Quenya speech has a strong rhythm to it, and sometimes the sounds in Quenya words change to maintain that rhythm.
6.1.1.1 Elision with two vowels: Quenya has a tendency to lose a vowel at the end of a word when the next word begins with a vowel, especially if the two vowels are similar. This process is called “elision”. We saw an example of this in our guided reading exercise from Chapter 2, Section §2.6:
The word tenna “until” appears before the word Ambar “World”. Since the next word begins with the same vowel, the final vowel of tenna disappears, leaving tenn’ Ambar-metta. The vowels don’t necessarily have to be the same, though: we also saw elision with elen síla lúmenn’ omentielvo “A star shines on the hour of our meeting” (LotR/81) in Chapter 1, Section §1.4. In this example, the final a of lúmenna “upon the hour” is lost before the vowel o.
Circumstances in which elision occurs include:
Speculative: For the last two cases, elision seem to be most likely in longer words if the result does not put the stressed syllables of the two words next to each other: márENN’ aLASsëo and VAnim’ ambaRÓnë are fine, but hrestann’ ailino is problematic because the two stressed syllables would be adjacent: hrestANN’ AIlino. In such cases I think elision is unlikely: hrestanna ailino “to the shore of a lake”. But hrestann’ ailinyo (hrestANN’ aiLINyo) “to the shore of his lake” is more likely to have elision because the stress patterns are amenable to it.
Elision may also occur if the words are particularly short and can be pronounced as a single unit; this is especially common if the following word is i “the”: minn’ i rotto “into the cave”, pronounced MINN’ i ROTto. Elision can occur with short words in other circumstances, especially with prepositions which tend to be weakly stressed, as with tenn’ Ambar (pronounced tenn’ AMbar) as seen above.
6.1.1.2 Elision with i “the”: In general, monosyllabic words like ni “me” and i “the” do not undergo elision with following vowels, otherwise they might become indecipherable or disappear entirely: i intya “the notion”. However, i “the” sometimes disappears when it follows a vowel, especially an i or e. We have two examples of this:
See section 6.4.1.3 Forms of the Definite Article for further discussion of alternate forms of i.
6.1.1.3 How do we know this? Tolkien did not discuss elision directly, so the rules above are deduced from attested examples. There are also some inconsistencies in how Tolkien applied elision. In Quenya prayers of the 1950s, Tolkien sometimes applied elision to suffixed pronouns (VT43/23). However, there are also examples in the same set of texts where elision does not occur and there are no other examples of elision of pronouns elsewhere in Tolkien’s later writings, so I consider this to have been a transient idea and ignore it.
The rule above where elision is prohibited in longer words if it would cause primary stresses to become adjacent is speculation on my part. It fits the available examples, but we have so few examples to work from that it could be wrong. However Quenya seems to have general preference for avoiding adjacent stressed syllables, so the proposed rule fits what we know about the language.
6.1.2.1 Vowel loss in long compounds: Given the number of suffixes used in Quenya speech, Quenya words can grow long. Compound words in Quenya in particular can be quite long, and with further suffixes they grow even longer. As such, there is a tendency to reduce the final syllables of compounds to make them shorter and more manageable. One common example of this is words ending in -nórë “land”, as in Valinórë “Vali-land” and Númenórë “West-land”, both of which have simplified forms Valinor and Númenor that are used more frequently.
In general, the final vowel is lost and any long vowel in the non-final syllable is shortened. If the result is a final consonant cluster, that cluster is also reduced but is restored in the stem form. For example quendë is an older word for “elf”, which in compounds reduces to -quen as in Moriquen (Moriquend-) “Dark elf”. Something similar happens with már “home”, whose vowel also shortens in the final syllables of compounds: Eldamar “Elf-home”, Valimar “Vali-home”.
6.1.2.2 Vowel shortening in suffixed words: When directional suffixes are added to longer words whose second-to-last syllable has a long vowel, that long vowel tends to shorten, as in Endorenna “to Middle-earth”, from Endórë “Middle-earth”. That’s because Quenya generally dislikes having adjacent heavy syllables, and one of the heavy syllables can be easily eliminated by shortening the vowel. This doesn’t happen with shorter words: lúmenna “upon [the] hour”.
6.1.2.3 Long-voweled monosyllables: In some cases reducing the long vowel is not practical, especially in monosyllables like ná “to be” or vá “don’t/won’t” where the long vowel is characteristic of the word. In such cases, the vowel may remain long, even when suffixes with consonant clusters are added: nántë “they are”, válmë “we won’t”. Short-vowel variants are also possible if the word is deemphasized: nantë, valmë.
Something similar can be seen for long-voweled monosyllabic nouns like má “hand” with possessive suffixes: márya “his/her hand”, mánta “their hand” (remember: mánta indicates one hand for each member of a group, so Quenya says “their hand” and not “their hands”).
Speculative: There is a special variant for the subject suffix -ssë when added to such words. The double-s in the suffix reduces to a single s when added to monosyllables like ná, á, vá and ui: násë “he is”, ásë “let him”, vásë “he mustn’t”, uisë “he is not”. The short pronoun suffix -s is also possible: nás, vás, uis, except for the indirect imperative ásë (“let him”), which requires -së.
6.1.2.4 How do we know this? Tolkien discussed the reduction of final syllables in compounds and vowel shortening in several places (PE17/106, PE21/32). The shortening of vowels when directional suffixes are added to nouns (Endorenna) was not discussed, but can be deduced from examples and fits Quenya phonology in general. The reduction of 3rd sg. -ssë after long-voweled monosyllable was described in a rough note from 1969:
When stem of noun or verb is long & long [?markers], double cons. was simplified after change of s > z. So nê-sse for nesse has nēse (VT49/28).
Likewise for the retention of the long vowel in má with possessive suffixes, as mentioned in a discussion of the negative verb lá:
lá is usually shortened to la before 2 consonants, according to the usual Q. procedure, but the long vowel can be retained, especially for additional emphasis, as in other cases where pronominal affixes follow a long vowel, as in márya “his hand” (PE22/160).
Tolkien vacillated on this approach. Examples like nanyë, nassë, nalmë with shortened vowels can also be found (VT49/30). Thus, it seems both nányë and nanyë are valid, as indicated in the note above.
