Ilkorin - a "lost tongue"?

    Originally, Tolkien did not imagine the history of the Elvish tongues in Middle-earth quite like the scenario we know from the published Silmarillion. According to the Silmarillion and other post-LotR sources, the Noldor and the Vanyar went over the sea and developed Quenya in Valinor, while some of the Teleri stayed in Beleriand, where their tongue evolved into Sindarin. But in Tolkien's older conception, Quenya was the language of the Vanyar (long called "Lindar") only; the Noldor developed the language that the Etymologies calls "Old Noldorin", while the tongue of the Teleri that were left in Beleriand evolved into Ilkorin. The Teleri in Beleriand were not called Sindar or "Grey-elves" in Tolkien's older scenario; they were Ilkorindi, "those not of Kôr" (a central place in Valinor). When the Noldor came from Valinor speaking "Old Noldorin", their language was influenced by the indigenous "Ilkorin" tongue and underwent drastic changes to become "Noldorin". (The last flicker of the idea that "the two tongues grew towards one another", before this concept was finally abandoned, is found in WJ:21, 24 - where the tongues in question are already Quenya and Sindarin.) Tolkien even considered the possibility that the languages became amalgamated. See LR:346.

    But further revisions were to follow. Tolkien decided that the "Noldorin" language he had already made was not the language of the Noldor after all. In fact, it turned out to be the language of the Teleri that had remained in Beleriand, who came to be called Sindar (replacing the term Ilkorindi), so Noldorin had to be renamed Sindarin. The Noldor, that were now conceived as speakers of Quenya just like the Vanyar, simply adopted the Sindarin tongue when they came to Middle-earth. There was no complex process of mutual influence and amalgamation. In other words, "Noldorin" > Sindarin usurped the place of Ilkorin as the indigenous Beleriandic tongue. So what happened to the Ilkorin tongue - did it disappear from the mythos altogether? Many words and names that Tolkien originally held to be Ilkorin, such as Esgalduin or the name of Túrin's friend Beleg, survived in the narrative texts - but after the revision they must probably be understood as Sindarin forms instead. Interestingly, the name Esgaroth known from The Hobbit occurs in the Ilkorin wordlist. Elrond was also thought of as an Ilkorin name when Tolkien first made it, but in the context of LotR it can only be Sindarin. The status of Ilkorin in the mature mythos is thus very questionable. Edward Kloczko has argued that Tolkien, rather than scrapping Ilkorin completely, turned parts of it into the obscure "northern dialect" of Sindarin, the tongue of the Mithrim; his original article is reproduced as an Appendix.

    Elementary Phonology

    The general sound of Ilkorin is reminiscent of Sindarin. As in Sindarin, the original unvoiced plosives have become voiced following vowels (e.g. t > d in adar "father" from the stem ATA; Sindarin also has adar; contrast Quenya atar). Original voiced plosives may become voiced spirants in the same position; cf. b > v in tovon "lowlying, deep, low" from the stem TUB; this is also the case in Sindarin. As in Sindarin, plosives generally become spirants following the liquids l, r: compare t in tôr "king" with th in balthor "Vala-king"; compare also dh in erdh "seed, germ" with d in the primitive form *eredê (stem ERÉD). The word erdh also illustrates one of the differences between Ilkorin and Sindarin. In primitive Elvish, there were many three-syllable words where all three vowels were the same except that the last was long, like *eredê "seed". Quenya dropped the middle vowel and shortened the last, *eredê yielding erde. Sindarin dropped the last vowel, *eredê becoming eredh. But Ilkorin dropped both the middle and the final vowel, so that *eredê became erdh.
              Long and short vowels are distinctive, as is demonstrated by the minimal pair gwen "girl" vs. gwên "greenness". In monosyllables the long vowels are marked with a circumflex, e.g. côm "sickness", môr "night", ôr "blood". In polysyllabic words, the accent is used instead; compare tôr "king" with its plural form tórin (not *tôrin). This system is also used in the spelling of Sindarin; in that language, it denotes that in monosyllables, long vowels tend to become especially long. This may also be the case in Ilkorin.

    The Noun

    Two inflections are found in the material, genitive and plural. A genitive ending -a is seen in the phrase Tor Tinduma "King of Twilight", a title of Thingol (THIN, TIN; cf. tindum "twilight"). The plural ending is -in, seen in adar "father" pl. edrin (ATA), aman "mother" pl. emnin (misreading "emuin" in the published Etymologies, stem AM1), Balthor *"Vala-king" pl. Balthorin (BAL), boron "trusty man" pl. burnin (BOR), gangel "harp" pl. genglin (ÑGAN), tôr "king", pl. tórin (TÂ/TA3, BAL), talum "ground, floor" pl. telmin (TALAM), thorn "eagle" pl. thurin (THOR/THORON). It will be noted that the ending -in causes a to umlaut to e, and in the case of polysyllabic words, the vowel in the syllable preceding the ending is lost (adar > edrin, aman > emnin, talum > telmin). In one attested case, plurality is shown by umlaut only, and the ending -in is not used: tal "foot", pl. tel "feet" (TAL).
              The vowel o becomes u in boron > burnin and thorn > thurin (read *thurnin?), but not in tôr > tórin (evidently because this ô is long and represents earlier long A).
              A plural genitive ending -ion seems to occur in the phrase Dor-thonion "Land of Pines"; cf. also thurnion "of eagles" in Torthurnion "King of Eagles". Cf. Quenya -ion.

    Verbs

    Only five verbs are known, and there isn't much we can say about them. All end in -a: góda- "to soil, stain", taga "he fixes, constructs, makes", tingla- "sparkle", toga "he brings", tolda "he fetches" (see wordlist below for references). As we see, taga, toga and tolda are glossed as 3. person masculine singular present-tense forms, while góda and tingla are glossed as infinitives. Since they display the same ending as the other forms, it seems likely that they, too, are really 3. pers. sg. present-tense forms - *"he soils, stains" and *"he (it?) sparkles". In Ilkorin, the 3. pers. sg. present tense may be the simplest form of the verb and is therefore used as the lexical form. In three out of five cases, Tolkien translated the Ilkorin verbs literally; in the remaining two cases, he used the English lexical form, the infinitive. The ending -a for "he" and "present tense" may indeed represent older endings where an explicit pronominal element was present; see taga in the wordlist below for further discussion.

    Past Participle

    The word thúren "guarded, hidden" from the stem THUR- "hedge in" seems to argue the existence of a past participle in -en, presumably the cognate of Quenya -ina. Notice that the stem-vowel is lengthened; this means that where the original vowel was a, it would change to ó in the past participle (since A, when it was long already in the primitive language, became long O in Ilkorin - see for instance tôr in the wordlist below). Original O in the verbal stem would likewise come out as long ú.

    Ilkorin vs. Doriathrin and Sindarin

    From the above it is clear that Ilkorin is very similar to Doriathrin, as would be expected since Tolkien conceived them as closely related languages. For instance, the two tongues share the plural ending -in and the genitive ending -a. Doriathrin and Ilkorin should be considered closely related dialects of the same language; indeed Tolkien sometimes seems to use the term "Ilkorin" with reference to all the dialects of Beleriand, including Doriathrin.
             But what is the relationship of Ilkorin to Sindarin, the language that usurped its place in the mythos? It is interesting to notice that when "Noldorin" became Sindarin, Tolkien did certain revisions that in some respects made the language somewhat more similar to Ilkorin. For instance, in "Noldorin" primitive initial l- and r- turned into unvoiced sounds lh- and rh-, but in Ilkorin, these sounds were unchanged - and this is also the case in Sindarin. Another change affects final w following a consonant: In "Noldorin" this sound remained unchanged; in LR:398 s.v. WEG, Tolkien noted that the "Noldorin" cognate of the Quenya name Elwë would have been Elw, but that no such form was in use. But in later Sindarin, we do find a Grey-elven form of the name Elwë, namely Elu - and this is also the form we would have expected in Ilkorin, in which language -w in this position did turn into -u. Cf. such an Ilkorin word as adu "double" as the cognate of Quenya atwa, LR:349 s.v. AT(AT). This word occurs in the Silmarillion as part of the river-name Adurant *"Doublecourse", and since in the published Silmarillion the language of Beleriand is Sindarin and not Ilkorin, adu must now be considered a Sindarin word - though it would probably have been *adw in Tolkiens earlier "Noldorin". Indeed one reason why Tolkien made "Noldorin" a little more similar to Ilkorin, thereby producing Sindarin, may have been that he wanted to keep many of the long-established names of places and people in Beleriand more or less unchanged - and this would not be possible if "Noldorin" was made the indigenous language of Beleriand just like this tongue had appeared in the Etymologies. So while "Noldorin" did in one way eat Ilkorin alive, it was not unaffected by this; Sindarin as we know it from later sources in some respects looks like the earlier "Noldorin" with an Ilkorin substrate!

    ILKORIN WORDLIST with etymological notes

    Ancestral forms "reconstructed" by Tolkien himself are not asterisked. In the primitive forms, long vowels are marked by circumflexes; the sources use macrons instead.

