Encoding of storyline events in Biblical Hebrew

This article first considers what is meant by perfective and imperfective aspect (sec. 1), then (sec.

2) discusses the aspects conveyed by different Hebrew verb forms. Section 3 reasons that wyyqtl forms should be viewed as the default way of encoding storyline events.1 Finally, section 4 con- cerns the significance of using other than the default verb form in narrative and procedural texts.

Aspect (Perfectivity)

Verbal aspect is a way of portraying an event (it ‘reflects the subjective conception or portrayal by the speaker’—Fanning 1990:31). A basic division is between imperfective and perfective aspects.

When imperfective aspect is used to describe an event, the event is portrayed as not completed. The verbal participle in Hebrew has imperfective aspect because it portrays events as in process or as not completed at the point of reference. Thus, in Ruth 4.1c, the participle ‘ober ‘[was] passing by’ indicates that the passing by is portrayed as an action that has not been completed at the point the story has reached. Boaz speaks while the kinsman-redeemer is passing by:

(4.1a) ûbo‘az ‘alâ hašša‘ar // (1b) wayyešeb šam

& Boaz went up the gate // & he sat there (1c) wahinneh haggo’el ‘ober

& behold the kinsman redeemer [was] passing by

(1d) ’ašer dibber-bo‘az // (1e) wayyo’mer

whom spoke-Boaz // & he said

‘Meanwhile Boaz went up to the town gate and sat there. When the kinsman-redeemer he had mentioned was passing by, Boaz spoke to him and said…’

When perfective aspect is used to describe an event, the event is portrayed as a whole (‘a com- plete and undifferentiated process’—Porter 1992:21). The qtl and wyyqtl verb forms in Hebrew have perfective aspect. It follows that, in Ruth 4.1a (above), the qtl form ‘alâ portrays the act of going up as a single journey, including its beginning and end. Though the use of the perfective presupposes that the journey was completed, it does not focus on the end of the journey; it simply views the journey as a whole. (The same argument applies to wayyešeb ‘& he sat’ [4.1b], dibber ‘spoke’ [4.1d)], and wayyo’mer ‘& he said’ [4.1e].)

The Aspect of Hebrew Verbs

English verbs convey both tense and aspect.2 Thus, ‘he went up’ in English is a past tense verb with perfective aspect, while ‘he was passing by’ is a past tense verb with imperfective aspect.

Hebrew verbs do not convey tense. Rather, when translating a Hebrew verb into a language whose verbs indicate tense, the tense has to be deduced from the context. Waltke and O’Connor (1990:459) show this by considering how the RSV translators of the book of Job render qtl verbs. They render them ‘as a past tense 252 times and as a present tense 244 times; in its bound form wqtl as a past 12 times, a present 23 times, and a future 14’.

1 I follow Bowling (1997:50-51) in referring to the basic verb forms of Hebrew as qtl, yqtl, wqtl, wyyqtl and verbal participle. Waltke & O’Connor (1990) call qtl (qatal) forms the ‘suffix conjugation’, yqtl (yiqtol) forms the ‘prefix conjugation’, and wqtl (waqatal) forms the ‘relative waw + suffix conjugation’. SIL International’s ‘Bible Analysis and Research Tool (BART) tags both qtl and wqtl forms as ‘p: perfect’. It tags yqtl forms as ‘i: imperfect’ and wyyqtl (wayyiqtol) forms as ‘w: waw consecutive’.

2 When a verb is marked for tense, it indicates a relationship between the time of the situation described and the time of speaking or some other point of reference in time.

The following chart distinguishes the basic verb forms of Hebrew:

Perfective

Imperfective

Without waw:

qtl

yqtl

Conjunctive:

wqtl*

wyqtl

Chain (consecutive):

wyyqtl

wqtl*

*The position of the accent distinguishes the two wqtl forms for the first person singular and second person masculine singular of some verbs (ibid. 520).

An example of a qtl form at the beginning of a clause without the conjunction waw is found in 1 Ki 21.12a (see Ruth 4.1a, 1d for qtl forms that are without waw because a preposed constituent precedes them). The event of proclaiming a fast is portrayed as whole (DO: direct object marker):

(21.12a) qar’û côm // (12b) wahošîbû ’et-nabôt baro’š ha‘am

they called fast // & they seated DO-Naboth at the head of the people

‘They proclaimed (qtl) a fast and seated (wqtl) Naboth in a prominent place among the people.’

