Quick Diachronics Challenge
Re: Quick Diachronics Challenge
I divided the languages into East and West according to isogloss #1. Maybe reconstructing each East and West might help. The West is closer to the proto-form concerning the articulation of consonants, but the East preserved the proto-word's "structure" better. Just like the Catalan mirror. Other thing's I've hinted:
- V² is not /a/
- no stops are involved
@gufferdk: A lot of your guesses for the isoglosses are correct, especially the later ones! I really enjoyed reading your post. I think you're really close!
@loglorn: Of course it is obvious then. But I tried to reconstruct it as if I didn't knew what changes etc. happened (of course I'm still biased because I still *know* them) and it was still fairly easy to guess although it was a bit iffy concerning a few changes. A lot of people came quite close to the proto-word already. I try my best to give good hints while still keeping it a challenge!
Native:
Learning: , , ,
Zhér·dûn a tonal Germanic conlang
old stuff: Цiски | Noattȯč | Tungōnis Vīdīnōs
Learning: , , ,
Zhér·dûn a tonal Germanic conlang
old stuff: Цiски | Noattȯč | Tungōnis Vīdīnōs
Re: Quick Diachronics Challenge
Hmm, C2 is a challenge. If it wasn't a stop then it's fortified in a lot of languages. This makes me think it was a fricative.ixals wrote:
I divided the languages into East and West according to isogloss #1. Maybe reconstructing each East and West might help. The West is closer to the proto-form concerning the articulation of consonants, but the East preserved the proto-word's "structure" better. Just like the Catalan mirror. Other thing's I've hinted:
- V² is not /a/
- no stops are involved
@gufferdk: A lot of your guesses for the isoglosses are correct, especially the later ones! I really enjoyed reading your post. I think you're really close!
@loglorn: Of course it is obvious then. But I tried to reconstruct it as if I didn't knew what changes etc. happened (of course I'm still biased because I still *know* them) and it was still fairly easy to guess although it was a bit iffy concerning a few changes. A lot of people came quite close to the proto-word already. I try my best to give good hints while still keeping it a challenge!
Proto-West: *zidʷaɾa
Proto-East: *hiɣɔla
If change 23 was C2 > /d/, then this group retained /d/ the longest (if it was /d/).
/dʷ/ > /w v b d l/ could be possible, but doesn't seem too likely. I can't think of a neater solution though.
Based on this, and assuming /ɣ/ isn't the middle consonant (although you haven't exactly hinted this was wrong), also that V2 isn't /a/ and C2 isn't a stop, I will say *ziðʷǝla. This is my third attempt, so I will leave it at that!
Re: Quick Diachronics Challenge
Trying to reconstruct from what you have there, West is very likely *zilaɾa. The reason for *zi has been explored in depth earlier. l seems likely to fortify to d randomly (#23) and l > (ɫ) > w isn't unattested. This also fits well with trying to explain the irregularities with #8,15 with regards to 12. as 8 then was simply l > w if you reconstruct C3 as *l in eastern. Given everything else, i also think that ɾ > n is more likely than the reverse. The vowels are so commonly a that I'll reconstruct those, though wawes might be messing with it. I could totally see proto-western reatining original V2 and then #3,4,5 messing things up across families.
Proto-eastern is harder. #2 affects all of it so the protoform might be something like *hiC2V2la with #1 being ɾ and l either metathesising or swapping phonetic realisation (a few Polynesian langs does that as counterintuitive as it sounds). Such a swap also qualifies a a "mean" sound change like you hinted. I was sorta onto this with my edit for #6 but now it seems clear that #6 is the stress shift I assigned to #1 originally. ɾ > g unconditionally is also literlly the first sound change from /ɾ/ that pops up in the index diachronica. Give the shape of #7 it seems likely that the velars are an innovation. This leaves V2. It isn't /a/, I doubt it's schwa given how commonly it's not dropped, and how no lang a schwa proper. I could see it being /ɔ/ but I like the formation of #3,4,5 not overlapping, so I'll guess /ɤ/ as centralisation to /ɘ/ (very short move) and lovering to /ɔ, a/ (ɤ > ɑ happened in Uralic) seems like likely sound changes.
This gives me final *hiɾɤla for eastern.
Which of the arrangements of the liquids is correct is anybody guess so I'll go with eastern and reconstruct *zilɤɾa.