Prosodic lengthening is a phenomenon that causes vowels to lengthen and is related to stress. Like stress, prosodic lengthening depends on whether the syllables in a word are heavy (with a long vowel, diphthong or two following consonants) or light (with a short vowel and a single consonant, or just a short vowel). This section discusses the basic mechanisms of prosodic lengthening; later sections will discuss the phenomenon along with the suffixes that are likely to cause it.
6.1.3.1 Conditions Causing Vowel Lengthening: Just as Quenya dislikes having two adjacent heavy syllables, Quenya also dislikes having too many light syllables in a row. Such a sequence of short-vowel syllables can occur when suffixes are added to a longer word. Vowel lengthening can be triggered when the base word is trisyllabic or longer, and one of the following is also true:
Up until this point in the course, we have avoided suffixes that are likely to trigger such lengthening, but we will see a number of them in this chapter. Using the vocalic noun culuma “orange” as an example, noun suffixes that are likely to trigger lengthening when added to longer nouns ending in a vowel include:
Using the consonantal noun Atanatar “Father of Men” as an example, suffixes that are likely to trigger lengthening when added to longer nouns ending in a single consonant include:
Consonantal nouns long enough to trigger prosodic lengthening are rare, and usually are compounds. Consonantal nouns whose stems end in consonant clusters also show no lengthening: Moriquen (Moriquend-) “Dark Elf” → Moriquendion “of Dark Elves”. Finally, the suffixes -i, -in, -ion do not trigger lengthening when added to e-nouns, since they replace the final vowel rather than adding to the base word: telumi, telumin, telumion.
6.1.3.2 Monosyllabic vs. Polysyllabic Suffixes: For monosyllabic suffixes like -va, lengthening is only triggered if both of the preceding syllables are light:
However, if the suffix is two syllables, it triggers lengthening even if the prior syllable is heavy. The genitive partitive plural suffix -lion is one such suffix:
The reason for this has to do with the rules for Quenya stress, which cannot fall further back than the third-to-last syllable. With a monosyllabic suffix, an original stress on the second-to-last syllable is still close enough to the end of the word to retain that stress and prevent lengthening: faLASseva. With a disyllabic suffix, the original stress is push far enough away from the end of the word that the original final vowel can accept the stress and lengthen as a result: falasSÉlion.
6.1.3.3 Lengthening Before Word Boundaries: Another way to think about prosodic lengthening is that Quenya “wants” the stress to mark the boundary between the word and a suffix, by falling on the last syllable of the word. This can’t happen if the base word is disyllabic: the first syllable is already stressed, so the last syllable can’t be. Likewise this can’t normally happen if the second-to-last syllable of the word is stressed because the original stress is again too close to the end of the word. As noted above, it does become possible with the suffix itself is two syllables, because the original stress is now far enough away from the end.
If the suffix itself already begins with a consonant cluster, lengthening is not necessary. For example, with telumenna “to the roof”, the stress already falls on the word boundary: teluMENna. If, however, a suffix with a single consonant is added, then the normal stress rules of Quenya would retract the stress to the third-to-last consonant: without any lengthening **telumeli “some roofs” would be stressed teLUmeli, not on the word boundary. Lengthening fixes this problem: teluméli gets the stress in the right place: teluMÉli.
Strictly speaking, when the suffix is two syllables, the added length is not required because the normal Quenya stress rules would put it into the right place: **telumelion without lengthening would still be stressed on the boundary: teluMElion. But lengthening happens anyway in Quenya, to further emphasize that boundary: telumélion.
All the examples discussed so far have marked the boundaries of a word and a suffix, but the same thing can happen in compounds as well, such as with Cuiviénen “Water of Awakening” = cuivië “awakening” + nén “water” (note that nén also shortens its long vowel in the final syllable of the compound, in keeping with the principles of section §6.1.2 above).
6.1.3.4 Retraction of Lengthening With Further Suffixes: As more suffixes are added to the word, the lengthened vowel may retract further back to mark the new boundary of the added suffix. For example, omentië “a meeting” lengthens its last vowel when the partitive plural suffix -li is added: omentiéli “some meetings”. The lengthened vowel moves back a syllable when the instrumental suffix -nen is added: omentielínen. Something similar happens with compounds: the genitive of Cuiviénen is Cuivienéno “of the Water of Awakening” (PE21/36). This does not happen for “combined” suffixes that are perceived as a unit, such as the genitive partitive plural suffix -lion, which is treated as a single disyllabic suffix.
6.1.3.5 How do we know this? Tolkien used prosodic lengthening regularly in his writings, but did not describe it directly beyond a few hints in his earlier writings: “Where the simple forms occur in long words the preceding vowel is usually long in inflected forms: as falmarin ‘sea-fay’ (masc.), pl. falmaríni” (PE14/43). The rules above are deduced from research into the attested forms that show vowel lengthening, and I believe this is the best explanation for this phenomenon. The vowel lengthening was first noted by Nancy Martsch all the way back in Vinyar Tengwar #13 published in 1990 (VT13/2) and I believe (but have not been able to confirm) that the term “prosodic lengthening” was first used by Alexander Zapragajev.
The key difference between the theory proposed above and other descriptions of the phenomenon I’ve seen elsewhere is that it highlights the role of word and suffix boundaries as the mechanism for where the lengthening occurs. It is very likely that prosodic lengthening originally arose from cases where long vowels were preserved before suffixes but shortened without them. But this alone cannot explain all cases, such as Atanatári, where the ancient form was Atanatar in the singular and Atanatarī in the plural. This is why I think prosodic lengthening is best treated as a pattern that preserves rhythmic stress in modern Quenya words to mark word boundaries.
Here are some vocabulary words having to do with warfare:
Instrumental Summary (Refresher): Here is a quick refresher on the instrumental for the next exercise:
Combine the following elements and indicate how the sounds would change in the combined forms. Capitalized words are names from Tolkien’s Legendarium.
To practice vocabulary words, translate the following to Quenya.
Answers are in Answer Key 6.1 at the end of this chapter.
Quenya actually has two different plurals, a group or total plural (using -i or -r) and a partitive or partial plural (“some”) using -li.
