(Conlangs) Q&A Thread - Quick questions go here

A forum for all topics related to constructed languages
User avatar
esoanem
hieroglyphic
hieroglyphic
Posts: 37
Joined: Tue 05 Sep 2017, 13:03
Location: Cambridge, UK
Contact:

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by esoanem » Mon 27 Nov 2017, 21:52

Wouldn't an alternative explanation for transparent etymologies in PIE but not later languages (alternative, or perhaps additional instead? to the replacement idea) simply be that we'd expect the default position to be transparent etymology (apart from a relatively small number of roots). At every stage in a language's development it has inherited forms with opaque etymology and new terms are coined with (mostly) transparent etymologies. Any etymology transparent in a daughter language will also be transparent in the parent language (with only a little bit of analogy or regularisation needed).

The key argument here is that a term with an etymology that's transparent in a parent language was not necessarily present in the parent. For example, the English word pontiff is from Latin pontifex which is transparently ponti-fex (ponti being the oblique stem of pōns, bridge) but -fex "-maker" itself is opaque but people might still notice the semantic relation to faciō "I do/make" and it's also opaque in Proto-Italic (in both cases the more naturally expected form would be *-faciō/*-fakiō similar to other nouns derived from the first person present active indicative of the verb) but becomes transparent in PIE with the both -fex and faciō apparently from *dhh1k- one with the ordinary nominal *-s and the other the relevant verb suffix.

So, none of it is transparent in English, part in Latin, and all in Latin (which also has a transparent etymology for pōns) but there's no reason whatsoever to believe the Indo-Europeans would have ever used the word **pónteh1dhh1ks.
My pronouns are they/them/their

:gbr: native | :esp: fluentish | :deu: learning | :fra: learning | :rus: learning | :ell: lapsed | :navi: lapsed | :con: making a bunch
Porphyrogenitos
cuneiform
cuneiform
Posts: 153
Joined: Sat 21 Jul 2012, 07:01
Location: Buffalo, NY

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Porphyrogenitos » Mon 27 Nov 2017, 22:47

esoanem wrote:
Mon 27 Nov 2017, 21:52
Wouldn't an alternative explanation for transparent etymologies in PIE but not later languages (alternative, or perhaps additional instead? to the replacement idea) simply be that we'd expect the default position to be transparent etymology (apart from a relatively small number of roots). At every stage in a language's development it has inherited forms with opaque etymology and new terms are coined with (mostly) transparent etymologies. Any etymology transparent in a daughter language will also be transparent in the parent language (with only a little bit of analogy or regularisation needed).

The key argument here is that a term with an etymology that's transparent in a parent language was not necessarily present in the parent. For example, the English word pontiff is from Latin pontifex which is transparently ponti-fex (ponti being the oblique stem of pōns, bridge) but -fex "-maker" itself is opaque but people might still notice the semantic relation to faciō "I do/make" and it's also opaque in Proto-Italic (in both cases the more naturally expected form would be *-faciō/*-fakiō similar to other nouns derived from the first person present active indicative of the verb) but becomes transparent in PIE with the both -fex and faciō apparently from *dhh1k- one with the ordinary nominal *-s and the other the relevant verb suffix.

So, none of it is transparent in English, part in Latin, and all in Latin (which also has a transparent etymology for pōns) but there's no reason whatsoever to believe the Indo-Europeans would have ever used the word **pónteh1dhh1ks.
Well, I think the issue at hand is that most of the vocabulary of English - the basic, everyday vocabulary, that is, the words that see the most frequent use - isn't transparent. E.g. not a single item on the English Swadesh list has a transparent etymology; all these words refer to basic concepts shared by virtually all human cultures and have been around for a very long time. Similarly, as Sapir points out:
Edward Sapir wrote: Contrast, for instance, the extensive and highly distinctive vocabulary concerned with the breeding and use of cattle (cow, ox, bull, steer, heifer, calf, cattle, beef, veal, butter, cheese, whey, curds, cream, to churn, to skim—all unanalysable terms of evidently considerable age) with the more meagre and less distinctive vocabulary of such an industry as, say, the growing of oranges. Linguistic evidence alone would make out a strong case for the greater age of cattle breeding and the dairy industry than of orange growing.
In fact, I'm not quite sure how your view differs from what I stated, aside from the point about most common vocabulary in English being non-transparent. Of course PIE speakers didn't have the term pontifex; the office of pontifex did not exist in their culture. Languages have different layers of vocabulary showing different degrees of transparency depending on their age; the oldest terms (save for loanwords) are the least transparent, typically referring to universal human concepts and very old cultural practices like (in English) basic agriculture and animal husbandry. We can see words in various stages of non-transparency, ranging from totally unanalyzable in the present day (bear, wolf), to showing hints of past compounding or derivation (woman, hussy), to having semi-transparent derivations from obsolete morphemes (werewolf, cranberry, spinster), to having only the beginning of the phonological changes that lead to non-transparency (blackguard/blaggard, fridge (still clearly connected to refrigerator and frigid - but for how long?), to totally transparent (night shift manager, defroster).
User avatar
sangi39
moderator
moderator
Posts: 2663
Joined: Thu 12 Aug 2010, 00:53
Location: North Yorkshire, UK

