Malo sam diskutovao o tome po Vikipediji…
Poenta je sledeća: 1) Vikipedija uvek prati postojeću praksu u literaturi, koja te glasove univerzalno zapisuje kao /ʃ ʒ/, a ne kao /ʂ ʐ/, 2) Vikipedija generalno daje relativno široku fonemsku transkripciju, gde se fini fonološki detalji apstrahuju zarad jednostavnosti i univerzalnosti zapisa. Već postoji generalni otpor "učenih" čitalaca "can’t understand that IPA gibberish", pa bi preterano inisistiranje na detaljnosti zapisa dovelo do komplikovanosti i nekonzistentnosti.
Koliko sam ja, amaterski, upratio problematiku, stanje je sledeće: Niko ko se detaljno bavio fonologijom srpskohrvatskog i ne tvrdi da su naši post-alveolari "kanonički" niti da su jednaki engleskim ili španskim. Ono što nedostaje da bi bili opisani kao retrofleksni je ta povijenost jezika (iako ste u pravu da su akustički daleko bliži poljskim [ʂ ʐ tʂ dʐ]).
Brus Moren [
link] te glasove u fonetskom zapisu daje kao
[ʃ̺] [ʒ̺], [t̺ʃ̺ʷ] i [d̺ʒ̺ʷ]tj. da su to apikalni (artikulisani samim vrhom jezika) i kaže:
As Miller-Ockhuizen and Zec (2003:165) have shown, differences in lip protrusion/compression “categorically distinguishes the two affricates” because lip protrusion causes a significant lowering of the spectral peaks associated with the frication noise of the affricates. As we will see in section 4, this lip rounding is best analyzed as a phonetic enhancement strategy, not as an indication of a phonological feature.
Zilke Haman (
link) se vrlo detaljno bavila komparativnom fonologijom retrofleksnih i post-alveolarnih glasova. Na primer,
ovde kaže da
Apart from Polish, Keating (1991: 35) classifies only Russian and Serbian as Slavic languages with retroflex fricatives, but merely on phonetic grounds. […]
If Keating’s claim (1991) is right that Serbian also has a retroflex fricative (recall the introduction), this might refute the hypothesis […]
As T.A. Hall (p.c.) points out, the Serbian system does not refute the hypothesis if one assumes that the change from /ʃ/ to /ʂ/ is a language-specific process in Serbian that occurred independently at a later stage.
Spomenuti Kiting je Keating, P. A. 1991. Coronal Places of Articulation. In The Special Status of Coronals: Internal and
External Evidence, eds. C. Paradis and J.-F. Prunet, 29-48. San Diego: Academic Press, pdf sken
ovde, a on sa sve slikom položaja jezik citira Miletić R., (1960) Osnovi fonetike srpskog jezika, Naučna Knjiga, Beograd. E do tog ne mogu da dođem.
Malgožata Ćavar (
link) sa Zagrebačkog univerziteta takođe se bavila komparativnom analizom poljskog i hrvatskog, i
ovde (u analizi zašto se u hrvatskom spajaju /č/ i /ć/) kaže:
the classification of /č/ and /dž/ is problematic even without considering the variation discussed in the paper, the symbols used in literature ranging between IPA /tʃ/ and /dʒ/ over less-specific American symbols /tš/ and /dž/ to IPA retroflex symbols /tʂ/ and /dʐ/. Since the goal of the paper is the discussion of the category internal phonetic variation, the orthographic symbols are used as least qualifying.