This is a long (almost 90 words) Quenya song found in LR:72 (also some fragments in LR:63). It was written about 1940. Tolkien provided no official title, but in the narrative for which he wrote the song, it is sung by a woman called Fíriel. Hence it is universally referred to as Fíriel's Song. The language of this song is what I would call "near-LotR-style" Quenya, or late "Qenya". It is not quite the same kind of Quenya as the language we know from LotR and later sources, but Tolkien was getting there. He had already come a long road since "Qenya" in its most primitive form first manifested in the Qenya Lexicon a quarter of a century earlier, in 1915.
Fíriel's Song, with Tolkien's translation interspersed:
1. Ilu Ilúvatar en káre eldain
a fírimoin
The Father made the World for Elves and Mortals
2. ar antaróta mannar Valion: númessier.
and he gave it into the hands of the Lords. They are in the West.
3. Toi aina, mána, meldielto - enga morion:
They are holy, blessed, and beloved: save the dark one.
4. talantie. Melko Mardello lende: márie.
He is fallen. Melko [Melkor] has gone from Earth: it is good.
5. En kárielto eldain Isil, hildin Úr-anar.
For Elves they made the Moon, but for Men the red Sun;
6. Toi írimar. Ilyain antalto annar lestanen
which are beautiful. To all they gave in measure the gifts
7. Ilúvatáren. Ilu vanya, fanya, eari,
of Ilúvatar. The World is fair, the sky, the seas,
8. i-mar, ar ilqa ímen. Írima ye Númenor.
the earth, and all that is in them. Lovely is Númenor.
9. Nan úye sére indo-ninya símen, ullume;
But my hearth resteth not here for ever,
10. ten sí ye tyelma, yéva tyel ar i narqelion,
for here is ending, and there will be an end and the Fading,
11. íre ilqa yéva nótina, hostainiéva, yallume:
when all is counted, and all numbered at last,
12. ananta úva táre fárea, ufárea!
but yet it will not be enough, not enough.
13. Man táre antáva nin Ilúvatar, Ilúvatar
What will the Father, O Father, give me
14. enyáre tar i tyel, íre Anarinya qeluva?
in that day beyond the end when my Sun faileth?
Since this is not LotR-style Quenya anyway, I have not regularized the spelling to the system Tolkien later used in LotR. Hence we have k rather than c, q rather than qu, and no diaereses over final e's etc. (táre, not tárë).
The text above incorporates some changes Tolkien made (mentioned, but not directly incorporated in LR:72). The variant readings are discussed in the analysis below. Only one of the changes Tolkien made is here ignored: for Melko (Melkor) in line 4, he decided to substitute Alkar. But this name for the diabolus of his mythos was eventually rejected, while Melko is still a valid alternative to Melkor in Tolkien’s later Quenya (see below).
One graphic feature cannot be reproduced here: In LR:72 (but not in LR:63), there is a dot under the final vowel of the word káre in line 1 and of the word íre in lines 11 and 14. We don't know precisely what this is supposed to mean. The best theory is perhaps that the dot indicates that these final vowels are to be elided and not pronounced (the next word begins in a vowel in all three cases, and in such an environment, elision could easily occur).
1. Ilu Ilúvatar en káre eldain a fírimoin "The Father made the World for Elves and Mortals": Ilu "the World" (the Quenya word seems to count as a proper name, so the article i "the" is not required). Ilu, ilu is defined both as "the world" (LR:47, 56, 72), "universe" (LR:361 s.v. IL), and as "everything, all, the whole" (of the universe also including God and all souls and spirits, which are not properly included in the term Eä; see VT39:20). Otherwise, the normal Quenya translation of "world" seems to be ambar (in LotR we find the word Ambar-metta, the End of the World). Ilúvatar is here translated "the Father", but this is of course the normal Quenya term for "God" (the name Eru, The One, being reserved for the most solemn occasions). Ilúvatar means "All-father", a compound of ilúvë "allness, the all" (related to ilu) and atar "father". en is a word that defies certain interpretation. It occurs twice in Fíriel's Song, and nowhere else. It does not seem to correspond to anything in Tolkien's English translation. One opinion is that en means "it", so that Ilu Ilúvatar en káre is literally "the world (-) Ilúvatar it made". But this seems very doubtful. Naturally, there have been attempts to connect this word en with the base EN in the Etymologies (LR:356), defined as "over there, yonder". Etym also states (LR:399 s.v. YA) that "en yonder (...) of time points to the future". Anthony Appleyard takes en to mean "then", and the suggestion has been put forward that its use in the phrase "the Father made the World for Elves and Mortals" indicates that the world was not to be "used" by the Elves and the Mortals immediately, but that some time (actually much time) passed between creation and the appearance of these groups. Thus, the element en would "point to the future". káre "made", past tense of kar- "make". This kind of past tense (formed by lengthening the stem-vowel and adding the ending -e) is often found in the earliest "Qenya Lexicon" (1915), where the form káre is actually listed (p. 45). However, it would seem that Tolkien decided to limit the use of this past tense formation. Láve as the past tense of lav- "lick" made it into Namárië in LotR, but the past tense of kar- "do, make" should rather be karne (carnë) in Tolkien’s later Quenya, and this past tense is actually mentioned in the Etymologies (LR:362 s.v. KAR). eldain "for Elves", dative plural of elda "elf". It seems that elda is here used in the general sense "Elf", though this term properly excludes the Avari. a "and"; the text otherwise uses ar (lines 2, 8, 10), and ar is attested in LotR-style Quenya. It has been suggested that a is preferred when the next word begins in f (though it is far from certain that this idea is valid in later Quenya). The first a in ananta "and yet, but yet" in line 12 may also be a prefixed conjunction "and"; see below. fírimoin "for Mortals", dative pl. of fírimo "mortal" (noun), in turn a nominalized form of the adjective fírima "mortal" (LR:381 s.v. PHIR). Tolkien eventually replaced the explicitly nominalized form fírimo with Fírima (nominal pl. Fírimar, "those apt to die", WJ.387); in other words, he simply used and inflected the adjective as a noun.
