[Q] adjectival agreement

[Paul Hopper]

I was wondering, considering Eddy Ruys's query, about well-studied
languages like French and German and their implications for the typology he
seeks to establish.

1. Standard German shows both number and gender agreement in attributive
but not in predicative adjectives. However, gender agreement is very
restricted. It occurs only if the adjective is (a) nominative/accusative,
(b) singular, and (c) not preceded by a determiner of the "der" class. If
an adjective is in an oblique case, or is plural, or is preceded by a
"der"-class determiner, the adjective does not agree in gender (der gute
Mann/die gute Frau/das gute Kind/die guten Maenner/die guten Frauen/die
guten Kinder).

2. French adjectives generally show agreement in gender but not in number
(le grand ba^timent/les grand' ba^timents; la grande maison/les grande'
maisons/les maisons sont grande'). However, there are exceptions, for
example if the adjective belongs to the small class of prenominal
adjectives and there is liaison between the adjective and the following
noun, a plural suffix will be heard (les beau-x arts, les haute-s e'tudes);
and adjectives that have an audible plural form (e.g., phe'nome'naux) will
be heard to "agree" with their noun in number. There may be other
exceptions I haven't thought of. (Obviously, all this refers only to spoken
French. Number agreement is alive and well in the written language.)

In both languages one _could_ say that from a typological perspective joint
gender-number agreement is found only in a minority of cases, and that the
norm is for there to be only number (German) or only gender (French)
agreement. BUT there could very easily be disagreement among linguists over
how to interpret the facts, and therefore whether to place French and
German in the plus or the minus column for one of the types of adjective
agreement. And if there is disagreement over the interpretation of very
well known languages like French and German, might there not be even more
uncertainty about languages that have not been studied quite so
exhaustively? Replicability, again, in a way.

- Paul Hopper


> On Thu, 2 Sep 2004, Ruys, Eddy wrote:
>
>> Dear colleagues,
>>
>> Would anyone know of languages (or a source which might list
>> them) with one of the following properties:
>> - adjectives agree for number but not gender. If yes, do they do so when
>>   attributive, predicative, or both?
>> - adjectives agree for gender but not number. If yes, do they do so when
>>   attributive, predicative, or both?
>> - adjectives agree for person (while being morphologically distinct from
>>   verbs). If yes, do they do so when attributive, predicative, or both?
>>
>> Thanks very much in advance,
>>
>> Eddy Ruys

---------------------------
Paul Hopper
Paul Mellon Distinguished Professor of the Humanities
Department of English
College of Humanities and Social Sciences
Carnegie Mellon University
Pittsburgh, PA 15213, USA
Telephone (412) 268-7174
Fax (412) 268-7989