Explaining sound change?

Discussion of natural languages, or language in general.
Post Reply
Grey
Sanci
Sanci
Posts: 17
Joined: Mon Jun 25, 2007 10:41 pm
Location: Europe

Explaining sound change?

Post by Grey »

Hello all, it's been a while since I've been to the ZBB; college got in the way and the military keeps me fairly busy. In that time, conlanging has fallen to the wayside, and though I dabbled in reading some interesting grammars here and there, I haven't managed to properly understand sound changes.
More specifically: I want to understand how to apply sound change to conlanging. I have both the LCK and the ALC, and have read the wiki article several times, so I might just be overthinking it. Regardless, I have a few questions for you fine people if you happen to have the time to spare.

1: I understand that sound changes will occur when criteria is met ( A = B under conditions C always). How complex can these conditions be? Most examples I've seen are pretty straightforward, but I'm no expert.

2: This is more of a personal opinion, but how many rules should I be looking at using at one time with the SCA? Do sound changes typically trigger further sound changes? If for instance, I include a sound change similar to the Anglo-Frisian law, can I expect further shifts to happen in some fashion? This is really where my trouble understanding comes from.

3: Would it be semi-realistic to place new roots into the SCA and apply later changes to them? For clarification: I'd like to intersperse new roots into the language to mimic borrowing from another language, is it realistic to skip a set of sound changes, and apply later ones to them?

4: Are there more/less common sound changes that occur? If so, where can I read about them? I am aware of the wiki for historic examples, but I'm asking more for general trends and I don't expect an in-depth answer to this, but perhaps your favourite sound changes for inspiration.
[quote="bricka"]
You aren't even using "as the LORD intended" as the LORD intended.[/quote]

User avatar
Zaarin
Smeric
Smeric
Posts: 1136
Joined: Sun Aug 15, 2010 5:00 pm

Re: Explaining sound change?

Post by Zaarin »

1. Certainly some sound changes can have complex environments, which means they'll be more difficult to reconstruct. Any case when you're reading a grammar that says "the environment of this sound change is not yet understood," the environment is probably complex. Can't think of any specific example at the moment, though.

3. Yes. One way to date loanwords is to note what sound changes have and have not affected them.

4. A few off the top of my head...Palatalization is extremely common (e.g., T/S/_Y, where T=dental consonant, S=sibilant fricative or affricate, and Y=front vowel). Loss of labiovelars (kʷ/k/_) is extremely common. Intervocalic lenition is extremely common (T/D/V_V, where T=voiceless stop, D=voiced stop; D/Ð/V_V, where D=voiced stop, Ð=voiced fricative). Loss or debuccalization of dorsal consonants (uvulars, pharyngeals, glottals) is very common. /n l r ɹ z/ all seem pretty interchangeable, with sound changes in many different directions happening in different languages. Combining these, I have one language with the shift s/z/V_[VB] (B=voiced consonant), then z/r/_ (with caveats--it also becomes /l/ in certain environments).
"But if of ships I now should sing, what ship would come to me,
What ship would bear me ever back across so wide a Sea?”

----
Smeric
Smeric
Posts: 1418
Joined: Tue Feb 15, 2011 11:15 pm

Re: Explaining sound change?

Post by ---- »

1. We should at least define complexity if we're going to have a formal discussion about the topic here. We might just say "complexity" in a sound change is merely how many parameters are necessary to trigger it, with parameters being the presence of certain segments, or location in the word, or presence of certain suprasegmental phenomena, etc. Sound changes can be triggered by several of these things at the same time. In some Sino-Tibetan languages for instance, the ultimate rhyme is produced by a combination of certain initial consonants being present, *and* certain final consonants being present in the proto-form.

2. Sound changes do often trigger other sound changes, especially when one sound change "puts pressure" on the phonemic structure of the language. E.g., if a sound change causes one phoneme to have many features in common with another, this can very likely encourage either a complete merger or a further split in a different direction in order to preserve a distinction that's felt to be valuable.

Sumelic
Avisaru
Avisaru
Posts: 385
Joined: Sat Mar 28, 2015 7:05 pm

Re: Explaining sound change?

Post by Sumelic »

Grey wrote:Hello all, it's been a while since I've been to the ZBB; college got in the way and the military keeps me fairly busy. In that time, conlanging has fallen to the wayside, and though I dabbled in reading some interesting grammars here and there, I haven't managed to properly understand sound changes.
More specifically: I want to understand how to apply sound change to conlanging. I have both the LCK and the ALC, and have read the wiki article several times, so I might just be overthinking it. Regardless, I have a few questions for you fine people if you happen to have the time to spare.

