Talk:Hungarian phonology

From Wikipedia, the free encyclopedia

You have new messages (last change).
This article is part of WikiProject Phonetics, an attempt to build a comprehensive and detailed guide to phonetics and descriptive phonology on Wikipedia. If you would like to participate, you can edit the article attached to this page, or visit the project page, where you can join the project and see a list of open tasks.


Umm, some of the last changes by the 219... IP-address have not been correct. The phonetic value of cs and dzs are tS and dZ, not those s/he mentions. Those are not the standard versions, they are considered rather childish. I will correct them next time if there is no argument against it.


I agree. The voiced and voiceless alveolo-palatal fricative(ɕ) does not occur in Hungarian, at least what I remember. I would delete it now, but it is such a pain in the ass to go through that table to find the individual glyphs, you can do it or I will do it later. By the way, who are you? InFairness 08:20, 1 Jun 2005 (UTC)

It was me , sorry :) Correction will follow in a day or two. --Sicboy 16:40, 2005 Jun 2 (UTC)

To InFairness: It’d be awesome if the western Wikipedias finally switched to UTF-8 already, wouldn’t it? -- Ralesk 16:43, 5 Jun 2005 (UTC)

Oy! Tell me! InFairness 22:04, 21 Jun 2005 (UTC)

Contents

[edit] Last changes by InFairness

Hi, thanks for the changes and improvement. However I don't aggree with you in the new phonetic equivalents for /o/, and /ö/, because the IPA symbol you use for /o/ is in fact between /a/ and /o/, but linguistic resources, particularly the ones written in English use this symbol for the /a/ (which use isn't correct as well, anyway). Thus I think [o] would be perfect for /o/. The same applies for /ö/, where the œ IPA symbol you use is much a much more opened version for the vowel, compare French oeuvre with Hungarian öv. This œ vowel is used in some Transsylvanian dialects, where tœr means 'he breaks', and tör (short!) means 'dagger' :)! So I recommend [ø] for the phonetic realisation for /ö/. --Sicboy 16:47, 2005 Jun 2 (UTC) ps sorry for some mistakes

Seconded and reverted to ɒ and ø. ö and ő are exactly the same in roundedness and position — at least in my Budapestan dialect; therefore I too support the ø notation for both (which is the sign used for the German ö as well, as far as I’m aware, and that sounds pretty much the same as the Hungarian one). The situation is the same with the o sound: it’s not “lax” when short, at least in the dialect I speak (which is the “de facto standard”, as unfortunate as that is (the lack of ë in it, for example :P)). “a” is an evil little sound, very hard to catch it right with IPA, but ɒ seems the most right. I have also fixed some of the z-curls that remained in the page. -- Ralesk 16:41, 5 Jun 2005 (UTC)
Okay, thanks for the job. Our task here is to describe the standard which behaves exactly as you have said it. --Sicboy 23:45, 2005 Jun 7 (UTC)

Thank you (both) for your input, and point well taken: Budapest, as far west as it is, is the "standard" dialect, so I guess that we should use their pronunciation. My father is from Bihar, but his parents were from Kassa and Kolozs. I almost certainly pronounce things differently than standard. That having been said, I will say that my ear does not hear ɒ (turned script "a" as you have it), but ɑ (script "a") from Hungarians from other areas. I do hear turned script "a" in speakers from certain areas, e.g. from the Tokaj region. I will also agree that the short "a" is a devilishly difficult sound to classify—it really falls between the closest approximations.

I must disagree, however, with the choice of æ for the sound of the short "e" (as listed in the edit summary). It could not be clearer to me that the sound for the short "e" is not æ, but between æ and ɛ. I think what we really need is a new symbol for the sound, because neither of the two really fit. But ɛ is closer to the actual sound than æ. Just think of how someone with a Hungarian accent would pronounce the word "accent": something closer to /ɛksɛnt/ than /æksænt/, showing that /ɛ/ rather than /æ/ is closer to the native Hungarian phoneme.

