Re: kappa aorists

From: Carl W. Conrad (
Date: Mon Oct 20 1997 - 18:18:31 EDT

I regret reproducing as much of the earlier thread as I do, but I think
some useful information is coming across here, and I want to be clear about
points at which Ward and I share a common view of the facts of Greek verb
morphology even if we prefer to formulate those facts differently--perhaps
because of Ward's narrower focus on Biblical Greek as opposed to my
interest in Homeric and Classical Greek and in the the larger diachronic
view of Greek verb morphology.

At 9:24 AM -0500 10/19/97, Ward Powers wrote:
>At 13:20 97/10/18 -0500, Carl wrote:
>>At 9:34 AM -0500 10/18/97, Ward Powers wrote:
>>>There are however three verbs in NT Greek which are kappa aorists - that
>>>is, in these verbs the aorist morph is KA instead of the standard SA. Thus
>>>we are talking here of aorist FORMS, not of aorist MEANINGS. This KA is
>>>simply an allomorphic variant of SA, and these verbs conjugate their aorist
>>>exactly like a verb with SA aorist morph.
>>My only point of qualification here would be with the assertion that the
>>"KA is an allomorphic variant of SA." It was originally a distinct
>>(archaic) type of aorist formation more akin to the athematic types with
>>alternating long and short vowels in the singular and plural--and yet it
>>tends to be irregular, as one may find --even in the classical period -- a
>>3rd plural EQESAN or EQHKAN (where the singular mode of conjugation has
>>been carried over to the plural, and likewise alternative forms of DIDWMI:
>Carl, thank you for your historical comments. But "allomorph" (from ALLOS
>and MORFH) is a linguistic term meaning "another form of a morph with the
>same meaning".

In turn I thank you for explaining the terminology you're using and
correcting my misapplication of it. I would guess (and certainly hope) that
this terminology is reasonably standard in use by linguists and not so
idiosyncratic as the terminology used to describe verbal aspect.

The terms "morpheme," "phoneme," and "phone" are by no means new to me,
although the term "allomorph" was. I do recollect Joshua Whatmough some 45
years ago objecting not to the use so much as to the spelling of
"morpheme," claiming that it really ought to be "morphome." That's an
argument I don't want to take up; it does seem to me that only biologists
make much of an effort to use Greek and Latin roots in a consistent fashion.

>These same linguistic categories apply to Greek. Thus the syllabic augment
>and the temporal augment are allomorphs of the Greek past time morpheme.

If I understand you rightly here, you are indicating equivalence of
allomorphs in terms of formal function within the constituent building
blocks of the form of a verb. You are NOT making any judgment about the
primacy of one allomorph over another, even IF one of these allormorphs may
happen to be earlier while another may be a later-emerging element used the
same way as another.

>And the morphs which indicate that a word form is aorist are allomorphs of
>the Greek aorist morpheme. These aorist allomorphs are:
>SA in most words;
>KA in kappa aorists;
>I.A (where the dot indicates an intervening liquid phone) in words with
>liquid roots where the vowel before the liquid is epsilon, as in EMEINA,
>from MENW (the iota before the nu and the alpha after it together indicate
>"aorist", and are termed a "discontinuous morph"); and
>A on its own in other liquid verbs and, irregularly, in the digamma verb
>CEW (e.g. ESURA, from SURW; EKRINA, from KRINW; ECEW, from CEW).
>For those with access to my "Learn To Read the Greek NT", I refer you to
>sections 4.57 and 4.59, page 56, and E4.77, page 301.

I'm going to try to get a copy; apparently it is not currently in print or
available from distributors in the continental U.S. I'm assuming that in
the last part of the last parenthesis above you meant to write "ECEA from

One matter on which I am not wholly clear in the above formulation is
whether you really mean to make these three forms fully equivalent. For
instance, you're not, I suppose, meaning to indicate that I.A is the
pattern for all liquid verbs, but only for those where, as you say, the
vowel before the liquid is epsilon. Obviously you would have to cite
additional allomorphs for liquid verbs with different vowels preceding the
liquid, as, EGHMA from GAMEW or EFHNA from FAINW. Of course the student
must know that these liquid aorists are built to the verb root generally
(as is also true of liquid futures), and not just to a present stem.

Moreover, it appears that you can include the aorist ECEA here because the
preceding vowel is an epsilon and the intervening consonant was once a
digamma. Of course there are other digamma verbs that behave
differently,like EPNEUSA from PNEW, etc.

Let me reiterate my question from above for clarity's sake: you are NOT
arguing that there is any historical relationship between aorists in -SA,
others in -KA, and still others in -I.A or E.A; what would you say of
those older thematic aorist stems so commonly conjugated in Biblical Greek
with alpha endings, such as HLQA, hUPHGA, EFAGA, etc.? Are these to be
deemed additional allomorphs of -SA?

Furthermore, does the fact that such -KA aorists as EQHKA may form a 1 pl
as EQHKAMEN or as EQEMEN alter the above description?--the problem, of
course, is that we may find in Hellenistic Greek (and for that matter, even
in classical Attic) concurrent forms of aorist active plurals based on the
-KA on the one hand or conjugated as athematic verbs on the other. I guess
what I'm really asking is whether this descriptive pattern is comprehensive
enough to cover the sometimes disparate phenomena of alternative forms of
the same verb.

