Polish Phonology.

Lorenc:

I'll jump on this thread for a related phonological question having to do with consonant softening.
I'm reading the "Concise Polish Grammar" by Ron Feldstein (to be found at http://www.seelrc.org/projects/grammars.ptml for free). My question is about the sound of softened (palatalised) /p/,/b/,/f/,/v/ and /m/, which this grammar book writes using a trailing apostrophe, e.g. as /p'/,/b'/ etc.
These sounds occur whenever in the spelling the letters p, b, (etc) are followed by an "i".
My question is: are these sound different, and how much different, from the cluster /pj/, bj/ etc.?
In other words, would there be a difference in sound between these pairs of words (the second of each pair is made up):
pięć - pjęć
biały - bjały
fiołek - fjołek
pawie - pawje
ziemia - ziemja

At pag. 49 of the grammar book there is a discussion of feminine names ending in -ia, where it is said that they can belong to two different classes. The first is constituted by native Polish words, where the ending -ia corresponds phonetically to /'a/ : e.g. ziemia is (I'll use the same transcription as the book) /źem'a/.
On the other hand in words of foreign origin -ia corresponds to /'ja/ : e.g. armia /arm'ja/.
The book also says that native Polish words admit /'ja/ as an alternate variant (I don't know if some speakers systematically do this, if it depends on the word or if it is a random thing).
Does all this make sense to you?
Do ziemia and armia rhyme for you?

As my native language is Italian the opposition between /nj/ and /ń/ or /lj/ and /l'/ is pretty clear to me. With some imagination I can contemplate /m'/ vs /mj/ but I'm not sure about the other pairs. That is, the difference in sound between /p'/ and /pj/ (etc.) seems tiny to to me. I don't have quality recordings or Polish speakers at hand so I turned to the forum :)
What do you think?

jump_bunnyThreads: 8 Posts: 332 Joined: May 14, 09	Edited by: jump_bunny Nov 24, 09, 01:42 / #
1. Vowels There are only six oral and two nasal vowels in the Polish Vowel System. Labial consonants can be followed by /ɨ/ (spelt <y>) and /i/. (Labial consonants are those which are articulated with: both lips (bilabial articulation), or: with the lower lip and the upper teeth (labiodental articulation). Apart from those, /i/ is usually pronounced when initials a word and after palatal (articulated with the tongue raised towards the hard palate) and alveolo-palatal consonants (articulated with the tongue behind the alveolar ridge), while /ɨ/ appears elsewhere. I mention them at the same time because they rhyme and often sound very much alike. Those differences are not shown in the spelling and most of native speakers don't realise them. Similar situation takes place when some vowels occure in particular places - but again - these distinctions are often not noticable so I will not ellaborate unless I'm asked to. All Polish oral vowels are monophthongs. (Which means their sound is a "pure" vowel sound, the articulation at both beginning and end is fixed - the sound doesn't tend to glide up or down towards a new position of articulation). The length of a vowel is not phonemic in Polish. (Which means that how long a vowel is pronounced does not change the meaning of a word). Oral vowels: <a>, <e>, <i>, <o>, <u> (<ó>), <y> Nasal vowels: <ę>, <ą>

ZiemowitThreads: 10 Posts: 1,063 Joined: May 8, 09	Nov 24, 09, 09:34 / #
jump_bunny: Nasal vowels: <ę>, <ą> What nasal vowel do you pronounce in the word awans?

LorencThreads: 6 Posts: 30 Joined: May 13, 08	Nov 24, 09, 15:30 / #
I'll jump on this thread for a related phonological question having to do with consonant softening. I'm reading the "Concise Polish Grammar" by Ron Feldstein (to be found at http://www.seelrc.org/projects/grammars.ptml for free). My question is about the sound of softened (palatalised) /p/,/b/,/f/,/v/ and /m/, which this grammar book writes using a trailing apostrophe, e.g. as /p'/,/b'/ etc. These sounds occur whenever in the spelling the letters p, b, (etc) are followed by an "i". My question is: are these sound different, and how much different, from the cluster /pj/, bj/ etc.? In other words, would there be a difference in sound between these pairs of words (the second of each pair is made up): pięć - pjęć biały - bjały fiołek - fjołek pawie - pawje ziemia - ziemja At pag. 49 of the grammar book there is a discussion of feminine names ending in -ia, where it is said that they can belong to two different classes. The first is constituted by native Polish words, where the ending -ia corresponds phonetically to /'a/ : e.g. ziemia is (I'll use the same transcription as the book) /źem'a/. On the other hand in words of foreign origin -ia corresponds to /'ja/ : e.g. armia /arm'ja/. The book also says that native Polish words admit /'ja/ as an alternate variant (I don't know if some speakers systematically do this, if it depends on the word or if it is a random thing). Does all this make sense to you? Do ziemia and armia rhyme for you? As my native language is Italian the opposition between /nj/ and /ń/ or /lj/ and /l'/ is pretty clear to me. With some imagination I can contemplate /m'/ vs /mj/ but I'm not sure about the other pairs. That is, the difference in sound between /p'/ and /pj/ (etc.) seems tiny to to me. I don't have quality recordings or Polish speakers at hand so I turned to the forum :) What do you think?