              -a genitive ending, seen in Tor Tinduma "King of Twilight", a title of Thingol (cf. tindum "twilight"). The primitive Common Eldarin genitive ending was -hô > -ô, derived from an "ancient adverbial element" HO meaning "away, from, from among" (WJ:368). The corresponding entry in the far earlier Etymologies seems to be 3O (3Ô) "from, away, from among, out of" (LR:360). Could primitive -ô come out as -a in Ilkorin? There are a few words where -ô may seem to develop in such a way (see adda, broga), but normally, final -ô is lost like other final vowels. - In the plural genitive ending -ion, the "genitival" element (< 3O or HO) appears as o; see -ion.
              adar "father", pl. edrin. Derived from a stem ATA, itself simply defined as "father" (LR:349). The Primitive Quendian form is given as atar, that must be seen simply as an extension of the stem itself. As in Sindarin, unvoiced stops (p, t, k) are voiced (to b, d, g) following a vowel, hence atar > adar. - The plural form edrin shows umlaut a > e, caused by the i of the plural ending -in (concerning which see separate entry). Also notice the syncope of the second vowel of adar in the inflected form. For similar contractions, compare aman pl. emnin, boron pl. burnin, gangel pl. genglin (q.v.)
              adda "father" (possibly hypocoristic, = *"dad"). Derived from the same stem ATA as adar above (LR:349). It would be tempting to believe that this is the cognate of Quenya atto, primitive *attô, evidently the stem ATA with a medial fortification t > tt and the masculine ending -ô. If so, the form adda would suggest that final -ô becomes -a in Ilkorin. Compare broga "bear" from primitive morókô - but in other examples, final -ô is lost without trace (see for instance benn, ber). Moreover, the word adda would seem to indicate that following a vowel, also double *tt becomes voiced dd in Ilkorin. Contrast Sindarin, in which language only single t would be voiced in this position, while *tt becomes th instead - cf. for instance peth "word" from earlier kwetta. However, another Ilkorin word does show a Sindarin-like development of *tt: brith "gravel" from b'rittê. It may be, then, that Ilkorin adda is not really the cognate of Quenya atto, but rather an affectionate form based on adar "father", formed later and not directly descended from the primitive language. If we see adda as the cognate of High-elven atto, we would have to assume that Tolkien changed his mind about the phonological evolution of Ilkorin while he was writing the Etymologies (which is of course not inconceivable; the editor in LR:346 refers to "divergent forms...between one part of the Etymologies and another").
              adu, also ado, "double". Derived from a stem AT(AT), which Tolkien defined as "again, back" (LR:349). It would seem that this is basically a simple stem AT that often appears as ATAT, the reduplication symbolizing the repetition. Adu, ado must come from simple AT, though. These Ilkorin words for "double" are apparently cognates of Quenya atwa of similar meaning. The primitive form, transparently meant to be *atwâ, combines the stem AT with the adjectival ending -wâ (concerning this ending, cf. for instance narwâ "fiery red", derived from the stem NAR1 "flame, fire", LR:374; see also alch, laig). After the loss of primitive final -â, the final semi-vowel of the resulting form *atw apparently turned into a full vowel -u, *atu then becoming adu after the voicing of post-vocalic *t (cf. adar from atar). It would seem that adu subsequently became ado (a similar change of final -u to -o occurred in Common Eldarin, but this Ilkorin development must be later). That this change occurred only when -u was final is suggested by the compound Adurant (rather than *Adorant) for *"Doublecourse", name of a river in Ossiriand which for a distance had divided streams (concerning the second element, see rant).
              alch "swan" (ÁLAK). Primitive form given as alk-wâ, derived from a stem ÁLAK "rushing" (LR:348). Alk-wâ would seem to be an adjectival formation (concerning the adjectival ending -wâ, see ado above). The primitive word was evidently an adjective with the same meaning as the stem: "rushing", later used as a noun "rushing (one)" and applied to an animal. Tolkien may have imagined that after the loss of the final vowel, the now final kw was de-velarized to k, the resulting form *alk then turning into alch because p, t, k following a liquid became spirants f, th, ch (= kh, [x]), just as in Sindarin (see UT:265, footnote). Compare words like Balthor, erdh, salch (but contrast the word tolda < tultâ-, where lt mysteriously becomes ld instead of lth; perhaps this is because this cluster lt occurred already at the very oldest stage, whereas the other combinations here listed arose only later, as a result of syncopes or compounding). Notice that kw was not de-velarized initially, so we have for instance cwess "down" (noun) from primitive kwessê (LR:366 s.v. KWES). In this respect at least, Ilkorin cannot function as a dialect of Sindarin after Tolkien's revisions of the history of the Elvish languages: In Sindarin, indeed in the entire Lindarin branch of the Elvish language family, primitive kw very early became p (WJ:375, cf. WJ:407 note 5). Alk-wâ so early became *alpâ that if we are to make Ilkorin a Lindarin language according to Tolkien's later ideas, this word would have had to become *alf, just like in normal Sindarin (in that language preferably spelt alph). Compare salch.
              aman "mother", pl. emnin (obvious misreading "emuin" in the published LR). Derived from a stem AM1 (LR:348), simply defined as "mother". Aman must represent an "extended" stem *aman, sc. AM1 with suffixation of the stem-vowel (by a Quenya term ómataina, vocalic extension) and a suffixed consonant -n. Compare boron, q.v., from BOR. The plural form emnin shows the same contraction and umlaut as in the examples adar pl. edrin, gangel pl. genglin.
              Argad "outside the fence", lands outside Doriath (the "fence" obviously being the Girdle of Melian). Also explicitly Argador "land outside the fence", sc. outside Doriath. (The forms Argad and Argador are mentioned in LR:349 s.v. AR2, Argador also in LR:358 s.v. GAT(H).) Concerning dor "land", see separate entry; as an Ilkorin word, the element gad "fence" is attested in this compound only, but Doriathrin has the same word: See gad in the wordlist appended to the article about Doriathrin for etymological discussion. The element ar- "outside" that is prefixed to Argad, Argador is derived from the stem AR2 (LR:349), not itself defined but probably meaning much the same as the Quenya preposition ara "outside, beside", the first word listed in this entry in the Etymologies. Compare the entry ar- in the Silmarillion appendix. In Etym it is stated that in Quenya and Ilkorin alike, this element was "purely local in sense", referring simply to spatial relationships. This was evidently the original meaning of this element as well, since it is further said that in Sindarin ("Noldorin"), this element developed a privative sense ("without"), as in arnediad (arnoediad) "without reckoning, numberless". Ilkorin and Quenya thus preserve the original meaning of this prefix.
              arn "red". Derived from a stem YAR, defined as "blood" (LR:400, the Ilkorin word for blood, ôr, is indeed derived from the same stem). The primitive form is undoubtedly meant to be *jarnâ (*yarnâ) with the common adjectival ending -nâ (sometimes used to derive passive participles); the primitive word must clearly have implied "bloody" or "blood-red". Perhaps this association would no longer be so strong in Ilkorin, the phonological evolution having made the words for "red" and "blood" somewhat divergent in form; the words arn and ôr are not as obviously connected as their primitive counterparts *jarnâ and *jara. Notice that the final -â of *jarnâ is lost without trace in arn; more remarkably, the initial j has also disappeared (as in ôr "blood").
              Aros name of river with reddish water, the southern river of Doriath. Derived from the same stem YAR "blood" as arn "red" above (LR:400). The "Noldorin"/Sindarin cognate is given as iaros, suggesting that the stem YAR as such (rather than a derivative like *jarnâ) was once prefixed in this word. The second element of the name, -os, is obscure. See the name of another river, Thalos, for some thoughts about this ending.
              ascar "violent, rushing, impetuous"; also Askar as name of a river (one of the tributaries of Gelion). The difference in spelling (c/k) seems unimportant; in the published Silmarillion, the name is spelt Ascar. Derived from a stem SKAR (LR:386), the root-sense of which is given as "tear, rend". The primitive adjective yielding ascar is given as askarâ, with prefixing of the stem-vowel (probably intensifying the meaning) and the common adjectival ending -â. As usual, final -â is lost without trace in Ilkorin.
              ass "cooked food, meat". Derived from the undefined stem AP (LR:349). The Quenya and "Noldorin"/Sindarin cognates (apsa, aes) would seem to indicate a primitive form *apsâ. There is not much more we can say about this form or about the ending -sâ; a few Quenya words do show the ending -sa, but they seem to have little in common semantically (in addition to the noun apsa, the cognate of ass, we have for instance the frequentative verb lapsa- "lick" and the adjective telepsa "of silver"). We may speculate that AP is a verbal stem *"cook", and that *apsâ is properly an adjective *"cooked", later used as a noun "cooked thing, something cooked" > "cooked food". For the assimilation *ps > ss in Ilkorin, compare tuss "thatch" from primitive tupsê (LR:395).
              Balthor, *"Vala-king" = Vala (pl. Balthorin is given). This form, listed in LR:350 s.v. BAL, contains the same elements as Quenya Valatar (Valatár-), though this Quenya word has a somewhat more specific meaning: it refers to one of the nine chief Valar (corresponding to the Aratar in Tolkien's later works), while Ilkorin Balthor just means "Vala" in general. The second element -thor is simply a form of tôr "king" (q.v. for etymological discussion); since tôr hear appears as the second, unaccented element of a compound, the long vowel becomes short, and since t here follows an l, it becomes th (see alch). The first element of Balthor represents bálâ, "Power, Vala", derived from the stem BAL, itself undefined in Etym (LR:350), though Tolkien suggests that the stem BEL "strong" is related (LR:352). A later source states that the Quenya word vala is properly a verb "has power" that was also used as a noun "a Power" (WJ:403); this may provide a good clue to the basic meaning of the stem BAL. The final vowel of bálâ "Vala" dropped out in the compound that produced Ilkorin Balthor; a primitive form is given (in LR:350) as bal'tar-, the apostrophe indicating the syncope. Actually the second a (in -tar) must have been long â for it to become ô > o, unless we are to assume that the word at one stage appeared as *baltr with a syllabic final -r, in front of which a vowel o subsequently developed (compare ungor < *uñgr < *uñgrâ). But if so, we would expect the plural form to be **Balthrin rather than Balthorin. Notice that there is no umlaut o > u in the plural (not **Balthurin, contrast for instance boron pl. burnin), indicating that o derived from another source than primitive *o is not affected by this umlaut.
              Balthronding name of Beleg's bow, more commonly called Belthronding, q.v. for discussion (LR:388 s.v. STAR, subentry STARAN).
              basgorn (for bast-gorn) "round bread". Mentioned in the Etymologies in the entry for the stem KOR "round" (LR:365); the element -gorn can readily be matched with the primitive adjective kornâ there listed (concerning the adjectival ending -nâ, see arn, caun). In Ilkorin, kornâ by itself would appear as *corn, but in a compound the initial consonant is apparently lenited: hence c > g. The element bast (here simplified to bas- to avoid the medial cluster *stg) is to be referred to the stem MBAS "knead" (LR:372). If primitive *mbastâ has the same meaning as its direct Quenya descendant masta, it could be either a verb "to bake" or a noun "bread"; the ending -tâ is a common verb-former, and the noun is perhaps derived from the verb. In Ilkorin as in Sindarin, primitive initial mb- is simplified to b-. The order of the compounded elements, "bread-round" with the adjectival element last, corresponds to the preferred order in Sindarin. In the entry MBAS in Etym, it seems that basgorn is indeed listed as a "Noldorin"/Sindarin word, not as an Ilkorin word as in the entry KOR. Of course, both languages could have the same word.
              bel "strength". Derived from a stem BEL "strong", that Tolkien tentatively compared to BAL, the source of bálâ "Power, God" (Quenya Vala). Concerning the latter stem, see Balthor. The primitive form of bel is given as belê; the ending -ê could be simply the stem-vowel suffixed and lengthened, but -ê is also an abstract ending.
              Beleg "the Strong", name of an Ilkorin bowman of Doriath. Derived from the same stem BEL "strong" as bel above (LR:352). This name would seem to be the same as the Sindarin/"Noldorin" word beleg ("mighty, great"), but remarkably, Tolkien noted that "this word is distinct in form from though related to Ilk. name Beleg". Obviously the words are not really distinct in form as they appear synchronically; perhaps Tolkien meant that the ancestral forms differed? This student cannot come up with any better etymology for the name Beleg than the one we can reconstruct for Sindarin beleg. In Etym, Tolkien first mentions a form bélek, which is probably to be understood as an extended variant of BEL, with the stem-vowel reduplicated and suffixed (so-called ómataina, vocalic extension) and a consonant -k added. Then follows the primitive form bélekâ, which is clearly this extended stem with the common adjectival ending -â. Bélekâ then yields Sindarin beleg via Old Sindarin beleka. After Tolkien revised the history of the Elvish languages, turning "Noldorin" > Sindarin into the language of the Elves of Beleriand (replacing Ilkorin), we can no longer insist of any distinction between the name Beleg and the common adjective "great, mighty".
              Belthronding (Bel-thron(d)-ding) name of Beleg's yew bow, mentioned in two different entries in the Etymologies (LR:352 s.v. BEL, LR:354 s.v. DING); in addition to this, a deviant form Balthronding is listed in LR:388 s.v. STAR, STARAN. In the entry DING, Belthronding is also broken up as Bel-thron(d)-ding. The first element bel- would mean "strong", like the stem BEL; -thron- means "stiff, hard" and also appears as an independent word in Ilkorin (see thrôn for further discussion), while the element ding is onomatopoeic ("twang"); it seems that this is also the Ilkorin word for "sound". So Belthronding would seem to mean *"Strong Stiff (thing that gives out a) Twang". The alternative form Balthronding replaces bel- "strong" with bal- "power" (see Balthor above concerning the stem BAL), hence meaning something like *"Powerful Stiff (thing that gives out a) Twang".
              benn "husband". Derived from a stem BES "wed" (LR:352); the primitive form is given as besnô, including the primitive masculine agentive ending -nô. Hence besnô is literally "wedder" = bridegroom. Notice the assimilation sn > nn, attested in this word only. Here, final -ô is lost without trace; in a couple of words, primitive -ô may seem to come out as -a in Ilkorin (see adda, broga).
              ber "valiant man, warrior". Derived from the stem BER "valiant"; the primitive form is given as berô, the masculine ending -ô being added to the adjectival stem to produce a noun "valiant one, valiant man". Again, final -ô is lost instead of producing -a as in adda (?) and broga.
              bereth "valor" (cf. El-bereth). Derived from the same stem BER "valiant" as ber above. And abstract ending -eth is known from Sindarin as well. We may refer bereth to *bereth- with some lost final vowel; this would be the stem BER in its extended form *BERE (with ómataina, reduplicated and suffixed stem-vowel) + a suffixed consonant -th. In Sindarin, it seems that the abstract ending -eth was later generalized and could be added to stems with any stem-vowel; we have no examples to show us whether this was also the case in Ilkorin.
              boron "steadfast, trusty man, faithful vassal", pl. burnin. Derived from a stem BOR "endure" (LR:353), specifically from an extended form that is given as bóron-, sc. BOR- with ómataina and the suffixed consonant -n, apparently followed by another vowel, lost in Ilkorin: cf. the hyphen at the end of bóron-. We might have expected the Ilkorin form to be *born instead, cf. thorn "eagle" from the stem THORON. In the plural form burnin, the second vowel of boron is syncopated; compare a similar contraction in adar "father" pl. edrin and other Ilkorin words. Why the change from o to u in the plural? (Compare thorn "eagle", pl. thurin.) Is this some kind of umlaut triggered by the vowel of the plural ending -in? Normally, we would expect the umlaut product of o to be ö or y.
              breth "mast, beech-mast". Derived from a stem BERÉTH (LR:352), apparently meaning "beech-tree" (this is the meaning of Telerin bredele, the first derivative listed). The stem BERÉTH would seem to be unconnected to BER "valiant". The primitive form of breth is given as b'rethâ; as usual, final -â has been lost in Ilkorin. B'rethâ is best taken as an adjectival form *"beechen", -â being a common adjectival ending (mentioned in WJ:382); later this adjective may have been used as a noun "that which pertains to beeches", hence "mast, beech-mast". B'rethâ must represent even older *beréthâ, but in the evolution of the Celtic-style languages of Tolkien's mythos, it is quite typical that the vowel of an unaccented initial syllable is lost, the syllable collapsing into a consonant cluster. Compare bril below.
              bril "glass, crystal" or (as part of the river-name Brilthor, q.v.) "glittering". Derived from MBIRIL (LR:372), said to be a "compound" of two distinct stems, MIR and RIL. The first of these, MIR, is not defined as such, but it yields Quenya mírë, Sindarin mîr "jewel" (LR:373). In the compound, MIR appears as MBIR, apparently showing initial fortification. The stem RIL signifies "glitter" (or, according to the Silmarillion Appendix, "brilliance") and is the source of the second element of Quenya Silmaril (translated "radiance of pure light" in Letters:148, the word translating -ril evidently being "radiance"). Bril must be referred to mbiríl-, later b'ril- after the loss of an unaccented vowel (cf. breth above); as usual, initial mb- is simplified to b- in Ilkorin (cf. basgorn above).
              Brilthor "Glittering Torrent" (river-name, one of the tributaries of Gelion). Listed in the Etymologies in the same entry MBIRIL (LR:372) as bril discussed above; also mentioned in the entry THOR, LR:393. The first element bril- would have the same derivation as bril "glass, crystal" above, though it is here translated "glittering" instead. The second element thor directly represents the stem THOR "come sweeping down", here used in an agental sense: "thing that comes sweeping down" = "torrent". Alternatively we can see -thor as a shortened form of the adjective thôr "swooping, leaping down" (probably from *thorâ), in which case Brilthor is actually Bril Thôr, "Glass/crystal (that is) leaping down" (Tolkien's translation "glittering torrent" being explicatory rather than literal). Notice that the final element of Balthor "Vala-king" is wholly unrelated; this -thor is just the form tôr (-tor) takes after the liquid l (lt becoming lth for phonological reasons).
