
and technical changes, and any attempt to unveil the inner
process of art creation is doomed like a search for the soul with
the anatomist's scalpel.
The only experience we can rely upon with confidence is this:
every creation in the arts is a personal act but must be prompted
and supported by collective needs. Nobody else could have
accomplished in exact1y the same way what Leonardo, Raphael,
and Michelangelo did; but they themselves could not have
achieved what they <lid, and how they did it, without the
powerful upheaval that we call the Italian Renaissance. Rembrandt is unique in his unmistakably personal style; and yet
he could not be imagined outside his native Holland and outside
the early seventeenth century. And this is true of every genius.
In the primitive world, with its tightly knit society and social
responsibility, the prompting and supporting power of the
2
Clyde Kluckhohn, Mirror tor man, New York, 1949, p. 59.
134
ON THE WAY
collective is even considerably stronger than it could be in the

individualistic culture, of the Later West.
THE ANTI-INDIVIDUALISTIC COHERENCE of primitive life and
the purposeful character of music within a world of magic ideas
seems to remove all primitive music from the concept of art
that we hold in the modem world. The question of sufficient
and insufficient skill and our complicated gamut of aesthetical
values can not apply where apredominant part of music is meant
to be powerful in a social, magic, or religious sense and to impress
the good and the evil forces in nature much more than the
audience on the village square. In a like manner, nobody would
expect or wish the intoning priest in a Catholic church to strive
for the laurels of opera tenors; nor is a congregation anywhere
supposed to sing with the discipline and tonal beauty of a welltrained concert chorus.
"The first empirical and detailed exploration of the interconnections between esthetic values and the more pervasive
standards and value orientations of a particular culture is,"
in Clyde Kluckhohn's words, "due to David P. McAllester's
study of social and esthetic values as seen in Navaho music." 3
The primitives themselves, at least on a post-neolithic level,
distinguish very definitely between a good and a poor performer
and good and poor compositions. They often admire a singer's
expanded range 4 or, as in Hawaii, his deep and powerful chest
resonance; 5 and on the N orth Pacific Puget Sound, certain
singers are said to have too rough a voice for love songs. 6 In
Kenya, the Gikuyu wornen - who themselves do not play the
flutes - listen with keenest interest and present a successful
flutist with food and drink as a sign of appreciation. 7 "In the
Trobiands (due SE of New Guinea), as with us, a tenor or barytone is sure of success with women. As the native put it: 'The
throat is a long passage like the wila (cunnus) and the two
8 Clyde Kluckhohn's foreword to David P. McAllester, Enemy Way music, Cambridge (Mass.), 1954.
4 Frances Densmore, Chippewa music, Washington, 1910, pp. 59, 6I.
Helen H. Roberts, Ancient Hawaiian music, Honolulu, 1926, p. 71.
6 Erna Gunther, in Willard Rhodes, Music 01 the American Indian Northwest
(Puget Sound), Washington, 1955, p. 10.
7 lomo Kenyatta, Facing Mount Kenya, London, 1953, pp. 94 ff.
ON THE WAY
135
attract each other. A man who has a beautiful voice willlike

women very much and they will like him.' " 8
An interesting statement on the people of Alor (East 1ndonesia)
tells us that "an inadequate performance is not publicly ridiculed,
but there is a certain amount of snickering behind the performer's back. Those with real ability persist in their efforts
and may be admired performers by the time they are in their
late thirties or early forties. Abilities naturally vary. By rough
estimate perhaps IO per cent of the mature men are recognized
as good singers." 9
Such distinction gives primitive music qualities of an aesthetic
order, indeed of art; it makes it independent from personal,
sexual, social, and religious functions: "the search for beauty
is universal in human experience." 10
The author found a similar deve10pment in the realm of the
dance, where an involuntary motor discharge, astate of frenzied
movement, and a ceremonial rite, were gradually transformed
"into a work of art conscious of and intended for observation. . ..
.The individual participant in a ceremony of this kind which
becomes a work of art cannot repudiate for himself the corresponding evolution into an artist." 11
1t also constitutes a step towards art when groups within a
village or from neighboring communities celebrate competitive
singing or playing feasts and the winners are awarded enthusiastic
applause or even material gifts (although in these games the text
is often more important than the melody). We know of many
such contests. They reach from Central Asia to 1celand,12 from
the Eskimo to South Africa (here both vocal 13 and on antelope
horns).14 The Alps, Corsica, Brittanny play an outstanding role;
and in ancient Greece the agones or eompetitive festivals gave a
prominent plaee to musical contests; indeed, Hellenic mythology
relates the eme1 penalties that Thamyris, Marsyas and Midas
had to pay for ehallenging the Muses and even Apollo.15
Bronislaw Malinowski, The sexuallile 01 savages, London, 1929, p. 478.
Cora du Bois, The people 01 Alor, Minneapolis, 1944, p. 136.
10 Herskovits, Gultural Anthropology, New York, I955, p. 234.
11 Curt Sachs, World history 01 the dance, New York, I937, pp. 2I8 ff.,
12 E. Emsheimer, Singing contests in Gentral Asia, in Journal 01 the International
Folk Music Gouncil, vol. 8 (I956), pp. 28-29.
18 Cf., e.g., Hugh Ashton, The Basuto, London, I952, p. 97.
14 Henri-A. Junod, Les Ba-Ronga, Neuchatel, I898, p. I47.
15 Curt Sachs, The rise 01 music in the ancient world, New York, I943, pp. 271 ff.
8
136
ON THE WAY
The use of the word art , in its turn, raises an inescapable,

embarassing question : can primitive music, so manifestly based
on inexorable tradition and collective mentality, ever have the
personal quality that we connect with the concept of art?
Yet this personal quality in music is expressly attested on
various levels of culture. Many melodies already in common use
show individual differences when performed by different singers
(who in scriptless civilizations are often the composers, too); and
in some of the higher cultures, a personal variant can be so
distinctive that fellow countrymen recognize the author. This
often goes beyond the life span of the inventor; in Central Asia
the melodies of great singers of the past are sung to this day,
"and they are given a name that unequivocally indicates their
authorship." 16 The same applies to Arnhem Land in North
Australia. 17
Anthropology teaches in fact that every single culture "includes individual and social poles which are distinguishable,
but not actually separable." The common social heritage of a
tribe must not "mislead us into overlooking the role of the
individual in modifying or adding to his cultural heritage.
Individuals not only fall to conform to ideal social patterns but
also initiate many changes which later become universally
accepted." 18
'Practical anthropology' is no less outspoken on this point
than 'theoretical anthropology.' "Every man," says Herskovits,
"lives as a member of a society, ordering his behavior and
shaping his thought in accordance with its pattern; yet except
in the rarest instances, this is not the wh oie story. At some time
in his life, however brief the moment, in some mode of conduct,
however slight its import, he asserts his individuality." 19
In art , "the directives laid down by any tradition al style
govern the artist even as he introduces change into its art forms.
In every society the artist is the experiment er, the innovator,
the rebe!. But he is an innovator only within bounds, for he is all
unwittingly influenced by factors that guide hirn in his creative
E. Emsheimer, l.c., p. 28.
A. P. Elkin, Arnhem Land music, in Oceania, vol. 24 (1953), p. 97, reprinted
as Oceania Monograph no. 9, Sydney, 1957.
18 David Bidney, Theoretical anthropology, New York, 1953, p. 28.
19 Melville]. Herskovits, Franz Boas, New York, 1953, p. 73.
16
17
137
ON THE WAY
experience, as they guide the behavior of all human beings in

every aspect of their lives ... "
"It is important to stress this function of the artist in inducing
change, where the art of nonliterate peoples is under discussion,
because stability, rather than change, is generally emphasized
when these art forms are being considered." 20
In music - and this fact, too, stands on the side of individuaIism - we are often told that the oldest songs 21 and dances 22
of a community were inspired in dreams. Here, we should remember that in primitive societies dreams are more intimately
interwoven with the reaIities of life and are more strongly under
their influence. 23 Whatever the exact nature of this gift of the
subconscious, it seems to be closely akin to the 'inspiration' of a
creative artist in the West. It makes up the unique and personal
character of a contribution, different from mere tradition and
also different from a mechanical, uninspired com-position of
conventional elements.
In a frlendly chat with Erle Wemer, we remembered that
Richard Wagner had expressed exactly the same idea when, in
the third act of the Meistersinger, Hans Sachs encourages young
Walter:
Des Menschen wahrster Wahn
Wird ihm im Traume aufgetan,
Und alle Poeterei
Ist nichts wie Wahrtraumdeuterei.
(Man's vision in his dream unfolds
That what his inner eye beholds,
And all poetization
Is merely dream interpretation).
(transl. C. S.)
MUSICAL INVENTION is in many archaic civilizations a matter
of course. It was hardly an exaggeration to call the Eskimo a
20 id., CuUural antkropology, p. 255. - This complex question is brilliantly treated
in A. L. Kroeber, Conligurations 01 cuUure growth, Berkeley, 1944, particularly in
the section on Genius, pp. 7-16.
21 Frances Densmore, l.c., p. 59. - id., The study 01 Indian music, in Smithsonian
Reporl 101' 1941, Washington, 1942, p. 538.
22 Curt Sachs, W O1'ld history 01 the dance, p. 53.
23 Richard Thurnwald, Des Menschengeistes Erwachen, Wachsen und Irren, Berlin,
1951, p. 121.
138
ON THE WAY
genuine people of musicians; every Eskimo must know the art

of composition; 24 and, in supreme contempt, a jealous Eskimo
woman would sneer at her rival: "She can't dance, she can't
even sing ... " 25 In theSouth African Ila and Tonga regions it is
incumbent on every young woman or man to compose personal
songs: a groom is asked to sing a melody of his own on the day
of his wedding,26 and Tonga men, when possessed by an evil
spirit, must during exorcising rites intone 'their' songs. 27
But it should once more be said: with a1l his creative abilities,
the individual is "born and reared in a certain cultural environment, which impinges upon him at every moment in his life.
From earliest childhood his behavior is conditioned by the
habits of those about him. He has no choice but to conform to
the folkways current in his group." 28
The individual and the social pole are counterbalanced against
one another when on the Andaman Islands in the Gulf of Bengal
"every man composes songs, and the boys begin to practise
themselves in the art of composition when they are still young.
A man composes his song as he cuts a canoe or a bow or as he
paddles a canoe, singing it over softly to himself, until he is
satisfied with it. He then awaits an opportunity to sing it in
public, and for this he has to wait for a dance. Before the dance
he takes care to teach the chorus to one or two of his female
relatives so that they can lead the chorus of women. He sings his
song, and if it is successful he repeats it several times, and thereafter it becomes part of his repertory of songs that he is prepared
to repeat at any time. If the song is not successful, if the chorus
and dancers do not like it, the composer abandons it and does
not repeat it. Some men are recognized as being more skilful
songmakers than others." 29 The individual, then, creates;
society accepts or refuses.
How personal, how anti-col1ective musical creation can be in
SocieU Neuchateloise de GeograPhie, vol. 54 (1948), p. 3.
25 Diamond Jennes, The people 01 the twilight, New York, 1928, p. 124.
28 A. M. Jones, Alrican music, Livingstone, 1949, p. 14.
27 Thurnwald, I.c., p. 132.
28 George P. Murdock, The science 01 cuUure, in American Anthropologist, vol. 34
(1932), reprinted in Hoebel, Jennings, Smith, Readings in anthropology, New York,
1956, pp. 314 ff.
29 A. R. Radcliffe-Brown, The Andaman Islanders, Cambridge (Mass.), 1933, p. I32.
ON THE WAY
139
primitive civilizations, shows in another habit of the Andaman

Islanders: "Every man composes his own songs. No one would
ever sing a song composed by any other person." 30
WE ARE SELDOM in a position to sense the personal contribution
in a music so remote from ours - just as a range of mountains,
seen from afar, appears to our eyes as a single compact and
undivided massif. In the chants of Hawaii, native singers often
point out distinct and personal traits where we, the foreigners,
only hear the same eternal cantillation on one pitch with merely
a cadential drop at the end of a poetical section. Tbis lack of
distinction, by the way, can be reversed: oriental musicians are
generally at a loss when asked to distinguish between the
personal styles, even a hundred years or more apart, of western
composers. I remember a musician from the Middle East who
after eleven years of musical studies in Germany confessed to me
that he was unable to notice any difference between a Haydn
and a Mahler symphony.
Only rarely are we allowed to witness creation at work.
It happens when the original pattern of one single step grows
to a two or three-step pattern. 'Growing' is understood as a
timid, occasional trait before the novel, larger pattern has taken
roots to stay. The new step, as a result of some personal urge,
may appear only once at the end of a melody after the traditional
one-step pattern has been safely established. Less often does the
singer yield to his urge right at the beginning, only to shyly
relapse into the conventional model.
The following section will discuss these developments in
greater detail.
A COMPOSER, the Eskimo say, is, and must be, the exclusive
possessor of bis song, since it carries bis inalienable SOul. 31
Such rights and restrictions are by no means rare or sporadic.
On Dobu in the western Pacific every man composes; "he is
proud of his creation, proud of its originality, and he has rights
to prevent others from using bis song, at least for a while. The
song-maker must give bis permission before his song is used
80
81
ibid., pp.
131
ff.
Zygmunt EstIeicher, l.c., p. 3.
140
ON THE WAY
for the dance." 32 In the primitive Northern Territory of Australia, the 'song-man' of each group owns all the melodies and
dances that he performs, and "no one can sing his songs without
his permission." 33 By the same token, a funeral song can be
the exc1usive property of an Indian chief or group,34 just as in
the Scottish Highlands a particular piper' s 'lament' is the
property of a particular clan. One of the Pawnee songs in the
Indian's Book had been the property of a man; but one day he
gave it to his brother by mutual adoption (the dosest human
relation), "bidding him to sing it when he needed help or proteetion." 35 The sollicitation of assistance through sounds
miraculously carried across unlimited space - as by playing the
'Magie Flute' and the horns of Oberon, Roland, and the departing Lohengrin - this ancient motif of saga finds a new
evidence on Amerindian soll.
The idea, intimately connected with the motif of a quid pro
quo relation between man and melody - anticipating the personal
Leitmotive in Wagner's Ring - finds a diminutive variation in
Uganda: every man of the Lango people has his own whistle
motif, not to be misused by anyone else. "An infringement of
this rule will certainly cause a violent quarrel, and may even
lead to bloodshed." 36 Evidently, such motifs are the precursors
of our somewhat degenerated family whistle-calls.
On the other hand, all melodies, though strictly taboo to unauthorized performers, may be legally 'copyrighted' for natural
heirs and for buyers.3 7 Actually, persons may for money or any
equivalent acquire the right to sing the melodies or dance the
choreographie inventions of other men. 38
In Australia, songmen often trade their songs because of some
elose kinship tie. But whether songs and dances are passed on
to other tribes or not, "they do serve to link tribes in friendship.
R. F. Fortune, Sorcerers 01 Dobu, London, 1932, p. 251.
A. P. Elkin introducing EFw FE 4439 (P 439).
34 Marius Barbeau, Asiatic survivals in Indian songs, in The Musical Quarlerly,
vol. 20 (1934) passim.
85 Natalie Curtis, The Indians' Book, New York, 1907, p. III.
86 Driberg, The Lango, London, pp. 124 ff.
87 Frances Densmore, Chippewa music, Washington, 1910, p. 60. - Cf. also: Erna
Gunther, in Willard Rhodes, Music 01 the American Indian Norlhwest (Puget Sound),
88 Richard Thurnwald, Die menschliche Gesellschaft, vol. I of Reprsentative Lebensbilder von Naturvlkern, Berlin, 1931, p. 38. - Curt Sachs, l.c., p. 219.
32
33
ON THE WAY
141
They are performed as mutual entertainment and as a way of

maintaining, strengthening and extending social relationships.
Thus music, while being aesthetic in content, is a social institution
of positive value." 39
.
One evening in 1930, when the sun was setting over the yellow
sands of the desert, a dozen Nubian oarsmen who were taking
me through the rapids of the Nile around the isle of Elephantine
sang endlessly in response form to the rhythm of the rows. 40
I followed with my pencil as best I could, running a race with
the night which so quickly falls in that latitude. But when I had
disembarked in Aswan, they demanded an extra bakshish as I
had written down their ditty and thus appropriated what
- they said - was rightfully theirs.
How far the concept of personal melodies as a line of business
reaches into orient al civilizations of the highest level, is told in
the story of a once famous musician in India who, being in
serious money difficulties, succeeded in mortgaging a certain
melody or rga, which was c10sest to his heart and a favorite of
his maharaja. Soon thereafter, when ordered by the sovereign
to sing it at court, he confessed what he had done. The deal,
apparently unusual, amused the prince; generously he paid up
the loan and, besides, rewarded the moneylender for his "keen
appreciation of the value of music." 41
Modern musical copyright was first enacted in 1831 at Washington D.C.42
In this context it might be appropriate to remember Herskovits' statement that "intangibles form an important category
of property in an societies, as witness the economic value placed
on patent rights, on good will, and on copyright in our own
. culture .. . [An] example of the importance of incorporeal
property inc1udes an assortment of rights called topati by the
Nootka of British Columbia - knowledge of family legends, a
ritual for spearing fish, honorific names of many kinds, the right
to carve certain designs on totem poles and grave posts, to sing
89 A. P. Elkin, Arnhem Land music, in Oceania, vol. 24 (1953), p. 106. Reprinted
in Oceania Monograph no. 9, Sydney, 1957.
40 Music example 35 in Curt Sachs, The rise 01 music in the ancient world,
New York, 1943, p. 95.
41 Herbert A. Popley, The music 01 India, Calcutta, 1921, p. 66.
42 Hans Abraham, in Willi Apel, Harvard Dictionary 01 Music, Cambridge, Mass,
1944, pp. 184 ff. (with bibliography).
142
ON THE WAY
certain songs, to dance certain dances, to perform certain

specific parts of certain rituals ... In the South Sea, personal
names, incantations, songs, charms, and family traditions all
figure as family wealth.'" 43 Therefore, "for the most part, those
who received training in composition, narration, and chanting
were usually of noble birth." 44
Melville J. Herskovits, l.c., p. I60.

id., l.c., p. I89. - Cf. for social restrietions as discussed in this seetion also
jaap Kunst, Sociologische bindingen in de muziek (inaugural oration), The Hague,
I953
48
44
FOUR
THE FATE OF
SECONDAL
AND
TERTIAL
PATTERNS
While tumbling strains developed by inner consolidation,

horizontal melodies chiefly evolved by enlargement. One-step
patterns reached out, upwards and downwards, usually by giving
citizenship to former occasional affixes. From there the process
went on to a conquest of more and more of the musical space. 1
A classical evidence of transition from a secondal one-step
to a secondal two-step pattern are the two oldest of the successive
styles in which the priests of India used to chant the sacred
texts from the Vedas. Early in times B.C. the chant, almost
certainly, knew only two notes or pitches; a higher one, udtta,
a lower one, anudiitta, and their.legato succession or ligature,
svarita. Pnini, patriarch of grammarians in the fourth century
B.C., has stated this expressly. Later, the svarita changed into a
third note which, appoggiatura-like, dropped to udiitta from a
tone or a semitone above but never fell directly to anudiitta, 2
possibly to avoid an indesirable leap:
Ex. 32*
J.. Il!O
, ) I J. i J I J J J I ,L I p. J J J IJ J J I ,J. I
Similar arrangements occur in other oriental cantillations of
high antiquity. A very fine example is the central creed of the
Jews in its Babylonian version - Harken Israel, the Lord our
God, the Lord is One - in which we readily distinguish the ap-
* Transer. C. S.
John Frederic Rowbotham, History of music, London, 1885, seems to have been
the first to see this phenomenon.
a Erwin Felber, Die indische Musik der vedischen und der klassischen Zeit, in
Abhandlungen der K. K. Akademie der Wissenschaften, Wien, 1912, no. 7. - EFw
FE 4431 (P 431) I 3.
1
144
SECONDAL AND TERTIAL PATTERNS
poggiatura-like BI7, being a quasi-svarita, the A as an udiitta, and

the anudtta on G (Example 24).
The process of widening the original one-step patterns is
rather common in the history of oldest music. Even typically
primitive tribes such as the Vedda in inner Ceylon often add a
third note to the two already used. And in this case, as well,
it will not do to simply count the notes. Such naive procedure
will yie1d no chronology, no grade of perfection, no rate of
civilization. A Vedda melody with three notes or two steps came
no later than a Rumanian folksong with only one step; nor is it
beUer; nor are the Vedda more advanced in culture than the
Rumanians. The tune is simply more complex.
Often is the original one-step quite c1early audible, while the
third note is just an addition without actual naturalization.
A characteristic example from the Vogul folIows:
Ex. 33*
The famous epic songs of the Yugoslav guslari often lasting

for an hour, also belong to this group, although the seconds
are rather short. 3
Sometimes we find group alternation, not just the alternation
of two notes, but of two one-step patterns, like the following
example from Bukaua in the Solomon Archipelago:
Ex. 34**
When in Hawaii either note sinks a seInitone in the second

half of the piece,4 transposing might find an explanation - a
possibly too rationalistic explanation - in the exhausting stretch
*
After A. O. Visnen, Wogulische und Ostiakische Melodien, Helsinki, 1937.

After Erich M. von Hornbostel, Die Musik auf den nordwestlichen SalomoInseln, in R. Thurnwald, Salomo-Inseln, vol. i (1912).
3 EFw FE 4506 (P 506) III 46.
4 Helen H. Roberts, Ancient Hawaiian music, in Bernice P. BiskOP Museum
Bulletin 29 (I926), pp. 76, BI.
**
145
of Hawaiian cantillation. The eyes of old singers are reported

to bulge with the effort.5
The tenacity of the nucleus is amazing. Affixes appear oftenest
at the beginning of a song or at its' end: the singer is bold and
pioneering either at the start when he is quite himself or else
in the coda when an audacious infringement does no longer
endanger the identity of the pattern. But the main stretch of
the melody follows tradition and is in many cases completely
free of affixes.
On the other hand, affixes at the right place cannot fail to be
eventually acG~pted with equal rights and to accrue the original
nucleus. In the following example from Estonia the kernel is
doubtless bp-a; but the affix c (much more than thelessimportant
infrafix g) has become strong enough to form the start and to
usurp a few of the melodie stresses:
Ex. 35*
A TERTIAL NUCLEUS can be widened by appending a second,

a third, or a fourth. But the most frequent, and a fascinating,
form of tertial two-step me10dy is the one usually found under
the doubtful nicknames 'fanfare' and 'triadic' melody.
In Europe, empty 'triads' are confined to the East, from the
Lapps down the map to Greece; filled, they expand all over the
continent. Both forms prevail in the two Americas and in Melanesia, while in Africa the filled type is more general. The Asiatic
distribution includes, in either form, tribes as primitive as the
Semang in Malaya, but it excludes the high civilizations of the
Far East and of India.
The two current names 'fanfare melody' and 'triadic melody'
are objectionable. The first is instrumental; it is too much
reminiscent of the trumpet and the bugle to suit a vocal pattern.
N either is the second name to the point. Accepted musical
terminology reserves the word triad for a chord of three simultaneous notes: a root or fundamental and, slightly less important,
*
6
After Walter Wiora, Europischer Volksgesang, Heemstede, 1952.

ibid., p. 72.
146
its third and its fifth. This wou1d by all means be a pardonable
loan - I myself do not hesitate to borrow from harmonic terminology when it facilitates quick understanding. But 'triad'
misleads; the name refers at best to tonal anatomy, with no
attention paid to the more important physiology.
Tonal physiology, as I see it, is less concemed with the static
structure of a scale than with the distribution of moving forces
and weight within a melody. It does not rest satisfied with
finding out what the distances between the three notes ofa
triad are or in what order they appear, but also wants to know
which note is the center of gravity and which of the two steps
is allotted the greater shaping power. While anatomy speaks
only of the major and the minor third that constitute a triad
C E G, physiology demands an answer to the two specific
questions : which of the three notes is the actual chief, and which
of the two steps C-E and E-G has the greater importance?
As astart, we look for the first and the final note and, if
possible, for cadential pauses inside. But we also must r-ecur to
mere statistics by counting how many time-units each of the
three skeletal notes and their infixes (if any) is given. This sounds
unpleasant to those who believe that music is a matter of the
spirit and not of deadly numerical facts. But in test and countertest the statistical method is with a few exceptions reliable
enough as long as the analyst knows its limits. 6 A comparison
of the starter-final and the statistical method will seldom lead
astray.
Let us take three examples of empty triads without infixes
or affixes. The first is a song of the Lapps:
Ex. 36*
3
~ (3 J JJ I J J J I" ~ d J JJ 1J J:11
The lowest note or root appropriates fourteen units or, in our
transcription, eightnotes; the third, six of them; and the fifth,
four or, including four rapid appoggiaturas, some more, say,
* After K..Tiren, :pie Lappische Volksmusik, Stockholm 1942, p. 103 nr. 60.
Cf. also: Linton C. Freeman and Alan P. Merriam, Statistical classi/ication in

anthropology: an application to ethnomusicology, in American Anthropologist, vol. 58
6
(1956), pp. 464-472.
147
six, units. No doubt the root prevails, and the two other notes
are about equal in weight between themselves. As a confirmation,
we find that the root acts as both the starter and the final. Thus,
the three notes form indeed a genuine triad in root position.
A lullaby of the Slavonic Kashub:
Ex. 37*
consists of only eight notes per verse, to be repeated indefinitely.

