The yiqtol/qatal contrast in ancient Hebrew

In a previous exchange, I touched on a controversial topic: how best to describe the tense-aspect-mood system of the verb in ancient Hebrew. For example, does yiqtol mark tense, aspect, or both, and in what contexts? I argued, if only as it were on the back of a napkin, that it marks tense, and is aspect-neutral; John Cook, who has researched the question in great detail, argues that it marks aspect, and is tense-neutral; Randall Buth, who has engaged the topic at length, argues that it marks tense and aspect in some instances, and tense only in others. Ken Penner commented that his doctoral research on Qumran Hebrew supports the view that yiqtol, at least in that corpus, marks the future – whether to call the future a tense or a mood being a separate question. Rob Holmstedt indicated his agreement with John Cook, and notes that the textbook he and John have put together, which I very much like, describes the verb in terms of aspect. In this post, I explore the yiqtol/qatal contrast in more detail.

An observation about nomenclature. Some grammar books and Hebrew teachers use terms like perfect and imperfect for the sake of convenience, not because they want to take sides in a debate. I remember being taught to refer to yiqtol and qatal as imperfect and perfect early on, but at the time, the terminology did not speak to me one way or another. It was only after I began to read in the field of linguistics that I started to ask questions and formulate hypotheses. This post, and others which follow, is not meant to cut off discussion, but to stir it up. In my previous post, I came out swinging. In this post, the boxing metaphor does not apply. I just want to take a closer look at the evidence.

The first thing to note about yiqtol is how seldom, relatively speaking, “free-standing” yiqtols occur. By free-standing, I mean yiqtols not introduced by  כי  or והיה כי or פן  or כאשר and so on. Yiqtols of all kinds, furthermore, are not nearly as frequently occurring as are the workhorses of biblical narrative, to wit: consecutive wayyiqtols and weqatals.

With respect to כי-introduced yiqtols and such, there is general agreement, I think, that they are modal. They deserve very careful study, but I leave them to one side here.

Like Randall Buth, I have trouble wrapping my mind around John Cook’s assertion that “Randy's examples with MXR do not prove that qatal is not past tense; they only prove that it is not non-past tense.” But I want to be teachable, so keep at me, John! That מחר does not occur with qatal remains telling in my book.

Here is another example: what happens with וְאַחַר / וְאַחַר כֵּן / וְאַחֲרֵי־כֵן? These examples are typical:

וְאַחַר בָּאוּ מֹשֶׁה וְאַהֲרֹן

Afterward Moses and Aaron entered.

      Exodus 5:1

וְאַחַר כֵּן יָבֹא הַכֹּהֵן לִרְאוֹת אֶת־הַבָּיִת

After that the priest will enter to examine the house.

     Leviticus 14:36

 Is the difference between Exodus 5:1 and Lev 14:36 one of aspect? I don’t see how that could be the case. The yiqtol/qatal difference marks a difference in tense.

Again:

וְאַחֲרֵי־כֵן יֵצְאוּ בִּרְכֻשׁ גָּדוֹל

After that, they will go out with great wealth.

     Genesis 15:14

וְאַחֲרֵי־כֵן קָבַר אַבְרָהָם אֶת־שָׂרָה אִשְׁתּוֹ

After that, Avraham buried Sara his wife

     Genesis 23:19

 Is the difference between 15:14 and 23:19 one of aspect? How is it possible to claim this?

Once more:

וְאַחַר נָפֹצוּ מִשְׁפְּחוֹת הַכְּנַעֲנִי

Afterward, the Canaanite clans spread out.

     Genesis 10:18

וְאַחַר יַשְׁקֶה אֶת־הָאִשָּׁה אֶת־הַמָּיִם

Afterward, he will make the woman drink the water.

     Numbers 5:26

 I’m leaving qotel out of the picture because I’m not convinced that Randall is right that qotel marks present tense in ancient Hebrew. But once again, I want to be teachable, so don’t give up on me yet, Randall!

My intuitive sense of how the verbal system works in ancient Hebrew is bound to be wrong in certain sectors. The question is: where? It's all well and good to be directed to the secondary literature, but there is no agreement on these matters there. Linguistically trained Hebraists agree among themselves no more than do philologically trained Hebraists. With respect to the issue at hand, I still think I’m right: free standing yiqtol/qatal marks a difference in tense.

