Quick Diachronics Challenge

[sangi39]

Alright, sorry again for the wait. Ideally, I'd like to have spent more time putting this together, but I didn't want to delay things even more. Hopefully this will be neither too easy nor too hard.

NW [ˈwĩːru] - N [əi̯nˈdʉu̯] - NE [ˌʝiːnʊˈlwaː]
W [jãĩ̯ˈroə̯] - C [ˈhei̯ndoː] - E [ɟeˈɲ̊ɔu̯]
SW [døsˈnœːry] - S [ˈzɔstr̩bi] - SE [ˈʑot͡ɕɛːf]

C = central, E = eastern, N = northern, S = southern, W = western

This 3x3 arrangement is meant to serve as a very rough "map" of where these nine related languages are currently spoken, although any historical migrations that may have taken place within this area have not been indicated.

Please let me know if there's anything you'd like me to clarify!

Spoiler:

As a preliminary guess... yeah, that's a tough one for a small sample of clearly related languages

NW and W seem more closely related to each other that to anything else. Same for N and C. N-C and W-NW then seem be most closely related to each other than anything else. Lets call this group "proto-Western", and it feels like C is fairly conservative on this front, with Proto-Western having something like [hei̯nˈdoː] or [hei̯nˈroː] (I'm assuming the stress is final here, but possibly wrongly, because of what looks like might have happen in the NE and E languages)

NE seem like an offshoot related to Proto-Western (forming a "Northern" group), with a proto-form of [heːnoˈroː] or [heːnoˈdoː]

The voiceless nasal in E would suggest that it was originally part of a cluster of a nasal followed by a voiceless sound, likely a palatal, which for now I'm just going to assume was [c], giving pre-E [ɟeˈɲcɔu̯] (and possibly older [ɟeɲVˈcɔu̯]

S and SE again look more closely related to each other than to anything else, with something like [ˈʑost͡ɕœvi] as the proto-form? (loss of the final fowel in SE followed by final consonant devoiced, while the vowel was retain in S with fortition of the [v], and some funkiness with the consonants)

SW really trips me it. It does seem most closely related to SE and S, possibly with an original [d] that lenited to [z], and the [y] appears to related to the [evi]~[ewi] of SE/S. But then the nasal is in the same position as it is the Northern group

I'm starting to wonder if the sibilant in SW and S might have come from older [ɕ] (which might then have come from the same sort of [ĩ̯] in W), meaning that Proto-Southern (SW, S, and SE) could have had [ˈɟoɕcœvi] from older [ˈɟoĩ̯cœvi] (from even older [ˈɟoɲcewi], or something like that). That would make E look like a Southern outlier, giving pre-Proto-Southern a proto form like [ˈɟeɲocewi], which also makes the palatalisation seem somewhat conditioned (maybe from older [denotewi]?)


So I'm going to guess:

Code: Select all

Proto-World > Macro-Northern > NE
                             > Northern Proper > C/N
                                               > NW/W
            > Macro-Southern > SW
                             > Southern Proper > E
                                               > S/SE

with [denotewi] as the proto-form

[Creyeditor]

My guess for the protoform is dwenstrib.

[zyma]

Thank you both for your guesses! I'm going to hold off on responding in any more detail for a bit, just in case anyone else might be interested in joining this round of the game.

[zyma]

Response to sangi39's first attempt:

Spoiler:

As a preliminary guess... yeah, that's a tough one for a small sample of clearly related languages

Ah, yeah, it seems that this has ended up being trickier than I intended. Of course, please let me know if there's anything that's not clear!

NW and W seem more closely related to each other that to anything else. Same for N and C.

I'm afraid that this is largely due to areal influence. NW-W and N-C are not valid first-level groupings.

N-C and W-NW then seem be most closely related to each other than anything else. Lets call this group "proto-Western", and it feels like C is fairly conservative on this front,

Although N-C-W-NW is not a valid grouping, I will say that C is not particularly conservative compared to its closest relatives.

with Proto-Western having something like [hei̯nˈdoː] or [hei̯nˈroː] (I'm assuming the stress is final here, but possibly wrongly, because of what looks like might have happen in the NE and E languages)

As a general note, the position of stress has shifted a few times in various languages over the course of the family's history.

