Material Hebreo
Material Hebreo
Material Hebreo
SIL eBook 26
Bryan L. Harmelink
Bryan L. Harmelink
SIL e-Books
26
2011 SIL International, Inc.
ISBN: 978-1-55671-287-6
ISSN: 1934-2470
Fair-Use Policy:
Books published in the SIL e-Books (SILEB) series are intended for scholarly research and
educational use. You may make copies of these publications for research or instructional
purposes free of charge (within fair-use guidelines) and without further permission.
Republication or commercial use of SILEB or the documents contained therein is expressly
prohibited without the written consent of the copyright holder(s).
Editor-in-Chief
Mike Cahill
ABSTRACT
This dissertation seeks to identify the uses and functions of yh!yw+ ~ in biblical
Hebrew. After an introduction to the topic and a review of the treatment of yh!yw+ ~ in the
grammars of biblical Hebrew, the theoretical framework of the Functional, DiscoursePragmatic model of linguistic analysis implemented here is presented. This model
focuses on the interaction of linguistic forms and their functions, exploring the choices
made by language users and the effects of these choices on their communication.
Before proceeding with the display of the occurrences of yh!yw+ ,~ fundamental
aspects of the biblical Hebrew verbal system are discussed in order to establish the
context within which yh!y+w~ is analyzed. After presenting a preliminary distribution of
yh!yw+ ,~ the occurrences are displayed according to its Verbal and Temporal uses. This
detailed classification lays the foundation for consideration of the discourse-pragmatic
functions of yh!yw+ ,~ which are discussed in the final chapter.
This research was motivated by questions that arose from examples of yh!yw+ ~
encountered in the biblical Hebrew text and from the discussion of its functions in the
literature. The analysis of the verbal occurrences, representing 53% of the total instances,
demonstrates the syntactic connection of yh!yw+ ~ in these cases. This analysis shows that
yh!yw+ ~ has DEICTIC features that indicate its involvement in the systems of personal, spatial,
and temporal reference. The identification of these DEICTIC features of the verbal uses
provides a significant semantic and cognitive link to its functions in temporal
expressions.
ii
iii
CONTENTS
Abstract
List of Illustrations
List of Tables
List of Abbreviations
Chapter 1 Introduction
PART I: Theoretical Background
Chapter 2 Historical Overview
2.1
Introduction
2.2
2.3
Introduction
3.2
The Analysis of yh!yw+ ~ in the Traditional Approach
3.2.1
Gesenius, Hebrew Grammar, 1813
3.2.2
Mller, Outlines of Hebrew Syntax, 1883
3.2.3
Harper, Elements of Hebrew Syntax, 1892
3.2.4
Weingreen, A Practical Grammar for Classical Hebrew, 1939
3.2.5
Joon-Muraoka, A Grammar of Biblical Hebrew, 1991
3.2.6
Analytical Summary of the Traditional Approach
3.3
The Analysis of yh!yw+ ~ in the Descriptive Approach
3.3.1
Lambdin, Introduction to Biblical Hebrew, 1971
3.3.2
Andersen, The Sentence in Biblical Hebrew, 1974
3.3.3
Waltke and OConnor, An Introduction to
Biblical Hebrew Syntax, 1990
3.3.4
Kelley, Biblical Hebrew: An Introductory Grammar, 1992
3.3.5
Seow, A Grammar for Biblical Hebrew, 1995
3.3.6
Chisholm, From Exegesis to Exposition, 1998
3.3.7
Schertz and Yoder, Seeing the Text: Exegesis for Students of Greek
and Hebrew, 2001
3.3.8
Ross, Introducing Biblical Hebrew, 2001
3.3.9
Pratico and Van Pelt, Basics of Biblical Hebrew, 2001
3.3.10
Analytical Summary of the Descriptive Approach
iv
3.4
The Analysis of yh!yw+ ~ in the Textlinguistic Approach
3.4.1
Preliminary Comments
3.4.2
Schneider, Grammatik des biblischen Hebrisch, 1974
3.4.3
Richter, Grundlagen einer althebrischen Grammatik, 1980
3.4.4
Bartelmus, HYH. Bedeutung und Funktion eines hebrischen
Allerweltswortes, 1982
3.4.5
Longacre, Joseph: A Story of Divine Providence: A Text Theoretical
and Textlinguistic Analysis of Genesis 37 and 39-48, 1989
3.4.6
Niccacci, The Syntax of the Verb in Classical Hebrew Prose, 1990
3.4.7
Eskhult, Studies in Verbal Aspect and Narrative Technique
in Biblical Hebrew Prose, 1990
3.4.8
Talstra, A Hierarchy of Clauses in Biblical Hebrew Narrative
3.4.9
Winther-Nielsen, A Functional Discourse Grammar of Joshua.
A Computer-assisted Rhetorical Structure Analysis, 1995
3.4.10
Exter Blokland, In Search of Text Syntax, 1995
3.4.11
Endo, The Verbal System of Classical Hebrew in the Joseph Story:
An Approach from Discourse Analysis, 1996
3.4.12
Hatav, The Semantics of Aspect and Modality, 1997
3.4.13
Van der Merwe et al, A Biblical Hebrew Reference Grammar, 1999
3.4.14
Rocine, Learning Biblical Hebrew: A New Approach
Using Discourse Analysis, 2000
3.4.15
Analytical Summary of the Textlinguistic Approach
4.3
Summary
Introduction
6.2
The Verb in Biblical Hebrew
6.2.1
Approaches to the Verbal System of Biblical Hebrew
6.2.2
The Multi-Dimensional Verb Analysis Implemented in this Study
6.2.3
Description of the Hebrew Verb Forms
6.3
Clause Syntax in Biblical Hebrew
6.3.1
The Role of Waw in Clause Syntax
6.3.1.1 Introduction
6.3.1.2 Nominal Conjoining
6.3.1.3 Verbal Conjoining
6.3.1.4 Other Functions of w+
vi
6.4
Narrative Time and the Hebrew Verbal System
6.4.1
Time and WAYYIQTOL Sequences
6.4.1.1 PROGRESSION
6.4.1.2 EXPANSION
6.4.1.3 COMPRESSION
6.4.1.4 RAPID SUCCESSION
6.4.1.5 INCLUSION
6.4.1.6 REGRESSION
6.4.1.7 CONCLUSION
6.4.2
Summary
Introduction
7.2
The Distribution of yh!yw+ ~
7.2.1
Charting the Distribution
7.2.2
The Relationship of Distribution and Genre
7.3
Summary
Overview
8.2
Uses of yh!yw+ ~ as the Main Verb in Independent Clauses
8.2.1
The EQUATIVE Use of yh!y+w~
8.2.1.1 To be
8.2.1.2 And it was so /k@-yh!yw+ ~
8.2.1.3 And thats the way it happened yh!Yw\ ~
8.2.1.4 Territorial Expressions
8.2.1.5 The Formulaic Expression hw`hy+-rb^D+ yh!yw+ ~
8.2.1.5.1 Occurrences of hw`hy+-rb^D+ yh!yw+ ~ in Genesis, Samuel-Kings,
and Chronicles
8.2.1.5.2 Occurrences of hw`hy+-rb^D+ yh!yw+ ~ in Isaiah
8.2.1.5.3 Occurrences of hw`hy+-rb^D+ yh!yw+ ~ in Jeremiah
8.2.1.5.4 Occurrences of hw`hy+-rb^D+ yh!yw+ ~ in Ezekiel
8.2.1.5.5 Occurrences of hw`hy+-rb^D+ yh!yw+ ~ in Jonah
8.2.1.5.6 Occurrences of hw`hy+-rb^D+ yh!yw+ ~ in Haggai
8.2.1.5.7 Occurrences of hw`hy+-rb^D+ yh!yw+ ~ in Zechariah
8.2.1.5.8 Variations of the standard Occurrences of hw`hy+-rb^D+ yh!yw+ ~
8.2.1.5.9 Analytical Summary of the FORMULAIC Use of yh!yw+ ~
vii
8.5
Uses of Wyh=Yw] ~
8.5.1
Verbal Uses of Wyh=Yw] ~
8.5.1.1 EQUATIVE Uses of Wyh=Yw] ~
8.5.1.1.1 As Auxiliary Verb with Participle
8.5.1.1.2 Quantity
8.5.1.1.2.1 Basic Quantity Statements with Wyh=Yw] ~
8.5.1.1.2.2 All the days of
8.5.1.1.2.3 Period of time
8.5.1.1.3 With Prepositions
8.5.1.1.3.1 With ta@
8.5.1.1.3.2 With b=
8.5.1.1.3.3 With K= Indicating Comparison
8.5.1.1.3.4 With l= Indicating Possession
8.5.1.1.3.5 With l= Meaning became
8.5.1.1.3.6 With l= Meaning became and l= Indicating Possession
8.5.1.1.3.7 With l= Meaning for
8.5.1.1.3.8 With yn@pl
= !
8.5.1.1.3.9 With <u!
8.5.1.2 DEICTIC Uses of Wyh=Yw] ~
8.5.1.3 DESCRIPTIVE Uses of Wyh=Y]w~
8.5.1.3.1 DESCRIPTIVE with ADJECTIVE COMPLEMENT
8.5.1.3.2 DESCRIPTIVE with QAL PASSIVE
8.5.1.4 Summary of the Uses of Wyh=Yw] ~
Overview
9.2
Introduction to Temporal Expressions in Biblical Hebrew
9.2.1
The Syntax of Temporal Expressions in Hebrew Grammars
9.2.1.1 Introduction
9.2.1.2 GKC, Gesenius Hebrew Grammar
9.2.1.3 Davidson, Introductory Hebrew Grammar ~ Syntax
9.2.1.4 Williams, Hebrew Syntax: An Outline
9.2.1.5 Joon-Muraoka, A Grammar of Biblical Hebrew
9.2.1.6 Lambdin, Introduction to Biblical Hebrew
9.2.1.7 Waltke and OConnor, Introduction to Biblical Hebrew Syntax
9.2.1.8 Pratico and Van Pelt, The Basics of Biblical Hebrew
9.2.1.9 van der Merwe et al, Biblical Hebrew Reference Grammar
9.2.1.10 Summary
9.2.2
Theoretical Background for the Analysis of Temporal Expressions
9.2.2.1 Speaker Deixis
9.2.2.2 Word Order and Narrative Strategies
9.2.2.3 Summary
ix
9.3
Temporal Expressions with yh!yw+ ~
9.3.1
With rj^a/^ yr\ja
& ^
9.3.1.1 After these things
9.3.1.1.1 Followed by WAYYIQTOL
9.3.1.1.2 Followed by QATAL
9.3.1.1.3 Followed by WE-X-QATAL
9.3.1.1.4 Analytical Summary of the After these things Use of yh!yw+ ~
9.3.1.2 After: rj^a^ yh!yw+ ~
& ^ yh!yw+ ~ + NOUN
9.3.1.2.1 After: yr}ja
9.3.1.2.2 After: yr}ja
& ^ yh!yw+ ~ Followed by a VERB
9.3.1.2.2.1 Infinitive Construct
9.3.1.2.2.2 QATAL
9.3.1.3 With /k@-yr\ja
& ^
9.3.1.3.1 Followed by WAYYIQTOL
9.3.1.3.2 Followed by QATAL
9.3.1.4 Summary of the Uses of yr\ja
& ^ yh!yw+ ~
9.3.2
With Prepositions
9.3.2.1 With B=
9.3.2.1.1 Occurrences of yh!yw+ ~ with B= + Infinitive Construct
9.3.2.1.1.1 Infinitive Construct Followed by WAYYIQTOL
9.3.2.1.1.2 Infinitive Construct Followed by WAYYIQTOL
with Intervening Clauses
9.3.2.1.1.3 Infinitive Construct Followed by QATAL
9.3.2.1.1.4 Summary of Infinitive Constructs Used with -B= yh!yw+ ~
9.3.2.1.2 Occurrences of B= yh!yw+ ~ with Specific Temporal Reference
9.3.2.1.2.1 Specific Temporal Reference Followed by WAYYIQTOL
9.3.2.1.2.1.1 With rq#Bb
) ^ / br\ub
# *
9.3.2.1.2.1.2 With awh!h^ tu@B* / ayh!h^ tu@B*
9.3.2.1.2.1.3 Infinitive Construct + tu@B= yh!yw+ ~
9.3.2.1.2.1.4 With aWhh^ <oYB^
9.3.2.1.2.1.5 With aWhh^ hl*yL
+ B
^ ^
` ^
9.3.2.1.2.1.6 With <h@h* <ym!YB
9.3.2.1.2.1.7 With <h@h* <yB!rh
~ * <ym!Yb
` ^
9.3.2.1.2.1.8 With NUMBER-h^ <oYb^
9.3.2.1.2.1.9 With hn`VB
* ^ + -h^ + NUMBER
9.3.2.1.2.1.10 With <oyB= + INFC
9.3.2.1.2.1.11 With tyu!bV
! B
= ^
9.3.2.1.2.1.12 At a Specific Time
9.3.2.1.2.1.13 With vd\jb
) = _____
9.3.2.1.2.1.14 With ym@yB!
9.3.2.1.2.2 Specific Temporal Reference Followed by QATAL
x
Introduction
10.4
xii
xiii
ILLUSTRATIONS
Illustration
Page
xiv
TABLES
Table
Page
xvi
ABBREVIATIONS
BDB .......................................Brown, F., S. R. Driver, and C. A. Briggs. 1996. The
Brown-Driver-Briggs Hebrew and English Lexicon: With
an Appendix Containing the Biblical Aramaic. Peabody,
MA: Hendrickson.
BHRG ....................................van der Merwe, C. H. J., J. A. Naud, and J. H. Kroeze.
1999. A Biblical Hebrew Reference Grammar. Sheffield:
Sheffield Academic Press.
GKC .......................................Gesenius, W. and E. Kautsch. 1910. Gesenius Hebrew
Grammar. Trans. and rev. by A. E. Cowley. 2d ed. Oxford:
Clarendon Press.
HALOT ..................................Koehler, L. and Baumgartner, W. 1995. The Hebrew and
Aramaic Lexicon of the Old Testament. Leiden: Brill.
JPS..........................................JPS Hebrew-English TANAKH. 2000. Philadelphia: The
Jewish Publication Society.
NASB.....................................New American Standard Bible. 1995. Updated ed.
Anaheim, CA: Lockman Foundation.
NIV ........................................New International Version. 1984. International Bible
Society. Grand Rapids: Zondervan.
NRSV.....................................New Revised Standard Version. 1989. Division of
Christian Education of the National Council of the
Churches of Christ in the U.S.A.
xvii
CHAPTER 1
INTRODUCTION
This is a study of yh!yw+ ~ one of the elements involved in the well-crafted ways in
which the biblical Hebrew text depicts the temporal dimensions of narrative. In The Art
of Biblical Narrative, Alter asks the following insightful questions:
Why at a particular juncture does the narrator break the time-frame of his
story to insert a piece of expository information in the pluperfect tense, or
to jump forward to the time of his contemporary audience and explain that
in those days it was the custom in Israel to perform such and such a
practice? Why does he pause to make a summarizing statement about the
condition of a character, as, for example, in the observation about Josephs
already established viceregal status just as the ten brothers arrive in
Egypt? Why at certain points is the regular rapid tempo of narration
slowed down to take in details of a kind for which in general no time is
allowed? (Alter 1981, 184-85)
Understanding how time is depicted is an important part of reading narrative, as
Fokkelman comments, The narrator is not obliged to provide indications of narrated
time, but he is quite capable of it. Whenever we hear this kind of information it is always
important (Fokkelman 1999, 36). It is possible to know that indications of narrated time
are important, but not really know what significance they have in the flow of the text.
For example, what is the function of a shift ahead in time or a reference to a previous
time? Many of the same questions raised by literary or narrative studies are of interest
here. The difference is that this study uses textlinguistic methods to analyze yh!yw+ ~ in the
1
2
pursuit of answers to these questions. The goal of this inductive, text-based study is not
to merely derive syntactic formulas or structural representations of the text, but rather to
use linguistic methods to explore the junctures and pausessuch as those referred to by
Alterto better understand the temporal shape of the text. So, in one sense this study is
about yh!yw+ ~ because it is the element in focus, but in another sense yh!yw+ ~ is only a small part
of what motivated this research into the temporal organization of text in biblical Hebrew.
The opening chapter of 1 Samuel provides many good examples of the types of
questions explored in this research. One of the first questions that emerges from reading
1 Sam 1:1-3 has to do with the fact that the book starts with yh!yw+ .~ Does this have any
significance for the whole book or is its function restricted to the first clause? Notice also
that yh!yw+ ~ occurs again in 1:2. What, if any, significance should be associated with two
occurrences in such close proximity?
<y]rp
` a
= # rh^m@ <yp!ox <y]tm
^ * rh
` -* /m! dj*a# vya! yh!yw+ ~ 1
.yt!rp
` a
= # [Wx-/b# WjT)-/B# aWhyl!a-$ /B# <j*ry{ -+ /B# hn`ql
* a
= # w{mv=W
hN`nP
] = tyn]Vh
@ ^ <v@w+ hN`j^ tj^a^ <v@ <yv!n` yT@v= w{lw+ 2
.<yd]ly* + /ya@ hN`jl
^ W= <yd]ly* + hN`np
] l
= ! yh!yw+ ~
hm*ym!y` <ym!Ym
` ! w{ryu!m@ aWhh^ vya!h* hl*uw* + 3
hOv!B= toab*x= hw`hyl^ j~Bz) l
+ w! + tw{jT
& v
^ h
= l
! =
.hw`hyl^ <yn]hK
& ) sj*np
+ W! yn]pj
= * yl!u-@ yn}b= yn}v= <v*w+
The next questions have to do with the four occurrences of w+ with a non-verbal
item. What is the function of w{mv=W, w{lw+, hN`jl
^ W= , and <v*w?+ How do these items interact
with the temporal organization of the opening of 1 Samuel? Where does the action of
1 Samuel begin? For example, what is the proper way to interpret the verb hl*uw* + in 1:3?
3
What is the relationship between this WEQATAL and the previous two occurrences of
yh!yw+ ?~
.tonm* h*yt#onb=W h*yn\B-* lk*lW= w{Tv=a! hN`np
] l
= ! /t^nw` + hn`ql
* a
= # jB^zY+ w] ~ <oYh^ yh!yw+ ~ 4
.Hm*jr
= ~ rg~s* hw`hyw~ bh@a* hN`j-^ ta# yK! <y]Pa
* ^ tj^a^ hn`m* /T@y] hN`jl
^ W= 5
In 1 Sam 1:4 another yh!y+w~ occurs, but this time with <oYh^. To what day or time
does the expression <oYh^ yh!yw+ ~ refer? Immediately following this is the first WAYYIQTOL
jB^zY+ w] ~ (other than the three yh!yw+ )~ , but this is quickly followed by /t^nw` ,+ raising questions
about how the preceding WAYYIQTOL should be read. The questions continue in 1:5, first
of all with /T@y] hN`jl
^ W= , which is in some way connected to /t^nw` ;+ secondly, the temporal
reference of the clause bh@a* hN`j-^ ta# yK! must be determined; and, finally in 1:5, the WEX-QATAL
rg~s* hw`hyw~ requires interpretation of the temporal nature of the QATAL as well as
4
textlinguistic understanding of its functions, recommending that yh!yw+ ~ be left
untranslated.1 Other descriptions of yh!yw+ ~ discuss its role in the text, but are limited in
scope. The most significant recent study specifically of yh!yw+ ~ is van der Merwes The
Elusive Biblical Hebrew Term yh!yw+ :~ A Perspective in Terms of its Syntax, Semantics, and
Pragmatics in 1 Samuel, but, as its title indicates, the study is limited to 1 Samuel.
The impetus for this study initially came from an even more limited corpus: the
book of Jonah. During a graduate course which focused on the translation of the Hebrew
text of Jonah, the five occurrences of yh!yw+ ~ found therein were discussed. It quickly
became apparent that the different occurrences of yh!yw+ ~ could not be merely dismissed and
left untranslated and it was clear that the syntactic environment and the pragmatic factors
of each occurrence required careful consideration. Increased curiosity led to looking into
the use of yh!yw+ ~ in Genesis, which then motivated the formulation of further questions.
The occurrences of yh!yw+ ~ in the Hebrew Bible can easily be found by computer
programs which allow this type of search; in a matter of seconds all the occurrences can
be displayed. In the attempt, however, to discern the parameters which motivate the use
of yh!yw+ ~ in the biblical text, this type of data-display is essentially useless since the
occurrences are extracted from the context in which they occur. For the type of analysis
carried out here, it was important that each occurrence of yh!yw+ ~ be encountered in its
For example, Ross (2001, 139-40) in Introducing Biblical Hebrew states: In older
Bible versions, the form yh!yw+ ~ was translated and it came to pass. Although the form can
sometimes be the main verb in a clause, it usually indicates simply that the narrated events
occurred in the past, and so the form need not be translated.
5
context, rather than merely listed in concordance-fashion. Repeated readings of extensive
sections of the Hebrew Bible, carefully taking note of its temporal organization, has
formed the basis of the observations and analysis summarized here.
If the main Hebrew grammars are consulted, two basic approaches to yh!yw+ ~ are
found which can be broadly characterized as follows: 1) yh!yw+ ~ means and it came to pass
and should be left untranslated because it is unnatural and clumsy in English; 2) yh!yw+ ~ is a
macrosyntactic marker which indicates simply that the narrated events occurred in the
past (Ross 2001, 139-40).
One of the questions that immediately arises is whether these approaches are
adequate or not. Regarding the first approach, there are numerous unnatural and clumsy
peculiarities in biblical Hebrew, as in any language, but this is certainly not sufficient
reason to leave them untranslated. Regarding the second approach, which appears to have
a degree of textlinguistic sensitivity, the question is whether markers that simply indicate
a past tense frame of reference even exist. If they do exist, would this adequately describe
how yh!yw+ ~ functions? Also, how can the uneven distribution of yh!yw+ ~ be explained? How
can the absence of yh!yw+ ~ be explained in contexts which are unambiguously past? If the
function of yh!yw+ ~ is to indicate past tense and certain past-tense narratives do not have any
occurrences of yh!yw+ ,~ how is its absence explained?
Within both approaches, an important question is whether every occurrence of
yh!yw+ ~ should receive the same treatment. Should certain occurrences be ignored as if they
were mysteriously in the consonantal text for no reason? The recommendation here is
6
not to resort to a mechanical rendering of every yh!yw+ ~ as and it came to pass
characteristic of a version like the NASB, but rather to carefully analyze the syntactic and
pragmatic parameters of the use of yh!yw+ ~ in the text in order to develop sensitivity to its
varied uses. To merely label yh!yw+ ~ as a discourse marker and then merely acknowledge its
presence as an indicator of past tense is to greatly underestimate the benefit that can come
from a thoroughgoing linguistic approach to the text. After the next chapter, which
outlines the current state of research on yh!yw+ ,~ the conceptual foundation for this linguistic
approach will be presented.
And finally, in these introductory remarks, it needs to be made clear that this is
not a mere academic exercise. To some, the way yh!yw+ ~ is handled in grammars of biblical
Hebrewor even how it is rendered in translationmay seem inconsequential at best.
As a linguist and translator, however, this is intolerable! When this study was in the early
stages as a mere idea, the implications of how yh!yw+ ~ is dealt with in translation seemed
significant; after processing and pondering all the data, awareness of the significance of
how yh!yw+ ~ is handled in translation has only increased.
CHAPTER 2
HISTORICAL OVERVIEW
2.1 Introduction
Good scholarship requires that any potentially useful procedure, approach, or
technique be used in the attempt to understand the complexities of the Hebrew text as
well as possible. If certain linguistic models are proving themselves to be beneficial in
the research of other languages, full consideration should be given to the implementation
of these models with their accompanying insights in ongoing research in biblical Hebrew.
Previous eras of scholarship should also be considered, even if the same type of
analytical tools of current research were not employed.
Notice that this is a textual study, which is the reason for the extensive text-incontext examples. It is recommended that all the examples be read and processed in each
section. This is the best way for the textual examples to make their own case. It is also
recommended that constant reference be made to the biblical Hebrew text from which the
examples are taken. However, if the readers time does not permit systematic study of all
the examples, each section includes a summary of the most pertinent examples.
10
Saussures Cours de linguistique gnrale is representative of the shift from this
diachronic perspective to a focus on synchronic studies. Saussure, commenting on the
priority of synchronic over diachronic analyses, stated that [t]he first thing that strikes us
when we study the facts of language is that their succession in time does not exist insofar
as the speaker is concerned (Saussure 1916, 81). Saussures interest was in how
language is actually used by speakers at a given period in time.
Linguistics was also heavily impacted by ethnography in the early twentieth
century. In American Linguistics, in particular, the work of researchers such as Edward
Sapir (1921) and Franz Boas (1911) was foundational. In contrast to the diachronic
model which had assumed that all languages were developing toward some ideal
structure, the characteristic of the school founded by Boas was its relativism. There was
no ideal type of language, to which actual languages approximated more or less closely
(Sampson 1980, 59). As a result, one of the hallmarks of American Linguistics is the
principle that every language is to be studied and described in its own right.
Another fundamental characteristic of Descriptive Linguistics as it developed
during the twentieth century is its focus on working from the data. As Leonard
Bloomfield stated, [t]he only useful generalizations about language are inductive
generalizations (Bloomfield 1933, 20). Within this perspective, linguistic terms are
defined in relation to each other in order to derive the linguistic system from the language
being studied. As Lyons comments:
each language is regarded as a system of relations (more precisely, a set of
interrelated systems), the elements of whichsounds, words, etc.have
no validity independently of the relations of equivalence and contrast
which hold between them. (Lyons 1969, 50)
11
In the middle of the twentieth century, a theoretical revolution took place that has
had an immense impact on the field of Linguistics. As Robins comments:
What is probably the most radical and important change in direction in
descriptive linguistics and in linguistic theory that has taken place in
recent years may be located in 1957, when Chomskys Syntactic
Structures was published, inaugurating the transformational-generative
phase of linguistics. (Robins 1968, 226)
Under the influence of Chomskys publications and teaching, the
transformational-generative school developed in directions not shared by others in the
field of descriptive linguistics. One of the basic conceptual distinctions in Chomskys
view is between competence and performance. In essence, this distinction represents the
two main schools that developed. Chomskys transformational-generative school focused
on competence, i.e., the more abstract, mental conception of language, whereas those not
involved in transformational-generative approaches typically focused more on
performance, i.e., actual speech.
Additionally, it was characteristic of Chomskys school to emphasize grammar at
the level of the sentence, seen in the phrase structure rules and transformations. The
sentence is certainly a valid level of linguistic analysis, but it has typically been out of the
approaches whose focus has been on actual speech and data that the more functional,
language-in-use models of discourse analysis have developed. In fact, Brown and Yules
brief definition of discourse analysis is simply that it is the study of language in use
(Brown and Yule 1983, 1), highlighting the communicative functions of language.
In contrast to what was happening in the Chomskyan school, the field research of
Pike, Grimes, and Longacre constrained them to develop models of linguistic analysis
12
that were descriptive in the sense of being inductive and focused on data obtained in
actual language-use situations (Pike 1967; Grimes 1974; Longacre 1996). Their models
and theoretical concepts were also developed with a fundamental awareness of the
context-sensitive nature of language. Pikes Language in Relation to a Unified Theory of
the Structure of Human Behavior in particular emphasizes the interrelated nature of
language and human behavior in general in the model known as Tagmemics.
In the shadow of Chomskys influence, there was significant conceptual
development in other schools of Linguistics that was laying an important theoretical
foundation for later models of language use. The work of Pike, Grimes, and Longacre
may not be the primary theoretical predecessor to later models, but their work is
representative of certain conceptual trends that would later converge with broader
currents of discourse analysis carried out within a functional approach to language.
One of the common denominators in the work of Pike, Grimes, and Longacre is
that their theoretical approaches were developed as they themselves were directly
involved in the study and analysis of a wide variety of non-Indoeuropean languages.
Their models were developed with a vivid awareness of language use in context. Also,
during the mid- to late-twentieth century, there was an increasing awareness in general
Linguistics of the social dimension of Language. Labovs Sociolinguistic Patterns (1972)
and Hymes Foundations in Sociolinguistics: An Ethnographic Approach (1974) are two
of the seminal publications in the field of Sociolinguistics, which is now one of the main
sub-fields within general Linguistics, with its particular concerns, methods and
objectives. The impact of Sociolinguistic studies has been such that many areas of
13
linguistic study have developed a greater sensitivity to the many dimensions of language
use.
Schools of Linguistics which are functional in approach are also the result of
increased attention to how language is used. Halliday (1973), Foley and Van Valin
(1984), and Givn (2001) are some of the main theoreticians in Functional Linguistics.
There are, of course, many nuances in the individual approaches, but the
theme unifying the various functional approaches is the belief that
language must be studied in relation to its role in human communication.
Language is thus viewed as a system of human communication, rather
than as an infinite set of structural descriptions of sentences. (Foley and
Van Valin 1984, 7)
Not only has the field of Linguistics been impacted by functional, language-use
models, but there has also been significant development of cognitive approaches. As
mentioned above, the important distinction between performance and competence gave
rise to approaches and schools which focused respectively on language use and the
mental aspects of language. It is not surprising, then, that in addition to the development
of functional approaches which tend to focus on performance and language use, a variety
of cognitive approaches have developed which are more directly concerned with matters
of competence defined as the human capacity to use and interpret language. Sperber and
Wilson (1995), Lambrecht (1994), Fauconnier (1985), Lakoff and Johnson (1980), and
Jackendoff (1994) are principal scholars promoting cognitive models of linguistic study.
In summary, the attempt to adequately account for the human linguistic capacity
drives linguistic theoreticians and analysts to keep refining their models and methods.
14
This brief sketch has highlighted only the most rudimentary concepts which underlie the
analysis implemented in the present study:
1) Descriptive, with a conceptual basis shaped by Tagmemics
2) Functional in its commitment to the study of language use in context; and
3) Cognitive in its attention to the interpretive processes involved in
understanding communication.
15
modern Linguistics, but the analytical trends and approaches eventually find their way
into the methodologies employed by biblical Hebrew scholars. Also, it is common for
vestiges of previous eras to be found in the research of a later era, especially in the period
of time prior to the more complete paradigm shift.
It is important to recognize that the early publications on biblical Hebrew
represent an early formative period of grammatical studies. This does not invalidate the
insights found in these publications, but rather is a reminder that the awareness of explicit
grammatical elements of biblical Hebrew has a long history. Khan, in his contribution to
the volume Hebrew Study from Ezra to Ben-Yehuda, makes the following pertinent
comment:
It can be said that grammatical elements are found in the early
masoretic tradition, which may go back as far as the Talmudic period. It is
important to notice, however, that the existence of these elements of
grammatical thought should not lead us to define the general activity of
the Masoretes of this period as grammar. The main purpose of their work
was still to preserve the text of Scripture rather than investigate the rules
of the language of Scripture. The use of grammatical categories was
ancillary to this purpose. (Khan 1999, 193)
Just as one can trace linguistic concepts like word classes back to the writings of
Aristotle, there is a long history of reflection on Hebrew. What one finds in later periods,
however, is a more conscious implementation of specific methods of analysis and
inquiry. Waltke and OConnors helpful chapter on the History of the Study of Hebrew
Grammar, discusses two periods of Hebrew study during which there was an everincreasing development of the grammatical analysis of Hebrew:
1) Medieval Jewish Studies (11th to 16th centuries)
16
2) Christian Hebrew Studies (16 to mid-18th centuries) (Waltke and OConnor
th
1990, 31-43)
The period of Medieval Jewish Studies has been referred to as the creative
period during which many issues of Hebrew morphology and grammar were debated
and refined. Scholars from this period typically drew upon knowledge of Arabic to help
describe areas of Hebrew grammar (Tene 1971, 1358).
The ad fontes drive of the Enlightenment gave rise to a new interest in the
classical languages, resulting in increased interest among humanists in the study of
Hebrew. Reuchlin is representative of this era, whose 1506 publication, Rudimenta
linguae hebraicae (1506), is representative of the shift, not only from Jewish to Christian
scholars, but also from Arabic to Latin as the language by which Hebrew was evaluated.
Reuchlins work, as well as the work of subsequent scholars, sets the stage for the
publication of the traditional grammars of biblical Hebrew, of which reprinted and reedited versions are still in use today. It is no accident that the study of biblical Hebrew in
the 18th and 19th centuries is characterized by the same kind of diachronic, developmental
conceptions that were typical of the study of language in general. As intellectual trends
shifted, so also did the conceptual basis for the study of Hebrew.
Subsequent to the periods discussed by Waltke and OConnor, the more recent
study of biblical Hebrew could be divided into three broad eras:
17
Traditional
Blau
Davidson
GKC
Joon-Muraoka
Kimhi
Weingreen
Descriptive
Andersen
Garrett
Kelley
Lambdin
Pratico and Van Pelt
Ross
Seow
Waltke and OConnor
Textlinguistic1
Buth
Long
Longacre
Putnam
Richter
Schneider
Talstra
van der Merwe
These refer to the authors of published grammars and monographs, which are
representative of the field in general. Other important monographs are not included in this list
because they do not deal with areas of analysis pertinent to the current study of yh!y+w.~
2
One approach to Hebrew studies that is not explicitly mentioned here is the
Comparative-Historical method which focuses on the reconstruction of earlier stages of the
Semitic languages. The main period of use of Comparative-Historical methods in biblical Hebrew
coincides with the Traditional and Descriptive eras described here, and definite influences and
perspectives are currently operative. A Textlinguistic approach does not necessarily conflict with
the principles or findings of comparative Semitics, but the synchronic focus characteristic of
Textlinguistic approaches would see Comparative-Historical considerations as part of the process
by which the language used in the text came into being. Once the text has been produced, the
diachronic processes are no longer in the purview of the analyst whose focus is on the final form
of the text.
18
The boundaries between these models are fuzzy in the sense that there is a
continuity and overlap of research that links them to each other. The shift from the
Traditional model to the Descriptive did not discard all previous research and start over;
in similar fashion, the Textlinguistic model does not ignore previous analysis, but rather
grows out of it. But the very nature of scientific revolutions is such that a shift in
paradigm implies that certain elements and perspectives from previous eras will not
necessarily be carried over. One of the motivations for paradigm shifts is an increasing
dissatisfaction with the prevailing model of research, which eventually leads to a major
shift.
The current state of affairs in the study of biblical Hebrew finds most scholars
employing the Descriptive approach, with strong conceptual connections to the
Traditional era. There are a number of scholars who employ textlinguistic insights or
terminology, but the core conceptual framework of their research is still within the
Descriptive model. An increasing number of scholars work within the Textlinguistic
approach, but a complete paradigm shift has not yet happened for the field of biblical
Hebrew studies in general.
It is time, however, to declare an end to the period of infancy referred to by
Waltke and OConnor (1990, 55). The flood of recent publications is indicative of the
maturing status of Textlinguistic or discourse studies of biblical Hebrew.
CHAPTER 3
SURVEY OF SCHOLARSHIP ON yh!y+w~
3.1 Introduction
yh!yw+ ~ has certainly not gone unnoticed in the history of research on biblical
Hebrew. The purpose of the following sections is to review and provide a summary of
the various descriptions of yh!yw+ ~ in these grammars and monographs. These grammars
and monographs are not studies of yh!yw+ ,~ but their descriptions of yh!yw+ ~ are representative of
the main ways in which yh!yw+ ~ has been analyzed. The three categories: Traditional,
Descriptive, and Textlinguistic discussed in the previous chapter are the main divisions
here. The approach one takes to Hebrew in generaland to the verbal system in
particularwill affect how yh!yw+ ~ is described. The focus of the discussion here is the way
yh!yw+ ~ is presented in the publications considered, with some critical analysis when it is
relevant.
To review an assortment of introductory grammars, scrutinizing how they deal
with yh!yw+ ,~ could be perceived as somewhat unfair. Some might correctly argue that an
introductory grammar cannot be expected to thoroughly describe every aspect of the
language being presented. Writing a grammar for instructional purposes involves an
19
20
agonizing process of selection, as the author seeks to explain the most pertinent
information in an efficient and pedagogically intuitive way. If the discussion of
something like yh!yw+ ~ does not receive extensive attention in an introductory grammar, that
is to be understood. It is not, however, the presence or absence or even the length or
brevity of comments regarding yh!yw+ ~ that are of interest here; the intention here is to
review the comments to discern the analytical perspective employed by the author(s). All
of this is done with the singular intent of working toward a better understanding of yh!yw+ .~
Any omissions and/or misrepresentations are the responsibility of the author of this study.
21
narratives beginning with any expression of time are not specific enough to provide a
clear picture of what yh!yw+ ~ is doing.
22
usage of yh!yw+ ~ in the earlier books indicates an awareness of possible diachronic factors in
its pattern of usage; and 3) the remarks about the form of the following verb mention the
context around yh!yw+ ,~ but there is no further delineation of the parameters of usage for
these verb forms. The syntactic connection of yh!y+w,~ certain diachronic considerations,
and verb form patterns with yh!yw+ ~ are all issues that are discussed in greater detail in
subsequent chapters.
23
contemporary with that time period. Close reading of Joon-Muraoka reveals many
characteristics that place it within the Traditional approach. The retention of Latin
translations in the following discussion of yh!yw+ ~ is telling:
Stative verbs present no particular difficulty; thus yh!yw+ ,~ used in the stative
sense, is equivalent to hy`h` and means et erat (it was), et fuit (it has
been). Used in the active sense, it is equivalent to hy`h` of action and
usually means et evEnit (it happened), et factum est (it came to pass);
sometimes, by misuse, both eveniebat (it would happen) and fiebat (it
would come to pass). (Joon-Muraoka 1991, 390)
The clearest indication of the perspective employed by Joon and retained by
Muraoka is the evaluative term misuse, which betrays a more prescriptive view typical
of the Traditional era rather than a fully descriptive view of Hebrew as a language on its
own terms.
In the discussion of the uses of the WAYYIQTOL form, Joon-Muraoka states that it
is sometimes used with the force of the French imparfait, i.e. frequentative action in the
past, but this use is irregular and improper (Joon-Muraoka 1991, 393). With specific
reference to yh!yw+ ,~ Joon-Muraoka states that this improper use is mainly found with
yh!yw+ ~ (Joon-Muraoka 1991, 394). The correct form according to Joon-Muraoka, for
example in Num 10:35 and 2 Sam 15:2, would be hy`hw* .+ These examples will be
discussed later; the point here is that Joon-Muraokas description of yh!yw+ ~ as an
introductory formula is within a traditional evaluative framework.
24
more recent approaches are already found in these publications, but consistent with the
general approach to language in that time period, the main concern is with presenting
representative cases or examples of the grammatical categories in use. Also typical of
these publications is the evaluation of yh!yw+ ~ as sometimes improper or as an example of
misuse.
25
examples start with hy`h,* which is not attested anywhere in biblical Hebrew. Lambdins
concern in these examples, however, is not in explicating the function of yh!yw+ ;~ his
concern is with the form of the verb that follows. For Lambdin, the mere presence of the
verb hy`h* seems to be of more consequence, stating that the verb hy`h* in a leading clause
requires special consideration (Lambdin 1971, 279). Unfortunately, even though
Lambdin demonstrates an awareness of certain syntactic parameters, understanding of
yh!yw+ ~ as a temporal modifier is not significantly advanced by Lambdins analysis.
26
mentioning that there are other options for initiating story-level episodes, nothing more is
discussed to more precisely define the use and function of yh!yw+ .~
27
The narrative use of imperfects with vav consecutive became so
commonplace that they were often used in this sense even without a
preceding governing perfect, especially with the imperfect forms of the
verb hy`h,` he was. (Kelley 1992, 210-11)
This comment reveals a perspective which is much more far-reaching than the
analysis of yh!yw+ .~ The implications of this comment are that the occurrence of a linguistic
entity such as yh!yw+ ~ has little or no functional motivation. By some type of developmental
process, the frequency of the WAYYIQTOL form is apparently so overpowering that certain
verbs begin to be used in new ways. This analysis is based on the assumption that the
proper pattern is that the WAYYIQTOL follows a preceding governing perfect and
indicates either consequence or sequence. This is characteristic of the waw consecutive
view which must explain why a WAYYIQTOL, which by definition is sequential, is found
without a governing perfect. The use of yh!yw+ ~ is only marginally in focus here, since
Kelleys concerns are more related to issues of governing sequences. Understanding of
yh!yw+ ~ is advanced very little by Kelleys grammar.
28
some event introduced by rv#aK
& ^ (even as) or yK! (when) or the like. (Seow
1995, 231)
One of the ways that this comment differs from the other Descriptive analyses
presented up to this point is its statement that yh!yw+ ~ typically introduces a past event. This
may reflect awareness of some of the early textlinguistic studies, such as the 1982 study
of hy`h* by Bartelmus (see 3.4.4 below) in which he discusses the function of yh!yw+ ~ as an
indication of past tense narrative. The specific ways in which yh!yw+ ~ is used, however, are
not addressed in any more detail beyond the citation above.
