Tense-mood-aspect system of ancient Hebrew

[zaw]

Hi!

I take as my point of departure the following thesis, which I agree with but would like other people's feedback on:

(1) yiqtol is the default future tense in ancient Hebrew.

According to some scholars, yiqtol is a tense-aspect in ancient Hebrew. On this view, sometimes the yiqtol’s imperfective aspect is suppressed, and sometimes it is not. I prefer to say that yiqtol sometimes marks future tense (and is aspect-neutral); sometimes aspect (for example, in contexts dominated by narrative past tense wayyiqtols); and sometimes, in conjunction with certain function words, mood. In fact, a case might be made that yiqtol is aspect-neutral, and that examples Randall Buth classifies as “past habitual” and “past continual” are weak modals, equivalent more or less to ‘would’ in English. Unusually, and only in poetry, yiqtol with or without an initial waw consecutive serves as a narrative past tense (e.g., in Deut 32:10-18).

It is also a fact that yiqtol is the default future tense in ancient Hebrew, corresponding to qatal as the default past tense. This is a shorthand way of saying that if one is talking to someone else in ancient Hebrew, it is correct to reach for a yiqtol form when beginning to speak about something one expects to happen in the future, and conversely, it is correct to reach for a qatal form when beginning to speak about something that belongs to the past. For example:

אָנֹכִי אֶעֱשֶׂה כִדְבָרֶךָ
(1) I will do as you have spoken.
Gen 47:30

אֵצֵא וְהָיִיתִי רוּחַ שֶׁקֶר בְּפִי כָּל־נְבִיאָיו
(2) I will go out and be a lying spirit in the mouth of all his prophets.
1 Kgs 22:22

עַד יִגָּמֵל הַנַּעַר וַהֲבִאֹתִיו
(3) When the boy is weaned, I will bring him.
1 Sam 1:22

Qatal examples:

אָבִינוּ מֵת בַּמִּדְבָּר
(4) Our father died in the wilderness
Num 27:3

רָאִיתִי אֶת־אֲדֹנָי נִצָּב עַל־הַמִּזְבֵּחַ
(5) I saw my Lord standing by the altar
Amos 9:1

נִשְׁבַּע יְהוָה בִּימִינוֹ
(6) The Lord swore with his right hand
Isa 62:8

(3) is interesting, because the TMA system of English does not use its default future tense in that kind of situation.

Randall Buth makes the argument that yiqtol is the default future tense when he notes “which [verb forms] are attested with a word like מחר tomorrow (52 occurrences in the Bible).” As he points out, yiqtol, consecutive weqatal, participles, and imperatives are attested with this verb, but not qatal. מחר עשה יי הדבר הזה ‘Tomorrow יי will do this thing’ is not ancient Hebrew; that would be מחר יעשה יי הדבר הזה (Exod 9:5).

Furthermore, ‘And when my glory passes by, I will put you in a cleft of the rock,’ which a language that systematically marks aspect might realize with an imperfective – perfective sequence, is realized quite otherwise than with a yiqtol – qatal sequence in biblical Hebrew:

וְהָיָה בַּעֲבֹר כְּבֹדִי וְשַׂמְתִּיךָ בְּנִקְרַת הַצּוּר
Exod 33:22

Perfective futures in ancient Hebrew are not expressed by qatal. They are expressed by yiqtol or consecutive weqatal (as in the preceding example). Therefore, the yiqtol-qatal contrast is not aspectual in nature.

Here are some Russian examples of perfective and imperfective futures. Don’t worry, you don’t need to know Russian to get the distinction:

Буду читать (imperfective future) статью, надеюсь, что прочитаю (perfective future)
I shall read/be reading the article and hope I shall get it finished.

Қогда я буду проходить (imperfective future) мимо аптеки, куплю (perfective future) табпетки от кашля
When I pass the druggist’s I shall buy some cough drops.2

So, it is clear that yiqtol and qatal are NOT tense-neutral, and/or primarily aspectual. Rather, yiqtol and qatal have a number of specific, context-sensitive usages. In conjunction with other discourse cues, they mark tense, mood, and omnipotentiality or lack thereof (qatal, like the so-called gnomic aorist in Greek, is used to mark omnipotentiality). If anything, yiqtol and qatal appear to be aspect-neutral. The terms imperfect and perfect are inappropriate.

I would love your feedback on this thesis. Toda raba!

 

[S1234]

Without checking context, your example 3 might have used העבר in the subordinate clause. See examples 31-32 in Randall Buth's article "The Hebrew Verb: A Short Syntax".

 

[Drink]

Well you are not the first to say this, and I believe you are mostly right.

However, I don't think #3 is a good example, because there is a modal element there as well.

 

[Sharjeel72]

I agree. In Ken Penner's dissertation on tense, aspect, and modality in the Dead Sea Scrolls, he demonstrates that the qatal-yiqtol opposition is one not of aspect, but of tense or modality (if future and habitual actions are modal). A waw-prefix is “conversive.”

Our professor teaches Biblical Hebrew yiqtol as modal, and qatal as past (except for semantically stative verbs).

zaw, you stated that the yiqtol-qatal contrast is not aspectual in nature as though it were an agreed-upon fact. I would like to hear your review of two works, one by John Cook (The Finite Verbal Forms in Biblical Hebrew Do Express Aspect.), and one by Rolf Furuli (New understanding of the verbal system of classical Hebrew : an attempt to distinguish between semantic and pragmatic factors (Book, 2006) [WorldCat.org]).

 

[mj99]

zaw, I could not disagree more. I think John Cook’s aspectual model is the best explanation given so far (note the reference that Sharjeel72 gave). He not only explains the Biblical Hebrew data but he, in my opinion, is the only one who has adequately dealt with the various linguistic models and cross-linguistic patterns.

You haven’t given any proof for your model. And if you want a model that is scientific, you’ll need more than a deep knowledge of the biblical texts and a comparison of one other language, you’ll need just as deep a knowledge of TAM systems -- and how they develop diachronically -- in Semitic and in non-Semitic languages. And again, you’ll spend an American PhD worth of time trying to better John Cook’s theory -- and I doubt it’s possible to produce an alternate theory that explains the data from all angles as elegantly as John Cook’s.

 

[Drink]

Could you explain why you believe John Cook to be correct against the opinion of many other scholars?

 

[JAN SHAR]

zaw, notwithstanding mj99’s remarks, I will add the following comments regarding your assertion that the yiqtol-qatal contrast is not aspectual in nature:

1) Your 6 examples are all direct speech. While the verb forms appear to contrast with each other in terms of tense in direct speech (i.e., qatal = past, qotel = present, yiqtol = future), this model does not work for non-speech.