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by sangi39 » Mon 27 Nov 2017, 23:09

Porphyrogenitos wrote:
Mon 27 Nov 2017, 22:47
esoanem wrote:
Mon 27 Nov 2017, 21:52
Wouldn't an alternative explanation for transparent etymologies in PIE but not later languages (alternative, or perhaps additional instead? to the replacement idea) simply be that we'd expect the default position to be transparent etymology (apart from a relatively small number of roots). At every stage in a language's development it has inherited forms with opaque etymology and new terms are coined with (mostly) transparent etymologies. Any etymology transparent in a daughter language will also be transparent in the parent language (with only a little bit of analogy or regularisation needed).

The key argument here is that a term with an etymology that's transparent in a parent language was not necessarily present in the parent. For example, the English word pontiff is from Latin pontifex which is transparently ponti-fex (ponti being the oblique stem of pōns, bridge) but -fex "-maker" itself is opaque but people might still notice the semantic relation to faciō "I do/make" and it's also opaque in Proto-Italic (in both cases the more naturally expected form would be *-faciō/*-fakiō similar to other nouns derived from the first person present active indicative of the verb) but becomes transparent in PIE with the both -fex and faciō apparently from *dhh1k- one with the ordinary nominal *-s and the other the relevant verb suffix.

So, none of it is transparent in English, part in Latin, and all in Latin (which also has a transparent etymology for pōns) but there's no reason whatsoever to believe the Indo-Europeans would have ever used the word **pónteh1dhh1ks.
Well, I think the issue at hand is that most of the vocabulary of English - the basic, everyday vocabulary, that is, the words that see the most frequent use - isn't transparent. E.g. not a single item on the English Swadesh list has a transparent etymology; all these words refer to basic concepts shared by virtually all human cultures and have been around for a very long time. Similarly, as Sapir points out:
Edward Sapir wrote: Contrast, for instance, the extensive and highly distinctive vocabulary concerned with the breeding and use of cattle (cow, ox, bull, steer, heifer, calf, cattle, beef, veal, butter, cheese, whey, curds, cream, to churn, to skim—all unanalysable terms of evidently considerable age) with the more meagre and less distinctive vocabulary of such an industry as, say, the growing of oranges. Linguistic evidence alone would make out a strong case for the greater age of cattle breeding and the dairy industry than of orange growing.
In fact, I'm not quite sure how your view differs from what I stated, aside from the point about most common vocabulary in English being non-transparent. Of course PIE speakers didn't have the term pontifex; the office of pontifex did not exist in their culture. Languages have different layers of vocabulary showing different degrees of transparency depending on their age; the oldest terms (save for loanwords) are the least transparent, typically referring to universal human concepts and very old cultural practices like (in English) basic agriculture and animal husbandry. We can see words in various stages of non-transparency, ranging from totally unanalyzable in the present day (bear, wolf), to showing hints of past compounding or derivation (woman, hussy), to having semi-transparent derivations from obsolete morphemes (werewolf, cranberry, spinster), to having only the beginning of the phonological changes that lead to non-transparency (blackguard/blaggard, fridge (still clearly connected to refrigerator and frigid - but for how long?), to totally transparent (night shift manager, defroster).
I think I'd agree with this. IIRC, as a whole, the majority of "stems" in, for example, English (even excluding borrowings) are derived but the majority of the most common, daily-used "stems" are underived "roots" (to use PIE terminology).