29
Chisholm makes one further comment about the use of yh!yw+ ~ to introduce a
parenthetical note:
Occasionally a wayyiqtol form, especially yh!yw+ ,~ introduces a parenthetical
note in the narrative. For example, 1 Kings 18:3 states that Ahab
summoned his palace administrator Obadiah. Verse 5 then records Ahabs
orders to Obadiah. But in between the narrator places a parenthetical note
about Obadiahs loyalty to the Lord (v. 4). This parenthesis is introduced
by yh!yw+ .~ For another example see 1 Chronicles 11:6, which inserts
parenthetical information about Joabs role in the conquest of Jerusalem.
(Chisholm 1998, 122-23)
This comment is indicative of the broad range of functions that are attributed to
yh!yw+ ~ in the literature. This is not the place for detailed discussion of this comment, but it
seems confusing to claim that yh!yw+ ~ can have what appear to be quite contradictory
functions. How can yh!yw+ ~ both begin a new narrative as well as introduce a parenthetical
note? These functions are discussed in Chapter 10, The Discourse-Pragmatic Uses of
yh!yw+ .~
3.3.7 Schertz and Yoder, Seeing the Text: Exegesis for Students of
Greek and Hebrew, 2001
Schertz and Yoders Seeing the Text: Exegesis for Students of Greek and Hebrew
is similar to Chisholms volume in both its purpose and perspective. The authors remark
regarding yh!yw+ ,~ that the pattern of yh!yw= ~ followed by a subordinate clause is a very
frequent boundary marker in Hebrew narrative (Schertz and Yoder 2001, 77). The
question that immediately arises is what type of boundary is marked in this way?
Schertz and Yoder also describe yh!yw+ ~ as an example of a transition construction that
signals the end of one region and the beginning of another (Schertz and Yoder 2001,
30
76). What is unclear from this descriptionas well as from the preceding discussionis
whether yh!yw+ ~ is signaling the end of what precedes it, signaling the transition to what
follows, or both. Once again, without more precisely defining the textual regions in
which yh!yw+ ~ operates, the function of yh!yw+ ~ will continue to elude readers and analysts.
31
Infinitive construct does not express time by itself. The time must be
determined from context. One way that the time of the clause is expressed
is with forms of hy`h* + wAw consecutive: yh!yw+ ~ indicating past time, and
hy`hw* + future time. (Ross 2001, 163)
In contrast to some previous grammars, Ross makes no mention in these
comments about yh!yw+ ~ s connection to the preceding or following narrative. Of primary
concern to Ross, it appears, is the function as temporal indicator. This actually goes hand
in hand with the recommendation to leave these forms of hy`h* untranslated. If yh!yw+ ~ and
hy`hw* + are doing nothing more than indicating past and future time respectively, why
should they be translated? It is interesting to notice that in the current analytical milieu
which tends to favor analysis of the Hebrew verbal system in terms of aspectual
distinctions rather than tense, that the mere yh!yw+ ~ and hy`hw* + forms of hy`h* are so
unambiguously assigned the function of indicating tense. It is true that infinitive
constructs do not indicate tense on their own, but is indicating tense really the function
that yh!yw+ ~ and hy`hw* + perform when they occur with infinitive constructs? Further
discussion of this question is found in Chapter 10.
32
The specific function attributed to yh!yw+ ~ is again that of temporal modifier,
indicating past tense. It is evident in this comment that Pratico and Van Pelt are also
concerned with what follows yh!yw+ ~ as seen in their remarks regarding narrative sequence.
The frequent use of yh!yw+ ~ with temporal clauses is also discussed in this grammar:
The form yh!yw+ ~ may also appear at the beginning of a temporal clause
within the sequence. When beginning a temporal clause, yh!yw+ ~ is frequently
& ^ and the whole
followed by a preposition or conjunction like K= or rv#aK
construction may be translated as and when. Words that designate time
are commonly a part of this type of construction. (Pratico and Van Pelt
2001, 196)
As temporal modifiers, in Pratico and Van Pelts view, yh!yw+ ~ and hy`hw* + are best left
untranslated, as seen in the following comment:
Because of their frequency in certain contexts, the temporal modifiers yh!yw+ ~
and hy`hw` + are best not translated in most occurrences, though you can still
translate the conjunction as and. Some will suggest, however, that yh!yw+ ~
be translated and it came to pass that and that hy`hw` + be translated and it
will be that. Given the frequency with which these temporal modifiers
will sometimes appear in a narrative sequence, however, it is often best to
avoid these translations in the interest of good English style. (Pratico and
Van Pelt 2001, 202)
In the interest of good English style, few would argue for always retaining the
and it came to pass that and the and it will be that renderings mentioned here.1
However, should good English style be the determining factor in translation decisions
like this? According to Pratico and Van Pelt, yh!yw+ ~ and hy`hw* + are best not translated
because of their frequency in certain contexts. Should the frequency of an item like this
play a decisive role in translation practice? The intention here is not to argue for the and
1
33
it came to pass that translation value, but rather to critically evaluate the implications
and ramifications of such a recommendation. As stated previously, this will be discussed
in more detail in Chapter 10.
34
biblical Hebrew. Certain essentially descriptive grammars may employ terminology used
in textlinguistic studies, just as a grammar that is fundamentally textlinguistic may have
terms and descriptions that are, for all practical purposes, identical to those found in
descriptive texts. The differenceand this is one of the most crucial issues at stake in
the present study of yh!yw+ ~lies at the level of presuppositions and basic perceptions of
language itself. This point will be argued in greater detail in Chapter 5, General
Theoretical Framework.
35
Schneider discusses the close connection in narrative of introductory yh!yw+ ~ with
temporal indicators (Schneider 1974, 252)2 and discusses the function of yh!y+w~ as a macrosyntactic sign: Macro-syntactic signs are words, particles, and expressions, in spoken
language, which function to indicate the large divisions of texts (Schneider 1974, 261)
[my translation].3 As a transition signal (Schneider 1974, 265) [my translation]4 yh!yw+ ~
occurs, according to Schneider, at the points of transition from the introduction to the
main part or before main events, which are particularly highlighted as such (Schneider
1974, 265-66) [my translation].5
As an indicator of tense, Schneider states that yh!yw+ ~ occurs at the beginning of
narrative sections. It is not linked to what precedes it, but rather it marks the following
text as narrative. (Schneider 1974, 265) [my translation].6 Many of the issues discussed
in later research are already found here in Schneiders seminal contribution to the study
of biblical Hebrew. His grammar makes claims regarding the function(s) of yh!yw+ ~ that will
In erzhlendem Kontext steht fast immer nach einleitendem yh!yw= ~ eine Zeitbestimmung
(Schneider 1974, 252).
3
Als bergangssignal steht yh!yw= ~ an Gelenkstellen der Erzhlung: beim bergang der
Einleitung zum Hauptteil oder vor Hauptereignissen, die dadurch als solche besonders
herausgehoben werden (Schneider 1974, 265-66).
6
36
be evaluated in subsequent sections of the present study. However, as will be seen in the
continuing review of other publications, some of the claims made by other authors seem
to state the exact opposite of what Schneider claimed.
37
Das yh!yw+ ~ in diesem Zusammenhang stets nur als Tempussignal verwendet ist
(Bartelmus 1982, 211-12).
38
Concerning yh!yw+ ,~ Longacre states that, [i]n general, wayh + a temporal phrase
marks an episode break in Hebrew narrative prose (Longacre 1989, 26). In the
hierarchical structure posited by Longacre, the episode is a major division larger than a
paragraph and part of the structure of the whole discourse or narrative. Longacre is here
claiming that one of the primary functions of yh!yw+ ~ with a temporal phrase is to indicate
the break between major parts of a biblical Hebrew narrative. Regarding this function
with temporal phrases, Longacre also comments that
[v]ery frequently this construction functions as a backreference to
previous material. In fact, such a reference is often found even in the rare
instances where a temporal expression is not preceded by wayh.
(Longacre 1989, 70)
For Longacre, the connection of yh!yw+ ~ + temporal expression is apparently with
what precedes as well as with what follows. The exact meaning of very frequently is
not clear, but one gets the impression that this is a basic function of this construction.
The reference to the rare instances where a temporal expression is not preceded by
wayh begs for further examination. Presumably, the backreferencing function is still
perceived with the temporal expression itself, but not the function of signaling an episode
break.
One of the concerns of Longacres approach to the study of narrative is the
detection of the profile of a discourse (Longacre, 1996, 2). Another crucial consideration
is the fact that most discourses formally mark (often by a variety of devices) a discourse
peak (Longacre, 1996, 2). Longacre defines peak as a zone of turbulence in regard to
the flow of the discourse in its preceding and following parts (Longacre, 1996, 38),
39
which reflects heightened action in the narrative, often marked by certain features or
clusters of features. This, in extremely brief form, is the theoretical concept that is behind
the identification of Gen 41 as a peak episode:
This peak episode (chap. 41) begins with an episode marker that indicates
a considerable lapse of time: wayh miqqec HnATayim yAmm, and it came
to pass at the end of two full years. There follows an unusual use of the
waw-conjunctive with the noun Pharaoh: and Pharaoh was dreaming.
Thus, the new embedded narrative establishes its new time horizon and
its new central participant. (Longacre 1989, 27)
Thus, it is clear that Longacre attributes a significant function to yh!yw+
~ at least
yh!yw+ ~ with temporal expressionsbut it is also evident from the reference to the wawconjunctive with the noun Pharaoh that yh!yw+ ~ + temporal expression is not the only device
that indicates peak in Longacres analysis. There is, however, no further detail regarding
the possible variations or nuances with different temporal expressions.
One of the other characteristics of Longacres approach is the discussion of verb
rank (Longacre 1989, 64-82) and the closely related discussion of the backbone or
mainline of a discourse. Longacre states that the storyline or the backbone of a
discourse in Biblical Hebrew is conveyed by use of clauses that begin with a wawconsecutive verb (Longacre 1989, 65). Waw-consecutive verbs, or preterites, occupy the
primary rankthey are the verbal clauses par excellence (Longacre 1989, 80).
Curiously, thoughand this will figure into subsequent discussion regarding the
~ is the following statement by Longacre that [t]he verb hAy, be,
functions of yh!yw+
even in its preterite form wayh, and it happened, does not function on the storyline of a
narrative (Longacre 1989, 80). As will be seen in the continuing review of other
40
publications, certain scholars claim, in contrast to Longacre, that yh!yw+ ~ does indeed have
the same mainlining function of other preterites or WAYYIQTOLs.
3.4.6 Niccacci, The Syntax of the Verb in Classical Hebrew Prose, 1990
Niccaccis 1990 study, The Syntax of the Verb in Classical Hebrew Prose, also
identifies text-level functions of yh!yw+ ,~ discussing its macrosyntactic function as
connecting circumstances and events with the main narrative thread (Niccacci 1990,
159). This study considers several macro-syntactic signs:
Besides these typical verb forms, in narrative and discourse there are
certain textual pointers, called macro-syntactic signs. These are elements
which mark the relationships among segments of the text. The main
indicator of narrative is wayehi; wehinneh chiefly marks discourse but also
functions in narrative, while weatta is exclusive to discourse. Mention
should also be made here of wehaya, an important macrosyntactic marker
in discourse which also occurs in the comment-sections of narrative.
(Niccacci 1990, 33)
According to Niccacci, then, the main indicator of narrative is yh!yw+ .~ As he states
elsewhere in Syntax of the Verb in Classical Hebrew Prose, wayehi is the supreme
macro-syntactic sign of narrative. This means that its presence is enough to mark the
passage as narrative (Niccacci 1990, 48). The implications of this claim need to be
carefully evaluated. In one sense, it seems more like a hyperbolic remark than a serious
claim that yh!yw+ ~ identifies the surrounding text in which it occurs as narrative. What about
other texts that seemingly exhibit the normal characteristics of narrative, but where yh!yw+ ~
does not occur? Is this function as macro-syntactic sign meant as a type of genre signal
like once upon a time? Niccaccis discussion of this function of yh!yw+ ~ cites Judg 11:1-5
as an example text; yh!yw+ ~ occurs in both 11:4 and 11:5 and Niccacci comments that the
41
two wayehis mark off the passage as narrative and at the same time indicate the two
basic components of the account: the war and Jephthahs role in it (Niccacci 1990, 49).
The second part of this statement reflects Niccaccis claim that [t]he textual function of
wayehi is to introduce a new element into the main narrative thread so that that element
becomes an integral and important part of the account (Niccacci 1990, 48). These are
significant claims that require further exploration.
With a slightly different slant on the function(s) of yh!yw+ ,~ Niccacci comments in
Lettura Sintattica della Prosa Ebraico-Biblica, that yh!yw+ ~ s function is to create
continuity throughout a text, so that the narration moves ahead at a regular pace9
(Niccacci 1991, 252) [my translation]. Similarly in Lettura Sintattica, Niccacci comments
that yh!yw+ ~ s function is to ensure cohesion throughout the various parts of a text, to
promote the textuality of the text10 (Niccacci 1991, 17) [my translation]. These
functions are related to those found in his Syntax, but the discussion of cohesion and
textuality moves beyond a mere label like macro-syntactic sign and identifies
possible effect[s] of its use in a particular text (Talstra 1978, 169). These claims will be
brought back into the discussion at a later point.
Il fatto che yh!yw+ ~ sia un mezzo sintattico che crea continuit allinterno di un testo,
comporta che la narrazione conserva un ritmo constante (Niccacci 1991, 252).
10
42
43
44
45
marker is more helpful than merely using the term discourse marker, but at the same
time, it does not help the reader know when yh!yw+ ~ has which function. The additional
comments about adverbial clauses being followed by infinitives, etc. help further specify
how and where yh!yw+ ~ is used, but the function in these environments is still unclear. To
clearly understand the uses and functions of yh!yw+ ,~ greater precision is needed.
46
elaborateness, in any case, is a general characteristic that describes most
features of peak. But since yhyw occurs so often in environments that are
not associated with peak or climax, one need only look at those instances
where yhyw stands at the beginning of main episodes in the Joseph story
additional data are needed to identify the cases in which the expression
functions as a peak feature. (Exter Blokland 1995, 49)
Two comments are in order here. First of all, it is questionable whether yh!y+w~ gives
the elaborateness to the passage described here by Exter Blokland. This is not to deny
that yh!yw+ ~ may have certain unique functions, but a term like elaborateness needs to be
more precisely defined. Often the perception of elaborateness is the result of the
outsiders perspective on what is very commonplace to the language user. At the same
time, however, it is certainly true that the speaker or writer is usually unaware of the
structural or functional complexity of seemingly mundane expressions. The issue here is
that elaborateness is too vaguely defined.
The second part of this comment also calls for further scrutiny. The function
attributed to yh!yw+ ~ of marking peak is not necessarily contradicted by its occurrences so
often in environments that are not associated with peak or climax, but the non-peak
occurrences definitely need to shape the way yh!yw+ ~ s function is defined. What exactly is
yh!yw+ ~ s role if it does indeed occur in both peak and non-peak? Exter Bloklands solution
is found in the following citation:
We have seen that episode boundaries can be marked by, e.g.,
(y)rja yhyw. But not all episodes are marked by such a grammatical or
text-syntactical marker, and even if they were, one would still be faced
with the question, whether an episode is embedded or not. The main
divisions of our text will have to be determined on the basis of subject
matter. (Exter Blokland 1995, 65)
47
For Exter Blokland, the fact that some episodes are not introduced by a textsyntactical marker like yh!yw+ ~ + temporal expression, leads to the conclusion that subject
matter or content determines the episode boundaries. This seriously argues against yh!yw+ ~
being assigned any unique function in marking episode boundaries. If content is really the
final arbiter of episode boundaries, the occurrence of yh!yw+ ~ becomes quite inconsequential
in terms of marking that boundary. The issue that faces the textlinguist at this point is to
investigate what has motivated yh!yw+ ~ in some episode boundaries and not in others. This
requires further analysis and will be commented on in Chapter 10.
3.4.11 Endo, The Verbal System of Classical Hebrew in the Joseph Story:
An Approach from Discourse Analysis, 1996
Endos study also deals with the function of yh!yw+ ~ in marking episodes or segments
of a text. The influence of Talstra, Schneider, and Niccacci is clear in the following
comment by Endo:
Talstra notes that yh!yw+ ~ (also hyhw) is used to mark the main segments of a
narrative, but also to distinguish the main story from the embedded
stories. Or, as Schneider and Niccacci explain, this verbal form may
function as a macro-syntactic sign (i.e. a marker of the relationships
among segments of the text.). (Endo 1996, 175-76) 11
In 3.4.8 Talstra was cited as having commented that yh!yw+ ~ distinguishes the main
story from the embedded stories. In similar fashion, Endo comments as follows:
Thus, the difference between the ordinary unmarked conditional/temporal
sentence and the yhyw or hyhw construction may involve the following two
11
60).
Endo cites Talstra (1978, 173); Schneider (1974, 265-66); and Niccacci (1990, 33, 48-
48
points: Firstly, in the latter case, because of this cleft construction, both
thematic and focal prominence are given to the temporal circumstance. In
this sense the function of this construction differs also from that of the
simple temporal clause with the temporal particle where the temporal
clause simply functions as a subordinate clause. (Endo 1996, 178)
Endos use of the terms thematic and focal prominence is evidence of influence
from certain conceptual trends within Linguistics, but the terms are used without clearly
defining what thematic or focal prominence means. The term emphasis (Muraoka 1983)
resonates at an intuitive levelit is easy to sense or recognize that a text consists of
information that varies in its relative prominencebut it is very difficult to precisely
define how emphasis or prominence actually functions in communication. Notice the
following comment by Endo:
Secondly, when the impersonal verb hyh as a sequential form with the
inherent meaning such as happen or become is combined with a
temporal circumstance especially with the particle, it emphasizes the
temporal setting of the incident with various senses depending on the
following particle. (Endo 1996, 179)
The function of emphasizing the temporal setting is identified by Endo as a
function of yh!yw+ ,~ but this raises the question of what this emphasis means or indicates to
the listener or reader. Also, what are the implications for the temporal clauses that do not
include yh!yw+ ?~ These issues will be dealt with in Chapter 10.
49
in 3 person masculine singular; 2) As a segmentational particle, marking
mainly temporal segmentation. (Hatav 1997, 70)
rd
50
Hatavs thought-provoking analysis is significant in its implementation of
theoretical concepts used in general Linguistics. Further discussion of these claims
regarding yh!yw+ ~ is postponed, however, until Chapter 10.
51
Many of these comments will be considered in the analysis in Chapter 10. The
syntactic distinctions made in this grammar are a significant contribution to the ongoing
study of yh!yw+ ,~ and the function assigned to yh!yw+ ~ of keeping things in the mainstream of the
narration will be considered in greater detail.
As examples of the categories used in BHRG, consider the following:
a. yh!yw+ ~ introduces the setting of a new scene or episode. yh!yw+ ~ signals that
the new scene or episode needs to be link [sic] to a preceding one on
which this new scene or episode follows in time. (van der Merwe, Naud,
Kroeze 1999, 332)
b. yh!yw+ ~ occurs in the setting of a new episode (seldom). yh!yw+ ~ precedes
reference to a state of affairs that (apparently) plays a pivotal role in the
subsequent episode.
c. yh!yw+ ~ occurs in the course of a scene. yh!yw+ ~ signals that a state of affairs
needs to be treated on a par with the mainstream events of the narration,
and that is not mere background information.
d. yh!yw+ ~ occurs at the conclusion of a scene. yh!yw+ ~ signals that a state of
affairs is the outcome of a preceding series of events. (van der Merwe,
Naud, Kroeze 1999, 333)
The attention to context is an important part of this grammars analysis. Similar to
previous remarks, yh!yw+ ~ is assigned a role in distinguishing the mainstream from the
background. The role of yh!yw+ ~ in setting is of particular relevance to the claim that yh!yw+ ~
marks a state of affairs that may play a pivotal role in the subsequent episode. These
claims will be dealt with in greater detail in Chapter 10.
52
Gen 17:1
stating that this verse is an example of a discourse bounded by yh!yw+ ~ plus a specification
of time (Rocine 2000, 299). In this case, Rocine goes on to explain that the
specification of time is the verbless clause which follows yh!yw+ ~ (2000, 299). This
53
comment reflects an analysis that sees yh!yw+ ~ as essentially detached from the temporal
expression. This raises questions about the syntactic connection of yh!yw+ ~ to the clause
which follows it, which is one of the main issues discussed throughout Chapter 9, The
Temporal Uses of yh!yw+ .~
54
observing what something like yh!yw+ ~ does; rather, the description is the foundation for
being able to anticipate or predict why or under what circumstances a certain element is
used. As Dik comments:
a theory of grammar should not be content to display the rules of
language for their own sake, but should try, wherever possible, to explain
these rules in terms of their functionality with respect to the ways they are
used and to the ultimate purposes of these uses. (Dik 1978, 2)
From this perspective, there are certain ways in which statements about linguistic
patterns or occurrences should be made. Statements that element X often occurs or that
element Y can occur in a particular environment may describe representative cases of
those linguistic elements, but more precise parameters are needed. This is not meant to
imply that every problematic case disappears and that all exceptions magically vanish,
but by broadening the contexts which textlinguistic approaches consider, a much greater
degree of predictability is made possible. These statements may be hypotheses, but the
objective is to clearly state the hypothesis so that it can be verified or refuted. As van der
Merwe insightfully comments, even if some hypotheses are refuted, our knowledge is
furthered by knowing what a construction does not mean (van der Merwe 1994, 39). The
problem with statements like often occurs and can occur is that they are not easily
verified or refuted.
Another facet of a predictive analysis is the necessary consideration of nonoccurrences. For example, an analysis may state that yh!yw+ ~ occurs with temporal clauses
that signal a new episode in the text. In order to account for all the dimensions of this use
of yh!yw+ ,~ however, the following questions must also be asked:
55
1) Are all new episodes initiated by yh!yw+ ~ + temporal clause? If not, then
the function of yh!yw+ ~ + temporal clause is not fully known until new
episodes without yh!yw+ ~ + temporal clause are brought into the analysis.
2) Is the only function of yh!yw+ ~ + temporal clause the introduction of new
episodes? If not, then the other uses of yh!yw+ ~ + temporal clause must be
considered to more fully discern under what conditions yh!yw+ ~ does in fact
occur with a temporal clause to initiate a new episode.
In Narrative Syntax, van Wolde comments that
most linguistic studies of Biblical Hebrew mainly focus on the analysis
of how the forms are organized. Only a few deal with the question of
motivation and ask why a specific form is used in a certain text. (van
Wolde 1997, 21)
The goal of textlinguistic studies, then, should be to move beyond how to why.
The goal of the present study is to contribute to this process.
CHAPTER 4
THE NEED FOR FURTHER STUDY OF yh!y+w~
It might seem from the preceding review of studies which deal with yh!yw+ ~ that there
is little more that could be said about this enigmatic little word. One of the goals here,
however, is to deal exhaustively with yh!yw+ ~ and in the process, further refine current
understanding of its uses and functions as well as extend the analysis into areas not yet
adequately explored.
There are many aspects of where and when yh!yw+ ~ is used that are not well-defined.
Part of the problem with previous studies is the untestability of what they claim
concerning yh!yw+ ~ s functions. To merely label it a macro-syntactic marker or Textdeiktikon does not answer many of the questions that come from seeing the diverse
ways in which yh!yw+ ~ is used. If the function of yh!yw+ ~ is Tempusmarker, another set of
questions arises. There are many past tense narratives that dont have any occurrences of
yh!yw+ ~ and certain texts seem to have many more than would be needed to simply mark past
tense. Questions like these and those discussed in the next three sections indicate that
there is much more to be understood about the uses and functions of yh!yw+ .~ These sections
are followed by a detailed discussion of van der Merwes analysis of yh!yw+ ~ in 1 Samuel.
56
57
Now it came about after the death of Moses the servant of the LORD, that the LORD spoke to
Joshua the son of Nun, Moses' servant, saying,
NIV: After the death of Moses the servant of the LORD, the LORD said to Joshua son of Nun, Moses'
aide:
NRSV: After the death of Moses the servant of the LORD, the LORD spoke to Joshua son of Nun, Moses'
assistant, saying,
Judg 1:1
WnL*-hl#uy& ~ ym! rm)al@ hw`hyB^ la@rc
` y= ] yn}B= Wla&vY= w] ~ u~vo% hy+ tom yr}ja
& ^ yh!yw+ ~
.w{B <j#Lh
* l
! = hL*jT
! B
= ^ yn]u&nK
~ h
= -^ la#
NASB:
Now it came about after the death of Joshua that the sons of Israel inquired of the LORD, saying,
Who shall go up first for us against the Canaanites, to fight against them?
NIV: After the death of Joshua, the Israelites asked the LORD, Who will be the first to go up and fight
for us against the Canaanites?
NRSV: After the death of Joshua, the Israelites inquired of the LORD, Who shall go up first for us against
the Canaanites, to fight against them?
1 Sam 1:1
<j*ry) -+ /B# hn`ql
` a
= # w{mv=W <y]rp
` a
= # rh^m@ <yp!ox <y]tm
^ r
* `h-* /m! dj*a# vya! yh!yw+ ~
.yt!rp
` a
= # [Wx-/b# WjT)-/B# aWhyl!a-$ /B#
NASB:
Now there was a certain man from Ramathaim-zophim from the hill country of Ephraim, and his
name was Elkanah the son of Jeroham, the son of Elihu, the son of Tohu, the son of Zuph, an
Ephraimite.
NIV: There was a certain man from Ramathaim, a Zuphite from the hill country of Ephraim, whose name
was Elkanah son of Jeroham, the son of Elihu, the son of Tohu, the son of Zuph, an Ephraimite.
NRSV: There was a certain man of Ramathaim, a Zuphite from the hill country of Ephraim, whose name
was Elkanah son of Jeroham son of Elihu son of Tohu son of Zuph, an Ephraimite.
58
2 Sam 1:1
.<y]nv
` = <ym!y` gl*qx
+ B
! = dw]D` bv#Yw} ~ ql@mu
* h
& -* ta# toKh^m@ bv* dw]dw` + lWav* tom yr}ja
& ^ yh!yw+ ~
NASB:
Now it came about after the death of Saul, when David had returned from the slaughter of the
Amalekites, that David remained two days in Ziklag.
NIV: After the death of Saul, David returned from defeating the Amalekites and stayed in Ziklag two
days.
NRSV: After the death of Saul, when David had returned from defeating the Amalekites, David remained
two days in Ziklag.
Ezek 1:1
rb*K-= rh^n-+ lu^ hl*oGh^-Eotb= yn]aw& ~ vd\jl
) ^ hV*mj
! B
& ^ yu!yb!rB
+ * hn`v* <yv!Ov=B! yh!yw+ ~
.<yh!Oa$ toar+m^ ha#ra
+ w# ` <y]mV
^ h
* ^ WjT=pn= ]
NASB:
Now it came about in the thirtieth year, on the fifth day of the fourth month, while I was by the river
Chebar among the exiles, the heavens were opened and I saw visions of God.
NIV: In the thirtieth year, in the fourth month on the fifth day, while I was among the exiles by the Kebar
River, the heavens were opened and I saw visions of God.
NRSV: In the thirtieth year, in the fourth month, on the fifth day of the month, as I was among the exiles by
the river Chebar, the heavens were opened, and I saw visions of God.
Jonah 1:1
.rm)al@ yT^ma
! -& /b# hn`oy-la# hw`hy+-rb^D+ yh!yw+ ~
NASB: The word of the LORD came to Jonah the son of Amittai saying,
NIV: The word of the LORD came to Jonah son of Amittai:
NRSV: Now the word of the LORD came to Jonah son of Amittai, saying,
Ruth 1:1
hd`Why+ <j#l# tyB@m! vya! El#Yw} ~ Jr\aB
* * bu*r` yh!yw+ ~ <yf!pV
= h
) ^ fp)v= ym@yB! yh!yw+ ~
.wyn`b* yn}vW= w{Tv=aw! + aWh ba*om yd}cB
= ! rWgl*
NASB:
Now it came about in the days when the judges governed, that there was a famine in the land. And a
certain man of Bethlehem in Judah went to sojourn in the land of Moab with his wife and his two
sons.
NIV: In the days when the judges ruled, there was a famine in the land, and a man from Bethlehem in
Judah, together with his wife and two sons, went to live for a while in the country of Moab.
NRSV: In the days when the judges ruled, there was a famine in the land, and a certain man of Bethlehem in
Judah went to live in the country of Moab, he and his wife and two sons.
Esth 1:1
.hn`yd]m= ha*mW@ <yr]cu
= w# + ub^v# vWK-du^w+ WDh)m@ El@M)h^ vorw}vj
= a
^ & aWh vorw}vj
= a
^ & ym@yB! yh!yw+ ~
NASB:
Now it took place in the days of Ahasuerus, the Ahasuerus who reigned from India to Ethiopia over
127 provinces,
NIV: This is what happened during the time of Xerxes, the Xerxes who ruled over 127 provinces
stretching from India to Cush:
NRSV: This happened in the days of Ahasuerus, the same Ahasuerus who ruled over one hundred twentyseven provinces from India to Ethiopia
The NASB consistently preserves at least now if not now it came about wherever
yh!yw+ ~ occurs at the beginning of these books, with the exception of Jonah. Other English
59
versions, such at the NIV and NRSV, leave no trace of the occurrence of yh!yw+ ,~ except
Esth 1:1 in NIV: This is what happened What does yh!yw+ ~ signal in this book-initial
position? Is it an indicator of narrative? If yh!yw+ ~ marks text as narrative, why does yh!yw+ ~
initiate certain books and not others which appear to have the same narrative character?
Obviously there is more than one way to start a book, but the analysis of yh!yw+ ~ needs to
take facts like these into account and avoid general statements about yh!yw+ ~ frequently
starting a book.
yT^ma
! -& /b# hn`oy-la# hw`hy+-rb^D+ yh!yw+ ~
.rm)al@
Jonah 1:1
Jonah 1:4
tolyl@ hv*Ov=W
rm)al@ tyn]v@ hn`oy-la# hw`hy+-rb^d+ yh!yw+ ~
Jonah 3:1
vm#Vh
# ^ j~rz{ K
+ ! yh!yw+ ~
Jonah 4:8
The occurrences at 1:1 and 3:1 are identical in structure, other than the use of
yT^ma
! -& /b# in 1:1 to introduce Jonah and of tyn]v@ in 3:1 to make it explicit that it is the
second time. Four of the five occurrences are followed by noun phrases, but the fifth is
followed by the preposition K= and an infinitive construct. To what extent do these details
of the syntactic environment affect the function of yh!yw+ ?~ Do these occurrences have any
60
structural significance within the book? For example, it could be argued that the
occurrences at 1:1 and 3:1 initiate two major divisions of the book. However, is this a
function of yh!yw+ ~ or would the same division be there because of the parallel structure even
if yh!yw+ ~ were not present? Do the other three occurrences of yh!yw+ ~ have any other function
in the book? Is there some significance in marking the storm, Jonah being in the fishs
belly, and the sunrise with yh!yw+ ?~ Are these pivotal events within the book that merit a
discourse marking? These are the kinds of questions that motivated further research.
61
62
current reference time of the narrative until it is updated or specified by another
temporal reference. Van der Merwes claim, then, is that yh!yw+ ~ signals and confirms that
the reference time of the construction with yh!yw+ ~ is the already established reference time.
According to this analysis in 1 Samuel, then, yh!yw+ ~ does not establish a new reference
time.
Van der Merwes conclusions are then divided into yh!yw+ ~ followed by a nominal
clause and yh!yw+ ~ preceding a temporal clause. He states that yh!yw+ ~ introduces a nominal
clause
that initiates the setting of a new episode in a narrative. In such cases it
normally points to the connection between major units in the Tanach.
Hence, it may be regarded as a device for establishing continuity at a
macro level; or
yh!yw+ ~ may also
introduce a nominal clause that concludes a scene or episode. In such
cases the fact that a state of affairs is the outcome of preceding events is
signaled. To put it differently, yh!yw+ ~ anchors a state of affairs to the timeline (i.e., story-line) of a narrative and establishes a continuity at an intrascene level between events and the state of affairs of a scene. (van der
Merwe 1999, 113)
In some respects, the claim that yh!yw+ ~ both initiates and concludes a scene or
episode may seem to be contradictory. What determines whether a particular occurrence
of yh!yw+ ~ is initiating or concluding a scene or episode? This will be discussed in detail in
subsequent sections.
63
64
4.1.6 Summary
Van der Merwes analysis of 1 Samuel highlights the complexity of yh!yw+ ~ s usage
in the Hebrew Bible. Even though the corpus provided by 1 Samuel seems quite
extensive, there are many questions about the use of yh!yw+ ~ that remain unanswered.
There are also several aspects to van der Merwes analysis that need clarification.
He states that yh!yw+ ~ signals that the reference time of a state of affairs is that of the
current reference time of a preceding temporally anchored event or events and that yh!yw+ ~
also allows the updating or specification of the reference time of an event by confirming
that the current reference time of the construction with yh!yw+ ~ is a preceding temporally
anchored event or state of affairs (van der Merwe 1999, 113). These claims need to be
evaluated after careful examination of the data.
65
though the present analysis focuses on yh!yw+ ,~ it cannot be adequately analyzed when
isolated from the web of which it is an integral part. This analysis also involves close
consideration of the syntactic components most integrally connected to yh!y+w~ in the text.
For example, the analysis of yh!yw+ ~ must take into account the syntactic shape of the clause
in which it occurs. One of the most unfortunate consequences of too quickly assuming
the textlinguistic function of an entity like yh!yw+ ~ is that its syntactic role or connection in
the immediate context can all too easily be overlooked.
At a broader level, yh!yw+ ~ must be analyzed in terms of how it relates to the various
features of narrative texture and temporal organization in biblical Hebrew. Whenever the
narrative functions of yh!y+w~ are discussed, yh!yw+ ~ must be analyzed as one of several
interrelated components of the text which work together to move it through its turns and
transitions. In this regard, yh!yw+ ~ must be analyzed along with the many WAYYIQTOL verb
forms in biblical Hebrew. As a member of the total set of WAYYIQTOL verbs, yh!yw+ ~ should
be expected to share similar functions with other members of the set, but without denying
that the grammaticalization of certain unique functions can occur. The fact that yh!yw+ ~ is a
WAYYIQTOL
66
This study is very data-oriented, based on the belief that working through
example after example is the best way for the general patterns of usage to become most
apparent. If only a few selective examples are given, it is more difficult to gain an
appreciation for and familiarity with the full spectrum of yh!yw+ ~ s uses. One of the other
benefits of displaying all the data is that it gives an objective starting point from which to
discuss the patterns and distribution of the linguistic items in question. It is easy to
assume that certain phrases are quite frequent unless the actual number of occurrences is
taken into consideration. For example, the specific phrase hL#ah
@ * <yr]bD
* h
+ ^ rj^a^ yh!yw+ ~
after these things (discussed in more detail in 9.3.1.1) occurs only eight times in the
entire Hebrew Bibleand five of these occurrences are in Genesis.2 While reading
Genesis, it is certainly reasonable to notice the relatively frequent occurrence of this
phrase, but this type of impression cannot be anything more than a preliminary
hypothesis until the rest of the biblical Hebrew corpus is considered. On the basis of
Genesis, this phrase may have appeared to be a significant temporal structuring device,
but this analysis is no longer possible when the other occurrences are taken into account.
An additional benefit of displaying all the examples is that it gives the reader
immediate access to the data, thus facilitating evaluation of the analysis and claims in this
study. In the study of something like yh!yw+ ,~ it is very easy to make general statements
based on limited data. Some of these statements may indeed be valid, but they need to be
verified in light of all the relevant data.
Gen 22:1, 20; 39:7; 40:1; 48:1; Josh 24:29; 1 Kgs 17:17; 21:1.
67
In certain sections, however, where the exact phrase is repeated numerous times,
just the references are given. This is the case, for example with the possessive phrases,
w{l-yh!yw+ ~ he had, especially since there is nothing in the broader context that affects these
verbal uses of yh!yw+ .~
68
The discourse-pragmatic analysis implemented here requires that yh!yw+ ~ not be
considered on the basis of the individual clauses in which it occurs, isolated from the
surrounding context. In this framework, yh!yw+ ~ will be considered within the whole textual
web of biblical Hebrew. Attention will also be paid to the narrative flow and the
function of other textlinguistic indicators in the global context in which yh!yw+ ~ occurs,
considering the WAYYIQTOL and WEQATAL forms in particular. The discourse distribution
and occasional clustering of occurrences of yh!yw+ ~ is another aspect that will require
analysis.
One of the other goals of this study is methodological in nature. The primary goal
is, of course, to analyze yh!yw+ ,~ but a secondary objective is that the analysis of yh!yw+ ~ be seen
as a case study to demonstrate the benefits of the multidimensional discourse-pragmatic
analysis implemented here (See section 5.2). This analytical model can be applied to any
aspect of biblical Hebrew. In fact, one of the greatest needs in current discourse studies of
biblical Hebrew is a comprehensive model capable of dealing with the multifaceted
features of the biblical Hebrew text. This may seem like an unattainable goal, but one of
the factors that has hindered wider acceptance of discourse studies is the perceived
fragmentary character of many individual studies of specific features. Of course, it is
possible to err in the opposite extreme if the model presented here is perceived as the
only model to answer all questions about biblical Hebrew discourse. The goal here is not
that ambitious, but at the same time what is implemented here is presented with the firm
conviction that satisfying answers to questions related to biblical Hebrew discourse
69
features can only be obtained by some type of multi-perspectival communicative model
of language in use.
Some may react to the eclectic nature of this model, but having a variety of
analytical tools at ones disposal is the best way to deal with the variety of features in the
biblical Hebrew text. Another potential criticism is that the analytical model appears to
be much more complex than even the Hebrew text itself. Complexity, in and of itself, is
not what validates a model of linguistic analysis, but all too often the inherent complexity
of language is minimized because of the automatic, subconscious way in which we as
speakers use language. Care needs to be exercised to avoid rejecting an analytical model
merely because of its apparent complexity. This, however, does not justify unnecessary
complexity in linguistic description; any analysis should always strive for simplicity
whenever possible.
It is also possible that some may dispute the emphasis on linguistic analysis
employed in this study, but the linguistic nature of the problem calls for the use of
appropriate linguistic tools. The reader is free to judge whether there is any benefit to be
derived from the use of these analytical perspectives and tools, but even if this study does
no more than further define the questions that remain about yh!yw+ ,~ a contribution will be
made. It is the goal and expectation of the author, however, that at the very minimum,
certain parameters for the study of an entity like yh!yw+ ~ will be at least clearly explicated.
In linguistic description, as in many scientific endeavors, fuzzy analysis is extremely
difficult to either refute or verify; the most verifiable analysis, then, is one which clearly
states its hypotheses and conclusions. This, then, is the goal here.
CHAPTER 5
GENERAL THEORETICAL FRAMEWORK
5.1 Introduction
The general theoretical framework discussed here lays an important foundation
for the analysis presented in the subsequent chapters. Two introductory matters require
attention, however, before getting into the main theoretical concepts: 1) The nature of
biblical Hebrew as a language and 2) the nature of linguistic description.
70
71
by reader and listener alike. Consequently, the study of biblical Hebrew requires a model
of linguistic analysis that focuses on language use.
The study of a linguistic item like yh!yw+ ~ assumes that the text reflects acceptable
patterns of usage and not misuse or improper usage. It is assumed that editing the text
did not violate its communicative integrity. By implication, then, all occurrences of yh!yw+ ~
are assumed to be in the text because they reflect how the writers and/or editors
understood and used yh!yw+ .~ This hypothesis may seem extreme, but it is the best way to
proceed with an analysis of yh!yw+ .~ If the assumption were made that certain occurrences of
yh!yw+ ~ did not coincide with acceptable patterns of usage, the analysis would not only have
to explain why these unacceptable occurrences of yh!yw+ ~ are in the text, but the integrity of
the entire text becomes suspect. On the other hand, if the integrity of the text including
all the occurrences of yh!y+w~ is assumed, the analysis is constrained to seek explanations for
those occurrences through careful implementation of whatever method of linguistic
analysis might be necessary. In the final analysis, if some occurrences are not adequately
explained, it will be the fault of the model and its implementation rather than the fault of
the text.
72
indeed be very complex, but one of the main problems is that the speaker of a language is
unaware of the complexity of the linguistic systems he or she uses without conscious
effort. This is not a license for uncontrolled abstractioncomplexity merely for the sake
of complexity is to be avoided at all costsbut a certain level of complexity has to be
expected. A conscious attempt has been made, however, to keep technical, theoryspecific jargon to a minimum.
73
74
systems requires awareness of the integral role those elements or features play in the
whole communication event.