An example of a yqtl form at the beginning of a clause without waw is found in Psalm 23.3b. The imperfective is appropriate because the guidance is viewed as an ongoing occurrence:

(23.3b) yanZenî bama‘galê-cedeq lama‘an ša

he guides me in paths-of righteousness for sake of his name

‘He guides me in paths of righteousness for his name’s sake.’

The following are examples of yqtl forms that are without waw because a preposed constituent precedes them. Since yqtl forms are imperfective, they are used to portray the events as not com- plete because they are still in the future (Ruth 1.17) or are habitual (Gen 2.6a):

(1.17a) ba’ašer tamûtî ’amût // (17b) wašam ’eqqaber

at where you will die I will die // & there I will be buried

‘Where you die (yqtl) I will die (yqtl), and there I will be buried (yqtl).’

(2.6a) wa’ed ya‘aleh min-ha’arec

& water arises from-the earth

‘but streams used to come up from the earth’

Bowling (p.c.) classifies 1 Ki 21.12b (above) as a *conjunctive *wqtl **clause. It is conjoined to a

wqtl clause, has the same subject, and is readily interpreted as having perfective aspect.3

Job 22.27b provides an example of a wyqtl clause; it is conjoined to a yqtl clause: (22.27a) ta‘tîr ’elayw // (27b) wayišma‘ekka

you will pray to him // & he will hear you

‘You will pray (yqtl) to him, and he will hear (wyqtl) you’

Many languages have a special verb form to encode storyline events in a narrative when no discontinuity is to be signalled, and Hebrew is no exception. Bowling (1997:52) applies the term ‘chain’ to such clauses. Two verb forms are found in such chains: wyyqtl and wqtl. I have already noted that the wyyqtl chain form has perfective aspect (sec. 1; this form is further discussed in sec. 3). I now consider the wqtl chain form.

3 There is no formal difference between conjunctive wqtl and wqtl chain forms in the third person. Driver (1892) suggested that the wqtl form has ‘lost its individuality and passed under the sway of the verb to which it is connected’ (Waltke & O’Connor 1990:465). However, see Gen 26.22 below for an exception; a wqtl clause with a different subject from the preceding qtl clause is interpreted as a wqtl chain form.

Like yqtl clauses, wqtl chain clauses have imperfective aspect. This is seen in Gen 2.6 (describing a habitual event in the past) and 12.12 (predicting future events):4

(2.6a) wa’ed ya‘aleh min-ha’arec // (6b) wahišqâ ’et-kol-panê-ha’adamâ

& water arises from-the earth// & it waters DO-all-surface-of the ground ‘but streams used to come up (yqtl) from the earth and water (wqtl) the whole surface of the ground’

(12.12a) wahayâ kî-yir’û ’otak hammicrîm // (12b) wa’ama

& it will be when-they see you the Egyptians // & they will say

‘When the Egyptians see you (yqtl), they will say (wqtl)…’

Bowling cites Ruth 1.11b as an instance in which a wqtl chain clause expresses the purpose of the event described in the previous clause (‘Qn’ = polar question marker):

(1.11a) ha‘ôd-lî banîm bame‘ay // (11b) wahayû lakem la’anašîm Qn still-for me sons in my womb // & they will be for you to husbands ‘Am I going to have more sons, who will become (wqtl) your husbands?’

Gen 26.22f-g is an example of a wqtl chain clause that expresses the result of the event described with a qtl form in the previous clause:

(26.22f)‘attâ hirZîb yhwh lanû // (22g) ûparînû ba’arec now has made room YHWH for us // & we will be fruitful in the land ‘Now the LORD has given us room (qtl) and we will flourish (wqtl) in the land.’

When a wqtl chain clause is preceded by an imperative, it presents a further imperative, which is to be performed after the first one. This is exemplified in Ruth 2.14:

(2.14) gošî halom // wa’akalta min-halleZem // waIabalta pittek baZomec approach here // & eat from-the bread // & dip your portion in the vinegar ‘Come over near, eat (wqtl) some bread, and dip (wqtl) it in the wine vinegar!’

