A subordinate or dependent clause is one that is not a sentence on its own, but modifies the larger context in which it appears: “the man who came here yesterday was very angry”. Many subordinate clauses begin with relative pronouns, as in the previous example: “who”. In English relative pronouns are the same as question words, but Quenya questions words (based on ma) are distinct from relative pronouns, which are based on i or ya.
Indeclinable i: Most attested relative clauses are formed with the pronoun i:
- Átaremma i ea han Ea “our Father who art in Heaven” (VT43/12).
- sív’ emme apsenet tien i úcarir emmen “as we forgive those who trespass against us” (VT43/12).
- alye arca meterme i naiquear “pray for us sinners (lit. for we who sin)” (VT43/27).
- Eru i or ilye mahalmar ea tennoio “Eru who is above all thrones for ever” (VT43/27).
- i carir quettar ómainen “those who form words with voices” (WJ/391).
- nai tiruvantes i hárar mahalmassen mi Númen “in the keeping of those who sit upon thrones of the West” (UT/305).
The pronoun i has a broad meaning. In the examples above, it functions as the subject of the subordinate clause, and essentially means “who” (both singular and plural). There are other examples where it functions as a subordinate conjunction “that”:
- umbe nin i hríve nauva urra (si loa) “I have a feeling that winter will be bad (this year)” (PE22/168).
- savin i Elessarno quetie naite “I believe that the words of A[ragorn] are true” (PE22/158).
In these examples i is not the subject of the subordinate clause, and simply joins the subordinate clause to the main clause. These examples are all form Late Notes on Verbs from 1969, but similar examples appear in the Quenya Verbal System (QVS) from the late 1940s:
- ᴹQ. me·merner i Túro tule “we wished that Turo come” (PE22/118).
- ᴹQ. le·merne i ni·karnéte “you wished that I had made (lit. made) it” (PE22/118).
In QVS pronominal subjects were prefixed to the verb rather than suffixed. In later syntax these phrases might be:
- mernelme i Túro tule “we wished that Turo come”.
- mernelye i carnenya sa “you wished that I had made (lit. made) it”.
Finally, there are some examples where i functions as the direct object of the subordinate clause:
- lá carita i hamil mára alasaila ná “not to do (in this case) what you judge good (would be) unwise” (LVS, PE22/154).
- ᴹQ. vahaia nóre ea i a-esta Valinor “far away (there) is a land [that is] called Valinor” (QVS, PE22/118).
- ᴹQ. manar i·lie i me·kenner “who are the people that we saw” (QVS, PE22/124).
In the first example i is the thing that “you judge good.” The second example uses late 1940s QVS syntax with prefixed pronominal suffixes and the indefinite subject ᴹQ. a; in later syntax it might become: vahaia ea nóre i mo esta Valinor. Here i is the thing that “one calls Valinor”, with the indefinite pronoun (a or mo) as the subject and i as the object of the subordinate clause. Likewise the third example would in later syntax become manar i lie i cennelme , with i being the object “that we saw”.
Thus it seems that i can be used (a) when it functions as the subject of the subordinate clause, (b) when it functions as the object of the subordinate clause or (c) when it functions as a subordinating conjunction. It seems i is indeclinable, and its referent can be both a person (our Father) or a thing (a land), and can be singular (Eru) or plural (us sinners). Tolkien briefly described the function of i in QVS:
But in the latter case (with nouns) a clause was far more usual, and could be used in all cases where the subject of the second verb was not the same as that of the first. A clause in such cases is introduced by i, before vowels in. The tense inside the clause depends on that of the first verb: the time of which becomes the present of the second verb (PE22/118).
In this context he was comparing i to the use of the infinitive, as in mernenye care sa “I wished to make it” vs. mernelye i carinye sa “you wished that I make it”. In the former, a subordinate clause is not required, because the subject of the second verb is the same as the first; in the latter a clause is required because the subjects (“you” vs. “I”) are not the same. In QVS, the relative pronoun i becomes in before vowels, but there is no sign of this in Tolkien’s later writing; see i úcarir and i Elessarno above.