2) S1234’s examples with מחר do not prove that qatal is not past tense; they only prove that it is not non-past tense. Both conclusions assume tense a priori: either qatal is past tense or it is non-past tense.

3) Most of your discussion is based on simple intuition, and all by people (you, S1234, Sharjeel72) whose native language is tense-based. No wonder the language appears to “work” as tense—that is the character of the metalanguage.

4) As a result, I find statistical studies such as Ken Penner’s and Furuli’s (despite the great differences between their conclusions and the generally much more sound linguistic foundation to Ken Penner’s) provide me with no more than a statistical tallying of their particular intuitive interpretation of the verb in its various contexts. Statistics give a false sense of objective proof in semantic study.

5) So why the differences of opinion on tense versus aspect with regard to qatal? I believe it is due to the semantically close relationship between past tense and perfective aspect as noted by Dahl 1985: 79 (available in pdf online: https://www2.ling.su.se/staff/oesten/recycled/Tense&aspectsystems.pdf). Dahl states that cross-linguistically ‘past time reference’ characterizes the typical use of perfective verbs.

(The case is similar to the English “will”: Is it future or modal? Linguists disagree, but I side with James D. McCawley that since certain statements about the future are judged by people to be “true” or “false,” there must be a non-modal future to be referred to by tensed “will.” Another way to argue the case is that future-time reference always accompanies the use of “will” whereas a modal sense is not always apparent. Similarly, perfective aspect always accompanies the use of qatal, whereas past temporal reference is typical but not exclusive of the form, on which see below.)

6) So how then can a case be made for perfective aspect at all? It would appear from Dahl’s observation that all perfective forms could easily enough be treated as past tense and then we would eliminate another one of those pesky TAM categories. My response is that there are two types of evidence germane to BH, and which are derived from cross-linguistic analysis and not intuitive interpretation:

a) The first is that the qatal form is regularly used for present-time reference performative statements, such as Gen 15:18: לְזַרְעֲךָ נָתַתִּי אֶת־הָאָרֶץ הַזּאֹת . This is directly contradictory to the notion that the qatal is past tense, whereas it makes sense with a theory that recognizes that qatal is perfective aspect that typically has past-time reference but in cases such as this can have a present-time reference.

b) Cross-linguistically past tense verbs and perfective verbs interact differently with stative predicates, thus providing an objective basis for distinguishing the two. The pattern is a privative marked one: past tense verbs with stative predicates always express states or inchoative events with past time reference, whereas perfective verbs with stative predicates express either past or present time reference, depending on the context. That is why we find in the Bible examples of ידע in qatal expressing present states “I know” (e.g., Gen 12:11: הִנֵּה־נָא יָדַעְתִּי כִּי אִשָּׁה יְפַת־מַרְאֶה אָתְּ ). By contrast, wayyiqtol always has a past temporal reference (e.g., Gen 3:7: .(וַיֵּדְעוּ כִּי עֵירֻמִּם הֵם

So, just to reply to your assertion, good-naturedly of course, I challenge you to explain away this important typological data with your intuitively-based tense interpretation.

For discussion of this typological data, see:

Bybee, Joan. 1998. “Irrealis” as a Grammatical Category. Anthropological Linguistics
40: 257–71.

Bybee, Joan L., and Östen Dahl. 1989. The Creation of Tense and Aspect Systems in
the Languages of the World. Studies in Language 13: 51–103.

Bybee, Joan L., 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.

Dahl, Östen. 1985. Tense and Aspect Systems. Oxford: Blackwell.

On yiqtol in particular, see John Cook's exchange with Jan Joosten in JANES:

Cook, John A. 2006. The Finite Verbal Forms in Biblical Hebrew Do Express Aspect.
Journal of the Ancient Near Eastern Society 30: 21–35.

Joosten, Jan. 2002. Do the Finite Verbal Forms in Biblical Hebrew Express Aspect?
Journal of the Ancient Near Eastern Society 29: 49–70.

 

[zaw]

Now that you’ve weighed in, JAN SHAR, I should just head for the locker room and call it a day.
But I’ll come back to the topic after I lick my wounds.

 

[Drink]

1) Your 6 examples are all direct speech. While the verb forms appear to contrast with each other in terms of tense in direct speech (i.e., qatal = past, qotel = present, yiqtol = future), this model does not work for non-speech.

Aside from the mere fact that direct speech makes up a rather small sample size, what would be your issue with using direct speech as a gauge for tenses in reality. After all, speech is the essence of language, and narratives are only a byproduct. In English, for example, it's common to use what's normally the present simple to narrate past events (though it's not as common in writing, but let's leave that aside). Does that mean we should ignore ordinary speech and use this narrative usage to evaluate the "true" meaning of the present tense?

2) S1234’s examples with מחר do not prove that qatal is not past tense; they only prove that it is not non-past tense. Both conclusions assume tense a priori: either qatal is past tense or it is non-past tense.

If you show that qatal is not used in the future tense, then you've clearly identified it as non-future. That's already a tense. Even you want to argue it also has aspect, you can't deny its tense here. This is why TAM models need to be viewed and described holistically and not split into three parts as if they're independent of each other.

3) Most of your discussion is based on simple intuition, and all by people (you, S1234, Sharjeel72) whose native language is tense-based. No wonder the language appears to “work” as tense—that is the character of the metalanguage.

Seems you've left me out

4) As a result, I find statistical studies such as Ken Penner’s and Furuli’s (despite the great differences between their conclusions and the generally much more sound linguistic foundation to Ken Penner’s) provide me with no more than a statistical tallying of their particular intuitive interpretation of the verb in its various contexts. Statistics give a false sense of objective proof in semantic study.

I actually agree with this point. While statistics are probably the only way to eventually reach a final conclusion, you really have to have a consensus over each and every data point for that conclusion to be worth anything.

5) So why the differences of opinion on tense versus aspect with regard to qatal? I believe it is due to the semantically close relationship between past tense and perfective aspect as noted by Dahl 1985: 79 (available in pdf online: https://www2.ling.su.se/staff/oesten/recycled/Tense&aspectsystems.pdf). Dahl states that cross-linguistically ‘past time reference’ characterizes the typical use of perfective verbs.

I believe it's primarily due to the rather misguided desire to see it as either or. Or to see one as "primary" and the other as "secondary". It's silly to do such things. If a verb form conveys multiple shades of meaning at once, and one of them is a time reference, and the other is an aspect, then that verb form indicates both tense and aspect. Nothing more needs to be said and any further discussion of which is more "primary" is mere fantasy.