Longacre (1994) points out that wqtl chain clauses are the default way of presenting steps in a procedure when no discontinuity is to be signalled. The procedure may be prescriptive, as in the descriptions in Leviticus of the rituals to be followed in making offerings (pp. 52-53). Alternati- vely, it may be ‘something done over and over again in the past’ (Bowling 1997:61), as in 1 Sam 17.34-35:

(17.34b) ro‘eh hayâ ‘abdaka la’abîw bacco’n

shepherd has been your servant for his father among the sheep

(34c) ûba’ ha’arî wa’et-haddôb // (34d) wanasa’ seh meha‘eder

& came the lion & with-the bear // & took away sheep from the flock (35a) wayaca’tî ’aharayw

& I went out after him

‘Your servant has been (qtl) a shepherd for his father. Whenever a lion or a bear came (wqtl) and seized (wqtl) a sheep from the flock, I would go (wqtl) after it…’

See section 4 on the status of wqtl chain clauses in narrative, and of wyyqtl chain clauses in procedures.

4 Because wqtl chain forms most often describe events that have not yet been realised, some grammarians prefer to say that they have irrealis mood. However, passages such as Gen 2.6b, 1 Sam 17.34-35 and Ex 1.19e show that they are also used to describe habitual realis events. That is why I classify them as having imperfective aspect, where imperfective is understood to include irrealis events because they too are not completed at the point of reference.

The wyyqtl form as the default way of encoding events in narratives

Two-way contrasts are very common in linguistics, and tend to be between a default and a marked member. A common mistake is to allocate a positive label not only to the marked member of the pair, but also to the default member. When this happens, although the label may describe the typical function of the default member, it will not account for all instances of its usage. I now apply this observation to the function of wyyqtl forms in narrative.

  1. Wyyqtl forms are not marked for chronological sequence

In his 1989 book on Joseph, Longacre made a positive claim about wyyqtl clauses in Hebrew: viz., that they move the story forward, with the events so presented in chronological sequence (p. 178

—as in Ruth 4.1b). However, Heimerdinger (1999) points out instances in which wyyqtl clauses describe events that are not in chronological sequence. One such is in Gen 7.22-23:

(7.22) kol ’ašer nišmat-rûaZ Zayyîm ba’appayw mikkol ’ašer beZarabâ metû all that breath-of spirit of life in its nostrils from all that on dry land died ‘All that breathed the breath of life in its nostrils, all that was on dry land, died.’

(23a) wayyimaZ ’et-kol-hayaqûm ’ašer ‘al-panê ha’adamâ

& he destroyed DO-all-the existence which on the surface of the ground

‘& was wiped out every living thing that was on the face of the earth…’

(23b) wayyimmaZû min-ha’arec & they were destroyed from-the earth ‘& they were wiped off from the earth.’

As Heimerdinger notes (p. 83), Gen 7.23b ‘cannot be described as advancing the action; it does not present a new event, but merely restates’ the action of 23a.

Wyyqtl clauses may even present flashbacks (pp. 86-89). Although some scholars dispute this, Num 1.47-49a seem pretty clear:

(1.47) wahalwiyyim lamaIIeh ’abotam lo’ hotpaqadû batôkam

& the Levites according to tribe of their fathers not were numbered among them

‘& the Levites according to their ancestral tribe were not numbered among them.’

(48) wayadabber yhwh ’el-mošeh le’mor // ’ak ’et-maIIeh lewî lo’ tipqod

& spoke YHWH to-Moses saying // only DO-tribe of Levi not you shall number

‘& YHWH had said to Moses, (49a) “Only the tribe of Levi you shall not number”.’

Longacre’s problem was that he described wyyqtl verbs as punctiliar (loc. cit.), which implies that events are viewed as completed, ‘a finished event being followed by another one’ (Heimer- dinger p. 86). The problem disappears if they are described as having default, perfective aspect. One can then observe that events presented in wyyqtl clauses are typically in chronological sequence when they form part of a narrative (the linear presentation of events ‘imitates the purported chronological sequence of events in the real world’—op. cit. 48), without requiring that this be so.