              brith "broken stones, gravel". The undefined stem BIRÍT is said to be unique to Ilkorin (LR:353). It looks like an extension of a simpler stem *BIR, but there is no material to throw any light on the more basic meaning. A primitive form is given as b'rittê, presumably representing even older *biríttê before the unaccented vowel of the initial syllable dropped out. The double tt of b'rittê is probably best taken as a medial fortification, the final consonant of BIRÍT being doubled (compare WJ:415, where a primitive word grottâ is said to be an "intensified" form of grotâ). Here, we see primitive tt become th, as in Sindarin; contrast (?) adda, q.v. The ending -ê sometimes occurs in the names of substances; cf. for instance mazgê "dough" (LR:371 s.v. MASAG) or srawê "flesh" (MR:350). "Broken stones" may also be considered a substance, as indicated by the additional gloss "gravel".
              Brithombar *"Land of Brithon" (LR:353 s.v. BIRÍT). Compound of the river-name Brithon, q.v., and bar "land", more literally "home" (notice assimilation nb > mb in Brithon-bar > Brithombar). As an Ilkorin word, bar is attested in compounds only (Brithombar and Dimbar), but it would obviously come from the stem MBAR "dwell, inhabit" (LR:372). It is possible that it would have a long vowel (*bâr) if it appeared as an independent word. (Compare thôr "sweeping" with short -thor in the river-name Brilthor.)
              Brithon "pebbly", river name. Derived from the same undefined stem BIRÍT as brith "broken stones, gravel" (q.v.) If the gloss is literal, this word must be derived from *biríttânâ > *b'rittânâ with a longer form of the adjectival ending -nâ (concerning which see caun).
              broga "bear". The stem MORÓK (LR:374) is not itself defined; it could be an extended form of MOR "dark, black" (Letters:382); perhaps we are to assume that the Elves' first acquaintance with bears involved black bears? The primitive form is given as morókô; the final -ô could be a third reduplication of the stem-vowel, or it may be an ending denoting a (male) animate. As often, the unaccented vowel of the initial syllable dropped out on the way to Ilkorin; there must have been an intermediate form *m'rokô, but mr was abandoned as an initial combination, becoming br instead (the same development occurred in Sindarin/"Noldorin"). As usual, post-vocalic k is voiced to g, and the final -ô seemingly survives as -a (cf. possibly adda < *attô above, but contrast benn < besnô and other words where it is simply lost like the other final vowels).
              burnin pl. of boron (BOR); see boron.
              caun "bowed, bow-shaped, bent". Another, apparently earlier form was coun. Derived from a stem KU3 "bow" (LR:365). The most primitive form is given as ku3nâ, sc. the stem KU3 with the adjectival ending -nâ (in UT:266, a word in -nâ is called as an "ancient adjectival form"). A later form kogna, perhaps "Old Ilkorin", is also mentioned: In Ilkorin, the primitive back spirant 3 (= gh) became a plosive g; compare go "from" (q.v.) from 3O. Another change affects the stem-vowel, u becoming o; this is evidently due to an umlaut caused by the final -a. The combination g + a nasal proved unstable, the voiced plosive later becoming a spirant gh following a vowel (see tûgh, tû) and then merging into the preceding vowel to produce a diphthong ou: kogna > *kogn > *koghn > coun. This diphthong eventually merged with au (found in many other words), producing caun. For a wholly parallel case, see daum (< do3mê).
              celon "river". Derived from a stem KEL "go, run (especially of water)" (LR:363). A primitive form kelu-n is given. Extensions in -n are not uncommon (see for instance boron), but instead of normal ómataina (reduplication and suffixing of the stem-vowel itself, which in this case would have produced *kele), an element u is suffixed to KEL. (Compare the middle vowel of the Quenya derivative celumë "stream, flow".) The Index to Unfinished Tales, entry Celos, actually mentions a root kelu- "flow out swiftly". This suffixed element -u cannot be fully explained; in WJ:411, Tolkien refers to telu as a "differentiated form of *TELE"; perhaps kelu- is similarly a "differentiated from" of *KELE. The u of kelu-n has become o in Ilkorin celon; this may parallel the change seen in ulgundô > ulgund > ulgon, ulion (q.v.) Perhaps unaccented u tended to become o in Ilkorin?
              côm "sickness". The stem KWAM (LR:366) is not itself defined. Quenya quámë, Sindarin paw and Ilkorin côm must all descend from a form something like *kwâmê (or conceivably *kwâmi > Common Eldarin kwâme). After long â became ô (compare ôr, tôr, q.v.), the initial consonant evidently lost its velarized quality, *cwôm becoming côm (compare gôd, q.v.). If Ilkorin is to be adapted as a Sindarin dialect according to Tolkien's later scenario, *kw would already have become p (as in the Sindarin/"Noldorin" cognate paw), so we would have seen *pôm instead.
              coun "bowed, bow-shaped, bent", alternative (evidently older) form of caun, q.v. (LR:365 s.v. KU3)
              cwess "down" (noun). The stem KWES (LR:366) is not defined; the primitive form is given as kwessê, apparently including a medial fortification s > ss (unless a longer ending -sê is present). The ending -ê may denote a substance (cf. brith "gravel" < *b'rittê), and down is indeed a substance, but the Quenya cognate of Ilkorin cwess (namely quessë) means "feather". If Quenya preserves the original meaning of *kwessê, the final ê may be simply the stem-vowel suffixed and lengthened. - Again, cw would have had to appear as p instead if Ilkorin is to be adapted as a dialect of Sindarin according to Tolkien's later scenario.
              dair "shadow of trees". Derived from a stem DAY "shadow" (LR:354). The Quenya word laira "shady" points to a primitive adjective *dairâ with a well-attested adjectival ending -râ (compare târâ "lofty"; see tôr). The same primitive form could have produced Ilkorin dair, but since the latter is a noun rather than an adjective, it may be better to assume a primitive noun *dairê; concerning the nominal (abstract or collective) ending -rê, compare for instance thêrê "look, face, expression" derived from the stem THÊ "look (see or seem)" (LR:392). - Notice that Ilkorin preserves primitive *ai unchanged, while it becomes ae in Sindarin (in the "Noldorin" of the Etymologies, Tolkien can be seen trying to make up his mind whether *ai yielded oe or ae, but the latter was his final decision).
              dang "sound", onomatopoeic. Variant of ding, q.v. (LR:354 s.v. DING; dang would obviously require a stem *DANG instead; DING is said to be a variation of TING, TANG [LR:394], so postulating a form *DANG corresponding to TANG like DING corresponds to TING is not implausible).
              daum "night-time, gloom". Derived from a stem DO3 with a variant DÔ (LR:354); this evidently indicates that the original stem was DO3 (in which 3 = a back spirant, in Orkish spelt gh); later, the weak consonant 3 was lost, and the preceding vowel was lengthened in compensation. The stem DO3/DÔ is not defined, but Quenya ló "night" comes directly from DÔ. Daum is clearly to be referred to do3mê with an ending denoting something abstract or intangible (this primitive form is mentioned in LR:355, in the entry DOMO). The development apparently parallels that of caun above, *do3mê becoming dogme (attested as an Old Sindarin word) and then daum; the form coun intermediate between caun and *kogn < kogna may suggest that there was likewise a form *doum intermediate between daum and *dogm < dogme.
              dem "sad, gloomy". The stem, similarly defined, is given in LR:354 as DEM, but this must be an error or a misreading, DIM being intended. The ancestral form of dem is given as dimbâ, not **dembâ, and in two other derivatives (dim and dimb, see below), the original quality of the stem-vowel still survives in Ilkorin. Dimbâ shows a medial fortification m > mb combined with the adjectival ending -â. Some might also argue that -bâ is here an allomorph of the longer adjectival ending -wâ (w becoming b after m?) In any case, there was a final -â that eventually caused the vowel i (in dimbâ) to umlaut to e, the final -â thus leaving its mark upon the word dem even after it was lost, like the original final vowels usually were in Ilkorin. Contrast dim "gloom, sadness" from dimbê, the original quality of the vowel persisting because there was no following -â to cause umlaut. - In the word dem from dimbâ, and likewise dim from dimbê, the cluster mb has been simplified to single m (probably after the loss of the final vowel: *demb > dem). Strangely, final -mb seems to persist in dimb (see below).
              Dilion - hypothetical Ilkorin form of Noldorin Gelion, not used (LR:359-360 s.v. GYEL, a stem that is not defined as such, but apparently having to do with joy and triumph). This example would seem to indicate that primitive initial gye- (gje-) comes out as di- in Ilkorin, so that the cognate of (say) Quenya yerna "old, worn" (primitive *gjernâ, stem GYER) would be *dirn. As for the name Dilion/Gelion itself, Tolkien made a note that his son rightly describes as "perplexing": "Gelion shorter name of a great river in E. Beleriand; a Gnome [Noldo] interpretation (this would have been Dilion in Ilkorin); cf. Ilk. gelion = bright, root GAL." What Tolkien is trying to say seems to be this: The name Gelion actually comes from the Ilkorin word for "bright", derived from the stem GAL. However, Tolkien imagined that the Noldor incorrectly associated this word with their own Noldorin word Gelion (for at this time, he still had not turned Noldorin into Sindarin), derived from the stem GYEL having to do with joy: Noldorin Gelion is translated "merry singer". That this was not the correct interpretation of the Ilkorin river-name, even though the Ilkorin and the Noldorin word had the same form, is shown by the fact that the true cognate of Noldorin Gelion (stem GYEL) would not have been Gelion at all, but Dilion: In Ilkorin, primitive gje- became di-, not as in Noldorin ge-. Gelion (q.v.) means "bright" instead, unrelated to the Noldorin word gelion.
              dim "gloom, sadness" (LR:354 s.v. DEM, but as explained in the entry dem above, the actual stem must rather be DIM). The primitive form is given as dimbê, which is the stem DIM with medial fortification m > mb and the abstract ending -ê or conceivably DIM with an ending -bê, allomorph of the abstract ending -wê (w becoming b following m). As in the case of dem above, the simplification mb > m likely took place only after the loss of the final vowels.
              dimb "sad" (LR:354 s.v. DEM). This synonym of dem (from dimbâ) must come from another adjectival form, with an ending that does not cause umlaut - say, *dimbi. Compare for instance the primitive adjective slindi "fine, delicate", derived from the stem SLIN; slindi thus shows a medial fortification n > nd that could be compared to m > mb in *dimbi. Strangely, *mb is in this case not simplified to -m after the cluster became final by the loss of the final vowels; contrast dim and dem above. Perhaps dimb is an early form, later becoming *dim (and then probably disappearing, since this adjective would then merge with the noun dim, leaving dem as the sole adjectival form). Dimb occurs in the place-name Dimbar, *"Sadland". Compare Christopher Tolkien's note on Dimbar in the Silmarillion Index, entry bar; the first element is there said to mean "sad, gloomy". Concerning the element -bar, see Brithombar above.
              ding "sound" (also dang). Essentially the stem DING, said (in LR:354) to be onomatopoeic, with no additional elements. DING is a variant of TING, TANG. See dang, Belthronding above.
              dôl "flat, lowlying vale". Does this gloss mean "vale that is flat and lowlying", or that dôl functions both as an adjective "flat" and as a noun "lowlying dale"? The word is in any case derived from a stem DAL "flat" (LR:353). One primitive form is listed as dâla; read perhaps *dâlâ since the final vowel is still in place in Quenya lára: A short final -a would have disappeared at the Common Eldarin stage. In Ilkorin, final A disappears whether it was short or long, but a long final -â could be identified with a well-attested adjectival ending (explicitly mentioned by Tolkien in WJ:382). Non-final long â in any case becomes ô, hence dâl- > dôl, perhaps an adjective "flat" (the original meaning of the primitive word) that also came to be used as a noun "lowlying vale".
              dor "land", isolated from Argador, Dor-thonion (q.v.) If dor appeared independently, it would probably have a long vowel: *dôr. In the Etymologies, the Eldarin words for "land" are derived from a stem NDOR "dwell, stay, rest, abide" (LR:376). No Ilkorin word is there listed, but dor would have the same origin as the identical Sindarin word: primitive ndorê (the ending -ê may sometimes denote a locality, compare taurê "forest", LR:391 s.v. TÁWAR). Notice, however, that Tolkien many years later derived the Eldarin words for "land" from a stem DORO "dried up, hard, unyielding" (WJ:413). However, this later source does confirm that the Primitive Quendian form was ndorê, now thought to be formed by initial enrichment d > nd. This is defined as "the hard, dry land as opposed to water or bog", later developing the meaning "land in general as opposed to sea", and finally also "a land" as a particular region.
              Dor-thonion "Land of Pines", place-name (LR:392 s.v. THÔN). Concerning dor "land", see separate entry above. Thonion is the genitive plural of thôn "pine", q.v.
              dorn "oak" (same word in Ilkorin and Doriathrin). The stem DÓRON is itself defined only as "oak" (LR:355). It is tempting to think of DÓRON as an extended form of DORO "hard, unyielding" (WJ:413), since oak-wood is hard - though the stem DORO is not in the Etymologies and only appears in a document written a quarter of a century later. Quenya norno and Sindarin doron together point to a primitive form *doronô, where the final -ô may simply be the stem-vowel reduplicated, lengthened and suffixed (-ô may also be a masculine ending, but this does not seem fitting here).
              duil "river". Listed under the stem DUI (LR:355), not itself glossed in the Etymologies, but in A Tolkien Compass p. 179 it is defined as "flow (in volume)". The primitive form is probably meant to be *duilê, originally an abstract formation "flowing". See VT39:16 concerning the abstract ending -lê; cf. also such an example as tuilê "spring-time" from TUY "spring, sprout" (LR:395); tuilê is literally simply an abstract *"springing, sprouting". In Ilkorin, duil has taken on a more concrete meaning than its primitive counterpart *duilê: "flowing" > "river".
              Duilwen place-name, *"Green River" (LR:355 s.v. DUI, LR:359 s.v. GWEN). See above concerning duil "river". The element -wen "green" would seem to be a lenited and shortened form of gwene, q.v.
              duin "water, river" (LR:355 s.v. DUI). See duil above concerning the most basic meaning of the stem DUI. The primitive form of duin "river" is given (in A Tolkien Compass p. 178-179) as duinê. A noun-former -nê is seen in a few primitive words, e.g. neinê "tear" from the stem NEI of similar meaning (LR:370). The primitive word ornê "(slender) tree" is stated to be related to the adjective ornâ "uprising, tall" (UT:266). In this word, -nê would seem to be a nominal ending corresponding to adjectival -nâ, an ornê being literally a "tall thing", used with reference to slender trees. It may be, then, that there was an adjective *duinâ "flowing (in volume)" that corresponds to a noun *duinê "thing that flows" = "water, river". - In A Tolkien Compass, duin is understood to be a Sindarin word, since Ilkorin was partly swallowed up by Sindarin when Tolkien revised the history of the Elvish tongues. Cf. Anduin "long river" in LotR. Also in Esgalduin, which name Tolkien originally thought of as Ilkorin.
              dûm "twilight" (short -dum in tindum, q.v.). Listed under a stem DOMO "faint, dim" (LR:354); a primitive form dômi- is listed (the hyphen suggesting that the word is incomplete?) The ending -i is rare (normally used to form colour-adjectives) and cannot be readily explained. In Quenya, dômi- "twilight" and do3mê "night" merged as lómë, but in Ilkorin they apparently remained distinct (appearing as dûm and daum, respectively). - This word provides our sole example of long ô becoming Ilkorin û, though this parallels the development in Sindarin.
              edrin pl. of adar, q.v. for discussion (LR:349 s.v. ATA).
              Eglath is in the Etymologies conceived as a form cognate with Quenya Eldar (LR:368 s.v. LED); according to LR:358 s.v. GAT(H) this is what the Ilkorins called themselves. Tolkien's ideas about the etymology of both Quenya Eldar and the Beleriandic word Eglath changed over time. In Etym he first associated Eldar with a stem ELED "go, depart, leave" (LR:356), transparently a stem-vowel prefixed variant of a simpler stem LED "go, fare, travel" (LR:368). The idea was that the Eldar were the elves that left Cuiviénen to go to Valinor (whether they ever got there or not). While this application of the word Eldar remained ever since, Tolkien soon changed its etymology to "Star-folk", introducing a new stem ÉLED, evidently to be understood as an extended form of EL "star" (LR:355). In the Silmarillion narrative the story emerged that Oromë originally called all the Elves by this name, but it was later applied only to those who followed him on the westward march. At this time, Tolkien probably imagined the primitive form of Quenya Elda to be the same as the one he gave in Letters:281 many years later: Eledâ "an Elf". According to LR:356, eled- had an alternative, transposed form edel-, the whole word presumably being *edelâ. After the loss of the middle vowel, bringing the consonants d and l into direct contact, the new cluster dl became Ilkorin gl as in Eglath. (In the "Noldorin" of the Etymologies, but not in Tolkien's later Sindarin, intervocalic dl also becomes gl.) The ending -ath seen in Eglath must be understood as a collective suffix, corresponding to a similar ending in "Noldorin"/Sindarin. However, Tolkien later derived Quenya Elda from primitive eldâ, an adjectival formation "connected or concerned with the stars" (WJ:360), and it is doubtful whether eldâ could come out as Egl- in Ilkorin. By this time, Tolkien had long since replaced Ilkorin with Sindarin as the language of the Elves of Beleriand. The term Eglath was carried over into Sindarin and is found in the published Silmarillion (chapter 5), where it is still said to be what the Elves of Beleriand called themselves, but now it is translated "the Forsaken People". Hence Eglath is no longer conceived as the cognate of Quenya Eldar "Star-folk". In WJ:365, a variant form Eglan (pl. Eglain, Egladhrim) is associated with a stem HEK, HEKE "aside, apart, separate" (cf. WJ:361).
              [El-bereth] ?"Heavenly Valor", personal name (LR:352 s.v. BER). It is not clear from the Etymologies whether this was intended as a name of Varda, the well-known application of the name Elbereth in Sindarin. In the Etymologies, Tolkien struck out El-bereth as an Ilkorin name, perhaps because of the emergence of a similar form as a name of Varda. Here, the second element is bereth "valor", q.v. for discussion. The element el- presumably means "heaven" or "sky" in light of two other Ilkorin words that were explicitly translated by Tolkien, Elrond "Vault of heaven" and Elthorn "eagle of sky". In the Etymologies, the element EL is defined not only as "star" but also as "starry sky" (LR:355), and the translation "heaven" or "sky" must depend on the latter gloss. It is said that in "Noldorin and Telerin", EL was confused with a distinct stem 3EL "sky" (LR:360), but this would probably come out as *gel- in Ilkorin, so it is difficult to see how the elements could have been confused in that language. - Concerning Tolkien's later, different interpretation of the name Elbereth as the Sindarin name of Varda, see barathi, elen-barathi in the wordlist appended to the article about Old Sindarin.