1: I understand that sound changes will occur when criteria is met ( A = B under conditions C always). How complex can these conditions be? Most examples I've seen are pretty straightforward, but I'm no expert.
As far as I know, conditions on sound changes can be quite restrictive, but I think you're right that they're usually straightforward. I think there usually are some conditions, though: I feel like conlangers often have too many unconditional sound changes of the format "phoneme A > phoneme B". Also, there can be isolated exceptions affecting specific words (though there still may be some semi-regular pattern to these exceptions). Two common examples of sound change from the history of English, Grimm's Law and the Great Vowel Shift, were both conditional: Wikipedia summarizes the conditions for Grimm's Law as "When two obstruents occurred in a pair, the first was changed according to Grimm's law, if possible, while the second was not," and the Great Vowel Shift did not affect Middle English /uː/ before labial consonants.
2: This is more of a personal opinion, but how many rules should I be looking at using at one time with the SCA? Do sound changes typically trigger further sound changes? If for instance, I include a sound change similar to the Anglo-Frisian law, can I expect further shifts to happen in some fashion? This is really where my trouble understanding comes from.
In general, some kinds of sounds in some kinds of positions seem to be more "stable" than others. (I don't know how well this holds up on a global scale, but it definitely seems this way for specific areas or languages.) For example, in English, word-initial nasal consonants like "n" and "m" have been fairly stable over time. But the nasal vowels resulting from the Anglo-Frisian nasal spirant law were unstable in the English phonological system and so the nasalization ended up being lost. If a sound change creates an "unstable" sound, then further sound changes are likely.
3: Would it be semi-realistic to place new roots into the SCA and apply later changes to them? For clarification: I'd like to intersperse new roots into the language to mimic borrowing from another language, is it realistic to skip a set of sound changes, and apply later ones to them?
Definitely. That's basically just the reverse of the process etymologists use to identify borrowings in real life.
4: Are there more/less common sound changes that occur? If so, where can I read about them? I am aware of the wiki for historic examples, but I'm asking more for general trends and I don't expect an in-depth answer to this, but perhaps your favourite sound changes for inspiration.
Hmm. This seems to me like another place where areal effects might be important. I'm familiar with what kinds of sound changes are common in European languages, but not so much languages from other continents. But there are certain changes that seem to be common worldwide and in all eras, such as palatalization of coronal or velar consonants before /j/ or /i/ and place assimilation of nasal consonants to a following plosive. For vowels, I guess coalescence of the diphthong /ai/ to /ɛ~e/ and/or /au/ to /ɔ~o/ is also pretty common. You might enjoy looking at the searchable version of the sound change database Index Diachronica.

User avatar
Zaarin
Smeric
Smeric
Posts: 1136
Joined: Sun Aug 15, 2010 5:00 pm

Re: Explaining sound change?

Post by Zaarin »

Sumelic wrote:I feel like conlangers often have too many unconditional sound changes of the format "phoneme A > phoneme B".
I'd say this might be regional. For example, Indo-European sound changes seem to involve a lot of conditionals, but Semitic (and I believe general Afro-Asiatic) languages have a lot of unconditioned sound changes--and even many of the conditional ones get leveled by analogy.
"But if of ships I now should sing, what ship would come to me,
What ship would bear me ever back across so wide a Sea?”

User avatar
KathTheDragon
Smeric
Smeric
Posts: 2139
Joined: Thu Apr 25, 2013 4:48 am
Location: Brittania

Re: Explaining sound change?

Post by KathTheDragon »

Actually, I'd expect the actual Semitic sound changes to be structurally similar to the Indo-European ones. The apparent exceptionlessness is entirely explicable by analogy, which can be attributed to the triconsonantal system. After all, if a conditional change obscures the root or the pattern, something is going to be generalised to restore the transparency.

User avatar
Zaarin
Smeric
Smeric
Posts: 1136
Joined: Sun Aug 15, 2010 5:00 pm

Re: Explaining sound change?

Post by Zaarin »

KathTheDragon wrote:Actually, I'd expect the actual Semitic sound changes to be structurally similar to the Indo-European ones. The apparent exceptionlessness is entirely explicable by analogy, which can be attributed to the triconsonantal system. After all, if a conditional change obscures the root or the pattern, something is going to be generalised to restore the transparency.
Fair point.
"But if of ships I now should sing, what ship would come to me,
What ship would bear me ever back across so wide a Sea?”

User avatar
Qwynegold
Smeric
Smeric
Posts: 1606
Joined: Thu May 24, 2007 11:34 pm
Location: Stockholm

Re: Explaining sound change?