  1. Wyyqtl forms are not marked as foreground events in narrative

Storyline events are typically presented not only in the perfective aspect, but also with a specific verb form: the simple past in English, the preterite in written Spanish, the perfect in spoken Castillian Spanish, the aorist in Greek, and a ‘narrative’ or ‘neutral’ form in many African languages. A danger is to treat this form as a marked way of presenting foreground events in sequence, rather than as the default way of describing storyline events.

For example, Longacre (1989) made a second positive claim about wyyqtl clauses: that they are the only form used for foreground events. All other verb forms then present background material of some type or another (in his 1994 article, Longacre notes some exceptions—pp. 71ff).

However, Heimerdinger argues (p. 77) that wyyqtl clauses describe both key events and trifling ones. An example is found in 2 Ki 4.37; the final clause (‘went out’) ‘is here a routine closure verb dismissing the participant’ (loc. cit.):

(4.37) wattabo’ // wattippol ‘al-raglayw // wattištaZû ’aracâ //

& she went in // & she fell at-his feet // & she bowed herself to the earth//

wattissa’ ’et-banah // wattece’

& she took up DO-her son // & she went out

‘She came in, fell at his feet and bowed to the ground. Then she took her son and went out.’

Longacre’s problem can be avoided if wyyqtl clauses are described not as the storyline band, but as the default or unmarked way of presenting storyline events (Dr. Longacre agrees—p.c.). ‘Typically, the body of a text is UNMARKED for prominence’ (Dooley & Levinsohn 2001:84).

Backgrounding in Hebrew

This section briefly discusses the significance of using other than the default verb form in narra- tives and procedures.

I argue elsewhere (Levinsohn 2000:173f) that there is a need to distinguish between verb forms that naturally convey background information in a particular discourse genre and those that are used in a marked way. In a narrative, for example, imperfective aspect tends to be used to present information of a background nature because it is the natural aspect to use to describe habitual or future actions. (Foley & Van Valin [1984:373] talk of an ‘inherent correlation’ between imperfec- tive aspect and background.)

However, there may be occasions when it is appropriate to portray a foreground event in narrative as not completed at the point of reference. Acts 4.31 in Greek provides a clear example (‘… and they were all filled [perfective] with the Holy Spirit and spoke [imperfective] the word of God boldly’). Luke portrays the people speaking the word of God boldly as an activity that resumed when they were all filled with the Holy Spirit and then was ongoing.

See also the use of wahinneh ‘& behold’ with a verbal participle in Hebrew to highlight the event concerned (discussed in Levinsohn 2009 §5.4.2), even though the participle has imperfective aspect (see sec. 1).

The primary reason for selecting a perfective versus an imperfective form, therefore, is aspectual: does the author wish to portray the event as a whole or as not completed at the point of reference? Only on rare occasions is an unexpected aspect used to create a special effect. The latter part of this section concerns such occasions.

The following chart presents the verb forms commonly found in the body of a narrative or a pro- cedure (i.e., excluding settings, non-events, etc.), together with their basic and special functions:

narrative

procedure

default theme line (no discontinuity marked) wyyqtl wqtl
discontinuity marked by preposing X: (w) X qtl (w) X yqtl
special effects: wqtl wyyqtl

I first discuss narrative and, in particular, the difference between (w) X qtl and wyyqtl forms in that genre. The basic difference between these two forms is not one of background versus fore- ground (see Bailey & Levinsohn 1992). Rather, the (w) X qtl forms are used at points of discon- tinuity, as when there is a switch of attention from one participant or situation to another (see Levinsohn 2009 §3.1). No such discontinuity is signalled by wyyqtl forms.

Gen 4.4b-5 illustrates this point. The event of v. 5a involves a switch of attention from Abel and his offering to Cain and his offering. This event is at least as important as that of 4b, even though it is presented with a w X qtl form, rather than a wyyqtl form:

(4.4b) wayyiša‘ yhwh ’el-hebel wa’el-minZatô

& looked YHWH to-Abel & to-his offering

‘Then the LORD looked with favor (wyyqtl) on Abel and his offering’

(5a) wa’el-qayin wa’el-minZatô lo’ ša‘â

& to-Cain & to-his offering not he looked

‘but on Cain and his offering he did not look with favor (qtl).