6.2.1.1 Function of the Partitive Plural: There are some interesting interactions in Quenya between the definite article i and the Quenya plural. A definite plural group (with i “the”) refers to a single (limited) group: i eldar matir massa “the elves eat bread” meaning a specific group of elves. Without an article the phrase becomes a general statement, referring to all of elvenkind: eldar matir massa “Elves eat bread”. In this context, the indefinite plural is also a total plural, referring to the entire class of Elves. The plural may still be qualified with other modifiers, as in eldar Lóriendo matir massa “Elves of Lórien eat bread”, but by default this phrase refers to all the elves of Lórien.
Quenya has a different plural suffix -li used to refer to part of a class; this is called the partitive or partial plural (Tolkien used both terms). Its meaning is roughly equivalent to English “some, many”, so eldali matir massa means “some/many elves eat bread”, which is less than all elves. Like English “some”, it also doubles as a way of specifying an indefinite group. For example, eldali túler i ostonna means “some elves came to the city”, that is some indefinite group of elves. You couldn’t say eldar túler i ostonna for an indefinite group, because that would imply all of elvenkind came to the city.
Speculative: Presumably once the indefinite group is mentioned, it would switch to the definite group plural for the rest of the discourse. For example: Eldali túler i ostonna. I eldar merner quetë i aranen. “Some elves (an unknown group) came to the city. The elves (the whole of the previously mentioned group) wanted to speak to the king”. The first instance uses the partitive plural -li for the indefinite group, and the second instance uses the normal plural -r and the definite article i for the now defined group.
6.2.1.2 Forming the Partitive Plural: To form the partitive plural with vocalic nouns, simply add -li to the final vowel. This will trigger prosodic lengthening if the base noun has three or more syllables, but not if the noun has its stress on the second-to-last syllable:
With several consonants, the partitive plural suffix undergoes assimilations: n, r, s all become l when the partitive plural suffix is added. The result is -lli for these as well as nouns ending in l:
Speculative: Other consonants have no suitable assimilations. Those, along with noun stems ending in consonant clusters, use a joining vowel e:
6.2.1.3 Definite Partitive Plurals: In English, “some” cannot be definite; you cannot say “the some elves”. You can, however, say “the many elves”, and this seems to be best English translation when the partitive plural is made definite: i eldali = “the many elves”. It seems this definite-partitive plural formation is rarely used, however, except in poetry where it may be useful for adding an additional syllable.
6.2.1.4 Partitive Plurals and Noun Cases: The partitive plural combines with the Quenya noun cases in various ways; we’ve already seen one instance in Chapter 5, §5.3.1 where we described the dative partitive plural suffix -lin = -li + n “to/for some ...”. Rather than covering all the possibilities here, each combination will be discussed along with each individual noun case.
6.2.1.5 How do we know this? Tolkien described the distinction between the general and partitive plural on numerous occasions (Let/178; PE17/62, 127, 135; PE21/73). For example:
... the two plural forms, the one casual and indefinite, the other particular and complete; so Eldar “Elves as a kind or people, or all the Elves concerned”; Eldali “some Elves, a lot of Elves” (PE17/135).
Thus in Q. Eldar (not with article!) = “Elves, The Elves, All Elves”; i Eldar = “(all) the Elves previously named” (and in some cases distinguished from other creatures); but Eldali “Elves, some Elves”. With Eldali the definite article is seldom used (PE21/73).
We do know the definite article was used at least occasionally with the partitive plural, since i falmalinnar “upon the many waves” appeared in the Namárië poem (LotR/377; RGEO/59). Given that the definite article “is seldom used”, I think it is reasonable to assume it is mainly a poetic devise.
The addition of the partitive plural suffix -li to vocalic nouns is straightforward: simply add the suffix to the vowel, the only complication being possible prosodic lengthening, as discussed above. With consonantal nouns things are less clear. We have only two examples: Casalli “some Dwarves” = Casar + -li (WJ/388) and elelli “some stars” = elen + -li (PE17/127). Both point to consonant assimilation rather than the use of a joining vowel. The other assimilations given above are speculative, based on what we know about the development of Quenya consonant combinations.
Assimilation becomes impossible when clusters are involved, and so a joining vowel is probably used, but we don’t know what it might be. The most common guess is that the joining vowel is e, as it is with the directional noun cases -(e)nna, -(e)llo, -(e)ssë. I think the same is true for individual consonants like t and m for which there are no suitable assimilations.
Here are some vocabulary words having to do with animals:
Partitive Plurals: Partitive plurals describe an indefinite group that is less than the entire class, using the suffix -li = “some, many”.
Translate the following to English:
Translate the following from English to Quenya. There are some new additional vocabulary words as well:
Answers are in Answer Key 6.2 at the end of this chapter.
Quenya has two distinct but similar noun cases, the genitive (using the suffix -o) and the possessive (using the suffix -va or -wa).
6.3.1.1 Function of the Genitive: The genitive case serves a function similar to the English preposition “of”, indicating relationships between things. It can be used to indicate:
The genitive typically follows the noun it is related to, but may precede it also: i atano má “the man’s hand”.
6.3.1.2 Forming the Genitive: The genitive is formed using the suffix -o, which combines with dual and plural suffixes as follows:
| Genitives | Sg. | Du. | Part. Pl. | Pl. |
|---|---|---|---|---|
| vocalic: cirya | ciryo | ciryato | ciryalion | ciryaron |
| e-noun: lasse | lasseo | lasseto | lasselion | lassion |
| consonantal: atar | ataro | ataruo | *atallion | atarion |
The plurals -ion, -ron and partitive plural -lion tend to trigger prosodic lengthening: Atanatárion “of Fathers of Men (Atanatar)”, culumáron “of oranges”, telumélion “of some roofs”. The genitive by itself may also trigger lengthening when added to long consonantal nouns: Ilúvatáro “of Ilúvatar”.
Genitives of nouns ending in -ië are awkward, since they would result in a sequence of three short vowels: márië → máriëo “of goodness”. To avoid this, a consonant n can be inserted before the genitive suffix; this will also trigger prosodic lengthening: máriéno (PE17/59). The extra n is unnecessary in the genitive plural: máriéron “of goodnesses (pl.)”.
6.3.1.3 How do we know this? As with other noun cases, most of what we know about how the genitive is formed comes from the so-called Plotz Letter, a 1966 or 67 letter from J.R.R. Tolkien to Richard Plotz which included a table of noun cases (VT6/14). That letter indicated that the genitive of a-nouns changed the final a to o: ciryo “of a ship”. We have other examples indicating the genitive of o-nouns remains unchanged, such as: Indis i·Kiryamo “The Mariner’s Wife” (UT/8). See below for a discussion of how we know the function of the genitive.
