LINGUIST List 28.2171

Tue May 09 2017

Review: Hebrew, Ancient; Discourse Analysis; Ling & Lit; Translation: Longacre, Bowling (2015)

Editor for this issue: Clare Harshey <clarelinguistlist.org>


Date: 08-Oct-2016
From: Geoffrey Sampson <sampsoncantab.net>
Subject: Understanding Biblical Hebrew Verb Forms
E-mail this message to a friend

Discuss this message


Book announced at http://linguistlist.org/issues/27/27-2349.html

AUTHOR: Robert E. Longacre
AUTHOR: Andrew C. Bowling
TITLE: Understanding Biblical Hebrew Verb Forms
SUBTITLE: Distribution and Function across Genres
SERIES TITLE: Publications in Linguistics 151
PUBLISHER: SIL International Publications
YEAR: 2015

REVIEWER: Geoffrey Sampson, University of Sussex

SUMMARY

In their book “Understanding Biblical Hebrew Verb Forms: Distribution and Function Across Genres”, Robert Longacre and Andrew Bowling aim to solve a large puzzle about the grammar of Biblical Hebrew, the language of all but one of the collection of books which is known to Jews as Tanakh and to Christians as the Old Testament. Briefly, Longacre and Bowling argue that a mystery about Hebrew verb forms, which has traditionally been addressed purely in terms of meaning and grammar, instead becomes tractable if discussed in terms of literary genre.

For speakers of European languages who study Biblical Hebrew, the grammar of that language in fact poses two interrelated puzzles, and Longacre and Bowling see genre as the key to both. What they have done can only be understood by reference to these puzzles, and I must ask readers to excuse me if my explanation takes more space than a book review would ideally consume.

First, verbs are conjugated in two ways, sometimes called the suffix and the prefix conjugations. For instance, the root of the verb for “keep” is sh-m-r (verb roots consist of consonants only, with the vowels varying depending on morphological context); the first person singular is -ti as a suffix and e- as a prefix, so we find the forms shamárti and eshmor in the two conjugations. The puzzle is about the meaning of the conjugations. Traditionally, scholars interpreted the system in terms of abstract logic and European grammatical categories, by calling suffix and prefix conjugations “past (or preterite) tense” and “future tense” respectively, so that shamárti was translated as “I kept” and eshmor as “I shall keep”. The obvious question about lack of a present tense was answered by pointing out that a participle, e.g. shomer “keeping”, could be used predicatively with present-tense meaning. The textbook from which I first began to learn Biblical Hebrew, published in 1955, presented the verb system in those terms.

Traditional grammarians recognized that not all uses of the Hebrew conjugations referred in practice to past or future time. In English “He flies to London tomorrow” is present tense but future in reference, and the first verb of “If you wanted, we could stay another night” is past tense but expresses a hypothetical present or future state. But mismatches between the Hebrew conjugation contrast and the contrast between past and future time are more pervasive than in English. Wilhelm Gesenius’s standard reference grammar of Hebrew (first published in 1813; for English version see Gesenius 1910) listed the diverse uses of each conjugation, but Longacre and Bowling see this as muddling together central and marginal uses, and failing to identify a consistent logic uniting the various uses of a given conjugation.

As the insights of structural linguistics became more widespread, scholars began to explain the Hebrew conjugations in terms of a wide range of tense, aspect, and modality (TAM) contrasts. (Some linguists call the suffix and prefix conjugations “perfective” and “imperfective” respectively; in order not to prejudge the issue, my glosses will represent them as Suff and Pref.) For instance, John Sawyer (1976: 82) says that Suff and Pref can express: past v. non-past time; single completed action v. habitual, continuous, or repeated action; or a fact (non-modal) v. a conceptual idea not necessarily realized in fact (modal). Sawyer says that “any one of these oppositions may receive the main emphasis, overshadowing or even eliminating the others”, adding that the context of a particular verb instance “normally makes the meaning clear”. But the latter remark sounds optimistic, and I believe few would claim that we can regularly predict which conjugation would be used in any given context. Gideon Goldenberg (who before his recent death might have been considered the doyen of Hebrew linguistics, being emeritus professor of linguistics at the Hebrew University of Jerusalem) commented (2013: 203) that “a constant flux of publications on Biblical Hebrew tenses continues, with repeated discussions of the question whether they indicate time or aspect.” Whether or not Longacre and Bowling’s approach is right, the conjugation puzzle certainly has not been definitively solved by their predecessors.