Sometimes, a (w) X qtl form is used when a wyyqtl form would have been expected. This usage can be discerned when the event that was described with the qtl form is in sequence with the event presented immediately before. A wyyqtl form would have been expected in such a context. By using the (w) X qtl form, the hearer’s attention is switched from one referent to another. This phenomenon is used on several occasions in connection with climactic events (Longacre 1989).

Gen 19.24 illustrates this phenomenon. At least some of the events of vv. 23-24 occur in chrono- logical sequence, yet all are presented with (w) X qtl forms, including the climactic event of 24:5

(19.21f) ‘He said to him [Lot], ‘… But flee there quickly, because I cannot do anything until you reach it.

That is why he called (qtl) the town Zoar.’

(23a) haššemeš yācā³ Kal- hā³ārec

the sun came out over the earth (23b) walôI bā³ cōKă

& Lot entered Zoar

(24) wyhwh himIîr Kal- saaōm waKal- Kămōrâ gāarîI wā³ēš…

& Yahweh caused to rain on Sodom & on Gomorrah sulphur & fire

‘The sun rose (qtl) over the land and Lot reached (qtl) Zoar and YHWH rained (qtl) burning sulphur on Sodom and Gomorrah…’

(W)qtl forms that begin clauses may be used to background events because they are of a prelimi- nary nature in relation to the storyline events that follow. This is illustrated in 1 Ki 21.12-13; the events of v. 12 (presented with (w)qtl forms) are readily interpreted as preliminary to and back- grounded with respect to the storyline events of 13 (presented with wyyqtl forms):

(21.12) qar’û côm wahošîbû ’et-nabôt baro’š ha‘am

they called fast & they seated DO-Naboth at the head of the people (13a) wayyabo’û šanê ha’anašîm banê-baliyya‘al

& came two the men sons-of worthlessness

‘They proclaimed (qtl) a fast and seated (wqtl) Naboth in a prominent place among the people. Then the two scoundrels came (wyyqtl)…’

Longacre (1994) notes that wqtl chain forms are also used to present the concluding event of an episode. Consider 1 Ki 20.21c, for instance, which ‘is the denouement of a story that beings in 20.1’ (p. 72):

(20.21a) wayyece’ melek yisra’el // (21b) wayyak ’et-hassûs wa’et-harakeb

& went out king of Israel // & struck DO-horses & DO-chariots (21c) wahikkâ ba’aram makkâ gadôlâ

& struck at Aram with slaughter great

‘The king of Israel advanced (wyyqtl) and overpowered (wyyqtl) the horses and chariots and inflicted (wqtl) heavy losses on the Arameans.’

5 See also Gen 15.18a, 38.25 and 44.4c (discussed by Bailey & Levinsohn [1992:202]).

I suggest that the imperfective is not used in the last clause of the episode to suggest that the king of Israel continued to inflict heavy losses on the Arameans. Rather, it indicates that the episode forms part of a larger story, with this event setting the scene for the next episode (vv. 22ff describe the efforts of the king of Aram to recoup his losses).6

I turn now to procedural texts. The difference between wqtl chain forms and (w) X yqtl forms is the same as that between wyyqtl and (w) X qtl forms in narrative. The (w) X yqtl forms are used in procedures at points of discontinuity, as when there is a switch of attention from one participant or situation to another. No such discontinuity is signalled by wqtl chain forms.

This is seen in the description in Lev 4 of the ritual for an offering for a sin committed in ignorance. Verses 6-7a use wqtl chain forms to set out the steps involving some of the blood (Longacre 1994:52-53). When attention switches to the rest of the blood (7b), however, a w X yqtl form is used, because of the discontinuity of topic:

(4.7a) wanatan hakkohen min-haddam ‘al-qarnôt mizbaZ qaIoret …

& shall put the priest from-the blood on-the horns of the altar of incense (7b) wa’et kol-dam happar yišpok ’el-yasôd mizbaZ …

& DO all-blood of the bullock he shall pour out at-the base of the altar

‘The priest shall then put (wqtl) some of the blood on the horns of the altar of … incense… The rest of the bull’s blood he shall pour out (yqtl) at the base of the altar…’