6.3.2.1 Function of the Possessive: The possessive case serves a function similar to the English apostrophe-s (’s), indicating possession, but it is also used for attributes and nouns-as-adjectives. Because of these two functions, Tolkien himself labeled this case the possessive-adjectival case (WJ/369). This case can be used to indicate:
The possessive typically follows the noun it is related to, but may precede it also: i atanwa macil “the man’s sword”.
6.3.2.2 Forming the Possessive: The possessive is formed using the suffix -va or -wa, which combines with dual and plural suffixes as follows:
| Possessives | Sg. | Du. | Part. Pl. | Pl. |
|---|---|---|---|---|
| vocalic: cirya | ciryava | *ciryatwa | ciryalíva | ciryaiva |
| e-noun: lasse | lasseva | *lassetwa | lasselíva | lassíva |
| consonantal: atar | atarwa | *atarúva | *atallíva | *ataríva |
For longer vocalic nouns, the possessive suffix -va is likely to trigger prosodic lengthening: teluméva “roof’s”.
6.3.2.3 Plurality of Possession: One interesting feature of the possessive case is that it behaves in some respects like an adjective (this was its original function). In particular, if the possessed noun is plural, the final -a in the possessing noun becomes -e just as if it were an adjective. Compare: i atanwa macil “the man’s sword” and i atanwe macili “the man’s swords”.
There are two possible dimensions of plurality, for both plural possessors and plural possessions. Either can be singular and plural independently:
6.3.2.4 How do we know this? Again, most of what we know about how the possessive is formed comes from the so-called Plotz Letter (VT6/14). The possessives suffix -wa is fairly well attested for consonantal nouns with individual consonants: andamacilwa (PE17/147), sindarinwa (LotR/1123). It is less clear how the possessive suffix behaves after consonant clusters. Vocalized -ua is currently my best guess, following the suggestions of Raccoon: https://feophan.github.io/tenguesta_ngoldorinwa/morph/nou/#fifth-declension-nouns. The next section discusses how we know the function of the possessive-adjectival case.
6.3.3.1 Distinguishing the Genitive and Possessive: A common shorthand to distinguish the genitive and possessive is that genitive -o is equivalent to English “of” and the possessive is equivalent to English apostrophe-s (’s), but this is misleading because in many circumstances English can use the two interchangeably. There is little difference between “a man’s sword” and “sword of a man” in English.
The clearest distinction between the two is that -va/-wa are used for possessions (things owned) and -o for most other relationships. A basic rules of thumb is that if the thing is a possession that could be given to someone else, use -va/-wa, otherwise use -o. For example, i atanwa macil “the man’s sword” can be given to someone else, and so uses the possessive. But i atano atar “the man’s father” or i atano má “the man’s hand” are relationships that cannot be given away, and so use the genitive.
However, the descriptive function of genitive -o and the adjectival function of -va are very similar, and in many circumstances either can be used. Both alcar Oromëo and alcar Oroméva means “Oromë’s glory” or “the glory of Oromë”. The only distinction between the two is that alcar Oromëo would imply that the glory was something originating from Oromë in the current situation, whereas alcar Oroméva would be the glory inherent to Oromë as a permanent attribute. In many circumstances this distinction is irrelevant, and the choice between the two purely aesthetic. When in doubt, genitive -o is used more frequently (with the exception of pure possession).
6.3.3.2 Definite Genitives and Possessives: Quenya frequently omits the definite article with genitive and possessive constructions in cases where English would use “the”, for example: má atano “the hand of a man”, macil atanwa “the sword of a man”. This is less strange than it seems, because if you were to use a proper possessive in English, it would also omit “the”: atano má “a man’s hand”, atanwa macil “a man’s sword”. The possessive or genitive can be said to fill the determiner slot, making the definite article unnecessary, the difference being that in Quenya, the genitive or possessive is allowed to follow rather than precede the modified noun.
If a definite article is used, it typically marks the definiteness of the modifying noun rather than the modified noun: i atano má “the man’s hand”, i atanwa macil “the man’s sword”. The same is true if the order is reversed: má i atano “the hand of the man”, i macil atanwa “the sword of the man”. With the genitive, the article appears before the definite noun (i atano) but with the possessive it appears before the entire phrase (i macil atanwa). This is because of the adjectival nature of the possessive suffix: Quenya (like English) does not like to have i “the” intervening between an adjective and its noun. If it does, it implies a “to be” statement: i macil i atanwa “the sword [is] the man’s”.
6.3.3.3 Loose Compounds as Genitives: It is possible to have a genitive-like construction simply by putting two words next to each other in a loose compound. Some examples Tolkien gave were cirya tyulma “a ship mast” (PE21/80) and Oromë róma “an Oromë horn” (WJ/368). The meanings are roughly the same as “a mast of a ship” and “a horn of Oromë”. English has a similar construction with roughly the same meaning. Quenya is particularly inclined to use this construction when the modified noun is indefinite, since ciryo tyulma “a ship’s mast” or tyulma ciryo “[the] mast of a ship” would imply a specific mast of the ship in question, the same way that “Orome’s horn” implies a specific horn in English, and “a horn of Orome” is needed to imply an indefinite horn.
6.3.3.4 The Last Declinable Word: Like English, adjectives in Quenya usually precede the noun they modify, but in Quenya they may occasionally follow it instead. Where such a thing happens, any noun case it is added at the end of the noun phrase, not just on the noun itself. This might happen, for example, when an adjective is used as an epithet: Aracorno Turca “Aragorn [the] Strong”. If a noun case were applied to this phrase, the case ending would be added to turca: macil Aracorno Turco “sword of Aragorn the Strong”, quenten Aracorno Turcan “I spoke to Aragorn the Strong”.
This rule cannot apply if the last word in a noun phrase already has a case ending: quenten heston i arano “I spoke to the captain of the king”. Since arano is already in the genitive, it cannot take another case ending, so the dative ending -n is applied to hesto “captain”.
6.3.3.5 How do we know this? Tolkien’s most detailed description on the function of both the genitive and possessive cases appears in the Quendi and Eldar essay written in 1959-60 (WJ/368-369). Most of the discussion above is based on that essay.
