Quick Diachronics Challenge
Re: Quick Diachronics Challenge
I'm gonna need some examples of German doing that. I am completely unaware of such a process.
At kveldi skal dag lęyfa,
Konu es bręnnd es,
Mæki es ręyndr es,
Męy es gefin es,
Ís es yfir kømr,
Ǫl es drukkit es.
Konu es bręnnd es,
Mæki es ręyndr es,
Męy es gefin es,
Ís es yfir kømr,
Ǫl es drukkit es.
- Inkcube-Revolver
- cuneiform
- Posts: 126
- Joined: 05 Nov 2015 23:20
- Location: Miami, FL
Re: Quick Diachronics Challenge
I second that. xD
I like my languages how I like my women: grammatically complex with various moods and tenses, a thin line between nouns and verbs, and dozens upon dozens of possible conjugations for every single verb.
- Creyeditor
- MVP
- Posts: 5091
- Joined: 14 Aug 2012 19:32
Re: Quick Diachronics Challenge
Okay, so one examples is <Chips> /tʃɪps/ which often becomes [tʃʏps], (actually [ʃʏps] in my dialect) . I don't think German low vowels act that way though.
Also re: aspiration. As an unconditional change tʰ → t makes a lot of sense, but word initially I think t → tʰ is more natural. Either way, it was a 50-50 chance.
Also, I think all of you played really well
Also re: aspiration. As an unconditional change tʰ → t makes a lot of sense, but word initially I think t → tʰ is more natural. Either way, it was a 50-50 chance.
Also, I think all of you played really well
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
But ɪ>ʏ is pretty much unconditional in stressed syllables isn't it? I've definitely heard it in words like the name Tim, where there's no post-alveolar sound near it. It would make sense though, as German ʃ, at least for me, is strongly rounded.Creyeditor wrote:Okay, so one examples is <Chips> /tʃɪps/ which often becomes [tʃʏps], (actually [ʃʏps] in my dialect) . I don't think German low vowels act that way though.
Also re: aspiration. As an unconditional change tʰ → t makes a lot of sense, but word initially I think t → tʰ is more natural. Either way, it was a 50-50 chance.
Also, I think all of you played really well
At kveldi skal dag lęyfa,
Konu es bręnnd es,
Mæki es ręyndr es,
Męy es gefin es,
Ís es yfir kømr,
Ǫl es drukkit es.
Konu es bręnnd es,
Mæki es ręyndr es,
Męy es gefin es,
Ís es yfir kømr,
Ǫl es drukkit es.
- Creyeditor
- MVP
- Posts: 5091
- Joined: 14 Aug 2012 19:32
Re: Quick Diachronics Challenge
I don't have it in <Tim>, I have it in <Tisch> though. So its bidirectional for me.
Also, some people might have it adjacent to labials, that would explain <Tim>. I once read a paper that had an accoustic explanation for it, mostly based on f3.
Also, some people might have it adjacent to labials, that would explain <Tim>. I once read a paper that had an accoustic explanation for it, mostly based on f3.
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
- Inkcube-Revolver
- cuneiform
- Posts: 126
- Joined: 05 Nov 2015 23:20
- Location: Miami, FL
Re: Quick Diachronics Challenge
That's pretty interesting. I'm pretty new to German, so stuff like this I'm just not normally privy to.
I like my languages how I like my women: grammatically complex with various moods and tenses, a thin line between nouns and verbs, and dozens upon dozens of possible conjugations for every single verb.
Re: Quick Diachronics Challenge
Interesting all that was. I guess we just showed here why comparative method isn't done with only one cognate set. Some changes, especially some environments, were not retrievable just from that cognate set. I would like to see the path, i.e. the sound changes involved, in the development of each result of the last segment, as it appears ultimately at pretty much every single POA from labial to glottal.
Re: Quick Diachronics Challenge
Here is the next round:
I wanted to limit myself on the sound changes but then I couldn't control myself. If anyone wants more hints, just say so! I'm curious what everyone comes up with.
