Quick Diachronics Challenge

A forum for all topics related to constructed languages
User avatar
sangi39
moderator
moderator
Posts: 2750
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: Quick Diachronics Challenge

Post by sangi39 »

ɶʙ ɞʛ wrote:
20 Jan 2020 05:05
My new guess: (maybe a 9-vowel system, /i ɨ u e ə o æ a ɒ/, to start with?)

*'pat'isæ
*aɲɟa'tɨk'a
*hila'mangjɒ
Spoiler:
So, to start, unfortunately, the reconstructed vowel system isn't correct.

Overally, in some parts you've gotten closer, but in others you've gotten further away. I can't really say much beyond that just based on this post alone.
If anyone else would like a second (or first) run at this, obviously, please do [:)]
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.

ɶʙ ɞʛ
sinic
sinic
Posts: 243
Joined: 02 Aug 2019 18:47

Re: Quick Diachronics Challenge

Post by ɶʙ ɞʛ »

If the original vowel system was just /i u e o æ a/, then:

*pat'isæ
*aɲɟa'tuka
*hila'mangjo, but *o was more like /ɒ/ in the protolang

opipik
roman
roman
Posts: 1366
Joined: 12 Mar 2015 19:41

Re: Quick Diachronics Challenge

Post by opipik »

Spoiler:
1
[ˈpatʃʼːe]
[aɲˈtʃika]
[ʃejˈɲago]

2
[ˈpatsʼiʃ]
[ɲaˈtsik]
[ʎeˈmaŋ]

protoforms: something like
*pats’iʃe
*ɲaˈtʃika
*ʎeˈmaŋgo

-

3 (groups with 1 & 2)
[ˈpaɟise]
[ɲæˈcikːa]
[leˈmeŋgo]

protoforms:
*pac’iʃe
*ɲæˈcika
*leˈmæŋgo

--

4
[ˈpʼaːtsʼæ]
[ɲeˈtuka]
[hilˈmæːŋa]

5
[ˈautsʼa]
[ɲiˈtuka]
[hilˈmaiɲa]

protoforms:
*'p’a:ts’æ
*ɲeˈtuka
*hilˈmæ:ŋa

-

6 (groups with 4 and 5)
[ˈpʼatis]
[ɲæˈtug]
[hilæˈmæŋ]

7
[ˈpadizæ]
[ãˈtuː]
[ilaˈmãː]

protoforms:
*'p'atisæ
*ɲæˈtug
*hilæˈmæŋ

--

8
[ˈbadis]
[jaˈtukʼa]
[ilˈmeŋgi]

9
[ˈbaris]
[ʑæˈtuga]
[ilˈmægi]

-

10 (groups with 8 & 9)
[ˈbalisa]
[dʑæˈtukːa]
[læˈmæga]

protoforms:
*'badis
*jæ'tuk’a
*ilæ'mæŋgi

general protoforms:
*'p’at’ise
*ɲæ'tuk’a
*ile'mæŋgo

I can't adequately explain the i/o discrepancy in the last vowel of the third word - Ukrainian did do o > (v)i though so I've reconstructed it as *o for now. (8/9 vs. 1/2/3).

User avatar
sangi39
moderator
moderator
Posts: 2750
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: Quick Diachronics Challenge

Post by sangi39 »

opipik wrote:
18 Feb 2020 14:03
Spoiler:
1
[ˈpatʃʼːe]
[aɲˈtʃika]
[ʃejˈɲago]

2
[ˈpatsʼiʃ]
[ɲaˈtsik]
[ʎeˈmaŋ]

protoforms: something like
*pats’iʃe
*ɲaˈtʃika
*ʎeˈmaŋgo

-

3 (groups with 1 & 2)
[ˈpaɟise]
[ɲæˈcikːa]
[leˈmeŋgo]

protoforms:
*pac’iʃe
*ɲæˈcika
*leˈmæŋgo
Spoiler:
While the three languages do form a valid grouping together (as mentioned above), you've unfortunately tried to split the group up in the wrong way, which I think may have influenced the reconstructions you've ended up with for the group as a while. They're close, but not quite right, and I would suspect grouping them differently might see your guesses getting even closer.