As Porphyrogenitos suggests, it's most often the case that the more basic the vocabulary the less likely it is to be derived and the more likely derived vocabulary is going to refer to less basic things or ideas. The less basic something is, the more likely the derivation is to be transparent, with time allowing for semantic drift after the term is coined.

At least with the Indo-European languages, what you'll then find is that the apparently "non-derived" roots, i.e. those roots that cannot be derived by any means in the modern form of the language, go back to derived stems. The result of this is that what we reconstruct of PIE shows a higher prevalence of derived stems for "basic" concepts than non-derived stems. The two conclusions here are a) PIE actually just did use derivation more*, or b) more and more of PIE's non-derived roots have been lost as the family has developed.
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.
Davush
sinic
sinic
Posts: 410
Joined: Sat 10 Jan 2015, 14:10

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Davush » Tue 28 Nov 2017, 11:03

Thanks - that has helped clear up some of the questions I had. I think I was assuming since so many reconstructed words were seemingly transparent derivations, PIE must have used derivation a lot more, but I suppose many words are just lost to history. I am planning on working backwards and making a proto-language, so this should help. I'll just have to make some words, and then 'lose' them. [:D]

I wonder if there are any reasons why a language might use more (or less) derivation, though. English probably seems to have less because of extensive borrowing, but Chinese for example makes great use of compounding, possibly because it has borrowed less from surrounding languages? Arabic also has derivation which seems far greater and more frequent than that of English.
Davush
sinic
sinic
Posts: 410
Joined: Sat 10 Jan 2015, 14:10

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Davush » Tue 28 Nov 2017, 17:14

Trying to work backwards and create a sketch of Proto-Qutrussan but a few sounds are tripping me up. How might I derive /dz/ from the following proto phonology:

pʰ tʰ kʰ
p’ t’ k’ ʔ
p t k
f s s’ ʃ ħ ʕ
m n ŋ
r l w j

a i u ɛ e ɔ o (+ length except /ɛ ɔ/)

Bearing in mind that /s'/ > /ts/ and /ts/ occurs between vowels in Qutrussan. I was thinking maybe /st'/ > /zd/ > /dz/ or similar, but other ideas would be welcome. It appears word initially, but I was thinking of having voiced consonants initially derived from the unaspirated series with a loss of previous unstressed vowel. I don't really want to alter the proto phonology too much at this stage and I want */ʕ/ as the only voiced fricative (or it might pattern more like ejective /ʔ/). I particularly like strange unexpected sound changes (like development of Armenian erk) and would like to try and incorporate one or two... [:D]
User avatar
sangi39
moderator
moderator
Posts: 2663
Joined: Thu 12 Aug 2010, 00:53
Location: North Yorkshire, UK

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by sangi39 » Tue 28 Nov 2017, 17:35

Davush wrote:
Tue 28 Nov 2017, 17:14
Trying to work backwards and create a sketch of Proto-Qutrussan but a few sounds are tripping me up. How might I derive /dz/ from the following proto phonology:

pʰ tʰ kʰ
p’ t’ k’ ʔ
p t k
f s s’ ʃ ħ ʕ
m n ŋ
r l w j

a i u ɛ e ɔ o (+ length except /ɛ ɔ/)

Bearing in mind that /s'/ > /ts/ and /ts/ occurs between vowels in Qutrussan. I was thinking maybe /st'/ > /zd/ > /dz/ or similar, but other ideas would be welcome. It appears word initially, but I was thinking of having voiced consonants initially derived from the unaspirated series with a loss of previous unstressed vowel. I don't really want to alter the proto phonology too much at this stage and I want */ʕ/ as the only voiced fricative (or it might pattern more like ejective /ʔ/). I particularly like strange unexpected sound changes (like development of Armenian erk) and would like to try and incorporate one or two... [:D]
Does geminate [j:] occur at all in the proto-language?
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
gestaltist
roman
roman
Posts: 1452
Joined: Wed 11 Feb 2015, 11:23

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by gestaltist » Tue 28 Nov 2017, 17:39

Davush wrote:
Tue 28 Nov 2017, 17:14
Trying to work backwards and create a sketch of Proto-Qutrussan but a few sounds are tripping me up. How might I derive /dz/ from the following proto phonology:

pʰ tʰ kʰ
p’ t’ k’ ʔ
p t k
f s s’ ʃ ħ ʕ
m n ŋ
r l w j

a i u ɛ e ɔ o (+ length except /ɛ ɔ/)