NE seem like an offshoot related to Proto-Western (forming a "Northern" group), with a proto-form of [heːnoˈroː] or [heːnoˈdoː]

NW-N-NE-W-C is indeed a correct grouping! Your reconstructions are pretty far off, I'm sorry to say. However, you are at least on the right track with [-n-r-] in [heːnoˈroː].

The voiceless nasal in E would suggest that it was originally part of a cluster of a nasal followed by a voiceless sound, likely a palatal, which for now I'm just going to assume was [c], giving pre-E [ɟeˈɲcɔu̯] (and possibly older [ɟeɲVˈcɔu̯]

This is partially correct! It was originally part of a cluster consisting of a nasal and a voiceless palatal consonant. I'll say that I'm not going to be particularly strict about the definition of "palatal" here.

S and SE again look more closely related to each other than to anything else, with something like [ˈʑost͡ɕœvi] as the proto-form? (loss of the final fowel in SE followed by final consonant devoiced, while the vowel was retain in S with fortition of the [v], and some funkiness with the consonants)

S-SE is a valid first-level grouping! Your reconstruction isn't too far off. [ˈʑoCCV-vi] is correct, though there's a segment missing. You're right about vowel loss + devoicing in SE & fortition in S, but otherwise, I'd say that S is actually more conservative when it comes to the consonants.

SW really trips me it. It does seem most closely related to SE and S, possibly with an original [d] that lenited to [z],

SW is not most closely related to SE-S, unfortunately, and [d] is an innovation.

and the [y] appears to related to the [evi]~[ewi] of SE/S. But then the nasal is in the same position as it is the Northern group

I'm not sure if this is what you were thinking, but I can say that the front rounded vowels in SW do not originate from [e~i] being rounded by adjacent labial/labialized consonants. A segment is missing from [evi]~[ewi], and [e] is incorrect.

I'm starting to wonder if the sibilant in SW and S might have come from older [ɕ]

Assuming you mean [-s-], that's correct!

(which might then have come from the same sort of [ĩ̯] in W), meaning that Proto-Southern (SW, S, and SE) could have had [ˈɟoɕcœvi] from older [ˈɟoĩ̯cœvi] (from even older [ˈɟoɲcewi], or something like that).

SW-S-SE is not a valid grouping. I really like what you've come up with here ([ɲ] > [ĩ̯] > [ɕ]), but regrettably, it's not what I had in mind.

That would make E look like a Southern outlier, giving pre-Proto-Southern a proto form like [ˈɟeɲocewi], which also makes the palatalisation seem somewhat conditioned (maybe from older [denotewi]?)

E-SW-S-SE is another correct grouping! With regard to [ˈɟeɲocewi], [ɟ-] is correct, as is the initial stress. [-o-] and [-e-] are both present, but not in these positions. Generally speaking, I don't believe it will be helpful to think in terms of conditioned palatalization here.

So I'm going to guess:

Code: Select all

Proto-World > Macro-Northern > NE
                             > Northern Proper > C/N
                                               > NW/W
            > Macro-Southern > SW
                             > Southern Proper > E
                                               > S/SE

with [denotewi] as the proto-form

You're right about the primary split in the family being between "Macro-Northern" and "Macro-Southern", although out of the lower-level groupings, only S-SE is correct. As for [denotewi], the segments [-e-], [-n-], [-o-], [-w-], and [-i-] are present in the original proto-form, though generally not in these positions. [-wV#] is correct, however.

[zyma]

Response to Creyeditor's first attempt:

Spoiler:

My guess for the protoform is dwenstrib.

Not an unreasonable guess, I feel, though it's unfortunately quite far from what I had in mind, especially with regards to syllable structure and the number of syllables. The segments [-w-], [-e-], [-n-], [-r-], and [-i-] are all present, although not in this configuration.

[Creyeditor]

My second guess is tonerwi, though I had a look at Sangi's feedback.

[ɶʙ ɞʛ]

This one's kind of off the wall.

