Quick Diachronics Challenge

A forum for all topics related to constructed languages
User avatar
sangi39
moderator
moderator
Posts: 3274
Joined: 12 Aug 2010 00:53
Location: North Yorkshire, UK

Re: Quick Diachronics Challenge

Post by sangi39 » 03 Nov 2018 18:06

Creyeditor wrote:
03 Nov 2018 11:12
Woops, totally forgot about this. So sorry. I would have time tomorrow to create a new challenge, if that's still okay.
I'm happy waiting as well [:)]
You can tell the same lie a thousand times,
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.

User avatar
Creyeditor
mongolian
mongolian
Posts: 4590
Joined: 14 Aug 2012 18:32

Re: Quick Diachronics Challenge

Post by Creyeditor » 05 Nov 2018 02:20

So I went a bit over the top when preparing this challenge, creating about 40 descendent languages, that I all like and care for. Since I thought that might be too much to cope with, I figured I could just split it up into three rounds. In the first round, you will see 12 cognates, in the second round I will add about 15, the same goes for the third round. Before I uncover them, some cognates will appear with black bars in the map. I hope you are okay with it. You get points for winning in each round, as well as having the best guess in general. Extra points for correct grouping or finding the Uhrheimat. Sorry again for being a month late.


Here's the challenge [:)]
Reduced size
Spoiler:
Image
Full size
Spoiler:
Image
Some notes on the notation. Superscript g /g/ indicates voicing on a click, doubling of a letter shows length. The rest should be standard IPA but feel free to ask.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]

User avatar
sangi39
moderator
moderator
Posts: 3274
Joined: 12 Aug 2010 00:53
Location: North Yorkshire, UK

Re: Quick Diachronics Challenge

Post by sangi39 » 05 Nov 2018 03:19

Just a preliminary run at this:
Spoiler:
Image
Spoiler:
The only group I think I can find a proto-word for is the red group in the south-east, something along the lines of *apəkdu.

The orange and green groups are tricky. I'm not sure how closely they are related too each other, with the nasalisation being either an inherited trait or the result of a "wave" change.

Similarly, I'm unsure as to the direction of migration, i.e. whether the major migration occurred east to west or west to east, but I've tried to make a guess at other migrations.
You can tell the same lie a thousand times,
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.

User avatar
GrandPiano
runic
runic
Posts: 2690
Joined: 11 Jan 2015 23:22
Location: Ohio, USA

Re: Quick Diachronics Challenge

Post by GrandPiano » 05 Nov 2018 05:17

My first attempt:
Spoiler:
Proto-word: *ãpəkʰdoo

Main island: *ãpəkʰdoo > *ãbikdo
*ãbikdo > *ãbikd > *ãbikt > *ãbitt > *ə̃bitt
  • *ə̃bitt > *əmitt > *əmett > əmets
  • *ə̃bitt > ə̃βitt
*ãbikdo > *ãbykdo > *ãbykd > *ãbygd > *ãbygg > *ãbỹgg > *ãmỹgg
  • *ãmỹgg > *œ̃mỹgg > œ̃mỹgg
    *ãmỹgg > ãnỹgg
Tiny island in the middle: *ãpəkʰdoo > *avəgdo > *vəgdo > *vəᶢǃo
  • *vəᶢǃo > vaᶢǃo
  • *vəᶢǃo > vəᶢǃo
Island to the right of the tiny island: *ãpəkʰdoo > *apəkʰdɯɯ
  • *apəkʰdɯɯ > *apəkdɯɯ > *apəktɯɯ > æpəktɯɯ
  • *apəkʰdɯɯ > *abəkʰdɯɯ > ɐbəkʰdɯɯ
Bottom island: *ãpəkʰdoo > *apəkdo > *apəido > afəiðo

Bottom right island: *ãpəkʰdoo > *ɑvəkro
  • *ɑvəkro > ɑvəkʟ̝̊o
  • *ɑvəkro > ɑvekro
Top right island: *ãpəkʰdoo > *ãməkdɯ > *aməɗɯ > *amɗɯ > anɗɯ
:eng: - Native
:chn: - B2
:esp: - A2
:jpn: - A2

