JF &IF~JM[& ml1JD)&l1&OO1]3JE~(Q)N ~;~

THE BOOK OF EL VEN TONGUES

Volume 2 Number4

Parma is a journal of linguistic studies of fantasy literature,

especially of the Elvish languages and nomenclature in the works
of J.R.R. Tolkien. It is published at irregular intervals as sufficient
material becomes available. Submissions are welcome!

STAFF and CREDITS

Christopher Gilson: Editor
Adam Christensen: Front Cover
Patrick Wynne: Back Cover
Jorge Quinonez: Typing Assistance

CONTENTS

The Unified Field Theory of Elvish, by Patrick Wynne ....... 2

Pronouns in Noldorin and Sindarin, by Tom Loback ......... 3

Itarille Quete, translated by Craig Mamock ................... 6

A Survey of Eldarin Pronouns, by Bill Welden ................... 8

23 and 31: Problematic Numbers in Elvish, by Arden R. Smith .... 16

Orc Military Organization and Language, by Tom Loback ....... 17

An Analysis of Firiel's Song, by Patrick Wynne ................. 21

RUNES: Letters to Parma . ................................. 45

Copyright © 1989 by Christopher Gilson. Individual articles and atrwork remain the property of
their authors. The works of ].R. R. Tolkien published through 1983 are © by George Allen & Unwin,
or else © by Frank Williamson and Christopher Tolkien as Executors of the Estate of ].R.R. Tolkien.
Submissions and comments should be sent to the Editor, 2310A Prince Street, Berkeley, CA 94705
 The Unified Field Theory when a subject has been previously alluded to and
ofEIvish -nte being used "where no subject is previously
mentioned", to use Tolkien's own words. For an
by Patrick Wynne example of extreme Conceptionist thinking at its
[from a letter to Bill Welden, 1 February 1989]
worst I would refer you to a recent issue of
quettar (no. 34, I think, but I can't locate my copy
Y?ur mention of Chris and how he is "always ... nght now) [no. 33, p.12 - Ed.] in which some-
trymg to find a way to make all of the published or one commented that most of "the Etymologies"
unpublished forms correct" brings up the in- was worthless as far as shedding light on the Q.
teresting topic of how to reconcile pre-LotR verb system since it was all so different from the
Quenya with LotR/ post-LotR Quenya. * There LotR material and that the best one could hope for
f~om "the Etym." was to pick it clean of any usable
seem to be two main schools of thought on this.
Chris's approach, in which one analyzes pre-LotR bIts of new vocabulary. This person obviously did
Quenya under the assumption that, theoretically at not, and likely never will, spend any time or effort
least, it can be seen to accord with the LotR/post- in trying to see how the verbs in "the Etymologies"
LotR material, might be dubbed the "Unified can be reconciled with LotR.
Field" schooL The other approach, and the one The main "con" of the "Unifist" school (I will
which currently seems to have the most adherents, not say "Unitarian"!) is that there can be no doubt
assumes by and large that (apparent) discrepancies at all that Tolkien's concept of Quenya did indeed
between the two masses of material are the result of change with time, and that when all is said and
changes in Tolkien' s conception of Quenya with done there will be pre-LotR material which will
the passage of time. This approach could be called have to be deemed wholly irreconcilable with the
the "Unimmaculate Conception" schooL LotR/post-LotR materiaL For example, I strong-
Both schools can be seen to have their pros and ly suspect that the early genitive sing. in final -n
cor:s. I used to be a staunch "Conceptionist" up was rejected in favor of that in -0. But from now
untIl Mythcon last summer, but after having struck on I will have to do an awful lot of long, hard
up an acquaintance with Mr. Gilson there and t?inking about variant forms and possible explana-
having exchanged several letters since, I will tlOns for them before I go declaring anything as
confess that I have now been converted to the "rejected". Take, for example, the early version of
"Unifist" way of thinking. So, I feel I have a fairly the Elvish greeting given in The Return of the
good perspective of life on both sides of the Shadow pg. 324: Eleni silir Iumesse omen-
linguistic fence, so to speak. It seems to me that tiemman, with its variant Elen sile ... In my
the main danger in the Conceptionist school is that days as a Conceptionist I would have thought auto-
it makes it too easy (for some people) to be a lazy matically: present tense sing. and plural endino-s -e
thinker. That was my problem, at any rate. For and -ir rejected in favor of later -a and -ar~ 1st
example, given -nte 'they' in the post-LotR person pI. possessive -mma rejected in favor of
co.rpus and -Ito 'they' in the pre-LotR corpus, I later -Iva, -Ima. Now I will not be so hasty. Of
blIthely assumed, without exerting too much course we see present tense 3rd pers. sing. verbs
mental energy, that -nte was a later, "correct" ending ~n -e in "The Etym.": tape 'he stops',
form which replaced the earlier (and now rejected) teke 'wntes', etc. Perhaps sile/silir differs form
sila in mood or aspect, hence a different inflexion
form -Ito. Chris recently pointed out to me that
(note also the short vowel in the earlier forms).
both forms are likely to be valid, -Ito being used Sile and sila might even be different verbs,
* This is perhaps a slightly better way to refer to the perhaps one < SIL and the other < THIL, and so
distinction than "published" and "unpublished". Certainly
they conjugate differently. Perhaps -mma is a
previously unknown mute on that point. Any or all
there is a lot of Quenya - Cirion's Oath, the latest version
of these suppositions may be wrong, but you get
of "OHima Markirya", etc. - which was written after
th~ idea .. I may sound overwrought when I say
LotR's publication and yet remained "unpublished" during thIS (and like any new convert to a cause or belief I
Tolkien's life, and one can probably feel almost 100% perhaps am liable to overzelousness), but each
certain that there would be nothing in this post-LotR word from the Master's pen is a precious thing,
unpublished material which would deliberately contradict the and I will not discard them lightly.
Q. seen in LotR.

2
 Pronouns in Noldorin and Sindarin
by 1"0 m loback

Cases: Subject Object PQssessive

Noldorin
Singular ho 'he' hon 'him' hono 'his'
he 'she' hen 'her' hene 'hers'
ha'it' ha'it' hana'its'
Plural huin'they' hin'them' hein 'theirs'
Sindarin
Singular hi 'this' hi 'this'
Plural hin 'these' hin'these'
Singular ha(?) 'that' ('that')
Plural hain 'those' hain 'those'

Singular e 'he' ('him') in 'his'

din 'his'
Ie 'thou' Ie 'thee' (,thine, thy')
.Im,mm.. (?) nin, nim(?)*
'I, myself' 'me, myself' ('mine')
('you') ('you') ('yours')
Plural ('we, ourselves') ('us') ('our')
?Royal e 'we/he' e 'us/he' in 'our/his'
• Formal usage vs. common: (J-chebln estel anlm, could read - 'not kept hope for me'.

Also while in nora lim, Asfaloth the lim can be interpreted otherwise and Glorfindel is speak-
ing apparently to Asfaloth; the total blank for 'you' is very suspicious - could it be that lim can be
interpreted to mean 'you'? Seems to fit with Ie and im/nim.
Further uin in letter seems odd. Could it be uin Echuir 'their Stirring' implying balance of text,
23 Feb etc.
Highly speculative fills might be:

?ha, hai 'that' ?i ?de ?ge 'him' obj.

?nini, nimi 'mine'
?Iene 'thine, thy'
?Ii 'you' lin, lim 'you' lini, limi 'your, yours'

Third Age Sind. may have dropped final vowel of possessives?

Following your suggestion that the Noldorin pronouns are arranged in case order, I offer up this tentative chart for Parma. By
studying case definitions, etc., in the Little Oxford Dictionary and assuming that Tolkien followed that order a more filled chart can be
made. Interestingly this led to a re-definition of hi, hln, haln from the Moria Gate inscription (re. 'these' and 'those' see Little Oxford,

3
and therefore hi ='this'). Overall look seems to contradict usage of e = 'he' and din ='his' in letter of Elessarto Samwise. Question:
Since gate inscription translation seems to indicate an "everyday" usage confusion between them and those such as we all make
at times, could that explain the anomaly of e = 'he' (perhaps a royal 'we' or a familiar 'he' or archaic form like theelthou) and in,
din = 'his' (perhaps adj. obj. plu. pass. of hono in spite of bess din sing.; or a later 3rd Age common usage indicating asimplification
of an obviously complicated situation?)? Orwhat? Up to a point Nold. and Sind. sections seem to fit together well especially Moria
Gate which was probably written nearer in time to Etymologies than Elessar's Letter.

P.S. Love to know what you think on this.

P.P.S. Can't say that I can chart all pronoun cases in English, perhaps you can. It seems JRRT's intention these have a form
in Nold.lSind. at the start of LotR and that there is some regular progression of forms including differing vowel changes in the
progressive change of a plural (Nold. 'they, them, theirs', Sind. 'this, that, these, those').
If huln is plural form of ho, he, ha, how does that affect plurals elsewhere, if at all?

Editor's Reply

When I suggested case assignments forthe 3rd person pronouns I was thinking only of the singular. The plurals have a pattern
which in the context of Etymologies seems more inviting. The vowels in the forms (which solely distinguish them from each other)
hY.i.n, hIn, h~n correspond to the phonetic pattern of plural nouns in Noldorin: amQ.n 'hill' pI. emu In, emyn; n~n 'water' pI. nIn;
adjir 'father' pI. ed~r, eder.) This suggests that the differences amoung the plural pronouns marked "gender" rather than case,
at least etymologically, thus:

Masculine ho 'he' hon hono huin'they'

Feminine he 'she' hen hene hin'they'
Neutral ha'it' hana hein 'they'

I suppose it should be added that this sort of paradigm might be susceptible to historical change, since a purely masculine 'they'
or feminine 'they' would occur less frequently than the neutral 'they', especially if the last included both pure plural of "it and it .
. ." and the common plural of "he and she (and others)". And I would rather leave haln 'them' translated as in the text and see
a connection with heln 'they' as plural of 'it'. These forms heln, hain could be viewed as phonetic variants of each other: ct. N
celr, S calr 'ship', Er~nlon 'scion of kings', Fornost Eriiln 'Norbury of the Kings'. But I am willing to admit that these could not
only come from the hypothetical PQ plural *han-i" but also from *han-ya-, *hen-ya-, *hin-ya. So it may combine with the ''function"
of Eng!. those as well as they, them. As for the demonstrative of proximity in Sindarin, I would relate them this way to the table:

Masculine ho 'he' hon hono huin'they'

Feminine he 'she' hen hene hin'they'
Proximity hi 'now' hin 'these'

This suggests the relationship between hi and hln is still our best clue for translating the difference between ho, he and hon, hen.
Since the final -n does not mark plural in 'he' or 'she' perhaps that is not the functional difference between 'now' and 'these'. Though
'now' does in a way mean 'this (time)" it can also be seen as equivalent of 'here (in time)'. Certainly hi is derived from root SI-
'this, here, now', just like ho < so, he < se. The full context of hln is Celebrlmbor teithant I thiw hin 'C. drew these signs' where
it seems the combination i ..• hin is what equates to English these. In other words when the demonstative is used as an adjective
it follows the noun which is also accompanied by the definite article. This same rule applies in Welsh grammar: y ty hwn 'this house'
lit. 'the house this', which if we try to put into idiomatic English while retaining the article would be something like 'the house here'.
In other words the demonstrative adjectives in this house, these signs are effectively equivalent to an adjectival form of here. If
we relate this to the personal pronouns he, she their adjectival forms are the possessive: his house, her house, his signs, her signs.
Notice that the form of these pronouns does not change between modifying a singular vs. plural noun. The difference between
this and these is in agreement with the number of the noun modified rather than the "antecedent" (if we can view 'here' as the
peculiar antecedent of hi, hin the way 'Dick' or 'Jane' would be the antecedents of ho, hon and he, hen), then it seems theoretically
possible to have hi 'here, now' with possessive hln 'of here, of now' = both 'this' and 'these'.

4
 Taking all this into account, I am inclined to view the probable way of saying 'his sign' and 'her sign' as tiw hon and tiw hen.
Of course 'his. sign' can mean 'a sign from him', 'a sign to him' or 'a sign about him'. I would also compare therefor the "objective"
1st person in A tiro n1n. Fanuilos! '0 guard~, Elbereth!' Beside ~is there is also in UTered e-mbar nln 'the mountains of!IlJ'.
home'. I would guess that this may be a nominalization of the possessive pronoun like 'mine' from 'my', 'hers' from 'her', 'theirs'
from 'their'. The literal translation of the phrase would be 'mountains of the home (that is) mine'. The "noun" form is used here
because the possessive modifies another possessive e-mbar '(of) the home', which is possessive purely by position in the phrase
following the main noun. The object cumpossessive form nin would modify the main noun if it were left short: ered e-mbar nin
'my mountains at home'. The (absolute) possessives in, din 'his' are further examples of this lengthened form, and possibly also
of the grammatical device of "deferential" vs. "formal" in the third person, which he mentions in Appendix F, II, as an aspect of some
Mannish speech habits of Third Age Middle-earth.
The connection of Noldorin ho, he, ha to Sindarin e 'he' may lie in an historical change. In unempatic syllables the sound
h tended to be lost in Sindarin. This would have lead to phonetic alternates 0, e, a of which the only homophone that had any utility
in its semantics (0 and a already had various other uses), was e a sometime variant of the word i, in, e, en 'the'. In Sindarin one
could perhaps either emphasise the gender of the subject, ho, he, ha or simply use e for any singular subject of whatever gender.
The possessive in is surely devised by analogy with nln, nln.
The breakdown of anim into *annim < an + *nlm supports an assumption of analogical *nim > im. Still we should keep in
mind that the parallel we are trying to explain is 1m 'I' (appositive both times) and lim which since it is adresses to a horse, and
the imperative verb is noro 'run, ride on', it seems that the sense of 11m in context is 'by yourself'. I could believe an element im
"self" derived from ING- 'first, formost', and related to the Q name Ingwe. Tolkien implies that *ngw would normally yield Noldorin
mb which we know in regular progression yields mm > "m" in Sindarin. It could be then that lim is from Ie + im. And the phrase
1m Narvi = 'Narvi's self', which satisfies perfectly the semantic requirements of the inscription, granted the possibility of Sindarin
idiom. But we can never lose sight that the fact of the matter is that 1m Narvl = 'I Narvi' in actual Elven usage.

5
 Quenya Version

Itarille quete: "Nainuvan, an

atarinya cena i umbarihyo tulie
Translated by ©IF®~@ [RaJ®[l'[1i)@©~ ant a r a 1 mindonihyasse; nain u-
van otsove 2 an herunya ataltie
t e r - m a h tal a 3 Melcor ar u-
Whilst going through the poem "1m Naitho" which entuluva marenna!" - an orerya
appeared in issue seven of Parma, I looked up the
part of "The Fall of Gondolin" from which the poem
ne 4 rUClna i lomeo nwalmesse.
was drawn so as to get a better idea of what was Tuor quete: "Ye! Itarille, ea
going on. Naturally, I began to consider ways in n e,S ar coianye; 6 sf tultu-
which I would have translated certain portions which van 7 atarilya sinome, nais 8
were rather difficult to render. Since I am an
Melcoro Mardissen!" Ar vanes
avowed quenyandll (perhaps even a quenyan-
dur), it was natural that I should do so in High-Elven u n d u 9 i ambo erve,1 0 rucina
terms, hence this piece. It is written as far as orehya i indiso 11 nainienen. Si,
possible in the Quenya evidenced in The Lord of the
il u v e a 12 orerya nainieo raumo-
Rings, and other places. I have tried to avoid,
insofar as it was possible, the use of vocabulary 1 'Highest, loftiest': an- superlative or intensive
from the volumes of The History fo Middle Earth, prefix, tara 'lofty'.
except where unavoidable and the word in question 2 'Seven times, to the seventh extent': otso
has related forms in the "final-form" corpus. 'seven', *-ve adverb-forming suffix.
3 mahta- 'fight' - Etym. MAK-; cpo macil,
Original Version maca r. I didn't translate this literally as I don't
(The Book of Lost Tales 2, p.187,
like NIB- &c., though I can't say why.
Unwin Paperbacks 1986 Edition)
4 Past tense of na- 'to be'.
Then said Idril: "Woe is me whose father S I prefer this to *ni by analogy with Ie, which
awaiteth doom even upon his topmost pinnacle; we are told is from Quenya; I think we have the
but seven times woe whose lord hath gone down same morpheme as for more appearing here.
before Melko and will stride home no more!" - for 6 By analogy with S cui-.
she was distraught with the agony of that night. 7 tulta- 'fetch' - Etym. TUL-; cpo utu lien,
Then said Tuor: "Lo! Idril, it is I, and I live; Entulesse, etc.
yet now I will get thy father hence, be it from the
8 Purely a subjunctive, of course; no optative
Halls of Melko!" With that he would make down
force.
the hill alone, maddened by the grief of his wife;
but she coming to her wits in a storm of weeping 9 undu 'down' - Etym. UNU-.
clasped his knees saying: "My lord! My lord!" 10 erve 'alone, on his own'. Based somewhat
and delayed him. Yet even as they spake a great hazardously on *elve 'like a star, in a starry
noise and a yelling rose from that place of manner' whence elvea 'starlike'. Other
anguish. Behold, the tower leapt into flame and possibilities are *ereve, *erave, or perhaps
in a stab of fire it fell, for the dragons crushed more plausibly *eryave (cp. erya in Etym. ER-).
the base of it and all who stood there. Great was 11 Possibly Indira; cpo plural olorl of 0105.
the clangour of that terrible fall, and therein 12 'Whole'; adj. formed from iluve by analogy
passed Turgon King of the Gondothlim, and for with lome/lomea, rave/ravea,
that hour the victory was to Melko. yalme/yalmea &c.

6
 t~~~~'i~~~~;I~,;;~\;'~;t)~~;~1"
sse maperye 13 telcohyar 14 que-
tala "Herunya! Herunya!" ar
~ .i!?t. . :.•\. ., ,. ,'. . ¥.•, . ,~. '~""'"
met t e 1 5 vaniehya. Lumesse -.r":'" . . .,..,:._",. \... . . . . ",
.....'1,.. ' . , ' ..•.
~ t~.:\~.,l...·"I't"·' .:.~'
";'. .....'
~,""':""
.t- ".
.~ "Ti·;:f:f..:·~·'~t .. ·-:-:;:::~7." <:~'.;:~:. ~.~:.', ~:
\"I!:,.•"........• ,<,.: .. ",v.;.\~, • ~

quetente rave ar nalme 1 6 tuler i

nwalmeo ardallo. Ye, i mindon
cap e 17 runyanna ar taltes naro
cirie, an i urul6ci racer sundahya
ar i rimbe i ne ara 18 se. 19
And a 20 ne i aica 21 taltieo
I am m a 22 yasse vane Turcanu i
o n dol i n d e rim b e 0 23 Aran ar
lumesse Melcor ne i nacil.24
Comments &c. welcomed.
<,;; :\'~f~ ... ~:. ...... .) \~

13 map- 'seize' - Etym. MAP-; cpo S mab

'hand'.
14 For use of teleo here as 'leg', compare
Teleontar, also TE'LEK- in the Etymologies. I~ )~:: ::'~';~'J '" t
1 5 'Ended" , from *metta- 'to end' , based on ~;.' .\~:~:~:. '-~
metta 'ending' (adj.) I think this can be used
transitively; if not, Etym. has a verb metya- 'put
an end to' (M ET -). A word related to S dar- is
another possibility.
16 'Clamour' - Etym. NGAL/NGALAM-. Cpo S
glam.
17 'Leapt'; cpo Seabed and Etym. KAP-. I've
translated this phrase literally, but am not sure
what was meant. [The OED gives an obsolete
metaphoric sense of 'burst' for the verb leap,
which may be meant here - Ed.]
18 'Beside'; a proposed preposition corresponding
to ar- given in the Appendix to the Silmarillion;
influenced by the form ara se.en in the Etymologies
(AR2.).
1 9 'It'; cpo note 5, above.
20 Anda means not just 'long', but 'great', also;
cpo adv. andave.
21 Here, 'fell'.
22 'A sound'; Etym. LAM-. Cpo glam, 10m.
23 Modelled on S Gondolindrim.
24 'Victor', seen in Hyarmendaeil; S degil.

7
 subject endings. Thus, in the sentence Nai tiruva-nte-s
i-nte 'May they guard it, those who sit' (UT 305) the
ending -nte is used, where the sentence *1 harar tiruva-
r 'Those who sit will guard it' would presumably require
the simple plural inflexion or.
That the ending -ote is not special in this respect can be
seen from the pair of sentences Nai hiruvalye 'May you
find' and Nai elve hiruva 'May even you find'. In this
case, the inflexi~n -lye is.used when the subject is not
previously mentioned, and is not used when it has been
by Bill Welden mentioned.
There is of a difference between the second and third
The Quenya Pronouns persons in general, in that the second inflexion is sufficent
to completely specify the subject of the sentence, and the
Most of our examples of Quenya pronouns occur in close third person is not; therefore the third inflection is used
combination with verbs and nouns. Subject markers in much more often with a seperate subject, so that the
verb constructs all have the -CCe-; and possesive elements placement of the subject before or after the verb becomes
in nominal constructs all have the form -CCa-. When a an issue.
noun with an attached possessive pronoun is expressed in The suggestion that elye is not simply the subject
the genitive case, the attached pronoun (if it is at the end of pronoun for 'you', but contains some element e- which
the word) takes the form -CCo. From the example sets of carried the meaning 'even' in Tolkien's gloss 'even thou
tie-w-nna 'upon your path' (UT 22, 51) versus hir- shalt find it' has not been raised recently, but perhaps needs
uva-lI.e. 'thou shalt find' (I 394), and omentie-lm..n. 'of to be addressed. In a language where the normal way of
our (exclusive) meeting' (I 90 - early editions - this expressing 'you will find' is hiruvalye, the unusual word
was an error for the inclusive form which appeared later - order elye hiruva (particularly when used in contrast to
see the more detailed discussion below) versus laituva- the normal word order) will place the focus of the sentence
~-t 'we will praise them' (III 231) we assume that the
on the word elye; as if in English we said 'Maybe you'll
consonant pair in a given subject marker is always the fmd it; maybe ~ will find it'. It is most likely this focus
same as the consonant pair in the corresponding possessive which is represented by the word 'even' in Tolkien's
pronoun ending, and construct hypothetical endings translation.
accordingly. This assumption has met with almost We therefore take elye to be the second person subject
universal acceptance in the elvish linguistic community. pronoun in Quenya. From the pair elye/-Iye we had
constructed hypothetical forms for all of the other subject
Here are the subject markers. pronouns, for example **enye 'I' from -nye. This process
resulted in an unconvincingly regular set of pronouns, and
MARKER MEANING REFERENCE was never taken very seriously. Now that we have another
-nye-, -n I III 250, III 245 subject pronoun, inye 'I' (LR 61), construction of the
-Ime we (exclusive) III 231 remaining forms can only be speculation.
*-Ive we (inclusive) I 90 -- later eds. The object pronouns stand in tantalizing relationship to
-lye you (reverential) 1394 the subject markers:
*-rye she (poss. he & it) 1394
-nte they UT 317 SUBJECT OBJECT MEANING REF
(OBJECT)
It appears from the examples that the first person singular -Dye *ni I/me 1394, PE 3
marker takes the form -nye when it is followed byan tye You/you (familiar) LR 61
object marker, but is shortened to -n when it occurs at the -lye Ie You/you (reverential) R 65
end of the verb. *-rye She (he? it?)/it III 250, PE 3
-s
It has occasionally been assumed that the second person -Ime me We (exclusive)/us I 394, LR 56
singular marker can be shortened to *-1 in the same way, *-lve We (inclusive)/us
but there is no evidence for this, and in Galadriel's Lament, -nte te, -t They/them III 231, L 308
the form hiruvalye occurs (I 394) showing that it is at
least possible for the long form to occur in final position. Each of the two patterns here occurs twice, providing a
Tolkien describes the ending' -nte as the 'inflexion of 3 relatively solid base for generalization. The singular subject
plural where no subject is previously mentioned' (UT 317), endings have the form -Cye, and the object pronoun -Ce or
and this has been taken to distinguish it from the other -Ci, using the same consonant. The plural subject pronoun

