Re: THEORY: Evolution of infixes/ablaut?

At 09:11 PM 3/15/2000 -0500, Padraic wrote:
On Wed, 15 Mar 2000, Eric Christopherson wrote:

 >Hi. I'm wondering if anyone has any information or even ideas about how
 >languages develop alternation inside of morphemes? That is to say, where
 >morphemes can be inflected or otherwise modified by changing, adding, or
 >deleting elements _inside_ the morphemes themselves, such as with infixes
 >and ablaut (vowel alternation). I'm really fascinated by the idea but I
 >can't figure out how the mind would allow a morpheme to be modified from
 >the inside -- just seems like morphemes "should" be concrete, unbreakable
 >elements to me. It's a bias in my conlanging instinct I guess :)

And of course, you've used vowel-change inflection 3 times so far. :)
Though no infixes that I can see. I'm curious, though: why would you
think a morpheme is inviolable?

I assume you mean the English ablaut; at first I thought you meant I had
used alternation in conlangs 3 times and wondered how you concluded that
:O) Anyway, I don't know. It just seems to me like a morpheme would be
fundamental and atomic, even though obviously the converse has been
demonstrated.

Certain ones, like umlaut changes in English, could be demonstrated
through successive stages of the language:
maniz > mani > meni > men; where i raises a
kind of thing. If your language is ancient enough and written, such
changes would be recorded, rather than inferred like in English.

Patrick Dunn added:
I *do* know how some vowel alterations arose. In some words, a suffix
causes an anticipatory change in a previoius vowel. So, say, in the
language Gluk (which I just made up on the spot, ain't I clever):

Yes, I've heard of this phenomenon before, and have already decided to use
it a lot in Dhak, but I haven't found any way to use it to arrive at
Semitic-style roots. I think that if I do go with Semitic-style roots I
will use anticipatory ablaut at a _later_ stage of the language.

Nik said:
Perhaps sometimes it's simple metathesis. Suppose that the plural infix
was -l-, placed before the final consonant. Well, it could've been that
at an earlier stage it was a suffix -l, and forms like, say, _pakl_
became _palk_. That's just a guess, tho, and couldn't explain all
infixes, either.

Ah, neat idea. I think I've heard of this happening in Navajo too, but I'm
not sure.

Another guess is that perhaps it started with only some words.
[snip]
Perhaps the old prefix was reanalyzed as part
of the root, and then -um- was generalized to all words, regardless of
whether they historically began with a prefix.

That's a good idea I will have to take into account also.

Quoth Roger Mills:
As well as think-thought, bring-brought et al., similarly in German/Dutch
that I know of- how about Swedish? Latin tangere-tactus et al. Greek?
probably. Sanskrit has a class n-infixing verbs like (IIRC) root /ruc/
'shine' > rinokti '3rd. sg.' So evidently of IE lineage, tho seemingly a
marginal procedure even at that remove.

Yep, there was a whole class of roots in PIE that had an <n>/<ne> infix in
certain inflections.

For years I've pored over a mini-family of langs. spoken in the islands east
of Timor-- they have very interesting ways of binding together
constituents/phonological phrases. E.g. verbal conjugation: Leti /mu+laa/
'you go' > mlwaa, compound /pipi/ 'goat' + (borr.) /duma/ '' 'sheep' >
pipdiuma 'sheep'. One investigator found that at least some of this
metathesis/infixing was due to fast-speech rules.

Same comments as to Nik above, on the word _palk_.

Pues Carlos escribio:
About infixation, I've noted that in my idiolect, the Spanish diminutive
suffix -ito is actually an infix:
[snip]
A Colombian trade mark uses _azuquita_ as diminutive for _azúcar_, as -ito
were a suffix after dropping the unstressed ending, but in my idiolect the
more natural form is _azuquitar_, with -it- infixed before the last and
unstressed vowel.

Wow, now this is interesting. I've thought a bit about this before, but
have never heard of anyone infixing in Spanish. Does anyone else say
_azuquitar_? (Sadly "small sugar" seems like a concept which would be very
rarely discussed.)

Thanks to everyone for the comments :)