Tolkien mentioned the differing behavior of the definite article with genitive and possessive in a couple places (PE17/13, PE22/80). The second of these has a useful set of examples:
In some cases the meanings of singular -o coalesced with -va: as e.g. in kirya tyulma “a ship-mast, ship’s mast, mast of an unspecified or any ship”; tyulma kiryo “the mast from some ship, of some ship”; tyulma i kiryo “the mast of the ship” or i tyulma kiryava (PE22/80).
This example is the basis of the notion above that the definite article i cannot intervene between the possessed noun and the possessive. It also illustrates the function of loose compounds as genitive-like constructions.
“The Last Declinable Word” rule was mentioned by Tolkien in notes appearing with Cirion’s Oath: “as is usual in Quenya in the case of two declinable names in apposition only the last is declined” (UT/317).
Here are some vocabulary words having to do with time:
Genitive Summary: The genitive noun case is for relationships similar to those with English preposition “of”:
| Genitives | Sg. | Du. | Part. Pl. | Pl. |
|---|---|---|---|---|
| vocalic: cirya | ciryo | ciryato | ciryalion | ciryaron |
| e-noun: lasse | lasseo | lasseto | lasselion | lassion |
| consonantal: atar | ataro | ataruo | *atallion | atarion |
The genitive is used for:
Possessive Summary: The possessive noun case is for relationships similar to those with English apostrophe-s (’s):
| Possessives | Sg. | Du. | Part. Pl. | Pl. |
|---|---|---|---|---|
| vocalic: cirya | ciryava | *ciryatwa | ciryalíva | ciryaiva |
| e-noun: lasse | lasseva | *lassetwa | lasselíva | lassíva |
| consonantal: atar | atarwa | *atarúva | *atallíva | *ataríva |
As with adjectives, the final vowel a of the possessive becomes ë when the thing possessed is in the plural: atanwë macili “a man’s swords”, atanívë macili “men’s swords”.
The possessive is used for:
This second descriptive use is often interchangeable with the genitive. The main distinction in descriptive use is that -va tends to be used for innate attributes, while -o is used of attributes applicable in the moment: alcar Oromëo = “Orome’s glory (in the moment)” vs. alcar Oroméva = “Orome’s (inherent) glory”.
Translate the following from English to Quenya. Pay particular attention to cases where Quenya would use genitive -o vs. possessive -va, as well as cases where it is ambiguous.
Translate the following from Quenya to English as vocabulary practice:
Answers are in Answer Key 6.3 at the end of this chapter.
This section discusses the differences between definiteness and indefiniteness in Quenya, and how the use of the Quenya definite article i differs from English “the”.
6.4.1.1 Function of the Definite Article: Before we can discuss the definite article, we must first discuss what it is used for in a sentence. Both English “the” and Quenya i are used to mark a definite or specific known thing within a discourse, as opposed to some unknown or unspecific thing. In both languages, if you were to say elda túlë i ostonna “an elf came to the city”, it would be some indefinite or unspecific elf, but if you were to say i elda túlë i ostonna “the elf came to the city”, it would mean the elf is specific and known to both the speaker and the listener.
The most common way for a noun to become definite is for it to have been introduced previously in the conversation. Consider:
The first mention of the elf is indefinite, because presumably he or she is unknown to at least one participant in the conversation. The second mention of the elf becomes definite, however, because it refers back to the (now specified) elf that was previously mentioned. Prior mention is not the only reason a word might be marked definite, however. In the second sentence i aranen “to the king” is also marked as definite, the implication being that there is some specific known king to whom the elf wished to speak, presumably the king of the city. If you were to say i elda mernë quetë aranen “the elf wanted to speak to a king”, then presumably the Elf did not care which king he or she would talk to. Consider also these two sentences:
The first sentence implies there are multiple windows in the room, and any of these windows may be opened. The second sentence implies there is one specific window that should be opened, either because it is the only one in the room, or it is the window next to the listener.
6.4.1.2 Definite Article Usage: Quenya uses the definite article less often than English. The circumstances where both English and Quenya use a definite article are when:
Both English and Quenya do not use the definite article with proper names, even though it has a specific referent: Elerondo but not **i Elerondo “the Elrond”, except perhaps when distinguishing multiple Elronds: i Elerondo Imladrisso “the Elrond of Rivendel” (as opposed to some other Elrond). However, one difference between English and Quenya is that English uses “the” to mark unique things like the Sun, Moon or sky, while Quenya does not:
Something similar happens with epithets (nicknames) following a name: English uses “the” but Quenya omits the definite article:
English also uses “the” with a determining adjective like first, last, highest, etc, while Quenya does not:
Note that the i in the Quenya sentence above is not the definite article, but rather is the indeclinable relative pronoun i “who, which, that”.
6.4.1.3 Forms of the Definite Article: The definite article almost always has the form i, and it is generally resistent to elision because otherwise it would vanish entirely. It sometimes causes elision for preceding vowels, however: ar’ i mindon “beside the tower”, á mat’ i massa “eat the bread”. There is also an alternate form of the definite article in which is used occasionally. We only have a couple examples of its use in Tolkien’s later writings:
For the second example, the phrase was deleted and the final version of the sentence had only (singular) i indo Sindicollo “the mind of Thingol”. Both these examples have two things in common: the following word begins with a vowel and the following word is plural. I think it is likely that in is an archaic plural form of the definite article, which is occasionally used before plurals beginning with vowels. There are plenty examples where in is not used, which makes me believe the use of in is optional: i eleni “the stars” (LotR/377), i arani “the kings” (WJ/369).
Finally, there is the unusual example:
Here ’n seems to be an elided variant of in, even though the following noun is not plural. Speculative: I think something different is going on here, because the use of the ordinary article would result in several sequential vowels: utúlië i aurë. This is difficult to pronounce, so my working theory is that when i is “squeezed” between two vowels it reduces to ’n. See the discussion in section 6.1.1.2 Elision with i “the” above.
6.4.1.4 How do we know this?: Tolkien did not describe Quenya’s rules for when it does or does not use the definite article i. The above is largely deduced from observations of how Tolkien used Quenya i in actual sentences. My own interpretation of these rules is based largely on Raccoon’s research into the topic, which is much more thorough than mine. See https://feophan.github.io/tenguesta_ngoldorinwa/syntx/art/.