              Elrond "Vault of heaven", name of Eärendel's [sic] son (LR:384 s.v. ROD). Concerning the first element el- "heaven", see El-bereth above. In the Etymologies, rond "vault" is derived from a stem ROD "cave", Quenya rondo indicating a primitive form *rondô, showing nasal infixion and an ending -ô that may be simply the stem-vowel suffixed and lengthened. Much later, when Elrond had long since become a Sindarin name, Tolkien derived the element rond from a stem RONO "arch over, roof in" instead (WJ:414). The primitive form would still be the same, *rondô, but now nd must be taken as a medial fortification of n; it cannot be produced by nasal infixion as when the stem was still conceived as ROD. In the same source, Tolkien translated Elrond as "Star-dome" rather than "Vault of heaven", the element el- now meaning "star" only, rather than covering both "star" and "heaven" as in earlier sources.
              Elthorn "eagle of sky", personal name. In LR:392 s.v. THOR/THORON, the name is given as "Elthor(o)n", probably indicating an Ilkorin form Elthorn and a "Noldorin"/Sindarin form Elthoron (since the "N"/S word for eagle is thoron, corresponding to Ilkorin thorn). Concerning the element el-, see El-bereth above; see thorn for a discussion of the last element.
              [emuin misreading for emnin, pl. of aman, q.v. (LR:348 s.v. AM1)]
              erdh "seed, germ". Derived from a stem ERÉD (LR:356) that is not itself defined, but to be understood as a stemvowel-prefixed variant of RED "scatter, sow". The primitive form of erdh is given as eredê; the ending -ê is probably just the stem-vowel suffixed and lengthened. The syncope brought original r and d into direct contact, and as in Sindarin, d becomes dh following a liquid r or l (see alch).
              Ermabin, Ermabrin "one-handed", name of Beren (LR:371 s.v. MAP). The prefixed element er "one" is obviously derived from the stem ERE "be alone, deprived" (LR:356; compare Eru, the One, as a Quenya name of God). For the middle element mab "hand", see separate entry. The endings -in, -rin turn "hand" into an adjective "handed". These adjectival endings would represent primitive -ina, -rina, the final -a being lost before it could cause umlaut (or we would see -en, -ren, as in Sindarin).
              esg "sedge" (LR:356 s.v. ESEK), obsoleting: [esg] "rustle, noise of leaves" (LR:357 s.v. EZGE). The stem EZGE was itself defined "rustle, noise of leaves"; this looks like a rearranged form of a simpler stem *SEG (compare EZDÊ, from which is derived the name of the Valië Estë, in relation to the verbal stem SED "rest": LR:357, 385). When Tolkien introduced the ESEK, he evidently imagined esg to be derived from either *eskê or *esekê. For the change sk > sg following a vowel, compare esgal (stem SKAL1) below. Curiously, this shift does not occur in the river-name Ascar (Askar).
              esgal "screen, hiding, roof of leaves". Derived from the stem SKAL1 "screen, hide (from light)". It is not clear where the prefix e- comes from; a similar prefix appears in esgar # 2 (see below).
              Esgalduin "River under Veil" of [?leaves] - Tolkien's handwriting was illegible (LR:386 s.v. SKAL1, LR:355 s.v. DUI, LR:357 s.v. EZGE).
              esgar (1) "reed-bed" (LR:356 s.v. ESEK). In the name Esgaroth, q.v. Esgar would seem to be esg "sedge" (see above) with some sort of collective ending, perhaps related to the Quenya collective ending -rë.
              esgar (2) "wound"? The word is not clearly glossed; Tolkien lists Quenya and "Noldorin"/Sindarin words for "wound", adding "cf. Ilk. esgar" with no further explanation. While esgar cannot be a direct cognate of the other words listed, it seems that it is intended to mean "wound" like they do. The stem SKAR (LR:386) is not defined as such, but its derivatives suggest a basic meaning "tear, wound", verb or noun. The e- prefix in esgar is curious, but compare esgal above. As in esgal (from the stem SKAL1), sk becomes sg following a vowel.
              Esgaroth "Reed-lake" (LR:356 s.v. ESEK). Concerning the first element, esgar "reed-bed", see separate entry. It would seem that -oth is the element translated "lake". No similar word or element of such a meaning seems to occur anywhere else.
              gangel ?"harp" (pl. genglin). The wording in LR:377 s.v. ÑGAN is not quite clear as to whether this Ilkorin word means "a harp", "harping" or even "harper", but gangel seems to be the cognate of Quenya ñandellë "little harp", Sindarin/"Noldorin" gandel, gannel "harp". The stem ÑGAN, also expanded ÑGÁNAD (with ómataina and suffixed -d), is defined "play (on stringed instrument)". Quenya ñandë "a harp" must represent *ñgandê, the ending -ê apparently being a noun-former here (was this originally an abstract "harping", since -ê often functions as an abstract ending?) The longer form ñandellë "little harp" (< *ngandellê) apparently has a diminutive ending attached, perhaps related to the feminine ending seen in Tintallë "Kindler" as a title of the goddess Varda. Though ñandellë is defined as "little harp", the Sindarin/"Noldorin" cognate gandel, gannel is simply glossed "a harp", suggesting that the final element was no longer felt to signify smallness in this language, gannel being perceived as a unitary word: this may also be true for Ilkorin gangel. In gangel we see the same shift nd > ng as in gwilwering "butterfly" = Quenya wilwarind-. This is an unusual development; in many words nd is unchanged in Ilkorin (e.g. lind, rond, thind, tindum, tund). In gangel, the shift nd > ng may be due to assimilation, nd being influenced by the initial g (or already the primitive initial *ñg). - The plural genglin shows the plural ending -in, q.v., combined with the regular syncope of the middle vowel and umlaut a > e caused by the i of the plural ending.
              Garthurian "Hidden Realm", "Fenced Realm" = Doriath; from gardh-thurian (LR:358 s.v. GAT(H), LR:393 s.v. THUR; notice that in LR:360 s.v. 3AR, Garthurian is said to be a Doriathrin word rather than Ilkorin). LR:393 s.v. THUR indicates that Garthurian is a compound of garth, gardh- "realm" and an element thurian "hidden". In LR:360 s.v. 3AR, Quenya arda "realm" is said to be derived from the stem GAR, defined as "hold, possess" (LR:357, there struck out, but on the next page GAR resurfaces as a variant of a stem 3AR - apparently of similar meaning). Quenya arda must therefore come from gardâ (this primitive form is actually mentioned in WJ:402), showing a medial fortification r > rd and the ending -â that may denote something inanimate; it may also simply be the stem-vowel suffixed and lengthened. Gardâ produced garth in Doriathrin (evidently gardâ > *gardh > *garth), and the same word evidently occurred in Ilkorin as well. - The element thurian "hidden" is obviously a kind of past participle based on the verbal stem THUR-, defined as "surround, fence, ward, hedge in, secrete". To explain the ending -ian we must probably assume a primitive verb *thurjâ- with a verbal ending that is very well attested (yielding Quenya -ya); to this verb the primitive adjectival/past participle ending -nâ has been added to produce *thurjânâ > Ilkorin thurian.
              gelion "bright", evidently the same word that is also the name of the river Gelion. Listed in LR:359 under GYEL; however, the stem is not GYEL; instead it is indicated to be GAL. The latter stem is listed as a separate entry in the Etymologies (LR:357), where it is said to be a variant of KAL "shine" (cf. LR:362), but no derivatives are listed. Gelion could represent primitive *Galjânâ, combining the adjectival endings -jâ and -nâ. - At a later stage, Tolkien noted that Gelion was among the "river-names [that] need revision to etymologizable words" (WJ:336), as if he himself could not think of any suitable etymology then. This may be connected to the rejection of Ilkorin as the language of Beleriand; Gelion now had to become a meaningful Sindarin word instead, or be dropped altogether. Various possible replacements are indeed listed in the source just quoted (WJ:336); Tolkien even considered postulating that the name was actually adapted to Grey-elven from Dwarvish Gabilân "great river". The published Silmarillion nonetheless retains Gelion as the name of the great river of Beleriand.
              go "from". Derived from a stem 3O "from, away, from among, out of". (LR:360; in a later entry, Tolkien noted that the stem WÔ "together" had no descendant in Ilkorin because it became identical to go- and hence did not survive: LR:399.) In a later source than the Etymologies, Tolkien gave the primitive stem meaning "from" as HO instead of 3O (WJ:368). Go- was also a patronymic prefix; Tolkien used the example go-Thingol "son of Thingol" - which is a strange example, since Thingol had no son, but only a daughter (Lúthien). But since the basic meaning is simply "from Thingol", this phrase could surely cover "daughter of Thingol" as well. - At the same time, Tolkien also stated that "initial h disappeared in Sindarin" (WJ:369), so it would be difficult to keep go as a Sindarin word if we were to adapt Ilkorin as a Grey-elven dialect.
              gôd "dirtiness, filth". The stem WA3 "stain, soil" (LR:397) yielded a primitive form wahtê; this is evidently Common Eldarin for even older (sc. Primitive Quendian) *wa3tê "a stain", the back spirant 3 later becoming h (here probably denoting [x] = German ach-Laut) by contact with the unvoiced plosive t. (Compare Primitive Quendian ma3-tâ yielding Common Eldarin mahtâ-; see LR:371 s.v. MA3.) The ending -tê is very rare, but can apparently be used to derive words for something that is produced by means of the verbal action denoted by the stem (compare kirtê "cutting, rune" from kir- "cut", WJ:396; if Tolkien's glosses for the stem WA3 - "stain, soil" - are taken as verbs rather than nouns, wahtê could be a similar formation). Waht- clearly became wât- at some early point, h being lost and the preceding vowel being lengthened in compensation, long â later becoming ô (cf. ôr, tôr), while -t was regularly voiced to -d following a vowel. This is not how primitive ht develops in Sindarin; instead of h being lost and the preceding vowel being lengthened, ht was assimilated to tt in Old Sindarin, later becoming th in Classical Sindarin (the cognate of Ilkorin gôd being gwath). - It is clear that original w- turned into gw- at some early stage (and here Ilkorin does follow Sindarin). Nearly all the words in gw- listed below come from stems in W- (only in the cases of gwên and gwene did the stem originally begin in GW-). Before o, gw was reduced to g- (hence gôd < *gwôt < *gwât- < *wât- < wahtê).
              góda- "to soil, stain". Derived from the stem WA3 of similar meaning (LR:397); a primitive form is given as wahtâ- (Common Eldarin for Primitive Quendian *wa3tâ-?), showing the common verbal ending -tâ. Tolkien probably imagined a development something like wahtâ > wâtâ- > *gwâta > *gwôta- > *góda-. Normally, final -â would not survive into Ilkorin as -a (cf. for instance gwedh [q.v.], not **gwedha, for primitive wedâ). The hyphen following góda- may suggest that it is not really final at all, but that some pronominal or inflectional ending would follow. Góda as an independent word may actually mean *"he soils, stains"; concerning the ending -a implying "he" and "present tense", see taga.
              gôr "soiled, dirty". Yet another derivative of WA3 "stain, soil" (LR:397); the primitive form is given as wa3râ with a well-attested adjectival ending (compare for instance ubrâ "abundant" from the stem UB "abound", LR:396). Wa3râ became *wârâ when the back spirant 3 was lost before a consonant and the preceding vowel was lengthened in compensation (compare *ta3râ > târâ, see tôr). Later, w turned into gw, but when â became long ô, the preceding gw turned into g. Compare gôd, góda- above.
              gwath "shade". This is also the gloss of the stem WATH (LR:397); no primitive form is given, but Old Sindarin ("Old Noldorin") watha points to *wathâ. Here, the ending -â is used to derive an inanimate noun (compare gwedh < wedâ "bond" below). In this and the following words, gw for primitive w- persists (instead of becoming g before o as in gôd, góda-, gôr above). - The form gwethion occurring in the name Urthin Gwethion (q.v.) may be the genitive plural of gwath; notice the I-umlaut turning a into e.
              gwau "wind". The relevant entry-head in the Etymologies reads WÂ-, WAWA-, WAIWA- (LR:397, defined as "blow"), apparently indicating a simple stem WÂ that also appears in reduplicated form WAWA, and the latter may also be I-infixed to produce WAIWA. Quenya vaiwa (archaic waiwa) and Sindarin gwaew must come from *waiwâ, but Ilkorin gwau rather represents WAWA.
              gwedh "bond". Primitive wedâ, derived from the stem WED "bind" (LR:397), shows an ending that is often adjectival but may also form inanimate nouns, as here. Wedâ is best taken as an impersonal agental formation: a "bond" being a thing that functions as a "binder". (Compare mapâ; see mab.)
              gwelo, also gwelu, "air, lower air" (said to be "distinct from the 'upper' air of the stars, or the 'outer' " - this refers to Tolkien's invented cosmology, or one stage of it). The stem WIL is defined "fly, float in air"; primitive wilwâ (> Quenya wilwa) by its form looks like an adjective (ending -wâ, see adu), but it is clearly intended to be a noun. Before the final -â disappeared, it umlauted i to e (compare rest, q.v., from rista-). After the loss of the final vowel, a form *gwelw may have occurred (initial w having become gw already); then the final semivowel -w turned into a full vowel u, producing gwelu, final -u later turning into -o, hence the form gwelo. Compare ado, earlier adu.
              gwen "girl". The stem WEN "maiden" (LR:398) also has a longer form WENED (with ómataina, sc. the stem-vowel reduplicated and suffixed, and a suffixed -D). This expanded stem is reflected in Quenya vendë (< *wendê), but Tolkien argued that there must also have been a simpler primitive form wen- (the hyphen suggesting some final vowel that was later lost), and he quoted the Ilkorin form gwen as evidence for such a form. This would seem to indicate that final -nd was not reduced to -n in Ilkorin (and this is confirmed by such examples as lind, rond, thind, tund, q.v.); hence gwen could not be a direct cognate of Quenya vendë, for if so it would appear as **gwend instead.
              gwên "greenness". This word, as well as gwene below, is derived from a stem where initial gw- is original and not a later elaboration of primitive w-. The stem GWEN is not itself defined - Tolkien only noted that it was not connected to WEN(ED) "maiden" (see gwen above) - but the derivatives of GWEN have to do with greenness and freshness. Ilkorin gwên is evidently a cognate of Quenya wén "greenness, youth, freshness", perhaps indicating a Primitive Quendian form *gwene, Common Eldarin *gwên (compare PQ kwene "person" > CE kwên, Quenya quén, WJ:360-361). Earlier long â, ô in Ilkorin become ô, û (see ôr, dûm), just as in Sindarin, and since Sindarin turns long ê into î, we might expect that an earlier form *gwên would come out as *gwîn in Ilkorin. However, this is not the case; gwên is simply unchanged. (Cf. terêwâ yielding Ilkorin trêw, q.v., not as in Sindarin/"Noldorin" trîw.) Long *í, *î is actually not attested in any Ilkorin word.
              gwene "green". Derived from the same undefined stem GWEN as gwên above (LR:359). This is surely the cognate of Quenya wenya "green, yellow-green, fresh", indicating a primitive form *gwenjâ. After the loss of final -â, the semivowel that remained at the end of *gwenj turned into a full vowel, evidently producing *gweni in early Ilkorin (cf. gwini, q.v., though this word was struck out). Later, final -i turned into -e, hence the attested form gwene (but probably still gweni- as the first element of compounds). This parallels the development of final *-w turning into -u and then -o in words like adu, ado "double" (but only adu- in compounds like Adurant, since -u is not there final). In compounds, the word gwene is actually only attested as the final element, in the name Duilwen *"Green River", in which -wen is apparently a reduced form of gwene. (A voiced stop following a liquid like l normally turns into a spirant, cf. the words alch, erdh, so at an older stage the form may have been *Duilghwen, gh subsequently being lost as in Sindarin. The word tûgh > tû, q.v., demonstrates that Ilkorin at one stage had the spirant gh and then lost it.)
              Gwethion *"Shadow-land"? This is a place-name possibly meant to replace Urthin; Tolkien's wording is not clear. Derived from the stem WATH "shade" (LR:397), this name shows the regular Ilkorin enrichment of initial w- to gw, plus the ending -ion occurring in the names of some countries; the i of this ending turns the stem-vowel a into e by umlaut. However, it is also possible that Tolkien intended a name Urthin Gwethion, that could be interpreted "Mountains of Shadows" if we take gwethion to be the genitive plural of gwath (q.v.)
              gwilwering "butterfly". The basic stem is WIL "fly, float in air" (LR:398); the rest of the word is somewhat obscure. Ilkorin gwilwering, Quenya wilwarin (wilwarind-), Telerin vilverin and "Noldorin"/Sindarin gwilwileth cannot all be cognates; the latter evidently represents a reduplicated stem WIL-WIL (combined with the feminine ending -eth). The middle element of the Ilkorin, Quenya and Telerin words, variously appearing as -wer-, -war- and -ver-, is difficult to explain; Quenya wilwarin suggests that the original middle vowel was a, in Ilkorin umlauted to e by the i in the next syllable. Quenya wilwarin may be analyzed as wilwa-rin, the first element meaning "air" (cognate of Ilkorin gwelo, q.v.) and the final element being obscure. If, on the other hand, we take -in(d)- to be an independent suffix, it may be analyzed as a feminine ending (cf. such names as Serindë), semantically corresponding to Sindarin -eth in gwilwileth. Unlike -eth, the -ing of Ilkorin gwilwering is apparently cognate with Quenya -ind-; for this (rare) correspondence ng = nd compare gangel "harp" = Quenya ñandellë. Perhaps this shift is due to assimilation (to g occurring elsewhere in the word) in both cases.
              gwing "spindrift, flying spray". Derived from an undefined stem WIG (LR:398); primitive wingê shows nasal infixion and an ending -ê that has no clear-cut meaning; here it may be relevant that it occurs in the words for a number of substances (see brith, maig). The derivation of wingê from WIG exactly parallels that of slingê "cobweb" from SLIG (LR:386, another undefined stem). In certain late writings, Tolkien considered removing Quenya wingë and Sindarin gwing (Ilkorin no longer existed) from the Elvish languages, making it a word of the Mannish language of the people of Bëor instead (PM:370; the original Bëorian form is given as wing). Tolkien's final decision, however, seems to have been that wing- was to remain an Elvish word. He now proposed an etymology by which it was related to a Quenya word winta- "scatter" (PM:377, note 24); this probably requires a stem *WIN, for the stem given in the Etymologies, WIG, could hardly be the base of a High-elven word winta-. A stem WIN was indeed listed in Etym (LR:399), but it was struck out and could for semantic reasons hardly have yielded a word for "spindrift" anyway (see gwini below).