>>While it may well be true that the second aorist form HNEGKON antedated
>>historically the form HNEGKA, the fact is that both are found in Homer, as
>>is also at least one other aorist conjugated in -A, namely ECEA (from CEW,
>>I think that a simpler statement about the facts of alpha conjugation is
>>that (a) it is already found in a couple aorists in Homer that also have
>>thematic "second" aorist conjugations; (b) it is already being used in
>>Homer with the Sigma aorist marker to constitute what is traditionally
>>called the "first aorist"; (c) it is used as early as Homer with the Kappa
>>aorist marker, although consistently only in the singular of DIDWMI,
>>TIQHMI, and -hIHMI;
>If I may say so, I would question the validity (certainly, from a
>linguistic perspective) of speaking of sigma as an aorist marker, and then
>referring to an alpha conjugation. The term "conjugation" means "pattern of
>conjugating", and this refers to the differences in the pattern when the
>verb forms for a tense are laid out in a table (a paradigm) for
>singular/plural, and first/second/third persons. On this basis, there are
>exactly three conjugations in NT Greek, First Conjugation (those taking the
>first aorist active set of endings), Second Conjugation (those taking the
>second aorist active set of endings), and Third Conjugation (MI verbs and
>the four other verbs taking the third aorist set of endings). Formation
>factors affecting the *stem* to which a set of endings is added do not
>actually affect the way a verb is conjugated, and thus have no connection
>with the conjugation of a verb.

I apologize for my sloppiness of language in the above description; you're
quite right--I should have referred to an alpha "conjugation"--although I
think I may rightly speak of a pattern of alpha endings that appear inthe
aorist active as -A, -AS, -E, -AMEN, -ATE, -AN. It is commonly believed
that this pattern of endings developed from vocalization of the secondary 1
sg. -M and 3d pl. -NT endings and spread throughout the other persons and
numbers. Be that as it may, these alpha terminations do not derive, it
would appear, from the Sigmatic/First Aorist, but rather the Sigmatic/First
Aorist is a combination of a PIE -S- aorist marker and these alpha endings.

I raise this issue because regardless of whether it is convenient to
classify these aorists ending in -A as "allomorphs" of each other, it does
not seem to be historically accurate to derive the others from a type in
-SA. I've referred to the concurrent Homeric forms HNEGKON and HNEGKA for
the aorist 1 sg. active--and there really is no way of knowing which of
these is oldest; one also finds concurrent EIPA and EIPON aorist 1 sg.
active in Homer; it may be that EIPA is secondary, but that is not readily
demonstrable so much as that it came later to be much more common.

>If one compares forms which differ only in tense such as
>feature which identifies the aorist form and differentiates it from the
>others is the presence of SA (in Slot 7, the aspect slot, of the verb's
>nine morph slots). This identification is a simple application of a
>principle of linguistic analysis to the forms of a verb, and enables SA to
>be recognized as the morph which carries the information, "aorist form",
>and thus to be identified as the aorist morph.

>Further examination of other aorist forms of a verb enables us to refine
>this to say that the alpha of the aorist morph elides before a following
>vowel, as in such forms as ELUSE(N), LUSON, LUSAI, and the subjunctive forms.

Here again I find a bit of a problem. I frankly doubt that there's any
elision of an -A- from a -SA- form, and I think one would be hard put to
demonstrate any such elision. It is easier to suppose that the alpha
originated in the indicative active and spread as an element additional to
the aorist -S- marker in several other indicative and non finite forms
(participles, infinitives, imperatives except for the aorist 2 sg. in -ON,
subjunctives; it appears in optatives in -SAIMI, SAIS, SAI but not in the
Attic optative 2nd sg. -SEIAS, 3d sg. -SEIE or 3d pl. SEIAN). That is to
say, I don't know that one can really say that the A in -SA- is a universal
aorist marker; I rather think that -S- is for the sigmatic type aorist and
that the additional -A- is not originally part of it. On the other hand, it
might be more convenient for pedagogical purposes to speak of "elision" of
the Alpha before an imperative ending -ON or before subjunctive vowels, so
long as one doesn't confuse that hypothetical elision with description of
what is actually involved.

Let me reiterate here once more, that my concern is probably more with the
historical development of the aorist forms than with how best to describe
them for pedagogical purposes. But even when teaching the aorist in
Biblical Greek, one must come to terms sooner or later with the fact that
the language does not fit quite so neatly into the slots as one might wish.
As the rabbit told Alice, those verbs are "an ornery lot" and one must
really struggle with them to master them.

>>>In any case, hHKW is not a kappa aorist - its aorist is hHXA.
>>Is this form hHXA (X = KS?) actually attested?
>Yes. hHXA occurs in the NT in Rev 2:25, in the form hHXW, 1p. aorist
>subjunctive after AN. This is cited in BAGD on hHKW, together with a ref.
>to its occurrence in POxy.933.13.

Carlton Winbery has already noted what I'd be inclined to say about the
appearance of hHXW in Rev: it's probably a future rather than an aorist
subjunctive. But I've had a chance to look at the latest LSJ/Glare and find
that there's quite a variety of alternative tense-forms of hHKW in
Hellenistic texts--all apparently late developments of new stems built upon
the stative present tense which alone is attested in earlier Greek.

>>>Please tell me - is anyone out there interested in this kind of info?
>>Well, Ward, there is at least One!
>Thank you. If there are others who confirm also (either on-list or
>off-list), I will continue to post similar occasional comments, when

By all means do. Yet once more I want to add that the perspective from
which I'm writing is not of how best to represent the facts of Greek verb
morphology for students of the Bible--and that is very important in its own
right--but rather how best to understand the emergence and spread of alpha
endings in the indicative past tenses of Greek over a period of 3000 years.
I'm interested in the diachronic problem, whereas I think Ward is focused
strictly upon synchronic description of Biblical Greek verb morphology.

Carl W. Conrad
Department of Classics/Washington University
One Brookings Drive/St. Louis, MO, USA 63130/(314) 935-4018
Home: 7222 Colgate Ave./St. Louis, MO 63130/(314) 726-5649 OR

This archive was generated by hypermail 2.1.4 : Sat Apr 20 2002 - 15:38:33 EDT