The occasional appearance of in was also not explained by Tolkien in his later writings. In his earliest iterations of Quenya grammar from the 1920s, in and n’ were variants of i “the” used before vowels and had nothing to do with plurality (PE14/42). However, shortly after this Tolkien introduced in as a plural definite article in the Sindarin branch of the Elvish languages (PE13/120). Tolkien largely abandoned the use of Quenya’s prevocalic n’ while retaining plural in for Sindarin. Our two known Quenya uses of in from Tolkien’s later writing are both (a) prevocalic and (b) plural, so I suspect that they represent remnants of an archaic plural definite article that had faded from use in Quenya but survived in Sindarin.
The definite article works differently with plurals. A definite group, i eldar “the elves”, implies a specific group of elves. An indefinite group, eldar “Elves”, implies the statement is applicable to any elf and hence is true of all of Elvenkind. Consider:
In the first, a specific known group of Elves is eating bread. In the second statement, since this group is not specified, the implication is that it is a general statement meaning all of Elvenkind eats bread. If you want to indicate an unspecified group that is less than all of Elvenkind, in English you use the word “some” and in Quenya you use the partitive plural suffix -li.
The partitive plural is usually indefinite, and only rarely is used with the definite article i. Speculative: When it is, however, the implication is that the definite group is large (“many”).
How do we know this?: See the discussion in section §6.2.1.5 on where the information we have on the partitive plural comes from, and how it ties to (in)definite plurals.
According to Tolkien the “article = ‘the’ [i] is usually not used when noun is defined by a possessive [-va] or genitive [-o]” (PE17/13). This is consistent with Tolkien’s use of the possessive and genitive noun cases, where the modified noun rarely has the definite article. Thus where English would say “I saw the captain of the orcs”, Quenya would say tirnen hesto i orcoron. Here hesto does not need an article, because it is already defined by the qualifier i orcoron.
This is less strange than it seems. If the sentence were rearranged slightly, English would also omit the definite article: “I saw the orcs’ captain”, tirnen i orcoron hesto. In English, the possessive (the orcs’) fills the determiner slot, making a second “the” redundant: “I saw the orcs’ **the captain”. The same is true of Quenya, but since the genitive or possessive can both precede or follow the modified noun, the definite article i is redundant either way: atanwa macil or macil atanwa both mean “a man’s sword”.
While the definite article is typically not used with the modified noun, it can still be used with the modifying noun to mark whether it is definite. For example, with the possessive suffix -va, the article i can still be applied to the possessor. Thus atanwa macil mean’s “a man’s sword” (some unspecified man) but i atanwa macil means “the man’s sword”, referring to some specific man. The same is true of the genitive: i atano má “the man’s hand” vs. atano má “a man’s hand”.
Things are a bit more complex if the order is reversed. With the genitive, the definite article precedes the modifying noun: má i atano “the hand of the man”. With the possessive, however, the definite article precedes the entire phrase: i macil atanwa “the sword of the man”. This is because the possessive functions more like an adjective and the article cannot intervene between a noun and its adjective. In fact i macil i atanwa would likely be interpreted as a “to be” expression: “the sword [is] the man’s”.
Despite the discussion above, the definite article is sometimes applied to a noun modified by a possessive or genitive if there is a more general principle requiring it. Consider:
Here coimas “waybread” is being used as an abstract concept, not referring to any particular waybread, and so is analogous to phrases like i macil (ná) turwa tamma ohto “the sword is a powerful tool of war”. As such, it requires the definite article. If it referred to actual bread rather than the concept of bread, the i would not be required, because it would be defined by the genitive: camnelvë coimas i(n) Eldaron “we received [the] waybread of the Eldar” or “we received the Eldar’s waybread”.
How do we know this?: See the discussion in section §6.3.3.5 on our sources of information for the relationship between the genitive/possessive and the definite article.
Here are some vocabulary words having to do with quantities:
Definite Article: The Quenya definite article is i “the”, but it is used less often than in English.
Translate the following to Quenya, paying particular attention to phrases where Quenya would omit the definite article even though English uses it:
Translate the following to English as vocabulary practice:
Answers are in Answer Key 6.4 at the end of this chapter.
This section discusses the Elvish numbering system, and how Elves go about counting and writing numbers.
The first twelve Quenya numbers are:
In the 1930s Tolkien used tolto for “eight” and cainen for “ten”. You might see both of these alternate forms in Neo-Quenya writing, but in the late 1960s he replaced these with toldo and quëan respectively. These numbers are used for counting things, but Quenya has two special alternate words for “one” and “two”:
The word er applies to a single thing that is unique (or is situationally unique), as opposed to min which applies to one out of a group or the first item counted in a list. If you were to say min macil “one sword” it would mean you want one sword out of a group of possible swords, but er macil would imply there is only one sword to begin with: “the one sword, the only sword”. Similarly, min oron would be one mountain in a mountain range but er oron would be one mountain standing alone; the number er is thus related to eressëa “lonely”. Similarly, yúyo “both” refers to pairs, so yúyo macil would mean “both swords” as a set, but macil atta “two swords” would mean any two swords will do.
The last example macil atta illustrates a peculiar feature of Quenya, a side effect of its dual formation: when referring to two of something, the singular form of the noun is used, which would be like saying “two sword” in English. See section 5.2.1.2 Noun Duals for further discussion. For numbers three and larger, the plural form is used: macili neldë “three swords”, macili canta “four swords”, etc.
Furthermore, for numbers two and larger, the number follows the noun rather than preceding it, as shown above. This makes numbers particularly prone to “The Last Declinable Word” rule (see section §6.3.3.4). However, only the number two (atta) can accept case endings; other numbers are indeclinable so the noun receives the case ending: quenten elda attan “I spoke to two elves” but quenten eldain neldë “I spoke to three elves”.
The numbers from the previous section, min, atta, neldë, etc., are cardinal numbers used for counting things. Quenya also has a set of ordinal numbers for ordering things:
The pattern for converting cardinals to ordinals is straightforward: simply change the final vowel to -ëa. The only exceptions are minya “first” and quainëa “tenth”, the latter because quëan [probably] becomes quain in compounds (see below). As discussed in section 6.4.1.2 Definite Article Usage, it seems Quenya (unlike English) does not use a definite article with determining adjectives like ordinals, so Quenya would say attëa elda ter i ando (without i “the”) whereas English would say “the second elf through the gate” using “the”.