(The case is similar to the English “will”: Is it future or modal? Linguists disagree, but I side with James D. McCawley that since certain statements about the future are judged by people to be “true” or “false,” there must be a non-modal future to be referred to by tensed “will.” Another way to argue the case is that future-time reference always accompanies the use of “will” whereas a modal sense is not always apparent. Similarly, perfective aspect always accompanies the use of qatal, whereas past temporal reference is typical but not exclusive of the form, on which see below.)

What you say about "will" always being accompanied by a future time reference is simply not true. "Will" is also pretty commonly used with an aspectual/modal meaning with a present time reference.

a) The first is that the qatal form is regularly used for present-time reference performative statements, such as Gen 15:18: לְזַרְעֲךָ נָתַתִּי אֶת־הָאָרֶץ הַזּאֹת . This is directly contradictory to the notion that the qatal is past tense, whereas it makes sense with a theory that recognizes that qatal is perfective aspect that typically has past-time reference but in cases such as this can have a present-time reference.

I don't believe there is any contradiction here. The fact that the same form of the verb has different meanings in different semantic categories of verbs does not indicate any synchronic need to reconcile them.

In Russian and other Slavic languages, for example, there are perfective verbs and imperfective verbs (which are determined lexically). The same verb form that serves as the present tense of imperfective verbs serves as the future tense of perfective verbs. Does this mean there is some sort of contradiction and therefore these tenses cannot be analyzed as tenses? No. They are tenses. And in fact we know they can't be aspects because the aspect is already covered by the lexical category. It can't be explained by mood either. It's just that the same exact form has two different meanings when used with two different lexical categories. One who wants to know the history can trace this back to some historical form that can explain this difference, but that has no bearing on the language as it is.

b) Cross-linguistically past tense verbs and perfective verbs interact differently with stative predicates, thus providing an objective basis for distinguishing the two. The pattern is a privative marked one: past tense verbs with stative predicates always express states or inchoative events with past time reference, whereas perfective verbs with stative predicates express either past or present time reference, depending on the context. That is why we find in the Bible examples of ידע in qatal expressing present states “I know” (e.g., Gen 12:11: הִנֵּה־נָא יָדַעְתִּי כִּי אִשָּׁה יְפַת־מַרְאֶה אָתְּ ). By contrast, wayyiqtol always has a past temporal reference (e.g., Gen 3:7: .(וַיֵּדְעוּ כִּי עֵירֻמִּם הֵם

This is a great point if you want to explain the origin of these forms. And in fact it's supported by comparative evidence. However, it doesn't really say much about Hebrew synchronically.

 

[S1234]

JAN SHAR wrote, "S1234’s examples with מחר do not prove that qatal is not past tense; they only prove that it is not non-past tense. Both conclusions assume tense a priori: either qatal is past tense or it is non-past tense."

Let's analyze JAN's statements by starting at the point of agreement:

"they [S1234’s examples] only prove that it [qatal - S1234] is not “non-past.”"

That admits that there is a time-component in there somewhere, that there is a time-feature in the verb. There is a positive interactive with ‘non-pastness.’ It is exactly the smoking gun that ‘aspect-only’ theorists pretend doesn’t exist. ‘Aspect-only’ would say that the context marks the time, like the word ‘tomorrow’ in the examples, and the verb would mark the aspects. But 52 to zero the Hebrew verb refuses to use a particular “aspect” with that TIME word. A-priori-ness has nothing to do with this conclusion. It is simply a false/bad prediction of “aspect-only” theory. That is why the Hebrew verb MUST be defined as a Tense-Aspect-Mood, not a pure aspect, not a pure mood, and not a pure tense. To wrangle about which of the three was first, is a wrangling about etymology, not meaning, to wrangle about which of the three is ‘more prominent’ is a subtle repetition of the same etymological philosophizing and is irrelevant to a language user and to the synchronic system. Once the whole TAM is in the simple verb system, the whole TAM is in the system.

People just don’t seem to understand the impact of closed systems in the language world, illustrated in Bickerton’s research into Creole and TAM. Creoles tend to start to morphogrammatize ‘perfective’ (so don’t be surprised to see ‘aspect’ at the base of Indo-European or Greek, or child development), but if the morpho-structure stops there, the verb system will use those “aspects” for time as well as mood. The meaning of a system is determined to a large extent by how many pieces the cake is divided, and the cake includes aspect, mood and TIME. Cross-linguistic tendencies are nice, but not water-tight and sometimes done by people who are ignoring “Bickerton’s cake.” Rajesh Bhat (Papers: Rajesh Bhatt, Linguistics, The University of Massachusetts at Amherst) warned people not to use his classification system as a water-tight predictor of semantics. Another example, if sequential tense systems typically have only one person-inflected verb structure, and we find two in biblical Hebrew, would that negate what Hebrew is?

 

[Aleppan]

zaw,

Thanks for calling this thread to my attention. I’m sympathetic to your proposal. I was taught that the basic meaning of yiqtol is to express the future. But to my mind that is simplification. In a discussion among grammarians I would rather define the basic meaning of yiqtol as the expression of irrealis: yiqtol means the process expressed by the verb is not (yet) begun at the moment of speaking (or at reference time); it is contemplated.

In reaction to some of JAN SHAR’s points I would say firstly that yes, future yiqtol forms occur in direct speech: where else would you expect them? Since narrative is situated at a point in the past, it does not allow the use of the simple future. Where the “future in the past” is to be expressed, in narrative, yiqtol is used (2 Kgs 13:14 Now Elisha had fallen sick of his sickness whereof he was to die [ימות]).

Secondly, I’m not a native speaker of English, but I don’t think it’s true “will” statement always refer to the future. JAN SHAR will say that, of course; but English “will” like BH yiqtol is also used in the expression of general truths and habitual processes.

I agree that qatal is not a past tense (although, again, I was taught that qatal expresses roughly the past). In my view, qatal is a perfect: it depicts the process as anterior to the moment of speaking (or to the reference time where this does not coincide with the moment of speaking). But a grammatical perfect is not the same as a perfective. Performatives, in my understanding, should not be used to argue grammatical meaning because the function is wholly dependent on the pragmatic context. נתתי can mean “I have given,” and does so very often; only the speech situation (the felicity conditions) may lend it a meaning that we render in English with a present tense.

Finally, and just for fun: wayyiqtol does not always have past temporal reference. Ps 45:8 אהבת צדק וַתשנא רשע, “Thou lovest righteousness, and hatest wickedness”. This usage occurs with the verbs נבהל ,ידע ,ירא ,גיל ,בטח and לאה.

