Linguistic Miscellany Thread

[Otto Kretschmer]

What was the accnetuation system like in Proto Indo Iranian? Was it a similar pitch accent like in PIE and Ancient Greek and Sanskrit?

How did it change in daughter languages i.e. Prakrits, Old Persian, Avestan etc

[WeepingElf]

What was the accnetuation system like in Proto Indo Iranian? Was it a similar pitch accent like in PIE and Ancient Greek and Sanskrit?

As Vedic largely preserves the PIE accent intact, we can logically conclude that it was the same in Proto-Indo-Iranian.

[hwhatting]

What was the accnetuation system like in Proto Indo Iranian? Was it a similar pitch accent like in PIE and Ancient Greek and Sanskrit?

Seeing how Vedic Sanscrit is normally assumed to have conserved the PIE accentuation system, the most reasonable assumption is that PIIr, as the intermediate stage, did as well. (EDIT: Ninja'd by elves )

How did it change in daughter languages i.e. Prakrits, Old Persian, Avestan etc

I know that later Sanscrit changed to a system with stress position conditioned by the length of the penultimate and antepenultimate syllable. Maybe Vijay knows more about the development in the Indic languages. Modern Persian has word-final stress, similar to the Turkic languages, which seems to be an areal feature; it also lost the old IE final syllables, which seems to speak for a similar situation to Brythonic, i.e. penultimate stress turning into final stress after losing the final syllable. But there are probably people who know the history better.

[Otto Kretschmer]

I am asking again about strong ultimate stress How might it change a language?

We have Caesar sancti urbis Romae Augustus

In normal Latin it was ['kaisar 'sankti 'urbis 'rome au'gustus]

With ultimate stress we have [kai'sar san'kti ur'bis ro'me augu'stus]

We get vowel reduction so [kɘi'sar sɘn'kti ər'bis rɘ'me əugə'stus]

Can we expect to see the same word shortening with dropping of unstressed syllabes as with word initial stress? Unlike the ultimat syllabes which carry only grammatical information, initial ones carry a lot of actual meaning of the word in IE languages.

[Pabappa]

i'd expect that if this happened, it would soon resolve itself back to having stress on the stem, or towards a stressless language (Ive seen French described as stressless). Stressing the least grammatically salient part of a word is unnatural. imagine if in english we said "THE crew OF four HAD gone TO shore". i just dont see it happening.

[Rounin Ryuuji]

We could probably arrive at similar changes if we dropped off most or all of the case endings (or deleted the vowels out of them, or if they're long vowels erode them to schwa), though we might still mark case with umlaut or consonant mutation, or both, note:

Augustus [aʊ'gʊs̺.t̪ʊs̺] > Ogustz [o'gʊʃt̪͡s̪] (nominative singular)
Augustum [aʊ'gʊs̺.t̪ʊm] > Ogust [o'gʊʃt̪] (accusative singular)
Augustī [aʊ'gʊs̺.t̪iː] > Ogystz̡e [o'gyɕ.c͡ɕə] (genitive singular, nominative, accusative, and vocative plural)
Augustōrum [aʊ.gʊs̺'t̪oː.rʊm] | Ogostuor [o.goʃ.t̪wor] (genitive plural, this form is weird, and might end up Oguoster, reflecting quantitiative metathesis)
Augustō [aʊ'gʊs̺.t̪oː] > Ogoste [o'goʃ.t̪ə] (dative and ablative singular)
Augustīs [aʊ'gʊs̺.t̪iːs̺] > Ogystz̡es [o'gyɕ.c͡ɕəs̺] (dative and ablative plural; the ablative will probably have been lost)
Augustō [aʊ'gʊs̺.t̪oː] > Ogoste [o'goʃ.t̪ə] (ablative)
Auguste [aʊ'gʊs̺.t̪e] > Ogøstz̡ [o'gøɕc͡ɕ] (vocative)

This example ends up an ugly consonant snarl, but there's no reason it would have to do that.

[Ephraim]

i'd expect that if this happened, it would soon resolve itself back to having stress on the stem, or towards a stressless language (Ive seen French described as stressless). Stressing the least grammatically salient part of a word is unnatural. imagine if in english we said "THE crew OF four HAD gone TO shore". i just dont see it happening.

