Zompist Bboard Again

Would it be sensible for /a/ to be realised as [ɒ] next to [ħ] in a three-vowel system? I’d say it seems plausible to me, but pharyngeals do weird things, so I’m not sure. (Apparently Arabic has [æ] in this position, but it has [æ] next to most other consonants too.)

bradrn wrote: ↑Fri May 12, 2023 2:48 am Would it be sensible for /a/ to be realised as [ɒ] next to [ħ] in a three-vowel system? I’d say it seems plausible to me, but pharyngeals do weird things, so I’m not sure. (Apparently Arabic has [æ] in this position, but it has [æ] next to most other consonants too.)

Counterintuitively, pharyngeals never (afaik) back vowels, even though /ʕ/ = /ɑ̯/. They consistently lower vowels, and sometimes also centralise them (e.g. in Ditidaht, Nuu-chah-nulth, Tsakhur, Udi), but they don't back or round them. Fronting effects on /a/ are attested in Salishan as well as Arabic so it seems to be crosslinguistic, for which reason I'd go with [æ] if not just [a].

Darren wrote: ↑Fri May 12, 2023 3:55 am

bradrn wrote: ↑Fri May 12, 2023 2:48 am Would it be sensible for /a/ to be realised as [ɒ] next to [ħ] in a three-vowel system? I’d say it seems plausible to me, but pharyngeals do weird things, so I’m not sure. (Apparently Arabic has [æ] in this position, but it has [æ] next to most other consonants too.)

Counterintuitively, pharyngeals never (afaik) back vowels, even though /ʕ/ = /ɑ̯/. They consistently lower vowels, and sometimes also centralise them (e.g. in Ditidaht, Nuu-chah-nulth, Tsakhur, Udi), but they don't back or round them. Fronting effects on /a/ are attested in Salishan as well as Arabic so it seems to be crosslinguistic, for which reason I'd go with [æ] if not just [a].

Emphatics force a backed /a/ in Arabic though—see DaabiT “officer”.

Darren wrote: ↑Fri May 12, 2023 3:55 am

bradrn wrote: ↑Fri May 12, 2023 2:48 am Would it be sensible for /a/ to be realised as [ɒ] next to [ħ] in a three-vowel system? I’d say it seems plausible to me, but pharyngeals do weird things, so I’m not sure. (Apparently Arabic has [æ] in this position, but it has [æ] next to most other consonants too.)

Counterintuitively, pharyngeals never (afaik) back vowels, even though /ʕ/ = /ɑ̯/.

Yes, this is similar to what I remember.

They consistently lower vowels, and sometimes also centralise them (e.g. in Ditidaht, Nuu-chah-nulth, Tsakhur, Udi)

This sounds interesting; do you have any references I can read through?

Man in Space wrote: ↑Fri May 12, 2023 5:35 am Emphatics force a backed /a/ in Arabic though—see DaabiT “officer”.

So does this mean emphatics behave differently to laryngeals? Makes sense — I recall hearing that emphatics are really uvularised.

bradrn wrote: ↑Fri May 12, 2023 6:52 am

They consistently lower vowels, and sometimes also centralise them (e.g. in Ditidaht, Nuu-chah-nulth, Tsakhur, Udi)

This sounds interesting; do you have any references I can read through?

There's this one on Ditidaht, this one for the Northeast Caucasian languages and this one for Nuu-chah-nulth. Interestingly, in Ditidaht (and the caucasian examples, and in arabic too) uvulars and pharyngeals pattern separately, but in Nuu-cha-nulth they pattern together. The former seems to be more common, but the latter could be taken as indirect precedent for your idea.

Darren wrote: ↑Fri May 12, 2023 7:24 am

bradrn wrote: ↑Fri May 12, 2023 6:52 am

They consistently lower vowels, and sometimes also centralise them (e.g. in Ditidaht, Nuu-chah-nulth, Tsakhur, Udi)

This sounds interesting; do you have any references I can read through?

There's this one on Ditidaht, this one for the Northeast Caucasian languages and this one for Nuu-chah-nulth. Interestingly, in Ditidaht (and the caucasian examples, and in arabic too) uvulars and pharyngeals pattern separately, but in Nuu-cha-nulth they pattern together. The former seems to be more common, but the latter could be taken as indirect precedent for your idea.

Very interesting indeed! Though I don’t see how they pattern together in Nuu-cha-nulth — their effects seem pretty separate to me. If anything, Ditidaht might give a stronger precedent for my idea, since it seems to have pharyngeals lowering [ɐ] to [ä]. It even has an overlap between /a/ and /o/ before coda [ʕ]!

Regarding cases and adpositions:

If I have a locative case but I want to be more precise then adpositions are employed, which seems perfectly natural. Say tree.LOC could mean 'in/on/at/near the tree' then the usual meaning would be 'in' as in "amongst the branches" like a bird as that is most likely to be what is meant. However, lets say I mean 'inside the tree' as in literally within the trunk - then the adposition 'in' + the locative could mean 'inside', no? The problem I have here is that most languages I look at seem to use the genitive or dative with adpositions, rather than a case relating to location. Does an adposition + locative make sense?

Jonlang wrote: ↑Tue May 16, 2023 4:24 pm Does an adposition + locative make sense?

I see no reason to not employ it. I would even use multiple locative cases (e.g. plain locative, ablative, and allative) with prepositions to make them more specific.

Travis B. wrote: ↑Tue May 16, 2023 5:30 pm

Jonlang wrote: ↑Tue May 16, 2023 4:24 pm Does an adposition + locative make sense?