Spoiler:
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
Since I didn't get a chance to try the last one, I will have a go at this, although I am not too familiar with how to work out sound changes that happened in waves (as opposed to the tree-model).ixals wrote:Here is the next round:
First hint, if needed: Get your surfboards ready, it's time for waves![/spoiler]
I wanted to limit myself on the sound changes but then I couldn't control myself. If anyone wants more hints, just say so! I'm curious what everyone comes up with.
If I'm approaching this in completely the wrong way, could somebody explain briefly how to go about reconstructing sound changes that happened in waves?
EDIT: I changed the format of my response to better fit a wave-model of sound changes.
Spoiler:
Re: Quick Diachronics Challenge
I'm a sound change noob, doing this for the learning by doing experience so watch me mess this up in 3, 2, 1, go:
Just looking at the map it seems pretty clear that the protoform was something along the lines of *SiCaLa, (As might be something else)
Going through for each element and mapping out the isoglosses leaves this map which didn't reveal anything amazingly useful:
The *S is definitely a coronal fricative of some description. I'll probably reconstruct it as *z as initial devoicing sounds more likely that hte oppositeThat the requires the following sound changes all of which seem reaosnably reasonable (some of them probably not in all environments):
z > s /#_ everywhere but (1), 2, 4, 6
z > j / _i in 1 (possibly with intermediaries)
s > ɕ / _i (probably) in 9-11, 25
ɕ > j in 11
s > h in 12-23
h > Ø in 13-15, 17, 20
The first vowel is very commonly /i/ so I reconstruct *i which remains mostly unchanged. Changes:
i > Ø / (unknown) in 9-11, 25
i > y (*C seems to be rounded/labial in a number of places so that probably causes this rounding. Probably part of a larger V[+front] > V[+front +rounded] / _C[+labial(ised)]) in 17-18
i > ɪ in 12
i > e in 19, 21-22 (*C seems to lower vowels. Some places have it as a rhotic so that may be the cause of this)
Additionally 13 has /i:/ which I guess comes from reducing *Ca to Ø with compensatory lengthening
*C is more problematic. It is reflexed as /d l w v b f ɾ r̝ k g ŋ ŋw ɾw/ as wel as nasalisation on the following vowel. It probably had a labial component but other than that I am not really sure. I'll reconstruct it as *ɾʷ? because as far as I can see on the Index diachronica both ɾ > g and w > g are attested, even unconditionally. It could also be *dʷ but the /d/s could also be ephentic in most places. Whether the /w v/ in 11, 13 comes from this is uncertain. For 13 I don't think so but I don't know for 11.
The first *a is reflexed as /a ɘ ɤ i ɔ ɔ̃ ɛ Ø/. The nasalisation in 19 is clearly form ŋɔ > ɔ̃. /a/ is the most sommon reflex so I'll go with that. Attempts at explaining the other reflexes:
/i/ in 14-15 is probably from being reduced because it's unstressed then assimilating to the preceeding vowel
/ɔ/ in 17-20 might come from the rounding on *ɾʷ? and raising from the preceeding vowel.
/ɛ/ in 21-22 might be a similar story without the rounding as they preserve rounding on the consonant. This probably also means that this change happened before the lowering of *i as I would imagine /i/ would be more likely to raise a vowel that /e/
/ɘ/ in 3, 25 comes from centralisation
/ɤ/ in 12 probably somes from centralisation, then backing
*L is reflexed as /ʀ ʁ n ɾ v l ʎ Ø/. The uvulars in 1, 5-6 works well as an areal change so the original probably wasn't uvular. I'll reconstruct *l because I feel like it.
The final vowel is commonly /a/ when attested, so I'll go with that and not bother writing out explanations for the edge cases because this has taken way too much time already.
The stress is either initial or second. I'll guess initial because I have no idea about how stress moves around.
This gives me a final *ˈziɾʷala. This is probably wrong but that's what I get for being a noob. One issue I have is that it means ɾ and l have swapped places in 23-24 which isn't completely unattested, but I don't want to spend any more time right now.