User avatar
CarsonDaConlanger
sinic
sinic
Posts: 250
Joined: 02 Nov 2017 20:55

Re: Quick Diachronics Challenge

Post by CarsonDaConlanger » 05 Nov 2018 16:49

Creyeditor wrote:
05 Nov 2018 02:20
So I went a bit over the top when preparing this challenge, creating about 40 descendent languages, that I all like and care for. Since I thought that might be too much to cope with, I figured I could just split it up into three rounds. In the first round, you will see 12 cognates, in the second round I will add about 15, the same goes for the third round. Before I uncover them, some cognates will appear with black bars in the map. I hope you are okay with it. You get points for winning in each round, as well as having the best guess in general. Extra points for correct grouping or finding the Uhrheimat. Sorry again for being a month late.


Here's the challenge [:)]
Reduced size
Spoiler:
Image
Full size
Spoiler:
Image
Some notes on the notation. Superscript g /g/ indicates voicing on a click, doubling of a letter shows length. The rest should be standard IPA but feel free to ask.
My computer can't view the images, if it's not an issue, could you link it in google drive? If not, could you list them in text? Sorry for the trouble.

User avatar
Creyeditor
mongolian
mongolian
Posts: 4590
Joined: 14 Aug 2012 18:32

Re: Quick Diachronics Challenge

Post by Creyeditor » 05 Nov 2018 17:42

They might be too big, sorry for the size. Here is a link, hope it works https://abload.de/img/histoxround1thfxb.png.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]

User avatar
sangi39
moderator
moderator
Posts: 3274
Joined: 12 Aug 2010 00:53
Location: North Yorkshire, UK

Re: Quick Diachronics Challenge

Post by sangi39 » 06 Nov 2018 16:16

Spoiler:
So, sticking with the same groupings I had in my preliminary work, I'd hazard a guess, using the data we have so far, that the proto word is something like *amikdu.

The nasal element is retained in the orange and green groups, where it is transferred to the preceding vowel (and to the following vowel in the green group, possibly in some nasal harmony-esque sort of way).

In the red group, this nasalisation is lost, with *m shifting to *b, giving Proto-Green *abəkdu, with the *b becoming /p/ in the north (looks similar to Tocharian, where the MOA distinction on plosives was lost entirely), while becoming /v/ in the southern islands. The initial *a was lost in the western subgroup, which also renders the interior cluster as a voiced alveolar click. Other groups handled this cluster differently, presumably with the southern group leniting the initial in a manner similar to that of Spanish and French, i.e. *k > *i, while the eastern group lenited the *d, at first to *r, which then seems to have become a voiceless lateral fricative in one language (I assume the change was alveolar > velar > lateral, although a shift to a lateral first, then variously to a rhotic vs. a velar sound doesn't seem unreasonable either).

In the north-eastern island, the cluster seems to have been simplified as well, although it's hard to tell in which direction. /ɗ/ is known to come from various sources, including /tʼ/ (which could have come from **kt), /ʔd/ (which could come from **kd), and even simply from /d/. The middle vowel seems to have been lost, and the retained nasal assimilates to match the POA of the following implosive.

The rounded vowels in the green group could be down to regressive assimilation, similar to umlaut, in this case *iCCu > *yCCu > *yCC, and I assume the geminate consonants in both the yellow and the green groups comes from cluster simplification, i.e. *kd > *kt > *tt and *kd > *gd > *gg, although given the /ts/ in one of the yellow languages, I'd be tempted to say that palatalisation may have played a role in this (similar to how Iceland egg is cognate with Crimean Gothic ada).

I'm still unsure of the movement of languages from the main island to the eastern islands, but given the retention of nasals in the west, I'd hazard a guess that the language family originates somewhere on the mainland.
You can tell the same lie a thousand times,
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.