Some may object to the expression, the whole communication event, with
reference to the written biblical Hebrew text. The intention is not to imply that the text is
the same as the whole communication event, but neither should the text be
underestimated for its ability to faithfully represent the communication event. Certain
aspects of the communication event are not accessible, but the fallacy of assuming we
have nothing because we dont have anything should be avoided at all costs.
The goal, then, is the best descriptive reflection of biblical Hebrew as a language
that was used in real communicative situations. Together with the other orientations that
follow, the goal of this description is to arrive at a principled description of Hebrew based
on an awareness of and interaction with categories used in general Linguistics for the
descriptive part of the analysis. In this descriptive orientation, the terms identification
(5.2.1.1), variation (5.2.1.2), and distribution (5.2.1.3) are evidence of the underlying
influence of Tagmemics, especially as discussed by Pike in Linguistic Concepts (Pike
1982). Longacre states in The Grammar of Discourse that [t]agmemics can offer itself to
the student of linguistics as an underlying, minimal theory (Longacre 1996, 269).
Indeed, one of the strengths of Tagmemics is the set of foundational concepts that were
conceived specifically for the task of linguistic description. These concepts are
implemented here, but the overall analytical approach is not limited to the Tagmemic
model.
75
5.2.1.1
Identification
Proper identification is essential to the description of any linguistic entity. In
traditional grammatical terms, this is the function of parsing a particular form in the
language. All the grammatical details are a necessary part of the full identification of a
linguistic entity. Identification involves cataloguing the contrastive features of the
various grammatical forms of biblical Hebrew,
such as the verb stems or binyanim. The
identification of forms provides a taxonomy of all
the forms used in the Hebrew Bible and is an
important foundation for other types of analysis.
5.2.1.2
IDENTIFICATION:
yh!yw+ ~
Pc-vqw3msXa
of hy`h*
Figure 2: Identification of yh!yw+ ~
Variation
This aspect of analysis considers the possible ways in which a given linguistic
entity may vary slightly from case to case without ceasing to be itself. For example, the
influence of the 'atnax will bring about slight changes in certain formstypically called
pausal formsbut these minor variations do not make the linguistic entity in question
become something else. For example, yh!yw+ ~ and
VARIATION:
normal:
yh!yw+ ~
yh!Tw= ~
pausal:
yh!Yw\ ~
yh!Tw# ~
76
historical, geographical, and sociological. In the analysis of biblical Hebrew, it is always
advisable to evaluate whether variations encountered in the text might be the result of
these influences, but the nature of the transmission of the biblical text makes it very
difficult to make conclusive statements.
5.2.1.3
Distribution
Also fundamental to the full analysis of any linguistic entity is an awareness of its
distribution throughout the text. An item may have a very limited distributionperhaps
occurring in only certain clausal or textual environmentsor there may appear to be a
very wide distribution; all of these factors are significant in the full analysis of a
linguistic item.
In the study of biblical Hebrew, the distribution of linguistic items has typically
been overlooked with the focus rather on the items frequency of occurrence. For
example, knowing only that za* occurs 141 times (Mitchel 1984, 10) is not nearly as
significant as also knowing its distribution in different contexts. An important aspect of
the distribution of an item like za* is how it patterns with other components in the systems
of temporal reference in the text. In the study of yh!yw+ ,~ distribution involves not merely
where it occurs in the text, but which type of occurrence of yh!yw+ ~ occurs in a given
location, whether there is any discernible pattern of occurrence, or whether there is any
departure from the typical pattern of occurrence, etc.
Identifying the number of occurrences is only the first step in describing the
distribution of an item like yh!yw+ .~ As a first glance at the pattern of yh!yw+ ~ s usage, it is
instructive to consider the difference in frequency by book. If it is assumed that yh!yw+ ~
77
typically occurs in narrative, this assumption appears to be confirmed by, for example,
the extremes of 130 occurrences in Genesis compared with 0 occurrences in Proverbs.
The relatively high frequency in Isaiah, Jeremiah, and Ezekiel might seem a little more
surprising, requiring further examination of the occurrences in these books since the
preliminary chart above does not distinguish between the various types of uses of yh!yw+ .~
Genesis
Exodus
Leviticus
Numbers
Deuteronomy
Joshua
Judges
1 Samuel
2 Samuel
1 Kings
2 Kings
Isaiah
Jeremiah
Ezekiel
Hosea
Joel
Amos
Obadiah
Jonah
Micah
total
yh!yw+ ~
yh!Tw= ~
yh!yw+ ~
123
7
130 Nahum
0
41
3
44 Habakkuk
1
1
0
1 Zephaniah
0
16
5
21 Haggai
2
7
0
7 Zechariah
9
62
0
62 Malachi
0
48
5
53 Psalms
5
55
12
67 Job
10
39
17
56 Proverbs
0
77
4
81 Ruth
5
55
2
57 Song of Songs
0
11
6
17 Ecclesiastes
0
43
3
46 Lamentations
0
64
7
71 Esther
6
1
0
1 Daniel
5
0
0
0 Ezra
1
0
0
0 Nehemiah
14
0
0
0 1 Chronicles
27
5
0
5 2 Chronicles
46
0
0
0
totals
779
Figure 4: Frequency of Occurrence of yh!y+w~ and yh!Tw= .~
yh!Tw= ~
0
0
0
0
0
0
1
1
0
2
0
0
1
1
0
0
2
5
1
85
total
0
1
0
2
9
0
6
11
0
7
0
0
1
7
5
1
16
32
47
864
Analyzing the distribution of linguistic items has implications for the process of
learning a language like biblical Hebrew. One of the typical methods for learning
vocabulary is according to frequency of occurrence. The advantage of this approach is
that it allows the beginning student to learn the words which he or she will likely
encounter in a wide variety of biblical passages. The student soon comes to a point,
however, where a contextual approach would be more beneficial. For example, the verb
am@f* to become unclean occurs 162 times in the Hebrew Bible and will be learned by
78
most students in standard Hebrew courses (Van Pelt and Pratico 2003, 105). The use of
this verb, however, is very contextual. The following chart shows the occurrences by
book:
Occurrences of Genesis, 3; Leviticus, 71; Numbers, 21; Deuteronomy, 2; 2 Kings, 4;
am@f* by book: Psalms, 2; Isaiah, 1; Jeremiah, 4; Ezekiel, 28; Hosea, 3; Micah, 1;
Haggai, 1; 2 Chronicles, 1
Figure 5: Occurrences of am@f* by book
When the student reads the Hebrew Bible, most of Genesis and all of Exodus will
be read without any occurrences of this word, whereas in Leviticus it is understandably
frequent. A contextual approach to vocabulary learning would at least alert the student to
these parameters of usage.
One of the unfortunate side-effects of learning biblical Hebrew vocabulary only
by frequency of occurrence is that it tends to foster the perception that frequency equals
importance. This Frequency Fallacy operates on the assumption that mere repetition is
significant. The recommendation to note any words that are repeated or that emerge as
motifs in a section or paragraph (Fee 2002, 80) reinforces the notion that repetition is
important. Repetition should not be ignored, but should be analyzed with a more
nuanced, contextual approach. Schertz and Yoder comment that even if a phrase or
clause is repeated exactly, it may have a different nuance in a different context (Schertz
and Yoder 2001, 49). This requires careful analysis of each occurrence of a word or
phrase before making exegetical decisions and conclusions.
Another aspect of the Frequency Fallacy is that significance is sometimes found
in the very infrequent, but strategically used word or concept or even in what is not
overtly stated. Giving a New Testament example, Stuart states that
79
[a]n excellent reminder of this is Jesus parable of the Good Samaritan in
Luke 10. He tells the parable in order to demonstrate what it means to love
neighbor as self, yet the parable does not contain the word love or
neighbor or selfeven though it contains powerfully the concept of
loving neighbor as self. (Stuart 2001, 21)
By implication, then, the repetition of words and phrases is only one feature to
consider in the whole analysis of the text. Full distributional analysis considers this
figure and ground nature of the use of linguistic items.
5.2.1.4
Summary
In terms of the present study of yh!yw+ ,~ these three analytical perspectives require an
exhaustive taxonomy of yh!yw+ ~ for its proper identification, taking any possible variants into
consideration. The factors which potentially affect the distribution of yh!yw+ ~ throughout the
text must also be explored. The analysis of a linguistic entity is incomplete unless all
aspects are taken into consideration.
5.2.2.1
The traditional distinction between morphology and syntax validly recognizes the
word as a basic linguistic unit at its level of the hierarchy.2 Starting with basic wordformation patterns, it is only logical to conceive of lower to higher levels of analysis
1
After this section had been written, while rereading Pikes Linguistic Concepts (1982),
it was discovered that Pike uses the term unit-in-context. In 5.2.1, my indebtedness to Pike is
acknowledged, but when I decided to use the expression a unit-in-context approach, I was not
consciously aware of the use of this term in Linguistic Concepts.
2
See Kenneth L. Pike and Evelyn G. Pike, Grammatical Analysis (1991, 98-99) for a
discussion of fundamental principles of word division which include the: 1) Isolatability of
words; 2) Relative mobility of words versus rigidity in the order of parts of words; and,
3) Noninterruptibility of a word by words.
80
based on the relative size of linguistic units. In some models of linguistic analysis, it has
been a matter of principle that these levels be analyzed separately and in order.
Newmeyer states that Hocketts principle of avoiding circularity in phonological analysis
became known as the prohibition against mixing levels in grammatical description
(Newmeyer 1980, 6). Certain analytical procedures do require clear separation of levels,
but this needs to be understood as an analytical abstraction. This type of abstraction is
like a solution of salt water which the chemist knows can be separated into its component
chemicals, but for the non-analyst it is just salt water. Language use is like the various
linguistic items in solution; the separation of these items into different levels is merely an
abstraction for analytical purposes.
The usefulness of these analytical abstractions is not denied, but this is not the
same as actual language in use. Language as it is being used is not confined to discrete
hierarchical levels, but is more properly conceived of as an intersecting network of
communicative functions. Discourse analysis is often referred to as an approach which
goes beyond the sentence, distinguishing it from other models which focus on
sentence-level grammar. It is unquestionable that describing the structure or grammar of
the hierarchical levels of language beyond the sentence is one of the typical concerns of
discourse analysis, but this characterization can give the impression that discourse is
merely another level of study that comes after the lower levels of the hierarchy have been
fully analyzed. In the view of language implemented here, however, discourse doesnt
refer to a discrete hierarchical level, but rather refers to the pervasive communicative
nature and functions of language which intersect with all the other components in the
81
network. Analysis needs to be carried out at each level, but full analysis will always be
provisional until the analysis of all the parts has been informed by the whole. The terms
bottom-up and top-down are sometimes used to describe how analysis proceeds from
lower to higher or from higher to lower levels, but the model presented here is better
described as a unit-in-context approach. As such, it cannot be characterized as either
bottom-up or top-down, but rather involves constant attention to part-whole
relationships within the text.
For example, in biblical Hebrew, the pronominal suffixes are typically analyzed at
the morphological level. This level of analysis is obviously necessary, but this is only
part of the total analysis since their narrative function is not restricted to the
morphological level. This is the motivation for the term unit-in-context, since the full
analysis of the unit which is realized at the morphological level requires examination of
its connection to the broader narrative context.
Then he gave them into the hands of
the Gibeonites, and they hanged them
in the mountain before the LORD, so
that the seven of them fell together;
and they were put to death in the first
days of harvest at the beginning of
barley harvest.
Pc-vqw3msX3mp Pp-ncfsc Pa-np Pcvhw3mpX3mp Pp+Pa-ncms Pp-ncbpc np
Pc-vqw3mp amd amscX3mp Pd Pcpi3mp Pc-pi3mp vHp3cp Pp-ncmpc ncms
Pp+Pa-amp ncfsc Pp-ncfsc ncmsc ncfp
<yn]ub
) G= h
] ^ dy~B= <n}TY= w] ~
2 Sam 21:9
hw`hy+ yn}pl
= ! rh*B* <u%yq!Yw{ ~
dj^y` <y]Tu
* b
= v
^ ! WlP=Yw] ~
ryx!q* ym@yB! Wtm=h% <h@w+
.<yr]uc
) = ryx!q= tL^jt
! = <yn]va
) r]B*
For example in 2 Sam 21:9, the use of the pronominal suffix on <n}TY= w] ~ and <u%yq!Yw{ ~
is not merely the result of the arbitrary application of an optional rule for encoding the
object. The morphological analysis of these verbs as Pc-vqw3msX3mp and Pcvhw3mpX3mp respectively, identifies the forms that occur in the text, but explaining
82
their use requires going beyond the surface forms to explore the contextual dimensions
which may influence the use of the pronominal suffix. Lambdin states, for example, that
[a] pronominal direct object may be suffixed directly to a verb rather than
to the object marker ta# (yt!a) etc.):
wta) yt!ya!r`
Whyt!ya!r=
I saw him.
Ht*a) gr~h*
Hg`rh
` &
He killed her.
There is no difference in meaning between the two constructions, though
there do appear to be stylistic preferences. (Lambdin 1971, 260)
In 2 Sam 21:9, the realization of the direct object as a suffix on these verbs is part
of the intricate scheme of participant reference and the relative levels of focus assigned in
the narrative. Preliminary research indicates that there are notable differences between
the following three, possibly four, ways of expressing the direct object:
Josh 10:40
Josh 10:41
pronominal suffix
Gen 18:7
explicit noun
Num 21:35
explicit pronoun
Jr\ah
* -* lK*-ta# u~vo% hy+ hK#Yw~ ~
hZ`u-^ du^w+ u~nr
} B
+ ^ vd}Qm
* ! u~vo% hy+ <K@Yw~ ~
bofw` Er~ rq*B-* /B# jQ^Yw] ~
w{Mu^-lK*-ta#w+ wyn`B-* ta#w+ w{ta) WKY~w~
It is beyond the scope of the present study to discuss these pronominal suffixes at
length, but the unit-in-context approach implemented here would seek to explain their use
on the basis of contextual factors rather than as stylistic preferences.
One of the other fundamental components of the Morpho-Syntactic Orientation is
an emphasis on syntax at all levels. In 1995, Lowery commented that
there remains much of syntax, and Biblical Hebrew syntax in particular,
which is not well understood. Could it be that this is because syntactic
usage is conditioned by yet another level of language, that of text, which
has yet to be taken into consideration? (Lowery 1995, 107)
83
This perceptive comment reflects the necessity of the analysis of the interaction
between morpho-syntax and textual phenomena. In many of the traditional and
descriptive grammars of biblical Hebrew, attention to syntax is minimal or limited to the
consideration of certain clause-level issues. One of the unfortunate consequences of
certain textlinguistic approaches is the tendency to move directly from the morphological
base of the traditional and descriptive approach to the analysis of textual features
circumventing important areas of phrase-, clause-, and sentence-level syntax that interact
with and are sensitive to textual phenomena. This is not a new idea; in fact, in Meeks
Presidential Address at the 1944 meeting of the Society of Biblical Literature, he
remarked:
We may be able to parse correctly every form in a passage and may know
the exact meaning of each word, but if we err at any point in the syntax,
even though slightly, our translation and interpretation cannot be correct.
For the correct understanding of the text nothing is so important as the
correct understanding of the syntax. (Meek 1945, 1)
The model implemented here emphasizes the need for thorough syntactic research
at all levels, investigating the pervasive interaction between linguistic items and their
context(s) of use. Foley and Van Valin comment as follows on the role that grammatical
or syntactic constructions can have in the broader textual context:
many of the morphosyntactic phenomena which have traditionally been
at the center of linguistic investigation, e.g., case marking, grammatical
relations, and clause linkage, are crucially involved in reference and
predication and therefore play fundamental roles in discourse tracking.
(Foley and Van Valin 1984, 2)
84
This interactive morpho-syntactic orientation is consistent with and builds on the
descriptive orientation described in the previous section. The goal is the full description
of the morpho-syntactic phenomena of biblical Hebrew in their contexts of use.
5.2.2.2
Syntactic Constraints
An important concept in this morpho-syntactic orientation is that certain
constructions in language are framed in particular ways that constrain syntactic patterns,
such as the use of verb forms. The proper interpretation of the verb forms of biblical
Hebrew presents a variety of challenges and requires careful evaluation of the syntactic
context in which a given form occurs. This includes not only text type or genre, but
information type as well. Consider the following example:
God called the light day, and the darkness
He called night.
Gen 1:5
hl*yl
+ * ar`q* Ev#jl
) w^ +
In some approaches, the QATAL and WAYYIQTOL in Gen 1:5 would be discussed in
terms of the assumed temporal contrast between these forms. However, there is very
little that can legitimately be said about the use of QATAL in the second half of this verse
unless the syntactic pattern is considered. Minimal significance can be attributed to the
contrast between WAYYIQTOL and QATAL in this verse if there is some sense in which the
occurrence of these forms is governed by the syntactic shape of the sentence in which
both of these forms occur. Very little is accomplished by merely giving a sentence like
this a label such as chiastic without, first of all, addressing why it has this syntactic shape
and secondly, investigating its distribution. It is also important to consider what other
way(s) this sentence could have been framed and then analyze how this would affect the
85
information status of the sentence. For example, if Gen 1:5 had been written with two
WAYYIQTOLs,
it would still report the same events, i.e., the naming of light and darkness
as day and night respectively, but the narrative framing would be different. The
WAYYIQTOL-QATAL sequence
of the two events, whereas two WAYYIQTOLs would give a SEQUENTIAL depiction of the
events.
Some analyses identify one of the main features of QATAL as ANTERIORITY. This
may apply in many cases and may clarify some otherwise perplexing occurrences of
QATAL,
but full analysis must take the narrative and syntactic context into consideration.
In the Gen 1:5 example above, the WE-X-QATAL in the second half cannot be analyzed
independently of the first half with the WAYYIQTOL. In this construction, the temporal
feature of ANTERIORITY that QATAL often has does not appear to be operative; rather, this
construction syntactically frames the two namings as co-temporal. Analysis of an entity
like the WE-X-QATAL always needs to reflect sensitivity to the context in which it occurs.
The WAYYIQTOL WE-X-QATAL frame discussed here is one of the reasons why
the sentence is an important area of analysis beyond the clause. The WE-X-QATAL here is
not the same as the sentence-initial WE-X-QATAL which has its own peculiar discoursepragmatic function. Both need to be analyzed in their broader context, but this is beyond
the scope of the present discussion.
Another example of the WAYYIQTOL WE-X-QATAL frame is 2 Sam 2:12-13
displayed below:
86
Now Abner the son of Ner, went out
from Mahanaim to Gibeon with the
servants of Ish-bosheth the son of
Saul. And Joab the son of Zeruiah
and the servants of David went out
and met them by the pool of Gibeon;
and they sat down, one on the one
side of the pool and the other on the
other side of the pool.
Pc-vqw3ms np ncmsc-np Pc-ncmpc np
ncmsc-np Pp-np npXd Pc-np ncmsc-np
Pc-ncmpc np vqp3cp Pc-vqw3mpX3mp
Pp-ncfsc np Pd Pc-vqw3mp acp Pp-Pancfs Pp-ams Pc-acp Pp-Pa-ncfs Pp-ams
yd}bu
= w^ + rn}-/B# rn}ba
= ^ ax@Y}w~
.hn`oub=G] <y]nj
~ M
& m
^ ! lWav*-/B# tv#B-) vya!
2 Sam
2:12-13
The WAYYIQTOL WE-X-QATAL frame spans verses 2:12 and 13. The NASB,
shown in the left column, renders 12 and 13 as if they were both WAYYIQTOL, but the
beginning of 2:13 in BHS is a WE-X-QATAL which highlights the temporal PARITY of the
actions of Abner and Joab. The PARITY of their actions is also indicated in 2:13 by wD`jy= ~
and hZ\m! hL#a@ and hZ\m! hL#aw@ .+ The narrative depiction of the tension of the
impending war is heightened by the parallel framing of Abner and Joabs meeting. This
type of analysis not only takes into consideration the syntactic shape of the clause at
hand, but also considers the narrative purpose or pragmatics of why the particular clause
is found where it occurs.
Subordination is another example of a category of syntactic patterns that requires
special attention. For example, clauses with rv#a& or yK! are typically followed by certain
verb forms. Familiarity with these patterns eventually develops into a degree of
anticipation of what will come next. For example, clauses with rv#a& may be followed by
a QATAL as in 2 Sam 11:16 ud~y` rv#a& or by a participle as in 1 Kgs 5:13 ax@y{ rv#a.&
87
When Joab guarded the city, he put
Uriah at the place where he knew
there were valiant men.
2 Sam
11:16
zr\ah
# -* /m! <yx!uh
@ -* lu^ rB@dy~ w+ ~
1 Kgs 5:13
(4:33)
These examples provide a good contrast between the function of the QATAL and
the participle that can be explained in terms of temporal reference. The QATAL in 2 Sam
11:16 reflects the fact that Joabs knowledge is prior to giving the order to Uriah and the
participle in 1 Kgs 5:13 gives present temporal reference. It is not difficult, however, to
conceive of the same clause in 1 Kgs 5:13 with a QATAL ax*y` rv#a& if the intended
temporal reference is past: which grew. These examples demonstrate the need for not
only recognizing the function of rva as a relative pronoun, but also for developing an
awareness of the syntactic patterns of the relative clauses it forms.
Awareness of these patterns is crucial for determining whether certain
occurrences are significant, since significance is typically the result of occurrences which
depart in some way from the standard patterns. If the clause being analyzed matches a
regular syntactic pattern, there is minimal exegetical significance. For example in 1 Kgs
5:13 above, the use of the participle for present temporal reference conforms to the
regular pattern. As such, it would not be legitimate to emphasize the aspectual features
of the participle in this type of occurrence. All of this underscores the need for an
88
interactive, unit-in-context morpho-syntactic analysis that builds on a full and thorough
description of biblical Hebrew.
1 Kgs
21:15
The actors of the passive Pual lQ^s% in this verse are the men who followed
Jezebels orders to kill Naboth so that Ahab could take possession of his vineyard. From
a functional perspective, analysis of the passive in 21:15 is not merely a matter of
identifying the form as Pual, but rather it involves considering the possible interaction of
linguistic forms and communicative functions (Foley and Van Valin 1984, 21) that may
89
have motivated the use of the Pual form in this verse. From this perspective, one possible
explanation for the use of the passive (Pual) rather than active (Qal) is that the passive
allows explicit reference to the action done to Naboth, but without giving further
prominence to the thugs themselves at this point in the narrative.
An analysis that merely parses verbal forms as Pual or Qal without consideration
of the contextual factors that shaped the author or speakers choice of one form or
another, is incomplete. As Dik comments:
a theory of grammar should not be content to display the rules of
language for their own sake, but should try, wherever possible, to explain
these rules in terms of their functionality with respect to the ways they are
used. (Dik 1978, 2)
Many functional models have shown themselves to be effective in the analysis of
the worlds languages; responsible study and interpretation of the biblical text calls for
the use of every means at our disposal to understand it.
The typological approach analyzes languages as representative members of the
whole set of human languages, comparing and contrasting structural similarities. The
typological orientation to language complements the descriptive approach discussed
above as seen in the following statement by Shopen in Language Typology and Syntactic
Description:
One must welcome the particular characteristics of a language that make it
a unique cultural artifact, and different from any other, but at the same
time, one will understand the workings of the language better, even in its
most distinctive traits, the more one knows what languages tend to be like.
(Shopen 1985, 1)
It is important to notice that the comparison of language features within a
functional-typological perspective differs greatly from the type of comparative approach
90
which assumes that one language is superior to another. The purpose of comparison in a
functional-typological approach is to see how the function of a particular item in one
language can improve our understanding of the function of a comparable item in another
language.
Another important aspect of the typological approach, as expressed by Miller, is
that cross-linguistic evidence demonstrates the plausibility of our analysis (Miller
1996, 200). Arguments for a particular analysis in one language find support in the
analysis of comparable linguistic features in other languages. As van der Merwe
comments:
to be successful in studying Biblical Hebrew from a discourse
perspective, the researcher must also allow imaginative hypotheses to be
formulated on the basis of observations of discourse phenomena in other
languages. (van der Merwe 1994, 39)
Even before the recent development of typological approaches, James Barr in his
Semantics of Biblical Language, argued similarly that a semantic method which is used
for Greek or Hebrew must be integrated with general linguistics as a whole, and must
therefore be open to relevant data for semantics of any language (Barr 1983, 25). In the
functional-typological orientation implemented here, biblical Hebrew is perceived and
studied as one member of the whole set of the worlds languages. As such, it may have
certain unique features, but it will also share features with other languages. Barr stated
that the isolation of Hebrew from general linguistics tends to heighten the impression of
Hebrew being quite extraordinarily unique in its structure (Barr 1983, 291). Awareness
of how other languages function can not only help understand the functions of linguistic
91
items in biblical Hebrew, but it can also help dispel the image of Hebrew as a strange
language.
92
14:1-2, where the kings Amraphel, Arioch, Kedorlaomer, Tedal, Bera, Birsha, Shinab,
Shemeber, and Zoar are all named, it seems clear that the un-naming of Pharaoh in
Exodus 1 is a pragmatically motivated choice, which results in diminished prominence of
the otherwise powerful Pharaoh.
5.2.4.1
distinctions or text-types that need to be taken into account when reading or analyzing
any text. For the reader, awareness of the genre of the text is one of the most important
starting points, since the genre of the text affects reading strategy. When readers start to
read a text, they make a conscious or unconscious genre identification, which involves
further expectations concerning what is to come (Longman 1996, 141). Genre, in
Longmans Literary Approaches and other literary analyses, typically refers to the
synchronic identification of the type of literature to which a text belongs (Longman
1996, 141). The approach implemented here recognizes the need to identify the genre of
the whole text, but greater attention to discourse-pragmatic considerations is needed. For
example, within a text classified as narrative which emphasizes that there is a succession
of events (Longman 1996, 141), there will typically be other text-types which
accompany the events. From a discourse-pragmatic perspective, genre is an integral part
of the communication process. The speaker or writer employs the characteristic features
of genre as a mechanism for accomplishing his or her communicative goals.
The model implemented here is indebted to Longacres description in The
Grammar of Discourse (Longacre 1996), which seeks to identify the types of textual
93
components which make up the whole narrative. Longacres approach posits Narrative,
Procedural, Behavioral, and Expository as the basic text-types in text (Longacre 1996,
10):
+Agent Orientation
NARRATIVE
-Agent Orientation
PROCEDURAL
Prophecy
How-to-do-it
+ Projection
Story
How-it-was-done
- Projection
BEHAVIORAL
EXPOSITORY
Hortatory
Proposal
+ Projection
Eulogy
Scientific paper
- Projection
+ Contingent
Succession
- Contingent
Succession
94
that it demands that the exegete attend as closely to the peculiarities of the text as to its
similarities (Longman 1996, 143).
The text-types posited by Longacre allow for greater precision than the two-way
distinction that Niccacci makes between narrative and discourse. For Niccacci, the basic
division in biblical Hebrew is between the narrative framework and reported speech using
the terms narrative and discourse for these two text-types (Niccacci 1994, 119). This is
an important distinction, but within these two major types, further classification is
necessary. The concept of fluidity allows for the shifts from one type to another
sometimes subtle and other times more obviousthat are used as rhetorical strategies in
the communicative development of the text. The model implemented here takes a
position between Niccacci and Longacre. The narrative-discourse distinction of Niccacci
is significant, but needs greater sensitivity to text types. Longacres model also has its
merits, but needs to be applied fluidly.
In the present analysis, an awareness of text-type is fundamental. There is
undoubtedly a crucial distinction between the two types which Niccacci refers to as
narrative and discourse, but the terms preferred here are narrative and reported speech.
More important than the actual terms, however, is sensitivity to the texts transitions and
movements which can affect the grammatical and syntactic realization of the text.
5.2.4.2
and coherence are features of all spoken or written communication. Cohesion refers to
the way in which the communication itself reflects syntactic and semantic
95
connectedness. This may vary in degree from one text to another, but the assumption
is that all communication seeks cohesion. Languages have a variety of mechanisms at
their disposal for building cohesion into the text and this needs to be carefully evaluated
for each language being studied. Halliday and Hasans Cohesion in English was the first
monograph to deal with the cohesive mechanisms of English, exploring the semantic
resources which are drawn upon for the purpose of creating text (Halliday and Hasan
1976, 10). Cohesion is assumed to be operative in every language even though the
language-particular mechanisms may vary. Berlins article Lexical Cohesion and
Biblical Interpretation explores some of the mechanisms employed in the biblical
Hebrew text (Berlin 1989, 29-39).
The notion of cohesion is crucial to the analysis of yh!yw+ .~ Anaphora (backward
reference) and cataphora (forward reference) are cohesive devices. One of the issues in
the analysis of yh!yw+ ~ is whether the connection is to what precedes or what follows (see
question 1 in 4.1.5). Further discussion specifically regarding yh!yw+ ~ is suspended,
however, until Chapter 10.
Coherence, on the other hand, is the connection of the text or communication with
the world in which language is used. A text is coherent to the degree that it appropriately
reflects the world in which the text is produced. Perceptions of coherence may vary
depending on the worldview underlying the text. In other words, awareness of the role
worldview can play in shaping a text is foundational. Even though there may be
incongruences with the analysts own worldview, coherence is assumed. This is not a
mere assumption of textual integrity because it is the biblical text; coherence is a
96
characteristic of human communication. The assumption of
coherence in communication is cognitively parallel to the
mental process used to interpret the fuzzy images reprinted
here (Churchland 1996, 110-12). When normal
communication takes place between members of the same
speech community, coherence is based on the speakers
Illustration 1: Don Quixote
figure, riding a horse.
communication is produced through one set of rules and interpreted through another. The
attempt to interpret is the minds search for coherence in spite of the fuzzy edges
resulting from the mismatch of rules.
One of the reasons why the study of a
languages cohesive devices is so
important is that they are one of the
primary means by which proper
interpretation of the text takes place. In
cross-linguistic communication, the
fuzzy edges will be filled in by the
interpreters own linguistic system. In the illustrations above, the gaps or fuzzy edges are
filled in by the mind based on perceptions of the real world. Similarly, in fuzzy
communication the mind fills in the gaps with the knowledge the interpreter brings to the
text.
97
5.2.4.3
Context-Sensitivity
One of the foundational concepts of the analysis presented here is that language is
not only used in context but language is also context sensitive. As Schiffrin states in the
introduction to Discourse Markers:
1.
2.
3.
4.
Context-free language use does not exist. Linguistic utterances may be extracted
from their original context, or language instructors may create what they think are
representative, context-free utterances, but the classroom or the pedagogical setting in
which they are created are contexts which shape the utterances. The automatic search by
language users for a plausible context for apparently anomalous linguistic itemsor the
search for coherenceis an increasingly accepted principle of linguistic analysis. As
Schiffrin states:
the role of context is so pervasive that it figures even in grammatical
analyses whose data consist of individual intuitions about isolated
sentences. Not only is the introspection which accompanies intuition
actually a special kind of cognitive context in and of itself, but (as teachers
of introductory syntax can no doubt attest) individuals are very adept at
imagining discourse contexts in which ungrammatical sentences find a
natural home. (Schiffrin 1988, 3)
The particular context within which language is used will affect its form and
shape. In the study of any language, including biblical Hebrew, all linguistic items must
be discussed in their various contexts of usage. Context includes not just awareness of
the content of the surrounding clauses, but it incorporates syntactic contexts as well as
narrative contexts. Syntactic contexts would be, for example, clauses with rv#a,& clauses
98
with yK!, as well as other dependent and independent clauses. Narrative contexts include
not only different text-types, but also the stage of development within a text. Different
stages of developmentsuch as aperture, nucleus, closuremay have certain
characteristic features. Episode boundaries, commented on so frequently in the review of
scholarship, are an example of a particular narrative context that are expected to have
certain characteristic features.
5.2.4.4
Choice
One of the results of the context-sensitive nature of language is that
hOv!b= hl*ka
= * yr}ja
& ^ hN`j^ <q*Tw* ~
1 Sam 1:9
ht)v* yr}ja
& w^ +
aS@Kh
! -^ lu^ bv@y{ /h@Kh
) ^ yl!uw@ +
.hw`hy+ lk^yh@ tz~Wzm=-lu^
Choice in narrative strategy is operative here. The use of yl!uw@ + is not the only way
the information about Eli being in the temple could have been conveyed. So far in
99
1 Sam 1, Eli has only been indirectly introduced as the father of Hophni and Phinehas
(1:3) and bv@y{ /h@Kh
) ^ yl!u@w+ keeps Eli in a secondary participant role. This is further
reinforced by rm@v) yl!uw@ + in 1:12 and then u~mV
@ y* ] aO Hl*oqw+ in 1:13. Eli comes into more
# j
= Y= w~ .~ The way the text
active participant status in the final part of 1:13 with yl!u@ h*bv
presents the participants is a product of choice.
5.2.4.5
Default
The notion of default is one of the most useful concepts for determining the
possible significance of the occurrence of a linguistic item. At the most basic level, to say
something is default is to say that it is the automatic or typically expected structure. In
the context-sensitive linguistic network of language, the concept of default must always
take into account the varying situational dimensions of language use. The concept of
default is not meant to posit a context-free basic structure in a prescriptive sense. Default
structures will vary depending on genre, register, dialect, and other contextual factors that
may impinge upon language use. Once these factors have been taken into consideration,
frequency of occurrence can help discern what is a default pattern or item. In any case,
judgment concerning whether or not a certain structure is default must only be made after
extensive familiarity with the biblical Hebrew text.
5.2.4.6
Markedness
A further concept that is fundamental to the discourse-pragmatic orientation is
100
crucial part of the interpretive process. Since the Old Testament was written by speakers
of biblical Hebrew who knew when to used Hebrews unmarked and marked patterns,
competence in understanding their communication in Hebrew, therefore, necessarily
involves developing an awareness of what was characteristic and what was not
characteristic in their use of Hebrew under certain contextual conditions. This has farreaching implications for exegesis. When the notion of markedness is implemented in
exegesis, the more unexpected or marked patterns in the text are given greater
significance than the characteristic or unmarked patterns. For example, what is the
characteristic way of referring to someone who is already one of the main participants in
a narrative? A full answer to this question would involve extensive discussion, but
assume for now that the following example from Gen 11:30 reflects a characteristic
pattern:
Sarai was barren; she had no child.
Pc-vqw3fsXa np afs Pd PpX3fs ncms
Gen 11:30
In this verse, Sarai is referred to by name like the reference to Abram in Gen 12:1:
the LORD said to Abram,
Pc-vqw3ms np Pp-np
<r`ba
= -^ la# hw`hy+ rm#aY{w~
Gen 12:1
Both of these examples appear in the narrative after Abram and Sarai have
already been introduced as main participants. This seems to be a typical pattern, but in
Gen 16:1 Sarai is referred to differently:
Now Sarai, Abram's wife had borne
him no children,
Pc-np ncfsc np Pn vqp3fs PpX3ms
Gen 16:1
101
Why does this more complete identification of Sarai as Abrams wife occur here
in the text? One possible explanation for the uncharacteristically full reference to Sarai
as Abrams wife, in this context, is to establish greater contrast between Sarai and Hagar
to make sure that the reader never loses sight of her crucial role as Abrams wife. Further
evidence of this distinction between Sarai and Hagar is seen in the explicit reference to
Ishmael in Gen 16:15 as the son which Hagar bore, rather than merely as Abrams son.
Sensitivity to markedness gives an added awareness of the purposeful use of
certain linguistic items that might otherwise seem to be nothing more than peculiar
variants in the text. The notions of choice, default, and markedness work together in such
a way that presence and absence are like two sides of the same pragmatic coin. Further
discussion of this important principle is found in Chapter 10.
102
representation of that communication is shaped by the cognitive processes which were
operative during the texts production.
One of the most fundamental cognitive notions that affects the analysts basic
perception of how language is used is the pervasive representational nature of all
language.
5.2.5.1
words such as nouns and verbs that refer respectively to things (objects and concepts) and
actions are not themselves the things and actions. They are linguistic representations of
those things and actions. This representational or metaphorical nature of language is one
of the most powerful, efficient characteristics of language. This does not bring into
question the reality of the real world or the referential nature of certain linguistic items,
but rather emphasizes the fact that the linguistic elements of language are not themselves
the entities they refer to.
Certain linguistic entities such as prepositions and connectors are seen in a new
light when analyzed from a metaphorical or representational perspective. For example, a
preposition like under not only represents the physical position of one object in relation
to another, but is also used metaphorically to represent the situation of being under the
control of something or someone. It is very common for prepositions to have this
representational function, which indicates the interconnectedness of the network of
spatial, temporal, and relational terms. In biblical Hebrew, then, it is no surprise that
103
prepositions are used spatially, temporally, and relationally. For example, b= is used both
spatially and temporally in Gen 6:4:
The Nephilim were on the earth in
those days,
<h@h* <ym!YB
` ^ Jr\ab
* * Wyh* <yl!pN! h
+ ^
Gen 6:4
104
the function of aspect is to allow the temporal dimensions of a situation
to be described from different points of view depending on how the
situation is intended to fit into the discourse. (Bybee 1985, 142)
So in Gen 1:5 referred to above, there is nothing inherently WAYYIQTOLish
about the first action, nor is the second action inherently QATALish. The use of
WAYYIQTOL
105
5.2.5.2
Information Structure
Another aspect of the cognitive dimension is the relationship between the
information structure of a text and the mental processes involved in its interpretation.3
The study of information structure
is not primarily concerned with the interpretation of words or sentences
in given conversational contexts, but rather with the discourse
circumstances under which given pieces of propositional information are
expressed via one rather than another possible morphosyntactic or
prosodic form. (Lambrecht 1994, 5)
In the information structure of a text, the various linguistic items which are used
to compose the text have different propositional or referential functions. Nominal items,
for example, refer to physical or conceptual entities in the world, while verbal items refer
to actions and states, and other linguistic items encode relationships and connections.
Still other items in the text are signals which aid the listener or reader in his or her
interpretation of the communication. These signals are not referential in the same way as
concepts and actions, but are essential for the proper interpretation of the text.
In the temporal organization of text, many time references are not referential in
the same way as other linguistic items, but are in the text to assist the reader/listener in
tracking properly with the temporal flow of the narrative depiction. With reference to
yh!yw+ ,~ it is important to know how it functions in the texts temporal organizationdoes it
maintain the same time frame of what precedes it or does it establish a new time frame?
Questions like this will be dealt with in Chapter 10.
3
The concepts implemented here are based on the model developed in Lambrecht (1994),
Information Structure and Sentence Form: Topic, Focus and the Mental Representations of
Discourse Referents.
106
The cognitive orientation implemented here has been particularly influenced by
Lakoff and Johnsons Metaphors We Live By (1980), Jackendoffs Patterns in the Mind:
Language and Human Nature (1994), Sperber and Wilsons Relevance: Communication
and Cognition (1995), Lambrechts Information Structure and Sentence Form: Topic,
Focus and the Mental Representations of Discourse Referents (1994), and Fauconniers
Mental Spaces: Aspects of Meaning Construction in Natural Language (1985). Rather
than go into further detail here, the reader is referred to these publications for broader
conceptual background.
5.3 Summary
Within this functional, typological, contextual and discourse-pragmatic
framework, text analysis requires sensitivity to the various intersecting networks within
which linguistic entities from every level function in communicative situations. The goal
is not just analysis of the discourse structure of a text, but attention to the communicative
nature of language use.
CHAPTER 6
ANALYTICAL PRELIMINARIES
6.1 Introduction
There are specific analytical preliminaries that require comment before getting to
the actual analysis of yh!y+w.~ The main sections in this chapter are:
6.2 The Verb in Biblical Hebrew
6.3 Clause Syntax in Biblical Hebrew
6.4 Narrative Time and the Hebrew Verbal System
108
109
entire history of the study of the Hebrew verb, but rather to give adequate background for
the present study of yh!yw+ .~
For greater detail on the history of the study of the biblical Hebrew verb, see
McFall (1982), Waltke and OConnor (1990), Hatav (1997), and Garr (1998).
writings and commentaries of Christian Hebraists until the grammars of Lee and Ewald
broke new ground in 1827 (McFall 1982, 17).
110
Ewalds description of the verb employed not tense but aspectual categories such
as completed and incompleted, and subsequently the terms perfect and imperfect
came into use to refer to the QATAL and YIQTOL forms respectively. The growing sense
that tense was not providing the explanatory key to the Hebrew verbal system led to
broad acceptance of the aspectual explanation. Rather than merely incorporate aspectual
insights into the tense model, however, aspect essentially replaced tense as the
explanatory category for the Hebrew verb. Just the fact, however, that aspect is a
significant grammatical category in some languages does not automatically mean that
aspect is grammaticalized in the same way in another language. This is exactly the same
kind of analytical fallacy into which the tense-based approaches had fallen by assuming
that only one such category could explain the full spectrum of the uses of the verbal
forms and how they relate to the Hebrew verbal system.
