Re: Voice, Transitivity, and Ergativity

Carl W. Conrad (
Wed, 8 May 1996 10:44:19 -0600

Sorry that I didn't send this to B-Greek--and also to the Hellenistic
Grammar list, where I should perhaps have addressed my concerns in the
first place, but in fact, they arose in response to Shaughn's particular
question over the voice of APOLLUMENOIS and SWZOMENOIS. I very much respect
Stephen's views on this subject, the more so in that he alone really
responded to this question, while the only other real response to the
issues I was raising came in an off-list exchange with another subscriber.
But if any of you who teach and/or think Greek grammar SHOULD happen to
care about this, I'd like to hear/read from them!

>Date: Wed, 8 May 1996 08:37:41 -0600
>To: Stephen C Carlson <>
>From: "Carl W. Conrad" <>
>Subject: Re: Voice, Transitivity, and Ergativity
>At 1:15 PM -0600 5/6/96, Stephen C Carlson wrote:
>But I didn't see it until late last night (5/7)--too late to
>respond--after I finally downloaded the digest from Monday and was able to
>read the B-Greek mail that I missed after being bumped off the list.
>> Carl W. Conrad wrote:
>> >(2) The Inadequacy of our Descriptive Terms for Voice
>> >
>> >I may be a voice crying against a majority or a voice crying in the
>> >wilderness (but I am not the herald of any savior from the quandary I
>> >perceive in this matter!) but I've come to the conclusion that there is
>> >something dreadfully inadequate about the way we identify verbs as
>> >"transitive" and "intransitive" or "absolute" and "active," "middle," and
>> >"passive." The absurdity of the terms "deponent," "quasi-deponent," "middle
>> >deponent," "passive deponent" and "quasi-passive" OUGHT to indicate to us
>> >that something is wrong with our categories and our descriptive
>> >terminology.
>> Maybe this is the reason why I favor the five-case system over the
>> eight-case system for Greek,
>Personally I don't; I prefer to use terminology that links morphological
>with syntactical categories: "pertinentive genitive, (OF)" "partitive
>genitive, (SOME OF/ANY OF)" "ablatival genitive (FROM)," "true dative
>(TO/FOR)," "instrumental-sociative dative (WITH)," "locative dative
>(IN/AT)." Does anybody actually teach eight cases WITHOUT linking the
>cases to morphology?
>> but I think it is very important to
>> reserve some terms of description that are completely dependent
>> on morphology and syntax, which the student can see immediately
>> in the text, rather than to incorporate its semantics.
>Well, I half agree and half disagree with this. I agree that it is
>important to link syntax to morphology as closely as is possible, but I
>also think it is important to be clear and to let students be clear about
>syntax and often meaning as well may be independent of morphology.
>One needs to know that BOULOMAI is equivalent to "I want," "I wish," and
>that its aorist is EBOULHQHN: one needs to learn the principal parts in
>order to be able to know the right conjugational pattern for it in the
>right tense. And I'm not satisfied with the terms "deponent" and "passive
>deponent" primarily because they seem to imply that the ENGLISH voice is
>the right one while the Greek is somehow screwed up in this point.
>I think the simpler truth of BOULOMAI/EBOULHQHN m-a-y b-e (I am
>theorizing, nothing more) that there were three types of aorist formation,
>the oldest being athematic with endings added to a stem vowel, another
>very old one being thematic with secondary endings, and the newer sigmatic
>with alpha secondary endings attached to a tense-sign -S-. As we know this
>newer type gradually came to supplant the thematic aorist so that alpha
>endings were attached to the root-stems used previously with thematic
>vowels--and we know that in the later history of the language these alpha
>endings came to dominate all past tense active indicative inflections,
>including the imperfect. But something of a like nature happened with the
>athematic aorist type when new forms began to emerge using a -QH/E- marker
>between verb root and secondary endings. Eventually these QHN/QHS/QH, etc.
>aorists came to be assigned a PASSIVE sense, but they did not have this at
>the outset, I think, and the marker was probably used to form aorists for
>some verbs like BOULOMAI that never had a passive sense. These have been
>CALLED "passive deponents," but however convenient it may be to have a
>category of "passive deponents" in which to pigeonhole them, the term
>"passive deponents" might better be replaced by something like
>> Thus, I would say that "transitive" and "intransitive" are syntactic
>> statements. If there is a direct object it is transitive, etc. This
>> is not perfect --, sometimes for rhetorical reasons the object is omitted.
>I agree that this may possibly be useful but that it is not perfect. One
>of the reasons I think it is very imperfect is that the term "object" is
>not very well defined. That's another big problem, but probably one that
>linguists have already attacked.
>> Also, I consider "voice" to a morphological concept. Verbs, then, are
>> active, middle, passive, or medio-passive.
>> The normal defintion of "deponent" is (middle/)passive in form but
>> active in meaning. This incorporates too much semantic knowledge.
>> I prefer to redefine it as verbs which lack the active voice
>> morphology.
>These are precisely the matters that seem to me to call for some creative
>reformulation of categories. As I see it, the problem with
>"Active,Middle,Passive" categories is that they postulate the "active" and
>"passive" as a natural opposition and have the devil of a time explaining
>the "middle" as something sort of "in between." But in fact, what we call
>"active" and "middle" are apparently the original Indo-European voices,
>while the "passive" is a relatively late, secondary development. So I'm
>wondering whether we could not perhaps come to a less-confused and
>less-confusing account of the morphology of voice and how that relates to
>the syntax of voice and, on the basis of that accounting, formulate new
>terminology that would simplify both the understanding and the teaching of
>the ancient Greek verbal voice.
>Am I just a nut too close to retirement to be thinking practically about
>the pedagogical problems involved? Or would this sort of reclassification
>perhaps better serve our understanding and our teaching?
>> > I was told that a new way of looking at these
>> >issues in terms of a category termed "ergative" and "non-ergative" was to
>> >be found in a book which I believe Rod Decker referred to in response to a
>> >different enquiry on B-Greek, a work by B. Comrie on proto-Indo-European
>> >linguistics. I want to look at that, now that the semester is over and
>> >there's some time free at last from the press of semester-ending rush.
>> In ergative languages, the direct object of a transitive verb has the
>> same case marker as the subject of an intransitive verb. This case,
>> termed the ergative, has a case marker distinct from that of the
>> absolutive case for subjects of transitive verbs. Basque is the only
>> ergative language in Europe.
>Thanks, Stephen. And now that I know what the source is (P. Baldi on I-E
>in B. Comrie, _The World's Major Languages_), I'm off to the library to
>check it out.

Carl W. Conrad
Department of Classics, Washington University
One Brookings Drive, St. Louis, MO, USA 63130
(314) 935-4018 OR