query: associative plurals via noun-verb disagreement

[Jan Rijkhoff]

As to the term 'noun-verb disagreement', it may be useful to point out that the term is typically used in a formal way (involving overt manifestations of true or apparent number markers). It seems that -- from a semantic point of view -- the term 'number disagreement' (or 'number discord') is not always used appropriately. 
I have argued that languages with what I have called set nouns (as opposed to e.g. singular object nouns, which are typically attested in European languages), there is in fact no number disagreement. In these languages, the NP is headed by a transnumeral noun ('set noun'), which designates a property of a SET of entities (a set can have any number of members, incl. '1'). In such cases the agreement marker on the verb refers to the whole set (i.e. not to the number of individual members in the
set), and since there is only one set, the agreement marker indicates singular number, as in (notice that some diacritics are missing):

Georgian (Harris 1981: 22)
(1)       sami    knuti     goravs
           three     kitten    roll:3SG
		'three kittens are rolling'

Lango (Noonan 1992: 168)
(2)	gulu  adek     otoo
		pot    three    3SG:die:PERF
		'three pots broke'

Oromo (Stroomer 1987: 107)
(3)	gaala   lamaani   sookoo   d'ak'-e
		camel   two          market   go-3SG.M.PAST
		'Two camels went to the market'

In each example, we see that a plural subject triggers singular agreement with the person/number marker on the verb (so not: a singular NP triggering plural number agreement on the verb - as in David's query). In my sample (50 lgs.), this phenomenon is only attested in languages with 'set nouns'. Properties of set nouns and other nominal subcategories (designating different 'Seinsarten') are discussed in more detail in my book on the noun phrase (OUP 2002/2004, pages 104-117; see also my article
'On flexible and rigid nouns', Studies on Language (2008) 32-3).

So the point is: if reference is made to a set entity, it does not seem correct to talk about 'number'** disagreement from a semantic point of view, since the agreement marker on the predicate may refer to just the set (singular) or to the size of the set (singular agreement in the case of a singleton set, otherwise plural agreement). 

** I write 'number' (i.e. with single quote marks), because I believe that the so-called number marker on a set noun does not specify a number value (SG/PL), but a kind distinction (i.e. we are actually dealing with a 'nominal aspect marker'): the speaker indicates what KIND of set is being referred to (e.g. a singleton set or a collective set).

Best - Jan Rijkhoff

Balthasar Bickel <autotype at UNI-LEIPZIG.DE> writes:
>David -- since you also asked for negative data: Belhare (Kiranti, Sinotibetan) does not allow ASPD. I tested this when exploring other  
>effects of 'disagreement'. (Disagreement in number, with a singular NP and a nonsingular agreement marker, results in meanings like 'one of them laughed'. I have a paper on this in Studies in Language, 2000).
>	-- Balthasar.
>
>On Nov 13  2008, at 15:53, David Gil wrote:
>> Dear all,
>>
>> I am interested in the cross-linguistic distribution of a  
>> construction type in which an associative plural meaning, eg. 'John  
>> and his associates', results from a singular noun triggering plural  
>> number agreement on the verb,
 .....
>
>>
>> Thanks and best wishes,
>>
>> David


Jan Rijkhoff
Dept. of Linguistics, University of Aarhus
Ndr. Ringgade 1, Building 1410, Office 338
DK-8000  Aarhus C, DENMARK
Map: http://www.au.dk/en/hum/lingvist/map.htm
Phone: (+45) 8942 6550   *   Fax (+45) 8942 6570
E-mail: linjr at hum.au.dk
Home page: http://person.au.dk/en/linjr@hum


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20081114/3adc3a0a/attachment.htm>