Quick Diachronics Challenge

[sangi39]

So is Click going to make the next challenge?

I'm working on it right now!

YAY!

[Click]

Here's the challenge, enjoy!

• L1 ɔˈtʃajər
  L2 ˈkéɪ̯ɦeɪ̯
  L3 ˈxɛːʃɪ
  L4 ˈʃtʃíːtê
  L5 ʃkɪ̯eːˈtɑ̏ː
  L6 ˈscɪ̯ɐd
  L6½ ʃəˈtsɛd
  L7 ˈsɪ̯ɔd
  L8 ʔaçóɪ̯ɾ
  L9 ǂʰɒ́ɪ̯ɾ

Language 1 is the westernmost language and is spoken along the west coast of the continent.

On the east coast there's a large bay. Languages 6, 6½ and 7 are spoken along its northern shore, and Languages 8 and 9 are found on the opposite side of the bay. Languages 4 and 5 are located in a river valley to the west of the bay.

Finally, Languages 2 and 3 are spoken in the rugged mountains between the west coast and the river valley.

[Creyeditor]

Just my first guess: *tʃa.ˈkʰe.tʷia

[Click]

That's not too bad for a first guess, but I feel you've made some wrong assumptions while reconstructing.
In general, it is a good idea to group related languages together and reconstruct their ancestor first because many languages derive same phonemes from different sources.

[sangi39]

Spoiler:

L1 ɔˈtʃajər
L2 ˈkéɪ̯ɦeɪ̯
L3 ˈxɛːʃɪ
L4 ˈʃtʃíːtê
L5 ʃkɪ̯eːˈtɑ̏ː
L6 ˈscɪ̯ɐd
L7 ˈsɪ̯ɔd
L8 ʔaçóɪ̯ɾ
L9 ǂʰɒ́ɪ̯ɾ

Proto-2-3 probably begins with a velar plosie and has a medial *ʃ (which voiced intervocalically in L2 and then backed to *ɦ). The first vowel could either have been a diphthong or a long vowel, but it seems likely that the second vowel was long, either *i: or *e:. So something like *ˈke(:/i)ʃi: or *ˈke(:/i)ʃe:

Proto-4-5 seems to be characterised by a level of palatalisation, with L4 going one step further, taking what could be an original *ʃkɪ̯ to *ʃtʃ. I'd guess that the original first vowel was *e: and the medial consonant was just *t. Assuming the diacritics are similar to Slovene and Serbo-Croation, then both final syllables have low or falling pitch but the height of the vowel is uncertain, as is backness. Maybe something like *a:, giving something like *ʃkɪ̯e:tà:. The *ʃ could either be the result of palatisation from the ɪ̯, or a German-like pre-plosive shift.

Proto-6-7 seems somewhat easier to reconstruct, possibly as something like ˈscɪ̯ɑd, with L7 losing the *c.

Proto-8-9 seems like a fun one. The palatal click in L9 likely comes from some sort of cluster simplification, possibly *ʔcʰ, with the *cʰ element becoming a fricative in L8. I'd guess Proto-8-9 was something like *ʔacʰɒ́ɪ̯ɾ.

That gives:

1 ɔˈtʃajər
2-3 *ˈke(:/i)ʃi: or *ˈke(:/i)ʃe:
4-5 *ʃkɪ̯e:tà:.
6-7 ˈscɪ̯ɑd
8-9 *ʔacʰɒ́ɪ̯ɾ.

Now, there are two ways I can see this going, either 1 and 8-9 both added a preceding vowel, or the vowel was there originally, lost in the "central" dialects and all 8-9 did was add the glottal stop (assuming it wasn't original as well).

Now, the fricative *x in L2 does point to the velar plosive in Proto-2-3 being aspirated, since, as far as I'm aware, it's pretty uncommon for plain plosives to becoming fricatives word initially, which points to there being an *s in the original proto-language.

As an inital guess, I'd say the original proto-form was something like *aˈske:te. The Initial vowel was lost in 2-3-4-5-6-7, but retained in the peripheral languages. All but 2-3 have a palatalised element, suggesting that palatalisation was probably an areal feature. The aspiration in 8-9 and 2-3 comes from the original *s becoming *h and then either dropping out (at it seems to have done in 1) or aspirating the following *k~*c. The original long vowel *e: becomes a diphthong basically everywhere, varying between *ei and *ie. The medial *t often voices between vowels, becoming *d, then *ɾ and finally *r in 1 (with the resulting *air becoming ajər) or palatalising in 2-3. The final vowel either lengthens or drops out altogether depending on the branch, only being retained in the "core" (2,3,4,5).

