I would say that there is a "deficiency" in Greek deponents, and that
the deficiency is that they lack the active voice morphology. The only
real issue, then, becomes: how do we can we distinguish whether a Greek
is truly lacking in the active voice or whether our evidence of Greek
is lacking at that point?
I suppose that the rule of thumb of "middle in form; active in meaning"
is a very rough attempt to make that distinction. What about a verb
like hEPOMAI "follow"? It is never found in the active, cognate to
the Latin deponent SEQUOR. However, its active from is seen prefixed
in ENNEPE "tell!" (= L. "inquit"), and cognates to the active voice
forms are found in English "see" and "say", all going back to the
Indo-European root *sekw- "to say/notice/see" The verbs hEPOMAI and
SEQUOR are understood to be a specialized sense of the IE middle
"keep in sight" => "follow" (specialized even in IE times).
I'm perfectly happy with term "deponent" (I reserve "defective" for
verbs lacking tense morphology, not voices). However, I'm not happy
with the idea that merely because a verb has a middle voice morphology,
it must have a "middle" meaning, and I suppose that is the attitude
that Carl and others are properly guarding against. Meaning is to be
inferred from usage and context. It is not to be deduced from
etymology and grammatical form if the evidence from usage and context
are to the contrary. In Greek, it is not infrequent that the middle
forms acquire a specialized meaning (e.g. LUOMAI "I ransom" from LUW
"I loose"), and deponent verbs are no exception, especially when the
original verb no longer exists.
As for the negative connotations of a term like "deponent" or
"defective," they trouble me no more than the terms "irrational,"
"fraction," "imaginary," or "transcendent" trouble me in mathematics.
-- Stephen C. Carlson : Poetry speaks of aspirations, firstname.lastname@example.org : and songs chant the words. http://www.mindspring.com/~scarlson/ : -- Shujing 2.35