              [gwini, later gwine "evening"] In the Etymologies, the entire entry for the stem WIN/WIND was struck out by Tolkien (LR:399). This undefined stem presumably meant much the same as the first word listed: primitive windi "blue-grey, pale blue or grey". Gwini, gwine is however derived from winjâ (spelt winyâ in LR:399). By its form, primitive winjâ would most likely be a verb or an adjective (ending -jâ); however, all its descendants in various languages are nouns for "evening". The development parallels that of gwene (q.v.): after the loss of the final vowel -â, the now final -j first became -i and then -e.
              gwo- "together" (LR:399 s.v. WÔ), prefix apparently occurring in older Ilkorin, but it was later lost because when w disappeared before o (cf. gôd), this prefix could no longer be distinguished from go, go- from" (q.v.) - The long vowel of the stem WÔ must have been shortened in this prefix (because it was unstressed?); otherwise, it would have become long û at the Ilkorin stage. Compare dûm, q.v., from dômi-.
              -in plural ending seen in adar pl. edrin, aman pl. emnin, boron pl. burnin, gangel pl. genglin, talum pl. telmin. The original Primitive Quendian language had a plural ending -î (as in kwendî "Quendi, Elves", WJ:360). There was also a plural element -m; LR:360 s.v. 3O makes reference to this "plural m". How and where this -m was originally used is not clear. It is tempting to assume that -in somehow represents -î-m, the two primitive plural morphemes being combined, though final or post-vocalic -m does not normally become -n in Ilkorin. This plural ending -in is also found in Doriathrin and Westron.
              -ion would seem to be the genitive plural ending, cf. Dor-thonion "Land of Pines", Torthurnion "King of Eagles", Urthin Gwethion *"Mountains of Shadows". Quenya has the same ending; in that language it represents the plural ending -i + o genitive marker + n another plural marker. See WJ:368, 407; cf. LR:360 s.v. 3O. We may assume that the Ilkorin ending has more or less the same etymology. See also -a (the singular genitive ending).
              laig "keen, sharp, fresh, lively". The stem LAIK is defined "keen, sharp, acute" (LR:367); a corresponding Quenya form is given as "laike", but this is almost certainly a misreading for laika (confused by the editor with the noun laike "acuteness, keenness of perception"? A form laika is actually mentioned in LR:368 s.v. LÁYAK, where the editor rightly supplies a cross-reference to the entry LAIK.) The corrected Quenya form points to a primitive form *laikâ with the frequent adjectival ending -â. Tolkien indicated that laig was not only descended from *laikâ, but also from laik-wâ "green", the latter being derived from the distinct stem LÁYAK (LR:368) and showing the adjectival ending -wâ (concerning which see adu). Both *laika and laik-wâ would become laig in Ilkorin, since after the loss of the final vowels, the now-final kw was evidently de-velarized to simple -k (compare alch, salch); hence laikâ and laik-wâ merged as laik, in turn becoming laig. There was evidently a semantic merger as well, the "keen, sharp, acute" of *laikâ being infused with the "green" of laik-wâ, resulting in the idea of something "fresh" or "lively".
              legol "nimble, active, running free". The stem LEK is defined as "loose, let loose, release" (LR:368). We are probably to assume a primitive form *lekla, that may be compared to hekla "any thing (or person) put aside" in relation to the primitive element HEKE "aside" (WJ:361). A *lekla would then be "any thing (or person) that is nimble, active, running free" (since the ending -la would seem to mean "thing or person"). Since legol is an adjective rather than a noun, it may also be referred to *leklâ, an adjectival form of *lekla, just like hekla is said to have an adjectival form heklâ (still WJ:361). After the loss of the final -a, -â, the final consonant of *lekl evidently became syllabic (pronounced lek-l, much like English "little" is pronounced lit-l), but such syllabic consonants were apparently disliked in the Eldarin tongues, for in all known cases a new vowel developed before them. In Ilkorin (and Sindarin) this vowel was o, hence *lekl > *lekol > legol, alternatively *lekl > *legl > legol if the voicing of post-vocalic unvoiced stops occurred before the new vowel developed (compare makla "sword" > *makl > Sindarin magl, later magol; other Ilkorin examples include tangol, ungol, ungor [q.v.] from earlier *tangl, *ungl, *ungr). Legol seems to occur in the name of the river Legolin; the ending -in could be an extra adjectival suffix (concerning this ending, see Ermabin)
              Lhinnon "musical land", a name of Ossiriand; the proper Ilkorin form seems to be Lindon, q.v. In the entry LIN2 (LR:369) the wording "Lindon, Lhinnon Ilk. name of Ossiriand" occurs. A literal interpretation of this would seem to indicate that Lhinnon is an Ilkorin word, so it has to be mentioned here, but Lhinnon definitely looks like "Noldorin" instead (the "almost-Sindarin" of the Etymologies; in LotR-style Sindarin we would expect *Linnon). Tolkien's intention may have been that Lindon is the proper form of the Ilkorin name and that Lhinnon is a "Noldorin" adaptation of it. It should be remembered that the Etymologies is for the most part Tolkien's rough working notes, not a work he thought others would ever read, so there was no reason for him to clarify things that would be perfectly obvious to him.
              lind "tuneful, sweet". Derived from a stem LIND "fair (especially of voice)" (LR:369); this stem cannot well be separated from LIN2 "sing" (according to the Etymologies, this stem was "originally GLIN", but WJ:382 mentions a primitive stem LIN with "primary reference to melodious or pleasing sound"; this agrees well with the meaning of Ilkorin lind). WJ:382 also mentions a reinforced form lind- that may well be equated with the entry LIND in the Etymologies. Lind corresponds to Quenya linda "fair, beautiful", pointing to a primitive form lindâ ("sweet-sounding") that is actually given in LR:386 s.v. SLIN. The ending -a is adjectival. - It should be noted that A-umlaut, whereby a final -â may turn a preceding i or u into e or o, respectively, does not occur in this word. (Contrast the "Noldorin" cognate lhend, Sindarin *lend, where the vowel has been umlauted. For an Ilkorin example of A-umlaut, cf. rest < *ristâ.) It seems that in Ilkorin, a vowel that is followed by a n is not subject to A-umlaut; cf. tund, tung corresponding to "Noldorin"/Sindarin tond, tong (see also tingla-). That only n and not nasals in general has the power to neutralize (or subsequently undo the effect of?) A-umlaut is evident from the word dem < dimbâ, where umlaut does occur despite the following nasal. It is also possible that A-umlaut originally did occur, lind indeed being *lend at one stage, but that a subsequent change turned e, o into i, u before n (and ñ = ng), as seems to be the case in the closely related Doriathrin language - this change incidentally undoing the effect of the earlier A-umlaut. Unfortunately we lack an Ilkorin example that could tell us whether original *o, *e (as opposed to the umlaut products of *u, *i) also turn into u, i before n, but cf. Doriathrin cwindor from kwentro. - In the Ilkorin word for "nightingale", murilind, the final element is probably a noun "singer" that is descended from *lindê instead; see murilind. We cannot tell whether a noun lind "singer" also appeared as an independent word in Ilkorin.
              Lindon, a name of Ossiriand. (Concerning the form Lhinnon, see separate entry.) Tolkien interpreted this name in several different ways over the years. In the Etymologies, it was defined as "musical land" ("because of water and birds"), hence to be referred to the stem LIN2 "sing" (LR:369); see lind above concerning this root. A primitive from Lindân-d is mentioned; this would seem to be the primitive adjective lindâ "sweet-sounding" (LR:386 s.v. SLIN) with an ending that is sometimes used to derive the name of countries (compare Sindarin Rochand, "Horseland, Rohan", Ossiriand "Land of the Seven Rivers"). - Later, when Ilkorin had been discarded as the language of Beleriand, Tolkien transferred the name Lindon to Nandorin (Green-elven) and derived it from primitive Lindânâ (WJ:385), which is clearly Lindâ "Linda, Elf of the Third Clan" + the well-attested adjectival ending -nâ. Lindânâ therefore means simply "(Land) of the Lindar", "Lindarin (Land)" (named after the Lindarin Green-elves that entered Beleriand and settled in Ossiriand).
              line "pool". Derived from the stem LIN1, that is similarly glossed (LR:369); the Quenya cognate linya indicates a primitive form *linjâ. The ending -jâ is most often used to derive adjectives or verbs, but here it is used to produce a noun (compare gwine "evening" < *winjâ). The development parallels that of gwini > gwine and gwene: linjâ > *linj > early Ilkorin *lini > later line.
              mab "hand". Not a direct cognate of Quenya má, though the stem MAP "lay hold of with hand, seize" is probably ultimately related to MA3, the stem yielding the Quenya word (both stems listed in LR:371). The primitive form is given as mapâ, the ending -â being simply a noun-former here (possibly the stem-vowel suffixed and lengthened). A mapâ is apparently *"that which seizes". Compare wedâ *"that which binds" = a bond; see gwedh.
              Mablosgen "emptyhanded", epithet of Beren who returned to Menegroth without the Silmaril, corresponding to Sindarin Camlost. Listed in LR:371 s.v. MAP, where we also find the first element of this name: mab "hand". The element los apparently represents *lustâ "empty", the cognate of Quenya lusta (LR:370, stem LUS); notice how the original final *-â causes u to become o by umlaut. The ending -en could be adjectival, representing primitive -inâ, but it is unclear where the intervening g (Mablosgen) comes from. Perhaps losg represents *luskâ rather than *lustâ, since -kâ is also an adjectival ending. An alternative form Mablothren is listed in the entry LUS, though it is somewhat unclear what language it belongs to; this plainly represents mab + lost + -ren, the latter perhaps being a variant of the adjectival ending -rin as in Ermabrin (another of Beren's epithets). In Sindarin, -ren is the normal form of the ending in question.
              maig "dough". Derived from a verbal stem MASAG "knead, make soft by rubbing, kneading, etc." (LR:371). MASAG looks like an expansion of a simpler stem *MAS, of which the synonymous stem MBAS "knead" (LR:372) may be a strengthened form. One primitive form of maig is given as mazgê; the ending -ê may sometimes signify a substance (see brith, gwing). Mazgê is therefore a *"kneaded substance". The oldest primitive form was probably *masgê, before the s of the stem MASAG was voiced to z by contact with the following voiced stop g. According to the information provided in WJ:402, the original form of the name of the Vala Estë was esdê, becoming ezdê in Common Eldarin; in light of this, mazgê may also be seen as a Common Eldarin form. - As in Old Sindarin, z followed by a consonant became i in Ilkorin, merging with the preceding vowel to produce a diphthong, in this case ai. Hence mazgê > maig.
              môr "night". Derived from the well-known stem MOR "dark, black", not explicitly glossed in LR:373, but see Letters:382. A primitive form mori is given in both of these sources, glossed "black" in LR:373 and "dark(ness)" in Letters:382. Since the ending -i is often used to derive colour-adjectives, the glosses "black, dark" are probably the most literal, but it is interesting that according to Letters, mori can also be an abstract "darkness". This explains how it could end up as a noun "night" in Ilkorin. (In LR:394, where a "Noldorin" loan-word from Ilkorin is discussed, Tolkien emphasized that "mori did not = 'night' in N", in a context implying that 'night' is precisely what it meant in Ilkorin.) Mori would become *more in Common Eldarin, which is still the form used in Quenya, but as usual, the final vowel has been lost in Ilkorin. At some late stage, the vowel of the resulting monosyllable *mor was lengthened to produce môr; we know this was a late change, for earlier long ô would have become û in Ilkorin (cf. dûm < dômi-). Such lengthening only occurs when there is only one consonant following the vowel (hence no lengthening in words like rant, tass), and in a few words (bel, tal) it mysteriously fails to take place. For a similar example of a late lengthening, see tâch.
              murilind or myrilind, also murlind or myrlind, "nightingale" (LR:373 s.v. MOR, see also LR:394). (The form "murulind" in LR:373 s.v. MOR is plainly a misreading for murilind as in LR:394; compare the alternative form myrilind mentioned just afterwards, and also Quenya morilindë mentioned just before.) The last element in all these Ilkorin forms, lind, evidently means "singer". If so it is related to, but not identical to the adjective lind "tuneful, sweet" (q.v.), that is descended from lindâ. Quenya morilindë indicates that the lind of murilind etc. descends from *lindê. Compare the (evidently Old Sindarin) word linde "singer / singing" that is mentioned in WJ:309, pointing in the same direction. The stem is transparently LIN, used with reference to "melodious or pleasing sound" (see lind for reference). In *lindê, the stem is reinforced to lind-; the ending -ê can be taken either as feminine (and then probably agental) or abstract. By the first interpretation, *lindê means "(female) singer", the counterpart of masculine *lindô (whence Quenya lindo "singer, singing bird", LR:369 s.v. LIN2). - The first element of these words for "nightingale", variously muri-, myri-, mur-, myr-, would seem to mean "night", and is therefore a variant form of môr "night" above. In LR:394, Tolkien implies that the first element of murilind represents primitive mori, as does môr. It would seem that mori developed in another way when used as the first part of a compound; the -i was not final and therefore was not lost. Mori- as part of *morilindê evidently became muri- because the i caused the original o to umlaut to u; concerning this umlaut, compare boron pl. burnin and thoron pl. thurin. It would seem that the umlaut affection continued and eventually turned u into y, since Tolkien indicated that murilind also (later, or in another dialect?) appeared as myrilind. The shorter forms murlind, myrlind are either shortened from murilind, myrilind or represent a shorter primitive form *morlindê with the stem MOR "dark, black" prefixed directly to lindê. If so, the umlaut o > u > y must here be caused by the i of the second element lind.
              olg "hideous, horrible". The stem ÚLUG (LR:396) is not defined; it could be an extended form of a simpler stem *UL- that may underlie the Quenya word ulca "evil" (as in henulca "evileyed", SD:68; cf. the Qenya Lexicon p. 97). The Telerin cognate ulga points to a primitive form *ulgâ with the adjectival ending -â. This former presence of this ending is indicated by the fact that original *u has been umlauted to o.
              ôr "blood". This is also the meaning of the stem YAR (LR:400). The Quenya cognate yár, becoming yar- with a short vowel before an ending, points to a Primitive Quendian form *jara (*yara), later becoming *jâr (with endings *jar-) in Common Eldarin. (Compare PQ kwene "person", CE kwên, kwen-, Quenya qúen, quen-: WJ:360, 361.) In Ilkorin, long â becomes ô (compare for instance côm, tôr) and initial j- is lost (compare arn "red" < *jarnâ). - In later sources, there turn up words for "blood" that cannot be derived from YAR: Sindarin sereg and Quenya sercë, indicating a primitive form *serekê (see sereg in the Silmarillion Appendix). In Ilkorin, this would have become *serch.
              oth "war". The entry-head OKTÂ occurring in the Etymologies (LR:379) is not so much a "stem" as a primitive word in itself. As confirmed by Tolkien's cross-reference, the actual stem is KOT (or according to one revision, KOTH) "quarrel" (LR:365). OKTÂ is a rearranged form of this stem, the stem-vowel being prefixed instead of occurring between the two first consonants (concerning such rearrangements, see esg). The ending -â may be just a noun-former here (it is normally used to derive concrete nouns, not as here an abstract). The primitive cluster kt comes out as th in Ilkorin, perhaps via assimilated *tt (compare brith from b'rittê for the development tt > th). This is somewhat different from the development in Sindarin, in which language oktâ becomes (*oktha > *outha > *outh >) auth.
              rant "flow, course of river" (also attested as the final element of the river-name Adurant). The stem RAT is glossed "walk" (LR:383); the meaning of the "Noldorin"/Sindarin verb rado "to make a way, find a way" may also be noted: A river-course is where the water makes its way. We are probably to assume a primitive form *rantâ, a nasal-infixed variant of ratâ "path, track", alternatively *rantê with the ending -ê sometimes denoting a locality (cf. dor "land" < ndorê).
              rest "cut" (noun). In the Etymologies, Tolkien made two entries for the stem RIS (LR:384), first glossing it "slash, rip" and then "cut, cleave". One primitive form is given as rista-, apparently a verb "cut" like its identical Quenya descendant; we should probably read *ristâ- with a long final vowel since a short final -a would not survive into Quenya rista-. The Quenya word rista may be used both as a verb "to cut" and as a noun "cut", and Ilkorin rest must also derive from *ristâ used as a noun. The final *-â caused umlaut before it was lost, changing *i to e (compare the umlaut *u > o in *ulgâ > olg, q.v.)
              rond "domed roof" (LR:384 s.v. ROD), also translated "vault" when appearing as a part of the name Elrond "vault of heaven". See Elrond concerning the various etymologies Tolkien proposed for this word; the one given in the Etymologies differs from the one given later, when Elrond had become a Sindarin name.
              salch "grass". The relevant entry-head in the Etymologies is SALÁK-(WÊ) (LR:385), which seems to indicate a stem SALÁK with an ending -wê; Quenya salquë and Old "Noldorin"/Sindarin salape likewise suggest a primitive form *salakwê. The ending -wê is somewhat surprising; LR:398 s.v. WEG mentions an "abstract suffix" -wê, but "grass" is obviously not an abstract. However, this ending also occurs on some concrete nouns, like atakwê "building" (LR:390). The development is apparently meant to be *salakwê > *salkwê > *salkw > *salk > Ilkorin salch. Compare alch "swan" from" alk-wâ. Again we run into the problem of kw/p if Ilkorin is to be adopted as a form of Sindarin; see alch.
              saum "drinking-vessel". The stem SUK means "drink" (LR:388); the primitive form is given as sukmâ, showing "a suffix frequent in the names of implements" (WJ:416, where a synonymous primitive word julmâ is mentioned, derived from another stem for "drink"). To better understand the development from primitive uk + nasal to Ilkorin au + nasal, we must realize that sukmâ would later appear as *sokma because the final A would umlaut the stem-vowel. Compare Ilkorin caun < coun (q.v.) from earlier kogna, ultimately derived from primitive ku3nâ, stem KU3. It may be that *sokma later appeared as *sogma, if (as in the evolution of Sindarin) k became g before a nasal. If so, the evolution from *sogma to saum would parallel that from kogna to coun > caun (and we must therefore assume that saum was *soum in early Ilkorin).