NW [ˈwĩːru] - N [əi̯nˈdʉu̯] - NE [ˌʝiːnʊˈlwaː]
W [jãĩ̯ˈroə̯] - C [ˈhei̯ndoː] - E [ɟeˈɲ̊ɔu̯]
SW [døsˈnœːry] - S [ˈzɔstr̩bi] - SE [ˈʑot͡ɕɛːf]

NW/N/NE/W/C: wi:ndo:~hwi:ndo:

NW: wi:ndo: > wĩ:do: > wĩ:ru
N: wi:ndo: > ɯi̯nˈdu: > əi̯nˈdʉu̯
NE: wi:ndo: > ji:nˈlwa: > ˌʝi:nʊˈlwa:
W: wi:ndo: > ji:nˈdo: > jai̯nˈdoə̯ > jãĩ̯ˈroə̯
C: hwi:ndo: > hɯi̯ndo: > ˈhei̯ndoː

E/SW/S/SE: djosnr̩w

E: djosnr̩w > ɟøçˈnɛw > ɟeˈɲ̊ɔu̯
SW: djosnr̩w > døsˈnɛru > døsˈnœːry
S: djosnr̩w > ˈd͡zɔstr̩v > ˈzɔstr̩bi
SE: djosnr̩w > ˈd͡ʑossɛrv > ˈʑot͡ɕɛːf

Overall protoform: ðys.nr̩w
Proto-N: ðys.nr̩w > θy:d.nr̩w~ðy:d.nr̩w > (h)wi:dnəw > (h)wi:ndo:
Proto-S: ðys.nr̩w > ðjos.nr̩w > djos.nr̩w

[sangi39]

I'll try to get a second guess in this evening. Had a bit of a busy week, haha

[zyma]

I'll try to get a second guess in this evening. Had a bit of a busy week, haha

No worries!

[sangi39]

I'll try to get a second guess in this evening. Had a bit of a busy week, haha

No worries!

Weeell I was wrong, and now I'm in Sheffield, lol

If you think you might want to go ahead and analyse second guesses, feel free Might be another week for me and don't want others missing out

[zyma]

I'll try to get a second guess in this evening. Had a bit of a busy week, haha

No worries!

Weeell I was wrong, and now I'm in Sheffield, lol

If you think you might want to go ahead and analyse second guesses, feel free Might be another week for me and don't want others missing out

I'm honestly fine with waiting. Especially since this game doesn't typically have very many players, I'd like to make sure that everyone who's interested in participating gets the chance to do so.

[Arayaz]

Very tentative guess:

Spoiler:

*ʝoinzwiɸ

Spoiler:

Code: Select all

      1   2   3   4   5   6
NW    Ø   wĩ  :   r   u
N     Ø   əi  n   d   ʉu  Ø
NE    ʝ   i:  nʊ  l   wa  :
W     j   ai  ~   r   oə  Ø
C     h   ei  n   d   o   :
E     ɟ   e   ɲ   Ø   ɔu  Ø
SW    [idk what to do for this one]
S     z   ɔ   st  r   bi  Ø
SE    ʑ   o   t   ɕ   ɛ:  f
*     ʝ   oi  n   z   wi  ɸ

Edit: I also have no idea how this game works, so if I did something wrong, tell me.

[zyma]

Response to Creyeditor's second attempt:

Spoiler:

My second guess is tonerwi, though I had a look at Sangi's feedback.

You're getting warmer, but the syllable structure and number of syllables are both still a bit off. [-o-] is indeed the nucleus of the initial syllable, and I'll say that [-n-r-w-] is correct as well. [-e-] and [-i-] are also present in the form that I have in mind, but [-t-] is not.

[zyma]

Response to ɶʙ ɞʛ's first attempt: 

Spoiler:

You're correct about the primary division in this family being between NW/N/NE/W/C and E/SW/S/SE. However, both of these branches can be further subdivided.

NW/N/NE/W/C: wi:ndo:~hwi:ndo:

I'll say that it shouldn't be necessary to reconstruct multiple forms (such as [wiːndoː~hwiːndoː]) at any point. You can if you'd like to, of course, but it wasn't my intention to make things that complicated.

[(h)wiːndoː] is actually pretty close to the protoform for one of the subbranches of NW/N/NE/W/C. Unfortunately, though, it's not very close at all to what I have in mind for Proto-NW/N/NE/W/C itself. The segments [-w-] and [-n-] are present, although not in this configuration. Neither [-iː-] nor [-oː-] is entirely correct or incorrect, though I'm not sure if I can be more specific at this point without potentially giving too much away.

