Re: "discontinuous affixes"

>Tom Wier wrote:
>> Well, why should we restrict the meaning to a root?  I mean, what's
>>happening
>> is this:  you have a prefix like "nala-" in your example, which has one
>>meaning,
>> is one morpheme, and than you insert another prefix inside that to alter the
>> word further, with the two constituent elements of the original prefix
>>remaining
>> otherwise intact...  what makes that any less of an infix than one that
>>goes on
>> a root?
>
>Hmmm .... I s'pose you *could* extend the meaning to being inside an
>affix, but I think I personally would analyze nala- as two prefixes,
>which usually (or always) co-occur, frequently adjacent to each other.
I'm inclined to agree with you, Nik - although there are indisputable
cases of an infix appearing inside another affix, suggesting that the
proper definition of "infix" is "an affix which appears inside another
morpheme", rather than "an affix which appears inside a root".

A nice example of an infix inside a prefix comes from Tagalog:  Stems
may be manipulated by adding the infix "-um-" before the first vowel,
or by reduplicating the initial CV of the stem (can't remember what
"-um-" and reduplication mean off the top of my head, unfortunately...).
When both apply to the same stem, reduplication applies first, after
which "-um-" is inserted within the reduplicate:

        stem:                           higan
        infixation:                     humigan
        reduplication:                  hihigan
        infixation + reduplication:     humihigan

Matt.

------------------------------------
Matt Pearson
mpearson@ucla.edu
UCLA Linguistics Department
405 Hilgard Avenue
Los Angeles, CA 90095-1543
------------------------------------