Bearing in mind that /s'/ > /ts/ and /ts/ occurs between vowels in Qutrussan. I was thinking maybe /st'/ > /zd/ > /dz/ or similar, but other ideas would be welcome. It appears word initially, but I was thinking of having voiced consonants initially derived from the unaspirated series with a loss of previous unstressed vowel. I don't really want to alter the proto phonology too much at this stage and I want */ʕ/ as the only voiced fricative (or it might pattern more like ejective /ʔ/). I particularly like strange unexpected sound changes (like development of Armenian erk) and would like to try and incorporate one or two... [:D]
I seem to remember the change *j > dz being reconstructed for some language group. If you can restrict it to some environment or regenerate /j/ from something else (diphthongs perhaps), that would be a quite cool way to derive /dz/, I think.
Salmoneus
MVP
MVP
Posts: 1236
Joined: Mon 19 Sep 2011, 18:37

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Salmoneus » Tue 28 Nov 2017, 20:01

Porphyrogenitos wrote:
Mon 27 Nov 2017, 22:47
In fact, I'm not quite sure how your view differs from what I stated.
I think their point is that just because we can reconstruct a transparent term to a parent doesn't mean that parent actually HAD that term. So not only do the transparent terms often survive when the opaque ones don't, but also some of the transparent words we reconstruct may not necessarily have actually existed in the parent. "Nutrigenic phosphorescence", probably does not descend from an actual phrase in Latin.

So, some of the transparent vocabulary in reconstructed PIE may just be parallel derivation of vocabulary in daughter languages - since it's transparent, a lot of it isn't surpising stuff. [This may also explain why some terms in PIE have such vague, thematic definitions - it may be that formally equivalent, straightforward derivatives with different meanings in the daughter languages were actually innovated in parallel and not present in PIE.

For instance, take *g'en1tis. This yields derivatives meaning "source" (in Greek), "son-in-law" (Slavic), "race" (Germanic), or "production" (Indic). Maybe in PIE it was a word that meant "the process of birthing and/or that which is birthed, whether literally or metaphorically". Or maybe it was never a word in PIE at all (even though we can reconstruct it), and each daughter just innovated it - not a startling innovation given g'en1- and the common suffix -tis - and they innovated it for different purposes.
Davush
sinic
sinic
Posts: 410
Joined: Sat 10 Jan 2015, 14:10

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Davush » Tue 28 Nov 2017, 23:07

Thanks! Qutrussan doesn’t have /jj/ so having that as a source for /dz/ fits very well!

Also thanks Salmoneus for the explanation.
User avatar
Vlürch
cuneiform
cuneiform
Posts: 96
Joined: Wed 09 Mar 2016, 21:19
Location: Finland

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Vlürch » Wed 29 Nov 2017, 10:14

I have a few questions about the names of grammatical cases:
1) What is the case called that denotes movement to the back of something? Eg. "to the back of the house"
2) What is the case called that denotes movement to the front of something? Eg. "to the front of the house"
3) What is the case called that denotes movement from the back of something? Eg. "from behind the house"
4) What is the case called that denotes movement from the front of something? Eg. "from the front of the house"

By analogy of antessive and postessive, the first two should be something like antelative and postlative, and there are a few results on Google that state those are exactly the right terms as far as I can tell, so logically the third one is postelative. For that, there's even a Wikipedia article, so it should be good to call it that. Still, it'd be nice if someone could confirm they're right, or at least if they make sense, and if not, tell me what the correct names are or at least suggest what some names that make sense would be.

The last one, though... wtf is it? I've tried googling everything I can think of but haven't been able to find anything that would give a name to that case except for anteelative, but that just sounds weird and there's only one result for it, so it's probably not right; I mean, even the most hypothetical things always have at least a few results, never just one... and even though that one result isn't conlang-related, it's obviously still just a practical name for the case for the purpose of describing that one language since there are no other results. I also found this list of cases that calls it excislative, but it doesn't seem like that term has ever been used anywhere and it has some weird ass stuff instead of well-established terminology so I don't really trust it.