8
endings have the fonn -CCe and the object pronoun - Ce Sun as She" (I 172). Such a statement would make no
where the object consonant is the same as the second sense if the pronouns for 'he' and 'she' were not different.
consonant in the subject ending. Although this statement is in the early chapters of the
Recalling that the sound s in Quenya changes to r book, and written against a linguistic background in which
between voiced elements, the ending -rye can be seen to Noldorin (as set out in the Etymologies) was the daily
follow this pattern as well. We can contruct two object language of the Elves, the footnote would have been easy
pronouns from this pattern: to excise or edit if Tolkien made it untrue.
In fact Noldorin does distinguish gender in the third
SUBJECT OBJECT MEANING person:
-nye ni I/me
*-tye tye You/you (familiar) s- demonstrative stem. su, SU, s6, so he (cf. -so
-lye Ie You/you (reverential) inflexion of verbs); si, si, se, se, she (cf. -se inflexion of
-rye *se, -s She (he? it?)/it verbs); Cf. N ho, hon, hono he; he, hen, bene she;
-Ime me We (exc1usive)/us ba, hana it; plurals huin, hin, bein. (LR 385)
-Ive *ve We (inc1usive)/us
-nte te, -t They/them The "·so inflexion of verbs" can be seen in the following
sentence (although it is disquised by the shift of s to r
This table is missing only the second person plural, and between voiced elements):
possibly gender variants of the thrid person singular and/or
pluar!. E man antavaro 'What will he give indeed?' (LR 63)
We could also construct object suffIxes for other personal
pronouns, such as ·n 'me' (for example, *ecenielyen Out of convenience (because I want a word for 'yes') I
'you heard me'), but certain markers could have no single take E to mean 'yes' (related to the verb to 'to be'), so that
consonant suffix (m cannot be final in Quenya, for the literal translation is 'Yes, what give-will-he?"
example), and there is a certain symmetry to restricting the It can be seen that the older "·se inflexion of verbs"
attached object markers to the third person only. (meaning 'she') is not much different from our assumed *-
It has also been suggested that -s and ·t represent not the rye 'she' inflection, differing only in the presence of the
singular and plural third person, but rather demonstrative sound y. Would we be justified in assuming an ending *.
pronouns -s 'this/thee' and -1 'that/those'. These meanings ryo 'he' and an ending *.rya 'it'?
are consistent with the roots in the Etymologies, SI- 'this' To begin with, every existing example of Quenya subject
(LR 385) and T A- 'that' (389), and adequately explain the markers has the fonn -CCe. The fonns ·ryo and -rya look
observed fonns (III 250, III 231). An early sentence of like noun endings. Would the presence of these fonns in
Quenya from the manuscript of The Lord of the Rings the language create too much ambiguity? The phrase i
[from the Marquette Collection] bolsters this argument: harar (UT 317) can be translated '(those) who sit' 01:. 'the
sitters', suggesting that the distinction between nouns and
En ni tuviet! 'I have found it! Lo!' verbs is less strict than in English; in which case the cues
provided by the endings on the words will be critical.
The word en we know to mean 'Lo!' (LR 356), and there Then there is the matter of the corresponding possessive
are several early examples where the word ni is clearly '1'. pronoun suffixes. If these all end in ·a, Quenya will
The word tuviet contains tuvie 'I have found', which already show some tendency to lose the third person gender
leaves ·t and meaning 'it' and not 'them'. If the distinction distinction, lending credibility to the suggestion that the
between ·s and -t is one of singular vs. plural, then the gender distinction is lost in other forms as well. The
presence of ·t here is surprising. If, however, it is a feminine form is almost certainly ·rya, based on the
distinction between 'this' and 'that', then either could be evidence of maryat 'her two hands' (I 394). A form *-ryo
considered an appropriate translation of 'it' . for the masculine ending is conceivable, but anything other
than *.rya for the neuter fonn would be surprising.
Gender in the Third Person Finally, if the grammatical consequences are acceptable of
adding a gender distinction in the third person indicated by
There is some evidence that Sindarin and Quenya make a
the vowel of the subject marker; and since Quenya shows
distinction in gender among pronouns, both singular and
plural, of the third person; although it is possible (based great regularity and analogical restructuring of its grammar,
however, mostly on lack of evidence) that this distinction we might expect to find the e/a/o distinction in the third
was abandoned as Tolkien's conception of the languages person subject markers generalized to the markers for other
developed. persons as well; for example, **utuvienyos rather than
The most straight-foward piece of evidence is Tolkien's utuvienyes 'I have found it' (III 250)?
own statement that "Elves (and Hobbits) always refer to the These three arguments are weak because they are based

9
primarily on the lack of evidence; but as arguments against inscription in its original Noldorin form, Tolkien added
hypothetical grammatical particles they are perhaps linguistic depth to the history of Middle-earth. We know
sufficient. that the dialect We know that the dialect of Rivendell was
The replacement of -Ito, the third person plural marker influenced by Quenya at least to the extent of borrowing
in an early version of Quenya (LT 270), with -nte in later one personal pronoun (R 65). Perhaps the form of the verb
versions is a shift toward the -CCe pattern, and may onen, incorporating the subject pronoun, is in the
represent a loss of a gender marker -0. In the earlier Rivendell Quenya. Although we might expect this same
Quenya, toi means 'they' (LR 72), and -ta is used once for dialect to be spoken in Eregion (since Rivendell was
'it (object form)' (LR 72). originally an outpost of Eregion [UT 238], perhaps the
These arguments apply only to subject markers in verbs West-gate inscription is ina more formal (and therefore
and to the possessive pronoun markers, where analogical slightly archaic), or in a less dialectal form of Sindarin.
levelling is more likely. It would not be surprising to find That Sindarin and Noldorin have different personal
distinct forms for different genders in isolated pronouns pronouns can be seen by comparing the Sindarin pronoun e
(and this could be the explanation of the comment about 'he' (discussed in more detail below) to the Noldorin ho 'he'
the Sun being referred to as She), but the one example we (LR 385). The Sindarin pronoun shows no vestige of the
have, the word te 'them', is perfectly consistent with the original *o/*e '(masculine)/(feminine), distinction which is
genderless ending -nte, and would if anything be so clear in Noldorin. Perhaps Sindarin, as well as Quenya,
masculine (referring to Frodo and Sam); although it has the has eliminated the masculine/feminine distinction in its
form we would expect of a feminine pronoun. pronouns.
The Noldorin pronoun hein 'them' (LR 385) can be seen The argument above might be taken to justify the
in the inscription of West Gate of Moria: assumption that the language of the West-gate inscription
was exactly the Noldorin of the Etymologies, but only if
1m Narvi hain echant the distinction was ignored between what was published
'I, Narvi, made them' (I 318-9) and what remained unpublished. In Tolkien's perspective
the material, once published, was for good and all a
The shift from ei to ai is known to have occurred in the reflection of the linguistic history of MIddle-earth,
Third Age Sindarin, and is not reflected in examples of although the full complexity of history (including
Noldorin from the Etymologies, so we would expect to sweeping political issues and minor scribal errors) still
find hain in place of hein in later Sindarin. allowed him to make some changes in the history behind
The West-gate inscription is, however, almost exactly the the evidence. The unpublished material was a source book
Noldorin which is found in the Etymologies. THis can be from which the published material could be derived. The
seen primarily in the regular Noldorin past tense echant, unpublished fact that Noldorin made a distinction of gender
which does not specify the person, and so must be among third person pronouns does not mean that the
accompanied by the personal pronoun im 'I'. The past language of the inscription did so, or even that any Eldarin
tense seen in onen 'I gave' (which incorporates the first language ever did so.
person pronoun) is nowhere seen in the Etymologies; and
most likely is Sindarin, the language which sprang
Noldorin during the development in Tolkien's conception The Second Person Plural Pronoun
of the linguistic history of Middle-earth. Note also that the
vocabulary of the West-gate inscription was taken from or There is more evidence bearing on the question of the
(as in the case of echant) added to the Etymologies. second person plural. In The Father Christmas Letters, [in
The restructuring of Noldorin into Sindarin took place the "Appendix"] there appears the following fragment of
very late in the writing of The Lord of the Rings. In an "Arctic":
early draft of the conversation between Faramir. Frodo, and
Sam in Henneth Annun, Farmamir says: "... all our names an ni vela tye ento 'until I see you next'
of town and field, hill and river are in ... the High-elven
tongue of the Noldor". This was written in 1944 (L 79), We now know (based on examinations of Tolkien's
more than five years after there is still no evidence of the manuscripts) that Tolkien's concept of Quenya went
Sindarin language -- lingua franca of the Elves in the First through a period in which the word T could be represented
Age _. specifically 1lQ1 the High-elven tongue of the Noldor by ni, at least when it immediately preceeded the verb (in
-- but still source for most geographical names in the West one case it is attached as a prefix). The word tye we know
of Middle-earth. as an object form of the second person familiar pronoun:
It is clear that Tolkien intended the West-gate inscription
to be Sindarin (III 398n, 400); and yet its style differs from tye-melane 'I love you' (LR 61)
that of other fragments of Sindarin. By leaving this

10
 This is from a different period in the development of the SUBJECT OBJECf MEANING
language, and so the object preceeds the verb while the -nye ni I/me
subject is attached as a suffix, but the gloss clearly *-tye tye You/you (familiar)
identifies tye as the object form of 'you'. -lye Ie You/you (reverential)
If we accept that Arctic is in fact Quenya, then the word -rye *se, -s She (he? it?)/it
tye in the first example refers to all of the children to
whom the letter is addressed, while in the ,second case it -Ime me We (exclusive)/us
refers to a single person. Here is a suggestion that Quenya -Ive *ve We (inclusive)/us
makes no distinction between singular and plural in the -lIe Ie You/you (plural)
-nte te, -t They/them
second person.
One more piece of evidence can be brought forth to
bolster this claim. In the unpublised epilogue to LotR, The subject ending -lIe, as with all of the other plural
the phrase genediad Drannail 'the Shire reckoning' subject endings, consists of a consonant followed by the
appears. The word Drann for 'the Shire' also appears and object pronoun.
has been adequately analyzed as Dor-ann 'Land of Gift'. It has already been mentioned that Tolkien used -Ito at
In fact, the Shire was a gift from the Dunedain to the one point rather than -nte for the third person plural
Hobbits; though it is amusing that the Dunedain in their pronoun suffix. Notice that if -nte is replaced by the
decline should liken themselves to the Valar and the Shire earlier -Ito in the table above, then all of the plural subject
endings consist of the sound I followed by (in the case of -
to Numenor.
It would be nice to analyzed Drannail as an inflection of Ito what might be) the object pronoun.
the form Drann. The gloss 'the Shire reckoning' suggests In the manuscripts, the form omentielmo replaced an
a genitive for the second component, but we know that the even earlier form **omentiemman. The ending **-n is
simple genitive was expressed only by word order in an earlier form of the genitive in Quenya, and **-mma
Sindarin. We will have to assume that the translation is must be 'our' corresponding to **-mme 'we'. At this point
not literal. in the development of the language, the series of endings
Since -I in Sindarin corresponds to -I in Quenya, one **-mme, -lIe, and -nte (if indeed they ever existed
possibility is that Dr ann a it shows an ending alongside each other at the same time) could have been
corresponding to the Quenya ablative suffix -lIo 'out of. analyzed as deriving from an original -n plus the object
The exact translation 'reckoning out of the Shire' is pronoun, with assimilation of the n to the following m or
acceptable in English, but doesn't quite fit: there is I. The I actually found in -Ime could be derived by
probably no movement of the calendar from the Shire to analogy from the first I of -lIe, or it could be borrowed
other parts of the world. It is also difficult to make the from the inclusive form -Ive where it actually represents
Sindarin ending -ail correspond phonologically to the the I in Ie, or it could even be the natural outcome of n
Quenya suffix -110. when it occurs befor_e ~.
The"Sindarin sound ai (from an earlier ei) only arises in Even if the form -lIe in the early version fo Galadriel's
environments where there is a ~ or an i in the following Lament is just an early version of the second person
syllable. singular, it can be seen from the pattern above that it fits
A more interesting possibility is that genediad more naturally withthe plural pronouns than with the
Drannail translates literally as 'your Shire reckoning' singular ones.
where Drannail shows a suffixed possessive pronoun like The situation observed would arise if the earliest Eldarin
the one in lam men 'my tongue' (S 363). The suffix -ail languages distinguished between a second person singular
could easily be cognate with the Quenya form *-Iya 'your'. (which later developed into Quenya tye ), and a plural
In the case of Drannail, however, the suffix must be (which yielded Ie); and at some time before the separation
plural, since it is the reckoning of the Hobbits as a whole, of the dialects which would later become Sindarin and
and not just of Samwise (to whom the sentence is Quenya, the singular/plural distinction came to be regarded
addressed). as a distinction between familiar and reverential. Formality
An early version of Galadriel's lament has the form -lIe is generally more appropriate when addressing groups than
in those places where the published version has -lye. This individuals; and while individuals have a strong connection
is either a change in the form of the second person to the physical world, what is common to a group is their
pronoun, or it is a different pronoun which, since we have
connectin to God, which is to say their divinity ("For
all of the other endings, would be the second person plural.
where two or three are gathered together in my name, there
Either way, it strengthens the present argument.
Note first how well the ending -lie fits with the patterns am I in the midst of them." -- Matt 18:20).
already set out, assuming that it represents the second In Quenya, then, the second person pronoun elle (subject
person plrual: ending -lIe) was analogically re-formed to elye (-lye) to
more closely match the other singular pronouns. It may be

11
that the form elle continued to be used for reverential that Quenya makes no distinction between dual and plural
plural, thus re-establishing the singular-plural distinction in its verbal subject (and possessive pronoun) markers; but
in Quenya for this one pronoun. that *-lve is a remnant of an ill.d. dual inflection,
The assumption that the Eldarin languages make no reinterpreted as a marker of the inclusive first person (a
distinction between singular and plural in the second distinction which was always most important in the dual).
person, while far from proven, has been shown here to be a Based on this assumption, we can construct the object
possibility; creates no difficulties in interpreting the pronouns *vet 'you and me' and *ve 'you, me, and
corpus; and provides us with a complete set of subject someone else'.
endings for Quenya. This assumption will be adopted for
the remainder of this article. Possessive Pronouns in Sindarin

The First Person Plural Turning for a moment to Sindarin pronouns, we have the
following possessive forms. The second and third examples
On the distinction between -lme and -lye, we have are taken from an unpublished letter from Aragorn to Sam:
statements by Richard Plotz and by Humphrey Carpenter:
e-mbar nin '(of) my home' CUT 54)
"... omentielmo, which means, literally, '... of mellyn in 'his friends'
our (my, his, her, NOT your) meeting.' Tolkien, on bess din 'his wife'
reflection, changed this to omentielvo, 'of our (my,
your, possibly his, her) meeting." (ItE 20) It is curious that there are two distinct forms in the same
letter which are translated 'his'. One possibility is that in
". . . Quenya makes a distinction in its dual is a special form of the word din which occurs following
inflexion, which turns on the number of persons words ending in the sound n. Such a situation could arise if
involved; ... omentielvo is the correct form .. ." the two words were perceived as a single unit (that is,
(L 447) mellyndin) with an internal sequence -nd-. The sequence
nd usually changed to nn in Sindarin (so that we would
These two statements are difficult to reconcile and neither have mellynnin) which might be interpreted as the loss
is authoritative. The second has a bit less credibility of the d.
because it seems to make no sense: if a dual inflection does It is, however, unlikely that Sindarin speakers would
not involve two people then it is not a dual inflection. confuse the sound n with the sound nn (a much more
Perhaps what Carpenter is trying to say is that Quenya important distinction in Sindarin than in English), which
makes a distinction in having a dual inflection (and such means that we might expect mellyn nin instead of
an inflection certainly depends on the number of people mellyn in; but the former certainly means 'my friends'.
involved). Plotz, on the other hand, implies that the -lvo The slightly surprising alternate form in can be expl~ned
form can refer to more than just two people; but perhaps as a dissimilation from the less surprising and less
what Tolkien told him was only that the form included the acceptible nino
first and second persons (consistent with its being a dual), Another possibility is that in means 'his' and din means
and that 'possibly his, her' in the citation above is an 'your'. In fact, the one time in occurs, it refers to Aragom
assumption on Plotz' part. (who is sending the letter), and every occurance of din
Thus, we can explain the two statements by assuming refers to Samwise (the recipient). Setting out the pronouns
that -lvo represents a dual pronoun meaning specifically under this assumption results in an intriguing pattern:
'you and me'; with Carpenter explaining the fact poorly,
and Plotz assuming that -lvo can refer to a third person as SUBJECf OBJECf POSSESSIVE QUENYA EQUIVALENT
well. The v in -Ivo, if it is derived from earlier *w, is im nin nin ninya 'my'
reminiscent of the older dual ending *·u. din Ie 'you (object form),
There is, however, a synchronic process in Quenya for e in ? 'he'
forming duals, at least in isolated pronouns, as can be seen
from the form met 'us two (not you)' (R 59) vs. me 'us
Note the similarity of the pronouns e 'he' and in 'his'
(not you)' (LR 56). Thus, an early specialized form
under this assumption. The Common Eldarin sound *d-,
meaning 'you two' would most likely give way in Quenya
to the newly constructed inclusive and exclusive 'we' would when it occurred initially, changed to I- in Quenya,
still need to be expressed by the pronoun itself. allowing Sindarin din to be related to Quenya Ie.
For this reason, and also (sontewhat irrationally) in order The evidence of Drannail 'your Shire', cited earlier,
to avoid getting into the question of what the exact form argues against this reconstruction, pointing to a proto-form
of all of those dual pronoun markers would be, I have in *1-. However, the dialect of Rivendell, as discussed
assumed here that Plotz' statement is exactly correct, and earlier, has taken many features from Quenya. Perhaps -ail

12
here is an early borrowing from the possessive pronoun Sindarin are built on the object form of the pronoun; or the
suffIx of Quenya. These suffIxed possessive pronouns stand word men means 'we' and is a subject pronoun in Sindarin.
in contrast (whether of dialect, or as stylistic options The odd form anim could be explained if it were necessary
within the dialect) to explicit possessive pronouns, as to resolve an ambiguity created by a phonological change.
shown by lammen 'my tongue' (S 363) vs. e-mbar nin Suppose that the object form object form corresponding to
'(of) my home'. This dichotomy of forms can be explained din 'my' vs. nin 'me'. Then the dative form 'for you
by assuming that the former is a borrowing from Quenya. (formal)' would be andin, or rather annin since the sound
There remains the question of why both in and din are group nd shifted to nn in most positions in Sindarin. This
translated 'his'. First of all, Tolkien clearly views later form would be indistinguishable from the compound
translation not as a mechanical process which can be driven an-nin 'for me' and it would not be surprising to fInd the
by a dictionary, but rather as a process of re-creation. The subject pronoun im stepping in to take the place of nin.
use of third person throughout the letter creates a sense of Even accepting the fact that ammen contains the pronoun
formality, but to Samwise in the second. This may be the 'us', it is still possible that *men is not the form that this
appropriate way to indicate formality in Sindarin, or pronoun would take in isolation. Objects of verbs show
perhaps din is already a formal pronoun (as its cognate is grammatical lenition in Sindarin (that is, they show
reverential). Either way, it does not seem necessary to lenition of their initial consonant regardless of their
assume that Tolkien's translations are literal. position in the sentence), for example lasto beth 'listen
Note that the Sindarin form nin and the Quenya form to the words' containing the object form of peth 'words'.
ninya can be taken as exactly cognate, pointing back to a Although pronouns often follow grammatical rules of their
proto-form *ninyan 'my'. This proto-form could be own (rules from an earlier stage of the language), it is quite
analyzed as containing *ni-ni- and showing a unlikely that the word for 'us' is *ven when it is not part
reduplication of the root NIT, but there is no precedent for of a compound. On the analogy of hain (which, by the
such a process within the Eldarin languages. A more likely way, probably shows a lenited s initially) 'them', we might
explanation is that we have *ni-n-yan 'which-me' (where even expect to find *vain. Such a form might even come
the n is perhaps an object marker), that is, 'which is for about purely from analogy, based on the fact that it looks
me'. It is likely, then, that the -nya in Quenya ninya more like a plural. For the moment, I will make the less
'my' is not the same in origin as the noun ending -nya radical assumption that the proper form is *ven.
'my'.
Putting these assumptions together, it is not Other Sindarin Pronouns
unreasonable to construct a proto-form *di-n-yan from
which the isolated Quenya form *linya 'thy' can be It is possible that there is another pair of pronouns in
derived. Aragorn's letter. The phrase i sennui Panthael
estathar aen is translated as 'who should rather be called
The Sindarin Datives Fullwise'. The word -i is 'who' (Quenya i 'who' -"UT
317), Panthael is 'Fullwise' (Quenya *quant- 'full' - R
Datives in Sindarin are constructed from compounds 59 - Sindarin Perhael 'Samwise, i.e., half-wise' - L
which include the preposition an 'for' followed by a 308 - the form Berhael shows grammatical lenition
personal pronoun: since it is the object of the imperative verb elgerio
'glorify!'); and estathar looks like the simple plural future
anim for me (ill 342) inflection of a verb esto 'to name' which would be cognate
am men for us (I 320; [The Return of the Shadow with Quenya esta 'to name' (LR 356). Thus, the whole
463]) sentence might be more literally translated as 'whom they
will properly name Fullwise'.
The forms anim 'for me' and ammen 'for us' both There remains the curious word aen. This replaced an
contain the preposition an 'for' in close combination with earlier word ge. It has been proposed that ge is an object
a personal pronoun. In the case of anim, we know that the pronoun cognate with Quenya tye 'you', and although this
pronoun im is the subject pronoun '1'. The compound proposal has met with some resistance, it now appears to
ammen can be divided an-men, but the form men looks be the best interpretation. The proto-form *kye would
more like an object pronoun (nin 'me'; hain 'them') than yield Quenya tye, and would give ce in Sindarin (cf.
like a subject pronoun (im '1'; e 'he') and in fact, could be Quenya tyelpe 'silver', Sindarin celeb); and this may, in
exactly cognate with Quenya me 'us'. In Quenya it is the fact be the subject form of the pronoun 'you'. Objects of
object pronoun on which the dative form is constructed the verb in Sindarin show grammatical lenition (as
(nin 'for me' from ni 'me'). . discussed earlier), so that ce in object position would
We are left with two alternatives: either the Sindarin form appear as ge.
anim 'for me' is unusual, and most dative forms in The word aen might be the word for 'him'. This

13
assumption is based on the vacillation between the second comment on it in advance of its availability to the rest
and third person which we have assumed with regard to of Parma's readers as my proposed Case system for
Samwise; but aen fits pretty well in the pronoun chart: Nold.lSind. pronouns is also appearing this issue and
they bear on one another.
SUBJECT OBJECT POSSESSIVE TRANSLATION
im nin nin lime Looks like Noldorin,
*ce *ge you/you Reads like Noldorin,
e *aen in he/him Sounds like Noldorin . . . .. must be Sindarin.