It gives five of them to the third, two to the fifth, and only a
single one to the 'root'; the note g aets also as a starter; and
from this principal note the melody reaehes twiee up to b and
only onee downward to e, whieh proves too little important
to make it a root. This so-ealled triad is functionally just a
eentrie melody of the kind to be examined in a following seetion.
Our third example, a song of the Vogul in northwestern
Siberia:
Ex. 38**
,) nn IJJJIJ J5JJIJJJIDJ5JJlffllhjjl
)=80
starts on the fifth and repeats this note to the amount of 24

units, while only I4 go to the third, and 4 to the presumptive
root. The fifth dominates; it forms a third with the middle note;
and the lowest note is again an infrafix at the distanee of a third.
Thus the same anatomieal strueture has three different
physiologieal aspects, .and the old suggestions that tribes who
sing in so-ealled triads might have a "latent feeling of harmony"
is Euroeentrie and untenable. The 'triads' are simply double
thirds.
Particularly eonvincing in this respeet are melodies in whieh
the so-called triad appears as two separated thirds: the first
seetion alternates between the 'third' and the 'fifth,' and the
* After Tiren, l.c.

** After Visnen, l.c.
148
second section, between the 'third' and the 'root.' 7 The double
third that a mechanical addition of the two steps makes believe,
is in fact a succession of two single thirds.
The general direction'of a me10dy is an important factor, too.
Double-third melodies of American Indians descend like tumbling
strains, while similar European folksongs start as a rule on the
lowest note.
Despite the non-triadic character of double thirds, the two
adjoining thirds form often an organic whole. They have different
sizes, one being (more or less) major, and the other one (more or
less) minor; the two thirds together form an approximate1y
perfect fifth. Non-perfect fifths, augmented or diminished, do
hardly occur.
How strong the urge of the perfect fifth can be, appears
from a klephtic (underground guerrilla) song of modern Greece:
Ex. 39*
Above the double third t-a-c', still another third, e', makes it a
tripIe third. As this e' forms a so-called changing note with the
affix d', it is via naturae flatted ; and being now an eb , it entails
an ab be10w where there had been a natural a before.
This example leads from a two-step to a three-step pattern,
from a double third to a tripIe third. The urge to expand the
range and yet to cling to the same basic interval can indeed be
so strong that singers feel compelled to enlarge their double third
(or, as we shall see in the following chapter, double fourth) by
adding, above or below, another third or fourth. In doing so,
they create 'chains,' such as tripIe, quadrupIe, quintuple, and
even sextuple thirds, and tripIe or quadrupIe fourths.
I eall them ehains, as our eustomary word seale would be a
misnomer. An actual seale is an organie whole, in which every
single note has a function of its own, as the tonic, the dominant,
* After Samuel BaudBovy, La chanson clephtique, in Journal ot the International

Folk Music Coucil, vol. 1 (1949), pp. 44 ff.
7 For an example, cf. M. Kolinski, in Melville J. and Frances S. Herskovits,
Suriname folk-lore, New York, 1936, p. 531, 1.
149
the subdominant, the leading note, the mediant, and so forth.

Hardly one of such functions exists in achain: while the singer
moves from interval number one to interval two and from there
to interval three, he often forgets about the first one and finds a
new orientation. His chain is not an organic whole, but an unorganized concatenation of similar intervals. The intervals are
similar, not identical, because the thirds in achain are now
major, now minor, to form every two a perfect fifth. It would
be admissible to describe such a tertial chain as interloclring fifths.
TRIPLE THIRDS, in their two basic forms C E G B and D F A C
have given music historians a troublesome problem to solve.
Still unaware of the concept 'chain,' analysers of medieval
melodies haveoperated with their old, familiar notions and,
somewhat embarrassed, have called them mixtures of the Dorian
and the Lydian Church mode or of our minor and major scales.
In doing sO,they ignored one basic fact: both the modes of the
Church and of the world outside the Gregorian realm have
octaval structures, while a tripie chain is septimal. Actually,
the tripie thirds, universal and prevalent in Europe both in
folksongs and in art music (including the Gregorian chant) were
mostly discontinued during the sixteenth century when the major
and the minor mode had been established to stay.s
An example from the so-called Turco-Sephardic version of
the Jewish liturgy, a Kol nidrei to open the eve of the Day of
Atonement, shows particularly weIl how a tertial nucleus, G-B,
is first widened to the lower third E and eventually also to the
upper third D:
Ex. 40.
We find similar tripie thirds ever so often on archaic levels

of Asia,9 in North Africa,lO in North and South America 11 way
* EFw FE 4505 (P 505) IV 14 (Transer. C. S.).
S
Curt Sachs, Tke I'oad to majOI', in Tke Musical Quarlel'ly, vol. 29 (1943), pp. 381-
404
e.g. Flores: ]aap Kunst, Music in FIOl'cs, Leyden, 1942, p. 43.

e.g. Berber (Morocco): Von Hornboste! and Lachmann in Zeitsckl'itt
gleichende Musikwissenschaft, vol. I (1933), p. 4.
11 e.g. Cocama (Upper Amazon): EFw FE 4458 (P 458) I 1.
9
10
/11,1'
Ver-
150
up to the Eskimo 12 as weIl as in Melanesia, Polynesia, and

New Guinea: 13
One of the Asiatic examples is "the" strange melody pattern

of the Palaung in Burma. All Palaung songs, we are told, are
sung to this same melody, whether they are grave or gay;
variations that we might hear are due to "the caprice of the
singer." The tune as we read it in L. Milne's book is obviously
written down by ear; but the transcription is convincing. Within
scale e g b c' d', the note b is three times held in long fermatas
at the beginning of each phrase. The singer, she says, "first takes
as long a breath as possible, in order to prolong the sound of the
first note, which should be at least ten seconds in duration." 14
requiring the wider range of a ninth,
are conceivably scarcer. In the two basic forms, CE G B D and
D F ACE, they are conspicuously absent from the Pacific;
they do not touch Asia save in Turkey, and reach only a few
North American tribes,15 inc1uding the Copper Eskimo,16 and
also American jazz.
Inversely, they play an important role in Bantu Africa and
are sung by the Hottentots as (usually unfilled) skeletons:
QUADRUPLE THIRDS,
Ex. 42**
~5(
..(1
~ J I J?
r=3 ,11 E~ ,J :1
W J 1J ~
~ r r p.
31:\
But their mainhabitat is Europe. Here, they occur as empty

skeletons in Finnish folk music, while they are filled and diatonic
* Marind-anim, Southwest New Guinea, after J. Kunst.
** After Moodie.
12 Helen H. Roberts and D. Jenness, Songs 0/ the COPP81' Eskimo, in Repon 0/ the
Canadian Aretie Expedition I9I3-I8, vol. 14, Ottawa, 1925, passim.
13 Jaap Kunst, De inheemse muziek in westelijk Nieuw-Guinea, (Mededeling XCIII
0/ the Royal Tropical Institute, Amsterdam), 1950, pp. 44-47.
14 (Mrs.) Leslie Milne, The home 0/ an East81'n clan, Oxford, 1924, pp. 303 ff.
15 e.g., Quileute and Zuiii (Densmore).
16 Helen H. Roberts and D. Jenness, l.c.
151
in other European countries. The pattern is so truly European

that it reappears again and again in art music. One example
comes from French polyphony of the mid-sixteenth century:
in the motet version of Claude Goudimel's Psalm 10, the tenor
cascades down from g' to e'e' and a /.17 We have evidences even
in contemporary music: a passage in Stravinsky's Petrouehka
(19II) ascends and descends in empty thirds from a' to b";
and the me10dy of the ninth Ludus in Carl Orff's Catulli Carmina
(1935) scales in thirds from d' to e".
QUINTUPLE THIRDS - C E G B D F or D F ACE G - are
Imrely European. They occur in leeland 18 and also in Serbia,
although in the Serbian piece that comes to my mind 19 the
Iowermost flfth is not subdivided into thirds. Of quintuple thirds,
as well, the polyphonie sacred music of the sixteenth century
provides a startling sampie: in the French Mass 20 of John
Shepherd (d. circa 1563), the two upper voices unite to form the
empty chaih d' /' a' c" e" g":
Ex. 43
'1 r g Ir U Ir u IJ U Ir n IJ97 1 J J IJJ

Indeed, late in the nineteenth century Mussorgsky gave to
the Song of the Gnat in his Boris Godunov exactly the same
chain in descending movement.
With an due reserve, mention could be made also of the
old zither' yamato koto of Japan, whose six strings form a quintuple tertial chain - d' /' a' c' e' g' 21 except that the three last
ones are tuned down by an octave.
Sextuple thirds do not occur in folksong, perhaps on account
of a range overtaxing the possibilities of untrained voices..
I found one in a Fantasia a 5 by William Lawes (1602-1645),
although to be a perfect example the lowest B should be flatted.
Courtesy Eleanor Lawry.
B. Thorsteinsson, Folkelig sang og musik paa Island, in Nordisk Kultur (1934).
19 Communication from Danica S. ]ankovic.
ao ed. H. B. Collins, London, 1927.
21 Sir Francis Piggott, The music and musical instruments 0/ Japan, 2nd edition,
London, 1909, p. III.
17
18
152
It will be good to remember that quintuple and sextuple

thirds are exelusively European. But (notwithstanding evidences
from other parts of the world) quadrupie thirds and lesser
tertial chains are European as weIl as Bantu African. Once more:
Europe and Africa, related in their preference given to thirds,
both in melody and in chordal organization, were tom apart by
a South Asiatic and Mediterranean wedge of quartal melodies.
MOST EXAMPLES, so far, were 'empty' skeletons. Only a few
showed 'infixes' of the kind discussed in our chapter on Onestep Melodies. Here, too, infixes mostly bisect the major thirds
and leave the minor thirds intact. The result is agamut of
wholetones and minor thirds. But it should be realized that the
infixes are fillers of inferior melodie importance and must not
be joined with their two immediate neighbors above and below
in what Hugo Riemann called a pyknon or density and believed
to be the melodie nueleus. This it is not, since the central note is
unmistakably a passage note without any structural significance.
This has not been the only mistake in interpreting the infixes
in tertial chains. Minor thirds plus wholetones are on the surface
so elose to the so-called pentatonic scales that, alas, most
authors have succumbed to the temptation of dubbing such
chains pentatonic. Indeed, pentatonism is one of the ob sessions
in recent ethnomusicology. This is a dangerous misnomer.
Pentatonie must be defined as an octave divided into five steps,
be they equal or unequal, just as tetratonic, hexatonic, heptatonic scales are octaves divided into four, six, seven steps. Hence
the term pentatonic applies perfect1y well to Scottish, old-Greek,
many Indian, and most East Asian scales, which are quartal in
structure and use the octave as an organic and vital interval.
But a tertial chain like A C d, E GaB has no octave in its
skeleton or at best as a mere infix witliot a structural role.
At a pinch, it is pseudo-pentatonic. The only similarity between
the two organisms is the lack of semitones (which is not a
necessary trait of pentatonic scales) ; they are, to use a Greek
term, anhemitonic or 'without semitones.'
did they develop from one

another? Are they independent? Do they reflect any racial
PENTATONISM AND HEPTATONISM -
153
idiosyncrasies? It means stirring Up a wasps' nest to approach

the question of their relations. Our literature is full of answers.
But most of the opinions we hear are rash, contradictory, and,
worst of all, 'plausible,' that is, as the writer at his desk thinks
it could have been. The author begs to desist from summing
them up 22 and to offer his own views for what they are worth.
Before giving an answer, albeit a tentative one, let us repeat
with great insistence: the old and easy-going equation of pentatonism and anheInitonism is no longer adInissible. The lack
of seInitones may occur in all kinds of structures, and, inversely,
a wide-spread type of pentatonic melody has semitones (and
major thirds). Unless this is clearly understood, we grope in
the dark.
Any answer must have the support of our most archaic
examples of priInitive music. On thumbing through my own
notations, I find in about eighty percent of all two-step melodies
two different distances: when the nucleus is a major second,
the other distance is either a Ininor third or else a seInitone;
when the nucleus is a Ininor second, the affix is generally a
wholetone apart; when the nucleus is a minor third, the other
distance is in most cases a wholetone or a majorthird. Evidently,
there has always been an inhibition against equality, which is in
a way reIniniscent of the inhibition of artists against squares
for the benefit of rectangles.
This elementary urge is even more obvious in three-step
patterns. At random, I present two primitive examples which
could easily be centupled. One comes from the Makushi in the
Roraima Mountains, British Guyana. It has the four notes
a g I d; gl is the nucleus, and the affixes keep at the distances
of a wholetone and a Ininor third. The genus is definitely composed of Ininor thirds and wholetones, without any seInitone.
The other example was sung by South African Lala. The four
notes are e" d c b,,; the nucleus is d b,,; theinfixccleavesthemajor
thirdintotwo wholetones; and the suprafix keeps at a distance
of a seInitone.
The two examples prove not only the existence of Iningled
steps on a very early level. They also show that our two most
e"
22 Arecent summary can be found in Constantin Brailoiu, Sur une melodie russe,
in Pierre Souvtchinsky (ed.), Musique Russe, vol. 2, Paris, 1953, pp. 329 ff.
154
SHCONDAL AND TERTIAL PATTERNS
frequent arrangements - anhemitonic and diatonie - have their

origins in very archaie times, long before melodie patterns grew
to the sizes of sixths and octaves.
It might be remembered that the word pentatonie can apply
only where, by definition, an octave has five steps. This requires
at least one structural fourth in an octave, that is, for example,
ce g c' or c j g c'. Chains of thirds with the major thirds filled in,
as c d e gab .... are anhemitonic, but not pentatonie, since the
octave appears as best as an infix but not as a structural tone.
The knowledge that anhemitonic and diatonie patterns are
prototypes forbids us
(r) to establish a chronology of the two;
(2) to derive the diatonie gender from the anhemitonic gender
by infixing two piens. The word and the fact apply to
Chinaonly.
(3) to attribute the two genders to 'race.'
When we observe how again and again the Chinese attempted
in vain to replace their pentatonie scale by introducing two
piens, this is not evolution or progress, but simply an abortive
change from an eastern to a western concept. And it is a matter
of strongly traditional culture, not of race or 'blood,' as any
American-born or American-reared Far Easterner will prove.
TWO OTHER FEATURES should be mentioned before we shelve
the tertial chains. In Europe, the feeling for third catenation
has been so strong that it led to some aspects inexplicable from
our modern diatonie viewpoints.
The first one is the so-called German choral dialect of the
Gregorian chant. When in the Roman chant a melody in the
first mode (on D) reaches out beyond the fifth, only to revert
at once to the latter, the singer makes instinctively this step up
and down as short as possible; he reduces it to a semitone
although this requires a flat against the purity of the mode.
As the old cantorial rule expressed it:
una nota super la

semper est canendum ja
which meant, in medieval solmization, that the note above a

must always be taken as b", as ja is the symbol of the semitone
155
(mi-la). But the Germans and Scandinavians, unaccustomed

to semitones, replaced them by their traditional thirds. In other
words, the Northerners transformed the typical pattern of the
first Church mode, dia bj, a, into the equally typical tripie
chain of thirds, dia c a. Inversely, melodies sung in the fourteenth century, and probably many of older times, too, avoided
the subsemitonium or semitone below the tonic, as in our customary g e c B c, and preferred a minor third, as in g e c A c, which
again opposes a tertial chain to a functional, modal scale with
its seventh or 'leading note.' In this case we speak of a Landino
third; but the great Francesco Landino (1325-1397) was by no
means its inventor. 23
I suggested long ago that in a similar way the four-line staff
of the Gregorian chant reflects, not the system of the church
modes, but the tertial chains of secular music: according to the
c1ef prescribed, the lines
dl a c'
I a c' e'
a c' e' g'
while the spaces in between are literally intended for the empty
spaces or the infixes between the skeletal notes.
IT MEANS MELODIC CONSOLIDATION, as opposed to the loose
concatenation of the 'chain,' when the singer, starting from a
double third, does not proceed to building a tripie third, but
reaches for a second only, be it major or minor. This second,
in turn, leads the melody back to the original fifth - a case
mentioned previously. Using our familiar terminology, the
skeleton would be F A C D (C) in major and ACE F (E)
in minor. Let us, once more, borrow from the nomenc1ature of
harmony and call this grouping a six-five pattern.
But here, again, we must rely on musical physiology, even
at the risk of disappointment. In most cases, as mentioned
before, the sixth is simply a 'changing note' above the fifth and
must be considered a suprafix. The given example:
28 An example from English folksong in Constantin Brailoiu, Sur une melodie
russe, l.c., p. 36I, no. I07.
156
Ex. 44
is a half-shouted briolage or incantation addressed to grazing

cattle in the French province Le Berry. Many other instances
are probably hybrids or concretions of two simpler patterns.
In a song of Alu, one of the Solomon Islands, Melanesia:
Ex. 45
&J &11 j r(li J J ~ I Jr [' I FJ J r I

J~
,j
the two elements are particularly obvious: a single empty fourth

on notes 3 and 6, and a triad on I 3 5.
It must have taken a long time before the sixth was firmly
established in the skeleton as apart with equal rights and duties.
This final stage is evident in the Gregorian chant and, probably
deriving from plainsong, in Philipp Nicolai's Wachet auf, ruft
uns die Stimme and similar Protestant chorales. But it is evident
in Asia as weIl. It reappears at the other end of the world as the
first of the Chinese modes or the ]apanese ryo; and between
these west-eastern extremes, there are many examples from
Tibet and various parts of Europe, including the archaic Faroes
due north of Scotland.
The Chinese occurrence of the six-five pattern has caused a
fatal mistake in analysis. In China, so we are told even in recent
books, aIl melodie modes derive from one original scale by toptail inversions, the lowest note being in this process cut off
and re-added an octave higher as the new top:
CD E GA
D E GA C
E GA CD
This looks heat but is very questionable: scales 2 and 3 have
quartal structures:
~,,..,-'--",,,
D E
E
GA
GA
C
CD
Pathe Part. 5607; cf. J. Can telou be (ed.), Anthologie des chants populaires
tranc;ais (Paris 1951), t.IV, p.l06
.. After Erich M. von Hornbostel,Die Musik auf den nordwestlichen Salomo-Inseln,
in Richard Thurnwald, Salomo-Inseln und Bismarck-Archipel, vol. I (1912), no. 41.
157
while the first scale, definitely non-quartal, constitutes a plain

hexachord in six-five form. Since this latter pattern also appears
not far from China, for instance in Indonesia to the south and
in Tibet to the west, the C scale must be supposed to belong to
some stratum different from that of the D and E scales.
ANOTHER FORM OF TRIADIC ENLARGEMENT is due to adding the
higher or the lower octave instead of the sixth. In this case, the
melody has the steps one-three-five-eight or, in the plagal form,
low five-one-three-five, such as ce g c' or G c e g. We met with
triadic octave skeletons in the seetion on tumbling strains. The
triadic octaves of the present section are different: they do not
cascade from the highest to the lowest note and therefore do not
recapture the upper octave in order to repeat the pattern.
Instead of stressing the octave, they usually rather accent the
triad as their natural nuc1eus. Good illustrations come from
New Guinea:
from Flores, the Amami Islands, and the western Bantu. A

combination of the authentie and the plagal form - D G B D G occurs in shaman's songs during the Peyote dances of the Huichol
in Mexico. 24 Though the Mexican Indians have to some extent
given way to white man's music, such influence is here quite
out of the question : it would show in heptatonic melodies but
not in a skeleton that at best may be heard on a bugle. The
elimination of white man's influence is corroborated by the
frequent occurrence of triadic octaves in songs of the Copper
Eskimo. 25 But there, again, we should not be satisfied with the
anatomical fact of a triadic octave, but proceed to the physiological facts. In such a skeleton as C E G C the bottom note C is
not at all a root or atonie; the functional stress is on E.
It is fascinating in this case of transition to watch the un After J. Kunst, A study on Papuan music, Weltevreden, 1931, p. 25.
EFw FE 4413 (P 413) II 3.
25 Roberts and Jennes, l.c., pp. 170 ff., 229.
24
158
certain, hesitant attitude of singers. Aiming now at the seventh

now at the octave, they alternately create now a tripie third
now a triadic octave. The following is a curious example: a
French rondeau from the Roman de la Rose (IZth century) first
establishes the triad, then turns to the seventh, but leaves it
immediately to catch the octave, only to return in haste to the
safer, wonted seventh. 26 A similar example of uncertain
octavation will be described in the following section on the Fate
of Quartal and Quintal Patterns.
A more primitive parallel, although in descending order, is
the followingsongoftheBuriat Mongols, 27 built on the quadrupie
third e' c' a f d. In one of its recorded variants it finishes correct1y
on d; in another one, the singer passes over the d and ends on c,
which is the lower octave of the starting note:
Ex. 47
In a similar way, chains can be arrested in their growth asa

step towards tonality. In an old German chorale, Nun bitten
wir den heiligen Geist,28 the principal note, starter as well as final,
is g, and the structural skeleton, the tertial chain e g b d'. The
melody, however, has a wider range; there are two additional
notes, one above and one be1ow. They could have been c and f
sharp, making the whole a quintuple third. Instead, they
abandon the tertial concatenation and repeat the highest note
d' an octave below, and the lowest note e an octave above.
26 Reprinted after Friedrich Gennrich in Heinrich Besse1er, Die Musik des MittelaUers und der Renaissance, Potsdam 1931, p. 104, ex. 61.
27 Ernst Emsheimer, The music ot the Mongo/s, part I (Reports trom the scientific
expedition to the norlh-western provinces ot China, pub!. 21), Stockholm, 1943, no. I.
28 From Constantin Brailoiu, l.c., p. 365, after Charles Schneider.
FIVE
THE FATE OF
QUARTAL AND QUINTAL PATTERNS
Double fourths, usually deseending, ean be paired in two

different ways, either by 'disjunction' or by 'eonjunetion.'
In eonjunetion, the two fourths share one eentral note, which
thus beeomes at onee the final of the upper and the starter of
the lower fourth. The total range of the double fourth is a seventh
or heptad. In disjunetion, the two fourths are separated by a
wholetone, and the total range is an oetave. Graphically:
,---.
eonjunct: d' a e
,---. ,..-...,
disjunct: d' a g d
In view of historical evidenee, especially from the Far East,

India, and ancient Greece, we ean hardly doubt that eonjunetion
marks an earlier phase of development than disjunction. Besides,
eonjunetion shows a more limited planning: the performer
eonsiders one tetrachord at a time; and when the urge for
enlargement creates another tetrachord, the new one starts where
the first has left off, without a dividing spaee between itself and
the older fourth, just as seeondal and tertial patterns are eonjunet.
The two fourths are simply added, not integrated in any higher
organization; the seventh is merely the sum of two eonjunet
tetrachords, not a melodie aim or a function. Exeept for a few
examples from the Orang Kubu in Sumatra,l Mongolia,2 and
Finland:
After A. O. Visnen, Wogulische und ostjakische Melodien, Helsinki, 1937, p. 5.