opipik wrote:
18 Feb 2020 14:03
Spoiler:
4
[ˈpʼaːtsʼæ]
[ɲeˈtuka]
[hilˈmæːŋa]

5
[ˈautsʼa]
[ɲiˈtuka]
[hilˈmaiɲa]

protoforms:
*'p’a:ts’æ
*ɲeˈtuka
*hilˈmæ:ŋa
Spoiler:
A valid grouping, and remarkably close in terms of your reconstructed forms. I think there's only one vowel wrong in the entire lot, otherwise it's spot on.

opipik wrote:
18 Feb 2020 14:03
Spoiler:
6 (groups with 4 and 5)
[ˈpʼatis]
[ɲæˈtug]
[hilæˈmæŋ]

7
[ˈpadizæ]
[ãˈtuː]
[ilaˈmãː]

protoforms:
*'p'atisæ
*ɲæˈtug
*hilæˈmæŋ
Spoiler:
As above, a valid grouping, including the higher grouping with 4 and 5, and the proto-forms you've given for proto-6,7 are, again spot on (completely in this case).

I've noted that you haven't given a reconstruction for proto-4,5-6,7, though, which I would have been interested in seeing [:)]

opipik wrote:
18 Feb 2020 14:03
Spoiler:
8
[ˈbadis]
[jaˈtukʼa]
[ilˈmeŋgi]

9
[ˈbaris]
[ʑæˈtuga]
[ilˈmægi]

-

10 (groups with 8 & 9)
[ˈbalisa]
[dʑæˈtukːa]
[læˈmæga]

protoforms:
*'badis
*jæ'tuk’a
*ilæ'mæŋgi
Spoiler:
A valid grouping in its entirety, i.e. 8,9, with 10 as a close relative on the side, forming a single 8,9,10 grouping higher up. I'd say that the proto-8,9,10 forms are relatively close, but have a couple of elements slightly off.

opipik wrote:
18 Feb 2020 14:03
Spoiler:
general protoforms:
*'p’at’ise
*ɲæ'tuk’a
*ile'mæŋgo
Spoiler:
I'd say these are about as equally close as Shimobaatar's guess. Between the two of you, two of your guesses are completely correct, and a couple of elements out on the third

opipik wrote:
18 Feb 2020 14:03
Spoiler:
I can't adequately explain the i/o discrepancy in the last vowel of the third word - Ukrainian did do o > (v)i though so I've reconstructed it as *o for now. (8/9 vs. 1/2/3).
Spoiler:
This one's, admittedly, a rather tricky correspondence, but I'd think wider comparison with more distant groups would be it clearer as to what the original final vowel might have been
Spoiler:
As with other guesses, wider comparisons, in general, might be helpful [:)]
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
sangi39
moderator
moderator
Posts: 2750
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: Quick Diachronics Challenge

Post by sangi39 »

ɶʙ ɞʛ wrote:
16 Feb 2020 22:19
If the original vowel system was just /i u e o æ a/, then:

*pat'isæ
*aɲɟa'tuka
*hila'mangjo, but *o was more like /ɒ/ in the protolang
I'd say this is closer that the guess before, but probably about as close as your first guess, overall.
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.

shimobaatar
korean
korean
Posts: 7465
Joined: 12 Jul 2013 23:09
Location: PA → IN

Re: Quick Diachronics Challenge

Post by shimobaatar »

I just want to say that I'd like to give it another shot, although finding the time to go back over things may be tricky.