I’m nor sure if I understand what you mean here. Plenty of languages can have the main word stress on inflectional affixes at least some of the time.

Here’s the combination the features ”Prefixing vs. Suffixing in Inflectional Morphology” and ”Fixed Stress Locations” on WALS:
https://wals.info/combinations/26A_14A#3/52.80/40.78

There’s not a lot of data for this combination, but there are both some languages that are strongly suffixing with fixed ultimate stress and one language that is strongly prefixing with fixed initial stress. The most common combination, strongly suffixing languages with no fixed stress, includes languages that can have word stress on inflectional endings, some of the time.

Of course, too much reduction of unstressed syllables may be problematic if the root syllables of content words are typically unstressed.

[hwhatting]

I think the question here is not whether stress can be regularly on inflectional affixes (it can, Turkic also has this), the question is whether the stress on the endings is so strong that it leads to vowel deletions in the stem. This is not the case in Turkic. In Russian, there are ending-stressed word classes resp. word classes that are ending-stressed at least in part of the paradigm, but then the vowel reduction in Russian mostly results in mergers, not in the deletion of vowels. I at least know of no case where (1) stress led to wide-ranging vowel syncope / apocope and (2) endings were generally kept while stem syllables were generally lost.

[Otto Kretschmer]

I think the question here is not whether stress can be regularly on inflectional affixes (it can, Turkic also has this), the question is whether the stress on the endings is so strong that it leads to vowel deletions in the stem. This is not the case in Turkic. In Russian, there are ending-stressed word classes resp. word classes that are ending-stressed at least in part of the paradigm, but then the vowel reduction in Russian mostly results in mergers, not in the deletion of vowels. I at least know of no case where (1) stress led to wide-ranging vowel syncope / apocope and (2) endings were generally kept while stem syllables were generally lost.

How strong is the stress? Finnish for example has an initial stress but it is light and does not result in vowel reduction

I am just thinking of a situation opposite to what happened in French in which strong initial stress resulted in reduction of unstressed syllabes and diversification of vowel inventory on (previously) stressed syllabes (the same thing also happened in Germanic languages) - rather than having strongly stressed root, we'd have strongly stressed inflectional affixes May we expect preserved inflectional endings and reduced root in such a scenario?

[hwhatting]

How strong is the stress? Finnish for example has an initial stress but it is light and does not result in vowel reduction

I don't see how your question relates to my remark? How strong is which stress?

[Moose-tache]

I am asking again about strong ultimate stress How might it change a language?

We have Caesar sancti urbis Romae Augustus

In normal Latin it was ['kaisar 'sankti 'urbis 'rome au'gustus]

With ultimate stress we have [kai'sar san'kti ur'bis ro'me augu'stus]

We get vowel reduction so [kɘi'sar sɘn'kti ər'bis rɘ'me əugə'stus]

Can we expect to see the same word shortening with dropping of unstressed syllabes as with word initial stress? Unlike the ultimat syllabes which carry only grammatical information, initial ones carry a lot of actual meaning of the word in IE languages.

The closest thing I can think of to this is Vietic. In Austro-Asiatic, strong final stress caused initial syllables to weaken in various languages, and in Vietnamese they are gone almost without a trace (they may have some influence on tone, I honestly don't remember). But these are still semantically rich syllables that are receiving final stress. In your example a lot of words would just end up identical because they're just suffixes. Plenty of language shift stress to the end of the stem, which could be a good compromise.

[keenir]

Quick question...

I know that English used to have Strong and Weak forms of various nouns and verbs...of which, we have goose - geese and run - ran as remnants (fossils, yes?)

My question is, at least in Modern English, can they be counted as Infixes? {not in the sense of for a book or a formal paper}

ie,
Prefix - rerun
Infix - ran
Suffix - running

[WeepingElf]

Quick question...

I know that English used to have Strong and Weak forms of various nouns and verbs...of which, we have goose - geese and run - ran as remnants (fossils, yes?)

Both are fossils of alternations which once were productive, but from different time stages. Goose/geese is an example of umlaut: the vowel is fronted before a former /i/ in the following syllable. This is a Germanic innovation. Run/ran is an example of ablaut: this is far older, inherited from Proto-Indo-European, originally an alternation between /e/, /o/ and zero, for reasons which are not clearly understood yet.