Erich M. von Hornbostel, Ueber die lVlusik der Kubu, in B. Hagen, Die OrangKubu auf Sumatra, Frankfurt alM., 1908, Phon. 18.
2 P. J. van Oost, La musique chez les Mongois des Urdus, in Anthropos, vol. rO/II
(19 I 5/ I 6), p. 364.
1
160
I do not know of any song where the voice Ieaps to the seventh
without touching the conjunction at a fourth from either end
of the heptad. Disjunction acts in a very different way. The two
tetrachords are placed a wholetone apart (which, seen from a
quartal viewpoint, is entire1y arbitrary) because this distance
integrates the two fourths in a higher organization: the octave.
With all due reserve, I want to draw the reader's attention
to a Greek parallel. According to Plutarch, as a literary, and
many vase paintings, as a visual testimony, the most archaic
Iyres of Greece had only three strings - called nete, mese, hypate,
or the high, the'middle, the Iow one - which were equally tuned
in two conjunct empty fourths. A little later, but still in preclassic times, the Iyre had four strings tuned in two disjunct
empty fourths: d' a g d against the previous d' a e. We do not
know whether or not these fourths were used as bare as they
appear or were stopped to provide all the notes of the oider,
microtone-Iess enharmonic gender. 3
Otherwise, such empty heptads, consisting of nothing but
the three skeletal notes occur quite often in Asia and in North
America: 4
Ex. 49*
Disjunct empty doubIe-fourths exist in episodes inside Mongolian and Korean songs and also in religious recitatives of
Italy.5
Mention of N orth American Indians and, in the footnote,
of the Pawnee draws our attention to an odd trait of doublefourths, frequent everywhere, but almost to the exclusion of
all other patterns in the hako ceremony of the Pawnee: 6 the
* Transcr. C. S.
a Sachs, The Rise 01 Music, l.c., pp. 209 ff.
4 Examples:
Asia: Tibet, Mongolia, Turkmen, Ostyak.
America: Iroquois, Hopi (B. I. Gilman, Hopi songs, in Journal 01 American Arckaeology and Ethnology, vol. 5, (1908), Pawnee (transcribed by Edwin S. Tracy for
Alice C. Fletcher, The Hako, Washington, 194, p. 50).
5 e. g. Constantin Brailoiu, Sur une m4lodie russe, in Pierre Souvtchinsky (ed.),
Musique Russe, val. 2, Paris, 1953, p. 380, 381.
6 Fleteher , l.c., passim.
161
voice leaps to the top note from the lower fifth, not, as one
should expect, from the lower fourth. Obviously, the fifth is a
better, more natural jumping board. This drive might ultimately
be responsible for changing the double fourth into a fifth-onfourth structure, as in the often printed skolion among the
seanty relies of Greek music (Example IO).
It marks adefinite vietory of the oetave when a heptad,
although still unmistakable as such, must tolerate the aeeretion
of an appendix or, with its official name in Greece, a proslambanomenos, a 'to be taken-on one,' to convert the heptad into
an octad. Appended above, it created in Greece a hyper-mode
(like Hyperdorian, Hyperphrygian, Hyperlydian); added below,
it gave rise to a hypo-mode (like Hypodorian, Hypophrygian,
Hypolydian) .
Good examples of pentatonic hypo-modes from outside Greece
are provided by the Buriat MongoIs :
Ex. 50*
While these ean be regarded as traits of higher civilization,

the accretion of an appendix was nevertheless prepared in the
primitive world. The following example from theIroquois Indians
is a tumbling strain out of two empty, eonjunct fourths with
an appended bottom note, from which the melody leaps back
to the upper octave: 7
Ex. 51
i>p r mim r pB I Yj'J! n)YI~jrJYJ;t11

)=IIZ
A less solidified conversion of the heptad occurs both in

primitive and in European music. In a melody of Bongili girls
in French Equatorial Africa, the melody descends stepwise in a
* After Ernst Emsheimer, The music 01 the Mongols, part I, in Reports trom the
Scientific Expedition to the N orthwestern Provinces 01 China, pub!. ZI, Stockholm,
1943, no. Z.
7 Cf. Gertrude P. Kurath, Local. diversity in Iroquois music, in Bulletin 01 the
Bureau 01 American Ethnology, no. 149 (1951), p. II9.
162
double tetrachord and then hesitates; it gropes its way one

second further to reach the lower octave but, unsteadily, seems
to probe the two possibilities. 8
Ex. 52*
We saw a parallel case in the preceding section. This is an

impressive illustration of what must have happened many times
before the octave won out.
Quartal chains achieve their most impressive climax in the
quadrupie fourth of the Sioux,9 which alternates between
disjunction and conjunction and spans two full octaves. An infix
subdivides each fourth into a (higher) third and a (lower)
wholetone. From the central note, the melody leaps back to the
upper octave for a new cascade: this pentatonic quartal chain
is in reality a genuine tumbling strain.
Quartal chains in such astate of thorough pentatonic organization lead from the primitive world to higher and highest
culture. They occur in a puzzling distribution: in the Indian
reservations of North America:
Ex. 53**
as well as in the northern countries of Asia, such as China, Japan

and Mongolia, and among the Chuvash, Vogul, and Cheremiss
- and again, as was the case before, they belong to Mongoloid
races. But for reasons yet unexplained, Scotland and Ireland j oin
the Mongoloids, as does the Solomon island Nisan in Melanesia.
This maybe a warning to keep away from the slippery dallying
with race theories.
Polynesia, Micronesia, Africa, and the European continent
are excluded.
* Transer. C. s.
*. Iroquois, Tutelo, after Gertrude P.
8 EFw FE 4402 (P 402) II x-
Kurath.
e Frances Densmore, Teton Sioux music, in Bulletin 0/ the Bureau 0/ American

Ethnology, no. 61 (1918), ex. 43.
163
Heptatonie quartal ehains are European. - especially Greek,

both ancient and modem, and Gregorian, as in the following
Gloria ad lib. I:
Ex. 54*
Altogether, the main territory of quartal struetures in the

primitive world is Indian N orth Ameriea. They appear seldom
in Melanesia or Negro Afriea and seem to be entirely absent from
Micronesia and Polynesia. In the southem half of Asia, including
the Caueasus and white Afriean cities from Moroeeo to Egypt,
the fourth is all-dominating. This statement, however, is true
only of art music. The trib3J. music of Sumatra for example,lO
is entirely seeondal and tertial, which seems to eonfirm that
the quartal layer is more reeent.
In Europe, parts of the East are quartal, but also Seotland
and Ireland. And so is the (original) Gregorian ehant. European
fourths are single, double, tripie, and, when double, either
eonjunct or disjunet; as a rule, they are not distinctly deseending.
CHAINS OF FIFTHS are rare. The reason is evident: when a
quintal melody expands, it eannot weIl afford to bypass the
oetave in order to pile up another fifth and henee prefers an
oetave-forming fourth above the initial fifth. In one melody of
the Lapps we have seemingly the double fifth - d-a-e'; but the
d is only an appoggiatura in front of every single note and acts
as a kind of interruped drone. Thus the singer leaps at the end
up to the single e' in what amounts to a ninth.
The Lapps, indeed, make their voices often leap a sixth, a
seventh, a ninth, and even a tenth; and in one of their melodies, 11
the two seetions form a double sixth.
There is adefinite example from Annam, beginning on d",
deseending to g', and, via d", aseending to a".12 This would be
an unambiguous double fifth. But the g' oeeurs only one time,
10
11
12
Liber Usualis, Paris/Roma, 1954, p.87.

EFw FE 4406 (P 406) I 1-3.
Kad Tiren, Die lappische Volksmusik, in Acta Lapponica, vol. 3 (1942), no. 246.
New Ox/ora history 0/ music, vol. I, London, 1957, p. 158, after Gaston Knosp.
164
and the example, non-phonographie, is open to doubt and

mistrust.
Otherwise, the urge to sing an aseending fifth is frequently
satisfied by simply skipping one of the thirds in a tertial ehain.
This happens often in the Gregorian ehant, as, for instanee, in
the initial formula of many melodies in the first mode or in the
antiphon 7 for the seeond Sunday in Advent:
Ex. 55*
The Inissing third is however restored after some time and

leaves no doubt that the pattern is a tertial ehain. Similarities
oeeur elsewhere: the map of distribution eharts Franee, including
Lorraine, fifteenth eentury Holland, sixteenth eentury Flanders,
Germany, Italy, Rumania, and Albania; Ashkenazic and Yemenite J ews and J aeobite Syrians; Mongolia; Canadian and Iroquois
Indians, and the Batwa pygmies in Congoland:
THE FREEDOM OF INDIVIDUALS mentioned in earlier sections

and also the ever varying eontacts of man eannot have left the
melodies as 'pure' as our rough classification makes believe.
Interpenetration of human groups, from single families to raees,
or even daily eontact of the two sexes within one village must
have led rather early to a duplicity, if not multiplicity, of melodie
impulses in the same person, indeed, in the same melody. Often,
tunes obeying two different impelling forees give an impression
of uneertainty. We diseussed sueh fluetuations in The Fate of
Seeondal and Tertial Patterns on page 149 when we saw the
**
Liber Usualis, Paris/Roma, 1954 P.486.

After Rose Brande!.
165
hesitant groping now for the seventh now for the octave in what
had started .as a conventional tripie third.
In the semitone-over-third tetrachords described both in the
One-Step Melodies and in the current section of the book, mere
attentive listening combined with a careful statistic analysis
shows vaccillation between the motor impulses of a fourth and
a major third, as if the singer were not quite decided as to where
to turn; and sometimes the fourth seems to be in competition
with a semitonal pattern. Inversely, the interpenetration of two
quite different urges is convincing in melodies of the Caribou 13
and the Smith Sound Eskimo:
where an (obviously older) one-step pattern g-f and a double

third f-a-c are in strong opposition and contention. The Pawnee
provide us with another informative example: of eleven measures,
eight are taken from the single third a-f, and only three, in the
middle, descend to d and C. 14
The following Rumanian folksong shows c1early that a secondal
impulse, yielding the step g-a, combines with a tertial impulse,
materializes on g-e, which again leads to the second e-d. The
melody is not secondal; nor is it tertial. As so many anthropological phenomena, it is due to two or three independent
stimuli:
"
JJ" HZ AB) JI JJJJ JJJV
I B5j jn JJtJ I
Some melodies from N ew Ireland, Melanesia, combine an

* After Christian Leden, Uebel' die Musik der Smith Sund Eskimos, in Meddelelser
om G1'Ilnland, vol. 152 (1952), ex. 19.

** Boceta, after Riegler-Dinu.
18 From Zygmunt Estreieher, La musique des Esquimaux-Cal'ibous, in Bulletin
de la Societe NeucMteloise de Geographie, vol. 54 (1948), p. 5.
14 Transeribed by Edwin S. Traey in Aliee C. Fleteher, The Hako: a Pawnee
ceremony, Washington, 1904, p. 107.
166
(upper) second with a (lower) fourth,15 and some from the Lapps,
an (upper) second with a (lower) fifth:
Ex. 59*
Add similar coalescences among the Celtic settlers of Cape

Breton Island, Nova Scotia,16 and the following gambling song
of the Klallam (Yuki Indians, California) 17 which starts as a
distinctly tertial pattern but changes in the third measure to an
equally unmistakable tetrachord with the note a as the common
pivot:
Ex. 60
The available mass of further examples defies enumeration.

Hybrid from the static viewpoint of the classifier, these coalescences are important steps from pristine narrowness to a freer
and richer expanse.
The most momentous form of coalescence, indeed of modulation, occurs in the province of quartal patterns. The term
modulation, it is true, is taken from western harmony. It
denotes a change of key due to the ambiguity of some chord:
belonging to two or more tonalities (as most of our chords), it
allows to continue in the old tonality as weIl as to glide into a
new one. There is however no reason why we should not use the
eonvenient word in a merely melodie eontext when a pattern
unexpeetedly ehanges to another one with the help of a common
note in which the singer, though often uneonsciously, gives a
new interpretation within the melodie organization.
We saw that double fourths are sung both in eonjunction
and in disjunction and that the former one of these structures
*
After Walter Wiora, Europischer Volksgesang, Heemstede, 1952, p. II.

Herbert Hbner, Die Musik im Bismarck-Archipel, Berlin, 1938, p. II.
16 EFw FE 4450 (P 450) I 3.
17 Frances Densmore in Zeitschrift fr Vergleichende Musikwissenschaft, vol.
(1933), p. 31 *.
15
167
represents most probably the older stage. Modulation appears

quite often as a shift from conjunction to disjunction and vice
versa. In this form it is well known in the Far East 18 and among
the relics of ancient Greece, such as the so-called Cairo Fragment
and the First of the Delphic Hymns. 19 But it also exists on a
primitive level in songs of N orth American Indians (example 53).20
Altogether, we might venture this supposition: at times when
disjunction was still less strongly settled than the older conjunction, the singer slipped at the turning point and relapsed
into the original habit of conjunction. Of the initial derailment
and uncertainty, later ages would have made a virtue in the
interest of greater abundance.
THESE OPPOSITE URGES are part of the delicate problem of
culture change. While the first half-century of Comparative
Musicology was predominantly, indeed excIusively interested in
the static aspects of tradition al standstill, a younger generation
of scholars investigate how this tradition is being challenged
under the impact of a growing interpenetration of countries and
cultures and the irresistible destruction of tribai life. . The
pioneers were basically concerned with the then present form of
music and resented bitterly (and often rightfully) the antitraditional transformations and distortions that music lik~ all
other forms of civilization of necessity underwent in the oneworld trends of the day. All tradition - in today's conception derives from a long process of influences and changes, both from
within and from without. This process is as old as culture itself;
and nowhere has it come to an end. Thus, the day we meet a tribe
cannot be taken as a key-date; nor must we ignore the inevitable
metamorphoses occurring under our very eyes.
This neo-historie concept is particularly strong in the United
States; for here the dash of western and non-western music has
been more vital and fateful than anywhere else. 21
Sachs, l.c., pp. 133 ff.
19 ibid. pp. 244, 245.
20 Kurath, l.c., p. 156.
A bibI. excerpt: B. MaJinowski, in L. Thompson, Fidjian Irontier, S. Francisco
1940, pp. xvii ff.-id., The dynamics 01 culture change, New Haven, 1945. - A. L.
Kroeber, Conligurations 01 culture growth, Berkeley, 1944. - C. Sachs, The commonwealth 01 art, New York, 1946. - E. G. Burrows, Native music 01 the Tuamotus, l.c.
pp. 97 ff. - W. Rhodes, Acculturation in N. American Indian music, in: S. Tax,
Acculturation in the Americas, Chicago, 1952, 127-132. - A. P. Merriam, The use 0/
music in the study 0/ a problem 0/ acculturation, in Am. A nthropologist, 57 (1955),
28-34 - A. Marinus, Tradition, evolution, adaptation, in Jourt!. 01 the Intern. Folk
Music Council, IX (1957), 15 ff.
18
21
SIX
CENTRIC MELODIES
All the melodies we met so far, and indeed the overwhe1ming
majority of melodie patterns, are steps or derive from steps.
Only in the seetion on One-Step Melodies did we describe a
recitativic variety of melodies based on a single, often repeated
note. On a somewhat higher level, steps are by no means absent,
but the melody, freely moving upward and downward, returns
again and again to the same note in the middle, which is often
starter and final as weIl as an ever recurring nucleus in the course
of the tune. Let us call such melodies 'centric.'
I have no claim to lying bare this type of melody: Robert
Lach defined it as early as 1913 1 and called it 'perihelial.' I
rather avoid this name. In the first place, it is incomprehensible
unless the reader knows Greek; in the second place, it seems out
of proportion to compare an often tiny strip of melody with the
solar system; and lastly, the astronomer, in Webster's words,
calls perihelion (the noun) "that part of the orbit of a planet or
comet which is nearest to the sun"; and this is not at all what
we want to express.
Historically, the centric melodies seem to be partly responsible
for the later concepts of finalis and confinalis in modal systems
and of tonic and dominant in harmonie music. These, too, have
a steady 'function' in their me1ody, without being parts of
some pattern of steps.
A plain example from the Lapps will easily show to what kind
of melody we give the epithet centric:
Ex. 61*
After Werner Danckert.

Robert Lach, Natur- und orientalische KuUufvIJlker ('Studien zur Entwicklungsgeschichte der ornamentalen Melopoie. Beitrge zur Geschichte der Melodie,' passim),
Leipzig, 1913.
.
1
CENTRIC MELODIES
169
Of the twenty-four eigth beats, sixteen go to the center f, and

only four to either the upper note gp and the lower note c. The
central note assumes two thirds of the entire phrase; and it
also serves as the starter and the final. No explanation is needed.
Many parallel examples can be drawn from the liturgy of
Babylonian J ews. One, the creed Harken Israel, has been given
earlier in this book as Example 24; we might add the beginning
of the Song of Songs, as chanted by the same Oriental-J ewish
group:
Ex. 62*
~P J n
mJ n DW
I) J
Dn n J) J J I
The relative chronology of centric patterns is in my opinion

given by two facts. First: they cannot have existed in the stratum
of one-step tunes, as they need at least one upper and one lower
step with the center as the common basis and pivot. Second:
. we know enough examples of centric melodies with only three
notes or two steps to provide a justification for placing the
origin of the whole type in the two-step layer.
On hearing a melody of two steps with the middle note accented
in weight, position, and frequency, we pause a moment to ponder
two possibilities: (1) is it a truly centric melody, or (2) is it a
one-step melody plus an affix above or below?
The example from the Lapps is unambiguous: the center has
four times more importance than either of the outer notes.
But other melodies present a different picture. Two examples
come to my mind, one from the very primitive Vedda in inner
Ceylon and one from the Mongolian lamas. Again we secure a
correct answer by the statistical method: in what number of
time units do the notes occur? In the Vedda song:
)UrFrrr I UrrrerlUrru r
~ +
After Idelsohn.
After Myers.
>
>
170
CENTRIC MELODIES
the central note, in all its recurrences, extends over 22 units,

the upper note over 6, and the lower note over I7. The evaluation
cannot be doubtfu1; numerically, the upper note counts hardly
at all, while the lower note moves in number so elose to the middle
one that the verdict is beyond hesitation: the Vedda melody is a
one-step tune with a suprafix. This decision, it seems, is confirmed by the important fact that the lower, not the middle note
is allowed to rest on a halfnote and even carries repeated sforzati.
The Mongolian piece is in a opposite condition: the middle note
has I9 units, the upper one II, and the lower one a mere 3. This,
too, is a one-step melody; but from time to time it has an
infrafix:
nJI J J JJ I J )J )1 J jJ JI J 1 JI~ J JJ JI J
Only where the middle note is essentially more frequent than

either one of the outside notes are we justified in speaking of
centric tunes.
The distance from the centre to the outer notes is of no
consequence. The Lappish example has a semitone above the
central note, and a wholetone below; others, from the South
Californian Luisefio Indians:
Ex. 65*
and also from German folksong, have a wholetone above and a

minor third below. This quality places the little Russian melody,
around which Constantin Brailoiu bui1t his study of scales: 2
the center g' qualifies as such with its 9 units against the 4 units
of a' and the three units of d'.
Innumerable centric tunes are double-thirds, concealed from
correct diagnosis by the usual misnomers 'triads' and 'fanfares.'
* After
Helen H. Roberts, Form in primitive music, New York, 1933, p. 24.