Post by Qwynegold »

Everyone has already answered these, but I'll provide my answers.
Grey wrote:1: I understand that sound changes will occur when criteria is met ( A = B under conditions C always). How complex can these conditions be? Most examples I've seen are pretty straightforward, but I'm no expert.
SCs can vary all the way from no conditions at all, to so complex conditions that very few words are affected by them. I think it's best to use a combination of SCs of different complexity.
Grey wrote:2: This is more of a personal opinion, but how many rules should I be looking at using at one time with the SCA? Do sound changes typically trigger further sound changes? If for instance, I include a sound change similar to the Anglo-Frisian law, can I expect further shifts to happen in some fashion? This is really where my trouble understanding comes from.
Yes. Languages tend to not want odd gaps in the phoneme inventory, but they also like to keep phonemes as distinct as possible, at the same time as the language should be easy to pronounce. There's lots of different things that shape the direction SCs take. So a language goes through a SC to fix a problem, but when the SC has been applied, it creates new problems instead, so you have more SCs trying to fix the new problems, which create yet new problems. I can't tell how many SCs you should have, but generally the more SCs you have the more the language changes. But it also depends a lot on what changes you have.
Grey wrote:3: Would it be semi-realistic to place new roots into the SCA and apply later changes to them? For clarification: I'd like to intersperse new roots into the language to mimic borrowing from another language, is it realistic to skip a set of sound changes, and apply later ones to them?
Yes. A word borrowed in a later stage of a language doesn't go through the old SCs. But you should also try to think in the perspective of your conpeople. When a word is borrowed, it's usually assimilated somehow. It can either be completely assimilated, so it completely confines to the phoneme inventory and phonological constraints of the recipiet language, or it can be only partially assimilated.
Grey wrote:4: Are there more/less common sound changes that occur? If so, where can I read about them? I am aware of the wiki for historic examples, but I'm asking more for general trends and I don't expect an in-depth answer to this, but perhaps your favourite sound changes for inspiration.
Some already mentioned palatalization and homorganic nasals. Two more examples are nasalization on vowels caused by following nasal consonants, and some nasalized vowels merging with one another. Languages often have fewer nasalized than oral vowels.
Image
My most recent quiz:
Eurovision Song Contest 2018

User avatar
Whimemsz
Avisaru
Avisaru
Posts: 690
Joined: Fri Jun 20, 2003 4:56 pm
Location: Gimaamaa onibaaganing

Re: Explaining sound change?

Post by Whimemsz »

Qwynegold wrote:
Grey wrote:1: I understand that sound changes will occur when criteria is met ( A = B under conditions C always). How complex can these conditions be? Most examples I've seen are pretty straightforward, but I'm no expert.
SCs can vary all the way from no conditions at all, to so complex conditions that very few words are affected by them. I think it's best to use a combination of SCs of different complexity.
Yes. I'm working on redoing all the Algonquian sound changes for the Correspondence Library/Index Diachronica, so here's a few examples of quite specific and/or complex changes from Algonquian languages (where I use "%" to signify "one syllable away from"):
  • Cheyenne: a short vowel in the second syllable is lengthened when before a Nasal+Obstruent cluster, if the first syllable has a short vowel (V → Vː /#...V̆%_NO) -- with the further complication that a resulting *ɛː before *r or *x is treated differently from inherited, original long *ɛː for the purposes of a later vowel shift
  • Cheyenne: a prepenultimate high tone vowel becomes a raised high if it follows a syllable with a high tone vowel and precedes a syllable with a surface low tone, and one or more voiceless vowel syllables may intervene between the two high tones (V˦ → V˥ /V˦(C0V̥C0)0%_%V˩%V)
  • Shawnee: m̥ʔ →Ø /#_C
  • Menominee: *ɛ preceding *k or *m raised to *i in the second syllable of a word if the first syllable contained a short vowel followed by a glottal stop (ɛ → i /#C0VʔC(C)_{k,m})
  • Menominee: an “even” short vowel is lengthened when preceding a consonant cluster (the first syllable of a word is “odd”; if the vowel of an odd syllable is short, the following syllable is even (= [-odd]); if an odd vowel is long, the following syllable is also odd; if a syllable is even, the next syllable is odd) (best I can do: V[-odd] → Vː /_CC)
  • Blackfoot: ɛ → o /p_m
  • Blackfoot: *i is lost in the second syllable of a word when preceding ss+Consonant, if the vowel in the first syllable is short (i → Ø /#(C)(C)V̆C_ssC)
  • Arapaho: *x becomes s before a front vowel or j, except when ɔ(ː) precedes (x → s /{!ɔ(ː)}_{i(ː),ɛ(ː),j}) (meanwhile, in the same environment before front vowels/j, *k became t͡ʃ and *m became b, regardless of the preceding vowel)
To be clear, the majority of changes in these languages are more straightforward and apply in more environments, but it's also not that hard to find more complex or very specific changes like this. A realistic set of sound changes will have a mixture, as Qwynegold says.

User avatar
Qwynegold
Smeric
Smeric
Posts: 1606
Joined: Thu May 24, 2007 11:34 pm
Location: Stockholm

Re: Explaining sound change?

Post by Qwynegold »

Whimemsz wrote:specific and/or complex changes from Algonquian languages
OMG @_@ The linguists who discovered these sound changes must be really clever.
Image
My most recent quiz:
Eurovision Song Contest 2018

Post Reply