[Pabappa]

i mean you could write an academic paper making your case for why ablaut & umlaut should be considered infixes, but it isn't the term we traditionally use, because the root vowel is being replaced rather than being supplemented. you could argue that /i:/ is underlyingly /u:i/ or something and, like i said, you could write a paper that way, but it'd be an argument for the sake of making an argument and wouldnt really add to anyone's knowledge of the way the language works today.

[keenir]

and wouldnt really add to anyone's knowledge of the way the language works today.

true...the question stems from two reasons:
1. the question occurred to me, and I thought better ask just to be sure.
2. was trying to think of how to explain Strong & Weak to my dad, after we watched Professor McWhorter's Languages A to Z on The Great Courses.

but from different time stages.

, but it isn't the term we traditionally use, because the root vowel is being replaced rather than being supplemented.

Thank you both.

[Nortaneous]

The closest thing I can think of to this is Vietic. In Austro-Asiatic, strong final stress caused initial syllables to weaken in various languages, and in Vietnamese they are gone almost without a trace (they may have some influence on tone,

Initials, not tone - the presence of an Old Vietnamese presyllable can condition lenition:
rắn 'snake' ~ Ruc /pəsíːɲ/
ven '(river)bank, near' < OC *tə.pˤe[n]

See Gong 2017. But this lenition is absent in some dialects.

[linguistcat]

I've been trying to learn about reduplication because I'd like to use it in my conlang, but I'm wondering if there much difference cross linguistically for root vs stem reduplication. This is an important distinction in my conlang's case because, being based on Old Japanese, verbs and verbal adjectives have 6 stems that they use in different circumstances and syllables were strictly (C)V, but some roots were consonant based. This is early in the separation so the conlang still would mostly follow OJ phonological rules. So, I'd have more variation reduplicating stems, but things would be simpler reduplicating the root and throwing in a connecting vowel to keep things CV.

[Rounin Ryuuji]

Given what I've experienced tinkering with Japonic linguistics, verbs tend to join using the ren'yōkei, so if a system of verbal reduplication were to develop out of Old Japanese specifically, I would expect the first occurrence of the verb to be in either the ren'yōkei or the izenkei, so the verb saku would either reduplicate saki-saku or sake-saku, and probably the former, given that older fossilised compound verbs use the ren'yōkei; alternately, it might use the te-form, from Old Japanese yielding sakitesaku (or a modern saitesaku, though I don't think the loss of certain occurrences of medial k was underway yet, and wouldn't necessarily have happened in every daughter language). The te-form is also useful if you want to negate one form but not the other, though I'm not sure if Old Japanese had a negative te-form, but it would be easy to innovate one, presumably with yodan -azite/-anite, nidan and ichidan -izite/-inite or -ezite/-enite, treating the negative suffixes -zu and -nu as further conjugable (since -nai wouldn't have developed yet).

You could also just have nai and -nakute/-naide develop in parallel, if you want it.

[linguistcat]

Thank you for the family specific response. That's about what I was thinking for reduplicating the root (and using the ren'youkei to connect them). I hadn't considered -te form. I think I would go with the ren'youkei then.

[vlad]

I am asking again about strong ultimate stress How might it change a language?

We have Caesar sancti urbis Romae Augustus

In normal Latin it was ['kaisar 'sankti 'urbis 'rome au'gustus]

With ultimate stress we have [kai'sar san'kti ur'bis ro'me augu'stus]

We get vowel reduction so [kɘi'sar sɘn'kti ər'bis rɘ'me əugə'stus]

Can we expect to see the same word shortening with dropping of unstressed syllabes as with word initial stress? Unlike the ultimat syllabes which carry only grammatical information, initial ones carry a lot of actual meaning of the word in IE languages.

Final stress and deletion of unstressed non-final vowels is found in a number of Oaxacan languages. E.g.

Chatino (Zenzontepec) [kesu] "net", [kʷiʔju] "flea"
Chatino (Teotepec) [ksu] "milk", [kʷʔju] "flea"

Mixtec [juwi] "mat", [nᵈɨkɨ̃] "seed"
Amuzgo [tswe] "mat", [nᵗkẽ] "seed"