So yeah, my guess would be *aˈske:te

[qwed117]

Basic phoneme property guess:
I think there were 4 "consonants".
C1 was some glottal consonant that later became clicklike
C2 was a shibilant or a velar (likely unvoiced)
C3 was another velar (likely voiced)
C4 was a dental or alveolar of some quality. (Likely voiced)

[Ashtăr Balynestjăr]

• L1 ɔˈtʃajər
  L2 ˈkéɪ̯ɦeɪ̯
  L3 ˈxɛːʃɪ
  L4 ˈʃtʃíːtê
  L5 ʃkɪ̯eːˈtɑ̏ː
  L6 ˈscɪ̯ɐd
  L7 ˈsɪ̯ɔd
  L8 ʔaçóɪ̯ɾ
  L9 ǂʰɒ́ɪ̯ɾ

My guess is something like this:

ˈtʃkájta

• Proto-Southwest *əˈtʃkájɾ
  • L1 ɔˈtʃajər
  • L8 ʔaçóɪ̯ɾ
  • L9 ǂʰɒ́ɪ̯ɾ

• Proto-Inland *skéjtʲa
  • Proto-Mountain *ˈhkéːʃe
    • L2 ˈkéɪ̯ɦeɪ̯
    • L3 ˈxɛːʃɪ
  • Proto-Northeast ˈskɪ̯éta
    • Proto-River-Valley *ˈʃkɪ̯étàː
      • L4 ˈʃtʃíːtê
      • L5 ʃkɪ̯eːˈtɑ̏ː
    • Proto-North-Bay *ˈscɪ̯ad
      • L6 ˈscɪ̯ɐd
      • L7 ˈsɪ̯ɔd

[ixals]

I haven't done a challenge in a while so I thought I'd do this one! I reconstructed it before looking at any of the other guesses.

West Coast:
L1: /ɔˈtʃajər/
> */ɔˈtʃajər/

Mountains:
L2: /ˈkéɪ̯ɦeɪ̯/ < /keːɟʰeː/
L3: /ˈxɛːʃɪ/ < /ˈkʰeːçe/
> */'keːceː/

River Valley:
L4 /ˈʃtʃíːtê/ < /ˈʃce̝ː.tɛ̂/
L5 /ʃkɪ̯eːˈtɑ̏ː/ < /ʃkʲeːˈtȁ/
*/ˈʃceː.ta(X)/ ~ /ˈskʲeː.ta(X)/

Northern Shore:
L6: /ˈscɪ̯ɐd/
L7: /ˈsɪ̯ɔd/ < /ˈscɪ̯ɔd/
> */ˈsc(e:)d/

Southern Shore:
L8: /ʔaçóɪ̯ɾ/ < /ʔacʰóɪ̯ɾ/
L9: /ǂʰɒ́ɪ̯ɾ/ < /əcʰɒ́ɪ̯ɾ/
> */(ʔ)acʰɒ́ɪ̯ɾ/

So my guess would be:

*/(a/ə)ˈskʲeː.d(e/ə)/

[Click]

Thank you for your interest in the challenge!

@sangi39:

Congrats on recognizing that accent development in Languages 4 and 5 is the same as in what you call Serbo-Croatian! As you've found out, in Language 4 accent evolved approximately as it did in Serbo-Croatian whereas Language 5 preserves the original accent.
I'll give you a hint with /ʃ/ in Proto-4-5 - the cause of palatalization is a certain segment that's also responsible for palatalization in Slavic languages.

The reconstructed Proto-2-3, Proto-6-7 and Proto-8-9 forms are very close to the actual ones but not quite correct.
Regarding Proto-8-9, your explanation of click development is spot on.

You've failed to identify two higher-level language groupings, though.


@qwed117:

I've already told you on chat that there are only two original consonants. Looking forward to a more elaborate guess!


@Ashtăr Balynestjăr:

You've been overzealous with grouping languages. Low-level subgroups have been reconstructed well enough (two are nearly correct, the other two not as much), but your trying to bring together disparate families has definitely led you astray a bit.


@ixals:

The reconstructions of Southern Shore and River Valley languages are very close to the actual ones but not quite correct. Northern Shore is also close enough, but you definitely should devote more attention to Mountains languages.

You've also missed two higher-level language groupings just like sangi did.


ANNOUNCEMENT

I've introduced an another language that's located between Language 6 and Language 7, we'll call it language 6½. The L6½ cognate is ʃəˈtsɛd.
I've edited the original post to reflect this change.

[sangi39]

@sangi39:

Congrats on recognizing that accent development in Languages 4 and 5 is the same as in Serbo-Croatian! As you've found out, in Language 4 accent evolved approximately as it did in Serbo-Croatian whereas Language 5 preserves the original accent.
I'll give you a hint with /ʃ/ in Proto-4-5 - the cause of palatalization is a certain segment that's also responsible for palatalization in Slavic languages.

The reconstructed Proto-2-3, Proto-6-7 and Proto-8-9 forms are very close to the actual ones but not quite correct.
Regarding Proto-8-9, your explanation of click development is spot on.

You've failed to identify two higher-level language groupings, though.