Tolkien gave us examples of Quenya numbers from thirteen to nineteen, though some of these only appeared in their ancient Elvish form so that modern forms need to be extrapolated:
Speculative: Tolkien did not describe larger numbers in his later writings, but he did describe them in the 1920s when the Quenya word for “ten” was cainen: yucainen, nel(de)cainen, can(ta)cainen “twenty, thirty, forty” (PE14/49). Extrapolating these to the 1960s paradigm where “10” became quëan/-quain:
Also based on Tolkien’s earlier writing, it seems Quenya writes smaller number first and then larger numbers, so lempë yuquain “five twenty” = 25 instead of the English order “twenty five”. This is also consistent with how Quenya digit are written in reverse order from English: “nine thousand two hundred sixty four” would be written in Quenya with the order 4629 (ôöòù) rather than 9264.
Quenya has two different numbering systems, regular decimal numbers (base 10) and duodecimal numbers (base 12). Quenya uses the same set of digits for both:
The digit “zero” is 0 (we don’t know its Quenya name). As noted above, Quenya digits were written in the reverse order from English numbers, so decimal value 1000 would be written ðððñ. Underdots could be used to mark a duodecimal number, so ðOðOðOñO would be 100012 (base-12) or 1728 as a decimal number. If it is necessary to distinguish them, decimal numbers can be represented with overdots. Thus ðTðTðTñT would be 100010 (base-10) = “one thousand”.
Beyond their digits, we don’t know much else about duo-decimal numbers in Quenya. There is a word yurasta “24” = two twelves in notes from the 1930s (PE14/17). Speculative: It is possible that -rasta plays a role for duodecimal numbers similar to that of -quain for decimal numbers, so that:
But these are just guesses with no real evidence, so I’d recommend sticking to decimal numbers for purposes of Neo-Quenya.
Much of the discussion on Quenya numbers above is drawn for several documents Tolkien wrote on Eldarin Hands, Fingers & Numerals from the late 1960s (VT47/3-17; VT48/4-14). The discussion of ordinals is based on notes on Eldarin numerals (VT42/24-27) associated with an essay on The Rivers and Beacon-hills of Gondor, also from this same period.
These documents introduced a new name for “10”, the word quëan or quain, the latter given in parentheses (VT48/6). Tolkien abandoned earlier cainen “10” because this form was identical to cainen “I laid”, the 1st person past tense of caita- “to lay” (VT48/12). Since quain was in parenthesis, I think quëan is the independent word for “10” as derived from primitive kwayam (VT42/24), whereas quain is a reduced form used in compounds (< kway’m), as evidenced by its appearance in quainëa “tenth” (VT42/25).
These 1960s documents also included words for 13 to 19 (VT48/21), but two were only given in their primitive Elvish forms: nelekwe “13” and kanakwe “14”. The modern forms *nelquë and *canquë must be extrapolated from these, with *nelquë supported by nelquëa “thirteenth”. The third form tolokwe “18” might be a slip for modern *toloquë, or it might be a primitive form; I personally think it is primitive, and the modern form would be *tolquë.
Numbers larger than 19 only appear in Tolkien’s earlier writings from the period where the word for “10” was kainen. The Early Qenya Grammar of the 1920s gives numbers 20 to 90, with the pattern: yucainen, nel(de)cainen, can(ta)cainen “twenty, thirty, forty”, etc. (PE14/49). These can be readily updated by changing -cainen to -quain, with some other minor tweaks.
The placement of small numbers before larger ones is consistent both with Tolkien’s 1920s number system and how Quenya digits work. The placement of numbers after the noun they modify was mentioned in notes from the 1960s, as well as the unusual use of singular nouns with the number 2: elen atta “two star” vs. eleni neldë “three stars” (VT49/44-45). This same note also describes how atta “two” follows “The Last Declinable Word” rule but other numbers do not.
The existence of the alternate duodecimal number system is something Tolkien mentioned several times, including in The Lord of the Rings itself (LotR/1107). We don’t know much about this system other than a discussion of tengwar digits, first published in the magazine Quettar issues #13 and #14. Aside from that, we only have a few isolated words like yurasta “twenty four = two twelves” (PE14/17). This makes any use of the duodecimal system very speculative.
The numeral glyphs presented in the Quettar article are reproductions by Christopher Tolkien, not Tolkien’s original writing, so they may not be correctly represented. They may also not be the only system Tolkien considered: the number digits in the King’s Letter written around 1950 are quite different in form; see the discussion at https://www.forodrim.org/daeron/md_kl_num.html.
Number Summary: Smaller numbers are:
Larger numbers are:
Two special numbers are used for solitary items and pairs:
Numbers larger than one follow the noun:
Ordinal numbers (mostly) change the final vowel to -ëa:
Combined numbers put the smaller digit first, followed by the larger one: canta lepenquain “four fifty”, instead of the English order “fifty four”.
Translate the following to English:
Answers are in Answer Key 6.5 at the end of this chapter.
Cirion’s Oath was the promise of allegiance between Gondor and Rohan given by Cirion, steward of Gondor, to then king of Rohan Eorl in the 26th century of the Third Age (UT/305). Tolkien only translated the end of the oath into Quenya. This text has several examples of places where Quenya omits the definite article i in circumstances where English would include it (see section 6.4 Definite vs. Indefinite). Before discussing the oath proper, we introduce some vocabulary:
We break up the text into four phrases to simplify analysis:
Vanda sina termaruva Elenna-nóreo alcar enyalien
ar Elendil Vorondo voronwë.
Nai tiruvantes i hárar mahalmassen mi Númen
ar i Eru i or ilyë mahalmar ëa tennoio.
