reconsidering the so-called prophetic perfect ibr 2017 prophetic...reconsidering the so-called...
TRANSCRIPT
1
Reconsidering the So-Called Prophetic Perfect
Daniel E. Carver
Lancaster Bible College
1.0 Introduction
The future time use of the suffix conjugation (SC) in Biblical Hebrew (BH) has been a
perennial challenge for biblical philologists since the time of the early Hebrew grammarians in
the Medieval Era. Abraham ibn Ezra, for example, observed that the prophets would
customarily use the SC, which he considered a past tense (עבר), to speak of future situations.1
Ibn Ezra, Qimḥi, and others essentially considered the future use of the SC a rhetorical device,
though they did not call it that. They suggested that the past tense was used for future situations
when they were firmly decided or considered to be absolutely certain – so certain that they are
thought of as already done.2 Over the last millennium, many scholars have wonder at the
apparent contradiction of the SC, often presumed to be a past tense, being used to refer to future
situations. However, Heinrich Ewald’s aspectual approach to the BH verbal system along with
Samuel Driver’s popularization of a modified form of Ewald’s approach would come to shape
the way philologists and biblical scholars of the 19th
and 20th
centuries would understand the
future time use of the SC.
*This is a paper presented at the annual meeting of the Institute for Biblical Research, in the Emerging
Scholarship on the Old Testament section, in Boston, Mass., on November 17, 2017.
1 W. Bacher, Abraham ibn Esra als Grammatiker, ein Beitrag zur Geschichte der hebräischen
Sprachwissenschaft (Strassburg: K. J. Trübner, 1882), p. 127 n. 6 “כמשפט הנביאים שידברו פעמים על עתיד בלשון עבר.”
2 Bacher, Abraham ibn Esra, p. 127; W. Chomsky, David Kimhi's Hebrew Grammar (Mikhlol):
Systematically Presented and Critically Annotated (New York, NY: Bloch, 1952), §77b; A. Bayly, A Plain and
Complete Grammar of the Hebrew Language (London: J. Ridley, 1773), p. 40.
2
Ewald’s theory of the BH verbal system began to take shape in the first few editions of
his Hebrew grammar, published in the 1820’s and 30’s. He took a radically different approach
from earlier and contemporary scholars, who all considered tense the primary signification of the
finite verb forms. He argued that the SC referred to situations that were completed (hence the
term perfect), that is “already finished, done, and therefore ... definite and certain.”3 Since the
form did not locate situations in any specific time, it could be used to refer to completed
situations in the past and present, and even to future situations that were by “the intention or the
imagination of the speaker” considered “already as good as done,” and “therefore as perfectly
unconditional and certain.”4
Meanwhile, another scholar was developing a new tensed approach to the BH verbal
system. Samuel Lee claimed that the SC was a past tense, and argued that when the SC referred
to situations that were not past from the speech time (i.e, not absolutely past), they were past
from another point in time (i.e., they were relatively past).5 He posited that the Hebrew authors
could metaphorically transport themselves and their audiences to a future time and describe
events from that point in time. Thus an event that was future from the speech time was past from
the point in time from which the author described it.6 He claimed that this was used in prophetic
utterances for the purpose of connoting certainty and assurance.7
3 G. H. Ewald, A Grammar of the Hebrew Language of the Old Testament (2
nd ed.), J. Nicholson (trans.),
(London: Whittaker and Co., 1836), §261.
4 Ewald, A Grammar of the Hebrew Language, §261.
5 S. Lee, A Grammar of Hebrew Language (2
nd ed.), (London: James Duncan, 1832), p. 332.
6 S. Lee, “Correspondence (Tenses of Hebrew Verbs),” Journal of Sacred Literature, 7 (1851): p. 470.
7 Lee, A Grammar of Hebrew Language, p. 351.
3
Later in the 19th
century, Driver took the aspectual approach of Ewald and added to it the
temporal transporting of Lee to create what is known today as the traditional description of the
prophetic perfect. According to Driver, the Hebrew authors would transport themselves to a
future time, which he called the ideal standpoint, and describe events that were still future and
undone from the speech time (i.e., the real standpoint), as if they were already completed.8
Although Driver’s approach is the traditional one in BH studies and it can be found in
commentaries and grammars from the late 19th
century to today, there has been a lot of
dissatisfaction with the category among scholars over the past century and a half, and especially
the past several decades. Yet it has remained one of the most under-examined categories in the
BH verbal system, as only a handful of studies have been dedicated to closely examining it.9
Perhaps the most obvious problem with the Prophetic Perfect is that there is, to my knowledge,
no parallel to the prophetic perfect as described by Driver in any other language, modern or
ancient.10
Moreover, a recent study by Rogland was able to explain many of the alleged
prophetic perfects with a modern relative tense theory. He cogently argued that many of these
SCs actually refer to situations that are future from the speech time, but are past relative to the
reference time.11
Most notably, this occurs in quoted speech that will take place in the future and
8 S. A. Driver, A Treatise on the Use of the Tenses in Hebrew, and some other syntactical questions (3
rd
ed.), with an Introductory Essay by W. R. Garr (Grand Rapids, MI: Eerdmans, 1998), §14.
9 M. H. Pullin, The Prophetic Perfect in Jeremiah 1-52 (unpublished M.A. thesis), The University of
Chicago, Chicago, IL (1932); G. L. Klein, “The ‘Prophetic Perfect,’” Journal of Northwest Semitic Languages, 16
(1990): pp. 45-60; M. F. Rogland, Alleged Non-past Uses of Qatal in Classical Hebrew, Studia Semitica
Neerlandica (Assen: Royal Van Gorcum, 2003); and to a lesser degree A. Andrason, “Future Values of the Qatal and
Their Conceptual and Diachronic Logic: How to Chain Future Senses of the Qatal to the Core of Its Semantic
Network,” Hebrew Studies, 54 (2013): pp. 7-38.
10 For a review of the alleged cases in ancient languages from the Near East and the East Mediterranean,
see D. E. Carver, A Reconsideration of the Prophetic Perfect in Biblical Hebrew (unpublished Ph.D. dissertation),
The Catholic University of America, Washington, D.C. (2017), pp. 56-60.