Also, R. said:

The situation is the same with the o sound: it’s not “lax” when short, at least in the dialect I speak (which is the “de facto standard”, as unfortunate as that is (the lack of ë in it, for example :P)).

I am not clear what you are referring to in the last part of the sentence: "(the lack of ë in it, for example :P)". Could you elaborate?

InFairness 22:04, 21 Jun 2005 (UTC)

Well, this seems like we could initiate a conference sometime :) And my English becomes worse and worse by entering the night akár magyarul is megbeszélhetnénk, de a tisztesség... So
a) the a-sounds. Well, the script "a" is the Open back unrounded vowel as you have pointed, while the most important characteristic of the Hungarian /a/ is the roundedness, compare it to the unrounded á. The difference between script "a" and normal [a] is not too important however. I doubt you hear script "a" from Hungarians from other areas -- that pronunciation is the shibboleth for the Palóc dialects, and for many Roma people, who tend to use Palóc (okay, Győzike and Be[a] [a]sszony is typical, if you have ever seen that weird tvshow:). However, these dialects usually use a long rounded (so turnedscript) [ɒ:] for the phoneme /a:/ (<á>), compare Salgótarján -x- "Sálgótárjaan" etc. But this use is restricted to these dialects (and to some Transsylvanian, Croatian etc. marginal dialects).
b) the e-sound: the choice between [æ] and [ɛ] is many times idiolectal, and it is not a very important question in Hungarian phonology: the most important feature is that it differs from the mid-vowel /é/, and dialectal /ë/, which is the short mid front vowel [e]. Those dialects however, that use /ë/ usually pronounce /e/ more open than it is pronounced in the standard. So /Szëged/: standard Szeged [sɛgɛd]; ë-dialects Szeged [segæd] ~ sometimes [segad]; ö-dialects Szöged [søgɛd], én [e:n].
There is a joke about these æ-e dialects: in Zalaegerszeg (western dialects are stable ë-dialects), the announcer in the train station says "Kérem vigyázzanak, vonat érkezik az Á-vágányra, az Á-vágány mellett kérem vigyázzanak", repeats twice, and when the train arrives, killing hundreds of uncautious people by the platform, s/he announces "Mondtam, hogy vigyázzanak az Á, mint Álámér vágány mellett!". Ugh
whoa, I think I'm too tired to be clear... --Sicboy 01:15, 2005 Jun 22 (UTC)

The o sound. It's not lax (never open o according to my ears). My parenthesed comment didn't have anything to do with the article content really, I just mentioned my dialect lacks ë and that's a bit sad. -- Ralesk 15:45, 22 Jun 2005 (UTC)

"That's a bit sad." Okay, I understand what you mean, but why is lacking a phoneme in your dialect is so sad :) Think of those poor western guys, having an ë... but lacking ú, ű, í at all! Now who's richer ;) --Sicboy 23:19, 2005 Jun 23 (UTC)
Once again, I will ask my question, hoping beyond hope that someone can answer it: I have seen the "e" with duplapont (ë) all over, but I never knew what sound it represents. Do either of you know?
Well, the answer is in this very article :) " This odd feature is solved in the old language and in dialects: there was/is an eighth short phoneme /ë/, which is just like the /e/ but it is mid, and its pronunciation is [e], in contrast with /e/ being [ɛ]. In dialects, this phoneme is found in the mid series, and the low /e/ in the low series." So pronounce ë like é, but short. I think most speakers pronounce the word és like it was ës, with the vowel length being lost. --Sicboy 2005 July 3 20:01 (UTC)
So written "ë" is pronounced [e] (in contrast to written "é", pronounced [e:]). Thank you for finally clearing that up. We (Sárándi) don't use "ë", at least that I've seen. Fascinating.