A brief response to feedback and a few new thoughts:
Spoiler:
sangi39 wrote:
11 Jan 2020 23:52
shimobaatar wrote:
05 Jan 2020 20:39
Proto-ABCD
*[ˈpʼatʼisæ] - *[ɲæˈtukːa] - *[hilaˈmæŋgo]
This is the closest anyone has been so far, and it's actually remarkably close (there's literally only two mistakes).
sangi39 wrote:
11 Jan 2020 23:52
Despite your lack of comfort about your reconstruction here, they're actually spot on [:D]
sangi39 wrote:
11 Jan 2020 23:52
As with Proto-B, the reconstructions are spot-on (well, actually, apart from the first vowel of Word 2).
Oh wow, I'm pleasantly surprised to hear that!

For Proto-C, was the second word *[ɲæˈtug] instead of *[ɲaˈtug]?
sangi39 wrote:
11 Jan 2020 23:52
As you'll see in your question about grouping, this one is correct in general, but there is another split in the group, as you have alluded to, so your difficulty in placing Language 1 is understandable.

Your second assumption about where the [-j-] came from in Word 3 of Language 1 was well spotted [;)]
sangi39 wrote:
11 Jan 2020 23:52
As with Group A, you've spotted that this one can be broken down into smaller groups, which could help reconstruction the words for Proto-D. Regardless of that, though, the reconstructions you have are pretty darn close.
sangi39 wrote:
11 Jan 2020 23:52
As I mentioned above, your current reconstructions are remarkably close.

Your thought on the Proto-A [-i-] in Word 1 is spot on, i.e. it's an example iotacism, but your reconstruction of the final vowel of Word 3 for Proto-ABCD? Perfect [:D] I'm sure you'll figure out what happened to the stressed vowels in Proto-B, but it looks like you're managing just fine without it [:)]

As I've mentioned to others, what might help is looking at how the groups might be more closely related to each other at older time-depths, which could help pin down more specifically what each proto-word might be (I think recognising older relationships might help you figure out some of the things you're unsure of, for example).
Thank you for all your feedback! Hopefully I can put it to good use.

Regarding the long vowels in Proto-B:

*[ˈpʼaːtsʼæ] - *[ɲeˈtuka] - *[hilˈmæːŋa] (Proto-B)
< **[ˈpʼatʼisæ] - **[ɲæˈtukːa] - **[hilaˈmæŋgo] (Proto-ABCD)

Perhaps the stressed vowel in the first word lengthened because it was originally in an open syllable? But then how does that explain *[æː] in the third word? My current reconstruction has stressed **[æ] before **[ŋg], so maybe something similar to late Old English's homorganic lengthening took place? Actually, though, looking at the modern languages, only two have [ŋg], so maybe that syllable was originally open in the proto-language? As in **[hilaˈmæŋo]? Although I'm not sure if I want to assume a change from [ŋ] > [ŋg] in two different languages (3 and 8), especially since they seem to be separated by a mountain range on the map.

I've also started doubting myself regarding my reconstruction of **[kː] in the second word, but I don't have the time at the moment to try to think of something better.

ɶʙ ɞʛ
sinic
sinic
Posts: 243
Joined: 02 Aug 2019 18:47

Re: Quick Diachronics Challenge

Post by ɶʙ ɞʛ »

My explanations for those:

Adjacent to original palatals, *a > æ and *o > *ə > a if unstressed. In proto-45, short *æ > e, while short *a > *æ after *i.

Original /ng/ cluster broke into /nVg/ in word 3, while assimilating to /ŋg/ elsewhere. In some of these, /ŋg/ reduced to /ŋ/, while stressed vowels in open syllables lengthened in proto-45.

User avatar
sangi39
moderator
moderator
Posts: 2750
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: Quick Diachronics Challenge

Post by sangi39 »

shimobaatar wrote:
18 Feb 2020 20:59
Spoiler:
sangi39 wrote:
11 Jan 2020 23:52
shimobaatar wrote:
05 Jan 2020 20:39
Proto-ABCD
*[ˈpʼatʼisæ] - *[ɲæˈtukːa] - *[hilaˈmæŋgo]
This is the closest anyone has been so far, and it's actually remarkably close (there's literally only two mistakes).
sangi39 wrote:
11 Jan 2020 23:52
Despite your lack of comfort about your reconstruction here, they're actually spot on [:D]
sangi39 wrote:
11 Jan 2020 23:52
As with Proto-B, the reconstructions are spot-on (well, actually, apart from the first vowel of Word 2).
Oh wow, I'm pleasantly surprised to hear that!