The second puzzle concerns the form for “and”, which is a prefix, w-. At least in the earlier books of the Old Testament, Hebrew is a verb-first language, so a co-ordinated clause will commonly prefix w- to the verb of the clause. The puzzle here is that the meaning of either verb conjugation is radically changed by adding w-. In his standard Hebrew textbook, Jacob Weingreen (1959: 90–1) gives an (invented rather than Biblical) example. (In quoting it I shall adjust one word, without affecting the point of the example.) As separate sentences, “The king kept the word of the Lord”, and “He judged the people in righteousness”, would respectively run:

shamar hammélek et-d’bar-JHWH
Suff.3ms.keep the.king Obj-word-JHWH

and:

shapath et-hagham bcédeq
Suff.3ms.judge Obj-the.people in.righteousness

But the co-ordinate sentence “The king kept the word of the Lord and he judged the people in righteousness” would be:

shamar hammélek et-d’bar-JHWH wajjishpoth et-hagham bcédeq
Suff.3ms.keep … w+Pref.3ms.judge Obj-the.people in.righteousness

And similarly, if the English were “The king will keep the word of the Lord and he will judge …”, the verb-forms would be Pref.3ms.keep but w+Suff.3ms.judge.

(In this review, Hebrew is transcribed in a broadly phonemic system, see www.grsampson.net/SHebTrans.pdf, which eliminates most of the diacritics and special symbols needed to reflect the subphonemic detail included in Masoretic script. ‘JHWH’ represents the name of God, which is forbidden to be uttered and hence is not given its own vowels in writing.)

The name of the Hebrew letter < w > is waw, and this use of it was called Waw Conversive, implying that the prefix caused the two conjugations to exchange their meanings. (The more usual present-day term is Waw Consecutive.) To many linguists, including Weingreen, it has seemed extraordinary that prefixing “and” could have such an effect. Some scholars, such as Kennett (1901), have attempted to explain the phenomenon as natural and logical, but I find Kennett’s explanation quite unpersuasive. The famous Hebraist Sir Godfrey Driver explained it in another way as a consequence of the alleged fact that Biblical Hebrew grammar had emerged from a mixture of two different Semitic languages (Weingreen 1959: 252–3). But, apart from the question whether mixture of languages could plausibly lead to such an effect in a single language, the idea of language mixture depends on beliefs about the origin of the Israelites which are not now taken very seriously (Rainey 2008).

Longacre and Bowling’s first step is, in effect, to eliminate the second puzzle by saying that Pref, Suff, w+Pref, and w+Suff should be seen as four conjugations, rather than two conjugations either of which can be preceded by “and”. That might sound like sweeping a problem under the table instead of solving it, but there are considerations making it reasonable (and Longacre and Bowling are not the only scholars to take this line). For one thing, it may be (Hetzron 1987) that the w- of w+Pref, at least, did not originate etymologically as the “and” morpheme but as a different element, which came to be reanalysed as “and” after it was reduced to the single consonant /w/ – and the w+Suff conjugation might then have been created by analogy. Also, the w+ forms can occur where “and” would seem out of place. Several Old Testament books begin with such a form, for instance Numbers opens with the words:

Waj’dabber JHWH el-Moshe bmidbar Xinaj …
w+Pref.3ms.speak JHWH to-Moses in.desert Sinai

The Authorized Version renders this as “And the LORD spake unto Moses in the wilderness of Sinai”. To those of us who are Christians, long exposure to such language might deafen us to the oddity of beginning a document with “And”; but it surely is odd, and some newer, linguistically-sophisticated translations omit the “And”. The range of meanings associated with w+Suff is not a perfect match for the Pref range, and vice versa. In some cases one can distinguish instances of w+verb-form where w- really does mean “and” from those where it merely identifies a conjugation, because the verb-forms differ in details of stress and vocalism, though this distinction is not maintained consistently.

Analysing conjugations as a four-member rather than two-member system still leaves us with the problem of defining the meanings of the conjugations. This is where Longacre and Bowling make their main contribution. The central idea of their book is that choice of conjugation for a verb is determined not (or not mainly) by the meaning of the individual sentence but by the genre of the discourse in which the sentence occurs. For a discourse in a given genre, many verbs will develop the “main line” of that discourse, and will be in the particular conjugation appropriate to that genre, though there may also be verbs in another conjugation to mark some special function (e.g., in the “narrative” genre, to express background information rather than material that moves the story forward). Longacre and Bowling explicitly reject the idea that a conjugation has a constant meaning independent of the genre in which an instance occurs.