User avatar
Creyeditor
mongolian
mongolian
Posts: 4590
Joined: 14 Aug 2012 18:32

Re: Quick Diachronics Challenge

Post by Creyeditor » 09 Nov 2018 21:22

Hey guys, I justed wanted to mention that I will leave you some feedback only when I will get to post the second round. I will give feedback only on the grouping and maybe the Urheimat, since these are potentially subject to change for each round. I will not comment on your guesses of the proto-word, since you might get around changing your guess based on the new langs anyway.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]

shimobaatar
korean
korean
Posts: 11633
Joined: 12 Jul 2013 22:09
Location: PA → IN

Re: Quick Diachronics Challenge

Post by shimobaatar » 10 Nov 2018 20:26

Creyeditor wrote:
05 Nov 2018 02:20
So I went a bit over the top when preparing this challenge, creating about 40 descendent languages, that I all like and care for. Since I thought that might be too much to cope with, I figured I could just split it up into three rounds. In the first round, you will see 12 cognates, in the second round I will add about 15, the same goes for the third round. Before I uncover them, some cognates will appear with black bars in the map.
Wow, this is a lot!

I've assigned each group a letter, moving from west to east (with one exception), for convenience.
Spoiler:
Group A:
-[œ̃mỹgg]
-[ãnỹgg]
< *[ãmỹgg] (Proto-A)

Group B:
-[ə̃βitt]
-[əmets]
< *[əmicc] (Proto-B)

A *[ãmỹgg]
B *[əmicc]
< **[ãmỹɟɟ] (Proto-AB)

Group C:
-[vaᵍ!o]
-[vəᵍ!o]
< *[vaᵍ!o] (Proto-C)

Group D:
-[afəiðo]

Group E:
-[ɑvəkʟ̝̊o]
-[ɑvəkro]
< *[ɑvəklo] (Proto-E)

C *[vaᵍ!o]
D [afəiðo]
E *[ɑvəklo]
< **[afakto] (Proto-CDE)

Group F:
-[æpəktɯɯ]
-[ɐbəkʰdɯɯ]
< *[apəkʰdɯɯ] (Proto-F)

Group G:
-[anɗɯ]

AB **[ãmỹɟɟ]
CDE **[afakdo]
F *[apəkʰdɯɯ]
G [anɗɯ]
< ***[am̥ækʰdu] (Proto-World)
Spoiler:
Proto-World ***[am̥ækʰdu]

Proto-AB **[ãmỹɟɟ] ([Vm̥V] > [ṼmṼ], [æ̃CCu#] > [ɶ̃CC#], [kʰd] > [gd] > [jd] > [ɟɟ], [ɶ̃] raises to [ỹ] over time)
Proto-CDE **[afakdo] ([m̥] > [ɸ] > [f], [aCæ] > [aCa], [kʰd] > [kd], [u] lowers to [o])
Proto-F *[apəkʰdɯɯ] ([m̥] > [p], medial vowel reduction, progressive rounding harmony, final vowel lengthening)
G [anɗɯ] ([Vm̥V] > [VmV], medial vowel deletion, progressive rounding harmony, [kʰd] > [kd] > [ʔd] > [ɗ], nasal place assimilation)

Proto-A *[ãmỹgg] (final depalatalization of [ɟɟ] to [gg])
Proto-B *[əmicc] (vowels denasalize, (pretonic?) vowel reduction, [y] unrounds to [i], word-final obstruent devoicing)
Proto-C *[vaᵍ!o] ([kd] > [kt] > [!], intervocalic voicing, loss of initial (unstressed?) vowels)
D [afəiðo] (medial vowel reduction, [VkC] > [ViC], [d] lenites to [ð])
Proto-E *[ɑvəklo] (medial vowel reduction, [a] backs to [ɑ], intervocalic voicing, [d] > [l])
F₁ [æpəktɯɯ] ([a] fronts to [æ], [kʰd] > [kʰt] > [kt])
F₂ [ɐbəkʰdɯɯ] ([a] becomes [ɐ], intervocalic voicing)