For Proto-C, was the second word *[ɲæˈtug] instead of *[ɲaˈtug]?
Spoiler:
It was indeed [:D]
shimobaatar wrote:
18 Feb 2020 20:59
Spoiler:
sangi39 wrote:
11 Jan 2020 23:52
As you'll see in your question about grouping, this one is correct in general, but there is another split in the group, as you have alluded to, so your difficulty in placing Language 1 is understandable.

Your second assumption about where the [-j-] came from in Word 3 of Language 1 was well spotted [;)]
sangi39 wrote:
11 Jan 2020 23:52
As with Group A, you've spotted that this one can be broken down into smaller groups, which could help reconstruction the words for Proto-D. Regardless of that, though, the reconstructions you have are pretty darn close.
sangi39 wrote:
11 Jan 2020 23:52
As I mentioned above, your current reconstructions are remarkably close.

Your thought on the Proto-A [-i-] in Word 1 is spot on, i.e. it's an example iotacism, but your reconstruction of the final vowel of Word 3 for Proto-ABCD? Perfect [:D] I'm sure you'll figure out what happened to the stressed vowels in Proto-B, but it looks like you're managing just fine without it [:)]

As I've mentioned to others, what might help is looking at how the groups might be more closely related to each other at older time-depths, which could help pin down more specifically what each proto-word might be (I think recognising older relationships might help you figure out some of the things you're unsure of, for example).
Thank you for all your feedback! Hopefully I can put it to good use.

Regarding the long vowels in Proto-B:

*[ˈpʼaːtsʼæ] - *[ɲeˈtuka] - *[hilˈmæːŋa] (Proto-B)
< **[ˈpʼatʼisæ] - **[ɲæˈtukːa] - **[hilaˈmæŋgo] (Proto-ABCD)

Perhaps the stressed vowel in the first word lengthened because it was originally in an open syllable? But then how does that explain *[æː] in the third word? My current reconstruction has stressed **[æ] before **[ŋg], so maybe something similar to late Old English's homorganic lengthening took place? Actually, though, looking at the modern languages, only two have [ŋg], so maybe that syllable was originally open in the proto-language? As in **[hilaˈmæŋo]? Although I'm not sure if I want to assume a change from [ŋ] > [ŋg] in two different languages (3 and 8), especially since they seem to be separated by a mountain range on the map.

I've also started doubting myself regarding my reconstruction of **[kː] in the second word, but I don't have the time at the moment to try to think of something better.
Spoiler:
You're spot on in thinking that the long vowels in B come from open syllables, but it's just that the exact timing is a little weird. So far, your reconstructions for the consonants in Proto-ABCD are all perfectly correct, so no need to second guess yourself on that front. I think the only thing wrong is a total of two vowels across the entire reconstruction for Proto-ABCD.
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
sangi39
moderator
moderator
Posts: 2750
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: Quick Diachronics Challenge

Post by sangi39 »

opipik wrote:
18 Feb 2020 14:03
Spoiler:
...
Spoiler:
Since I gave Shimobaatar a similar hint, your reconstruction for the vowels in the ultimate proto-language are spot-on, but you've gotten two consonants wrong (sort of... look at word 3 and see whether a loss or a gain is more likely).
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
sangi39
moderator
moderator
Posts: 2750
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: Quick Diachronics Challenge

Post by sangi39 »

ɶʙ ɞʛ wrote:
18 Feb 2020 22:05
My explanations for those:

Adjacent to original palatals, *a > æ and *o > *ə > a if unstressed. In proto-45, short *æ > e, while short *a > *æ after *i.