NW: wi:ndo: > wĩ:do: > wĩ:ru
N: wi:ndo: > ɯi̯nˈdu: > əi̯nˈdʉu̯
NE: wi:ndo: > ji:nˈlwa: > ˌʝi:nʊˈlwa:
W: wi:ndo: > ji:nˈdo: > jai̯nˈdoə̯ > jãĩ̯ˈroə̯
C: hwi:ndo: > hɯi̯ndo: > ˈhei̯ndoː

I wish I could comment more on the sound changes you've assumed here, but that's hard to do when the starting point ([(h)wiːndoː]) is incorrect. You don't have to, but it may be helpful to try identifying the lower-level groupings here. 

E/SW/S/SE: djosnr̩w

[dj-] is incorrect. It's a reasonable guess, but I think it may be throwing you off a bit when it comes to reconstructing the original protoform for the entire family. 

You're on the wrong track with the syllabic consonant here, but there is an [-r-]. In fact, [-o-n-rw-] is correct. 

E: djosnr̩w > ɟøçˈnɛw > ɟeˈɲ̊ɔu̯
SW: djosnr̩w > døsˈnɛru > døsˈnœːry
S: djosnr̩w > ˈd͡zɔstr̩v > ˈzɔstr̩bi
SE: djosnr̩w > ˈd͡ʑossɛrv > ˈʑot͡ɕɛːf

[ɟøçˈn-] is correct for the protoform of one of the lower-level groupings within this branch, the one containing E (though not necessarily E alone). Otherwise, what I said above about it being difficult to comment more on specific sound changes applies here as well. 

Overall protoform: ðys.nr̩w
Proto-N: ðys.nr̩w > θy:d.nr̩w~ðy:d.nr̩w > (h)wi:dnəw > (h)wi:ndo:
Proto-S: ðys.nr̩w > ðjos.nr̩w > djos.nr̩w

In general, I would advise against ignoring the placement of stress here. Anyway, [-n-r-w-] (though with a non-syllabic [-r-]) is correct, but none of the other segments are. This is overall rather far off, I'm afraid, especially in terms of syllable structure and the number of syllables.

[zyma]

Response to Üdj's first attempt:

Spoiler:

Very tentative guess:

Spoiler:

*ʝoinzwiɸ

I will say that you're off in terms of syllable structure and the number of syllables, but this is probably the closest anyone's gotten so far! [-o-i-n-w-] is correct, and although [ʝ-] isn't exactly what I had in mind, it may be close enough. Also, generally speaking, I'd advise against ignoring stress here.

Spoiler:

Code: Select all

      1   2   3   4   5   6
NW    Ø   wĩ  :   r   u
N     Ø   əi  n   d   ʉu  Ø
NE    ʝ   i:  nʊ  l   wa  :
W     j   ai  ~   r   oə  Ø
C     h   ei  n   d   o   :
E     ɟ   e   ɲ   Ø   ɔu  Ø
SW    [idk what to do for this one]
S     z   ɔ   st  r   bi  Ø
SE    ʑ   o   t   ɕ   ɛ:  f
*     ʝ   oi  n   z   wi  ɸ

Disregarding SW, the correspondences you've proposed for the initial consonant (or lack thereof) are spot-on! Unfortunately, given the discrepancies between your reconstruction and the form that I have in mind, I'm not sure how much more I can say here without potentially giving too much away about the original form or subdivisions within the family. More specifically, my version would require more than six slots, so it's hard to say whether the segments/features you've placed in most of your slots really correspond to one another or not, since most of your slots correspond to multiple slots on my end. Hopefully that makes sense.

Edit: I also have no idea how this game works, so if I did something wrong, tell me.

Well, you could have fooled me! You've attempted to reconstruct the most recent common ancestor of the set of words I provided, which is what this game's all about. There isn't one single "correct" way to play, though if you wanted, you could try identifying some of the lower-level branches/subdivisions within the hypothetical language family represented by these nine words. Working in stages like that might be easier than trying to reconstruct the original protoform directly from the nine "modern" forms. Again, though, you can play however you'd like.