Also, after review and discussion with my dad, we can agree that "turned script a" is close enough to the actual sound, though not exactly, of the short "a" to merit its use as opposed to "script a". However, there is still the issue of pronunciation of the short written "ö". Apparently, in Bihar dialect (or whatever it is we actually speak) the short written "ö" is indeed pronounced as [œ] and even, on occasion and depending on word placement, [Œ] (!); whereas the long written "ő" is pronounced with almost an (almost diphtong-like) off-glide [ø:(Y)]. Of course, this is the way that I learned it, and it sounds most natural to me, hence the difference in opinion.

I am glad to see, though, that you did change the IPA for the short written "e" from [æ] to its correct [ɛ].

Now...what about [ɨ]? Nevermind...

Well I'm glad I could convince you about the a-probelm :)
I'm equally glad to have convinced you about the [æ] problem! :)

Well the ö-problem is not a problem anymore as well, read my first note in this conversation from part: "this œ vowel is used in some Transsylvanian dialects". It seems your dialect is a transition between the Alföld and Mezőség dialect, the latter is the one that has the tendency to open ö to œ, as well as opening o to a, such as magyaró for mogyoró etc. However this tendency (sound change) is complete only in some villages. The ë>e change is also part of this short-mid --> short-low change, but this change has a much wider territory: the Tisza/Northern Plain dialect (Debrecen, Hajdú-Bihar, etc.), the Mezőség dialect (central Transsylvania, Kolozsvár, ...) the Northeastern dialect (Sátoraljaújhely, Zemplén county, base of standard), some Székely dialects, and the standard with the Budapest colloquial.

BTW: In English I believe you are referring to a "sound shift", not "sound change".
Oh, yeah, I'm only en-2, now everyone understands why... Even in my specialty, grr.
The diphtongization of the ó, ő, é (so long-mid) vowels is typical for many dialects, base rule is that in the West, you get uo, üö, ië (now you know what ë is:); while in the East you can hear ou, öü, ëi respectively. As your Bihar dialect is Eastern, it is not surprising that you hear [øy] for ő, as I bet, you'll find [ei] and [ou] as well if you concentrate on your (or your father's) speech ;) And yes, you have a comment on a change in the article, with "
I don't need to concentrate very hard...it is quite clear in his speech as well as mine. As clear as a southerner's draaaaawwl.
The ɨ: there's not much difference between that, and the ɯ, and to be honest, no one knows which was pronounced (and when) in Ancient or Old Hungarian. It would be intuitional to think that the change was in direction ɯ > ɨ > i. But who knows.
I had read somewhere that the use of the written Latin "Y" originally transcribed the sound "ɯ/ɨ", in the same way that the "y" represents ɯ in modern Polish (which tends towards [I] at times). Also, it is possible that ɨ or ɯ are preserved from early Uralic--Estonian, Veps, and maybe a few other Baltic Finnic languages retain the sound as well, though in Finnish, Ingrian, Karelian, etc. it has migrated to [ø] or [i].
Well, I think, the written Latin "Y" was just an orthographical pecularity, like names written as Eötvös, Dessewffy, etc. The old ɨ/ɯ in the Ancient Uralic is a very thin supposition, and I think modern Estonian õ is not derived from that, and this is easy to prove by comparing Estonian words with õ to Finnish words, usually having e,ö,i and things like that :) but anyway, this is very far from the topic... Old Hungarian /ɯ/ has stems usually from PFU/PU words like inni which derives from *junni compare Finnish juota (????) and Iranian words like híd compare Ossetian hɯd (??), and a lot of Turkish words
As for your comment on the page "Certain conservative dialects (e.g. the Bihar dialect) preserve it unchanged or as /ɪ/ (as in the English word "it").", I'm afraid to tell you that it is incorrect: there are no dialects preserving the old ɯ like that, at least Hungarian linguistic sources don't mention them, and believe me: if they found one, everyone would know about that in 5 minutes :D. There is an /ɯ/ phoneme in very archaical Csángó, old-Kórógy dialects, but those seem to be an innovation (well okay, I don't know too much about the latter, it is extinct). And the [ɪ] in Hungarian is an allophone of [i], as there is no distinction between these (unlike in English). I think that after one concentrates way too much to his/her own pronounciation, he/she can hear everything, so I can convince you that alma is pronounced körte after some time :)
That's why listening to others' pronunciation is important. I am glad to be surrounded by a number of other native speakers in my area, most from Budapest. BTW: The /ɯ/ phoneme (unrounded [u]) in Csángó is, from what I understand, not attested in earlier varieties, but is a (fairly) recent introduction. It seems more likely to me that it represents the influence of local Vlach dialects (written â, î, sounding [ɨ]).
Yes, the Csángó ɯ is a recent introduction from Romanian (yes, let's be PC! :), and that is another argument that Hungarian has no traces for Old Hungarian /ɯ/ other than the suffixation of these words (hidat, iszok, etc.). --Sicboy 21:23, 2005 July 16 (UTC)

