In defense of etymological speculation

[Fred Shapiro]

On Tue, 13 Aug 2002, Paul McFedries wrote:

> I've been deleting the "bulldyke = bulldog-like" thread), if a "bull"
> is a large, mature male of certain animal species, and also therefore
> means "a large or agressive man," then it's not at all surprising that
> a person with a strongly mannish appearance or demeanor (a bull) who
> is also a lesbian (a dyke) should be called a "bulldyke." (The RHHDAS
> appears to back me up here by showing the etymology as "[_bull_ +
> DYKE]".)

I like Paul's implicit argument for an Occam's Razor approach to etymology
-- why go for elaborate explanations hypothesizing exotic processes of
word-formation when simpler explanations are available?  But Paul's
explanation is actually worse than Jerry's in that Paul's one contradicts
the historical evidence I have uncovered.  "Bulldyker" is attested in
1906, "bulldyke" in 1931, "dyke" in 1931.  It's of course possible,
particularly with this very-underground type of word, that "dyke" came
first and no early usages have been found, but that's not the way the
evidence is looking.  The RHHDAS etymology is based on the known evidence
before my recent discovery, which renders that etymology doubtful.

I feel sometimes that in etymological discussions I am speaking a
different language from most everyone else, a language of evidence rather
than of speculation.

Fred Shapiro


--------------------------------------------------------------------------
Fred R. Shapiro                             Editor
Associate Librarian for Public Services     YALE DICTIONARY OF QUOTATIONS
  and Lecturer in Legal Research            Yale University Press,
Yale Law School                             forthcoming
e-mail: fred.shapiro at yale.edu               http://quotationdictionary.com
--------------------------------------------------------------------------