Couple of edits: minor mistakes corrected
Just looking at the map it seems pretty clear that the protoform was something along the lines of *SiCaLa, (As might be something else)
Going through for each element and mapping out the isoglosses leaves this map which didn't reveal anything amazingly useful:
Spoiler:
z > s /#_ everywhere but (1), 2, 4, 6
z > j / _i in 1 (possibly with intermediaries)
s > ɕ / _i (probably) in 9-11, 25
ɕ > j in 11
s > h in 12-23
h > Ø in 13-15, 17, 20
The first vowel is very commonly /i/ so I reconstruct *i which remains mostly unchanged. Changes:
i > Ø / (unknown) in 9-11, 25
i > y (*C seems to be rounded/labial in a number of places so that probably causes this rounding. Probably part of a larger V[+front] > V[+front +rounded] / _C[+labial(ised)]) in 17-18
i > ɪ in 12
i > e in 19, 21-22 (*C seems to lower vowels. Some places have it as a rhotic so that may be the cause of this)
Additionally 13 has /i:/ which I guess comes from reducing *Ca to Ø with compensatory lengthening
*C is more problematic. It is reflexed as /d l w v b f ɾ r̝ k g ŋ ŋw ɾw/ as wel as nasalisation on the following vowel. It probably had a labial component but other than that I am not really sure. I'll reconstruct it as *ɾʷ? because as far as I can see on the Index diachronica both ɾ > g and w > g are attested, even unconditionally. It could also be *dʷ but the /d/s could also be ephentic in most places. Whether the /w v/ in 11, 13 comes from this is uncertain. For 13 I don't think so but I don't know for 11.
The first *a is reflexed as /a ɘ ɤ i ɔ ɔ̃ ɛ Ø/. The nasalisation in 19 is clearly form ŋɔ > ɔ̃. /a/ is the most sommon reflex so I'll go with that. Attempts at explaining the other reflexes:
/i/ in 14-15 is probably from being reduced because it's unstressed then assimilating to the preceeding vowel
/ɔ/ in 17-20 might come from the rounding on *ɾʷ? and raising from the preceeding vowel.
/ɛ/ in 21-22 might be a similar story without the rounding as they preserve rounding on the consonant. This probably also means that this change happened before the lowering of *i as I would imagine /i/ would be more likely to raise a vowel that /e/
/ɘ/ in 3, 25 comes from centralisation
/ɤ/ in 12 probably somes from centralisation, then backing
*L is reflexed as /ʀ ʁ n ɾ v l ʎ Ø/. The uvulars in 1, 5-6 works well as an areal change so the original probably wasn't uvular. I'll reconstruct *l because I feel like it.
The final vowel is commonly /a/ when attested, so I'll go with that and not bother writing out explanations for the edge cases because this has taken way too much time already.
The stress is either initial or second. I'll guess initial because I have no idea about how stress moves around.
This gives me a final *ˈziɾʷala. This is probably wrong but that's what I get for being a noob. One issue I have is that it means ɾ and l have swapped places in 23-24 which isn't completely unattested, but I don't want to spend any more time right now.
Couple of edits: minor mistakes corrected
Last edited by gufferdk on 19 Mar 2017 19:44, edited 2 times in total.
Warning: Anything I post may be ninja-edited up to half an hour after posting.
Re: Quick Diachronics Challenge
I basically agree with gufferdk to a large extent; except on the rhotic-lateral order. I think it's more likely that stress moved from antepenult to the penult. This lead to the fortition of a velar l or labiovelar w to a velar plosive g.
*'zilara.
I feel like metathesis would be more likely with this. But I don't think metathesis occured. I think /l/ might be more tapped and thus entering some flaporhinoglottophillia.
*'zilara.
I feel like metathesis would be more likely with this. But I don't think metathesis occured. I think /l/ might be more tapped and thus entering some flaporhinoglottophillia.
