Monday 4 May 2009

Translating a qatal/yiqtol sequence in Ps 24:2 (drawing on Niccacci)

Ps 24:2 reads as follows:

כִּי־הוּא עַל־יַמִּים יְסָדָהּ

וְעַל־נְהָרוֹת יְכוֹנְנֶהָ

We have here a typical example of a qatal // yiqtol sequence, often found in Hebrew poetry, much to the frustration of translators. That qatal refers to the past is clear, but how do we translate the yiqtol (underlined)?

A translation found in some of the older commentators (Briggs, Kittel, Bäthgen) translates it in the present tense:

for he has founded it upon the seas
and upon the rivers he establishes it.
(Bäthgen even saw cosmological significance in the distinction:"Das Perf. יסדה geht auf die Schöpfung, das Imperf. יכוננה auf die Erhaltung.”).

But this isn't the view of most modern scholars. Gesenius opted for an even more unlikely translation: he lists יכוננה as an example of the frequentive use of yiqtol in the past rather than the present (§107.1a; giving the odd translation: "was continually being established").

All English translations, including most modern commentators, translate it in the past ("established").

I'm trying to grasp this from the perspective of Alviero Niccacci's fascinating essay (Eisenbrauns, 2006) which, as far as I can see, is the first attempt to consistently translate the verbs in poetry along the same lines as in prose. It would appear that for Niccacci, yiqtol never refers to the present. The present tense is always referenced by either a verbless or a participle clause. This leaves us with yiqtol for either the future or the past. The future translation is well known (e.g. Ps 6:10: “The Lord has heard my supplication; the Lord will accept my prayer). When the qatal/yiqtol sequence refers to the past,

they signal a shift from main-line, punctual information (qatal) to secondary-line, repeated/habitual/explicatory/descriptive information” (yiqtol), p. 253.
So, in sum, on Niccacci's reading, יְכוֹנְנֶהָ

  1. cannot be present (because yiqtol never refers to the present);
  2. could theoretically be future, except that it is excluded due to the context (i.e. it doesn't make sense; are there formal indicators as well?)
  3. is probably past, and given the semantics in the Psalm probably either communicates either secondary-line descriptive or explicatory information. If explicatory, it wishes to clarify the process by which God founded the earth on the seas; if descriptive, it wishes to represent the event more “graphically,” even to “celebrate” it, giving the event “depth of field” (to uses Niccacci's terms). Here, the yiqtol would have a “relief function.”

The best translation for this that I can think of is the gerundive, which expresses the "dynamicity” of an action, i.e.:

for he has founded it upon the seas
establishing it upon the rivers.

My questions:

  1. Does Niccacci's theory hold water?
  2. Does the gerund do justice to the form?
  3. What's your favourite translation?

By the way: in a footnote, Niccacci notes that a first attempt was made in this direction in his article "Syntactic Analysis of Jonah," which, thankfully, is available online.

16 comments:

Bob MacDonald said...

Hi Phil - I wonder what sorts of opinions you will receive on this question. My own untrained opinion is that you are asking for a linear aspect to temporal ideas that may be completely out of context for the writers of the day.

for he set it on seas
and in torrents established it

'Set' and 'established' are not 'past' but descriptive of an action that represents the power of the subject rather than the finality or completeness of the verb.

I often see such verbs translated as continuing present and subject to several modalities.

I have not read Niccacci - but 'consistent' is a difficult and not necessarily relevant attribute of translations.

I hope you get some more helpful responses - I have seen arguments over verbs extend to reams of email.

John Hobbins said...

Personally, I don't find Niccacci's approach convincing.

Two more plausible explanations come to mind. Post-positive yiqtols in older poetry are, often enough, equivalent to iprus (the preterite; others call this a perfect) in Akkadian. There even seem to be a few cases of prepositive yiqtol in archaic poetry.

Work through Deut 32 and you will see what I mean.

In later classical Hebrew, post-positive yiqtols with a preterite sense are limited to wayyiqtol and 'az yiqtol.

Another possibility, the one I prefer. I don't think that yiqtol is a future tense in the strict sense as Niccacci seems to think.