              tâch "firm, stiff, solid". Since this word is quite synonymous with Quenya tanca and "Noldorin"/Sindarin tanc, and is moreover derived from the same stem TAK "fix, make fast" (LR:389), it would be tempting to assume that all three words are cognate. Tanca, tanc are most likely to be derived from *tankâ (with nasal infixion and adjectival -â). The spirant ch of the Ilkorin word probably represents older *kk; compare th from earlier *tt in brith < *b'rittê. But can we turn *tankâ into *takka? An assimilation nk > kk is not implausible as such (it occurs in Adûnaic, SD:420). But later in the same entry in the Etymologies, Tolkien derived Ilkorin tangol "pin, brooch" from tankla. Here, nk does not become kk, or we would expect *tachol (as in "Noldorin"/Sindarin). It seems, then, that tâch cannot be a direct cognate of tanca, tanc < *tankâ. Rather we must refer this Ilkorin word to a distinct primitive adjective *takkâ, perhaps the stem TAK with the adjectival ending -kâ (cf. for instance poikâ "clean" from the stem POY, LR:382), alternatively TAK with a medial fortification k > kk and the adjectival ending -â). - When the final vowel had been dropped so that only a monosyllable remained, the stem-vowel was lengthened to produce tâch. This change was late, since early long â produces Ilkorin ô (see ôr, tôr). For another example of a late lengthening of the stem-vowel in a monosyllable, see môr.
              taga "he fixes, constructs, makes". The stem TAK meant much the same: "fix, make fast" (LR:389). There would be a primitive verbal stem *taka- or *takâ-, but how did the final vowel survive into Ilkorin when final -a, -â is otherwise lost? It should be noted that Tolkien's gloss for taga includes a pronoun, "he fixes". In the primitive language, the pronoun "he" (sô or so) also appeared as a suffix; Tolkien made a reference to the "-so inflexion of verbs" (LR:385 s.v. S-). In the primitive language, "he fixes" might therefore be *takâ-so, *takaso. After the loss of the final vowels, we would be left with *takas; thereafter, post-vocalic lenition set in. If the development of Ilkorin parallels that of Sindarin, single post-vocalic s became h at the same time k became g; hence we would have *takas > *tagah, but final -h did not survive, so in Ilkorin, the word appears as taga. Final -a thus survived because it was not final at an earlier stage, when final vowels were lost. The form taga probably covers the entire third person singular: "he, she, it fixes" (since primitive *takaso, *takase and *takasa, with endings for "he, she, it" respectively, would all end up as taga in Ilkorin). Compare toga "he brings", tolda "he fetches".
              taig "deep". Tolkien's notes on the etymology of this word are vague. It the Etymologies, it is listed in the entry for the stem AYAK "sharp, pointed" (LR:349). A Quenya word aiqua "steep" is tentatively suggested to be related to "Noldorin" oeglir "range of mountain peaks" (in later Sindarin aeglir). Quenya aiqua would most likely represent primitive *aikwâ, a shortened form of the stem AYAK + the adjectival ending -wâ. (As Tolkien says, the other words can only be "related" and not direct cognates, for *aikwâ would produce "Noldorin" **oeb-, Sindarin **aeb-, not oeg-, aeg-.) In Ilkorin, aikwâ would come out as *aig. The unexpected appearance of the initial consonant seen in taig Tolkien explained by postulating that this Ilkorin word was "blended with tára, see TÂ". LR:389 lists a stem TÂ, TA3 "high, lofty, noble" with a derivative târâ "lofty" (the ending -râ is adjectival; see gôr), Quenya tára. It does not make much sense that a Quenya word as such could influence an Ilkorin word, so tára in LR:349 s.v. AYAK is better taken as some early Beleriandic form descended from târâ. However, it is not quite obvious how the meanings "steep" and "lofty" could merge to produce the meaning "deep".
              Taiglin, river-name (= Teiglin in the published Silmarillion). In the Etymologies, is not clear from the wording in the entry LIN1 (LR:369) what language this word belongs to (it is mentioned after the Quenya word ailin "pool", but Taiglin is obviously not Quenya, since this language cannot have g in this position). The name looks like Ilkorin, though, and the Etymologies lists many other Beleriandic river-names that are explicitly identified as Ilkorin (Adurant, Aros, Askar/Ascar, Brilthor, Brithon, Gelion, Duilwen, Esgalduin, Legolin, Thalos). In the Etymologies, it seems that Taiglin is intended to mean *"deep pool" (taig "deep", q.v., plus an element -lin evidently derived from LIN1 "pool", the entry where the name occurs in the Etymologies). Later, when this river-name had to become a Sindarin word because Ilkorin had been discarded as the language of Beleriand, Tolkien interpreted it quite differently: The (Old Sindarin?) elements were now said to be taika "boundary" and linde "singer / singing", the latter being used as the name "of many rivers of quick course that make a rippling sound" (WJ:309); hence the name would mean "boundary river". As Tolkien noted in the same place, taika + linde would rather produce Taeglind in Sindarin (and *Taiglind in the Ilkorin of the Etymologies). As mentioned above, the published Silmarillion nonetheless has the form Teiglin; after Tolkien's revisions, this probably has to be understood as some dialectal Sindarin form.
              tal "foot", pl. tel. The stem TAL is itself glossed "foot" (LR:390). The Quenya cognate tál becomes tal- with a short vowel when an ending is added. This may suggest a Primitive Quendian form *tala, Common Eldarin *tâl, tal-, Quenya tál pl. tal-. (Compare PQ kwene "person", CE kwên, kwen-, Quenya qúen, quen-: WJ:360, 361.) However, Common Eldarin *tâl should have yielded Ilkorin **tôl (compare ôr, q.v.). Perhaps Ilkorin tal "foot" is simply the stem TAL with no additions; it is still strange that we do not at least have **tâl with a long vowel, since the vowel of monosyllables is lengthened in a number of other words (see môr, tâch, thôr, but contrast bel - are we to assume that lengthening does not take place before l?!) Whatever the case, the plural form tel is obviously due to i-umlaut caused by the primitive plural ending -î: *talî. It is not clear why the longer Ilkorin plural ending -in is not used in this word (**telin).
              talum "ground, floor", pl. telmin. Derived from a stem TALAM "floor, base, ground" that is said to be an extended form of TAL, the stem for "foot" (LR:390). TALAM shows ómataina (reduplication and suffixation of the stem-vowel) + a suffixed -m. Quenya talan and "Noldorin"/Sindarin talaf point to a primitive form *talama (the final vowel cannot be reconstructed with certainty; it may also have been -e or -o). Following the syncope of middle vowels and the loss of final vowels, a form *talm would arise in the branch of Elvish leading to Ilkorin, and this form may still be reflected in the plural telmin. But it would seem that the final nasal of talm became syllabic (constituting a syllable by itself, tal-m), and a vowel developed before it to make a regular syllable. In Ilkorin, the vowel developing before final syllabic consonants is normally o (see tangol, tovon, ungor, ungol), but here we see u instead; perhaps this is the vowel that regularly develops before syllabic m (no other examples in Ilkorin, but compare Tolkien's early "Gnomish" language, where a word like telm "roof, sky" has an alternative form telum; see the Gnomish Lexicon p. 70). - The plural form telmin perfectly fits the pattern exemplified by the examples adar pl. edrin and aman pl. emnin: the plural is formed with the ending -in, triggering umlaut a > e in the stem of the noun itself, and the noun is contracted by omission of the second vowel (or, as we argued above, the plural telmin may reflect the simpler word *talm that evidently existed before the vowel u developed before the final nasal).
              tangol "pin, brooch". Derived from a nasal-infixed variant of the stem TAK "fix, make fast" (LR:389) combined with the ending -la: tankla. This ending seems to imply only "thing" or "person" somehow related to the meaning of the stem; see the entry legol about the example hekla. Sometimes, words for implements are derived by means of this ending, like tekla "pen" from TEK "write" (LR:391). A tankla is not precisely an implement (unlike takmâ "thing for fixing" from the same root, see taum below), but a thing that is itself "fixed", namely a brooch. Tankla is the only published primitive word to combine the ending -la with a nasal-infixed variant of the stem. It would seem that k, sandwiched between the voiced consonants n and l, became voiced to g (see tingla- concerning a possible parallel development *ntl > *ndl); later, after the loss of the final vowels, the final consonant of *tangl evidently became syllabic, and then a vowel o developed before it to produce tangol. (See legol concerning such developments.)
              targ "tough, stiff". Derived from an undefined stem TÁRAG (LR:390); the primitive form is given as targâ, apparently including the adjectival ending -â (WJ:382). Notice that while the unvoiced stops p, t, k turn into spirants f, th, ch following the liquids l, r (see alch), a voiced stop like g is not affected. (In Sindarin, g following a liquid apparently first became a spirant *gh paralleling the development of other stops in such a position. Later final gh following a consonant became -a; hence the "Noldorin"/Sindarin cognate of targ is tara, apparently representing older *targh.)
              tass "pin". The primitive form is given as taksê, defined as "nail", so in this case, there is a semantic change as well as a change in form. The stem TAK means "fix, make fast" (LR:389). In a number of primitive words, the ending -sê seems to denote something that is made by the verbal action of the stem, e.g. sjadsê (syadsê) "cleft, gash" in relation to its stem SYAD "shear through, cleave" (LR:389). But in taksê, the ending seems to denote an implement, a "nail" (the meaning of the primitive word) being a thing that is used to fix something. There seems to be no other example of -sê being used with such a meaning (this is rather how the ending -mâ is normally used; cf. taum below). In Ilkorin tass, original ks has assimilated to ss (which is also how primitive ps comes out in Ilkorin: see ass, tuss).
              taum "holder, socket, hasp, clasp, stable". This word, like tass above, is derived from the stem TAK "fix, make fast" (LR:389). The primitive form is given as takmâ, "thing for fixing". This is a very literal translation; see saum concerning the ending -mâ often occurring in the names of implements. As suggested in the entry saum, it may be that k followed by a nasal was voiced to g at some stage, so that there was an intermediate form *tagma; later, when g following a vowel became a spirant gh (see tûgh), the combination *-aghm- evidently proved unstable and became aum.
              taur 1. "great wood, forest"; 2. "wood" (material). The first of these descends from taurê "great wood, forest". The meaning of taurê primarily reflects the stem TÁWAR "wood, forest" (LR:391; taurê = *taw'rê; in a later source, Tolkien gave the stem for "wood" as TAWA [VT39:7], of which TÁWAR would be an expanded form). Except for this taurê, there are few examples of primitive words in -ê denoting localities, but ndorê "land" may be noted (LR:376 s.v. NDOR, also WJ:413). Tolkien noted that taurê was blended with taurâ "mighty" from the quite distinct stem TUR "power, control, mastery, victory" (LR:395); this taurâ shows A-infixion (according to VT39:10 an intensive formation) and the adjectival ending -â. Hence, taur was used with reference to great forests. But in Ilkorin, taur was also used for "wood" as a material, and according to LR:391 s.v. TÁWAR, this was descended from the distinct primitive form tawar, though taurê and tawar merged as taur in Ilkorin. Presumably tawar first became *tauar, the vowel following the diphthong later dropping out to produce taur.
              tel pl. of tal, q.v. (LR:390 s.v. TAL)
              telf "silver". This word alone indicates that in Ilkorin, Tolkien uses the spelling f for final [f], not ph as in "Noldorin"/Sindarin. The stem KYELEP itself means "silver" (LR:366-367); in the Etymologies, Tolkien hinted that there was also a side-form TELEP, but this may also be seen as the form KYELEP took in Telerin. In PM:366, a stem KYELEP "silver" is mentioned among the four stems for metals that are common to all Eldarin languages (the others denote gold, iron and copper), The primitive form of the word for "silver" is given (in Letters:426) as kjelepê (there spelt kyelepê). The final -ê may be the base-vowel suffixed and lengthened; on the other hand, one of the functions of the ending -ê is to denote a substance (cf. for instance mazgê "dough", LR:371 s.v. MASAG, or srawê "flesh", MR:350). Interestingly, primitive kj- comes out as t in Ilkorin (as in the Telerin of Aman: the Telerin cognate of telf is telepe, telpe). This is markedly different from Sindarin, in which language kj- became c (kjelepê yielding celeb). Primitive kj- probably first became ty (tj), cf. Quenya tyelpë, ty in turn being simplified to t. - The loss of the second vowel of kjelepê brought l and p into contact, and as usual, an unvoiced stop turns into a spirant following a liquid: lp > lf as in telf. See alch.
              telmin pl. of talum, q.v. (LR:390 s.v. TALAM)
              thall "steep, falling". The stem STAL is itself defined "steep" (LR:388); the primitive form, given as stalrê, is somewhat surprising, since the ending -rê is normally used to derive abstract or collective nouns. It may be that stalrê is a misreading for *stalrâ with a well attested adjectival ending (as for the endings -rê and -râ, compare dair). Notice the assimilation lr > ll in thall; initial st becomes th (presumably = þ, sc. th as in English think). In Sindarin and Quenya, initial st- developed in the same way (þ subsequently becoming s in Noldorin Quenya).
              thalos "torrent", also used as name of a river in Ossiriand, Thalos apparently being thought of as the "Torrent" par excellence. Derived from the same stem STAL "steep" (LR:388) as thall above. The ending -os is somewhat obscure (it also occurs in another Ilkorin river-name, Aros); we must assume that the primitive form was something like *stalossê (while Aros perhaps descends from *jarossê). The ending -ssê may sometimes denote a locality; cf. its Quenya descendant -ssë in such words as aicassë "mountain peak", apparently derived from the adjective aica "sharp": LR:349 s.v. AYAK. Hence *stalossê = ?"steep place" (and *jarossê = ?"red place"). It is not clear where the middle vowel of *stalossê would come from, though.
              thavon "carpenter, wright, builder". The stem STAB as such is not defined; the words derived from it revolve around the theme of architectural constructs made of wood: chambers, halls, wooden pillars (LR:387-388). Thavon seems to be the cognate of Quenya samno (but not directly cognate with "Noldorin"/Sindarin thavron; though this word seems more similar, it actually employs a distinct ending). Two primitive forms are listed, stabnô and stabrô; Ilkorin thavon (and Quenya samno) are clearly to be referred to the former. The ending -nô in stabnô is masculine and often agental; cf. for instance tirnô "watcher" in relation to the stem TIR "watch, guard" (LR:394; tirnô occurs in the compound khalatirnô, khalatirno "fish-watcher"). The development must have been *stabnô > *thavno > *thavn after the loss of the final vowels. Tolkien may have imagined that the now final *-n became syllabic (constituting a syllable by itself, *thavn being pronounced *thav-n, like English oven is pronounced ov-n). Then, as in the case of final syllabic l and r, a vowel o developed before it to produce thavon. Compare tubnâ > *tov-n > tovon, q.v.
              thind "grey, pale", also Thind as the name of "Elwë's brother". (Thind is the same person as Thingol, King of Doriath; while Tolkien later made Elwë/Elu a name of Thingol himself, "Elwë" here corresponds to the character Tolkien later called Olwë, Lord of the Teleri. See LR:217, note 23, about these revisions.) Both of these words are derived from a stem THIN (LR:392), not itself defined but yielding words having to do with grayness: it is suggested that it may be related to TIN "sparkle, emit slender (silver, pale) beams", LR:393. Both the adjective thind and the name Thind clearly reflect a strengthened stem thind-, a medial fortification turning N into ND. However, this is where the similarities end, for while the adjective and the name could no longer be distinguished in Ilkorin, they actually represent quite different formations at the primitive stage. The original Primitive Quendian colour-adjective is given as thindi (both in LR:392 and WJ:384; notice that primitive th is aspirated t, in Ilkorin becoming th as in English think; this parallels the development in "Noldorin"/Sindarin). That the final -i is not just the stem-vowel suffixed is seen from the fact that other colour-adjectives with other stem-vowels than i show the same ending (e.g. karani "red", LR:362 s.v. KARÁN). The name Thind, on the other hand, must descend from *Thindô with the masculine ending -ô, still reflected in the Quenya and Telerin cognates: Sindo, Findo. - In the Etymologies, it is stated that Thind was "later in Doriath called Thingol" (LR:392 s.v. THIN); see below.
              Thingol (masc. name). Derived from the stem THIN (LR:392), not defined as such, but it is suggested to be a variant of TIN "sparkle, emit slender (silver, pale) beams". THIN yields words for "grey, pale, evening, fade". This entry in the Etymologies implies that Thingol's name in the primitive language was *Thindô *"Grey One" (primitive form not given as such, but compare Quenya Sindo, Telerin Findo). *Thindô yielded Ilkorin Thind, q.v. above. But according to the same source, Thind was later called Thingol as a compound of Thind (Thin-) and gôl (-gol), the latter element meaning "wise" (derived from the same well-known stem ÑGOL "wise, wisdom, be wise" [LR:377] that is also the source of the Quenya word Noldo). However, Tolkien eventually rejected this explanation of the second element in the name Thingol. In later sources, the name Thingol is interpreted "Grey-cloak" (so already in LotR Appendix A: "Lúthien Tinúviel was the daughter of King Thingol Grey-cloak...") In MR:385, the second element of Thingol (Quenya Sindikollo) is said to be kolla, which is defined as "borne, worn, especially [when used as a noun] a vestment or cloak". (The final -a of kolla is replaced by the masculine ending -o in the name Sindikollo.) It seems that kolla is a Quenya word; the primitive form can tentatively be given as *kolnâ, sc. a stem *KOL "bear" (cf. Quenya colindo "bearer" in Cormacolindor "Ring-bearers", LotR3:VI ch. 4, translated in Letters:308) with the adjectival/past participle ending -nâ. If the masculinized form kollo descends from a word that existed already in the primitive language, this would be *kolnô. Primitive *Thindikolnô would come out as Sindikollo (or *Sindikoldo) in Quenya, and evidently it would also become Thingol in the Ilkorin of the Etymologies. In Sindarin, in which language initial k (c) is often lenited to g when a word appears as the second element in a compound, *kolnâ or *kolnô would indeed become -gol in this position. A similar lenition C > G is seen to occur in Ilkorin, cf. basgorn "round bread" (bast "bread" + corn "round").
              thôn "pine-tree", gen. pl. thonion in Dor-thonion "Land of Pines" (concerning dor "land", see separate entry). The stem is given as THÔN, not itself defined (LR:392). It is strange that the vowel is indicated to be long, for a primitive form *thôn- would have produced Ilkorin **thûn (cf. dûm, q.v., from dômi-). Rather the primitive form must be *thon-, probably with some final vowel that was later lost, the vowel of the resulting monosyllable *thon being lengthened in Ilkorin to produce thôn. Compare mori > *more > *mor > môr, q.v.; cf. also thôr below. Since this lengthening only occurred in monosyllables, it is not surprising to see that the vowel remains short in the genitive plural thonion. - The Etymologies suggests that thôn was borrowed into early "Noldorin" and later became thaun, but while "Noldorin" normally closely corresponds to later Sindarin, the Sindarin word for "pine-tree" is not thaun, but the same as the Ilkorin word: Thôn. When Tolkien discarded Ilkorin as the language of Beleriand, he carried over into Sindarin a number of Ilkorin names, including Dor-thonion (in Grey-elven the ending -ion is better seen as adjectival, since Sindarin has no genitive endings).