Looking at the sound changes (I'll try to go through them all later, #16 might actually be voicing of the initial with /z/ in 2 being a later seperate change. While this goes the other way around from what would be expected it would explain how C1 > h is such a quick change and why there is no isogloss seperating z from s,h. This gives the alternate reconstruction *silɤɾa.
EDIT: Forgot about stress. From the shape of #6 I'll guess it was on the second syllable in the protoform.
Proto-eastern is harder. #2 affects all of it so the protoform might be something like *hiC2V2la with #1 being ɾ and l either metathesising or swapping phonetic realisation (a few Polynesian langs does that as counterintuitive as it sounds). Such a swap also qualifies a a "mean" sound change like you hinted. I was sorta onto this with my edit for #6 but now it seems clear that #6 is the stress shift I assigned to #1 originally. ɾ > g unconditionally is also literlly the first sound change from /ɾ/ that pops up in the index diachronica. Give the shape of #7 it seems likely that the velars are an innovation. This leaves V2. It isn't /a/, I doubt it's schwa given how commonly it's not dropped, and how no lang a schwa proper. I could see it being /ɔ/ but I like the formation of #3,4,5 not overlapping, so I'll guess /ɤ/ as centralisation to /ɘ/ (very short move) and lovering to /ɔ, a/ (ɤ > ɑ happened in Uralic) seems like likely sound changes.
This gives me final *hiɾɤla for eastern.
Which of the arrangements of the liquids is correct is anybody guess so I'll go with eastern and reconstruct *zilɤɾa.
Looking at the sound changes (I'll try to go through them all later, #16 might actually be voicing of the initial with /z/ in 2 being a later seperate change. While this goes the other way around from what would be expected it would explain how C1 > h is such a quick change and why there is no isogloss seperating z from s,h. This gives the alternate reconstruction *silɤɾa.
EDIT: Forgot about stress. From the shape of #6 I'll guess it was on the second syllable in the protoform.
Last edited by gufferdk on 22 Mar 2017 08:04, edited 2 times in total.
Warning: Anything I post may be ninja-edited up to half an hour after posting.
Re: Quick Diachronics Challenge
@gufferdk: Congratulations, you won! The metathesis was the "mean" sound change I was referring too and therefore the proto-form was silɤɾa~siɾɤla indeed!
Native:
Learning: , , ,
Zhér·dûn a tonal Germanic conlang
old stuff: Цiски | Noattȯč | Tungōnis Vīdīnōs
Learning: , , ,
Zhér·dûn a tonal Germanic conlang
old stuff: Цiски | Noattȯč | Tungōnis Vīdīnōs
Re: Quick Diachronics Challenge
Woo. I did not expect to do so well in my first round . I should really go to bed right now, so if noone minds me posting the next challenge then I'll wait until some time tomorrow with that.ixals wrote:@gufferdk: Congratulations, you won! The metathesis was the "mean" sound change I was referring too and therefore the proto-form was silɤɾa~siɾɤla indeed!
Warning: Anything I post may be ninja-edited up to half an hour after posting.
- Creyeditor
- MVP
- Posts: 5091
- Joined: 14 Aug 2012 19:32
Re: Quick Diachronics Challenge
So, I'm gonna go segment by segment for each half. I am still not convinced that we need full CVCVCV for the proto-word.
West: *zilɘra ~ ʑlɘra ~ zilra ~ zilɘr
C1:All frincatives, except j, which may be epenthetic after all. Palatalization might come from following i. Word initial devoicing is more common than word initial voicing, so the proto-C1 should be voiced. This yields: *z
V1: Most of them i, so I guess *i. Another possibility would be a palatalized C1 and no V2 at all.
C2: Three groups, laterals, labials and d. All of them could develop from *l, via w and syllable initial fortition, if we assume unconditional l → ɫ.
V2: If it's not *a, it has to be *ɘ. Don't see why there should be such bizarre lowering, but hey, there's been worse in my challenge. Could very well be epenthetic than.
C3: What links ʀ and n? Obviously only *r, which has the trillness of the former and the POA of the latter. Also *r → n and r → ʀ are defintely attested.
V3: This can be *a or epenthetic. Nothing else.
East: *hiʀʷɤlʲa
C1: Is very obviously h, which in some language got elided. I think I will have to assume it was **s at some point, to link it to the Western languages.