Qatal has two major uses in ancient Hebrew, it seems to me: (1) for something viewed as completed (and therefore past in terms of tense); (2) for something viewed in its completeness (omnitemporal, equivalent to the gnomic aorist in Greek).

Yiqtol has three major uses: (1) for something viewed as yet to be initiated (and therefore future in terms of tense); (2) for something that is viewed as initiated but still continuing (a "pseudo-present" in terms of tense); (3) for something that is viewed as continuative or iterative in the past - this last is best understood as a modal use (= "would" in English). In fact, yiqtol has other modal uses.

In short, I think the older commentators got it right (minus the superfluous theologizing):

he founded it on the seas,
he establishes it on the rivers.

"Establishes" on this view is like the "historical present" in English. Something that took placed in the past is *narrated* of if it was taking place in the present.

We associate tense switching of this kind with an oral register. "I went over to my Mom's house. Then she says to me, " . . ."

Like enallage, tense-switching is frequent in ancient Hebrew poetry.

Now I've opened up several cans of worms! Don't be afraid to ask questions.

Phil Sumpter said...

Hi Bob,

thanks for your thoughts. You evoke a bunch of questions from my side:

My own untrained opinion is that you are asking for a linear aspect to temporal ideas
I'm not sure how I'm doing that. See my comments to John below. I've "improved" my translation there. It's central to Niccacci's approach that poetry, in contrast to prose, is segmented and not linear.

'Set' and 'established' are not 'past' but descriptive of an action that represents the power of the subject rather than the finality or completeness of the verb.
Are you saying that there is no temporal reference at all? I can't imagine how that would make sense in context and how the past tense ought to convey that information.


'consistent' is a difficult and not necessarily relevant attribute of translations.
Perhaps, but I mean translation in the sense of "helping us understand the Hebrew," and not in the sense of "what is most appropriate for non-Hebrew readers."

John,

I'd like to thank you for interacting with me like this. I am a newcomer to translation/interpretation and appreciate you taking the time. I've still got a lot to read (theoretically I'm still only at the "text critical" phase of my exegesis; it was the suggestion by BHS to ammend יְכוֹנְנֶֽהָ to כּוֹנֲנָהּ that got me started on this whole thing semantics thing). For now, I'll try and stick with the little I've read of Niccacci and see how far it gets me.

Before I start, may I ask for one clarification? Which of the three meanings of yiqtol are you applying to יְכוֹנְנֶֽהָ? Your use of the "historical present" doesn't seem to fit any of them ... It's not (1) about to happen, neither (2) an action either in progress or in repetition (it's a one off event), nor (3) an action in progress or being repeated in the past (à la Gesenius) ... Bäthgen's talk of Erhaltung (preservation, sustainment) might fit (2), but yours ... ?

In fact, it seems to me that your translation comes a lot closer to what I am suggesting! Despite the different verb forms we are using (they aren't really tenses), the effect is actually very similar.

You opt for the "historical present," which is one use of the so-called "present simple." In his awesome book The English Verb, Michael Lewis tries to uncover the basic semantic distinction of this form of the verb. Lewis prefers to call it "the first form" (as opposed to the "second form," the so-called "past-simple") because ultimately these two forms are not about time at all. Lewis summarises the basic semantic content of this form as follows, concluding with a reference to your "historical present":

"Paradoxically, not only is the present simple not about Present Time, but it is not about time at all. This is a characteristic which it shares with "the past simple." Each of these forms is "simple" in the sense that the speaker sees the events described as single, simple entities, unities, totalities. The "present simple" and "past simple" contrast with each other in that in the first the speaker sees the event as immediate, and with the "past simple" the event is seen as remote. These are not essentially temporal concepts. What matters, for all uses of the simple tense forms, is that the speaker, at the moment of speaking conceptualises the action described as total, complete. As so often ..., we find that a relatively unusual use of a form helps make this clear. In a sequence of correspondence in the newspaper, it would be natural to write:

C.B. Stewart (June 4th) misses the point of my earlier letter ... ... In the correspondence column, misses conceptualises C.B. Stewart's letter as a whole, and clearly wishes to reply to it. ... misses clearly refers to an event in Past Time. ... The essential characteristics of an event described by a "present simple" verb form are:

- the speaker sees the event as a single, total entity. Temporal references are irrelevant.
- the speaker sees the event as a matter of fact.
- the speaker sees the event as immediate, rather than remote (in which case the "past simple" is chosen)." (66; my emphasis)

Key here is that he first form (including the "historical present") expresses what, from the speaker's point of view, is "factual immediacy." It explains the rhetorical use of the historical present in making a story "come alive."

