Re: Voice and Morphology (was hHKW/hHKASIN)

From: Don Wilkins (
Date: Mon Oct 27 1997 - 13:08:58 EST

At 11:25 PM 10/27/97 +1100, Ward Powers wrote:
>The relationship of aspect to time did not come into it at any point.
>However, you have asked me several times to spell out where I stand in
>relation to this matter of aspect. I am not trying to dodge the issue of
>where I stand. I may say that I took no part in the recent controversy on
>b-greek re aspect. My answer now arises out of my involvement with
>questions of morphology which to me are crucial. I hold that the form of a
>word indicates (by various morphological devices) both time meaning and
>aspect meaning, and that these are two different things (indicated by
>different morphs).

Unfortunately my comments appeared to be hostile toward you (though not
intended to be so, I assure you), and I did not mean to imply that you were
dodging the issue. And in fact like you, I saw no reason to bring aspect
into your current discussion, and was just hoping that you would clarify
your purpose and position.
>Now, it is perfectly possible to use verbs in contexts which are at
>variance with what their morphology is saying. ... It is perfectly possible
to argue a case that certain
>indicative aorist forms are being used in a context where they mean the
>present, the future, or "timeless". Or to argue against such a case. All
>this lies in the realm of semantics and exegesis. I am operating (in this
>discussion) at the level of morphology.

I would agree. It turns out that I have recently become aware of the
*opposite* situation, i.e. where the "tense" (as defined by morphology)
appears to pull adverbs or other time indicators toward the verb and away
from their conventional meanings.

>"Tense" as a term originally related to time. In discussing Greek verbs I
>use it (like most others, I believe) as a convenient way of referring to
>different patterns of inflection, which can then be discussed. When used of
>such particular patterns of inflection, "tense" does not imply anything
>about my opinion of the time factor involved.

Again, I would agree, and like others I have pointed out that the terms
"tense" and "aspect" are used in ways contradictory to their meanings
throughout the verb system; we simply use the terminology as a shorthand
method of refering to verb properties.

>My research leads me to the conclusion that (and again I draw attention to
>the fact that I am discussing the morphology of koine Greek) verb roots
>possess inherent aspect (which has nothing to do with time). Verb roots of
>the First Conjugation are inherently durative, and those of the Second
>Conjugation and Third Conjugation are inherently punctiliar. The -O/E-
>joining vowel (which occurs in Slot 7) is a neutral morph meaning "no
>change in a verb's aspect". The first aorist is formed by replacing the
>neutral morph with the punctiliar morph -SA-, meaning "switch aspect to
>punctiliar", and in the perfect active it is replaced by -KA- meaning
>"switch aspect to perfect".

I think these are reasonable conclusions, though it may not be possible to
verify them.

>When the future morph sigma (epsilon after liquids) occurs in a verb form
>(it goes into Slot 6), it switches the form from durative aspect to future
>time. The standard grammars (and I do not place Porter in that category:
>his approach is a very long way removed from "standard") affirm that "the
>future is the only tense which expresses only a level of time and not an
>Aktionsart, so that completed and durative action are not distinguished".
>(This is cited from BDF p.178, #348.) I have not found any evidence to
>contradict this statement and I wholly accept it.

We may disagree here. The similarity in thought and morphology between the
future indicative and the aorist subjunctive indicates that the meaning may
be one of potentiality for a punctiliar action. My own experience indicates
that fut. ind. is inherently punctiliar.

>Morphologically, there is a one-to-one relationship between the presence of
>the augment (temporal or syllabic) in a verb form and that form being
>therefore and thereby past time. This morphological feature has got
>absolutely nothing whatsoever to do with aspect: there is a "past time"
>flexion for each of durative (imperfect), punctiliar (aorist), and
>perfective (pluperfect).

We agree here. As you may know, the recent debate on aspect involved the
notion that the augment, from the (late?) classical to modern period, was
obsolete and meaningless, at least in practice.

>[We can note that verbs beginning with a vowel and a few others can take a
>temporal or syllabic augment as a perfective morph in lieu of
>reduplication, and that the past time morph can be omitted from the
>pluperfect in the style of some authors, but these facts do not affect my

Agreed on all points.

>An aorist indicative contains the past time morph, and its form is
>therefore both past time and punctiliar aspect. All other moods (or modes,
>if we were to use A. T. Robertson's much more accurate term - see p.306 of
>his "Short Grammar") lack the past time morph and therefore are not
>morphologically displaying any time feature.

Agreed. I should perhaps add for claification that you would include (as
already indicated) the imperfect and pluperfect as having a past time morph.
Whether one recognizes the augment as being linguistically relevant or not,
its use clearly indicates that the indicative is the mood/mode of time.