If there were any results for antlative that weren't on a random spam site, it could make sense to call the second one that and then the last one antelative, but since there's only one result on Google for the former and it seems to be on a random spam site, and there are results calling the second one antelative... well... you know, using the same term for the opposite of what it's used for by others would be really annoying and I don't want to be intentionally annoying when I'm annoying enough already when I'm trying not to be annoying. [:P]

Why is conlanging so painful sometimes.
User avatar
Pabappa
hieroglyphic
hieroglyphic
Posts: 41
Joined: Sat 18 Nov 2017, 02:41
Contact:

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Pabappa » Wed 29 Nov 2017, 15:32

You could use -ablative as the suffix instead of -elative , and therefore prevent vowel elision .... and this preserves nearly the same meaning as the original.
Image
Davush
sinic
sinic
Posts: 410
Joined: Sat 10 Jan 2015, 14:10

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Davush » Wed 29 Nov 2017, 16:37

What’s the view on uncomplete/unstable/sporadic sound changes? For example I want /o/ to become /a/ at the end some words and remains /a/ in others but there aren’t really any conditioning environments, it seems more or less random. I know some changes can stop before they are fully complete throughout the lexicon but that doesn’t seem a good explanation either since it’s all common words. Are there any attestations of sound changes which aren’t very regular or aren’t applied consistently?

(I have so many questions recently... [:D] )
User avatar
esoanem
hieroglyphic
hieroglyphic
Posts: 37
Joined: Tue 05 Sep 2017, 13:03
Location: Cambridge, UK
Contact:

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by esoanem » Wed 29 Nov 2017, 17:29

You could potentially justify it as having two intermediate varieties, one continuing final /o/ and the other unrounding it to /a/ and then have the final variety be some sort of mixed lect, either descending from one and borrowing some words from the other, or potentially as a full-on koiné of the multiple intermediates.

These are both only appropriate in certain historical contexts so may not be useful to you and, in each case, you probably ought to try and think about which words have which reflex rather than have it be truly random.
My pronouns are they/them/their

:gbr: native | :esp: fluentish | :deu: learning | :fra: learning | :rus: learning | :ell: lapsed | :navi: lapsed | :con: making a bunch
User avatar
Creyeditor
mongolian
mongolian
Posts: 3948
Joined: Tue 14 Aug 2012, 18:32

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Creyeditor » Wed 29 Nov 2017, 17:54

Vlürch wrote:
Wed 29 Nov 2017, 10:14
I have a few questions about the names of grammatical cases:
1) What is the case called that denotes movement to the back of something? Eg. "to the back of the house"
2) What is the case called that denotes movement to the front of something? Eg. "to the front of the house"
3) What is the case called that denotes movement from the back of something? Eg. "from behind the house"
4) What is the case called that denotes movement from the front of something? Eg. "from the front of the house"
Grammar writers mostl come up with names for new cases on the spot, based on some (more or less) Pseudo-Latin. So no pressure for conlangers to do any better. Martin Haspelmath tried to establish some systematicty in naming locative cases, but it hasn't really caught on.

Code: Select all

         'in'          'on'          'at'
location  in-essive   super-essive   ad-essive
goal      in-allative super-allative ad-allative
source    in-ablative super-ablative ad-ablative
If you added yours to this you would probably get: post-allative, ante-allative, post-ablative and ante-ablative. That looks like a reasonable solution to me. Another option, that I might even prefer, is to just treat these cases as if they were adpositions , less fnctional, more lexical. You could just say that these are the adpositions expressing 'to behind', 'to front' 'from behind' and 'from front' and state that they are bound morphemes.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :fra: 4 :esp: 4 :ind:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
User avatar
gestaltist
roman
roman
Posts: 1452
Joined: Wed 11 Feb 2015, 11:23

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by gestaltist » Wed 29 Nov 2017, 18:53

Davush wrote:
Wed 29 Nov 2017, 16:37
What’s the view on uncomplete/unstable/sporadic sound changes? For example I want /o/ to become /a/ at the end some words and remains /a/ in others but there aren’t really any conditioning environments, it seems more or less random. I know some changes can stop before they are fully complete throughout the lexicon but that doesn’t seem a good explanation either since it’s all common words. Are there any attestations of sound changes which aren’t very regular or aren’t applied consistently?