11 Rogland, Alleged Non-past Uses of Qatal, pp. 58-62.
4
in reference to dreams or visions that were seen and experienced in the past. One of the most
important conclusions in Rogland’s study is that the so-called prophetic perfect is actually a
collection of other, unrecognized or misunderstood uses of the SC.
In a recent study, I have proposed other linguistically reasoned uses of the SC that
together with Rogland’s relative tense approach explain all of the alleged prophetic perfects in
the prophetic literature. As a result, it is apparent to me that the term prophetic perfect is no
longer of substantive value for describing any use of the SC. In this paper, I briefly present one
of the categories I have proposed. The occupation of this paper is with the irrealis use of the SC
without the conjunction waw.12
In the following sections of this paper, I outline the linguistic
argument I have made in my recent study. The argument begins with the expression of irrealis
mood in BH, and is followed by a discussion of weqatal and the use of WO to indicate irrealis
verbs. This is followed by examples of the irrealis use of the SC in the prophetic literature.
2.0 Irrealis Mood in Biblical Hebrew
BH expresses modality in two ways. First, through a modal system, which involves
lexemes that express modality (e.g., English would, may, can and Hebrew יכל), and second
through mood, which involves what is basically a binary system, realis : irrealis.13
Realis
situations are actual, meaning that they refer to a particular situation or to a situation that is
directly perceived, whereas irrealis situations are not actual, as they may refer to a virtually
infinite number of hypothetical, possible, potential, and contingent situations.14
12
The consonant waw is here representative of the reflexes of the conjunction *wa in BH.
13 Cf. F. R. Palmer, Mood and Modality (2
nd ed.), Cambridge Textbooks in Linguistics (Cambridge:
Cambridge University Press, 2001), pp. 4-5.
14 See Palmer, Mood and Modality, pp. 1-3; J. A. Cook, Time and the Biblical Hebrew Verb, Linguistic
Studies in Ancient West Semitic (Winona Lake, Ind.: Eisenbrauns, 2012), pp. 249-271.
5
Every irrealis situation expressed by a grammatical construction may be categorized as
one of two types. The first is event which deals with potential or “not actualized” events.15
Event modality may be further divided by how the factors relate to the relevant individual. In
expressions of deontic modality, the relevant factors are external (permission, obligation,
commission, etc.), while in expressions of dynamic modality, the relevant factors are internal
(ability, willingness, etc.).16
The second type is propositional which deals with the “speaker’s
attitude to the truth-value or factual status of the proposition.”17
There are two kinds of
propositional modality, evidential and epistemic. While evidential modality expresses situations
that are perceived through the senses, epistemic modality is based on inference from one or many
factors that lead the relevant individual to the proposition. When the factors and the
circumstances surrounding a given situation cause one to have a low level of certainty regarding
the truth of the proposition, the proposition is considered speculative. However, propositions
that are considered to have a high degree of certainty are either assumptive or deductive.18
The
difference between these is that assumptive is drawn from “generally known evidence,” while
deductive is drawn from “observable evidence.”19
As a result, deductive propositions are viewed
with a slightly stronger sense of certainty than assumptive propositions.20
English typically
15
Palmer, Mood and Modality, p. 8.
16 Palmer, Mood and Modality, pp. 7-10.
17 Palmer, Mood and Modality, p. 8.
18 See Palmer, Mood and Modality, pp. 24-31. Cf. J. L. Bybee, R. D. Perkins, & W. Pagliuca, The
Evolution of Grammar: Tense, Aspect, and Modality in the Languages of the World (Chicago, IL: University of
Chicago Press, 1994), pp. 179-180; S. Cantarini, W. Abraham & E. Leiss, “Introduction,” in S. Cantarini, W.
Abraham & E. Leiss (Eds.), Certainty-uncertainty – and the Attitudinal Space in Between, pp. 1-25, Studies in
Language Companion Series, 165 (Amsterdam: John Benjamins Publishing), pp. 4-5.
19 Palmer, Mood and Modality, p. 24.
20 Bybee et al., The Evolution of Grammar, p. 179; Palmer, Mood and Modality, p. 25.
6
distinguishes these three kinds of epistemic modality by using may or might for speculative
propositions, will for assumptive, and must for deductive.21
BH expresses irrealis situations in several ways. In addition to certain modal lexemes
(e.g., יכל) and particles that mark hypothetical and other irrealis clauses (e.g., לו), BH also uses
verb forms to express irrealis situations. The long prefix conjugation (LPC) and the SC (most
frequently in the construction weqatal) are used to express situations with a wide variety of event
and propositional modalities.22
There is also a directive-volitive system, which does express
modality but is not technically part of the mood system of BH.23
The directive-volitive system
includes the jussive (i.e., the short prefix conjugation [SPC]), the imperative, and the so-called
cohortative. Unlike the LPC and SC, these verb forms are almost completely semantically
restricted to directive-volitive values.24
3.0 Weqatal and the Irrealis Use of the Suffix Conjugation
The vast majority of the irrealis uses of the SC occur in the construction weqatal.
Although many scholars have argued that the weqatal construction expresses imperfective
aspect,25
Joosten has strongly argued that every use of the construction (apart from the so-called
weqataltis) can be explained as irrealis. For more than two decades, he has compiled linguistic
21
Palmer, Mood and Modality, p. 25.
22 Cf. J. Joosten, “Biblical Hebrew w
eqāṭal and Syriac hwā qāṭel Expressing Repetition in the Past,”
Zeitschrift fur Althebraistik (1992): p. 13; also, cf. Cook, Time and the Biblical Hebrew Verb, pp. 237-256.
23 Palmer, Mood and Modality, p. 5.
24 Cook, Time and the Biblical Hebrew Verb, p. 255.
25 E.g., H. Bauer, “Die Tempora im Semitischen,” Beitrage zur Assyriologie und semitischen
Sprachwissenschaft, 81 (1910): p. 35; T. D. Andersen, “The Evolution of the Hebrew Verbal System,” Zeitschrift
fur Althebraistik, 13 (2000): pp. 35, 40; D. O. Moomo, “The Imperfective Meaning of Weqatal in Biblical Hebrew,”
Journal of Northwest Semitic Languages, 31/1 (2005): pp. 89-106; A. Van de Sande, Nouvelle perspective sur le
systeme verbal de l’hebreu ancien: Les formes *qatala, *yaqtul et *yaqtulu, Publications de l’institut orientaliste de
Louvain, 57 (Louvain: Institut Orientaliste, 2008), p. 276.