2. ar antaróta mannar Valion: númessier. "and he gave it into the hands of the Lords. They are in the West": ar "and". antaróta "he gave it". The elements are clearly anta, the stem of the verb "give" (mentioned in LR:341 s.v. ANA1), -ro pronominal suffix "he" (accented -ró-), plus an ending -ta "it" (cf. LR:389 s.v. TA). This ending may have been replaced by -s in LotR-style Quenya; cf. a word like utúvienyes "I have found it" (utúvië-nye-s "have found-I-it") in LotR, not *utúvienyéta. Strangely, there does not seem to be any past tense marker in antaróta, though it is translated "he gave it". Perhaps this is actually an example of a "historic present", antaróta really meaning "he gives it" (the present tense marker -a would be invisible when suffixed to a stem already ending in -a, such as anta). mannar "into (the) hands". This is má "hand" with the plural allative ending -nnar (sg. -nna) "to, into" (plural to denote several hands). Notice that the long á in má becomes short a before the consonant cluster nn; it seems that Quenya phonology normally does not permit a long vowel before a cluster (it may be that ry in the word máryat "her hands" in Namárië somehow counts as a single consonant, palatalized r, allowing á to remain long before it). Valion "of the Lords" (Vali, Valar). In Tolkien's vision of Quenya, the form Vali always was an alternative to Valar as the plural form of Vala "angelic power, god" (LR:350 s.v. BAL, QL:99). Here, the plural genitive ending -on has been added to express "of the Lords" (surely Valaron would also be possible). The Quenya text has no article i "the" before Valion, though an article is supplied in Tolkien's English translation of this word ("of the Lords"). It seems that Quenya does not use the article before plural words referring to entire peoples or "species"; cf. a much later example, the sentence Valar valuvar "the will of the Valar will be done" (WJ:404). It will be noted that there is not really an article before the word "Valar" in the Quenya sentence, though there is one in Tolkien's translation. númessier "they are in the West". This word provides the first example of a remarkable grammatical devise that is employed five times in Fíriel's Song: the stative verb ending. It is indeed the most characteristic feature of this text; there is no trace of this ending in any other available document, except only the question Man-ie? "What is it?" from the narrative Fíriel's Song was originally a part of (LR:59). As for númessier, the underlying word is clearly númesse, "in the West", the locative of Númen "West" (the final -n of Númen being displaced by the locative ending -sse "in"). But to this word, an ending -ie "is" (pl. -ier "(they) are") is added, producing the form númessier "(they) are in the West" This ending -ie clearly corresponds to the independent word ye "is", also found later in Fíriel's song (írima ye Númenor = "lovely is Númenor", sí ye tyelma "here is ending"; see below). A good example of a stative verb is provided by the word márie "it is good" in line 4, derived from mára "good" (LR:371 s.v. MA3, MAG; notice that the stative verb ending -ie seems to displace the final vowel of a word it is added to). Is the stative verb ending valid in LotR-style Quenya? It has been used by at least one writer, Ivan Derzhanski, in his poem Lá ilqua i maltie kalta ("Not all that is gold shines" - notice maltie "is gold" from malta "gold"), published in Vinyar Tengwar #38. However, I would not use this ending myself. As Anthony Appleyard points out, Tolkien "likely rejected `-ie' as 'is' because `-ie' has too many other uses, risking ambiguities". Notice that in Namárië in LotR, the phrase "lost is" is expressed as vanwa ná, not **vanwie.
Instead of Valion: númessier "of the Lords. They are in the West", Tolkien originally wrote Valion númenyaron, translated "of the Lords of the West". The literal meaning of the word númenyaron would seem to be, not "of the West", but "of (the) western ones" - referring to people or lands. The Etymologies lists an adjective númenya "western" (LR:376 s.v. NDÛ). Here it is inflected as a noun, with the plural ending -r and the plural genitive ending -on. A late source confirms that the genitive case can be used to describe the relationship between rulers and the ruled (people or land): Elwë, Aran Sindaron/Lestanórëo, "Elwe, King of the Sindar/of Doriath)" (WJ:369).