[ɶʙ ɞʛ]

Second attempt:

NW [ˈwĩːru] - N [əi̯nˈdʉu̯] - NE [ˌʝiːnʊˈlwaː]
W [jãĩ̯ˈroə̯] - C [ˈhei̯ndoː] - E [ɟeˈɲ̊ɔu̯]
SW [døsˈnœːry] - S [ˈzɔstr̩bi] - SE [ˈʑot͡ɕɛːf]

NW/N/NE/W/C = ɥi:nro: ~ ɥi:nəro:

E/SW/S/SE - ʑoɕnɛrwi

Initial proto-form: joi.ner.we:

Spoiler:

To proto-north:

jo we > jø wɤ > ɥe wo /ɥei.ner.wo:/
ei > i:, wo > o /ɥi:.ner.o:/
Elision/reduction of short vowels /ɥi:n.ro:/~/ɥi:nə.ro:/

To NW:
ɥ > w /wi:n.ro:/
o: > u when not stressed /ˈwi:n.ru/
Vn.C > Ṽ.C /ˈwĩ:ru/

To N:
ɥ > j /ji:n.ro:/
nr > nd /ɥi:n.do:/
ji > i /i:n.do:/
i: o: > ɨi̯ u: > əi̯ ʉu̯ /əi̯nˈdʉu̯/

To NE:
ɥ > j /ji:nə.ro:/
ə > ʊ when next vowel is rounded /ji:nʊ.ro:/
#j > #ʝ /ʝi:nʊ.ro:/
r > l /ʝi:nʊ.lo:/
o: > wa: /ˌʝi:nʊˈlwa:/

To W:
ɥ > j /ji:n.ro:/
i: o: > ai̯ oə̯ /jai̯n.roə̯/
Vn.C > Ṽ.C /jãĩ̯ˈroə̯/

To C:
ɥ > j /ji:n.ro:/
#j > #h /hi:n.ro:/
nr > nd /hi:n.do:/
i: > ei̯ when stressed /ˈhei̯ndo:/


To proto-south:
j > ʑ /ʑoʑ.ner.we:/
Random coda devoicing? /ʑoɕ.ner.we:/
e e: > ɛ i /ʑoɕ.nɛr.wi/

To E:
#ʑ > #ɟ /ɟoɕ.nɛr.wi/
o > ø > e when next vowel is front /ɟeɕ.nɛr.wi/
ɕn > ɲ̊ /ɟe.ɲ̊ɛr.wi/
wi > u /ɟe.ɲ̊ɛr.u/
Deletion of r /ɟe.ɲ̊ɛu/
ɛu̯ > ɔu̯ /ɟeˈɲ̊ɔu̯/

To SW:
#ʑ > #ɟ /ɟoɕ.nɛr.wi/
o > ø when next vowel is front /ɟøɕ.nɛr.wi/
wi > y /ɟøɕ.nɛr.y/
Palatals shift to alveolars /døs.nɛr.y/
Rounding harmony /døsˈnœ:ry/

To S:
Sonorants fortite to stops after other consonants /ʑoɕ.tɛr.bi/
Palatals shift to alveolars /zos.tɛr.bi/
Deletion of ɛ, e o > ɛ ɔ: /ˈzɔstr̩bi/

To SE:
Sonorants fortite to stops or fricatives after other consonants /ʑoɕ.tɛr.vi/
ɕt > t͡ɕ /ʑo.t͡ɕɛr.vi/
Deletion of final vowel /ʑo.t͡ɕɛrv/
Deletion of r, compensatory lengthening /ˈʑot͡ɕɛːv/
Final devoicing /ˈʑot͡ɕɛːf/

E/SW vs S/SE seems to be a clear subgrouping, with protoforms /ɟøɕ.nɛr.wi/ for E/SW and /ʑoɕ.tɛr.vi/ for S/SE.

[Arayaz]

Attempt #2:

Spoiler:

*gʲoisnr̩wif

Reasoning:

Spoiler:

Part 1: Beginnings

Northern languages begin with something followed by /i/. The nasalization in NW & W = the /n/ in N & NE.
The ancestor of the northern languages probably began with a semivowel, likely rounded, that could also become /ə/ or palatalize to /ʝ/. I'll propose *ɥ there, but that's tentative.
The h- of C is a clue. Perhaps *ʍ or something?
I'll say something like *?in'do:. Not sure where the length in NE comes from though. (The question mark is a question, not a glottal stop.)
(As for NE, I'll say the wa: is breaking of *o:, the l is from *d, and the ʊ is epenthesis.
W jãĩ.'roə is likely just i: > ai and the same breaking as N and the same d > r as NW.
E ɟ- suggests some other palatal for the *?. I'll update my suspicion to *g or maybe *gʷ or *gʲ. We see the same breaking, too. Why the *N is ɲ and voiceless is a mystery to me.
Perhaps *gin'to: instead?
Now, using what you told me in the response to my last guess, I'll update to *gʲoin'to:. The *w you told me was present will be found later.
All of the southern ones are weird.
SW has stress on the nœ:, but I doubt it's from *do:, especially as it's followed by something else. Definitely vowel harmony of some sort happening, probably frontness.
SW d- is from *gʲ > *dʒ > d. The ø is from *oi, in one of many possible ways. The s is weird, and I don't know what it might come from. The *t definitely became the r, unless affixation is involved. The y is from *o: > *u and getting fronted by the harmony. The œ: is also inexplicable, at least for now.
S z- is from *gʲ > *dʒ > ʒ > z. The ɔ is from *oi, losing the offglide, and *o > ɔ is definitely not the least believable sound change I've invoked here.
The s is showing up again, so I'll add it to the protoform: *gʲoin'sto:.
The -rbi is also very odd.