7
evidence from ancient and modern languages and his conclusion is that none of the uses of the
weqatal construction, including the past iterative, are imperfective; rather, they are all irrealis.26
While from a synchronic point of view the weqatal construction clearly expresses irrealis
situations, one might wonder how it came to have these irrealis values. There is no compelling
comparative Semitic evidence to suggest that this construction developed from anything other
than the conjunction *wa and the SC,27
neither of which is inherently modal.28
Cook has cogently argued that the SC acquired modal values through the process of
context-induced reinterpretation.29
He argued that SCs that occurred in modal contexts were
reinterpreted as modal verbs, and once the reinterpretation was accomplished, the SC could be
used in other contexts with a modal value. Although many others had suggested that the
construction weqatal developed from its use in the apodoses of conditional propositions, Cook
added a critical piece of the puzzle arguing that it was the use of the SC in apodoses and protases
that spurred the reanalysis of the SC as an irrealis verb form.30
Recognizing that the SC in
conditional protases was also reanalyzed is very important for two reasons: (1) only in this ways
can on account for variety of modalities that the SC is used to express in BH, and (2) it suggests
that the reanalyzed constituent was the SC and not the construction weqatal.
26
E.g., J. Joosten, “Biblical Hebrew weqāṭal and Syriac hwā qāṭel,” pp. 1-14; J. Joosten, “Do the Finite
Verbal Forms in Biblical Hebrew Express Aspect?” Journal of the Ancient Near Eastern Society, 29 (2002): pp. 53-
68; J. Joosten, The Verbal System of Biblical Hebrew: A New Synthesis Elaborated on the Basis of Classical Prose,
Jerusalem Biblical Studies, 10 (Jerusalem: Simor LTD, 2012), pp. 288-308; see also, Cook, Time and the Biblical
Hebrew Verb, pp. 249-256.
27 Cf. V. J. J. DeCaen, On the Placement and Interpretation of the Verb in Standard Biblical Hebrew Prose
(unpublished Ph.D. dissertation), University of Toronto, Toronto (1995), pp. 122-126.
28 A. Andrason, “The BH Weqatal A Homogenous Form with No Haphazard Functions (Part Two),”
Journal of Northwest Semitic Languages, 38/1 (2012): p. 4.
29 Cook, Time and the Biblical Hebrew Verb, p. 251; cf. B. Heine, U. Claudi, & F. Hunnemeyer,
Grammaticalization: A Conceptual Framework (Chicago: University of Chicago Press, 1991), pp. 71-72.
30 Cook, Time and the Biblical Hebrew Verb, p. 252.
8
4.0 Word Order and Modal Verbs
Over the past several decades, scholars have observed that modal verbs have a strong
tendency to occur in clause-initial position. In 1987, Niccacci argued that the jussive and
indicative uses of yiqtol are distinguished by word order (WO).31
He claimed that, apart from a
handful of syntactic environments, the jussive always occur in clause-initial position. His
conclusions were affirmed in later studies by Revell and Joosten.32
Revell’s study concluded
that “[a] verb can be marked as modal in the corpus by its form (imperative, or short or affixed
imperfect), by its position (initial in its clause) or by co-occurrence (with following נא or
preceding אל).”33
Revell also noted that sometimes volitives (i.e., imperatives and jussives) are
preceded by certain clausal adverbs (e.g., עתה or אך), which are not to be considered part of the
clause (i.e., they are syntactically detached). He considered them connecting particles that mark
“logical transition.”34
Shulman, in her 1996 dissertation, affirmed Revell’s argument in the corpus of Genesis-
Kings. She found that directive-volitive verb forms occur in clause-initial position nearly every
time. Imperatives are clause-initial 1454 times out of 1515 occurrences (96%), non-negated
jussives are clause-initial 96 times out of 102 occurrences (94%), and cohortatives are clause-
initial 192 times out of 197 occurrences (97%).35
All of the non-initial volitives36
follow a
31
A. Niccacci, “A Neglected Point of Hebrew Syntax: Yiqtol and Position in the Sentence,” Liber Annuus,
37 (1987): pp. 7-19.
32 E. J. Revell, “The System of the Verb in Standard Biblical Prose,” Hebrew Union College Annual, 60
(1989) pp. 1-37; J. Joosten, “A Neglected Rule and Its Exceptions: On Non-Volitive yiqtol in Clause-Initial
Position,” in G. Geiger (Ed.), ʼΕν παση γραμματικῇ και σοφια: En pāsē grammatikē kai sopiā: Saggi di linguistica
ebraica in onore di Alviero Niccacci, ofm (Jerusalem: Franciscan Printing Press, 2011), pp. 213-219.
33 Revell, “The System of the Verb in Standard Biblical Prose,” p. 32.
34 Revell, “The System of the Verb in Standard Biblical Prose,” pp. 19-20.
35 A. Shulman, The Use of the Modal Verb Forms in Biblical Hebrew Prose (unpublished Ph.D.
dissertation), University of Toronto, Toronto (1996), pp. 241-248.
9
constituent that was preposed for pragmatic purposes, mostly for focus or topicalization.37
The
data provided by Shulman demonstrates that, apart from pragmatically preposed constituents and
clausal adverbs, modal verbs occur in clause-initial position.38
Holmstedt used the correlation of modal verbs and clause-initial position to explain the
irrealis use of the SC when following conditional particles, such as כי and םא , and in the weqatal
construction.39
Essentially, Holmstedt has argued that weqatal is simply the conjunction we- and
an irrealis SC.40
Furthermore, Kawashima has argued that this construction was synchronically
viewed as an irrealis SC and the conjunction we-, based on the uses of SCs that follow אם ,גם ,או
.אם and ,לא41
He referred to the irrealis use of the SC without the conjunction we- as an
“‘orphaned’ converted tense.” The correlation of clause-initial WO and irrealis mood is also
affirmed by the statistically supported tendency of the realis SC to occur in non-initial position in
unmarked, main clauses.42
It is this very tendency that has led many scholars, who take a text-
linguistic approach to the verbal system, to view the realis SC as an indication of background or
36
Which does not include those that followed a clausal adverb. She followed Revell’s suggestion that
clausal adverbs are syntactically detached.