Phew. A long quote.

The (correct) implication for your translation (in contrast to all Bible's known to me and most commentators) is that the "establishing" is given a different flavour to the "founding." God's founding is presented as "factually remote," whereas God's establishing is presented as "factually immediate." This immediacy created by your "historical present" tense is, in my opinion, similar to the gerund I am using.

The "gerund" (a problematic term) is basically the -ing form without "be." Like the so-called "present simple," the -ing form is also often mistakenly called the "present participle," even though it can be used to refer to the past ("she walked out the room smoking") and also "gnomically" ("Smoking is bad for you"). In my opinion, all uses of -ing in English (regardless of whether nominally, adverbially, adjectivally, or verbally) emphasise the dynamic quality of an action, understood as extended in time. In focussing on the dynamic quality of the action, especially when devoid of temporal markers such as the auxiliary "be", the action is made "dynamically immanent." Hence my translation:

he founded it on the seas,
[indeed] establishing it on the rivers.

I add "indeed" in order to (1) eradicate any false sense of temporal succession between the cola and (2) because I get the feeling that the second colon "intensifies" the concepts found in the first (I need to look into that, cf. Alter).

In sum: "establishes" or "establishing" (but not "established" or "was establishing") better gets at the dynamic of the verb form, though I still reckon that -ing is better. The historical present is usually used to tell stories, isn't it? Here there is no story, no narrative description of what happened. It is a vivid celebration of a single event in the past.

Thoughts? I'll be turning my comments here into an independent post.

Anonymous said...

Phil,

Great post. We attempt to briefly answer your three questions.

1. Niccacci's theory holds too much water and it sinks. In IBHS 31.3, Waltke-O'Connor give many examples of yiqtol as present and include discussion on whether or not these yiqtols are really present tense. Are yiqtols always present? Not at all, but to say that not one yiqtol in the whole Hebrew Bible refers to a present tense action is incorrect.

2. It's one option. A simple present wold do, but if you want to belabor the "extended through time" aspect (assuming its there), then it does quite well.

3. Any translation that doesn't use "river" and instead uses something more like "currents (of the 'seas' in line 1)"


After reading your post, I looked over on the bookshelf and realized that Biblical Hebrew in its Northwest Semitic Setting was one of the few hard copies we brought to Africa with us. Re-reading Niccacci's essay now.

D&T

Phil Sumpter said...

Hi D&T,

thanks for your comments! I'd love to hear your response the article after you've read it. I do like it ... though mainly because of a biased propensity to want to treat the Bible as a self-contained unit devoid of diachronic diversity (and yes, I know this is wrong! I spend much of my time posting on why this is wrong).

In response to your answers:

1. I may have misread Niccacci. My comments are based on the table in the article you own. I haven't read his Syntax of the Verb, which is highly respected. I can't imagine he would have just overlooked Waltke and O'Connor, so either I've misunderstood him or he has an answer. What I find particularly interesting is his accounting for Deut 32 (fn. 24), where he reads the yiqtols as in some sense "conditional." What do you think of that!? And, on the basis of the table in the article, have I misread him?

2. I'm not so much interested in the extending through time aspect (I don't think that the verb wants to express establishing as either a recurring event or one long extended one), but on the dynamic quality the verb form seems to lend it (on Niccacci's approach). It's more the dynamism, hence me quoting Niccacci who calls it "celebratory." I'm not sure how to really get that accross in English.

3. I actually agree with you. I was just lazy here and copied and pasted from ESV.

John Hobbins said...

Phil,

I've read a fair bit of Niccacci, both in Italian and in English, with respect to prose, and found him insightful in several ways. I'm just not sure his work on poetry is as helpful.

As I see it, "establishes" is a case of (2) in my rough-and-ready taxonomy: the action is viewed as initiated but still continuing (a "pseudo-present" in terms of tense).