The paradigm shift has been so strong that Hebrew is sometimes referred to as a
tenseless language. Before proceeding, however, the term tenseless requires
explanation. This term should be understood as referring to a system in which the verbal
forms themselves are not morphologically marked with explicit tense indicators. As an
example of explicit tense indicators, consider the following verbs from Mapudungun:
amun
I went, go
amu-a-n
I will go
dungun
I spoke, speak
dungu-a-n
I will speak
The -a- in both examples is the explicit indicator of the future, contrasted with
non-future. In the Hebrew forms QATAL and YIQTOL, there are regular morphological
patterns, but the question is what the difference in form indicates. Those who use a term
like tenseless mean that the suffix-prefix contrast in form is not an indication of tense.
111
Indeed, as Waltke and OConnor comment, Biblical Hebrew has no such simple tense
forms (Waltke and OConnor 1990, 458). Unfortunately though, the use of a term like
tenseless can lead to widespread misunderstanding and misrepresentation, giving the
impression that tense or temporal marking is not even a part of the biblical Hebrew
system. This results in Waltke and OConnors observation that most translators fly by
the seat of their pants in interpreting the Hebrew conjugations and what others have
observed that the tenses are used promiscuously (Bayly 1782, 22). One of the most
intriguing comments regarding the perplexing verbal system of biblical Hebrew is by
Goldfajn:
Ever since I started looking into this subject I have been intrigued by the
evident discrepancy which one perceives between, on the one hand, the
relative ease with which most readers of BH seem to understand the
temporal indications of the BH verbs and, on the other hand, how difficult
it has proved to come up with adequate accounts which would explain the
temporal ordering of BH. (Goldfajn 1998, 2)
As mentioned above, aspect has essentially replaced tense as the descriptive
category for the biblical Hebrew verb, but this seems to be based on the assumption that a
verbal system exclusively marks only tense or only aspect. Tense and aspect, however,
are interconnected systems that have been artificially separated in linguistic descriptions
of verbal systems. As Payne states:
Tense, aspect, and mode are sometimes difficult to tease apart. In fact, it
may be that linguists have thought of these three categories as distinct only
because they are somewhat distinct in the classical languages and in IndoEuropean generally. (Payne 1997, 234)
112
In practice, tense theories admit degrees of aspectual nuance and aspectual
theories admit that the verbs have certain temporal features which are contextually
defined. The multidimensional analysis implemented in this study does not require that
there be any binary opposition between tense and aspect, since it is not expected that only
one category such as tense or aspect will be sufficient for full analysis of the Hebrew
verb. Tense and aspect are not mutually exclusive categories as if a tense system were
incapable of indicating aspectual notions or as if an aspectual system were incapable of
expressing temporal distinctions.
The attempts to describe the biblical Hebrew verbal system as if the verbs only
indicated tense or aspect have typically been made by imposing the expectations of some
linguistic system foreign to Hebrew. There is no theoretical reason why a verbal system
can only mark either tense or aspect; there is also no theoretical reason why the
distinctions of tense and aspect must be encoded morphologically in the verbal forms
themselves. More detail will come in the following sections, but in the model
implemented here tense and aspect are potential dimensions of all verbal forms.
Mood is another category used to capture the nuances of verbal systems.
Descriptions of biblical Hebrew have started taking mood into account. Indicative of the
overlap between the categories of aspect and mood is the following comment by
Chisholm: it is difficult to reduce the essence of the imperfect to a single concept, for it
encompasses both aspect and mood (Chisholm 1998, 89). Mood or modality is a
necessary dimension of the verb since all verbal forms characteristically indicate some
type of modality.
113
In general linguistics, the acronym TAM is
Aspect
verb
Mood
Tense
Applied to Hebrew, the verbal system would not be analyzed in exclusive terms
of tense, aspect, or mood; each verb would be considered for its potential temporal,
aspectual, or modal features. To TAM, however, another dimension needs to be added,
namely that of Pragmatic Role in order to represent the context-sensitive way in which
verb forms indicate temporal, aspectual, modal, and pragmatic features. TAMP, then,
incorporates Pragmatic Role as an integral part of the description of the verb form helps
systematize the contextual dimensions of usage in text. The full description of a verb,
therefore, is its TAMP profile, since no single dimension captures the whole description.
For example, within an aspectual system, the WAYYIQTOL is sometimes defined as
indicating SEQUENTIALITY. If this is seen as the essential character of the WAYYIQTOL, all
non-sequential uses become exceptions. For example, certain WAYYIQTOLs have a
summarizing character that is not sequential, whereas other WAYYIQTOLs clearly do
indicate a chronological progression of events.
SEQUENTIALITY
is an important feature of
the WAYYIQTOL, but not the defining characteristic. Likewise, if ANTERIORITY is seen as
114
the defining feature of the QATAL, apparent exceptions will need to be explained. In a
dynamic, context-sensitive view, certain features will be highlighted in certain contexts
and will not be in focus in others. Under certain pragmatic or narrative conditions, the
SEQUENTIALITY
favor of other features which come into play under those conditions. This does not mean,
however, that SEQUENTIALITY and ANTERIORITY cannot be primary or predominant
features of the WAYYIQTOL and QATAL, but they are not the only ones.
Consider 2 Sam 11:14-15 in this regard:
In the morning David wrote a letter
to Joab and sent it by the hand of
Uriah. He had written in the letter,
saying, Place Uriah in the front line
of the fiercest battle and withdraw
from him, so that he may be struck
down and die.
2 Sam
11:14-15
Notice that bT)kY= w] ~ occurs in both 11:14 and 11:15. After the yh!yw+ ,~ there are three
WAYYIQTOLs.
narrative clock ahead in somewhat similar fashion. This, however, is not the case, as
plotted on the following graph:
Text:
bT)kY= w] ~
jl^vY= w] ~
Time:
bT)kY= w] ~
jl^vY= w] ~
bT)kY= w] ~
Figure 8: Narrative Time
bT)kY= w] ~
115
The bT)kY= w] ~ in 11:14 and at the beginning of 11:15 refer to the same action and
cannot in any way be construed as indicating SEQUENTIALITY. Analysis of the
WAYYIQTOL
needs to account for these occurrences as well as for those which are indeed
sequential in nature. One of the ways to work toward an answer to what is happening
with this WAYYIQTOL is to ask 1) what other narrative options were available? and 2) how
would the narrative depiction have been affected by the use of another form? If, for
example, it were a WE-X-QATAL dw]D` bt^k* rp#Sb
@ W^ , what would change?1 The answer to
this question depends, of course, on what function is assigned to the WE-X-QATAL in the
overall verbal system in biblical Hebrew narrative. Niccacci states that when the WE-XQATAL
is used, the narrative flow is broken to provide information required for the
communication of the actual narrative (Niccacci 1990, 40). Based on this analysis by
Niccacci, dw]D` bt^k* rp#S@bW^ in 2 Sam 11:15 would break the flow of the narrative,
providing important information about the letter. It is crucial to observe, however, that
the clause initiated by the WAYYIQTOL bT)kY= w] ~ also provides information required for the
communication of the narrative, i.e., the message of the letter. In BHRG 346-50, the
authors comment on the [s]emantic-pragmatic functions of fronting, stating that
[t]he fronted complement or adjunct signals that an entity is introduced,
activated or reactivated to function as the topic of an utterance. The event
referred to by means of the predicate of that utterance is not discourse
active. (van der Merwe, Naud, Kroeze 1999, 347)
The sequence dwd btk rpsbW of waw + noun (object) + qatal + noun (subject) is
attested, for example, in Josh 2:25 hw`hy+-/t^n` lWbg+W and Judg 7:24 /oud+G] jl^v* <yk!al
* m
= W^ .
116
Based on van der Merwes discussion, the hypothetical WE-X-QATAL in 2 Sam
11:15 would be an example of reactivating an entity. The effect of this reactivation is to
heighten the level of focus associated with the nominal item in the WE-X-QATAL and this
appears to be precisely why the text does NOT have a WE-X-QATAL here. Even though the
WAYYIQTOL
in 2 Sam 11:15 seems anomalous, its use maintains the focus on David as
agent without shifting focus to the letter itself. This is a good example of a WAYYIQTOL
being used where SEQUENTIALITY is not the feature in focus.
117
other vowels of that language. The consequence of
front
high
central
[i]
back
[u]
mid
low
[a]
system by its
front
central
back
relationship in the
high
[i]
[u]
mid
[e]
[o]
low
[a]
system has a greater number of vowels, less variation is possible because each vowel
operates within a more restricted part of the total matrix.
It is important to recognize, however, that there is no difference in communicative
efficiency between languages with different vowel systems. At an abstract level of
acoustic phonetics, a vowel in the five-vowel system may appear to be more precisely
defined in contrast to one from the three-vowel system, but this neither impairs the ability
of the speaker of the three-vowel system to speak precisely, nor does a five-vowel system
enhance the ability of the speaker in any other language.
In parallel fashion, verbal systems may differ in terms of the number and type of
forms attested, but each language will be able to use its system to efficiently
communicate. When one language is compared to another, or when translation from one
language to another is attempted, the differences may seem insurmountable, but each
language within its own system is capable of efficient communication.
118
Just as a vowel from a three-vowel system cannot be equated with one from a
five-vowel system, there should be no expectation that the verbal forms of one language
will directly map onto the verbal forms of another. Each system needs to be understood in
its own right before comparisons are made. In terms of verbal systems, this means that
the study of forms and conjugations needs to be carried out within the whole system,
realizing that languages differ as to what is grammaticalized or encoded in the various
subsystems. For example, the verbal forms in Mapudungun, a language spoken in Chile
and Argentina encodes a two-way distinction between non-future and future. This does
not imply, however, that speakers of Mapudungun make no distinction between present
and past. The present-past distinction is not grammaticalized in the actual verb forms, but
other mechanisms are used to contextually indicate past or present temporal reference.
In the description of the verbal system of biblical Hebrew, the verb forms have
often been analyzed in isolation from the whole system within which they function. In
1990, in The Syntax of the Verb in Classical Hebrew Prose, Niccacci recommended that
a verb form needs to be studied in texts, not in isolation but in connection with all its
associated linguistic markers (Niccacci 1990, 10), but much remains to be done before
this type of (con)textual study is fully integrated into the study of biblical Hebrew.
What needs to be avoided at all costs is the insinuation that the use of the verb
forms is random or that the biblical Hebrew verbal system is impoverished, as seen in the
following quote from GKC:
While the Hebrew verb, owing to these derivative forms or conjugations,
possesses a certain richness and copiousness, it is, on the other hand, poor
in the matter of tenses and moods. The verb has only two tense-forms
(Perfect and Imperfect), besides an Imperative (but only in the active),
119
two Infinitives and a Participle. All relations of time, absolute and relative,
are expressed either by these forms (hence a certain diversity in their
meaning) or by syntactical combinations. (GKC 1910, 117)
Even though this standard grammar acknowledges a certain richness of the
Hebrew verb, the implication is that of inadequacy compared to some other linguistic
system.
The complexity of the verbal system is seen in that there is no single category
which adequately explains the choice of verb forms. The approach here is, in a sense, to
embrace the temporal, aspectual, and modal dimensions of the biblical Hebrew verb,
combining them all with discourse-pragmatic considerations of the narrative role of the
verb in context. Certain syntactic or discourse-pragmatic contexts will have their
preferred verb forms, certain optional verb forms if other factors come into play, and also
certain verb forms which are excluded from those contexts.
120
WEQATAL:
YIQTOL:
WAYYIQTOL:
QOTEL:
this term refers to the participle, which is an integral part of the verbal
system even though it shares certain characteristics with the noun. The
temporal value most typically associated with QOTEL is the present.
Context-sensitivity does not, however, open the Hebrew verbal system up to the
kind of apparent randomness of usage that is characteristic of certain analyses. For
example, Kelley states that
[t]ime (tense) is not inherent in the form of a Hebrew verb, but is
determined by the context in which it stands. Therefore, the same verb
form may be translated as past in one context, as present in another, and as
future in still another. (Kelley 1992, 82-83)
The problem with this statement is not necessarily its truth value, but rather the
unqualified way in which it states the possible translation values for the Hebrew verb
forms. One of the unfortunate results of this type of statement is reflected in Waltke and
OConnors comment that [m]ost translators, we think it is fair to say, fly by the seat of
their pants in interpreting the Hebrew conjugations (Waltke and OConnor 1990, 55).
Talstras comment is also insightful:
The impression one gets from this practice of translation is that exegetes
tend to exploit the existing theories of tense and aspect according to the
needs of an ad hoc textual interpretation. (Talstra 1997, 91)
121
But context-sensitivity is not flying by the seat of ones pants or ad hoc
interpretation. A context-sensitive approach to the biblical Hebrew verbal system
provides a principled basis for understanding the uses and functions of the verb forms.
In addition, the form WE-X-QATAL merits special comment. The X in this form
represents some other element, typically a noun, that occurs with w+, followed by a QATAL
form of the verb. This is a more accurate description than saying it is a WEQATAL
interrupted by a noun, because the WE-X-QATAL is more than just a WEQATAL with an
interposed noun. In the narrative structure of the biblical Hebrew text, WE-X-QATAL
performs an important function. All of these forms will be discussed in more detail in the
following chapters.
function
form
form
L1
L2
Hebrew will not necessarily have direct equivalents in other languages into which the
biblical text is translated. As shown in Figure 11, the challenge that faces the reader,
interpreter, or translator is, first of all, to understand the function of the connections and
transitions in the biblical text; then, secondly, to understand the function of the receptor
languages connections and transitions before, finally, deciding on the appropriate form
122
in the receptor language to most appropriately communicate the function of the original
text. This is, in fact, one of the most basic principles of translation.
Specifically with regard to yh!yw+ ,~ this
function
function
form
L1
form
L2
untranslated, however, short circuits the process in Figure 11. The move has been made
directly from L1 form to L2 form without mapping L1 function to L2 function, as in Figure
12. In English, the form and it came to pass is arguably clumsy, and as such may appear
to be superfluous in the text. It is sometimes argued that a more literal, direct translation
is not as susceptible to error as a more dynamic translation, but literal translations often
transfer forms into the receptor language that carry very little meaning. The repetition of
and it came to pass gives the Old Testament text a different feel, but does not help the
reader know why it is in the text. The opposite approach which would leave yh!yw+ ~
untranslated does not fare much better, since some English versions dont leave even a
trace of its occurrence. In both approaches the end result is very similar, since whatever
function yh!yw+ ~ may have is not represented in the text.
yh!yw+ ~ is just one of the many items that requires careful scrutiny at the level of
clause syntax. Full attention to clause syntax involves careful examination of the
differences between independent and dependent clauses, as well as the patterns of all
clause types such as relative clauses, temporal clauses, conditional clauses, concessional
123
clauses, etc. Clause sequences and the forms of the verbs found within them are also
important areas of the study of biblical Hebrew.
Introduction
Analysis carried out in a traditional/descriptive approach to the study of biblical
Hebrew will typically discuss a linguistic entity like w+ as a conjunction and then list
various meanings. For example, Pratico and Van Pelts discussion presents the basic
form of the conjunction w+ (and, but, also, even) (Pratico and Van Pelt 2001, 43), listing
the various meanings that may be necessary to properly render w+ in its various contexts.
One of the difficulties with this type of definition is that it seems perplexing to the
beginning learner that a word like and can also mean but. Often, in the attempt to
keep the description simple and brief, there is no discussion of the parameters by which
the choice between and and but needs to be made. It is presumably not the intention
of authors like Pratico and Van Pelt to make it appear that any meaning is possible
wherever w+ occurs, but unfortunately this unintended result is too often the case. Even if
the learner has an intuitive sense of the contexts which might call for one or another
meaning, it would be better if the learner were alerted to the fact that the varied senses
are indeed context-sensitive. Millers The Pragmatics of waw as a Discourse Marker in
Biblical Hebrew Dialogue is a good example of research which explores the functions of
w+ in text (Miller 1999). It is exactly this type of research in other areas of clause syntax
that will advance understanding of the syntax of the biblical Hebrew text.
124
w+ is a functor, and is best identified by its function rather than its meaning. From
this perspective, then, w+ functions as a conjunction, syntactically conjoining either similar
or dissimilar items. This more general function of w+ as conjunction avoids the need to say
that w+ means and, but, or, etc. This is not mere semantics, but rather reflects an
analytical perspective that focuses on morphosyntactic function in order to simplify the
description of certain problematic linguistic items. It is not a different w+ with different
meanings, but the same linguistic item for which the syntactic function is contextsensitive. This perspective is also beneficial for the analysis of w+ at the clause level. By
focusing more on the function of w+ rather than a lexical meaning such as and, the
tendency will be to look for the clause level function rather than to default to translating
every w+ with and.
Cognitively, the analysis of w+ does not focus on its meaning, but rather on its
potential effect on the interpretation or processing of the text. From this perspective, the
occurrences of w+ will be read in terms of the connection(s) they establish and maintain
throughout the text. As Halliday and Hasan state, the function of conjunction is a
specification of the way in which what is to follow is systematically connected to what
has gone before (Halliday and Hasan 1976, 227).
The description of w+ as conjunctive or disjunctive states a fundamental distinction,
but there are many context-sensitive nuances that need further analysis. The following
three sections discuss some of the basic functions of w+.
125
6.3.1.2
Nominal Conjoining
One of the most straightforward examples of w+ in its function as nominal
<y]mV
^ h
* ^ ta@ <yh!Oa$ ar`B* tyv!ar}B=
Gen 1:1
Jr\ah
* * ta@w+
(or), or DISJUNCTIVE (but). In this syntactic pattern, w+ will not mean God
created the heavens OR the earth. In nominal conjoining, the ALTERNATIVE and
DISJUNCTIVE
ALTERNATIVE
a man or a woman
Exod 21:28
There are, of course, variations on this basic pattern and numerous examples of
more complex structures. For more detail on the syntax of w+ with nominal items see, for
example, BHRG 31.
6.3.1.3
Verbal Conjoining
When w+ conjoins verbal items (V w+ V), it is crucial to take careful note of the verb
forms which are linked in this way. The conjunctive function of w+ is operative when the
successive verb forms are identical in formthe most typical example of this in narrative
are sequences of WAYYIQTOL. This is where w+s function of conjoining coordinate
syntactic items is fundamentalw+ should not be read as and or then with every
126
WAYYIQTOL, but rather should be seen as the syntactic indicator which signals the
conjoining of these similar verbal items.
When coordinate verbal items are conjoined by w+, caution needs to be exercised to
avoid superimposing the logico-temporal structure of another language. Consider Exod
4:31:
So the people believed; and when
they heard that the LORD was
concerned about the sons of Israel
and that He had seen their affliction,
then they bowed low and worshiped.
<u*h* /m@aY& w~ ~
Exod 4:31
la@rc
` y= ] yn}B-= ta# hw`hy+ dq^p-* yK! Wum=vY= w] ~
<y`nu
+ -* ta# ha*r` yk!w+
WdQ=Yw] ~
.WWj&Tv
^ Y= w] ~
The four WAYYIQTOLs in Exod 4:31 are linked by the w+. The question to ask is on
what basis did the translators decide to use when in English? The point is not that
every independent clause in Hebrew must be represented by an independent clause in
English (or any other language). However, since Hebrew does have a means of explicitly
) k
= ,! there needs to be a principled basis by which these
stating when they heard: u~mv
translation decisions are made. In English, when they heard then they bowed low,
implies a particular logical and temporal dependency that is not an inherent part of the
syntactic arrangement of the Hebrew linkage of WAYYIQTOL forms.
It is appropriate to avoid a mechanical rendering of every w+ with and, but it is
important to proceed on the basis of awareness of clause syntax and the function of w+. A
crucial difference in approach can be seen in the following two alternatives. The
translator might say:
127
1) At some abstract, literal level, w+ seems to mean and, so the proper gloss for
every occurrence is and. The repetition of and, and, and does not sound like
good English, so some occurrences of w+ can be translated then or so and
others can be left out so that the English is not too clumsy.
2) The function of w+ is to conjoin coordinate nominal or verbal items. And is
usually the appropriate translation in compound nominal and verbal phrases.
Some occurrences of w+, however, indicate other types of syntactic relationships.
The function of these syntactic relationships needs to be determined so that an
appropriate, equivalent relationship can be expressed in the translation of the
Hebrew clause.
In the model implemented here, the second approach is preferred, since the
function of linguistic items is fundamental to their proper analysis. A good example
showing different uses of w+ is Gen 24:30:
When he saw the ring and the
bracelets on his sister's wrists, and
when he heard the words of Rebekah
his sister, saying, This is what the
man said to me, he went to the man;
and behold, he was standing by the
camels at the spring.
Pc-vqw3msXa Pp-vqc Po-Pa-ncms PcPo-Pa-ncmp Pp-ncfdc ncfscX3ms Pc-PpvqcX3ms Po-ncmpc np ncfscX3ms Ppvqc Pd-vpp3ms PpX1cs Pa-ncms Pcvqw3msXa Pp-Pa-ncms Pc-Pi vqPms PpPa-ncmp Pp-Pa-ncfs
<z\Nh
\ -^ ta# ta)rK
+ ! yh!yw+ ~
Gen 24:30
128
case of the conjoining of two coordinate verb forms, i.e., two WAYYIQTOLs. The final
occurrence is not the standard nominal or verbal conjoining, but is rather an example of
one of the other functions performed by w+, discussed in the following section.
6.3.1.4
Other Functions of w+
The N w+ N and V w+ V examples in the above sections are only a part of the overall
function of w+. In both cases, it is crucial to keep in focus that w+ conjoins coordinate
nominal elements or coordinate verbal elements. When w+ is not used to syntactically link
coordinate elements, its function changes. This is not a different w+, but rather is evidence
of the context-sensitivity of w+ in a different syntactic environment. The typical use of w+ in
this syntactic environment is the WE-X-QATAL as seen in the following example:
All the persons who came from the
loins of Jacob were seventy in
number, but Joseph was already in
Egypt.
Exod 1:5
In this example and others like it, the w+ does not link either coordinate nominal or
coordinate verbal elements. The use of the WE-X-QATAL primarily indicates that the
relationship between the verbal element of this clause (hy`h)* and the preceding verbal
element (yh!yw+ )~ is not the same as the relationship between two verbs of the same form,
such as two WAYYIQTOLs conjoined by w+. What effect would there be on the narrative if
the second clause of Exod 1:5 were <y]rx
` m
= b
! = [s@oy yh!yw+ ?~ The NASB reflects the WE-XQATAL
by using but and placing already in italics: but Joseph was already in Egypt.
Presumably, if this clause had yh!yw+ ,~ it would be translated and Joseph was in Egypt.
The use of more than one yh!yw+ ~ in close proximity is attested as in Gen 39:2, so the use of
129
yh!yw+ ~ in this last clause of Exod 1:5 would be conceivable. The question, however, is what
the effect would be on the narrative depiction of the events referred to in these clauses.
Questions like this underscore the need for the present study. Further discussion is
postponed, however, until later sections of this study.
however, is not necessarily synchronized with the events as they happened. The temporal
130
organization of a text is a representation of the event world, not an exact synchronized
match. The following sequences of WAYYIQTOLs are examples of this mismatch.
108
4
3
11
126
the WAYYIQTOL.
6.4.1.1
PROGRESSION:
1 Sam 3:45
All the WAYYIQTOL verbs move the narrative along in temporal progression. The
events are depicted as if they were a series of points along the same line.
131
6.4.1.2
EXPANSION:
Wc*ul
@ = rm#aY{w~ qj*xy= ] /u^Yw~ ~
Gen 27:37
The speech event to which /u^Yw~ ~ and rm#aY{w~ refer is one and the same, thereby
making the relationship between the two WAYYIQTOLs non-SEQUENTIAL. One of the
questions that emerges from this example is whether this is merely a case of hendiadys
(lit., one through two) or whether there is some function that would not be realized with
just /u^Yw~ ~ or rm#aY{w,~ but not both verbs together. As a first step toward answering this
question, it is instructive to consider the frequency of occurrence of /u^Yw~ .~ This is an
example of how frequency of occurrence is a point of entry into other areas of analysis
which are more qualitative and pragmatically oriented.
Out of 126 third person occurrences of hn`u,* only 11 occur without an
accompanying verb. In certain occurrences, for example 1 Chr 21:26, the verb hn`u* is
used for a non-verbal response, in this case with fire. In other cases, the response is not
reported speech like the other 115 cases. This is just a preliminary glance at the data, but
these are examples of the types of contextual factors that need to be considered. This
analysis indicates, however, that the majority of occurrences are indeed two WAYYIQTOLs
used non-sequentially to refer to one event.
132
6.4.1.3
COMPRESSION:
of time
yw]l@ tyB@m! vya! El#Yw} ~
Exod 2:1-2
RAPID SUCCESSION:
<yv!du
` & dyz]nW+ <j#l# wc*ul
@ = /t^n` bq)uy& w~ +
Gen 25:34
This goes beyond the more common sequence of two WAYYIQTOLs considered
above as EXPANSION. These sequences depict a series of actions, creating an effect of
rapid succession. Even
though narrative time
is typically less than
event time, these
event world:
textual depiction:
Figure 15: Depiction of Events in Narrative
133
sequences are even more tightly compressed.
The following chart shows the types of sequences attested in the Hebrew Bible:
Sequence of 3
WAYYIQTOLs
Gen 25:17; Exod 1:7; 2:16;
Josh 8:14; 9:4; Judg 6:29;
9:27; 18:21; 19:4, 10, 14;
Ruth 2:14; 1 Sam 15:12;
18:25; 2 Sam 1:12; 17:23;
1 Kgs 19:21; 2 Kgs 7:8;
10:9; Jer 26:21; Ezek 16:7
Sequence of 4
WAYYIQTOLs
Gen 43:31; 1 Kgs 19:8;
2 Kgs 6:23; Neh 9:25
Sequence of 5
WAYYIQTOLs
Gen 25:34
These sequences have a distinct aural impact because of the repetition of the Yw~
prefix of the WAYYIQTOL form.
6.4.1.5
INCLUSION:
INCLUSION
Jr\ah
* -* lu^ cm@rh
{ * rc*B-* lK* uw~gY+ w] ~
Jr}Vh
) ^ Jr\Vh
# -^ lk*bW= hY`j^bW^ hm*hB
@ b
= W^ [ouB*
Gen
7:21-23
.<d`ah
* * lk)w+ Jr\ah
* -* lu^
wyP*aB
^ = <yY]j^ j~Wr-tm^vn= ] rv#a& lK)
.Wtm@ hb*rj
` B
* # rv#a& lK)m!
hm*da
` h
& * yn}P-= lu^ rv#a& <Wqy+h-^ lK*-ta# jm^Yw] ~
<y]mV
^ h
* ^ [ou-du^w+ cm#r-\ du^ hm*hB
@ -= du^ <d`am
* @
Jr\ah
* -* /m! WjM*Yw] ~
.hb*TB
@ ^ w{Ta! rv#aw& ~ j~n-{ Ea^ ra#Vy* w] ~
What is the relationship between the WAYYIQTOLs here? In 7:21-22 the first
WAYYIQTOL,
uw~gY+ w] ,~ is the main verb. But when the next WAYYIQTOL, jm^Yw] ,~ at the
beginning of 7:23 is read, the expectation of SEQUENTIALITY is not fulfilled. Upon closer
134
examination, it becomes evident that jm^Yw] ~ is included within the event frame set up by
uw~gY+ w] .~
6.4.1.6
REGRESSION:
This example, discussed above in 6.2.1, differs from INCLUSION since there is an
intermediate WAYYIQTOL, jl^vY= w] ,~ that moves the narrative along the timeline. The
second occurrence of bT)kY= w] ,~ then, goes back in time.
Now in the morning David wrote a
letter to Joab and sent it by the hand
of Uriah. He wrote in the letter:
Place Uriah in the front line of the
fiercest battle and withdraw from
him, so that he may be struck down
and die.
2 Sam
11:14-15
CONCLUSION:
In certain WAYYIQTOL series, the final occurrence concludes and summarizes the
series. Consider the following example from 1 Sam 31:4b-6:
So Saul took his sword and fell
on it. His armor bearer saw that
Saul was dead, and also fell on
his sword and died with him.
Thus Saul died with his three
sons, his armor bearer, and all
his men on that day together.
1 Sam 31:
4b-6
135
WAYYIQTOL, but the function is not to indicate the typical SEQUENTIALITY associated with
the WAYYIQTOL.
6.4.2 Summary
One of the main implications of the preceding discussion is that the assumption
that all WAYYIQTOLs are SEQUENTIAL is not congruent with the data. If all WAYYIQTOLs
are not the same, then it follows that not every yh!y+w~ can be expected to be the same. Just
as the preceding categories of WAYYIQTOL bring out the context-sensitive nature of this
form, so also the analysis of yh!yw+ ~ requires a context-sensitive approach: not every yh!yw+ ~
will be performing the same function in the text.
CHAPTER 7
DISTRIBUTIONAL ANALYSIS OF yh!y+w~
7.1 Introduction
In analysis of the type presented here, it is extremely important to have a
principled basis on which to evaluate the different occurrences of a linguistic entity. At
one extreme is the approach which assigns one gloss to the item being studied wherever
that item occurs. In the more context-sensitive approach advocated here, it is expected
that a slightly different sense will often be assigned to a linguistic entity based on the
context in which it occurs. What is needed, however, is a set of criteria for evaluating the
different occurrences while at the same time avoiding reading too much into each
different occurrence. There needs to be some control mechanism to prevent ad hoc
interpretation of each new occurrence.
Distributional analysis involves locating each occurrence, but this differs
significantly from the type of statistical analysis that would perhaps calculate the number
of occurrences per 1,000 words of text. The type of distributional analysis recommended
here proceeds on the basis of quantitative data and moves to qualitative analysis of the
possible patterns which emerge from how the occurrences are distributed throughout the
text. This analysis requires constant sensitivity to all levels of linguistic context.
137
138
31
31
30
24
23
23
19
19
15
13
13
11
09
08
08
07
05
05
03 07
1 2 3
22
34
67 34
18
17
30 16
29
52 32
25
17
08 32 17
30 03 14
26 22
30 27 14 30 23
16
03 31 12 13 19 02 11
17
17 13 20 20 22 20 08 30 13 43
05
02 23 01 10 06 10 01 10 07 01 12 01 14 12 20 01 15 11 01 01 10 25 10 06 24 03
4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
CHAPTERS
139
Genesis, continued
19
18
17
16
15
21
14
20
13
19
12
18
11
15
10
13
09
11
08
10
07
07
06
29 06
05
28 05
04
27 05 51
03
24 02 13
02
07 02 20 08 21
28
01
23 01 02 01 01 35 02 24
20 01 15 09
# 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50
CHAPTERS
Exodus (44)
05
04
03
23
02 21 11
01 05 10
# 1 2 3
24
04
03
4
51
41
27
41
22
19
21 14 24
30 17 24 13
16
30 29
28 10 13 10 22 29 15 20 02 10 12 13 16
18
19 28
5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
CHAPTERS
Exodus, continued
05
04
03
02
27
01
13 24 17
# 35 36 37 38 39 40
CHAPTERS
Leviticus (1)
01
01
# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
CHAPTERS
140
Numbers (21)
06
52
05
43
04
37
03
36
02
12
35 25
23
32
01
43
01 06 11 01
31 07
41 02 19
16
12
# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
CHAPTERS
Deuteronomy (7)
01 03 16
23
11
05
24 05
# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
CHAPTERS
Joshua (62)
08
13 41
07
11 33
06
10 33
05
27
27
30
09 25
04
20
24
29
07 18 40
03
18 13 16 25 20
25 18 10 02 10 20
02
14 11 08 15 24 16 11
23 02 05 01 02 10
01 01 05 02 01 01 08 05 14 01 01 01 16 01 05 01 12 01 04 17 01 29
# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
CHAPTERS
Judges (53)
07
06
05
04
03
02
01
#
28
19 27
14 18
03 04 10
1 2 3
12
39
11
40
39
07
38
35
20 25 05
27 15 33
29
17 21 04
25 09 27
05 14 20 15 17 16 01 02
07 06 26 42 04 04 09 02 11 01 04 01 01 46 04
4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
CHAPTERS
1 Samuel (67)
08
07
06
05
04
03
02
01
#
30
52
29
25
19
19
14
20
18
14
15
10 23
42
04
10
13
15
09 20 35
17 38
02 36 09 05 10 10 02 26 27 11
14 21 06 09 27
26 06 37
16 25
01 17 02 01 09 01 02 01 01 11 11 10 01 10 06 01 07 24 02 06 02 02 07 01 01 08
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
CHAPTERS
141
2 Samuel (56)
07
06
05
04
03
02
01
#
15
23
14
02 17 06
06
02 11 02
04 02
01 01 01 04 13 01 01
1 2 3 4 5 6 7 8
20
20
27 38 32
19
16 36 12
18
14 30 30 07
08 26 18
02 18 23 02 27 07 10 15 19
01 01 03 01 01 16 21 06 03 01 19 11 09
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
CHAPTERS
1 Kings (81)
09
29
08
27
07
26
06
24
05
21
04
12
03
11
02
39
06 11
01 04 15 18 01 02 01
# 1 2 3 4 5 6 7
45
40 34
44
29 31
36
25 24
29
28
24 23
17 27 40 27
20 30 20
31 17 17 39 17
54
15 22 20 28 18 08 07 29 16 33
17 10 26 04 20 06 25 29 11 07 04 26 15 32
10 01 14 03 02 04 06 21 01 02 01 13 12 01 02
8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
CHAPTERS
2 Kings (57)
07
06
05
04
03
02
01
# 1
40
25 30
18 26
27 11 25 21
28
11 20 08 08 24 20 15 25
25 37
27
09 15 08 07 20 18 05 09 11
12 07 09 35
11
25
01 05 06 02 05 16 03 22 07 03 07 21 05 05 03 01 01 04 03 01 01
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
CHAPTERS
Isaiah (17)
02
18
13
01
25 01 05
02
07 03
11
# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
CHAPTERS
Isaiah, continued
02
38
19
01
01 01 04
18
15
08
# 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66
CHAPTERS
142
Jeremiah (46)
05
04
03
02
01
#
27
13
23
11
08
23
16
04
06
17
12
19 12 09 13
03 01
03 16 01 05 03
10 04 08 01 30
26 01 12 11 01 06
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
CHAPTERS
Jeremiah, continued
05
13
04
07
03
06
02
04 07 08
31
01 04 01 07 01 22
04
# 39 40 41 42 43 44 45 46 47 48 49 50 51 52
CHAPTERS
Ezekiel (71)
05
04
03
02
01
#
25
03
01
1 2
22
16
16
03
3
26
34 11
21
34 07
23
27
17
23 06
13 23
20
23 23
14 08 12 19 06
02 06 17 10 15
11 20 17 21
01 01 01 08 06 13 01 01 02 01 01 01 01 14 01 01 01 01 01 01 01 01 01 17 01 01 01 01 01 01
4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
CHAPTERS
Ezekiel, continued
05
04
03
02
15
01 16 07 01
# 36 37 38 39
48
CHAPTERS
Hosea (1)
01
11
# 1 2 3 4 5 6 7 8 9 10 11 12 13 14
CHAPTERS
Jonah (5)
02 04
01 01 01 01 08
# 1 2 3 4
CHAPTERS
Habakkuk (1)
01 03
# 1 2 3
CHAPTERS
143
Haggai (2)
01 03 20
# 1 2
CHAPTERS
Zechariah (9)
05
13
04
12
03
08
02
04 18
01
08 09 01 01
# 1 2 3 4 5 6 7 8 9 10 11 12 13 14
CHAPTERS
Psalms (6)
01
19 09
# 1 18 33
11 03
69 76
22
94
CHAPTERS
14
118
Job (11)
05
04
03
02
01
#
13
06
05
03
03 01
16
31
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
CHAPTERS
Job, continued
05
04
03
13
02
12
01
07
# 36 37 38 39 40 41 42
CHAPTERS
Ruth (7)
03
02
01
#
19
01
16
01 17 08 13
1 2 3 4
CHAPTERS
Lamentations (1)
01
37
# 1 2 3 4 5
CHAPTERS
150
144
Esther (7)
03
15
02
08
02
01 01 07 04 01
# 1 2 3 4 5 6 7 8 9 10
CHAPTERS
Daniel (5)
03
02
01
#
21
16
(Aramaic) 15
06
02
1 2 3 4 5 6 7 8 9 10 11 12
CHAPTERS
Ezra (1)
01
04
(Aramaic)
# 1 2 3 4 5 6 7 8 9 10
CHAPTERS
Nehemiah (16)
04
10
03
09
02 04 38 06 16
19
01 01 01 33 01 01 01 01 17
03
# 1 2 3 4 5 6 7 8 9 10 11 12 13
CHAPTERS
1 Chronicles (32)
06
05
04
03
02
01
# 1
26
22
03
2
06
06
05
21
29 03 14 04
13
26 03 06 02
11 07
09 51
08 06
25 01 01 01 01 05 08 03 01 24
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
CHAPTERS
2 Chronicles (47)
04
03
02
01 14
# 1 2 3 4
12
23
13
26
11
10 32
18 16 21
11
25 12 02
05 31 29 19 12 11 14 11
32
27
09 07 01 13 02 02 01 15 07 05 03 01 01 09 08 04 03 05
08 26 25 19
5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
CHAPTERS
145
much into these generalizations. For example, yh!yw+ ~ should not be read as if its only
function is to indicate that the text in which it occurs is narrative. Even though genre is
one of the factors that needs to be taken into account, the function of yh!yw+ ~ cannot be
reduced to being the indicator of narrative. The chart of Leviticus is instructive in this
regard. The only occurrence of yh!yw+ ~ in the book of Leviticus is in 9:1, in the only past
narrative section of the book. But it is important to not overstate the significance of this
occurrence of yh!yw+ .~ yh!yw+ ~ may be in Leviticus 9 because this chapter is narrative, but the
reverse is not necessarily true that Leviticus 9 is narrative because yh!yw+ ~ occurs there.
It is also important not to overstate the significance of the absence of yh!yw+ .~ For
example, there are no occurrences of yh!yw+ ~ in chapters 12, 17, 21, 26, and 29 of 1 Samuel.
At the broadest level, there is no apparent difference in text-type that differentiates these
chapters from the rest of the book. These chapters are not less narrative because yh!yw+ ~
does not occur in them. yh!yw+ ~ is only one strand woven into the complex narrative
tapestry.
One of the questions that emerges from this distributional analysis is whether
there is any significance to clusters of yh!yw+ .~ For example, in 1 Sam 18, there are eight
occurrences when most of the surrounding chapters have only three or four. Clusters can
be seen in other books as well. The first step in answering this question would be to
separate the verbal occurrences from the temporal ones and then determine whether any
significant pattern emerges. This, of course, as stated previously, needs to take all levels
of linguistic structure and context into consideration.
146
These questions are attempts to explore whether any kind of profile of the text of
1 Samuelor any other bookcan be derived by looking at the pattern of occurrence of
yh!yw+ .~ yh!yw+ ,~ however, is not the only motor which keeps the text moving ahead. Other
mechanisms that move the text along or indicate transitions in the text need to be
considered along with yh!yw+ .~
7.3 Summary
One of the most important observations made in this occurrence-by-occurrence
study of yh!yw+ ~ is that the syntax of each of its occurrences needs to be carefully examined.
One of the inherent dangers in saying that yh!yw+ ~ performs a textlinguistic or discourse
function is that it can lead to the improper assumption that it functions only at that level.
There are numerous instances of yh!yw+ ~ functioning as a simple verb. The analyst must
determine when a particular occurrence is verbal or when a different function is being
performed. This is the goal of the following chapters.
CHAPTER 8
THE VERBAL USES OF yh!y+w~
8.1 Overview
As the WAYYIQTOL form of the verb hy`h,* yh!yw+ ~ often behaves like or carries out the
functions of a typical WAYYIQTOL in the biblical Hebrew text (see Niccacci 1990, 60).
Out of the total number of 864 occurrences of yh!yw+ ,~ 458 or 53% function as verbs. The
following sections display the various uses of yh!yw+ ~ in its basic function as a verb. One of
the questions in the full analysis of yh!yw+ ~ is whether it is just a verb in these cases, or
whether it might also have some other function, such as signaling divisions within the
text. This question is fundamental to the ultimate objective of this analysis, but judgment
is momentarily suspended until the diverse occurrences of yh!yw+ ~ as a verb can be
examined. Examining all the occurrences is the best way to work toward an answer to
this question.
A verbal use of yh!yw+ ~ is defined as an occurrence of yh!yw+ ~ (or yh!Tw= )~ as the nuclear
verb of the clause where the person, number, and gender match that of the subject of the
clause. The use of yh!yw+ ~ as a full verb has been recognized by many scholars and
147
148
grammarians. Difference of opinion does exist, however, concerning the status and
function of these occurrences as a verb.
