Wikipedia:Reference desk/Archives/Language/2018 July 3

Language desk
< July 2 << Jun | July | Aug >> July 4 >
Welcome to the Wikipedia Language Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.


July 3

edit

Languages using the beginning of the word to determine the stressed syllable

edit

In languages with predictable stress the procedure ("algorithm") for determining the stressed syllable starts in almost all cases from the end of the word. Thus according to article Stress (linguistics): Armenian words are stressed on the ultimate (last syllable), Polish and Quechua words are stressed on the penultimate (second from last) and Macedonian words are stressed on the antepenultimate (third from last). Even when stress is not necessarily on a particular syllable from the end, the various rules for determining where it goes start from the end of the word and move backwards, for example Latin. The only exception seems to be languages where the stress is invariably on the first syllable, such as Hungarian or Czech.

Are there any languages that do not necessarily accent the first syllable but where the procedure for determining the stressed syllable refers to the beginning of the word, for example languages that would regularly stress the second syllable from the beginning, or that would have a rule that "the first syllable is stressed if it is long, if not it is the second syllable" (the mirror image of the Latin rule), or "the stressed syllable is the first syllable that is long starting from the beginning of the word"...

If there aren't, is there a general reason why? Thanks. Basemetal 15:55, 3 July 2018 (UTC)[reply]

