About that question on hyponyms

[Ben Zimmer]

At 12:33 AM 10/17/01 -0500, you wrote:
>Hello,
>
>About that question on hypernyms / hyponyms,
>
>>1) Is there a specific term for terms that serve both as terms for a genre
>>and terms for members of that same genre (e.g., "man" when used to refer
>>to both male homo sapiens and all homo sapiens)?
>
>(Sorry, I erased the original message and those that followed).
>
>If we regard man1 and man2 as different entries in the lexicon, man1 and
>man2 can be just treated as homonyms. This could explain why "All men are
>created equal" is acceptable -- though demode and ideologically
>hypocritical -- while "Half of all men can bear children and breastfeed"
>is very strange, to say the least.
>
>I believe "generalization" as the common process by which a term comes to
>cover a broader group of referents can apply in this case as well in which
>the broader group is the entire class. The reverse is "specialization". So
>in man1 / man2 historically we've got one of those (I don't know which; in
>Romance languages I would say the process was specialization, since Latin
>generic "homo" covered both "vir" and "femina"). But, if one doesn't want
>to use the expression "generic term" or "general" term for this hypernym,
>then syncronically we could talk about hypernym/hyponym homonyms, or
>something funny like that. An HHH, triple H.

This is more easily expressed in terms of markedness theory (see Jakobson,
Kurylowicz, Silverstein).  "Man" as an unmarked term follows a typical
pattern: it can refer either to an entire generic category or else to a
subcategory that contrasts with a marked term ("woman").  According to the
OED, there are cognates in all Germanic languages that are similarly
unmarked (i.e., the cognates all can denote either ‘human being’ or ‘adult
male human being’).  The two senses coexisted in Old English, though the
generic sense was originally the principal one (gender distinctions were
more commonly coded by the lexical pairs "wer"/"wíf" and
"wæpman"/"wífman").  Eventually in Middle English "man" overtook "wer" in
the non-generic sense of 'male human being', with "wífman" (<"woman")
retained for 'female human being'.  But since "man" has continued to be
used in its generic sense as well, the structural asymmetry of the
markedness relationship has been read by some as inherently sexist.  This
is quite similar to the development of "he/his/him" as an unmarked personal
pronoun.

--Ben


Refs:

Jakobson, Roman. 1983. "The Concept of Mark." In: _Dialogues_, with
Krystyna Pomorska. Cambridge, Mass.: MIT Press.  Reprinted in: Jakobson,
Roman. 1990.  _On Language_, edited by L.R. Waugh and M.
Monville-Burston.  Cambridge, Mass.: Harvard University Press.

Kurylowicz, Jerzy. 1949. "Linguistique et theorie du signe." In _Readings
in Linguistics_ Vol. 2, edited by E.P. Hamp, F.W. Householder, and R.
Austerlitz. Chicago: Univ. of Chicago Press.

Silverstein, Michael. 1985. "Language and the Culture of Gender: At the
Intersection of Structure, Usage, and Ideology." In _Semiotic Mediation_,
edited by E. Mertz and R.J. Parmentier. New York: Academic Press.


___________________
Benjamin G. Zimmer
Ph.D. Candidate
Dept. of Anthropology
University of Chicago