It is important to keep in mind that the following sections present a wide variety
of uses and/or meanings associated with yh!yw+ ,~ but this rather taxonomic presentation
should not be construed as implying that there are different types of yh!yw+ .~ Every
occurrence is the same verb hy`h,* but English uses a variety of verbs to express the
contextual nuances of meaning. The kind of etic detail in the following sections may
seem tedious, but it is necessary for the full description of yh!yw+ .~ Some of the categories
may appear subjective. In fact, some of the categories may not seem to reveal significant
distinctions that benefit the current analysis of yh!y+w.~ However, if future research is
facilitated by these categories, then an important objective has been met.
The early stages of analysis for this study began by making the most basic
division possible between yh!yw+ ~ s occurrences as the main verbal element in independent
clauses and the occurrences in dependent temporal constructions. Each subsequent
review of the occurrencesboth as found in context in the Hebrew Bible and as data
being analyzed hereresulted in increasingly finer distinctions based on both syntactic
and contextual considerations.
The examples throughout the next chapters all have the following format:
translation
Hebrew text
Reference
In almost all examples, at least one full verse is given in the second column from
the right. The left-most column displays both a translation and the morphological
149
information for the text example from the Westminster Grammatical Tags database.1 The
translation is based on the NASB, modified according to the analysis or point being made.
This information is included with each example to make them more accessible to readers
at different levels of proficiency in biblical Hebrew.
Gen 39:2
The purpose of this section is to analyze the various ways in which yh!yw+ ~ carries
out this function as a normal verb. As previously stated, the divisions found in this part
of Chapter eight are not intended to imply that these are different types of yh!yw+ ~ ; the
intention rather is to display the semantic range of yh!yw+ ~ s uses. This section is divided
into the following five categories:
8.2.1 The EQUATIVE Use of yh!yw+ ~
8.2.2 The EXISTENTIAL Use of yh!yw+ ~
8.2.3 The DEICTIC Use of yh!yw+ ~
8.2.4 The DESCRIPTIVE Use of yh!yw+ ~
150
Isa 23:3
In this example, the SUBJECT is known from the context and the noun phrase
which follows yh!Tw= ~ is the NOMINAL COMPLEMENT.
The occurrences of yh!yw+ ~ are divided into different sections based on the
appropriate contextual meanings for the verb. In most of these occurrences, yh!yw+ ~ has the
meaning to be or some other contextually appropriate verb, such as come or stay. In fact,
almost all of these occurrences could possibly be woodenly or literally rendered with
was, but it is important to see the broad semantic range that hy`h* encompasses. The
sub-sections here are:
8.2.1.1 To be
8.2.1.2 And it was so
8.2.1.3 And thats the way it happened
8.2.1.4 Territorial Expressions
8.2.1.5 The Formulaic Expression hw`hy+-rb^D+ yh!yw+ ~
8.2.1.6 Expressions of Age
8.2.1.7 Quantity
8.2.1.8 With Prepositions
151
8.2.1.1
To be
The first set of examples makes it clear that the verbal function is not limited to
yh!yw+ ,~ the masculine form, but this function is also performed by yh!Tw= ,~ the corresponding
third person feminine singular form, as shown in the first example.2
The beginning of his kingdom
was Babel and Erech and Accad
and Calneh, in the land of
Shinar.
w{Tk=lm
^ m
= ^ tyv!ar} yh!Tw= ~
.ru*nv
+ ! Jr\aB
# = hn}lk
= w^ + dK^aw^ + Er\aw# + lb#B*
Gen
10:10
.<yd]ja
* & <yr]bd
* W+ tj*a# hp*c* Jr\ah
* -* lk* yh!yw+ ~
Gen 11:1
wyr`ja
& m
^ @ w{Tv=a! fB@Tw^ ~
.jl^m# byx!n+ yh!Tw= ~
Gen
19:26
25:27
hm*hB
@ b
= W^ <d`aB
* * j~rP
} ) tu)Bu
% b
= a
^ & /yj!v= yh!yw+ ~
Exod
9:10
Gen
Deut
.la@rc
` y= ] yf@bv
= ! dj^y~ <u* yv@ar` [S@at
^ h
= B
! =
33:5
<yr]ys!ah
& * tyb@B= /j@of yh!yw+ ~
Judg
16:21
Even though this is the case, the term WAYYIQTOL and the form yh!yw+ ~ will be used
consistently throughout to represent both the masculine and the feminine forms.
152
the name of his firstborn was
Joel, and the name of his
second, Abijah; they were
judging in Beersheba.
1 Sam
8:2
Pc-vqw3msXa ncms-ncmscX3ms
Pa-ncms np Pc-ncmsc ncmscX3ms
np vqPmp Pp np
hd\CB
* ^ dw]D` rt@SY* w] ~
1 Sam
vd\jh
) ^ yh!yw+ ~
20:24
.loka$l# <j#Lh
# -^ lu^ El#Mh
# ^ bv#Yw} ~
yn]la
@ v
* T
= ! hM*lw* + la@Wmv= rm#aY{w~
1 Sam
28:16
/orb=jB
# = <yn]B* dw]dl
` = Wdl=Y}w~
2 Sam
.tl!au@rz+ Y+ h
] ^ <u^ny{ j!al
& ^ /onm=a^ w{rokb= yh!yw+ ~
3:2
la@rc
` y= -] lK*-lu^ El#m# hm)Ov= El#Mh
# ^ yh!yw+ ~
1 Kgs 4:1
1 Kgs
Wbv=Yw} ~ qc#Md
# ~ Wkl=Yw} ~ <t*a) dw]D` gr{hB
& ^
11:24
.qc#Md
* B
~ = Wkl=mY= w] ~ Hb*
Isa 23:3
Isa 29:13
153
<yr]ouc= ha*lm
@ = hd\Ch
* ^ tq^lj
= # yh!Tw= ~
11:13
8.2.1.2
1 Chr
chapter of Genesis; the other occurrences are in Judges and 2 Kings. This use with /k@ not
only summarizes what immediately precedes, but also, as the next WAYYIQTOL in the text,
it incorporates this summarization as an event in the narrative. Narratively, this
establishes, especially in Gen 1, a powerful pattern of confirming that what was spoken
was accomplished.
God made the expanse, and
separated the waters which were
below the expanse from the
waters which were above the
expanse; and it was so.
lD}bY= w~ ~ u~yq!rh
` -* ta# <yh!Oa$ cu^Yw~ ~
Gen 1:7
<y]Mh
^ ^ /yb@W u~yq!rl
` * tj^T^m! rv#a& <y]Mh
^ ^ /yB@
./k@-yh!yw+ ~ u~yq!rl
` * lu^m@ rv#a&
<y]mV
^ h
* ^ tj^Tm
^ ! <y]Mh
^ ^ WwQ*y] <yh!Oa$ rm#aY{w~
Gen 1:9
./k@-yh!yw+ ~ hv*BY* h
~ ^ ha#rt
` w@ + dj*a# <oqm*-la#
av#D\ Jr\ah
* * av@dT
+ ^ <yh!Oa$ rm#aY{w~
w{nym!l= yr]P= hc#u) yr]P= Ju@ ur~z\ u~yr]zm
+ ^ bc#u@
./k@-yh!yw+ ~ Jr\ah
* -* lu^ w{b-w{ur+z~ rv#a&
Gen 1:11
154
let them be for lights in the
expanse of the heavens to give
light on the earth; and it was
so.
<y]mV
^ h
* ^ u~yq!rB
+ ! tr{oam=l! Wyh*w+
Gen 1:15
./k@-yh!yw+ ~ Jr\ah
* -* lu^ rya!hl
* =
Gen 1:24
<y]mV
^ h
* ^ [ou-lk*lW= Jr\ah
* * tY~j-^ lk*lW=
Gen 1:30
/r\GB
{ ^ rm#Xh
# ^ tZ~G-] ta# gyX!m^ yk!na
{ * hN}h!
HD`bl
^ = hZ`Gh
] -^ lu^ hy\hy= ] lf^ <a!
yT!ud
= y~ w` + br\j) Jr\ah
* -* lK*-lu^w+
.T*rB
+ D
^ ] rv#aK
& ^ la@rc
` y= -] ta# yd]yB
` = u~yv!ot-yK!
hZ`Gh
] -^ ta# rz~Yw` ~ tr`jM
( m
* ! <K@vY= w~ ~ /k@-yh!yw+ ~
.<y]m* lp#Sh
@ ^ aolm= hZ`Gh
] -^ /m! lf^ Jm#Yw] ~
Judg
6:37-38
155
This is the word of the LORD
which He spoke to Jehu, saying,
Your sons to the fourth
generation shall sit on the
throne of Israel. And so it was.
2 Kgs
la@rc
` y= ] aS@K-! lu^ ;l= Wbv=y} <yu!yb!r+ yn}B=
15:12
./k@-yh!yw+ ~
Notice the position of /k@-yh!yw+ ~ in Judg 6:37-38. In all other instances, it occurs
immediately before a sillq, whereas in 6:38 it occurs as the first in a series of four
WAYYIQTOLs.
Even though the meaning is the same, this does not have the same
ru^VB
^ ^ <u*h* w{ta) Wsm=rY+ w] ~ /K@ w{l-yh!yw+ ~
.tm)Yw` ~
2 Kgs
7:20
8.2.1.3
156
Also My bread which I gave you,
fine flour, oil and honey with
which I fed you, you would offer
before them for a soothing
aroma; so it happened,
declares the Lord GOD.
Ezek
j~yr}l= <h#yn}pl
= ! WhyT!tn^ W+ EyT!lk
= a
^ h
$ # vb^dW+ /m#vw# `
16:19
.hw]hy+ yn`da
{ & <a%n+ yh!Yw\ ~ j~jy) n]
Psa 33:9
A third example occurs in Lam 3:37 with yh!Tw# .~ There is no 'atnax in BHS, but the
form is similarly affected by the pausal environment.
yh!Tw# ~ rm^a* hz\ ym!
Lam 3:37
.hW`x! aO yn`da
{ &
8.2.1.4
Territorial Expressions
yh!yw+ ~ is used twenty-one times in expressions referring to territories and borders, as
yn]un& K
~ h
= ^ lWbG+ yh!yw+ ~
10:19
Gen
.<d\Qh
# ^ rh^ hr`ps
* = hk*aB
& ) av*Mm
@ ! <b*vo* m yh!yw+ ~
Gen
10:30
157
The border of the sons of
Reuben was the Jordan.
/D}rY+ h
~ ^ /b@War+ yn}B= lWbG+ yh!yw+ ~
13:23
<y]nj
~ M
& m
^ ! <l*Wbg+ yh!yw+ ~
Josh
13:30
Josh
Josh 15:1
<t*jP
) v
= m
= l
! = <y]rp
~ a
= -# yn}B= lWbG+ yh!yw+ ~
Josh
16:5a
./oyl=u# /r{ojtyB@-du^
Josh
16:5b
hV#nm
~ = hF@ml
^ = lr`oGh^ yh!yw+ ~
Josh 17:1
tt*mk
= M
= h
! ^ rv@am
* @ hV#nm
~ -= lWbg+ yh!yw+ ~
Josh 17:7
hM*Yh
` ^ wyt*ax
) t
= ) yh!yw+ ~
Josh 17:9
Josh
17:10
Josh 19:1
Josh
19:10
158
Their territory was to Jezreel
and included Chesulloth and
Shunem
Josh
19:18
[v*ka
= w^ + /f#bw# ` yl!jw& ~ tq^l=j# <l*WbG+ yh!yw+ ~
19:25
<l*Wbg+ yh!yw+
/D}rY+ h
~ ^ wyt*ax
) t
= ) yh!yw+ ~
<t*lj
* n& ~ lWbG+ yh!yw+ ~
<y]rp
` a
= # hF@Mm
^ ! <l*ro` g yr}u* yh!yw+ ~
Josh
21:20
Josh
19:41
Josh
19:33b
Josh
19:33a
Josh
hr}cu
= # <yT@v= <yr]u* <l*ro` G yh!yw+ ~
Josh
21:40
1 Chr
6:51
(6:66)
8.2.1.5
The phrase hw`hy+-rb^D+ yh!yw+ ~ is one of the typical, formulaic expressions for the
message coming to a prophet. This is a subset of the EQUATIVE category of yh!yw+ ~ with a
noun as grammatical SUBJECT, in every case being the word of the Lord which came.
8.2.1.5.1
The term FORMULAIC refers to the repetitive nature of this statement. Before
assuming, however, that it has narrative-structuring significance, it is important to
159
consider all the occurrences and their distribution. For example, in all of Samuel, there is
only one occurrence. The next seven occurrences are in 1 Kings, but there does not
appear to be any pattern that seems significant in the global organization of the book.
The final occurrence in this set is in 2 Chronicles.
the word of the LORD came to
Samuel, saying,
15:10
1 Kgs
6:11
1 Sam
/j*lV
= h
% -^ la# <yb!vy= { <h@ yh!yw+ ~
1 Kgs
13:20
1 Kgs
.rm)al@ av*uB
= -^ lu^ yn]nj
` &-/b#
16:1
17:2
.rm)al@ yB!vT
= h
! ^ WhY`la
! -@ la# hw`hy+-rb^D+ yh!yw+ ~
1 Kgs
21:17
1 Kgs
17:8
1 Kgs
.rm)al@ yB!vT
= h
! ^ WhY`la
! -@ la# hw`hy+-rb^D+ yh!yw+ ~
1 Kgs
21:28
160
the word of the LORD came to
Shemaiah the man of God,
saying,
hw`hy+-rb^D+ yh!yw+ ~
.rm)al@ <yh!Oa$h-* vya! Why`um
= v
^ -= la#
2 Chr
11:2
Gen 15:1
<r`ba
= ^ ar`yT!-la^ rm)al@ hz\jM
& B
^ ^ <r`ba
= -^ la#
.da)m= hB@rh
+ ^ ;r+kc
* = El* /g}m* yk!na
{ *
Pd Pa-ncmp Pa-acp vqp3ms ncmscnp Pp-np Pp+Pa-ncms Pp-vqc Pdvqi2ms{1}Jm np pi1cs ncms
PpX2fs ncmscX2ms vha Pd
rm)al@ Why`uv
= y^ -+ la# hw`hy+-rb^D+ yh!yw+ ~
Isa 38:4
161
Isaiah shows that the prophetic genre does not automatically require its frequent
occurrence.
Further analysis of Isaiah would need to take into account, for example, the use of
the verb rb*d` in Isa 8:5; 16:13, 14; 20:2; 22:25; 24:3; 25:8; 37:22; 40:5 and 58:14. Isa 8:5
reads as follows:
.rm)al@ dou yl^a@ rB@D~ hw`hy+ [s#Yw{ ~
Isa 8:5
There are forty-nine occurrences of rm^a* hK) which would also need to be
considered. The occurrences are as follows:
rm^a* hK)
in Isaiah:
Isa 7:7; 8:11; 10:24; 18:4; 21:6, 16; 22:15; 28:16; 29:22; 30:12, 15; 31:4;
36:4, 14, 16; 37:3, 6, 10, 21, 33; 38:1, 5; 42:5; 43:1, 14, 16; 44:2, 6, 24;
45:1, 11, 14, 18, 17; 49:7, 8; 49:22, 25; 50:1; 51:22; 52:3, 4; 56:1, 4;
57:15; 65:8, 13; 66:1, 12
Figure 18: rm^a* hK) in Isaiah
To study the occurrences of rb*d` and rm^a* hK) would obviously require more
extensive analysis of the whole book of Isaiah, which is exactly the point. The
occurrence of the formulaic yh!yw+ ~ is only a small part of the whole network of reference to
the word of the Lord in Isaiah. Any determination of whatever significance this one
occurrence might have cannot be made by merely comparing Isaiahs one occurrence to
twenty-nine occurrences in Jeremiah (the next section). The whole web of the word of the
Lord expressions in Isaiah must be scrutinized; then the significance of hw`hy+-rb^d+ yh!yw+ ~
in Isa 38:4 can be considered. Even within Isaiah, the function of this formulaic
expression needs to be carefully examined. Regarding the significance of hw`hy+-rb^d+ yh!yw+ ~
in Isa 38:4, Wildberger comments that
162
[s]uch an arrangement for introducing a message from Yahweh is found
nowhere else in Isaiah 1-39. But it makes perfect sense in this context. We
learn that Isaiah does not go to Hezekiah of his own accord, but acts as a
result of a specific command from Yahweh. (Wildberger 1991, 447)
This comment acknowledges the unique occurrence of this expression in Isaiah,
but does not seem to take into account the rb*d` and rm^a* hK) expressions which also are
evidence of the Lord communicating with Isaiah. As stated above, a more extensive
analysiswhich is not possible herewould consider the whole web of expressions in
order to determine whether there is indeed some unique function that can be attributed to
hw`hy+-rb^d+ yh!yw+ ~ in Isa 38:4.
8.2.1.5.3
Jer 1:4
Jer 1:11
Jer 1:13
Jer 2:1
Jer 13:3
Jer 13:8;
16:1;
18:5;
24:4
163
The word of the LORD came to
Jeremiah after Hananiah the
prophet had broken the yoke
from off the neck of the prophet
Jeremiah, saying,
hy`mr
= y+ -] la# hw`hy+-rb^d+ yh!yw+ ~
Jer 28:12
rm)al@ Why`mr
= y+ -] la# hw`hy+-rb^D+ yh!yw+ ~
Jer 29:30
rm)al@ Why`mr
= y+ -] la# hw`hy+-rb^D+ yh!yw+ ~
Jer 32:26
tyn]v@ Why`mr
= y+ -] la# hw`hy+-rb^d+ yh!yw+ ~
Jer 33:1
.rm)al@ hr`FM
* h
^ ^ rx^jB
& ^ rWxu* WNd\ou aWhw+
rm)al@ Why`mr
= y+ -] la# hw`hy+-rb^D+ yh!yw+ ~
Jer 33:19
rm)al@ Why`mr
= y+ -] la# hw`hy+-rb^D+ yh!yw+ ~
Jer 33:23
Jer 34:12
rm)al@ Why`mr
= y+ -] la# hw`hy+-rb^D+ yh!yw+ ~
Jer 35:12
Why`mr
= y+ -] la# hw`hy+-rb^d+ yh!yw+ ~
Jer 36:27
<yr]bD
* h
+ -^ ta#w+ hL*gM
] h
= -^ ta# El#Mh
# ^ [r{c= yr}ja
& ^
.rm)al@ Why`mr
= y+ ] yP!m! EWrB* bt^K* rv#a&
rm)al@ ayb!Nh
` ^ Why`mr
= y+ -] la# hw`hy+-rb^D+ yh!yw+ ~
Jer 37:6
<ym!y` tr\cu
# & JQ@m! yh!yw+ ~
Jer 42:7
.Why`mr
= y+ -] la# hw`hy+-rb^d+ yh!yw+ ~
rm)al@ sj@nP
+ j
^ t
= B
^ = Why`mr
= y+ -] la# hw`hy+-rb^d+ yh!yw+ ~
Jer 43:8
Regarding Jer 28:12, Keown, Scalise, and Smothers comment that Jeremiah
spoke again only after having received a fresh revelation of the Lords
word. In v 12, then the word-event formula and the temporal clause do not
appear simply pro forma; they serve an important narrative function
showing that neither Hananiahs oracles nor his sign-act (28:1-11) had
canceled or revoked the message given through Jeremiah in chap. 27.
(Keown, Scalise and Smothers 1995, 27)
164
This comment uses the term important narrative function, but it is clear, in the
view of these authors, that the context within which this function is performed is limited
to chapters 27 and 28. There is no mention at all of the switch from yl^a@ to Why`mr
= y+ -] la#.
Based on this pattern with yh!yw+ ,~ 28:12 appears to be a significant transition point in the
book. Carroll comments that by himself Jeremiah cannot refute Hananiah, though he
may make some canny observations; he must receive the divine word in order to do that
(Carroll 1986, 546), but there is no mention of the switch from first to third person
reference. This would need to be explored in a more complete analysis of the formulaic
expression with yh!yw+ ~ in Jeremiah.
Also, within the overall reference to the word of the Lord in Jeremiah, the
following occurrences of hy`h* need to be taken into account:
The words of Jeremiah the son
of Hilkiah, of the priests who
were in Anathoth in the land of
Benjamin, to whom the word of
the LORD came in the days of
Josiah the son of Amon, king of
Judah, in the thirteenth year of
his reign.
<yn]hK
& h
) -^ /m! WhY`ql
! j
= -! /B# Why`mr
= y+ ] yr}bD
= ]
Jer 1:1-2
./m!yn` B
+ ! Jr\aB
# = totn`uB
& ^ rv#a&
wyl*a@ hw`hy+-rb^d+ hy`h* rv#a&
! y{ ym@yB!
hd`Why+ El#m# /oma*-/b# WhY`va
.w{kl=ml
* = hn`v* hr}cu
= -# vOv=B!
Why`mr
= y+ -] la# hw`hy+-rb^d+ hy`h* rv#a&
.torX*Bh
^ ^ yr}bD
= -] lu^
Jer 14:1
165
For each time I speak, I cry
aloud;I proclaim violence and
destruction,because for me the
word of the LORD has resulted
in reproach and derision all day
long.
ar`qa
= # dv)w` sm*j* qu*za
+ # rB@da
~ & yD}m-! yK!
Jer 20:8
sl#ql
# W= hP*rj
+ l
# = yl! hw`hy+-rb^d+ hy`h-* yK!
.<oYh^-lK*
/oma*-/b# WhY`va
! y{l= hn`v* hr}cu
= # vOv=-/m!
Jer 25:3
Jer 32:6
Jer 39:15
.rm)al@ hr`FM
* h
^ ^ rx^jB
& ^ rWxu* w{ty{hB
= !
Why`mr
= y+ -] la# hw`hy+-rb^d+ hy`h* rv#a&
Jer 46:1
.<y]oGh^-lu^ ayb!Nh
` ^
ayb!Nh
` ^ Why`mr
= y+ -] la# hw`hy+-rb^d+ hy`h* rv#a&
.hZ`u-^ ta# hu)rp
+ ^ hK#y~ <r\fB
# = <yT!vl
= P
! -= la#
Jer 47:1
166
That which came as the word of
the LORD to Jeremiah the
prophet concerning Elam, at the
beginning of the reign of
Zedekiah king of Judah, saying:
ayb!Nh
` ^ Why`mr
= y+ -] la# hw`hy+-rb^d+ hy`h* rv#a&
Jer 49:34
hY`qd
! x
+ ! tWkl=m^ tyv!ar}B= <l*yu@-la#
.rm)al@ hd`Why+-El#m#
8.2.1.5.4
<ym!y` tu^bv
= ! hx@qm
= ! yh!yw+
Ezek
3:16
Ezek 6:1;
7:1,
11:14;
12:1
167
In the morning the word of the
LORD came to me, saying,
.rm)al@ rq#BB
) ^ yl^a@ hw`hy+-rb^d+ yh!yw+ ~
12:8
Ezek
Ezek
12:17,
21, 26;
13:1;
14:2, 12;
15:1;
16:1;
17:1, 11;
18:1;
20:2;
21:1, 6,
13, 23;
22:1, 17,
23; 23:1
tyu!yv!Th
= ^ hn`VB
* ^ yl^a@ hw`hy+-rb^d+ yh!yw+ ~
Ezek
.rm)al@ vd\jl
) ^ rocu*B# yr]yc!uh
& * vd\jB
) ^
24:1
Ezek
24:15;
25:1;
27:1;
28:1, 11,
20; 30:1;
33:1, 23;
34:1;
35:1;
36:16;
37:15;
38:1
168
analysis of how they may contribute to the overall reference to the word of the Lord in
Ezekiel.
(On the fifth of the month in the
fifth year of King Jehoiachin's
exile, the word of the LORD
came expressly to Ezekiel the
priest, son of Buzi, in the land of
the Chaldeans by the river
Chebar; and there the hand of
the LORD came upon him.)
tyv!ym!jh
& ^ hn`Vh
* ^ ayh! vd\jl
) ^ hV*mj
! B
& ^
./yk!yo` y El#Mh
# ^ tWlg`l=
Ezek 1:23
yz]WB-/B# laq@zj
+ y# -+ la# hw`hy+-rb^d+ hy`h* hy{h*
rb*K-= rh^n-+ lu^ <yD]cK
= ^ Jr\aB
# = /h@Kh
) ^
.hw`hy+-dy~ <v* wyl*u* yh!Tw= ~
vd\jl
) ^ dj*aB
# = hn`v* hr}cu
= #-yT@vu
= B
^ = yh!yw+ ~
Ezek
26:1
vd\jl
) ^ rc*u* <yn}vB
= ! yr]c!uB
& * tyr]yc!uh
& * hn`VB
* ^
Ezek
29:1
Ezek
29:17
Ezek
30:20
169
In the eleventh year, in the third
month, on the first of the month,
the word of the LORD came to
me saying,
yv!yl!VB
= ^ hn`v* hr}cu
= # tj^aB
^ = yh!yw+ ~
Ezek
31:1
Ezek
32:1
vd\jl
) ^ rc*u* hV*mj
! B
& ^ hn`v* hr}cu
= # yT@vB
= ! yh!yw+ ~
Ezek
32:17
8.2.1.5.5
In Jonah, there are two occurrences of hw`hy+-rb^D+ yh!yw+ ~ at 1:1 and 3:1. In the first
! -& /b# the son of Amittai, which is not
occurrence in 1:1, Jonah is introduced as yT^ma
repeated in 3:1 when the LORD speaks to Jonah the second time, made explicit by tyn]v.@
The word of the LORD came to
Jonah the son of Amittai saying,
the word of the LORD came to
Jonah the second time, saying,
Rm)al@ yT^ma
! -& /b# hn`oy-la# hw`hy+-rb^D+ yh!yw+ ~
Jonah 1:1
Jonah 3:1
It is easy for the reader of Jonah to be tempted to say that the occurrence of this
formulaic expression in 1:1 and 3:1 has definite structural significance in this book.
Stuart comments that
170
[i]f one has listened or read carefully, there can be no doubt that the story
is, as it were, starting over. Once again Jonah has heard the word of
Yahweh. Jonah is back where it all started. (Stuart 1987, 482)
It is important to realize, however, that the division of Jonah into two parts from
1:1-2:11 and 3:1-4:11 involves more than just the nearly verbatim repetition of the
formulaic expression hn`oy-la# hw`hy+-rb^D+ yh!yw+ .~ It is not insignificant that the expression
occurs twice, but a similar division could be made by other expressions as well. Also
involved in the fresh start in chapter 3 is the repetition of the command to Jonah:
h*yl#u* ar`qW= hl*odG+h^ ryu!h* hw}ny+ n]-la# El@ <Wq and the contrast in Jonahs response to the
command. The expression hw`hy+-rb^D+ yh!yw+ ~ is only one component in the new start in
chapter 3.
8.2.1.5.6
There are two occurrences in Haggai. The hw`hy+-rb^D+ yh!yw+ ~ is standard, but what
follows it, especially the yG^j-^ dy~B= in 1:3, is unique to this book. 2:20 states explicitly
tyn]v@ a second time.
the word of the LORD came by
Haggai the prophet, saying,
.rm)al@ ayb!Nh
` ^ yG^j-^ dy~B= hw`hy+-rb^D+ yh!yw+ ~
Hag 1:3
Hag 2:20
171
Not surprisingly, in Haggai there are also occurrences of hw`hy+-rb^d+ hy`h.*
In the second year of Darius the
king, on the first day of the sixth
month, the word of the LORD
came by the prophet Haggai to
Zerubbabel the son of Shealtiel,
governor of Judah, and to
Joshua the son of Jehozadak, the
high priest, saying,
yV!Vh
! ^ vd\jB
) ^ El#Mh
# ^ vw\y`rd
+ l
` = <y]Tv
^ = tn~vB
= !
Hag 1:1
Pp-ncfsc afd Pp-np Pa-ncms Pp+Pancms Pa-ams Pp-ncms ams Pp+Pancms vqp3ms ncmsc-np Pp-ncfscnp Pa-ncms Pp-np ncmsc-np ncmsc
np Pc-Pp-np ncmsc-np Pa-ncms Paams Pp-vqc
vd\jl
) ^ dj*aw# + <yr]cu
= B
# = yu!yb!VB
= ^
Hag 2:1
.rm)al@ ayb!Nh
` ^ yG^j-^ dy~B= hw`hy+-rb^D+ hy`h*
yu!yv!Tl
= ^ hu*Br
* a
+ w^ + <yr]cu
= # B=
Hag 2:10
In Haggai, there are also five occurrences of rm^a* hK), found at Hag 1:2, 5, 7; 2:6,
11, that would need to be considered in full analysis of these expressions.
8.2.1.5.7
In Zechariah, four of the six occurrences explicitly state that the word of the Lord
came to me, one occurrence states to Zechariah in the third person, and the remaining
occurrence in 8:1 is the most general, stating only that the word of the Lord came.
the word of the LORD came to
me saying,
Zech 4:8
172
the word of the LORD came to
me saying,
Zech 6:9
Zech 7:4
.rm)al@ hy`rk
+ z^ -+ la# hw`hy+-rb^D+ yh!yw+ ~
Zech 7:8
Zech 8:1
Zech
8:18
vw\yr
` d
+ l
` = <y]Tv
^ = tn~vB
= ! yn]ym!Vh
= ^ vd\jB
) ^
Zech 1:1
hy`kr
= B
\ -# /B# hy`rk
+ z^ -+ la# hw`hy+-rb^d+ hy`h*
.rm)al@ ayb!Nh
` ^ oDu!-/B#
Zech 1:7
vw\yr
` d
+ l
` = <y]Tv
^ = tn~vB
= ! fb*v= vd\j-) aWh
Why`kr
= B
\ -# /B# hy`rk
+ z^ -+ la# hw`hy+-rb^d+ hy`h*
.rm)al@ ayb!Nh
` ^ aoDu!-/B#
El#Mh
# ^ vw\yr
` d
+ l
` = uB^ra
+ ^ tn~vB
= ! yh!yw+ ~
vd\jl
) ^ hu*Br
* a
+ B
^ = hy`rk
+ z^ +-la# hw`hy+-rb^d+ hy`h*
.wl@sk
= B
! = yu!vT
! h
= ^
Zech 7:1
173
8.2.1.5.8
Three occurrences depart enough from the pattern that they are considered
separately here. In 1 Kgs 12:22 and 1 Chr 17:3 <yh!Oa$ occurs rather than hw`hy+. Notice
that the occurrence in 1 Chr 22:8 is the only one that inverts the order of hw`hy+-rb^D+ and
the recipient, inserting yl^u* after yh!yw+ .~ Also included in this sectione is the only
occurrence of hw`hy+-rb^d+ hy`h* in Daniel, even though the FORMULAIC expression with yh!yw+ ~
does not occur in this book.
But the word of God came to
Shemaiah the man of God,
saying,
hy`um
= v
^ -= la# <yh!Oa$h* rb^D+ yh!yw+ ~
1 Kgs
12:22
aWhh^ hl*yL
+ B
^ ^ yh!yw+ ~
1 Chr
ym!vl
= ! ty]b^ hn\bt
= -! aO
.yn`pl
* = hx*ra
+ ^ T*kp
= v
^ * <yB!r~ <ym!D` yK!
17:3
1 Chr
22:8
174
The FORMULAIC use of yh!yw+ ~ does not occur in Daniel, but hw`hy+-rb^d+ hy`h* does.
in the first year of his reign, I,
Daniel, observed in the books
the number of the years which
was revealed as the word of the
LORD to Jeremiah the prophet
for the completion of the
desolations of Jerusalem,
laY}nD
] ` yn]a& w{kl=ml
* = tj^a^ tn~vB
= !
Dan 9:2
<yn]Vh
* ^ rP^sm
= ! <yr]pS
* =B^ yt!ny{ B!
ayb!Nh
` ^ hy`mr
! y+ -] la# hw`hy+-rb^d+ hy`h* rv#a&
.hnv
` * <yu!bv
= ! <]lv
^ W* ry+ tobr+jl
* = twaOm^l=
8.2.1.5.9
175
the location of occurrence of each expression is plotted, then the pattern of distribution
can be analyzed.
8.2.1.6
Expressions of Age
A further use of yh!yw+ ~ is in statements of age or the passing of time. The two basic
patterns of this reference to age are: yh!yw+ ~ + S + -/B# and yh!yw+ ~ + N-ym@y+ + NUM. This use of
yh!yw+ ~ is remarkable for its very low frequency in the Hebrew Bible.
8.2.1.6.1
The shape of these occurrences is yh!yw+ ~ followed by the persons name as SUBJECT,
which is then followed by the expression of age formed by /B# + the number of years. In
these instances, the persons age is stated as circumstantial information forming the
context for the next event.
Noah was five hundred years
old,
Gen 5:32
Gen 17:1
8.2.1.6.2
hn`v* <yu!Br
* a
+ -^ /B# qj*xy= ] yh!yw+ ~
Gen
25:20
hn`v* <yu!Br
* a
+ -^ /B# wc*u@ yh!yw+ ~
Gen
26:34
Rather than refer to the age of a person when some other event happened, these
occurrences are backward-looking, summarizing statements of the length of a persons
life.
176
<yn]v* ub^v# wyY`j^ yn}v= bq)uy& -~ ym@y+ yh!yw+ ~
Gen
47:28
Gen 5:23
Gen 5:31
The occurrence of the singular form yh!yw+ ~ with the plural ym@y+ days may raise
questions regarding agreement in number, but the collective character of the expression
ym@y,+ especially as ym@y-+ lK* all the days, poses no problem for the singular verb. There are
some cases, however, of ym@y-+ lK* Wyh=Yw] ~ as in Gen 5:4. See 8.5 for further discussion of
occurrences with Wyh=Yw] .~
8.2.1.6.3
Other references to a persons age occur with the sequence w+ + N + -/B# + NUM.
There are eleven occurrences with this sequence. These examples highlight the contrast
in narrative strategy between w+ and yh!yw+ .~
Noah was six hundred years old
when the flood of water came
upon the earth.
Pc-np ncmsc-afs afp ncfs Pc-Pancms vqp3ms ncmp Pp-Pa-ncbs
Gen 7:6
177
Abram went forth as the LORD
had spoken to him; and Lot went
with him. Now Abram was
seventy-five years old when he
departed from Haran.
Gen 12:4
Gen
16:16
Gen
17:24
Gen 21:5
wc*u@ bq@uB
& ^ tz\ja
# ) w{dy`w+ wyj!a* ax*y` /k@-yr}ja
& w^ +
bq)uy& ~ w{mv= ar`qY= w] ~
Gen
25:26
.<t*a) td\lB
# = hn`v* <yV!v!-/B# qj*xy= w] +
w{dm=uB
* = hn`v* <yv!Ov=-/B# [s@oyw+
yn}pL
= m
! ! [s@oy ax@Yw} ~ <y]rx
` =m-! El#m# hu)rP
+ ^ yn}pl
= !
Gen
41:46
.<y]rx
` m
= ! Jr\a-# lk*B= rb)u=Yw~ ~ hu)rp
+ ^
hn`v* <yn]mv
) -= /B# hv#mW)
hn`v* <yn]mv
) W= vOv*-/B# /r{ha
& w^ +
.hu)rP
+ -^ la# <r`Bd
= B
~ =
Exod 7:7
178
Aaron was one hundred twentythree years old when he died on
Mount Hor.
Num
33:39
hn`v* <yr]cu
= w# + ha*m-@ /B# hv#mW)
Deut
34:7
hn`v* hn\mv
) W= <yu!vT
= -! /B# yl!uw@ +
.toar+l! loky` aOw+ hm*q* wyn`yu@w+
1 Sam
4:15
However, in 2 Sam 2:10 which states that Ish-Bosheths age is forty, there is
neither w+ nor yh!yw+ .~
Ish-bosheth, Saul's son, was
forty years old when he became
king over Israel, and he was
king for two years. The house of
Judah, however, followed David.
2 Sam
2:10
.dw]d` yr}ja
& ^ Wyh* hd`Why+ tyB@ Ea^
There will be further discussion of examples like these and other issues related to
narrative strategy in Chapter 10.
8.2.1.7
Quantity
The basic structure of these examples is: yh!yw+ ~ + NP + NUM. The NOUN PHRASE (NP)
in these occurrences is sometimes realized by nothing more than a noun, such as w{ryv! in
1 Kgs 5:12. lK* all is occasionally an additional element, as in Exod 1:5; Num 3:43; Judg
20:46; Josh 8:25; and 1 Chr 21:5. The NP may also be a participle used nominally as in
Judg 20:46 and Josh 8:25. The occurrence in Num 31:32 is an example of an expanded
179
NP:
ab*Xh
* ^ <u^ Wzz+B* rv#a& zB*h^ rt#y\ j~oql=Mh
^ ^ the booty that remained from the spoil
which the men of war had plundered, in which the complex nominal structure zB*h^ rt#y\
* ^ <u^ Wzz+B* rv#a.&
j~oql=Mh
^ ^ is further modified by the relative clause ab*Xh
The important point here is that yh!yw+ ~ is the verb of the clause. For example, in
Exod 1:5 yh!yw+ ~ is followed by a NOUN PHRASE as SUBJECT and then the statement of
quantity is realized as an ADJECTIVE PHRASE COMPLEMENT.
vp#n` <yu!bv
= !
ADJECTIVE PHRASE
COMPLEMENT
yh!yw+ ~
as SUBJECT
VERB
NOUN PHRASE
8.2.1.7.1
Quantity Statements
vp#n` <yu!bv
= ! bq)uy& -~ Er\y\ ya@xy= { vp#n-\ lK* yh!yw+ ~
Exod 1:5
rK*K! <yr]cu
= w# + uv^T@ hp*WnT=h^ bh^z+ yh!yw+ ~
Exod
38:24
rP^sm
= B
! = rk*z` rokB=-lk* yh!yw+ ~
Num
<y]nv
~ = <h#yd}qp
% l
= ! hl*um
= w^ ` vd\j-) /B#m! tomv@
3:43
.<y]ta
* m*W <yu!bv
= w! + hv*Ov= [l#a# <yr]cu
= w# +
Num
31:32
.<yp!la
* -& tv#mj
@ w& ~ [l#a# <yu!bv
= w! +
180
The half, the portion of those
who went out to war, was as
follows: the number of sheep
was 337,500
Pc-vqw3fsXa Pa-ncfs ncms PavqPmp Pp+Pa-ncbs ncmsc Pa-ncbs
afsc-afp ams Pc-amp ams Pc-afsc
amp Pc-afsc afp
ab*XB
* ^ <ya!xY= h
{ ^ ql#j@ hx*j$Mh
# ^ yh!Tw= ~
Num
31:36
Num
<yp!la
* & tu^bv
= ! [l#a# <yv!Ov=W [l#a#
31:43
.toam@ vm@jw& ~
hw`hyl^ Wmyr]h@ rv#a& hm*WrT=h^ bh^z-+ lK* yh!yw+ ~
Num
lq#v* <yV!mj
! w& ~ toam@-ub^v= [l#a# rc*u* hV*v!
31:52
hV*mj
! w& ~ <yr]cu
= # /m!yn` B
+ m
! ! <yl!pN= h
{ -^ lk* yh!yw+ ~
vya! [l#a#
Judg
20:46
Pc-vqw3msXa ncmsc-Pa-vqPmp
Pp-np amp Pc-ams ams ncms
Josh 8:25
2 Sam
24:9
.[l#aw* ` hV*mj
! & w{ryv! yh!yw+ ~
5:12
1 Kgs
5:27
1 Kgs
/ax)-yp@la
= ^ tu^bv
= ! Whn}qm
= ! yh!yw+ ~
Job 1:3
181
Joab gave the number of the
census of all the people to
David. And all Israel were
1,100,000 men who drew the
sword; and Judah was 470,000
men who drew the sword.
Pc-vqw3ms np Po-ncmsc ncmsc-Pancms Pp-np Pc-vqw3msXa ncmscnp ams amp Pc-afs ams ncms
vqPms ncfs Pc-np afs afp Pc-amp
ams ncms vqPms ncfs
8.2.1.7.2
<u*h-* dq^pm
= ! rP^sm
= -! ta# ba*oy /T@Yw] ~
<yp!la
* & [l#a# la@rc
` y= -] lk* yh!yw+ ~ dyw]D-` la#
1 Chr
21:5
<h#yP!-la# <d`yB
` = <yq!ql
= ^m&h^ rP^sm
= ! yh!yw+ ~
Judg 7:6
<ym!Yh
` ^ rP^sm
= ! yh!yw+ ~
1 Sam
27:7
2 Sam
2:11
.<yv!dj
` ( hV*vw! +
hl*um
= w* ` hn`v* <yv!Ov= /B#m! <Y]wl
] h
= ^ Wrp=SY* w] ~
<yr]bg* l
+ ! <t*OG+lg= l
| = <r`P*sm
= ! yh!yw+ ~
1 Chr
23:3
182
Moreover, David and the
commanders of the army set
apart for the service some of the
sons of Asaph and of Heman and
of Jeduthun, who were to
prophesy with lyres, harps and
cymbals; and the number of
those who performed their
service was:
hd`bu
) l
& ^ ab*Xh
* ^ yr}cw* + dyw]D` lD}bY= w~ ~
<ya!yB!Nh
+ ^ /WtWdyw] /m*yh@w+ [s*a* yn}bl
= !
