Reconstruction talk:Proto-Balto-Slavic/ṓˀstei

Accent

@Kwékwlos Where did you get the accent from? Lithuanian and Slavic give no information, Lithuanian in particular because all root verbs have mobile accent, which is not always inherited. The Latvian broken tone ˆ is a reflex of a mobile pattern, which implies accent on the infinitive ending. —Rua (mew) 13:15, 4 May 2019 (UTC)

@Rua Hmm... it was months ago after generalizing the fact that PIE disyllabic verbs never have the accent on the second syllable. I wasn't aware of such changes in Balto-Slavic, so I assigned the accent on the first syllable. Does it mean that it will be unmarked or transferred to the second syllable? Kwékwlos (talk) 13:18, 4 May 2019 (UTC)

Regarding mobility in general, and the accent placement that results from it, the rules are relatively simple:

1. Fixed-accent stems never have the accent on any ending, at all.
2. Mobile-accent stems have the accent on the ending in some forms, and on the first syllable of the stem in others.
3. The forms in which mobile-accent stems have ending accent happen to include all the lemma forms, so the nominative singular and the infinitive. A caveat applies in consonant stems, where the nominative singular has no ending (or at least, not one that is its own syllable).
4. The accent in Lithuanian and Slavic is further modified by De Saussure's law, Meillet's law, Dybo's law and Ivšić's law. You have to take that into account when comparing accents across the languages. The operation of these sound laws allows the possibility that all languages agree, and yet none of them preserve the original accent.

The accent on the verb is a more complicated matter in PBS, in particular because none of the languages preserve the accent intact in all verbs. As I mentioned, Lithuanian makes all root verbs like this one mobile, so that the original distinction is lost. Slavic, too, has generalised mobility across some stems, namely those ending in an obstruent (including this one). Of the non-obstruent verbs in Slavic, Jasanoff remarks that there seems to be a correlation between a verb being thematic in PIE and mobile accent in Slavic, whereas verbs that appear to have been originally athematic more often have fixed accent in Slavic. But this is not a hard rule. The fact that there's a lot of levelling, and so many exceptions, makes it difficult to make hard rules for Balto-Slavic verbs. —Rua (mew) 13:35, 4 May 2019 (UTC)

Also, the relationship between mobility in PIE and that of later PBS is very complicated, so you can't take PIE and assume the accent didn't move or that it remained fixed. The existence of countless mobile o- and ā-stems in Balto-Slavic attest to that; such nouns couldn't even be mobile in PIE, yet they are in BS. Rather, the trigger for mobility seems to have been forms with a word-final accent.

In verbs it gets even more complicated, since we see lots of verbs that arise from PIE simple thematic presents, which should have had fixed initial accent in all cases and thus no word-final accent. Yet somehow most of them end up mobile. Another complicating factor is that verbs are actually an amalgamation of a whole variety of paradigms, which each might have had different accentuation in PIE: present, aorist, infinitive, participles etc. It appears that these generally all adapted to the accentuation of the present stem in Balto-Slavic, so that a mobile present would always have a mobile aorist and infinitive. This is necessary to account for the different accentuation we see in the infinitive, which comes from an ending-accented case form of a PIE noun in *-tis. If this kind of accent adaptation in the infinitive hadn't occurred, we'd expect all infinitives to have final accent, whereas in reality they have final accent when the present is mobile, stem accent when the present is fixed. —Rua (mew) 13:45, 4 May 2019 (UTC)