Constantin Brailoiu, Sur une melodie russe, in Pierre Souvtchinsky (ed.), Musique
Russe 11, Paris, 1953, p. 330 no. I.
Z
CENTRIC MELODIES
171
A case of particular interest is the double-fourth, in which the

central note, the 'dynamic mese' of the Greeks, is the hinge and
meeting point of the two tetrachords. Ancient Greece provides
an evidence in the so-called Second Delphic Hymn, where the
center acts at once as the starter and as the final. Related
examples are frequent in the liturgy of the Syrian J acobites 3
and also in songs of Rumania and the Ostyak :
Ex. 66*
Although the center acts usually both as a starter and as a

final, this is not compulsory. At times, it begins the melody, but
does not conclude it; at times it ends it, but does not start it.
One of our most primitive examples, from Central New Guinea:
Ex. 67
~rrrf'11
is a tiny motif of only four notes on three pitches; the two outer
notes, at the distance of a minor third, make the beginning with
a single beat each, while the central note is touched twice and
stands for three beats. The example is remarkable because the
central note comes last. But the two outer ones above and below
cannot be thought of without amental anticipation of the
coming center.
A great many centric melodies can be safely derived from such
one-tone recitations as we met in the First Part of this book.
I think of melodie configurations whose central note is so often
repeated that any initial, intermediary, or final deviations are
almost negligible. Evidence of such centric repetition is nowhere
as strong as in the so-called Gregorian psalm-tones which provide
the models for performing the psalms in the Catholic Church.
The center, termed the tenor, tuba, or repercussa, is repeated as
After Visnen.
8 Cf. no. 1 in A. Z. Idelsohn, Der Kirchengesang der ]akobiten, in Archiv fr Musikwissenschaft, vol. 4 (1922), p. 369.
172
CENTRIC MELODIES
many times as the text requires ; an initium precedes the tenor

from below, a Ilexa allows the voice to relax on a lower pitch
before the mediatio describes a semi-cadential turn around the
center. Then, the tenor 'proceeds with the second line of the
text and ends in a terminatio:
Ex. 68
These tenor melodies of the Catholic Church lead us from the

simple two-step forms of centric singing to richer structures.
We find them without going back to the Primitives or to archaic
liturgies; there are many examples of two notes above and two
notes below the center in the treasure house of the nineteenth
century. One is the introduction of Schubert's Ninth Symphony,
another, the slow movement in the First Symphony of Brahms:
Ex.6g
IJ\
r r rl J. ) f' Ir ~ r Irr r IJ. ) r Ir' (' I
Therewith we have left the realm of simple two-step centricity

- one step up, one step down. Practically, the number of circumcentric steps is illimited. A good illustration is the following
prayer of the Jacobite Syrians 4 with the range of a minor
seventh and the ever recurring center C sharp:
Ex. 70
But in making our way through the province of wider centric

melodies, we must make sure not to call a melody centric unless
its focus is more or less in the middle of its span. It is not centric
when in principle it reaches upward except for a leading note
below the center.
4
A. Z. Idelsohn, l.c., p. 372.
CENTRIC MELODIES
173
The importance of the centric type in instrumental melodies

has been mentioned at the end of our section on instruments.
The terminology of ancient musical instruments and their
tunings and scalar systems seeins to corroborate the vital
impact of a tonal center. The Greeks expressed their musical
system and the instrument on which they studied music by the
same word lyra; they also called the string and the individual
pitch by the same word chorde; and they arranged their whole
material of pitches around a mese or 'center,' which served as a
common name to denote three different concepts: the middle
string of the lyre, the center of the musical space (as the 'thetic'
-mese), and the center of each of the modal octaves (as the 'dynamic' mese). India knows in a similar sense a madhyam;
and even the modern West speaks popularly of 'middle C,'
taken from the piano keyboard, as a more descriptive name for
what otherwise and more professionally is called the one-lined c'.
Frequently we find in early instrumental music a centripetal
quality resulting from this middle position: a higher and a lower
note, or several of them, play around a central note as butterflies
play around an attractive flower. This even occurs where the
player needs not lean over to hit the more distant slabs of a
xylophone or metallophone; wind and stringed instruments are
often fingered in a constant return to some immutable center.
Centering melodies reach a cllmax in the neckbreaking eoloraturas in whieh oriental and Mediterranean pipers or fiddlers
dissolve their tunes past reeognition - provided that there is a
melodie idea behind the foamy easeades. This dizzying, aerobatic
style includes Pakistan and India and, no doubt, also ancient
Greeee, where one poet aeeused an aulos or oboe player for being
'loquacious. '
OUR CLASSIFICATION of melody patterns eomes at this point
to an end. An intrieate system with all possible ramifieations
of melodie forms ean be found in one of Kolinski' s reeent papers. 5
With all due respeet for sueh eourageous endeavors, I believe
that detailed classifications are desirable and necessary in the
5 Mieczyslaw Kolinski, Tke stl'uetul'e 01 melodie movement, a new .metkod 01 analysis,
in Miseelanea de Estudios dedieados al DI'. Fernando Orliz, La Habana, I9S6, pp. 881918.
174
CENTRIC MELODIES
realms of nature objects and of human artifacts but impractical

and useless in the immaterial world of human expression. Indeed,
even if it were fruitful to catch static forms in a coherent system,
dynamic, kinetic types' elude c1assification beyond a mere
minimum. Open to permanent change, growth, and mutation,
they are doomed to lose their live individualities when dissected
and labeled with a deadly terminology.
SEVEN
POLYPHONY
The word polyphony marks the performance and perception

of more than one note at a time. Western terminology distinguishes two basic concepts of polyphony. One is 'harmony' or 'vertical'
polyphony: we hear simuletaneous sounds or 'chords' in a lawful
sequence of tension ('dissonance') and relaxation ('consonance').
The other concept is 'counterpoint' or 'horizontal' polyphony:
we hear a lawful coexistence of voice parts or simultaneous
melodie lines. Actually the two concepts overlap: counterpoint
obeys the laws of harmony, and good part-writing in harmony
. requires contrapuntal skill.
Both concepts have long been claimed for the medieval and
later West exclusively and thought to be beyond the minds and
means of Antiquity, the Orient, and the Primitive world. In a
narrower sense this is true. When we speak of part-writing, we
imply the salient point: most of our harmony and counterpoint
depend indeed on notation and careful elaboration. They are,
not exactly what the eighteenth century called 'learned,' but
at least 'literate' music and therefore out of reach of non-western
civilization.
In a wider sense, rudimentary polyphony as an improvised
spice of simple monody is almost all-present. Way back to the
Eskimo, there is hardly a single civilization without at least the
rudiments of some sort of mingling notes and parts. Indeed,
three large seetions of the world,
(1) the Malayo-Polynesian area, including Melanesia, New
Guinea, Indochina, and the ancient Far East,
(2) the Caucasian area with radiation to the northwest and
the west,
(3) Central and South Africa,
have created polyphonies on a level to be compared with the
part-writing of medieval Europe.
176
POLYPHONY
A study of these eoincident sounds yields baffling results. 1

Against our inherited miseoneeptions, today's monophony is
here and there in the oriental and primitive worlds an end
stage of what onee was polyphonie. The Far East with its early
surrender of the polyphonie so-ealled Right Musie 2 gives
examples as good as do the destinies of early Gregorian ehant
and the transition from multipart to solo madrigals at the end
of the sixteenth eentury. And twiee during the seventeenth and
eighteenth eenturies European musicians abandoned a polyphonie stile antico for a monodie stile moderno.
Our doeumentation is far from exhaustive; strange things
have happened in reporting musical facts. To refer to a single
example: the northwestern N ootka Indians, so Morris Swadesh
relates, sing a sustained drone and two moving parts, whieh an
oetave duplieation by women might bring altogether six parts. 3
But in the same book, Helen Roberts eonfesses that Edward
Sapir's phonographie reeordings "do not offer eonclusive
evidence as to what actually happens, since they were generally
given by only one informant singing into the recorder (!)." 4
It is not possible to squeeze the many different forms of
simultaneous otherness into a neat historical sequence. Reaching
down into the primitive world of prehistory and up into the
western world of today, they overlap, grow complex, and again
yield to simplicity, so much so that any attempt at chronology
is doomed from the very beginning.
1 Erich M. von Hornbostel, Ueber Mehrstimmigkeit in der aussereuropischen
Musik, in Kongressbericht der Internationalen Musikgesellschaft, Wien, 1910.
Marius Schneider, Geschichte der Mehrstimmigkeit I, Berlin, 1934. - id.,
Die musikalischen Beziehungen zwischen, Urkulturen, Altpflanzern und Hirtenvlkern,
in Zeitschrift f#r Ethnologie, vol. 70 (1938), pp. 287 ff. - id., Kaukasische Parallelen
zur mittelalterlichen Mehrstimmigkeit, in Acta Musicologica, vol. U (1940), pp. 52-61.
- id., Ist die vokale Mehrstimmigkeit eine Schpfung 'der Altrassen, ibid. vo1 23
(1951), pp. 40 ff.

Z. Estreicher, La PolYPhonie chez les Esquimaux, in Journal de la Societe des Americanistes, N.S., vol. 37 (1948), p. 259.
M. Antonowytsch, Die Mehrstimmigkeit in den ukrainischen Volksliedern, 'in
Kongress bericht der Internationalen Gesellschaft fM Musikwissenschaft, U trecht,
1952, pp. 55-64
EFw P 504 I 3.
For a practically exhaustive bibliography see: Jaap Kunst, Ethnomusicology,
The Hague, 1959, p. 241 S.V. multi part music.
2 Curt Sachs, The rise of music in the ancient world, New York, I943, p. 46.
8 Helen H. Roberts and Morris Swadesh, Songs of the Nootka Indians 01 Western
Vancouver !sland, in Transactions 01 the American Philosophical Society, N.S. vol.
45, part 3 (I955), p. 324.
4 ibid., p. 222.
POLYPHONY
177
At random, we start from a type reminding, superficially,

of western harmony and especially of the medieval conductus:
Polynesian chords. The word chords is handy if inadequate, as
nothing indicates a harmonie feeling in our sense, with the
tension and relaxation of dissonance and consonance from chord
to chord. On the westem Polynesian islands, each note of a
melody is often accompanied by a coincident bass note. The two,
beating strictly together, are 'isochronous.' There are usually
only two parts - melody and bass. When there happen to be
three, the third part duplicates one of the two in the octave.5
.. Wherever, in or outside Polynesia, the two parts keep the
same distance throughout, we speak of parallel singing - whether
it is intentional or else unintentional and hence not meant to
be polyphonie.
A PARALLEL, as this seetion defines it, is tM simultaneous
singing of a melody on two or three noticeable different pitches.
The voice parts proceed at a certain interval and keep it mostly
unchanged throughout a piece or only for a shorter stretch.
In the simplest case, the interval amounts to an octave and
results in the East as in the West from the natural distances
of men's, women's, and children's voices. Fifth and fourth
parallels are almost as frequent as octaves. The West knows
them from the older form of the organum in medieval holy day
liturgies of the Roman Church. But we also hear them in folksinging instead of the unison that the singers intend and believe
to render.
.
Examples from extra-European fifth and fourth parallels are
too numerous to encourage cataloguing. As two hardly known
evidences I would like to mention the solemn fourths in the
chant of Japanese Buddha priests:
Ex. 71*
EFw FE 4449 (P 449) (Transer. C. S.).

5 E. G. Burrows, Songs 01 Uvea and Futuna, in Bernice P. Bishop Museum Bulletin
183, Honolulu, I945.
178
POLYP}{ONY
and those of the archaic Takasago in the inner, 'inaccessible'

mountains 'of Formosa:. The most ancient among the Takasago
clans, according to a J apanese source, sing "in chorus, using the
most beautiful chords.'~ 6 When I mentioned this mystic polyphony. to Jaap Kunst, he produced on the spot a dubbed rei.:
cording f this enigmatic music which he happened to have in
his luggage. 7 It was a rare treat to listen to the "most beautiful
chords" sung by a men's chorus in an organ-like hum without
text. Actually, they were long drawn-out notes in a slowly
rising chain from an awe-inspiring low Du to d, at first in microtones and later in widening steps, with the accompaniment of a
second voice part, climbing, too, but lagging behind and almost
independent, so that the listener percieves ever different intervals,
of a second and a third, a fourth, a fifth, and a sixth, now in
painful friction, now in blissful consonance. One can hardly
imagine a truer musical rendition of the seeds which germinate
in the dark of the soil, laboriously grope to the light, and then,
in the open, are free to thrive. For this is exactly what the eerie
chorus means: acharm to help the growth of the millet.
We know or can infer that fifth and fourth paralleis are
sometimes retrogressive traces of richer choral forms. The
Chinese mouth organ, today reduced to simple parallels, once
played pentatonic chords or clusters 8 as it still does in the
venerable gagaku music of the J apanese court; and the organum
of the medieval Church, far from being an awkward beginning,
is likely to be an offshoot of ancient chordal practice, possibly
passed on by the paraphonistae or 'fifth-singers' of the papal
schola cantorum in Rome. 9
Third paralleis, known in the history of European music from
the medieval gymel = cantus gemellus or 'twin song' of England,
are in folksongs actually used all over Europe.
Small wonder that European explorers, finding all the way
through Bantu Africa the same kind of parallel thirds, partly
8 Takatomo Krirosawa, The musical bow of the Bunun tribe in Formosa, in Toy8
ongaku kenkyft, Dec. I952, p. 2.
7 Afterwards publishedby Constantin Brailoiu in the World collection of recorded
folk music, vol. V rec. no. 1.
8 Curt Sachs, The rise 01 music, l.c., p. 146.
9 atto Kinkeldey; in a paper read before the Annual Meeting of the American
Musicological Society at Princeton, December I955, tried to identify them as drone
sinl;l"ers.
POLYPHONY
179
major partly minor, thought of European influence via the

Mediterranian or colonial settlements. It was perplexing to find
harmony where harmony had no rightful place. Unfortunately,
we are always prone to give chronological and causal priority to
things learned first. We wronged the Bantu, however. There
is not the slightest doubt that the third paralleis are just as
African as they are European. Negro melody, as we saw, is
greatly built on tertial structures, and some of their pre-European
instruments, such as the Madagascan tube-zither of bamboo,
are tuned in consecutive thirds. 10
An even more conc1usive argument is the fact that parallel
thirds occur on essentially lower, pre-Bantu levels, We meet
them for instance on the western Caroline Islands in Micronesia
as a frequent feature of a very archaic musical style. 11
Ex. 72
The Carolinian melody consists sometimes of only three notes a

second apart from each other - music could hardly be more
primitive. And despite such primitivity, the accompaniment
seems to form alternately major and minor thirds like ours. It is
true that in the case of consecutive thirds, we must question the
printed transcriptions, no matter how excellent otherwise. For
our ears are never more biased than where they are conditioned
by western similarities.
Even if we hear them correctly, neither the Carotine nor the
Bantu parallels are exactly harmony in the sense of western
'functional' chord writing. But there is little doubt that this
go-together represents the seed of what in Europe grew to
chordal concepts.
Conc1uding our survey, we meet with a startling kind of
parallel, almost unbearable to our ear, in a wide stretch from
Micronesia to South Africa and the southeastern quarter of
Europe: in the Pacific, we record, from east to west, the western
10 Curt Sachs, Les instruments de musique de Madagaseal', in UniversiU de Paris,
Travaux et Memoires de I' Institutd'Ethnologie 28, Paris, 1938, p. 53. - Cf. EFw FE 4451
(P 451) II I, FE 4504 (P 504) I I.

11 George Herzog, Die Musik der KaroUnen-Inseln, Hamburg, 1936, passim.
180
POLYPHONY
islands of Polynesia,12 the Melanesian Admiralty Archipelago,

and the Micronesian West Carolines:
Ex. 73
in Indonesia, the Moluccas and Flores; 13 in Africa, the Babira

(Bantu), the Mambuti (pygmies), and the Zulu (Bantu again) in
the South. There in a constant grating friction, the one or two
steps of the melody are accompanied at the minimal distance
of an often nondescript second. Strangely enough, the second
paralleis recur in Russia, in Istria at the northern tip of the
Adriatic Sea,14 and in Bosnia. 15 They do however not occur in
modern Lombardy, as has been asserted, although, according
to Franchinus Gaffori, a contemporary, they were used in the
fifteenth century for mourning services at the Cathedral of
Milan.1 6 The dosest kinship between European and non-European folkmusic is once more apparent.
How innate and unconscious these seconds are, though incomprehensible to the western ear, can be gathered from the
already quoted letter that Dr. Walter Lurje kindly wrote to me
in 1949 from Puerto Rico. Having been a Chinese army doctor
from 1939 to 1946, he had witnessed time and again how recruits
from primitive tribes in inner China, a few hundred at a time,
were every morning on raising the flag compelled to sing the
(westernized) national anthem. "At first it was impossible to
find a trace of the melody in their singing or, better, howling.
But aftera week or so, the howling had abated, and the men
12 E. G. Burrows, Songs 0/ Uvea and Futuna, in Bernice P. Bishop Museum
Bulletin I83, Honolulu, I945.
13 Jaap Kunst, Music in Flores, Leyden, I942, p. 8.
14 Ludwig Kuba, Einiges ber das istro-dalmatinische Lied, in Bericht d. III.
Kongress der Internationalen Musikgesellsckajt, I909, pp. 27I-276.
15 EFw FE 4434 (P 434) II 5.
16 Ernest T. Ferand, The "howling in seconds" 0/ the Lombards, in The Musical
Quanerly, vol. 25 (I939), pp. 3I3-324. - Cf. also: Gustave Reese, Music in the Renaissance, New York, I954, p. -I79.
POLYPHONY
181
sang in parallel seeonds. After a few more days, oeeasional

fifths appeared, and about a week later they sang in perfeet
fifths throughout. After a while, some oetaves eropped up, and
ultimately they sang in oetaves throughout. I never heard
thirds, fourths, or sixths. This happened again whenever a new
bateh of rookies eame, and always in the same sequenee." 17
Whether or not this interesting development is eompulsory
and universal eannot be stated without similar reports from
other eountries.
A warning is in plaee not to hold 'harmonie' thirds against
'non-harmonie' seeonds. In primitive music, we must not speak
of harmonie relations in our sense; besides, the two intervals
are not as strictly segregated as on our pianos and often run
into one another.
The ease of seeond paralleis must be dismissed from the realm
of vertical hearing and dissonanee. Otherwise we eould not find
in a eountry as musical as Lithuania actual folksong Canons 18
with eonsistent elashes of semitones and wholetones:
Ex. 74
ft
Attention, strietly horizontal and foeused on the re-entry of

the theme in another voiee every four measures, eompletely
ignores the eonstant frietions on our vertically trained ears.
DRONES OR SUSTAINED NOTES, above or below the melody,
are sometimes deseribed as pedal notes. But this term, borrowed
from European music, must not be used; it is too elose to the
17 Walter Lurje's letter is translated in ]oumal 01 the Amel'ican Musicological
Society, vol. 3 (x950), pp. 292 ff.
18 EFw FM 49 (P w09) I x, with transcription by Jonas Balys in the accompanying notes on p. 4.
182
POLYPHONY
specific practice of holding a foot on one of the pedal keys of an

organ and too remote from what it means in the primitive world.
Besides, it easily suggests some note far below the melody and
becomes incongruous when - as so often happens in primitive
and orient al music - this note lies in the range of the melody
or even above.
Sustension, highly important in Middle Eastern music and
indispensable in India, has a comparatively small impact on
primitive singing. When a Kubu woman in Sumatra sustains
one shrill high note while a man sings a simple two-step melody,
she is unlikely to consider a two-part coperation.1 9 This is
more probable in the case of the Lifou women in the Loyalty
Islands who throw the same discordant note into each of the
endless repetitions of the short melodic1e that the chorus incessantly sings. 20 The primitive drone that has impressed me
most is held in a song of the Temiar in Malaya: it impersonates
a priest whom the mountain spirits warn in his trance to beware
of diseases that the white men might bring to his people. The
narrowmelody, on thestep f-g, moves above an eerie, intermittent
drone on c. 21 In another Temiar song, the static one-note drone
becomes a quartal-step ostinato on g-d, while the melody alternates between a and g:
Ex. 75*
tlrvm
r r
f r r
f r r
On the level of high civilization, the drone was given a focal,

indispensable role in the Indian subcontinent; 22 and from there
it reached the folk music of Southeast Europe: with the Turkish
conquest of the Balkans, drones affected the Yugoslavs; a longnecked lute of Middle-Eastern provenience, the tamburica, often
* EFw FE 4460 (P 460) I 4 (Transcr. C. S).
19 Erich M. von Hornbostel, Die Musik der Kubu, in B. Hagen, Die Orang-Kubu
aul Sumatra, Frankfurt am Main, I908, phonogram no. 25.
20 Curt Sachs, The rise 01 music in the ancient world, New York, I943, ex. 27,
transcribed by the author from a recording in the Musee de l'Homme.
21 EFw FE 4423 (P 423) I 6.
22 Sachs, The rise 01 music in the ancient World, New York, I943, p. I81.
POLYPHONY
183
accompanies their songs with a high-pitched, ever repeated

note,23 as do the instruments in India.
This is of course not the sustained drone which expands below
the melody in an uninterrupted horizontal. In the Polynesian
Pacific we even find what I would like to call a synchronous drone
following the melody beat on an unchanging pitch 24 (cf. p. 177).
Drones are found everywhere in a westward belt across the
Pacific, India, the Caucasus, the Muhammedan countries of the
West, and the eastern parts both of Africa and of Europe.
Additionally, they occur in Mediterranean Antiquity and wherever bagpipes, hurdy-gurdies, and the Mediterranean fiddle
('lira') are played. In the Orthodox Georgian Church, the drone
changes its pitch from stretch to stretch,25 as does the vox
principalis of organa in the Roman Church of the twelfth
century. The distribution of drones seems to coincide with that
of quartal patterns.
At times, as we saw, the drone consists in two alternating
notes, transforming the simple drone into an ostinato ground.
A recorded example comes from the Kpelle in Liberia; 26 a
three-note ostinato, played by a bard's accompanying fiddle in
East Africa, has been published by Erich M. von Hornbostel:
Ex. 76*
The Naga tribes in Assam, Burma, establish their chordal

drone as the ostinato repetition of a harmonie cadence formula,
known to musicians as
7)5
2
From Alrican Negro music, in Africa, vol. I (1928), p. 20.