I see no reason to not employ it. I would even use multiple locative cases (e.g. plain locative, ablative, and allative) with prepositions to make them more specific.

That's the idea!

Do you have a true adposition class of words? Because if derived from nouns (as these things often are), you could use a noun meaning, say, "(the) inside", mark that with the locative, and mark "tree" with the genetive (so it's "at the tree's inside / at the inside of the tree").


JAL

Then of course there is the traditional Germanic way of doing things where there's both a "dative" and "accusative" case, and both are used with some adpositions, each of which has a somewhat different (but related) meaning.

Travis B. wrote: ↑Wed May 17, 2023 8:46 amThen of course there is the traditional Germanic way of doing things where there's both a "dative" and "accusative" case, and both are used with some adpositions, each of which has a somewhat different (but related) meaning.

Though that tradition has died out in almost all Germanic languages save German and perhaps regional dialects of other languages. In German it mostly signifies movement vs. stationary (at least for pps like "über"). In Dutch, that same distinction is made by changing the word order from P NP to NP P. Note sure about North Germanic languages.


JAL

jal wrote: ↑Wed May 17, 2023 9:22 am In Dutch, that same distinction is made by changing the word order from P NP to NP P.

That sounds really interesting… any idea how it arose? I’m not sure I’ve ever heard of any other language where the same word can be used as both preposition and postposition.

bradrn wrote: ↑Wed May 17, 2023 9:26 amThat sounds really interesting… any idea how it arose? I’m not sure I’ve ever heard of any other language where the same word can be used as both preposition and postposition.

I think German does it too, with certain prepositions. An example in Dutch:

Hij liep in de kamer
He walk.PST in the room
He walked inside the room

Hij liep de kamer in
He walk.PST the room in
He walked into the room

I think in the latter case, the German word order is the same (i.e. postponed adposition), but like English it uses a directional adposition (hinein).

Note I'm not entirel sure what the contemporary anaysis of this construction is, it might well be analyzed as an adverb in the latter case, using a resultative construction (e.g. cf. "Hij sloeg de lamp kapot" "He broke the lamp" (litt. "He hit the lamp broken").


JAL

No, German has:

Er ging im Zimmer. 'He walked in the room.' (dative) vs. Er ging in das Zimmer (hinein). 'He walked into the room.' (accusative). The hinein is optional; it will most likely become mandatory in future German when the case system collapses (confusion of accusative and dative is common enough in colloquial German for a saying Mir und mich verwechsle nicht 'Don't confuse me (dat.) and me (acc.).').

Ah, thanks for setting me straight.


JAL

Be my guest. While Dutch and German are quite close to each other, they sometimes do things differently.

jal wrote: ↑Wed May 17, 2023 10:41 am I think in the latter case, the German word order is the same (i.e. postponed adposition), but like English it uses a directional adposition (hinein).

Second opinion on German: certain Northern German sociolects have a combination of (a) obligatory reduction of weak definite (i.e. non-anaphoric) articles and (b) either phonological or morphological neutralization of dative and accusative forms in at least in some contexts. My personal dialect has phonological neutralization for masculine forms but close friends of mine that I grew up with had a full-blown morphological neutralization in the same context.

That means for me in this specific context 'rein' is obligatory. The next two sentences can be differentiated by the postpositional thing-y but it's not obligatory for what used to be dative case.

Er ging i[ŋ] Kreißsaal rein.
Er ging in den Kreißsaal rein.
he went in DEF.M.ACC maternity.room into.it
'He entered the maternity room.'

Er ging i[ŋ] Kreißsaal (drin).
Er ging im Kreißsaal (drin).
he went in DEF.M.DAT maternity.room (in.it)
'He walked inside the maternity room.'

In a way, this creates a contrast between a cirvumposition-like pattern P NP P and a prepositional/circumpositional pattern P NP (P). This is famously parodied in the TV show Dittsche where phrases such as the following occur. I think these are clearly a parody, at least I haven't heard it IRL.

auf[ə] Flasche auf
auf der Flasche auf
on DEF.F.DAT bottle on
'on top of the bottle'

The special thing here is that both instances of the adposition in the circumpositional pattern are identical.

bradrn wrote: ↑Wed May 17, 2023 9:26 am

jal wrote: ↑Wed May 17, 2023 9:22 am In Dutch, that same distinction is made by changing the word order from P NP to NP P.

That sounds really interesting… any idea how it arose? I’m not sure I’ve ever heard of any other language where the same word can be used as both preposition and postposition.

Well, there's Latin cum 'with', mostly a preposition but a postposition with personal pronouns. I thought tenus was also both, but quick searches don't back me up on that.

Sanskrit ā has different meanings depending on case and whether it's a preposition or a postposition.

Was reading today about the training of scribes in Akkadian society, and the lessons they had to copy and learn from, and I had a thought:

Divide at least a handful of verbs according to if somebody actually does them...ie, "[the king] Sargon built the gagon wall." (gagon was a community scholars used to think was composed of nuns...but its actually more complicated than that)

Of course Sargon's not going to personally build the wall himself...unless this is a movie and he has to do everything. So it would be

"[the king] Sargon built-3P the gagon wall." (Sargon oversaw the wall's construction, or gave the command to "make it so")

...and in time, maybe this could generalize to also include recipients or beneficiaries of the action/object...such as:

"The children built-3P the house." (it was built for the children)
vs
"The children built-1P the house." (it was built by the children)