A₁ [œ̃mỹgg] ([ãCỹ] > [æ̃Cỹ], [ɶ̃Cỹ] > [œ̃Cỹ])
A₂ [ãnỹgg] ([m] > [n])
B₁ [ə̃βitt] ([VmV] > [ṼβV], [cc] depalatalizes to [tt] word-finally)
B₂ [əmets] (vowels are lowered in closed syllables, final degemination, [c] > [ts])
C₁ [vaᵍ!o] (no change)
C₂ [vəᵍ!o] ((pretonic?) vowel reduction)
E₁ [ɑvəkʟ̝̊o] ([l] assimilates in POA and voicing to [k], [ʟ̥] > [ʟ̝̊])
E₂ [ɑvəkro] ([l] > [r])

Just to clarify:

-Is stress not relevant? It doesn't seem to be marked anywhere.
-Is the [i] in [əi] non-syllabic, or is this a sequence of two syllables?

User avatar
Creyeditor
mongolian
mongolian
Posts: 4590
Joined: 14 Aug 2012 18:32

Re: Quick Diachronics Challenge

Post by Creyeditor » 11 Nov 2018 15:55

shimobaatar wrote:
10 Nov 2018 20:26
-Is stress not relevant? It doesn't seem to be marked anywhere.
-Is the [i] in [əi] non-syllabic, or is this a sequence of two syllables?
1. Stress is only marked when relevant, i.e. when it is not predictable. That was my idea.
2. It's a diphthong, but I would actually argue that the schwa might be non-syllabic.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]

User avatar
Creyeditor
mongolian
mongolian
Posts: 4590
Joined: 14 Aug 2012 18:32

Re: Quick Diachronics Challenge

Post by Creyeditor » 14 Nov 2018 01:49

It's time for the second round:

Original size:
Spoiler:
Image
Reduced size
Spoiler:
Image
Link to the original file: https://abload.de/image.php?img=histoxround2a3fsx.png
Link to the reduced file: https://abload.de/image.php?img=histoxround25uf1s.png


And here's the feedback for all of your guesses in the first round. As always, feel free to read or to ignore your own and everyone else's feedback.
@sangi
Spoiler:
Your light green and orange group are valid low level groupings. Your red group on the other hand does not correspond to any valid genealogical grouping. There are also several groupings that you did not find yet. Your Urheimat is still off, but you have identified several shared inventions correctly that should lead you to a correct grouping.
@GrandPiano
Spoiler:
You have identified all correct low-level grouping that were detectable in round one. Your main island family is also a valid grouping so far. You have missed several higher and intermediate level groupings in round one.
@shimo
Spoiler:
All of the grouping you assumed is valid so far. You have identified all detectable low level grouping in round one correctly. Your AB and DEC groups are also correct groupings. Still, you have missed several groupings both below and above the level of these families.
I will not indicate any preliminary/intermediate result, i.e. who of you is doing best in round one. I think you are all doing great actually [;)] I hope everything is clear and okay so far. If not, feel free to ask.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]

User avatar
sangi39
moderator
moderator
Posts: 3274
Joined: 12 Aug 2010 00:53
Location: North Yorkshire, UK

Re: Quick Diachronics Challenge

Post by sangi39 » 14 Nov 2018 18:31

If it's okay, can I have until about Tuesday afternoon (in the UK) to work on round 2? I'm going away for a few days and don't want to miss out.
You can tell the same lie a thousand times,
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.

User avatar
Creyeditor
mongolian
mongolian
Posts: 4590
Joined: 14 Aug 2012 18:32

Re: Quick Diachronics Challenge

Post by Creyeditor » 14 Nov 2018 20:47

That's totally okay. I am doing the challenge for you guys, so I can wait for you (sg). I wanted to let you guys have at least one week for the second round anyway, so no big deal.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]

User avatar
GrandPiano
runic
runic
Posts: 2690
Joined: 11 Jan 2015 23:22
Location: Ohio, USA

Re: Quick Diachronics Challenge

Post by GrandPiano » 18 Nov 2018 21:59

One of the words appears to be transcribed afə̻xdo, with a square below the ə. As far as I know, a square below an IPA letter indicates laminal articulation, which I don't think makes sense for a vowel. Is this a mistake, or is it a notation that I'm not familiar with?
:eng: - Native
:chn: - B2
:esp: - A2
:jpn: - A2