V1: All front vowels, mostly high and/or unrounded, so I guess it's *i. But I have to mind the rounding, where did it come from? The following consonant?
C2: Ranges from stop over nasal to trill and from alveolar to velar. This is also an environment where we would expect both lenition (because intervocalic) and fortition (because syllable onset). I am gonna go with ʀʷ, the exact opposite argumentation.
V2: We got high, mid and low vowels. Diphthongs and Monophthongs. Front, central, and back vowels. This is really tough. If the previous consonant is the trigger of rounding, we could assume a back unrounded vowel here. The following consonant could act as a trigger of raising and fronting to i. So, *ɤ.
C3: This seems to be l again, but palatalized lʲ.
V3: Looks like it is *a again.
All: *zirʷɘrʲa
C1: As I said, I have to assume that there is some stage where the east languages had **s, which means proto-C1 is *z.
V1: Is clearly *i.
C2: Is pretty tough. It could have been rʷ, which backened in the east and became a lateral in the west. Again there must have been some kind of intermediate stage, because rounding did not survive in the east.
V2: Is definitely unrounded, not high and not front. So either *ɤ or *ɘ or something in between.
C3: Really unsure abouth this one. Maybe rʲ?
V3: Is *a.
I feel like I am pretty off the track again. Sorry about that.
*zirʷɘrʲa
West: *zilɘra ~ ʑlɘra ~ zilra ~ zilɘr
C1:All frincatives, except j, which may be epenthetic after all. Palatalization might come from following i. Word initial devoicing is more common than word initial voicing, so the proto-C1 should be voiced. This yields: *z
V1: Most of them i, so I guess *i. Another possibility would be a palatalized C1 and no V2 at all.
C2: Three groups, laterals, labials and d. All of them could develop from *l, via w and syllable initial fortition, if we assume unconditional l → ɫ.
V2: If it's not *a, it has to be *ɘ. Don't see why there should be such bizarre lowering, but hey, there's been worse in my challenge. Could very well be epenthetic than.
C3: What links ʀ and n? Obviously only *r, which has the trillness of the former and the POA of the latter. Also *r → n and r → ʀ are defintely attested.
V3: This can be *a or epenthetic. Nothing else.
East: *hiʀʷɤlʲa
C1: Is very obviously h, which in some language got elided. I think I will have to assume it was **s at some point, to link it to the Western languages.
V1: All front vowels, mostly high and/or unrounded, so I guess it's *i. But I have to mind the rounding, where did it come from? The following consonant?
C2: Ranges from stop over nasal to trill and from alveolar to velar. This is also an environment where we would expect both lenition (because intervocalic) and fortition (because syllable onset). I am gonna go with ʀʷ, the exact opposite argumentation.
V2: We got high, mid and low vowels. Diphthongs and Monophthongs. Front, central, and back vowels. This is really tough. If the previous consonant is the trigger of rounding, we could assume a back unrounded vowel here. The following consonant could act as a trigger of raising and fronting to i. So, *ɤ.
C3: This seems to be l again, but palatalized lʲ.
V3: Looks like it is *a again.
All: *zirʷɘrʲa
C1: As I said, I have to assume that there is some stage where the east languages had **s, which means proto-C1 is *z.
V1: Is clearly *i.
C2: Is pretty tough. It could have been rʷ, which backened in the east and became a lateral in the west. Again there must have been some kind of intermediate stage, because rounding did not survive in the east.
V2: Is definitely unrounded, not high and not front. So either *ɤ or *ɘ or something in between.
C3: Really unsure abouth this one. Maybe rʲ?
V3: Is *a.
I feel like I am pretty off the track again. Sorry about that.
*zirʷɘrʲa
Edit: Damn I am late
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 2 3 4 4
Ook & Omlűt & Nautli languages & Sperenjas
Papuan languages, Morphophonology, Lexical Semantics
"Thoughts are free."
Produce, Analyze, Manipulate
1 2 3 4 4
Ook & Omlűt & Nautli languages & Sperenjas
Papuan languages, Morphophonology, Lexical Semantics
- Creyeditor
- MVP
- Posts: 5091
- Joined: 14 Aug 2012 19:32
Re: Quick Diachronics Challenge
Where did the word initial z's come from?ixals wrote:@gufferdk: Congratulations, you won! The metathesis was the "mean" sound change I was referring too and therefore the proto-form was silɤɾa~siɾɤla indeed!