~ And then he realized: the NW & W r wasn't from the *t.

Part 2: Redemption

In light of this, NW -ru is definitely from what I'll guess is *rwi. I think the *r also becomes the d in N and C, and the NE l.
So: *gʲoinrwi. (For the SW & S s, I'll just say it's from the *n somehow?) I find this a more plausible medium.
SW œ: = epenthesis + rounding harmony.
*w > b in S.
SE ʑ is obviously from *gʲ. The tɕ is strange, probably related to the -st- in S. ɛ: is *wi > *oi > *e: > ɛ:, with some sort of -ATR in unstressed syllables. The -f I can't explain. Probably an affix. But, I'll say there was a *f at the end, which was lost in most cases but remained in SE.

A last-minute decision to add in a *s in before the *n.

[ɶʙ ɞʛ]

/bump

Just making sure that we can revive the challenge at some point.

[zyma]

My response to a recent PM about this thread:

I was originally just trying to wait a bit before responding to the latest guesses in order to give anyone else who might be interested in the game a chance to join (wishful thinking on my part TBH), but then I unexpectedly got very busy with IRL/offline stuff for the past few months & lost track of time. That thread has actually been on my mind again recently, though, so fingers crossed that I'll be able to respond to your guesses soon. Thank you for your continued interest in playing.

It shouldn't be too much longer.

[zyma]

Response to ɶʙ ɞʛ's second attempt: 

Spoiler:

Many thanks again for your patience and continued interest in playing. I've tried to be pretty forthcoming with information here, but please feel free to ask for clarification on anything if needed.

NW/N/NE/W/C = ɥi:nro: ~ ɥi:nəro:

Of these two forms, [ɥiːnəroː] is closer. The sequence [-nər-] is correct. As a reminder, in my response to your previous guess, I confirmed that the segment [w] is present somewhere in the NW/N/NE/W/C protoform. To clarify something else I said last time, neither [iː] nor [o:] is correct, but there is an instance of [oi̯] here. 

E/SW/S/SE - ʑoɕnɛrwi

The form I had in mind was [ˈɟoçnarwe], but I feel like this is close enough. Well done! 

Initial proto-form: joi.ner.we:

You're very much on the right track here. The medial [-e-] is off in terms of vowel quality and the final vowel is off in terms of quantity. Also, [-oi̯-] should be [-oCi-], and one other vowel is missing as well. Aside from the lack of stress marking, though, everything else is correct.

To proto-north:

jo we > jø wɤ > ɥe wo /ɥei.ner.wo:/
ei > i:, wo > o /ɥi:.ner.o:/
Elision/reduction of short vowels /ɥi:n.ro:/~/ɥi:nə.ro:/

There was some kind of vowel reduction, but this is otherwise all incorrect.

I will say that there is a stage between Proto-North and the modern languages. In other words, the NW/N/NE/W/C branch can be further subdivided. You don't have to reconstruct the intermediary forms, of course, but it might help.

To NW:
ɥ > w /wi:n.ro:/
o: > u when not stressed /ˈwi:n.ru/
Vn.C > Ṽ.C /ˈwĩ:ru/

The first two are incorrect, but the last bit about nasalization is correct.

To N:
ɥ > j /ji:n.ro:/
nr > nd /ɥi:n.do:/
ji > i /i:n.do:/
i: o: > ɨi̯ u: > əi̯ ʉu̯ /əi̯nˈdʉu̯/

[-nr-] > [-nd-] is correct. [-uː] > [-ʉu̯] is also correct, though not [-oː] > [-uː]. Everything else is incorrect. 