3. Toi aina, mána, meldielto - enga morion: "They are holy, blessed, and beloved: save the dark one": Toi "they", a pronoun found in this song only (also in line 6). In Tolkien’s later Quenya, the word for ”they, them” (of persons) is té or te; in LotR we also find te for object "them", and it may also cover subject "they". aina "holy". mána "blessed". meldielto *"they are beloved". This is yet another stative verb, derived from melda "beloved, dear" (LR:372 s.v. MEL). The ending -ie for "is, are" has been added, regularly displacing the final vowel of melda. The whole phrase toi aina, mána, meldielto is perhaps a sort of abbreviation for *toi ainielto, mánielto, meldielto, with all the adjectives turned into stative verbs with full pronominal inflection: The adjectives aina and mána get a free ride with meldielto, so to speak; it is understood that the stative verb ending -ie and the pronominal element -lto apply to the whole series of adjectives. This is probably also the explanation why the adjectives aina and mána are not inflected: Here they ought to be plural, and in this "Qenya" variant, plural adjectives take the ending -r: Compare toi írimar, *"they [are] beautiful", in line 6 (írima = "beautiful, lovely", see line 8). When we don't have *toi ainar, mánar here, it is certainly because the stative verb meldielto is anticipated. The whole series may be taken as a kind of loose compound (*toi aina-mána-meldielto, "they holy-blessed-beloved-are"). The ending -lto used here, as a (semantically superfluous) extra "they" at the end of the sentence, is also found in LT1:114: Tulielto, "they have come". This word dates from the very early period (before 1920), and it is interesting to see that the ending -lto "they" was still valid about twenty years later, when Fíriel's Song was written. However, it does not seem that this ending made it into Tolkien’s later Quenya: A post-LotR text, Círion's Oath, points to -nte as the pronominal ending "they" in the later incarnations of the language (UT:305, 317), though the ending -lte (closer to -lto) also occurs in Tolkien’s late material. enga "save" (= except). This word is attested here only. One late essay mentions hequa as a Quenya word for "leaving aside, not counting, excluding, except" (WJ:365), but of course, enga may still be valid. morion "the dark one" (Melko, Melkor). This word is also found in the earliest "Qenya Lexicon" (1915), where it was defined as "son of the dark" (p. 62), but it is not there clear what it refers to. The ending -ion could mean "son" also in Fíriel's Song (as well as in later Quenya), but since the word is translated "the dark one", it is possible that Tolkien had another etymology in mind here: *mori-on(d)-, sc. morë (mori-) "dark" (cf. LR:373 s.v. MOR) combined with the masculine ending -on(d)-.
4. talantie. Melko Mardello lende: márie. "He is fallen. Melko [Melkor] has gone from Earth: it is good". (In the first version of the song, the word order was Mardello Melko instead of Melko Mardello, but this does not affect the sense.) talantie "he is fallen". The translation suggests that this is yet another stative verb, derived with the now-familiar ending -ie "is" from an adjective *talanta "fallen". No pronominal element actually meaning "he" seems to be present. However, since -ie is also the perfect tense ending, we should consider the possibility that talantie is no stative verb at all, but rather the perfect tense of the stem talat- used for "slipping, sliding, falling down" (Letters:347). The past tense of such a stem may be *talante with nasal infixion, and the perfect possibly talantie. Nasal infixion does not seem to occur regularly in the perfect tense (cf. for instance irícië, not **irincië, as the perfect tense of ric- "twist", VT39:9). However, we do have a few examples of nasal infixion in the perfect tense, such as oantie as the perfect tense of auta- "go away, leave" (WJ:365). It is possible, then, that talantie is intended as a perfect tense, literally meaning *"he has fallen" rather than "he is fallen". Melko is the name of the diabolus in Tolkien's mythos, normally called Melkor in later texts, but MR:350 confirms that Melko is still a valid form in Tolkien’s late Quenya. However, Tolkien's interpretation of the name differed over the years. In the earliest source, the Qenya Lexicon of 1915, Melko was simply defined as "God of Evil" (p. 60), with no etymological considerations. In the Etymologies of the mid-thirties, the name is derived from a stem MIL-IK having to do with lust and greed (LR:373). But in his later years, Tolkien stated that Melko simply means "Mighty One" (MR:350). The longer name Melkor means "Mighty-rising", "He that Arises in Power". Mardello *"from earth", the ablative ending -llo "from" being added to a noun mar (stem mard-) "earth", a connecting vowel e appearing between the stem and the ending to avoid an impossible consonant cluster (cf. the e before -nna "to" i Elendilenna "to Elendil", PM:401). The noun mar "earth" also occurs in line 8. lende "went". This irregular past tense of a word for "go" is attested in various places, though there are some discrepancies concerning the present tense of this verb. In the Etymologies, the form lende is first given as the past tense of both lesta- (LR:356 s.v. ELED), then as the past tense of linna- (LR:368 s.v. LED), both of these verbs meaning "go". The late essay Quendi and Eldar introduces yet another verb "go", lelya-, though the past tense is still lende (WJ:363). - The whole phrase Melko Mardello lende is literally "Melko from earth went", though Tolkien's translation reads "Melko has gone from earth"; the Qenya text uses the past, not the present tense. márie "it is good", actually simply "is good", a stative verb derived from mára "good" (LR:371 s.v. MA3, MAG) with the ending -ie "is".