[edit] Palatal vs palatalized

gy is not palatal (IPA [ɟ] -- which sounds almost like cs [tʂ-ligature]), but palatalized ([dʲ]). This is something completely different! The same holds for ny, and I'd be very surprised if it weren't also true for ty, though I haven't been to Hungary for years, so I'd appreciate it if a native speaker could help me out here.

Maybe it should be mentioned that ly was [lʲ] in some earlier age, which explains why it isn't spelled j.

I think you're wrong: d is dental, while gy is palatal. If gy were only palatalized, based on d, its place of articulation was the same as d, only the tongue surface were raised. However, at gy, the tongue tip touches the palate more backward.
Sounds like gy is the palatalized version of a laminal-alveolar d, like the one Spanish has, instead of of the dental d that Hungarian really has. Would sound identically anyway, despite the slightly different articulation.
On the other hand, ly denoted (and in some dialects, still denotes) the palatal lateral approximant, IPA [ʎ], not [lʲ]. -- Adam78 23:39, 2 August 2005 (UTC)
Well, in those dialects, does it sound like Italian gl (palatal) or like Slovenian/Serbocroatian lj (palatalized)?
Yes, gy, ty and the others are palatal. The cs [tʂ-ligature] sound mentioned can be found in Polish (ć). It is also palatal, however it is not a stop, rather an affricate, that makes the difference. --Sicboy 17:43, 2005 August 3 (UTC)
[tʂ-ligature] is cz in Polish and č respectively ч in the other Slavic languages. It is almost a palatal stop -- which [c] would be; the difference is that the tongue is released forwards instead of downwards, so that an affricate results. Ć is its palatalized version, pronounced like ci in AFAIK some kinds of Italian.
Did I forget to sign, or did someone accidentally erase my signature? David Marjanović 22:30 CET-summertime 2005/8/5
I see you are Slavic, but I just don't aggree, as I can speak some Polish, after family reasons, and beside listening to relatives, and TVPolonia (OMG :) I've read some books about this language, and everything coincides: cz is [tʃ] (dental affricate) and ć is [ʨ] (palatal affricate) or sometimes [c] (palatal stop). See Polish language, this is not disputed there. --Sicboy 00:44, 2005 August 8 (UTC)
Uhh ohh, and yes. I can't speak Croatian / Serbian, but when listening to them, I can't differentiate between ć and č (dž and đ) (but I'm sure they are somehow different...), so the situation there is -- I think -- different from Polish. --Sicboy 00:48, 2005 August 8 (UTC)
I've heard the Bosnians don't tell them apart, but elsewhere ć č đ dž are [tɕ t̠s̠ dʑ d̠z̠], like Polish ć/ci cz dź/dzi dż. (The diacritic is for "retracted", see Postalveolar consonant.)
A few days ago I heard two Hungarians talking. One said [dʲ] once, not [ɟ] or [ɟj]. Of course I don't know what dialect that was...
David 2005/9/27 18:00 CET-summertime
I think that could have heard just some performantial variation. Anyway, difference between [dʲ] and [ɟ] is quite small and unimportant, since there is no phonological distinction between them in Hungarian. :) --Sicboy 22:56, 27 September 2005 (UTC)

[edit] New sections

Is there someone who could write about assimilation, intonation (esp questions v statements) and phonotactics (esp consonant clusters), please? Gailtb 06:47, 25 October 2005 (UTC)

[edit] palatal etc.

