Re: polysynthetic languages

On Friday, September 19, 2003, at 02:30  AM, Chris Bates wrote:

>> A polysynthetic language is a language that inflects the verb to the
>> point that
>> free word order is possible. They often also allow noun-verb
>> incorporation and
>> can thus express an entire sentence in a single word, like this:
>>
>> Xek'ûla'ikêlama.
>> Xe-k'ûla-'ikê-la-ma.
>> want.to-house-make-3SOI-1SSA
>> I want to build a house.
>>
>>
>>
> Often, I would say that this degree of synthesis actually enforces
> pretty strict order of morphemes... when a sentence can be one long
> word, word order isn't really that relevant is it? All I mean is... I
> wouldn't characterise polysynthetic languages by the ability to have
> "free word order", since the order of the building blocks of sentences
> often seems as strict as the strictest isolating language. I'm sure
> that
> in your lang for instance I couldn't write
>
> k'ûla-'ikê-ma-la-Xe
>
> and still mean that I want to build a house, even if it makes sense. I
> thought what Christophe said about french being polysynthetic was quite
> interesting actually... and it made my wonder again like I have before,
> what apart from the position of stress is the difference between a very
> isolating and a very agglutinative or polysynthetic language? I mean,
> what's to stop me analysing your example above by:
>
> Xe k'ûla 'ikê la ma.
>
> Five words of an isolating language with strict word order, and saying
> that only one of them is stressed? If stress is the only difference
> between an isolating and a polysynthetic language then it seems like
> the
> distinction is over emphasized.
There can be other differences. Stress just happens to be a good
example of a phonetic category which is most often associated with word
status. If a word has a single main stress, counting up the stresses in
a sentence can give you an idea of how many words it contains. There
are problems with it, of course. Clitics, which for some analysts are
separate words, don't receive stress, and compounds often receive more
than one stress.

Other phonetic or phonological clues for word status could be found.
For example: final-vowel devoicing is only *word*-final in Shoshoni
(not syllable-final), intervocalic lenition in Paiute (p > v, t > r,
etc) only occurs within a word and not across word boundaries, vowel
harmony is often restricted in its application to vowels within the
same word, and so on. Many phonological processes in the world's
languages are domain-specific in this way; 'word' is one domain within
which such processes can apply.

Dirk
--
Dirk Elzinga
Dirk_Elzinga@byu.edu

"I believe that phonology is superior to music. It is more variable and
its pecuniary possibilities are far greater." - Erik Satie

Reply