 

[S1234]

Hi Aleppan,

I, too, am sympathetic to your “future,” a.k.a., “irrealis.”

I think that I see you doing what Joüon did almost a century ago, using terms that are incongruent, in order to implicitly expose the complexity/simplicity of the system. Joüon juxtaposed perfect/future (an aspectual term and a time term) while you are juxtaposing perfect / irrealis (an aspect and a mood).

Someone once used a term ‘indefinite,’ meaningless by itself and too לועזית for a classroom, in order to include both time and aspect, and implicitly mood, in the label of the Hebrew yiqtol. I think that it is helpful for all concerned to follow Bickerton’s advice not to try to ‘straightjacket’ something into one dimension of a multi-dimensional usage and reality. The problem or power of the biblical Hebrew TMA is that it “under-differentiates,” but still covers all the bases, including TIME. In one sense it is using a binary switch in a 3+ parameter world.

Actually, Biblical Hebrew uses an emerging three-position switch, since I agree with you that the participle, contrary to Arabic, had already become a real present tense in First Temple Hebrew. (Though a couple of modal lexemes חפץ, יכול , lagged behind and only expanded into the three-layered TMA at the end of the First Temple e.g. 1 Kgs 21:6 אם חפץ אתה ... also LBH. (And בינוני יכל only in Biblical Aramaic).

For the record, we use אינני חפץ ,אני חפץ in our class along with a binary אני אוכל, לא אוכל . We have debated going ‘First Temple only’ which would result in something like a present situation: חפצתי אך לא אוכל “I want to but I can’t.” (Maybe they’ll adopt it this summer, I’ll talk with teachers again. They let in words like כבר.) But the point isn’t time-machine purity, but rapid language acquisition for biblical Hebrew, especially for those not starting from modern Hebrew.

 

[Aleppan]

Hi S1234,

The participle of יכל may be attested in Arad (Aharoni) 40:14:
איננו יכלם לשלח
we can’t send

This would be First Temple (colloquial?). But the reading is admittedly doubtful.

 

[S1234]

Thank you for that, aleppan. I knew I was forgetting something out there. The reading is actually solid for יכלם , it is only the אי]ננו ] that is partial. It just reinforces how the participle had taken over for the present tense, because חפצים HAFETSIM and יכלים YEXOLIM were among the last holdouts.

 

[Drink]

How much evidence is there that they were in fact hold-outs though? I think it would be pretty hard to make such fine distinctions about individual lexemes, especially given the fact that חפץ is not a super frequent lexeme, and for יכל, there isn't much semantic distinction between a concrete present and a hypothetical present/future.

 

[JAN SHAR]

S1234 and Aleppan,

I can’t resist responding to your previous post.

1) S1234, your argument is a non sequitur. Although I claim that qatal-yiqtol form an aspectual opposition, it does not follow that I deny the BH verb the ability to express TIME. Indeed, I am in full agreement with your reasonable claim that the BH verbal system is Tense-Aspect-Mood; to wit, wayyiqtol is past (narrative) tense, qatal-yiqtol opposition is aspectual (perfective-imperfective), and VS ordered qatal and yiqtol, as well as the imperative, are irreal mood. You’ve chosen to ignore a most important datum in Dahl’s observation, that prototypically perfective verbs have past TIME reference.

2) Your dismissive comments regarding “wrangling about etymology” is misplaced, as is your trivializing of cross-linguistic tendencies. In the first case, typologists are now recognizing that, to quote Moravcsik, “Indeed, the only possible causal explanation for a language system is by reference to history: how a given system evolved from something else” (2007: 38). Thus, historical explanation cannot be dismissed from the discussion of the Hebrew verb. The results from those who claim a “synchronic only” approach over the past century have proven as much. In the second case, in the absence of native speakers I would posit that cross-linguistic tendencies are the closest we come to an objective basis for analyzing ancient verbal systems. Thus, we should have to come up with better dismissals of the validity of these tendencies than intuitive or traditional interpretations or pragmatic arguments regarding the easiest approaches to teaching and learning. While pragmatic decisions are bound to play a part in language teaching, they should not be confused with accurate descriptions of the language (this goes to Aleppan’s points as well; e.g., even though I was taught qatal and yiqtol as aspectual / modal (based on word order) in my Hebrew courses, the professors did give us students—mostly native speakers of tense-prominent English—rudimentary clues to begin translating these forms that are based in the grammar of our native languages, such as use past tense or a form of the English Perfect to render qatal, and use present or future to render yiqtol.

3) In response to your comments, Aleppan, your example from 2 Kgs 13:14 (Now Elisha had fallen sick of his sickness whereof he was to die [ ימות ]) demonstrates that point. I can’t think of any sample languages that allow a future-tense marked verb form to function with past reference — even if it is future-in-the-past (I would appreciate if anyone does know of an example). I think that examples such as these support instead taking yiqtol as aspectual (imperfective) or as modal, as you have; other factors lead me to argue the former rather than your latter option.

4) Point taken that perfect and perfective are not the same thing. However, perfect forms do develop into perfective forms, and in the process they may not lose their earlier perfect meaning. Thus, our professors explained to us that qatal is perfective (prototypically with past time reference, hence translatable by Past Tense in English and other tensed languages), but that it has held on to its earlier perfect meaning, so that it expresses both depending on the discourse context, verb sequence, etc.

5) I cannot agree though with your dismissal of the importance of qatal in performative statements. Perhaps though I could have been more clear. The issue is not simply one of temporal reference, but also aspect: performatives cross-linguistically use punctiliar type verb conjugations (such as perfectives) rather than durative or progressive. Thus, in English the person presiding over a wedding will say “I (hereby/now) pronounce you . . .” but not “I (hereby/now) *am pronouncing you . . .” Thus, the evidence goes towards arguing that qatal expresses perfective aspect (even if one still claims that it is past tense); but further, the fact that the performative statement has present time reference remains a valid argument against a past time interpretation of the form. (Unless you persuade me otherwise; I’m not entirely clear on why you object to the argument).