When I converted the tables to wiki syntax, I changed some of the phonetic values per the Handbook of the International Phonetic Association. There it describes ty and gy as being most commonly palatal affricates, but palatal stops in formal speech. Peter Ladefoged has nice pix showing that they truly are dorso-palatal, and not palatalized coronals. As for the postalveolars, the Handbook transcribes them as palato-alveolar [ʃ ʒ], but that doesn't mean much; it could simply be that those are the most convenient symbols to use. Since the authors didn't discuss this point, it shouldn't be counted as evidence as to whether they are like Polish [ʂ ʐ]. (One operating principal of the IPA is that symbols may be used as convenient, so "[c]" is often used for [tʃ].) kwami 07:40, 25 October 2005 (UTC)

Well, maybe Ladefoged is not right here, okay, I know that is blasphemy, however, I can assure you, that if ty and gy are palatal stops in formal speech, then every Hungarian is speaking High Literary Hungarian :). Ooops, exceptions are Felsőőr dialects (in Austria), and some csángó dialects in Moldva, those pronounce an affricate, just like in the way Old Hungarian is supposed to have pronounced them. But nobody else. --Sicboy 02:50, 31 October 2005 (UTC)

The waveforms of affricates are pretty clear, so I doubt the researchers are wrong. What may be happening is that they come off as stops when a word is spoken is isolation, say as a demonstration of how it is to be pronounced, but as an affricate in running speech. That would be hard to hear, because our brains filter out anything that's not supposed to be there, but would be easy to capture in a recording. A lot of English words are like this. For example, him is normally pronounced [im] by people who otherwise pronounce their aitches, but if you told them this, most would insist the [h] is there. kwami 03:02, 31 October 2005 (UTC)

Okay, but if this is the case, I can't even think what the [c] stop is? It can't be more stopped than the Hungarian ty. Maybe the researchers were curious about the [c] sound, and not the Hung. /c/ phoneme. Anyways, I have never seen the phonetical representation [mɑɟʝɑr] which can be seen on the Hungarian language page. Moreover, even if we accept that [ɟ] and [c] are a little bit affricated, they are not like [ɟʝ] and [cç], these pronunciations are highly stigmatised as foreign accent, and not as substandard, as your remark states. I have never heard different pronounciations of /c/ and /ɟ/ of speakers of Hungarian, being politicians, writers or janitors.

I never said it was substandard. When I was in Hungary, ty and gy sounded like affricates in colloquial speech, but I never picked up much of the language, and my observations are pretty worthless. All I can tell you is what other linguists have said. This is from the IPA Handbook:

The style of speech illustrated is that of Educated Colloquial Hungarian of the 1990s as spoken in Budapest. [...] The speech on the recording is that of a male speaker in his fifties with an academic background, using a somewhat advanced style of speech as spoken in a formal communicative situation.

The transcriptions of ty and gy in the recorded passage are consistently affricated [cç] and [ɟʝ]. However, the author (Tamás Szende of the Institute of Linguistics, Hungarian Academy of Sciences, Budapest) states that "In formal style /cç, ɟʝ/ are mostly pronounced as palatal stops, i.e. as [c] and [ɟ]."

Ladefoged states,

Figure 5.20, based on x-ray data in Bolla (1980),[*] shows that Hungarian c and ɟ have the root of the tongue more advanced than ç and ʝ. This may be because the stops (and the nasal ɲ) require an articulatory gesture in which the tongue has to be raised considerably higher, as if the aim were to push the tongue through the roof of the mouth, as we noted in chapter 2.
*Bolla, Kálmán. 1980. Magyar Hangalbum (A Phonetic Conspectus of Hungarian). (Magyar Fonetikai Füzetek (Hungarian Papers in Phonetics) 6). Hungarian Academy of Sciences, Budapest.