5. En kárielto eldain Isil, hildin Úr-anar. "For Elves they made the Moon, but for Men the red Sun": The mysterious word en reappears from line 1, and in a similar context. kárielto "they made". The pronominal ending -lto "they" reappears; cf. meldielto in line 3 (and tulielto in LT1:114). Removing the pronominal ending, we are left with kárie as the verb "made". In line 1, the past tense of the verb "made" is káre instead. The stem is of course KAR (LR:362), so kárie is formed by lengthening the stem-vowel and adding -ie, but this looks rather like a perfect formation (in Tolkien’s later Quenya the stem-vowel would normally be prefixed as an augment: akárie, acárië "has made"). Does kárielto really mean "they have made" rather than simply "they made"? In a late source, Tolkien states that "the forms of past and perfect became progressively more closely associated in Quenya" (WJ:366), so perhaps it is sometimes permissible or even preferable to use an English past tense form to render a Quenya perfect? eldain "for Elves", dative plural of elda, as in line 1. Isil "the Moon" (the Quenya word is a proper name and does not require the article). At this point, Tolkien's translation has a "but", but there is nothing that corresponds to this word in the Elvish text. (Cf. the "missing" conjunction and in line 10.) hildin "for Men", a dative plural. Hildi, "Followers", was an Elvish name of Mortal Men as the Second-born of Ilúvatar, the Elves being the Firstborn. Later, Tolkien used the form Hildor instead (sg. *Hildo; see LR:248 and WJ:387), and Hildor is used in the published Silmarillion. The dative plural corresponding to nominative Hildor would have been *Hildoin, and if Tolkien had ever translated Fíriel's Song into LotR-style Quenya, he would probably have replaced hildin with this form. Úr-anar "the red Sun" (a kind of proper name, hence no article in Quenya). Anar is the Quenya word for "sun" (cf. Anarinya "my sun" in the last line); the prefixed element úr means "fire" (see LR:396 s.v. UR, in the original version of this entry), so Úr-anar is literally "[the] Fire-sun". (Cf. another Quenya name of the sun, Úrin.)
6. Toi írimar. Ilyain antar annar lestanen "which are beautiful. To all they gave in measure the gifts": The relative sentence "which are beautiful" clearly isn't a literal translation of the Elvish text; toi írimar simply means *"they [are] beautiful": Toi "they", as in line 3. írimar "beautiful", plural to agree with "they"; the sg. form írima is found both in line 8 (there translated "lovely") and in the Etymologies (LR:361 s.v. ID, where the gloss is "lovely, desirable"; the latter would seem to be the etymological meaning). In this "Qenya" variant, the plural form of adjectives is formed with the ending -r (that is also used to form the plural of nouns and verbs). Another example, written about the same time as Fíriel's Song, is the sentence ilya...maller raikar "all...roads [are] bent" in LR:47 (sg. raika "crooked, bent, wrong" occurs in LR:383 s.v. RÁYAK). In many versions of Quenya, adjectives in -a form their plurals in -ë (*írimë, *raikë/raicë) instead of by the ending -r. Toi írimar is a nominal sentence, *"they beautiful"; there is not actually any word for "is" connecting the adjective with the pronoun (and neither is a stative verb used here; *toi írimier or even *írimielto would presumably have been possible constructions). An actual word for "is", ye, occurs in line 8. ilyain "to all". This was an emendation; Tolkien first wrote ilqainen, a form that in some respects makes little sense. Ilqa would seem to be the word for "all", and line 8 of Fíriel's Song confirms this, since ilqa there appears with no ending. In later Quenya, we find ilya rather than ilqa as the word for "all" (and Tolkien even replaced a form of ilqa with a form of ilya here); the Etymologies lists both ilya and ilqa, there glossed "all, the whole" and "everything", respectively (LR:361 s.v. IL). The ending -inen is very surprising. Since Tolkien's translation reads "to all", we must interpret -inen as a kind of dative (or conceivably allative) ending, but in LotR-style Quenya, -inen is the plural instrumental ending. The corresponding singular ending -nen actually occurs in Fíriel's Song, in the noun lestanen "in (by) measure" later in this line. Other words occurring in this song (eldain, hildin, fírimoin) demonstrate that the dative plural in -in had already come into place in Tolkien's vision of Quenya, so it is not surprising that he changed ilqainen to ilyain. One has to wonder whether he originally confused the dative and the instrumental, writing ilqainen where he meant ilqain. antalto "they gave". Cf. antaróta "he gave it" in line 2. Originally Tolkien wrote simply antar here; this would be anta- (the stem of the verb "to give", LR:341 s.v. ANA1) with the plural ending -r, here translated "they"; the simplest plural ending was used instead of the longer pronominal ending -lto, as in kárielto, meldielto in line 3 and 5. However, Tolkien changed his mind and brought in the longer ending for "they" after all, emending antar to antalto. Just as in the case of antaróta "he gave it" in line 2, it is puzzling that there seems to be no real past tense marker in the word antar > antalto, though Tolkien's translation once again employs a past tense form: "they gave". In the context of Tolkien’s later Quenya, I would definitely take antar to be a present-tense form (antalto would probably be antaltë or antantë in later versions of Quenya, but this would still be a present tense). May this be another example of "historic present", the literal meaning being "they give" rather than "they gave"? annar "the gifts" (there isn't really any article in the Quenya text, but this noun is determined by the genitive Ilúvatáren in the next line: "Ilúvatar's gifts" = "the gifts of Ilúvatar"). Annar is the pl. form of anna "gift" (LR:348 s.v. ANA2). lestanen "in measure". Here we see the instrumental ending -nen, still valid in LotR-style Quenya. The form lestanen thus means "in/by measure". The noun *lesta "measure" is not otherwise attested, unless it is to be equated with the first element of the Quenya name of Doriath, Lestanórë (WJ:369). This means "Girdle-land", Quenya lesta "girdle" corresponding to Sindarin lest (Lest Melian "the Girdle of Melian", WJ:225). A semantic connection "girdle/border/border-line/clearly defined boundary/measure" may be barely plausible.
