Re: Infixing in interlinears

Roger Mills said:
> Mark P. Line wrote:
>
>
>>
>>       http://www.eva.mpg.de/lingua/files/morpheme.html
>>
> One of the more useful sites mentioned recently.
Yep. Google turns up a number of related articles as well (by Lehmann,
among others).


>> So assuming your example root 'fish' means "cumquat" in English, and
>> assuming that the -il- infix is left-peripheral, your example in Leipzig
>> glossing would be:
>>
>>       f<il>ish
>>       <DIM>cumquat
>>       'little cumquat'
>>
>> If you needed to treat -il- as right-peripheral instead, you'd gloss it
>> as
>>
>>       cumquat<DIM>
>>
> But if it were R-peripheral, wouldn't that imply that the infix form was
> **-li-? Given that both the infix vowel and the stem vowel happen to be
> the same in this case, it's open to debate. Any other ex. -- perhaps "dash
> ~?dilash or ?dalish" '(little) mango"-- would show which it is.
I thought these new-fangled math-head terms left-peripheral and
right-peripheral had to do with constituent order (which is rather fuzzy
in the case of infixes). (Rule of thumb: if the term is not used in either
the SIL glossary or in Payne, it's probably an artifact of somebody's
formalism.)

I would expect the form of the infix to co-evolve with the way in which it
patterns phonotactically with the root or stem. If it occurs between the
initial C and the following vowel in a predominantly CV language, I'd
expect the infix to be VC, for example.


> I have an on-going argument (in my head, that is) with some of my
> Indonesianist colleagues over this.  In a certain group of languages,
> there's a nominalizing infix that works as follows:
> herun (vb.) > henerun (n.) 'exchange'
> pali (Vb) > panali (n) 'anchor(age)
> hukum (vb.) > hunukum (n) 'judge(ment)' etc.
>
> Wedded I guess to the Gospel truth that in synchronic analysis you
> _mustn't_
> peek at the history, many persist in claiming the infix to be -nV- (at
> least
> they get the V-harmonizing part right), whereas the slightest knowledge of
> AN history would show that the infix is -Vn- and is the local form of the
> Object Focus *-in- seen in Philippine langs. </rant>

Hehe. Why include historical data when it's so much easier to limit
yourself to a bit of introspective mysticism? Sometimes we can be happy if
they even use synchronic data from actual native speakers.

Clearly, a more broadly-based model of AN languages will be simpler if
these infixes are treated as -Vn- plus the new V-harmonization pattern.


-- Mark