6) Finally, comments on English ‘will’ not having future time reference and wayyiqtol sometimes not having past time reference both deal with gnomic or generic type statements. It has been argued elsewhere that gnomic statements (cross-linguistically—sorry S1234) allow for a wide range of verb tenses (see esp. Carlson and Pelletier 1995). Thus, Gross some years ago already noted the use of wayyiqtol in gnomic expressions. That said, I would argue that ‘will’ in gnomic expressions portrays the event in a particular light: as a future prediction of what will happen based on the way the world ‘works’ (there are several different models for explaining gnomics, but the point is valid in any case). As for the Ps 45:8 example, I don’t think that wayyiqtol is non-past there. I render it in English, “You have loved righteousness and hated wickedness. Therefore, God your God has anointed you . . .” The perfect interpretation of the qatal holds for the wayyiqtol as well, just as in the case of flashback story-lines in BH narrative, where the initial qatal expresses past perfect and the following wayyiqtol past narrative forms continue the storyline (e.g., Gen 39:14: וַיְהִי כִּרְאוֹתָהּ כִּי־עָזַב בִּגְדוֹ בְּיָדָהּ וַיָּנָס הַחוּצָה׃ or 2 Kgs 13:13–20). On this phenomenon, see both Randall Buth’s 1994 article and John Cook's 2004 article:

Buth, Randall. 1994. Methodological Collision Between Source Criticism and Discourse Analysis: The Problem of "Unmarked Temporal Overlay" and the Pluperfect/Nonsequential wayyiqtol. Pp. 138–54 in Biblical Hebrew and Discourse Linguistics, ed. Robert D. Bergen. Winona Lake, IN: Eisenbrauns.

Cook, John A. 2004. The Semantics of Verbal Pragmatics: Clarifying the Roles of Wayyiqtol and Weqatal in Biblical Hebrew Prose. Journal of Semitic Studies 49/2: 247–73

Moravcsik, Edith A. 2007. What is Universal about Typology? Linguistic Typology 11: 27–41.

 

[Drink]

I think that claiming that perfective verbs prototypically has past time reference is basically the same thing as saying that purely aspectual perfective verbs do not exist (or are rare). The fact that we call them "perfective" is irrelevant. If they have a past time reference, then they have a tense. Simple as that. It would be worth looking into these cross-linguistic perfectives and determine whether they truly are "primarily aspectual" in the first place, and whether it might make more sense to just call them perfect or preterite. This is a huge problem with cross-linguistic analysis: that each language was analyzed by a different person with different methodology.

Also your point about "will" being gnomic seems kind of as using "gnomic" to dismiss the data points. In some languages, the imperfective present tense is used for such meanings, and the future cannot be used. So you can't just call it "gnomic" and exclude it from further analysis. It may be "gnomic", but gnomic is just another modal component of the TAM system.

 

[zaw]

S1234 and Aleppan,

This idea that the participle by itself marks the present tense bothers me. If that is what you two are saying, I have problems with that. The topic is certainly worth returning to.

I want to draw everyone’s attention back to some statements by JAN SHAR:

The case is similar to the English “will”: Is it future or modal? Linguists disagree, but I side with James D. McCawley that since certain statements about the future are judged by people to be “true” or “false,” there must be a non-modal future to be referred to by tensed “will.” Another way to argue the case is that future-time reference always accompanies the use of “will” whereas a modal sense is not always apparent. Similarly, perfective aspect always accompanies the use of qatal, whereas past temporal reference is typical but not exclusive of the form, on which see below.

JAN SHAR put all of that within parentheses, which is what scholars tend to do with their best thoughts. But I still don’t see why it’s wrong to suggest that yiqtol in ancient Hebrew and “will” constructions in English are alike in many ways. Conversely, I do not find it helpful to explain the verbal system of ancient Hebrew by analogy with verbal systems in which aspect is regularly marked, such as those of Russian and ancient Greek.

For the rest, I’m not convinced that qatal נתן in JAN SHAR's Gen 15 example is best analyzed as a present tense performative. I understand there to be past reference: “To your offspring I have assigned this land.” It’s a very interesting case, because the semantics of what is being talked about allows for “staging” along opposite lines. It would have been possible to relate the very same action in future terms; indeed, that is what is done in Gen 12:7 (yiqtol!). The promise could just as well have been related in present performative terms, but that, in ancient Hebrew, would have been with הנה + the participle (cf. Jer 32:3).

JAN SHAR, I agree with you about the importance of cross-linguistic comparisons and I love historical explanations, but not all, to say the least, are especially convincing.

 

[Aleppan]

OK, JAN SHAR,

Here is another example of wayyiqtol not referring to the past:

אָנֹכִי אָנֹכִי הוּא מְנַחֶמְכֶם מִי־אַתְּ וַתִּירְאִי מֵאֱנוֹשׁ יָמוּת

I, I am he who comforts you; why then are you afraid of a mere mortal who must die?

Isa 51:12

This is not gnomic, nor is Ps 45:8, in my understanding. Here it is again

אהבת צדק וַתשנא רשע

Thou lovest righteousness, and hatest wickedness.

Ps 45:8

 

[JAN SHAR]

Hi Aleppan,

Without doing some research on it, my initial reaction to the example is that I agree with you that it is not gnomic. But you will agree that it is one of the very few examples one could dig up of wayyiqtol + stative with non-past temporal reference, and I would be inclined to argue further that you have a distinct possibility of confusion between wayyiqtol and yiqtol: “Who are you that you are afraid” is exactly the context that we would find what Joüon calls an “indirect volitive” yiqtol, is it not? Alternatively, perhaps it is best to follow the LXX, which renders it as having a past temporal reference (aorist indicative): “you were afraid . . .”

 

[Aleppan]

Dear JAN SHAR,

It’s a pleasure discussing with you. I agree with all three of the points you make.

 

[JAN SHAR]

Aleppan,

Couldn’t find some good examples to demonstrate the “inappropriateness” of wayyiqtol in the usual rendering “that you are afraid . . . ,” but leave it to Delitzsch to pull out the relevant examples: Ex 3:11 מִי אָנֹכִי כִּי אֵלֵךְ אֶל־פַּרְעֹה and Judg 9:28 מִי־אֲבִימֶלֶךְ וּמִי־שְׁכֶם כִּי נַעַבְדֶנּוּ . In light of these examples, I would argue even more strongly that in Isa 51:12 וַתִּירְאִי should be amended to a weyiqtol or else interpreted as past with the LXX evidence.

 

[Aleppan]

I wouldn’t say wayyiqtol is inappropriate in Isa 51:12. Several verbal forms may fit one and the same syntactic slot. A question may be followed by a modal clause:

מָה־אֱנוֹשׁ כִּי־תִזְכְּרֶנּוּ Ps 8:5

What is the human being that you should remember him?

Or it may be followed by wayyiqtol:

מָה־אָדָם וַתֵּדָעֵהוּ Ps 144:3

What is the human being that you do know him?

Here of course the past meaning is feasible: “you have taken cognizance of him” (if this is English).

With verbs like ירא and ידע , the “immediate past” and the present are very close to one another.

