LINGUIST List 5.399

Wed 06 Apr 1994

Sum: Generics

Editor for this issue: <>

Directory

1. Michael Kac, Summary on Generics

Back in February I posted the following query:

>Are there languages in which there is some kind of overt
>morphological or lexical marking to indicate genericity -- i.e. in
>which e.g. *the lion* qua generic is formally distinguished from
>*the lion* qua singular definite description?

I received a number of interesting replies, summarized below
with apologies for the delay in doing so. The answers support the
contention of Oesten Dahl (1988) -- who drew my attention to his
paper -- that there's no language with a strictly generic article.
Herewith, then, the promised summary:

Julie Auger -- who also drew my attention to Dahl's paper --
argues in a paper of her own (see References, below)that in
colloquial French the distal demonstrative pronoun *c,a* has
developed into a generic marker as used in sentences like

 (1) Les hommes, c,a parlent tout le temps.
 the men that talks all the time
 'Men talk all the time.'

 (2) Les hommes, ils parlent tout le temps.
 they
 (ambiguous as bet. generic and nongeneric interp.)

In her analysis, *c,a* is associated syntactically with the verb but
forces a generic construal on the subject.

Greg Carlson notes that Manfred Krifka, in his introduction to a
book in press, reports on a number of German dialects (including
Bavarian) in which there are two forms of the definite article, one
but not the other of which is used generically. The primary
function of the 'long' form is to refer anaphorically while the
'short' form is used to refer to something already in the
background and with proper names. However, the short form (but
not the long) is used for generic reference. Exx.:

 (3) Da Schnapps is daia.
 Schnapps is expensive. (generic)

 (4) I hab a Bia un an Schnapps bschdait. Dea/*Da Schnapps
 war daia.
 'I have ordered a beer and a Schnapps. The Schnapps was
 expensive.'

A similar pattern also exists in Frisian. This is the closest anybody
came to identifying a specifically generic form of the article,
though Dahl's contention is still supported since the short form
isn't used *only* to mark genericity.

Carlson (seconded by Amy Uhrbach) also cites work by
Portersfield and Srivastav, who report that the definite article in
Indonesian is omitted in a variety of circumstances, including
generic reference. Bill Croft adds that there are some Micronesian
languages -- e.g. Mokilese (Sheldon 1976) -- that work this way.
(So, according to Ingo Plag, do many creole languages. See
Bickerton 1981.) Krifka also notes that some languages (e.g.,
Swahili) have morphological markings on verbs when the intent is
to make reference generic properties or activities (such as
habitual activity) rather than to specific events.

Kjetl Hauge describes a pattern which is exhibited by certain
varieties of Norwegian in certain styles. Compare

 (5) a. den unge mann-en
 DEM PRONOUN young man-ART (postposed)

 b. den unge mann

(5a) is anphoric, (5b) generic. Further details in Lundeby 1965. (A
further note: Danish uses only the former type of construction,
Swedish only the latter.)

Michael Newman points out that in that most exotic of all
languages, English, the common-sex *they* can be used to at least
strongly suggest genericity in some contexts; thus, compare

 (4) You take a guy like Morrow that's always snapping his/their
 towel at other people's asses ...

where the case with *his* favors a reading on which you're talking
about Morrow himself and the one with *they* favors one on
which you're talking about the class of guys like Morrow.

Norbert Strade has contributed some examples in Finnish, where
genericity can be marked through the case of the adjective:
Finnish can make the distinction through the case of the adjective
in predicative constructions:

 (6) kahvi on hyva"
 'coffee is good' (noun in nom.sg. + copula + adjective in
 nom.sg.)

 (7) kahvi on hyva"a"
 the coffee is good (adjective in partitive sg.).

 (8) leijona on hyva"
 the lion is good

 (8) leijona on hyva"a"
 lion [understood as lion-meat] is good

I posted the query because genericity is not something I've
thought about much, but I've been thinking about it of late for
two reasons. One is that I was at the time of the posting gearing
up for a course in which we'd be working through Ojeda's
*Linguistic Individuals*, which has what I consider a particularly
ingenious (and satisfying) treatment of definite generics in
English. The other is that I've been at work on a paper dealing
with singular terms from a somewhat different point of view and
trying to assimilate definite generics to ordinary definite
descriptions in a way something like the one Ojeda suggests.

Thanks to everyone quoted above, and also to Ariel Cohen, John
Cowan, Kiyoshi Ishikawa, Marion Kee, and Ernest McCarus for
their interest.

References

Auger, J. 1993. Syntax, semantics and *c,a*: on genericity in
 colloquial French. Penn Linguistics Review.

Barlow, M. 1992. A Situated Theory of Agreement. Garland.

Bickerton, D. 1981. Roots of Language. Ann Arbor, MI: Karoma.

Carlson, G. 1992. Project description, ms. University of Rochester.

Dahl, Oesten. 1985. Tense and Aspect Systems. Oxford: Blackwell.

------------- 1988. The expression of the episodic-generic
 distinction in tense-aspect systems. In M.
 Krifka ed., Genericity in Natural Language
 (Proc. of 1988 Tuebingen Conference, SNS-
 Bericht 88-42, Univ. of Tuebingen 95-106)

Harrison, Sheldon P. 1976. Mokilese Reference Grammar.
 Honolulu: University Press of Hawaii.

Lundeby, Einar 1965. Overbestemt substantiv i norsk og de andre
 nordiske spr}k, Universitetsforlaget.

Michael Kac