Response to Ward Powers re Voice (1)
Carl W. Conrad
cwconrad at artsci.wustl.edu
Wed Nov 7 08:48:32 EST 2001
I hope that the length of this will not deter interested parties from
reading and sharing in the discussion; I found it difficult to digest what
Ward is saying here in this format, although it is perfectly clear that he
continues to promote the "traditional" understanding of ancient Greek voice
system. I'll try to trim some of the repetitious material and focus upon
just a few points. And I think it would be wise also to partition my
response into at least two parts; perhaps I too shall require a week or
more for a full response--although, quite honestly, I think the
investigation will continue onwards for some time.
At 12:21 AM +1100 11/7/01, B. Ward Powers wrote:
>With the discussion of "voice" continuing apace on b-greek , and now under
>more than one subject heading, I note that some list members are indicating
>their inclination towards accepting the position re middle and passive verb
>forms which is being advocated by Carl.
Really? All I've observed is that Iver and I have had an illuminating
exchange over the questions involved. I have had some interesting exchanges
off-list about these questions, but I'm not aware of any significant
acceptance of my proposals.
>Against this, I would contend that the case for which Carl is arguing has
>not been made out: the evidence is just not there to support it.
I would rather say that most of the evidence has as yet still not been
explored; I'm still busily examining the data and the texts bearing on the
matter; I don't claim to have demonstrated anything very surely other than
that the concept of deponency--insofar as it is used to refer to ANOMALOUS
semantic function in traditionally-termed "MP" and "P" morphoparadigms is
bankrupt. What I have argued (and attempted to present some evidence to
support) is that the semantic value of so-called "deponent" verbs is not
inconsistent with the observed semantic values for the morphoparadigms in
which they appear--and that this is the case whether we're talking about
those traditionally-termed "middle deponents" (present in -MAI, aorist in
-MHN) or those traditionally-termed "passive deponents" (present in -MAI,
aorist in -QHN or if we're talking about the significantly large number of
verbs with "active" present-tense morphoparadigms but "middle" future
Now, to respond to some of Ward's points:
Let me state clearly, if anyone misunderstands my position, that I am in
agreement with Iver (and I think with Bryant Williams also) that we need to
maintain a clear distinction between SEMANTIC FUNCTIONS of active, middle,
and passive. The question, I think, is whether we are rightly describing
the way the Greek verb's morphoparadigms represent these semantic functions
(For sake of argument, I'll limit this to the Koine Greek verb, so long as
it's granted that the verb system is indeed in flux during the period in
which Koine Greek was spoken and written). In particular, is it the case
that the -QH- forms regularly and consistently indicate passive semantic
function? and that, when they don't appear to represent passive semantic
function, the numbers of such apparent anomalies are insignificant--or at
least, not significant enough to do anything with them other than lump them
together into a sizable category of inexplicable anomalies.
>Yes, this is a clear statement of our divergent viewpoints. I hold to a
>basic correlation (certainly, as an initial starting point for examining
>the use and meaning of verb forms) between: the ELUSA-type forms and the
>concept of active voice; the ELUSAMHN-type forms and the concept of middle
>voice; and the ELUQHN-type forms and the concept of passive voice.
> . . .
>The fact is that a three-way differentiation DOES exist morphologically in
>Greek between the flexions of the aorist forms ELUSA, ELUSAMHN, and ELUQHN,
>and correspondingly between the future flexions for LUSW, LUSOMAI, and
Granted that LUW is a convenient verb to use to draw up a complete taxonomy
of regular Greek verb forms--and that I've used the present middle-passive
in its classical Attic form to illustrate how passive semantic function
readily comes to be attached to middle morphoparadigms--I still have to
lodge a bit of a protest against employment of this taxonomy as
representative of the Koine Greek verb. That taxonomy of morphoparadigms of
LUW is in fact something of a mannikin of the Greek verb-forms actually
found in Koine Greek usage. In the database of the GNT in fact, as I've
noted before, there are no aorist middle forms of LUW. Moreover, although
the aorist in -SA and -SAMHN is surely the most common morphoparadigm in
use for the aorist "active" and "MP" in the Koine, nevertheless there are a
great number of verbs with aorists in -ON and -OMHN--and also in -HN (and a
couple in -WN and -UN), and I rather suspect that the survival of passive
semantic function associated with -MHN/SO/TO morphoparadigms would be found
among these older types of aorist than among the "weak" or "first" aorist
I did mention the interesting form of APWLETO that surely seems to be used
in a passive sense in 1 Cor 10:10. Let me throw in another. In an off-list
exchange with Kimmo Huovila I asked him whether he thought there was any
semantic difference between EGENETO and EGENHQH. I trust that Kimmo will
not object to my citing his immediate response and then his respone
following further reflection:
At 11:00 PM +0200 11/6/01, Kimmo Huovila wrote:
>Kimmo Huovila wrote:
>> > (CC) And would you claim that there's any semantic
>> > difference between the concomitant Koine forms EGENETO and EGENHQH?
