Zompist Bboard Again

Zaarin wrote: ↑Fri Aug 16, 2019 12:53 pm

Max1461 wrote: ↑Fri Aug 16, 2019 7:31 am How plausible is debaucalization of voiceless coda stops to [ʔ] only in post-vocalic position? So [pot] > [poʔ] but [post] > [post].

Many dialects of English, my own for instance, do exactly that. (Well, with the caveat that /t/ is usually debuccalized after a resonant, too.)

In careful speech in my English dialect this is precisely what happens (whereas in very careful speech /t/ does not debuccalize at times by itself). On the other hand, in everyday speech, final /st/ frequently is reduced to [s], and before a vowel or approximant /t/ by itself is frequently lost.

Hi all,

This is my first post here on ZBB. I'm looking for ideas for how /tʰ/ can change in different environments. I currently have /tʰ/ changing to /r̥/ when word-initial, word-final, and intervocalically

/tʰan natʰ atʰa/ -> /r̥an nar̥ ar̥a/

But my current issue concerns how /tʰ/ will change in consonant clusters, both within word-roots and across word boundaries when words collapse together. The phonological inventory is:

/p t k pʰ tʰ kʰ pʲ tʲ kʲ pʷ tʷ kʷ/
/s t͡s/
/m n ŋ/
/r l ɣ/

Below is a list of all the clusters I'm currently dealing with. I currently have /pʰ/ turning to /f~ɸ/ more or less everywhere, and /kʰ/ leniting to /x/ which itself becomes /ɣ/ intervocalically, and /h/ elsewhere. Would this suggest having /tʰ/turn to /s/ in most of the cluster environments?

Any help in this regard would be most appreciated.


[Edit: P.S. I'm disinclined to blithely lose the aspiration word-finally, as this would/could create push-/pull-chains with the unaspirated series. But I'm open to suggestions.]

/tʰ/ > /s/ in most cluster environments would indeed make a lot of sense if /pʰ/ and /kʰ/ also usually become fricatives. I'd personally tend to include a few exceptions though, for instance /tʰ/ > /t/ when adjacent to at least /s/ but quite possibly all coronal consonants, and/or /tʰ/ > /r̥/ > /r/ when followed by a voiced consonant (and maybe have /rl/ > /ll/ after that).

A nice additional possibility for your list of clusters might be fortition of /ɣ/ > /ɡ/ > /k/ in clusters before any other changes, which might result in something like e.g. /tʰɣ ɣtʰ/ > /sk ks/.

Has /tʰ/ > /s/ ever happened withut an intermediary non-sibilant fricative?

Also welcome holbuzvala! Have some pickles and tea!

mèþru wrote: ↑Fri Aug 23, 2019 12:57 pmHas /tʰ/ > /s/ ever happened withut an intermediary non-sibilant fricative?

Isn't that kind of what happened to Hebrew words in Yiddish?

EDIT: Also Finnish: root vete- 'water' → vesi

Hebrew never had /tʰ/, and Hebrew did have at one point non-sibilant coronal fricatives

But it did have /t/, which changed to /s/ in Yiddish without an intermediary non-sibilant fricative, didn't it? Why should it make a difference whether it had /t/ as opposed to /tʰ/ or whether it had non-sibilant coronal fricatives?

I had always assumed that the change in Hebrew was /t/ > /θ/, and then /θ/ > t in Sephardic pronunciation while /θ/ > s in Ashkenazi. Yiddish, I dont know about, but I assume it borrowed the pronunciation of Ashkenazi. Possibly both under influence of surrounding languages (no /θ/ in German or most European langs). Do we know whether the original /t/ sound was dental? if it was dental it almost certainly would have shifted to /θ/.

The Finnish change had a palatal intermediate, i think. There is /t/ > /ts/ > /s/ all over the place in German, though.

Vijay wrote: ↑Fri Aug 23, 2019 2:54 pm

mèþru wrote: ↑Fri Aug 23, 2019 12:57 pmHas /tʰ/ > /s/ ever happened withut an intermediary non-sibilant fricative?

Isn't that kind of what happened to Hebrew words in Yiddish?

EDIT: Also Finnish: root vete- 'water' → vesi

This is a general process of *t -> s before /i/ and likely went through /ts/.

mèþru wrote: ↑Fri Aug 23, 2019 12:57 pm Has /tʰ/ > /s/ ever happened withut an intermediary non-sibilant fricative?

10.3.5.2 Proto-New Caledonia to Jawé t̪ t̪ʰ t tʰ → t tʰ c s
30.3.1.1.3 Proto-Gbe to Proto-Fon {ts,tʰ} {dz,dʱ} → s z
30.3.1.1.5 Proto-Gbe to Proto-Phla-Pherá {ts,tʰ} {dz,dʱ} → s z
39.1 Proto-Tanoan to Jimez {tʰ,tsʰ} → ʃ

Pabappa wrote: ↑Fri Aug 23, 2019 4:19 pm I had always assumed that the change in Hebrew was /t/ > /θ/, and then /θ/ > t in Sephardic pronunciation while /θ/ > s in Ashkenazi. Yiddish, I dont know about, but I assume it borrowed the pronunciation of Ashkenazi. Possibly both under influence of surrounding languages (no /θ/ in German or most European langs). Do we know whether the original /t/ sound was dental? if it was dental it almost certainly would have shifted to /θ/.