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 2 3 4 4
Ook & Omlűt & Nautli languages & Sperenjas
Papuan languages, Morphophonology, Lexical Semantics
"Thoughts are free."
Produce, Analyze, Manipulate
1 2 3 4 4
Ook & Omlűt & Nautli languages & Sperenjas
Papuan languages, Morphophonology, Lexical Semantics
Re: Quick Diachronics Challenge
s/z/_ maybe only word initially. but /s/ is indeed the best reconstruction between /z/ and /h/Creyeditor wrote:Where did the word initial z's come from?ixals wrote:@gufferdk: Congratulations, you won! The metathesis was the "mean" sound change I was referring too and therefore the proto-form was silɤɾa~siɾɤla indeed!
- Creyeditor
- MVP
- Posts: 5091
- Joined: 14 Aug 2012 19:32
Re: Quick Diachronics Challenge
Damn, it's kind of what German does/did Again
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 2 3 4 4
Ook & Omlűt & Nautli languages & Sperenjas
Papuan languages, Morphophonology, Lexical Semantics
"Thoughts are free."
Produce, Analyze, Manipulate
1 2 3 4 4
Ook & Omlűt & Nautli languages & Sperenjas
Papuan languages, Morphophonology, Lexical Semantics
Re: Quick Diachronics Challenge
Creyeditor wrote:Damn, it's kind of what German does/did Again
I'm just amazed at that virtually unreconstructible V2 being /ɤ/
Spoiler:
Re: Quick Diachronics Challenge
I have no idea whether this is too hard, too easy or reasonably reasonable in difficulcy. I'll post more hints if necessary:
Big picture:
Small version:
There are four subfamilies (edit: deleted the word "main", there are only 4 families).
The changes primarily follow the tree model though there are a few areal changes. Areal changes never leave the upper-level subfamily they are in.
There is one consonant from the original word that has been changed in all languages, though in different ways. I think it should still be guessable based on the variety of reflexes.
Big picture:
Spoiler:
Spoiler:
Edit: Updated the maps due to missing nasalisation on 4 forms
Hints:There are four subfamilies (edit: deleted the word "main", there are only 4 families).
The changes primarily follow the tree model though there are a few areal changes. Areal changes never leave the upper-level subfamily they are in.
There is one consonant from the original word that has been changed in all languages, though in different ways. I think it should still be guessable based on the variety of reflexes.
Last edited by gufferdk on 22 Mar 2017 16:56, edited 2 times in total.
Warning: Anything I post may be ninja-edited up to half an hour after posting.
Re: Quick Diachronics Challenge
Here is everything that happened for everyone who's interested:
Spoiler:
Why would I not pick a change that happened in my native tongue?Creyeditor wrote:Damn, it's kind of what German does/did Again
Hahaha, I'm sorry for that!qwed117 wrote:I'm just amazed at that virtually unreconstructible V2 being /ɤ/
Edit: I just saw your new post, gufferdk. I'll try the challenge later when I've got enough time!
Native:
Learning: , , ,
Zhér·dûn a tonal Germanic conlang
old stuff: Цiски | Noattȯč | Tungōnis Vīdīnōs
Learning: , , ,
Zhér·dûn a tonal Germanic conlang
old stuff: Цiски | Noattȯč | Tungōnis Vīdīnōs
Re: Quick Diachronics Challenge
Wave 7 is fortition, though not your garden variety fortition. Something like uvularization + fortition.
22 and 23 are also fortition, of a tamer sort. 22 is actually quite common a change both ways.
I will have a go at gufferdk's shortly, but i just wanted to say that above beforehand.
22 and 23 are also fortition, of a tamer sort. 22 is actually quite common a change both ways.
I will have a go at gufferdk's shortly, but i just wanted to say that above beforehand.
Re: Quick Diachronics Challenge
I made a notable screwup while making the map. The second vowel in nysˤøː, nysˤø͡ə, nysˤøw and nysˤew should be nasalised so nysˤø̃ː, nysˤø̃͡ə̃, nysˤø̃w and nysˤẽw are correct. I updated the map in the original post.
Warning: Anything I post may be ninja-edited up to half an hour after posting.
