Q. elision grammar.

Q. elision grammar.

In Quenya a short final vowel is sometimes lost before another word beginning with a vowel, a phenomenon known as elision. Examples appear for many combinations of vowels:

Where these losses occur, they are for vowels that carry little-to-no semantic weight. Examples include:

As a further example, at one point Tolkien seems to have explained the conjunction ar “and” as the result of ancient aða or asa, which became ara reduced by elision to ar’ before vowels and then generalized to the form ar before consonants: “Quenya ‘and’ ar is for as(a), in sandhi [alternate form based on position] before vowel” (PE17/41). There are, however, examples similar to the above where elision does not occur:

This inconsistency makes it hard to determine a pattern, but I think elision is likely to occur in the following circumstances (from most to least likely):

I also think elision is most likely when (a) the two vowels are identical, (b) it involves two short words that can be pronounced as a unit, or (c) for longer words when the result would not cause the primary stress of both words to become adjacent. I think the elision of pronouns in Quenya prayers of the 1950s were a transient idea, and I wouldn’t use it in Quenya writing generally (there are counter-examples within the prayers themselves, and no examples outside the prayers). I also think elision probably occurs more freely in poetry, especially when it aids the poetic meter. I think it is likely that elision is never fully required, and that you always have the option of pronouncing all vowels when speaking slowly and distinctly.

Elision and Stress: We don’t know how elision affects stress, but I am of the opinion that the stress patterns of the word do not change, so that lúmenn’ is pronounced lúMENN’ and not LÚmenn’.

Elision with i “the”: It seems that the definite article i is generally not elided before vowels, even when the following vowel is identical: sustane Manweo súle ten i indo Sindicollo “*the spirit of Manwe blew unto the heart of Thingol” (NM/239). However, i might be elided if the preceding vowel is i or e. We have two examples of this:

In the first example, the preposition mi “in” simplifies to with long í when preceding a definite article (mi i); Tolkien glossed as “in the” for the prose version of Namárie making it clear it included the definite article. Since mi is the only monosyllabic preposition ending in i, it is likely unique in this respect; any polysyllabic prepositions like imbi would likely undergo elision instead.

The second example is especially interesting, since ’n is a particularly abnormal form for the definite article. There is at least one other example of a variant definite article in being used before a vowel: mana i·coimas in·Eldaron “what is the ‘coimas’ of the Eldar”. Thus it seems in cases where the article i is “squeezed” between two vowels it might further reduce to ’n: utúlie i aurëutúlie in aurëutúlie’n aurë. Hat tip to Raccoon on the notion that this ’n is a definite article; I first got the idea from his discussion of the definite article, though my guesses on the mechanics of its appearance are not exactly the same as his.

Elision with ar “and”: We have one example of elision of the infinitive of an a-stem verb before the conjunction ar “and”: cuita’r pare “live and learn” (PE22/154). It is not clear why this is cuita’r and not cuit’ ar (analogous to á þak’ i fende above), since the pronunciation would be the same. Perhaps in this case the final vowel of the verb is considered more semantically significant than the vowel of ar. In any case, it seems that for pronunciation the two a’s would merge in this circumstance as well.

This kind of elision of the initial letter of ar is not a universal phenomenon, however: sínen i·anda nyarne metta ar taina andaurenya na quanta “*with this the long tale ends and my extended long day is complete” (Minor-Doc/1955-CT).

Conceptual Development: There are examples of elision in Tolkien’s earlier writing, but not enough information to determine if there were any conceptual shifts on how and when it occurred. Elision does seem to be more frequent in the 1950s and 60s than it was earlier, however.

One significant exception to this are the subject pronoun prefixes from the Early Qenya Grammar of the 1920s, which underwent modification or elision when appearing before verbs beginning with vowels (PE14/86). There is no evidence of such elision in Tolkien’s brief restoration of subject prefixes in the late 1940s (for example in the Quenya Verbal System, PE22/99-127), but that could just be lack of examples.