              thôr "sweeping, leaping down". It is not clear from the wording in LR:393 whether this word is intended to be Ilkorin or "Noldorin"/Sindarin; it may occur in the river-name Brilthor (q.v.), that is explicitly said to be Ilkorin. In any case, thôr is probably to be referred to a primitive adjective *thorâ, derived from the stem THOR "come swooping down". After the loss of the final vowel, the resulting monosyllable *thor was apparently lengthened to thôr; compare môr, tâch. Another word from the same stem that may or may not be Ilkorin is thórod "torrent", evidently representing an extended stem *thórot-.
              thorn "eagle" (also in the names Elthorn "eagle of sky" and Thorntor *"Eagle-king"), pl. thurin, and evidently genitive pl. thurnion in Torthurnion, q.v. The relevant entry-head in the Etymologies (LR:392), namely THOR-, THORON-, clearly indicates a simple stem THOR with an extended form THORON (showing ómataina, sc. a reduplicated stem-vowel, plus a suffixed consonant -n). As mentioned in the discussion of thôr above, Tolkien assigned the root-sense "come swooping down" to the basic stem THOR (LR:393), and this can of course be applied to eagles. The syncopated form thorn is what we would expect in Ilkorin; however, the plural thurin is a bit strange. The change o > u is apparently an umlaut caused by the vowel of the plural ending -in (compare burnin as the pl. of boron, q.v.) But why does the n of thorn disappear in the plural form thurin? Is this plural formed from the simple stem THOR instead of the expanded form THORON? The example burnin indicates that a final -n does not regularly drop out when the plural ending -in is added. It may be that thurin is a misreading or misprint for *thurnin, and the genitive plural thurnion seems to support this theory.
              Thorntor evidently = Sindarin Thorondor, "King of Eagles", literally *"Eagle-king" (LR:392 s.v. THOR/THORON). The elements are thorn "eagle" + -tor "king" (uncompounded tôr); see thorn, tôr for discussion. Notably, the second element of the compound is not lenited; since *corn becomes -gorn in basgorn (q.v.), we might have expected -tor to become -dor here, as in the Sindarin form of the name.
              thrôn "stiff, hard". The basic stem is STAR "stiff", with an extended form STARAN (sc. STAR with -n suffixed to ómataina, the stem-vowel reduplicated). The primitive form of thrôn is given as starâna. The ending -a could be adjectival (but we would then expect long -â), or it may be just another reduplication of the stem-vowel. Notice the lengthening of the middle vowel of starâna; surely this vowel also received the accent at some stage, and the unaccented vowel before it dropped out: starâna > *st'râna. Compare such a primitive word as b'rittê "gravel"; the stem BIRÍT indicates that the older form was *biríttê (see brith); cf. also Ilkorin trêw from terêwâ, apparently accented on ê. From *st'râna we get directly to Ilkorin thrôn, the changes st > th, â > ô and final vowels > zero all being familiar. - In compounds, thrôn may be shortened to thron (as in thron-ding below).
              thron-ding *"stiff-twang", element in the name Balthronding, Belthronding (LR:388 s.v. STARAN); see Belthronding.
              thúren "guarded, hidden". This is the past participle based on the verbal stem THUR "surround, fence, ward, hedge in, secrete" (LR:393). A "Noldorin"/Sindarin past participle equivalent in meaning, thoren, is derived from tháurênâ later in the same entry; yet tháurênâ does not seem to be the ancestral form of thúren, for it is difficult to get from primitive áu to Ilkorin ú. A primitive form *thûrinâ seems much more likely. Compare such a past participle as Quenya rácina "broken" vs. the verb rac- "break" (MC:222-223); rácina would represent primitive *râkinâ, a form wholly parallel to *thûrinâ: the past participle is in both cases formed with the ending -inâ combined with lengthening of the stem-vowel. In Ilkorin, *thûrinâ becomes thúren because the final -â, before it was lost, umlauted the i in the syllable before it to e (cf. for instance dem concerning this umlaut).
              thurin pl. of thorn, q.v. (LR:392 s.v. THOR, THORON)
              tim "spark, star" (the first gloss is the most literal, "star" being a secondary application). Tolkien suggested that the stem TIN "sparkle, emit slender (silver, pale) beams" (LR:393) was related to THIN, the stem yielding words for "grey" (see thind above). The primitive form tinmê by its form looks like an abstract formation; concerning the abstract ending -mê, cf. for instance julmê "drinking, carousal" in relation to the verbal stem JULU "drink", WJ:416). Of course, there is no great semantic leap from an abstract *"sparkling" to a concrete noun "spark". It may be that nm early assimilated to mm in the branch of Elvish that leads to Ilkorin, tinmê becoming *timme and later *timm > tim (while nm was dissimilated to nw in Quenya, *tinmê producing High-elven tinwë). - Later, after Ilkorin had been discarded as the language of Beleriand, Tolkien spoke of tim as a Sindarin word (MR:388).
              tindum is glossed both "starlight, twilight" (LR:393 s.v. TIN) and "starry twilight" (LR:354 s.v. DOMO). The final element is plainly just a shortened form of dûm "twilight", see separate entry. The prefixed element tin- reflects the stem TIN "sparkle, emit slender (silver, pale) beams", that is also the source of an Ilkorin word for "star" (see tim above). Since tindum corresponds to Quenya tindómë and "Noldorin"/Sindarin tinnu, it would seem that this compound was made already in Common Eldarin: primitive *tindómi? - Tor Tinduma "King of Twilight", a title of Thingol (LR:393 s.v. TIN, LR:392 s.v. THIN); tindum here has the genitive ending -a.
              tingla- "sparkle", verb. This is a strange formation. The stem TIN also means "sparkle" or "emit slender (silver, pale) beams" (LR:393), but this entry in the Etymologies provides no help regarding the ending -gla-. Conceivably, tingla- could be the cognate of the Quenya form tintila- "sparkle" attested in Namárië in LotR. We must then imagine a development something like this: primitive *tintilâ- is first syncopated to *tintlâ-; the t, sandwiched between the voiced consonants n and l, is itself voiced to produce d; the resulting form *tindla- then becomes tingla-, since a shift nd > ng is seen in a few Ilkorin words (see gangel, gwilwering). The final -a of tingla does not change the stem-vowel to e by umlaut (as in rest < *ristâ) because there is an n following the stem-vowel; compare lind (not **lend) < lindâ.
              tiog "thick, fat". Derived from a stem TIW, similarly defined (LR:394). The primitive form is given as tiukâ, -kâ being an adjectival ending (cf. poikâ "clean" from POY, LR:382; see also tâch). As usual, a final semi-vowel in the stem turns into a full vowel when followed by a consonant, hence TIW-kâ > tiukâ for *tiwkâ (and likewise poikâ for *poykâ, *pojkâ). The final -â umlauted the u to o before it was lost: tiukâ > *tioka > *tiok > Ilkorin tiog.
              toga "he brings". The stem TUK likewise means "draw, bring" (LR:395); we are clearly to assume a primitive verbal stem *tukâ- (the â umlauting the u to o), to which the ending -so "he" was added; cf. taga "he fixes, constructs, makes" for further discussion of the development.
              tolda "he fetches". The ending -a ("he" + present tense) would represent *-aso as in taga (and toga above). The stem TUL means "come, approach, move towards (point of speaker)" (LR:395); the formation tultâ- "make come" is perhaps our best example of the verbal ending -tâ being used with a causative meaning. In the Ilkorin form tolda-, the final -a has umlauted the original u to o as we would expect (cf. toga above), but the change lt > ld is somewhat surprising. In our other example, lt becomes lth instead: Balthor "Vala-king" (Bal + tôr). This, of course, is a compound and therefore represents l + t in secondary contact, whereas tolda represents a case where the combination lt existed already in the primitive language (tultâ); this may account for the differing Ilkorin developments of this cluster.
              tôr "king", only used of Thingol (LR:389 s.v. TÂ, TA3), but the plural form tórin "kings" was also used of the Valar (LR:350 s.v. BAL). The original stem was apparently TA3, glossed "high, lofty; noble", later becoming TÂ when the back spirant 3 was lost and the vowel was lengthened in compensation. A primitive adjective târâ (presumably representing even older *ta3râ) is mentioned, defined as "lofty"; the ending -râ is adjectival (cf. Quenya laira "shady" form *dairâ, see dair; see also ungor). While Ilkorin tôr could be derived from târâ itself, Tolkien also listed a primitive noun târo, apparently a personalized form of târâ (the masculine ending -o, more commonly -ô, replacing the final -â). Târo "king" ("only used of the legitimate kings of whole tribes") is the form quoted as the source of Ilkorin tôr, the changes being perfectly regular: final vowel lost and long â becoming ô. In various phrases and compounds tôr is shortened to tor, apparently because it is not fully accented: Tor Thingol "King Thingol" (LR:390 s.v. TÂ, TA3), also compounded Torthingol (LR:392 s.v. THIN). Tor Tinduma "King of Twilight", a title of Thingol (LR:393 s.v. TIN, LR:392 s.v. THIN; see tindum), Balthor "Vala-king" (LR:350 s.v. BAL, -tor becoming -thor following the liquid l), Thorntor *"Eagle-king", King of Eagles, = Sindarin Thorondor (LR:392 s.v. THOR, THORON).
              tóril a title of Melian, evidently meaning "queen". Listed in the same entry TÂ, TA3 as tôr "king" (LR:389; see above) and clearly derived from it by means of a feminine ending. There are no other examples of feminine -il in Ilkorin, but compare Doriathrin Thuringwethil "(woman of) secret shadow" (LR:393 s.v. THUR) and Quenya tavaril "female dryad" (LR:391 s.v. TÁWAR).
              Torthingol "King Thingol" (LR:392 s.v. THIN), the title tor (tôr) "king" being directly prefixed to the name Thingol (compare uncompounded Tor Thingol in LR:389 s.v. TÂ, TA3).
              Torthurnion evidently = S Thorondor, "King of Eagles" (LR:392 s.v. THOR, THORON). The elements are tôr "king" (q.v.) and thurnion "of Eagles"; see thorn.
              tovon "lowlying, deep, low". The stem TUB (LR:394) is not defined, but all of its derivatives revolve around the same theme as the gloss of tovon. The primitive form is given as tubnâ "deep", the ending -nâ being adjectival. Undoubtedly we are to assume a development like tubnâ > *tobna (the final vowel causing A-umlaut so that u becomes o) > tobn, tovn > tovon. As in the case of thavon (q.v.), the final -n evidently became syllabic, and a vowel o regularly developed before it.
              trêw "fine, slender". The stem TER is apparently verbal, since it is glossed "pierce", though no verbal derivatives are listed for any Elvish language (LR:392; an extended form TERES is also listed, but seems to be irrelevant for trêw). The primitive form is given as terêwâ, glossed "piercing, keen"; notice the semantic drift that has taken place between the primitive language and Ilkorin. Terêwâ must represent an extended stem TERE with ómataina (suffixed stem-vowel), here actually terê- with lengthened ómataina, and an ending -wâ that is normally adjectival (concerning this ending, see adu). Here, the ending -wâ takes on an almost participial force, being used to derive a word for "piercing" from a verbal stem meaning "pierce". It seems clear that it was the middle vowel of terêwâ that received the accent, and as in the case of other Ilkorin words, the unaccented vowel before it drops out (compare brith < b'rittê < *birittê, or thrôn < starâna).
              tûgh, tû "muscle, sinew, vigour, physical strength". The stem TUG is not defined; all of its derivatives have to do with things that are somehow strong or taut. The primitive form is given as tûgu; the ending -u may occur in the names of body-parts: cf. mbundu "snout, nose, cape" (LR:372 s.v. MBUD) and ranku "arm" (LR:382 s.v. RAK). Therefore, it may be that "muscle, sinew" is the original, basic meaning of tugû, from which the more abstract meanings "vigour" and "physical strength" developed. - Post-vocalically, original g became a back spirant gh in Ilkorin, like all voiced stops turned into spirants in this positions. However, the sound gh was lost in later Ilkorin, as the alternative form tû would indicate (this parallels the development in "Noldorin"/Sindarin, that also has tû).
              tund "tall". Derived from an undefined stem TUN (LR:395); the primitive form tundâ shows a medial fortification n > nd + the adjectival ending -â. Again we see that a vowel followed by an n is not affected by A-umlaut; hence this u is not umlauted to o (contrast olg, tovon above). Compare tung below; cf. also lind (rather than *lend < *lindâ).
              tung "taut, tight", (of strings:) "resonant". Concerning the stem TUG (LR:394), see tûgh, tû above. The primitive form is given as tungâ, showing nasal infixion and the same adjectival ending -â as in tundâ (see tund above). Taken together, the words lind, tund and tung (< lindâ, tundâ, tungâ) seem to confirm that a vowel followed by an n is not affected by A-umlaut (or we would have seen **lend, **tond, **tong). Alternatively, as we suggested in the entry lind, it may be that these forms did exist at one stage, but a subsequent change turned e, o into i, u before n (and ñ = ng), as seems to be the case in Doriathrin - this change incidentally undoing the effect of the earlier A-umlaut.
              tuss "thatch". The stem TUP (LR:395) is not defined, but probably has to do with some kind of roofing; compare the Quenya verb untúpa "down-roofs" = "covers" in Namárië in LotR (for the translation "down-roofs", see the interlinear rendering in RGEO:67). The primitive form of tuss is given as tupsê; this is a quite unique example of the ending -sê being used to derive a word for a construction. Notice that as in ass < apsâ, the primitive cluster ps is seen to assimilate to ss in Ilkorin (as does ks; cf. tass < taksê).
              Uduvon name of Melko[r]'s vaults in the North; = Quenya Utumno. The stem TUB is not defined as such, but all of its derivatives have to do with something that is deep or low-lying (LR:394). In the entry for this stem in the Etymologies, the primitive name of Utumno was given as Utubnu. This form cannot be fully explained; utub- is clearly a stemvowel-prefixed variant of TUB, but the ending -nu does not seem to occur in any other primitive word mentioned by Tolkien. Apart from this difficulty, the Ilkorin form Uduvon arises perfectly regularly: After the loss of the final vowels we would have *utubn; unvoiced stops become voiced and voiced stops become spirants following a vowel, resulting in a form *uduvn, and finally a vowel o develops before the final, syllabic -n to produce Uduvon. Compare another word derived from the same stem, tubnâ "deep" becoming Ilkorin tovon via *tobn > *tovn. - In a later source, Tolkien derived Quenya Utumno from Utupnu, translated "the Deep-hidden" (MR:69). Utupnu would probably have yielded Ilkorin *Udubon, but by the time this new primitive form was proposed, Tolkien had discarded Ilkorin as the language of the Elves of Beleriand, replacing it with Sindarin.
              ulgund, ulgon, ulion "monster, deformed and hideous creature". Derived from a stem ÚLUG (LR:396) that is not itself defined; it could be an extension of ULU "pour, flow" listed on the same page, but then the semantic connection would be obscure. Perhaps ÚLUG is somehow related to Quenya ulca "bad, evil" instead (henulca "evil-eyed", SD:68). Whatever the basic meaning of the stem may be, the primitive form of the word for "monster" etc. is given as ulgundô; the ending -ndô seems to be masculine (a nasal-infixed variant of -dô as in ñgolodô "Noldo"?) The three variant Ilkorin forms listed evidently developed in the same order: Ulgundô first yielded ulgund simply by the loss of the final vowels. Later, final -nd in polysyllabic words was evidently simplified to -n, so perhaps Elrond is a slightly archaic form (there is no trace of such a simplification in monosyllables; see gwen for a list of words that would have been affected). At the same time, the unaccented u became o, ulgund thus turning into ulgon (compare celon, q.v., from kelu-n). Later still, the cluster lg + a vowel became li- (this did not happen finally, cf. olg, not **oli). Perhaps lg first became *lgh; see taum for another example of the back spirant gh turning into a vowel (*-aghm > aum).
              ungol "darkness". The stem UÑG is not defined in the Etymologies (LR:396), but its derivatives have to do with darkness and gloom. The immediate ancestor of ungol is clearly meant to be *ungl, the l being syllabic so that a vowel o regularly developed before it (see legol). The l of *ungl would represent a remnant of a longer ending. The complete word may have been *uñgla; see legol concerning the ending -la. However, considering the abstract meaning ("darkness"), it may be more probable that Tolkien intended the primitive form to be *uñglê, since "nouns made with the ending -lê seem properly to have been universal and abstract" (VT39:16). Notice that the Ilkorin word ungol as such is not directly cognate with the first element of Quenya Ungoliantë ("Noldorin"/Sindarin adaptation: Ungoliant). This is a compound of Quenya ungo "dark shadow" (< *uñgô or *uñgu) and liantë "spider"; see LR:386 s.v. SLIG for the latter. - It may be that Tolkien altered the meaning of the stem UÑG, making it refer to spiders instead of darkness and gloom (if so, he probably reinterpreted the name Ungoliant, since he originally intended the second element to mean "spider"). What suggests that such a revision had taken place is the Quenya word ungwë. In the Etymologies, it was defined as "gloom", but in LotR Appendix E (in the list of the names of the Tengwa letters), ungwë is said to mean "spider's web" instead.
              ungor "black, dark, gloomy". Derived from the same undefined stem UÑG (LR:396) as ungol above. Again, this stem can be seen to refer to darkness and gloom (rather than spiders) at the time Tolkien wrote the Etymologies. Ungor is undoubtedly meant to represent *uñgrâ, the ending -râ being adjectival (cf. târa "lofty", see tôr, and Quenya laira "shady" form *dairâ, see dair). When the form uñgr, ungr arose after the loss of the final vowels, r became syllabic, and as in the case of final syllabic l, a vowel o developed before it to produce Ilkorin ungor.
              Urthin Gwethion a place-name listed in LR:397 s.v. WATH. The wording in this entry is not clear; Tolkien first wrote Urthin and above it Gwethion, but following Urthin the "Noldorin"/Sindarin form Eredwethion was added (= Ered Wethrin in the published Silmarillion). Urthin Gwethion would seem to be the Ilkorin equivalent of Eredwethion. Urthin is apparently "mountains", plural of *orth, derived from the stem ÓROT "height, mountain", in turn an extended form of ORO "up; rise; high" (LR:379; orth is attested as a Doriathrin word). Gwethion may be seen as the genitive plural of gwath "shade, *shadow", hence "mountains of shadows".
              usc "smoke". Derived from an undefined stem USUK; a primitive form is given as us(u)k-wê, the ending -wê denoting an abstract (see LR:398 s.v. WEG). If USUK is a verbal stem "reek, belch", us(u)k-wê could be basically a verbal noun "reeking, belching", later used in a more concrete sense for the smoke itself. It seems that after the loss of the final vowels, the now final -kw became -k in Ilkorin (compare alch, for *alk, from *alk-wâ "swan").
             