7. Ilúvatáren. Ilu vanya, fanya, eari, "of Ilúvatar. The World is fair, the sky, the seas,": The "Qenya" genitive Ilúvatáren logically goes with the previous line, completing its meaning: annar...Ilúvatáren "the gifts...of Ilúvatar". (Here the name "Ilúvatar" is also used in Tolkien's English translation of Fíriel's Song; in lines 1 and 13 it is translated "the Father" or "o Father".) In this "Qenya" variant, the genitive ending is still -n (here with a connecting vowel e since **Ilúvatarn is impossible). In LotR-style Quenya, -n came to be the dative ending, while the genitive has the ending -o instead (Tolkien did this change while he was writing LotR: in one early variant of Namárië, "Varda's" was Vardan, changed to Vardo). The later genitive of Ilúvatar (in -o) is actually attested; the Silmarillion Index, entry "Children of Ilúvatar", mentions Híni Ilúvataro as the Elvish equivalent of this phrase. We might have expected *Ilúvatáro with a long á in the last syllable of atar, since this noun seems to lengthen its final vowel when an ending is added (atár-). Cf. the old genitive (= later dative?) form Ilúvatáren, and also the word Atanatári "Fathers of Men" (*"Manfathers") in the Silmarillion, chapter 20 (also in PM:324). However, the plural of atar is given as atari (not atári) in the Etymologies (LR:349 s.v. ATA), so perhaps Tolkien was simply changing his mind back and forth. Whatever the case, the lengthening of the final vowel of atar does occur in the word Ilúvatáren in Fíriel's Song. Ilu "the World", as in line 1. vanya "fair" (cf. the nominal pl. Vanyar as the name of the First Clan of the Elves). Ilu vanya is another nominal sentence, lacking any actual word for "is": *"The world fair." fanya "the sky". The translation is somewhat unusual; otherwise, fanya is glossed "cloud". In the Etymologies (LR:387 s.v. SPAN), the word fanya is defined as "cloud", derived from a stem having to do with whiteness. The pl. form fanyar in Namárië is also translated "clouds". eari "the seas", pl. of ear "sea". Neither fanya nor eari is preceded by any actual definite article, despite the translations: "the sky", "the seas". Contrast i-mar in the next line.
8. i-mar, ar ilqa ímen. Írima ye Númenor. "the earth, and all that is in them. Lovely is Númenor": i-mar "the earth". Here the definite article i "the" really appears. Infrequently, Tolkien connects it to the following word by means of a dot or, as here, a hyphen. (However, Namárië in LotR has i eleni, not i-eleni, for "the stars".) Mar is the shortest form of the noun "earth"; the stem is mard-, seen in the ablative Mardello in line 4. In later Quenya, the normal word for "earth" seems to be cemen, kemen; cf. Yavanna's title Kementári "Earth-queen". ar "and". ilqa "all" (cf. ilqainen "to all"; see comment on line 6 above). ímen "that is in them", literally perhaps simply "in them", or even *"of them". This is a most peculiar form; it cannot be related to any other 3. person plural form in the published corpus. Fíriel's Song otherwise uses the endings -lto, -r or the independent pronoun toi for "they"; later sources have te and the ending -nte. My best guess is that Tolkien intended ímen to be 1) a demonstrative stem í- "that" (related to the article i "the") combined with 2) the primitive plural element -m (turning "that" into "those") and 3) the ending -en, conceivably the same as the genitive ending seen in Ilúvatáren "of Ilúvatar" in line 7. Hence ilqa ímen = *"all of those (ones)" = "all that is in them". But ímen certainly isn't a word I would recommend to people writing in LotR-style Quenya. Írima "lovely" (pl. írimar; see comment to line 6 above). ye "is". This is the sole occurrence of an independent word for "is" in Fíriel's Song, but it is obviously related to the stative verb ending -ie. Furthermore, the word yéva in lines 10 and 11 is obviously the future tense of ye. However, it does not seem that the word ye made it into Tolkien’s later Quenya. Instead, Tolkien reverted to his original choice for "is", the word ná. This word occurs already in the Qenya Lexicon of 1915 (p. 64), reappeared in the Etymologies of the mid-thirties (LR:374 mentions NÂ as the "stem of verb 'to be' in Q") and was finally fixed as the Quenya word for "is" by appearing in Namárië in LoTR (sí vanwa ná...Valimar, "now lost is...Valimar"). Númenor "Númenor" (Westernesse; númen = "west").