1 Chr
25:1
<y]Tl
* x
= m
! b
= W! <yl!bn* B
+ ! torN{kB
! =
.<t*db
` u
) l
& ^ hk*al*m= yv@na
+ ^ <r`Ps
* m
= ! yh!yw+ ~
Pc-vhw3msXa np Pc-ncmpc Pancbs Pp+Pa-ncfs Pp-ncmpc np Pcnp Pc-np Pa-ncmp Pa-vnPmp Ppncmp Pp-ncmp Pc-Pp-ncbd Pcvqw3msXa ncmscX3mp ncmpc ncfs
Pp-ncfscX3mp
ryv!-yd}Ml
= m
% = <h#yj@a-& <u! <r`Ps
* m
= ! yh!yw+ ~
.hn`omv=W <yn]omv= <y]ta
^ m* /yb!Mh
@ -^ lK* hw`hyl^
1 Chr
25:7
2 Chr
<yc!bK
* = ha*m@ <yl!ya@ <yu!bv
= ! rq*B* lh*Qh
* ^
29:32
8.2.1.7.3
183
The weight of the gold earrings
that he requested was 1,700
shekels of gold, besides the
crescent ornaments and the
pendants and the purple robes
which were on the kings of
Midian, and besides the neck
bands that were on their camels'
necks.
Judg 8:26
<yn]rh
{ C
& h
^ -^ /m! db^l= bh*z` toam@-ub^vW= [l#a#
lu^v# /m*Gr
` a
+ h
^ * yd}gb
+ W! topf!Nh
+ w^ +
yr}aW+xB
^ = rv#a& toqn`uh
& -* /m! db^lW= /y`dm
+ ! yk@lm
= ^
.<h#yL@mg^ +
bh*Zh
` ^ lq^vm
= ! yh!yw+ ~
10:14
8.2.1.8
1 Kgs
2 Chr
9:13
.bh*z`
With Prepositions
An EQUATIVE verb like hy`h* is commonly used in the expression of spatial
8.2.1.8.1
With ta@@
All five occurrences in the following set are of yh!y+w~ accompanied by the
preposition ta@. The infrequent occurrence of this use of yh!yw+ ~ is quite remarkable.
God was with the lad,
Pc-vqw3msXa ncmp Pp-Pa-ncms
ru^Nh
~ -^ ta# <yh!Oa$ yh!yw+ ~
Gen
21:20
184
The LORD was with Joseph,
Pc-vqw3msXa np Pp-np
Gen 39:2
Gen
39:21
Josh 6:27
Judg 1:19
aB@jt
^ m
= ! <yh!Oa$h* tyb@B= <T*a! yh!yw+ ~
2 Chr
.Jr\ah
* -* lu^ tk#lm
# ) hy`lt
= ^uw& ~ <yn]v* vv@
22:12
In this example, yh!yw+ ~ is followed by ta + the 3mp pronominal suffix: he was with
them in the house of God hidden. The phrase aB@j^tm
= ! <yh!Oa$h* tyb@B= <T*a! is a complex
nominal phrase functioning as the ADJECTIVE COMPLEMENT.
8.2.1.8.2
With la#
There are only two instances with la#, one with yh!Tw= ~ and one with yh!yw+ .~
God did so that night; dryness
came to only the fleece, and all
the ground was covered with
dew.
Pc-vqw3msXa ncmp Pd Pp+Pancms Pa-pi3ms Pc-vqw3msXa-ncms
Pp-Pa-ncfs Pp-ncmscX3fs Pc-Ppncmsc-Pa-ncbs vqp3ms ncms
aWhh^ hl*yL
+ B
^ ^ /K@ <yh!Oa$ cu^Yw~ ~
Jr\ah
* -* lK*-lu^w+ HD`bl
^ = hZ`Gh
] -^ la# br\j-) yh!yw+ ~
.lf* hy`h*
Judg 6:40
185
lWav*-la# hu*r` hw`hy+ j~Wr yh!Tw= ~
1 Sam
19:9
These occurrences are noteworthy since it is the preposition lu^ which is used in
the majority of references to the Spirit of the Lord coming upon someone. So, why is the
preposition la# used? Is there some particular nuance that la# communicates better than
some other option?
First of all, the use of la# in Judg 6:40 needs to be compared with 6:37:
If there is dew on the fleece only,
and it is dry on all the ground,
then I will know that You will
deliver Israel through me, as
You have spoken.
HD`bl
^ = hZ`Gh
] -^ lu^ hy\hy= ] lf^ <a!
Judg 6:37
yT!ud
= y~ w` + br\j) Jr\ah
* -* lK*-lu^w+
.T*rB
+ D
^ ] rv#aK
& ^ la@rc
` y= -] ta# yd]yB
` = u~yv!ot-yK!
In both cases in Judg 6:37, lu^ is used, which seems to make it the expected
preposition in 6:40. The use of la# instead of lu^, however, adds an element of surprise by
stating it in a different way. Caution needs to be exercised to avoid reading too much
into the use of a different preposition, but the assumption is that some factor motivated
the authors choice.
There are five instances of lu^ yh!Tw= ~ used to refer to the Spirit of God coming or
being upon someone (listed below under the preposition lu^), so the question naturally is
what the use of la# indicates in 1 Sam 19:9. There may be no significant difference, but
1 Sam 19:9 could be translated an evil spirit from the Lord came to Saul to reflect the fact
that the text has la# rather than lu^.
8.2.1.8.3
With B=
The pattern of these occurrences is yh!yw+ ~ with optional explicit SUBJECT, followed
by B= + Noun, most frequently expressing location, such as in, on, among, and the
references to hw`hy+-dy~ typically understood as against.
yr]xM
= h
! ^ wyn`da
{ & tyb@B= yh!yw+ ~
Gen 39:2
Pc-vqw3msXa Pp-ncmsc
ncmpcX3ms Pa-np
/olM*B^ Er\Db
\ ^ yh!yw+ ~
Exod
4:24
<y]rx
` m
= ! Jr\a-# lk*B= <D`h^ yh!yw+ ~
7:21
hm*hB
@ b
= W^ <d`aB
* * <N`Kh
! ^ yh!Tw= ~
Exod
8:13 (16)
hm*hB
@ b
= W^ <d`aB
* * <N`Kh
! ^ yh!Tw= ~
Exod
Exod
8:14 (17)
.hl*yl
+ * <yu!Br
* a
+ w^ + <oy <yu!Br
* a
+ ^ rh*B* hv#m) yh!yw+ ~
Exod
24:18
hw`hy+ td~uB
& ^ hp*GM
} h
^ ^ yh!Tw= ~
Num
31:16
Jr\ah
* -* lk*B= w{um=v* yh!yw+ ~
Josh 6:27
hw`hy+ td~uB
& ^ [g\Nh
\ ^ yh!yw+ ~
Josh
22:17
la@rc
` y= B
] = qj)-yh!Tw= ~
Judg
11:39
186
187
he returned the silver to his
mother, his mother took two
hundred pieces of silver and
gave them to the silversmith who
made them into a graven image
and a molten image, and they
were in the house of Micah.
Judg 17:4
5:9
1 Sam
<yT!vl
= p
! = hd}c=B! hw`hy+-/ora& yh!yw+ ~
<yv!dj
` ( hu*bv
= !
1 Sam
6:1
<yT!vl
= P
! B
= ^ hw`hy+-dy~ yh!Tw= ~
7:13
<yT!vl
= P
! -= <u! bogB= hm*jl
* =Mh
! ^ dou-yh!Tw= ~
<yT!vl
= P
! -= <u! bogB= hm*jl
* =Mh
! ^ dou-yh!Tw= ~
1 Kgs
5:11
2 Sam
21:19
2 Sam
21:18
1 Sam
<u*br
= y` -` ta# tym!hl
* = hm)Ov= vQ@by^ w+ ~
1 Kgs
qv^yv!-la# <y]rx
~ m
= ! jr~bY= ]w~ <u*br
= y` ` <q*Yw` ~
11:40
188
Obadiah was on the way,
behold, Elijah met him, and he
recognized him and fell on his
face and said, Is this you,
Elijah my master?
w{tar`ql
= ! WhY`la
! @ hN}hw! + Er\DB
\ ^ Why`db
+ u
^ ) yh!yw+ ~
rm#aY{w~ wyn`P-* lu^ lP)Yw] ~ Whr}KY! w~ ~
1 Kgs
18:7
.WhY`la
! @ yn]da
{ & hz\ hT*ah
^ ^
<r`a& hn}jm
& ^ ta@ WZb)Yw` ~ <u*h* ax@Yw} ~
<yr]uc
) = <y]ta
^ s*w+ lq#vB
# = tl#s-) ha*s= yh!yw+ ~
2 Kgs
7:16
.hw`hy+ rb^dK
+ ! lq#vB
# =
/m!yn` B
+ ! ru^vB
^ = aWh-yh!yw+ ~
Jer 37:13
hy`ml
= v
# -# /B# hyY`ar
! y+ ] w{mv=W td|qP
! = lu^B^ <v*w+
ayb!Nh
` ^ Why`mr
= y+ -] ta# cP)tY= ]w~ hy`nn+ j
~ -& /B#
.lp@n{ hT*a^ <yD]cK
= h
^ -^ la# rm)al@
Pc-vqw3msXa-pi3ms Pp-ncms np
Pc-Pd ncmsc ncfs Pc-ncmscX3ms
np ncmsc-np ncmsc-np Pc-vqw3ms
Po-np Pa-ncms Pp-vqc Pp-Pa-np
pi2ms vqPms
Psa 76:3
Jonah 2:1
tolyl@ hv*Ov=W
laY}nD
] ` hd`Why+ yn}Bm
= ! <h#b* yh!yw+ ~
.hy`rz+ u
~ w& ~ la@vy* m! hy`nn+ j
~ &
Dan 1:6
189
Three occurrences merit special comment:
Gen 39:2 The preceding context creates the conditions for 39:2 to not
have an explicit SUBJECT. The possibility of Subject elision is an
important contextual factor that needs to be taken into account for the
proper reading of yh!yw+ .~
Exod 4:24 Based on the preceding comment regarding Subject elision,
Exod 4:24 is analyzed in the same way. The occurrence of Er\Db
\ ^ yh!yw+ ~ is a
locational rather than a temporal use of yh!yw+ .~
Psa 76:3 The order in most of the occurrences of yh!yw+ ~ with B= is that the
location marked by B= follows the noun being modified. Psa 76:3 inverts
this order in the first part of the chiastic structure of the complex clause.
This occurrence of yh!yw+ ~ follows normal rules for elision of the verb in the
second half of the complex clause, which confirms its status as a verb.
Additionally, in Exod 7:21; 8:13, 14; Num 31:16; Josh 22:17; and 2 Sam 21:18,
19, the noun which follows yh!yw+ ~ has the definite article. The presence of the definite
article is the main reason that these occurrences are not categorized as EXISTENTIAL. In
the NASB, Exod 8:13 (16) reads: and there were gnats on man and beast and 8:14 (17)
reads: so there were gnats on man and beast, but rather than an EXISTENTIAL sense,
these occurrences should be rendered in a way that takes the definite article into account:
the gnats were on man and beast. Notice in the EXISTENTIAL occurrences in section
8.2.2, that the nouns following yh!yw+ ~ are not prefixed with the definite article.
In one instance, the preposition B= occurs with yh!y+w,~ but with a different function
altogether. In this instance, taz{b* yh!yw+ ~ because of this, the preposition is analyzed as
CAUSAL
190
Joab the son of Zeruiah had
begun to count them, but did not
finish; and because of this,
wrath came upon Israel, and the
number was not included in the
account of the chronicles of
King David.
1 Chr
aOw+ la@rc
` y= -] lu^ [x#q# taz{b* yh!yw+ ~
27:24
<ym!Yh
` -^ yr}bD
= ] rP^s=mB
! = rP*sM
= h
! ^ hl*u*
.dyw]D` El#Ml
# ^
hu)rp
+ ^ yn}B= EotB= hu)rP
+ ^ tyB@ tb^ng| + yh!yw+ ~
1 Kgs
11:20
8.2.1.8.4
With /yB@
/y @
The only occurrences of yh!yw+ ~ with /yB@ are of either strife or peace coming between
two parties. These occurrences could also feasibly be translated existentially: there was
strife, there was peace, depending on contextual factors.
strife came between the
herdsmen of Abram's livestock
and the herdsmen of Lot's
livestock. Now the Canaanite
and the Perizzite were dwelling
then in the land.
Pc-vqw3msXa-ncms Pp vqPmpc
ncmsc-np Pc-Pp vqPmpc ncmsc-np
Pc-Pa-np Pc-Pa-np Pd vqPms
Pp+Pa-ncbs
<r`ba
= -^ hn}qm
= ! yu@r{ /yB@ byr]-yh!yw+ ~
fol-hn}qm
= ! yu@r{ /yb@W
.Jr\aB
* * bv@y{ za* yZ]rP
] h
= w^ + yn]un& K
~ h
= w^ +
Gen 13:7
191
peace came between Israel and
the Amorites.
yr]ma
) h
$ * /yb@W la@rc
` y= ] /yB@ <olv* yh!yw+ ~
14
1 Kgs
5:26
8.2.1.8.5
1 Sam 7:
In the following cases, the SUBJECT of yh!yw+ ~ may be understood from the context as
in Gen 19:14 or explicit, as in Num 11:1. yh!yw+ ~ S is then followed by K= + the standard of
comparison, which may be realized as a noun, adjective, or participle. In two examples,
Isa 29:11 and Ezek 3:3, there is a prepositional phrase as an additional component in the
expression; in both cases the prepositional phrase immediately follows yh!Tw= .~
Lot went out and spoke to his
sons-in-law, who were to marry
his daughters, and said, Up,
get out of this place, for the
LORD will destroy the city. But
he was to his sons-in-law as one
jesting.
wyt*nb
{ = yj@qO
= wyn`tj
* -& la# rB@dy~ w+ ~ fol ax@Yw} ~
<oqM*h-^ /m! WaX= WmWq rm#aY{w~
Gen
19:14
yn}za
+ B
* = ur~ <yn]na
+ t
) m
= K
! = <u*h* yh!yw+ ~
Num
<B*-ru^bT
= w! ~ w{Pa^ rj^Yw] ~ hw`hy+ um^vY= w] ~ hw`hy+
11:1
.hn\jM
& h
^ ^ hx@qB
= ! lk^aT)w~ hw`hy+ va@
192
the young man was to him like
one of his sons.
wyn`Bm
* ! dj^aK
^ = w{l ru^Nh
~ ^ yh!yw+ ~
17:11
Judg
1 Sam
hj*nm
+ ! w{l Wayb!h-@ aOw+ Whz|bY= w] ~
10:27
.vyr]jm
& K
^ = yh!yw+ ~
hn`va
) r]bK
* = yh!Tw= ~
1 Kgs
13:6
toxWj br\qB
# = hj*WSK^ <t*lb
* n= ] yh!Tw= ~
Isa 5:25
Isa 9:18
va@ tl#ka
) m
& K
^ = <u*h* yh!yw+ ~
(19)
<Wtj*h# rp#Sh
@ ^ yr}bd
= K
] = lK)h^ tWzj* <k#l* yh!Tw= ~
Isa 29:11
qotm*l= vb^dK
+ ! yp!B= yh!Tw= ~
Ezek 3:3
Hos 7:11
193
as one the trumpeters and the
singers were to make themselves
heard with one voice to praise
and to glorify the LORD, and
when they lifted up their voice
accompanied by trumpets and
cymbals and instruments of
music, and when they praised
the LORD saying, He indeed is
good for His lovingkindness is
everlasting, then the house, the
house of the LORD, was filled
with a cloud,
u~ym!vh
= l
^ = <yr]rv
& ) ml
= w^ + <yr]xx
= j
) M
& l
^ ^ dj*ak
# = yh!yw+ ~
todh)lW= lL@hl
^ = dj*a-# loq
2 Chr
5:13
<y]Tl
^ x
= m
! b
= W! torx=xj
) B
& ^ loq <yr]hk
* W= hw`hyl^
bof yK! hw`hyl^ lL@hb
^ W= ryV!h^ yl@kb
= W!
.hw`hy+ tyB@ /n`u* al@m* ty]Bh
^ w^ + w{Ds=j^ <l*oul= yK!
The following two occurrences are listed separately because of the alternate order
of the components. In both of these examples, K= + NOUN precedes rather than follows the
S of yh!yw+ .~ This minor variation in order does not make a significant difference at this
level, but it reflects sensitivity to contextual factors.
If only you had paid attention to
My commandments! Then your
well-being would have been like
a river, And your righteousness
like the waves of the sea.
yt*wx
{ m
= l
! = T*bv
= q
^ h
= ! aWl
Isa 48:18
;m#olv= rh*Nk
` ^ yh!yw+ ~
.<Y`h^ yL@gK
~ = ;t=qd
* x
+ w! +
wyt*um
) K
= ! ;yu#m@ ya@xa
* x
$ w# + ;u#rz+ ~ lojk^ yh!yw+ ~
.yn`pL
* m
= ! w{mv= dm@Vy* -] aOw+ tr}Ky* -] aO
Isa 48:19
194
Another variation on the comparative use is K= indicating an approximate quantity,
as seen in the following examples. This use of K= is referred to as APPROXIMATION by
Williams 257 (1976, 47).
That first slaughter which
Jonathan and his armor bearer
made was about twenty men
within about half a furrow in an
acre of land.
hn`va
) r]h* hK*Mh
^ ^ yh!Tw= ~
1 Sam
14:14
br\uh
* -* du^ hd\CB
* ^ fQ@lT
^ w= ~
Ruth
.<yr]uc
) = hp*ya@K= yh!yw+ ~ hf*Ql
@ -! rv#a& ta@ fB)jT
= w^ ~
2:17
8.2.1.8.6
w{l-yh!yw+ ~
Gen 12:16; 26:14; 30:43; Judg 10:4; 12:9, 14; 17:1, 12;
1 Sam 16:21; 1 Kgs 10:26; 11:3; 2 Kgs 17:3; Job 42:12,
13; 1 Chr 2:22; 2 Chr 1:14; 11:12; 15:5.
I have
Pc-vqw3msXa-PpX1cs
<h#l* yh!Tw= ~
Gen 11:3
yl!-yh!yw+ ~
Gen 32:6
195
the land was Pharaohs
Pc-vqw3fsXa Pa-ncbs Pp-np
I have
Pc-vqw3msXa-PpX1cs
hu)rp
+ l
^ = Jr\ah
* * yh!Tw= ~
Gen
47:20
yl!-yh!yw+ ~
Exod
15:2
Num
.<yu!bv
= w! + vm@j* toam@ vv@
31:37
Josh
13:16, 25
hV#nm
~ = fb#v@ yx!jl
& ^ hv#m) /T@Yw] ~
.<t*ojP=vm
= l
! = hV#nm
~ -= yn}b= hF@m^ yx!jl
& ^ yh!yw+ ~
Josh
13:29
<h#l* yh!yw+ ~
Josh 15:2
Josh 17:2
Manasseh had
Pc-vqw3msXa Pp-np
hV#nm
~ l
= ! yh!yw+ ~
17:11
<h#l* yh!yw+ ~
Josh
18:12
<h#l* yh!yw+ ~
/r{ha
& ^ yn}bl
= ! yh!yw+ ~
Josh 19:2
Josh
21:4, 10
Pc-vqw3msXa Pp-ncmpc np
Josh
Judg
14:20
<yd]ly* + hN`np
] l
= ! yh!yw+ ~
1 Sam
1:2
dw]dl
` = <yd]bu
* & <oda$-lk* yh!yw+ ~
2 Sam
8:14
196
And was like a daughter to him.
Pc-vqw3fsXa-PpX3ms Pp-ncfs
tb^K= w{l-yh!Tw= ~
12:3
la@rc
` y= -] ta# <yT!vl
= P
! l
= ^ hm*jl
* m
= ! dou-yh!Tw= ~
Solomon had
Pc-vqw3msXa Pp-np
2 Sam
21:15
2 Sam
yl! /u*vm
= ! hw`hy+ yh!yw+ ~
2 Sam
22:19
yj!al
* = yh!Tw= ~
1 Kgs
2:15
hm)Ov=l! yh!yw+ ~
1 Kgs
5:6, 29
1 Kgs
la@rc
` y= B
] = Jq*Yw` ~ dd`h& rv#a& hu*rh
` -* ta#w+
11:25
2 Kgs
24:1
.w{B-dr`mY= w] ~
my mother is my grave
Pc-vqw3fsXa-PpX1cs ncfscX1cs
ncmscX1cs
yr]bq
= ! yM!a! yl!-yh!Tw= ~
<yv!Nl
` ^ <v@-yh!Tw= ~
hw`qT
= ! lD~l^ yh!Tw= ~
Jer 20:17
Ezek
23:10
Pc-vqw3fsXa-ncms Pp+Pa-ncfp1
Job 5:16
1 Chr
2:22
There is no need for agreement between yh!Tw= ~ and <v@ since the SUBJECT of yh!Tw= ~ is the
woman referred to in the 3rd person feminine ending.
197
la@mj
= r
= y~ l! tr\ja
# ^ hV*a! yh!Tw= ~
1 Chr
2:26
hm)Ov=l! yh!yw+ ~
Solomon had
Pc-vqw3msXa Pp-np
2 Chr
9:25
as*al
* = yh!yw+ ~
Asa had
Pc-vqw3msXa Pp-np
2 Chr
14:7
w{dy`B= hk*lm
* M
= h
^ -^ ta# hw`hy+ /k#Yw` ~
2 Chr
17:5
Jehoshaphat had
Pc-vqw3msXa Pp-np
18:1
WhY`Zu
] l
% = yh!yw+ ~
Uzziah had
Pc-vqw3msXa Pp-np
2 Chr
26:11
da)m= hB@rh
+ ^ dobk*w+ rv#u) WhY`qz! j
+ y! l! yh!yw+ ~
2 Chr
bh*zl
` W= [s#kl
# = w{l-hc*u* torx*aw) +
32:27
8.2.1.8.7
2 Chr
<ym!cb
* l
= w! + hr`qy* + /b#al
# W=
.hD`mj
= # yl@K= lk)lW= <yN]gm
] l
* W=
The combination l= yh!yw+ ~ is used to express the notion of one thing being turned
into another, or becoming something other than what it was. The combination of yh!yw+ ~
with l= meaning become is not treated in most grammars. The grammars by Lambdin
(1971) and Putnam (2003) are, however, two notable exceptions.
man became a living being.
Pc-vqw3msXa Pa-ncms Pp-ncfs afs
Gen 2:7
198
for them the bricks became
stone, and tar became mortar2
Pc-vqw3fsXa PpX3mp Pa-ncfs Ppncfs Pc-Pa-ncms vqp3ms PpX3mp
Pp-ncms
/b#al
* = hn`bL
@ h
= ^ <h#l* yh!Tw= ~
.rm#jl
) ^ <h#l* hy`h* rm*jh
@ w^ +
Pc-vqw3msXa Pp-ncms
it became a staff
Pc-vqw3msXa Pp-ncms
it became a serpent.
Pc-vqw3msXa Pp-ncms
vj*nl
` = yh!yw+ ~
Exod 4:3
hF#ml
^ = yh!yw+ ~
Exod 4:4
/yN]tl
^ = yh!yw+ ~
Exod
7:10
Pc-vqw3msXa Pp-ncmp
<y]ml
* = yh!yw+ ~
db@u-) sm^l= yh!yw+ ~
Josh 7:5
Josh
16:10
<yh!Oa$ tD~rj
+ l
# = yh!Tw= ~
1 Sam
14:15
Deut
26:5
it became water.
Gen
49:15
Pc-vqw3msXa Pp-ncmsc-vqPms
it became a serpent;
Gen 11:3
The meaning here is became in the sense of replaced, or took the place of.
1 Sam
22:2
199
The victory that day was turned
to mourning for all the people,
for the people heard it said that
day, The king is grieved for his
son.
2 Sam
19:3
(19:2)
tn\ks
# ) El#Ml
# ^ yh!Tw= ~
taF*jl
^ = hZ\h^ rb*Dh
` ^ yh!yw+ ~
1 Kgs
12:30
<u*br
= y` ` tyB@ taF^jl
^ = hZ\h^ rb*DB
` ^ yh!yw+ ~
1 Kgs
13:34
1 Kgs 1:4
hl*lq
* l
= w! + hM*vl
^ W= hB*rj
+ l
* = <k#xr
= a
+ ^ yh!Tw= ~
Jer 44:22
Ep#hl
# = yh!Tw= ~
16:34
tj^rs
~ ) /p#g\l= yh!yw+ ~
Pc-vqw3fsXa ncmp
he became chief.
Pc-vqw3msXa Pp-ncms
Ezek
17:6
it became a vine
Ezek
/p#g\l= yh!Tw= ~
Ezek
17:6
hn`yq!l= yh!Tw= ~
Ezek
19:14
var{l= yh!yw+ ~
1 Chr
11:6
<yd]bu
* & dyw]dl
` = <r`a& yh!yw+ ~
1 Chr
18:6
There is a degree of semantic overlap between be and become, but part of the
difference can be described as one of perspective at the time of speech. Become
encodes the speakers perspective on a certain state being initiated, whereas be encodes
the perspective on an already existing state. When the combination l= yh!yw+ ~ is used, the
200
speakers perspective is unambiguously marked as referring to the initiation of a state. In
Gen 19:26, .jl^m# byx!n+ yh!Tw= ~ wyr`ja
& m
^ @ w{Tv=a! fB@Tw^ ,~ this perspective is significant,
indicating that when Lots wife looked back, she was a pillar of salt. It is obvious that at
one moment Lots wife was a living, breathing human being and that at another moment
she became a salt statue. What is being argued here on the basis of the absence of l= is
that the writers perspective seems to indicate that the transformation into salt was so
instantaneous that when she turned to look back she already was a pillar of salt. In
contrast to this example, in Gen 2:7, hY`j^ vp#n\l= <d`ah
* * yh!yw+ ,~ the becoming is in focus,
indicated by l= in vp#n\l.= English versions have many instances of yh!yw+ ~ without l=
translated as become, but this is questionable in several places. The context of yh!yw+ ~
without l= may indicate that the meaning become is appropriate, but this needs to be
determined on a case by case basis.
8.2.1.8.8
hV*al
! = yl!-yh!Tw= ~
Gen
20:12
hV*al
! = w{l-yh!Tw= ~
Gen
24:67
/b@l= Hl*-yh!yw+ ~
Exod
2:10
201
it became a snare to Gideon and
to his family
Judg 8:27
/h@kl
) = w{l-yh!yw+ ~
Judg 17:5
hV*al
! = w{l-yh!Tw= ~
25:42
hj*nm
+ ! ya@cn= { <yd]bu
* l
& ^ dw]d`l= ba*om yh!Tw= ~
<yd]bu
* l
& ^ dw]dl
` = <r`a& yh!T=w~
Pc-vqw3fsXa-PpX3ms Pp-ncfs
He became my salvation
Pc-vqw3msXa-PpX1cs Pp-ncfs
2 Sam
8:6
2 Sam
8:2
1 Sam
hV*al
! = w{l-yh!Tw= ~
2 Sam
11:27
2 Sam
23:19
hu*Wvyl! yl!-yh!yw+ ~
Isa 12:2
Isa 63:8
yb!bl
* = tj^mc
= l
! W= /occ*l= yl! ;yr\bd
* + yh!yw+ ~
Jer 15:16
yd]ya@-<oyb= yn]WmD+qy^ +
Psa 18:19
Pc-vqw3msXa ncmpcX2ms
ncmscX2ms PpX1cs Pp-ncms
.yl! /u*vm
= l
! = hw`hy+-yh!yw+ ~
vpi3mpX1cs Pp-ncms-ncmscX1cs
Pc-vqw3msXa-np Pp-ncms PpX1cs
It became my reproach
Pc-vqw3fsXa Pp+Pa-ncfp PpX1cs
yl! topr`jl
& ^ yh!Tw= ~
Psa 69:11
bG`cm
= l
! = yl! hw`hy+ yh!yw+ ~
Psa 94:22
He became my salvation.
Pc-vqw3msXa-PpX1cs Pp-ncfs
hu*Wvyl! yl!-yh!yw+ ~
Psa
118:14
hV*al
! = w{l-yh!Tw= ~
Ruth
4:13
tn\ma
# l
) = w{l-yh!Tw= ~
Ruth
4:16
202
rc*l= <h#l* yh!yw+ ~
8.2.1.8.9
1 Chr
11:21
In two examples, the l= preposition is used with the sense of to, for. This is the
DISTRIBUTIVE
la@rc
` y= -] lk*l= la@Wmv=-rb^d+ yh!yw+ ~
4:1
1 Sam
1 Kgs 5:2
ta@ tq#xl
# * [s#Kh
# ^ rK^K! ta^m= yh!yw+ ~
Exod
<yn]da
` & ta^m= tk#rP
{ h
* ^ yn}d+a^ ta@w+ vd\Qh
) ^ yn}da
+ ^
38:27
./d\al
* * rK*K! rK*Kh
! ^ ta^m=l!
hm*ojh^-ta# hn\bN= w] ~
Hy`xj
= -# du^ hm*ojh^-lK* rv@QT
* w! ~
.tocu&l^ <u*l* bl@ yh!yw+ ~
Neh 3:38
(4:6)
203
He was determined to seek God
in the days of Zechariah, who
had understanding through the
vision of God; and as long as he
sought the LORD, God
prospered him.
Why`rk
+ z^ + ym@yB! <yh!Oa$ vr{dl
+ ! yh!yw+ ~
<yh!Oa$h* ta)rB
+ ! /yb!Mh
@ ^
2 Chr
26:5
.<yh!Oa$h* w{jyl!xh
= ! hw`hy+-ta# w{vr+D` ym@yb!W
Pc-vqw3msXa Pp-vqc ncmp Ppncmpc np Pa-vhPms Pp-vqc Pancmp Pc-Pp-ncmpc vqcX3ms Po-np
vhp3msX3ms Pa-ncmp
The occurrence in 2 Chr 26:5 merits further comment. The literal sense is he
was to seek. GKC 114i comments on this occurrence, giving he set himself to seek
as the translation (GKC 1910, 348). The NASB has rendered this he was determined to
seek God, whereas the NLT states Uzziah sought God. There is insufficient data to
make a decisive statement, but it appears that a translation like the NASB better reflects
the fact that this clause does not begin with the simple WAYYIQTOL of vr~d`.
= !
8.2.1.8.11 With yn}pl
= ! and they represent two different
There are two occurrences of yh!yw+ ~ with yn}pl
= .! In 1 Sam 19:7, the sense is essentially literal: in his presence and in
senses of yn}pl
2 Kgs 5:2, yn}pl
= ! is used figuratively to indicate a position of servitude, being literally she
was in the presence of Naamans wife.
Then Jonathan called David,
and Jonathan told him all these
words. And Jonathan brought
David to Saul, and he was in his
presence as formerly.
Pc-vqw3ms np Pp-np Pcvhw3msXa-PpX3ms np Po ncmscPa-ncmp Pa-acp Pc-vhw3msXa np
Po-np Pp-np Pc-vqw3msXa PpncbpcX3ms Pp-Pd Pd
dw]dl
` = /t*no` hy+ ar`qY= w] ~
hL#ah
@ * <yr]bD
* h
+ -^ lK* ta@ /t*no` hy+ w{l-dG\Yw~ ~
lWav*-la# dw]D-` ta# /t*no` hy+ ab@Yw` ~
.<ovl=v! lomt=aK
# = wyn`pl
* = yh!yw+ ~
1 Sam
19:7
204
and she waited on Naaman's
wife.
2 Kgs 5:2
Zech
7:12
u~yq!rl
` * lu^m@ loq-yh!yw+ ~
Ezek
./h#yp@nk
+ ^ hn`yP#rT
~ = <d`mu
= *B= <v*ar{-lu^ rv#a&
1:25
El#Mh
# ^ vr\okl= tj^a^ tn~v=-du^ laY}nD
] ` yh!yw+ ~
Dan 1:21
205
8.2.1.8.15 With lu^
The preposition lu^ with yh!yw+ ~ is most typically translated on, upon. yh!yw+ ~ could
feasibly be translated was in these occurrences, but in most cases some other translation
is more contextually appropriate.
the rain fell upon the earth for
forty days and forty nights
Pc-vqw3msXa Pa-ncms Pp-Pa-ncbs
abp ncms Pc-abp ncms
<oy <yu!Br
* a
+ ^ Jr\ah
* -* lu^ <v#Gh
\ ^ yh!yw+ ~
Gen 7:12
.hl*yl
+ * <yu!Br
* a
+ w^ +
Jr\ah
* -* lu^ <oy <yu!Br
* a
+ ^ lWBM^h^ yh!yw+ ~
Gen 7:17
<r`Tw* ~ hb*Th
@ -^ ta# Wac=Yw] ~ <y]Mh
^ ^ WBr+Yw] ~
.Jr\ah
* * lu^m@
<yr]uh
* -# lu^ <yh!Oa$ tT^j! yh!yw+ ~
Gen 35:5
rq#Bh
) ^ ty{hB
= ! yv!yl!Vh
= ^ <oYb^ yh!yw+ ~
Exod 19:
16
24:2
Num
/h#yb!a& tj^Pv
^ m
= ! hF@m-^ lu^ /t*lj
* n& ~ yh!Tw= ~
Num
36:12
Judg 3:10
206
the Spirit of the LORD came
upon Jephthah
11:29
hd\Ch
* ^ yn}P-= lu^ vb^d+ yh!yw+ ~
<yh!Oa$ j~Wr lWav* yk@al
& m
= -^ lu^ yh!Tw= ~
Pc-vqw3fsXa Pp-ncms np
1 Sam
19:23
2 Sam
12:30
2 Kgs
3:15
1 Sam
19:20
1 Sam
14:25
Judg
wyT*jT
= ^ EOm=y-] rv#a& rokB=h^ w{nB=-ta# jQ^Yw] ~
hm*jh
) -^ lu^ hl*u) Whl@uY& w~ ~
2 Kgs
3:27
la@rc
` y= -] lu^ lodG`-[x#q# yh!yw+ ~
.Jr\al
* * Wbv%Yw` ~ wyl*um
* @ Wus=Yw] ~
Pc-vqw3ms Po-ncmscX3ms Pancms Pr-vqi3ms PpX3ms Pcvhw3msX3ms ncfs Pp-Pa-ncfs Pcvqw3msXa ncms-ams Pp-np Pcvqw3mp Pp-PpX3ms Pc-vqw3mp
Pp+Pa-ncbs
Isa 9:5
Ezek 1:3
3:22
Pc-vqw3fsXa Pp-ncms np
Ezek
32:27
Ezek
1 Chr
20:2
207
the dread of the LORD was on
all the kingdoms of the lands
which were around Judah, so
that they did not make war
against Jehoshaphat.
tokl=mm
= -^ lK* lu^ hw`hy+ dj^P^ yh!yw+ ~
2 Chr
hd`Why+ tobybs
! = rv#a& toxr`ah
& *
17:10
tokl=mm
= -^ lK* lu^ <yh!Oa$ dj^P^ yh!yw+ ~
2 Chr
20:29
.la@rc
` y= ] yb@yo+ a
2 Chr
<yB!xu
^ h
& -* ta#w+ <yr]va
@ h
& -* ta# Wdb=uY^ w~ ~
24:18
<*l
! v*Wryw] hd`Why+-lu^ hw`hy+ [x#q# yh!yw+ ~
rv#aK
& ^ hq*rv
} l
= w! + hM*vl
^ =
hu*wz` l
+ ! <n}TY= w] ~
2 Chr
29:8
WhY`qz! j
+ y! + byv!h@ wyl*u* lm%gk
+ -! aOw+
2 Chr
32:25
.<*l
! v*Wryw] hd`Why+-lu^w+
208
8.2.1.8.16 With <u!
In four cases, yh!yw+ ~ occurs with <u! with. There appears to be no significant
difference in sense between ta# and <u!.
HALOT states
express the divine presence (HALOT 1995, 839), but the occurrences of <u! yh!yw+ ~ are too
infrequent to lend much support to this position. To have the status of formula, a much
higher frequency of occurrence would be expected. Note also regarding frequency, that
1 Kgs 8:17 and 2 Chr 6:7 are parallel passages, so essentially count as one occurrence.
Also, ta# yh!yw+ ~ (8.2.1.8.1) is used to refer to the divine presence.
let us arise and go up to Bethel,
and I will make an altar there to
God, who answered me in the
day of my distress and has been
with me wherever I have gone.
<V*-hc#ua
$ w# + la@-tyB@ hl#un& w~ + hm*Wqn`w+
Gen 35:3
yt!rx
` * <oyB= yt!a) hn\uh
) * la@l* j~Bz@ m
+ !
.yT!kl
= h
* * rv#a& Er\DB
\ ^ yd]Mu
* ! yh!yw+ ~
Pc-vqi1cp{1}Ct Pc-vqi1cp{1}Cm
np Pc-vqi1cs-Pd ncms Pp-ncms PavqPms PoX1cs Pp-ncms ncfscX1cs
Pc-vqw3msXa PpX1cs Pp+Pa-ncbs
Pr vqp1cs
1 Kgs
8:17
Pc-vqw3msXa Pp-ncmsc np
ncmscX1cs Pp-vqc ncms Pp-ncmsc
np ncmpc np
2 Chr 6:7
.la@rc
` y= ] yh@Oa$ hw`hy+ <v@l= ty]B^ tonb=l!
Pc-vqw3msXa Pp-ncmsc np
ncmscX1cs Pp-vqc ncms Pp-ncmsc
np ncmpc np
dyw]D` yk@rd
+ B
~ = El^h* yK! fp*vo* hy+-<u! hw`hy+ yh!yw+ ~
.<yl!uB
* l
= ^ vr~d` aOw+ <yn]va
) r]h* wyb!a*
2 Chr
17:3
209
8.2.1.9
occurrences give a good idea of the varied uses of yh!yw+ ~ and demonstrate the need for
careful examination of the context and syntactic environment. The purpose of the
preceding categorization of these occurrences is to establish the syntactic connection of
yh!yw+ ~ as a verb. This does not eliminate the possibility that many of these occurrences also
have certain narrative functions. Discussion of these possible functions does not come,
however, until Chapter 10.
210
flood came is not given an EXISTENTIAL translation, There was the flood, because of the
definite article on lWBM^h.^ Notice the other instances of the definite article as well (See
also the discussion at the end of 8.2.1.8.3.)
One of the typical uses of EXISTENTIAL verbs is the introduction of an entity in the
narrative and the nouns introduced in this way do not occur with the definite article
(compare loq-yh!yw+ )~ . This restriction would also apply to the definiteness inherent in
construct chains and names of people and places. There are also considerations regarding
the narrative structure that affect the decision to not translate certain occurrences as
existentials.
8.2.2.1
EXISTENTIAL
Occurrences of yh!yw+ ~
roa-yh!yw+ ~
Gen 1:3
Gen 1:5,
8, 13, 19,
Pc-vqw3msXa-ncms Pcvqw3msXa-ncms
23, 31
Jr\aB
* * bu*r` yh!yw+ ~
Gen
12:10
Jr\aB
* * bu*r` yh!yw+ ~
la@tyB@m! Wus=Yw] ~
ht*rp
` a
= # aobl* Jr\ah
* -* tr~bK
= ! dou-yh!yw+ ~
Gen 26:1
Gen
35:16
.HT*dl
+ B
! = vq^Tw= ~ lj@r` dl#T@w~
toxr`ah
& -* lk*B= bu*r` yh!yw+ ~
Gen
41:54
<y]rx
~ m
= ! Jr\a-# lk*B= hl*pa
@ &-Ev#j) yh!yw+ ~
Exod 10:
22
211
there was a great cry in Egypt
Pc-vqw3fsXa ncfs afs Pp-np
<y]rx
` m
= B
! = hl*dg{ + hq*ux
* = yh!Tw= ~
30
Ev#jh
) w^ + /n`uh
* # yh!yw+ ~
Exod 14:
20
Exod 12:
Num 9:6
yn]Dh
` ^ tj^Pv
^ M
= m
! ! hu*rX
+ m
* ! dj*a# vya! yh!yw+ ~
Judg 13:2
<y]rp
` a
= -# rh^m@ vya!-yh!yw+ ~
Judg 17:1
Judg 17:7
<yp!ox <y]tm
^ r
* h
` -* /m! dj*a# vya! yh!yw+ ~
1 Sam 1:
1 Sam
9:1
1 Sam
14:15
21:1
2 Sam
tg~B= hm*jl
* m
= ! dou-yh!Tw= ~
2 Sam
21:20
2 Sam
21:20
212
there was a heavy shower
Pc-vqw3msXa ncms ams
1 Kgs
18:45
/orm=vB
) = lodG` bu*r` yh!yw+ ~
h*yl#u* <yr]x* hN}hw! +
2 Kgs
6:25
<yn]mv
) B
= ! romj&-var{ toyh$ du^
.[s#k-* hV*mj
! B
& ^ <yn]oyyr}j& bQ^h^ ub^rw{ + [s#K#
Pc-vqw3msXa ncms ams Pp-np PcPi vqPmp PpX3fs Pp vqc ncmsncms Pp-amp ncms Pc-ncms Pancms ncmpc-ncfp ncmp Pp+Paams-ncms
So I prophesied as I was
commanded; and as I
prophesied, there was a noise,
and behold, a rattling; and the
bones came together, bone to its
bone.
yt!yW}x% rv#aK
& ^ yt!aB@nw] +
Ezek
37:7
Pc-vnp1cs{2} Pp-Pr vPp1cs Pcvqw3msXa-ncms Pp-vncX1cs PcPi-ncms Pc-vqw3fp ncfp ncfs PpncfscX3ms
Jonah 1:4
Hab 1:3
<yf!pV
= h
) ^ fp)v= ym@yB! yh!yw+ ~
Ruth 1:1
Jr\aB
* * bu*r` yh!yw+ ~
hd`Why+ <j#l# tyB@m! vya! El#Yw} ~
w{Tv=aw! + aWh ba*om yd}cB
= ! rWgl*
.wyn`b* yn}vW=
1 Chr
20:5
Neh 8:17
tg~B= hm*jl
* m
= ! dou-yh!Tw= ~
1 Chr
20:6
213
8.2.2.2
Based on this pattern, certain questions about the EXISTENTIAL use of yh!yw+ ~ are inevitable.