The fact that ge does not have a final ·n can be attributed In fact, the Moria Gate inscription is all Noldorin, and
to the fact that its function as an object is adequately further, it ought to be. Bill's assertion that Tolkien
marked by the lenition which is not shown by any of the intended the West-gate inscription to be Sindarin is
other object pronouns. supported by text references that do not support it.
The Sindarin clause (i sennui Panthael estathar Finally, the key point of his asertion is based on a
aen) does not really need an object, since the relative misreading of the Noldorin cases (as is Gandalf's).
pronoun i already serves that function; but it is possible to Ought to be Nold. The inscription was written by a
shoehorn one into the English translation ("he whom they Noldorin prince of the Feanorean line, a mere (for an
will properly name Fullwise"), so perhaps its presence in elf) 800 or so years after the First Age, making it closer
the Sindarin version is acceptable. to 1st than the 3rd Age. Culturally and politically we
would expect Nold. Text referenced does not say
this is Sind.; it says: "The words are in the elven-
Conclusion tongue of the West of Middle-earth in the Elder
Days", which if we accept Etymologies, logically
By stretching the available evidence, we are now in a means Nold. The reference to the appendix only
position to propose a complete set of personal pronoun says that the inscription "illustrates a MODE of 'full
affixes for Quenya, and to suggest what most of the writing' .... All the vocalic LETTERS of Sind. are
isolated Sindarin pronouns are. shown." Mode and lang. are two separate things; you
I began the exercise of laying out all of the certain fonns can write English in any number of Elven modes and
and as many hypothetical fonns (of less than thorough this does not change English into Elvish, nor does it
implausibility) as were needed to fill in the empty spaces change Nold. into Sind. Neither does the second
in between, but the interested reader will no doubt have part of the statement make it Sind. just because the
begun to do just this already. As an example, let me vocalic letters of Sind. are present (especially if the
suggest the fonn dn for 'your' in Sindarin, although this vocalic letters of Nold. and Sind. correspond). Bill's
is at least as dubious as ge 'you'. mis-read of Nold. cases is shown in my proposed case
It is clear that we have only begun to scratch the surface system, Nold. for 'them' is hin, hein is 'their', the plural
of the volume of material which has become available to possessive. And this bit of information is of
us in the last few years. With the exception of a few very STUPEFYING SIGNIFICANCE to every one who for
basic words, the vocabularies of Sindarin and Quenya are the last twenty years has accepted this as the largest
now to the point where they can used in the same ways single example of Sind. in the corpus. 1
that real languages are used. Hopefully, this article has
begun the process of reconstructing these remaining words, 1 Stupefying is not the adjective I would choose here.
the final keys to a fluent speaker's understanding of the There is further evidence to support the contention that
Eldarin tongues. Tolkien thought of the language of the inscription as
Sindarin. This is implied indirectly when in Appendix F, after
explaining what Eldarin means and that "in this history ...
all the Elvish names and words are of Eldarin form", he
begins the next paragraph: "Of the Eldarin tongues two are
Comments found in thjs book: the High-elven or Quenya, and the
by Grey-elven or Sindarin." I admit this does not explicitly
Tom Loback deny that more than two Eldarin tongues are found in
LotR, but if that were the case, and one of the longest
samples of Eldarin were actually neither Quenya nor
Bill Welden kindly sent me a preview copy of his Sindarin, this would be a rather peculiar turn of phrase. And
survey of Eldarin pronouns, which he has submitted there is more evidence.
to Parma. I am taking this as an opportunity to In Letter #347 after stating that the "Ienitions or

14
 In my case system I suggest that hain might mean Doriathrin and, according to Christopher Tolkien,
'those' because hin is translated as 'these' and the written in 1951!, after LotR and the so-called
grammatic'ally correct usages are the singular 'this' and restructu ring.
'that', plural 'these' and 'those' cases. I reasoned that Curiously, as a believer in the Gnom.lNold.lSind.
Gandalf makes a common mistake in the usage, whole, I find myself refuting a separatist view by
interpreting it as 'them'. Bill inadvertantly gives the further demonstrating the separation between Nold.
perfect explanation for this unlikely slip by Tolkien - and Sind. The conclusion I find myself reaching is
that it is no slip by Talk. at all; but one by Gandalf, who that there is going to be less and less Sind. in the
not being completely familiar with Nold. cases, corpus, and more and more dialects that are: Nold.,
translates the case as the Sind. pronoun for Ilk., and Dar.
'them'.2 The significance is that Talk. did this One more paint, Bill says that Rivendell was
intentially to point out the differences between originally an outpost of Eregion, which is also
dialects in the language GROUP called Sindarin in the misleading, especially in terms of the dialect to be
3rd Age. Bill is right, but for the wrong reasons and expected there. It is founded by Elrond, coming from
wrong for the right reasons. I believe this shows that Lindon and joined by Celeborn (presumably leading a
Tolkien intended that the whole of his language Sindar faction) from Eregion. the leadership of these
corpus (excepting, perhaps, parts struck through in two Doriathrin Elves, plus the mixed nature of
Etymologies and elsewhere) is one whole, Lindon's population, would suggest a stronger
developing in a natural way. And that it is probably Dor.lllk.lFal. element in this dialect than that found in
wrong to try to force the languages into Eregion; considering the pOlitics and events of the
compartmentalized eras in his writings, at least in the 1st Age.
case of Gnomish/Noldorin/Sindarin/etc. More to the point, I think, is that Gandalf's error
I offer one more example of the retention of the shows the very close relationship between Nold. and
dialects from Etymologies to refute another of Bill's 3rd Age Sind. in that they both use a pronoun case
arguments: that of his claim that "The restructuring of system in which some of the forms are the same, if not
Nold. into Sind. took place very late in the writing ... " their meaning (e.g. N HAIN = 'those', SHAIN = 'them').
The place name Androth, dwelling place of non-Nold. I think a fairly complete case system can be made for
Elves of Mithrim (UT p.19 and probably meaning 'long Nold. and I wouldn't be surprized if the case system
cave', implying the continued, though underground for Sind. tums out to be Dor. or Ilk. Further, I think the
existence of the river from Lake Mithrim to Drengist, much disputed Luthien fragment from Lays of
shown on the earliest Silm. map), which does not 8eleriand ought to be looked at from a dialectal point
appear in Etym., is neither Nold. or Sind. in form, but of view in the context of its place in the history of M.e.
to see what lang. the majority of its words and forms
'mutations' of Sfindarinl were deliberately devised to
belong to.3
resemble those of W[elsh]" though with differences, he
goes on to discuss certain examples illustrating those
differences, including: "ost-giliath 'fortress of stars' in
which the second noun functions as an uninflected genitive
shows no mutation. Cf. ennyn Durin." Now since this last
example has an uninflected genitive, 'doors of Durin', it
makes perfect sense in context itit is Sindarin. But if 3 Having rather intrusively debated certain of Tom's
Noldorin has survived as a separate tongue with a different points, it is only fair to admit that I wholeheartedly agree
grammar from S., then even if they happen to share with his concept of Sindarin as a 2nd and 3rd Age Quenya
constructions that do show lenition, it would make little term for a GROUP of mutually intelligible dialects, so called
sense to cite Noldorin to explain the rules for nonlenition in because most of the speakers who talked to each other
Sindarin. - Editor using it were Sindar 'Grey-elves' in an ethnic sense. "The
Exiles, dwelling among the more numerous Grey-elves,had
2 Looks like good English
adopted the Sindarin for daily use; and hence it was the
Sounds like good English
tongue of all those Elves and Elf-lords that appear in this
Understandable as good English ...
history." (App. F) To me this images native Noldorin
... must be a common mistake.
speakers choosing to speak Doriathrin, but does not
What I do not understand about· this convoluted exclude their importation or reinvention of many nonnative
interpretation of Tom's is, if I made them is the wrong way words. We are talking about Noldor after all. So I find an
to use the pronoun them, then when would it be correct to "Eregionian" dialect of Sindarin with a heavy admixture of
use this English pronoun? - Editor Noldorin forms quite plausible. - Ed.

15
 acceptable in light of the Elvish preference for sixes
and twelves, but it does make one wonder about the
morphological structure of kainen and whether
Problematic Numbers in Elvish Tolkien continued to construct his Elvish numbers in
a similar way in later years. It is unfortunate that
by Arden R. Smith the word "twenty-third" in Elessar's letter to
Samwise in the unpublished epilogue to The Lord of
the Rings appears in th~ final clause of the letter,
The numbers one through ten are very often among
which does not appear in the Sindarin text.
the first words that a person learns when beginning
Another interesting number, however, appears in
the study of a foreign language. Until the
both the English and Sindarin texts of this letter,
publication of The Lost Road, however, the scholar
namely nelchaenen 'thirty-first'. Chris Gilson in-
of Elvish would have been unable to recite a
formed me that this is the correct form of the word
complete list of these ten basic words in either
as it actually appears in the manuscript, rather
Quenya or Sindarin. Unfortunately, an extreme
than Jim Allan's problematic reading, nechaenen.
paucity of data still remains when one progresses
This word can be broken down as follows. Nel-
beyond the first twelve cardinal numbers. All we
ovbiously means 'three' (L R, p. 376, and
have between the numbers 12 and 1000 are the
elsewhere). The -chaen- in the middle of the word
cardinal "23" and the ordinal "thirty-first", and
seems to be a mutated form of *caen 'ten', derived
the curious forms of these two words seem to create
form KAYAN- (d. kainen) rather than from
more questions than they answer about the
KAYAR· (whence Noldorin caer, see LR, p. 363).
formation of Elvish number words above twelve.
The spirantization of the c can be explained by the
The first of these two words, leminkainen '21',
same rule that changed Telerin alp to Sindarin
appears in an entry from the Qenya Lexicon headed
by lemin 'five' and also including lempe 'ten' alph, that stops turned to spirants after 1 and r (UT,
p.265). It could also be a mutation required by the
(BLT1, p. 246). Christopher Tolkien writes, "The
grammatical relationship between nel and caen,
choice of '23' suggests that this was my father's age
which combine to form the word for 'thirty'. The
at the time" (ibid.), but since this word bears an
uncanny resemblance to the name of a "reckless final -en is evidently an ordinal suffix, d. dolothen
'eighth' and edwen 'second' in the same letter.
adventurer" (as W. F. Kirby's glossary describes
This, however, presents a problem, since this
him) in the Kalevala(1907), i.e. Lemminkainen, it
would give nelcha:enen the meaning 'thirtieth',
also seems likely that Tolkien gave the word for '23'
(his age) this particular form in order to create some rather than 'thirty-first'. I suggest this solution:
sort of connection between the Finnish hero and nelchaenen is an error. Since nelchaenen does not
appear in the tengwar text of the letter but only in
himself. After all, Tolkien (from German tollkiihn
the transcription into Latin letters, we only have
'foolhardy') would be a fitting name for a "reckless
one reading of the word available to us. That one
adventurer." reading, unpublished and therefore possibly
But why is the word for '23' related to the word for uncorrected, could possibly be in error. Since
'five'? The second element of the word, kainen, is nelchaenen is translated as 'thirty-first', I believe
the later Quenya word for 'ten' (LR, p. 363). It that 'one and thirtieth' was intended, *min a
cannot have that meaning here. Even though it nelchaenen or some variant thereof, from which
would be possible for the language to have two Tolkien inadvertently omitted the *min a. Granted,
words for 'ten', i.e. kainen alongside the afore- my theory may not be correct, and I would
mentioned lempe, the word leminkainen with appreciate other solutions to the problem.*
kainen meaning 'ten' could only mean 'fifteen' (5 + However, we cannot be sure how numbers larger than
10) or 'fifty' (5 x 10). Since lemin is definitely 'five', twelve were constructed in Elvish until more of
23 can only be broken down as either (5 + 18) or (5 x Tolkien's own words on the subject become available.
4.6). Since five does not go into twenty-three evenly, * Possibly there is a grammatical difference in the use of the
kainen in this case can only mean 'eighteen' (which ordinals, not reflected in their translations, whereby nelchaenen
Bill Welden noted at Mythcon XIX). This is quite is '30th (after)' rather than '31 st (of)' - Editor

16
 HOST {40,000 to 50,000 strong}
_ _ 1 _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _

1 1
ARMY {20,000 to 30,000 strong} ARMY
a=G@lYa=G ~ a=G@lYa=G ooa~ _1 _ _ _- - - : - - - -
1 1 1
a=G@lYa=Goo@OO LEGION {1 O,OOO} LEGION AUXILIARIES
-1--------------- __ _
1

Ore Military Organization COMPANIES {50 to 1000 strong}

and Language From the text (5. p.116) it can be inferred that a HOST is
commanded by a servant of Morgoth more powerful than an Orc,
By Tom Loback after the Dagor Aglareb. Salrogs, sorcerers, and dragons
supplant the Orc, who are relegated to captains of lesser
commands. Estimated strengths on the T.O. are derived from
comparison statements of battle strengths made in the texts
The grand overview of Christopher Tolkien is a and a demographic study of Elven population in Myth/ore 51
monumental project that mostly enlights and delights. (p.34).
However, it cannot cover every aspect in detail, The expectation is that much of this postulation can be
even in ten volumes. Those readers who crave the supported by information from various places within Tolkien's
sustaining and extension of the secondary world, mythology. The many Elven dictionaries and etymologies reveal
should choose an aspect and explore it in context of much, once it is established what is sought. Now the T.O. can be
, the whole conception and throughout the entire drawn with secondary world terminology that supports it.
corpus. But, any aspect thus studied will lead to the
hope and expectation that its framework can be HOTH(HOST)
filled in. Often the student is disappointed to come to
seeming dead ends. Other times, the expectation --------1------------ 1
that something ought to be there, and the search for HOTHRI (ARMY) hothri 'captain' HOTHRI
it, yields surprizing results. 1 ? Othrod 'Orc-Iord '
_1--_--_ _ _- - -
1 1
A continuing study of the military history of the HOSTAR (?LEGION) or HAl HOSTAR
First Age of Middle-earth has led to theories on _I ________________ _
demographics, social organization, geography and
language use. While trying to establish Orc military 1

HAl (?COMPANY) or HOST (GROSS 144)

organization, a framework was postulated that had many gaps in 1--_ _ _ _- __ _
it when it was first applied to the secondary world. The
1
expectation that some of these gaps might be filled, led to the
DOG (SOLDIER) or SNAGA (?SLAVE)
search for expected, specific things, terms and definitions;
often in quite unexpected and unlik~ly places. Though lacking in
Comparing estimated strengths of a F.A. Orc-host as a
complete consistency, these findings do make sense and reveal
standard military formation with Sauron's force before the
an exciting glimpse of the depth of J.R.R. Tolkien's subcreation.
gates of Mordor at the end of the Third Age is revealing. The
Men of the West have some 6,000 troops. Sauron's force is
Orc military organization is here posited as being similar in described as "ten times and more than ten times their number".
structure to that of the Republican Roman military organization If the ARMY of the Easterlings is about 20,000 and the HOST
of our own world. The Table of Organization [T.O.] for Orcs, of Mordor 40,000 to 50,000 that gives a total of 60,000 to
using the English terms employed by Tolkien, looks like this: 70,000, certainly fitting the Uten times and more than ten
times·. From this it seems as there might be little overall

17
change to Orc military organization since the First Age. Orcs. Certainly the RUMHOTH ('Romans' LT II p.294) and
GLAMHOTH ('Orc-hosr, S. p.360) suggests a general parallel
The definitions of the Elven terminology are also intriguing. between the Romans and the Orcs on some level. The Republican
Roman Consular army was made up of two pairs of legions, each
HOTH 'host' (S. p.360), 'horde' (LR p.364) (LT I pair containing one Roman and one Allied legion. The Allied
p.264) While it did not originally mean 'host', its evolement to legion was composed of soldiers drawn from a TRIBE of the
'host' corresponds with the idea of an evolving use of the word Italians allied to Rome. The strength of the Consular army was
host to describe Orc forces larger than armies. usually 18,000 to 20,000, which compares favorably to the
strength of an Orc army derived from the comparative
HOTHRI 'army' (LT I p.264) This is an early indication estimates based on K.P.H.E. (Myth/ore 51 p. 34). While it may
that there is a distinction between a HOST and an ARMY. seem unfair to step outside Middle-earth to analyze an aspect
Although hoth is not yet defined as 'host', the ·rl ending seems of it, an exception is made here due to Tolkien's own use of
to indicate that this word is a subordinate, or diminutive of Rumhoth. Further it should not be taken that the Roman model
hath. will hold up in all detail.

HOTHRON 'captain' (LT I p.264) Contains hoth + r + on HAl This Black Speech term is usually defined as
(on from kano S. p.360) and is also clearly a military rank, unlike 'folk', but it may also derive from GWETH/GWAITH
Othrod. (LR p. 398) through corruption and C to G to H to O. It is used
both as Uruk·hal and Urukhal and is most often used when
OTHROD a lord of the Orcs (LT I p.181) Contains the referring to groups of large warrior Orcs. This suggests a
elements: oth (from hoth) + r (perhaps) + rod ('lord'). This particular formation such as troop, company, regiment regiment
could be taken as a political title or as a military rank. If or legion. Uruk·hai might be defined so: UR (,great', LR 396)
political, it could indicate a high chieftain of a group of closely UK ('all' from DURBATULUK I p.267) HAl ('regiment,
related Orc-tribes making up a host or an army. Under the company, legion') or as 'Ali-tail-folk'. 'Company' would seem its
proposed T.O. this is used as a questionable military rank, as the likeliest definition and that would appear to be the standard
lord of an Orc-host would be a creature of greater will than an subdivision of the LEGION. COMPANIES range in size from a
Orc. As a rank it might be something nominally superior to or probable minimum of 50 to a speculative maximum of 1000
about equal to a captain. warriors.

The above terms are not the Black Speech, but rather the HOST 'gross, 144' (LR p.364) This is relateg to
terms applied to the Orcs by their enemies. From The Lord of HOSTA· 'to collect, gather, assemble' (MC p.233) and the
the Rings (RotK B.P. p.248), it is known that in the forces of stem KHOTH·. They would appear to be the roots of most of
Mordor, the Orcs were identified by numbers. Previous the above and show much tinkering by Tolkien throughout the
evidence suggested that the Ord T.O. remained generally corpus.
unchanged after Morgoth established it. Assuming this is the
case for Orcs all over M.e., something of interest appears in the DOG 'warrior' (LR p.375) This would appear to be a
Black Speech of the Misty Mountains - the name Azog (H. generally used term for 'soldier' or 'warrior' in some form, in all
p.33). Azog is the commander of the Orcs in the Battle of dialects.
Azanulbizar. As such (if the Orcs are also numbered) he would
be the first soldier, number one Orc, so to speak. As this is a SNAGA 'slave' (?SNAR 'tie', LR p. 387, perhaps
dialecct, a case can be made that Azog is the corruption of Ash used in the sense 'tied' or 'bound') This word has been
dog (,one soldier') following the line that Bolg =Boldog (LR defined as 'slave' or 'thrall', but may also be used by higher
pp.375,377). ranking Orc to indicate a status (or size) difference with a
lesser Orc. It could also be a universal term (it is used both by
Hostar ?'Iegion' (LT II p.340) This is defined as 'tribe' in Isengarders and Mordor Orcs, II p.54, p.181) to indicate the
QL and can possibly be the term needed for LEGION. Numerous lowest ranking soldier or to differentiate between a warrior
feferences, especially in The Lays of Be/eriand, inCicate that and a common soldier.
the LEGION is a SUb-division of an ARMY or HOST (LoB
pp.334,337,340). If so, this may mean that this military While this T.O. is useful for gaining an idea of the numerical
formation is rooted in the cultural and social structure of the strength of Orc forces in any given battie, the descriptions

18
must be read most carefully. Close inspection will show that the considering that the wars in M.e. continue over thousands of
word HOST is used to describe, at times: all the Orc forces in years and hundreds of battles with so much detail, and realism in
general; more than two armies, and an Orc force that is regards to tactics and detail. That the author kept a T.O. in
significantly greater than a single army but less than two mind, which would allow him to more easily assess what his heroes
armies. An example of the last case would be an Orc Army with and their armies were up against, at any given moment; seems a
a large contingent of auxiliary wolves and wolf-riders attached. simple and elegant solution.
Boldog's host, in its final manifestation, would fit this
description. Also, it would appear that wolves and wargs and
their riders were not part of the regular formation, but
attached to it at need. Both the term GAURHOTH (LoB 337)
and the fact that the wolves are considered one of the 5 armies
in the Battle of Five Armies support this.
The adjectives describing an ARMY or a HOST in the text
are not idly chosen, but are purposely used to modify and detail
aspects of the size description - as would be expected of
Tolkien.
It would be interesting, and most revealing of Orc culture
and social structure, to see how much could be garnered about
arc-tribes as they seem to be integral to the whole structure.
Certainly this is so in the Roman system. In fact, the word
TRIBE is from the Latin TRIBUS meaning: one of the three
bodies into which the Romans were originally divided (TRES,
TRIA = 'three'). Hence, the Roman legionary officer, Tribune.
If Morgoth made the Orcs in mockery of the Elves, perhaps
he copied the Great Awakening in which the Elven mass
awakening was soon divided into the Three Kindreds: the
Vanyar, the Noldor and the Teleri. Morgoth's incomplete
understanding of the plan of lIuvatar for his children lends
support for the speculation that Orcs might be spawned. If
Morgoth continued to raise the Orcs en masse, in tribal groups
of three, that could go a long way to explain: tribal differences
and hatreds, language divergence, and the lack of female Orcs.
If combined with the opinion that Orcs are semi-independent,
physical manifestations and extensions of the original evil
incarnate; they would then be animations of Morgoth's residual
will in M.e., needing a greater will to direct them after the exile
of Morgoth. They would be much like Aule's Dwarves are before
they are given free will by lIuvatar (see especially pp. 279,280
RotK Bal. paperback ed.) This lack of independent will or
existence would account for the lack of mercy shown defeated
Orcs and the lack of any offer of redemption form a fallen
state. An offer made to Gollum, Sauron and even Morgoth. In
Morgoth's limited view, he needed only slaves and warriors to
obtain more slaves. Why would he bother with such extraneous
niceties as family and motherhood?
However interesting and speculative this may be, it seems
Tolkien intended to leave the origins of the Orcs a matter of
contradiction and mystery. Nof so their army organization.
That appears to be established fairly early on and retained
throughout the history of M.e. It seems a logical conclusion,

19
"Never have I been so moved by any musical piece as to make me want to weep . .. "
- Steven Kingsley Fairchild
Cynthian Publications is pleased to announce the release of

THE TALE OF THE RINGS OF POWER

Text by J .R.R. Tolkien, Music by Thomas Peterson
--The Tale of the Rings of Power" poster
This 22" x 30" poster featuring beautiful, never-before-published illustrations by recognized Tolkien artist,
Patriek Wynne. was inspired by Thomas Peterson's new, critically-acclaimed musical saga The Tale of the
Rings 01 Power based onthe mythology of J.R.R. Tolkien. The illus~rations span the entire history of the Rings
of Power, from "The Forging of the One Ring" to "The Last Riding of the Keepers of the Rings."
The posters, each autographed by the composer and artist, are available for $7.99; plus $3 shipping and handling.
To order posters, or to request more information about The Tale of the Rings of Power, write to:
Cynthian Publications
P.O. Box 8004
Grand Forks, ND 58202-8004
(Please allow 4-6 weeks for delivery.)
 An !IDx@fl(JJ]@UtfflaUffD@USf /f!JD@fl@(JJ]@!JD !iJffD&JUSf@U@ of
~[~U®a?)® ®@ail®
by
Patrick Wynne
Jan. 27, 1989
I have probably already mentioned that I wrote draft #1 of an analysis of FS
last fall (completed Oct. 2, to be exact, with 17 fun-packed pages). Basically I
will now be doing draft #2 right here before your very eyes (drum roll, please).
Enclosed for convenient reference (tho' I suspect you have made up one of these
yourself) is a transcript of FS withe lines numbered and Tolkien's translation
written beneath, plus a transcription fo the emendations made to the poem, and
a few other random bits 0' Quenya to be found knocking about in "The Lost Road".