Comments

[Ken Penner]

I agree for the most part with where you're going, John, but should make a couple of cautionary comments.

First, it is probably possible to find a couple of examples or counterexamples to any of the major hypotheses regarding the semantics of the qatal-yiqtol opposition. This is how the tense model was "disproven": by providing a few examples of yiqtol for past reference and qatal for future reference. What is needed is an examination of a large sample that is not hand-picked to prove a point. I am inclined to guess that such an examination would lead to the rejection of the aspect hypothesis for Biblical Hebrew, but I make that guess based on my work with the Dead Sea Scrolls, not on any systematic examination of biblical texts.

Second, what you describe as a tense distinction could at least as easily be considered a distinction of modality. Of the three yiqtol statements you used as examples above, only one is a prediction; the other two are prescriptions.

I should also note that John Cook makes the best case I've seen for aspect, but I still find it unconvincing. I take issue with every one of the points in his comment to the earlier post. I am puzzled by a grammar that describes the qatal-yiqtol in terms of aspect (whole vs. in-progress), then says the perfect is used predominantly for past-time events, and the imperfect for non-past time events. It glosses the imperfect into English as a perfective (!) future: "he will attend". I don't see what is "in-progress" about this translation of the ostensibly "imperfective" yipqod. Would not "he is attending" or even "he will be attending" convey the imperfectivity better? Is this not how we would express non-past and in-progress actions in English?

Ken M. Penner

[JohnFH]

Ken,

I agree that one must be careful not to cherry pick the data. I should have been clearer that this post was based on systematic research covering all occurrences of yiqtol and qatal with axar, axar ken, etc. There are further cases of predictive yiqtol with axar.

It is also true that future tense and modality are ihterrelated with one grammatical form used to indicate both. That seems to happen in a lot of languages.

[John Cook]

Hi John,

Well you have managed to create and keep astir an interesting discussion. You note your interest over nomenclature. I would encourage you to call for care in distinguishing between "marking" and "expressing." All of us here in the discussion and most listening in (to retain the aural metaphor) would agree that the BHVS can express a full range of Tense-Aspect-Mood meanings. But this is different from the issue of the marking of individual verb forms. There are some who may argue that the latter (i.e., what meaning is expressed by the verbal conjugations) to be pointless, but this is to open up the door to discourse-pragmatic theories that run roughshod over syntax, semantics, and even sometimes morphology. I side with those who remain committed to the idea that words do mean "something" in isolation, otherwise we would be violating the most basic linguistic principle of compositionality (namely, that the meaning of an expression is based on the contribution of the constituent parts). Thus, it is important to ask what the contribution of the individual verbal conjugations are to the TAM meanings they express.

Thus, in your examples I wholeheartedly agree that there is an opposition of time going on, though I would prefer to lable it "temporal reference" rather than "tense," because the latter implies verbal morphological marking. The perfective qatal expresses past temporal reference a majority of the time; yiqtol expresses future temporal reference a majority of the time. These facts do not mean that these conjugations are marked for past tense and future tense respectively, just as the fact that modal verbs have future time reference does not automatically mean they are marked for future tense.

Making this important distinction goes a long way toward showing the common ground amongst us all regarding which verbs can express which meanings in BH. It is with regard to the explanation of what each conjugation contributes to these TAM expressions that we differ with one another.

[JohnFH]

Thanks, John, for pointing out the common ground amongst us all.

I like your suggestion that we use the terminology of temporal reference rather than tense so as not to decide the question before the race is run.

[John Cook]

Hi Ken,

I hadn't seen your post before I posted my comments. I appreciate your insightful remarks on my work. I appreciate your being unconvinced by yiqtol as imperfective expressing future, but it is indeed supported by cross-linguistic evidence, namely, that imperfectives can express general (not just imperfective) future (see Bybee, Perkins, and Pagliuca 1994: 275–76). We should bear in mind that the major difference between progressives and imperfectives is the generalization of meaning in the latter case. Hence, it is not entirely suitable to render the imperfective yiqtol with future time reference with English Future Progressive "he will be attending." Note similarly the ability of English present and present progressive to express future when in a future context: "I travel/am traveling to Kentucky next week."

