Q. talat-stem verbs grammar.
The talat-stem verbs are derived from primitive verbal roots of the form √TALAT, but are classified differently from the basic verbs because phonological forces resulted in very different conjugations. In the end, their conjugations ended up mostly the same as the half-strong verbs, whereas the most irregular talat-stem verbs tended to abandon their original inflections in favor of the weak conjugations. As such, it is often convenient to categorize talat-stem verbs among the derived verbs, even though strictly speaking they are based directly on a primitive verbal root.
The talat-stem verbs are rare, since most primitive verbal roots were biconsonantal (√KAT). The best known verb in this class is talta- “to slip, slide down, fall”, and Tolkien generally used this verb as the example of inflections for this class (PE17/186; PE22/164). In “modern” Quenya this verb’s conjugation is:
According to notes from 1964 (PE17/186), its ancient past tense was regularly formed from the root √TALAT via nasal-infixion: ✶talantē > talante. Likewise its perfect was ✶atā̆latie > ataltie as the result of the Quenya syncope. The ancient aorist/stem form would have been ✶talati > talte and its present ✶talatā > talta. These were probably confused with larger classes of weak and half strong verbs, so that the talat-stem present came to be perceived as its stem/aorist, while its present was modified to match the weak present conjugation: taltea.
However, in notes from the early 1950s Tolkien said a was the final vowel in talat-stem even in Common Eldarin:
The vowel ă was used (a) in basic verbs of dissyllabic kalat-form: as talat-, talta- “slips down”; peles-, pelsa- “fences round” (PE22/134).
Furthermore, in Late Notes on Verb Structure from 1969, the scenario for the present tense development seems to be reversed. The ancient present for talat-stems was ✶tal[a]taya > taltea whereas the formatives had (archaic?) ia-presents like ortia (PE22/164), so that the formative presents were the ones that adopted weak conjugations like ortea (maybe, see PE22/164 note #103). The talat-stems inflections from 1969 were otherwise the same as 1964 except the perfect was derived directly from the past: atalantie. The net result in both 1964 and 1969 (and the 1940s, see below) was that these talat-stem conjugations effectively merged with those of the half-strong formatives.
Unlike the formatives whose stems always ended in ya or ta, the talat-stems could have a variety of consonant clusters, resulting in some peculiar past tenses: carpa- → past carampe (PE17/126) or [ᴹQ.] sulpa- → past sulumpe (PE22/115). These abnormal inflections were generally reformed to match weak conjugations, for example past sulumpe became sulpane. Thus the talat-stems probably disappeared as a distinct verb class, either merging with the conjugations of half-strong verbs, or abandoning all irregularities to become ordinary weak verbs.
Conceptual Development: Triconsonantal roots were even rarer in the 1910s and 1920s than they were in Tolkien’s later writing, so finding verbs that might be of this class in the Early Period is very difficult. However, the verb ᴹQ. talta- < ᴹ√TALAT did appear in The Etymologies from the 1930s, and Tolkien explicitly mentioned the existence of talat-stem verbs in Quendian & Common Eldarin Verbal Structure in 1940s (PE22/98). He described them in more detail in the Quenya Verbal System (QVS) written shortly afterwards, lumping them together with the half-strong formative verbs (PE22/113). As he described their conjugations in QVS, the talat-stems were essentially identical to the formatives except that they had a greater tendency to adopt weak past tenses (PE22/114-115).
One factor in the preservation of talat-stem conjugations was its use in one of the names for Númenor: Atalantë “Downfallen” (S/281). Tolkien was clearly pleased that this name resembled “Atlantis”, and his retention of the nasal-infixed past tense talante for talta- was almost certainly motivated in part by his desire to retain this name.