Line 1
lIu lIuvatar en kare eldain a flrimoin
"The Father made the World for Elves and Mortals,"

lIu : 'the World'. In Ouenya the nominative and accusative cases became identical in the spoken
language and were distinguished by context and word order. Firiel's Song is in Spoken Ouenya rather than in
Sook Ouenya, as can be seen by such accusative forms in the poem as IIUi annar rather than SO forms ~Iu,
annaL Tolkien's rearrangement fo "Namarie" into "a clearer and more normal style" in The Road Goes Ever
On shows that sUbject-verb-object was the usual arrangement. However, line 1 of FS would indicate that the
order object-subj~ct-verb was also used to distinguish subject form object (admittedly, in this line context
plays an important role as well; regardless of modern-day thinking on the origin of God, the reading of lIu
lIuvatar en kare as 'The World made the Father' would seem nonsensical to a Numenorean). lIu could have
been placed after kare for a more normal S-V-O word order which would still maintain the meter, which is
iambic with seven feet per line (more about which later). But doubtless the poet placed lIu at the beginning of
the line for the wonderful word-music it then creates with the following word: lIu lIu- . .. See also the
discussion of tare in line 13. 1

1 This and other select notes through line 4 are from a 3rd draft of Patrick's circa Feb. 2 - April 15.
On the original content of this paragraph I commented to Patrick that there was still presumably a
grammatical way of saying 'The World made the Father' (even if understood as an untrue statement) and
that the "normal" S-V-O way would be the opposite of Iluvatar en kare IIu, namely Ilu en kare
Iluvatar. In other words the normal order S-V-O implies that O-V-S is ungrammatical, because subject
to the opposite (often nonsensical) interpretation. But that leaves 4 other basic sentence structures
possible, of which the choice between O-S-V and S-O-V is the issue in our particular sentence, and
seems to be made in Quenya just as it is made in English: the World the Father has made rather than the

21
 lIuvatar : 'the Father'. Or more fully The Father of All' or The Father of the Universe'.
en : Bill's theory that a 'indeed' is cognate with ea 'let it be' is very interesting. Assuming an in FS to
be cognate with a and ea, this would lend support to the supposition that it is an emphatic auxiliary verb or
emphatic particle: I. en kare 'the Father did make'. In a previous letter I mentioned the possibility that an in
FS is equivalent to an 'there, look! yonder'. If this is indeed the case, it may serve a function somewhat
different than merely providing emphasis of the 'do, did' variety. Tolkien tells us in The Etymologies that "en
yonder of time points to the future". So at first glance it may seem odd to have an used in conjunction w/a
past tense verb. Or is it odd? If en in FS is the future-pointing en 'yonder', perhaps it is a method of
indicating the perfect tense: I. en kare 'the Father has made', the purpose being to emphasize the present and
future results of lIuvatar's past act of making, i.e. God made the World for Men and Elves, and the World is now
and will continue to be for Men and Elves. This is much along the lines of the use of Sl 'now' with a pa.t. verb in
"Namarle" to form a perfect tense: sf Varda ortana 'Now Varda has uplifted'. Even if en derives from the
verb 'to be' rather than from an 'yonder', it could still perform this "perfective" function in FS. Of course,
Tolkien translates an kare as pa.t. 'made' rather than perfect 'has made', although we do see that T.'s
translations are not always excruciatingly literal.
kare : 'made'. I haven't much to add here, other than to wonder what the significance is of the
underposed dot beneath the final vowel. 2 It also crops up (or crops under) in ire and enyare. These are all

Father the World has made. C.G.

2 In Fernand Masse's Handbook of Middle English this abbreviatory symbol in particular has two
uses: (1) to indicate in grammatical discussion the phonetic pronunciation of "close e in contrast to
open e", which is symbolized by an underposed hook (no sign of which occurs in Tolkien) and (2) in
metrical discussion to indicate [final] e "in verse, is silent in contrast to" e with 2 dots over or diaeresis
mark, (which symbol .is. known in LotR), when it is pronounced as an unstressed mid-central vowel. In
Middle English with much looser standards of grammar and spelling. with certain final vowels in the
process of decay, to different extents between different dialects, final e was often written where the
reader did not pronounce it, so that "silent e" became a spelling convention, and excrescent final e for
metrical expedience probably became a poetic convention, at least where -some dialect still sustained
them "legitimately". Compare modern English poetic pronunciation of silent e in the ending -ed. (An
interesting example of this may occur in the 3rd line of the Ring Poem, if syllable count is intended to
be regular in it.) The situation of Quenya with a spelling standard at least (or at least tentative ones at
various periods), would be different from Middle English, but probably Tolkien's usage of underposed dot
would be analogous, as is the use of the double dots.
I count 4 instances of this underposed dot in Firiel's Song. I also count 4 lines that have an odd
number of syllables: #1 (15), #11 (17), #13 (15), #14 (15). All the rest have 14. Unfortunately the
positions match only in 3 of the 4 cases, but if one of the dots is misplaced then these numbers
themselves would corroborate the hypothesis that the underposed dot means the final e is not
pronounced. So it is remarkable that in the case of kare eldain, ire ilqa, and ire Anarinya this
"silent" e is followed by a vowel. The exception is enyare tar in the last line. But we do not need the
elision of this e for the meter to come out iambic. In the preceding line, however, which also has an extra
syllable, almost immediately above on the physical page, we have tare antava with vowel following the
final e, whose elision wou'ld perfect the metre. I dislike using textual revision to build a theory, but I
think it is the most plausible possibility in this instance. Of course the elision is optional on the part of
the songster, or else rule-determined with nonelision (or hiatus) optional, for we must account for sere

22
similar in that we have long vowel followed by an r followed by final e with the subscripted dot. Hmm ...
eldain : 'for Elves'. Dative pI. of elda 'Elf.' As if you didn't know.
a : 'and'. In all other occurrences in this poem the form ar is used. Perhaps a is used before words
beginning w/a consonant, ar before vowels (all occurrences of ar in FS are before vowels). Note A yonya
inye tye-mela 'And I too, my son, I love thee', which is consistent wlthis pattern. But in "Namarie", ar
is always followed by a word beginning wla consonant. Perhaps the alar variation was a unique feature of the
Numen6rean dialect of Quenya (and that the Numen6reans did indeed speak their own unique brand of Q. is shown
by T.'s statement that mortals as a rule failed to use the Q. duals). Cirion's Oath is consistent on this point,
but alas Aragorn, that most Numen6rean of Numen6reans, says ar Hildinyar (and he was quoting Elendil, who
was so Numen6rean he positively squeaked). See the discussion of SI in line 10.
firimoin : 'for Mortals'. Dative plural of firimo 'a mortal, a Man'. The forms
Frrlmar 'Mortals' (used in The Silm.) and Frrlmor (used in Lost Road) need not be considered mutually
exclusive, since they are slightly different in derivation, i.e. Frrlmar is a substantive use of an adjective
flrlma 'mortal', whereas Flrimor is the plural of a real, live noun frrlmo 'a mortal Man', which in turn was
derived from the adjective trrima. Both forms may have been coexistent, which one was used being
determined by whim of the individual Quenya speaker. Another example of this might be the peaceful
coexistence of two plurals for Vala 'Power, a God': "... and these are they whom we now call the Valar (or
the Vall, it matters not)." (BolT I, pg. 58)

Metrics: I am no hot ticket when it comes to metric analysis of poetry, but I

have felt compelled to give FS a try anyway. As mentioned above, FS is iambic,
w/lines of 7 feet. FS is not purely iambic, of course, T. throwing in the
occasional trochee and/or anapcest to make things interesting. And so it is
w/Line 1, which has both a trochee and an anapcest in the first and sixth feet,
respectively: (I will use a macron to indicate long vowels, and acute and grave
accents for primary and secondary stress respectively)

trochee anapa3st
, , , r , , ,
Line 1: lIu Iliulvatar I en kalre elldain a filrimoin
1 23 4 5 6 7

Line 2
ar antar6ta mannar Valion : numessier.
"and he gave it into the hands of the Lords. They are in the West."

ar : 'and'
antar6ta : 'he gave it'. The first element is the verb anta- 'give' (LR-348), used here as a past
tense, 'gave', with no change in ·form. Anta is also used as a past tense in both versions of line 6: antar,

indo without the dot and with the proper syllable count in the line. Note that at first sight this might
seem to weigh against the proposed explanation of the symbol. But if it were supposed to refer to some
phonetic difference we woald still find elision of e to be the best explanation of the meter in those lines
that have an odd number of syllables, so we would have to explain sere indo anyway. C.G.

23
antalto 'they gave'; and also in the first version of "Ollima Marklrya" (MC-221): mandulaml anta marl
Ambalar 'the East gave black shadows out of Hell' (Tolkien's translation glosses anta as 'raised'). I have
seen it theorized that anta 'gave' in FS is an example of the historical present. But why then is anta the only
present tense used historically in this poem? (Cf. kare, karler, karlelto, len de, all standard-looking past
tenses.) I suspect that anta is indeed the ordinary past tense form of anta- 'give', though I cannot explain
why we do not see **ante or **antane. Anta is not the only present-looking past tense form in the corpus:
note kautaron 'bent' (MC-216) and numetar 'went down in the West' (MC-221 )..
The middle element -ra 'he' is a masculine 3rd person singular inflection, which is also seen in antavaro
'he will give' (LR-63). It must be one and the same with the -so 'he' inflection of verbs mentioned in "The
Etymologies" (LR-385) under the demonstrative stem S-, with the usual transition of s to r in Quenya
intervocally.
The final element -ta 'it' is an object suffix which is listed in "The Etymologies" as an independent word ta
'that, it' (LR-389).

mannar : 'into the hands'. Allative pI. of rna'hand', It is interesting that the allative is
here used in the sense 'into'. Note a similar usave in both versions of the Eressean fragment: kilyanna 'in
Chasm, into Chasm'. The use fa the allative here (Le. mannar) in connection w/anta 'gave' lends support to
your idea about the substitution of the allative for the dative in certain instances.
In LR pg. 373 we see another way of saying 'into', using prepositions inflected for case (!) - mir, minna
'to the inside, into'. So evidently one could say mlr kllya or minna kilya instead of ki Iyan na. No doubt
this principle could be applied to other prepositions as well - Nyano lende nunna kaimanyal

Val ion : 'of the Lords', Genitive of Vali, one of the two plural forms of Vala
'God'.
numessier : 'They are in the West', This one really bowled me over when I first realized
what was going on. What we have here is nume or numen 'West' in the locative, numesse, to which has been
added a suffixed form of ye 'is' : n"messe + yer 'they are' > n"messler. Note the loss of final vowel in
numesse and the change from y to I in yer/-Ier. This suffixed form fa ye crops up frequently in FS and
Lost Road:

meldielto 'they are beloved' < melda 'beloved, dear' + yelto 'they are' (ye + -Ito)
marie 'it is good' < mara 'useful, fit, good' + ye 'it is'
hostainh~va 'it will be collected' < hostalna 'gathered, collected' + yeva 'it will be'
man-ie 'what is it?' < man 'who? what?' + ye 'it is'
and perhaps:

talantie 'he is fallen' < *talantea 'downfallen' (cf. atalantea 'ruinous, downfallen',
Mon/Crit pg. 223) + ye 'is' (the subject in reality being marion 'the dark one' at the end of the preceding
line).
A similar usage may occur w/na 'is', which might be one and the same Withe past participle ending in -na.

Valion numenyaron : 'of the western Lords', This was the predecessor to Valion:
n"messler, and as stated in an earlier is one of our few clear examples of an adjective made to agree wlits
noun in both number and case.

24
 Metrics: Line 2 is entirely iambic
, , , ,
ar anltarolta manlnar Vallion : I niimeslsier.
2 3 4 5 6 7

Line 3
Toi aina, mana, meldielto - enga morion :
"They are holy, blessed, and beloved: save the dark one."

Toi : 'They'. As you surely surmised by my remarks above during my discussion of antarota, I'm
swayed to your way of thinking about the semantic distinction between -Ito and ante. Surely to in tol is the
same element seen in -Ito, probably derived from TA- 'that'. The I in tal is probably a plural marker, and if
to in itself is therefore not inherently plural, I find myself wondering if the I in -Ito is a plural marker as
well, < LI- 'many' or IL- 'all', so that -Ito originally meant 'all those guys' or 'those many'? Also note the
other plural inflection -Ime 'we', which also is preceded by I. If me is inherently plural, -Ime could be
derived analogically from -Ito.
All this I-ish talk brings to mind something [else.] Isn't it true that in some European languages the 2nd-
pers. pI. form came to be used as the "polite" form (for both sing. and pI.) while the sing. form was the
familiar form? I am thinking of French vous from the Latin 2nd person plural vos, for example. Perhaps
the same thing occurred in Quenya, an original 2nd-person plural -lye becoming a "reverential" form for both
2nd-pers. sing. and pl., while the original 2nd-person sing. -tye became the familiar form.
I should also note that -mma 'our' in omentlemman 'of our meeting' (Return/Shad. pg. 324) might have
the "plural m" mentioned in The Etymologies as its first m, and this might also be the source of the n in ante
« *-mte?) On the other hand, the element te in ante may itself result from ta 'that, it' + pI. -I, *tal > teo
This would not be the first example of the Elves shoveling on extra plural markers: cf. falmalln nar. (the
declension sheet [Beyond Bree March 1989, p. 7] marks the final r in the multiple plural allative as optional;
it may be included in "Namarle" for ease of pronunciation w/following Imbe).·
This brings me to te 'them' in a lalta te, lalta te '0 praise them, praise them'. Before you made the
scales fall from my eyes, I had assumed this to be the accusative of ante. But since there are abundantly
mentioned antecedents for te in the preceding lines, te must be the accusative form of toi. The tal> te shift
is phonetically plausible, for tho' I can't think of an 01 > e example in Elvish off hand there are abundant
examples to be had elsewhere, e.g. Greek olkonomla > English economy. It may be that the acc. form of -
nte is te as well, but ante probably has as little use for an accusative form as it does for a possessive form.

aina : 'holy'. mana : 'blessed'. Since both of these adjectives modify a plural subject tal, one
would expect plural forms alnar, manar. Note in Line 6 we have Tal frimar. The answer to this enigma
seemingly lies w/meldielto. As mentioned above, this consists fo meld(a) + yelto, and note that yelto is
suffixed to the singular form of the adjective; we do nt see **meldarielto. Aina and mana are therefore
kept in the sing. to keep them in accord w/melda, their plurality being made abundantly clear by tal and -Ito
squatting on either side like bookends. We do have the rather odd situation here of tal and -Ito both occurring
in the same sentence, a double subject as it were: 'They (are) holy, blessed, and beloved are they'. Here we
can see that -Ito is acting very much like a 3r2. person pI. inflection rather than as a pronoun suffix, much in
accord w/your own theories I would say. A more normal way of constructing this sentence might be Toi
aina, mana, meldler, but meldlelto is clearly used for the extra syllable it provides, necessary to
maintain the iambic meter. If pronoun suffixes sometimes act more like inflexions, there are also instances of
inflexions acting more like pronoun suffixes: note the use of -r for 'they' in karler 'they made' and antar

25
'they gave' in Lines 5 and 6. Perhaps this was considered bad style - T. changed these forms to karielto and
antalto. Tol, besides its obvious metrical function in this line, serves to provide a subject which prevents
confusing alna and mana as adjectives used substantively : aina, mana meldielto could be mistaken for
'the holy one, the blessed one, they are beloved'. The presence of tol makes it clear this is not the case.
enga : 'save'. That is, 'except for'. Where does this derive from? Might it be related to en
'there, look! yonder', in the sense of isolating something exceptional?3
morion : '(the) dark one'. Nothing too exceptional here though! The first element is adj. more
'black' (I'm assuming more is an adj., while m6re - wllong 0 - is a noun), or its combining form morl- as
in morimalte 'blackhanded'. the ending -on '-one' must derive from ONO- 'beget', a root which begat Q
onna 'creature', Nan.
Metrics: Another entirely iambic line
, , , ,
Toi ailna, malna melldiellto - enlga molrion :
2 3 4 5 6 7

Line 4
talantie. Mardello Melko lende : marie.
"He is fallen. Melko has gone from Earth: it is good."

talantie : 'He is fallen'. One possible explanation for this form has already been given in the
discussion of numessler in line 2, q.v. It is also possible that talantie is a verb in the perfect tense, 3rd
person singular, 'has fallen', the subject being morlon 'dark one' at the end of the preceding line (for the use
of the colon in this poem to indicate the subject, see tare in Line 12).
It is well known that the perfect tense is marked by -Ie (at the same time it is important to remember that
-Ie serves a wide variety of other functions in Quenya, appearing in the past tense plural: ullier 'they
poured'; the gerundial/infinitive: enyalle 'recalling'; certain plural adjectives; laurie 'golden'; an ending. in
feminine names: Nessanle; a copula suffix: marie 'it is good'; etc.) It is not readily apparent form the
published material, however, what stem the perfect inflection is added to. This does become clear in the
unpublished material found at Marquette, where one finds the perfect form lendien 'I have come', which shows
that the perfect tense is formed from the past-tense stem (Iend-) rather than from the present-tense stem
(linn- or lest-). Also compare the present tense tulln 'I come' (short vowel) with the past tense tll Ie
'came' and perfect tense utullen 'I have come' (both with long vowels).
In the first draft of the Eressean Fragment (LR-56) we find the 3rd person singular past tense form
atalante 'down-fell'. This would yield a perfect form atalantie, which is said to be a prefix meaning
·complete" (LR-390). That this initial a- is something other than the perfect augment seen in avanier 'have
passed away' and other forms is indicated by its use in a past tense form atalante (as well as in the past
tense form ataltane in the second version of the Eressean Fragment).
Such augmentless perfects as talantle and lendien (also note from The Book of Lost Tales 1
tullelt0 4 'they have come', tulier 4 'have come', antullen 'hath returned') show that use of the augment

3 Perhaps prefix en- 'back, again, reo' (as in entul- 'come back, come again, return') added to root
3AR 'have, hold' yielded verb stem *engar- 'hold back, exclude, reserve', which was used in a stock
imperative construction like engar(a) nelde atani 'exclude three men' (similar the English use of
'save'), and separation from the verbal conjugation lead to reinterpretation of the root r as an inflection
only appropriate with plural object. Thus plural *engar > singular *enga. C.G.

26
in the perfect tense was not considered de rigueur. Since the past tense plural was also marked by -ie, this
might be seen as leading to confusion: would lendielto mean 'they came' or 'they have come'? It may be that
the importance of a clear distinction between past and perfect tenses did not loom quite so large in the Elvish
mind as it does in our own. One does not have to search far to find languages which do not distinguish tenses
which we consider basic. Japanese, for example, does not distinguish between past and perfect: Ikimaslta
can be translated as either 'went' or 'has gone'. Finnish has no separate future tense, and uses the present to
indicate that concept. In cases in Elvish where a clear pasVperfect distinction was deemed vital, no doubt the
augment would be used: elendielto 'they have come'. There were other methods of clarification available as
well, such as the use of sl 'now' and perhaps en (see the discussion of en in line 1). Indeed, the availability of
methods of pasVperfect distinction other than the augment would be vital in the case of verbs beginning with a
vowel, where there seems no convenient way to add an augment, as with ullier 'poured'. In such a case si
could be used: SI ullier '(they) now have poured'.

Mardello : 'from Earth'. Ablative of mar 'Earth', seen in Line 8 (inflectional

stem m a rd e-). In its most basic sense, mar (or mar) means 'home' or 'dwelling', and the variety of
uses to which this word is put is nothing short of astonishing. It is used to refer to everything from a ship
(Eambar), to domed buildings (oromardi), to entire countries (Eldamar), and in this poem to the Earth
itself. The more usual cosmological term for the Earth was ambar (inflexional stem ambaro-), though its
avoidance in this line is understandable since the ablative ambarollo would not fit the meter. Ambar could
have conceivably been used as the first word in Line 8 instead of I-mar, assuming that beginning that line w/a
trochee rather than an iamb would fit the artist's vision, but no doubt once mar was used instead of am bar in
Line 4 the poet withes to remain consistent.

Melko : At this point the name was probably interpreted as 'The Lusty One, He Who Lusts' < *Mailiko <
MIL-IK-. Also note Q maile 'lust', mallea 'lustful', and the similarity between the Noldorin form Maeleg to
N mael 'lust', maelui 'lustful'. These words perhaps had sexual as well as propertarian connotations
(remember Luthien before Morgoth? - "Then Morgoth looking upon her beauty conceived in his thought an evil
lust, and a design more dark than any that had yet come into his heart since he fled from Valinor.")
lende : 'has gone'. Literally, 'went'. The pa.t. of either IInna or lesta- (the latter may
have been rejected). Note the Sind. verb anglennatha 'will approach' in the Unp. Epilogue. The Q equivalent of
this would be analinnuva (ana- 'to, towards' + linna 'go' + uva). Also note elle, eller 'came' in
"Nleninque", which must derive somehow from ELED- 'go, depart, leave' (but this meaning for this root was
rejected) or LED- 'go, fare, travel'. Lende is translated as a perfect tense in this line due to its placement
between two verbs referring to the present: talantie 'he is fallen' and marie 'it is good'. Another example
of the "implied perfect".
marie : 'it is good'. As stated previously. this is derived from mara 'good' + ye 'it is'. "The
Etymologies" says that mara derives from the proto-form *magri 'useful, fit, good (of things)' < MAG-
'use, handle' Here the definition has been expanded beyond the mere utility of a physical object to include the
moral good of an abstract event, namely the ousting of Melko. No doubt the similarity to mana 'blessed'
helped in this extension of the original meaning of mara.

Metrics: another fully iambic line:

, , '\ '\ , '\

talanltie. I Mardeilio Mellko lenlde: malrie

12 3 4 5 6 7

4 Curiously, these have short vowels, all the more curious since antullen has a long vowel. [P.W.]

27
Tolkien's emendation of Mardello Melko to Melko Mardello apparently has
the result to change the third foot into a trochee:

trochee
, .. , , ,
talanltie. I Melko I Mardeilio lenlde: malrie.
12 3 4 5 6 7

I say "apparently" because I am not sure that the weak stress on the final sy"able of talantle can legitimately
be used as a metrical stress under these circumstances, namely being immediately followed by a primary
stress on the first syllable of Melka. Tolkien writes in his metrical analysis of "Namarle" in Road Goes
Ever On (pg. 61) that weak stress as in Andune, etc." I also note that trochees only occur in "Namarie" as
the first foot of a line, not internally. Anyway, if the final weak stress in talantle cannot be used, then the
emended line works out to only six feet, the second and third being anapcests:

anap. anap.
, ~, , , ,
talanltie. Mellko Mardeilio lenlde: malrie.
1 2 3 4 5 6

Obviously, the emendation was intended to provide some metrical variety (a" those nonstop iambs do get
monotonous after a while). T. may also have wanted to put this sentence into a more normal word-order, i.e.
subject first, and in the process put Melka next to tal antle, which refers to him.5

Line 5
Eldain en karier Isil, nan hildin U'r-anar.
"For Elves they made the Moon, but for Men th_e .red Sun;"

Eldain : 'For Elves'. Dative plural fa elda 'elf'.

en karier " 'they made'. Both here and with antar in line 6, plural -r performs a function
virtually identical to that of the inflection -Ito. The emended versions of lines 5 and 6 emphasize this: there
karler and antar become karlelto and antalto. This pronominal use of -r seems unique in the corpus, with
the exception of numessier 'they are in the West' (although Vallon practically acts as its subject).

5 Probably both metrical interpretations of the emended line are possible, depending on the purpose
and style of the line in context. That 13 of the 14 lines all have 14 syllables suggests that syllable
pairing or dimeter is part of the scheme and that the "trochaic" interpretation of this and line 6 is part
of the effect. In both cases it begins a new sentence with a rhetorical shift (here between Melko falling
and his being gone), so that putting stress on the final syllable in taillntie may tend to yield a pause
before the trochee, with the time made up by the more rapid pronunciation of the metrically unstressed
pair -ko Mar-. But it may be this works precisely because the alternative exists of "anaprestic"
reduction in these lines of. 3 feet into 2, if a rhetorical pause were undesirable here. This would explain
both revisions since what Tolkien's observed stricture, that secondarily stressed syllables cannot be
metrically unstressed save before a primary stress, means both original lines cannot be reinterpretted
with two anaprests vs. iamb + trochee + iamb. C. G.