>> Not necessarily. I could approach the meaning from both middle and
>> passive viewpoints. I think this is a case of overlap. Now, this does
>> not mean that an analysis finegrained enough would not find a semantic
>> difference - either in style, register, sosiolinguistics etc. but these
>> may now be well beyond our reach. It's hard for me to believe that no
>> distinction could be teased out (if we could analyze lots of native
>> speakers from different areas, ages and social groups), but generally
>> speaking, I would not claim any (significant) difference, and for the
>> insignificant difference that they presumably was, I do not know what it
>> might have been.
>Upon further reflection, I think that neither is semantically middle.
>Perhaps we should label both passive. Middle and passive semantics are
>sometimes close and it is easy to see a historical shift, but
>synchronically EGENETO is not semantically middle. It seems that even
>with the rise of the passive, it never drove the passive function
>completely out of the older middle-passive paradigm.
I was really quite surprised at this. Aware though I am that EGENETO is
regularly translated in the Vulgate with FACTUM EST, the perfect tense of
FIERI which functions as the "passive" of Latin FACIO, yet it seemed to me
that EGENETO doesn't usually carry the sense of "was made." Or does it
perhaps do so more often than we might suppose? Is there a real semantic
difference between EGENETO and EGENHQH?
What about DUNAMAI? This is tricky because its future is regularly "middle"
DUNHSOMAI and its aorist regularly passive HDUNHQHN. Does the verb mean "I
can/am able"? or does it mean "I am enabled"? I don't know how the other
morphological taggers have interpreted the GNT evidence, but I was frankly
shocked to find that Accordance shows 210 forms of DUNAMAI and lists only 4
of these as middle, all of them future "MP" forms, while it lists 206 forms
as passive, including 6 future "MP" forms and all 9 aorists in -QH-. While
I don't doubt that some of these forms marked as passive really do mean "be
enabled" yet I'm really surprised that all those presents in -MAI/SAI/TAI
are tagged as passive. I'm curious what Dale Wheeler will say about
this--and what Kimmo will say about this too, which is why I've cc'd both
At least one possible implication of this is that the semantic boundaries
between middle and passive are not so clearly drawn as we might suppose.
Must we always assume that a verb that is passive semantically is the
patient and/or experiencer of an action performed by some other?
Just a word at this point about A.T. Robertson's discussion of the future
passive from the big book, (A.T. Robertson, _A Grammar of the Greek NT in
the Light of Historical Research_. 4th ed., 1923).
>Carl cites the information given by A T Robinson, for whose work (as for
>Carl's) I have the utmost respect. Robinson says:
><<The future passives without certain passive sense are illustrated by the
>ANAKALUFQHSOMAI (Mt. 8:11),
>APOKRIQHSOMAI (Mt. 25:37),
>EPANAPAHSETAI (Lu. 10:6),
>QAUMASQHSOMAI (Rev. 17:8),
>KOIMHQHSOMAI (1 Cor. 15:51),
>ENTRAPHSONTAI (Mk 12:16),
>METAMELHSOMAI (Heb. 7:21),
>FANHSOMAI (Mt. 24:30),
>FOBHQHSOMAI (Heb. 13:6).>>
>This is a list of nine instances where (Robinson says) passive forms are
>used <<without certain passive sense>>.
>Actually, I cannot find ANAKALUFQHSOMAI in Matthew 8:11 (or anywhere else
>in the GNT for that matter) - perhaps Robinson may have been using a
>different text, though I do not find reference to such a textual variant at
>that point either. But I do find ANAKLIQHSONTAI there, in Matthew 8:11, so
>maybe it's a typo. There is another typo: Mark 12:16 should read Mark 12:6.
Yes, this was a typographical error in my part (it's not very easy copying
a printed text with lists of verb-forms and textual references): it should
have been APOKLIQHSONTAI in Mt. 8:11.
>Yet in some of these instances, I can see why a writer may have felt a
>passive form was appropriate. Especially where there is a causative idea,
>with the implication that someone or something is the cause of the event
>referred to. Thus in Matthew 8:11 those who come and "recline at table"
>with Abraham, Isaac and Jacob are "caused to recline at table" (i.e., given
>the right to recline) - it is not their own doing.
>But just suppose that in looking at these nine future verbs I have been
>blessed with an overactive imagination, and these are all, morphologically,
>real future passive forms with definitely non-passive sense. And, moreover,
>Robinson says these are "illustrative" of his point, so there could be
>others. Even so, they are a small number of instances, and do not nullify
>the general recognition that futures with -QH- are passive.
I shall neither cite nor argue these instances of -QH- futures. I expect to
be forwarding to the list later the text of Robertson's discussion of the
-QH- futures from his chapter on Voice under Syntax, pp. 818-20. What I
find most astounding in A.T. Robertson is the depth of his insight and his
command of the diachronic aspects of this question, and I am encouraged by
his calm and gentle demonstration of the perversity (my word, not his) of
traditional terminology and analytic categories for description of voice in
the ancient Greek verb.
More anon, including my own comments on the conclusions Ward sets forth at
the end of his lengthy message.
Carl W. Conrad
Department of Classics, Washington University (Emeritus)
Most months: 1647 Grindstaff Road/Burnsville, NC 28714/(828) 675-4243
cwconrad at artsci.wustl.edu OR cwconrad at ioa.com
More information about the B-Greek