For example, are the people, entities, and situations introduced in this way by yh!yw+ ~ of
significance at the broader textual level? This is an important questionthe very kind of
question this whole study is designed to explorebut discussion is postponed until
Chapter 10.
One of the examples above requires special comment. Notice the use of the
definite article in Exod 14:20:
there was the cloud with the
darkness
Ev#jh
) w^ + /n`uh
* # yh!yw+ ~
Exod
14:20
In 8.2.1.8.3 the discussion of the use of the definite article and EXISTENTIAL
occurrences of yh!yw+ ~ stated that occurrences with the definite article would not be
translated as EXISTENTIALS. This case in Exod 14:20 is different, however.
1) The cloud is already known from the context: compare Exod 13:21 where the
first reference to /n`u* occurs without the definite article, consistent with the normal
pattern of usage.
2) There is no nominal or adjective complement in the clause: as seen in 8.2.1 and
8.2.4, the instances of EQUATIVE and DESCRIPTIVE uses of yh!yw+ ~ have either a
nominal or adjective complement.
These contextual factors support the translation of Exod 14:20 as EXISTENTIAL.
214
EXISTENTIAL there
like There is X or There are X and Y, whereas statements with DEICTIC there are
phrased X is there, referring to an entitys location rather than its existence.
8.2.3.1
DEICTIC
Occurrences of yh!yw+ ~
Gen
39:20
<yr]Wsa& El#Mh
# ^ yr}Wsa&-rv#a&
.rh^Sh
) ^ tyb@B= <v*-yh!yw+ ~
<oy <yu!Br
* a
+ ^ hw`hy+-<u! <v*-yh!yw+ ~
Exod
hl*yl
+ ^ <yu!Br
* a
+ w^ +
34:28
<yv!dj
` ( hu*Br
* a
+ ^ <ym!y` <v*-yh!Tw= ~
Judg 19:2
2 Sam
13:38
hl*odg+ hp*GM
} h
^ ^ <v*-yh!Tw= ~
:7
2 Sam 18
<yv!du
` & ha*lm
@ = hd\Ch
* ^ tq^lj
= # <v*-yh!Tw= ~
2 Sam 23
:11
215
The poles were so long that the
ends of the poles of the ark could
be seen in front of the inner
sanctuary, but they could not be
seen outside; and they are there
to this day.
<yD]Bh
^ ^ yv@ar` War`Yw} ~ <yD]Bh
^ ^ Wkyr]aY& w~ ~
2 Chr 5:9
aOw+ ryb!Dh
+ ^ yn}P-= lu^ /ora*h-* /m!
.hZ\h^ <oYh^ du^ <v*-yh!y+w~ hx*Wjh^ War`y}
8.2.3.2
The DEICTIC occurrences of yh!yw+ ~ are not very frequent, but they are another
example of the need for sensitivity to the context and syntactic environment.
1 Kgs 4:1
DESCRIPTIVE:
Gen 11:30
SUBJECT
VERB
yr~c*
SUBJECT
yh!Tw= ~
VERB
216
8.2.4.1
DESCRIPTIVE
Occurrences of yh!yw+ ~
Gen 4:2
.hm*da
` & db@u) hy`h* /y]qw^ + /ax) hu@r{ lb#h-# yh!yw+ ~
hr`qu
* & yr~c* yh!Tw= ~
Gen
11:30
Gen 38:7
j~yl!xm
= ^ vya! yh!yw+ ~
Gen 39:2
ha#rm
+ ^ hp@yw] ra^t-) hp@y+ [s@oy yh!yw+ ~
Gen 39:6
Gen 50:9
dr`Bh
* ^ EotB= tj^Ql
^ t
^ m
= ! va@w+ dr`b* yh!yw+ ~
Exod
9:24
.yogl= ht*yh
+ * za*m@ <y]rx
~ =m! Jr\a-# lk*B=
vm#Vh
* ^ aB)-du^ hn`Wma$ wyd`y` yh!yw+ ~
17:12
dj*a# /K*vM
= h
! ^ yh!yw+ ~
da)m= hl*odG+ <yr]uN* h
+ ^ taF^j^ yh!Tw= ~
<yT!vl
= P
! -= lu^ hq*zj
` & hm*jl
* =Mh
! ^ yh!Tw= ~
1 Sam
14:52
2 Sam
2:17
1 Sam
2:17
Exod
36:13
Exod
hK*ra
| & hm*jl
* M
= h
! ^ yh!Tw= ~
2 Sam
3:1
217
the conspiracy was strong,
Pc-vqw3msXa Pa-ncms ams
JM!a^ rv#Qh
# ^ yh!yw+ ~
15:12
Jr\ah
* -* lk* yn}P-= lu^ tyx@pn) ` hm*jl
* M
= h
! ^ <v*-yh!Tw= ~
Er\DB
\ ^ tk#lv
# m
= % w{tl*bn= ] yh!Tw= ~
1 Kgs
13:24
2 Sam
18:8
2 Sam
aB@jt
^ m
= ! hw`hy+ tyB@ HT*a! yh!yw+ ~
.Jr\ah
* -* lu^ tk#lm
# ) hy`lt
= ^uw& ~ <yn]v* vv@
2 Kgs
11:3
El#Mh
# -^ ta# hw`hy+ uG~ny~ w+ ~
w{tm) <oy-du^ ur`xm
) = yh!yw+ ~
2 Kgs
15:5
El#Mh
# -^ /B# <t*oyw+ tyv!pj
= h
* ^ tyb@B= bv#Yw} ~
.Jr\ah
* * <u^-ta# fp@v) ty]Bh
^ -^ lu^
Isa 22:7
.hr`uV
= h
* ^ Wtv* tv) <yv!rP
` h
* w^ +
Pc-vqw3msXa ncmsc-ncmpcX2fs
vqp3cp ncms Pc-Pa-ncmp vqa
vqp3cp Pa-ncmsXd
truth is lacking
Pc-vqw3fsXa Pa-ncfs vnPfs
tr\Du
\ n= \ tm#ah
$ * yh!Tw= ~
Isa 59:15
Jer 23:10
218
But there was another great
eagle with great wings and much
plumage; and behold, this vine
bent its roots toward him and
sent out its branches toward him
from the beds where it was
planted, that he might water it.
<y]pn^ K
` = lodG+ lodG` dj*a-# rv#n\ yh!yw+ ~
Ezek
hn`pK
= * taZ{h^ /p#Gh
\ ^ hN}hw! + hx*on-br~w+
17:7
w{L-hj*lv
= ! wyt*oYl!dw` + wyl*u* h*yv#rv
( *
.Hu*Fm
* ^ togr|um
& @ Ht*oa toqv=hl
^ =
rob-yt@Kr
= y+ B
~ = h*yt#rb
{ q
= ! WnT=n] rv#a&
Ezek
<yl!lj
* & <L*K% Ht*rb
` q
% = tobyb!s= Hl*hq
* = yh!yw+ ~
32:23
.<yY]j^ Jr\aB
# = tyT!j! Wnt=n`-rv#a& br\jB
# ^ <yl!pn= {
Job 1:3
Neh 5:1
1 Chr 2:3
tyn]un& K
~ h
= ^ u~Wv-tB^m! w{l-dl^on
hw`hy+ yn}yu@B= ur~ hd`Why+ rokB= ru@ yh!yw+ ~
.Wht@ym!yw+ ~
w{Ma!w+ wyj*am
# @ dB*kn= ] JB@uy= ~ yh!yw+ ~
.bx#uB
) = yT!dl
+ y^ ` yK! rm)al@ JB@uy= ~ w{mv= ha*rq
+ *
1 Chr 4:9
219
Jahath was the first and Zizah
the second; but Jeush and
Beriah did not have many sons,
so they became a father's
household, one class.
yn]Vh
@ ^ hz`yz]w+ var{h* tj^y-~ yh!yw+ ~
1 Chr
23:11
.tj*a# hD`qp
% l
= ! ba* tyb@l=
2 Chr
ur`xm
) = tWvp=jh
* ^ tyB@ bv#Yw} ~
26:21
8.2.4.2
occurrences of yh!yw+ ,~ but the relatively low frequency of occurrenceonly twenty nine
timesis somewhat unexpected. There are, of course, other types of descriptive phrases
without yh!yw+ ,~ such as those with an adjective in what is typically referred to as its
predicative usage, but the preceding examples are the only ones with yh!yw+ .~
220
hy`h* hf*lu
* w& ~ ha*B* vm#Vh
# ^ yh!yw+ ~
Gen
Gen
42:35
hl*ouh* hl#um
& ^ la@Wmv= yh!yw+ ~
aWhh^ <oYh^m@ dw]D-` ta# /w}u) lWav* yh!yw+ ~
lyK!cm
= ^ wk*rD
+ -` lk*l= dw]d` yh!yw+ ~
<ym!Yh
` -^ lK* dw]D-` ta# by}a) lWav* yh!yw+ ~
1 Sam
23:26
lWav* yn}Pm
= ! tk#ll
# * zP*jn= \ dw]d` yh!yw+ ~
<yr]fu
= ) wyv*na
` w& ~ lWav*w+
.<c*pt
= l
* = wyv*na
` -& la#w+ dw]D-` la#
1 Sam
18:29
1 Sam
18:14
1 Sam
18:9
1 Sam
7:10
Exod
19:19
15:173
221
So David reigned over all Israel;
and David was administering
justice and righteousness for all
his people.
la@rc
` y= -] lK*-lu^ dw]D` EOm=Yw] ~
.w{Mu^-lk*l= hq*dx
` W= fP*v=m! hc#u) dw]d` yh!yw+ ~
2 Sam
8:15
hw\jT
& v
^ y= -] rv#a& var{h-* du^ aB* dw]d` yh!yw+ ~
2 Sam
yv^Wj w{tar`ql
= ! hN}hw! + <yh!Oal@ <v*
15:32
.w{var{-lu^ hm*da
` w& ~ w{Tn+TK
* % u~Wrq* yK!ra
+ h
^ *
la@rc
` y= ] yf@bv
= -! lk*B= /odn` <u*h-* lk* yh!yw+ ~
2 Sam
19:10
Pc-vqw3msXa ncmsc-Pa-ncms
vnPms Pp-ncmsc-ncmpc np
<yz]ra
` & yx@u& hm)Ov=l! /t@n{ <oryj! yh!yw+ ~
.w{xp=j-# lK* <yv!orb= yx@u&w~
1 Kgs
5:24 (10)
/j*lV
= h
% -^ la# <yb!vy= { <h@ yh!yw+ ~
1 Kgs
13:20
rb@u) El#Mh
# ^ yh!yw+ ~
1 Kgs
20:39
1 Kgs
/K@ la@rc
` y= -] El#m# wyl*a@ rm#aY{w~
20:40
.T*xr
= j
` * hT*a^ ;f#Pv
* m
= !
2 Kgs
2:11
222
One was felling a beam, and the
axe head fell into the water; and
he cried out and said, Alas, my
master! For it was borrowed.
Pc-vqw3msXa Pa-ams vhPms Pancfs Pc-Po-Pa-ncms vqp3ms Pp-Pancmp Pc-vqw3ms Pc-vqw3ms Pi
ncmscX1cs Pc-pi3ms vqsms
hm*jh
) -^ lu^ rb@u) la@rc
` y= ] El#m# yh!yw+ ~
2 Kgs
6:26
2 Kgs 6:5
hy`jh
$ -# rv#a& ta@ El#Ml
# ^ rP@sm
^ = aWh yh!yw+ ~
2 Kgs 8:5
hy`jh
$ -# rv#a& hV*ah
! * hN}hw! + tM@h-^ ta#
Ht*yB@-lu^ El#Mh
# -^ la# tq#ux
# ) Hn`B-= ta#
taz{ El#Mh
# ^ yn]da
{ & yz]jG& } rm#aY{w~ Hd`c-* lu^w+
.uv*yl!a$ hy`jh
$ -# rv#a& Hn`B-= hz\w+ hV*ah
! *
vya! <yr]bq
= ) <h@ yh!yw+ ~
13:21
2 Kgs
19:37
2 Kgs
17:28
2 Kgs
Isa 37:38
<g`Bt
* P
= -^ ta# ac@n{ rx^lM
= h
# ^ yh!yw+ ~
Dan 1:16
223
The people of the land were
discouraging the people of
Judah,
Ezra 4:4
hS*dh
~ -& ta# /m@a) yh!yw+ ~
Esth 2:7
Pc-vqw3msXa ncms-Pa-ncbs
vpPmp ncfdc ncms-np
1 Chr
15:29
w{Mu^-lk*l= hq*dx
` W= fP*vm
= ! hc#u) yh!yw+ ~
18:14
hl*um
= l
* -= du^ ld}gw` + El@h) fp*vo* hy+ yh!yw+ ~
2 Chr
17:12
1 Chr
rh*Nh
` -^ /m! <yk!lM
* h
= -^ lk*B= lv@om yh!yw+ ~
.<y]rx
` m
= ! lWbG+ du^w+ <yT!vl
= P
! = Jr\a-# du^w+
2 Chr
9:26
224
The two angels came to Sodom
in the evening as Lot was sitting
in the gate of Sodom. When Lot
saw them, he rose to meet them
and bowed down with his face to
the ground.
br\uB
# * hm*ds
{ = <yk!al
* =Mh
^ ^ yn}v= Wab)Yw` ~
Gen 19:1
<d{s-= ru^vB
^ = bv@y{ folw+
<t*ar`ql
= ! <q*Yw` ~ fol-ar+Yw~ ~
.hx*ra
+ * <y]Pa
^ ^ WjT^vY= w] ~
225
the distinction is clear between yh!yw+ ~ which is an actual verb and yh!yw+ ~ that has no verb
function in the individual sentence and as a result does not act like a normal WAYYIQTOL
(Niccacci 1990, 159). The question, however, is whether it is possible to categorize yh!yw+ ~
in this way. Niccacci argues that any yh!yw+ ~ that has a macrosyntactic function has no
function at all in the single sentence (Niccacci 1990, 159). This seems to imply that any
yh!yw+ ~ that does in fact have a function in the single sentence is excluded from having any
macrosyntactic function. Consequently, none of the verbal occurrences in this chapter,
by Niccaccis definition, could possibly perform a macrosyntactic function. The first
matter at hand, however, is to clearly define what macrosyntactic means and then what
constitutes a macrosyntactic marker. This will be dealt with in Chapter 10.
Exod
7:12
226
And the man and his wife were
both naked and were not
ashamed.
w{Tv=aw! + <d`ah
* * <yM!Wru& <h#yn}v= Wyh=Yw] ~
Gen 2:25
.Wvv*Bt
) y= ] aOw+
.hq*br
= l
] W= qj*xy= l
] = j~Wr tr~m) /`yy\hT
= w! ~
Gen
26:35
One of the important reasons for considering the occurrences of Wyh=Yw] ~ is that they
provide supporting evidence for the verbal uses of yh!yw+ ~ since the same basic patterns
occur. The more limited sample of these occurrences does not contain the same diversity
of examples, but the categories of verbal uses are quite parallel. This is not surprising
since the only real difference between yh!yw+ ~ and Wyh=Yw] ~ is that of singular versus plural. For
the yh!yw+ ~ categories which are not represented in the Wyh=Yw] ~ examples, one only needs to
have a plural SUBJECT and Wyh=Yw] ~ fits fine. The parallel nature of the uses of Wyh=Yw] ~
confirms the status of yh!y+w~ as a verb.
It should also be noted that not only is the verbal usage parallel, but there are also
parallel questions which emerge from examining the Wyh=Yw] ~ examples. For example in
1 Chr 12:40, Wyh=Yw] ~ has a summarizing sense of so then, or with the result that (see this
example below under 8.5.1.2, DEICTIC Uses of Wyh=Yw] .~
They were there with David
three days, eating and drinking,
for their kinsmen had prepared
for them.
Pc-vqw3mp-Pd Pp-np ncmp ams
vqPmp Pc-vqPmp Pp-vhp3cp
PpX3mp ncmpcX3mp
1 Chr
12:40
(39)
227
8.5.1.1
Uses of Wyh=Yw] ~
8.5.1.1
EQUATIVE
8.5.1.2
DEICTIC
8.5.1.3
DESCRIPTIVE
EQUATIVE
Uses of Wyh=Yw] ~
Uses of Wyh=Yw] ~
Uses of Wyh=Yw] ~
hb*Th
@ -^ /m! <ya!xY= h
{ ^ j~n-{ yn}b= Wyh=Yw] ~
Gen 9:18
./u^nk
` = yb!a& aWh <j*w+ tp#yw` ` <j*w+ <v@
.hq*br
= l
] W= qj*xy= l
] = j~Wr tr~m) /`yy\hT
= w! ~
26:35
Gen
Gen
36:11
Gen
36:22
Gen
46:12
228
These then are the sons of Levi
by their names: Gershon and
Kohath and Merari.
Num
.yr]rm
` W= th*qW= /ovr+G} <t*mv
) B
= !
3:17
tj^Pv
^ m
= ! hl*vl
@ = <t*jP
) v
= m
= l
! = hd`Why+-yn}b= Wyh=Yw] ~
Num
yx!rP
+ h
^ ^ tj^Pv
^ m
= ! Jr\pl
# = yn]lV
* h
@ ^
26:20
.yj!rZ+ h
~ ^ tj^Pv
^ m
= ! jr~z\l=
Pc-vqw3mp ncmpc-np PpncfpcX3mp Pp-np ncfsc Pa-np Ppnp ncfsc Pa-np Pp-np ncfsc Pa-np
yn]rx
{ j
= h
# ^ tj^Pv
^ m
= ! /r{xj
= l
# = Jr\p-# yn}b= Wyh=Yw] ~
Num
.yl!Wmj*h# tj^Pv
^ m
= ! lWmj*l=
26:21
D+ra
+ ^ ul^b-# yn}b= Wyh=Yw] ~
Num
.ym!uN& h
~ ^ tj^Pv
^ m
= ! /m*un& l
~ = yD]ra
+ h
^ * tj^Pv
^ m
= ! /m*un& w~ +
26:40
hK*lm
= W! hl*gj
+ w* + hx*rt
+ ! hl*jm
= ^ hn`yy\hT
= w! ~
Num
.<yv!nl
` = /h#yd}d{ yn}bl
= ! dj*p=lx
* = tonB= hu*nw{ +
36:11
hd`Why+-yn}b= hF@ml
^ = hx@qm
= ! <yr]uh
* # Wyh=Yw] ~
la@xb
= q
= ^ hB*gN+ B
\ ^ <oda$ lWbG+-la#
Josh
15:21
.rWgy`w+ rd\uw@ +
u~Wv-yK!lm
= W^ yw]vy= w] + /t*no` y lWav* yn}B= Wyh=Yw] ~
1 Sam
hr`yk!Bh
= ^ <v@ wyt*nb
{ = yT@v= <v@w+
14:49
.lk^ym! hN`fQ
^ h
= ^ <v@w+ br~m@
229
Then Hadad died. Now the
chiefs of Edom were: chief
Timna, chief Aliah, chief Jetheth,
Pc-vqw3msXa np Pc-vqw3mp
ncmpc np ncms np ncms np np
ncms np
dd`h& tm*Yw` ~
<oda$ yp@WLa^ Wyh=Yw] ~
1 Chr
1:51
1 Chr
2:25
la@mj
= r
= y~ + rokB= <r`-yn}b= Wyh=Yw] ~
.rq#uw@ ` /ym!yw` + Ju^m^
1 Chr
2:27
1 Chr
2:28
<k#vw# ` /y`ja
= ^ ud`ym!v= yn}B= Wyh=Yw] ~
.<u*yn]aw& ~ yj!ql
= w! +
1 Chr
7:19
var{h* hy`bj
= r
^ + rz\uy# l!a-$ yn}b= Wyh=Yw] ~
1 Chr
yn}bW= <yr]ja
@ & <yn]B* rz\uy# l!al
$ # hy`h-* aOw+
23:17
.hl*um
= l
* = Wbr` hy`bj
= r
^ +
1 Chr
12:22
.ab*XB
* ^ <yr]c* Wyh=Yw] ~
(21)
230
ht*cu
* * yd]y` hL#a-@ lK*-ta#w+
Isa 66:2
It is merely coincidental that the majority of the preceding examples name the
sons born to a particular person. This is not to be construed as a meaning of Wyh=Yw] ,~ but
rather should be seen as one of the typical types of statements which is expressed with
hy`h.*
8.5.1.1.1
Wyh=Yw] ~ functions together with the plural participle in the following examples,
expressing continuous action in the past.
The cherubim had their wings
spread upward, covering the
mercy seat with their wings, with
their faces toward each other;
the faces of the cherubim were
toward the mercy seat.
hl*um
= l
^ = <y]pn^ k
` = yc@rP
+ ) <yb!rK
| h
= ^ Wyh=Yw] ~
vya! <h#yn}pW= tr\PK
) h
^ -^ lu^ <h#yp@nk
+ B
^ = <yk!ks
= )
Exod
37:9
.<yb!rK
| h
= ^ yn}P= Wyh* tr\PK
) ^h-^ la# wyj!a-* la#
<yu!tm
= ^ hZ\h-^ <u*h* yr}Va
= m
^ = Wyh=Yw] ~
.<yu!Lb
* m
% = wyr`Va
* m
% W=
Isa 9:15
(16)
231
At the beginning of their living
there, they did not fear the
LORD; therefore the LORD sent
lions among them which were
killing some of them.
2 Kgs
toyr`ah
& -* ta# <h#B* hw`hy+ jL^vy^ w+ ~
17:25
.<h#B* <yg]rh
+ ) Wyh=Yw] ~
2 Kgs
<yn]rm
{ V
= h
) ^ Wcu* rv#a& tomB*h^ tyb@B= WjyN]Yw~ ~
17:29
2 Kgs
tomb* yn}hK
& ) <t*oxq=m! <h#l* Wcu&Yw~ ~
17:32
2 Kgs
21:15
1 Chr
6:17 (32)
232
For the cherubim spread their
wings over the place of the ark,
so that the cherubim made a
covering over the ark and its
poles.
<y]pn^ K
` = <yc!rP
+ ) <yb!WrK=h^ Wyh=Yw] ~
2 Chr 5:8
2 Chr
24:14
ryu!m@ <yr]bu
= ) <yx!rh
` * Wyh=Yw] ~
2 Chr
30:10
.<B* <yg]ul
! m
= W^ <h#yl@u& <yq!yj!cm
= ^ Wyh=Yw] ~
ryu!m@ <yr]bu
= ) <yx!rh
` * Wyh=Yw] ~
2 Chr
30:10
.<B* <yg]ul
! m
= W^ <h#yl@u& <yq!yj!cm
= ^ Wyh=Yw] ~
233
but they continually mocked the
messengers of God, despised His
words and scoffed at His
prophets, until the wrath of the
LORD arose against His people,
until there was no remedy.
<yh!Oa$h* yk@al
& m
= B
^ = <yb!ul
! =m^ Wyh=Yw] ~
2 Chr
du^ wya*bn! B
+ ! <yu!Tu
= T
= m
^ W! wyr`bD
* + <yz]obW
36:16
.aP@rm
+ ^ /ya@l-= du^ w{Mu^B= hw`hy+-tm^j& tolu&
El#Mh
# ^ tn~v= hd`dn+ ` aWhh^ hl*yL
+ B
^ ^
Esth 6:1
<ym!Yh
` ^ yr}bD
= ] tonr{kZ= h
] ^ rp#s-@ ta# ayb!hl
* = rm#aY{w~
.El#Mh
# ^ yn}pl
= ! <ya!rq
` n= ] Wyh=Yw] ~
2 Chr
<yr]kc
= ) Wyh=Yw] ~ hw`hy+-tyB@ td~obu& tk#al#m=
24:12
The data display here is merely the first step toward analyzing the narrative
function of these examples. Further analysis would need to consider the context of each
occurrence and explore the factors which motivated the use of Wyh=Yw] ~ with the participle.
8.5.1.1.2
Quantity
234
8.5.1.1.2.3 Period of time
8.5.1.1.2.1 Basic Quantity Statements with Wyh=Yw] ~
The basic structure of these examples is: Wyh=Yw] ~ + NP + NUM, parallel to the
structure of the same type of statement with yh!yw+ .~
While Israel was dwelling in
land, that Reuben went and lay
with Bilhah his father's
concubine, and Israel heard of
it. Now there were twelve sons of
Jacob
Pc-vqw3msXa Pp-vqc np Pp+Pancbs Pa-pi3fs Pc-vqw3msXa np Pcvqw3ms Po-np ncfs ncmscX3ms Pcvqw3ms np Pc-vqw3mp ncmpc-np
amd ams
awh!h^ Jr\aB
* * la@rc
` y= ] /K)vB
= ! yh!yw+ ~
wyb!a* vg\ly# P! hh*lB
= -! ta# bK^vY= w] ~ /b@War+ El#Yw} ~
Gen
35:22
la@rc
` y= ] um^vY= w] ~
<t*dl
{ o= T la@rc
` y= ] rk)B= /b@War+-yn}b= Wyh=Yw] ~
rP^sm
= B
! = <t*ba
) & tyb@l= <t*jP
) v
= m
= l
! =
hn`v* <yr]cu
= # /B#m! rk*z-` lK* <t*OG+lg= l
| = tomv@
.ab*x* ax@y{ lK) hl*um
= w^ `
/b@War+ hF@ml
^ = <h#yd}qP
% =
.toam@ vm@jw& ~ [l#a# <yu!Br
* a
+ w^ + hV*v!
Num
1:20-21
235
So all the numbered men of the
sons of Israel by their fathers'
households, from twenty years
old and upward, whoever was
able to go out to war in Israel,
even all the numbered men were
603,550.
Pc-vqw3mp ncmsc-vqsmpc ncmpcnp Pp-ncmsc ncmpcX3mp Ppncmsc amp ncfs Pc-PdXd ncmscvqPms ncbs Pp-np Pc-vqw3mp
ncmsc-Pa-vqsmp afs-afp ams Pcamsc amp Pc-afsc afp Pc-abp
<t*ba
) & tyb@l= la@rc
` y= -] yn}b= yd}WqP=-lK* Wyh=Yw] ~
hl*um
= w^ ` hn`v* <yr]cu
= # /B#m!
Num
1:45-46
.la@rc
` y= B
] = ab*x* ax@y-{ lK*
toam@-vv@ <yd]qP
% h
= -^ lK* Wyh=Yw] ~
.<yV!mj
! w& ~ toam@ vm@jw& ~ <yp!la
* & tv#Ov=W [l#a#
Num
.<yV!mj
! w& ~ toam@ vm@jw& ~ <yp!la
* & tv#Ov=W
1:46
<t*jP
) v
= m
= l
! = <h#yd}qp
% = Wyh=Yw] ~
Num
.<yV!mj
! w& ~ toam@ ub^v= <y]Pl
^ a
= ^
4:36
<t*ba
) & tyb@l= <t*jP
) v
= = ml
! = <h#yd}qP
% = Wyh=Yw] ~
Num
4:40
<t*jP
) v
= m
= l
! = <h#yd}qp
% = Wyh=Yw] ~
Num
.<y]ta
* m*W <yp!la
* & tv#Ov=
4:44
<yp!la
* & tn~mv
) = <h#yd}qP
% = Wyh=Yw] ~
Num
.<yn]mv
) W= toam@ vm@jw& ~
4:48
Num
<yt!Mh
@ ^ db^Lm
= ! toam@ ub^vW=
17:14
(16:49)
hp*GM
} B
^ ^ <yt!Mh
@ ^ Wyh=Yw] ~
Num
.[l#a* <yr]cu
= w# + hu*Br
* a
+ ^
25:9
236
These are the families of the
Reubenites, and those who were
numbered of them were 43,730.
<h#yd}qp
% = Wyh=Yw] ~ yn]bW@ ar|h* tj)Pv
= m
= ! hL#a@
Num
26:7
[l#a# <yr]cu
= w# + hv*Ov= <h#yd}qp
% = Wyh=Yw] ~
Num
hl*um
= w* ` vd\j-) /B#m! rk*z-` lK*
26:62
la@rc
` y= ] yn}B= EotB= Wdq=Pt
* =h* aO yK!
.la@rc
` y= ] yn}B= EotB= hl*j&n~ <h#l* /T^n-] aO yK!
Pc-vqw3mp vqsmpcX3mp ams Pcamp ams ncmsc-ncms Pp-ncmscncms Pc-PdXd Pp Pn vup3cp Ppncmsc ncmpc np Pp Pn-vnp3ms
PpX3mp ncfs Pp-ncmsc ncmpc np
<yr]uy* + ty~rq
+ B
! = /ora*h* tb#v# <oYm! yh!yw+ ~
hn`v* <yr]cu
= # Wyh=Yw] ~ <ym!Y`h^ WBr+Yw] ~
1 Sam
7:2
.hw`hy+ yr}ja
& ^ la@rc
` y= ] tyB@-lK* WhN`Yw] ~
qz\bB
* = <d}qp
= Y= w] ~
[l#a# toam@ vOv= la@rc
` y= ]-yn}b= Wyh=Yw] ~
1 Sam
11:8
tonyd]Mh
= ^ yr}c* yr}un& -~ ta# dq)pY= w] ~
1 Kgs
dq^P* <h#yr}ja
& w^ + <yv!Ov=W <y]nv
~ = <y]ta
^ m* Wyh=Yw] ~
20:15
.<yp!la
* & tu^bv
= ! la@rc
` y= ] yn}B-= lK* <u*h-* lK*-ta#
One slightly different example is Judg 16:30, where the quantity is expressed by a
plural adjective <yB!r~ together with /m! to express the comparative:
237
And Samson said, Let me die
with the Philistines! And he
bent with all his might so that
the house fell on the lords and
all the people who were in it. So
the dead whom he killed at his
death were more than those
whom he killed in his life.
<yT!vl
= P
! -= <u! yv!pn= ~ tomT* /ovm=v! rm#aY{w~
<yn]rS
` h
= -^ lu^ ty]Bh
^ ^ lP)Yw] ~ j~kB
) = fY}w~
Judg
16:30
Pc-vqw3ms np vqi3fsXa{1}Jm
ncfscX1cs Pp-np Pc-vqw3msXa Ppncms Pc-vqw3ms Pa-ncms Pp-Pancmp Pc-Pp-ncmsc-Pa-ncms PrPpX3ms Pc-vqw3mp Pa-vqPmp Pr
vhp3ms Pp-ncmscX3ms amp Pp-Pr
vhp3ms Pp-ncmpcX3ms
Gen 5:4
Gen 5:5
hn`v* hr}cu
= # <yT@v= tv@-ym@y-+ lK* Wyh=Yw] ~
Gen 5:8
Gen 5:11
238
All the days of Kenan were nine
hundred and ten years, and he
died.
Gen 5:14
la@ll
= h
^ m
& ^ ym@y-+ lK* Wyh=Yw] ~
Gen 5:17
hn`v* <yu!vt
= w! + vm@j*
.tm)Yw` ~ hn`v* toam@ hn\mv
) W=
hn`v* <yV!vw! + <y]Tv
^ = dr\y-\ ym@y-+ lK* Wyh=Yw] ~
Gen 5:20
Gen 5:27
Gen 9:29
hn`v* <y]ta
^ m*W <yn]v* vm@j* jr~t-# ym@y+ Wyh=Yw] ~
./r`jB
* = jr~T# tm*Yw` ~
Gen
11:32
.hn`v* <yn]mv
) W= hn`v* ta^m= qj*xy= ] ym@y+ Wyh=Yw] ~
Gen
35:28
One exception to the above pattern is in Gen 23:1, which has yY}j^ instead of the
more frequent ym@y.+
Sarah lived one hundred and
twenty-seven years; these were
the years of the life of Sarah.
Pc-vqw3mp ncmpc np afs ncfs Pcamp ncfs Pc-afs ncfp ncfpc ncmpc
np
hn`v* <yr]cu
= w# + hn`v* ha*m@ hr`c* yY}j^ Wyh=Yw] ~
.hr`c* yY}j^ yn}v= <yn]v* ub^vw# +
Gene
23:1
239
8.5.1.1.2.3 Period of time
In two examples, <ym!y` Wyh=Yw] ~ occurs referring to a period of time.
The captain of the bodyguard
put Joseph in charge of them,
and he took care of them; and
they were in confinement for
some time.
Gen 40:4
.rm*vm
= B
! = <ym!y` Wyh=Yw] ~ <t*a) tr\vy* w+ ~
<l*lv
* -= ta# zb)l* w{Mu^w+ fp*vo* hy+ ab)Yw` ~
2 Chr
<yr]gp
` W= vWkr+W br{l* <h#b* Wax=mY= w] ~
20:25
Pc-vqw3msXa np Pc-ncmscX3ms
Pp-vqc Po-ncmscX3mp Pc-vqw3mp
PpX3mp Pp+Pa-ncms Pc-ncms Pcncmp Pc-ncmpc afp Pc-vpw3mp
PpX3mp Pp-Pd ncms Pc-vqw3mp
ncmp ams vqPmp Po-Pa-ncms Pp
ams-pi3ms
8.5.1.1.3
With Prepositions
Because of the essentially parallel nature of the following uses with those already
discussed in the sections which deal with yh!yw+ ~ above, there is minimal comment on the
data with Wyh=Yw] .~
8.5.1.1.3.1 With ta@
When they saw him, they
brought thirty companions and
they were with him.
Pc-vqw3msXa Pp-vqcX3mp
PoX3ms Pc-vqw3mp amp ncmp Pcvqw3mp PpX3ms
Judg
14:11
240
8.5.1.1.3.2 With b=
From Kibroth-hattaavah the
people set out for Hazeroth, and
they remained at Hazeroth.
Num
.torx@jB
& ^ Wyh=Yw] ~
11:35
rB*dM
+ B
! ^ la@rc
` y= -] yn}b= Wyh=Y]w~
Num
.tB*Vh
^ ^ <oyB= <yx!u@ vv@qm
) = vya! Wax=mY= w] ~
15:32
<]lv
^ W* ry+ El#m# qd\x-# yn]da
{ & u~mv
) k
= ! yh!yw+ ~
Josh 10:1
Gen
29:20
241
When he came to Lehi, the
Philistines shouted as they met
him. And the Spirit of the LORD
came upon him mightily so that
the ropes that were on his arms
were as flax that is burned with
fire, and his bonds dropped from
his hands.
w{tar`ql
= ! Wuyr]h@ <yT!vl
= p
! =W yj!l-# du^ ab*-aWh
hw`hy+ j~Wr wyl*u* jl^xT
= w! ~
Judg
15:14
<yT!vP
= K
! ^ wyt*ouorz+-lu^ rv#a& <yt!bu
) h
& * hn`yy\hT
= w! ~
.wyd`y` lu^m@ wyr`Wsa$ WSM^Yw] ~ va@b* Wru&B* rv#a&
la@rc
` y= ] yt!ax*m* rB*d+MB
! ^ <yb!nu
` K
& ^
Hos 9:10
tv#Bl
) ^ Wrz+NY` w] ~ rouP=-lu^b^ WaB* hM*h@
.<b*ha
( K
* = <yx!WQv! Wyh=Yw] ~
<yP!vl
% W= <yP!jl
% = hV*a! jq^l* ryk!mW*
dj*pl
= x
* = yn]Vh
@ ^ <v@w+ hk*u&m^ w{tj)a& <v@w+
1 Chr
7:15
.tonB* dj*pl
= x
* l
= ! hn`yh
\ T
= w! ~
1 Chr 8:3
ba*om-ta# EY~w~
.hj*nm
+ ! ya@cn= { dyw]dl
` = <yd]bu
* & ba*om Wyh=Yw] ~
1 Chr
18:2
242
And it had strong branches fit
for scepters of rulers, And its
height was raised above the
clouds So that it was seen in its
height with the mass of its
branches.
<yl!vm
= ) yf@bv
= -! la# zu) toFm^ Hl*-Wyh=Yw] ~
Ezek
ar`Yw} ~ <yt!bu
) & /yB@-lu^ w{tm*oq HB^gT
+ w! ~
19:11
.wyt*Yl
{ D
! ` br{B= w{hb=gb
` =
Ht*oja& hb*yl!ha
( w* + hl*odG+h^ hl*ha
( * /t*omv=W
Ezek
23:4
.hb*yl!ha
( * <]lv
^ W* ryw] hl*ha
( * /orm=v) /t*omv=W
<yr]uy* + ty~rq
+ ! yb!a& lb*ovl= <yn]b* Wyh=Yw] ~
1 Chr
.tojn|Mh
= ^ yx!j& ha#rh
{ *
2:52
1 Chr
dyw]dl
` = <yd]bu
* & <oda$-lk* Wyh=Yw] ~
18:13
<ym!y` <y]tn^ v
` l
= ! yh!yw+ ~
2 Sam
roxj*lu^bB
^ = <olv*ba
= l
^ = <yz]zg+ { Wyh=Yw] ~
13:23
<y]rp
` a
= -# <u! rv#a&
.El#Mh
# ^ yn}B-= lk*l= <olv*b=a^ ar`qY= w] ~
243
8.5.1.1.3.5 With l= Meaning became
For each one threw down his
staff and they turned into
serpents. But Aaron's staff
swallowed up their staffs.
Exod
.<t*Fm
) -^ ta# /r{ha
& -^ hF@m^ ul^bY= w] ~ <n]yN]tl
^ = Wyh=Yw] ~
7:12
<t*a) ul^bT
= w! ~ h*yP!-ta# Jr\ah
* * jT^pT
= w! ~
Num
26:10
Judg 1:30
w{Br+qB
! = yn]un& K
~ h
= ^ bv#Yw} ~ lOh&n~ yb@vo= y-ta#w+
.sm^l* Wyh=Yw] ~
/olY`aB
^ = sr\j-# rh^B= tb#vl
# * yr]ma
) h
$ * la#oYw~
Judg 1:35
la@rc
` y= -] ta# <B* toSn~l= Wyh=Yw] ~
hW`x-! rv#a& hw`hy+ tw{xm
= -! ta# Wum=vy= h
] & tu^dl
~ *
.hv#m-) dy~B= <t*oba&-ta#
Judg 3:4
244
The sons of Benjamin gathered
together behind Abner and
became one band, and they
stood on the top of a certain hill.
Pc-vtw3mp ncmpc-np Pd np Pcvqw3mp Pp+Pa-ncfs afs Pcvqw3mp Pp ncms-ncfs afs
rn}ba
= ^ yr}ja
& ^ /m!yn` b
+ -! yn}b= WxB=qt
^ Y= w] ~
lu^ Wdm=uY^ w~ ~ tj*a# hD`ga
| l
& ^ Wyh=Yw] ~
2 Sam
2:25
hd`Why+ yr}uB
* = ru^bT
= w! ~ yP!a^w+ yt!mj
* & ET^Tw! ~
Jer 44:6
<*l
! v*Wry+ toxj%bW=
.hZ\h^ <oYK^ hm*mv
* l
= ! hB*rj
+ l
* = hn`yy\hT
= w! ~
la@rc
` y= -] yn}b= Wlu$h-# rv#a& [s@oy tomx=u-^ ta#w+
hd\Ch
* ^ tq^lj
= B
# = <k#vb
= ! Wrb=q* <y]rx
~ M
= m
! !
