Re: Small Derivational Idea

On Mon, Feb 23, 2009 at 2:28 PM, David J. Peterson <dedalvs@...> wrote:
> On Feb 23, 2009, at 2∞03 PM, Garth Wallace wrote:
>
>> On that note, does anyone know how natlangs determine where in a
>> morpheme an infix may be inserted?
>
>
> It's based on phonological constraints usually (languages don't
> know about "morphemes".  They're not a very useful theoretical
> device, in my opinion).  So, for example, in Tagalog it comes
> after the onset, or just appears as a prefix if there isn't one.
> There was an interesting paper on the psychological cohesion
> (if that's the term I want) of onset clusters using infixes in Tagalog.
> Originally, there were no onset clusters in Tagalog, but contact
> has brought words with onset clusters into the language, e.g.
> "gradwet" (from English "graduate").  With /-um-/ there were
> two possibilities:
>
> (1) gumradwet
> (2) grumadwet
>
> Speakers consistently prefer (2).
>
> I can only recall seeing infixes that infixed right after the first
> consonant or syllable (prefixing infixes), or right before the
> last consonant or syllable (suffixing infixes).  It's theoretically
> possible (though likely unattested) for a language to do almost
> anything, including:
>
> (3) Infixing after/before a certain number of phonemes.
> (4) Infixing as close to the direct center as possible.
> (5) Infixing after an even syllable, or suffixing to monosyllabic words.
> (6) Infixing after the lowest alphabetic letter.
>
> Etc.
>
> -David
Thanks!

I don't think I'm ready to jettison morphemes. Some things seem tough
the explain without them, like clitics and incorporation.