9. Nan úye sére indo-ninya símen, ullume; "But my heart resteth not here for ever,": nan "but". In the Etymologies (LR:375 s.v. NDAN), the Quenya word for "but" is nán with a long vowel. In LotR-style Quenya, nán could be interpreted "I am" (ná- + the pronominal ending -n "I"), so when writing Quenya texts, I actually prefer the form nan from Fíriel's Song to avoid any possible confusion. úye *"is not". This is the word ye "is" (as in írima ye Numenor "lovely is N." in the previous line) with the negation prefix ú- "no, not" (LR:359 s.v. GÛ). Evidently because of this example, many writers use this prefix to express negation in their Quenya texts (e.g. hónya ú-hiruva sére "my heart will not find rest" in a poem by Arandil Erenion). This is also what Nancy Martsch teaches in her Basic Quenya. Personally I usually prefer to use the independent word lá "not" (LR:367 s.v. LA), since the prefix is often somewhat cumbersome. sére *"rest", noun. In LR:385 s.v. SED, a word sére "rest, repose, peace" is listed. The literal meaning of the Elvish text seems to be, not really "my heart resteth not", but rather *"there is not [any] rest for my heart". indo-ninya "my heart". Tolkien originally wrote hondo-ninya, with another word for "heart". According to LR:364 s.v. KHO-N, the Quenya word hón (= hondo?) refers to the physical rather than the metaphorical heart, so this may be why Tolkien decided to go for indo instead. The word indo "heart, mood" is also listed in the Etymologies (LR:361 s.v. ID), but the suffixed element ninya is attested here in Fíriel's Song only. It would seem to mean "my", but in later Quenya, the pronominal ending "my" is -nya (tatanya "my father", UT:191, so "my heart" would probably become indonya; compare Anarinya "my sun" in the final line of Fíriel's Song itself). To make sense in this context, indo-ninya "my heart" would have to be the indirect object, and in such a highly inflected language we would expect a dative marker to indicate the meaning "(there is no rest) for my heart", but no dative element seems to be present. símen "here". This word is attested here only, but the elements are transparently sí-, a form of the stem SI "this, here, now" (LR:385), plus men "place, spot" (LR:372 s.v. MEN), hence literally "this place". (In LotR we find sinome for "in this place", but símen may still me valid.) In line 10, the shorter word sí is used for "here", but sí is translated "now" both in the Etymologies (LR:385 s.v. SI) and in Namárië in LotR. ullume "not...for ever". Compare another word having to do with time, yallume "at last" in line 11, literally *"in that time". The Etymologies lists a word lúme "time" (LR:370 s.v. LU), and the element -lume occurring here is surely related, while the prefix u- is certainly more or less identical to the negative prefix ú- (as in úye above). Perhaps ullume means something like "not [for all] time". The double ll in ullume may suggest that the prefixed element u- "not" originally ended in some consonant that was later assimilated to produce a double consonant. One negative element UMU is mentioned in LR:396; perhaps we are to assume that ullume represents something like *umlume?
10. ten sí ye tyelma, yéva tyel ar i narqelion, "for here is ending, and there will be an end and the Fading,": ten "for", a word that is attested here only; Namárië in LotR has an instead (an sí Tintalle...máryat ortanë "for now the Kindler...has uplifted her hands"). sí "here". As noted above, this word is used for "now" in other sources, and line 9 has símen for "here". Writers probably should not use sí for "here", since it would normally be understood as "now". ye "is", as in line 8 (írima ye Númenor). tyelma "ending", a word that is attested here only. However, it is obviously derived from the stem KYEL "come to an end" (LR:366); original KY became ty in Quenya. Tyelma could be regularly derived from *kyelmâ (*kjelmâ), but the ending -mâ is somewhat surprising, for it is normally used to derive nouns denoting concrete things, often implements (WJ:416). By its etymology, tyelma would most likely mean "thing used to end (something)", not "ending" as an abstract. This is far from conclusive, but one has to wonder whether tyelma is a misreading for tyelme, since -me is a well attested abstract ending. It would not be the first case of e and a being confused by editors trying to read Tolkien's handwriting. yéva "there will be", literally simply *"will be". (Before "there will be", Tolkien inserts a conjunction "and" in his translation, but there is no conjunction in the Elvish text - only a comma.) Yéva is the future tense of ye "is"; see the comment on antáva in line 13 for a possible reconstruction of the grammatical rules Tolkien held in mind when writing Fíriel's Song, and for my opinion on how they relate to the system used in LotR-style Quenya. The verb yéva can be suffixed as a stative verb ending, -iéva, seen in hostainiéva in line 11. tyel "an end" - literally simply "end"; Quenya has no indefinite article "a, an", and when translating Quenya to English you simply have to supply an indefinite article where English grammar demands one. (Compare the beginning of the Elvish greeting "a star shines upon the hour of our meeting": elen síla... "[a] star shines...") The noun tyel "end" is obviously related to tyelma "ending"; unlike tyelma it is listed in the Etymologies (LR:366 s.v. KYEL), there with an alternative, longer form tyelde. The word tyel also occurs in the last line of Fíriel's Song. ar "and", i "the", narqelion "Fading". In the Etymologies (LR:366 s.v. KWEL), the more literal gloss "fire-fading" is provided, the prefixed element nar- "fire" evidently referring to the warmth of the Sun (Quenya Anar). The Etymologies provides the additional gloss "autumn". In Fíriel's Song, this word normally used for "autumn" seems to be used with a wider reference: the "autumn" or "fading" of the world, the End drawing near. In LotR Appendix D, we find Narquelië as the name of the month October; this may be the LotR-style Quenya equivalent of narqelion. (An abstract word in -ion would be most unusual in LotR-style Quenya, while abstracts in -ië are common.)
