ENDUW/ENDUOMAI (correction/clarification) Middle and passive
Carl W. Conrad
cwconrad at artsci.wustl.edu
Wed Oct 31 14:26:23 EST 2001
At 5:17 PM +0100 10/31/01, Iver Larsen wrote:
>Thanks, Carl, for responding to my query. Let me comment on the new data.
>You said about ENDUW:
>> The fact is that there are NO aorist passive (-QH-) forms in the GNT for
>> this verb; I got my figures from Accordance but have found that all 6 of
>> the forms tagged as aorist passives are in fact aorist middles and not one
>> of them can be considered passive in form or meaning. Upon rechecking I
>> find the following forms given in these instances:
>> Lk 24:49 ENDUSHSQE; 1 Cor 15:53 ENDUSASQAI (2x), 1 Cor 15:54 ENDUSHTAI
>> (2x); Col 3:12 ENDUSASQE. Obviously all of these are middle and I think
>> there's no doubt that these are the right readings in each
>> instance. What a
>> re-check of the Accordance data shows is that in each of these instances
>> there is a variant that is identified as passive; I can't imagine such
>> forms very well, but I guess they'd be something like ENDUQHTE (subj. 2
>> pl.), ENDUQHNAI (2x), ENDUQHi (2x), and ENDUQHTE (imptv 2nd pl.). That's
>> all that I can imagine is meant by the tagging, although I think it's
>> utterly conceivable that such forms as these were ever
>> grammatically viable.
>> The upshot: ENDUOMAI is the standard form (27x in GNT) of this verb with
>> the sense "dress oneself," but the active also appears, 3x only (Mt 27:31
>> and Mk 15:20 ENEDUSAN AUTON TA hIMATIA; Lk 15:22 ENDUSATE AUTON) in the
>> sense "put clothes upon a person".
>I am not familiar with Accordance, but could it be the case that although
>these six verbs are middle in form, they were tagged as potentially passive
>in meaning? I realize you discarded the possibility that they could be
>passive in meaning, but I am not so sure.
Accordance is the Macintosh version of the Gramcord program, except that I
believe it is more powerful and versatile than the corresponding Windows
As you mention the possibility that these aorist middle forms may be tagged
as an INTERPRETATION of the forms as passive, it occurs to me that you may
well be right. I hadn't even considered that because I'd assumed that the
program would not acknowledge an aorist middle paradigm to be understood as
passive; in fact it tags the form of APWLETO in the phrase we were
discussing previously as a middle although the context pretty clearly
indicates it must be understood in a passive sense.
>The active ENDUW has two objects in the accusative as can be seen above in
>Mk 15:20, one is the experiencer, the person who is clothed, and the other
>object is in semantic terms the "patient", what the person is clothed in or
>Now the passive transformation as I understand it will make the agent
>implicit. The agent is still hanging in the background and could be made
>explicit again by a hUPO or similar preposition. Furthermore, one object is
>transformed to become subject and the other object is retained. This means
>that the experiencer fills the subject slot and the patient remains as the
>only possible object. The passive transformation is from "I clothed him in a
>dress" to "He was clothed in a dress (by me)."
>You mentioned Luke 24:49:
>hEWS hOU ENDUSHSQE EX hUYOUS DUNAMIN - until you are clothed in power from
>Here the subject is the experiencer you-plural, and the remaining object is
>the patient: power. The implied agent is the Father or Jesus. It is
>theoretically possibly to supply hUPO QEOU/EMOU. The EX hUYOUS is if not the
>agent then the source or realm where the power will come from.
>The semantic difference between the middle and passive of these forms is
>that in the middle the subject is agent-experiencer combined, and the
>patient is object. In the passive the subject is the experiencer, and the
>patient is object. The agent is implied with the potential of being made
>If I say a sentence with the meaning of middle like "I clothed myself with
>this dress" or "I have put on this dress" it is not possible to make
>explicit another agent, because the agent is already explicit, although
>combined with the experiencer. You cannot say "I clothed myself with this
>dress by me/him/her/someone".
>But if I make a passive statement like "I was clothed with this dress" then
>I am still the experiencer, but not the agent. You can supply "I was clothed
>with this dress by him/her/someone".