>The information that is given to us morphologically enables us to identify
>accurately the FORM of a verb. It is open to a person to argue that in a
>particular usage the MEANING is at variance with the form. In a given
>instance he may well convince me. But the onus is (I would hold) on the one
>who wants to argue against what the morphology says, to show that meaning
>overrides form on the occasion in question.

Well said. Unfortunately the meaning is a subjective determination, and the
situation becomes difficult if one does not recognize the possibility of
alternative translations.

><SIGH!> Ah well. I would dearly like to think that there were others on the
>list who noted my protest against present practices of teaching middle
>forms as being passive, and were weighing up the validity of the case I was
>making out for this.
I sympathize with you. Teaching beginning Greek is like teaching alpine
skiing, and it appears that we have to take some liberties in terminology
etc. (like learning snow-plough or using shorty skis) to get somewhere.

>As I had explained that I was setting out the results of my own research in
>various areas of application of linguistic principles to Greek, and was
>summarizing material published in my grammar "Learn To Read The Greek New
>Testament", to which I sometimes gave a cross-reference, I had thought it
>would be understood that my posts to b-greek expressed my conclusions and
>opinions. But if this needs to be more expressly expressed: Would all list
>members please note that if I should post any future contributions to this
>list I am speaking only for myself and giving my own judgements in the matter.

Again, I sympathize, and I should have been much more sympathetic at the
beginning. However, I think you'll find that telling everyone up front what
your perspective is will not get the job done. We all tend to forget (at
least those who like myself are prone to human error), and I think it is
useful and necessary to reiterate when we are expressing our opinions,
especially since there are times when we really are attempting to make
axiomatic statements.

>>...To be precise: When you want to express a passive in the future or
>punctiliar, you have the forms available, containing QH (in the subjunctive
>and participle, this is in the short-vowel form QE - an allomorph of QH).
>There is a one-to-one relationship between the forms used to express the
>passive future or aorist and the presence of QE/QH in all such forms: so
>QE/QH is the passive morph.

Here again a point of clarification may be in order: there are a number of
so-called second passives in which the Q is absent and only the vowel of the
morph remains.

>>The other side of this coin is that the pronoun endings which LUSOMAI and
>ELUQHN have in all their forms, across the board, are to be recognized as
>being MIDDLE endings, which distinguish these forms from active and passive
>>But when you want to express a passive in the durative or perfective, you
>do not have separate, distinctive forms available, because the language did
>not develop them. What Greek did was: use the middle forms with passive

That is what we have assumed, at any rate. I suspect that we have yet to get
to the bottom of this. Theoretically, to the Greek mind these forms could
have a special middle meaning.

>>It is my contention that THAT is how voice forms should be taught to Greek
>students: that is, that of the four subsystems which exist in the Greek
>verb, two have passive forms and two do not. The two that do not, durative
>and perfective, use MIDDLE FORMS with PASSIVE MEANING when passive meaning
>is required.

That is reasonable. Of course no matter what we do we reach a dead end at
some point (e.g. a student asking the obvious question, "Why *didn't* Greek
develop more passive forms?").

>>Thus I register a protest against teaching (say) LELUMAI as being the
>passive perfect of LUW. There are three faults with this. First, students
>miss seeing (or at least, are not being taught to see) the middle ending
>patterns which run across all four subsystems, with of course the variants
>between the "past forms" set of endings (imperfect, aorist, pluperfect) and
>the "non-past" set of endings (present, future, present perfect).
>>Second, students are having to learn forms and paradigms as separate and
>on their own, instead of seeing them as part of the greater whole (LELUMAI
>as part of the total MIDDLE structure) and allowing the identification of
>morphs in the word to guide them. (On this approach, you cannot use -MEQA
>to simply tell you "1st p. plural middle" if you have learnt LELUMEQA as
>being passive.)
>>Thirdly, your initial reaction to seeing such a form will be to take it as
>a passive and you may not even think to take into account that it could as
>easily be a middle.

This is a problem on several levels, in that there is a great deal of
duplication between different morphological forms of the verb, and as you
indicate, one can easily forget to consider all the possibilities. Having
Perseus or some other powerful parser parse a form now and then can be an
eye opener, because you'll be surprised very often at how many different
possibilities there are.

>>So this I recommend:
>>1. Teach students to recognize all middle forms as morphologically middle,
>across all four verb subsystems.
>>2. Teach them that when they encounter a middle form (which they can be
>shown to recognize from its morphs), check if the form contains QHS-. If
>so, the sigma in Slot 6 indicates "future", while the QH, the passive morph
>(in Slot 5), switches what would otherwise be a middle to being passive.

Don't forget that sneaky -H- variation.

Don Wilkins

This archive was generated by hypermail 2.1.4 : Sat Apr 20 2002 - 15:38:34 EDT