To understand stress in Quenya you need to know the difference between “heavy” syllables and “light” syllables; Tolkien often called these “long” syllables and “short” syllables. A “light” syllable is one that contains a single short vowel and is followed by zero or one consonant: ta or tan. A “heavy” syllable is one that is not light, that either (a) contains a long vowel or diphthong or (b) is followed by two or more consonants: tán, tain, tand. Only true diphthongs (ai, oi, ui; iu, eu, au) make up a heavy syllable. Two vowels in hiatus make up a pair of syllables, not one heavy syllable: tie (or tië) has two light syllables: ti-e.
As a general rule, stress in Quenya words must fall on the heavy syllable closest to the end of the word, except (a) it cannot fall on the last syllable if there is more than one and (b) it cannot be further back than the third-to-last syllable. If there are no heavy syllables towards the end of the word, stress falls on the first syllable, or on the third-to-last syllable for longer words. Thus:
- In disyllables the stress is on the first syllable: LAS-se “leaf” (lassë).
- In trisyllables the stress is on the second syllable if it is heavy: ni-QUES-se “snowflake” (niquessë).
- Otherwise the stress is on the first syllable: TYA-li-e “play” (tyalië).
Longer words are stressed like trisyllables, except that syllables before the last three don’t factor in. Thus the stress is on the second-to-last syllable (if it is heavy) or the one before it (if the second-to-last is light): las-se-LAN-ta “autumn” (las-se-LAN-ta) vs. a-LAI-ti-en “I have praised” (alaitien). As this last example shows, inflected forms typically follow the same stress rules as uninflected forms, but see the entry on prosodic lengthening for some extra complications for inflected forms.
The stress rules described here are identical to those appearing in The Lord of the Rings appendix E (LotR/1116), but more detailed descriptions appear elsewhere in Tolkien’s body of work. In more phonetic writing, Tolkien used a macron to mark long vowels: ī, ē, ā, ō, ū. This freed him to use the accent to mark stress. He used acute accent (´) to mark primary stress and a grave accent (`) to mark secondary stress. I generally use the same conventions when discussing Elvish word-stress to avoid switching notations. Tolkien described stress in some detail in his analysis of the Namárië poem (Galadriel’s Lament) in the book of songs The Road Goes Ever On (RGEO) he published in collaboration with Donald Swan (here accents mark stress, not length):
The stresses employed metrically were those used in the normal pronunciation of Quenya. The main (high-toned) stress was originally on the first syllable of all words, but in words of 3 or more syllables it had been moved forward to fall on the penultimate syllable [second-to-last], if that were long; if it were short, it fell on the antepenult [third-to-last] irrespective of length, (as in éleni) (RGEO/60).
This note indicates stress coincides with higher intonation in Quenya, much as it does in English. These notes are of particular interest because Tolkien also addressed the placement of the secondary stress:
The initial syllable usually retained some degree of stress. In long words, especially recognized compounds, it was, though lower in tone, often equal in force to the main stress: as in óromárdi, fálmalínnar, etc. It was weaker when it immediately preceded the main stress, as in Àndū́ne, ṑmáryo, Tìntálle, Ròméllo; and in such cases, if it was short it became unstressed, as in avā́nier (Compare E. almighty, ècónomy, éconómical). The weaker stress can be employed as the metrical stress, or in the place of the unstressed element, according to their position. They are used as unstressed syllables only when immediately followed by a main stress as in Andūne, etc.
Final vowels were normally short and unstressed, in words of more than one syllable, if they followed the main stress as in lassi, linte, yulma, etc. But they had nearly all formerly been long vowels (or they would have disappeared) so that in the very frequent case of words ending in two short syllables, as ūnōtime, tellumar, lumbele, hīsië, etc., they received a light stress that could be used metrically (RGEO/60-61).
Thus the first syllable receives a secondary stress unless (a) it immediately preceded the main stress and (b) was a light syllable. The secondary stress was stronger if separated from the main stress, weaker if adjacent (and if adjacent, could not be used for metrical stress in poetry). In trisyllabic or longer words ending with two light syllables, the final syllable had a light secondary stress that could be used in metric poetry. Tolkien gave a version of the poem with primary and secondary stress marked, which provides a number interesting examples; in these examples | marks the metrical divisions (RGEO/68):
- yḗni| ū̀nṓt|imè |ve rā́m|ar ál|daròn “long years numberless as the wings of trees”
- sī̀ mán| i yúl|ma nín| ènquán|tuvà “who now shall refill the cup for me”
- an sī́| Tìntál|le Vár|da Ói|olós|sëò “for now the Kindler, from Mount Everwhite”
Note how the “small” monosyllables like ve, i, an (prepositions and the definite article) are generally unstressed in the phrase, but monosyllable with long vowels like sī “now” are stressed, as are more “significant” monosyllables like man “who” and nin “for me”. This phenomenon of stressed (and unstressed) monosyllables is mentioned in other contexts: “The Quenya (stressed) separate [pronoun] forms were Sg. 1a ní, 2a tyé b lyé, 3 sé …” (VT49/51, from notes on pronouns written in the late 1960s). Quenya independent pronouns generally appear with short vowels elsewhere, so presumably the vowel was lengthened when the pronoun was stressed within a phrase. The following sentence written around the same time provides a clue for when this might occur:
- melin sé apa la hé “I love him but not him [the other]” (VT49/15).
