Re: OSV Italian Particles

Tim Smith wrote:

>>"VO languages have prepositions;   OV languages have postpositions.
>>VO languages have wh movement;   OV languages do not.
>>In VO languages, the order is N-Adj;   in OV languages, ADJ-N.
>>In VO languages, the order is head-noun relative clause;   in OV languages,
>>rel-N.
>>In VO languages, the AUX precedes the verb;  in OV languages it is V-AUX."
>>
>>Jim here, again, with this commentary:
>>
>>Deviation from these tendencies is NOT necessarily unnatural.  English, for
>>example, is VO, but has ADJ-N.    Generally, the more of the above
>>generalizations a language violates, the stranger (typologically) it looks.
>>However, a conlang should not be dismissed as a picture of an impossible
>>language merely because its grammar doesn't conform to tendencies listed
>>above.
>
>Some deviations from these patterns are more common than others.  For
>example, the order of noun and attributive adjective correlates only very
>weakly with other word-order characteristics.  Thus there are lots of
>languages that are basically VO but Adj-N (like English) or basically OV
>but N-Adj (like Basque).
I wouldn't say that there are "lots of languages" which are N-Adj and OV.
Basque is the only one I can think of.  You're right, though, that
Adj-N + VO order is more common that it 'should' be, according to
Greenberg's typology.

Of course, when it comes to noun modification, there's the added
complication that in some languages (most) adjectives are basically
verb-like, and adjectival modification shares properties with
(reduced) relative clauses, while in other languages (most) adjectives
are basically noun-like, and adjectival modification shares properties
with compounding.  I wonder if the distribution of Adj-N vs. N-Adj
order would be more consistent if that fact were taken into account...

Matt.