Original /ng/ cluster broke into /nVg/ in word 3, while assimilating to /ŋg/ elsewhere. In some of these, /ŋg/ reduced to /ŋ/, while stressed vowels in open syllables lengthened in proto-45.
Spoiler:
Unfortunately the development of palatals happened after Proto-123, so no affect on unstressed vowels such as that. Similarly the affect on vowels in Proto-45 is less to do with length and umlaut (and you've sadly started with the wrong vowels).

No cluster breaking in Word 3, I'm afraid, either, but you're right about reduction to /ŋ/ in some languages, and the lengthening of stressed vowels in open syllables for Proto-45 [:)]

To give the same note I just gave Shimobaatar and opipik, across all three words in the proto-language, you've got the stress placement right, 7 (I think) consonants right, and 7 of the vowels, but you've got the number of syllables wrong in Word 2
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.

shimobaatar
korean
korean
Posts: 7465
Joined: 12 Jul 2013 23:09
Location: PA → IN

Re: Quick Diachronics Challenge

Post by shimobaatar »

Quick revision of the vowels:
Spoiler:
**[ˈpʼatʼise] - **[ɲæˈtukːa] - **[hileˈmæŋgo] (Proto-ABCD)

ɶʙ ɞʛ
sinic
sinic
Posts: 243
Joined: 02 Aug 2019 18:47

Re: Quick Diachronics Challenge

Post by ɶʙ ɞʛ »

sangi39 wrote:
18 Feb 2020 22:22
ɶʙ ɞʛ wrote:
18 Feb 2020 22:05
My explanations for those:

Adjacent to original palatals, *a > æ and *o > *ə > a if unstressed. In proto-45, short *æ > e, while short *a > *æ after *i.

Original /ng/ cluster broke into /nVg/ in word 3, while assimilating to /ŋg/ elsewhere. In some of these, /ŋg/ reduced to /ŋ/, while stressed vowels in open syllables lengthened in proto-45.
Spoiler:
Unfortunately the development of palatals happened after Proto-123, so no affect on unstressed vowels such as that. Similarly the affect on vowels in Proto-45 is less to do with length and umlaut (and you've sadly started with the wrong vowels).

No cluster breaking in Word 3, I'm afraid, either, but you're right about reduction to /ŋ/ in some languages, and the lengthening of stressed vowels in open syllables for Proto-45 [:)]

To give the same note I just gave Shimobaatar and opipik, across all three words in the proto-language, you've got the stress placement right, 7 (I think) consonants right, and 7 of the vowels, but you've got the number of syllables wrong in Word 2
The /j/ also caused that vowel shift and existed in the protolanguage.

If there were no other palatals besides /j/ in the protolanguage, word 2 could be *anja'tuka, with an epenthetic *d in some cases resulting in *dʑ, but in other cases reducing to *j or *ɲ.

User avatar
sangi39
moderator
moderator
Posts: 2750
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: Quick Diachronics Challenge

Post by sangi39 »

ɶʙ ɞʛ wrote:
20 Feb 2020 05:47
sangi39 wrote:
18 Feb 2020 22:22
ɶʙ ɞʛ wrote:
18 Feb 2020 22:05
My explanations for those:

Adjacent to original palatals, *a > æ and *o > *ə > a if unstressed. In proto-45, short *æ > e, while short *a > *æ after *i.

Original /ng/ cluster broke into /nVg/ in word 3, while assimilating to /ŋg/ elsewhere. In some of these, /ŋg/ reduced to /ŋ/, while stressed vowels in open syllables lengthened in proto-45.
Spoiler:
Unfortunately the development of palatals happened after Proto-123, so no affect on unstressed vowels such as that. Similarly the affect on vowels in Proto-45 is less to do with length and umlaut (and you've sadly started with the wrong vowels).

No cluster breaking in Word 3, I'm afraid, either, but you're right about reduction to /ŋ/ in some languages, and the lengthening of stressed vowels in open syllables for Proto-45 [:)]

To give the same note I just gave Shimobaatar and opipik, across all three words in the proto-language, you've got the stress placement right, 7 (I think) consonants right, and 7 of the vowels, but you've got the number of syllables wrong in Word 2
The /j/ also caused that vowel shift and existed in the protolanguage.