To NE:
ɥ > j /ji:nə.ro:/
ə > ʊ when next vowel is rounded /ji:nʊ.ro:/
#j > #ʝ /ʝi:nʊ.ro:/
r > l /ʝi:nʊ.lo:/
o: > wa: /ˌʝi:nʊˈlwa:/

The bit about [-ə-] > [-ʊ-] is basically correct. [j-] > [ʝ-] is correct, though [ɥ-] > [j-] is not. [-r-] > [-l-] is correct as well. For the last change, what I had in mind was actually [-oə̯] > [-waː].

To W:
ɥ > j /ji:n.ro:/
i: o: > ai̯ oə̯ /jai̯n.roə̯/
Vn.C > Ṽ.C /jãĩ̯ˈroə̯/

Only the last point is correct. 

To C:
ɥ > j /ji:n.ro:/
#j > #h /hi:n.ro:/
nr > nd /hi:n.do:/
i: > ei̯ when stressed /ˈhei̯ndo:/

[-nr-] > [-nd-] is correct. [j-] > [h-] is close enough. 

To proto-south:
j > ʑ /ʑoʑ.ner.we:/
Random coda devoicing? /ʑoɕ.ner.we:/
e e: > ɛ i /ʑoɕ.nɛr.wi/

[j-] > [ʑ-] is probably close enough. Everything else is incorrect. 

To E:
#ʑ > #ɟ /ɟoɕ.nɛr.wi/
o > ø > e when next vowel is front /ɟeɕ.nɛr.wi/
ɕn > ɲ̊ /ɟe.ɲ̊ɛr.wi/
wi > u /ɟe.ɲ̊ɛr.u/
Deletion of r /ɟe.ɲ̊ɛu/
ɛu̯ > ɔu̯ /ɟeˈɲ̊ɔu̯/

I won't count [ʑ-] > [ɟ-] as incorrect, even though, as mentioned above, [ɟ-] is present in the southern protoform on my end. [o] > [ø] > [e] is correct, but I'm afraid the conditioning is not, as the vowel in the following syllable was not front. [-ɕn-] > [-ɲ-] is basically correct. [-wi] > [-u] is basically correct as well, though I have it as [-we] > [-w] > [-u]. You're also correct about the deletion of [-r-], but the preceding vowel was not [-ɛ-]. 

To SW:
#ʑ > #ɟ /ɟoɕ.nɛr.wi/
o > ø when next vowel is front /ɟøɕ.nɛr.wi/
wi > y /ɟøɕ.nɛr.y/
Palatals shift to alveolars /døs.nɛr.y/
Rounding harmony /døsˈnœ:ry/

I won't count [ʑ-] > [ɟ-] as incorrect, even though, as mentioned above, [ɟ-] is present in the southern protoform on my end. [o] > [ø] > [e] is correct, but I'm afraid the conditioning is not, as the vowel in the following syllable was not front. You're right about the palatal to alveolar shift. SW did develop vowel harmony, but based on backness, not rounding. 

To S:
Sonorants fortite to stops after other consonants /ʑoɕ.tɛr.bi/
Palatals shift to alveolars /zos.tɛr.bi/
Deletion of ɛ, e o > ɛ ɔ: /ˈzɔstr̩bi/

You're right about fortition and shift from palatals to alveolars, though I will say that I have [-β-] as an intermediary stage between [-w-] and [-b-]. [-o-] > [-ɔ-] is correct, and while the vowel in [-Vr-] > [-r̩-] wasn't [-ɛ-], you're basically correct there as well. 

To SE:
Sonorants fortite to stops or fricatives after other consonants /ʑoɕ.tɛr.vi/
ɕt > t͡ɕ /ʑo.t͡ɕɛr.vi/
Deletion of final vowel /ʑo.t͡ɕɛrv/
Deletion of r, compensatory lengthening /ˈʑot͡ɕɛːv/
Final devoicing /ˈʑot͡ɕɛːf/

You're pretty much spot-on here. 

E/SW vs S/SE seems to be a clear subgrouping, with protoforms /ɟøɕ.nɛr.wi/ for E/SW and /ʑoɕ.tɛr.vi/ for S/SE.

You're right about the subgroupings. Additionally, you're pretty close with the protoform for S/SE, but a bit further off for E/SW. [ɟøçn-r-] is correct, though.