Newsgroups: sci.lang
Path: cantaloupe.srv.cs.cmu.edu!europa.chnt.gtegsc.com!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: <D6LEvo.Bnz@eskimo.com>
Sender: usenet@eskimo.com (News User Id)
Organization: Eskimo North (206) For-Ever
References: <1995Mar31.164904.9015@grace.rt.cs.boeing.com> <3ljtaq$erc@usenet.srv.cis.pitt.edu> <D6DAv9.n24@eskimo.com> <D6J46F.C1H@cogsci.ed.ac.uk>
Date: Thu, 6 Apr 1995 02:57:24 GMT
Lines: 68

In article <D6J46F.C1H@cogsci.ed.ac.uk>,
Ivan A Derzhanski <iad@cogsci.ed.ac.uk> wrote:
>In article <D6DAv9.n24@eskimo.com> rickw@eskimo.com (Richard Wojcik) writes:
>>In fact, we are probably agreed that the present cyrillic system is
>>superior to a purely one-to-one correspondence for Russian, which would
>>require more symbols to represent the palatalized/nonpalatalized split in
>>the consonantal system.
>
>Would a system not count as one-to-one if, in recognition of the fact
>that palatalisation is a secondary feature, it marks it with a diacritic
>on the corresponding letter instead of using a second set of unrelated
>letter shapes?

Technically, the system would not be one-to-one, but "technically" is
pedantic here.  Diacritics are excellent ways to capture secondary
articulations.  Russian cyrillic is interesting for some of the
idiosyncracies it has developed--e.g. spelling rules governing what vowel
symbols can follow inherently soft consonants (ch, shch) and inherently
hard consonants (sh, zh), etc.

>...
>Regarding those place names where _y_ appears in syllable-initial position,
>I vote that they be counted as words of the language, because they decline
>(that is, when one of them doesn't, it's because its final prevents it from
>so doing, not the initial or medial _y_).  This is a criterion which only
>works in one direction: not every Russian noun declines, but everything
>that declines like a Russian noun is one.

Right, but I would consider foreign loan words to be "Russian" even when
they don't decline.  The relevant point is that they must conform to
constraints on Russian articulation, which has more to do with the mental
program that coordinates articulatory gestures during speech.
Morphological constraints--e.g. declensions--are not really motivated by
articulatory needs.  I take phonology to be just the same domain that
Trubetzkoy, Baudouin, and Sapir took it to be--the domain of
articulatory/perceptual constraints on speech, not the phonemic patterns
associated with morphemes.

>This would mean that the distinction between the two allophones of the
>phoneme /i/ has been phonemicised.
>
>Now [N] also exists as an allophone of /n/ before velar stops, but
>the distinction between [n] and [N] isn't likely to become phonemic
>any time soon, even under an avalanche of English loanwords.  How is
>the case of [n N] different from the case of [i y]?  Does the fact
>that the latter distinction is reflected in the orthography make
>all the difference?

Interesting comments.  I wasn't aware that nasal assimilation was making
headway in Russian, but it isn't surprising.  Historically, /n/ came to be
juxtaposed to a consonant via vowel deletion word-internally.  It is very
unusual for a language with nasal+consonant clusters to resist
assimilation.  

>And if an English speaker pronounces _Xhosa_ with a click or _Nahuatl_
>or _Klingon_ with a lateral affricate, does that make those consonants
>phonemes of (his variety of) English?

That depends.  It is interesting that Russians pronounce those initial /y/
words with [y] in far greater frequency than English speakers pronounce
"Xhosa" with a click or "Nahuatl"/"Klingon" with lateral affricates, isn't
it?  I'm not sure that it matters much if you can pin the cause on the
existence of a letter.  It may be unusual for an alphabet to play a role in
causing a phonemic split, but that doesn't necessarily make the split
itself less real or interesting.
-- 
Rick Wojcik  rickw@eskimo.com     Seattle (for locals: Bellevue), WA
             http://www.eskimo.com/~rickw/