Also of interest is the fact that the tense of the subordinate clause is determined relative to the tense of the main verb, not unlike the infinitive. In particular, in mernelye i carinye sa the subordinate clause (with an aorist verb) is in the same time frame as the main verb, whereas with mernelye i carnenya sa the subordinate clause (with a past tense) is even further in the past, hence the translation: “that I had made it” despite the fact that the subordinate clause’s verb is only in the simple past rather than the pluperfect.
Declinable ya/ye: In addition to the very generic relative pronoun i, Quenya has a more specific pronoun ya. Its references can only be things, and in the examples where it appears it is declined into a noun case:
- yassen tintilar i eleni “wherein the stars tremble” (LotR/377).
- sana wende nieninquëa yan i wilyar antar miquelis “that maiden like a snowdrop to whom the air gives kisses” (PE16/96).
In the second example, the thing given kisses is the snowdrop, not the maiden, so despite the gloss “whom” it still refers to a thing rather than a person. There is at least one phrase where ya seems to be used as a subordinate conjunction, however:
- sí ar lúmesse ya firuvamme: násie “now and at the hour of our death [lit. that we die]: Amen” (VT43/28).
In the prior version of this sentence, the relative pronoun was yá with long á (VT43/27), so it is possible that rather than relative pronoun ya “what”, this was the subordinate conjunction yá “when” (lit. “when we die”), also seen in yá hríve mene, ringa ná “when winter comes, it is cold” (VT49/23). It seems likely that short ya is used as a (neuter) relative pronoun, whereas long yá functions as subordinate conjunction “when” vs. the more generic subordinate conjunction i “that”.
In addition to neuter ya, Quenya also has a personal/animate relative pronoun ye “who(m)” which can likewise be inflected into noun cases: genitive yeo, allative yenna, ablative yello (VT47/21). Compare inanimate ya “what”/animate ye “who” with inanimate sa “it”/animate se “he, she”. Interestingly, this “inflectable who” ye had a plural form of i:
who rel. per[sonal] ye, pl. i
that rel. imp[personal] ya
yenna leltanelyes “to which [?whom] you sent him”
yeo / yello / ion / illon camnelyes “from whom you received it” (VT47/21).
This plural form of ye may have influenced the function of the indeclinable relative pronoun i. Unlike indeclinable i, both ya and ye are inflected for noun case and number. Indeed, where they appear in example sentences from Tolkien later writing, they are inflected for noun case (not so in Early Qenya, see below). This may be the distinction between indeclinable i and declinable ya/ye: the latter are used where case inflection is required.
Origins of ya and i: Tolkien discussed the origins of ya in notes associated with Galadriel’s song (NGS) written in the late 1950s or early 1960s:
ya-, used in Quenya as stem of relatives, being originally a demonstrative referring back to something behind, or previous in time (PE17/66).
Indeed, the root ᴹ√YA appears in The Etymologies from the 1930s with the gloss “there, over there, back/ago (of time)”, and is also the basis for words like yára “old”. As a relative pronoun, it may have originally meant “that thing previously referred to”. The use of ya as a relative pronoun dates all the way back to the Early Quenya Grammar (EQG) of the 1920s, though at this earlier stage it was indeclinable:
The indeclinable relative pronoun is ya, which is either to be understood in any relation, or, very frequently, is defined by demonstrative or pronominal or adverbial words inside the relative clause (PE14/54).
The etymology of i as a relative pronoun is less clear, but its use must have been ancient because it serves a similar function in Sindarin. Most likely it was connected to the definite article i “the”, whose root in The Etymologies was glossed: ᴹ√I “that (deictic particle)”. It was also mentioned in the Gnomish Lexicon of the 1910s as the “root of relatives” (GL/50), though in the contemporaneous Qenya Lexicon this root was glossed “here it is” (QL/41). It may have been influenced by the plural i of ye “who”, as noted above.