Finally, a wyyqtl form may be used towards the end of a procedure to describe the step that leads to the climax of the procedure. This is illustrated in 1 Sam 17.35. As noted in section 2, David is describing a series of events that were done over and over again in the past. They are presented with wqtl chain forms. However, the act performed by the wild animal that leads to the climactic event performed by David uses a wyyqtl form:7

(17.35) wayaca’tî ’aZarayw // wahikkitîw // wahiccaltî mippîw//

& I went out after him // & I struck him // & I delivered from his mouth //

wayyaqom ‘alay // waheZezaqtî bizqanô // wahikkitîw // wahamîtîw

& he rose against me // & I seized by his beard // & I struck him // & I killed him ‘I would go (wqtl) after it, strike (wqtl) it and rescue (wqtl) the sheep from its mouth. When it turned (wyyqtl) on me, I would seize (wqtl) it by its hair, strike (wqtl) it and kill (wqtl) it.’

See also 1 Sam 2.16 (cited by Longacre 1994:62). It is the protest by the man who brought the offering (v. 16a), which is introduced with a wyyqtl form, that leads to the climactic retort by the priest’s servant (v. 16b—using a wqtl chain form).

6 This employment of an imperfective to point forward may be compared to the use of the historical present in Greek to present the concluding event of an episode when the event points forward to a later episode (Levinsohn 2000:206, 245). See also Longacre’s observation that the wqtl chain form of wayhî ‘it happened’ is used to point forward to a climactic event (1994:84, 95).

7 Cross-linguistically, it is common for the event that immediately precedes the climax to be backgrounded in some way; see Levinsohn 1991:150 on this phenomenon in Inga (Quechuan).

References

Bailey, Nicholas A., and Stephen H. Levinsohn. 1992. The function of preverbal elements in independent clauses in the Hebrew narrative of Genesis. Journal of Translation and Textlinguistics 5(3):179-207.

Bowling, Andrew C. 1997. Another brief overview of the Hebrew verb. Journal of Translation and Textlinguistics

9:48-69.

Dooley, Robert A., and Stephen H. Levinsohn. 2001. Analyzing Discourse: A Manual of Basic Concepts. Dallas: SIL International.

Driver, S.R. 1892. A Treatise on the Use of the Tenses in Hebrew. 3rd edition. Oxford: Clarendon Press. Fanning, Buist M.

  1. Verbal Aspect in New Testament Greek. Oxford: Clarendon Press.

Foley, W. A., and R. D. Van Valin. 1984. Functional Syntax and Universal Grammar. Cambridge: Cambridge University Press.

Heimerdinger, Jean-Marc, 1999. Topic, Focus and Foreground in Ancient Hebrew Narratives. Journal for the Study of the Old Testament: Supplement Series 295. Sheffield, England: Sheffield Academic Press Ltd.

Levinsohn, Stephen H. 1991.Variations in tense-aspect markers Among Inga (Quechuan) dialects, In Language Change in South American Indian Languages, ed. by Mary Ritchie Key, 145-65. Philadelphia: University of Pennsylvania Press.

———. 2000. Discourse Features of New Testament Greek: A Coursebook on the Information Structure of New Testament Greek. 2nd edition. Dallas: SIL International.

———. 2009. Self-Instruction Materials on Narrative Discourse Analysis. Online at http://www.sil.org/~levinsohns. Longacre, Robert E. 1989. Joseph: A Story of Divine Providence: A Text Theoretical and Textlinguistic Analysis of

Genesis 37 and 39-48. Winona Lake: Eisenbrauns.

———. 1994. Weqatal forms in Biblical Hebrew prose. In Biblical Hebrew and Discourse Linguistics, ed. by Robert

D. Bergen, 50-98. Dallas: Summer Institute of Linguistics.

Porter, Stanley E. 1992. Idioms of the Greek New Testament. Sheffield: JSOT Press.

Waltke, Bruce K., and M. O’Connor. 1990. An Introduction to Biblical Hebrew Syntax. Winona Lake, Indiana: Eisenbrauns.

Encoding of storyline events in Biblical Hebrew - Maxime Seveleu-Dubrovnik