(I have so many questions recently... [:D] )
My go-to handwavium is saying that some part of vocabulary or some specific feature was borrowed from a non-dominant dialect if I want to have sporadic sound change.
User avatar
Pabappa
hieroglyphic
hieroglyphic
Posts: 41
Joined: Sat 18 Nov 2017, 02:41
Contact:

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Pabappa » Wed 29 Nov 2017, 19:39

Davush wrote:
Wed 29 Nov 2017, 16:37
What’s the view on uncomplete/unstable/sporadic sound changes? For example I want /o/ to become /a/ at the end some words and remains /a/ in others but there aren’t really any conditioning environments, it seems more or less random. I know some changes can stop before they are fully complete throughout the lexicon but that doesn’t seem a good explanation either since it’s all common words. Are there any attestations of sound changes which aren’t very regular or aren’t applied consistently?

(I have so many questions recently... [:D] )
Sporadic sound change is one of my favorite things to do with a conlang.

I have nearly the exact same sound change in Pabappa, where a word-final /-o/ always becomes /-a/ unless preceded by another /o/ earlier in the word. Since /o/ was rare to begin with, this exception is not triggered very often. Nevertheless, many Pabappa roots do end in /-o/, and most of them do not have the conditioning environment of another /o/ earlier in the word. There was no subsequent sound change creating a new final /-o/, so why do these words exist? The answer is: grammar.

In Pabappa, all word-final /-o/ indeed became /-a/, as stated above, with a tiny number of exceptions. However, this sound change effectively applied only to nouns, because verbs in Pabappa never occurred in bare form: they always either had an inflection on, or occurred as the first part of a compound word. In either case, the stem-final /o/ was never exposed to the environment that triggered the sound change, and was therefore preserved. The sound change *did* apply to stems which were commonly used to form both nouns and verbs; for example pusa "host, caretaker; to hug, greet". This came from an earlier puso-, which first split into two stems because of the shift, and then later merged into the shifted form pusa by analogy.

Later on, after the sound change had ceased operation, some Pabappa verb stems came to be used for nouns, of which the nominative case always had a zero-morph marker. These had been spared from the shift because at the time, they had never been used to form nouns.
Therefore, some Pabappa nouns came to end in /-o/ again.

-------

In other conlangs, I do similar things. In early Poswa, a verb could end in any of the six vowels. At one point, though, a sound change took place that caused all of the verbs to collapse into what scholars called /i/-stems and /u/-stems. (These names were of little meaning to the speakers because those vowels were always elided; however, they had distinct effects on surrounding phonemes.)

Later on, some of the consonant stems also came to end in vowels, which set up a new inflection paradigm for verbs ending in /i/ and /u/. Some of the original vowel stems then switched to this paradigm, whose inflected forms were usually one syllable longer, while others remained. Some of those that remained were reanalyzed as /a/-stems, because /a/ had been the commonest vowel all along and was still present in nouns transparently related to many of these verb stems. Since the /a/-stem conjugation was a syllable shorter than all others, the commonest verbs tended to be reanalyzed as /a/-stems, even in some cases when there was no etymologically related noun to analogize with. Thus, for example, the inherited verb stem mi- "see" became ma-, whose /a/ disappears in all inflected forms of the verb.

----
With Khulls I do the same thing yet again, but this time the change is internal to the noun stem, and does not rely on verb inflection. Here, the /a/-stem and /ə/-stem nouns came to merge in most of their inflected forms early in the history of the language. Later, Khulls evolved a new set of affixes beginning with -/w/, which attached readily to the inherited final /ə/, and soon eliding it; but had no affected on inherited final /a/, historically a much more resistant vowel in Khulls diachronics. Thus yet again the desire to save a syllable caused many, but not all, /a/-stem nouns to be reanalyzed as /ə/-stems, and those most likely to change were those seeing the most frequent use.

-------

Not all of my ideas would work in every language. I think the Pabappa setup is the most straightforward, and the most similar to the one you're looking for. It could be used in any language that has one class of words that is sometimes used in bare form (thus triggering the sound change), and another, "sheltered" class that never is (thus preventing the sound change). Once the sound change is over the protected words can be freely used again.

