Tolkien’s terminology regarding weak verbs is somewhat inconsistent. Sometimes he used this term to describe a specific verb class, and other times he used it to contrast weak versus strong past tenses within a given verb class. In one place, he used it to describe an entire category of verb classes, some with weak conjugations and some without (PE22/113). In this lexicon, I use the term “weak verbs” for the specific group of verbs that use weak conjugations, where (1) the final vowel in the verb stem is a and (2) the past tense is exclusively formed with the suffix -ne added to the stem. Even here the terminology is a bit ambiguous, because (a) weak conjugations tended to invade other verb classes and (b) some “weak” verbs may have variant “strong” inflections (especially irregular past tenses) resembling those of other verb classes.
There are two groups of verbs that most frequently make use of weak conjugations:
- Causative verbs produced by adding the causative suffixes -tā or -yā to a verbal stem: tulta- “to send for, fetch, summon, (lit.) cause to come”.
- Non-verbal derived verbs produced by adding a suffix to some non-verbal stem: ninquita- “to whiten”.
The conjugations for these two groups are close enough that I treat them as a single class (but see below for other possibilities). The weak verb conjugations for the five common Quenya verb tenses are:
- Stem: tulta- “to fetch, send for, make come”.
- Aorist: tulta “fetches”.
- Present: tultea “is fetching”.
- Past: tultane “fetched”.
- Perfect: utultie “has fetched” (or utultanie).
- Future: tultuva “will fetch” (or tultauva).
These conjugations are mostly based on Late Notes on Verb Structure from 1969 (PE22/157, 164), but the future forms incorporate other sources (PE17/77).
Of the above, the weak perfect form is the least certain. Forms like utultie are the norm for causative verbs in Tolkien’s 1969 notes, and are the closest fit to how perfects were formed in Common Eldarin and in other verb classes. In these notes, Tolkien gave the “natural” development of ta-causative perfects as orortā̆iye > oro[r]taye > -ëe > orortie (PE22/159), and there are several examples fitting this pattern:
- Stem caita → perfect acaitie (PE22/159).
- Stem orta → perfect orortie (PE22/159).
- Stem tulta → perfect utultie (PE22/164).
In other examples, the weak perfects of causatives seemed to be formed using the past tense:
- Past caine → perfect acainie (PE22/157).
- Past ortane → perfect ortanie (PE17/77).
- Past tulta- → perfect utultanie (PE22/157).
This also seems to be the pattern for non-causative weak verbs:
This inconsistency might be explained by the following quote from the Quendi and Eldar essay written around 1960 (WJ/366):
The form perfect avānie is regularly developed from *a-wāniiē, made in the prehistoric period from the older perfect form of this type *a-wāwiiē with intrusion of n from the past (the forms of past and perfect became progressively more closely associated in Quenya).
Thus perfect forms like utultanie are probably the result of “progressively more close association” of past and perfect tenses in Quenya. Given the variations of perfect patterns within individual verbs, it seems likely Tolkien was vacillating between these two ideas. Of the two patterns, the formation of perfects from past tenses seems to be the earlier pattern, seen in causatives and other weak verbs all the way back in the Quenya Verbal System of the 1940s:
The proper future forms of weak verbs are also somewhat ambiguous. Most examples of derived verbs ending in -a have futures like hentuva, with -uva replacing the final a. But in Late Notes on Verb Structure from 1969 Tolkien had some alternate future inflections specifically for causative verbs: either ortúva with long ú (PE22/159) or ortauva with -uva added to the final a (PE22/164). In the currently published corpus, these -úva and -auva futures for weak verbs appear only in notes from 1969 so they may be a very late expirements, but we could also be missing unpublished examples from the intervening years.
Conceptual Development: Using past tenses as a guide, there are signs of weak verbs dating all the way back to the Qenya Lexicon of the 1910s. There are verbs with past tenses formed with just the suffix -ne as in ᴱQ. lampa- “beat, hit”, past lampane (QL/51). However, as noted by Thorsten Renk in his article on the Quenya Past Tense (QPT), many derived verbs show varying past tenses of different formations, both weak and strong:
- ᴱQ. avalta- “strip”: weak past avaltane vs. strong past avalante (QL/34).