Thus, to take one of John's examples, Gen 15:14: יֵצְאוּ has future temporal reference because of the context of future prediction that began, in this case, with יִהְיֶה in the previous verse. (I am of the opinion that היה is a tensed verb form: yiqtol for future and qatal for past. Longacre (2003: 64) rightly notes that the verb 'to be' is a special case in many languages.

Additionally, in light the alternative branching model of Bybee and Dahl versus Kurylowicz (see my discussion of them in Cook 2006),* the expression of progressiveness versus a more general imperfective meaning is more apparent in past time reference in which the imperfective contrasts with the typically past-time referencing perfective. Hence, in non-past time an imperfective can express either general present/future temporal reference or more specifically a progressive sense. The fact that there are nearly no (I can't pull an example of the top of my head, can anyone else) future progressive yiqtols is in large part due to the marginality of future progressives in all language communication. When was the last time any of you English speakers used a Future Progressive expression?

*Just to make the argument explicit, the contrast between Kurylowicz's models of Semitic and Bybee and Dahl's model of verbal systems in their cross-linguistic data comes down to a dispute over which is more basic to verbal systems, tense or aspect. Kurylowicz argues that former, stating that languages must distinguish tense before aspect (logically and etymologically) in their verbal systems. By contrast, Bybee and Dahl make several arguments that aspect is more basic in the world's verbal systems, and propose a branching model as representative of over half the languages in their data set. The model consists of a primary perfective/imperfective contrast and a secondary tense distinction for the imperfective. As I've stated in earlier comments, they find no reason to have an explicit tense distinction for perfective verbs, because they prototypically have past time reference, where they contrast with past imperfective verbs. By contrast, the imperfective can have general non-past tense/aspect reference because it does is not in contrast with an explicit perfective form in most cases. Dahl cites Classical Arabic as a prime example, in which qatala is perfective and limited mostly to past temporal reference, and yaqtulu is the general imperfective, which can have a general non-past sense, and kana yaqtulu (i.e., the periphrastic imperfective construction) that is marked for past imperfective in contrast to perfective (past) qatala.

Bybee, Joan, Revere Perkins, and William Pagliuca
1994 The Evolution of Grammar: Tense, Aspect, and Modality in the Languages of the World. Chicago: University of Chicago Press.
Cook, John A.
2006 The Finite Verbal Forms in Biblical Hebrew Do Express Aspect. Journal of the Ancient Near Eastern Society 30: 21–35.
Longacre, Robert E.
2003 Joseph: A Story of Divine Providence. A Text Theoretical and Textlinguistic Analysis of Genesis 37 and 39-48. 2nd ed. Winona Lake, IN: Eisenbrauns.

[Randall Buth]

Briefly, first to John C, then to John FH.

a. Still no explanation as to why there is no QATAL with maHar data.
b. Now no explanation as to why the different verbs 'ba' and 'yavo' were chosen. Why did the author do what he did? John FH appears correct that these are not being chosen to mark aspect.
c. I am quite unsatisfied with "imperfective" perfective futures. It's been many a year since I've read the studies cited, what do they say about modern Greek, a truly aspect-sensitive language? How do we know that Bickerton is not right about these studies, too, perhaps they are citing others who are ignoring the "Bickerton cake"?

and for John FH:
"I’m leaving qotel out of the picture because I’m not convinced that Randall is right that qotel marks present tense in ancient Hebrew. But once again, I want to be teachable, so don’t give up on me yet, Randall!"

eneni marpe velo arpe.
Here is something that might help:
Try to list five examples of actual, present tense YIQTOL, that are not questions or poetry (often reinterpretable as generic/habitual.) As mentioned, I don't accept "uxal", since that is a specific modal lexeme, mixing with the modal-friendly side of yiqtol.

[JohnFH]

Randall,

does Dahl represent a consensus point of view when he interprets the classical Arabic verbal system as at root aspectual? Or is there a continuing debate in that field that mirrors the one in ours?

I have an Arabist friend I could ask, but you might know the answer off the top of your head.