CONSONANT MUTATIONS

IN CONCEPTUAL EVOLUTION OF NOLDORIN/SINDARIN PHONOLOGY

by Ryszard Derdzinski
ABSTRACT: Consonant mutations were characteristic for all stages of conceptual evolution of J.R.R.
Tolkien's Welsh-sounding language which evolved from early Goldogrin to Sindarin. In my essay I
want to compare the consonant mutations in Welsh and the consonant mutations in early stages of
conceptual evolution of Welsh-sounding language of J.R.R. Tolkien's legendarium.

KEYWORDS:soft mutation (lenition), hard mutation ('stop mutation'), nasal mutation, phonetic and
grammatical mutation, Goldogrin, Noldorin, Sindarin

INTRODUCTION

In 1955 in his letter to the Houghton Mifflin Co. J.R.R. Tolkien wrote: The 'Sindarin', a Grey-elven
language, is in fact constructed deliberately to resemble Welsh phonologically and to have a
relation to High-elven (i.e. Quenya ) similar to that existing between British (properly so-called,
sc. the Celtic languages spoken in this island at the time of the Roman Invasion) and Latin.1 His
special interest in - and true love of – Cymraeg2 began in Tolkien's childhood in early 1900s when for
the first time he saw Welsh place-names on the coal-trucks near his home in Moseley, then odd and
curious to the boy, but also very beautiful and mysterious.3
Celtic-sounding language of Tolkien's legendarium emerged in the very beginnings of his sub-
creation.4 Around 1914 to 1917 young Tolkien began to devise two related, though unlike languages
which were called Goldogrin and Qenya. "Gnome-speech and Elfin of the Eldar", as he wrote in the
Book of Lost Tales (I, 48),5 were constructed to resemble two real languages: Welsh and Finnish.6
Tolkien's legendarium – stories from the Book of Lost Tales – were meant to provide a historical
context for newly devised languages. In conceptual evolution of Tolkien's languages Goldogrin
became Noldorin in 1920s (which evolved through 1930s and 1940s becoming language of most
names and place-names in The Lord of the Rings) and Sindarin in the beginning of 1950s (which
emerged during the final writing of Appendices for The Lord of the Rings).

In my analysis I will compare Welsh consonant mutations with mutations in Tolkien's Celtic-
sounding language which evolved from Goldogrin to Sindarin.

I. CONSONANT MUTATIONS IN WELSH

Welsh, like other Celtic languages (e.g. Breton, Cornish, Manx, Irish) is characterized by CONSONANT
7
MUTATIONS. It has three types of consonant mutations: SOFT (LENITION), NASAL (NASALIZATION) AND SPIRANT.

1
The Letters of J.R.R. Tolkien, p. 219.
2
i.e. Welsh.
3
For details see J.R.R. Tolkien. A Biography by Humphrey Carpenter, p. 33-34. Tolkien wrote about his attitude towards
Welsh in the essay English and Welsh [in:] J.R.R. Tolkien, The Monsters and the Critics and Other Essays, pp. 162-197.
4
Sub-creation is a term from Tolkien's theory of literature. God is the Creator of His Creation and a writer, as a person
created by God as his likeness, is sub-creator or his sub-creation. For details see J.R.R. Tolkien's On Fairy Stories [in:] J.R.R.
Tolkien, The Monsters and the Critics and Other Essays, pp. 109-161.
5
I use bibliographical abbreviations devised by Christopher Tolkien for his Index to The History of Middle-earth published in
2002. I-XII are twelve parts of The History of Middle-earth.
6
For information about similarities between Goldogrin/Noldorin/Sindarin and Welsh see Sindarin Grammar and Dictionary by
Jim Allan [in:] An Introduction to Elvish by Jim Allan, pp. 49-62.
7
Some authors add here MIXED mutation (see A Welsh Course at www.cs.brown.edu\fun\welsh). For detailed analysis of Welsh
mutations see John Morris-Jones, A Welsh Grammar: Historical and Comparative. Oxford 1913.
SOFT MUTATION (LENITION)

