the meaning of GENERIC in linguistics

[Laurence Horn]

At 12:44 AM -0500 3/5/05, RonButters at AOL.COM wrote:
>Larry Horn's comments (reproduced below) were for me thought-provoking , aND
>I need to think about them some more. (Thanks, Larry, for the summary of
>aspects of the Horn-Kleinedler LSA paper, which I think you were
>indeed kind enough
>to send me some time in the past.) However, I hope Larry will give me some
>clarification of just how he is using the term GENERIC in
>linguistics for words
>where trademarks are not involved, because he seems to be using the term
>GENERIC in a way unusual to either linguistics or the law.
>
>With respect to trademark law, the term "generic" relates to features of
>individual words--trademarks are as special kind of proper noun.
>This is the sense
>in which it is usually used in lexicography, to my knowlege. In linguistics,
>to my knowledge, "generic" is a syntactic property of nouns rather than an
>inherent feature of individual words. That is to say, "man" and "guy" are
>intrinsically neither generic nor nongeneric; rather, they are
>construed as "generic"
>or "nongeneric" in particular linguistic contexts. This is the definition
>that I find in all introdcutory linguistics textbooks and dictionaries of
>linguistics, which usually gok on to give such stock examples as the
>following:
>
>GENERIC: Man is an animal that nurses its young.
>GENERIC: A man should always open a door for a lady.
>NONGENERIC: A man opened a door for a lady.
>NONGENERIC: The man was an animal who tried to nurse his young.
>
>So I totally agree with Larry that the noun MAN "isn't quite the same as
>'kleenex' " with respect to genericness; indeed, lexically, MAN is always
>potentially either generic or nongeneric, and so is GUY, except
>insofar as MAN or GUY
>may be a trademark, which is perfectly possible (though I don't know of any
>myself off the top of my head). Syntactically, MAN is either generic or
>nongeneric, depending on the syntax. That is to say, there is indeed a major
>difference between words such as GUY or MAN and words such as
>KLEENEX: brand names are
>intrinsically generic proper nouns, and people give them some kind of special
>status based on that fact.

Having decided I've over-posted on this thread, I determined to go
cold turkey and don't want to open up any new cans of worms.  And I
fear I can't close this one that easily; there are a number of issues
you bring up above that I'll have to defer any response to.  But a
brief point of clarification on the issue below:

>"autohyponymous" is a term that I am not familiar with, so I have some
>trouble following Larry's argument below (how a word can be
>hyponymous to itself is
>not clear to me).

"autohyponymy" is a term I've been using since 1984 for the situation
arising in precisely the kinds of cases I brought up in connection
with that word ("Yankee", "dog", "Frau", "guys", "cow", etc.), in
which a lexical item--typically through either semantic broadening or
narrowing--has two (or more) senses, one of which includes another.
Hyponymy is the relation "collie" or "bitch" bears to "dog"
(identifiable with unilateral entailment--all collies/bitches are
dogs but not vice versa.  But if a bitch is a female dog, its
counterpart for a male dog ('a male animal of the family Canidae', in
the AHD4's terms) is "dog".  Thus "dog" [AHD4's sense #3] is a
hyponym of "dog" [AHD4's sense #1], hence an autohyponym. Similarly,
the 'Anglo/WASP New Englander' is a hyponym of the 'New Englander'
"Yankee", which is a hyponym of the 'northerner' "Yankee", which is a
hyponym of the 'U.S. inhabitant' "Yankee"--concentric layers of
autohyponymy.

This term is more transparent than "genericide", I'd argue--but I
promise I won't here.  But I'm willing to grant it's not totally
transparent, since we're really talking about one sense of a word
being a hyponym of another sense of the same word, not literally a
word or sense that's hyponymous to itself.


>  I agree, though, that speakers' use of KLEENEX to refer to
>tissues in general often times "reflects speakers' knowledge of their
>meaning" as (1) trademarks (proper nouns) and as (2) the thing that
>the trademark
>(proper noun) most frequently refers to (i.e, facial tissues)--indeed, I that
>hass been pretty much central to my argument all along! On the other hand,
>speakers' use of ZIPPER to refer to a type of fastener "reflects"
>only on the latter
>kind of knowledge.
>
>Why Larry and Steve would call MAN a "quasi-generic"--and what the difference
>is for them between "quasi-generic" and "pseudo-generic"--I'd also like to
>know about.
>
>
The relevant notion of generic in this case is not the grammatical
one (which distinguishes the bare singulars allowed with "woman" and
"man"--as in "Man is the only animal that rapes" on either the
(putative) generic-human or generic-male-adult-human sense--from the
impossible bare singulars with other count nouns barring this use
(*Dog is a mammal, *Boy tends to be taller than girl, *Guy drinks
beer).  Rather, it's the one that comes up in discussions of "he/man"
language, in which "generic" seems too ensconced to uproot in the
sense of 'sex-neutral'.   The issue of (intended or purported)
sex-neutrality overlaps perniciously with that of morphological
genericity (the bare singular illustrated above) as well as
(ordinary) semantic genericity, of the type Ron alludes to.  I will
(reluctantly) elaborate if asked, but maybe this is enough for now.

Larry