I understand what you're saying, Iver, and I am WIDE OPEN to viewing this
aorist "middle" paradigm as capable of expressing both middle and passive
sense, just as the "middle" paradigm does in all the other
tenses--including the future of verbs that don't have exclusively the
-QHSOMAI type paradigm. But that is precisely what the traditional
grammarians seem determined to deny when they insist that there is a clear
and distinct semantic differrence indicated by the middle and the passive
paradigms--whereas my claim is that the MHN/SO/TO forms in the aorist are
simply the older "middle/passive" morphological paradigms whereas the
QHN/QHS/QH forms are the newer "middle/passive" morphological paradigms
>Is this not what you have been saying, Carl, that the mark of the passive
>meaning is that it is possible to make explicit the agent? That would not be
>possible for the middle meaning, because the agent is already explicit and
>combined with the experiencer.
Yes--or more precisely, that MAI/SAI/TAI (+ MHN/SO/TO) paradigms and
QHN/QHS/QH paradigms BOTH bear the ambivalent focus upon the subject,
whether as experiencer, beneficiary, or object, and that Greek usage makes
the clear distinction of a passive ONLY--or at least normally--by making
explicit an agent or instrument (Aristotle's "efficient cause").
>For the other five instances you mentioned, one can similarly argue that God
>is the implied agent and therefore the meaning is or at least could be
I agree that is possible, and in some cases even plausible.
>I can imagine several instances of this and other verbs in the middle form
>where it would not be clear from the context whether the meaning was middle
>or passive. The only difference is whether the agent is implied and could be
>made explicit or whether the agent is joined in union with the experiencer,
>and it doesn't seem to be significant whether the tense is present,
>imperfect, aorist, future, or perfect.
That is exactly what I am thinking. Let me add another point that I may
have mentioned before. I'm saying that APEKRIQHN means exactly the same
thing as APEKRINAMHN ("I responded"), but in the GNT forms of APEKRIQHN
appear 104x while forms of APEKRINAMHN appear only 7x. Although it is
argued that "In the middle (seven instances in the aorist) the verb
connotes a solemn or legal utterance. This is in keeping with the genius of
the middle voice, for a legal defense is more than a mere response--it
involves a vested interest on the part of the speaker" (Wallace, p. 421,
citing the older BAGD--but BDAG doesn't note any such distinction), I think
that distinction is specious; the use of APEKRINATO may be "archaic" but I
really think it's not significantly different from using the older form
AFIHMI instead of the newer form AFIW. It really does need to be taken into
account that NT Greek is NOT a stable language but one in flux (as is
probably true of any living language).
>When it comes to grammatical tagging, I am wondering whether it would not be
>simpler - and linguistically more accurate - to drop the whole question of
>deponency - thank you, Carl - keep the three tags A, M and P but make the
>distinction between M and P on contextual grounds only, not morphology, for
>all tenses. In that case, one would find a number of instances where a form
>is unquestionably middle in meaning, others which are unquestionably passive
>in meaning, but many which would need double tagging as either middle or
>passive. (I have suggested this to Tim Friberg, but whether he takes the
>suggestion, I don't know. There may be disadvantages that I don't see at the
>What do you think?
Well, I certainly think that the notion of deponency ought to be dropped,
precisely because it entails too much baggage--too many misconceptions that
can be proven to be misconceptions. I might add that I have been in
discussion with Tim Friberg about this but don't know what will come of it.
It would be SOME progress if we could use A(ctive), M(iddle) and P(assive)
ONLY to refer to the paradigms--if there were some awareness that these
names don't necessarily mean that an "active" paradigm necesarily means
that a verb in it is really active. It would be better, in my view, to go
all the way and draw the distinction between "basic" paradigm (W/EIS/EI
etc., with a parenthetical note,"formerly termed 'active'") and
"subject-focused" paradigms (both MAI/SAI/TAI etc. and QHN/QHS/QH, with a
parenthetical note, "formerly termed "middle-passive" or "passive." It
seems to me that IF we can teach students even now that that "aorist
active" comes in three varieties represented by EPAIDEUSA, ELABON, and
EGNWN, we ought to be able to teach them some day that "subject-focus"
comes in two varieties represented by APOKRINOMAI/APEKRINAMHN and
APEKRIQHN--and that for almost all of the verbs in the GNT we'll only see
one or the other of those alternative varieties in the aorist.
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