It seems that when the pronoun is emphasized, (“him but not him), the pronoun is stressed, much like in English. In Quenya this added stress is accompanied by vowel lengthening. This might be an intermediate tier of emphasis between regular unstressed pronouns and fully emphatic pronouns:
- melinyes or melin se “I love him” [simple statement of fact].
- melin sé apa la lyé “I love him but not you”.
- melin elye “It is you that I love [above all others]”.
This variation between short and long vowels is also seen in verbal particles like verbal particles like á/a (imperative) and lá/la (negative). It seems this lengthening is noted in writing, and may serve as a visual indicator of added emphasis in much the way that bold does in the English sentences above. This added long vowel mark probably appears only on emphasized elements: melin se apa lá lye “I love him but not you”.
The system described above (main stress on second-to-last syllable if heavy, otherwise third-to-last) applies to almost all Quenya words but is not completely universal. In The Lord of the Rings appendix E Tolkien said “In words of two syllables it falls in practically all cases on the first syllable (LotR/1116)”, but “practically all” is not the same as “all”. Probably the biggest exception is the suffixed imperative like “avá (stressed on the last syllable)” (WJ/371). I suspect it was stressed on the last syllable when used as a standalone interjection: norá “run!”, avá “don’t!”. It probably followed normal stress patterns when part of a larger phrase: áva quete ya carin “don’t tell me what to do [lit. don’t say what I do]”.
Conceptual Development: Tolkien established the Quenya stressed system very early and stuck with it throughout his life. A system basically (but not entirely) identical to the one above was described in his Qenyaqesta document written in the 1910s:
The accent falls always on some syllable which is not the final syllable (except in accented monosyllables of course). In dissyllables the first syllable is always accented.
In trisyllables the first syllable is always accented, where possible by the laws stated below, i.e. in practice, if the second syllable is not long. It, however, very frequently is: a very common word-rhythm being ˘ ´ ˘.
In quadrisyllables and more the stress (chief) can never go farther back than three syllables, except in one case, that of quadrisyllables where the two internal syllables consist of 2 short vowels in hiatus. This is also the only case where a short final vowel bears a strong secondary stress. Areanor is accented Áreanòr, but Areanóre is accented Àreanóre, and Pakalane is accented pakálane.
The secondary accent normally is on alternate syllables in either direction from the main stressed syllable (normally the antepenult). If this vowel is a short final one the secondary stress is comparatively weak (PE12/26-27).
Tolkien described the same basic system again in the Outline of Phonetic Development from the 1930s:
After the above changes and before PQ or AQ there occurred the Q. accent shift. The main accent became fixed and possibly somewhat less forcibly stressed (no further vowel obscurations or losses occurred except in final syllables). It now fell on the penult where that was long; otherwise it fell on the antepenult (PE19/59).
In this document he also discussed secondary stress:
The secondary stresses were usually reorganized and governed by the main stress; but on same principles as those described above for the archaic period — except that a secondary accent (often[?] with low tone) could stand on an initial syllable immediately before a long main stress, as in cṑmállo above.
The secondary accent falling on the initial syllable, or any non-initial long syllable, that was separated (by an intervening stressless long or one or more stressless short) from the main was strong (PE19/60).
Also:
In Tarquesta pronunciation the highest tone-pitch also accompanies as a rule the main stress in words pronounced “normally”, se. in isolation, or in circumstances where no special significant tonal modifications were required. An exception to this rule is provided by long words, especially compounds, in which a very strong secondary accent fell on the initial syllable, while the actual main stress fell later (often on a derivative syllable). In such cases the initial secondary was in normal circumstances higher in pitch than the succeeding main stress: e.g. kà¹rpalimái²te, Vā̀¹linṓ²re (PE19/61).
Again, this is essentially the same system described for the Late Quenya conceptual period of the 1950s and 60s, with the added information that the secondary stress on the initial element of a compound could have a higher pitch than the second element with the main stress.