 

[S1234]

Thank you, JAN SHAR, for trying to deal with the data and explaining yourself. For the record, I too love cross-linguistic explanations. They just need to be done correctly, and they cannot override a specific language’s system. So to substance. You stated:

"S1234, your argument is a non sequitur. Although I claim that qatal-yiqtol form an aspectual opposition, it does not follow that I deny the BH verb the ability to express TIME. Indeed, I am in full agreement with your reasonable claim that the BH verbal system is Tense-Aspect-Mood; to wit, wayyiqtol is past (narrative) tense, qatal-yiqtol opposition is aspectual (perfective-imperfective), and VS ordered qatal and yiqtol, as well as the imperative, are irreal mood. You’ve chosen to ignore a most important datum in Dahl’s observation that prototypically perfective verbs have past TIME reference."

This evades the point that I made about qatal. You have just included time with vayyiqtol, for that I congratulate you, (though you seem to deny that a vayyiqtol / veqatalti opposition exists. more below.) But I was talking about qatal in the previous note, and it is qatal that needs an answer from you first. The מחר evidence suggests that both vayyiqtol and qatal have time features inside them, not just vayyiqtol. (this is only the tip of the iceberg, of course, because most of those ‘conversational’ yiqtol referring to future events appear to default as perfective, more on perfective futures below.)

It is certainly not a ‘non-sequitur’ to say that you deny a time component to qatal. You just did so again in the quote. You did not explain why there is no qatal with מחר but instead accused me of a non-sequitur. Now what do logicians call that? Respectfully, this might be called “evading the question.” But you still haven’t given an answer. (PS: there are some answers, not good ones in my eye, but there are always ‘options.’)

And Dahl’s comments are not relevant to this. Why? Because true perfectives can be also used in future contexts. Far from being confused by ‘past’ and ‘perfective’, some of us are keeping them clearly in view. Note κα γραψω ‘I will write’ modern Greek perfective future. (unambiguously not imperfective future ‘I will be writing,’ which would be κα γραφω). There is no problem with the close correlation of past and perfective in languages around the world. I only have problems with claiming that there is no time in the qatal / yiqtol contrast. That is what the מחר evidence was showing. So, for Dahl’s comments one can only say .מה לי ולו

Finally, far from being ‘intuition’ or ‘pragmatics,’ this is what the language teaches about itself, through actual attestation and usage, which is how everyone learns their own language. Even ancient Hebrew speakers. (Wow, what a novel idea, the language teaches itself.) If qatal / yiqtol were pure aspect markers, then ancient Hebrew would accept מחר באתי*. I just don’t like following a system that mispredicts onesidedly in a 52 to zero fashion, especially when future systems themselves have a leaning to perfectivity. Note again modern Greek where this perfectivity is morphologized, something not very common cross-linguistically in comparison to marking within past systems. But not having perfectivity commonly marked cross-linguistically in future morphology does not rule it out of Greek. Incidentally, ancient Greek grammarians grouped the aorist (simple past and perfective) with the future, aspectually. Makes sense, since both developed σ morphology.

So back to the issue, it is qatal that never occurs with מחר, showing that time is a feature included within qatal, causing it to react with מחר.

And it is veqataltí that is the opposite of vayyiqtol, and that is frequently in complementary distribution with X-yiqtol and both with (vayyiqtol vis-à-vis X-qatal [including with לא ]). (Exodus 25-40 is a classic.) Bickerton’s cake would suggest that if vayyiqtol contained a time feature, then veqataltí would likely have a time feature.

If your system were to freely predict * מחר באתי , then I am afraid that we would be using two different languages. One of us would be building a ‘leaning tower of Pisa’ (warning, here comes some intuition: I still believe that never using a system is a sure way to build a system that doesn’t work. I’ve seen that a lot in Africa, where outside linguists would often try to fit a language to a theory. There are even quite a few ‘aspect’ languages cited in cross-linguistic studies, that on closer inspection were not ‘aspect-only’, just under-differentiated “Bickertonian cakes” that outsiders needed to label as non-Indo-European).

And for something really fun for you to shoot at, in most contexts where yiqtol has future reference the situation covers the ‘whole event’ and would receive perfective marking in an aspectually sensitive language like Greek. Greek is very sensitive to aspect marking, Hebrew is much less sensitive to aspect. Now I will admit that these last observations are interpretations, mappings of situation to form, but they are an iceberg in size.

And because of the iceberg size of this, I would agree with zaw, enough of ‘aspect-only’ explanations of qatal-yiqtol. In old-fashioned metalanguage: they were wrong.

When we understand how biblical Hebrew works we can turn to its typology cross-linguistically and explain where it fits, where it doesn’t fit, and pose reasons why. That is really a lot of fun. You will remember that even Bhat recognizes mixed-scales within typologies.

 

[zaw]

S1234,

I think your point about complementary distributions is very important.

Like you, I try to understand a system from the inside out. I have my doubts about cross-linguistic explanations most of the time, but I find cross-linguistic analogies - and dis-analogies - very helpful. The contrast between ancient Hebrew and ancient Greek when it comes to aspectual differentiation, for example, is instructive. But you know and could explain that much better than I.

 

[JAN SHAR]

S1234,

I will make this brief since your argument is mainly aimed at my take on qatal (I’m not going to get into weqatal now). I think we must be talking past each other (most innocent interpretation). First, I have never in print nor elsewhere claimed that BH was an aspect-only language, and yet that is what you seem to be attacking me on. Wayyiqtol is past TENSE, VS qatal and yiqtol express IRREAL MOOD. But, as long as it is still recognized that qatal-yiqtol is the core opposition of the system, that core opposition is aspectual.

Second, I haven’t heard from you a good reason to dismiss Dahl’s observation that past temporal reference is a “secondary feature” of perfective verb forms. This explains perfectly why qatal does not combine with מחר , because past temporal reference is implied by perfective verbs in many languages, including, I would say, Hebrew.

Third, if you are going to continue to simply throw out the מחר case, then I should just continue to throw at you all those instances of ידעתי ‘I know’ and the like. Why are they not all past tense if qatal is marking past tense?

Thanks again for a great exchange. Always stimulating!

 

[S1234]

JAN SHAR,

You didn’t answer the מחר data, for the umpteenth time. Why don’t you just admit that the Hebrew qatal-yiqtol is a mixed tense-aspect?

Once you accept a mixed tense aspect then ידעתי is not a problem. When focusing on time one says יודע / יודעת , when focusing on decisiveness one says ידעתי ; if you want it Englished, ‘I have realized,’ ‘I fully know,’ respectively. Yes this is a classic ‘perfect’ something-completed-in-the-past with present results. In a binary system that is frequently encoded in the “past.” (Note quotation marks, I do not claim that qatal is a pure past. It’s only the straightjacket people, to paraphrase Bickerton, that would do that. Yet sometime I hear you trying to misread me that way!).

