Reconstruction talk:Proto-Finnic/niittü

@Tropylium: Contrary to your edit, this must be a *-u noun, not a participial form, for three reasons:

1. If it were a participle, it would have originally been *niit'üt. The original passive II participle must have been *-ttUt, despite what Hakulinen claims. Many dialects show final gemination in this ending, and South Ostrobothnian data shows forms like otettuen.

Indeed, forms without a final *-t must be an areal Northern Finnic innovation, for they are only found in Finnish, Ingrian and Karelian, not even in Veps, despite *-nut losing its final -t in -nu (!).

2. Nothing of the sort *niit'üt is found in any descendant.

3. The verb is not an action noun, but rather a deverbal entirely comparable to *juttu and *laulu. As *juttu fundamentally means "what is talked" and *laulu "what is sung", *niittü would have meant "what is mowed" (with a general imperfective aspect), which fits better semantically anyway than "what has been mowed".

85.76.64.127 14:58, 18 January 2023 (UTC) 85.76.64.127 14:58, 18 January 2023 (UTC)

I would follow the view of e.g. Laanest 1975 that endings pointing to *-ttut have been, on the contrary, analogically formed after the example of the active past participle *-nut. This readily explains the presence of *-t, while there is no reason why it should be elsewhere lost without any trace. Of course this analogy looks very widespread and could have existed as a variant already in common Finnic. (There may have been something else going on too: Eastern Votic has for some reason -tuɢ (and also -nuɢ), with /k/ and not /t/.)

The type juttu does not mean 'what is talked' etc., but fundamentally rather 'act of talking'. This distinction is a somewhat vague for intransitive verbs, but quite clear looking at other transitive verbs. E.g. *pestäk (“to wash”) → pesu is 'act of washing' and not 'that which is washed'; *maksadak (“to pay”) → maksu is 'act of paying' → 'payment, means of paying' and not 'that which is paid, merchandise' (bearing also in mind that, contrary to e.g. English pay, in Finnic this takes traded object as its primary object and the means of payment as its secondary object).

Lastly, in a verb like this that already ends in *-tta-, the distinction between *-u and -(t)tu will be invisible; but other derivatives of this sort like tuttu from *tuntedak clearly require the participle to provide the extra *-t- (and they, too, show no sign of an additional final **-t, I think due to fossilizing early enough to escape the *-(t)tu × *-nut → *-(t)tut analogy). --Tropylium (talk) 18:05, 18 January 2023 (UTC)

The loss of the final *-t would of course also be analogy, mainly by *-u derivatives, which would have been reinforced by the fact that the participle of this type is not that often seen in inflected forms and mainly exists in the nominative singular as part of the periphrastic passive perfect and pluperfect forms. As for tuttu, it appears to only be found in Finnish, Ingrian (→ Votic, only in Luzhitsy) and Karelian, so it is a moot point (the form *tuttaba is older though, and definitely an old participle: *tuttu could be by analogy). 85.76.64.108 19:02, 18 January 2023 (UTC)

To elaborate, the final *-t may have been lost either through regular change (in some regions, basically most of Finnish, c.f. Kettunen #32). This only leaves a gap for Ingrian, Karelian and many eastern dialects (those that have -nut or -nt: honestly, for the latter all bets are off, an example could be *-tut > *-tt > *-t and then reintroduction of *-u). If one believes in the theory that those have spread from the west, an early loss during the spread would explain it. 85.76.64.108 20:11, 18 January 2023 (UTC)

Even further, we see -o in some dialects of Estonian for the sense "meadow", e.g. in the Risti and Vaivara dialects, the latter known for preserving final *-o. This suggests a variant *niitto, and likewise *niittU is found in the sense "mowing" in at least Jögöperä Votic. Given that there is common -O/-U variation for deverbal derivatives like this in Finnic, I would very much argue the evidence weighs more towards an U derivative than a participle form. As to the participle question, the older books ignore the recent evidence, and I would argue that the forms showing a final *-k as you mentioned are evidence for an elided final consonant, which based on other evidence can well be reasoned to have been a *t. Lest we forget that this lets us connect the II participles etymologically, which aligns with the I participles showing such a pattern more clearly. A brave idea would be to say that *-ttut itself is a contraction of *-tt-nut, rather than a common *-ut present in both endings. 85.76.139.37 22:23, 27 February 2023 (UTC)