28
 Isil : 'the Moon'. Here in the accusitive. "The Etymologies" gloss this as 'the Sheen' < THIL-, the
initial 1- being an "intensive prefix where I is base vowel."

nan : 'but'. "The Etymologies" give two forms, na and nan, but we have no examples of the usage of
the first form.
hildin : 'for Men'. Dative of hildi, literally 'followers'. I formerly assumed that
Hildorien consisted of Hlldor 'Men' plus -len 'land'. But Lost Road pg. 245 shows that Hlldorien was
once contemporaneous with hlldl. Therefore the first element can't be the plural Hlldor, although -ien 'land'
still works as the final element. The or which follows hard on the heels of Hild- may be ore 'a rising', which
also occurs in anarore 'sunrise'. OL defines ore as 'the dawn, Sunrise, East'. So Hild(i)-or(e)-ien is 'The
Eastern Land of Men', or if you interpret ore more generally as 'a rising', then it's 'Land of the Rising of Men'.
Certainly ore alludes to the ~rise as well, since "At the first rising of the Sun above the earth the younger
children fa the world awoke in the land of Hlldorien" (LR pg. 245), so Hlldorlen also means 'The Sunrise
Land of Men'. Here we see the Elvish love of subtle multiple meanings. The singular of hildi is hil, as shown
by its use in Tarkll, from *tara-khil.
U'r-anar : 'the red Sun'. Which element means 'red' and which 'Sun' is ambiguous, since Anar
and Ul\r both occur in the corpus as solar names. No UR- derivatives occur with the meaning 'red', but there
are piles of them meaning 'hot', 'fiery', 'blazing', etc. NAR- 'flame, fire' has color-oriented derivatives: 0
narwB 'fiery red', N narw, naru 'red' (note NarodOm 'The Red Vale' and Narosir 'The River Redway' in
Return of the Shadow). Still, given the use of Anar as 'Sun' in line 14, and considering our poet's love of
consistency, -anar here is probably intended as 'Sun', and U'r- as 'red hot' or 'fiery red'.

Meter: How do you stress a hyphenated word like U'r-anar? It makes the most sense to regard it as a
single word: O'ranar. Treating it as two words, with the main stress falling on the first syllable of anar,
raises hob with the meter. Treating hyphenated forms as a single word means that i-mar in line 8 and i-
narqellon in line 10 receive a main stress on the definite article: Imar, Inarqellon. This strikes me as a
little odd.
In line 5 we again have the juxtaposition of a weak and a main stress:_ k.s'rier I'sil. If this weak stress
can be used metrically, we have a line of seven feet:

trochee trochee
, , ,

Eldain I en kalrier IISil, I nan hilldin Or-Ianar.

1 23 4 5 6 7

If this weak stress cannot be used, that yields another line of six feet:

trochee anapeest anapeest

, , , ,
Eldain I en kalrier Iisil, nan hilldin Or-Ianar.
1 2 3 4 5 6

Emendation to line 5: En karielto eldain Isil, hildin U'r-anar. Tolkien

changed karier to karielto and omitted nan. He also juggled the word order a
bit and tidied up the meter, which is now seven iambic feet:

29
 , , ,
"
En kalriellto elldain IlsH, hill din Or-Ianar.
2 3 4 5 6 7

Line 6
Toi frimar. IIqainen antar annar lestanen
"which are beautiful. To all they gave in measure the gifts"

Toi : 'which'. Tolkien's translation indicates that tol can be used as a relative pronoun. But the
transcription of the Quenya, with a period after U'r-anar and Tol capitalized, implies that Tol irlmar is an
independent sentence rather than a subordinate clause, i.e. 'They are beautiful'. By using 'which' instead of
'they', Tolkien may have intended to make it clear that the poet is referring to the Moon and Sun as beautiful
and not to Men, Elves, or the Valar. However, the Quenya still remains ambiguous as to the referent of tol,
which may be intentional. Certainly the poet would not disagree that Men, Elves, and the Gods were indeed
irimar.
frimar : 'beautiful'. 'Are' is merely implied. Here we have a plural adjective formed with -r
rather than a vowel shift, irlma > plural irlme. Also note ralkar, plural of ralka 'bent', in the second draft
of the Eressean Fragment. Lalqall occurs as the plural of lalqa 'green' in "Earendel". You can probably
form adjectives using any of these three methods: frlma > frlme, irlmar, or irlmeli (cf. va n I mtlion <
vanlma). I wonder if you can form plural adjectives using the dual endings: (rlmat eldu 'two beautiful
Elves'? Bu the Third Age it apparently was usual to use the vowel-shift method, and -r and -II tended to be
reserved for the plural adjectives used substantively. Are there any examples of a vowel-shift plural
adjective used substantively?
In "Narqellon" there seem to be plural adjectives ending in -ai, for example mallnal pI. of mallna
'yellow', and slldal plural of silda 'gleaming'. Paul may be correct that these are accusatives, but they could
be nominatives as well, Book Quenya or archaic forms which led to the later vowel-shift forms: malinai >
maline.
I doubt umeal in lines 3 and 17 is the plural of umea 'evil', given the pastoral, if somewhat melancholy,
tone of the poem, with its description fo the Noldorin folk dancing (IIIta lie noldorlnwa) to the music of
pipes playing (rotser slmpetalla) while a fine grey rain (hlswa tlmpe) flows through the trees (sirilla
ter I-aldar). It may be a variant form of uvea 'abundant, in very great number, very large' « U B-
'abound'). Line 3, ve sangar voro umeal, may mean 'like ever-abundant throngs' (of leaves, lasser, given
in line 2). 'Like ever-evil throngs' just doesn't cut it.
In line 5 IIntulllnd(ov)a might derive from lin- 'many' and tuilindo 'swallow', plus the adjectival
suffix -va seen in turuva 'wooden' and uruva 'like fire', and so mean 'having many swallows' or 'like many
swallows': AI IIntuilindova Lasselanta 'Ah! The falling of the leaves, like many swallows .. .' Compare
this with the description of falling elm-leaves in lines 83-85 of "Kortirion among the Trees" (BoLT I, pg. 34):

And wanly borne on wings of amber pale (Oikta ramavoite malinai?)

They beat the wide airs of the fading vale
And fly like birds across the misty meres.

Getting back to frlma, r should note that it appears in Unfinished Tales as the basis for the "right name" of
Tar-Meneldur : "rlmon, 'The Beautiful One'.

30
 IIqainen : 'To all'. This cannot be the instrumental plural of the noun IIqa 'everything', which is
all-inclusive in its singular form and unlikely to have a plural. The translation 'to all' indicates a noun IIqa Ine
'every~', which stands in contrast to IIqa 'every1bl.o..g.'. It appears here in the dative. 6
antar : 'they gave'.
annar : 'the gifts'. Accusative plural of anna 'gift', which looks like a past participle 'given' used
substantively as 'a thing given'.
lestanen : 'in measure'. Evidently the instrumental of a noun lesta 'measure', which does not
occur elsewhere. Under the root ELED- 'go, depart, leave' "The Etymologies" give lesta- 'to leave', which
Tolkien rejected in favor of IInna 'go' as the present tense fo lende. He also gives a root LETH- 'set free',
related to LED- 'go, fare, travel'. Could any of these forms be related to lesta 'measure'? 7

Meter: Seven iambic feet:

, , , , ,
Toi ilrimar. IlIqailnen anltar anlnar lesltanen
1 2 3 4 5 6 7

Emendation to line 6: Toi irimar. lIyain antalto annar lestanen.

Tolkien emended antar to antalto and replaced ilqainen with i Iyain, dative
plural of ilya 'all, the whole', used here substantively as 'for everyone'. The
emended line presents the too-familiar problem of weak stress followed by
main stress: j'rimar. I'lyain.

6 This iIqaine seems rather ad hoc albeit possible. It may be that plural iIqai is used to
distinguish 'each one' receiving in her or his particular measure vs. the collective ilqa 'all' receiving in
the measure of the whole group. This would then require a syntactic explanation of the use of
insturmental ilqainen. The best I can devise is that we have an appositive connection with the
instrumental lestanen or that ilqainen is the "instrument" of the verb underlying lesta- 'measure',
i.e. 'to all in measure' = in the degree to which each one measures up. Either construction may have been
syntactically "difficult" enough to motivate the revision of the line. It is perhaps worth mentioning that
our only direct evidence for labelling the case forms in -nen, -inen as instrumental is the
abbreviation "I." used by Tolkien in the Book Quenya declension of cirya (see Beyond Bree,March '89)
and, as Paul Nolan Hyde reminds us, this could as easily stand for instructive which is one of the 15
cases of Finnish. In Sir Charles N. Eliot's Finnish Grammar (Oxford 1890, p. 160), he says "The
instructive is used to express the instrument or the manner in which an action is performed." Arthur H.
Whitney's Teach Yourself Finnish in Lesson 11 calls this "The Instructive or Instrumental Case". So the
significant point is not precisely what T. meant by "I." but rather that the related concepts of
instrument, manner, means can be expressed consistently by the same inflected form. C.G.
7 We may have a (Mannish?) euphemism here for 'death' as 'departure', with metaphoric reference to
its limitation or "measuring'" of Man's life. Compare the allusion to mortality as the "Gift of Men" with
the idea of antar annar here. A cliche use of lesta 'departure' = 'death' may have contributed to its
suppletion by linna in the conjugation of the literal verb 'to go'. C. G.

31
 trochee
, , ,
Toi ilrimar. Illyain I antallto anlnar lesltanen
1 2 3 4 5 6 7

or:

anapmst anapest
, , , , ,
Toi ilrimar. IIlyain antallto anlnar lesltanen
1 2 3 4 5 6

Line 7
lIuvataren. lIu vanya, fanya, eari,
"of lIuvatar. The World is fair, the sky, the seas,"

lIuvataren : 'of lIuvatar.' The examples you brought to my attention of Finnish subject nouns
in the genitive/dative -n have convinced me of two things: 1) There is a connection between the Quenya
genitive singular -n and the dative -no You can see the semantic similarity in a sentence such as Makll tanya
na Elwen; 'That sword is Elwe's' and 'That sword is for Elwe' express very similar concepts. Perhaps
genitive -n arose from a specialized use of the dative. 2) The mysterious subjects marked by -n (vean et
at.) are related to the genitive/dative -no I lay no claims to understand this relationship yet; and while not
ordinarily clairvoyant I do foresee a lot of Finnish grammar-study in my future.
You have also set me to thinking about the structure of the Quenya genitive plural and its similarity to that
of Finnish. I have previously assumed that the genitive plural broke down as follows:

Stem + Plural marker + GenitiV'e. m. + Plural m. 8

kiryaron 'of ships' = kirya + r + o + n
lassion 'of ships' = lass + + o + n

However, the Finnish genitive plural is constructed differently:

Stem + Plural m. + Linking element9 + Gen. m.

maiden 'of the lands' = ma- + i + de (te) + n

If Quenya follows the Finnish example, we may reanalyze klryaren, lasslen thus:

Stem + Plural m. + Linking element + Gen. m.

kiryaron 'of ships' = kirya + r + 0 + n
lassion 'of ships' = lass + + 0 + n

8 As in -lIe.!l. -ssen.. [P.W.]

9 So called in Teach Yourself Finnish, implying -fa- bears no semantic significance. Your grammar
hyupothesizes that -fa- is "apparently" the nominative plural -f.

32
In this interpretation 0 merely serves to link the genitive marker to the plural in such a way as to avoid
confusion with the dative plural (Iassln) or to avoid producing a form difficult to pronounce (ciryarn). I'm
not sure what you intended by bringing to my attention the elision of -te- in some Finnish genitive plurals,
such as jalka-iten > jalkojen. Perhaps in Quenya plural subject nouns ending in -n are genitive plurals in
which the linking element has been elided: oromandion, wlngildion > oromandl'n, wingildi'n?
Finally, compare lIuvataren with Hinl lIuvataro 'Children of lIuvatar' (Silm., pg. 400). Is the short
a in lIuvataro a typographical error, the correct form being *lIuvataro?
lIu : 'The World'.
vanya : 'is fair'. Again, "is" is merely implied.
tanya : 'the sky'. This gloss of fanya seems unique. Elsewhere Tolkien defines it as "cloud".
Fanyamar 'Cloudland' was the cosmographic term for the upper region of Vista 'Air', so this poetic
extension of meaning seems reasonable.
eari : 'the seas'. Plural of ear 'sea'.

Meter: Seven feet:

trochee
, , \

IIQlvatalren. Illu vanlya, fanlya, elari,

123 4 5 67

Line 8
i-mar, ar i1qa (men. I'rima ye Numenor.
"the earth, and all that is in them. Lovely is Numenor."

i-mar : 'the earth'. We find prefixed articles in I-mar and l-narq-uQllon but not in i tyel. Wtty?
The reason might be metrical. The stresses in imar and inarqellon may have suited our poet's purposes
better than those in unhyphenated I mar and i narqelion.
ar : 'and'. ilqa : 'all'.
(men : 'that is in them'. This is kith and kin with simen 'here' in line 9. The element -men in
both must be men 'place, spot', so simen is literally 'in this spot'. The first element in imen is probably the
"deictic particle" 1- 'that' or one of its derivatives I 'which' or I 'the', so that fmen is literally 'in which
place'.

I'rima : 'Lovely'.
ye : 'is'. Quenya has a veritable swarm of "beN's There is na, which besides functioning as a copula
also means 'exist'. Anwa 'real, actual, true' derives from ANA- 'be, exist' and seems to be a past participle
'been, existed' (compare vanwa 'gone', past participle of vanya- 'go'). Is it significant that both roots for
na, ANA- and NA., are identical to roots meaning 'to, towards'? does this relate to na's function as copula
of joining a subject to, towards a predicate complement? Na also appears cognate with the past participle
ending ·na and may also be related to the past-tense marker -ne. In this last regard I note you translated ns
as "has been".
Next comes ea, which has emphatic overtones. Again I note your translation, "is (for sure)". The Sindarin
equivalent may occur in the Unpublished Epilogue, in the phrase I sennul Panthael estathar aen 'who

33
should rather be called Fullwise'. The earlier draft has ge instead of aen. My interpretation is that sennul is
an adverb 'rather' (derivation unknown), estathar is a passive infinitive 'to be named' (note Quenya esta-
'to name'). and aen/ge is 'is'. Therefore a more literal rendition of the phrase is 'who is rather to be named
Fullwise'. The use of an emphatic form of 'to be' in this context makes sense.
Etymologically, Tolkien seemed to be hesitating between two original roots:

* A VA- > O. ea, Sind. aen

*GAV A- > O. ea, Sind. ge

Note the similarity with the roots for 'sea' and their derivatives:

* A V AR- > O. ear, Sind. aear, aer

*GAVA 'awe, dread' > O. ear, Sind. gaer

Ea and aen/ge might also be related to the root AYAN- 'holy' and its derivatives alnu/aini 'holy one, pagan
god' and alna/alre 'holy'. I read recently that Old Norse ass 'god' and the name of the Celtic divinity Aesus
both derive from the Indo-European root meaning 'to be'.
Then there is yeo There may be a connection with ye 'Io!', which is perhaps literally 'There it is!' Also
related may be QL's yeta 'look at'. If you can look at a thing, it must be or exist. In that case ye could also
mean 'Look!' Ye's interesting habit of suffixing itself to other words means that from now on we'll have to
examine all forms ending in -Ie with ye in mind. Is ye the source of the gerundiallifinitive ending -Ie? I like
your idea that the latter element in Inye and elye is ye 'is'.
Quenya also has two negative verbs, 'to not be', derived form the negative stems UGU- and UMU-. The
first-person singular present-tense forms are uln, umln 'I do not, am not'. Two third-person singular
present-tense forms occur, uye in line 9 and ume in the compound laume 'no indead not, on the contrary'
(literally 'no, it is not'). The past tense is ume, and the future tense uva occurs in line 12.
Numenor : 'Westland',

Meter: Seven feet:

trochee
, , , , ,
i-mar, I ar illqa ilmen. ilrima I ye NOlmenor.
1 234 5 6 7

Emendation to line 8: Tolkien wrote Vinya ("the Young") above Numenor as

an alternative, but this would not work metrically unless substantial changes
were made to the rest of the line as well.

Line 9
Nan uye sere indo-ninya simen, ullume;
"But my heart resteth not here for ever;"

Nan: 'But'.
uye sere indo-ninya : 'my heart resteth not'. I'm sure you are familiar with the
Finnish "verb of negation" (en. et. e/. and friends) and how it is used with the stem of a verb to form the

34
negative in the present tense. The stem of 'to rest' is lepa(t)a-, and so 'it does not rest' is se ei lepaa.
Line 9 seems to use this same construction, with indo-ninya 'my heart' being subject and the negative verb
uye 'is not, does not' being followed by sere, apparently the stem of the verb 'to rest' (as in serin 'I rest'.
For a structurally similar stem with a long vowel and final -e, note lute in "Ollima Markirya": Man
kiluva kirya . . . lute 'Who shall see a ship leave?').
Until recently I assumed that you form the negative of a verb in Quenya simply by prefixing u- to the
affirmative form. That's the way it's done in Sindarin: u-chebin 'I have not kept.' And so you saw
uhirielto 'they did not find' in the sample form my "Darkening of Valinor" translation. 10 But having
searched I cannot find an example of u- used this way in Quenya; it only appears there as a prefix in nouns and
adjectives. Now I think uhirlelto may be incorrect, and that a form using the verb of negation should be used.
The present tense of 'they do not find' would be uyelto hira. But what would the past-tense form be?
You have probably noticed that the Finnish and Quenya verbs of negation differ in one major respect: the
Finnish forms only convey the semantic concepts of negativity and person (en 'I-not', et 'you-not'),
whereas the Quenya forms also convey the concept 'to be' (or 'not to be', if you will). Therefore Q uin is
sufficient in itself to mean 'I am not', while Finn. en requires the addition of the stem of the verb 'to be',
o/e-. This also means that the Quenya forms can inflect for tense (uva, ume) while Finnish forms do not. In
Finnish negative constructions it is the auxiiary verb which indicates the tense rather than the verb of
negation: en syo 'I don't eat', en syonyt 'I wasn't eating' (in the latter case the aux. verb is the active
past participle). I think that Quenya would deviate from the Finnish model on this point, and that the verb of
negation would indicate tense while the auxiliary verb remained the same (i.e. the stem). So I shall emend
uhirielto to umlelto hira. The future tense 'they will not find' would be uvalto hira.
There is another possible way to interpret uye sere Indo-ninya. "The Etymologies" list the noun sere
'rest'. If we assume sere in line 9 t be this noun, we encounter a problem: 'there is not peace my heart' or
'my heart is not rest' is nonsense. You could surmount this problem by reinterpreting indo-ninya as 'for my
heart', i.e. Indo + dative -n + -inya 'my'. Finnish possessive pronoun suffixes are similarly added after the
case ending: talo 'house' > talossa 'in the house' > talossanl 'in my house'. This is not the usual noun +
possessive pronoun + case ending order seen in Quenya, as in tlelyanna 'upon thy path'. However, what is
usual in a language is far narrower than what is allowable, especially in poetry. You're unlikely to hear
someone say "Today I saw friends three" in ordinary cnversation, and_ yet we unblinkingly accept as
grammatical the line "and he called for his fiddlers three" in "Old King Cole". Metrically indo-nfnya is-far
more suitable than indonyan.

simen : 'here'.
ullume : 'for ever'. Clearly related is another time-word in line 11, yallume 'at last', which may
be a form of yalume 'former times'. This suggests that -Iume in ullume and yallume is lume 'time' or
some derivative of the root LU-. The first element ullume might be one of the negative stems UGU- or
UMU-, in which case ullume literally means 'timeless'.
Meter: seven iambic feet:
, , , ,
"
Nan Cllye selre inldo-ninlya silmen, uiliume;
1 23 4 5 6 7

Emendations ot line 9: The original version had hondo-ninya instead of

indo-n inya. Hondo-' is the inflectional stem of hon 'heart (physical)'. The word also had a broader
metaphorical application, as seen by its use in Huore 'Heart-vigour' (intended as 'Courage' rather than 'Having

10 Readers: Look for this next issue! - Editor

35
a Healthy Cardio-Vascular System'), slneahonda 'flint-hearted', and hon-maren, 'heart of the house', the
fire which always burned (as a heart never ceases to beat while life endures) in the house of Elendil.
Sinea 'flint' above brings to mind slnqe in "Narqellon". You will recall that sinea was originally tlnga
'flint' in Treebeard's epithet, a form related to tlnwe 'spark' and tlneo 'metal', all from TIN- 'sparkle'.
"The Etymologies" state that TIN- is "a variant of and in any case affected by THIN-", a root with
derivatives referring to dim grey light: slnde 'grey', sinye 'evening', sinta- 'fade'. Sinqe might belong in
this group, perhaps derived from THIN- as tlnwe from TIN-: *thln-kwe > slnqe 'pale grey spark,
glimmer'. It could be a verb, either a past-tense form in -e or one fa those present-tense forms ending in -e
(note the preference for forms ending in -e : slnde, sinye; and from TIN-, tine 'it glints'). Eldamar could
be the subject in that line: V'ematte slnqe Eldamar 'like something glimmers Eldamar'. Slnqe might
even be an adjective modifying Eldamar : slnq' Eldamar 'pallid Eldamar'.
Also note slnqltalla in line 14, which looks like the present participle of a verb slnqit-, as simpetalla
in line 9 is probably the participle of simpet- 'to pipe', and slrilla in line 11 is of sir- 'to flow'.
Laiqanlnwa 'green-blue' might act as an adverb modifying slnqltalla : 'gleaming palely green and blue'?
* * *
April 2, '89
Line 10
sa
ten ye tyelma, yeva tyel ar i-narqelion,
"for here is ending, and there will be an end and the Fading,"

ten : 'for', That is, 'for this reason, because'. This probably derives from TE3-/TEN- 'line,
direction' in the sense of 'reason for the direction of events'. Tenna 'unto, up to, as far as' must also derive
from this root: *te3-nna, *tefi-nna 'toward the direction'. The etymological reasoning resembles that of
mennal 'until', literally 'toward the place that'.
s! : 'here'. "The Etymologies" give si, sin as 'now' rather than 'here', and si is 'now' in Galadriel's
Lament as well. Elsewhere the word for 'here' is simen (line 9) and sinome (simane in the early drafts of
Aragorn's oath). However, in Sindarin Sl is usually the word for 'here', leading me to suspect that here in
line 10 we have a Sindarinism which has crept into Numen6rean Quenya. th_e .form a 'and' used before a word
beginning with a consonant (except h) may be another borrowing from Sindarin.
ye : 'is'
tyelma : 'ending', The noun-ending -rna most often occurs in words referring to concrete objects
(Nancy Martsch once equated -rna with rna 'hand', giving it the sense 'thing that can be held or touched Withe
hand'): kalma 'lamp', tarma 'pillar', tyulma 'mast' etc, Yet here we have an example showing -rna appears
in abstract nouns as well; tyelma 'ending', I.e. 'the process of coming to an end' rather than 'the physical end-
part of an object' (the word one would use for 'end-part' would appear to be telle 'rear'). Another example of
an abstract use of -rna which comes to mind is alma 'good fortune, weal, wealth', and there are probably
others.

yeva : 'and there will be', 'And' does not occur in the Quenya. Given the long vowel in yeva, one
might expect the present tense to be *ye, a fa ns. Was the vowel shortened to distinguish ye when used as
'it is' from its use as an exclamation ya 'Look! There it is!'? Note that ye is like anta in that it forms the
future tense with -va rather than -uva (**yeuva is certainly phonetically possible; cf. leuka 'snake',
neuma 'snare', etc.) Might ye also mimic anta in having a past-tense form identical to the present: ye 'it
was'?
tyel : 'an end'.
ar : 'and',
i-narqelion : 'the Fading', "The Etymologies" define this as 'fire-fading, autumn', the first

36
element being nar(e) 'flame' with short vowel in a closed syllable. Oelion 'fading' derives from KWEL-
'fade, wither'. The ending -Ion here may be a gerundial suffix; cf. Narquelle 'Sun-fading', the name used in
Third-Age Quenya for October, which ends in the gerundial suffix -Ie. Oellon might also derive from *qele
'fading' (cf. quelle 'fading' in T.A. Quenya) plus the augmentive suffix -on 'great'. In any case, narqellon is
used in this line as a euphemism for death, the flame nare which is fading being Anar the Sun, used in this
poem (and probably in Numen6rean thought in general) as a symbol of mortal man's span of life upon the earth.
It is to be remembered that Men awoke in Hlld6rlen only at the first rising of the Sun. Note how the use of
narqelion in this line is echoed in line 14: (re anarlnya qeluva 'when my Sun faileth'. Herendil's comment
on hearing the last line of this poem is: "Melka cometh back, they say, and the king shall give us the Sun
forever." (LR-63).