Note that this would be the case whether they were pure stops or affricated stops. Elsewhere he describes the sounds as being true palatals, and not palatalized alveolars. But in SOWL he never addresses the issue of affrication. kwami 01:12, 2 November 2005 (UTC)

Okay, good, but I still pray that we don't use [cç] and [ɟʝ], even if this is the very phonetical fact. Just like the way we don't use [z̻] and things like that for the dental consonants. Hey, do you really want to use [ɒ̜̽] in the section below consistently for the /a/? Sure, it is more precise, the page could mention it once, but I think using these in the Hungarian language page is very much overcomplicated. And I still think that Szende is not right here, there is no stilistic or sociolinguistical difference in pronouncing gy and ty. But maybe I'll be asking around about this question in the Institute of Linguistics. --Sicboy 22:42, 6 November 2005 (UTC)

[edit] Sounds of a

Open front unrounded vowel gives Hungarian bal as an example for IPA [a]. This seems wrong given the information on this page, which suggests that if the spelling is regular it should be [ɑ] or [ɒ], depending on which of those symbols is right. Also, áll is given as an example for [ɑ] on Open back unrounded vowel. Are these wrong, or are there some complications not described on this page?--JHJ 17:48, 4 November 2005 (UTC)

In the IPA Handbook, <á> is described as a long open central vowel, [äː], while <a> is described as a near-open, near-back, partially rounded vowel, [ɒ̜̽]. However, the symbols used in the Handbook are [aː] and [ɑ] (with [ɒ] as an alternate). That is, they don't bother with the diacritics for a broad transcription. The question we want to ask is how precise we want to be. kwami 22:23, 4 November 2005 (UTC)
So the examples on the pages mentioned above were the wrong way round? I've removed them.--JHJ 22:32, 4 November 2005 (UTC)
The bal example was totally wrong, the áll example is not so bad, however, again, it can be hard to distinguish between open back and open central (or even open front) vowels. Traditional Hungarian phonology states /á/ to be back (because in vowel harmony it is back), but it is not worth too much. The Structural Grammar of Hungarian (Strukturális magyar nyelvtan, ed. F. Kiefer) analyses its articulation to be central, however its phonological behaviour to be back. --Sicboy 22:49, 6 November 2005 (UTC)

[edit] End of a word

Are there any vowels which cannot occur at the end of a Hungarian word? --84.61.41.214 08:43, 7 May 2006 (UTC)

Short <ö> (/ø/) "cannot occur" at the end of native Hungarian words and short o, ü (IPA /y/) & u are quite rare. --194.152.154.2 19:10, 26 May 2006 (UTC)

[edit] Different meanings and pronunciations of "mentek"

I completely agree with mëntëk "you all go" and mëntek "they went", but can someone confirm that he/she pronounces "I save" as mentëk? I have never ever heard this before (I use mëntëk for "I save", mëntesz for "you save" etc.).

(I do not want to dispute the "mentek" that means "mentesek" because I do not use the "mentek" form. But I would probably read it as either mëntëk or mentëk because I use mëntës for the singular of "mentesek" but I heard some people say mentës.)

It is unfortunate that many people say that they can not hear the difference between e and ë, even though they use the two sounds correctly. So I may not get too many replies for my question, but please, if you do hear the difference, tell us: Which pronunciation do you use? -- BarroColorado 18:55, 29 March 2007 (UTC)

I grew up in Budapest and I use /ɛ/ only, so I can only rely upon MorphoLogic's Helyësen software, which does indeed list ment for 'he saves'. I hope that helps somewhat. By the way, I am the exact opposite of what you described above: I can pronounce and distinguish between the two sounds just fine, but I am not able to use them correctly in everyday speech. KovacsUr 21:46, 31 March 2007 (UTC)

Retrieved from "http://en.wikipedia.org../../../h/u/n/Talk%7EHungarian_phonology_dfda.html"