37 Cf. Shulman, The Use of the Modal Verb Forms in Biblical Hebrew Prose, pp. 241, 246.
38 In the Amarna letters from Canaan, volitive verb forms are always clause-initial, apart from a few cases
in which constituents are preposed; K. J. Baranowski, The Verb in the Amarna Letters from Canaan, Languages of
the Ancient Near East (Winona Lake, IN: Eisenbrauns, 2016), p. 160. This is of particular significance since the
normal Akkadian practice is for all verbs to occur in clause-final position; A. F. Rainey, Canaanite in the Amarna
Tablets: A Linguistic Analysis of the Mixed Dialect Used by the Scribes from Canaan. Morphosyntactic Analysis of
the Particles and Adverbs (Vol. 3), (Leiden: Brill, 1996), p. 272.
39 R. D. Holmstedt, The Relative Clause in Biblical Hebrew: A Linguistic Analysis (unpublished Ph.D.
dissertation), University of Wisconsin – Madison, Madison, WI (2002), pp. 136-139; Cook (Time and the Biblical
Hebrew Verb, pp. 249-250) added the particles לו and לולא to this list.
40 His argument has been further supported by Cook, Time and the Biblical Hebrew Verb, pp. 235-237.
41 R. S. Kawashima, “‘Orphaned’ Converted Tense Forms in Classical Biblical Hebrew Prose,” Journal of
Semitic Studies, 55/1 (2010): pp. 31-34.
42 See R. D. Holmstedt, “The Typological Classification of the Hebrew of Genesis: Subject-Verb or Verb-
Subject,” Journal of Hebrew Scriptures, 11 (2011): p. 10.
10
offline material.43
Ultimately, the evidence strongly suggests that WO is used as a strategy for
disambiguating the realis and irrealis uses of the SC. This WO includes clause-initial position
and instances in which the SC follows a clausal adverb44
or the conjunction we-.
Several studies mentioned above have noted various clausal adverbs that occur before
irrealis verbs, but are syntactically detached. As Moshavi has explained, this means that they are
syntactically required to occur first in the clause, regardless of the normal WO or any constituent
preposing. Moshavi noted that the conjunction we-, clausal adverbs (inter alia לכן ,הנה ,אך ,כי, and
.and negative particles all fit in the category of particles that are syntactically detached ,(על כן45
At this point, it may be helpful to consider a few examples. The examples below
illustrate the use of clausal adverbs and pragmatically preposed constituents in modal clauses.
(Translations of the preposed constituents are in bold).
(a) Jer. 28:7 מע־נא הדבר הזה אך־ש
“Only, hear this word!”
(b) Gen. 19:12 עת לה ו הש ע יך אל ים אלה ר אמ ר אש כ
“And now, do everything that God said to you!”
(c) Ps. 119:37 י ך חינ רכ ד ב
“Keep my life46
in your ways!”
43
E.g., D. M. Gropp, “The Function of the Finite Verb in Classical Biblical Hebrew,” Hebrew Annual
Review, 13 (1991): pp. 45-62; R. Buth, “The Hebrew Verb in Current Discussions,” Journal of Translation and
Textlinguistics, 5/2 (1992): pp. 91-105; P. J. Gentry, “The System of the Finite Verb in Classical Biblical Hebrew,”
Hebrew Studies, 39 (1998): pp. 7-39.
44 In addition to the studies cited above, see Kawashima, “‘Orphaned’ Converted Tense Forms.”
45 A. Moshavi, Word Order in the Biblical Hebrew Finite Clause: A Syntactic and Pragmatic Analysis of
Preposing, Linguistic Studies in Ancient West Semitic, 4, M. O’Connor & C. L. Miller (Eds.), (Winona Lake, IN:
Eisenbrauns, 2010), pp. 68-76; cf. N. P. Lunn, Word-Order Variation in Biblical Hebrew Poetry: Differentiating
Pragmatics and Poetics, Paternoster Biblical Monographs (Eugene, OR: Wipf & Stock, 2006), pp. 7, 54-59.
46 A dynamic rendering; woodenly translated as make/keep me alive.
11
(d) Jer. 22:3 גר מנ ום ית ו אל נוה ו מ אל־ת ד סו אל־תח כי נק ם ו פ ש ההז ום במקו אל־ת
“And do not oppress the alien, orphan, or widow; do no violence (to them)! And do
not spill innocent blood in this place!”
(e) Jer. 11:14 עד־העם הזה פלל ב ת אתה אל־ת ו
“But you – do not pray for this people!”
In example (a), the imperative follows a syntactically detached clausal adverb, meaning that
syntactically the imperative is clause-initial. Similarly, example (b) begins with a syntactically
detached clausal adverb, but it also has an object clause ( ל יךאל ים אלה ר אמ ר אש כ ) that is preposed
for pragmatic purposes. The next example has a pragmatically preposed prepositional phrase
(PP). Example (d) has two preposed object clauses, and in example (e), the pronominal subject
is preposed.47
Thus far, I have outlined an argument that the irrealis SC occurs in clause-initial position
which is why it most often appears in the weqatal construction. I have also suggested that the
irrealis SC, without the conjunction we-, can occur after clausal adverbs. It should be noted that
these very features are cited in the traditional literature as features by which interpreters can
identify prophetic perfects. Ewald claimed that the SC with future time reference often occurs in
clause-initial position,48
and many scholars have claimed that prophetic perfects often follow
certain clausal adverbs, such as הנה,49
.לכן and ,כי 50
Additionally, prophetic perfects are said to be
47
Cf. Moshavi, Word Order in the Biblical Hebrew Finite Clause, p. 65.
48 G. H. Ewald, Syntax of the Hebrew Language of the Old Testament, J. Kennedy (trans.), (Edinburgh: T.
& T. Clark, 1879), §135c; cf. Driver, Use of the Tenses in Hebrew, §14β-γ.
49 C. P. Caspari, Beitrage zur Einleitung in das Buch Jesaia und zur Geschichte der jesaianischen Zeit
(Berlin: Gebauersche Buchhandlung, 1848), p. 68 n.; W. E. Pearson, The Prophecy of Joel: Its Unity, Its Aim and
the Age of Its Composition (Leipzig: Theodor Stauffer, 1885), p. 5; Ewald, Syntax of the Hebrew Language, §135c;
J. A. Hughes, “Another Look at the Hebrew Tenses,” Journal of Near Eastern Studies, 29/1 (1970): p. 21.
12
interspersed with weqatals and LPCs,51
which would be expected if the SCs in question were
actually irrealis.