Soft mutation (lenition) is a remarkable phonological and morphonological innovation of Celtic

language-family. Lenition means "softening".8 This name reflects a phonetic mutation in which
"hard" or unvoiced consonants like p or t change to "softened" (lenited) or voiced b, d, while original
b, d are further "softened" to spirants: v, dh. In other words – using definition of Jim Allan – lenition
refers to the loosening and slackening of the pronunciation of a consonant, so that it becomes more
vocalic. Voiceless consonants may be "softened" to voiced consonants, stops may be "softened" to
spirants and spirants may be "softened" to approximants.9
Lenition occurs when the mutating consonant is intervocalic - it is placed between two vowels -
or when it lies between a vowel and a sonant (a syllabic consonant). In Celtic family lenition (in
phonological meaning) took place before the disappearance of unaccented vowels, reduction of
final sounds and simplification of consonant clusters. Disappearance, reduction and simplification
mentioned above did not affect lenition, but they suppressed its phonetic requirements. However in
anlaut (in initial consonants) lenition has been determined by the morphosyntactical context, i.e. it
has occurred (and still occurs) only after particular parts of speech (e.g. lenition of initial consonant
in a noun following its adjective in Welsh, like in hen ddyn 'old man': dyn 'man' > ddyn). This
restriction must have occurred in the time when the phonetic requirements of lenition still existed.
Consequently lenition of initial consonants must have had three phases: (1) phonological phase –
lenition determined by the phonetic context (intervocalic consonants), (2) morphonological phase –
lenition determined by both phonetic and morphosyntactic context, (3) morphosyntactic phase –
after the reduction of final sounds lenition is determined only by the morphosyntactic context.
J.R.R. Tolkien in his writings uses two terms describing these two types of lenition: PHONETIC LENITION
(lenition in phonological meaning) and GRAMMATICAL LENITION (lenition in morphosyntactic meaning).10
According to Leszek Bednarczuk phase (2) took place in Proto-Celtic period.11 Welsh lenition
patterns can be seen below:

limited to
contexts

basic p t c b d g gw m rh ll
mutated b d g f dd - w f r l

NASAL MUTATION (NASALIZATION)

Nasal mutation is typical for all Celtic languages. Final nasal consonant affects initial p, t, k, b, d, g
of the next word. Similarly as in lenition of anlaut, nasal mutation of the initial consonant in Celtic
family has been determined by the morphosyntactical context, i.e. it has occurred (and still occurs)
only after particular parts of speech (e.g. nasal mutation of initial consonant in a noun which
follows preposition yn 'in' in Welsh, like in yn nhŷ 'in house', yng Nghymru 'in Wales'). Welsh nasal
mutation patterns are as follows:

basic p t c b d g
mutated mh nh ngh m n ng

SPIRANT MUTATION

Origin of spirant mutation in Welsh is gemination of initial voiceless consonants. Similarly as lenition
and nasal mutation it has been determined by the morphosyntactical context, i.e. it has occurred
(and still occurs) only after particular parts of speech (e.g. spirant mutation of initial consonant in a

8
Word lenition comes from Latin lenire 'to soften, alleviate' and was first recorded in English in 1910-1915, few years before
Tolkien began to devise his Goldogrin.
9
Cf. Jim Allan's Sindarin Grammar and Dictionary, p. 57.
10
Cf. The Letters, p. 426, note ‡ which reads: though of phonetic origin, they are used grammatically, and so may occur or
be absent in cases where this is not phonetically justified by descent. In his text from c. 1917 Lam na nGoldathon (Parma
Eldalamberon XI, p.7): This […] may be referred to under head of "grammatical mutation" for it was generalized to a rule
and is now used in many cases not justified purely on phonological grounds.
11
Cf. Języki celtyckie [in:] Języki indoeuropejskie, Part II, Warsaw 1988, p. 652)
noun following its possessive in ci chi 'her dog'). Below I present patterns of spirant mutation in
Welsh:

basic p t c
mutated ph th ch

My short overview of Welsh consonant mutations in the general Celtic background is not
sufficient for serious linguistic research, but it has given us a couple of terms and ideas necessary in
analysis of consonant mutations in Celtic-sounding language of Tolkien's Middle-earth.

