Talk:Polish phonology

Actual Polish pronunciation

• y is close mid
• cz, dż, sz, ż are post-alevolar, not retroflex; retroflex pronunciation is considered as speech defect
• c, dz, cz, dż, ć, dź are stop consonants
• and c, cz, ć pronunciation Has nothing to do with t

bananowiec 10:30, 25 February 2023 (UTC)

Changing analysis of nasals

"Recent sources present for modern Polish a vowel system without nasal vowel phonemes, including only the aforementioned six oral vowels." Is that because the pronunciation of these sounds changed in the 20th century? Or did this change occur much earlier, and wasn't it until recently that linguists started to acknowledge it? Steinbach (talk) 19:05, 26 August 2023 (UTC)

Antepenultimate stress

> Some loanwords, particularly from classical languages, have the stress on the antepenultimate (third-last) syllable. For example, fizyka (/ˈfizɨka/) ('physics') is stressed on the first syllable. That may lead to a rare phenomenon of minimal pairs differing only in stress placement: muzyka /ˈmuzɨka/ 'music' vs. muzyka /muˈzɨka/ – genitive singular of muzyk 'musician'. When further syllables are added at the end of such words through suffixation, the stress normally becomes regular: uniwersytet (/uɲiˈvɛrsɨtɛt/, 'university') has irregular stress on the third (or antepenultimate) syllable, but the genitive uniwersytetu (/uɲivɛrsɨˈtɛtu/) and derived adjective uniwersytecki (/uɲivɛrsɨˈtɛt͡ski/) have regular stress on the penultimate syllables. Over time, loanwords tend to become nativized to have a penultimate stress.

This is just plain incorrect in contemporary speech. It's very outdated. Pronounciations like /ˈfizɨka/ and /ˈmuzɨka/ are pretty much unheard of, and for many generations now.

"Over time, loanwords tend to become nativized to have a penultimate stress" is how you could describe this process a century ago, the process is complete now. 90.254.230.66 (talk) 16:43, 22 November 2023 (UTC)

The process obviously not complete. That would be the case only if there were no more words with irregular stress. But in fact the number of such words might be higher now than ever before, simply because the number of borrowings is higher. If initial stress in muzyka and fizyka is now "unheard of" (is it though?), this simply shows that the process is still active, but not at all that it is "complete". 2.207.102.112 (talk) 00:00, 24 November 2024 (UTC)

Number of palatalizations

"four Proto-Slavic palatalizations and two further palatalizations that took place in Polish and Belarusian."

So six palatalizations in total? It would be nice to know which source was used for this sentence. Most places I can find mention three Proto-Slavic palatalizations — two regressive palatalizations and the progressive palatalization) — not four. In Old Polish it says that the further palatalization that took place in Polish is sometimes called the "fourth Slavic palatalization". Even then, what are the two remaining palatalzations that took place in Polish and Belarussian? SKOgoras (talk) 08:20, 6 April 2024 (UTC)

Vowel charts

First of all, do we need that many charts? IMO we should pick one and just list the various realizations mentioned by various sources in the article. That's how it's done in other phonology articles.

I analyzed my speech in Praat (see this chart which, if I understand Wiki Commons rules correctly, I shouldn't upload there), and Rocławski's chart is the most accurate (and the richest), though not without flaws. Does Sawicka 1995 feature a vowel chart? The realizations she mentions ([ʉ, ɵ, æ]) are very much like those I use, at least in some contexts (e.g. after /j/, as in "już" [jʉʂ], "jutro" [ˈjʉtɾɔ], "Julia" [ˈjʉljæ̈] - note that it doesn't matter what follows the vowel. I have [ɵ̞] for the first /ɔ/ in "jojo" [ˈjɵ̞jɔ̈] and [æ̈] in "Jaś" [jæ̈ɕ] and "jak" [jæ̈k] (and "Julia" above) - again, it doesn't matter what follows /a/. AFAIK this can only be near-open in my speech, I don't use [a] at all, it's an alien phone I'd have to learn to use if I wanted to speak, say, RP. A central [ä] often (or very often - I'm too lazy to determine that) comes out as [ɐ] in my speech (like Greek /a/) - I can't be alone in this, and I see no mention at all of this in the article ([ɐ] is mentioned in two different contexts there). I think that my default phone here is [ɐ]. Furthermore, my /ɛ/ is near-front [ɛ̈] (in accordance with Wells's observation - it's spot on) in non-palatal contexts (as in "te" [tɛ̈]) and between a hard and soft consonant (as in "weź" [vë̞ɕ]), where it's raised to [ë̞]. This is where a (near-)merger with /ɨ/ can take place, and Rocławski seems to describe it. I have a fully front vowel only after palatals, as in "nieś" [ɲeɕ] and "siekać" [ˈɕe̞kɐ̟tɕ].

My /u/'s can become near-close almost by chance, as in "stój" [stʊj] and "ciut" [tɕʊ̈t]. This isn't exactly like, say, German /ʊ/ which is less rounded than this. My allophony of /ɔ/ matches the way Rocławski describes it almost perfectly - it can be fronted or raised, or both. It can become unrounded [ʌ] when not in contact with soft consonants and it can become close-mid back rounded [o] in "iłołupek" [iwoˈwupɛ̈k] (thanks to Piotr Rybka and his blog for teaching me that word) and in one pronunciation of "no" [no ~ noː] (and probably also in some other words, either way I have this phone available to me when I speak and I don't feel to be completely non-native to my idiolect like, say, [a] and [ɶ]).

My allophony is more Russian-style than most sources seem to suggest is the norm. If we wanted to transcribe the way vowels change in contact with palatals then using ⟨ʉ, e, ɵ, æ⟩ is the way to go. They're really good and accurate symbols for that. Even when [ʉ, ɵ] are back-central, rather than fully central, [e] is mid, rather than close-mid and [æ] is front-central, rather than fully front, it's still a reasonable transcription - and much better than using diacritics. Sol505000 (talk) 20:15, 26 November 2024 (UTC)