11. íre ilqa yéva nótina, hostainiéva, yallume: "when all is counted, and all numbered at last,": íre "when" (not as a question word, but used to point to a specific time: "when all is counted" here, and "when my Sun faileth" in the last line). To writers, this is perhaps the most valuable vocabulary item Fíriel's Song provides; it is attested here only. Anthony Appleyard has pointed out that íre seems to contain the element -re, -rë "day" (final element in many similar compounds - cf. for instance mettarë, in LotR Appendix D said to be the name of the last day of the year, transparently meaning "end-day", since metta means "end"). It may be that íre ilqa yéva nótina is literally *"the day all will be counted". However, writers have used íre for "when" in the most general sense, and until we have any evidence either way, there is little reason to criticize this. ilqa "all", or "everything"; see comment on ilqainen (changed to ilyain) in line 6 above. yéva "is", or literally "will be", as in line 10: we are talking about future events. nótina "counted". This is the past participle of a verb not- "count". This verb is listed in the Etymologies (LR:378 s.v. NOT), there glossed "reckon", but the stem NOT itself if defined "count, reckon". Cf. also the related word únótimë "uncountable, numberless" in Namárië in LotR. The past participle is regularly formed with the ending -ina and lengthening of the stem-vowel: o becomes ó. (This lengthening does not occur when there is a consonant cluster following the stem-vowel, cf. *hostaina in the following discussion - not **hóstaina.) hostainiéva "[will be] numbered". This is the sole example of a future-tense stative verb. The underlying word is clearly a past participle *hostaina, formed (with the same ending -ina as in nótina) from a verb hosta-. This verb is listed in the Etymologies (LR 364 s.v. KHOTH), there glossed "to collect". *Hostaina would then mean "collected"; here the translation is "numbered" instead, but the semantic gap between these glosses isn't too wide. *Hostaina is then turned into a stative verb by suffixing -iéva, the future-tense variant of the stative verb ending, with the same relationship to the independent word yéva as the present-tense ending -ie has to the independent word ye. Like the ending -ie, -iéva displaces the final vowel of a word it is added to: hostainiéva, not **hostainaiéva. yallume "at last". This word is attested here only. As noted above, the final element must be related to lúme "time" (LR:370 s.v. LU). Indeed a very similar word, yalúme, is listed in the Etymologies - but this word means "former times" (LR:399 s.v. YA). Yallume must be assumed to be lúme "time" with a prefixed element that would mean something like "that", hence "at that time" = "at last". The Etymologies does provide a word yana "that" (LR:399 s.v. YA), but it is there said that with reference to time, this word refers to the past: "the former". Yallume could be derived from yana-lúme > yan-lúme, but according to the information provided in the Etymologies, this ought to mean "that time (in the past)". In Fíriel's Song, yallume clearly refers to the future. It would seem that Tolkien, at the time he wrote this song, did not assume that yana had past rather than future reference. I would not recommend the word yallume "at last" to writers, especially since safer alternatives are easy to construct (e.g. *mettassë, "in (the) end", locative of metta "end").
12. ananta úva táre fárea, ufárea! "but yet it will not be enough, not enough": ananta "but yet". The Etymologies gives a-nanta with a hyphen, glossed "and yet, but yet" (LR375 s.v. NDAN). It would seem that the prefixed element a means "and" (compare a in line 1), while nanta means "yet", though we don't know whether this word can be used independently. úva "it will not be", literally simply *"will not be". This seems to be the future tense of the same negative verb that in the Etymologies is listed in the first person aorist: uin, "I do not, am not" (the ending -i- denoting aorist and the pronominal suffix -n meaning "I", leaving u- as the stem; for the future formation úva, compare yéva from ye). táre *"(in) that day". There is nothing corresponding to this word in Tolkien's translation, but it seems to combine ta "that" (LR:389 s.v. TA) and are "day" (LR:349 s.v. AR1). Another word for the same is enyáre in line 14. fárea "enough". The word occurs in the Etymologies (LR:381 s.v. PHAR), but there the first vowel is short: farea "enough, sufficient". ufárea "not enough". This is simply fárea "enough" with the common negation prefix, though it usually appears with a long vowel: ú-. Perhaps Tolkien did not want to have two long vowels following one another, though this occurs in Namárië in LotR (únótima "uncountable, numberless" - hence I think we should read úfárea for "not enough").