The Finnish change had a palatal intermediate, i think. There is /t/ > /ts/ > /s/ all over the place in German, though.

Yes, the Hebrew > Yiddish t > s change was via the begadkefat [θ]. This is why it doesn't occur initially (e.g., "Torah" is Toyre, not *Soyre.)

(Also, remember that [even ignoring the primarily German base] Yiddish is not a linear descendant of spoken Biblical Hebrew in the sense of inheriting sound changes from it normally; the Hebrew/Aramaic component of Yiddish is Hebrew/Aramaic words pronounced with an Ashkenazi accent beginning a millennium and a half after Hebrew ceased to be a living spoken language, and borrowed into Yiddish--they're loan words).

dhok wrote: ↑Fri Aug 23, 2019 6:02 pm

Vijay wrote: ↑Fri Aug 23, 2019 2:54 pm

mèþru wrote: ↑Fri Aug 23, 2019 12:57 pmHas /tʰ/ > /s/ ever happened withut an intermediary non-sibilant fricative?

Isn't that kind of what happened to Hebrew words in Yiddish?

EDIT: Also Finnish: root vete- 'water' → vesi

This is a general process of *t -> s before /i/ and likely went through /ts/.

But /ts/ isn't a fricative and mèþru didn't say anything about needing an unconditional change.

mèþru wrote: ↑Fri Aug 23, 2019 3:07 pm Hebrew never had /tʰ/, and Hebrew did have at one point non-sibilant coronal fricatives

There's good reason to believe the non-emphatic unvoiced plosives in Phoenician and Akkadian were aspirated; I wouldn't be surprised the same was true in Biblical Hebrew.

@cedh, methru, Vijay, Pabappa, dhok, Nortaneous, Whimemsz

Thanks for the feedback! If I have /tʰ/ go to /s/, it'll probably be via a dental fricative to it doesn't create a push chain with the already extant /ts/. But, I'll also allow the voiceless trill in some cluster environments because I think that'd be fun.

I have a rule where phonation spreads forwards in consonant clusters for all clusters, e.g. akda→akta, iɹkə→iɹgə. Is this plausible? If so, then what should happen to consonants like /ʔ/, /ɹ/, /m/, /l/, which may not have a voiced or voiceless counterpart in the language? (I know that I could just posit allophonic [ɹ̥], [m̥] etc. in clusters, but a cluster like /ɣʔ/ is still problematic since /ʔ/ doesn’t really have a voiced counterpart.)

bradrn wrote: ↑Mon Aug 26, 2019 5:35 am I have a rule where phonation spreads forwards in consonant clusters for all clusters, e.g. akda→akta, iɹkə→iɹgə. Is this plausible? If so, then what should happen to consonants like /ʔ/, /ɹ/, /m/, /l/, which may not have a voiced or voiceless counterpart in the language? (I know that I could just posit allophonic [ɹ̥], [m̥] etc. in clusters, but a cluster like /ɣʔ/ is still problematic since /ʔ/ doesn’t really have a voiced counterpart.)

It's perfectly realistic for the voicing/devoicing rule to only apply to non-glottal obstruents or something, leaving /ʔ m n ɹ l/ unchanged. edit: if you want, you could also just delete /ʔ/ in clusters.

Max1461 wrote: ↑Mon Aug 26, 2019 6:56 am It's perfectly realistic for the voicing/devoicing rule to only apply to non-glottal obstruents or something, leaving /ʔ m n ɹ l/ unchanged

Well, that’s what I’m doing now, but I wanted to know if there’s any alternatives.

edit: if you want, you could also just delete /ʔ/ in clusters.

I quite like that solution! I might end up doing that.

In Poswa historically all liquids were devoiced after a voiceless stop (almost always /p/). This wouldnt make any difference, even to native speakers, since it never produced any new phonemes. howewver it interplays with a later sound change where /l r/ > /ž b/ in most environments but shift to /š p/ instead if they are voiceless. thus, i wouldn t worry too much about this rule unless you either want a very detailed writeup or are planning to make use of allophonic variation to derive later phonemic shifts.

What's more likely in a word-final situation?

/(V) r p/ ---> /(V) r/

or

/(V) r p/ ---> /(V) p/

Raphael wrote: ↑Thu Aug 29, 2019 5:49 am What's more likely in a word-final situation?

/(V) r p/ ---> /(V) r/

or

/(V) r p/ ---> /(V) p/

Did you mean

Vrp# > Vr

vs

Vrp# > Vp?