It is a one-off event that occurred in the past, but it is still presented (that is the operative word) as if we are watching it happen as it is ongoing. Does that make sense?

The reason I prefer "establishes" to "establishing" is that the second gives the impression - per Niccacci - a shift from main-line to secondary-line information. But I don't think that is the case at all. On the contrary, there is "intensification" going on, with a G (yasad) - D (konen) sequence. The yiqtol after a qatal may also heighten things, as a historical present does in English. The shift from qatal to yiqtol in Ps 2:1 in that sense is similar (but not otherwise, since in Ps 2:1-2, the tense-switching also expresses a temporal sequence). Another interesting example (but once again, a temporal sequence is expressed): Ps 104:5-9.

For another instance in which similar alternative present themselves, see Ps 29:9. It is possible to translate:

Yahweh's voice convulses the terebinths /
he stripped the forests bare.

No temporal sequence here. Just differences in "stage effects." The yiqtol describes a past event as if it were ongoing. The wayyiqtol goes back a past narration mode, as elsewhere in the psalm.

For the rest, I'm not sure that a sharp distinction can be made between narrative description and celebratory description. The more difficult question, it seems to me, is what to make of the subordination of 24:2 to 24:1. The verbless - and "timeless" - clauses that frame the psalm and occur at intervals correspond to rhetorical peaks: 24:1.6.8.10.

Finally, don't take anything I say too seriously. I mean it seriously, but I'm not confident enough about my approach to recommend it as a be-all and end-all. Better to learn from many teachers on these issues.

Bob MacDonald said...

Phil - thanks for these questions. After 2 years, I feel I have a beginning of a mechanical appreciation for the forms of the verbs but after 62 years I have a life's appreciation of the nature of time. When I said linear, I over-read your question, thinking you were looking for the traditional mathematical linearity of time. Nothing to do with the poetry. I was wrong to think that - your grammar is deep - but the results are one word in a translated poem. That is the tough part. You recognize that in English, time and tense are far more subtle and flexible that a simple line - and you have good names and adjectives (yes I am beginning to like adjectives in some hands) for them.

John in his comments has threaded the needle very well for me. His justifications are precise without being over definitive. Personally I like torrents rather than rivers - but with respect to the verbs, the 'tense' switch gives intensification and makes present the archaic creative moment. That for me is the nature of time - continuous creation in the hand of God. And floods or torrents are what we experience - in them we are established. Present - past - and ongoing. An imperfective completeness.

I am reading my son-in-law's new book on Hazlitt and the history of romantic grammar and spelling books - it is fascinating how words in their forms and rules determine politics also.

Anonymous said...

Phil,

I think it is most likely simply a preterite and that the bi-colon is straight synonymous parallelism. As John mentioned, preterites are very common in the older poetry and it is a common stylistic feature continuing into classical poetry to alternate qatal/yiqtol across a seizura. Thus I would prefer to translate both as simple past. Regardless, I don't see any real semantic or pragmatic distinction being made between the two by the alternation.

I dislike Niccaci's view because he seems to be importing prose discourse features onto poetry, but poetry doesn't work quite the same way. Like John, I don't think a gerund works because it implies that the second cola is background information.

Pete

John Hobbins said...

This is a fun discussion. One thing I am confident of: Pete's proposed spelling of caesura should be immediately be adopted by all serious scholars.

I would much rather call a c-section a seizura.

Anonymous said...

Hee hee. I never had Latin.

Pete

Phil Sumpter said...

Balshanut,

thank you Pete for your comments. You may well be right. I had decided to really dive into this topic and read up on the diachronic/synchronic dimensions, until I remembered that technically I'm supposed to be doing text-criticism, so I'm having to shore up what I've learned so far and get back to it later.

Your position seems to be the mainstream one. Could you tell me why there was a morphological change at all then? Though John said that there was an archaic preterit in older poetry, he also said he preferred the view which saw semantic significance in the use of the forms (which, as far as i can see, puts him in a similar boat to me, contra someone like Adele Berlin, who felt that in examples of "grammatical parallelism" the verbs ceased to have temporal reference).