Another point of overlap between my argument and the traditional descriptions regards
the notion of certainty, which is ubiquitously cited in literature. According to my explanation,
the majority of alleged prophetic perfects are SCs that express situations with assumptive or
deductive, epistemic modality. Both of these are used to express situations with a degree of
confidence or certainty, but which it is depends on the kind of information the speaker/author has
or what he has experienced.
5.0 Examples
There are three criteria that must be met in order to demonstrate that alleged examples of
prophetic perfects are actually irrealis uses of the SC. The verb must have (1) a contextually
determined modal meaning, which in prophetic literature is most often very closely associated
with (2) a contextually determined future time reference. The final criterion is (3) the verb must
occur in a clausal position appropriate for a modal verb (as outlined above); namely in clause-
initial position, following a clausal adverb, or with various pragmatically preposed constituents.
In this section, I illustrate that many alleged examples of the prophetic perfect are in fact irrealis
uses of the SC. The sections below have a small sample of examples from my prior study that
occur in the prophetic literature. I briefly discuss each example, but a full discussion is not
possible here. Each of the examples in this section was argued at length in my study, and the
interested reader should refer to that work for further details.
50
J. C. L. Gibson, Davidson’s Introductory Hebrew Grammar: Syntax (4th
ed.) (Edinburgh: T. & T. Clark,
1994), §59b.1.
51 A. B. Davidson, Introductory Hebrew Grammar; Hebrew Syntax (3
rd ed.) (Edinburgh: T. & T. Clark,
1902), §41.1; Gibson, Introductory Hebrew Grammar, §59b; cf. Driver, Use of the Tenses in Hebrew, §113.1.
13
The first clause-initial example is one of the most regularly cited of the alleged examples
of the prophetic perfect.52
(f) Num. 24:17 נו א ר א ל נו אשור ה עת א ו ל יעק ב כוכ ך דר וב רק א ו ק ב מ טם ו רא שב ש י ל מ
“I see it, but not now; I behold it, but not near! A star will go out (OR: march) from
Jacob, and a scepter will rise from Israel!”
Based on the broader context of Balaam’s oracles and the immediate context, including clauses
prior and following the clause of interest (דרך כוכב מיעקב), the situation referred to by the verb דרך
is unambiguously future from the ST. The ancient versions (LXX, Vulgate, Targum), as well as
modern translations (ESV, NRSV, NASB, NIV, NAB, HCSB, NKJV, etc.) and scholars,53
affirm
that this verb has future time reference. The prophecy’s content begins with an irrealis SC (דרך),
expressing the situation with assumptive, epistemic modality. The verb is then followed by a
series of irrealis SCs, in the construction weqatal, in vv. 17-18 (והיה ,והיה ,ומחץ ,וקם).54
The next example comes from Isaiah 28, which begins with a “woe” directed at Ephraim
(v. 1). The following verses (vv. 2-4) describe the destruction that is coming to Ephraim, thereby
providing the explanation for the woe.55
52
J. G. Murphy, The Elements of Hebrew Grammar; Together with an Appendix on Chaldee Grammar
(London: David Nutt, 1857), §81; Driver, Use of the Tenses in Hebrew, §14α; F. R. Blake, A Resurvey of Hebrew
Tenses, Scripta Pontificii Instituti Biblici, 103 (Rome: Pontifical Biblical Institute, 1951), §10; Klein, “The
‘Prophetic Perfect,’” p. 50; Joosten, The Verbal System of Biblical Hebrew, p. 207; cf. B. K. Waltke & M.
O’Connor, An Introduction to Biblical Hebrew Syntax (Winona Lake, IN: Eisenbrauns, 1990), p. 490; Andrason,
“Future Values of the Qatal and Their Conceptual and Diachronic Logic,” p. 20.
53 E.g., Klein, “The ‘Prophetic Perfect,’” p. 50; R. D. Cole, Numbers, The New American Commentary, 3b
(Nashville, TN: Broadman & Holman Publishers, 2000), p. 426.
54 For discussion of the reading קם in 1QM 11:6, see S. Holst, Verbs and War Scroll: Studies in the Hebrew
Verbal System and the Qumran War Scroll, Studia Semitica Upsaliensia, 25 (Uppsala: Uppsala Universitet, 2008),
pp. 101, 105. For discussion of v. 19, see Carver, A Reconsideration of the Prophetic Perfect, pp. 172-173.
55 M. A. Sweeney, Isaiah 1-39: With an Introduction to Prophetic Literature, Forms of the Old Testament
Literature, 16 (Grand Rapids, MI: Eerdmans, 1996), pp. 359, 362.
14
(g) Isa. 28:2 נ ץק חז ה ה אמ נ ו ז י לאד ם כ ב ק ער ש ד בר ר ם כ ט ים מ זר יר ם כב פ י ט נ ים ש ץ לא יח ה י ר דב
“Behold, a strong and mighty one belongs to the Lord; like a storm of hail, a gale of
destruction; like a storm of strong, flooding waters; he will hurl (them) to the earth
with power!”
Some translations and commentators have translated הניח with past time reference, but that makes
no sense in this context. First of all, this is evidently a “threat”56
and, moreover, the LPC in v. 3
and the weqatal in v. 4 further affirm that the time reference of these verses in future. Many
commentators,57
translations (e.g., NRSV, NKJV, NIV, and NET), and ancient versions (LXX,
Targum, Peshitta)58
interpret הניח with future time reference. Traditionally, the SC in v. 2 has
been considered a prophetic perfect,59
but I propose that, with the future time reference and
clause-initial position (in the fourth colon of the four cola in this verse),60
this is an irrealis use of
the SC expressing the situation with assumptive, epistemic modality.
56
H. Wildberger, Isaiah 28-39: A Continental Commentary, T. H. Trapp (trans.), Continental
Commentaries (Minneapolis, MN: Fortress Press, 2002) p. 3.