jump_bunnyThreads: 8 Posts: 332 Joined: May 14, 09	Edited by: jump_bunny Nov 25, 09, 14:04 / #
Lorenc: I'll jump on this thread for a related phonological question having to do with consonant softening. I'm reading the "Concise Polish Grammar" by Ron Feldstein (to be found at http://www.seelrc.org/projects/grammars.ptml for free). My question is about the sound of softened (palatalised) /p/,/b/,/f/,/v/ and /m/, which this grammar book writes using a trailing apostrophe, e.g. as /p'/,/b'/ etc. These sounds occur whenever in the spelling the letters p, b, (etc) are followed by an "i". My question is: are these sound different, and how much different, from the cluster /pj/, bj/ etc.? In other words, would there be a difference in sound between these pairs of words (the second of each pair is made up): pięć - pjęć biały - bjały fiołek - fjołek pawie - pawje ziemia - ziemja At pag. 49 of the grammar book there is a discussion of feminine names ending in -ia, where it is said that they can belong to two different classes. The first is constituted by native Polish words, where the ending -ia corresponds phonetically to /'a/ : e.g. ziemia is (I'll use the same transcription as the book) /źem'a/. On the other hand in words of foreign origin -ia corresponds to /'ja/ : e.g. armia /arm'ja/. The book also says that native Polish words admit /'ja/ as an alternate variant (I don't know if some speakers systematically do this, if it depends on the word or if it is a random thing). Does all this make sense to you? Do ziemia and armia rhyme for you? As my native language is Italian the opposition between /nj/ and /ń/ or /lj/ and /l'/ is pretty clear to me. With some imagination I can contemplate /m'/ vs /mj/ but I'm not sure about the other pairs. That is, the difference in sound between /p'/ and /pj/ (etc.) seems tiny to to me. I don't have quality recordings or Polish speakers at hand so I turned to the forum :) What do you think? When you pronounce the vowel i, the middle part of your tongue goes up, to your palate. In Polish, there are consonants pronounced with similar upward movement of the tongue if they're followed by the vowel i. They are called palatalised consonants or soft consonants. The important rule of the Polish phonetics is that a consonant is always soft before the vowel i. Most books don't use any special symbol to mark hard consonants but use the ['] symbol to mark softness only when needed. This means, they often don't indicate the softness of the consonant followed by the vowel i because then this is obvious, this must be softened. The labials (and labio-velars) p, b, f, w, m are always soft before i, e.g. biskup - 'bishop'. The groups pi, bi, fi, wi, mi may mark [pj, bj, fj, wj, mj] as well – but only in many words of foreign origin and usually in geographical names. We do not use j after labials in spelling and this causes the difficulty. The difference between [p'] and [pj] is that in [pj] there is no palatalisation of p and you can hear normal j, not only a transient sound as it is with [p']. I hope this helps :) Ziemowit: What nasal vowel do you pronounce in the word awans? From what I know, there is no nasal vowel in this word but only a nasal consonant n.

Lyzko	Nov 25, 09, 14:29 / #
Marvellous thread, guys! Have been waiting for something like this forever. Just a point or two. First, whilst Polish vowels are indeed 'phonemic', as Jumping Bunny said, the language most assuredly has its homophones (Bóg/buk, czy/trzy etc...), as does English (whine/wine, lain-supine form of 'lay'/lane etc..) and many other languages. So, while Polish appears to foreign learners as myself to be a remarkably consistent language, compared, say with English or French, even German, it nonetheless contains irregularities between its othography and its pronunciation, as the above instances 'pięć' vs.'pjęć' etc.. seem to indicate. A Germanic linguist by training, I came relatively late to the Slavic languages, and therefore am reliant on this and other fora, as well as many of the standard texts in the field, e.g. Jakobson, for further guidance:-) Loof forward to hearing from you!

