R: part of speech correlation

[Paolo Ramat]

Dear Matt,
by no means you are dead-wrong! On the contrary: I think that your
hypothesis to elaborate a (partial) typology on the basis of noun-prominent
(see the denominal verbs in Latin and Engl.: <to bottle>, <to ship> etc.)
vs. verb-prominent lgs. (see compounds of the <andro'-phonos>, <homi-cida>-
type (?) and the North American lgs.: e.g. Cayuga <ka-ya'-'ta:> 'Bild',
related to <ka-te,hta:> 'es ist ausgebreitet': Sasse,'STUF'46/1993:210)
deserves to be seriously tested. A preliminary point is in my opinion that
NOUN, VERB and other 'parts-of-speech' are not water-proof boxes (see
Ramat,"Linguistics" next issue), so that it is perhaps safer to speak of
'tendentially verb-prominent lgs.' or 'tendentially noun-prominent lgs'.
In other words I believe that we have to do with categories having
prototypical and less prototypical representatives, in different linguistic
types, so that the first step of your typological approach should define the
concepts of NOUN and VERB, both on the formal and functional plane.
Best wishes for your research!         Paolo
-----Messaggio originale-----
Da: Masayoshi SHIBATANI <matt at KOBE-U.AC.JP>
A: LINGTYP at LINGUIST.LDC.UPENN.EDU <LINGTYP at LINGUIST.LDC.UPENN.EDU>
Data: giovedì 20 maggio 1999 1.50
Oggetto: part of speech correlation


>Hi Everybody!
>
>   My student and I are looking for possible correlations between parts
>of speech system and grammatical constructions and syntactic phenomena.
>
>   There are languages, like Chinese and Japanese and a lot others,
>which group verbs and adjectives together, isolating nouns from this
>grouping. On the other hand, Latin and English and others, group nouns
>and adjectives together, isolating verbs from this grouping.
>   Our questions is if there are any grammatical correlations of this.
>We are entertaining a hypothesis that we may be able to identify
>languages being either noun-prominent or verb-prominent. The former
>'elaborates' noun phrases, or their referential function, such that   we
>may expect them to have a distinct topic marker/construction,   double
>subject constructions, where a NP is predicated over by a clausal
>predicate, valence-increasing passives, and perhaps tendency toward
>dependent-marking, all of which, we expect, would be less favored among
>verb-prominent languages, which, we assume, emphasize the predicational
>function of the verb.
>   Are we dead-wrong?  Help!
>
>Matt Shibatani
>