homophones (was Re: "Tom Swifties" (Was Re: Prescriptivism))

[Laurence Horn]

At 2:41 PM -0800 12/4/05, Jonathan Lighter wrote:
>Gorge here on "Tom Swifties" :
>
>   www.fun-with-words.com/tom_swifties_history.html
>
>   JL

which, among other interesting items, happens to mention in passing:

There is a special kind of homonym called a homophone. Homophones are
homonyms which are spelled differently.

There seems to be no consensus on what exactly counts as a homonym
and what counts as a homophone.  I think we've discussed this in the
past, but non-conclusively.  I can imagine the use of "homophone"
implicating a difference in spelling on the assumption that "homonym"
is the unmarked term for sound-alike unrelated lexical items that may
or may not differ in spelling, but here it seems to be claimed that
homophones must differ in spelling.  I think a lot of us *would* take
the verb "bear" ('endure' or 'carry') and the noun "bear" ('ursus')
to be homophones (as well as homonyms), as well as either "bear" and
the adjective or verb "bare".  But now I'm not sure what the range of
standard usage is, especially since the dictionaries I've consulted
have different takes on whether "homonym" itself implies orthographic
identity, as well as whether "homophone" forbids it.  AHD4, for
example, has

HOMONYM: one of two or more words that have the same sound and often
the same spelling but differ in meaning, such as ["bank" vs. "bank"]

HOMOPHONE: one of two or more words, such as "night" and "knight",
that are pronounced the same but differ in meaning, origin, and
sometimes spelling

But that can't be right--do we really want to say that "metal" and
"mettle" are NOT homophones, given that they have the same origin?
And what about "bear" and "bear" above, or "bank" and
"bank"--homonyms but not homophones?

Anyone have a strong intuition or religious belief about this,
whether or not supported by any evidence?

larry