User avatar
Creyeditor
mongolian
mongolian
Posts: 4590
Joined: 14 Aug 2012 18:32

Re: Quick Diachronics Challenge

Post by Creyeditor » 19 Nov 2018 08:56

GrandPiano wrote:
18 Nov 2018 21:59
One of the words appears to be transcribed afə̻xdo, with a square below the ə. As far as I know, a square below an IPA letter indicates laminal articulation, which I don't think makes sense for a vowel. Is this a mistake, or is it a notation that I'm not familiar with?
Sorry, it's supposed to be a voiceless vowel, so a ring below.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]

User avatar
sangi39
moderator
moderator
Posts: 3274
Joined: 12 Aug 2010 00:53
Location: North Yorkshire, UK

Re: Quick Diachronics Challenge

Post by sangi39 » 19 Nov 2018 15:41

Spoiler:
Image
Spoiler:
So, I've tried to group languages together at the lowest levels I think make sense, and I'll deal with them moving roughly north to south, east to west:

A:
(a) /ә̃mit/
(b) /ә̃βitt/
(c) /әmets/
< *әmicc, with (c) branching off earlier (proto-c *әmetɕ perhaps, with proto-ab as *ә̃mitt with a laminal denti-aveolar /t/).

B:
(a) /œ̃mỹgg/
(b) /ãnỹgg/
< *ãmỹgg, where the /œ̃/ in B.a comes from regressive rounding and the /n/ in B.b seems to be the result of a simple shift of /m/ > /n/.

C:
(a) /ɔɔmiʒdʲ/
(b) /umiʒdʲ/
(c) /amiʒʒ/
< *amiʒdʲ. I suspect, given the tendency in surrounding languages for the vowel in the first syllable to be /a/ that /a/ is the original vowel in Proto-C, having raised in the east, where the group is split in two (proto-ab ɔmiʒdʲ where the final consonant cluster is maintained, but the initial vowel rises and backs, and proto-c, where the initial vowel is retained, but the final cluster simplifies, possibly through -ʒdʲ- > -ʒdʒ- > -ʒʒ-).

I suspect /amðð/ is in some way related to Group C, losing the middle vowel, and then possibly fronting the final cluster, possibly though an intermediate *zz.

I'm struggling more to place /ambigi/ and /ˈʔamɨkk/ although I get the feeling their exact placement will become much more clear once the languages between them, A and B are revealed.

Over the sea to D:
(a) /vag!o/
(b) /vәg!o/
< *vag!o ~ *vәg!o (looking at this one purely in isolation, the vowel could really go in any direction.

E:
(a) /ɔukdɯɯ/
(b) /akʷdɯɯ/
< *aukdɯɯ ~ *awkdɯɯ, the *u~w rounding the preceding vowel in (a) but the following consonant in (b).

F:
(a) /ɒhokdo/
(b) /ɒɸɔkdo/
(c) /afə̥kdo/
(d) /afəiðo/
< *ɒfəkdo ~ ɒɸəkdo, with clear groupings into a northern group (proto-ab ɒɸokdo, where the rounding of the first vowel spreads to the second) and a southern group (proto-cd *afəkdo, where the rounding of the first vowel is lost). The other option is of course that the rounding is an innovation is proto-ab as a result of being adjacent to a bilabial, and the move to /h/ in (a) does make me thing the plan is to hint at “this feature has become phonemic in this language”, but I think it's safe to say the first vowel is definitely low and back.

G:
(a) /æpəktɯɯ/
(b) /ɐbəkʰdɯɯ/
< *apək(ʰ)dɯɯ. This one was actually a bit tricky. The front vowel does seem more likely to be /a/ than anything else, fronting in (a) while rising in (b), the first consonant a /p/ since intervocalic voicing is more common (although annoyingly (a) could have gone through a sort of Tocharian MOA levelling), then *-əkdɯɯ, but I don't know whether the aspiration in (b) was an innovation or something that was lost in (a). It's a surprisingly tricky pair when taken on their own.