Examples (elided)
aistan’ “*blessed” [← aistana] {aistan’ >>} aistana elye {mitta >>} mika nísi ✧ VT43/28
’r “and” [← ar] kuita’r pare ✧ PE22/154
ep’ [← epë] ancalima ep’ eleni, arcalima ep’ eleni ✧ PE17/56
ep’ [← epë] A anamelda na ep’ ilya ✧ PE17/57
’n “the” [← ] Utúlie’n aurë! ✧ S/190
’n “the” [← ] Aiya Eldalië ar Atanatári, utúlie’n aurë ✧ S/190
’n “the” [← ] Utulie’n aurë! ✧ WJ/166
’n “the” [← ] Aiya Eldalië ar Atanatári, utulie’n aurë ✧ WJ/166
imb’ “among” ← imbi (plural) ankalima imb’ illi ✧ PE17/91
lann’ “(to a point) beyond” ← lanna lendes lann’ i sír ✧ PE17/65
on’ [← ono] {ono >>} on’ et·á·rúna me va·úro ✧ VT43/23
m’ [← me] mal ám’ etelehta ulcullo: násië ✧ VT43/23
métim’ “last” [← métima] métim’ auressë ✧ MC/222
métim’ “last” [← métima] Man kenuva métim’ andúne?, Man kenuva métim’ andúnie? ✧ MC/222
#Nand’ “valley” ← #nan(do) Nand’ Ondolunkava ✧ PE17/28
pell’ [← pella] Átaremma meneldëa [>> i ëa pell’ Ëa] ✧ VT43/13
sív’ “*as” [← sívë¹] sív’ emme apsenet tien i úcarer emmen ✧ VT43/20
tenn’ “unto” [← tenna¹] Sinome maruvan ar Hildinyar tenn’ Ambar-metta ✧ LotR/967
ten [← tenna¹] Sustane {i sul >>} Manwëo súle ten i indo {Sindicolluo >>} Sindicollo ar he lastane ar carnes ✧ NM/239
tenn’ [← tenna¹] Sínome maruvan tenn’ ambarmetta ... arhildinyar ✧ PE22/147
tenn’ [← tenna¹] et // {sinon...} sillumello // {tenna} yéni {apo yéni} pa yéni {pa} // ter yénion yéni tenn’ ambarmetta ✧ VT44/36
Véne’ “*virginal” [← #vénëa] á Véne’ alcare ar manquenta ✧ VT44/10
aly’ [← á] 2nd-sg-polite ono· va úro aly’ eterúna me ✧ VT43/23
ruxal’ “crumbling” [← #ruxa-] active-participle ruxal’ ambonnar ✧ MC/222
lúmenn’ “on the hour” [← lúmë¹] allative elen síla lúmenn’ omentielmo ✧ Let/265
lúmenn’ “on the hour” [← lúmë¹] allative Elen síla lúmenn’ omentielvo ✧ LotR/81
þak’ “close” [← sac-] aorist á þak’ i fende, mekin ✧ PE22/166
turun “mastered” [← turu-] passive-participle A Túrin Turambar turun ambartanen ✧ S/223
turún’ “mastered” [← turu-] passive-participle A Túrin Turambar turún’ ambartanen ✧ UT/138

Element In


ᴹQ. elision grammar.

Examples (elided)
kenn’ “until” [← tenna¹] Símane maruvan, ar hildinyar kenn’ Iluve-metta ✧ SD/56
tenn’ “until” [← tenna¹] Sinome nimaruva, yo hildinyar tenn’ Ambar-metta ✧ SD/56
ankalim’ [← #kalima] intensive Kuivië, kuivië! ankalim’esselínen ✧ SD/51

Element In


ᴱQ. elision grammar.

Examples (elided)
air’ [← aira] i·air’ anūre ✧ QL/31
kal’ “light” [← kala¹] I·kal’antúlien ✧ LT1/184
en’ [← en-] ✧ QL/35
em’ [← en-] ✧ QL/35
ailim’ [← ailin²] ✧ PE16/66
’n “the” [← ] i mitta ’n·felda aksínen, i·ner i·táralda ’n·Noldolion ✧ PE14/48
’n “the” [← ] i mitta ’n·felda aksinen, i·ner i·taralda ’n·Noldolion ✧ PE14/81
laiq’ “green” [← laiqa] {ondoise laiqa kainer >>} kaire laiq’ ondoisen kirya ✧ PE16/62
laikv’ [← laiqa] käire laikv’ öndöisen kirja ✧ PE16/72
laiq’ [← laiqa] kaire laiq’ ondoisen kirya ✧ PE16/74
lir’ [← liri-] Ómalingwe lir’ amaldar ✧ VT40/8
oilim’ [← oilima] oilim’ ambar {ie >>} ien {oilima >>} oilin ✧ PE16/62
qant’ “all” [← qanta] qant’ i lie telerinwa ✧ PE16/90
V’ [← ve] V’ematte sinqi Eldamar ✧ VT40/8

Element In