    APPENDIX:

    Edward Kloczko's theory that Tolkien recycled parts of Ilkorin as North Sindarin

    This article originally appeared in Tyalië Tyelelliéva #9 (October 1996) and is reproduced with Mr. Kloczko's kind permission. Some very small changes have been made to adapt it to other Ardalambion texts (like references being given in the form "WJ:400" rather than "400 WJ"). The letter @ represents an open variety of o; in Tolkien's writings (and Kloczko's original article) it is represented by o with a comma-like diacritic attached. The vowel @ is always long in this article, and was marked as such with a macron in the text in Tyalië Tyelelliéva. The abbreviations PQ and CE mean Primitive Quendian and Common Eldarin.

    Ilkorin and North Sindarin (Mithrim)
    When Etymologies was published, I was thrilled! - like every Vinya-Lambengolmo, - but at the same time quite puzzled. I did not know at that time what to do with the Ilkorin (and Doriathrin) words. I am, and was, against mixing these languages. Even if I quickly discovered that Ilkorin and Goldogrin looked very much alike, I also knew that Noldorin was to be called Sindarin later. What on earth (or Middle-earth) had Tolkien done with Ilkorin in the 1950's? Had he totally disregarded it, like Taliska, when he changed the history of the Elvish languages? Last year, finally, according to the War of the Jewels, we know that Sindarin possessed, according to Tolkien's latest discoveries, three dialects before the coming of the Etyañgoldi: the dialect of Doriath (the most archaic), that of the Falas, and Mithrim, also called North Sindarin.
              Let us look at what we know (from the published text) of this Mithrim dialect: "the diphthongization of ô and the opening of intervocalic m did not occur" (WJ:400). And so North Sindarin Arum = Sindarin Araw (the Vala Oromë); this means that final North Sindarin -um = Sindarin -aw from *-@m(æ). Now let us have a look in Etymologies. We see that not a few Ilkorin words with um = Noldorin words with aw; e.g., Ilkorin daum = Noldorin daw (LR:354). We know also from the above, that intervocalic m stays m in North Sindarin, just as in Ilkorin; PQ. *tinmê > Noldorin tinw but > Ilkorin tim (LR:393); or as Tolkien could have later put it: CE *tinmê > Common Telerin *tinmæ > [?Old Sindarin] *tinm(a) > North Sindarin tim(m), *tinmh > Sindarin tinw. I don't think that the use of tim, instead of tinw, as a Sindarin word in MR:388 contradicts the above: "Sindarin tim, gil referred properly to the Valinorian imagines [on the Dome of Varda]." For tim, being a Sindarin word referring to the Valinorian images, could only be a late (semantic) loan, since the Sindar did not have information about the Dome (except maybe from Melian). The Exiled Noldor learned first North Sindarin, since they came first to that region (Mithrim), and so it was they who introduced in the late First Age Sindarin tim from North Sindarin, with the sense 'star from the Dome of Varda'.
              We know also from WJ:414 that n@v, n@f is North Sindarin (< nâbâ), but is in Sindarin (Doriathrin and Falassian) nauv > naw. This means that CE *â > @ in North Sindarin (written ô or ó), but CE *â > au in Sindarin. It is exactly what we have in Etymologies: e.g., *târ > Noldorin taur versus Ilkorin tôr (LR:389).
              It looks to me that many Ilkorin words in Etymologies can now be considered, in fact, to be North Sindarin words.

    Editorial P.S. by H. K. Fauskanger: For words that it would be difficult to adopt as "dialectal Sindarin", see alch, côm, cwess, salch (because kw fails to become p) and go (because primitive 3, H becomes g instead of disappearing). But then Kloczko does not argue that Ilkorin as such was turned into North Sindarin, but only that Tolkien in North Sindarin may have recycled some words and ideas from Ilkorin.

    Ardalambion Index