- Creyeditor
- MVP
- Posts: 5091
- Joined: 14 Aug 2012 19:32
Re: Quick Diachronics Challenge
So, my guess, before doing the work and before looking at the hints is: *kni.tʃor
Edit: An here's my second guess, based on a more close look: *gnyts'ode
Here are the groupings that I assume (without arreal effects)
Here are the Proto-Words:
Here are the groupings that I assume (without arreal effects)
Spoiler:
Spoiler:
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 2 3 4 4
Ook & Omlűt & Nautli languages & Sperenjas
Papuan languages, Morphophonology, Lexical Semantics
"Thoughts are free."
Produce, Analyze, Manipulate
1 2 3 4 4
Ook & Omlűt & Nautli languages & Sperenjas
Papuan languages, Morphophonology, Lexical Semantics
Re: Quick Diachronics Challenge
I reconstructed *gn̩ʲcʼoðə which is really similar to Creyeditor's guess but looking at his answer, I think I'll change my guess to *gnycʼodə. The south-western islands were a big problem for me as well as the last consonant in the north. The south (and the east) was my favourite area to reconstruct tbh!
Native:
Learning: , , ,
Zhér·dûn a tonal Germanic conlang
old stuff: Цiски | Noattȯč | Tungōnis Vīdīnōs
Learning: , , ,
Zhér·dûn a tonal Germanic conlang
old stuff: Цiски | Noattȯč | Tungōnis Vīdīnōs
Re: Quick Diachronics Challenge
Hmm. I think *gnʉc'ɔde could work... It's basically the same as what Creyeditor and Ixals wrote
Spoiler:
- Creyeditor
- MVP
- Posts: 5091
- Joined: 14 Aug 2012 19:32
Re: Quick Diachronics Challenge
Interesting. I found this challenge pretty tough, but we all arrived at similar results.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 2 3 4 4
Ook & Omlűt & Nautli languages & Sperenjas
Papuan languages, Morphophonology, Lexical Semantics
"Thoughts are free."
Produce, Analyze, Manipulate
1 2 3 4 4
Ook & Omlűt & Nautli languages & Sperenjas
Papuan languages, Morphophonology, Lexical Semantics
Re: Quick Diachronics Challenge
You are all very close but none of you got it exactly right. Creyeditor was the closest, they was only off on one consonant, the consonant that I hinted wasn't reflexed in its protoform anywhere. Since none of you got that consonant right I think going on would just be a case of random guessing and you were close anyway so I'll just reveal the protoform, which was gnys'ode.
Your grouping was off though. Black, light green and cyan are related but your grouping of languages into them is wrong becuase areal changes are obscuring things. Same story with pink and purple as well as white and red. White/red together belong somewhere else than pink/purple. Dark green, brown and blue are correct but shouldn't go together with yellow. If anyone is interested I'll post all the sound changes.
Your grouping was off though. Black, light green and cyan are related but your grouping of languages into them is wrong becuase areal changes are obscuring things. Same story with pink and purple as well as white and red. White/red together belong somewhere else than pink/purple. Dark green, brown and blue are correct but shouldn't go together with yellow. If anyone is interested I'll post all the sound changes.
Warning: Anything I post may be ninja-edited up to half an hour after posting.
Re: Quick Diachronics Challenge
I got half way through this, so I carried on as if the answer hadn't been revealed. My guess was *ᵑɡiwsħode, with /e/ or /i/ for the final vowel, which fasn't *too* far off, I suppose.gufferdk wrote:You are all very close but none of you got it exactly right. Creyeditor was the closest, they was only off on one consonant, the consonant that I hinted wasn't reflexed in its protoform anywhere. Since none of you got that consonant right I think going on would just be a case of random guessing and you were close anyway so I'll just reveal the protoform, which was gnys'ode.
Your grouping was off though. Black, light green and cyan are related but your grouping of languages into them is wrong becuase areal changes are obscuring things. Same story with pink and purple as well as white and red. White/red together belong somewhere else than pink/purple. Dark green, brown and blue are correct but shouldn't go together with yellow. If anyone is interested I'll post all the sound changes.
I thought /ᵑɡ/ could give rise to most of the initial reflexes, but /gn/ makes sense too. I chose /sħ/ because I could see this fortifying or becoming /sʕ/ or similar, and then moving to /sˤ/.
I used your hint at four language families which I divided as: