Discussion in 'Magyar (Hungarian)' started by Encolpius, May 12, 2010.
Hello, do you have any idea where to put Hungarian in this list? Thanks.
I think Hungarian uses mora timing, but I'm not sure.
Maybe you don't have to put Hungarian in that list at all...
If you do, I'd think (after wiki's other explanations as well) it's not mora timing because we don't pronounce two vowels instead of long ones, nor have we final sounds pronounced separately...
I think syllable-timed is a more likely choice.
As that article suggests, a good indicator is the prevalence of reduced vowels in stress-timing languages. Since (as far as I know) Hungarian has no reduced vowels… it must be a syllable-timing language. But, as the article also suggests, this doesn't really say anything about the duration of syllables in Hungarian.
See also: Routledge dictionary of language and linguistics
If it was syllable-timed, how would Radnóti have written all those beautiful hexameters (in "időmértékes verselés")? Just asking, not rhetorically, just out of curiosity.
I'm not an expert in this but just from reading a couple of things about it (see the links provided in this thread), I have the impression that one doesn't have anything to do with the other, Csaba.
When I tried to look for something relevant to this topic I found this partial quote from a book entitled Phonetic and phonological aspects of geminate timing by William Hallett Ham
"Morton et al. hypothesize that perceptual isochrony is organized around what ... of isochrony in mora-timed languages, such as Hungarian, Finnish, Japanese, ..."
Unfortunately this is from a page that google books did not allow me to see (243) and as such does not look entirely conclusive. Does anyone by accident have the complete book? I still suspect that there is some correlation between metric poetry and isochrony albeit I admit I have not found a definitive source to back this up. Metric poetry is either based on stressed/unstressed syllables or long/short syllables and in Hungarian it is based on the latter rather than the former. Also, for example, 'i' and 'í' are the same sound except their temporal duration so syllables based on i need to be significantly shorter in time than those on í.
Hm, found this in a research description from University College London (they are comparing Japanese and Hungarian from some points of view)
One of the specialities of Hungarian is that you can have a short vowel in a stressed syllable or a long one in another that doesn't have an accent on it.
Also, beware of the "common knowledge" (at least in Hungary) that vowels with long accents are long vowels (= vowels pronounced for a longer time than their short equivalents). It is especially not true in the case of á and é.
Also, dont't forget that in metric poetry you get a long syllable not only with the help of a long vowel but also with a short one followed by 2 (or more) consonants.
I simply brought up what I thought was a counter-example to the idea that all syllables have roughly the same length in Hungarian. I did not overlook the metric definitions of length but rather omitted them from my already crowded post. However, e.g. fülel and füllel do sound like they are different temporal durations do they not? I would tend to agree with your first post, Hungarian fits in none of the categories because it does not feel like füllel is 1.5x longer in time than fülel (as it were in a mora-based timing).
I would be interested to know why W. Hallett Ham suggested such a thing that seems so obviously not the case... (Do you know anything about him?)
I could read a bit from one of his books here (I don't know whether it'll work for you).
Unfortunately, I haven't found out what the point is in the whole thing but at least one of his sources (Szende) is surely reliable.
Thank you Zsanna, his book looks interesting, too bad that only parts are available of section 4.3 about Hungarian. His approach looks scientific to me and he does not seem to arrive at a simple conclusion, at least based on what I could read.
Page 110 ends with the affirmation that sült and hűlt do not take equal time to pronounce, which I agree with, this would be a problem for both mora-timed and syllable-timed models.
Hungarian is definitely NOT a stress-timed language as a stressed syllable will not sound longer than a non-stressed syllable just because it is stressed (all other properties being unchanged), and there are stressed-short, stressed-long, unstressed-short, and unstressed-long syllables.
It is definitely NOT syllable-timed because, taken in isolation, a syllable can either be short or long, where a long syllable takes roughly twice as much time to pronounce than a short syllable. Also, percieved syllable length also depends on what follows a VC or CVC syllable with a short vowel in it. If it has a syllable starting with a consonant, it will be perceived as a long syllable:
- pap [one short syllable]
- papné [two long syllables; the first one is perceived long here]
The length of the first short vowel DOES NOT CHANGE but the syllable is perceived to become longer by the addition of the time it takes to pronounce a combination of two consonants [pn] rather than just a single consonant [p] as in:
- papé [one short plus one long syllable].
I have read a book on Hungarian prosody that actually introduces, in addition to "purely short" and "purely long" syllables, a third syllable of, let's say, 1.5 units of time, to cover 'short' syllables that become long as a result of their contexts.
I'd say that Hungarian could be seen as a mora-timed language if we take the liberty to introduce one-mora and two-mora syllables, agreeing that a one-mora syllable can assume a time value of two moras in specific contexts.
However, overall, I'd also say that Hungarian has, if anything, a fourth type of isochrony, adding that:
(1) the speaker's actual performance has little to do with the 'ideal' rhythm of the language; and that
(2) when it comes to writing verse, Hungarian, as a language into which practically all its vernacular poets translated from other languages of greater prestige, frequently with the intention of preserving the original form/meter, has developed its own solutions (conventions) to best reflect (or most successfully imitate) widely different foreign forms of poetry, from classical Greek and Latin metric poetry to Old English stress-timed forms, and much more.
All in all, I see a problem with the idea of isochrony in that syllable lengths are not precisely defined in a language; all they need to be is different enough to carry the semantic information. There are only two realms where the ACTUAL time-length of the syllables matters:
(1) in hypothetical, 'ideal-case' language theories such as the theory of isochrony; and, more importantly,
(2) whenever language is applied to music (such as in recited poetry, sung lyrics, or rap).
But in this latter case (2), the effect may be practically ANYTHING: Hungarian may be used as though it were a stress-timed language or a syllable-timed language or even an untimed, free-flowing language with all but local contrasts between actual syllable lengths.
Separate names with a comma.