Nahuatl and CJ [vowel length question]

[David Robertson]

Note to Henry K (non-linguists, hit "snooze"),

You asked as an aside whether there is a phonemic contrast (difference in
meaning of otherwise identical words) involving short vs. long vowels in
CW.  I think that's a great question.

It's clear that each word in CW has been long established with a particular
accepted pronunciation of its vowels.  In some words you find a "long /i/",
the vowel in English "seed", while in others you find "short /i/", the
vowel in English "sit".  CW examples are /t'Lip/ 'slip [et al.]' (short i)
and the Kamloops Wawa word /slip/ (presumably long i) 'sleep'.  Other
vowels could be said to have long and short variants as well: /u/ with the
vowels of English "too" and "foot", and /a/ with the vowels of
English "car" and "cut".  Maybe even /o/ and /e/.

Maybe the pairing we see is due to the so-called "tense versus lax"
distinction that's traditional in describing English.  (Note: Tense vs. lax
might not be phonologically justified, last I heard.)  I feel English has
influenced the vowel system of CW enormously.

But the tough part is that there are so few "words" in CW.  By which I mean
there are only around 500 distinct morphemes -- though of course there are
many times more words once you count combinations thereof that have
nonpredictable or lexicalized~idiomatic meanings.  The small lexicon makes
it difficult to find minimal pairs!

(Not very explanatory note for the non-linguists: Minimal pair = two words
in a language, differing only in one feature/sound.  [For example,
differing in vowel length or differing by just one consonant.]  The
difference in sound is what makes the two words have different meanings.
This sort of test is the classic way of deciding whether a particular
sound, or feature like vowel length, is a phoneme in the language, one of
the basic "atoms" of linguistic analysis.)

The riddle that all of this seems to present is whether we can prove that
CW has certain phonemic contrasts, not just in vowel length but also for
example in aspirated vs. unaspirated stops.  I haven't methodically checked
yet, but it might even be hard to "prove" that ejectives (the popping
sounds) are separate phonemes!

I'm very interested in the theoretical issues that this raises.

-----Why should certain contrasts be apparently maintained in this
language, in the form of quite precisely defined pronunciations, if there's
no phonological justification for those distinctions?

-----Surely we can simply decide that there's a reasonable, if not
conclusive, amount of evidence for such distinctions' existence in CW --
but then, how do we justify our claims about the nature of those
distinctions?  E.g. why not say CW indeed has tense vs. lax vowels (or even
retracted vs. advanced tongue root), rather than long vs. short ones?

-----If certain distinctions are unprovable, does each word have its own
independent phonology (lexically specified for a given, ridiculously
precise, pronunciation)?

-----And if we're taking certain distinctions on faith rather than on CW-
internal evidence, aren't we now vulnerable to claims that CW has no
phonological system of its own, or one which contains only underspecified
archiphonemes?

Darn it, I'm going to have to put together some data and write a paper.

LhExhayEm (or //laxayam//)  :-) ,

--Dave R.

To respond to the CHINOOK list, click 'REPLY ALL'.  To respond privately to the sender of a message, click 'REPLY'.  Hayu masi!