23 EFw FE 4434 (P 434) II 2.
24 Cf. E. G. Burrows, Songs 01 Uvea and Futuna, in Bernice P. Bishop Museum
Bulletin 183, Honolulu, 1945.
25 EFw FE 4504 (P 504) I 3.
26 EFw FE 4465 (P 465) I 5.
184
POLYPHONY
while the soloist proceeds uneoncernedly. Therewith we are

very elose to 'alternation.'
ALTERNATION AND CANON. Intermittent sustentions of a note
are early steps on the way to the two world-wide devices of
alternating performance: antiphony and response or respond
(cf. pp. I27 ff.). Antiphony is the alternation of two choruses, with
or without overlapping; and response, the alternation of a
chorus and a soloist. Either form keeps the performers from
exhaustion and the listeners from boredom ; they satisfy the
aesthetie sense of neat artieulation and symmetry and add
dramatie spiee in connecting performers and audience.
Such alternation, in itself a matter of structure, has developed
to polyphony in Asia and in Africa. On a Stone Age level, J enissei
Siberians in their responses of the shaman and his aides, the very
primitive pygmies in the jungles of Malaya,27 Samoans arid
Australians,28 and the diminutive tribes in the swamps between
the sources of the Nile and the Congo:
Ex. 77*
as well as Bushmen and Dama - they all developed overlapping

responses into regular canon singing. After this initial shock,
it was a lesser surprise to hear the women of the Nage tribe in
West Flores in Indonesia sing such canons over double drones on
tonic and fifth that the men sustain. 29 The primitive canons of
the Mafulu in Australian New Guinea 30 seem to be merely what
*
After M. Kolinski.
M. Kolinski, Die Musik der Primiti1lstmme auf Malakka, in Antkropos, vol. 25
(1930), pp. 588 ff.
28 A. P. Elkin, Arnhem Land music, in Oceania, vol. 24 (1953), p. 97, reprinted
in Oceania Monograph No. 9, Sydney, 1957.
29 The beginning of such a canon, transcribed by ]aap Kunst, is reprinted in Curt
Sachs, The rise of music in the ancient world, New York, 1943, p. SI. - See, for more
Florinese canons with and without drones, ]aap Kunst, Music in Flores, Leyden,
1942, pp. 52, 77, 78, 81 and 86.
80 Robert W. Williamson, The Mafulu, London, 1912, p. 217.
27
POLYPHONY
185
the Germans caIl Stimmtausch or alternate presentation of the

leading theme and its accompaniment in either voiee.
How spontaneous, unpremeditated, and nearly unconscious
this kind of polyphony can be, is described in the continuation
of Dr. Lurje's letter: "When the stage of the octave was reached
(sometimes even before, in the stage of fifths) some soldiers
failed to start in time and therewith produced canons in two or
more voiee parts at the distance of a half, two, or more
measures." 31
It is evident that the strict rules of consonance in western
canon writing do not apply to unvoluntary canons. They do not
even apply to intentional folk rounds on European soil, mentioned before in this section on describing the Lithuanian rounds.
They do not apply, because the listener's attention, focused on
succession instead of coincidence, is obviously interested in the
regular re-entries of the theme, but not in the vertical dux-comes
concords which our harmonie training makes so important.
Still, nothing could be a more impressive warning against
the prejudice of a 'plausible' evolution from simple to complicated forms. Nothing shows more c1early how much the concepts
of simplex and complex are biased and unreliable.
derived from the Greek word hiteros, different, is a vague and noncommittal expression. It covers, or should
cover, a11 possible types of otherness in voiee coperation, between the opposite extremes of unison and of invertible counterpoint ; it should designate the slightest deviation from a tolerably
accurate unison as weIl as a Bachian quadrupie fugue - they aIl
are 'other-sounding.' From the beginning of this century, when
earl Stumpf revived the old Greek term (whieh the Greeks
themselves had never clearly defined), 32 the word served basieaIly to denote the simultaneous appearance of a theme in two
or more voice parts with a freedom that the nature of the
competing voiees or instruments and the players' fancy might
prompt. Taken in this sense, the customary octave parallelism
HETEROPHONY,
Lurje, I.C.
32 earl Stumpf, Geschichte des Konsonanzbegritfes (1897), in Abhandlungen der Kgl.
Bayerischen Akademie der Wissenschaften, PhilosoPhisch-Philologische Klasse, vol. 21
(1901), p. 1. - Guido Adler, Ueber Heterophonie, in Jahrbuch der Musikbibliothek
Peters, vol. 15 (1909), pp. 17 ff.
31
186
POLYPHONY
of 'celli and doublebasses in a modem score would be heterophonie when the heavier basses simplify some rapid passages
for reasons of technique.
This definition is however both too wide and too narrow~
The divergence of voice parts can obviously be either unconscious
or conscious, according to whether the performers are not aware
of creating heterophony, and conscious, when heterophony is
an intentional enrichment.
Unconscious heterophony is, psychologically speaking, a nonpolyphonie type of music. The performers as wen as the listeners
accept it as homophonie; they ignore occurring consonances
and dissonances and even tolerate, as unimportant, careless
entries, retarded conc1usions, and the haphazard lengthening
or shortening of notes. Any congregational participation in
modem church music provides examples, even when the organ
and a professional choir support the singers. Such anarchie
singing would be unbearable if intention and attention were
focused on satisfactory sense perception, meaning, on art.
Instead, we behold the performance as an idea, in the philosophical sense of the word - an idea in which perceptive elements
like rhythmic precision and pure intonation are repressed. Apt
to detract from the sacred words and the mood of devotion,
they seem irrelevant or even undesirable.
We might compare such unconscious heterophony with the
natural, leisurely walk of a group of people who move in the
same direction without keeping unchangeable abreast or caring
for equal steps. These they leave to the unnatural, rigorous
unison of marching paraders.
Where a single voice or instrument is being accompanied,
heterophony begins to be conscious. Good examples are found
in Europe, Asia, and Africa, where the old-time bards still chant
their ancient epical poetry and relate the immortal deeds of
national heros. Here as elsewhere the same melodie idea is allpresent in either element of the performance, the singer's voice
and the supporting fiddle or whatever stringed instrument is
used. But there is never a rigid unison. So much are the two
parts distinct individuals - even when the singer is his own
accompanyist - that strict coincidence would seem artificial,
empty, and dead. Free walking side by side is essentially more
POLYPHONY
187
natural than drill parading in a stiff-Iegged left-rightleft-right. The

accompaniment is not only free to proceed or to rest. As a rule,
the singing voice delivers its melody in a simple, almost sober
form, while the accompanying mstrument presents the same
melody profuse1y embroidered and dissolved in florid coloraturas.
This kind of heterophony is best characterized as 'simultaneous'
or 'coincident variation.' I t is irrelevant whether or not coincident
variation leads to dissonances and grating frictions; vertical
hearing must be discounted almost completely.
A normalization of such heterophonies - again in a purely
horizontal sense - can be observed in Japan, where the accompanying instrument follows the singer in free variation at the
respectful distance of an eightnote without disturbing or confusing the listener with its random con- and dissonances.
Curiously enough, Richard Wagner has paralleled this Japanese mannerism in the score of Lohengrin (EIsa in act I scene 3:
"Mein Schirm") by forcing the soprano voice to follow the
c1arinet at the distance of an eight 33 - a passage that our
performers in their vertical training rarely bring out satisfactorily.
If heterophony is the simultaneous variation of one theme in
two or more parts, practically all non-European ensembles,
inc1uding orchestras, are heterophonic. Certainly they are
heterophonic not always in the purest sense. Gagaku, the classic
J apanese court music, of which finally an excellent edition in
full score has been published,34 assigns to flutes and oboes
heterophonic parts, and sometimes the lutes and kotos join them.
But the gongs and drums seem independent, and the mouth
organs, forming clusters in the highest range and perpetually
driving the pitch upward, determine the general impression
almost more than the heterophonic lines.
The most widely known Far Eastern type of orchestra is the
gamelan of Indonesia, in which the essential, kernel- forming
instruments, gong chimes and metallophones, are often built
in three sizes each, an octave apart from one another, to play
the same melodie idea at the same time in a fast, a moderate,
Musical examples in Curt Sachs, Rhythm anti tempo, New York, 1953, pp. 66, 67.
Score 0/ Gagaku, transcribed by Sukehiro Shiba, vol. 1 (Tokyo, 1956). Reviewed
by Rose Brandel in Journal 0/ the American Musicological Society, vol. 10 (1957),
pp. 39-44
8a
84
188
POLYPHONY
and a slower reallzation, if not in what our western eounterpoint ealls diminution and augmentation.
Rigid diminution and augmentation, on the other hand,
oeeur as dialogues of two hand-beaten drums in the refined
eharnber music of India: arnong the many sophistieated types of
duos, one eonsists in the simultaneous playing of some metrieal
pattern or tala and, on the other drum, its augmentation in the
ratio 2 : 1. 35
Not everywhere does heterophony attain to deviees so artful
or even at symphonies as rieh and eolorful as we admire them
in Java and Ball. Still- mueh as heterophony is neither harmonie
nor quite polyphonie and seemingly anarehie - the wilful
maladjustment of similar melodie lines has often a particular
eharm in its blissful impression of personal freedom against
meehanistic bondage.
Vocal polyphony of an ahnost harmnie type oceurs in the
music of wandering minstrels in South Rhodesia 36 as well as
in Southeast Europe. Croatian mixed choruses proeeed in unison,
octave; and third parallels, but finish the stanza with a full
ehord on the subtonium and resolve it into an empty octave. 37
Indeed, we find similar peasants' ehoruses in many parts of
the Austrian Alps,38 Slovakia,39 the Caueasus, the Ukraine and
Russia.
An analysis of these strange and enehanting improvisations
yields, for all their differenees, a few reeurring traits. All principal
cadenees end in unison or in an oetave, the latter rarely with an
added fifth; and a leading note in the preceding ehord remains
unsharped. The number of participating voiees goes seemingly
up to five; but only two or three are aetual parts, with Stimmtausch, or alternation of singers, in one or two of them. At least
one part is a drone, whieh however shifts by a tone upwards or
downwards after an undefined number of measures. This shift
is obviously eonnected with the ehanging paralleis that the upper
voices form - now fifths now fourths. Third paralleis, too, ean
Curt Sachs, The rise 01 music in the ancient world, New Yark, 1943, p. 190.
A. M. Janes, Alrican music, Livingstane, 1949, p. 12, pI. I B.
37 Cf. EFw FE 4434 (P 434) I I.
38 Viktar Karda, Genuine lolk PolYPhony in the Austrian AlPs (summary), in
Journal 01 the International Folk Music Council, val. 9 (1957), pp. 9 ff.
39 Frantillek Palaczek, Slovakian lolk song, in Journal 01 the International Folk
Music Councit, val. 9 (1957), p. 13.
86
3e
POLYPHONY
189
be observed, but they might belong to another stratum. 40

Altogether, every single element in this polyphony is arehaie and
even primitive; the drones, the thirdless eadenees, and the
parallel as the essential form of tonal coincidenee. These elements
are known from the Pacifie in the East to the West eoast of
Afriea and, aeross the Atlantie, in the older Ameriean settlements
of Negros. Whatever the roots of this polyphony may be, it is
quite definitely not a retroeessive adaptation of European
polyphony.
We eannot be equa,lly sure in the ease of Russian polyphony;
the Orient has no prototype for sueh artful deviees as eontrary
motion (with the voice parts moving in opposite direetions).
Here, an influenee from the (later) polyphonie settings of the
Orthodox Chureh 41 is by no means impossible.
Whatever form it assumes, folksinging in parts must be due
to a special gift and trend. The geographie distribution of part
singing seems to eonfirm this eonc1usion.
Finally, we must turn to the most widely quoted and most
eontroversial evidenee in the field that we are diseussing: the
passage in Plato's Laws (7; 812 D-E), which defines the 'heterophony' of Greeee as answering short by wider steps, high by
lower pitehes both in eonsonanee and dissonanee, and also as
adding various adornments. The passage holds a key position
in the weighty question as to whether or not the Greeks knew
polyphony. Still, the meaning of Plato's explanation is not at all
unambiguous. Uneonscious heterophony is not even diseussible
sinee Plato deseribes the elements as an intentional enriehment
of the bare melody; neither does he mention any melodie theme
that the voice parts have in eommon. Quite to the eontrary the
passage does insist on wilful eontrast throughout: on steps of
different widths, on notes of different pitehes, and upon dissonanees and eonsonances, whieh remind much more of modern
contrapuntal devices than of the random coincidences of heterophonie playing.
40 M. Antonowytsch, Die Mehrstimmigkeit in den ukrainischen Volksliedern, in
Kongressbericht der Internationalen Gesellschaft fr Musikwissenschaft, Utrecht I952
(I953), pp. 55 ff. - EFw FE 4505 (P 505) I 3. - E. Lineff, Velikorusskiia piesni, S.
Petersburg, I904. - id., Peasant songs of Great Russia as they are in the folk's harmonization, 11, Moscow, I9II. - Viktor Korda, l.c.
41 About them cf. Gustave Reese, Music in the Middle Ages, New York, I940,
pp. I03 ff.
190
POLYPHONY
As a matter of course, this does not mean that the Greeks

had anythingin the way of late-westem counterpoint or hannony,
with all the. rules of part-writing, resolution of dissonances,
and interdiction of panille1s. But it might be appropriate to
quote at least two evidences: Athenaios in the Sophists at
Dinner 14: 618 (c. 200 A.D.), who wams a piper to keep the
voice parts distinct without probtemata when playing with "this
girl"; and pseudo-Longinus (probably in the first century A.D.),
who asserts that the melody is "usually sweetened" by the
accompaniment of either a fifth or a fourth. Speaking repeatedly
of the pleasant character of consonances - and this in a strict
gradation from absolute to lesser perfection - the Greeks as weIl
as the Romans were to a certain degree prepared for vertical
hearing. But the current concept of heterophony coIlapses
when and where its freedom and naivety are subjected to the
critical control of vertical awareness.
In considering this, we realize that the use of the term heterophony should be more precisely defined than it has been done
in the past. Terminology seems to have the choice of quite a
number of possibilities: (1) heterophony is in every composition
in which 'other notes' are heard at the same time, inc1uding a
simple drone with a melody, but also inc1uding modem polyphony and harmony; (2) aIl forms of 'otherness' except the
lawful polyphony and harmony of the West, but inc1uding
almost all part-performing of the western Middle Ages; (3)
simultaneous variation - a choice that would of necessity assign
the more complicated forms of the Orient to polyphony proper.
We look over the numberless evidences extant from Polynesian
canons to the ubiquitous drones; from the archaic court music of
the Japanese emperor to the 'Andalusian' orchestra of the
Moroccan sultan; from the formal, regular part-writing of the
Later Ages. forward to the contradictory forms of the twentieth
century and back to the medieval types in church and folksong.
Is it not impractical and against our common sense to cut off the
orthodox part-writing of a few centuries in the official music of
Europe from all the part singing and part playing in the overwhelming majority of countries and ages - unless we put emphasis
on the word writing?
This is a highly important, indeed the most important criterion:
POLYPHONY
191
only European polyphony of the second millennium A.D., and

actually only a certain part of it, has been written down instead
of being improvised. If we realize that this polyphony owes its
forms and its very existence to notation and elaboration at the
desk, we see a fourth choice of terminology: heterophony is
every type of part-performance left to tradition and improvisation - contrapunto alla mente as against res facta.
Possibilities (1) and (2) should be eliminated as unsuitable;
(3) is covered by the title 'simultaneous variation,' a term much
less ambiguous and therefore a better choice than 'heterophony.'
Thus possibility (4) appears to be most commendable from the
Viewpoints of music history and of easy, unmistakable definition.
In establishing this definition, we must not allow ourselves
to be lured into pedantry. Some 'scriptless' heterophonies Japanese court music for example - were actually, although not
flawlessly, notated; and, inversely, Bach was able to improvise
fugues. But the beginnings of the Mikado's gagaku - a thousand
or more years ago - were not contrived with a writing brush in
. the hand: J apanese notation, as all the musical scripts of the
Orient, followed as a mere memorial help whenever a preceding
oral tradition was in danger of oblivion; and improvising fugues
is an organist's tour de force, impossible without a thorough
training in written counterpoint. On the other hand we see in
Europe notated music appropriating improvised forms: the
sixth chords of the so-called English discant and the subsequent
fauxbourdon, or the popular fifth paralleis of the sixteenth
century, especially in the Italian villanella, not to forget our
modern swings and their scores.
Notated or not, such forms are heterophonic as long as they
derive from improvisation and preserve its unmistakable spirit. 42
48 Essential parts of these paragraphs on heterophony are published in German

language in Musik in Geschichte und Gegenwart, vol. 6 (1956), col. 327-330.
EIGHT
CROSS- OR POL YRHYTHM
Onee upon a time I was sitting in the garden. The lilacs and
the lilies of the valley united their sweet aromas to a smiling
fragrance of spring. A friend eame, remarked on the beautiful
scent and unconcemedly lit a cigarette. This time, the odors
did not merge to form a eonsonanee; here were the flowers,
there was the poignant tobaeco smoke, foreign to one another
and almost hostile. Yet I took them in, each in its own right;
I focused my attention now on the blooms, now on the cigarette,
without attempting at a fusion and without needing one. It was
a pleasant coexistence, not an integration, of different olfaetory
pereeptions.
After a while we left the garden. On our walk we passed in
front of two ehurches, one, its door open, letting us hear the
organ and the ehoir, the other, sending down the solemn c1ang
of its bells. The bells did not agree with the ehoir and the organ,
either in tempo or in harmony or even in pitch. Nevertheless,
the eoincidenee was deeply moving and beautiful. Half consciously we feit that any adjustment in tempo, pitch, and harmony
would weaken the three-dimensional power of unresolved
discordanee.
Some other day, a parade passed through a near-by street.
The marchers kept time with the music; but the lookers-on
strolled leisurely about, and a few children raced prestissimo to
overtake the silent paraders and get at the band. There was a
steady, organized tempo on the drive and a wholly unorganized,
careless stir on the sidewalks. Yet it was one picture, one gay
ado, one colorful, festive impression.
Just as unproblematic, natural, and attractive as the free
combinations in the garden, chureh, and marching episodes can
the rhythmical, para-rhythmical merger be in primitive and often
193
in oriental music. Indeed, our generic term music is misleading.

For it implies an all-embracing concept of the various elements
that reach our ears and therewith a common rule for each of
them. Such a concept is often absent in early civilizations;
singing might obey one rule, and striking, another law-even
when neither activity exists without the other. The simultaneous
perceptions and acts of singing, playing, and dancing are often
kept apart as being different, unrelated sensations, which do
not call for fusion or mutual adaptation; As I earlier wrote:
"The bliss that the rhythmic organization of c1appers or drums
conveys can be feit without in the least encroaching upon the
coincident song." 1 The drummer might be slower, or else he
might be faster. Frances Densmore, studying the music of Y uman
and Yaqui Indians, found that the drummer was in 49 percent
of all the songs slower or faster than the singer. 2 Such incongruities occur elsewhere, too; the voice and the drum have each a
tempo and a rhythm of their own 3 (which does not exc1ude an
occasional speeding up or a slowing down); or, in other cases,
the drummer is strict, and the singer is not. In North Australia,
too, the clicking tap-sticks are at times kept at a tempo different
from that of the song. 4
A similar independence can still be found on the level of high
civilization. In the c1assical mask drama of Japan, the no play,
a strange instrumental group is seated on the stage (like in
Stravinsky's Histoire du Soldat). It is made up of one transverse
flute and two or three drums - one stick-beaten (often absent)
and two hand-beaten. "As a rule, the drummers strike an even
rhythm though the voice is free. Now and then the flute joins
in and soars above the voice; but its melody is neither cordinated
nor even correlated to the song: the two parts are not supposed
to be heard together, but to coexist, in a magical, not in an
aesthetic, sense." 5
One might advance that, like any recurring motion of the
Curt Sachs, Rhythm and tempo, New York, 1953, p. 43.
a Frances Densmore, Yuman and Yaqui music, in Bulletin 01 the Bureau 01 A mBrican
Ethnology, no. IIO (193Z).
Societe Neuch4teloise de GeograPhie, vol. 54 (1948), p. 4. - Cf. EFw FE 4444 (P 444)
I I, FE 4445 (P 445) II 7.
4 A. P. Elkin, Arnhem Land music, l.c., p. 99.
5 Curt Sachs, The rise 01 m1fSic in the ancient world, New York, 1943, p. 137.
1
194
body, percussion invites a regular time, "hut that singing has

no such urge: ,hythmus in C01'po,is motu est. 6 And as rhythm
is indeed essential in the body's movement, and not in music,
such rhythmic independence can here and there even separate
a dance from its accompanying song. Loango Negros, so we are
told, keep the two activities at different speeds, and eyewitnesses
who saw the traditional t,atta of Hellenie women on Easter
Mnday at Megara make a similar statement. 7 The two examples
are quoted with due reserve.
ACTUAL POLYRHYTHM is of an entirely different nature. In
using this term, I realize its ambiguity. While most competent
authors in the field give it to part music in which the parts
proceed in different rhythm (just as in polyphony they move
in different melody lines), we find it here and there intended
to cover successive changes of rhythm in the same melodie line.
This is a doubtful terminology. When a composer writes a fivebeat melody in equal measures of five beats, his work is ll.lOnorhythmic; when he prefers to notate exact1y the same melody
in alternating two-beat and three-beat measures, it becomes
polyrhythmic. This makes no sense. The reader might feel safer
with the unmistakable word cross rhythm for simultaneous
differences in several parts.
Gross-rhythm, in contrast with successive differences within
one voice part, represents, as happens so often in the visual arts
as weIl, a conscious, purposeful enrichment created by the joint,
not segregated, perception of contradictory rhythmical patterns.
Within primitive, oriental, and ancient-European civilizations,
polyrhythm or cross-rhythm belongs exclusively to instrumental
music, even in the highly sophisticated music of Greek Antiquity.
Voices act collectively, very probably under the binding force
of cOlIlIl}on texts; they sing in unison or, when the sexes and the
ages coperate, in parallels. Instruments, free from textual ties,
preserve a remarkable independence, both against the voices and
among themselves. They cordinate disparity and similarity,
law and freedom; and in the fusion of contrasting and even
disagreeing elements, they make e plu,ibus unum.
G
Marcus Fabius QJ.lintilianus, lnstitutionum oratoriarum Ubri duodecim, 9 : 4.

Curt Sachs, World history 01 the dance, New York, 1937, pp. 176,241.
195
HEMIOLA, or 'one and a half,' is the simplest form of heterorhythm, both as a change from measure to measure and as a
cross-rhythm. WeIl known in western art music too, it derives
from the ambiguous character of slx time units as 2 X 3 and
3 X 2: the same series of notes might yield now a 6/8 now a
3/4 time. This ambigu;ty is of course facilitated and normalized
in styles where the accents are less pronounced than they are
in the modern West. As a cross-rhythm, the hemiola pattern
would be:
3/4: x . x . x . x . x . x . x
6/8: x .. x .. x .. x .. x
If, as before, we accept the dance movements as one of the
voice parts and their music as the other part, Yugoslav folk
dancesgivea rich illustration of such crossing. The steps coincide
with the time units of the melody but often devolve in periods
so different that the accents do not necessarily agree. It is not
yet polyrhythmic when 24 steps of dancing match 2 X 12 beats
of melody; the latter is merely repeated for the second part of
the dance. But cross-rhythm arises when the dance has four
tripIe measures against three quadrupIe measures of the melody;
or, in more interesting cases, 6 X 5 steps of dancing against
5 X 6 beats of melody; or 3 X 4 steps against 2 X 6 beats; or
2 X 12 against 3 X 8. Sometimes the dance and the melody
reach a common ending only after many repetWons of either one;
the maximum observed so far is a dance of five steps against a
melody of sixty-four time units: they meet for the first time
after 320 beats. 8
The paradise of polyrhythmic music is Bantu Africa. "Whatever be the devices used to produce [polyrhythms]," says one of
its noted experts, "in African music there is practically always
a dash of rhythms: this is a cardinal principle. Even a song which
appears to be mono-rhythmic will on investigation turn out to
be constructed of two independent but strictly related rhythmic
patterns, one inherent in the melody and one belonging to the
accompaniment ... but it must always be remembered that the
African normally makes no noticeable physical stress on any
8 Danica S. and Ljubica S. ]ankovic, Pravitno u nepravitnome, in Zvuk, I955,
pp. 65-79
196
note and sings all the notes in a steady outpouring of even tones
in a legato style." 9
Western-trained musicians feel too easily tempted to speak
of syncopation in this rave1 of accents and voice parts. Syncopes,
as we use this term, deviate from the regular beat and therewith
heighten its power, as they are exceptions confirming the rule.
In African cross rhythms, on drums as well as on xylophones,
the allegedly syncopating voice part does not contradict or
enhance the regular, normal pattern. Just as it would do in
Japanese heterophony, it rather adds to the whole a second
main beat of its own, independent from that of the neighboring
part. And with the main beat, the whole rhythmical flow is
shifted in the narrowest stretto - as a rule by half a time unit.
There are then actually two (or more) rhythms with equal
rights.
To think of an analogy in the visual arts is more than a mere
simile: while the Later Ages of Europe, from the Renaissance on,
have a strictly unified space illusion within a painting or arelief,
with one common vanishing point in perspective, this unification, so self-evident to modem man, is absent from many,
indeed, from almost every work of the Orient and Europe's own
Middle Ages. Here, the eye must often roam and refocus from
episode to episode and from object to object.
Individually, African rhythms are very simple and become
confusingly involved only in their concurrence - especially when
ternary and binary groupings dash - in such extraordinary
ensembles as the seventeen xylophones that the younger Junod
describes. 10 And in all this seeming confusion we will not forget
how rightly Von Hornbostel once characterized the essent;al
contrast between the rhythmic conception of the Negros and
that of the West: "We proceed from hearing, the Negros from
motion." 11 Hence, it may happen that, in reversal of our downbeats and upbeats, their stress is a powerful tension, which shows
in a sudden lift of the body and the striking arm, while the
9 A. M. Jones, Alrican rhythm, in Alrica, vol. 24 (I954), pp. 27, 28.
10 Henri-Philippe Junod, The mbila, in Bantu Studies, vol. 3 (I927-I929), pp. 275285.
11 E. M. von Hornbostel, Alrican Negro music, in All'ica, vol. I, I928, p. 26. John Blacking, Some notes on a theory 01 A/rican rhythm advanced by El'ich von
Hornbostel, in All'ican Music, vol. I (I955), no. 2, pp. I2-20.
197
resulting drop and sound eomes only after the 'beat' as a relaxation. This also oeeurs in Indian Ameriea. l2 We ourselves eould
often witness sueh sht when groups of 'Belgian' pavers rammed
the stones in arhythmie sequenee..
I ncidentally: German versifieation calls the poetical (downbeat) aecent Hebung or 'lift'; and in full agreement, although
as the result of a Roman mistake and against the original meaning
in Greeee, the modem aeeent or downbeat is generally ealled an
arsis, whieh means a lift onee more. l3
This being as it is, our ears do not easily register what N egro
motion dictates. Sinee unraveling sueh polyrhythms from
phonographie reeordings is next to impossible. A. M. Jones
devised an electrograph in whieh the fingers of performing
drummers and clapperers were eonnected with peneils marking
individually everybody's eontribution on a rotating band of
paper. l4
One more attempt of individualization has been made in the
phonographie field: on a fine reeording for Ethnic Folkways,
entitled Drums 01 the Yoruba 01 Nigeria, William Baseom
separately reeorded the five parts of a drum set. l5 This is another possibility of transeribing eorreetly. But the objeetion is
that the recording fails to indieate whether the five drummers
enter on the same beat or, which is more probable, at eertain
distanees.
The western hemisphere has inherited an essential part of
this drum polyphony, if only in a limited area. With the SudaI).ese
Yoruba slaves, trios of rhythmieally different Afriean drums
entered Cuba and are to this day the baekbone of indigenous
music in that island, often complieated an~ enriehed by the
eoperation of a small bell (cencerro), a gourd rattle (maraca) ,
and a seraped ealabash (guiro), in six independent parts. l6
Looking from African drums to those of India, we find in
12 Carl Stumpf, Liede1' der BeUakula-Indianer, in Vierteljahrsschrift ff,r Musikwissenschaft, vol. 2 (1886), p. 49, and in Sammelbnde jf,1' Ve1'gleichende Musikwissenschaft, vol. I (1922), p. 92.
13 CUIt Sachs, Rhythm and tempo, New York, 1953, pp. 128 ff.
14 A. M. lones, Aj1'ican music, Livingstone, 1949, pp. 63-77.
15 EFw FE 4441 (P 441) 11 2.
16. EFw FE 4403, 4407,4410, 4435, 4440, 4461, 4500, 4502 (P 403, 407, 410, 435.
440, 461, 500, 502). - Cf. also: Fernando Ortiz, Los instrumentos de Ja mUsica ajrocubana, 5 vols., La Habana, 1952-55.
198
either country the same importance given to single and to

