zompist bboard

You could make them voiceless nasals.

Sotho has *MP *MB > p_h p_>.

Ideas:
Glottalisation/pharyngealisation of plosives followed by rhinoglottophilia.
C > /h/ > /h̃/
/h/ + vowel > /h/ + nasal vowel, followed by the transfer of nasalisation to the next consonant

Didn't Basque have something like *VnV > *Vh̃V > VV̯n?

Does this set of changes look legit?

pʰ > ʰp > xɸ, x _[C#]
tʰ > ʰt > xθ, θ _[C#]
kʰ > ʰk > xː > x, xk V_V

It does to me. As I understand it, preaspiration isn't very common, but I think those changes are still plausible.

I'm curious about the assymetry between ʰp > x and ʰt > θ in _[C#]. Any particular reason?

It looks wrong to me for the same reason.

I didn't even notice that. Oops.

KathTheDragon wrote:I'm curious about the assymetry between ʰp > x and ʰt > θ in _[C#]. Any particular reason?

I had two thoughts in that regard:

1. I wanted /ɸ/ to be asymmetric in the group: where /θ x/ can occur in isolation (somewhat more frequently than these three sound changes suggest due to some later sound changes), /ɸ/ always occurs in the cluster /xɸ/.
2. I found acoustically /xɸ/ very easily elided into /x/ in these contexts, whereas /xθ/ seemed to favor /θ/--not very scientific, but it "felt right" to my tongue, if you will.

I've tested number 2 on myself and found it isn't true for me.

I want a pushchain of kʷ > k > ć > č > t > ṭ > pʲ > p > pʷ ... for a vertical vowel inventory. /pʲ/ i slabiodental and therefore has to be spread-lipped, which causes a similar sound to palatalization and is seen as the same by the speakers. ṭ is true dental. I dont see any reason why not to do thius since every indicular link in the chain is attested and there would be very few coalescences (/pʷ/ was rare) but IM not sure mouth shape can change as fluidly as does tongue poisition.

Zaarin wrote:Does this set of changes look legit?

pʰ > ʰp > xɸ, x _[C#]
tʰ > ʰt > xθ, θ _[C#]
kʰ > ʰk > xː > x, xk V_V

Preaspiration to /x/ is reasonable, but intermediate *xp *xt fricativizing further seems odd. Clusters like these seem to often resist even fairly general fricativization. Compare Proto-Germanic: *pt *kt > *ft *xt and not **fθ **xθ, or modern Greek: φθ χθ: pʰtʰ kʰtʰ > fθ xθ > ft xt.

Is it plausible to split a uvular series off from a velar series? If so, how?

StrangerCoug wrote:Is it plausible to split a uvular series off from a velar series? If so, how?

Sure - back velar consonants adjacent to back or more specifically low back vowels, then do something that removes the conditioning environment, e.g. merging low back and low front vowels.

Tropylium wrote:

Zaarin wrote:Does this set of changes look legit?

pʰ > ʰp > xɸ, x _[C#]
tʰ > ʰt > xθ, θ _[C#]
kʰ > ʰk > xː > x, xk V_V

Preaspiration to /x/ is reasonable, but intermediate *xp *xt fricativizing further seems odd. Clusters like these seem to often resist even fairly general fricativization. Compare Proto-Germanic: *pt *kt > *ft *xt and not **fθ **xθ, or modern Greek: φθ χθ: pʰtʰ kʰtʰ > fθ xθ > ft xt.

Hmm, that is a fair point.

Would it be possible to have sound changes first result in /s̺ z̺/ > /ʃ ʒ/, and then have /ʒ dz/ > /z/, so that the most common sibilants in the language are /ʃ/ and /z/?

Knit Tie wrote:Would it be possible to have sound changes first result in /s̺ z̺/ > /ʃ ʒ/, and then have /ʒ dz/ > /z/, so that the most common sibilants in the language are /ʃ/ and /z/?

s z > ʃ ʒ, no problem.
dz > z, no problem

I am, however, a little skeptical of ʒ > z but not ʃ > s.

Is there a separate/ s/ that survives, or is this language entirely without plain /s/? I think Hungarian has/ S z/ as the most common but also has /Z s/ in comparable quantities.

/Z/>/z/ moving alone would be most likely if it shifted to a more open articulation than the voiceless . C.f. d turning dental in some langs but leaving t behind.

Hungarian has plenty of /s/ but not so much /ʒ/. EDIT: Or at least that's my impression. I'm pretty sure about the /s/ part but a bit more uncertain about /ʒ/.

Soap wrote:Is there a separate/ s/ that survives, or is this language entirely without plain /s/? I think Hungarian has/ S z/ as the most common but also has /Z s/ in comparable quantities.

/Z/>/z/ moving alone would be most likely if it shifted to a more open articulation than the voiceless . C.f. d turning dental in some langs but leaving t behind.

The proto-language had /t tˤ d dˤ s sˤ z zˤ/ which became /t ts d dz ʃ s ʒ z/ later on. So yeah, there's a plain /s/, it's just rarer than the /ʃ/ and considered marked by the native speakers. And funnily enough, the /z/ really is very strongly dental in the standard dialect.

Actually, speaking of coronals, what are the people's favourite reasonably-bland coronal consonant systems? I've been stumped as to what I want my conlang's to be in the end and so I think I'd just give somebody here the sincerest form of flattery.

In terms of general phonology, there's a strong voicing distinction in both plosives and fricatives and historical presence of pharyngealised coronals that I'm trying to eliminate through diachronic changes. In terms of particular phonology, as mentoned above, there're /t tˤ d dˤ s sˤ z zˤ ʃ ʃˤ ʒ ʒˤ ɾ ɾˤ l lˤ/ historically, which become /t ts d dz ʃ s ʒ z ʃ ʃ ʃ ʃ ɾ ʒ l l/ later on in the current version, respectively, though I don't quite like how the result sounds.

Welp, my last question seems to have been to vague, so how about this one:

I want to make Bantu, specifically Swahili, influence to cause development of this post-nasal fortition & voicing:

N(f v p) → mb
Nt → nd
N(s z ts) → ndz
N(ʃ ʒ j) → ɲdʒ, with [dʒ] not being otherwise found anywhere else
N(k g x ɣ) → ŋg
Nw → ŋgw
N(l ɾ) → ndr, which is the only three-consonant cluster allowed in the language

Also due to Bantu influence, the nasal clusters above will simplify to just nasals when preceded by another nasal cluster in the next syllable, e.g.

/imbindi/ → [imbini]
/anrinʃu/ → [andriɲdʒu] → [andriɲu]
/kanɣanwu/ → [kaŋgaŋgwu] → [kaŋgaŋu]

And the resulting /ɲ/ and /ŋ/ are not phonemic, but merely allophones of /m/ or /n/ in the complex environment described above.

Does that look good?

How do these changes look?

ts tɬ tɕ→ s ɬ ɕ (but ts' tɬ' tɕ' preserved)
(later)
p’ k’ q’ → pf’ kx’ qχ’ → f’ x’ χ’
f' → f

Assuming an initial inventory without ejective fricatives?

I like the first line, and I'm hard-pressed to come up with a reason to reject the rest.