Longacre has for many years been writing about discourse genre as a factor in grammar (in various languages). He is not the only scholar to have invoked it in the attempt to explain Hebrew conjugations: Alviero Niccacci (e.g. 1994: 119ff.) makes a similar move. But for Niccacci there is just one relevant genre contrast, between “historical narrative” and “direct speech”. Longacre and Bowling recognize a total of nine genres. For each genre there is a conjugation appropriate to “mainline” clauses, and other conjugations appropriate to various types of non-mainline clause. Thus mainline verbs in “narrative” genre take the w+Pref form; in “exposition” they either take the participle form or the clause is verbless; in “complaints” and “laments” mainline verbs are in Suff form; and so forth.

The bulk of Longacre and Bowling’s book works out the details of this system, and illustrates it, via successive chapters on each of the genres they recognize, giving abundant Old Testament quotations to show how they apply their scheme to the respective genres. They make things easier for the reader by colour-coding the conjugations in their Hebrew quotations.

The primary concern of the Summer Institute of Linguistics, as it used to be called (now “SIL International”), with which Longacre is affiliated and which published the book, is to support evangelism by translating the scriptures. Consequently it is natural that much here is about how various uses of the Hebrew conjugations would most sensitively be rendered into English. Linguists will be more interested in the nature of the conjugation contrasts within Hebrew itself than in translation problems. But the emphasis on translation does not get in the way of explaining Longacre and Bowling’s ideas about the way Hebrew verb forms work.

EVALUATION

It seems fair to regard this book as the culmination of Robert Longacre’s career. It displays a remarkable depth of familiarity with the Hebrew scriptures, and it would be impossible for anyone interested in that body of writing to read the book without encountering new insights.

Nevertheless, there is a fundamental problem about the Longacre approach, which I am not the first to raise. Let me illustrate it via an admittedly caricatured analogy. English expresses futurity using modal verbs, but some European languages, such as French, have a special tense – to avoid question-begging, let me call it the ‘sera’ tense. One fairly distinctive prose genre is weather forecasts. I expect it is true (though I have not checked) that verbs in French weather forecasts tend to be in the ‘sera’ tense. If so, one could imagine a linguistic analysis in the Longacre style explaining that the ‘sera’ tense should be defined as the tense appropriate to mainline verbs within the weather-forecast genre. But that would miss the point, surely? Weather forecasts do not make heavy use of the ‘sera’ tense just because they are weather forecasts. They do so because the job of a weather forecast is to describe things that are going to happen after the time when the forecast is written, and the function of the ‘sera’ tense is to talk about the future. ‘Sera’ is the right tense for weather forecasts because it is the future tense – it is not called “future tense” because it occurs a lot in weather forecasts.

Similarly, it seems to me that if the Hebrew Suff form is regularly found in mainline clauses of complaints and laments (for instance), we will hope to identify something about the meaning of the Suff conjugation which makes it the natural choice for that genre. Just defining Suff as “the complaint and lament conjugation” (and similarly for the other Hebrew conjugations) would not feel like a solution to the puzzle we began with.

Perhaps I am being too demanding here. After all, this language has been studied about as intensively as any, yet if Longacre and Bowling’s solution to the puzzle is not accepted, it appears that no-one has a clearly better solution. Let me give another analogy, more favourable to these authors. French has two perfect tenses, the passé simple and the passé composé – for instance, with the verb for “go”, ‘alla’ versus ‘est allé’. The passé simple is used in writing, the passé composé in speech, and writing versus speech could well be described as a genre contrast. If that is a satisfactory account of the French tenses, why should we ask for more than Longacre and Bowling offer in the case of Hebrew conjugations?

Certainly I could not claim that there exists any meaning difference between ‘alla’ and ‘est allé’, let alone a meaning difference from which it follows that one form is appropriate for writing and the other for speech. But still it seems that a linguist will want to get behind the differential distribution of the two forms to discover why they should be distributed differently. And in the French case one can. As I understand it, the passé simple was an older form which fell out of use in speech and was replaced by the passé composé, but writing conventions were conservative (as they often are): speakers who are educated enough to learn to write respectably also learn to use the conjugation which is now obsolete in the spoken language. That account of the two French tenses does not refer to meaning, but it does explain the difference between the tenses in a way that I am not sure Longacre and Bowling’s exposition achieves for the Hebrew case. One can see that speech and writing are such different activities that it is only to be expected that conventions might differ between them, but it is not obvious why, for instance, lamenting would naturally involve different grammar from narrating, and so on for seven other genres.