If there were no other palatals besides /j/ in the protolanguage, word 2 could be *anja'tuka, with an epenthetic *d in some cases resulting in *dʑ, but in other cases reducing to *j or *ɲ.
Spoiler:
Ah, fudge, I appear to have misspoken. The development of some palatals didn't occur until after Proto-123
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.

shimobaatar
korean
korean
Posts: 7465
Joined: 12 Jul 2013 23:09
Location: PA → IN

Re: Quick Diachronics Challenge

Post by shimobaatar »

shimobaatar wrote:
18 Feb 2020 23:27
Quick revision of the vowels:
Spoiler:
**[ˈpʼatʼise] - **[ɲæˈtukːa] - **[hileˈmæŋgo] (Proto-ABCD)
Was this any closer?

User avatar
sangi39
moderator
moderator
Posts: 2750
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: Quick Diachronics Challenge

Post by sangi39 »

shimobaatar wrote:
22 Feb 2020 23:26
shimobaatar wrote:
18 Feb 2020 23:27
Quick revision of the vowels:
Spoiler:
**[ˈpʼatʼise] - **[ɲæˈtukːa] - **[hileˈmæŋgo] (Proto-ABCD)
Was this any closer?
Hmmmmm.... It's been four days since I did my last lot of replies, right? I think that might be enough time to say that:

[ˈpʼatʼise]
[ɲæˈtukːa]
[hileˈmæŋgo]

Are all spot on [:D] Well, as close as. I think the vowel of the initial syllable in Word 3 could be reconstructed as /i/ and it honestly wouldn't make a difference, but was conceived as /e:/

For a while, it was a running tie between yourself and opipik (I think opipik made a more definitive decision on how to group 1-2-3 and 8-9-10, but you were already coming to some good conclusions on that front, while still not explicitly pinning it down), but you got their first.

Image

(larger image https://i.imgur.com/yAR5qNm.png)

There are few sounds changes here and there (like the exact timing of vowel lengthening in early 4-5), since I can't seem to find the final version any more, but I think most of them are there. You can see one of my favourites, for example, of having a vowel reduce to schwa, and then have that go in two different directions, which explains the /i/~/a/ alternation in 8-9 and 10 [:P]
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.

ɶʙ ɞʛ
sinic
sinic
Posts: 243
Joined: 02 Aug 2019 18:47

Re: Quick Diachronics Challenge

Post by ɶʙ ɞʛ »

Interesting, I thought the /i/-/a/ alternations came from a diphthong.
Last edited by ɶʙ ɞʛ on 25 Feb 2020 21:34, edited 1 time in total.

shimobaatar
korean
korean
Posts: 7465
Joined: 12 Jul 2013 23:09
Location: PA → IN

Re: Quick Diachronics Challenge

Post by shimobaatar »

Wait, do I not get a chance to provide something for the next round?

User avatar
sangi39
moderator
moderator
Posts: 2750
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: Quick Diachronics Challenge

Post by sangi39 »

shimobaatar wrote:
25 Feb 2020 19:35
Wait, do I not get a chance to provide something for the next round?
That's the way we've typically done things, yeah [:S] It goes to the winner, unless they let it skip over them.
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.

ɶʙ ɞʛ
sinic
sinic
Posts: 243
Joined: 02 Aug 2019 18:47

Re: Quick Diachronics Challenge

Post by ɶʙ ɞʛ »

Oops, you should present yours first.

shimobaatar
korean
korean
Posts: 7465
Joined: 12 Jul 2013 23:09
Location: PA → IN

Re: Quick Diachronics Challenge

Post by shimobaatar »

Just wanted to say that I am working on something for this thread, but I've been having a hard time feeling satisfied with anything I come up with. Sorry for the wait.

Post Reply