Spoiler:
- Creyeditor
- MVP
- Posts: 5091
- Joined: 14 Aug 2012 19:32
Re: Quick Diachronics Challenge
Okay, so here is a list of final segments that originated from *k. h, nx, n̥, ç, ʃ, ʂ, ʔ, ∅, t, q, kʷ, p, c, tʃ.loglorn wrote:I would like to see the path, i.e. the sound changes involved, in the development of each result of the last segment, as it appears ultimately at pretty much every single POA from labial to glottal.
h was a direct word final debuccalization into a fricative.
nx is a combination of the nasality that split from the nasal vowel and a postnasaly fortified *h
the voiceless nasal is another outcome of the combination *nh, which is sensible I think.
I already talked about ç at length in an earlier post.
ʃ is a development from ç. A similar change is attested in various German dialects.
ʂ comes from ʃ. Nothing special here.
ʔ is again a debuccalization of a coda consonant. This time it turns into a stop.
∅ is plain old coda consonant deletion.
t is a emergence of the unmarked change, where coda consonants change to the coronal PoA, because it's the unmarked one.
q is backing of *k adjacent to a non-high back vowels. Attested somewhere in (eastern or western) Africa. Keep in mind that the vowel change was pretty much unpredictable.
kʷ is rounding of velar plosive *k adjacent to round vowels. Pretty common. Keep in mind that the vowel change was pretty much unpredictable.
p comes from *kʷ, just like in Celtic.
c is a palatalization of *k adjacent to front vowels.
tʃ is a later development from c.
tl;dr: palatalization after front vowel yields {c,tʃ}, debuccalization yields {ʔ,h} and the rounded back nasal vowels trggers labialization/backing to {p,kʷ,q} and {t} is just the return to an unmarked PoA in the syllable coda.
My new guess is *zindra. I did some reconstruction and there are only very few unpredictable third vowels, so I guess the Proto-Word only had two.
16-18: hikaʎ
12-14: hirva
8-11: ɕwar
3-5,7,: zibar
1-2,6,24-25: zidar
19-21: hengola
23-24: sidra
I choose *d because I think d → ð →v is attested. Also I think the velars come from a rather long chain of sound changes dr → dl → tl → kl. This also fits nice with the palatal lateral.
Obviously I have to assume vowel insertion/epenthesis and metathesis at some point.
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
It was so interesting to read all of your guesses and theories! I didn't think you'd guess some changes right at the first try but then it was funny to see that some parts are quite off. The first change might be a bit mean so that's the most interesting thing to look at!
Here is a crowded map of all the isoglosses as a second hint. They're in the right order so 1 happened first, then 2, then 3 etc.
I'll also say that Creyeditor's guess is the furthest away from all your guesses and that there are no stops in the proto-word as well. Maybe that will help a bit more to come up with a closer example. The second vowel is not easy to guess just by looking at the daughter languages so I'll accept any vowel that's close to it (as some already come close enough I guess). Good luck!
Here is a crowded map of all the isoglosses as a second hint. They're in the right order so 1 happened first, then 2, then 3 etc.
Spoiler:
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
Well that means my initial guess with *g as the middle consonant is out.ixals wrote:It was so interesting to read all of your guesses and theories! I didn't think you'd guess some changes right at the first try but then it was funny to see that some parts are quite off. The first change might be a bit mean so that's the most interesting thing to look at!
Here is a crowded map of all the isoglosses as a second hint. They're in the right order so 1 happened first, then 2, then 3 etc.I'll also say that Creyeditor's guess is the furthest away from all your guesses and that there are no stops in the proto-word as well. Maybe that will help a bit more to come up with a closer example. The second vowel is not easy to guess just by looking at the daughter languages so I'll accept any vowel that's close to it (as some already come close enough I guess). Good luck!Spoiler:
I think if part of a word isn't really reconstructible to any level of certainty based on the available words, then all well-reasoned guesses are equally as acceptable, otherwise it just becomes pure guesswork if the original was say /i/.
It looks like /d/ and /ʁ/ happened quite late and are restricted to only a few examples.
The 1st sound change looks like it affected the middle consonant based on the East-West distinctions.
East has: ŋ g k ɾ r
West has: w v b (d)
Sound change 2 looks like it happened to the initial consonant based on the East-West distinction again. East has /h/ or no initial consonant.
Sound change 7 splits the Eastern group to those with /ŋ g k/ as medial and those with other.