And for Dahl, I do recognise that perfectives generally line up with pasts in binary verb systems. there is no point of argument here. But if, in zaw's examples, the perfective is chosen because the context is PAST and the imperfective is chosen because the context is future, not because of marking the aspectual view of the wholeness or completeness of the event, then that is temporal, then there is a time feature within the “tense / aspect”. Östen Dahl might be one of those that Derek Bickerton would classify with those who have retreated from the real data to pastel overlays.

All of this stems from the lack of a good linguistic term for a basic binary opposition in a TAM system. A lot of linguists use ‘aspect’ for that, but sloppily, and forgetting what they are talking about, so that they proceed to misapply to real data and real languages. theoretically, one could propose ‘aspect’ for this “tense + aspect” term, as long as it was remembered that a time feature may be included in such an “aspect.” But that would ruin the term ‘aspect’ as a pure parameter term for TAM, and another term would need to be developed for true ‘aspect.’

So will you come out and admit that the מחר data, as well as confirming contextual data, shows that the Hebrew qatal-yiqtol is a tense-aspect (or aspect-tense, it really doesn’t make any difference)?

 

[JAN SHAR]

Now, when it comes to discussing the verb with you, S1234, I find us falling again and again into arguments over nomenclature. But beneath this I think there lies a fundamental difference of approach—one of discourse-functional versus semantic.

For the record (and in keeping with all that I have previously said and written), the BH verbal system is a tense-aspect-mood (TAM) system in two regards:

(1) It can indicate a full range of notional meanings traditionally categorized under TAM, such as temporal location of a situation (tense), temporal constituency of a situation (aspect), and the role a speaker wants a situation to play in the discourse (mood / modality—just to use Bybee’s definition; others are possible);

(2) The verbal forms in the system are morphologically marked for tense (e.g., wayyiqtol is past tense), aspect (e.g., qatal is perfective aspect and yiqtol is imperfective), and mood/modality (imperative and jussive are deontic mood, VS qatal and yiqtol are more generally irreal mood).

And sure, I admit that the מחר data show that the BH verbal system can indicate tense. I never denied that. But this is where you want to end the inquiry, with the notion that the BH verbal system can functionally express tense or aspect, depending on context and speaker strategy. But this leaves unanswered how we know which is being indicated—tense or aspect? If we answer “context” then it threatens to become a viciously circular argument, as are Weinrich’s discourse approach (background to Schneider, Talstra, and Niccacci) and Longacre’s model (Hatav has criticized him of being circular).

Weinrich, for instance, argues that the verb forms in European languages served to indicate discourse type, but how do we know which discourse type they indicate unless we have already determined the discourse type independently of the verb forms, in which case, what possible reason could there be for signaling the discourse type with the verb forms? Thus, the verb form is stripped of all semantic significance, and we devolve into arguments like Baayen’s (https://www.sfs.uni-tuebingen.de/~hbaayen/BaayenCV10-14.pdf), that qatal has NO semantics, but signals a disconnect between discourse entities.

Similarly, note your statements above:

the perfective is chosen because the context is PAST and the imperfective is chosen because the context is future, not because of marking the aspectual view of the wholeness or completeness of the event,

Here you actually admit my position by labeling the forms PERFECTIVE and IMPERFECTIVE. But I can agree with your statement only halfway, because you don’t see any meaningful connection between the choice of the perfective verb for past temporal reference (and similarly the imperfective for future). If there is no connection, then why have different verb forms? They must contribute something to the expression; they are not just arbitrarily chosen!

Such discourse-functional arguments violate Frege’s principle of compositionality (Compositionality (Stanford Encyclopedia of Philosophy)), whereby we are required to ask what the individual verb forms contribute to the utterance. I wholeheartedly agree with Fleischmann:

"The pragmatic functions of tense-aspect categories in narrative are not arbitrary; rather, I see them as motivated extensions of the meanings of those categories, extensions that, according to the view of grammar as ‘emergent’ may ultimately contribute to a reshaping of the basic meanings.” (1990: 23)

Thus, the way I see it (and no doubt you will correct my perception; but for the sake of those “listening” to this exchange . . .), you are content to say that the verb forms express a range of TAM meanings depending on context, discourse type, speaker strategy, etc. (call it what you like), whereas I am interested in getting beyond this “arbitrary” assignment of discourse function to the verb forms by explaining how the contribution of the semantics of the verb conjugations themselves to the variety of TAM expressions they appear in.

Thus, I argue that the qatal conjugation is marked for perfective aspect, because that identification is the most coherent and comprehensive explanation of the data: Why does qatal indicate past temporal reference regularly? Because it is perfective aspect, which denotes an event as undifferentiated entity, typical of narrating past events. Thus it has a implied meaning of past. But, given that qatal also expresses present temporal reference, such as with stative lexemes, performative expressions, etc., it is less problematic to explain its contribution to each utterance as being perfective aspect than to say it contributes past tense in one instance and perfective aspect in another. This latter judgment just seems sloppy and incomplete to me. We can understand the language better!

Fleischman, Suzanne. 1990. Tense and Narrativity. Austin, TX: University of Texas Press.

 

[S1234]

JAN SHAR said:

This explains perfectly why qatal does not combine with מחר , because past temporal reference is implied by perfective verbs in many languages, including, I would say, Hebrew.

I can read this as meaning that Hebrew “perfective” is in fact a TENSE-ASPECT, or ASPECT-TENSE if you wish. But if you say that, then it would miscommunicate to claim that Hebrew qatal-yiqtol has no time component within it. Your Dahlian definition of “aspect” is already including time within the aspect. (I am not sure if Dahl says or demands that, one of the problems of quoting outside ‘authorities’). But in any case, your definition now includes time with “aspect.”

If you would simply highlight that, would make it transparent, there would not be a problem. Since the discussion proceeds as though there is a difference, I end up assuming that you don’t really use or include the time component when you yourself say ‘aspect.’ And as mentioned, if “aspect” includes time, then there are two terms out there, creating confusion. I would rather keep aspect for what it is, and then be precise, so that we don’t have perfective futures marked as “imperfective.” That really is a non-sequitur, or an inversion of the ‘aspect’ term through a hidden re-definition that uses ‘future’ inside itself in order for the “imperfective [sic]” to be used for a perfective without implying imperfectiveness. Weird. Linguistics is supposed to take us beyond such double-speak.