Meter: Seven iambic feet

, , , , , , ,
ten si I ye tyellma, yelva tyell ar i-Inarqellion,
1 2 3 4 5 67

Line 11
ire ilqa yeva not ina, hostainieva, yallume:
"when all is counted, and all numbered at last,"

ire : 'when'. The element -re occurs as a suffix in many words referring to time: ire 'when', tare
'then', enyare 'in that day', and yare 'former days'. This element is probably one and the same with re, the
word for the Eldarin solar day. Note that enyare and yare are glossed as 'in that day' and 'former days',
respectively. Likewise, tare might be literally glossed as 'in that day' « TA- 'that') and ire as 'in which
day' « deictic particle 1-; compare this with (men, discussed in line 8).
The Sindarin equivalent of (re appears in Luthien's song in Lays of Beleriand pg. 354: Ir Ithil am men
Eruchfn I menel-vir sila diriel, When the Moon shines silver, a heavenly jewel for us, the Children of
God, (and) has watched over (us) .. .'

ilqa : 'all'
yeva : 'will be'
notina : 'counted'. This is the past participle of not· 'reckon' (LR-378) and is morphologically in
perfect accord with the various past participles seen in the latest version of Oilima Markirya, which
Christopher Tolkien writes "comes ... from the last decade of my father's life". That is, we see lengthening
of the stem vowel and the addition of -ina, hence not- > n6tina and rak 'break' > rakina 'broken'. We are
not given the uninflected stem of rukina 'confused, shattered, disordered', but it is very likely to be ruk-.
Tolkien translates yeva n6tlna as "is counted", but yeva is in the future tense 'will be' and so a more literal
translation would be 'will be counted' or 'will have been counted'. What we have here is a compound tense, a
future passive periphrastic (watch those Quettar-readers head for the bushes).
hostain ieva : 'and all numbered'. 'And' is not present in the Quenya, and lJqa 'all' only
occurs previously in the line before yeva n6tlna. Hostalnleva is another future passive periphrastic verb,
this time with the auxiliary yeva attached in suffix form -leva to the past participle hostalna 'numbered',
with a literal meaning of 'will be numbered' or 'will have been numbered'. The verb hosta- is only defined
elsewhere (MC-223, LR-364) as 'gather, collect, assemble' rather than 'to number', but in "The Etymologies"
the derivative noun hosta is defined as 'large number'. Note that the final a in the verb stem hosta- is

37
retained in the participle hostalna, and compare this with the stems not-, rak-, which have no final vowels,
and their participles n6t1na, raklna.
yallume : 'at last'. See the discussion of ullume in line 9. I will add here that the double I in
yallume may derive from a proto-form *yira-Iume 'former-time' (yara 'ancient, belonging to or
descending from former times', LR-399), which contracted to *yarlume and then to yallume (as ullume
may derive from earlier *uglume or *umlume). If yallume does indeed have a meaning similar to yalume
'former times', one might cock an eyebrow at its use in a sentence referring to events in the distant future.
But it should be remembered that yaya n6tlna and hostalnlaya may be future perfect passive verbs,
referring to events which will have been completed, over and done with, at some time in the future: 'when all
will have been counted, and all will have been numbered in former times.'

Meter: Thanks for the info on the significance of the underposed dots. Elision of the final e in ire is an
absolute necessity for this line to work metrically:

anapaest anapaest
, , , , , ,
"
ire illqa yelva noltina, hosltainielva, yailiume:
1 2 3 4 5 6 7

Without the elision you end up with a line of eight feet, the first four of which are trochaic.
I agree completely with your comment that "the elision is optional on the part of the songster, or else rule-
determined with nonelision (or hiatus) optional." Assuming the transcriptions of lines 1, 2, 13, and 14 given in
LR-63 represent an accurate record of what Elendil and Herendil heard on that occasion, we can see that Firiel
herself chose to sing the song without any elision at all in lines 1 and 14 (Le. on pg. 63 we see no underposed
dots or apsotrophes).

Line 12
ananta uva tare farea, ufarea!
"but yet it will not be enough, not enough."

ananta : 'but yet'. "The Etymologies" give this (LR-375) in hyphenated form a-nanta with the
additional definition 'and yet', making it clear that the initial a- is the short form of ar 'and' discussed in line
1. The middle element is of course na, nan 'but, on the contrary, on the other hand'. The final element might
be ta 'that, it' (LR-389), giving ananta the literal meaning 'and on the contrary that .. .' Another possibility
is the allative suffix -nta seen in some of the "Secret Vice" poems, e.g. tollallnta 'upon hills' (MC-214) and
sapsanta 'into a grave' (MC-221), hence ananta = 'and to the contrary'.
uva : 'it will not be'. Future tense of the negative verb. See the discussion of ye in line 8 and uye
sere indo-nlnya in line 9.
tare : 'then'. For the etymology of this word see ire in line 11. 'Then' is not included in Tolkien's
English translation of this line, and neither is it included in the translation of line 13 in which tare also occurs,
casting some doubt on this interpretation. Another possibility is that tare is a demonstrative pronoun < T A-
= 'that (thing or person),. Way back when in your letter of Oct. 27, 1988 (commenting on the translation of
Tom Peterson's title inscription) you compared ten sf ye tyelma 'for here is ending' and yeya tyel 'there
will be an end' as subjectless clauses with ananta uya hire tarea 'but yet it will not be enough' as a
subjectless clause (if it is indeed subjectless).11 There is an important difference between the first two

11 What I said was: "There is the pronoun tar 'thither' in Etymologies under T A -. I suppose you have

38
clauses and the last clause, namely ten si ye tyelma and yeva tyel involve predicate nouns whe reas
ananta uva tare farea has instead a predicate adjective, farea 'enough, sufficient' < fare 'sufficiency,
plenitude'. It is clear from ye tyelma and yeva tyel that in Quenya the copula used alone with a predicate
noun and no explicit subject is a perfectly grammatical construction. However, it may be that the copula used
with a predicate adjective does require the presence of an explicit subject, and that may be the function tare
serves in this instance. This is precisely the case in Esperanto. If you wish to use the adjective sullt;a
'farea' in translating 'it will not be enough' you must provide an explicit subject: flo ne estos sulifa; tio
= 'that thing' or 'that state of affairs'. You can omit the explicit subject, but in that case you cannot use the
adjective suflfa but instead must use the adverbial form sufife: ne estos sulife, 'it will not be
enough'.
There is marie 'it is good' in line 4 to be contended with in this regard. As noted in my discussion fo line
4, this appears to be a suffixed form of the copula ye added to the adjective mara. It would appear, however,
that the clause Mardello Melko lende which precedes marie acts as its subject (a fact which you point out
in your most recent letter: "However we view the morphology of the verb the subject is whatever noun ends
the previous clause ... or the clause itself in the case fo marie, where the preceding word is the verb in that
clause.") The subject function of Mardello Melko lende in relation to marie is apparently what Tolkien
meant to indicate by the colon after len de - 'Melko has gone from earth, (which) is good'; and likewise with
Valion : numessier 'of the Lords, (who) are in the West' and morion : talantie 'the dark one, (who) is
fallen' .

farea, ufarea : 'enough, not enough', As noted above the adj. farea derives from the noun
fare 'sufficiency, plenitude, all that is wanted'. "The Etymologies" give the form as farea with short a. The
short u in ufarea is puzzling as well (cf. un6time).

Meter: All iambic, seven feet:

noticed ore as a temporal ending in the "pronoun" ire 'when' = '(at) which time'. So tare would be '(at)
that time' = 'then', and uva tare farea translated 'it will not be enough' is literally 'will not be then
enough', We could argue that tare corresponds to 'it' in the translation (and perhaps it does serve a ..,.
similar function) but it is just a marker here, as say in it is raining; it will be sunny; it was a dark and
stormy night; it is I. It does not stand for anything, but simply allows the copula with a predicate
(nominative or adjective) to constitute a sentence. In other constructions there serves the same marking
function: there is enough for everyone; there will be rain tomorrow; there is no fool like an old fool.
This is the way yeva tyel is translated, 'there will be ending' with nothing corresponding to 'there' in
the Quenya. There are historical reasons peculiar to English no doubt, but nothing logically imperative
in pronouns with the meanings of it and there being used as these kinds of abstract markers. Because
Quenya has verbs inflected for each person it does not even require a separate word subject in every
clause (which ~ the grammatical "reason" that it or there fill the subject slot in these subjectless
clauses in English, which does.)
"But there might be situations in Quenya where a grammatical gap needs to be filled by a word
without a seemingly essential meaning, In ten sf ye tyelma, yeva tyel both clauses are subjectless.
But ananta uva tare farea which is subjectless follows a clause whose subject is iIqa. Perhaps the
use of tare 'then', placing~ the clause in the time-frame of yallume 'forever', helps to indicate that the
"subject" has changed, the absense of explicit subject not referring back to ilqa, but indicating another
pure predicate clause." C.G,

39
 ananlta alva talre falrea, I ufalrea!
2 3 4 5 6 7

Line 13
Man tare antava nin lIuvatar, lIuvatar
"What will the Father, 0 Father, give me"

Man : 'What'. This interrogative pronoun also occurs in man-Ie 'What is it?' and E man antavaro
'What will he give indeed?' In Galadriel's Lament it is used as 'who?' : si man I yulma nln evquantuva?
'Now who will refill the cup for me?' In most cases context would be sufficient to indicate which meaning was
intended.
tare : As in the peceding line, tare here presents us with more than a single possibility. It may mean
'then', in which case this line reads 'What then will the Father, 0 Father, give me.' If tare is a demonstrative
pronoun, what purpose does it serve in this sentence? It may be to prevent confusion - Man antava nin
alone could easily be misinterpreted as 'Who will give me?' (and if laire is 'then', it would seem that the line
could still be easily misinterpreted as "Who will then give me?") After reading your comments on the use of
word-order in Quenya to distinguish subject from object, I would agree that besides the "standard" S-V-O
order Quenya also made use of O-S-V (the purpose of course being to emphasize the object by giving it primacy
of place), as seen in line 1: lIu lIuvatar en kare 'God made the World .. .' Tare may be used in this line as
a stand-in for the subject (later specified as lIuvatar) so that the object (man) - subject (Ure) - verb
(antava) order wil make it clear that man is the object and not the subject. 12
antava : 'will give'. See discussion of antar6ta in line 2.
nin : 'me'. This is a dative form ni-n, literally 'for me' or 'to me'.
lIuvatar, lIuvatar : 'The Father, 0 Father'. The first lIuvatar is the subject fa antava
(or at least acts as a further specification of tare), whereas the second is considered a vocative, '0 Father!',
i.e. a direct address to lIuvatar. This gives the line an interesting doubre ·implication. Man tare anniva
nin lIuvatar 'What will the Father give me?' is an impersonal rhetorical question; but the addition of the
second, vocative lIuvatar has the effect of implying a direct question or plea to God, "What will you give
me, 0 Father!" without expressing this plea directly or overtly. Typical Numen6rean restraint in such matters
of religion, as you pointed out in a previous letter with regard to Cirion's Oath and the sidelong way in which it
refers to the Valar and lIuvatar. In fact, if tare is indeed a demonstrative pronoun, its use in this line as a
sort of unspecified subject the exact nature of which is clarified later on in the sentence is reminiscent of the
use of -nte in Cirion's Oath, where it acts as the unspecified subject of tlruvantes (i.e. there is no previous
referent to clarify who -nte refers to), the specific identity of which is provided afterwards: I harar
'(those) who sit' and I Eru 'the One'. Of course lIuvatar is previously referred to in Ffriel's Song in lines 1
an 7, but by the time line 13 rolls around it would no longer be apparent that lIuvatar was the referent of
tare.

Meter: I am as wary as you (well, almost) when it comes to assuming the texts as we have them contain
typos and/or other errors, but I think there is great merit in your hypothesis that the underposed dot in
enyare in line 14 actually belongs in line 13 under the final e in tare. This would yield a perfectly iambic
line:

12 Another sentence in which the customary nature of the O-S-V order is relied on to distinguish subject from
object is Hul oilima man klluva 'Who shall see the last evening?' (MC-214). [P.W.]

40
 Man tare I antalva nin IIiOlvatar, IIiOlvatar
2 3 4 5 6 7
The line still works metrically without the elided e (and this appears to be the way Ffriel herself chose to
render the line), save that the second foot becomes an anapaest:
anapaest

Man tarle antalva nin IIiOlvatar, IIiOlvatar

2 3 4 5 6 7

Line 14
enyare tar i tyel, Ire Anarinya qeluva?
"in that day beyond the end when mt Sun faileth?"

enyare : 'in that day'. There is merit in your theory that this form derives from the future
element en plus yare 'former days' to mean literally 'future former days'. This would parallel the use of a
form of yalume 'former times' with verbs in the future passive perfect in line 11. Just as Quenya sometimes
uses a word referring to the past with a verb in the future tense to imply the future perfect: enyare ire
Anarinya qeluva 'in that (former) day when my sun will have failed.'
There are other etymological scenarios as well. The first element might be enya 'middle', with the sense
being extended to 'in the middle of' and thence to simply 'in'. It could also be an adjective *enya 'that' derived
form EN- 'over there, yonder' (an alternative form to enta 'that yonder'). The latter element could be either
the temporal suffix -re as in ire, or it could also be are 'day'. It is interesting to note that if you reverst the
elements from enya-are to are-enya and submit this latter to the standard vowel shifts observable in the
development of Sindarin words from Common Elvish forms, you end up with the form erin, and of course erin
'on the day' is indeed a Sindarin word occurrig in the Unpublished Epilogue.
tar : 'beyond'. In LT2-347 (entry Tarulthorn, Taruktarna 'Oxford') is found the root TARA with
Qenya derivatives tara- 'cross, go athwart' and tarna 'crossing, passage'.· The preposition tar 'beyO'fld'
must derive from this root as well. "The Etymologies" do not give the root TARA, but in the entry for the
root TA- 'that' one finds the form tar 'thither', a word which ends in the locative/allative suffix -r seen
elsewhere in such words as mir 'to the inside, into' (LR-373), ranar 'in the moon' (MC-213, 214), and yar
'to whom' (MC-215). It is possible that tar 'beyond' is merely tar 'thither' used in a special prepositional
sense; at any rate it seems likely that TARA '(go) across' is derived from or related to TA- 'that', in the
sense of 'go over to that place'.
"The Etymologies" also give a root THAR- 'across, beyond', which is clearly related to TARA, but tar
cannot derive from the former since Common Elvish th yielded s in Quenya and so one would expect *sar.
Also note the preposition ala 'beyond, after' (MC-214, 221) and the postposition pella 'beyond' (perhaps
derived from "pelala, present participle of pel- 'go around'?)
i tyel : 'the end'.
Ire: 'when'
Anarinya : 'my Sun'. This is a metaphor for "my Life". See discussion of I-narqellon in line 10.
qeluva : 'faileth'. Tolkien translates this as present tense when it is in fact future tense: 'will fade'
or 'will wither'. It will be noted that anta- 'give' has the future form antava whereas qel- 'fade' has
qeluva. 13 It may be a rule that verbs with stems ending in a vowel form the future tense simply by the
addition of -va (w/lengthening of the final stem-vowel), hence hosta- 'to collect' > "hostava (also note the

13 qel- is not given in the corpus, alas. But it seems likely, given the root KWEL-.

41
retention of the stem-vowel in hostalnleva}; whereas verbs with stems ending in a consonant form the future
tense by the addition of -uva, hence slr- 'to flow' > *siruva.

Meter: As noted in the metrical analysis of line 13, the underposed dot in enyare may be an error. At
any rate there seems no good reason to elide this e, since the following word begins with a consonant rather
than a vowel and the line will not work out as iambic with this e elided. The e in ire, on the other hand, is
followed by a word beginning with a vowel and its elision is necessary to maintaining the iambic meter:

, , , ,
enyalre tar I i tyel, I ire Alnarinlya qelluva?
2 3 4 5 6 7

And now, believe it or not, we are almost at the end of the Excruciating
Analysis. I say "almost", because there are still a few things which should be
noted about Ffriel's song as a whole.
First of all, the structure of the poem gives strong evidence of a Numen6rean belief in numerology. It is
significant that this poem has 14 lines, and that all the lines have 14 syllables, except for line 11 which has 16
syllables (and assuming one may elide the final e in tare in line 13, which was necessary to yield 14
syllables). There are several references in The Hobbit which indicate that it was a common belief among the
races of Middle-earth that 14 is a lucky number. The number fa members or Thorin's expedition to Erebor was
carefully chosen to total 14; when the dwarves show reluctance to accept Bilbo as a companion, Gandalf acidly
comments: "You asked me to find the fourteenth man for your expedition ... Just let anyone say I chose the
wrong man or the wrong house, and you can stop at thirteen and have all the bad luck you like" (H-27). In his
conversation with Smaug, Bilbo says of himself, "I was chosen for the lucky number (H-235), and later on
Smaug says "Why not say 'us fourteen' and be done with it, Mr. Lucky Number?" (H-237). Evidently the belief
in lucky 14 was so wide spread even dragons were aware of it.
Tolkien was wont to insert bits of Elvish lore into his Father Christmas letters, and on the third page of the
Introduction t Fe Letters (I cannot believe they didn't number the pages in that book), beneath the drawin~ of
the reindeer-drawn sleigh "Father Christmas" makes an interesting remark: "I am coming from the north you
see - & note NOT with 12 pair of deer, as you will see in some books. I usually use 7 pair (14 is such a nice
number)."
Where did this belief originate? It certainly must derive from the fact that the Valar were 14 in number:
"The Lords of the Valar are seven; and the Valier, the Queens of the Valar, are seven also." (S-25). Under the
entry Valar in the appendix to Lost Tales 1 (LT-272) it is said that the Qenya words valin, valimo 'happy'
and vald- 'blessedness, happiness' derive from the same root VALA as Valar, Vall. Similar connections
between the Valar and luck are given in GL, including the expression i-walt ne Vanion 'the luck of the Valar'.
By modelling his poem on 14, our poet probably hoped to acquire some of I-walt ne Vanion for himself ant to
quiet his troubled heart, perhaps even his lifespan.
Line 11 with its anomalous 16 syllables could be a corruption of an original
14-syllable line. The likeliest reading of this original line is:
", , ,,, ", , '\

ire illqa yelva noltina, I hostailna, yailiume:

1 2 3 4 5 6 7
'when all will be counted, (and) numbered, at last,'14
----------------------
14 Isn't it ironic that the sale flaw in the numerical structure of the poem would occur in a line referring to

42
which achieves its 14 syllables by omission of the copula suffix -h~va fro m
hostain iE~va. The insertion of this superfluous copula into this line may have
been due to scribal error or was the addition of a later bard who meant to
"correct" the grammar of the line while failing to comprehend the numerological
importance of maintaining 14 syllables per line (Ffriel herself does not seem to
comprehend this, singing lines 1, 13, and 14 without elision and hence with 15
syllables each).15
Who wrote Ffriel's Song, and when? Clearly Ffriel did not write it, for it is said to have been "made by
men, long ago" (LR-62) and it is clear from the way the song is referred to in LR-62-63 that by F(riel's time
the song was a "golden oldie" familiar to every Numen6rean. The song expresses discontent over Man's
mortality, and yet there is no trace of bitterness in the references to the Valar or the Elves. It is therefore
most likely that it was composed sometime during the reigh of Tar-Minastir, for as it says in LotR, \11-316:
"The first sign of the shadow that was to fall upon them appeared in the days of Tar-Minastir, eleventh king ...
. He loved the Eldar but envied them. The Numen6reans ... began to yearn for the West and the forbidden
waters; and the more joyful was their life, the more they began to long for the immortality of the Eldar."
According ot "The Line of Elros" in Unfinished Tales (U-220), Tar-Minastir's reign lasted from Second Age
1731 - 1869. The year 1731 given as the first of Tar-Minastir's reign conflicts with information given
elsewhere. "The Tale of Years" has Tar-Minastir on the throne in S.A. 1700, stating that is the year in which
"Tar-Min astir sends a great navy from Numenor to Lindon." (111-364). "The History of Galadriel and Celeborn"
puts him on the throne even earlier: "In 1695, when Sauron invaded Eriador, Gil-galad called on Numenor for
aid. Then Tar-Minastir the King sent out a great navy ... " (UT-239). So the most accurate date we can put to
the composition of Ffriel Song is to say that it was probably sometime during the period S.A. 1695 - 1869.
"The Line of Elros" also has this to say about how Tar-Minastir came by his royal name (U-220): "This
name he had because he built a high tower upon the hill of Dromet, nigh to Andunie and the west shores, and
thence would spend great part of his days gazing westward." Is it possible that Minastir himself was the
author of Ffriel's Song, written as he sat in his tower gazing yearningly to the West?
Ffriel's Song is referred to as "an even-song" (LR-62), and one can see the poignancy of singing it at
sunset, when Anar, used by the poet as a symbol fo Man's life, fades away and dies in the West, where dwell
the enviable Elves and Valar in immortal bliss.
Finally, there evidently was an altered version of Ffriel's Song current in the days of Elendil and Herendil,
a version with far darker implications than the ancient original (Elendil says on LR-66 that Sauron has been in
Numenor for 44 years; "The Tale of Years" dates Sauron's coming to Numenor as S.A. 3262, which would
indicate the events in "The Lost Road" take place in S.A. 3306. This fits in with "The Tale of Years" which
states: "3262 - 3310 Sauron seduces the King and corrupts the Numen6reans.") Herendil refers to this
altered song in LR-63: "They sing it otherwise now. Melko cometh back, they say, and the king shall give us
the Sun forever." Also note Elendil's comment in LR-68: "The old songs are forgotten or altered; twisted into
other meanings. It

Here ends the EXCRUCIATING ANALYSIS. If it has passed from the high and the
beautiful to darkness and ruin, well that's just the sort of thing you've come to
expect from me anyway.

everything being counted and numbered?

15 Another grammatical emmendation would presumably be ire ilqa notina, hostainiha,
yallume, i.e. 'when all counted, will have been assembled, at last'. This has the "advantage" of being
syntactically subtle, ambiguous as to when the 'counting' takes place, and so susceptible to scribal or
bardic "clarification" by repeating the yeva from the previous clause. C.G.

43
 to understand; to give some idea of how different the
languages are, compare these two sentences, both meaning
"He is going back to Wales on a bicycle":
Breton: Mont a ra deus Bro-Gembre war
warc'h-houarn.
Welsh: Mae e'n mynd i Gymru ar ben beic.
Even so, as you point out, it is a great deal easier for
someone with a knowledge of Welsh to learn Breton from
David Doughan scratch than for an English-speaker without any Welsh.