II. CONSONANT MUTATIONS IN GOLDOGRIN (c. 1917)

First known linguistic description of Gnomish, on that time know as Goldogrin, is a lexicon from c.
1917 entitled I Lam na nGoldathon ('The Tongue of the Gnomes', known also as the Gnomish Lexicon
- GL) and accompanying grammar referred to as Gnomish Grammar (GG). Both texts were published
in 1995 in Parma Eldalamberon Number XI.12 Both are contemporary with The Book of Lost Tales
(part I and II of The History of Middle-earth).13

SOFT MUTATION (LENITION)

Tolkien created several grammatical mutations in his new language. Lexemes collected in GL prove
that phonological lenition didn't occur in early phases of evolution of Gnomish (no mutation of
intervocalic consonants before reduction of final vowels in ancient Goldogrin period. An example
can be Goldogrin ram 'wing, pinion' (PE 11:64), Qenya rāma 'wing' (GL 12:97) – in Noldorin/Sindarin
it would have been **rauf [r'auv] (in fact Noldorin form of 'wing' is rafn < *ramna).
Goldogrin lenition has morphonological context. In GG it is named "interiorchanges" (PE 11:7) – in
1917 new phonological term lenition was almost unknown (see note 8). Sometimes Tolkien uses the
term "softening" (cf. entry i· in PE 11:50). 14 Tolkien wrote that "interiorchange" or "softening" might
be referred to as "grammatical mutation" for it was generalized to a rule and is now used in many
cases not justified purely on phonological grounds. It is used in a good many other cases besides
that of the article [i · 'the']. Goldogrin lenition is probably identical to soft mutation (lenition) in
Early Noldorin (see below). GG is accompanied by the table of "interiorchanges":

basic c cr cl cw t tr p pr pl d
mutated ·g ·gr ·gl ·gw ·d ·dr ·b ·br ·bl ·dh

basic dr g gw gl gr b br bl h
mutated ·dhr ·’ ·’w ·’l ·’r ·v ·vr ·vl ·ch

Note: Sign ’ signifies the lenition product of initial g – a vanishing sound. In Noldorin it is represented in
writing by the gasdil 'stopgap' (V:298, 354, 357). In GG is called by Tolkien Vau. Vau (or Wāw, Vav) means
'hook' and is the sixth letter of Hebrew alphabet.

The exceptions in Goldogrin grammatical lenition are:

12
The Gnomish Grammar and Lexicon [in:] Parma Eldalamberon Numer XI, 1995, edited by Christopher Gilson, Patrick
Wynne, Arden R. Smith and Carl F. Hostetter
13
Similarities and differences between this early Goldogrin and later Sindarin were described in an interesting text by
Christopher Gilson, Gnomish is Sindarin. The Conceptual Evolution of an Elvish Language. [in:] Tolkien's 'Legendarium'.
Essays on 'The History of Middle-earth edited by Verlyn Flieger and Carl F. Hostetter, Greenwood Press 2000.
14
J.R.R. Tolkien used to mark the mutation by placing the symbol · between active and passive element of mutation, e.g.
na·Ngoldathon.
 (1) initial d-, b- and g- which come from original *-nd, *-mb and *-ng like in Goldogrin dor
'land' : i·ndor 'the land' (and not ** i·dhor), Belca 'Melko' : i·Mbelca 'Melko' (and not
**I·Velca), golda 'gnome' : i·ngolda 'the gnome' (and not ** i·’olda);

(2)all words beginning in gi- (by generalization irrespective of the ultimate etymology after
i· N. or D.) like in Goldogrin gilim 'winter' : i·ngilim 'the winter';

(3)many words in go-, ga- (unaccented).15

Interesting is that Goldogrin adjectives in singular follow nouns without mutation (unlike in Sindarin
where adjectives following nouns undergo lenition).

STOP MUTATION (HARD MUTATION)

Stop mutation (so named in Tolkien) occurs after a(d) 'into' (PE 11:17). It belongs to grammatical
mutations too. It is probably identical with hard mutation in Early Noldorin (see below).