And then why in the world would anyone object to a “tense-aspect” or an “aspect-tense”? That keeps things transparent in the term and helps prevent nonsense like students who might say “it’s an aspect, so there is NO time involved.”

 

[zaw]

S1234 and JAN SHAR,

Perhaps this is what it boils down to:

For you, JAN SHAR, the qatal / yiqtol contrast is, fundamentally, one of aspect, but past temporal reference is implied by the use of a perfective form, the qatal, in specific cases;

For you, S1234, the qatal / yiqtol contrast marks, fundamentally, both time and aspect; in specific instances, the aspectual dimension is suppressed.

I think you two agree on quite a bit.

Meanwhile, though the three of us pooped out in this discussion long before the two of you, Sharjeel72, Aleppan, and zaw continue to think that yiqtol is future / modal, and aspect-neutral.

 

[S1234]

zaw,

Nice summary.

I am very happy with Authors C, E, and B’s aspect-neutral future modal, that is exactly what I see for yiqtol and ve-qatal outside of past contexts. Within past contexts I see both habitual and incomplete examples (even supported by etymological predictions based on comparison with Arabic), so I use the ‘aspect’ word rather than ‘modal’. Randall Buth's tense-aspect is short for ‘tense-aspect-mood,’ and he discusses mood separately under specifically modal morphology. And a lot of that chapter is taken up with explaining how qatal~yiqtol can be used for mood. However, the summary glosses over the original incongruity, the Dahl-John Cookian “perfective” that implies itself into a past context but is NOT used in independent future sentences to mark perfectivity.

JAN SHAR, I am grateful for the stimulation to engage and clarify positions. Hopefully, things clarified will not retreat. You said:

Thus, I argue that the qatal conjugation is marked for perfective aspect, because that identification is the most coherent and comprehensive explanation of the data: Why does qatal indicate past temporal reference regularly, because it is perfective aspect, which denotes an event as undifferentiated entity, typical of narrating past events. Thus it has a implied meaning of past. But, given that qatal also expresses present temporal reference, such as with stative lexemes, performative expressions, etc.

It still appears that you are using ‘implied’ in order to deny actual existence, and once again have left out the fact that such a view would allow (‘would predict’ in some linguistic circles) a qatal to mark perfective futures. The מחר data is only the tip of thisiceberg. When we have a false prediction, we have a less-satisfactory theory. You would counter that Dahl would allow you to use ‘imperfectives’ for future perfectives, so the ‘perfective’ is not necessary. But if you did that, then you would have a time-based, AD-HOC footnote, and you surely don’t want to press me to use the ‘A’-word. [For Semitists: AD-HOC is very strong language within generative linguistics ]

As for “sloppy”, we can take it up with the ancient speakers, or most any binary tense-aspect-mood system in the world. All languages have points of weakness and ambiguity, which is why they continually change. Theory must allow for what’s there and then generate neither too little nor too much.

“Incomplete” would belong to a “perfective” theory with false predictions and to one that would seem to deny a synchronic vayyiqtol~veqatal dichotomy, too. (I am aware that etymologically and comparatively the dichotomy is a neo-structure. But etymology is NOT semantics. And if one allows some time in vayyiqtol, and if veqatal functions in a dichotomous relationship with vayyiqtol, then one has just added time to veqatal, too.)

“Better” should require a comprehensive semantics (yes, semantics, not pragmatics, I agree with ‘compositionality’) in regard to the data. “Better” would then be able to eat cake (Bickertonian), and have it, too.

So I see our disagreement as based on internal dynamics of attested data and structural oppositions (with both qatal~yiqtol and vayyiqtol~veqatal having a binary TAM with tense a feature) versus a particular application of external theory. External theories are not monolithic and can be applied differently, but one cannot change the language.

 

[Ali Smith]

6) So how then can a case be made for perfective aspect at all? It would appear from Dahl’s observation that all perfective forms could easily enough be treated as past tense and then we would eliminate another one of those pesky TAM categories. My response is that there are two types of evidence germane to BH, and which are derived from cross-linguistic analysis and not intuitive interpretation:

a) The first is that the qatal form is regularly used for present-time reference performative statements, such as Gen 15:18: לְזַרְעֲךָ נָתַתִּי אֶת־הָאָרֶץ הַזּאֹת . This is directly contradictory to the notion that the qatal is past tense, whereas it makes sense with a theory that recognizes that qatal is perfective aspect that typically has past-time reference but in cases such as this can have a present-time reference.

But is it not possible that it is not a performative utterance? If so, it could be translated using the past tense:

בַּיּ֣וֹם הַה֗וּא כָּרַ֧ת יְהֹוָ֛ה אֶת־אַבְרָ֖ם בְּרִ֣ית לֵאמֹ֑ר לְזַרְעֲךָ֗ נָתַ֙תִּי֙ אֶת־הָאָ֣רֶץ הַזֹּ֔את מִנְּהַ֣ר מִצְרַ֔יִם עַד־הַנָּהָ֥ר הַגָּדֹ֖ל נְהַר־פְּרָֽת׃
(בראשית טו יח)

On that day Y----- made a treaty with Abram saying, "To your offspring I have given this earth from the river of Egypt until the big river on the mountain of Euphrates.

 

[JAN SHAR]

No, it is clearly a performative utterance.

 

[Drink]

Nothing's ever so clear in an ancient language that has no native speakers. There are several plausible ways to interpret this. Even if one seems more likely than others, it certainly can't be said to be "clear".

 

[JAN SHAR]

The author of the 52-to-nothing hypothesis (who says that Hebrew verbs show tense first and mood/aspect/whatever later) may have neglected to take into consideration that the SC forms that he has counted are, I assume, all non-waw-consecutive, while the waw-consecutive+PC would by definition be preceded by the conjunction—whether that different morpho-syntactic distribution has anything to do with the conclusion that he reached I do not know, but it’s something to keep in mind. I seem to remember the few so-called “future perfects” that show up in Numb 19 being free forms and I don’t remember right off hand any waw-consecutive + PC forms having that function.

 

[Drink]

Note that past-in-the-future and future-in-the-past is still tense and not aspect. Either one can be either perfective or imperfective. Tense is the placement of the action in time, aspect is the continuity of the action or lack thereof, regardless of when the action is placed in time.

Looking at Numbers 19:2 (which is what I assume you were referring to, but correct be if I'm wrong), I find it interesting that English, which is not normally considered to be "aspectual" (which is a pretty meaningless classification, since English marks both tense and aspect), would also use a past tense verb (such as the past perfect) in these past-in-the-future sentences, and not the future perfect.