13. Man táre antáva nin Ilúvatar, Ilúvatar "What will the Father, O Father, give me": man "what". This word is used for "what" here and in Elendil's question in LR:61 (immediately following Fíriel's Song in the narrative): E man antaváro? "What will he give indeed?" In later sources, the word man is used for "who": cf. the question sí man i yulma nin enquantuva? "who now shall refill the cup for me?" in Namárië in LotR. The word man = "who" also occurs many times in the Markirya poem. While we cannot be absolutely certain that man doesn't cover both "who" and "what", the word is only attested with the meaning "who" in LotR-style Quenya. PM:395, 402 seems to indicate that Tolkien later decided that the Quenya word for "what" is actually mana. Thus, man may now be unambiguous, meaning "who" only. táre *"in that day". Here, as in the previous line, this word is not actually translated by Tolkien. Another word for "in that day", enyáre, occurs in the next line, and that word is translated; perhaps Tolkien wanted to avoid repetition. antáva "will...give", the future tense of the verb anta- "give"; other forms occur in lines 2 and 6. When Tolkien wrote Fíriel's Song, he seems to have used the following grammatical rules when forming future tenses: If the stem of the verb ends in a vowel, lengthen it and add the ending -va: hence antáva from anta here, and yéva (-iéva) "will be" from ye (-ie) "is" in lines 10 and 11. If the stem of the verb ends in a consonant, add the ending -uva, as in qeluva from qel- in line 14 (see below for meaning). In his later forms of Quenya, Tolkien seems to have expanded the use of the longer ending -uva, and it is not certain that the shorter ending -va survived into later Quenya at all (perhaps because Tolkien didn't want it to be confused with the ending of the possessive case?) It LotR, we find laituva as the future tense of a verb laita-, while the rules apparently used in Fíriel's Song would have produced *laitáva instead. The new rules, as far as we can figure them out, seem to be simplified: 'The future tense is formed with the ending -uva. If the stem of the verb ends in a vowel, this vowel is dropped before -uva is added.' So in LotR-style Quenya, we should perhaps read *antuva for antáva, following the pattern of laituva. nin "me" (= dative, *"to me"). This is ni "I" with the dative ending -n. This pronoun also occurs in the question sí man i yulma nin enquantuva? "now who shall refill the cup for me?" in Namárië in LotR. Ilúvatar "the Father", literally "All-father" - see the comments on line 1. Ilúvatar "o Father". The divine epithet is repeated; Tolkien's translation suggests that the second occurrence is to be understood as a vocative: God is addressed directly. (The vocative particle "o" is in Quenya a, but it is not used here, though Tolkien's translation reads "o Father".)
14. enyáre tar i tyel, íre Anarinya qeluva? "in that day beyond the end when my Sun faileth?": enyáre "in that day". This word is attested here only. It seems to be a compound of *enya and are. The latter means "day" (see note on táre in line 12), while *enya is evidently an adjective formed from the base EN (not to be confused with the attested adjective enya < endya "middle" from the distinct stem ÉNED). As noted above, Etym states (LR:399 s.v. YA, cf. LR:356 s.v. EN) that "en yonder (...) of time points to the future". So *enya-are > enyáre is evidently "(on) that (future) day". tar "beyond". This word is found here only; Namárië in LotR has pella instead, placed after the noun as a postposition (Andúnë pella, "beyond the West"; cf. also elenillor pella "from beyond the stars" in the Markirya poem). Whether tar is still a valid word for "beyond" in LotR-style Quenya is therefore uncertain and perhaps doubtful. i "the", tyel "end", as in line 10. íre "when", as in line 11. Anarinya "my Sun". This is Anar "sun" (LR:348 s.v. ANÁR) with the ending -nya "my", that seems to prefer i as its connecting vowel when it is added to a noun-stem ending in a consonant (other pronominal endings should perhaps have e as the connecting vowel, though we lack good examples). qeluva "faileth", actually future tense *"will fail": This verb as such is not listed in the Etymologies, but it must surely be referred the stem KWEL- "fade, wither" (LR:366). This stem could yield a Quenya verb qel- (or quel- according to Tolkien's later spelling). Here it appears with the future-tense ending -uva. Since Tolkien elsewhere in Fíriel's Song uses the shorter ending -va, it may be that he thought of qeluva as the stem qel- with the ending -va + a connecting vowel u when he wrote this song. The longer ending -uva seems to have become universal in his later conception of Quenya (see the comments on antáva in line 13). In any case, the form qeluva (queluva) would certainly be valid in LotR-style Quenya.
Speaking of LotR-style Quenya, how would Fíriel's Song go in that language? Well, here is my rather tentative suggestion (I deliberately try to use words known from later sources, and in some cases bring the Quenya wording closer to Tolkien's translation):
Ilu Ilúvatar carnë Eldain ar
Fírimain
The Father made the World for Elves and Mortals
ar antanéses mánnar Valaron: ëaltë Númessë.
and he gave it into the hands of the Lords. They are in the West.
Naltë ainë, mánë ar meldë - hequa morion:
They are holy, blessed, and beloved: save the dark one.
alanties. Melkor Mardello lendë: nás mára.
He is [/has] fallen. Melkor has gone from Earth: it is good.
Carneltë Eldain Isil, Hildoin Úr-anar,
For Elves they made the Moon, but for Men the red Sun,
yar nar írimë. Ilyain antaneltë lestanen i annar
which are beautiful. To all they gave in measure the gifts
Ilúvataro. Ilu ná vanya, fanya, ëari,
of Ilúvatar. The World is fair, the sky, the seas,
i cemen, ar ilya ya ëa taissë. Írima ná Númenor.
the earth, and all that is in them. Beautiful is Númenor.
Nan lá ëa sére indonyan sinomë tennoio,
But my heart resteth not [lit. there is not rest for my heart] here for
ever,
an sinomë ëa tyelma, ar euva metta ar i narquelië,
for here is ending, and there will be an end and the Fading,
írë ilya nauva nótina, ar ilya hostaina, i mettassë:
when all is counted, and all numbered at last,
ananta úva tárë fárëa, úfárëa!
but yet it will not be enough, not enough.
Mana tárë antuva nin Ilúvatar, Ilúvatar
What will the Father, O Father,
enyárë i metta pella, írë Anarinya queluva?
give me in that day beyond the end when my Sun faileth?
(The word tárë *"in that day" is still left untranslated.)