As for the relative pronoun ye “who”, it was probably derived from ya “what” by analogy with animate vs. inanimate se and sa.
Other subordinate conjugations: In addition to relative pronouns i (indeclinable) and ya/ye (declinable), Quenya had a number of specialized subordinate conjugations such as yá “when” mentioned above: yá hríve mene, ringa ná “when winter comes, it is cold” (VT49/23). One conjunction of particular interest is sa “that”, which seems to have a more general function:
- merin sa haryalye alasse nó vanyalyë Ambarello “I hope that you have happiness before you pass from the world” (Merin Sentence).
- [ᴹQ.] ni qente (sa) e·tule “I said (that) he was coming” (PE22/119).
The second example is from QVS of the late 1940s, and in later syntax might be qenten sa tulis. In QVS Tolkien said “In reporting or stating a clause is introduced by sa or nil” rather than i, but this is our only clue to its distinction from i. In the small number of examples where it appears, sa is used only as a subordinate conjunction and not as a relative pronoun. The conjunction sa also seems to follow the same rule for tense as the conjunction i, with the tense of the subordinate verb relative to that of the main verb. The English paraphrasing “I said that he came” (past, past) would be expressed in Quenya as qenten sa tulis (past, aorist).
Neo-Quenya: Given the various options discussed above, here is how I recommend forming subordinate clauses in Quenya.
- The indeclinable pronoun i (personal and neuter, singular and plural) is used when the relative pronoun is the subject of the subordinate clause: cenin i atan i túle noa “I see the man who came yesterday”; cenin i atani i túler noa “I see the men who came yesterday”.
- The declinable pronoun ya is used when the relative pronoun is declined into a noun case or is the object of a preposition: istan i osto yasse i atan mare “I know the city where [in which] the man lives”; istan i ostor yassen i atan amárie “I know the cities where [in which] the man has lived”.
- The personal relative pronoun ye (declinable, plural i) is used instead of ya when the referent of the subordinate clause is a person: istan i elda yenen i nér parne “I know the elf from whom the man learned”; istan i eldar inen i nér parne “I know the elves from whom the man learned”.
- Where the relative pronoun is the object of the subordinate clause, indeclinable i can be used: cenin i atan i cenil “I see the man whom you see”; cenin i aldar i cenil “I see the trees that you see”.
- I suspect it is also possible to use ye/ya as objects of the clause, but there are no attested examples and this syntax may be archaic (remnants of the classical accusative case): cenin i atan ye cenil “I see the man whom you see”; cenin i aldar yar cenil “I see the trees that you see”.
- The basic subordinate conjunction is i, which is used when the subordinate clause has no referent in the main clause: merin i hirilyes “I hope that you find it”. Optionally, sa may be used instead: merin sa hirilyes, and sa is required in a few special cases, most notably reported speech.
If the subordinate clause follows the main clause, the tense of the subordinate verb is relative to that of the main verb so: istan i tulis “I know that he comes” (aorist/aorist) and istan i túles “I know that he came” (aorist/past) but sinten i tulis “I knew that he came” (past/aorist) and sinten i túles “I knew that he had come” (past/past), where in the last example the past of the subordinate clause is further in the past than the main clause, equivalent to English’s use of the pluperfect.
It is less clear what would happen if the subordinate clause preceded the main clause. My intuition is that in that circumstance, the subordinate clause would be declined into the tense of the time frame under discussion, not relative to the main clause: i atan i túle náne hlaiwa “the man who came was sick” (past/past) vs. i atan i utúliéne náne hlaiwa “the man who had come was sick” (pluperfect/past).
This assumption is based on Tolkien’s phrasing: “the tense inside the clause depends on that of the first verb, the time of which becomes the present of the second verb” (PE22/118), which seem to imply that the word order matters. The examples given in QVS are also analogous to the behavior of the infinitive, which likewise derives its tense from the main verb and must follow it. However, it is possible my assumption is wrong, and that the tense in the subordinate clause is always relative to the main clause and thus remains indeterminate until the main clause is expressed.