Yes -- Hopi_language#Stress is one. HenryFlower 21:14, 3 July 2018 (UTC)[reply]
Thanks. The Hopi rule is exactly my second hypothetical example (provided syllable length in Hopi is determined as in Latin). Btw, that rule is not (as I claimed) the mirror image of the Latin rule. That would be the rule that the second syllable is stressed if it is long, otherwise the third syllable. I'm intrigued by your knowledge of Hopi stress: Have you studied Hopi? Did you one day read that article and were able to recall how Hopi stress went when you saw my question? (That in itself would be remarkable). Or have you randomly looked through WP trying to answer my question and got lucky? (That would be even more remarkable). Basemetal 22:55, 3 July 2018 (UTC)[reply]
The discussion's got way beyond me below, but just to answer -- nothing so impressive, I'm afraid. I was just intrigued enough by your question to do some googling. :) HenryFlower 08:10, 4 July 2018 (UTC)[reply]
What querie(s) did you use? I wish I was able to use Google as effectively. Basemetal 09:25, 4 July 2018 (UTC)[reply]
I thnk my initial search was for "stress counted from first syllable", which took me to this page, where the Directionality section mentions Hopi. HenryFlower 09:47, 4 July 2018 (UTC)[reply]
Thanks. That Cardiff page is interesting in itself (as well as these two tables on the same site). Trying to reproduce your results in Google Books (before you'd answered) I noticed this (Harry van der Hulst, ‎Rob Goedemans & ‎Ellen van Zanten (eds), A Survey of Word Accentual Patterns in the Languages of the World, Walter de Gruyter, 2011) which may be of interest in the context of this discussion. Basemetal 10:30, 4 July 2018 (UTC)[reply]
Basemetal -- in the book Metrical Stress Theory, Bruce Hayes discusses typological factors which would make such word-initial quantity-sensitivity unlikely -- trochaic (initial prominence) stress-generating constituents tend to be either insensitive to syllable weight or contain two syllables of equal weight, and to be assigned starting from the beginning of a word, while iambic (final prominence) stress-generating constituents tend to be sensitive to syllable weight (often containing two syllables of unequal weight), and to be assigned at the end of a word. And extrametricality also generally applies starting from the end of a word. AnonMoos (talk) 00:07, 4 July 2018 (UTC)[reply]
So what about the Hopi stress rule? Theoretically it shouldn't exist. Isn't that a bit of a problem? Is Hopi mentioned in Metrical Stress Theory? Basemetal 00:31, 4 July 2018 (UTC)[reply]
It's greatly oversimplifying to say it "shouldn't exist". Much of linguistic typology is concerned with explaining why some patterns are more common than others among the world's languages. Hopi is briefly discussed as having a word-initial iambic stress-generating constituent. By itself it doesn't particularly refute anything in the Hayes book (only if Hopi-like patterns were much more common than they actually are among the world's languages would there be a problem). AnonMoos (talk) 01:55, 4 July 2018 (UTC)[reply]
It's only rules that are weight sensitive that are unlikely to work forward, correct? So there are languages where say the 2nd syllable, or the 3rd syllable, is stressed, just as there are languages where the 1st syllable is stressed (as such rules don't take syllable weight into account)? Basemetal 05:09, 4 July 2018 (UTC)[reply]
In prosodic theories with constrained or "parametrized" stress-generating constituents, it's hard to get antepenultimate-syllable stress without invoking extrametricality. Since extrametricality very rarely operates at the beginning of words, main stress on a word's third syllable would be predicted to occur in few if any languages... AnonMoos (talk) 05:28, 4 July 2018 (UTC)[reply]
Ok. How about languages with stress on the 2nd syllable? Basemetal 05:33, 4 July 2018 (UTC)[reply]
Sure, just construct an iamb at the beginning of each word. If there's no contrast between light and heavy syllable weights, then you get second-syllable stress. If there is a contrast between syllable weights, then you get a Hopi-type system... AnonMoos (talk) 05:53, 4 July 2018 (UTC)[reply]
And what languages do have such a 2nd syllable stress rule? Thanks, btw, for all this; I didn't know you were also well-read on theoretical linguistics (which is not my case, I must unfortunately admit) besides being knowledgeable in historical linguistics, IE, Hebrew... Basemetal 06:17, 4 July 2018 (UTC)[reply]
I'm really not well-versed in all areas of theoretical linguistics (I don't know or care about Chomsky's latest syntactic theory), but due to past research interests, I have a physical paper copy of the Bruce Hayes book.
From pages 266-267 of the Hayes book, it appears that at least some forms of Southern Paiute, Araucanian, and Dakota have basic second-syllable stress (though there may be complications in the details). AnonMoos (talk) 09:25, 4 July 2018 (UTC)[reply]
Thanks. Would it be fair to say (from what you could gather from the cases mentioned in Hayes's book) that the most common forward stress pattern (stress rule that starts from the beginning of the word) by far is stress on the first syllable? Basemetal 10:29, 4 July 2018 (UTC)[reply]
Looking through the Hayes book more carefully, I see that it doesn't contain as much frequency information as I thought. He discusses two or three things which might be considered theoretical possibilities, but which are very rare or non-existent in real languages, but otherwise leaves most frequency stuff aside (I may have been confusing it with another source). Nevertheless, I believe that what you're asking is true (though I can't really support it from the Hayes book). AnonMoos (talk) 07:31, 5 July 2018 (UTC)[reply]
Thanks for looking through the book. If you go through this table (for which let me thank Henry Flower again) you certainly finds lots of "1L", i.e. basic 1st syllable stress (often with complications, as you say, described in the "Comments" column) (and to my surprise also lots of "1R", basic last syllable stress) and in that particular table they do seem to be represent the single most common forward pattern. But that's not really frequency information as there are some 5000+ languages. The data very often comes from Hayes. If you're curious about the sources here's the bibliography. Basemetal 11:02, 6 July 2018 (UTC)[reply]
Responding to the original post: Mongolian, although this view seems to be disputed/outdated. --Theurgist (talk) 14:40, 4 July 2018 (UTC)[reply]
From Uralic languages#Phonology: In many Uralic languages, the stress is always on the first syllable. From Hungarian language#Prosody: Primary stress is always on the first syllable of a word, as in Finnish and the neighbouring Slovak and Czech. (These don’t really answer your question since you’re interested in the stress on the second syllable, but I think they’re interesting and related.) Loraof (talk) 19:39, 4 July 2018 (UTC)[reply]
Early Germanic languages such as Old English had stress on the first syllable of a word-stem, but that didn't always mean on the first syllable of a word. For one thing, compound words had two word-stems, and therefore two stresses. Also, prefixes could be added to a stem. Most prefixes were unstressed, but some (mainly the prepositional ones) were given a stress of their own... AnonMoos (talk) 07:31, 5 July 2018 (UTC)[reply]