NASAL MUTATION

Nasal mutation is in Tolkien's Gnomish Grammar referred to as "n-mutation" (PE 11:12, 17). It is an
"initial grammatical mutation" (PE 11:12). According to PE 11:17 this "n-mutation" occurs in genitival
expressions – with or without a·, an- which is 'genitive prefix'. We can also add here na·(2), pl. nan-
'genitive sg. & pl. of i(n)' (PE 11:7, 59), which is exemplified in the expression: fôs a·Ngalmir 'bath
of Sun' and i·fôs na·Ngalmir 'the bath of the Sun' (Goldogrin Galmir is 'Sun') (PE 11:12) in contrast to
lenition in fôs ’Almir 'bath of Sun' which is a genitive without prefixal flexion (PE 11:12). Goldogrin
nasal mutation probably agrees exactly with nasal mutation of Early Noldorin (see below).

LIQUID ASSIMILATION

Assimilation is any process in which a vowel of consonant becomes phonetically similar or identical
to another in its vicinity.16 Following the liquids l, r, Goldogrin changed plosives to spirants. J.R.R.
Tolkien wrote about this phenomenon describing Sindarin phonology: The much-changed Sindarin of
Middle-earth turned the stops to spirants after l and r.17 Comparing Goldogrin with every phase of
the evolution of Tolkien's Welsh-sounding language, we can see that this LIQUID ASSIMILATION is a
constant feature of its phonological system.

III. CONSONANT MUTATIONS IN 'EARLY' NOLDORIN (c. 1920)

'Early' Noldorin is a term introduced by J.R.R. Tolkien (probably in 1950s) for a his Welsh-sounding
language from the period after compiling the Gnomish Lexicon and Gnomish Grammar.18 The
representative phonological and grammatical description of this language is a text entitled Early
Noldorin Grammar (under this title from 1948 and referred here to as ENG)19 or Lam na·NGoluith
'Language of the Gnomes'. This text is written on Leeds University candidates' examination paper.
Internal evidence indicates that it must have come from c. 1920. Tolkien planned to write an
appendix to ENG which was to explain the sound laws of Noldorin, but it was never completed.
Happily the next conceptual phase of the grammar of Noldorin written in the 1930s would begin
with a comprehensive treatment of the historical phonology of the language.20 Probably it will be
published in the next issue of Parma Eldalamberon.
First part of ENG is devoted to consonant mutations.21 It was the first time when Tolkien
presented all consonant mutations in one complete table. 'Early' Noldorin had three kinds of
consonant mutations: (1) SOFT (i.e. LENITION), (2) HARD (i.e. STOP), (3) NASAL. They were described in
detail in the previous chapter. Below I present tables with all sound-changes included in Tolkien's

15
Details can be found in GG on p. 8.
16
Roger Lass, Old English. A Historical Linguistic Companion, p. 255.
17
UT:265, footnote.
18
Cf. PE 13:119.
19
Ibidem.
20
Ibidem, p. 120, footnote 1.
21
Ibidem, p. 120.
original text, but I rearrange the order of consonants following the table of soft mutations in GG
(see above) and I add consonant clusters (like cr, cl, tr, pr, dr, gl, br, bl) which were presented in
that table:

SOFT MUTATION (LENITION)

basic c cr cl cw t tr p pr pl d
mutated ·g ·gr ·gl ·gw ·d ·dr ·b ·br ·bl ·dh

basic dr g gw gl gr b br bl h
mutated ·dhr ·’ ·’w 22 ·’l ·’r ·v 23 ·vr ·vl - 24

HARD MUTATION (STOP)

basic c cr cl cw t tr p pr pl d
mutated ·ch ·chr ·chl ·chw ·th ·thr ·f ·fr ·fl ·d

basic dr g gw gl gr b br bl h
mutated ·dr ·g ·gw ·gl ·gr ·b ·br ·bl ·h

NASAL MUTATION

basic c cr cl cw t tr p pr pl d
mutated ·ch ·chr ·chl ·chw ·th ·thr ·f ·fr ·fl ·n

basic dr g gw gl gr b br bl h
mutated ·nr ·ng ·ngw ·ngl ·ngr ·m ·mr ·ml ·h

IV. CONSONANT MUTATIONS IN LATE NOLDORIN (1930s – 1940s)

We know little about this period of the Welsh-sounding language of Tolkien’s legendarium. The
only phonological evidence of that “Late Noldorin” can be found in loose notes and one longer text
entitled The Etymologies. We can also trace the evolution of the Noldorin phonology in the sketches
of the most important work-in-progress of that period which was The Lord of the Rings. There exists
“a comprehensive treatment of the historical phonology of the language”25 from 1930s, but it still
waits for publication. Fortunately the editors of the Parma Eldalamberon promise that the next