H:
(a) /ɑvəkʟ̝̊o/
(b) /ɑvəkro/
(c) /havəkjo/
< *ɑvəkro, where /ʟ̝̊/ and /j/ in (a) and (c) respectively come from *r (I think at least /r/ > /j/ is attested, might be in Burmese, and I'm sure I've seen /r/ > /ʟ̝̊/ somewhere, might be similar to how Hiw ended up with its prestopped velar lateral). The initial /h/ in (c) might be something similar to the prosthesis seen in some Slavic languages where a “semivowel” (/j/ or /v/) is inserted in front of initial vowels (/h/ in this case, through an intermediate maybe).

I:
(a) /anɗɯ/
(b) /amɓɯ/
(c) /ɒmədɯ̀/
(d) /amə̰ðɯ/
< *aməʔdɯ, again, split into two groups, a northern one (proto-ab *amɗɯ, where the medial cluster simplifies through POA assimilation, regressive in (a) but progressive in (b)) and a southern one (proto-cd *aməʔdɯ, the glottal stop yielding in creaky voice in (d) but some sort of tonal system in (c)). In the northern group, the *-ʔd- cluster simplified becoming an implosive *-ɗ-, after which the preceding schwa was dropped, while in the southern group the glottal stop was maintained, but dropped out affecting the word in other ways later on.



That gives us:
Proto-A: *әmicc
Proto-B: *ãmỹgg
Proto-C: *amiʒdʲ
Proto-D: *vag!o ~ *vәg!o
Proto-E: *aukdɯɯ ~ *awkdɯɯ
Proto-F: *ɒfəkdo ~ ɒɸəkdo
Proto-G: *apək(ʰ)dɯɯ
Proto-H: *ɑvəkro
Proto-I: *aməʔdɯ

There's definitely some sort of larger western grouping of A, B, and C going on that involves patalisatation somehow, as well as a grouping of D, E, F, G and H going on in the south-east, leaving I out there on its own (it seems pretty closely related to the western group because of the *m, but it lacks palatalisation, so either, it's more closely related to the western groups, but split off before palatalisation or it's more closely related to the south-eastern groups, but split off before denasalisation).

Anyway, Proto-D likely comes from some older Pre-Proto-D *vakdo ~ *vәkdo where the cluster simplified into a voiced click, which makes it look a lot closer to Proto-H and Proto-F now:

Pre-Proto-D: *vakdo ~ *vәkdo
Proto-F: *ɒfəkdo ~ ɒɸəkdo
Proto-H: *ɑvəkro
< *ɑfəkdo, with a western group (Proto-DF *ɑfəkdo) and an eastern group (Proto-H *ɑvəkro). Intervocalic voicing happening more than once seems fine, so I'd be okay reconstructing *f here rather than *v, the initial vowel being lost in D, and the *d leniting to *r in H.

Proto-G and Proto-E look particularly close to each other, especially with the same vowel ending, so maybe:

Proto-G: *apək(ʰ)dɯɯ
Proto-E *aukdɯɯ ~ *awkdɯɯ
< *abəkdɯɯ where the *b shifted to *w in the south with the following schwa dropping out, but being retained in the north. This suggests that Proto-G might in fact be *abəkdɯɯ, with the voicelessness in the north and the aspiration in the south both being innovations, not retentions as previously thought (although I did leave the option open that this might have been the case, just didn't expect to make the wrong choice twice [:P] ).

A, B, and C are still a pain. Palatalisation seems to be a common feature, suggesting that it might be from a shared ancestor (Proto-ABC), so let's see where that takes me:

Proto-A: *әmicc
Proto-B: *ãmỹgg
< *amyɟɟ ~ *amiɟɟ where the palatal geminate backed in B (it's a rare change, but I've heard it said that Egyptian Arabic /g/ does legitimately come from /ɟ/) and devoiced in A. I'm not sure where the rounded vowel in Proto-B came from, though. Could be spread of rounding from the *m in the same way that nasalisation would spread out later.

Proto-AB: *amyɟɟ ~ *amiɟɟ
Proto-C: *amiʒdʲ
< *amiɟdʲ. This would seem to look like the shift of palatalisation as a secondary articulation to some “palatal” consonant in Proto-C, while a merger between and occurred in the northern languages.