Looking at the issue “from the other end”, consider the opening words of Psalm 23:

JHWH roghi lo’ echxar
JHWH my.shepherd not Pref.1s.lack
“The Lord is my shepherd, I shall not want”

The first clause is verbless, the second has a verb in the Pref form. Longacre and Bowling say that the psalm exemplifies the genre “exposition with lyric overlay”. Verbless clauses are their mainline form for exposition, which covers the first clause, and it seems that the “lyric overlay” concept allows them to avoid treating the second clause as subordinate or exceptional (I do not find this part of their discussion entirely clear). But is the reference to genre not a hindrance rather than help to understanding here? The first clause asserts that two entities are the same, and it seems to me that verbless clauses are the usual Hebrew means of doing that, irrespective of genre. The second clause describes a future state of affairs, and we have seen that “future” has often been taken as a central meaning of the Pref conjugation. What does “exposition with lyric overlay” buy us?

Furthermore, Goldenberg (loc. cit.) explains part of our puzzle in a way that overlaps to some extent with Longacre and Bowling’s account, but without invoking genre. He tells us that some languages, particularly languages of the Niger–Congo family, have a specific “narrative tense”, used in recounting a sequence of events where it is not necessary for separate verbs to carry independent indications of time, because it is understood that successive verbs each express “what happened next”. In these languages, narrative-tense forms often include material suggesting that the verb is subordinated or linked to the preceding verb. Goldenberg sees w+Pref as a narrative tense in Hebrew (he does not suggest it originated through contacts with Niger–Congo languages), and he points to the w+ as a similar “linking element”, originally “and” but by Biblical times grammaticalized as part of the narrative conjugation. (If Goldenberg is right about this, presumably Hetzron’s idea about the etymology of w+Pref would have to be wrong.) For Goldenberg it is not that a given stretch of prose contains a lot of w+Pref forms because it is a narrative: it is a narrative because it contains a lot of clauses telling “what happened next”.

Are these merely alternative ways of saying essentially the same thing? I believe they are more than that. Consequently, despite the rich detail and deep knowledge manifest in Longacre and Bowling’s discussions of Biblical examples, I am sceptical about whether they can be seen as dissolving our puzzle.

At times I wondered whether the authors themselves are totally convinced. They write at one point that the linguist “is not flying off into the wild blue literary yonder when he insists that the discourse structure of a language needs to enter into the description of its verb system”. The extravagance of the metaphor suggests recognition that the theory may at least verge on the “wild”. But I would rather read a book which acknowledges a degree of uncertainty about an interesting but perhaps implausible idea, than one which implies that only idiots could doubt it. Anyone interested in the linguistics of Biblical Hebrew should certainly read this book and decide for himself where he stands on the authors’ theory.

REFERENCES

Gesenius, Wilhelm. 1910. Gesenius’ Hebrew Grammar, as edited and enlarged by the late E. Kautzsche. Oxford: Clarendon Press.

Goldenberg, Gideon. 2013. Semitic languages: features, structures, relations, processes. Oxford: Oxford University Press.

Hetzron, Robert. 1987. Hebrew. In Bernard Comrie, ed., The world’s major languages, 686–704. London: Croom Helm.

Kennett, R.H. 1901. A short account of the Hebrew tenses. Cambridge: Cambridge University Press.

Niccacci, Alviero. 1994. On the Hebrew verbal system. In Robert D. Bergen, ed., Biblical Hebrew and discourse linguistics, 117–37. Dallas, Texas: Summer Institute of Linguistics.

Rainey, Anson. 2008. Shasu or Habiru: who were the early Israelites? Biblical Archaeology Review 34.51–5.

Sawyer, John F. A. 1976. A modern introduction to Biblical Hebrew. Stocksfield, Northumberland: Oriel Press.

Weingreen, Jacob. 1959. A practical grammar for Classical Hebrew, second edn. Oxford: Oxford University Press.


ABOUT THE REVIEWER

Geoffrey Sampson graduated in Oriental Studies at Cambridge University in 1965, and studied Linguistics and Computer Science as a graduate student at Yale University before teaching at the universities of Oxford, LSE, Lancaster, Leeds, and Sussex. After retiring from his Computing chair at Sussex he spent several years as a research fellow in Linguistics at the University of South Africa. Sampson has published in most areas of Linguistics and on a number of other subjects. His most recent book is a new edition of ''Writing Systems'' (Equinox, 2015).

Page Updated: 09-May-2017