57 O. Kaiser, Isaiah 13-39: A Commentary, The Old Testament Library (Philadelphia, PA: Westminster
Press, 1974), pp. 236, 240; J. N. Oswalt, The Book of Isaiah, Chapters 1–39, The New International Commentary on
the Old Testament (Grand Rapids, MI: Eerdmans, 1986), p. 502; J. Blenkinsopp, Isaiah: A New Translation with
Introduction and Commentary (Vols. 3), Anchor Bible, 19 (New York, NY: Doubleday, 2000), p. 385; W. A. M.
Beuken, Isaiah II: Isaiah 28-39 (Vol. 2), Historical Commentary on the Old Testament (Leuven: Peeters, 2000), p.
11; B. S. Childs, Isaiah, Old Testament Library (Louisville, KY: Westminster John Knox Press, 2001), p. 201.
58 The reading of 1QIsa
a reflects an interpretation of the MT and a linguistic-stylistic update, rather (והניח)
than an alternative reading; cf. E. Tov, Textual Criticism of the Hebrew Bible (3rd
ed., revised and expanded)
(Minneapolis, MN: Fortress Press, 2012), p. 256. This is to be considered on a par with the LXX translating the
verb with a Future.
59 Driver, Use of the Tenses in Hebrew, §14α; Caspari, Beitrage zur Einleitung in das Buch Jesaia, p. 68 n.;
E. J. Young, The Book of Isaiah: The English Text, with Introduction, Exposition, and Notes; Chapters 19-39 (Vol.
2), The New International Commentary on the Old Testament (Grand Rapids, MI: Eerdmans, 1972), p. 266 n. 5.
60 Cf. H. Wildberger (Isaiah 28-39, p. 10) who noted that “[i]n v. 2bβ the imagery is abandoned” since it is
not the storm or the strong one but “Yahweh himself” who throws down.
15
My first example of an irrealis SC following a clausal adverb is another very commonly
cited alleged example of the prophetic perfect.61
Isaiah 5 begins with the song of the vineyard
(vv. 1-7) and following verses (vv. 8-24) describe the analogical relevance of the song for Israel.
In vv. 8-24 there are several marked transitions from the current (i.e., described mostly in present
time) sins of the people to the future judgment that is coming on them as a result. The first
transition is made by a direct speech from the Lord (vv. 9-10), while the others are marked with
the conjunction לכן (vv. 13, 14, and 24).62
(h) Isa. 5:13 י ל ב י מ דעת לכן גלה עמ
“Therefore, my people will go into exile for lack of knowledge.”
In light of the context and transitional markers, this verb undoubtedly has future time reference.63
This use of the SC is best explained as an irrealis use, expressing the judgment of Israel with
future time and epistemic modality.64
Another example of an irrealis SC found in a judgment context occurs in the fifth chapter
of Jeremiah.
61
Driver, Use of the Tenses in Hebrew, §14α; C. F. Keil and F. Delitzsch, Commentary on the Old
Testament (vol. 7; Peabody, MA: Hendrickson, 1996), p. 111; G. B. Gray, A Critical and Exegetical Commentary on
the Book of Isaiah, I-XXVII: Introduction and Commentary on I-XXVII (Edinburgh: T. & T. Clark, 1912), p. 92; E. J.
Young, The Book of Isaiah: The English Text, with Introduction, Exposition, and Notes; Chapters 1-18 (Vol. 1), The
New International Commentary on the Old Testament (Grand Rapids, MI: Eerdmans, 1972), p. 211 n. 14; Klein,
“The ‘Prophetic Perfect,’” p. 52.
62 G. V. Smith, Isaiah 1-39, The New American Commentary (Nashville, TN: B & H Publishing Group,
2007), pp. 174-175; Keil & Delitzsch, Commentary on the Old Testament (vol. 7), p. 111; cf. J. D. Nogalski,
Interpreting Prophetic Literature: Historical and Exegetical Tools for Reading the Prophets (Louisville, KY:
Westminster John Knox Press, 2015), pp. 59-63.
63 In addition to those who consider גלה a prophetic perfect, see H. Wildberger, Isaiah 1-12: A Continental
Commentary, T. H. Trapp (trans.), Continental Commentaries (Minneapolis, MN: Fortress Press, 1991), p. 189;
Smith, Isaiah 1-39, pp. 174-175; Van de Sande, Nouvelle perspective sur le systeme verbal de l’hebreu ancien, pp.
332-334; Andrason, “Future Values of the Qatal and Their Conceptual and Diachronic Logic,” p. 21.
64 The same is true for the SC in Isa. 5:14, but space prohibits discussion of this verse; see Carver, A
Reconsideration of the Prophetic Perfect, pp. 179-180.
16
(i) Jer. 5:6 כ כן על י ם ה י ה אר א ער מ ד ערבותב ז שד ם י
“Therefore, a lion from the forest will slay them; a wolf of the deserts will destroy
them.”
The first five verses of this chapter depict the prevalence of sin among the people of Judah.
Their sin is metaphorically described as breaking the yoke and tearing up the bonds (v. 5). As a
result, judgment will come on them (v. 6). Most translations (e.g., NRSV, NASB, NAB, NIV,
ESV, NKJV; cf. Targum) and scholars65
recognize the future time reference of the SC, which is
affirmed by not only the context of coming judgment but also the LPC that appears in the
following clause.
In this section, the primary focus of our discussion on the preposed constituents is on the
syntactic classes of the constituents, and not on their functions (e.g., focus or topicalization). In
the following examples, some of the preposed constituents are syntactically dislocated from the
clause (i.e., in extraposition). These will be distinguished from the preposed constituents that are
not syntactically dislocated.
My first example comes from the Oracles Against the Nations in Jeremiah.
(j) Jer. 49:21 ים עקה ב ץ צ לם רעשה האר פ קול נ מע קולה מ ש סוף נ
“From the sound of their fall the earth will shake; a cry – at the Reed Sea its sound
will be heard.”
65
Driver, Use of the Tenses in Hebrew, §14α; J. Bright, Jeremiah: A New Translation with Introduction
and Commentary (2nd
ed.), The Anchor Bible, 21, W. F. Albright & D. N. Freedman (Eds.) (Garden City, NY:
Doubleday, 1965), p. 36; R. K. Harrison, Jeremiah and Lamentations: An Introduction and Commentary, Tyndale
Old Testament Commentaries (Downers Grove, IL: InterVarsity Press, 1973), p. 75; R. P. Carroll, Jeremiah: A
Commentary, The Old Testament Library (Philadelphia, PA: Westminster Press, 1986), p. 174; L. C. Allen,
Jeremiah: A Commentary, The Old Testament Library (Louisville, KY: Westminster John Knox Press, 2008), p. 71.