Would you also argue that this form of parallelism is an indicator for the antiquity of the poem?

I dislike Niccaci's view because he seems to be importing prose discourse features onto poetryNiccacci is consciously drawing on his own work in prose in order to understand poetry. The idea isn't so foreign, though, is it? In an unpublished PhD by Cristian Rata, he cites Hendel by saying:

" as in most other languages, poetry extends and plays on normal linguistic rules and forms." (R. S. Hendel, Margins of the Hebrew Verbal System,” 153.) Therefore, to understand the verb in poetry, it is important to begin with a good understanding of the system in prose, “from which poetry takes its point of departure." (ibid.)

I don't think a gerund works because it implies that the second cola is background information. I added the word "indeed" in order to balance this, making it: "indeed, establishing it upon ocean currents." Does that help? Niccacci also uses this option (p. 258), though he still maintains that "the line of information with x-yiqtol/weqatal does not stand on the same level with the line of information with qatal/wayiqtol, but the former is subservient to the latter -- it specifies it in different ways according to various context situations." (257) I'm not sure if the concept of "second-line information" is always "background" for Niccacci, because the functions he ascribes to it don't sound very background to me (repetition, description, explication etc.; or does "background" just mean "subservient"?). In terms of these functions, I still don't see why we can't say that the second colon is not "second-line". John talked about the second colon intensifying the meaning of the first. I agree, but then wouldn't that make this an example of the second colon clarifying or qualifying or dramatizing the first in some way? It may not be subordinate in terms of the sequence of what happened or in terms of clarification of how it happened, but it is subordinate (Niccacci says "subservient) in terms of dramatizing what went before. In that sense, it is kind of parasitic ... isn't it? In short, I don't get why the second colon can't be subservient. Or am I missing something?

Niccacci gives what I think are a number of similar examples from Psalm 78, which I've now posted here along with my thoughts.

Thanks for popping by!

Phil Sumpter said...

John, (sorry, I'm posting responses in the opposite order in which I answered them!)

Thanks for getting back to me on this, John.

Better to learn from many teachers on these issues.
If only I had them! You're helping me out a lot, so thank you!

Does that make sense?
I get the concept. In fact, as I said in this post, I get the feeling that we are saying similar things. Whereas you talk of "pseudo-present," I talk of "dynamically immanent." You talk of "intensification," which I also mentioned. I felt that my understanding of was supported by that fact. I don't really get how the second colon being "secondary-line information" would contradict that, but then I haven't read enough of Niccacci to know what he really means by that. Your description of the qatal-yiqtol sequence "hightening things," the different "stage effects" it creates, sounds like Niccacci's language of "celebration," "relief function," and "depth effect." Of the various types of information the non-punctual yiqtol can convey (for Niccacci: repeated/habitual/explicatory/descriptive information), your description seems to fit the "descriptive" category (as I expressed in my post). I'm not sure why we seem to agree ... either I'm missing something or I've misunderstood Niccacci or I'm just confused! All are real possibilities

I searched van der Merwe's description of the function of the yiqtol to see where your (our?) definition would fit in his scheme and I couldn't find any one category. It seems that our view represents a mixture of everything on his list apart from the future tense interpretation.

You comments on the structure of Ps 24 really are interesting ... I'd love to go into them but that may divert the direction of the conversation too much ... (e.g. isn't verse 4 timeless as well? And how are 8 and 10 timeless? They are asking a concrete question to a single individual in "time and space"). In terms of my exegesis of Ps 24, I'm still stuck in the text critical phase. Next I'll be dealing with poetic structure so perhaps I'll have more opportunity to interact with you then ...

Anonymous said...

Phil, sorry for the delay. I'm a bit busy.

Could you tell me why there was a morphological change at all then? I assume you mean why use qatal then switch to yiqtol if they mean the same thing. This is one of the great mysteries of Hebrew poetry, and no I can't tell you why. My personal feel is that it is a bit of showing off. Learned people would know that yiqtol can be a simple past tense, heck they probably still made the connection to the wayyiqtol as a simple past tense in narrative. There are enough semantic clues between the cola to show that the poet is describing the same event in the past tense (even John's historical present recognizes that the event time is in the past) so the "author" (whatever that means for a Psalm) feels free to throw in a yiqtol preterite for a little variation. FWIW, I think that the suffix -ha instead of -ena also tips that it is a short-form preterite rather than an "imperfect".
In general I think we have been mislead a bit by structuralism into thinking that every apparent opposition must have some significance. Sometimes people just like to vary the way they say things, especially in poetry.