If I had to take a stab at it, I'd hazard a guess that there's some Pre-Proto-ABC stage that looks something like *amigʲdʲ(i).

That gives:

Proto-ABC: *amiɟdʲ (earlier *amigʲdʲ(i))
Proto-GE: *abəkdɯɯ
Proto-DFH: *ɑfəkdo
Proto-I: *aməʔdɯ

At a guess, I'd say ABC is its own branch, with GE, I and DFH off in the east. Proto-I, if this is the correct grouping, seems to have split off from the eastern group earlier than the rest, with DFH diverging from GE later on. Giving:

Proto-GE: *abəkdɯɯ
Proto-DFH: *ɑfəkdo
< *abəkdɯ (southern). This does make me question qhat I said of intervocalic voicing in the D and H branches of DFH, so perhaps Proto-DFH is *ɑvəkdo

Southern: *abəkdɯ
Proto-I: *aməʔdɯ
< *aməkdɯ (eastern) (glottalisation in I, denasalisation in Southern)

Eastern: *aməkdɯ
Proto-ABC: *amiɟdʲ (earlier *amigʲdʲ(i))
< *amekdɯ ~ *amekdɨ

Annoyingly, I'm still thinking the urheimat is somewhere on the western island, I suspect somewhere in the region of /ambigi/ and /ˈʔamɨkk/, but that was actually more or less my original guess without seeing those languages, so now I suppose I'll have to wait to see what languages turn up between AB and C (which I get the feeling will answer the question of how exactly those are related) and AB and /ambigi/ and /ˈʔamɨkk/ (which I think might be able to answer the question of what exactly that medial cluster is).
You can tell the same lie a thousand times,
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.

User avatar
cedh
MVP
MVP
Posts: 378
Joined: 07 Sep 2011 21:25
Location: Tübingen, Germany
Contact:

Re: Quick Diachronics Challenge

Post by cedh » 19 Nov 2018 17:34

Spoiler:
My guess for the earliest proto-word (after reading Sangi's latest reconstruction) would be *ambəkdɨ, with a prenasalised stop and two central vowels.

User avatar
GrandPiano
runic
runic
Posts: 2690
Joined: 11 Jan 2015 23:22
Location: Ohio, USA

Re: Quick Diachronics Challenge

Post by GrandPiano » 20 Nov 2018 02:22

Creyeditor wrote:
19 Nov 2018 08:56
GrandPiano wrote:
18 Nov 2018 21:59
One of the words appears to be transcribed afə̻xdo, with a square below the ə. As far as I know, a square below an IPA letter indicates laminal articulation, which I don't think makes sense for a vowel. Is this a mistake, or is it a notation that I'm not familiar with?
Sorry, it's supposed to be a voiceless vowel, so a ring below.
Oh, OK. I figured that might be it.

I’ll try to post my next attempt within the next few days.
:eng: - Native
:chn: - B2
:esp: - A2
:jpn: - A2

User avatar
GrandPiano
runic
runic
Posts: 2690
Joined: 11 Jan 2015 23:22
Location: Ohio, USA

Re: Quick Diachronics Challenge

Post by GrandPiano » 20 Nov 2018 22:46

Here's my next attempt, though I'm not very confident about my proto-word:
Spoiler:
I start with northernmost of the southeastern islands, which appears to have the most conservative forms, and move outward from there.