17
Although some commentators have tried to force a non-future time interpretation of these
verbs,66
the majority of commentators recognize that the SCs have future time reference, even
though they have not been able to explain how these verbs could refer to future situations.67
I
propose that these are irrealis SCs in clauses with preposed prepositional phrases (PP). The WO
of the parallel cola in this verse is PP-V-S, with the added variation that in the second colon there
is a dislocated noun (צעקה) that is resumptively referenced with a pronominal suffix (on קולה). It
is important to note, however, that the expected irrealis WO of V-S is still intact.
The next example illustrates a syntactically restricted deviation from the expected WO
for irrealis clauses. This occurs when there is an independent pronominal subject.68
Ezekiel has
two long descriptions of the metaphorical adultery and prostitution of Samaria and Jerusalem in
chapters 16 and 23. The resulting judgment in each chapter draws to a close with an indication
that the people will bear (√נשא) the consequences of their sins.
(k) Ezek. 23:49 נ נת נהו ו כ מת לוליכ י וחטא ן עליכ ז ש ן ג ת אינה ת ידע נ י אנ י כ ם ו הו י אד הי
“And they will set your vileness upon you, and you will bear the guilt of your
idolatry. And you will know that I am the Lord Yahweh.”
(l) Ezek. 16:58 ת מת א תך ז א ך תועבות ו שאת ת א י א ים נ הו ם נ הי
66
E.g., Rogland, Alleged Non-past Uses of Qatal, p. 103; Bright, Jeremiah, p. 330; W. L. Holladay,
Jeremiah 2: A Commentary on the Book of the Prophet Jeremiah, Chapters 26-52, P. D. Hanson (Ed.), Hermeneia,
A Critical and Historical Commentary on the Bible (Philadelphia, PA: Fortress Press, 1989), p. 371.
67 E.g., Carroll, Jeremiah, pp. 804-805; F. B. Huey, Jeremiah, Lamentations, The New American
Commentary, 16 (Nashville, TN: Broadman & Holman, 1993), p. 403; W. Brueggemann, A Commentary on
Jeremiah: Exile and Homecoming (Grand Rapids, MI: Eerdmans, 1998), p. 458; G. H. Parke-Taylor, The Formation
of the Book of Jeremiah: Doublets and Recurring Phrases, Society of Biblical Literature Monograph Series
(Atlanta, GA: Society of Biblical Literature, 2000), p. 155; J. M. Bracke, Jeremiah 30-52 and Lamentations,
Westminster Bible Companion (Louisville, KY: Westminster John Knox Press, 2000), pp. 137-138; L. Haney,
“YHWH, the God of Israel...and of Edom? The Relationships in the Oracle to Edom in Jeremiah 49:7-22,” in J.
Goldingay (Ed.), Uprooting and Planting: Essays on Jeremiah for Leslie Allen (pp. 78-115), Library of Hebrew
Bible/Old Testament Studies (New York, NY: T. & T. Clark, 2007), p. 109; Allen, Jeremiah, pp. 490, 493.
68 Cf. Moshavi, Word Order in the Biblical Hebrew Finite Clause, p. 65 n. 1.
18
“Your vileness and your abominations – you must bear them, declares Yahweh.”
There are two things to note regarding the WO in 16:58. First, the phrase “your vileness and
your abominations” is syntactically dislocated from the clause, and is resumptively referenced
with a pronominal suffix. Second, there is an independent pronominal subject, which frequently
occur in first position.
The context of the passage and the LPC in 23:49 strongly suggest that the SC in 16:58
has future time reference, and many scholars have recognized that it refers to a situation that is
future from the ST.69
Moreover, several scholars have translated the verb with a modal force,
translating “you must bear.”70
I propose that in light of the persistent and unrepentant vileness
and abominations of Jerusalem, נשאתי expresses that which is “epistemically necessary,”71
i.e.,
that there is no other possibility but for Jerusalem to bear the consequences of their sins. In other
words, the SC expresses a situation with future time reference and deductive, epistemic modality.
Perhaps one of the most interesting cases of pragmatic preposing is for purpose of
creating parallelism. Imperatives and jussives express deontic modality and their unmarked WO
is clause-initial. However, there are cases in which imperatives and jussives occur in non-initial
position because of parallelism. Some patterns do not disrupt the unmarked WO for irrealis
69
E.g., G. A. Cooke, The Book of Ezekiel, The International Critical Commentary (Edinburgh: T. & T.
Clark, 1936), p. 178; L. C. Allen, Ezekiel 1-19, Word Biblical Commentary, 28 (Dallas, TX: Word Books, 1994), p.
232; H. D. Hummel, Ezekiel 1-20, Concordia Commentary (Saint Louis, MO: Concordia 2005), p. 457; see also,
NIV.
70 W. Zimmerli, Ezekiel 1, R. E. Clements (trans.), F. M. Cross & K. Baltzer (Eds.), Hermeneia, A Critical
and Historical Commentary on the Bible (Philadelphia, PA: Fortress Press, 1979), p. 333; W. Eichrodt, Ezekiel: A
Commentary, The Old Testament Library (Philadelphia, PA: Westminster Press, 1970), p. 198; M. Greenberg,
Ezekiel, 1-20: A New Translation with Introduction and Commentary, The Anchor Bible, 22-22A (New York, NY:
Doubleday, 1983), p. 273; also, NRSV, NAB; cf. D. I. Block, The Book of Ezekiel: Chapters 1-24, The New
International Commentary on the Old Testament, R. L. Hubbard, Jr. (Ed.) (Grand Rapids, MI: Eerdmans, 1997), p.
512.
71 Palmer, Mood and Modality, p. 89.
19
verbs, such as abab and aabb. However, chiastic parallelism (e.g., abba) does disrupt the
unmarked WO.72
For example,
(m) Jer. 17:18 ם יום רעה יא עליה רם הב ברון שב נה ש ש ומ
“Bring a day of calamity upon them! And break them (with) double destruction!”
(n) Jer. 18:23 כפר על אל פניך אל עונם ת ל חטאתם מ י ו ח מ ת
“Do not forgive their iniquity, and do not blot out their sin from before you!”