Now, could there be nuance in the second verse? Sure. But this is the problem with the BH verbal system in general, how do we formally demonstrate that there is nuance beyond the argument "This is how we could translate it in English?". I think it is safer from what we have to assume that it is simple past.

Would you also argue that this form of parallelism is an indicator for the antiquity of the poem?No. A later author can always pull from the earlier language. There is nothing else in the poem that jumps out as early to me as particularly early. It seems like good classical poetry.

Niccacci is consciously drawing on his own work in prose in order to understand poetry. The idea isn't so foreign, though, is it?I certainly agree that the verbal forms aren't magically transformed to tenseless-aspectless "blobs" by virtue of their appearing in poetry so that they can be used where ever you want. There must be some relation between their use in poetry and their use in prose. My real problem with Niccacci (and Schneider/Talstra on whom he builds) is his top-down approach to the verb which puts priority on its discourse function. To me the discourse function flows from the morpho-syntax of the verb combined with the natural contour of narrative/speech, which is usually a linear sequence. In poetry the contour of discourse is different. It is built not on sequence, but repetition and nesting. Therefore, while the morpho-syntax of the verb should be similar, there is no reason to assume that the discourse functions will be.

In short, I don't get why the second colon can't be subservient. Or am I missing something?Well, I can't argue that it can't be subservient, I just think it is safer to see it as straight parallelism.

Phil Sumpter said...

Pete,

I appreciate your feedback here, especially as you have so little time.

I think that the suffix -ha instead of -ena also tips that it is a short-form preterite rather than an "imperfect". That's a very interesting idea, I've not come across that before. Someone recently sent me part of his PhD in which he argues for the presence of a "short-form preterite," though I haven't read it yet. W. Schneider calls the -enah ending a Modus-Endung (modal ending), though I'm not sure what it has to do with modality.

Sometimes people just like to vary the way they say things, especially in poetry.Certainly a possibility. Thanks for the thought. I do wonder about Hebrew though ... in prose words and roots are often repeated in such a way that the English translator feels uncomfortable and feels the need to show a bit of lexical variety in order to ease what to an English ear sounds monotonous.

My real problem with Niccacci (and Schneider/Talstra on whom he builds) is his top-down approach to the verb which puts priority on its discourse functionI wish I could respond to this more intelligently, but unfortunately I'm way too much of a novice. Just some random thoughts that occur to me:

Niccacci wrote an article called "Analysing Biblical Hebrew" (which you no doubt know) in which he affirms what you say: narrative is linear, poetry is segmented. When he turns to the verb in poetry, he does take this into account, at least ostensibly. He says that the function of the verb is similar, as long as one takes into account the segmented nature of poetry. Whether he does justice to the various discourse forms is another question, which I can't answer.

I have to say, I do have a soft-spot for synchronic analysis in the realm of linguistics and am wary of comparison with cognate languages. One example for me is the comparison of the "perfect form" (have+past participle) in English, French and German. All the three languages are geographical neighbours, chronologically co-temporal, and are very closely related. Yet the forms, which are exactly the same, are pretty different when it comes to meaning. The meaning of the present perfect in English can only be gauged by comparing it to all the other forms in the language and seeing how they relate differentially, not by comparing it to German and French and seeing how it functions there.

All the best!

Evedyahu said...

I know I am late to this discussion - I was reading it before reading Niccacci's article. If the form is short (as Balshanut suggests, and it seems to me that he is right). Then it is most definitely a preterite and I would have no problem translating it as a past.

In my opinion, when the yiqtol and qatal are used without difference in translation, the yiqtol is ALWAYS short (even if it cannot be differentiated as is the case with most forms now)! Chris

Phil Sumpter said...

Hi Chris,

thanks for your thoughts, they're always welcome here. Having finished Pete's (Balshanut's) thread, I find myself coming over to your view of things.