Lyzko	Nov 25, 09, 20:47 / #
Lorenc, Italian too is an almost completely 'phonetic' (along with phonemic) language: all vowels in the standard language are pronounced as nearly exactly as written, so far as I can tell, and I speak only a bit. Secondly, those less than transparent cases, e.g. 'Puglia' (poollyah NOT puglya), with a silent 'g-sound', are always the same. Furthermore, even the exceptions look regular. Finally, the diacritical accents, such as in 'citta' with an acute sign above the 'a', guide the learner as to where the stress lies. Polish 'accent marks' on the other hand, give no such hint as to syllabic stress, but instead, to the quality of the given vowel (Kraków = krahkuff, NOT krakawff) or consonant (świat = sschviyaht, NOT suuiyaht).

cinekThreads: 1 Posts: 264 Joined: Nov 16, 07	Nov 26, 09, 12:29 / #
Lorenc: In other words, would there be a difference in sound between these pairs of words (the second of each pair is made up): pięć - pjęć biały - bjały fiołek - fjołek pawie - pawje ziemia - ziemja No. There's no difference in normal speaking. One could pronounce it a way to make the difference hearable, but it would not sound naturally. Cinek

ZiemowitThreads: 10 Posts: 1,063 Joined: May 8, 09	Edited by: Ziemowit Nov 26, 09, 12:31 / #
Lorenc: My question is about the sound of softened (palatalised) /p/,/b/,/f/,/v/ and /m/, which this grammar book writes using a trailing apostrophe, e.g. as /p'/,/b'/ etc. These sounds occur whenever in the spelling the letters p, b, (etc) are followed by an "i". My question is: are these sound different, and how much different, from the cluster /pj/, bj/ etc.? I don't think there is any difference now. The letter i is used to mark the softeness of the preceding vowel. It is a matter of convention that we use the letter i rather than the letter j to mark the softeness of a consonant (the Russians, for example, use a special soft sign for that). It is true that before 1939 they wrote the j in foreign words such as armja. Later this distinction was abandoned as there was no real need for that (it remains [sic!] in the declined cases as we spell armii, but we spell ziemi). Maybe the pre-1939 Poles pronounced it as armja, so it should sound a bit like arm'ja (saying the soft m first and adding the j after it) or it should sound as arm+ja, which would altogether be different from arm'a (= armia) indeed; today it would sound ridiculous. I don't think you will find any words in contemporary Polish that use the spelling pj, bj, mj, fj in them. Or if you do find them, please let me know. You will, however, find words like wjechał, and this shows that the pronounciation w+j is possible along with the pronounciation wi (w') as in the word wiecha; the pronounciation is different in both, although may be the same when you tell wjechał quickly.

LorencThreads: 6 Posts: 30 Joined: May 13, 08	Edited by: Lorenc Nov 29, 09, 01:55 / #
Thank you all for the comments and sorry for the late reply. Lyzko: yes, Italian spelling is very regular and predictable, overall I'd say a little more than the Polish one. Italian is almost univocal in the passage "oral word"->"written word"; that is to say, if I hear an Italian word there is only one way I could possibly write it down, with the (unique?) exception of the /kw/ sound which could be written down either as "cu" or "qu". This makes dictation exercises easy and confined to the very first years of elementary-school, and the concept of a spelling bee essentially impossible. On the other hand in Polish it would be at times "possible" to write a word in more than one way, especially playing with the end-of-word devoicing, the rz/ż pair, ę/en (etc) and with the rules that prohibits syllables to contain a mix of voiced and unvoiced consonant. Things are different in the "written word->sound" where in Italian there are a few areas where minor ambiguities as to how to read a word may arise, namely with the letters s,z,e,o which admit (sometimes and only in some particular word positions) two possible readings. Polish is in this area very precise, at least when one has learned the voicing/devoicing rules: a Polish written word can be read in only one way... a part from the ziemia/ziemja thing which was the reason for this thread :-) So, I'll end this digression and go back to my original question. By the way: my concern was not with spelling, but just on the phonetic of softened p,b,f,w,m and l. In effect I wasn't sure about their very existence as opposed to the succession consonant + j. Now I see that they do exist and I would like to know if I can pronounce them reasonably correctly. They say a picture is worth a thousand words and I think in this case the same applies to a recording... I made a little experiment. I recorded myself saying the following words: pięć, biały, fiołek, pawie, ziemia, lis. In the first series of words I tried to use _my idea_ of soft consonants for p,b,f,w,m,l. In the second series I said them as I would say them normally, with the consonant + j sound. Here it is, for those of you who'd like to lend an ear to it (300KB): homepages.ucl.ac.uk/~ucapllo/Soft-Consonants.mp3 Do they sound different to your ears? Which series sounds more correct? Ziemowit, you asked about Polish words using the spelling with pj, bj, mj, fj. These are the results of a quick search on the PWN-Oxford dictionary: pj: finds only "maskotka Igrzysk Olimpjskich" which I think is a mistake in the dictionary for Olimpijskich bj: finds loads but only with word-initial obj- , including objaw, objąć, objazd mj, fj : none Thank you, Lorenc