Almost certainly, some true sporadic change could accompany this conditioned shift, but I've yet to need this in my own work because I've never detailed my languages down to the level of dictating which words can and cannot be used outside of their listed part of speech, or likewise for the triggering conditions of the other changes.
Image
Davush
sinic
sinic
Posts: 410
Joined: Sat 10 Jan 2015, 14:10

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Davush » Wed 29 Nov 2017, 21:56

Thanks - the Pabappa system is a nice way to do it, but I don't think Qutrussan had such inflectional categories (if you can call them that in Pabappa?). I know you mention in Khulls that the most commonly used words were subject to change more than others - I had always thought it was the other way round, i.e. the more commonly used the word, the more likely it is to resist analogy/levelling? (Of course that isn't to say the diachronics for Khulls couldn't lead to what you have described, my question is more general.)

@Gestaltist: Thanks, I was thinking of going down the dialectal borrowings path, but I don't really want to as Qutrussan (or even Proto-Qutrussan) is supposed to become the dominant dialect pretty early in its history, and so most borrowing will be from it.

I guess I was asking if sporadic change is attested in the sound change history of any language to a significant degree? It seems it's commonly presented as if sound change works robotically, but when you look at a 'case study' of say PIE to Latin, you find many strange and unexpected 'twists'. I don't know if these can really be considered wholesale sporadic changes though?
User avatar
esoanem
hieroglyphic
hieroglyphic
Posts: 37
Joined: Tue 05 Sep 2017, 13:03
Location: Cambridge, UK
Contact:

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by esoanem » Thu 30 Nov 2017, 00:30

English has a few with the reflex of OE /æ/, <sc> /ʃ/, and <ċ> /tʃ/ (respectively as both /æ/ & /a/, /ʃ/ & /sk/, and /k/ & /tʃ/). The situation here is fairly well understood as a case of multiple intermediate lects. In particular, the OE standard was West Saxon; Modern English descends primarily from an Anglian variety (most likely Mercian) but features many words with West Saxon forms (with unexpected reflexes of /æ/) and Norse-influence Northumbrian forms (with unexpected reflexes of <sc> & <ċ>).

Latin also has a few. French and Spanish (the two I'm familiar with) both show a stage of 'confusion' between <v> and <b> with the two getting shuffled moreorless randomly between the two regardless of their classical value. Likewise, the degree of lenition of intervocalic plosives varies. We have a chain shift [p t c] -> -> Ø but there are words where some consonants make no jumps, or two, or just one, I don't think there's any suggestion of conditioning here instead though.
My pronouns are they/them/their

:gbr: native | :esp: fluentish | :deu: learning | :fra: learning | :rus: learning | :ell: lapsed | :navi: lapsed | :con: making a bunch
User avatar
Ahzoh
korean
korean
Posts: 5997
Joined: Sun 20 Oct 2013, 01:57
Location: Tom-ʾEzru lit Yat-Vṛḵažu

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Ahzoh » Thu 30 Nov 2017, 05:12

What are ways to develop word-initial geminates? I asked this question on the other board and was told it could arise from elision of first syllables like in Malayalam, vowel syncope, and this thing Japanese did to get "tte" from "to iu".
But what are other ways, especially involving hetero-organic clusters?

Also, is /s/ > /a/ ever possible like "ism(a)" > "iam(a)"?
Image Ӯсцӣ (Onschen) [ CWS ]
Image ʾEšd Yatvṛḵažaẇ (Vrkhazhian) [ WIKI | CWS ]
Porphyrogenitos
cuneiform
cuneiform
Posts: 153
Joined: Sat 21 Jul 2012, 07:01
Location: Buffalo, NY

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Porphyrogenitos » Thu 30 Nov 2017, 06:08

Well, pretty much by any other way you get geminates, but just at the start of a syllable.

Deletion of a word-initial vowel before a geminate:

attama > ttama

Deletion of vowels (due to whatever reason) between two already identical consonants:

titama > ttama

Assimilation of a cluster and then deletion of an initial vowel:

aktama > attama > ttama

Assimilation of a word-initial cluster:

ktama > ttama
mbama > bbama
rtomo > rromo

Development of initial geminates in certain syntactic contexts due to word-initial mutations resulting from sandhi effects:

ad kasa > ak‿kasa
ad > a
ak‿kasa now reanalyzed as a kkasa

Then musu, coined centuries after a lost its final /d/, becomes mmusu after a because it's a regular morphosyntactic process now

And I think s > a would be plausible if /s/ was shifted to /h/, which is both possible and extremely common. From there coda /h/ could vocalize to /a/, especially if it first became [ɦ]
Post Reply