multiple drums and the same fondness of rhythmical patterns
and cross rhythms. We also find the characteristic delight in
improvised variations.
The most striking Afro-Indian concordance is almost conclusive: Uganda's kings possessed a quasi Indian or Burmese
chime of twelve independent drums, tuned in the native scale
of five equal steps per octave and spanning altogether two
octaves and a ample wholetone. 17
But there are differences as weil. The first is that the Hindu
tune their drum-skins carefully in octaves, fifths, or fourths
against one another and strike them in different spots in order
to distinguish dry and meilow sounds. The N egros, possibly
afraid of diverting attention from rhythmic sensation, are often
satisfied with a vaguer turung in high and low and middle pitch
(which however admits exceptions, such as Uganda's set of
twelve or the unmistakable G D of the Yoruba drums in Cuba).18
The second difference (even though the longs cannot well be
sustained on a drum) is India's strict adherence to additive
metrical patterns, such as long - short - short (3+2+2) or
short - long - short (2+3+2). These are tradition al and inviolable, unless the drummer tries to vary the pattern in artful,
acrobatic circumscriptions. Contrariwise, Africa has rhythm by
stress, and the unstressed beats between the accents are metrically indifferent. Cross rhythms emerge in India by combining
one traditional pattern or tla with its own 'augmentation' to
double note values or with some other pattern (which might,
in an almost African way, cause non-coincident 'bars'). But the
first stresses of either one pattern coincide, while the N egros are
particularly fond of shifting the entries in one or several parts.
In India, the drum pattern follows in principle that of the voice;
in Africa, it is completely independent - if it accomparues at
all: the Negros have drum ensembles in their own right.
The result of this brief comparison is: the Indian drum is
almost melodie and much less 'percussive' than that of Africa.
It foilows 'additive' meter, while the African drummer rather
obeys 'divisive' rhythm by stress.
17 J. Kyagambiddwa, African music tram the Souree 01 the Nile, I.c., pp. II4 ff. - A
beautiful recording on English Decca LF II20 (rec. by Hugh T. Tracey) I I.
18 EFw FE 4502 (P 502) III I5 and FE 44IO (P 4IO).
199
These differences wou1d hardly allow to relate the drums of

either region directly, although they leave the door ajar for
tracing a common origin somewhere in the middle between
their habitats. Pages 77 and 78 of my History 01 Musical Instruments refer to a parallel between the extraordinary shape of an
ancient Babylonian foot drum and one of modem East Africa
and to the similarity of religious rituals devoted to the sacred
drums of either civilization. At the end of that passage, mention
is made of archaeological finds in Mesopotamia and ethnological
facts in modem Africa. With all due reserve, the possibility of a
fork relation radiating from an ancient Babylonian (not Sumerian) center may be suggested. 19
19 See, in this connection, also Jaap Kunst, The origin 01 the kemanak, in Bijdl'agen
tot de Taal-, Land- en Volkenkunde, Deel II6 (1960) p. 263-269.
NINE
The high artistic development of African drum and xylophone

music draws our attention to the beginnings of musical professionalism.
The neat separation of amateurs and professionals, a pillar
of modern musicallife, presents a concept not applicable in truly
primitive culture. Here, the professional cannot exist because
he depends on specializing forms that his society has not yet
acquired. And by the same token his antipode, the dilettante
in our sense cannot find a place either. In the village, more or
less every member of a tribe participates in performing or even
composing songs as an essential function of tribai life. He is not
an amateur or self-complacent imitator of professionals, but a
full-fledged and exclusive carrier of all the music in his community.
A specialist, although at best a semi-professional, is the songman of an Australian tribe, who leads the tribai songs and dances,
acts as the soloist in response forms, and often is the licensed
owner of the tribai repertory. The medicine man in shamanistic
societies is, to be sure, a professional - perhaps the first and only
professional in his group - and singing and drumming are
essential elements of his services. He is a singing professional,
but not a professional singer; his magic incantations require
no special musical training.
The professionals that we find in more complex, albeit still
scriptless societies are the by-product of an economic merger.
In the older of the two basic layers, all mankind lives on nature
in a way of simple appropriation: men hunt, and women gather
food, collecting the fruit of trees or smaller animals that the
beach or the woods provide without resistance. In the later layer,
mankind takes care of nature with an intention to improve and
201
increase what it provides; men, as a rule, convert their random

hunting into a weil planned breeding of herds, and women
proceed from haphazard picking to sowing seeds and reaping
the crops. N eedless to say that breeding does not exc1ude the
older hunt, nor agriculture the pristine food coIlecting.
There has been a trend to unite the two higher forms of
economy by an interpenetration of herdsman tribes and of
planters, in which the cattle drivers, alert, predaceous, bellicose,
and experienced in fighting both cattle thieves and beasts of
prey, contributed defense and security, while the working
peasants provided arieher variety of food and sundry handicrafts. This symbiosis led forcibly to the formation of hereditary
c1asses and castes as weil as to a strict division of labor. Only in
this kind of social atmosphere was an actual profession of
musicians possible. But also music itself, its contents and
techniques, indeed its very scope did profit from the new form
of society. Economical strata created very different forms of
religions, of world outlook, of social ideas, indeed of emotions.
In short, mentallty and, hence, the urge and content of selfexpression through art became essentially different from what
it had been in a tribal world, and in this difference it grew
complex and centrifugal.
THE PROFESSIONALS in these complex societies must be carefully c1assified; they work in two fields with little common ground.
The first of the two groups connects directly with the preceding section on cross-rhythm. A growing sophistication of
instrumental music in Asia and Africa acted of necessity as
a dividing factor; persons with a gift for music gave time and
energy to study and performance at the cost of other work, and
men of lesser talent stepped back. Participants in African crossrhythms are more often than not at least semi-professionals,
unless they enter a king's or chieftain's band and acquire the
status of full professionals. Carefully selected from a particular
c1~n 1 for skill and reliability, they are conscientiously trained
by reputed masters or by their own fathers, whom they often
succeed in their positions.
1 In Uganda the Hippopotamos Clan (Ralph Linton, The tree 01 cuUure, New York
1955, pp. 450-45I).
202
African trumpeters are in a similar situation; and, just as

their conlreres in Europe up to the eighteenth century, they
were only in special cases a1lowed to play with other, 'inferior'
musicians,2 - a likeness suggesting common origin in the Middle
East.
The xylophones in Africa, too, were entrusted to professionals,
since their tremendous requirements of technique exclude
amateurish trifling. 3 As the performers toil in two shifts, the
music goes on relentlessly till dawn while men and women keep
dancing around the triangle that the usual three instruments
form in the Iniddle of the village square. 4
are connected with such players by
the mastery of some artful instrument; otherwise they are
completely set apart. Though often applauded for their skill in
accompanying themselves on the fiddle, harp, or lyre, they have
their center of gravity in being the sounding memories of their
peoples and in bringing to life their genealogies, myths, and
histories. We are reminded of the divine aoidos in Homer, of
Demodokos, "the sacred master of celestical song dear to the
Muse, who gave his days to flow with mighty blessmgs, mix'd
with mighty woe, with clouds of darkness quench' d his visual
ray, but gave hirn skill to mise the lofty lay." 5
What we read in the immortal dactyls of ancient Greece
applies to almost all of modern Eurasia and to parts of Africa,
to the laoutars of Rumania and the regs of Hungary, to those
who recite the Finnish kalevala and the guslari of Yugoslavia. 6
They all are painstakingly prepared for this art: in ancient
Ireland, the bards or reciting poets (who, according to Irish
tradition, had come from the Orient in times B.C.) 7 needed
at least twelve years of study before becoIning full-fledged
THE SINGERS OR BARDS
ollamhs. 8
2 (Margaret Trowell and) K. P. Wachsmann, Tribal crafts oj Uganda, London,
I953, p. 356.
3 Evidenced as early as the sixteenth century by Fr. ]oo dos Santos (I586), as
quoted in Percival R. Kirby, The musical instruments oj the native races oj South
Ajrica, ]ohannesburg, I953, p. 47.
4 EFw FE 4402 (P 402) I 8.
5 Odyssey, book 8, transl. Alexander Pope.
6 EFw FE 4506 (P 506) III 46.
7 Michael Conran, The national music oj Ireland, 2nd ed., London, I850, p. I9.
B William H. Grattan Flood, A history oj Irish music, Dublin, I905, p. 2.
203
Thinking of bards evokes the picture of empty, sightless eyes,

"quench'd with clouds of darkness." The singer's blindness has
often been pointed out but never explained, except by associating blindness, introversion, clairvoyance, wisdom, and
age. 9
Sociologically, the bard, as reciter of myth, history, and
genealogy, belongs predominantly in herdsmen cultures: "there
seems to be some intimate connection between a preponderant
cattle economy and epic poetry." 10 Both bards and instrumentalists are respected members of the society, especially where,
at least in principle, they work without pay.ll "The status
of a harp-player [bard] in Ganda [Uganda] is very exalted."
K. P. Wachsmann tells us that most of them are officials "of
the royal household, but chiefs of sufficient wealth and importance may keep a harpist without incurring the King's
displeasure. There are very few of these men and the manual
skill required in Ganda is considered beyond the reach of the
ordinary person." 12 In Central Asia, as well, the singer is
"everywhere received with reverence." 13
It is in this spirit that in ancient Egypt the pharao invited
the court musicians to build their tombs in the royal cemetery 14
- even though I suspect that this benevolence was more than
probably due to the typically pharaonic desire to have his
musicians at hand in the life to come.
Like the instrumentalists, the bards as well are preceded
by a semi-professional stratum. Ernst Emsheimer, discussing
the nomadic cattle-driver civilizations of Central Asia, tells us
of such harp-playing bards, but wams it might not be quite
accurate "to refer to them as 'professional singers,' as it is only
very rarely that they can earn their living exclusively by singing.
They do not form aseparate group, but make an appearance only
9 Cf. Hans Hickmann. Le metier de musicien au temps des Pharaons, in Cahiers
d'Histoire Egyptienne, Le Caire, 2/I954, pp. 299 ff.
10 Ralph Linton, The tree 0/ cuUure, New York, I955, p. 259.
11 Cf., e.g., ]. H. Nketia, The role 01 the drummers in Akan society, in Alrican
Music, vol. I (I954), pp. 34-43.
12 K. P. Wachsmann, Ha'Tp songs l'Tom Uganda, in Journal 01 the International
Folk Music Council, vol. 8 (I956), p. 24.
13 E. Emsheimer, Singing contests in Cent'Tal Asia, in Jou'Tnal 0/ the International
Folk Music Council, vol. 8 (I956), p. 27.
14 H. Hickmann, l.c., p. 257.
204
PROFESSIONAL MUSle AND MUSICAL SYSTEMS
in connection with certain festivals, after which they resume

their usual way of life as pastoral nomads." 15
THE DEGENERATED FORM OF BARDISM is minstrelsy. There is
good reason to suppose that minstrelsy, like many other traits
in African culture, has come to the Negros from the Mediterranean coast. Ethiopian minstrels accompany themselves on
Arabian fiddles; minstrels of the Cameroons illustrate recitals
in a way reminiscent of the Bedouins in Tunesia. Robert Lachmann gives us a detail description how in Tunesia one of them,
playing a programmatic piece on the flute, enacted from his
squatting position a horse shying from a lion or a girl arranging
her hairdo and sash or grinding barley in the circle of her friends. 16
In the Cameroons, "the travelling minstrel occupies a place
of favor ... The old traditions are passing, but the minstrel
remains as popular as ever. When he appears in a community,
crowds gather to hear him. His traditional garb consists of a
bark-fiber breech cloth, a kind of skirt made of monkey tails
or strips of antelope skin, and leg rattles ... On his arms are
fastened raffia cuffs and more monkey tails. His head-dress is
prepared with feathers ... The minstrel is not only a story teller
and a musician, he is a carrier of news from distant places ...
In all of the minstrel's 'news' and stories, singing plays a major
part. ... In the narration of a long tale he may talk for two or
three minutes, then break in with the appropriate song, and
again revert to the story. Sometimes, when the story calls for
dramatic action, the minstrel will give his instrument to an
assistant and, while the music continues, act out the drama with
gestures and dance steps. He is at his best with a group of supporters to act as his chorus. Sometimes he brings his own with
hlm. Other times he gets impromptu assistance from the villagers
whom he is. visiting ... men to sing, women to help with handclapping, stick clapping and the shaking of rattles." 17
But welcome and respect are not necessarily one.
Henri Labouret has given us a vivid picture of professional
performers in the western French Sudan, of their social position
l.c., p. 26.
Robert Lachmann, Die Weise vom Lliwen und der pythische Nomos, in Festschrift
jr ]ohannes Wolf, Berlin, 1929, pp. 97-101.
17 Edwin Cozzens introducing EFw FE 4451 (P 4Sl) (in free re-arrangement).
16
18
205
and of their duties. They are not artists but artisans; and as
such they are not independent in our sense but caught in the
meshes of a rigid system of despised castes. If they hold the rank
of real 'musicians' and fiddlers, they belong to the higher crafts,
together with weavers, leather workers, smiths, and founders of
precious metals, and are allowed to intermarry with them. But
the lower dass of entertainers form the bottom caste of society
and are not permitted to marry outside their dass. These griots,
as the French call them, importune the rich with either glorifications or insults, depending on whether their victims are openhanded or stingy. They often roam from village to village in
gangs of about a dozen under a chief who is at the same time a
seasoned historianand genealogist and knows to their last details
the alliances, hostilities and conflicts that unite or oppose the
families and the villages of the country. With this knowledge,
they sing long, high-flown praises of the leading persons in the
village and often achieve to reconcile, and probably just as often,
to spur iIitertribal strife. l8
From other sources we learn that the Wolof in the French
Sudan, divided into a hierarchy of caste-like dasses very much
like those in medieval Europe, give their musicians, minstrels,
praisers, and jesters the lowest status. They all are stereotyped
as lazy beggars and buffoons who lack pride and modesty.1 9
Indeed, they are regarded in some ways as untouch'ables: they
are not allowed to eat out of the same dishes with members of
other dasses and must not be buried in the village graveyard.
In earlier times it was not even sufficient contempt to deny them
burial in a graveyard: a British source of I745 reports disgustingly that the corpses of African minstrels, instead of being buried,
were placed upright on their feet in hollow tree trunks for slow
putrefaction. 20
Although the latter treatment is unmatched, a11 these descriptions evoke familiar pictures. In ancient Greece as well,
18 Henri Labouret, Paysans d'Afrique occidentale, Paris, x94x, p. X34. - SampIes
on EFw FE 445X (P 45x) I 3, 4, II x and FE 4462 (P 462) I x, 3, 4, 5, II x, 2, 5 David Ames, introducing EFw FE 4462 (P 4621. - Cf. also ]aap Kunst, Ethno
musicology, 3rd ed. The Hague, X959, p. 238 S.v. griots.
19 Ames, l.c.
ao Thomas Astley, A new general collection of voyages and travels, London, X745,
vol. 2, pp. 277-9, quoted from Richard Wallaschek, Primitive music, London, x893.
p. 66, and Anfnge der Tonkunst, Leipzig, x903, p. 69.
206
Homer's godly rhapsode yielded in post-Homerie times to a

minstrel who more and more "came to be unfavourably regarded." 21 And alongside with the minstrel, instrumental
players took the stage. At the Pythian games of 586 B.C., the
piper Sakadas played a program pieee on the famous fight of
Apollo and the dragon - in its eontent and strneture not unlike
the tune of the lion mentioned on page ~o4.
Rhapsodes as well as pipers and lyrists degenerated to boastful,
inflated virtuosos, applauded by ten thousands with thunderous
eheers and pampered by hysteriealladies. 22
We ean draw a parallelless remote than ancient Greeee between the misery of the Afri ean minstrel and the Spielmann
and juggler in Europe, a misery that with exeessive eontempt,
segregation, and outlawry lasted one thousand years. 23
To be eomplete beyond the interests of the present book, we
should include the professional wallers..., mostly women - whom
we find from the Middle East along the Mediterranean via
Corsica and up to Ireland : "Did ever a wall make man's marrow
quiver and fill his nostrils with the breath of the grave, like the
ululu of the North or the wirrasthrue of Munster?" 24 A beautiful
sampie from Rumania, a bocet, has been reeorded by Bela Bart6k
and is aeeessible in a Folkways reeord: 25
Ex. 78*
J"' 60 O,.fy,nallttJn"hlghUO
@JI9Inllt.lJll~jJIJJJ)JJJID I
N or should we forget that upon entering the house of J aims,
Jesus "saw the minstrels" (St. Matthew 9 : 23) and "them that
wailed greatly" (St. Mark 5 : 38).
though modest in size and seope, is
the inevitable tool and eompanion of professional music. Learning
A MUSICAl.. VOCABULARY,
EFw FE 4419 (P4I9) II 9.

The Oxtord classical dictionary, Oxford, 1949-53, p. 765.
22 Curt Sachs, The rise 01 music in the ancient world, New York, 1943, p. 271.
23 Curt Sachs, Our musical heritage, 2nd ed., New York, 1955, pp. 74 ff.
24 Thomas Davis, after Grattan Flood, l.c., p. I.
25 The Greek distinguished between the epikedeion, over the dead body and the
threnos in memory of the dead. Cf. E. Reiner, Die rituelle Totenklage der Griechen,
1938.
21
207
and teaching the craft, much as it might be imitation of the

teachers and, in return, the master's correction, is unthinkable
without at least the rudiments of a terminology as a means of
communication. And terminology in itself includes the opposite
concepts of distinctiveness and affinity, of seeing differences and
similarities; and therewith classification and system are born.
Examples are frequent in Africa. The Baganda in the East,
despite their scriptless culture, used a rather involved terminology, and the South African peoples have names for every note.
These words are actually names of notes, since they appear in
the vocabulary of the xylophone as weIl as in that of the flute; 26
but there is little doubt that originally, before becoming abstract,
they had been the distinctive names of sound-producing agents
like the individual slabs of xylophones. This is particularly
evident in Madagascar. 27 But then, the ancient Greeks adopted
six of their seven note-names from the strings of the lyre, as the
high, the next-to-high, the middle, the next-to-middle, the low,
the next-to-Iow; and aseventh note they called lichanos or
index finger. Indeed, they covered with the same word chorde
both astring and a note .
. Besides the .. names of .notes, some African peoples have
distinct scale theories. The instrumental scale of the Gwere in
Uganda has four, in principle equal steps in the octave, and that
of the Ganda had five intentionally equal steps in the octave,28
in other words: an Indonesian salendro gender.
A second surprise: the scale can be 'high' and 'low,' 29 an
indication whieh, in connection with the arrangement of the
note names in either one, leaves little doubt that the Lower
Scale was 'authentie' and the Higher Scale 'plagal,' one starting,
say, from C, and the other one from G. The coupling of authentie
and plagal modes occurs from Indonesia in the east to the
Mediterranean in the west, including ancient Greece, the Gregorian chant, and Byzantine cantillation. If such dualism bewilders in the case of isotonie scales, where it is seemingly a case
of mere transposition or change of gravitation and not of a
Percival R. Kirby, l.c., passim.
Curt Sachs, Les instruments de musique de Madagascar, Paris, 1938, p. 63.
28 K. P. Wachsmann, l.c., pp. Z3, Z4.
29 ]oseph Kyagambiddwa, Alrican music Irom the sources 01 the Nile, New York,
1955, pp. 19 ff.
26
27
208
different function as in the case of scales with steps of different

sizes, the Uganda pattern is obviously paralleled in certain
respects in the likewise Javanese and Balinese saUnd1'o scales
and their palet. 30
As to the terminology of rhythm: in many parts of Africa,
rhythmic formulas for drums have individual names, somewhat
like the various Indian tlas. Similarly, the language of the
Nootka in the Northwest (Vancouver Island) has at least twenty
verbs expressing the various ways of drumming, such as 'beatbeat-rest,' or 'beat-beat-beat-rest,' or 'with a steady, even beat';
also 'sounding split' for part singing. 31
Polynesian singers use different terms for each of the two or
three voice parts in their polyphony,32 And special names for
the individual instruments of an ensemble are common in
Africa. 33
As a by-way, Bantu Africa has devised pitch syllables to
replace t1;le particular text of a piece - a terminology elose to
both the vocal solmisation of the West and to the distinctive
syllables that Arabian drummers give the timbres of their beats:
dum, muffled, dim, less muffled, tik, less elear, tak, elear,34
The LokeIe, near StanleyviJ1e in the Belgian Congo, when
shouting messages across a larger body of water in substitution
of a slitdrum, replace the various syllables of the texts by a ke
or le for the lower pitch and a ki or li for the higher pitch: the
words tokolokolo twatoala, or 'little sticks of firewood,' are shouted
as Kekelekeke kikelike. The Duala in the Cameroons substitute
to for ke,35
Similarly, the drummers of India have a special syllable for
every pitch and timbre of their instrument and learn their craft
by memorizing the syllables of the various patterns and to recite
30 Cf. Mantle HOOd, The nuclear theme as a determinant 0/ pa!et in Javanese music,
Groningen, 1954.
S1 Helen H. Roberts and Morris Swadesh, Songs 0/ the Nootka Indians 0/ western
Vancouver Island, in Transactions 01 the American Philosophical Society, N.S. vol. 45,
part I (1955), p. 326.
3a E. G. Burrows, Songs 0/ Uvea and Futuna, in Bernice P. Bishop Museum Bulletin
183, Honolulu, 1945, p. 81.
S3 J. F. Carrington, Individual names given to talking-gongs in the Yalemba area
01 Belgian Congo, in African Music, vol. I no. 3 (1956), pp. 10-17.
S4 Cf. for a Central Javanese parallel: Jaap Kunst, Music in Java, The Hague,
1949, vol. I, pp. 204 ff.
3S John F. Carrington, A comparative study 01 some Central Ajrican gong-languages,
Ph D. dissertation, University of London, Brussels, 1949, p. 40.
209
them in a neck-breaking tempo before attacking their drums. 36