Lyzko	Nov 29, 09, 17:44 / #
My suspicions precisely, Lorenc! For this reasons, Poles and Italians have numerous vocalic/sonic similarities with English, in addition to having more problems in understanding (..if not necessarily in SPEAKING) English. LOL

Lyzko	Dec 2, 09, 14:01 / #
Incidentally, Albanian too seems a most phonetically (if also phonemically as well as phonologically!) 'regular' language, despite an overabundance of alphabetic representation and a stress-movable 'schwa-sound', often NOT indicated by any accent marks:-) Seems no language has a monopoly on transparency!

Lyzko	Dec 5, 09, 14:47 / #
I recently met an Albanian man from Tirane who tried learning Polish on his onw (having already studied Russian in public school!) and told me that he wished he had started off with Polish instead, as he found the latter so phonetically pronouncable in contrast with Russian which he found almost impenetrable. He also studied English at university and said furthermore that Russian stress, ellisions and vowel mutations is far more similar to English than either to Polish or Albanian. If learning a third language, the man went on to say that it would definitely be either Italian or Rumanian (..but interestingly enough, not Greek)! Just an aside:-)

Lyzko	Dec 6, 09, 21:06 / #
Hi back, Lorenc! My money's on Turkish of this bunch. I've been seriously engaged in learning it on my own, and haven't come across an orthographic/pronunciation exception yet-:) Am still looking and will post again when I do. LOL The rest I'm cursorily familiar with and have to say that of the Slavic languages, Serbo-Croatian probably has even Polish beat out here. Hungarian for me remains a bit of an enigma wrapped up in a puzzle. While the spelling sure looks phonetically consistent, speaking what I read and writing it as I hear presents more difficulties than either of the above languages when I started learning them. My dictations in beginner Polish were usually pretty near perfect. With Hungarian? Lots and lots of red from my teacher, even after the fourth week. By the end of my first half semester of Polish, dictations were the least of my problems. The aspect system proved the biggest headache by far.

Lyzko	Dec 6, 09, 21:21 / #
My most recent forrays into Finnish suggest that, while regular, i.e. consistent in vowel as well as consonant quality, it's pronunciation is scarcely transparent to an outsider with zero working knowledge of the language. Just be looking at those diacreses and vocalic combinations, I wouldn't have had a clue. Italian, on the other hand, seemed effortless to pronounce after a mininal amount of coaching, having never even been to Italy before learning to speak it a little. I would have fared far worse in Finnish, I fear!

Lyzko	Nov 29, 09, 17:46 / #
Mistyped previous post. I meant, of course, Italian and Polish have numerous such similarities with one another, indeed NOT with English!!!!! Therefore, quite the opposite from what I posted just now. Got distracted, I guess, when keyboarding too quickly))) Sorry:-)

ZiemowitThreads: 10 Posts: 1,063 Joined: May 8, 09	Dec 2, 09, 13:44 / #
Now that I come to think of it again, in seems to me that there isn't really any "certain consonants+j" thing in Polish. Most consonants before the "j" are just soft ones as they would have been before the "i". The reason why we write wjechać or objaw or objazd is that we can't write it as wiechać or obiaw or obiazd because there exist a principal rule in Polish which says that we should stick to the original writing of core words in compound words such as objaw. The core word in it is jawa/jawić, so any of the derivative words should retain the j in them: ob+jaw, z+jawa. We pronounce objaw as if it were written obiaw; an interesting thing is with the word zjawa - I would never say we pronounce it as if it were written ziawa, as ziawa and zjawa would be different pronounciations. It is perhaps here where the problem "consonant+j" versus "soft consonant" is clearly visible. P.S. "Olimpjskich" is definitely a mistake.

LorencThreads: 6 Posts: 30 Joined: May 13, 08	Dec 6, 09, 17:29 / #
Hi Lyzko, I cannot comment on Albanian but some time ago I discussed with some people (for fun) which languages have the most regular spelling. There are many which are practically perfectly regular, with a one-to-one correspondence between sound and spelling: for example Turkish, Serbo-Croatian, Finnish and Hungarian should all be very very regular.