- ᴱQ. kanda- “blaze”: weak past kandane vs. strong past kande (QL/47).
- ᴱQ. peanta- “enjoin”: weak past peantane vs. strong past peane (QL/72).
Since we don’t know much about Early Qenya verb classes, it is hard to say what is going on. But it seems likely there were a variety of verb classes in the 1910s which had a tendency to regularize into a weak conjugation with suffixal -ne. Similar variations can be seen in the past tenses of “non-stem” [derived] verbs in the Early Qenya Grammar (PE14/58) which may represent early verb classes.
The verb classes in their later form emerged by the 1940s. In Quendian & Common Eldarin Verbal Structure from the mid-1940s, Tolkien discussed the “Formation of other Verbal Stems”. He did not label any specific group as “weak verbs”, but he did say:
The true derivatives did not allow the intrusion of any element, and their ā vowel was originally always long. ortā “to raise”, ortāt(e) “raises it”, ortānē “lifted” (PE22/98).
This group was contrasted with formative verbs where inflections did intrude into the stem: ort-ă with past oro-n-t-ē. It seems fair to say these “true derivatives” were properly “weak verbs”. Tolkien went on to say:
This -tā was specially frequent in forming causatives from adjectives: *niñkwi “white, pale”: niñkwitā- “to whiten”. Causatives from verbs more often used yā: as tulyā- “send hither” (PE22/98).
This last statement was reversed in the Quenya Verbal System (QVS) written shortly afterwards, where -ya was favored for intransitive/formative verbs (PE22/115). QVS described “weak” verbs in more detail, but used the term loosely for all non-basic verbs. QVS listed two more specific verb classes that are a closer fit to how this entry labels weak verbs (PE22/114):
4) Causatives with suffixes -tā́, -yā́, made from all kinds of stems …
5) Derivatives of various kinds made from adjectival and noun stems, with or without suffix …
The conjugations of these two classes were similar, and both show weak past tenses with suffixal -ne (PE22/117):
- 4) Aorist orta, present imperfect ortalya, past ortane, perfect ortanie or (more usual) ortanelye, future ortáva, gerund ortáre.
- 5) Aorist ninqinta, present imperfect ninqintalya, past ninqintane, perfect ninqintanelye, future ninqintuva, gerund ninqintie.
Just like causatives versus other weak verbs in the 1960s (see above), there were some variations in the perfect and future forms in QVS between these two subgroups of weak verbs.
Neo-Quenya: Until recently, many Neo-Quenya authors treated all verbs ending in a as part of a single large class of verbs: the derived or “a-stem” verbs. Most older Neo-Quenya courses conjugated this unified class more or less like weak verbs as described above. Variations were generally assumed to be either irregular verbs or conceptual vacillations. The publication of PE22 in 2015 overturned our understanding of Quenya derived verbs, however, revealing that there were a lot more verb classes in Quenya than originally thought. Even now things are complicated because it is often difficult to determine precisely which class a given verb belongs to.
I recommended treating the weak conjugations above as the “default” conjugation for verbs whose stems end in -a, using other conjugations only where they are known, most notably for half-strong verbs. For simplicity, I recommended having causatives use the weak conjugations as well. A good argument can be made that the causatives should be their own verb class, but exactly what the specialized forms should be is unclear, since Tolkien himself was still changing his mind late in his life.
One late experiment seems to be the introduction of -úva or -auva futures for causative verbs, as mentioned above. For simplicity, I recommend ignoring this change, which complicates future forms. However, in the late 1960s Tolkien considered having causative perfects conjugated like most other derived verbs so that utultanie >> utultie. I do recommend adopting this change, since it makes perfects more regular. I personally generalize it to other weak verbs as well.
I’m less committed to this recommendation on perfects than I am to the recommendation on futures, however. It seems that from the 1940s up through at least the mid-1960s, conjugating weak perfects from the past tense was the normal pattern. A solid argument can be made for ignoring these late (and possibly tentative) changes to weak perfects and retaining the earlier pattern. My main reason for recommending forms like utultie is this is more or less the established practice in Neo-Quenya, and there enough vacillations on Tolkien’s part that it is hard to figure out what he was aiming at.