It seems to be a bewildering practice when, evidently to
facilitate learning, the apprentice must master two crafts
instead of one (for memorizing aseries of syllables and reproducing them at highest speed is indeed a craft in its own right).
But such methods occur rather often, in the West as much as in
the preliterate world. In Uganda, every Lango has a personal
whistle motif ... which may be memorized by a few words, a
catch or a phrase of a private song. 37 Indeed, did not centuries
of European music impose the Guidonian hand on tyros? A
similar hand had been used in India for the same purpose of
locating the pitches of melodies, and in China, for marking the
inflections of speech.
Did they not add to this tactile bridge the vocal bridge of
solmisation up to the modem 'tonica do' method in order to
read and remember melodies ? And do we not ourse1ves substitute
adequate texts for military bugle calls and symphonie themes
to assist the memory of soldiers and appreciation adepts ? Who,
-in the U.S.A., does not know the cello theme from the B minor
symphony under its verbal counterpart "This is the symphony
that Schubert wrote and never finished" ?
A text is more readily leamed when sung to a melody, and,
inversely, a melody can be better fixed in our memories when
it follows a text. This daily experience proves the advantage of
what I might be allowed to call the concomittant method of
learning.
88 I am indebted to Rosette Renshaw of Montreal for providing me with an

original tape on which the current patterns of Indian drum rhythms are spoken as
weIl as struck.
87 Driberg, The Lango, London, p. 124.
TEN
'PROGRESS' ?
The picture this book has endeavored to draw is strange

enough. All over the world, from the Eskimo to the Fuegians,
from the Lapps to the Bushmen, people sing and shout and bleat
with voices wild or monotonous; they scream and mumble,
nasalize and yodel; they squeak and howl; they rattle, clapper,
and drum. Their tonal range is limited, their intervals are foreign,
their forms short-winded, their inventive capacaties, it seems,
rather deficient, their traditional shackles a11 powerful. Is it
permissible to call these noises music, if the word denotes the
sacred art of Bach and of Mozart?
And if it is music - how many steps were needed to lead from
the humble, anonymous inventor of palaeolithic songs through
untold thousands of years to the divinized genius of the Later
Ages with his boundless imagination and master technique,
how many steps to climb from an ever repeated, unassuming
scrap of melody to modem music dramas and symphonies ?
Beholding these steps, these numberless changes in style,
ideas, craftmanship, and social connotations, the reader must
have had the dangerous slogan 'progress' in mind.
But was it actually progress that marked the long peregrination
through ages and cultures?
Not long ago, the answer would have been a wholehearted
Yes.
"The dominant theory of ethnology at the time Boas entered
anthropology was that culture- or society, as it was variously
phrased - had evolved, and was evolving from simple to complex
forms, from lower to higher modes of life. It was, like a11 theories,
a product of its times. It is commonly assumed to have been an
offshoot of the Darwinian hypothesis ... " 1
1 Melville
J. Herskovits, Fl'anz Boas, New York,
1953. p. SI.
PROGRESS'?
211
We no longer believe in a neat evolution from low to high,

a constant development from unassuming simplicity towards an
ever growing complication, from infancy through boyhood and
maturity to a doting old age - and to what next?
Those were the days when ancient Egyptian art was disdainfuUy judged as a not yet matured precursor of Greek and Roman
c1assicism, and Romanesque architecture as a somewhat lowly
preparatory step towards the dizzying Gothic cathedrals, which in
due time led thewaytothenobleperfection of Renaissance building.
There occurred a few mishaps, however: evolution from poor
to better had a number of disturbing flaws. In the official,
c1assicistic mentality of the nineteenth century, the arts of
Greece and the High Renaissance were 'perfect.' The Romans
ranked not quite as high; the Middle Ages acquired in Italy the
disreputable name of Gothic or barbarian; and the Baroque was
repellent to c1assicistically refined tastes. N either did Frank
Lloyd Wright evolve from Phidias, or Picasso from Zeuxis. There
were connections and disconnections, attractions and repulsions;
in short, there were fluctuations but no evolution to an ever
greater perfection. The evolutionists found themselves on an
artificial flight of stairs, on which they climbed up from step to
step while the landing of highest perfection should be at the upper
end but actually was, according to their own aesthetics, two
thousand five hundred years behind and below the step on which
they were climbing.
When we exc1ude the science concept of evolution from the
humanities, we do by no means believe in astandstill in the arts.
On the contrary, we are aware of unending change. I quote
Malinowski: "In the widest sense of the term, culture change is a
permanent factor of human civilization; it goes on everywhere
and in all times. It may be induced by factors and forces spontaneously arising within the community, or it may take place
through the conta.ct of different cultures." Indeed, all elements
of culture, art inc1uded, must change, under the impact of tribal
interpenetration, of talent, or of the kind of spontaneous variation
that biologists call a mutation.
Before examining such alterations, we must realize that they
appear in three fundamental forms: culture graft, progress,
and simple change.
212
'PROGRESS' ?
INORGANIC GRAFT is well documented in the musical

culture
;
of the utterly primitive Orang Kubu in Sumatra. Their instru.,.
ments - carefully manufactured drums and whistle flutes with
fingerholes, both of unmistakably Malay provenance - are inorganically pasted upon a merely vocal, pre-instrumental culture.
They are like an attic without stories built on a basement. And
so is the musical contribution of these instruments: songs of a
ripe pentatonic structure and complicated rhythms independent
from the tunes they accompany. After removing the Malay
veneer, we find, as expected, characteristics of oldest age: a
horizontal alternation of two notes a second apart or else a
passionate strain tumbling down from a fortissimo peak, at
times united in duets of either sex, with the women's strain
cascading down and the men's cantillation steadily shuttling
between two notes. 2
THE WORD PROGRESS, an too often indiscriminately used or
misused and oftener misunderstood, is semantically so rich that
Webster's New International Dictionary (2nd edition, I950) gives
it no less than seven definitions. The first three and the last three
do not apply to our case. The fourth one does; and it is excellent:
progress means "advance to an objective, a going or getting
ahead."
The salient point is: progress implies advance to an objective
or final aim.
Franz Boas - and many others with hirn - recognized accordingly "progress in adefinite direction in the development of
invention and knowledge. If we should value a society entirely
on the basis of its technical and scientific achievements it would
be easy to establish a line of progress which, although not uniform, leads from simplicity to complexity. Other aspects of
cultural life are not with equal ease brought into a progressive
sequence." 3
Of "technical and scientific achievements," dosest to our
lives and concerns is medicine with its incontestable possibility
and record of progress. Its definite objective being the relentless
2 Erich M. von H,?rnbostel, Ueber die Musik der Kubu, in B. Hagen, Die OrangKubu auf Sumatra, Frankfurt alM., 1908, pp. 245-257.
3 Franz Boas, Anthropology and modem life, quoted from Melville J. Herskovits.
Franz Boas, New York. 1953. p. 99.
PROGRESS'?
213
fight against sickness and death, every new insight, technique,

discovery in diagnosis and treatment is a step towards this goal.
To switch to the contrary: hunt and war have the definite
objective of killing, and, seen from this viewpoint, atomic bombs
and radar-fired guns are doubtless progressive when compared
with David's slingshot and the crossbows of Agincourt.
In a book on ethnomusicology, we cannot a\Toid adding the
example of our principal tool: the phonograph. Its objectives
are a faithful reproduction of sound, undisturbed by distortions
and improper noises; avoidance of interruption in the midst
of a piece in both recording and backplaying; mechanical
reliability; and as little peril of damage as possible. Unbreakable,
long-playing micro-groove disks and magnetic tape-recording
represent a remarkable progress when compared with Edison's
fragile cylinders and his annoying crank action.
But already in a field like language, not technical and not
scientific and yet belonging to usefulness, matters are less
obvious. Its objective is indubitable: speech is the exhaustive
.communication of facts, emotions, and thoughts within a certain
group of men. Much actual progress had to be made before the
dawn of history, before vowels, consonants, and other sounds
formed syllabies, words, and sentences. What an advance from
the inarticulated grunt in the world of anthropoid primates to
the highly developed and differentiated languages even of our
most archaic, primitive contemporaries! But once the objective
of perfect communication and understanding is reached, the only
further objective left is a continuous adaptation to the ever
changing needs of those who speak it. In this adjustment there
is no trace of the once so cherished trend from simplicity to
complexity. Finnish has no less than fifteen cases in its declensions, and archaic languages of the Indo-European family,
like Sanskrit, Lithuanian, or Russian, have still eight, with
vocative, ablative, locative, and instrumental; English, one of
the most recent languages, needs more or less but one.
It is no different with vocables. "Eskimo has twenty or more
precise words for conditions of snow; and the Tokelau islanders
in northwestern Polynesia have nine names for distinct stages
of ripeness in a coconut, their main food." 4 In our languages,
4
William Howells, Back 0/ history, Garden City, ;[954, p. 68.
214
'PROGRESS' ?
family relationship terms, such as cousin or brother-in law, are

rather poor and vague when compared with those of primitive
societies, where family ties and marriage taboos play a complicated legal and social role unknown in the West: the Baiga in
North India have no less than ninety-eight terms of this kind.5
Such constant adaptation to specific trends and conditions of
culture has little bearing on perfection.' The Authorized King
J ames version is neither more nor less perfect than Shakespeare
or Shelly. The British language has changed and will continue
doing so with English-speaking men, their ways of life, and their
scope of thought.
In art, the concept of an objective is even more limited. No
absolute goal hovers in some immeasurably distant future and
seems to draw nearer with every step we take. If at all we may
speak of a goal, it can only mean a sublimation, the purest
fulfilment of an urge to express ourselves - an urge that changes
its character almost perpetually and differs in every area of
culture, not only between the West and the East, but even between Milan and Venice, between the nord of France and its
midi, and between the Germanies to the north and to the south
ofthe Main.
As a consequence, there cannot be a steady, straight evolution
from childish beginnings to an ever more perfect art, as evolutionists once dreamed. There is rather a bewildering sequence
of sudden changes by leaps and bounds, indeed, a constant
reversal to older, new, and foreign ideals. Everyone of these
changes leads to perfection in its own right - perfections that we
cannot compare because they are based on incomparable changes.
'Functional' architecture in our time is so infinitely simpler
than the often overdecorated, redundant palaces of the Baroque
or, half a millennium earlier, the Gothic cathedrals whose walls
dissolve in a dizzying polyphony of ogives, columns, buttresses,
roses, statues, and tracery. But the Renaissance, which bridged
the gap between the Gothic age and the Baroque, was in its sober
simplicity almost as bald as our modem architecture. Or, to
recall something different, yet related: compare the unassuming
hairdo of ladies of our day with the incredible complication of
the coiffures in which' Italian society delighted at the end of the
8
Verrier Elwin, The Baiga, London, 1939, pp. 527 ff.
'PROGRESS' ?
215
fifteenth century; look at the sudden about-face around 1500

that led to the sleek and sober parting of Mona Lisa's hair; and
hold this against the sophisticated hairdress of the folIowing
generation. What a curious way of progress I
Music history offers striking parallels with the quite unprogressive alternation of complex and simplex. How sudden was
the turn from the almost unbelievable crossrhythms of the
French generation around 1400 6 to the simplicity of the subsequent masters of Burgundy in the age of Binchois and Dufay;
or the sober cl~ty of Peri's and Monteverdi's monodie recitativo
style about .1600 after the contrapuntal excesses of Italo-Flemish
polyphony; or the triekle of the style galant after the grandiose
flow of Bach's and Handel's fugues.
There is no steady evolution from simplicity to complication;
nor is there areverse development. The sequence of styles seems
to be confused, capricious, arbitrary - until we begin to see the
iron law behind the happenings. The interested reader may find
in my Commonwealth of Art (New York, 1946) a tentative
answer. First: subsequent generations set against one another
in their tastes; and second: in supragenerational cycles (like
the ones that we call HelIenistic, Romanesque, Gothic, Romantic)
the beginnings are of necessity simple and strict, and the endings
complex, profuse, naturalistic - only to yield in a rapid reversal
to new simplicity and strictness at the entry of another cycle.
All this is inherent; it cannot be called progress.
The word progress has indeed little meaning in this eternal
zigzag. There was change, not progress, from Raphael to Velasquez and Rembrandt, from Bach to Mozart or Beethoven, from
the Gothic masters to Lorenzo Bernini and Christopher Wren.
And yet they all are woven in the untearable fabric of history,
as each of the masters reflects and expresses the endless renewal
to which we have given this name.
A rather recent reversal assailing the concept of 'history'
altogether and accusing it of misrepresentation 7 is already
obsolete: "historical perspective is again coming into the fore6 Cf. Willi Apel, French secular music 01 the late lourleenth century, Cambridge
(Mass.), I950.
7 Cf. e.g. Fritz Bose, Musikalische VlJlkerkunde, Freiburg, I953 [or do I misread
him?].
216
'PROGRESS' ?
ground of comparative musicology." 8 Unless one confuses

history and evolution, history, the Greater Memory of mankind
and our only account of lapsing time and of change, is an etemal
function of the mind, whether it appears in nebulous sagas of
yore or in the modem search for provable facts and convincing
interpretation. Much as we know that there has not been a neat
evolution of the arts from infancy to maturity and old age; that
the Gothic cathedrals did not sire the palazzi and duomi of the
Renaissance; and that Palestrina did not lead the way to Wagner,
Schoenberg, and Stravinsky - still the history of the arts is an
eloquent document of mankind's changing urge and behavior,
in other words, of our ever changing taste.
Let us not reject the good with the bad and not mistake
historical facts and connections for progressive evolution.
Progress, without doubt, is never completely absent from art.
Although there cannot be an all-time goal, every period, in its
characteristic, novel mentality, has an urge of specific selfexpression, setting for art a temporary goal of its own. Such
new ideal imposes on artists an arduous fight for perfection or
progress in one direction. A new technique towards a new expression must be leamed in steady, laborious efforts, by generations as well as by individuals, who in toilsome years improve
their senses and hands before they master the craft. The artists
of the early Renaissance, still steeped in the irrealistic, antirealistic ideals of the Middle Ages but aImost suddenly confronted
with the new ideal of three-dimensional realism and unified
space as adefinite objective, had to make progress in that
special field before they were able to reproduce with convincing
power what they saw with their eyes. The gigantic strife for
mastering the laws of perspective in the fifteenth century was
'progress.' In a similar way, the evolution from Peri's often so
dry and wooden recitatives to the subtle flow and expressive
intenseness of Monteverdi's stile rappresentativo was progress too.
But the Renaissance itself, as an attitude and a style, was not
an advance or a progress against the preceding Gothic mentality;
nor was the austere monodie style of the 1590es a progress from
the luscious, mellow polyphony of Palestrina, Lassus, or Byrd.
8 Bruno Nettl, Change in tolk ond primitive music, in Journal ot the American
Musicological Society. vol. 8 (1955). p. 109.
'PROGRESS' ?
217
Whatever you prefer according to your personal taste, these

styles are so different that any unbiased comparison of values
is impossible. Progress exists at best within a limited span; as to
the total of art, there is no progress, no regress, but simply
otherness.
My readers might accept these judgments as sound and to the
point and yet be reluctant to delete the handy word progress
from comparing an Eskimo melody and a symphony of Beethoven
or whatever to them represents the all-time peak of musical
creation.
There is no doubt: songs of the primitive are in many respects
more limited than the 'great' music of the modern West to a
degree of excluding comparison. They have nowhere the breath
and breadth of our symphonies ; they lack in the third dimension
that the vertical expanse of harmony adds to the horizontal
stretch of melody; they are monotonous and almost drab against
the glittering orchestration of the romantic age; and they are
narrow in scope when we think of the wealth of emotions,
. thoughts, and images that the West conveys in its manifold
forms, in the intimacy of quartets and trios and the overwhelming mass effects of orchestral music, in the playful charm
of dances and the profundity of adagios, in the pleasantness of
catchy melodies and the 'learned' sophistication of tripie and
quadrupie fugues.
Our music has become wider, no doubt. It has become wider,
as man, its creator, has widened the scope of bis life. His God
rules a boundless world, not just a few huts around an almost
timeless village square; the earth is open to bis railroads, planes,
and liners; and hundreds of trades and vocations shape bis mind
in hundreds of molds. And with the world we rule, our problems
have grown. We are sceptical; we are in doubt about ourselves
and our progress. Somehow we are lost and 10m in the universe
whose dimensions we can no longer express in miles but at best
in sideral years; lost in a wealth of knowledge that yields more
questions than answers; lost in our individualism, wbich is strong
enough to uproot us and estninge ourselves from our fellow men
and yet too weak to prevent incredible, bloody mass psychoses.
The age of Rousseau longed for the bliss of living outside the
pernicious web of western civilization. The paradise that Rous-
218
PROGRESS'?
seau saw in the world of elose-to-nature man was certainly a

dream with little reality. Still, his anguish wa'S justified and would
have been deeper had he anticipated the next two hundred years
of western destiny.
We have lost our paradise; and, with it, we have lost music as
an unseverable part of life itself. Rarely do we real;ze our loss
and its magnitude. We are too hopelessly blinded and deafened
by our assumed progress towards enormous symphony orchestras,
giant concert halls, ticket offices, managers, and publicity, and
we take for granted that there must be dilettantes and professionals, insatiable fans as well as indifferent and outright unmusical persons, music lovers and music haters - and we do not
see that this is not progress but simply change in aims and
values.
1T 1S THE OTHER MAN who knows and understands: the Oriental
and even the Primitive. And once in a while he speaks out and
opens the eyes of those who want to see.
When one day an arctic traveIer played a recorded song by
one of the most famous European composers (siel) to an Eskimo
singer, the man smiled somewhat haughtily and stated: "Many
many notes, but no better music." 9
Yury Arbatsky told me in 1953 a related story: in Belgrade
he once had taken an excellent Albanian folk musician to Beethoven's Ninth Symphony. Asked how he had liked it, the man
hesitated and at last, after a couple ofbrandies, gave the astonishing answer: "Fine - but very, very plain (lepo ali preprosto)."
The Albanian was neither arrogant nor incompetent. He just
had a different standard. The unified, oversimplified, divisive
rhythm could not possibly satisfy his eastern ears - exactly as
"to an illiterate African the regular crotchets and quavers in a
piece of music would appear rather dull." 10 The man must have
feit frustrated by the total absence of all the enlivening, sensitive
spices of 'additive' rhythm, where measures of, say, nine beats
2
2
3 units, or
would appear as picturesque sums of 2
groups of ten as sums of 7
I
2 - not even to mention the
delicate, complex, ever-varying, and oddly stirrlng accompani-
+ +
+ + +
Christian Leden, Ueber die Musik der Smith Sund Eskimos, Copenhagen, 1952.
Phillip Gbeho, Music 01 the Gold Coast, in Alrican Music, vol. I, no. I (1954),
p.63
9
10
'PROGRESS'?
219
ment of hand-beaten drums. Instead of such refinement, he was

presented with western 'divisive' rhythm, where quite mechanically a stress preceded every two or three unstressed units of
equal duration, as one-two-three, one-two-three - an inlpoverishment of European music owed to the growing impact Qf chordal
harmony.
The delicate art of connecting meaningful chords in a harmonie
sequence of dissonant tension and relaxation, vital and expressive
even to western ears for hardly four hundred years, left the
oriental man utterly unmoved. The constant change of orchestral
clor and intensity up to nerve-wrecking fortissimi must have
been little significant and starkly confusing if not repulsive to a
man whose native music was intended for a small and intimate
audience only and not for often gigantic concert halls.
The little episode illustrates in a striking way the urgent
problem at the bottom of this book. Music is not the universal
language. Even the Westerner has to learn his own musical
idiom as if it were a foreign tongue. Otherwise we would not need
Music Appreciation in high schools and colleges, nor. would so
alarming a number of Introductions to Mtisic and Listening
Guides flood our book market. White man is convinced he stands
on the peak of the world - including our music, of which we are
so rightfully proud. He should learn to realize that, once more
including music, we can be as primitive in other peoples' eyes
as they appear in ours. "There are many European traits that
seem uncouth, impolite, incomprehensible or downright 'primi.tive' to a sophisticated 'native'." 11
THE BELGRADE STORY is one of many. When I discussed its
implications with Eric Wemer, he volunteered a similar experience: a highly educated Chinese lady whom he had introduced
to Mozart's Requiem politely acknowledged its beauty and
craftsmanship but thought little of its ...... "superficiality."
The verdict superficial passed on Mozart is somehow disarming. Yet, it might be wise to take the lady's judgment
seriously and, instead of delivering riposte from our own position, conjure up the Chinese magi of the past.
In the words of Confucius (55I-478 B.C.), "a vulgar-minded
11
J.
Manchip White, Antkropology, New York, 1955, p. II9.
220
'PROGRESS' ?
man's perfonnance is loud and fast, and again fading and dim,
a picture of violent death-agony. His heart is not hannonically
balanced; mildness and graceful movements are foreign to him
... ," and again: "The noble-minded man's music is mild and
delicate, keeps a unifonn mood, enlivens and moves. Such a man
does not harbor pain or mourn in his heart; violent and daring
movements are foreign to him." L Pu-we, the poet, "was able
to speak of music only with a man who has grasped the meaning
of the world." Confucius himself knew how to play such a music.
As he once struck "the ch'ing, a man who passed his house
exc1aimed: 'this heart is full that so beats the sounding stone: "
And there is the story of Cheng, who after three years of
endeavor laid his zither down, sighed, and said: "It is not that I
cannot bring the melody about. What I have in my mind does not
concern strings; what I aim at is not tones. Not until I have
reached it in my heart can I express it on the instrument." 12
This is truly Chinese; but not Chinese alone. Far away in
South Africa, a Baronga-boy once played his pristine, onestringed musical bow with the slender stick while the missionary
was listening. Sudderily he began to suck the wooden bow with a
pensive air. "Why are you doing this?" inquired the missionary.
And the boy: "To allow that which I say in my heart to glide
into the instrument." 13
Neither are, what we hear from China, just legends of olden:
the present time has still preserved a certain awareness of the
sacred music of the ancient. Frau Marie du Bois Reymond, in a
musical diary kept during her stay in China from I909 to I9II
and now my treasured possession, writes about the eh'in or
koto-like zither, which to this day is never played without
reverent solemnity. Before sitting down, she relates, the perfonner
washed his hands and burned a few incense candles, because, to
the educated, Chinese, playing the eh'in is not only art in our sense
but rather a mystic, sacred act through which he might summon
a supernatural being. "There are ideas and emotions in his
perfonnance that are hidden from us Westerners," adds Frau
Du Bois.
This modest sentence can and must be expanded: the ideas
18
18
Curt Sachs, The l'ise 01 music in the ancient W01'ld, New York, 1943, p. 106.
Henri-A. Junod, Les chants et les contes des Ba-Ronga, Lausanne, 1897, p. 22.
'PROGRESS' ?
221
and emotions, the essence, the soul of all non-western music is

indeed hidden from Westerners - just as the soul of our own so
magnificently written scores is inaccessible to Easterners. And
again it must be underlined that music is not the universal
language.
In realizing this fact, we have found the answer to our initial
question. The advance from the oldest one-step sing-song to our
giant scores is a tremendous gain in scope. Is it also a gain in
depth and significance? Depth is a matter of sodal and personal
empathy and beyond the reach of musical sdence; significance
depends on the various cultures and their emotional needs.1 4
This statement should exclude any misunderstanding. Neither
do I praise the past, nor do I chide the present. I do not exalt
the Primitive or despise the Westerner. I do not want to exchange
what we possess for what we have lost; I do not trade the B minor
Mass for an Eskimo melody; I am a denizen of this world and of
this time as much as my most critical reader. I write without
nostalgia, without mourning for the paradise lost that Rousseau's
age beheld in the pristine purity ascribed to dawning culture.
The phenomena I mention are facts, not dreams.
One of these facts is the infinitely wider scope of western art
music; its wealth of artistic means from the inimitable single
Stradivari to the giant orchestras and their saturated, everchanging colors; the concept of a space-suggesting third dimension achieved through counterpoint and chordal harmony;
the broader, breathtaking forms extending to symphonies of
more than an hour duration and a dramatic development; and
an area of expression that comprises all human emotions from
a cool and almost unemotional courtly elegance to the unfathomable depth of humaneness and religious experience.
A second, sobering fact is the limited sodal significance of
this venerable art. High civilization, in all its facets, drives of
necessity to diversification, and diversification leads to estrangement - estrangement from person to person, from group to group,
and hence from audience to audience. Statistically speaking,
oUf art music, with all its concerts, recordings, and radios,
reaches but a smalI, almost insignificant layer of mankind, and
14
The subject has received extensive attention in Leonard B. Meyer, Emotion
ana meaning in music, Chicago, 1956.
222
PROGRESS'?
even within this limited stratum its impact is very uneven.