Broken down literally this is:
Vanda sina termar-uva Elenna-nórë-o alcar enyalie-n
Oath this endure-will Elenna-land-’s glory memory-for
ar Elendil Vorond-o voronwë
and Elendil Faithful-of faith
Nai tir-uva-nte-s i hára-r mahalma-sse-n mi Númen
May watch-will-they-it those (are sitting)-(pl.) throne-on-(pl.) in West
ar i Eru i or ilyë mahalma-r ëa tennoio
and the One who above all throne-(pl.) exists forever
First Phrase: Vanda sina termaruva Elenna-nóreo alcar enyalien. The beginning half of the first phrase is simple: vanda sina termaruva “this oath will endure”, the only unusual bit being the placement of sina “this” after the noun vanda “oath”. The second half is more tricky. The verb termar- “to endure” is an intransitive verb and cannot have a direct object, but it can have an indirect object: enyalien “for memory”. The preceding alcar is a loose compound used as a genitive (see section §6.3.3.3), so alcar enyalien “for glory-memory” = “for memory of glory”. The preceding Elenna-nórëo is an ordinary descriptive genitive meaning “Elenna-land’s”, so the closest translation that would be valid English is: “for Elenna-land’s glory-memory”.
This phrasing is pretty awkward English, however, so revising the phrase to use the preposition “of” (and also reversing the order) would result in “for memory of [the] glory of [the] land Elenna”; the definite article i is missing from the Quenya phrase because Quenya does not require a definite article for words governed by a possessive or genitive (see section §6.3.3.2). The entire phrase would translate as “This oath will endure for memory of the glory of the land Elenna”. Note that Elenna “Starwards” is another name for Númenor.
Second Phrase: ar Elendil Vorond-o voronwë. The second phrase has the name Elendil Voronda “Elendil [the] Faithful”, with Elendil being his name and Voronda “[the] Faithful” being an adjectival epithet, without i “the” because the definite article is not required before epithets in Quenya (see section §6.4.1.2). The phrase follows “The Last Declinable Word” rule (see section §6.3.3.4), so the genitive suffix -o is added to Voronda, changing the final a to o as usual (see section §6.3.1.2). The entire genitive phrase applies to voronwë “faith”, so that the meaning is “and Elendil the Faithful’s faith”. Again the English wording can be made less awkward by using “of” and reversing the order: “and the faith of Elendil the Faithful”.
Third Phrase: Nai tiruvantes i hárar mahalmassen mi Númen. The verb in the third phrase is tiruvantes, decomposing to tir-uva-nte-s = “they will watch it”. The preceding nai indicates this is a wish (see section §3.4.3), so a more complete translation would be “may they watch it”. The “it” being watched is the oath (vanda), and the “they” is specified by the remainder of the phrase: i hárar mahalmassen mi Númen “who sit on thrones in West”. Note how the verb hárar “sit” is plural to agree with its indirect referent -ntë “they”.
The word mahalmassen “on thrones” is a locative plural (mahalma-sse-n). In theory “in West” could be locative as well. However, Quenya feels awkward using repeated locatives, the same way English feels awkward with repeated prepositions. As such, the phrase uses the preposition mi “in” instead. Note that mi Númen “in [the] West” lacks the definite article i because it describes a unique location, and universally unique things do not use the definite article in Quenya (see section §6.4.1.2). A full translation of the phrase would be “May they watch it, [those] who sit on thrones in the West”; English requires an additional “those” to be grammatically correct. Based on context the “they” are clearly the Valar of the Uttermost West.
Fourth Phrase: ar i Eru i or ilyë mahalmar ëa tennoio. The i Eru is the main noun of the fourth phrase, but this noun has no associated verb because Eru is another participant in the watching of the third phrase: “they (the Valar) and Eru” are both watching the oath. It is odd that Eru is preceded by the definite article i, but here Eru is used as a title rather than a name, so that i Eru = “the One” much like how in English you might say “the Lord” in reference to God.
The remainder of the phrase further qualifies the identity of Eru: i or ilyë mahalmar ëa tennoio “who above all thrones exists forever”. Thus the translation of the fourth phrase is “and the One who above all thrones exists forever”.
Putting the above together, a full translation of the oath would be:
This oath will endure for memory of the glory of the land Elenna,
and the faith of Elendil the Faithful.
May they watch it, [those] who sit on thrones in the West,
and the One who above all thrones exists forever.
Tolkien’s actual translation was:
This oath shall stand in memory of the glory of the Land of the Star,
and of the faith of Elendil the Faithful,
in the keeping of those who sit upon the thrones of the West
and of the One who is above all thrones for ever.
This translation is somewhat loose, with various minor prepositional changes to make the English flow better. In two places Tolkien was quite liberal with the translation. The English phrase “of the Land of the Star” is based on the English meaning of the name Elenna “Starwards”, so “of the land Starwards” becomes “of the Land of the Star”. The English phrase “in the keeping of” is a rewording of “may they watch it” with the same general sense but a completely different set of words.
Warfare:
Animals:
Time:
Quantity:
Numbers:
Large Numbers:
Miscellaneous:
6.7.2.1 Quenya Vowel Shifts Summary:
6.7.2.2 Partitive Plural Summary: Partitive plurals describe an indefinite group that is less than the entire class, using the suffix -li = “some, many”.
6.7.2.3 Genitive Case Summary: The genitive noun case is for relationships similar to those with English preposition “of”:
| Genitives | Sg. | Du. | Part. Pl. | Pl. |
|---|---|---|---|---|
| vocalic: cirya | ciryo | ciryato | ciryalion | ciryaron |
| e-noun: lasse | lasseo | lasseto | lasselion | lassion |
| consonantal: atar | ataro | ataruo | *atallion | atarion |
The genitive is used for:
6.7.2.4 Possessive Case Summary: The possessive noun case is for relationships similar to those with English apostrophe-s (’s):
| Possessives | Sg. | Du. | Part. Pl. | Pl. |
|---|---|---|---|---|
| vocalic: cirya | ciryava | *ciryatwa | ciryalíva | ciryaiva |
| e-noun: lasse | lasseva | *lassetwa | lasselíva | lassíva |
| consonantal: atar | atarwa | *atarúva | *atallíva | *ataríva |
As with adjectives, the final vowel a of the possessive becomes ë when the thing possessed is in the plural: atanwë macili “a man’s swords”, atanívë macili “men’s swords”.
The possessive is used for:
This second descriptive use is often interchangeable with the genitive. The main distinction in descriptive use is that -va tends to be used for innate attributes, while -o is used of attributes applicable in the moment: alcar Oromëo = “Orome’s glory (in the moment)” vs. alcar Oroméva = “Orome’s (inherent) glory”.
6.7.2.5 Definite Article Summary: The Quenya definite article is i “the”, but it is used less often than in English.
Tengwar answers:
Tengwar answers:
Tengwar answers:
Tengwar answers: