Re: Conciseness

Quoting "Mark P. Line" <mark@...>:

> A *phonological* word has to have at least one syllable, but it could
> easily map to three or more grammatical words. Morphemes represented by a
> single segment or by suprasegmental processes will tend to be from closed
> classes, so the point you're trying to make is probably that open-class
> morphemes will almost always have at least one syllable and that therefore
> a prototypical simple transitive clause constructed from open-class
> morphemes will have at least three syllables.
>
> I think that's probably a fair assumption, although there are languages
> with subsyllabic open-class morphemes -- and at least one of them could
> probably manage to provide a simple transitive clause with open-class
> morphemes that has less than three syllables. So saying that such clauses
> can't really exist is probably overstated, strictly speaking.
Vowel droping to avoid hiatus is supposed to be pretty common, isn't it?

In a language which has it you might get away without subsyllabic open-classers;
imagine that the word for "cat" is _ka_, the one for "hunt" is _a_ and the one
for  "dog" is _alg_, and that we're not seeing much morphology - _ka a alg_ "cat
hunts dog" might then come out as [kalg]. I know of no language doing this to
such an extent, but should be possible, not?

                                                  Andreas