RE: Classify deponents (PEEVISH)

From: Carl W. Conrad (
Date: Wed Jun 30 1999 - 08:34:09 EDT

At 8:39 AM +0200 6/30/99, Wieland Willker wrote:
>I wanted to restrict this question to the usage in a computer program. Which
>tagging would you find most useful?
>I can't believe that you, Carl would think that a simple labeling of ALL
>middle AND passive forms to a general "middle/passive" class is the optimum.
>I agree that an extra deponent class is problematic and so do feel the folks
>at Bibleworks.
>But they try to assign some deponents according to meaning as passive, with
>all others remaining in the middle, though they are active in meaning. I
>think this is problematic, too.

Yes, I know they do, and I would agree that in not a few instances, perhaps
even MOST instances, that's easily done and there would be a pretty clear
consensus among grammarians as to which forms should be designated middle
and which forms should be designated as passive. Nevertheless there are
numerous instances of so-called "deponents" about which grammarians will
agree only about the morphology of a verb form such as HGERQH or PEPAUTAI
and then squabble interminably over whether HGERQH means "rose" or "was
raised," whether PEPAUTAI (as in 1 Peter 4:1) means "has been made to cease
from sin" or "has reached a state of cessation from sin."

My own thinking about voice in ancient Greek verbs has in fact been
radicalized. I have long believed that the term "deponent" is
misleading/obfuscating because it suggests that the forms of the Greek verb
do not fall into the neat voice-categories that one imagines (probably from
one's native language or from what one imagines to be the categories of the
target language) and are therefore somehow irregular. I rather think the
truth of the matter is that the categories of Greek voice have not been
adequately described or understood in the first place. Because POREUOMAI
doesn't have an active form we think it is somehow irregular; of course it
really is an intransitive verb, so active and passive are not meaningful
terms to describe it. What about AISQANOMAI: this is a transitive verb: it
does take an object, but it has no active forms: does that mean that it is
defective? My sense is that our whole notion of voice in Greek verbs is
laid on a false foundation insofar as it takes for granted that "active
voice" is a transitive verb category and is the normal form in which every
verb should appear at base.

My own thinking is that our terminology to describe the phenomena of Greek
voice are skewed from the outset. I think we should drop the whole question
of transitivity and intransitivity as bearing any essential relationship to
the morphology of the Greek voices; I think we should understand the
"active" morphology as the default voice form originally denoted by the PIE
endings -M,-S,-T,-ME,-TE,-NT and that we should understand the "middle"
morphology as a "marked" voice form indicating "subject intensity" in the
verb--whether this means that the subject acts upon itself or that the
subject is the recipient of action initiated outside the subject--which is
to say, this "subject-intensive" morphology may, in the right
circumstances, be understood to function as transitive passive verb. But I
think that the MEANING inherent in the "middle" morphology is not strictly
middle or passive but simply "subject-intensive."

The other new departure in my thinking--since I first described my
perspectives in a document entitled "Observations on Ancient Greek Voice
(LONG!)" posted to B-Greek on May 27, 1997--is that the -QHN,-QHS,-QH
aorist and -QHSOMAI,-QHSHi,-QHSETAI future morphologies ought to be
understood as the standard forms in later ancient Greek for
"subject-intensive" meanings in these tenses--and that it is just as wrong
to delimit these forms fundamentally to distinctly transitive and passive
function as it is to distribute the -MAI,-SAI,-TAI (-MHN,-SO,-TO) forms in
the other tenses to distinctly transitive middle or distinctly transitive
passive functions.

I realize, of course, that this is anything but helpful in terms of a
software program that one expects to tag all the verb forms in the GNT (or
any other ancient Greek documentary text, for that matter) as active,
middle, or passive. My point is that in too many instances that kind of
tagging is arbitrary or questionable; it may SEEM helpful to the student
who wants a quick authoritative answer to the parsing question, but
ultimately it's the same sort of crutch that a "trot" or interlinear
translation is: one bows before someone else's authoritative opinion and
thinks no longer about the question WHY a subject-intensive
("middle/passive" or "middle/reflexive") morphology is being employed here
rather than a default ("active") morphology.

I find myself with respect to these questions torn between pragmatic
pedagogical concerns for presenting the morphology of the Greek voice
system in its relationship to meaning in a simple, perspicuous pattern that
is easy to learn and understand and concerns for understanding what the
real functions of this morphology are/were in ancient Greek. I honestly
believe that our whole approach to ancient Greek voice morphology is
confounded by our endeavor to hammer square pegs into round holes, to adapt
the ancient Greek voice morphology to our understanding of our modern
languages in terms of transitivity of verbs and straightforward
differentiation of all verb forms as either active or passive in form and
meaning. I honestly wish that we could come to an understanding of how the
ancient Greek voice system really functions in its own terms and that we
could develop a new and more appropriate descriptive terminology for
morphologically differentiated forms. IF we could do that, I 'fantasize'
that we might even ultimately make the ancient Greek voice system
intelligible and teachable.

But this is why "I, Carl, would think that a simple labeling of ALL
middle AND passive forms to a general "middle/passive" class is the
optimum." As I said in my earlier post, a student must learn the principal
parts of all verbs that are in any way irregular, must know that a present
tense AKOUW has a future AKOUSOMAI which is NOT passive, must know that a
present tense DUNAMAI has an aorist HDUNHQHN which is NOT passive. So how
should AKOUSOMAI and HDUNHQHN be intelligently tagged? My guess is that we
will go on calling AKOUSOMAI a "deponent" future and HDUNHQHN a "deponent"
passive, go on saying that both of them are "active" in meaning. We shall
muddle on in our present way of doing Greek and manage fairly well, I
guess. Nevertheless a verb like EGEIRW is problematic: apparently it really
is a transitive verb, since it can be used in the "active" voice with a
direct object in the sense of "awaken" or even "raise from the dead."
EGEIROMAI is more problematic: is it really transitive and
reflexive--meaning "I wake myself" or is it intransitive--meaning "I
awaken"? And what of HGERQH? Shall we want to say that in some instances
this is really passive ("he was awakened/raised from the dead") while in
other instances it is really intransitive ("he awoke/rose from bed/rose
from the dead")? Is a clear differentiation of these forms in a software
parsing program going to contribute to a student's understanding of how
these forms actually work?

Carl W. Conrad
Department of Classics, Washington University
Summer: 1647 Grindstaff Road/Burnsville, NC 28714/(828) 675-4243

B-Greek home page:
You are currently subscribed to b-greek as: []
To unsubscribe, forward this message to
To subscribe, send a message to

This archive was generated by hypermail 2.1.4 : Sat Apr 20 2002 - 15:40:31 EDT