Image

Proto-word: *hampəkʲʰdo
  • Proto-Green (north): *hampəkʲʰdo > *hampəkʰdo > *ampəkʰdo > *ampəkʰdɯ
    • Proto-Pink (east): *ampəkʰdɯ > *ãpəkʰdɯ > *apəkʰdɯ > *apəkʰdɯɯ
      • Proto-Red (east): *apəkʰdɯɯ
        • *apəkʰdɯɯ > *apəkʰtɯɯ > *apəktɯɯ > æpəktɯɯ
        • *apəkʰdɯɯ > *abəkʰdɯɯ > ɐbəkʰdɯɯ
      • Proto-Gray (center): *apəkʰdɯɯ > *apəkdɯɯ > *aβəkdɯɯ > *aβkdɯɯ > *aukdɯɯ
        • *aukdɯɯ > ɔukdɯɯ
        • *aukdɯɯ > akʷdɯɯ
    • Proto-Yellow (east): *ampekʰdɯ > *ampəkdɯ > *ambəkdɯ > *aməkdɯ > *aməʔdɯ
      • Proto-Purple (east): *aməʔdɯ
        • *aməʔdɯ > *amə̰dɯ > amə̰ðɯ
        • *aməʔdɯ > *amədɯ > *ɒmədɯ > ɒmədɯ̀
      • Proto-Brown (east): *aməʔdɯ > *aməɗɯ > *amɗɯ
        • *amɗɯ > amɓɯ
        • *amɗɯ > anɗɯ
  • Proto-Orange (east): *hampəkʲʰdo > *ampəkdo > *ambəgdo > *ãbəgdo > *avəgdo > *vəgdo > *vəᶢǃo
    • *vəᶢǃo > vəᶢǃo
    • *vəᶢǃo > vaᶢǃo
  • Proto-Blue (east): *hampəkʲʰdo > *ampəkdo > *ãpəkdo > *apəkdo > *aɸəkdo
    • Proto-Green (south): *aɸəkdo > *aɸokdo > *ɒɸokdo
      • *ɒɸokdo > ɒhokdo
      • *ɒɸokdo > ɒɸɔkdo
    • Proto-Lime (south): *aɸəkdo > *afəkdo
      • *afəkdo > *afəxdo > afə̥xdo
      • *afəkdo > *afəido > afəiðo
  • Proto-Lime (east): *hampəkʲʰdo > *hampəkdo > *hambəkdo > *hãbəkdo > *habəkdo > *havəkdo > *havəkro
    • Proto-Gray (east): *havəkro > *avəkro > *ɑvəkro
      • *ɑvəkro > ɑvəkro
      • *ɑvəkro > ɑvəkʟ̝̊o
    • *havəkro > havəkjo
  • Proto-Red (west): *hampəkʲʰdo > *ampəkʲdo > *ampikdo > *ambigdo > *ambigdø > *ambigdy
    • Proto-Yellow (west): *ambigdy > *ambigd > *ambikt > *ambitt > *əmbitt
      • *əmbitt > *əmitt > *əmett > əmets
      • *əmbitt > *ə̃bitt > ə̃βitt
      • *əmbitt > *əmitt > *əmit > ə̃mit
    • Proto-Orange (west): *ambigdy > *ambiggy
      • Proto-Brown (west): *ambiggy > *ambiggi
        • *ambiggi > ambigi
        • *ambiggi > *amiggi > *amigg > *amikk > *amɨkk > ˈʔamɨkk
      • Proto-Blue (west): *ãmỹgg
        • *ãmỹgg > ãnỹgg
        • *ãmỹgg > œ̃mỹgg
      • Proto-Purple (west): *ambigdy > *ambigdi > *ambiʒdi > *ããbiʒdi > *aamiʒdi
        • Proto-Pink (west): *aamiʒdi > *ɔɔmiʒdi > *ɔɔmiʒdʲ
          • *ɔɔmiʒdʲ > ɔɔmiʒdʲ
          • *ɔɔmiʒdʲ > *ɔmiʒdʲ > umiʒdʲ
        • Proto-Lavender: *aamiʒdi > *aamiʒd > *aamiʒð
          • *aamiʒð > aamiʒʒ
          • *aamiʒð > *aamiðð > *aamðð > amðð
Last edited by GrandPiano on 21 Nov 2018 02:29, edited 3 times in total.
:eng: - Native
:chn: - B2
:esp: - A2
:jpn: - A2

shimobaatar
korean
korean
Posts: 11633
Joined: 12 Jul 2013 22:09
Location: PA → IN

Re: Quick Diachronics Challenge

Post by shimobaatar » 20 Nov 2018 23:48

I'll try to make an attempt soon, too.

Post Reply