Newsgroups: sci.lang
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!news.alpha.net!mvb.saic.com!eskimo!rickw
From: rickw@eskimo.com (Richard Wojcik)
Subject: Re: Russian vowel  bI
X-Nntp-Posting-Host: eskimo.com
Message-ID: <D5zCBv.64F@eskimo.com>
Sender: usenet@eskimo.com (News User Id)
Organization: Eskimo North (206) For-Ever
References: <3kcq7c$167k@news.ccit.arizona.edu> <3kpp90$phs@news.ycc.yale.edu>
Date: Sat, 25 Mar 1995 04:55:06 GMT
Lines: 88

In article <3kpp90$phs@news.ycc.yale.edu>,
Ken Balakrishnan <kbalakri@minerva.cis.yale.edu> wrote:
>Hung J Lu (hlu@helium.gas.uug.arizona.edu) wrote:
>
>: One first question. I can swear that the vowel
>: bI (is this transliterated as "y"?) is pronunced
>: very differently in isolation and when following
>: a consonant, like in  TbI  or  BbI. What is going
>: on? (what would be the IPA symbols, I believe in
>: one case it is a hyphenated i)
>
>It is indeed transliterated as "y", and, like all other vowels in Russian 
>or in any other language I know, its precise pronunciation varies 
>depending on the phonetic environment.  It's a high back unrounded vowel, 
>but I don't know what the IPA symbol is.  Actually, from a phonological
>standpoint, the "y" is not a full-fledged phoneme but the variant after 
>hard consonants of the vocalic phoneme +high, -rounded.

It is more complicated than that.  Native grammarians have debated this
point for decades, with the St. Petersburg/Leningrad school arguing for
separate phonemes and the Moscow school arguing for a single phoneme.  The
progenitor of phonological theory, Baudouin de Courtenay (who founded the
St. Petersburg school), originally argued for a single phoneme.  Whatever
the historical facts, it now appears that the two sounds can be used to
contrast words.  It is true that a preceding consonant dictates [y] if it
is unpalatalized and [i] if it is palatalized, but [y] and [i] appear to be
able to distinguish words when there is no preceding consonant to govern the
front/back character.  What we really have is a case of partial phonemic
overlap, something like the English intervocallic flap.  (Well, maybe.  I
am not a native speaker.  This is my current perception of where the debate
stands.)

>: And I have no idea about rules for diphtongs...
>: I think the ending ...aya (with the backwards R)
>: is pronounced only as a schwa, as in moskovskaya.
>: (my book says that unstressed ya is pronounced
>: as yi... I don't think this is true in this 
>: example).
>
>Again, you're right.  The pronunciation rules often do not apply to 
>grammatical endings.  The unstressed ending "-aya" should be pronounced 
>with two schwas, the nominative ending "-ya" (as in "volya") with one.

Russian, like English, has a very active phonological process of vowel
reduction.  (The original analyst of Russian vocalism was, I believe, the
19th century linguist Potebnia.)  Vowel reduction varies wildly in
different dialects (cf. the work of Durnovo).  The basic (perhaps
oversimplified) facts are as follows for literary Russian:
     1)  There is only one stress per word, no secondary stresses as in
         English.
     2)  All unstressed vowels reduce to schwa, except word-initial vowels
         and those before stress (i.e. "pretonic" vowels). 
     3)  Palatalized consonants cause the following vowel to "front", which
         means that unstressed "ya" and "ye" tend to sound like "yi".
     4)  The effect of palatalization on a following vowel is less
         pronounced in the high vowels /i/ and /u/, which has led some
         phonologists to claim that unstressed Russian syllables can only
         have three phonemes: /i/, /u/, and /a/. 
This is why the city Pyatigorsk is pronounced something like [p'it'igOrsk],
with stress on the final syllable.  The adverb for "good", xorosho, is
pronounced [x@rashO], with stress on the final syllable, but the adjective
xoroshoe is pronounced [xarOsh@y@], because the pretonic vowel cannot
reduce fully to schwa [@] (Note the stress shift to the second syllable in
the adjective).  In summary, Russian reduces unstressed vowels, as does
English, to schwa.  The trick to remember is that a palatalized consonant
causes the schwa to "front"--i.e. become more like [i]--especially in
syllables that precede the word stress.

The unique "strong" behavior of pretonic vowels in Russian has an
interesting effect in fast and casual speech styles.  If you are speaking
quickly, you tend to omit or delete the weakest vowels.  Thus, a native
Russian speaker will tend to pronounce "good" (xorosho--[x@rashO]) as
[xrashO], with the initial vowel deleted.  English speakers tend to speak
with alternating stress.  That is, every other syllable, counting from the
primary stress, has a secondary stress.  Unlike Russians, English speakers
tend to weaken the pretonic vowel and strengthen the one before the
pretonic vowel.  Therefore, when English speakers pronounce xorosho
[x@rashO], they tend to reverse the reduction: [xar@shO].  In fast speech,
an English accent is invariably [xarshO], as opposed to the native
[xrashO].  This greatly affects the comprehensibility of speech between
English-speaking learners of Russian and native Russians.  The two language
groups tend to preserve different vowels in fast and casual styles, even
though the two languages share the trait of having vowel reduction in
unstressed syllables.

-- 
Rick Wojcik  rickw@eskimo.com     Seattle (for locals: Bellevue), WA
             http://www.eskimo.com/~rickw/
