adjectives and word classes

[Kazuto Matsumura]

There seem to be four morphosyntactic classes in Japanese color terms:

     bare form    attributive        adverbial
I)   siro         siro-i             siro-ku       'white'
II)  midori       midori-no          midori-ni     'green'
     hai+iro      hai+iro-no         hai+iro-ni    'gray'
III) ki+iro       ki+iro-i           ki+iro-ku     'yellow'
                  ki+iro-no          ki+iro-ni
IV)  mas+siro     mas+siro-na        mas+siro-ni   'snow-white'


I) At least three more words belong to this class: kuro-i 'black', aka-i 'red', 
   ao-i 'blue'. They are all morphemically simple in their bare form.

II) Most morphemically complex color terms ending in +iro 'color' belong to
   this class.

III) This is a hybrid class. Besides ki+iro, at least one word belongs to 
   this class: cha-iro 'brown' (literally: tea-color) [Though Japanese tea
   is typically greenish yellow, you know Chinese oolong tea is brown :-)]

IV) Prefixed color terms derived form class I words belong to this class: 
   mak+kuro 'coal-black', mak+ka 'deep red', mas+sao 'deep blue'

Though some speakers may reject the forms "ki+iro-i", etc. as "unacceptable",
i-adjectival forms are amply attested  for both "ki+iro" and "cha+iro" in
my 24 mil. word corpus of literary texts. Someone may want to check whether
this is also true in other genres too.

Kazuto Matsumura
University of Tokyo


Martin Haspelmath <haspelmath at EVA.MPG.DE> wrote:
(2006/04/04 20:00)

>The same is true in German, though to a lesser extent: The colour 
>adjectives /rosa/ 'pink' and /lila/ 'purple/violet' do not inflect, at 
>least in the standard language (colloquially, they tend to exhibit 
>ordinary adjectival inflection, following an additional consonant /-n-/: 
>/rosanes/ etc.).
>
>These cases illustrate a larger point: It is very difficult or even 
>impossible to make a principled distinction between "parts of speech" 
>(the subject of David Gil's query) and "inflection classes". Most 
>descriptions of German would not say that /rosa/ and /lila/ belong to a 
>different part of speech than /schwarz/ 'black' and /rot/ 'red'; 
>instead, they would say that they belong to a special inflection class 
>of adjectives. But according to Dixon 2004 (the paper in the volume 
>"Adjective classes"), any morphosyntactic distinction can be regarded as 
>sufficient to set up a separate part of speech. I'm sure there are 
>languages that are described as having a verb/adjective distinction, but 
>where the only difference is that verbs inflect but adjectives don't.
>
>So it isn't clear to me that David's query is different from the 
>question whether there are further languages where different colour 
>words have different morphosyntactic properties.
>
>Martin

>-- 
>Martin Haspelmath (haspelmath at eva.mpg.de)
>Max-Planck-Institut fuer evolutionaere Anthropologie, Deutscher Platz 6	
>D-04103 Leipzig      
>Tel. (MPI) +49-341-3550 307, (priv.) +49-341-980 1616

--
Kazuto Matsumura
kmatsum(a)tooyoo.l.u-tokyo.ac.jp
http://www.kmatsum.info/introd/