Spoiler:

Hmmm... the only sound change I can think of in the history of the Slavic languages that led to *ʃk is the Ruki sound change (although there's also progressive palatalisation, but reconstructing a preceding short *i and extending the change to affect *s as well might be unlikely.

I had considered higher level groupings, but I found them somewhat difficult to pin down. 4-5-6-7 seemed like the most likely, with a possible proto-word resembling *'skɪ̯eta and then potentially 1-8-9 with a proto-word resembling *a'caiɾ but with one being on the western extreme and another being on the eastern extreme I really wasn't comfortable grouping them together. 2-3 definitely stands alone as a separate branch, though.

EDIT: The only other grouping I could think of was to some sort of Proto-456789, but the branching there seems hard to pin down.

[qwed117]

*əsˈkɛːdɛ mebbe?
(Yes I know I'm 100% copying sangi. )

[Click]

@sangi39:

You're thinking in the wrong direction when it comes to development of /ʃ/. What I was trying to hint was a certain phoneme that palatalized the consonant to its left and was later lost in unstressed position.

The higher-level groupings are dead on though.


@qwed117:

The proto-language has a basic /a e i o u/ vowel system with diphthongs.


I might give some hints later.

[sangi39]

@sangi39:

You're thinking in the wrong direction when it comes to development of /ʃ/. What I was trying to hint was a certain phoneme that palatalized the consonant to its left and was later lost in unstressed position.

The higher-level groupings are dead on though.

Spoiler:

Huh, seems the option I thought was less likely was more likely Hmmm... Could be an initial *e but I'm not sure.

Another thing I've just noticed, which might be interesting, is that L2, L3, L4 and L5 have long vowels (or a diphthong in the case of L2)
in a stressed open syllable. This makes me think that in the original proto-language, the middle vowel was short, but was allophonically long in open, stressed syllables.

Wait, the proto-word isn't something like *aj'ske:.tej, is it?

Spoiler:

Or as'ke:.tej

[qwed117]

Slavic yers come from u and i right?
*uskede

[Click]

@qwed117:

You're right about yers, but the matter is that in this language the yer caused palatalization and that's unlikely to come from /u/.


@sangi39:

Your observation concerning the middle vowel is correct, but your new guess is less accurate than the one before.

[qwed117]

@qwed117:

You're right about yers, but the matter is that in this language the yer caused palatalization and that's unlikely to come from /u/.


@sangi39:

Your observation concerning the middle vowel is correct, but your new guess is less accurate than the one before.

Hmm, I thought both i and u degraded to /ɨ/ before palatization???
(Click told me palatization is regressive in the family. I hate you.)

*iskeda

[ixals]

Well if the proto-language had /a e i o u/, it didn't have extra short vowels so I guess /i u/ are used in this case.

/usikeːda/ > /usikeːda/ > /ə́sʲəkʲeːda/ > /asʲkʲeːda/?

That wouldn't explain the difference between /ʃəˈtsɛd/ though. Maybe it depended on the language which "yer" was weak?

/usikeːda/ > /ə́sʲəkʲeːda/ > /asʲkʲeːda/ > e.g. L1 /ɔˈtʃajər/
/usikeːda/ > /əsʲə́kʲeːda/ > /sʲakʲeːda/ > e.g. L6½ /ʃəˈtsɛd/

I'm confused now haha, good job!

Edit: I also changed the last vowel to /a/ because if this is going down the Slavic root, I thought /e/ would palatalise the preceding consonant which is does not in any daughter language.

[sangi39]

@qwed117:

You're right about yers, but the matter is that in this language the yer caused palatalization and that's unlikely to come from /u/.


@sangi39:

Your observation concerning the middle vowel is correct, but your new guess is less accurate than the one before.

Spoiler:

i'ske.te, as a compromise guess, then, with the initial *i becoming *ə in 189, remaining an extra-short *ĭ everywhere else,
causing palatalisation of the following *s in 45. The final *e would drop after causing intervocalic voicing in 189 and later on in 67,
lengthening in 2 (just because it's open) and 5 (because of the stress shift after lowering to *a before the stress shifted).

Spoiler:

6.5 is really messing with how I'd handle Proto-67. The initial ʃəts- makes it seem closer to 45, which I would have put down to cluster breaking or even metathesis, e.g. *ĭsc- > sĭc- > sĭtʃ- > sĭts- > ʃĭts- > ʃəts-.

[Click]

Guys, the Slavic thing is confined to Languages 4 and 5. In those languages, palatalization is also regressive (consonants palatalize in front of a front vowel, not after one).

HINT

There is no *s in the proto-word. has it origins in an allophonic voiceless vowel that became a fricative.

[qwed117]

Guys, the Slavic thing is confined to Languages 4 and 5. In those languages, palatalization is also regressive (consonants palatalize in front of a front vowel, not after one).

HINT

There is no in the proto-word. has it origins in an allophonic voiceless vowel that became a fricative.

I hate you
*i'keti (no reason to have a d tbh)