First, congratulations on PE 7, which not only looks [My thanks for the correction: I do hate to spread
good, but is full of thought-provoking matter about Elvish misinformation. What I should have meant to say was "!lQ..U1lS.
languages. For now, I'll just comment on a couple of cause lenition in the following adjective if the noun is feminine"
things involving the Welsh language. On pg. 37 you and certainly added that this is not all there is to lenition in
remark that "adjectives cause lenition in the following Welsh. This would not have illustrated the point I was making
noun if that noun is feminine". Sorry, no. In normal quite so straightforwardly, but still it is true that Welsh lenition
modern Welsh, an attributive adjective invariable follows depends in part on grammatical gender, which must simply be
the noun (the only exceptions being hen [= 'old'], priJ memorized for most nouns. I was expressing the hope that
[= 'main, chief], and unig [= 'only'], all of which cause "mutationis" mutandis Tolkien might have kept the grammatical
lenition in a following noun, whatever its gender). changes of consonants in Sindarin free of rules that we could
Attributive adjectives following a feminine noun do 1enit, never learn to apply properly simply from the select examples of
e.g. dyn mawr ('a big man'), gwraig Jawr ('a big Sindarin included in his writings. - C.G.]
woman'). However, just to confuse the issue, the situation
is quite different in medieval Welsh (as well as some forms
Jenny Coombs
of deliberately archaic modem Welsh, such as "strict metre
poetry"). There it is more usual (though by no means Thank you for the Parma and your letter; they were both
invariable) for the adjective to precede the moun, in which
very interesting. Is the elen sile lumesse
case the noun always lenits (mawr ddyn, Jawr wraig)
omentiemman which you quoted from the early
... confused? You will be when you learn that adjectives
manuscripts as published in "Return of the Shadow"? I
used precatively always lenit after the particle yn (again
very much enjoyed -Parma; it's beautifully presented,.,}Vith
irrespective of gender), and lenition also occurs in the direct
all those typefaces and letter styles!
object of a simple-form verb, e.g. gwelais ddyn (= 'I Concerning Corma Laire Quenyasse. I agree
saw a man') - cf. Sindarin lasto beth lammen. Quenyasse does mean literaly "in Quenya", but would
This all goes to show that lenition/soft mutation in seriously question its suitability here! The locative surely
Welsh is a large and complex subject - as it is also, refers to physical situation, as in mahalmasse = "on a
mutatis mutandis, in Sindarin, Goldogrin, etc. throne"; the chances of its happening to correspond to an
You rightly note (pg.31) the claims and counter-claims English idiomatic phrase with no physical reference are
about the mutual intelligibility of Welsh and Breton. I fairly remote. In what sense can a poem be said to be in,
personally have heard some Welsh-speakers hold forth at or on an intangible concept like language? I suggest the
about the days of their youth (the 1920's and 1930's) when instrumental instead, i.e. Quenyanen. This suggests "by
the the valleys of South Wales were regularly visited by means of, using Quenya". There are precedents for using
the instrumental to translate nonphysical "in". ego
"Sioni Wynons" (= "Johnny Onions") - Breton
surinen = 'in the wind'; likewise in Latin, the ablative
onion-sellers on bicycles - whose Breton , it is claimed,
(which is frequently instrumental in sense) is often
could be understood quite easily by the locals. However,
appropriate, as in lingua Latina = 'in Latin'.
since the poeple who make these claims usually have I assume "allative" in the second line of Parma p.26 to
considerate difficulty themsel~es in understainding even be a misprint for "dative", which ham in clearly is both in
simple Breton, I strongly suspect that the "Sioni Wynons" form and sense.
were actually attempting to speak Welsh ... Cetainly, I I find toia rather clumsy for 'their', and also ambiguous;
as a fluent (non-native) Welsh-speaker fmd Breton difficult it could refer either to the dwarf-lords or the halls of stone,

45
although I suppose common sense would imply the latter. turieneninya = 'by the ruling of me', 'by my being ruled'
Patrick Wynne's suggestion seems neater, and more in (using the "form of a suffix -inya [which the 1st. pers.
keeping with the spirit of Quenya, which rejoices in sing. takes] when added to a consonant stem", as given in
forming great long compounds with twenty suffixes at the your "First Person Pronoun Possessive in High-Elven"). I
end! (Ondo-rondo-nta-sse-n = 'stone-cave-their-in (pl.)' don't know how seriously one should take those; I admit
is really rather splendid.) Does not Tolkien's comment that they were getting progressively more unlikely. Still, it's a
the verb suffix -nte is used "where no subject is thought, and you've got to expect some monsters like
previously mentioned" apply equally to the other person ***turieneninya when you translate five English words
suffixes? Where there is a subject, the bare stem of the into one Quenya!
verb is used (eg. Firiel linta); where there is no subject, By the way, is nute te, in the penultimate line of the
the personal inflection is used (eg. lintarye). Where this poem, a misprint for nutie, or have I missed its
would lead to ambiguity, the personal pronoun itself could significance?
be used in apposition to the noun, as you have explained in Thanks again for Parma; I look forward to the next
"1m Naitho"; so I might say lintanye (where the suffix issue!
makes the person clear), but inye Yenne linta (the
pronoun showing that I am making an observation about [Jenny's argument that Carma Laire Ouenyanen would be 'the
myself). This does not affect the use of the possessive - Ring Poem in Quenya' in is quite compelling, and had Jorge
nya suffix with nouns; so might not possessive -nta be suggested it for his title I would not have disagreed in my
legitimate? position as his assistant and "corrector" in rendering that
I admire the way you and Jorge have tackled the "to find English poem into Quenya. But at the same time I do not feel
them" etc. phrases; I had wondered how you would do that. comfortable with idea that the locative in Q. only "refers to
But I find myself tending to agree with Patrick Wynne physical situation", as she seems to suggest. So I did not feel
about the use of the dative. I can't really see that compelled to insist that Jorge ought to change it.
enyalien's being subordinate to a verb rather than to a Certainly as separate theoretical question this deserves study.
noun makes much difference. And all the gerunds in the We know that a noun that can have "nonphysical" reference does
Corma Laire govern direct objects (te), as enyalien not thereby lose any of its case·formations in the paradigms of
does, so there is no difference there. Mine Corma its language. And we know from considering English alone that
hirien te does not seem to me to be ambiguous; 'for one the analogy whereby the grammatical parrallel between say I
ring to find them' would rather be mine Corman (dat.) conveyed an Idea In latin and I ferried a passenger In my boat
hirien teo If the gerunds are nominative, then the phrases emerged in the history ot our tongue, is obviously independ~nt of
would have to be understood 'one ring the finding of them' specific "idiom", because it is a natural characteristic of sentient
etc., where the gerund is in apposition to the noun: 'one expression. In Sanskrit, which like Quenya has both a specific
ring (being) the finding of them'. 'One ring for the rmding instrumental and locative case-form, the later combines among
of them' makes better sense, to me at least; although its many uses, reference to the place of an action, the time of an
admittedly the whole ring poem does not contain one action (cf. Q lumesse), or the circumstances in which an action
complete sentence in any case. takes place. I feel that the question ought to be, given the
To complicate matters a little further while I'm about existence of forms Ouenyanen vS. Ouenyasse, what would the
it, what about incorporating the direct objects into the difference be used for? - C.G.]
gerunds? The object of a finite verb is of course
incorporated as a suffix (eg. u-tuv-ie-nye-s = 'I have Craig M arnock
found it'), and although gerunds are rather special, operating
both as verbs and as nouns, I don't know of a compelling
Comment on : Corma Laire Quenyasse
reason why they shouldn't be able to incorporate object
sufixes after the usual case endings; the case endings would
take the place of the person endings in a finite verb. It TITLE : Corma Laire :: I don't like these two words
would certainly make it tidier. I suggest turient(e) = just placed together like this, though I can't give any good
'for the ruling of them', etc. SiJpilarly with other cases and reason why. In English The Ring Poem' means 'the Poem
objects : turiot = 'of the ruling of them both', 'of their about the Rings', which I think would be better expressed
both being ruled' (I'm a bit uncertain about the formation via some Q. inflection, though it's hard to say which one
of a genitive from nominative -ie, but never mind), since, as noted directly below, it is easy to make good

46
cases for just about every one in just about every (Hist. Middle-earth, Vol. 4) and refers to a specific area in
situation. 1 the sky. Although the comparison is not of course exact,
Quenyasse :: I'd prefer Quenyanen here, as it is 'The it is as if we said "Three Rings for the Elven-Kings under
Ring Poem (written) in Quenya', i.e. using Quenya, the mesosphere'. Menel would appear to be the most
though the instrumental is the "tricksiest" case to accurate replacemant 3
implement.
Line 2 (Otso Nauco·heruin toia ondorondossen):
Line 1 (Nelde Cormar Elda-harnin nu elenarda): beruin :: Misprint here, either heru or her, not
haran :: As noted in my accompanying letter, I don't like **heru.4
using words from another "variety" of Eldarin where a toia :: An awful lot of the stuff appearing in LR &
paraphrase (or construction) is possible or (even worse) other places is inconsistent with "Classical" Quenya, and
where a perfectly good word already exists (Le. is extant) this is one of them. I can't imagine why you {Chris}
for that variety. This element does of course appear in the should object to *-nta from ·nte, paralleling -Ima & •
canon for "Classical" {LotR} Q. as a part of haranye Ime. The point that "no subject is previously mentioned"
(and with closely related word in harma), but whether this has been brought up before (Quettar #28, p. 8) but not
means 'lord or king of a specified region' (see TA, Etym.) answered. It may be that noninflected forms were used (i.e.
is open to great doubt, especially as the related S. forms the subject-verb-object arrangement rather than verb +
begin with arm, and we have many examples of Quenya subject + object), though textual criticism would be
words which also begin with arm. Since no base ** AR- necessary to back this up.
occurs in Etym. (all forms coming form GHAR-
3 The reason mesosphere sounds unusual here is the fact that
n,
{3AR- I'd suggest that the (Q.) forms in har- have the
it uses a modem borrowing of the ancient Greek meso-which
meaning of 'something (to be) treasured', whilst those with
most English speakers would have to "look up" to understand its
royal & kingly connotations were transferred to base ...jAR,
meaning, and even realizing that mesosphere = 'middle sphere'
the two being mutually exclusive. There is a wide
does not pin down what the word refers to. On the other hand
selection of possible alternatives (aran, cano/u, her,
any Quenya speaker would know that elenarda = elen + arda =
heru, &c.) depending on the connotation required. 2 'star-realm' and that it refers to the sky. To take an example
elenarda:: This is actually a proper name given in the closer to home if in a different sphere, the compound black·bird
list of cosmological words following the "Ambarkanta" would mean something much more specific to a bird-watcher than
1 There are some Quenya proper names that consist of 2 nouns it does to the average English speaker for whom it is generic and
simply juxtaposed. For example Caron Ololaire 'Mound (o~ Ever- refers to any black bird.. To treat elenarda as a strictly teclJ.pical
summer', with no cases. - Editor term like English mesosphere is to impose a modem (specializing)
2 This is interesting speculation on the meta-history of these view of vocabulary usage that need not apply to Quenya at all. -
words in the hypothetical conception of the author, but I find the Editor
conception displayed in Etymologies too subtle and realistic to 4 Not a misprint. The thinking behind this form was that if
hastily assume that T. abandoned it when no unequivocal evidence one were to form an ablative say from Mr one would have to
of a new conception is forthcoming. You gloss over the fact that supply some connecting vowel before the ending ·110. Generally
under the entry 3AR 'have, hold' he also mentions that the this vowel is assumed to be e whence ·Mr~1I0, on the evidence of
"related GAR, GARAT, GARAD were much blended in Eldarin", nom. ear 'sea', abl. Ear~lIo; nom. Endor, End6re 'Middle-earth',
and he gives alongside of and coexisting with the Q. forms in har·, allative Endor~nna. But even if the vast majority of ·consonant-
including harma, haryon '(heir), prince', haran, several Q. forms in stem" nominatives show connecting vowel e in other cases we
ar·, including arda 'realm', armar 'goods', aryan 'heir'. I see would expect some exceptions. For example riel 'crowned maiden'
seems to have connecting vowel a, judging by the genitive
nothing in "Classical" Quenya or Sindarin to contradict the
Altarlello. If the connecting vowel were e we would get "rielleo.
hypothesis that Q *aran derives from haran by analogy with
Risky to guess which nouns are the exceptions, but her is
related arda, aryOnlharyon. I would accept AR· as a Quenya and
certainly unusual in having an alternative nominative heru with this
Sindarin root, and harm as a Quenya stem that developed a sense particular final vowel. I consequently posit abl. herullo, dative
more or less as you describe, in harma at least. But the idea that plural heruln, etc. Other solutions to the original problem might
historically these various forms all derive from primitive roots be proposed, and indeed Jorge and I went back and forth on this
3AR·/GAR· as explained in Etymologies, still seems plausible to one. In the end we split the difference. (See line 4 of the poem.)
me.- Editor -Editor

47
 A problem that I have encountered before is possible Line 3: (Nerte Atanin fairenen umbartar) :: When I
ambiguity in the texts, taken to mean whichever backs up first read this, it came out as "Nine for men fated by the
the speaker, but I think your "'inflection' (Tolkien does not phantom". Mortal men doomed to die" is a
call it a 'pronoun')" comment is the most outrageous pleonasm in any case.6
instance of this that I have seen! The element is clearly Atanin :: I'd have thought the more obvious translation
given as a pronoun, and the following phrase (i h3rar) is here for 'mortal men' was firima (QS, Ch. 12), giving
clearly dependent upon it for meaning, so the "passive
firimain here. 7
marker" bit is out
All the justification you give would appear to have no reference to mannar Vallon 'into the hands of the Lords' in the
rationale other than that you dislike *-nta and prefer preceding sentence, and 'the Moon and the Sun' referring to the
*toia, but as far as this writer is concerned, -nta wins, earlier objects Isil and U'r-anar), is a usage of a pronoun that we
hands down. "Ondorondontassen" is a bit lolloping, have been searching for ever since we leamed that ·nte was used
though ondomardintassen might be more wieldy.5 specifically when the subject has !1Q1 yet been mentioned. Since
there is no conflicting evidence for a different 3rd person plural
5 Appearances can be deceiving. I do not dislike *-nta per se. pronoun comparable in meaning and usage to toi attested in later
I would be willing to accept a phrase like *tlrienmnen I harar'in Quenya I cannot imagine why you {Craig} assume Tolkien
their keeping who sit upon' as a valid parallel to t1ruvantes I harar abandoned this one.
, they will keep it who sit upon'. But given this possibility, it is As for my emphasis on Tolkien's use of the word "inflection" in
remarkable that Cirion does D2t use this locution, even though T. his explanation of ·nte, I was not trying to infer that "inflection"
gives the exact English equivalent in his idiomatic translation of excluded the possibility of a pronominal inflection. I assumed that
the Quenya. Adifference in Quenya and English idiom is not the reader would know such an inference is specious. Since one of
surprising, and Tolkien seems habitually to display such the ways Patrick's proposal of ·nta differed from mine of tala
differences. But their im plications should be studied, not was that his was an infiection and mine was not, I was concemed
glossed over with statements like "I can't imagine why" . I with the opposite question of whether pronouns could come in
assumed (precipitously) that any student of Quenya pronouns both inflectional aru1 noninflectional forms in Quenya, and if so
reading Cirion's oath would have noticed Cirion's disfavoring of how extensively. If elye were exceptional, then Tolkien might
*·nta, and perhaps too tersely (and apparently unconvincingly) unambiguously refer to a pronominal inflection like -nte
offered my own explanation. generically as a "pronoun" (the same way Patrick, Craig and I all
It is a mistake to view the evidence of Quenya from The Lord do on occasion.) Tolkien's sticking to the altemative generic
of the Rings and later writing as the self-contained display of a "inflection" might p"o~sibly corroborate the existenc2 of
complete grammatical system. There are too many gaps, as has unrecorded pronouns that were not inflections. The case for this
been thoroughly demonstrated by analysis starting with An might be stronger if ·nte were only inCidentally pronominal in this
Introduction to Elvish and continuing from futher evidence, in context, since this would enhance the plausibility of some
the pages of Quettar and currently in Beyond Bree. Once the alternative or other. Hence my brief allusion to what else the
"deductive" ingenuity of the purists has panned out, the only suffix -nte might mean in parallel contexts.
place left to turn to for further insight into Tolkjen's Rnally as to Craig's suggestion that word order is supposed to
grammatical conception is to the pre-LotR evidence. Their distinguish between whether the noun the pronoun refers to
caveats are well-taken, and we certainly should not assume that comes before or after, I think the logical conslusion is that ·Ita =
every piece of Q(u)enya penned by Tolkien is consistent with 'their' somebody mentioned and -nta = 'their' sombody about to
LotR. But the opposite approach is also oversimplistic. Even if be mentioned, is the obvious interpretation of everything Tolkien
"an awful lot of the stuff ... is inconsistent", the objective wrote, insofar as it relates to possessive 3 plural inflection in
student will not dismiss evidence merely on the suspicion of Q(u)enya. This would leave the word free itself in its own
inconsistency. sentence, to be placed anywhere required by the sentence syntax.
Craig does not give any specifics regarding the inconsistency -C.G.
he finds in tal 'they (are), which (are),. The phonetic shape is 6 I agree this is a valid second interpretation of the line, a
valid: compare nal for diphthong in a monosyllable. The possibility sort of "subliminal" key to the spell, insofar as it worked on
of separate-word subject pronoun~ is guaranteed by elye 'even Mortal Men. Sauron of course is the "phantom". - C.G.
thou'. The use of a third plural pronoun simply to repeat or 7 I thought it best to avoid the redundancy of "mortal" and
reference some plural noun from the previous discourse (for the "to die". One of the things that makes this tolerable in the
two uses of tal at LR :72, the subjects are 'the Lords' in English is that the alliteration of mortal men and Iloomed to Ille

48
 fairenen :: Faire appears in the Classical Q. corpus Mordor :: (groan) [thank-you]. No wonder Jorge needed
("The Last Ark") as 'phantom, disembodied spirit', so I'd convincing! I'd prefer something else in here (from the
object to the use of a homonym here, in any case. 8 Classical corpus), but that's just personal taste.
umbartar :: Even after looking at the notes I still see
this as a verb-fonn. Some pleonasm or other is required to Line 6 (Mine Corma turie te ilye, Mina Corma
fill up this line, though, along the lines of 'doom, fate'. hirie te) : turie :: I'd agree with your last note (No.4)
I'd prefer ambartime here (i.e. whole line Nerte about this, though I'd say this was infinitive, rather than
Firimain ambartime) since their fate is not evil in any gerundial.
hirie :: ¥ou missed an opportunity here, I think. We
way, just supremely indifferent (to them).9
can't rhyme find/bind them, but if you used tuvie here,
you'd have three of the four verb-fonns used in lines 6 & 7
Line 4 : (Mine Morna Heren morna mahalmasse) :
as tu*ie.
mine :: I'd agree with the use of mine here (as opposed
to deriv. of "ER) due to minya.
Line 7 (Mine Corma tucie te ilye ar mi mornie
nute te) : nute :: typo, nutie.
(Whole line):: Very good; sonorous, very nearly a perfect
iambic metre.
Comment on "First Person Possessive in High Elven"
Line 5, 8 : (Mi Nore Mornandor yasse Mordor
I found the article interesting, though this may have been
caitar) : Nore :: A[nother] way round the "Land of Black-
for personal reasons as much as anything else. (I recently
land" problem you cite in your footnote would be to use
discovered that my surname, Marnock, contains such an
arda, since you're as much referring to the region as the
"embedded possessive" as you hypothesise for senya).
people.
The article basically comes down to a simple question: is
Mornandor :: Since this is a name, it would always be
senya literally "my child", or is "my child" merely a way
stressed on the second syllable, which I don't like. Also,
of idiomatically translating the Quenya? I would opt for
the use of the -ndor fonn would indicate the antiquity of a
the latter, i.e. that senya is as adequately translated by
name, which I don't feel likely here. Taking both of the
"dear child" or "beloved child" as by "my child". The
above into account, I'd have said M 0 rna nor,
evidence you give for your point of view is based on a
Mornanore. lO possible contraction of *yondo·nya to yonya. Yonya
splits the line into two units that the readers perceive as is not a piece of evidence on which I should like to base a
sequential and saying more than what they would perceive from theory about final-ior~ (LotR) Quenya - the two w'7>rds
mortal men doomed by mortality, which says nothing more than following yonya (which is explicitly glossed as 'my son')
men doomed by their mortality. In terms of how the alliteration are inye tye-mela, which are plainly not final Q.,
would strike the Quenya listener, saying 1irlmain 1airenen though there are close resemblances. This is the value of
umbartar with two related words next to each other, would have "The History of Middle-earth"; to provide us with
been like saying mortals by mortality doomed or giers to f!.Ying additional infonnation about the Eldarin languages of the
fated. I preferred the sort of mirroring of sounds in ner1e final stage - if that is what we are studying; if not, they
atanln and falrenen umbatlar with the center words not to link to provide a greater field of study - but they are not
closely associated in sense or derivation. - C.G. the same thing, and to (implicitly) say they are is incorrect.
S'Nhy? Do you believe Quenya has no homonyms? No where is yonya said to be a possible parallel, rather
9 Whatever ambarta· is in turun' ambartanen 'by doom than "the real thing". I don't object to the use of LR &c.
mastered' (looks like a noun to me), that is what umbarta· is
10 "Taking both of the above into account" is no more than
intended to be here, with just the negative prefix added. I think
taking either one individually. We do not regret chOOSing a three
where you go wrong is the assumption that a verb in phonetic
form cannot also be the form of a noun: cf. lanta·, both noun and syllable name that is stressed in the fashion of only a minority of
verb stem. As for the fate of men being evil, we were trying to such words in Quenya. We assume that Sauron was capable of
present the picture as Sauron would paint it: death an evil to be imitating (and affecting) the antiquity implied by using -ndor:
avoided by accepting a Ring, the subtext being subserviance to the specific analogies are Valandor and Numendor possibly the
the evil spirit of Sauron himself. I think this represents him as latter another suggestion by Sauron in imitation of the former.
the devious liar he must have been in person. - C.G. - Editor

