[Haskell-cafe] Re: OT: Languages

[wren ng thornton]

Tillmann Rendel wrote:
> wren ng thornton wrote:
> > Indeed. The proliferation of compound words is noteworthy, but it's 
> > not generally considered an agglutinative language. From what (very 
> > little) German I know compounds tend to be restricted to nouns, as 
> > opposed to languages like Turkish, Japanese, Korean,...
> 
> Yes, compounds are restricted to nouns in German. But as I understand 
> it, agglutinative relates more to the inflection system than to the 
> lexicon anyway.

In general, I'm not sure I draw a distinction there. What belongs in the 
grammar vs what belongs in the lexicon is rather fluid and depends on 
both the language and the theory in question; whereas the phenomenon is, 
I think, easily identifiable (if not always easily definable). That is, 
the distinction between agglutinative vs fusional is typological rather 
than theoretical.

The distinction has to do with information content per morpheme (or 
compositional vs idiomatic information construction). For determining 
this, root/base morphemes are included just as much as inflectional 
morphemes. The distinction between what is "root" vs what is 
"inflection" is a spectrum and not always clear cut, especially in 
agglutinative languages. In languages like Japanese which lacks spaces, 
this difficulty is highlighted by the fact that it's not always clear 
whether something is a "word" or a "phrase" (and hence whether the 
latter major segment contains base morphemes, or is "only inflection").

Though yes, the distinction is most clearly observed by looking at 
verbal inflections. And now we're really far off topic :)

-- 
Live well,
~wren