How do stress patterns change over time?

If you're new to these arts, this is the place to ask "stupid" questions and get directions!
Post Reply
Toropshpii
rupestrian
rupestrian
Posts: 11
Joined: 13 Sep 2015 23:58

How do stress patterns change over time?

Post by Toropshpii »

I've been trying diachronic conlanging for one of my languages, and some of the sound changes are depenent on which syllables are stressed and which syllables aren't. I have an idea of what I want the stress patern for the proto-language to be, but I have no clue how this pattern would change over time. Should each sound change intoduce a whole new pattern, should I keep the pattern the same throughout the diachronic process, or is there a more complicated way this should be dealt with?
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: How do stress patterns change over time?

Post by eldin raigmore »

If the stress-pattern is quantity-sensitive ("weight"-sensitive),
and what counts as "heavy" doesn't change between the protolang and the daughterlang,
sound-changes could change some syllables from heavy to light, or from light to heavy;
and that could change the stress-pattern.

Perhaps likelier; maybe the proto-lang's stress-pattern was quantity-insensitive (stress was determined solely by the syllable's position in the word, first or last or second or penultimate or whatever); but the daughter-lang's stress-pattern is quantity-sensitive. This could mean that sometimes the stressed syllable in the protolang counts as a light syllable in the daughter-lang, or an unstressed syllable in the protolang counts as a heavy syllable in the daughterlang; and that could make stress shift.

Or, perhaps as likely, maybe the change is the other way; the proto-lang's stress system is weight-sensitive, but the daughter-lang's is not.

Another possibility might be the introduction of superheavy syllables.
Maybe some word, in the protolang, has a stressed heavy syllable right next to (probably just before, maybe just after) a light-as-can-be unstressed syllable, perhaps just a CV or VC syllable in which the vowel is very "reduced".
In the daughter-lang, the unstressed vowel has disappeared completely and the formerly-unstressed-syllable's only consonant has been absorbed into the stressed, formerly-heavy syllable's coda (or onset, but probably coda).
Maybe syllables created that way count as "superheavy".
Maybe the daughterlang has other syllables, not created that way (possibly created by other sound-changes), that now also count as "superheavy", now that there is such a thing as "superheavy".
Maybe the daughterlang has a rule that any superheavy syllable must be stressed, no matter where in the word it occurs.
Maybe there was no such rule in the protolang.
Maybe this means some words, especially words not wholly inherited from the protolang -- coined more recently, or borrowed more recently -- now have the stress occurring on syllables other than those stress would have been assigned to by speakers of the protolang.

I don't know if any of that has actually happened in real life natlangs.

Did it help anyway, I hope?
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: How do stress patterns change over time?

Post by Creyeditor »

Toropshpii wrote:I've been trying diachronic conlanging for one of my languages, and some of the sound changes are depenent on which syllables are stressed and which syllables aren't. I have an idea of what I want the stress patern for the proto-language to be, but I have no clue how this pattern would change over time. Should each sound change intoduce a whole new pattern, should I keep the pattern the same throughout the diachronic process, or is there a more complicated way this should be dealt with?
To me one interesting method for diachronic stress is the following. Start with a regular stress system like 'stress the second to last syllable'. In the first part of the diachronic development, the syllable count might be altered but the absolute position of stress stays the same, e.g. final i-deletion in stráti > strát, faráti > farát but not in stráta > stráta or málat > málat.
In the next step you look at the forms after the first part of your diachronic development (in our case strát, farát, stráta, málat) and think about the simplest generalization that a speaker of your conlang could make to derive most of the data. For our example words above this might be 'stress the last syllable if its heavy, otherwise stress the second-to-last. Note that this generalization does not explain all of the data, so some forms have to be adjusted. In our case málat > malát. Now you start from the beginning again, i.e. you do sound changes, look at the results, make generalization. If you make a lot of sound changes without making new generalization you might end up with very irregular stress systems, if you make a new generalization often enough, you will end up with a more regular system probably.
Another interesting change that has ocured in natlangs and just came to my mind is the domain of stress that can shift. In some Indonesian varieties stress on the word level is more prominent in others it's stress on the phrase level.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
User avatar
GamerGeek
sinic
sinic
Posts: 345
Joined: 17 May 2017 18:10
Location: The Universe
Contact:

Re: How do stress patterns change over time?

Post by GamerGeek »

stráta > stráta or málat > málat.
:wat:
Sumelic
greek
greek
Posts: 566
Joined: 18 Jun 2013 23:01

Re: How do stress patterns change over time?

Post by Sumelic »

GamerGeek wrote:
stráta > stráta or málat > málat.
:wat:
It's an example of the simplest kind of change: leaving things the same.

Creyeditor's post describes the following situation:

Stage 1 of the language has the following phonological words: stráti, faráti, stráta málat
Stage 2 of the language has the following phonological words: strát, farát, stráta, málat

Stress has become phonemicized because some words have lost final syllables, while others haven't.
User avatar
sangi39
moderator
moderator
Posts: 3024
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: How do stress patterns change over time?

Post by sangi39 »

Eldin and Creyeditor, from what I can tell, have mentioned some of the most common reasons for stress shifting, and in some cases simply becoming irregular.

In Latin, for example, stress was placed on the penultimate syllable when it was "heavy" (closed, containing a long vowel or both), but shifting back one syllable if the penult was "light" (was both open and contained a short vowel). In Vulgar Latin, however, vowel length was lost so stress was placed irregularly on either the penultimate or antepenultimate syllable.

Later Romance languages took this further. Deletion of final vowels lead to languages like Spanish having stress either on the final syllable or the penultimate syllable (as in Creyeditor's example the stress on the original word hasn't changed, but the word has been shortened so relatively it has), and in French vowel/syllable deletion has meant that stress (at the level of individual words) is now almost always final.

Weight-dependent stress could probably develop by means of analogy, again following Creyeditor's example (vowel deletion creates instances of stressed heavy syllables that create a new perceived pattern and most or all similar words shift stress to match it).

One thing I still need to look up is how Latin went from the word-initial stress pattern of Old Latin to the (ante)penultimate stress pattern of Classical Latin.
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.
Post Reply