50
to back up theories - I do it myself - but this piece of shorten the base vowel by analogy over time, but the
evidence is analogical, not actual, which is the way it's matter of onya is left largely unexamined. Under the
presented. 11 "haplology" you propose, the form of the substantive
I've said rather more on yonya that I meant to, but I would have to be **6n, a word which I ftnd unlikely. It
think "admissability" of the HME evidence is an couldn't be **6na or **6ne since for your haplology you
important (and contentious) point. What does it for me, would require the epenthetic -i-, i.e. that the word was
though, is Erusen. This clearly shows a long vowel originally *oniniii, so that the second -ni- could be
which is most unusual for Quenya (palantir is the only syncopated. 13
other I can immediately summon to mind, though there are In summary, I think that both senya and onya are
probably one or two others). This form clearly shows the derived in the way you suggest at the end of footnote 4 (p.
long vowel in monosyllables which Jan van Breda 14) where you say that, "Possibly onya ... was modelled
(Quettar, No. 24) calls "the ftrst vowel lengthening rule", directly on the verb stem on- 'beget' and the shape of the
which is seen more often in Sindarin than in Quenya The word yonya," i.e. that it contains -ya, the "suffix of
vowel in senya is short, which suggests to me that the endearment". (Undo my view, expressed above.) 14
elemant in senya is not the same as that in Erusen,
like primitive ·h~r vs. ·heru with short vowel whenever the root
though they are of course Closely related. Long vowels are
was manifest in a word of more than one syllable. This means that
not found in Q. syllables which end in two consonants; cpo
the oldest pattem for senya to conform to if primitive is the
are & asta, both from ...JAS. The fact that senya shows
short vowel in ·sen-ini-A since this is not a monosyllable. But we
a short vowel suggests that it is a separate form, and not a
contraction of *seninya. This is not particularly strong
should consider if we explore this issue noncentral to my theory,
and ask ourselves why Erus~n is plural. A possibility is pre-
evidence,12 since such a syncopation would probably LotR "plural m" given in Etymologies under ...J30- as part of
11 The word yonya was quite clearly identified as evidence from the explanation of Q. partitive ending -on, clearly retained in
Lost Road, at the very beginning of my article, so as to avoid any Quenya genitive plural endings like aldaron and elenion, with
confusion or specious inference on the part of the reader, such as phonological principle (PQ m > Q word-final n) also displayed in
Craig rather cantankerously suggests. The article is not, in any sandastan 'shield-barrier' < ·thandA 'shield' and ·stama- 'bar,
case, "a theory about final-form Quenya", as Craig supposes out exclude'. But this would suggest singular ·se> ·sA > Q ·se 'child',
of thin air. Rather, as indicated in my first footnote, I examine all whence the interpretation of senya as se + -nya would be a
the evidence in parallel, hypothesizing a unified conception "for natural source of analogy. But the rarity of plural forms like ·sen
the sake of argument", which means to find out what logical (if Erusen is idiomatic and not an archaism in Q.) might lead to a
deductions follow from the possibility that Tolkien had developed collective sense like EngHsh offspring has and coincidental
a conception of the 1st person possessive in writing The Lost interpretation of sen-va as sen + -va. This is all compatible with
Road which he retained in LotR and later. my thesis - C. G.
It is Craig who misleads the reader when he mentions things 13 We should not confuse phonologically determined
like "inye tye-mela, which are plainly not final Q'" This plainly is syncopation with isolated and analogical haplology. Leonard
inappropriate. The present verb-form mel a 'love' is ~ the Bloomfield (Language) gives as an example of haplology, Latin
same grammatical conception as sUa 'shines', the conjunctive stipendiurn < •stipl-pendiurn where the repetition pi pe is
pronoun inye 'I too' is probably the same conception as elye 'even simplified to one syllable even where the vowels in the two
thou', and for 2nd person object pronoun tye there is no late syllables differ, or else we can view this as shortening of ip ip, in
evidence one way or another. Where is the plain inconSistency which case ono-nya might be a source for onya. But I do bQl want
that Craig'S words infer? to suggest that I agree with you that *6n is "unlikely" as a Quenya
There is incidental evidence from Sindarin to corroborate the word-C.G.
retention of yonya in Tolkien's later conception, namely Gildor
14 The problem with Craig's explanations taken as a whole at
IngloriQn and ErelniQn 'Scion of Kings' (usually known by his
this point, is that by separating "Classical" Quenya, as a
sumame Gil-galad). That *Ion is 'scion' or 'son' here is confirmed
by the unpublished epilogue to LotR (on display at Marquette different conception from what went before, we cannot use
during Mythcon 1987) with the word lonnath used by Aragom Anardllxa as an explanation of yonya and yet with identical
to refer to the 'sons' of Samwise . ..:.. C.G. context and semantics to later senya, onya we would naturally
12 This is really ghost evidence. I think Craig realized his assume the same syntactic "explanation" or an intemal
reasoning was leading back to my theory when he changed connection, even if there is a partial external difference. This
subject. The theory of Jan van Breda would imply altemations suggests that T. had an explanation for yonya, senya, onya before

51
 You raise an interesting point with regard to Tarinya, part of your article running from the last two lines on
and one which I had not previously considered. I'd agree p.17., to the paragraph break in the second column on
with you here that the fIrst element is the titulus of the p.18., where you bring in aran as 'lord of a region' to set
Numenorean Rulers, but why it should be used for the up your translation of atar aranya. I would disagree
ruler's consort (the fIrst Ruling Queen was of course yet to entirely with your process where you class aran as a later
be born) is a question that needs responding to; in origin it form of haran by analogy with arda. Arda is from
is probably the mode of address for the King ("my liege" is --JGAR, whence also S gar, gardh, gardhon (S ard is
probably the closest we can get to it in English), and it from --JGHAR, or possibly --JGHA'RAT, whence also
later became the norm for addressing (probably) the arth, I think); harm a, har-, haranye are from
immediate Royal Family, though it was perhaps reserved *--JKHAR (I fmd the gh. (3-) > h· sequence evidenced in
for the Monarch and his/her Consort (and the Heir Elect?). the Etymologies a highly unlikely one for "Classical"
Something I've noted about your writing before is the Quenya); aran is a bona fide Quanya word from --JAR,
mass of sources you tie to something from all stages of the whence also arna - the fact that the only derivative of an
development of the Eldarin languages, and present them as extension of --JAR, --JARAT, is Quenya (arato) surely
being related internally rather than externally as different gives the lie to your etymologies. No nastiness intended!
manifestations of the same idea. 15 This is true for the These happenings & re-modellings are external, not

he thought of -ya used as mark of "endeannenr added to proper internal.1 6

names. Since he translates yonya 'my son', even if the idea of Actually, I think one of the more interesting aspects of
endeannent, clear in context, is primary (and it could be arar aranya was missed, the arrangement of it. Reading
m
secondary), he is aware of the sense of 'my' and so must have it brought to mind one of the comments in the note to the
thought of its relation to Atarlnya 'my father' and Indo-nlnya 'my oath in "Cirion & Eorl", where it is said "adjectives used as
heart'. For me the possibility of analogy underlying the internal a 'title' or frequently used attribute of a name are placed
shifts in inflection is an interesting one because we know of its after the name, and as is usual in Quenya in the case of two
abundant use in real languages but cannot always be certain of declinable names in apposition only the last is declined."
the plausibility of specific instances. I am naturally interested in This clearly has implications for atar aranya. Although
how traditional or intuitive Tolkien's ideas in this area are as aran is not an adjective, it would clearly (1 seem to be
illustrated by Quenya. This body of evidence has all the earmarks using the word "clearly" an awful lot} appear to be an
of a breeding ground for analogy: duplication of sound and attribute of atar. Although -ya is not part of the nominal
ambiguity of meaning. The sequence yenda : yo~ : atari.llYi > declension system, the same "rule" for words in apposition
yooma> yOOY.a : yoni!D :: II'IdoDY,a >I~ is such adassic and would appear to hold, so I'd say "(my) dear father and Qng"
typical example of this kind of change (if correct), that was more appropriate a translation than "milord father". 17
alternative possibilities without equal verve pale. This becomes general laws of probability, absent a showing that he threw out
an especially attractive analogy when it can be supported by a more Elvish words than he retained. I think the most logical
perceived ·nlnya < dative nln + adjectival-ya = 'of mine'. I assume methodology is to gather all the evidence on a particular subject
an intersecting internal development, a possibility T. seemed to before deciding if some of that evidence is invalid - C. G.
relish-C.G.
16 This is a distinction without a difference. There are no
15 So long as we stick to the words Tolkien actually wrote, "remodellingt internal or external. The only paradoxical change
then I think an honest but humble assessment of the relative is haran » aran which hardly requires a new explanation of harma.
linguistic skill of professor and student, will assume that a The root ··--JKHAR· explains nothing that the previous
pattern which the latter observes existed in the mind of the conception did not explain. There is no Sindarin evidence to
maker. It is when we resort to paraphrase and analogy on our own support this new etymology which therefore is unnecessary. The
that we must reserve attribution to the author, though merely to new word harar 'sit upon' seems to fit with the derivation of
the extent his own words can be better interpretted. To take hanna from ·ghar- 'have, hold', thus 'treasure' = 'thing held'; 'sit
those words and divide them in half and say everything written upon, occupy' = 'hold. possess'. And I see no obstacle to viewing
before 1940 should be assumed abandoned unless positively the new fonns aran, arna, arato as parallel in etymology to the old
proven otherwise, and claim that any theory based solely on post- words arda, armar, aryon, arwa - C.G.
1940 evidence is thereby more "likely· to be true: this sort of 17 I would say that if aran is a "title" or attribute here, then
methodology would be misdirected, and misconstrue Tolkien's 'lord father' would be a better translation than 'father and lord',
own methods of conservation and evolution of ideas, as well as the which is conjunctive not attributive - C. G.

52
'King' because of Letter #345, where Tolkien gives Comment on "Notes on
possible bull-names :: Aramund is glossed 'Kingly Bull', the History of the Elven Languages"
and is immediately followed by Tarmund, glossed 'NQbk
Bull'. It would appear that the distinction you spend half a I found your response to Stanley's article excellent, a good
page pointing out was later reversed. 18 I may be exposition of the facts (if that's the right word!) we have
misrepresenting you, but I couldn't help feeling that you concerning consonant mutation in Sindarin. It was
decided on your translation of atar aranya first, and then surprising how much could be deduced from "Pedo
went looking for proof. Not that there's anything wrong mellon", though I can't help thinking it would have been
a lot simpler if Feanor had provided his tenkele with
with that - I do it all the time! 19
quotation marks!
Finally (three pages later!), I'd like to come to -ninya.
I've often tried to come up with an etymology for
You give an ingenious origin for it but (ref. my comments
danwedh, without much success. I liked your solution,
about yonya) I don't think it can be placed here in a study
but don't quite agree with it. NDAN- is 'backwards,
of later forms. I feel you have been affected overmuch by
contrary', as contrasted with ATCAT)- which is 'back,
the resemblance between -Ci)nya and -ya Cwhich to be
again'; also I had assigned WED- to being an external
sure, do share similar origins). My final table would look
predecessor of *.,JWAD, whence fmal-form Quenya vanda,
like this: though your etymology sounds very good, and I can't think
of anything better.
-ya : Suffix of Endearment Another etymological question is raised by Haudh-en-
Ndengin, which clearly shows a singular article. I
Anardilya, atar aranya, senya, onya couldn't swear to it, but I think this is the only example
we have of this. The Etymologies give a word dangen
'slain', but follows it with Haudh i Ndengin, clearly
{that word again!} showing a plural form. I'd explain this
-Ci)nya : 1Sl Person Possessive as a plural noun which has come to be used as a singular
(or at least not explicitly as a plural). I'd derive an
hildinyar, Atarinya, Tarinya adjective *ndakna from .,JNDAK C+ *-na, see arna,
"Lost Road forms" : aina &c.), meaning 'slain'. This came to be used as a
yonya, Anarinya, hondo/indo-ninya noun meaning 'some-one who has been slain' (the other
derivatives of the n-extension would not have such a
meaning: *ndakni ~ould mean 'that which slays':-'and
18 I cannot find any reversal here. Aramund is purely *ndakno 'one who slays' (see in this respect the contrast
metaphoric and alludes merely to the royal nomenclature of Arnor between hilde & hiIdo), i.e. a lethal weapon of some
in the Third Age, where the idea of territorial kingship of sort, and a killer). This would be *dang in S., and the -
lordship of a territory or people, was the only form of absolute in plural of it gives us the required word. (It is hard to say
leadership known. Tarmund does not refer to anything more than at present whether the ending causes secondary affection, or
inherited "highness·. Among Men it referred to inherited divine not;20 if not, the fact that it occurs here points to the
blood, and the right to rule those Men of extended lifetime form's age as a unit.) The word's being a unit may have
granted Numenor to live in. In their "kingdoms· in exile, the occurred later (as you suggest in the article), or it may have
lineage of tarle had dwindled in the public knowledge to a mere already been so at that time.
legend. In the case of application to bulls, perhaps descent from
the "Kine of Araw· might be the fairy-tale allusion that is Comment on "Chart of Body Parts"
suggested - C.G.
19 The meanings I earlier considered for atar aranya are on I liked this - a different kind of article. Actually, looking
record in Parma #6, so your speculation here is idle, more
revealing of your own methods than penetrating into mine. In any 20 I'd say no; lack of vowel affection is partly the point of using
case one has to go back and forth between theory and evidence, this plural form as opposed to the more normal mutation plural.
until one is familiar with everything written by Tolkien and every I'd say (cp. UT Index) that cerln is not fr. cOr +·In but is a later
possible connection worth considering, before one ought to stop "manifestation" of the word karin appearing in BoL T1
looking for clues - C.G. Craig's note

53
at it, I was surprised by just how much physical vocab. din:: Where is this from? [The epilogue to LotR
there was. It always seems that you can never find a which T. decided not to include in the published text -
specific word when you want it, whether or not it appears Ed.]
in the corpus. What annoys me, though, is the durinthi :: I don't agree with the lenition here, but it's
unevenness, e.g. the number of words for 'darkness, gloom' hard to give reasons why.
&c. -there's a multiplicity available in this or that area,
but none for the others. Line 5 : odogranaithos :: I like this. Who cares if it
may not be totally exact? (I don't think you can impose
Comments on 1m naitho those sorts of rules Qn something synthetic, in any case.)

I'm not really into Sindarin (or this synthetic "Low- Line 7 : Nalos :: I wouldn't say this was a verbal form,
Elven" - rather a good term, I think. What about an as seems to be required.24
Elvish term for it? How's about Sundarin? (ouch!)), so
the following are more meta-comments (on your notes) Line 9 : NOTES:: The S. in "The Lay of Leithian
rather than comments on the poem itself (though it does Recommensed" is annoying, in that we can't say for sure
seem that most of the poem is commented on, reappearing what it means. But I'd disagree with the identification of
in the notes). I think it would have been helpful were the the Moon with the giIthonieI for several reasons.
source-text to be identified, rather than letting the reader Firstly, the word implies making; this is the meaning of
find it for themself. 21 the second element, equivalent to -tal- in Tintalle,
where the -n- has assimilated to the following -1-; cpo
Line 1 : 1m naitho :: I'm not quite sure what the last tano, Cirdan, Etym: TAN -. Secondly, the -iel
sentence of the comment on this phrase. I appreciate most element is feminine. A later alteration by Tolkien made it
of the other points you make, but an objective pronoun non-specific, but at the time of the formulation of the word
followed by an imperative looks strange whichever way it was feminine; no examples are in fact found of -iel
you cut it. The obvious translation of 'I will lament' marking male names (not as far as I can remember,
(though it lacks any sense of compulsion, which is I think anyway). So the name means "the female maker ('kindler'
the reason for the form shown here) is naithathon. 22 in this sense refers to the original making of the star, when
it first began to shine, not to when they shine anew each
Line 3 : uf :: I think uf implies motion of some sort. night) of stars", which obviously excludes Rana. 25
It's not clear if this is so.23 room; looked . . .the Window. Offhand I cannot think of any
preposition that "implies motion". Usually they tend to be
21 I was writing for the reader who was interested in figuring directional or relational so as to accommodate usage with a
out the poem as a poem rather than as a translation, for it was variety of verbs - C.G.
not intended to be an accurate or even representational
241 totally disagree with this. Tolkien gives the gloss 'sinking,
"translation" of anything in English. I was trying to write as a
setting, slope', which cleverly and clearly shows that this word is
Sindarin speaker might write, at leasure in Tol Eressea, as I tried
both verbal noun and "concrete" noun. The glosses show this, as
(rather feebly) to explain in my final note - C.G.
one is purely verbal 'sinking', one a form whose participle/gerund
221m is not objective, but subjective: 1m Narvi haln echant 'I has verbal (setting sun) and just barely verbal (setting of the
Narvi made them', in the gate inscription. On the use of stage/play) meaning, and one hardly verbal at all: look at the
imperative with explicit subject, T. says that a verb-form ending (mountain) slope. The idea of 'sinking' as if going into a deep
in -0 is the imperative "of all persons·. We know from multiple place without being able to help it, seemed appropriate here, not
examples that the imperative verb by itself is used for 2nd
only for the imagined fate of Tuor in Hells of Iron, but also
person. So the imperative in the other persons seems to require
echoing what happens to her father below her in a tower falling
personal pronouns to distinguish them, which is what the last
sentence of my comment on 1m naitho means. I would probably in -e.G.
any case never use nalthathon in a poem, because it sounds 25 I too ascribed to the interpretation that Tlntalle contained
atrocious - e. G. tan· until the form t1nta 'cause to sparkle' was published. I
23 The glosses of this Gnomish preposition are 'out of, forth assume that in Tolkien's final conception the very primitive
from' (BLT2, under Ufedhln). These compound prepositions in element "ta- meant 'make'. kly phase of a "language" of
English can both be used with or without motion: walked . . .the speaking peoples will include the idea of creating things (and

54
Line 10 : lin:: I'd agree with lin as analogous to nin, given in LotR as "Limlight". Its chequered history (i.e.
but as far as I remember the second pers. forms in I were of the name) is discussed in note 46 to "Cirion and Eorl"
borrowed from Quenya, used as "reverential" or some such. (Un where it seems clear that the fIrst element is Elvish,
I doubt (if I recall correctly) if Tuor would say this to ldril. whatever the 2nd is, or was. An earlier note would give its
The situation in S. with regard to 2llil pers. pronouns is meaning as 'swift', which would either be understood as an
probably like English, in that the originally plural form is epithet for Asfaloth or as an adverb, which seems suitable.
used as both pI. & sing., and (although by different routes Your note on hin. I'd connect this with the base given
in each case - note however that each are connected with in Etymologies as SI-, being a plural, lenited form of
religion, something which was a concern of JRRT) the *sen, which is < *sina 'this (adj.)' which also> Q sina.
sing. forms have a "heightened" sense. 26 The alteration of the -e- to the -i- is probably by analogy
(lin) I found the idea that lim was 'thyself a nice one. with the adjectival ending -en (e.g. remmen, &c.),
A well-known phrase in the Glasgow dialect is "Gaun though some kind of "haplology" as you suggest in your
yerseU", a term coming from soccer, meaning roughly 'go "First Person Possessive" article may also be present.
for it!', 'go on!'. I would connect lim with the river-name Another possibility is that the form is a unique one; the
possible association with e(n) is a valid one.
ideas), so an element as pervasive as this (there are numerous
derived verbs in the element -ta) is logically explained by a basic Line 11 : fuinon:: As far as I can make out, you're
idea like this. The sense that to make or 'build' a building or using this word adverbially. I don't know if this [is] likely
artifact, involves 'raising' the materials, suggests an association or not for S(undarin), though I can't fInd any objection to
with the ta- words for 'high'. But the basic idea of 'make' implies it on any literal grounds.
both 'cause' and 'do', as well as 'create'. The variety of
derivatives in tao, tan-, taro, is not surprising. The idea of Line 13: NOTES:: I don't agree that the 'by' can be
'kindling', which connects with the (extemally) old idea of the assumed from word-order. I think some particle is more
"Tale Fire" (TOn a Gwedrln), may be the intemally oldest idea likely in here. This is hard to render, though, so some
that T. envisions associated with tan-. other paraphrase would be a better move. Not really -
The suggestion that -lei is feminine for the post-LotR lay
what about "na,,?27
recommenced, is contradicted by mirlel'shining like jewels', -dlrlel
din:: I think you get this from Timlviel's Song. But if
'gazing', which show that gender of derivative names is
it is pronominal, it would be "him" or "his" (poss. n.), not
irrelevant. The word gllthanlel is uncapitalized in the Lay and
could easily refer to the Moon temporarily hiding and then the possessive adjective, despite its suitable form. 28
"rekindling" the stars as it passes with its greater light. 'th.
know (and perhaps the scientifically minded Elves knew) it is only Line 14 : gruinon :: I don't like the addition of -on to
the relative intensity of the Moon that seems to dim the nearby 27 It is not a question of what is more likely. As I said in my note
stars and that it passes between us and those stars that it "lamentation of his wife" ~ mean "lamentation by his wife". It
completely obscures. But to Luthien perhaps it seemed that the
can also mean "lamentation for/about his wife". Both meanings
Moon "rekindled" the stars after snuffing them as it passed - are possible, and neither is more likely than the other. Together
c.G 'by' and 10r' constitute the QD.!J: possible interpretations of the
26 I do not think that the fact that religion was a concem of genitive with these particular nouns. These are facts, not
Tolkien's lends much weight to the assumption that the variety of assumptions: simple observation about the way the genitive
Sindarin "used by the High Elves ... marked in high style and works in all languages that have it - C.G.
verse by the influence of Quenya, which had been originally their 28 I do not follow Craig'S reasoning here. But anyway I use the
normal tongue", which included the dative pronoun "Ie, the
word the same way it is used in the unpublished material, to
reverential 2nd person sing." (NT: 64-5) was ousted from
modify another noun. The word in the Lay occurs with a verb, si
nonreverential function by the original plural 2nd person. Since lath a galadh lasta din! something like 'now blossom and leaf listen
we have no example of the 2nd person plural in Sindarin, it seems ... !' Possibly this blank could be filled by 'to this' referring to
pure guesswork to speculate about its interaction with the the song itself, and conceivably this could be connected with din
singular. In any case, though I adlT\it that J do not really know what 'his'. But perhaps it would be likelier to connect the former din
it is like to be married to an Elf, I cannot imagine a grammatical with Aman Din 'Silent Hill' (and maybe dI-nguruthas
restriction on the way Tuor talks to his beloved Idril, that 'overwhelmed in dread of death, beneath death-horror'). The
requires him IlQ1 to address her with reverence - C.G. sense of lasta din would be something like 'listen in awe'- e.G.

55
 .................. '"
" ................ '" .... ; ~

an adjective; in both of the cases you cite, it has been added

~~
to nouns. 29
Line 15 : sein-hael :: I like this, so I'll restrain my
urging to nitpick about sael being used. .................. :
Line 16 : ruithant:: An interesting rationale, though
there would have to be a lot of verb stems ending in -at
for it to be plausible. 30
urant:: I think that uradant would sound better here,
but that's just personal taste} 1
Line 18 : oronte :: I'd derive erio fro *orya-. "orie
is newer than ortie, some sort of analogical reshaping" :
What does this mean?32
29 I was partially supporting this by the noun ·ruin 'fire' in
Orodruin - C. G.
30 It is a common misconception that linguistic analogy must be
based on numerous examples. In fact a single form reinterpreted
is theoretically sufficient to produce a new marker for an old
category. An example comes right from English. The universal
3rd person singular present tense ending is now ·s, but the Q!ljy
such form that goes back to Old English is is. The 3 sing. pres.
was in all other verbs expressed by ·th or·t .
The sole evidence of thelthant is sufficient "rationale" for
rulthant. We know from this that the ending ·ant was used for
some verbs in Sindarin. What I was trying to do in my discussion
of echant was remind the reader of the implications of
Etymologies about this form, and the possible source for ·ant
which this information entails. Nor is this all the evidence of this
kind of development in Noldorin: cf. trevant pa.t. of trevedi
'traverse' (BAT), hant pa.t. of hedi (KHAT· 'hurl'), pant 'full'
beside pathro 'fill' (KWAT), atlant 'oblique, slanting' beside
atlanno 'to slope, slant' (TALAT). Since we do not have to
assume ·ant was universal, I think these are sufficient forms to
render plausible the analogical development of teithant, orthant,
gwedhant, etc. - C. G.
31 That would also destroy the scansion - Ed.
32 Because Exilic Noldorin erlo 'rise' is closest, among the Old
Noldorin verbs in form and sense, to orle (but as you pOint out
does not derive directly from it) there is reason to assume
analogical erl-o, of which erl· comes from orle. This implies that
ortle> orle is the correct chronology of these forms (unless we
assume artie is a "faddish" form devised in Old Noldorin and
dying out there, never competing successfully with orle), or at
least the more plausible chronology.
The precise analogy would be something like this: ON oro : f·· .. ··············~~ ,
. . . . . . . . . . . . . . . . . u . . . .'
.
'mountain' might suggest the divisi~n or·t6be, while orot! :: ~ ~
.: ::
'mountains' might suggest ort-6be. The simultaneous existence ....
of these three words would suggest the altematives ort-/or·, .
!:
:

and either of ort·le or or·le from the other by analogy - C.G.

56
Gr;shtl~kh, tAke t:\ leiter: "Four RinJS
for -the. Elveh.kin:Js" •.. 110, no, no,
better make 'thAT "Two R,n~s~'
mA~ [" Ive "lniJs ,••.
be "~o· ,-
"0,
~_""'SIl