Side by side with art music, and numerically to an essentially
higher degree, we are daily exposed to aUshades of popu1ar and
non-descript sheet music for shaJlow entertainment; and the
total suppression of these characteristic products is probably a
pious fraud. An annex in our accounts of modem music also
should indicate the enormous percentage of persons unmusical
or indifferent who have no part in the delight and edification that
'high-brow' music gives its devotees. Our music is, alas, a marginal
phenomenon.
To this sketch of music in western sodety we have opposed
the picture of primitive music - a picture certainly poorer when
drawn from the viewpoint of art, but always meaningful,
dignified, and hence never vulgar. Primitive music is not compe11ed to fight the indifferent and the unmusical with 'appreciation' or to appease critics, board members, and managers
with tarne and carefu11y balanced programs. Everyone in a tribe
is part of th;s music; everyone sings, and many enrich the inherited stock by creations of their own. No rite in religious
services, no secu1ar act can be performed without the appropriate
songs; and as a vital part of sodal existence music is firmly
integrated in primitive life and an indispensable part of it.
The awareness of this deep-rooted contrast will not change
our destined ways. We cannot escape from the culture that we
ourselves have made. But seeing and weighing the differences
between the two musical worlds might help us in realizing that
our gain is our loss, that our growth is our wane. It might help
to understand that we have not progressed, but simply changed.
And, when seen from a cu1tural viewpoint, we have not always
changed to the better.
A NOTE ON BIBLIOGRAPHY
Although many readers are less interested in an author's
thought and text than in the titles that he might compile in a
bibliographical appendix, I feIt I should not unduly increase the
size, the weight, and the price of this volume with the dead
ballast of lists.
A pertinent ethnomusicological bibliography can be found in
Jaap Kunst's comprehensive Ethnomusicology, 3rd edition (The
Hague, I959, pp. 79-2I5, which, together with a Supplement
(The Hague, I960) contains over five thousand items. I also
might refer the reader to Charles Haywood, A bibliography
of North American folklore and folksong (New York, I95I), and
to the Bibliography of Asiatic musics, published in N otes,
vols. 5-8 (I947-5I) under the leadership of William Lichtenwanger and Richard A. Waterman.
The non-musicalliterature I had to absorb - anthropology,
travel diaries, sociology, cuIture history - can hardly be caught
in a catalogue that pretends to be complete and meaningful.
Ample footnote quotations will provide the reader with all the
information he needs.
All the same, it seems to be indicated to refer to a few comprehensive surveys of the whole field:
Fritz Bose, Musikalische Vlkerkunde, Freiburg i.Br., I953;
Jaap Kunst, Ethnomusicology, The Hague, 2nd ed. I955, 3rd
ed. 1959; Supplement, The Hague, I960.
Bruno Nettl, Music in primitive culture, Cambridge (Mass.),
1956.
INDEX OF NAMES
Aalst, J. A. van II7.

Abraham, Hans 141.
Abraham, Otto 10, 29, 30, 53.
Adler, Guido II, 34, 185
Ambros, August Wilhelm 7,8.
Ames, David 205.
Amiot, J oseph 9.
Anderson, V. C. 21.
Antonowytsch, Miroslav 176, 189.
Apel, Willi 141, 215.
Appun 20.
Arakchiev, D. 32.
Arbatsky, Yuri 69, II8, 218.
Archytas 67.
Aristoxenos 67.
Ashley Montagu, M. F. 43
Ashton, Hugh 79, 135.
Astley, Thomas 205.
Athenaios 190.
Avenary, Hanoch 100.
Azara, Felix de 39.
Besseler, Heinrich 158.

Bidney, David 136.
Binchois, Gilles 215.
Blacking, J ohn 196.
Boas, Franz 86, 114, 210, 212.
Boethius 5.
Bose, Fritz 28, 66, 215, 223.
Brahms, Johannes 57, 172.
Braidwood 40.
Brailoiu, Constantin 53, II4, 153,
155, 158, 160, 170, 178.
Brandei, Rose 55, 89, 98, IIl, 164,
187.
Bcher, Karl 34, II2.
Bcken, Ernst 114.
Bukofzer, Manfred 102.
Burney, Charles 5, 6, 8.
Burrows, Edwin G. 37, 53, 54, 63,
68, 69, 71, 87, 93, 130, 167, 177,
18o, 183, 208.
Byrd, William 216.
Bach, J. S. 191,210,215.
Baker, Theodore II.
Baldus, Herbert 90.
Balys, Jonas 181.
Barbeau, Marius 2, 49, 52, 53, 140.
Bart6k, Bela 120, 206.
Bascom, William 197.
Baud-Bovy, Samuel 148.
Becher, Hans 39.
Becking, Gustav 78, 126.
Beckler, Hermann 87.
Beethoven, Ludwig van 215, 217,
218.
Begun, S. J. 19.
Beier, Ulli 35.
Belaiev, Victor 69.
Berliner, Emil 18.
Bernini, Lorenzo 215.
Carrington, John F. 98, 208.

Chapple, L. W. 42.
Charlemagne 45.
Classe, A. 36.
Coclicus, Adrian Petit 46.
Cole, Sonia 41, 77.
Collins, H. B. 151.
Condominas, Georges 42, 106, 107.
Confucius 219.
Conran, Michael 202.
Coon, C. S. 42.
Couperin le Grand, Fran~ois 50.
Courlander, Harold 98.
Cowell, Henry 85, 88, 95.
Cozzens, Edwin 204.
Curtis, Natalie 33, 71, 140.
Dahlback, Karl 21.
INDEX OF NAMES
225
Dalberg, F. H. von 10.

Danckert, Werner 73, 168.
Dante 114.
Davies, E. Harold 51.
Davis, Thomas 206.
Debussy, Claude 2.
Densmore, Frances 14, 17, 31, 32,
38, 57, 68, 69, 81, 83, 84, II9,
134, 137, 140, 150, 162, 166, 193
Dobrizhoffer, Martin 53, 87.
Drger, Hans-Heinz 92.
Driberg, J. 140, 209
Du Bois, Constance Goddard 34,
69
Du Bois, Cora 128, 135.
Du Bois Reymond, Marie 220.
Dufay, Guillaume 89, 121, 2105.
Dzhambu13
Gershwin, George !I8.

Gerson-Kiwi, Edith 26.
Gheerbrant, Alain 84, 93, 97.
Gilbert, Dorothy R. 36.
Gilman, Benjamin Ives 13, 30, 31,
52, 160.
Gluck, Chr. W. von 77.
Goethe, Johann Wolfgang von 78.
Goldmark, Peter C. 18.
Goudimel, Claude 151.
Grattan Flood, William H. 202, 206
Greni, Liv 36.
Griffin, James B. 17.
Grocheo, Johannes de 123.
Grove, George 36.
Grtzmacher, M. 21.
Gunther, Erna 134, 140.
Gurvin, Olav 21.
Ebert, Max 74.

Edison, Thomas Alva 10, 13, 213.
Elkin, A. P. 51, 75, 81, 87, 136, 140,
141, 184, 193
EIlis, Alexander John 12, 13, 20,
24,25
Elwin, Verrier 3, 129,214.
Emsheimer, Ernst 3, 135, 136, 158,
161, 203.
Estreicher, Zygmunt 59, 62, 69, 91,
138, 139, 165, 176, 193
Haddon, Ernest 104.

Hagen, B. II, 159, 182, 212.
Hagen, Carl II.
Hall, Jody 72.
Handel, George Frederic 50, 215.
Haslund-Christensen, H. 3.
Hassler, Hans Leo 50.
Hawkins, Sir John 5, 6, 8
Haydn, Joseph 139.
Haywood, Charles 223.
Heine, Heinrich 50.
Heinitz, Wilhelm 123.
Heizer, Robert F. 40.
Helmholtz, H. L. F. 12, 13, 20.
Hemingway, Ernest 95.
Herodotos 46.
Herrlinger, G. 76.
Herskovits, Frances S. 33, 148.
Herskovits, Melville J. 17, 19, 33,
40, 44, 86, 106, 135, 136, 141,
142, 148, 210, 212.
Herzog, George 14, 35, 36, 60, 66,
71, 75, 76, 77, 79, 86, 87, II6,
121, 126, 127, 179.
Hickmann, Hans 203.
Hivale, Shamrao 3, 129.
Hoebel, E. A. 17,39,4,44,86,138.
Hoerburger, Felix 126.
Homer 206.
Hood, Mantle 208.
Hornbostel, Erich M. von 10, 12,
14, 20, 26, 29, 30, 34, 53, 57, 62,
66, 67, 81, 86, 92, 101, 102, 125,
Fagg, Bernard 106

Felber, Erwin 38, !I9, 143.
Ferand, Ernst T. 180.
Fetis, Franc;ois-Joseph 7, 47
Fewkes, Jesse Walter 13.
Fleteher, Alice Cunningham 32, 58,
66, 71, 131, 160, 165.
Ford, James A. 17.
Forkel, Johann Nicolaus 6.
Fortune, R. F. 2, 96, 127, 140.
Freeman, Linton C. 146.
Frer-Haimendorf, Christoph von
I, 2, 4.
Gaffori, Franchinus 180.
Garfield, Viola E. 2, 52.
Garrod 40.
Gbeho, Philip 218.
Geiger, Bernhard !I9.
Gelatt, Roland 18.
Gennrich, Friedrich 158.
226
INDEX OF NAMES
128, 144, 149, 156, 159, 176, 182,

183, 196, 212.
Howells, William 38, 83, 213.
Hudson, Arthur Palmer 76.
Hbner, Herbert 88, 166.
Hullah, John F. 7, 10.
Husmann, Heinrich 26.
Idelsohn, A. Z. 64, 73, 169, 171,
17 2 .
Jankovic, Danica S. 151, 195.
Jankovic, Ljubica S. 195.
Jenness, Diamond 69,138,150, 157
Jennings 17, 39, 40, 44, 86, 138.
Jones, Arthur Morris 35, 43, II3,
130, 138, 188, 196, 197.
Jones, Trevor A. 51.
Jones, William 10.
Junod, Henri A. 57, 83, 104, 125,
128, 135, 220.
Junod, Henri-Philippe 196.
Kaufmann, Walter 68.
Kenyatta, Jomo 95, 109, 134
Kinkeldey, Otto 178.
Kirby, Percival R. 36, 202, 207.
Kluckhohn, Clyde 133, 134
Knosp, Gaston 163.
Koch-Grnberg, Theodor 86.
Kolinski, Mieczyslaw 33,45, 59, 71,
148, 173, 184.
Korda, Victor 188, 189.
Krause, K. C. F. 7.
Kroeber, A. L. 40, 68, 96, 137, 167.
Kuba, Ludwig 180.
Khn, Herbert 74.
Kunst, Jaap 8, 12, 14, 15, 17, 20,
23, 37, 43, 55, 59, 66, 74, 102,
104, 124, 142, 149, 150, 157, 176,
178, 18o, 184, 199, 25, 208, 223.
Kurath, Gertrude 161, 162, 167.
Kurosawa, Takamoto 178.
Kuttner, Fritz A. 21, 34, 107.
Kyagambiddwa, Joseph 35, IIO,
128, 198, 207.
Laborde, J. B. de 6.
Labouret, Henri 37, 205.
Lach, Robert 34, 71, 124, 168.
Lachmann, Robert 66, 114, 149,
24
Lambertus, Magister 88.

Landino, Francesco 155.
Lasso, Orlando di 15, 50, 216.
Launis, Armas 78.
Lawes, William 151.
Lawrey, Eleanor 151.
Leden, Christian 18, 69, 90, 165,
218.
Leonardo 133.
Levy-Strauss, Claude 43.
Libby, Willard F. 40.
Lichtenwanger, William 223.
Lineff, Eugenie 30, 189.
Linton, Ralph 76, 8o, 86, 15, 106,
132, 201, 203.
LIoyd, Lucy 68.
Longinus (pseudo) 190.
Loomis, A. 21.
Lottermoser, W. 21.
L Pu-We 34, 220.
Lurje, W. 4, 180, 181, 185.
Machaut, Guillaume de 121.
Mahler, G. 139.
Malinowski, Bronislaw 81,135,167,
2II.
Marais, Marin 50.
Marcel-Dubois, Claudie 101.
Marinus, Albert 167.
Martini, Padre 5, 6.
McAllester, David P. 52, 57, 64, 86,
134
Merriam, Alan P. 33, 69, 86, 109,
147, 167.
Merriam, Barbara W. 86, 146.
Metfessel, Milton E. 21.
Meyer, Leonard B. 221.
Michel Angelo 133.
Milne, Leslie 83, 150.
Moberg, Carl Allan 36.
Monteverdi, Claudio 38, 215, 216.
Moodie 150.
Mozart, W. A. 210, 219.
Murdock, George P. 39, 138.
Musser, Claire Omar 108.
Mussorgsky, Modeste 151.
Myers, Ch. S. 169.
Napoleon 9.
Nettl, Bruno 72, 121, 122, 223.
Nguyen Van Huyen 128.
Nicolai, Philipp 156.
INDEX OF NAMES
Nketia, J. H. 203.
Obresc):lkoff, Chr. u8.
Oost, P. J. van 159.
Opler, Morris Edward 74,79.
Orff, Carl 151.
Ortiz, Fernando 59, 173, 197.
Osgood, Cornelius 17.
Page du Pratz, Le 44.
Palestrina, G. P. da 15, 216.
Pnini 143.
Peary, Robert W. 90.
Pepper, Rerbert 37.
Phidias 2 u.
Phillips, Philip 17.
Philol00.
Picasso, Pablo 211.
Piggott, Frances T. 98, 151.
Pike, Kenneth L. 35.
Plato II2, 189.
Plutarch 120, 160.
Poloczek, Frantisek u8.
Pope, Alexander 202.
Popley, Rerbert A. 141.
Poulsen, Valdemar 18.
Ptolemy 67.
Pythagoras 5.
Quintilianus, Marcus Fabius 92,
II2, 194.
Radc1iffe-Brown, A. R. 138.
Radin, Paul 81, 124.
Raphael 133, 215.
Raven-Rart, R. 167.
Redly, Bishop 16.
Reese, Gustave 180, 189.
Reimer, Dietrich 91.
Reiner, E. 26, 28, 206.
Reiner, Marcus 26.
Rembrandt 133, 215.
Renshaw, Rosette 209.
Revesz, Geza 34.
Rhodes, Willard 15, 134, 140, 167.
Riegler-Dinu, Emil 65, 125, 165.
Riemann, Rugo 152.
Ringbom, Nils-Eric 73.
Roberts, Relen Reffron 37, 56, 69,
70, 77, II5, 134, 144, 150, 157,
170, 176, 208.
Ross, J ohn 69, 90.
227
Rouget, Gilbert 131.

Rousseau, Jean-Jacques 9, 217,
218, 221.
Rowbotham, J ohn Frederic 143.
Rubin, Meyer 40.
Sachs, Curt 4, 8, 26, 46, 55, 56, 57,
58, 63, 64, 67, 71, 78, 79, 84, 88,
90,92,96,101,13,15,17,110
114, 115, 117, 118, 120, 121, 125,
126, 127, 128, 135, 137, 140, 141,
143, 149, 160, 162, 167, 176, 177,
178, 179, 182, 184, 187, 188, 193,
194, 197, 206, 207, 220.
Sachs, Rans 50, 137.
Sande, G. A. J. van der 74
Santos, Joo dos 202.
Sapir, Edward 176.
Savart, Felix 24.
Schaeffner, Andre 37, 80, 92, 97,
106, 107.
Schmidt, Wilhelm 101.
Schneider, Charles 158.
Schneider, Marius 34, 176.
Schoenberg, A. 216.
Schubert, Franz 78, 172, 209.
Scott, Geoffrey 9.
Seeger, Charles 21.
Seewald, Otto 92.
Seikilos 58.
Selenka, M. 18.
Shakespeare, William 214.
Shelly, Percy B. 214.
Shepherd, John 151.
Shiba; Sukehiro 187.
Si1cher, Paul 50.
Smith, G. E. 17, 39, 4,44,86,138.
Souvtchinsky, Pierre 153, 160, 170.
Spector, Johanna 84.
Spier, Leslie 86.
Stanley, Douglas 85, 88.
Steenhoven, Geert van den 62.
Stein, R. 90.
Stravinsky, Igor 2, 120, 151, 193,
216.
Strelnikov, J. D. 125.
Stumpf, Carl II, 13, 14, 16, 20, 52,
55, 57, 66, II5, 185, 197
Suess, Rans 40.
Suwanto, Raden 90.
Swadesh, Morris II5, 176, 208.
Swanton, J ohn R. 44.
228
INDEX OF NAMES
Tacitus 84.
Tax, s. 167.
Tessmann, Gnther 128.
Thompson, Laura 167.
Thorsteinsson, B. 151.
Thurnwald, Richard 8, 41, 44, 95,
101, 137, 138, 140, 144, 156.
Tiren, Karl 146, 147, 163.
Tracey, Edwin S. 32, 63, 104, 16o,
165.
Tracey, Hugh T. 17,55,58,71, 130,
131, 198.
Trowell, Margaret 101, 202.
Visnen, A. O. 121, 144, 147, 159,
17I.
Victoria 15, 16.
Villoteau, Guillaume-Andre 9, 10.
Vinaver, Chemjo 69.
Virgil II7.
Vogelweide, Walter von der 45.
Vries, Hessel de 40.
Wachsmann, Klaus P. 74,87, 91,
101, 13, 14, 202, 23, 207.
Wagner, Richard 38, 68, 77, 137,
140, 187, 216.
Wallaschek, Richard 2, 12, 32, 36,

49, 2 5.
Waterman, Richard A. 223.
Weber, C. M. von 9.
Webster 212.
Weisbecker, A. 68.
Werner, Eric 137, 219.
Werner, Heinz 72.
Wertheimer, Max 18, 113.
Westphal, R. 8.
White, J. Manchip 17, 42, 48, 219.
Wilhelm, Richard 34.
Williamson, Robert W. 184.
Wingert, Paul S. 2, 52.
Wiora, Walter 15, 59, 64, 66, 145,
166.
Wolf, Hugo 77.
Wolf, Johannes 123, 204.
Wren, Christopher 215.
Wright, Frank Lloyd 211.
Wnsch, Walther 78.
Young, R. W. 21, 26.
Zelter, Kar! Friedrich 78.
Zeuner, Frederick E. 40.
Zeuxis 211.

Curt Sachs, Jaap Kunst History of Music

Uploaded by

Copyright:

Available Formats

Curt Sachs, Jaap Kunst History of Music

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Curt Sachs, Jaap Kunst History of Music

Uploaded by

Copyright:

Available Formats

What is the book about?

What is the book about?

What is the editor's role?

What is the editor's role?

THE WELLSPRINGS OF MUSIC

Photo by Joseph Breitenbach

ISBN 978-94-015-1059-2 (eBook)

Copyright I96I by Martinus Nijhott. The Hague. Netherlands

EDITOR'S PREF ACE

When Irene Sachs asked me to take charge of the practically

typical mild "Sachsonian" humor and illustrated with a wealth

THOUGHTS AND METHODS

IH. The ethnomusicologist's workshop

Field work - phonograph, gramophone, tape - tone

IV. The question of origin

Theories - tone languages - mouth musie - drum languages

V. The oldest music: tumbling strains

VI. The oldest musie: one-step melodies

VII. Conservatism and magie

VIII. Voeal mannerisms

Von Hornbostel's and Merriam's descriptions - European,

X. Rhythm and form

Growth, personality, art

Chapter Four I THE

FATE OF SECONDAL AND TERTIAL

THE FATE OF QUARTAL AND QUINTAL

Disjunction and conjunction - quartal chains - chains of

Chapter Six !CENTRIC

Gregorian tenor melodies - the concept of

Chapter Seven / POLYPHONY

Independent accompaniment - hemiola - "syncopation" Afro-Indian concordances

PROFESSIONAL MUSIC AND MUSICAL SYSTEMS

THOUGHTS AND METHODS

Western music, the pride of our culture, is no longer what

Christoph von Frer-Haimendorf, The naked Nagas, London, 1939. p.

THOUGHTS AND METHODS

pp. 406 ff.

THOUGHTS AND METHODS

Such music cannot be bought in stores, but comes from

Reports Irom the scienti/ic expedition to the Norlhwestern provinces 01 China,

VIII, Ethnography 4, Stockholm, 1943, pp.

THOUGHTS AND METHODS

11 Von Frer-Haimendorf, l.c., p. 55.

THOUGHTS AND METHODS

II. THE ADVENT OF THE ETHNOMUSICOLOGIST*

The two pillars of European education during the Middle

THOUGHTS AND METHODS

Their best music - as he wrote in the Preface - His said to be

THOUGHTS AND METH ODS

THOUGHTS AND METHODS

pronunciamentos such as: "Assyrian music seems (!) never to

THOUGHTS AND METHODS

After these early attempts, it was an epochal moment when

THOUGHTS AND METHODS

free of the preconceived ideas of a European. Villoteau, it must

aFCI ~Et U F:1

THOUGHTS AND METHODS

And so it eame to pass that his "regularized" version proeeeded

WE NEED NOT follow up the subsequent attempts of the

THOUGHTS AND METHODS