22
Tolkien adds here ngw.
23
In Tolkien's spelling occurring in ENG bh which in fact is [v].
24
Lenition of h > ch, does not occur in this table.
25
Cf. PE 13:119.
issue of this periodical will contain the linguistic materials from the 1930s. It is possible we will
soon know much more about Tolkien’s conception of the consonant mutations in those years.
All we can say now about consonant mutations in Late Noldorin of 1930s and 1940s is that the
phonological rules did not alter, and we can apply here the rules from 1920s described above. There
is no need to present any tables here.

V. CONSONANT MUTATIONS IN SINDARIN (from 1950s)

Unfortunately Tolkien did not publish any phonological description of the “mature” stage of his
Sindarin. In 1972 in a letter to Richard Jeffery he wrote: I have not bothered to explain the S. (i.e.
Sindarin) lenitions in the Appendices, already overloaded, because I am afraid they would have
been passed over, or have been felt unintelligible and tiresome, by practically all readers, since
that is the normal attitude of the English to Welsh.26
The most interesting description of the consonant mutations in Sindarin is the text by Helge K.
Fauskanger Sindarin - The Noble Tongue27. As Fauskanger writes: We will attempt to describe the
various mutations, as well as they can be reconstructed. The actual evidence being scanty, we must
in many cases fall back on our general understanding of Sindarin phonology to fill the gaps. What
follows is based on a thorough analysis (mainly conducted by eminent Sindarist David Salo), but
future publications may well prove it wrong in some respects. However, the most frequent
mutations (soft and nasal) are relatively well attested, so that we can reconstruct the rules with
some confidence. Fauskanger (after David Salo) adds two additional mutation rules called Mixed
Mutation and Liquid Mutation.
Fauskanger’s Liquid Mutation is reconstructed following the rules of the liquid assimilation
described above (chapter II). We can read in Fauskanger’s text: This mutation represents a leap of
faith. It is not mentioned, alluded to or directly exemplified anywhere in the published material;
yet our general understanding of Sindarin phonology seems to demand it. Examples given by the
author of the Sindarin – The Noble Tongue have no evidence in Tolkien’s published papers and seem
to be improbable. Tolkien never wrote about such grammatical mutation in his works, letters etc.
Welsh – the main inspiration of Sindarin – does not contain such a mutation too. All we can do is to
wait for the publication of the Noldorin materials from 1930s. They should explain this question.
Mixed Mutation of Fauskanger is also a phenomenon with no evidence in Tolkien’s papers. The
only examples of the mutating consonants in the way which is described as the “mixed” mutation
come from a very problematic text which belongs to the liquid period of the very “late” Noldorin or
very “early” Sindarin (the specialists are not sure if we can call this stage of the Welsh-sounding
tongue Noldorin or Sindarin) – The King’s Letter28.

The complete analysis of the consonant mutations in Sindarin will be possible after the further
publication of the “Late” Noldorin and Sindarin materials found in Tolkien’s papers. I hope the next
years will provide us with plenty of the new and interesting information about Tolkien’s languages.

Copyright ©2004 Ryszard Derdzinski

Quotations from the works of J.R.R. or Christopher Tolkien are the copyright of their publishers and/or the
Tolkien Estate, and are used here with their kind permission. The word TOLKIEN is a registered trademark of
The J.R.R. Tolkien Estate Limited. The characters and scripts of Tolkien’s invented languages and works in
those languages are the copyright of the Tolkien Estate.

First published as PDF on April 14th, 2004

26
The Letters..., p. 426.
27
http://www.uib.no/People/hnohf/sindarin.htm#mutations
28
IX:128,129.