9 is the loss of final vowel. So /hildɾa/ and /silr/ must come from *silra with the /d/ being epenthetic.
10 is obviously the umlaut sound change giving medial /i/.
/hildɾa/ is annoying because it looks like it participated in one major sound-change which differentiates East-West, but not another.
Ill update my guess to *ziɣʷala or *ziɣola. Maybe the middle consonant was part of a cluster such as /lɣ/ or /ɾɣ/.
I'm not entirely happy with my guess because /ɣ/ > /l/ requires quite a lot of intermediate steps. Maybe the ones with medial /l/ are more conservative, but I can't tell from this map.
Re: Quick Diachronics Challenge
Well, it is reconstructable, but not that easily just because the languages south of 4 shifted it to /a~ɐ/ and that's almost half of all the languages. The other languages might offer some more hints at the exact value of V2. But in the end, it could be /i/ (although it isn't) since only the daughter languages are known. Who knows, maybe the proto-word was /t͡ɬʼʌ̰˨kʲːr̩˦qɛə̯˥˩ʘ/?Davush wrote:I think if part of a word isn't really reconstructible to any level of certainty based on the available words, then all well-reasoned guesses are equally as acceptable, otherwise it just becomes pure guesswork if the original was say /i/.
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
Sometimes in these challenges (and real languages), the proto-form isn't 100% reconstructable from the available words either due to not enough reflexes, or sound changes which have been obscured due to layers of other sound changes. That's why I think the path has to be at least mostly reconstructibleixals wrote:Well, it is reconstructable, but not that easily just because the languages south of 4 shifted it to /a~ɐ/ and that's almost half of all the languages. The other languages might offer some more hints at the exact value of V2. But in the end, it could be /i/ (although it isn't) since only the daughter languages are known. Who knows, maybe the proto-word was /t͡ɬʼʌ̰˨kʲːr̩˦qɛə̯˥˩ʘ/?Davush wrote:I think if part of a word isn't really reconstructible to any level of certainty based on the available words, then all well-reasoned guesses are equally as acceptable, otherwise it just becomes pure guesswork if the original was say /i/.
Your example: if the medial vowel was /i/ in the proto-word, this wouldn't be reconstructable based on the data you have given us, because the only reasonable explanation for /i/ in the daughter languages is a few isolated umlaut-type things. There's no way we could arrive at /i/ other than just a random guess. If the proto-word was /t͡ɬʼʌ̰˨kʲːr̩˦qɛə̯˥˩ʘ/, then it's not really a diachronics challenge anymore...
Re: Quick Diachronics Challenge
So guessing at the nature of each isogloss:
The biggest problem for my reconstruction is V2. Given #3-5, the only unaffected form is 12, so I'm going to guess *ɤ is the value for V2. I might go through my reconstruction again and see if there's more that could do with updating later.
Edits: minor updates to the list of sound changes, mostly due to clerical errors.
Spoiler:
Edits: minor updates to the list of sound changes, mostly due to clerical errors.
Last edited by gufferdk on 21 Mar 2017 20:02, edited 3 times in total.
Warning: Anything I post may be ninja-edited up to half an hour after posting.
Re: Quick Diachronics Challenge
No, I only marked those that were shared between at least two languages. Changes that only affected one language, happened after all the isoglosses though, but only a few languages had a single change that seperated them a little bit more, but the older forms of those are quite easy to reconstruct imo. When comparing /heˈɔ̃.le/ with neighbouring words like /he.ŋwɛˈla/ and /iˈŋɔ.la/, the missing consonant in /heˈɔ̃.le/ is obvious I think.
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
Everything is obvious when you know the answer.ixals wrote:No, I only marked those that were shared between at least two languages. Changes that only affected one language, happened after all the isoglosses though, but only a few languages had a single change that seperated them a little bit more, but the older forms of those are quite easy to reconstruct imo. When comparing /heˈɔ̃.le/ with neighbouring words like /he.ŋwɛˈla/ and /iˈŋɔ.la/, the missing consonant in /heˈɔ̃.le/ is obvious I think.