In examples (l) and (m), the verbs occur in the expected, clause-initial position in the first colon
of each verse (the A-lines), but occur in clause-final position in the respective second cola (the
B-lines). The WOs of the second cola were disrupted (i.e., constituents were pragmatically
preposed) in order to create the chiastic parallelism.73
When the irrealis SC occurs in parallel with another verb form, it is usually the LPC, as
in Jer. 5:6 (see above, example [i]), or a weqatal, as in Isa. 19:8.
(o) Isa. 19:8 לו כל אב אור חכה ו יכי בי ל ת על מש ר מ כ שי מ ר ני ופ ללו פ מ ם א מי
“And all those who cast a hook in the Nile will mourn; and those who spread out a net
on the face of the water will languish.”
The WO pattern is abcbca (a = V; b = NP; c = PP) with a modification in the first colon as the
PP “breaks” the construct chain.74
The first colon has the unmarked WO, but in the second
colon, the NP and the PP were preposed for poetic purposes.
The future time reference of 19:1-10 is unambiguously indicated by the formulaic marker
Participle and the many subsequent weqatals and LPCs describing the disaster that will + הנה
72
In the examples below, an extra space is put between cola to indicate my understanding of the lineation.
73 It is typically, but not always, the WO of the second colon that is disrupted. Cf. Lunn, Word-Order
Variation in Biblical Hebrew Poetry, p. 275: “It is primarily within the B-line of synonymous parallelisms that
word-order variation as a purely stylistic or rhetorical device ... is admissible.”
74 See Waltke & O’Connor, An Introduction to Biblical Hebrew Syntax, p. 140.
20
befall the Egyptians. Most translations (e.g., ESV, NRSV, NASB, NIV, NKJV, etc.), ancient
versions (LXX, Targum, Vulgate), and modern commentators75
interpret the SC with future time
reference, but only my explanation can account for this. The verb אמללו is an irrealis SC without
the conjunction we- that expresses the situation with future time reference and epistemic
modality.
Another example occurs in Jeremiah 31. The relevant literary unit is 31:23-26. The first
verse begins with a formulaic marker (כה אמר יהוה צבאות) and describes the promising future that
will be when the Lord restores the fortunes of his people. The future time reference of the unit is
further indicated by a LPC in v. 23 and a weqatal in v. 24. Verse 25 continues the hopeful
description of the future.
(p) Jer. 31:25 ש עיפה י נפ וית ר י ה כל כ י ו לאת ש דאבה מ נפ
“For I will give (the) soul of the weary (its) fill, and every soul (that) has languished I
will fill up.”
The first irrealis SC occurs after a clausal adverb (cf. examples above) and the second (מלאתי)
occurs at the end of the second colon. This verse has a modified chiastic parallelism76
with the
pattern abcbca (a = V; b = S; c = modifying phrase). There is a difference in the modifying
phrases in that the first colon has an adjective, while the second has an asyndetic relative
clause.77
Modern translations (e.g., NRSV, NAB, ESV, and NIV) and scholars78
have
75
E.g., Keil & Delitzsch, Commentary on the Old Testament (vol. 7), p. 232; Oswalt, The Book of Isaiah,
Chapters 1–39, p. 364; Blenkinsopp, Isaiah, p. 312; Smith, Isaiah 1-39, p. 357.
76 Cf. D. R. Jones, Jeremiah: Based on the Revised Standard Version, New Century Bible Commentary
(Grand Rapids, MI: Eerdmans, 1992), p. 397.
77 Bright, Jeremiah, p. 276; J. A. Thompson, The Book of Jeremiah, The New International Commentary on
the Old Testament (Grand Rapids, MI: Eerdmans, 1980), p. 576.
78 Driver, Use of the Tenses in Hebrew, §14β; Thompson, The Book of Jeremiah, p. 576; Carroll, Jeremiah,
p. 605; Holladay, Jeremiah 2, p. 154; G. L. Keown, P. J. Scalise, & T. G. Smothers, Jeremiah 26-52, Word Biblical
21
recognized the future time reference of these SCs, and I propose that they are irrealis SCs
expressing these situations with epistemic modality. While the first follows a clausal adverb, the
second occurs in clause-final position on account of parallelism.
6.0 Conclusions
In my previous study of the prophetic perfect in Biblical Hebrew, I have argued that what
has been known for the last century and a half as the prophetic perfect does not exist, and that
those uses of the SC that were previously lumped together into this category actually have other
uses. In this paper, I have briefly presented the argument for one of those uses that has been
inappropriately categorized as the prophetic perfect. I have argued that the SC is used without
the conjunction we- to express epistemic, irrealis situations.
Research over the past few decades has shown that modal verbs nearly always take
clause-initial position, and that this was used as a strategy for disambiguating realis and irrealis
verbs. Research has also shown that modal verbs are not the first constituent in a clause in
certain syntactic environments, such as when a clause is begun by a clausal adverb or the
pragmatic preposing of (a) constituent(s). Based on the criteria of contextual modal meaning
(and future time reference) and WO, I have argued in my previous study that 58 alleged
examples of the prophetic perfect are actually epistemic, irrealis uses of the SC. In this paper I
have given a brief overview of my analysis of 10 of the 58 examples argued at length in my
previous study.
In conclusion, this paper makes several important contributions. First, it provides an
introduction to my approach, which employs diachronic explanation and synchronic description
of the language, to a use of the SC that has gone without a sound explanation for more than a
Commentary, 27 (Waco, TX: Word Books, 1995), p. 124; W. McKane, A Critical and Exegetical Commentary on
Jeremiah, 26-52, The International Critical Commentary (Edinburgh: T. & T. Clark, 1996), p. 808.
22
millennium. Second, it labels and explains semantic categories and syntactic parameters for the
irrealis use of the SC without the conjunction we-. An additional implication concerns the
interpretation and translation of Hebrew Bible. It has long been recognized that many of the
irrealis SCs have future time reference and sometimes even a modal meaning, but scholars have
failed to explain how the SC could express these situations. In light of the greater linguistic
framework provided by my approach, interpreters and translators can have a great understanding
of the language of prophecy (and even poetry more generally) as well as a heightened sensitivity
to some of the more subtle aspects of the BH verbal system.