Josh
24:32
.hl*jn& l
~ = [s@oy-yn}bl
= ! Wyh=Y]w~
lau#rz+ Y+ m
] ! dw]D` jq^l* <u^ny{ j!a-& ta#w+
1 Sam
.<yv!nl
` = w{l /h#yT@v-= <G~ /`yy\hT
= w! ~
25:43
2 Chr
EOm=-du^ <yd]bu
* l
& ^ wyn`bl
* =W w{l-Wyh=Yw] ~
36:20
.sr`P* tWkl=m^
245
And served their idols, which
became a snare to them.
106:36
Psa
hu#r{ yl!Bm
= ! hn`yx#WpT=w~
Ezek
.hn`yx#WpT=w~ hd\Ch
* ^ tY~j-^ lk*l= hl*ka
= l
* = hn`yy\hT
= w! ~
34:5
hw]hy+ yn`da
{ & <a%n+ yn]a-* yj^
Ezek
yn]ax) hn`yy\hT
= w! ~ zb^l* yn]ax)-toyh$ /u^y~ aO-<a!
34:8
vWuyw] yn]Vh
@ ^ hz`yz]w+ var{h* tj^y-~ yh!yw+ ~
1 Chr
23:11
.tj*a# hD`qp
% l
= ! ba* tyb@l= Wyh=Yw] ~
246
8.5.1.1.3.7 With l= Meaning for
The others came out from the
city to encounter them, so that
they were trapped in the midst of
Israel, some on this side and
some on that side; and they slew
them until no one was left of
those who survived or escaped.
<t*ar`ql
= ! ryu!h-* /m! Wax=y` hL#aw@ +
Josh 8:22
<n`za
+ -* ta# WFh!-aOw+ Wum=v* aOw+
Jer 7:24
.<yn]pl
* = aOw+ roja*l= Wyh=Yw] ~
<h#yr}ja
& m
^ @ aobl* br`aM
= h
^ -^ ta# bs@h@ <u*br
= y` w` +
2 Chr
.<h#yr}ja
& m
^ @ br`aM
= h
^ w^ + hd`Why+ yn}pl
= ! Wyh=Yw] ~
13:13
247
8.5.1.1.3.9 With <u!
Saul chose for himself 3,000
men of Israel, of which 2,000
were with Saul in Michmash and
in the hill country of Bethel,
while 1,000 were with Jonathan
at Gibeah of Benjamin. But he
sent away the rest of the people,
each to his tent.
la@rc
` Y= m
] ! <yp!la
* & tv#Ov= lWav* w{l-rj^bY= w] ~
cm*km
= B
! = <y]Pl
^ a
= ^ lWav*-<u! Wyh=Yw] ~
1 Sam
13:2
1 Sam
22:2
1 Kgs 1:7
.hY`nd
] a
{ & yr}ja
& ^ Wrz+uY= w~ ~ /h@Kh
) ^ rt*yb
` a
= # <u!w+
Pc-vqw3mp ncmpcX3ms Pp np
ncmsc-np Pc-Pp np Pa-ncms Pcvqw3mp Pd np
8.5.1.2
DEICTIC
Uses of Wyh=Yw] ~
The DEICTIC use of Wyh=Yw] ~ is parallel to the yh!yw+ ~ section above. The same use of <v*
is found in both.
248
I turned and came down from
the mountain and put the tablets
in the ark which I had made;
and there they are, as the LORD
commanded me.
/ora*B* tj)Lh
% -^ ta# <c!aw* ` rh*h-* /m! dr}aw@ ` /p#aw@ `
Deut
.hw`hy+ yn]Wx
~ ! rv#aK
& ^ <v* Wyh=Yw] ~ yt!yc!u* rv#a&
10:5
Josh 4:9
<yn]hK
& h
) ^ yl@gr
+ ~ bX^m^ tj^T^ /D}rY+ h
~ ^ EotB=
Pc-afd afs ncfp vhp3ms np Ppncmsc Pa-np Pp ncmsc ncfdc Pancmp vqPmpc ncbsc Pa-ncfs Pcvqw3mp Pd Pp Pa-ncms Pa-ams
Ruth 1:2
hm*yT
+ G* ] <yt!ra
{ B
@ h
= ^ Wjr+bY= w] ~
.hZ\h^ <oYh^ du^ <yr]G` <v*-Wyh=Yw] ~
2 Sam
4:3
2 Sam
2:18
249
But the poles were so long that
the ends of the poles could be
seen from the holy place before
the inner sanctuary, but they
could not be seen outside; they
are there to this day.
vd\Qh
) -^ /m! <yD]Bh
^ ^ yv@ar` War`Yw} ~ <yD]Bh
^ ^ Wkr]aY& w~ ~
1 Kgs 8:8
1 Chr
12:40
(39)
8.5.1.3
DESCRIPTIVE
Uses of Wyh=Yw] ~
There are two sections of DESCRIPTIVE occurrences. In the first, the examples are
essentially identical to the descriptive uses of yh!yw+ ~ discussed above:
8.5.1.3.1
DESCRIPTIVE
w{Tv=aw! + <d`ah
* * <yM!Wru& <h#yn}v= Wyh=Yw] ~
Gen 2:25
.Wvv*Bt
) y= ] aOw+
Wyj=y] <ya!yc!Nh
+ ^ <h#yl@a& Wrm=aY{w~
hd`uh
@ -* lk*l= <y]m-^ yb@av
& w) + <yx!u@ yb@fj
= ) Wyh=Yw] ~
.<ya!yc!Nh
+ ^ <h#l* WrB=D] rv#aK
& ^
Josh 9:21
250
David answered the priest and
said to him, Surely women
have been kept from us as
previously when I set out and the
vessels of the young men were
holy, though it was an ordinary
journey; how much more then
today will their vessels be holy?
1 Sam
21:6 (5)
Pc-vqw3msXa np Po-Pa-ncms Pcvqw3ms PpX3ms Pp Pd-ncfs vqsfsPpX1cp Pp-Pd Pd Pp-vqcX1cs Pcvqw3mp ncmpc-Pa-ncmp ncms Pcpi3ms ncbs ncms Pc-Pc Pp Pa-ncms
vqi3ms Pp+Pa-ncms
2 Kgs
tomb* yn}hK
& ) <t*oxq=m! <h#l* Wcu&Yw~ ~
17:32
2 Kgs
17:41
<t*ba
) & Wcu* rv#aK
& ^ <h#yn}b= yn}bW=
.hZ\h^ <oYh^ du^ <yc!u) <h@
1 Chr
8:40
251
hj*Wr hV*vw! + <yu!vT
= ! <yn]Mr
) h
] * Wyh=Yw] ~
Jer 52:23
.byb!s* hk*bC
* h
= -^ lu^ ha*m@ <yn]oMr]h-* lK*
8.5.1.3.2
DESCRIPTIVE
The second section of two occurrences of Wyh=Yw] ~ with the QAL PASSIVE are unique.
This uniqueness, however, should not be attributed to some peculiarity of Wyh=Yw] .~ The QAL
PASSIVE
<t@ym!yw+ ~ /k@-yr}ja
& ^ u~vo% hy+ <K@Yw~ ~
<yx!u@ hV*mj
! & lu^ <l@tY= w] ~
Josh
10:26
.br\uh
* -* du^ <yx!uh
@ -* lu^ <y]WlT= Wyh=Yw] ~
<]lv
^ W* ry+ w{tyB@-la# dw]d` ab)Yw` ~
<yv!gl
+ P
^ ! <yv!n-` rc#u# ta@ El#Mh
# ^ jQ^Yw] ~
2 Sam
20:3
tr\mv
# m
= -! tyB@ <n}T=Yw] ~ ty]Bh
^ ^ rm)vl
= ! j~yN]h! rv#a&
ab*-aO <h#yl@aw& ~ <l@Kl
= k
= y^ +w~
.tWYj^ tWnm=la
= ^ /t*m% <oy-du^ torr|x= hn`yy\hT
= w! ~
Pc-vqw3msXa np Pp-ncmscX3ms
np Pc-vqw3ms Pa-ncms Po afs-ncfp
ncfp Pr vhp3ms Pp-vqc Pa-ncms Pcvqw3msX3mp ncmsc-ncfs Pcvdw3msX3mp Pc-PpX3mp Pnvqp3ms Pc-vqw3fp vqsfp Pp-ncms
vqcX3fp ncfsc ncfs
These examples are evidence of the limited corpus of data provided by the
Hebrew Bible, which is an important consideration in any study of this type. The
252
existence of only two unique occurrences does not necessarily make them unusual. The
Hebrew Bible does not provide examples of every possible structure that normal usage of
the language would have produced. This is an important point for the student or
grammarian to consider when tempted to make claims regarding what Hebrew can and
cannot say. It is appropriate to make statements about what does and does not occur in
the text of the Hebrew Bible, but caution needs to be exercised in making claims about
the language as a whole when there is a limited corpus.
8.5.1.4
evidence for the function(s) of yh!yw+ ~ as a verb. What is equally instructive and even more
intriguing is the absence of any temporal uses of Wyh=Yw] .~ This highlights the unique role of
yh!yw+ ~ in temporal constructions, which is the complex issue dealt with in the next chapter.
CHAPTER 9
THE TEMPORAL USES OF yh!y+w~
9.1 Overview
Besides the verbal use of yh!yw+ ,~ the other major use is in temporal clauses. Again,
the occurrences are categorized according to the syntactic environment where yh!yw+ ~
occurs. The primary purpose of this chapter is to present the remaining 408 occurrences
of yh!yw+ ,~ but at different points throughout the display of the data certain comments and
observations are necessary. More detailed discussion of the functions of yh!yw+ ~ is found in
Chapter 10.
One of the added dimensions to the analysis in this chapter is the increased
concern with the context in which the temporal expression with yh!yw+ ~ occurs. In the verbal
uses, for example, in Gen 12:10,
There was a famine in the land;
so Abram went down to Egypt to
sojourn there, for the famine
was severe in the land.
hm*yr
+ x
~ m
= ! <r`ba
= ^ dr\Yw} ~ Jr\aB
* * bu*r` yh!yw+ ~
.Jr\aB
* * bu*rh
` * db@k-* yK! <v* rWgl*
Gen
12:10
the analysis of yh!yw+ ~ is primarily concerned with its local use as an integral component of
* * bu*r` yh!yw+ ~ is an independent
the verbal clause: the WAYYIQTOL form of hy`h.* Jr\aB
253
254
clause, but the temporal expressions considered in this chapter are dependent clauses,
giving them a very different syntactic relationship with what follows. For this reason, the
temporal expressions need to be analyzed not merely in terms of their own internal
structure, but in terms of how they connect to adjacent clauses. This does not deny the
importance of careful examination of the context in which independent clauses occur, but
the analysis of the dependent, temporal clauses introduces complexities which are not
involved in the verbal uses of yh!yw+ .~ This is why attention needs to be given to whether
the temporal expression with yh!yw+ ~ is followed by a WAYYIQTOL or QATAL. One
dimension is the set of components which occur with yh!yw+ ~ in the temporal expression
itself; the connection of the entire temporal expression with the next clause is another
dimension of this analysis.
The main divisions of this chapter are:
9.2 Introduction to Temporal Expressions in Biblical Hebrew
9.3 Temporal Expressions with yh!yw+ ~
9.4 Summary of the Temporal Uses of yh!yw+ ~
Section 9.3 is the heart of this chapter, containing all the data displayed in the
following categories:
9.3.1 With rj^a/^ yr\ja
& ^
9.3.2 With Prepositions
9.3.3 With rv#aK
& ^
9.3.4 With yK!
255
9.3.5 With Specific Temporal Reference
9.3.6 Occurrences of yh!yw+ ~ with Adverbs
The number of occurrences of yh!yw+ ~ in each category varies widely. In 9.3.2, for
example, two of the largest sections are those which display the uses of yh!yw+ ~ followed by
the prepositions B= and K=. The summary which concludes this chapter sets the stage for
the discussion in Chapter 10, the Discourse-Pragmatic uses of yh!yw+ .~
Introduction
Introductory grammars and textbooks on syntax cannot be expected to be
256
clauses, he discusses the role of prepositions, saying that the choice of B= or K= is governed
by the aspectual nature of certain verbs. He states, however, that a catalog of uses would
serve no purpose in an elementary grammar of this sort (Lambdin 1971, 129). The
decision to not include a catalog of uses in an introductory grammar is understandable,
but unfortunately, information which is deemed too advanced or detailed for the
elementary grammar seldom finds its way into another publication. Also, a mere catalog
of uses is of limited value if there is no discussion of the function of the linguistic items
listed therein. The problem lies, however, not in the amount of space or pages which deal
with the topic, but in the atomistic perspective that is typically reflected in many
grammatical and syntactical descriptions, i.e., presenting examples of grammatical
categories without considering their connection to the system of the language as a whole.
For example, Davidsons Syntax states, [w]hen is expressed by B=, K= with infin.,
& ^ with finite verb (or nominal cl.) (Gibson 1994, 157). It is important, of
or by yK!, rv#aK
course, to know what items like yK! mean, but to say that all four items listed here
mean when does not help the student know when to use which when. In a sense, this
description provides a type of lexical equivalent. If the goal were to merely assign lexical
equivalents for each linguistic item in Hebrew, this might be adequate, but when the goal
is understanding the use and function of these items, a greater depth of understanding is
needed.
At the level of the narrative or discourse as a whole, understanding the use and
function of temporal expressions is fundamental. From a functional perspective, analysis
not only involves cataloguing the linguistic items involved in temporal expressions, it
257
also requires exploring the different contexts of use and the factors which motivate and
govern the use of the different temporal expressions. The motivation to explore the
functions of, for example, the prepositions B= and K= with the infinitive construct is the
analytical principle of choice (5.2.4.4). Choice implies that there is some functional
difference that motivates the use of B= or K= in certain contexts. This indicates, then, that
both B= and K= cannot mean exactly the same when. Any description of these prepositions,
even in introductory grammars, should reflect the different nuances of meaning and
usageor at least alert the student to their context-sensitive nature.
The following brief review of GKC, Davidson, Williams, Joon-Muraoka,
Lambdin, Waltke and OConnor, Pratico and Van Pelt, and van der Merwe, reveals that
very little attention has been given to the variety of temporal expressions in biblical
Hebrew. Most of the discussion in these grammars is dedicated to the lexical level, with
some attention given to the basic syntax, for example, of the combination of prepositions
with infinitive constructs to form temporal expressions.
9.2.1.2
relations of time simply by juxtaposition (GKC 1910, 501) of, for example, a series of
WAYYIQTOL
verb forms. and the conjunctions used to introduce temporal clauses (GKC
1910, 502), such as yK! and rv#a.& In addition, other conjunctions such as <u!, w)mK=, and
rv#ak
& ^ are presented.
The use of prepositions with the infinitive construct, however, is only included as
one of three additional, secondary remarks. According to GKCs section on infinitive
258
constructs, these constructions are equivalent to temporal clauses, stating that [t]his use
of the infinitive construct is especially frequent in connexion with B= or K= to express timedeterminations, especially after yh!yw+ ~ (GKC 1910, 347). Later on, the following statement
is also made: The infinitive with B= may usually be rendered by when, as, or whilst; the
infinitive with K= by when, as soon as. (GKC 1910, 503) Examples are given, but there
is no further discussion of any difference in usage between B= and K= with the infinitive
construct.
The use of B= and K= with infinitive constructs is indeed a frequent means employed
in biblical Hebrew to express temporal relations. The numerous occurrences of the
prepositions B= and K= with the infinitive construct following yh!yw+ ~ are displayed in later
sections of this chapter. The display of all the data is the focus of this chapter, but
comments and observations are included to guide the reader through the extensive
sections of examples. Also, GKCs claim regarding the temporal role of juxtaposition
will also be discussed, but this is postponed until Chapter 10.
9.2.1.3
prep. and infin (Gibson 1994, 157) and further that [t]emporal clauses or phrases are
commonly preceded by yh!yw+ ~ or hy`hw* ,+ punctuating a narrative or discourse time-wise
(Gibson 1994, 157). The reference to the function of punctuating a narrative timewise
seems significant, but unfortunately it is not explained.
259
Regarding the use of prepositions with the infinitive constructs, Davidson states
that [w]hen is expressed by B=, K= with infin., or by yK!, rv#aK
& ^ with finite verb (or nominal
cl.) (Gibson 1994, 157), but there is no discussion of the parameters that govern the
& ^ are all used to mean
choice of B= or K= with the infinitive. To state that B=, K=, yK!, and rv#aK
= ,! za*m,@
when helps separate these items from the others that are considered (yr@j&a,^ yn}pl
du^, and yD}m)! (Gibson 1994, 157-58), but it obscures the fact that B=, K=, yK!, and rv#aK
& ^ are
not all the same in usage and meaning. As discussed in 9.2.1.1, certain decisions have to
be made to restrict the amount of information given in an introductory grammar, but
proper distinctions in meaning and usage should be made.
9.2.1.4
260
In his discussion of temporal clauses, Williams repeats the example given in GKC
164b1 of [s]imple juxtaposition, meaning when, e.g. rm#aT)w~ w{tq)vh
= l
^ = lk^Tw= ,~ When
she had finished giving him a drink, she said (Gn 24:19) (Williams 1976, 83). This
point will not be dealt with in detail here, but as stated above, this claim has important
ramifications that will be discussed in Chapter 10.
9.2.1.5
made to get at the reason for the difference in meaning by expressing their particular
nuance in temporal clauses. The description of B= indicating the inclusion of an action in
the time of another seems more intuitive than that of K= indicating that the time of one is
like that of the other. This is an attempt to extend the lexical meaning of the
prepositions to their use in temporal expressions.
There is significant cognitive support for maintaining the connection between the
lexical meaning and the temporal nuances, but it is unclear what is meant by K= indicating
that one time is like another. The discussion of the examples with K= later in this chapter
provide the basis for evaluating these claims. This matter will be discussed in further
detail in the next chapter.
261
9.2.1.6
Both prepositions are common in this usage, but with certain verbs,
especially um^v* and ha*r,* K= is by far the preferred preposition. There are
reasons, founded in the aspectual nature of these verbs, which govern this
choice, but a catalog of uses would serve no purpose in an elementary
grammar of this sort. (Lambdin 1971, 129)
Even though Lambdin recognizes that certain factors are involved in the choice of
B= or K=, the examples give the impression that both prepositions are used as when. The
comment that the real issue is verbal aspect is intriguing, but unfortunately, he provides
no further discussion. This comment will, however, be discussed below in 9.3.2.2.1.5
after the display and discussion of the occurrences of both B= and K= with infinitive
constructs.
One further comment by Lambdin regarding temporal clauses has particular
significance for the present study. Lambdin states that
[t]he verb hy`h* in a leading clause requires special consideration. By virtue
of its double meaning be/become it may be used to describe a nonpunctual past tense situation (e.g. there was a famine in the land). If a
narrative sequence begins with a clause containing the verb hy`h* (or yh!yw+ )~ ,
the real nature of the sequence is not clear until we reach a continuing
verb. (Lambdin 1971, 279)
262
The comment regarding the need to consider the continuing verb in a narrative
sequence initiated by yh!y+w~ has important implications for the analysis of yh!yw+ ~ presented in
the upcoming sections of the present chapter. This comment reflects an awareness of the
complex nature of the narrative web which is also what motivated the separation of the
temporal occurrences of yh!yw+ ~ into categories based on the type of verb which directly
follows.
9.2.1.7
temporal clauses are discussed. First of all, in 36.2.2, the temporal use of the infinitive
construct with the prepositions B=, K=, and l= is presented. Specifically with reference to B=
and K= (Examples 2-7 in 36.2.2b), they state that B= denotes in general the temporal
proximity of one event to another, K= more specifically the more immediately preceding
time (Waltke and OConnor 1990, 604). With the limited examples they give, it is
difficult to tell exactly what is meant by this distinction between B= and K=. There is ,
unfortunately, no further discussion of the temporal use of the infinitive construct with B=
and K=.
There is, however, a separate discussion of the preposition l=, stating that its
temporal use is to mark a point in time or an extent in time (Waltke and OConnor
1990, 607). Again, limited examples are given without much discussion of the usage of l=
beyond stating that signalling a point is chiefly associated with the verb pny to turn
263
(Waltke and OConnor 1990, 607). The uses of l= with infinitive constructs are displayed
in section 9.3.2.3.1 below, followed by further discussion of its uses.
Secondly, in 38.7, in the chapter entitled Subordination, they comment that
[t]he majority of dependent temporal clauses are formed with an infinitive
introduced by a preposition. There are a variety of other temporal clause
types, however, introduced by other particles. These may be classified
according to the temporal relation of the main clause situation and that of
the subordinate clause. (Waltke and OConnor 1990, 643)
The ten examples they provide are of different temporal particles which indicate
contemporary, later, or preceding temporal situations, referring to the temporal relation
of the main clause situation and that of the subordinate clause (Waltke and OConnor
1990, 643). For example, according to Waltke and OConnor, du^ indicates a later
situation.
The third section deals with clausal adverbs, discussing deictic and independent
temporal adverbs. These adverbs are listed with their corresponding lexical meanings,
but no examples of usage are given. Even though a variety of temporal expressions are
catalogued in Waltke and OConnor, there are numerous issues of their uses and
functions that are not touched on in their volume on syntax.
9.2.1.8
264
Regarding the time reference of the infinitive, Pratico and Van Pelt state that the
temporal modifiers yh!yw+ ~ and hy`hw` + provide the context for determining the temporal
value (Pratico and Van Pelt 2001, 245). According to this grammar, the temporal
modifier yh!yw+ ~ signals past tense narration and the temporal modifier hy`hw* + signals future
tense narration (Pratico and Van Pelt 2001, 245), providing the temporal value for the
otherwise atemporal infinitive construct. This claim regarding the function of yh!yw+ ~
requires further analysis and will be discussed in Chapter 10.
In the section dealing with Exceptions to Word Order with Verb First (23.3),
Pratico and Van Pelt have two more comments regarding temporal expressions. The first
is:
While the verb does usually stand first in a sentence or clause, it may also
be preceded by an adverb of time, an adverbial phrase, the word hN}h!
(behold), a temporal modifier (yh!yw+ ~ or hy`hw` )+ , an expression that provides
context or circumstantial information or an independent personal pronoun
for emphasis. (Pratico and Van Pelt 2001, 272)
The assumption of verb-first word order is valid for specific types of clauses in
certain contexts. Pratico and Van Pelts three examples do indeed have verbs in the first
position, but there is no consideration given to the fact that the examples come from very
different text-types. The first example is from narrative and begins with a WAYYIQTOL,
whereas the other two examples are from poetry. The verbs in these poetic examples are
a QATAL and an imperative. At first glance, this may seem inconsequential or it may be
the understandable result of Pratico and Van Pelts decision to present only the most
basic issues of sentence structure (Pratico and Van Pelt 2001, 272), but it is crucial that
even the most basic examples be carefully selected from the same text type. To say that
265
the three examples in 23.3 are evidence of verb first word order describes the surface
form, but does not take into account the syntactic and contextual factors that motivated
the occurrence of these verbs in first position. When claims are made that certain
syntactic patterns or word orders are characteristic of a language, extreme care needs to
be exercised to control the variables in order to make the most valid statement possible.
The desire to present the most basic issues of sentence structure is valid, but certain
fundamental concepts and theoretical notions must always be operative. All of the
examples in 23.3 should have come from narrative, or some comment should have been
made acknowledging the different text-type sources from which the examples came.
The second comment in 23.3 is that the verb may be preceded by a temporal
clause beginning with yh!yw+ ~ or hy`hw` .+ Two examples are given, one with B= and the other
with K=, which give the impression, unfortunately, that very little difference, if any, exists
between these two prepositions. This is congruent with the statement cited above that
the prepositions B= and K= are translated either when or while, which also indicates
that these prepositions are assumed to be interchangeable. This matter will not be
discussed further here, but will be dealt with later in this chapter after the occurrences of
B= yh!yw+ ~ and K= yh!yw+ ~ have been displayed.
9.2.1.9
many uses of the various prepositions, the temporal uses of B= and K= are discussed, stating
that: 1) The preposition B= + infinitive construct often refers to events that provide the
temporal frame of an event or events referred to in a subsequent sentence (van der
266
Merwe, Naud, Kroeze 1999, 281) and 2) K= + infinitive construct is used to indicate that
an event referred to in the main clause following the temporal clause with the K= +
infinitive construct immediately follows it in time (van der Merwe, Naud, Kroeze 1999,
284). These comments reflect the type of distinction between B= and K= found in GKC and
Joon-Muraoka, but they are expressed more clearly here in BHRG. The display of the
occurrences of -B= yh!yw+ ~ and -K= yh!yw+ ~ later in this chapter will provide a good basis for
evaluation of this proposed distinction. This matter will then be discussed in more detail
in 9.3.2.2.1.5.
One of the insightful comments in BHRG that reflects awareness of the important
role of syntax is that a distinction must be made between the preverbal field (Vorveld)
and main field (Hauptveld) of a BH verbal clause (van der Merwe, Naud, Kroeze
1999, 337). An adjunct of time (van der Merwe, Naud, Kroeze 1999, 339) is given as
an example of the type of item that can occupy the preverbal field, claiming that it is
typically used to provide the temporal point of orientation of the subsequent event(s)
(van der Merwe, Naud, Kroeze 1999, 339).
The significance of BHRGs distinction between the preverbal and main fields
may not be immediately apparent here, but these concepts have important implications
for understanding the function of yh!yw+ ~ in temporal expressions. More detailed discussion
will come in later sections.
9.2.1.10 Summary
The preceding review of how temporal clauses are dealt with in the selected
grammars reveals basic agreement as to the linguistic items in biblical Hebrew that
267
participate in the expression of temporal relations. There is significant variation,
however, in the meanings attributed to the prepositions B= and K= with infinitive constructs.
Some of the grammars appear to make no distinction in meaning, while others seek to
differentiate them. In the grammars that attempt to differentiate B= from K= with infinitive
constructs, the explanation of the use of B= typically seems easier to understand than the
explanation of K=. As stated above, however, BHRG is an exception to this statement. The
following chart displays various views of B= and K=:
Grammar
Joon-Muraoka:
Waltke and OConnor:
van der Merwe et al:
B=
inclusion of an action in
the time of another
denotes in general the
temporal proximity of one
event to another
refers to events that
provide the temporal frame
of an event or events
referred to in a subsequent
sentence
K=
the time of one is like that
of the other
denotes more specifically
the more immediately
preceding time
indicates that an event
referred to in the main
clause following the
temporal clause with the K=
+ infinitive construct
immediately follows it in
time.
One of the common characteristics in the grammars reviewed in this section is the
attention paid to the connection of the temporal clause to the main or independent clause.
Another feature they share is the lack of attention paid to the connection of the temporal
clause to the broader textual context.
From this review, the following issues surface in the analysis of temporal
expressions:
1) Greater clarity is needed in differentiating the uses and functions of the
prepositions with infinitive constructs
268
2) Further exploration is needed of the connection of the temporal clauses
to the broader context
3) The role of yh!yw+ ~ needs further analysis to evaluate the claim that it is a
temporal modifier, providing the temporal value for the infinitive
construct
It will be important to keep these issues in mind as the following sections move
through the different categories of yh!yw+ ~ s uses in temporal expressions.
Speaker Deixis
Speaker deixis refers to the complex systems of reference within speech and
narrative that express the speaker or narrators spatial or temporal point of reference
relative to what is being spoken or narrated. The speaker has many deictic mechanisms
and systems at his or her disposal for making spatial and/or temporal reference. The
speaker and or narrator makes selections out of the set of possible expressions to best
accomplish his or her narrative strategy.
One of the most basic concepts is that events in a narrative, by their very nature,
make relative temporal reference to each other. In biblical Hebrew narrative, a series of
269
WAYYIQTOL verbs, as stated in 6.4.1.1, depicts events as if they were a series of points
along the same line.
Each WAYYIQTOL moves along the temporal dimension of the narrative, with each event
establishing a new Reference Time. Even though there are some examples of
REGRESSION,
270
text. In biblical Hebrew, for example, the Reference Time is established by each
successive WAYYIQTOL in the text. Departures from this pattern are significant and raise
questions about their function and temporal reference relative to the WAYYIQTOL pattern.
One of the most common departures from the WAYYIQTOL series is some type of structure
with a QATAL form. Hatav makes the observation that the QATAL does not introduce or
update the Reference Time (Hatav 1997, 80). The temporal reference of the QATAL is
relative to that established by the WAYYIQTOL, typically prior to or anterior to the
established Reference Time. When a QATAL occurs in a series of WAYYIQTOL verbs, the
temporal reference is temporarily moved back relative to the established Reference Time
in the context. This concept is crucial to the temporal interpretation of the examples in
the following sections where the temporal expression is nearly identical, but the
following verb is a WAYYIQTOL in one case and a QATAL in another.
When they saw him, they brought
thirty companions to be with him.
Pc-vqw3msXa Pp-vqcX3mp PoX3ms
Pc-vqw3mp amp ncmp Pc-vqw3mp
PpX3ms
Judg 14:11
hK*lM
= h
^ ^ rT@sa
= -# ta# El#Mh
# ^ toar+k! yh!yw+ ~
Esth 5:2
.fyb!rV
+ h
^ ^
Based on the fundamental principle that linguistic systems are not random, but
rather are intricately interactive, context-sensitive systems, the assumption is that
271
occurrences of QATAL and WAYYIQTOL are motivated by factors relevant to narrative
strategy. The precise motivations for each case may not be accessible to the modern
reader or analyst, but the alternative is much more unsatisfactory, requiring the
conclusion that the QATAL and WAYYIQTOL are sometimes used interchangeably for no
apparent reason. These concepts need to be kept in min throughout the extensive sections
of data displayed later on in this chapter.
9.2.2.2
the constants is that word order is context-sensitive. This is one of the reasons that
claims for basic word order need to be carefully qualified with respect to genre, the
syntactic environment, and narrative context. Whole books need to be written about
word order phenomena in biblical Hebrew, but certain elementary observations need to
be presented here before proceeding with the display of the data in this chapter.
The specific word order issue to be introduced here is the WE-X-QATAL, which is
briefly discussed above in 6.3.1.4. For example, Gen 4:1 begins with a WE-X-QATAL.
Now the man had relations with
his wife Eve, and she conceived
and gave birth to Cain, and she
said, I have gotten a manchild
with the help of the LORD.
When this is seen next to Gen 4:17, comparison of the WE-X-QATAL and
WAYYIQTOL
is facilitated.
Gen 4:1
272
Cain had relations with his wife
and she conceived, and gave
birth to Enoch
Gen 4:1
There are several dimensions of this word order variation. The proposal here is
that both the WAYYIQTOL and the WE-X-QATAL are motivated by narrative strategy.
Specifically with reference to the WE-X-QATAL, analysis of its occurrences requires
attention to all three components, the w+, the nominal item, and the QATAL in order to work
toward an answer to questions like why Gen 4:1 has WE-X-QATAL and Gen 4:17 starts
with a WAYYIQTOL.
9.2.2.3
Summary
The goal in these preliminary comments is to alert the reader so that careful
discussed. Observations regarding the syntactic patterns and word order variations are
made following these sections of data.
273
chapters eight and nine is the fact that the verbal occurrences of yh!yw+ ~ in chapter eight are
independent clauses, whereas those which are presented in this chapter are dependent
clauses. For this reason, the description in chapter eight deals primarily with the
elements in the individual clause within which yh!y+w~ occurs, and now, the description must
take into account not only the dependent clause in which yh!yw+ ~ is found, but also the other
clause(s) to which yh!yw+ ~ is connected. As Gropp comments, [t]he construction wayh K/B- + infinitive construct syntactically requires a main verb to follow (Gropp 1995,
203). Consequently, the following sections of occurrences of yh!yw+ ~ are classified by
whether the next verbal form following yh!yw+ ~ is a WAYYIQTOL, or a QATAL, or a WE-XQATAL.
By making these classifications, certain questions about the patterning of yh!yw+ ~ can
be explored, seeking to discern whether there are pragmatic motivations for these
patterns.
The verbal uses of yh!yw+ ~ (Chapter 8) are not surprising or unusual even though the
semantic range of hyh may exceed that of some other verbs. However, the use of yh!yw+ ~ in
temporal expressions tends to stretch the notion of verb, since in many contexts it is
as many have claimedindeed clumsy and awkward to translate these occurrences of
yh!yw+ ~ as a verb. For example, consider two possible translations of 1 Sam 20:35:
hd\Ch
* ^ /t*no` hy+ ax@Yw} ~ rq#Bb
) ^ yh!yw+ ~
1 Sam
20:35
274
In terms of English style, the second option is arguably more colloquial, but it
must be acknowledged that the first option alerts the English reader to the presence of
yh!yw+ .~ This does not mean, however, that now it came about gives the reader access to
the function of yh!yw+ .~ In fact, it may appear to be nothing more than a strange Hebraism
when translated this way. Part of the awkwardness of translating yh!yw+ ~ as a verb when it
occurs in temporal expressions may be because it is not functioning as a verb in these
occurrences.
This is where the crucial question is not What is yh!yw+ ?~ but What is yh!yw+ ~ s
function? The answer to the first question is that yh!yw+ ~ is a verbas chapter eight has
clearly demonstrated. The answer to the second question is that yh!yw+ ~ may indeed be a
verb, but its function in temporal expressions may not be to state the propositional
content and it was or and it happened.
Syntactically, the yh!yw+ ~ of temporal clauses does not have the same connection that
yh!yw+ ~ has in its verbal occurrences. This, in fact, is one of the most remarkable differences
between the verbal and temporal uses of yh!yw+ .~ Verbal yh!yw+ ~ is directly linked to the clause,
as stated in 8.1, as the nuclear verb of the clause where the person, number, and gender
match that of the subject of the clause. The temporal use of yh!yw+ ~ does not have the same
syntactic link to the clause with which it occurs. This raises an important issue in the
analysis of yh!yw+ ~ s function in temporal expressions since certain ones can occur with or
without yh!yw+ .~ If all temporal expressions occurred with an obligatory yh!yw+ ,~ the analysis
275
would be much more straightforward. This, of course, is not the case, since the following
contrast is attested:
yv!yl!Vh
= ^ <oYB^
yv!yl!Vh
= ^ <oYb^ yh!yw+ ~
Gen 22:4
Gen
34:25
The question of what yh!y+w~ contributes to Gen 34:25 that is not in Gen 22:4 is
unavoidable. If yh!yw+ ~ is not a necessary part of all temporal expressions, then why is it
used where it does occur? This is no longer a matter of clause syntax, but requires
exploration of the possible discourse-pragmatic motivations for its use. This is one of the
reasons why it is necessary to not only study the occurrences of yh!yw+ ~ itself, but also the
various constructions which occur with it.
Even though yh!yw+ ~ with a temporal expression does not have the same syntactic ties
to the clause as it does in its verbal uses, this does not mean that yh!yw+ ~ used temporally has
lost every connection to yh!yw+ ~ used verbally. It is helpful to consider yh!yw+ ~ from the
standpoint of the process of grammaticalization. According to Hopper and Traugott,
[w]hen a content word assumes the grammatical characteristics of a function word, the
form is said to be grammaticalized (Hopper and Traugott 1993, 4). In the process of
assuming another function, it is common for there to be some loss of certain aspects of
the previous forms function and meaning (Hopper and Traugott 1993, 2-4).
In its verbal uses, yh!yw+ ~ or yh!Tw= ~ is the required nuclear verb of the independent
clause in which they occur. When yh!yw+ ~ occurs in temporal expressions, several
morphosyntactic differences are noted:
276
1) yh!yw+ ~ is the only form attested in temporal expressions. yh!Tw= ~ never
occurs
2) yh!yw+ ~ has no explicit SUBJECT in the temporal clause
3) yh!yw+ ~ has no nominal or adjectival complement in the temporal clause
It is not possible to explain or reconstruct the precise circumstances under which a
verb like yh!yw+ ~ becomes grammaticalized or takes on another function. What is possible,
however, from a functional-typological perspective, is
1) to identify the functions of hy*h* as an equative verb or copula that make
it a likely candidate for yh!yw+ ~ s functions in temporal expressions, and
2) to find cross-linguistic support in other languages where a similar
phenomenon occurs
The grammaticalization of deictics is discussed by Hopper and Traugott, stating
that deictics may be used for metalinguistic functions involving clause reference in order
to achieve overt linking of clauses (Hopper and Traugott 1993, 178). The deictic use of
yh!yw+ ~ is proposed as one of the cognitive links or associations between the verbal uses and
the temporal uses. Hopper and Traugott cite Swahili, Japanese, and Chickasaw as
evidence of other languages in which copula constructions are grammaticalized as clause
linkers (Hopper and Traugott 1993, 179). This does not mean, however, that the DEICTIC
meaning there is is directly transported from the verbal to the temporal use, but rather
that there is a cognitive association in DEICTIC function. This is very similar to how w+ is
described in 6.3.1.3, in the sense that it is the function and not the lexical meaning that is
in focus. Likewise, with regard to yh!yw+ ,~ it is the DEICTIC function that needs to be kept in
focus in the analysis of its use in temporal expressions.
277
The Cognitive Orientation introduced in 5.2.5 provides helpful concepts for
describing this deictic function of yh!yw+ ,~ which will be discussed in more detail in Chapter
10. The point here is to state this hypothesis as a way of explaining the connection
between the verbal and temporal uses of yh!yw+ .~ This chapter now proceeds with the
display of the occurrences of yh!yw+ ~ used in temporal expressions.
278
2 Sam 17:21
Infinitive Construct:
Gen 25:11
Noun Phrase:
<T*kl
= #
<h*rb
` a
= ^ tom
yr}ja
& ^ yh!yw+ ~
Gen 22:20
Noun Phrase:
Judg 16:4
Adverb:
hL#ah
@ * <yr]bD
* h
+ ^
/K@
After each section which displays the data, a brief statement is made about the
preceding context of each example. More detailed discussion of these sections is found in
the summary in 9.3.1.4.
9.3.1.1
phrase hL#ah
@ * <yr]bD
* h
+ ^ rj^a,^ literally after these things, found in Gen 22:1, 20; 39:7; 40:1;
48:1; Josh 24:29; 1 Kgs 17:17, 21:1.
9.3.1.1.1
Followed by WAYYIQTOL
<h*rb
` a
= l
^ = dG~Yw| ~ hL#ah
@ * <yr]bD
* h
+ ^ yr}ja
& ^ yh!yw+ ~
rm)al@
wyn`da
{ -& tv#a@ aC*Tw! ~ hL#ah
@ * <yr]bD
* h
+ ^ rj^a^ yh!yw+ ~
.yM!u! hb*kv
= ! rm#aT)w~ [s@oy-la# h*yn\yu@-ta#
Gen
22:20
Gen 39:7
279
After these things Joseph was
told, Behold, your father is
sick. So he took his two sons
Manasseh and Ephraim with
him.
hL#ah
@ * <yr]bD
* h
+ ^ yr}ja
& ^ yh!yw+ ~
Gen 48:1
Josh
24:29
Gen 39:7
Gen 48:1
Josh 24:29
These examples provide an interesting contrast in Gen 22:20 and Gen 48:1. The
WAYYIQTOLs
Gen 22:20
<h*rb
` a
= l
^ = dG~Yw| ~
Gen 48:1
[s@oyl= rm#aY{w~
The Hofal WAYYIQTOL in Gen 22:20 is perhaps more expected in this context than
the Qal WAYYIQTOL in Gen 48:1. This is not the place to enter into a lengthy discussion,
but the unusual nature of the Qal in Gen 48:1 needs to be acknowledged and there needs
to be some consideration of possible motivating factors. GKC 144d lists Gen 48:1 as an
example of the third person used to indicate an indefinite personal subject (GKC 1910,
280
460) as in English they say or in similar expressions in the Romance languages. This is
an important area of further research, but is beyond the scope of the present study.
9.3.1.1.2
Followed by QATAL
In three cases, the first verbal element to follow the temporal expression is a
QATAL.
hL#ah
@ * <yr]bD
* h
+ ^ rj^a^ yh!y+w~
hp#ah
) w* + <y]rx
~ m
= -! El#m# hq@v=m^ Waf=j*
.<y]rx
` m
= ! El#ml
# = <h#yn}da
{ l
& ^
Gen 40:1
hL#ah
@ * <yr]bD
* h
+ ^ rj^a^ yh!y+w~
1 Kgs
ty]Bh
* ^ tl^uB
& ^ hV*ah
! -* /B# hl*j*
17:17
hL#ah
@ * <yr]bD
* h
+ ^ rj^a^ yh!y+w~
yl!au@rz+ Y+ h
] ^ tobn`l= hy`h* <r\K#
1 Kgs
21:1
ba*ja
= ^ lk^yh@ lx#a@ lau#r+zy+ B
] = rv#a&
./orm=v) El#m#
Joseph is in prison
1 Kgs 17:17
1 Kgs 21:1
The back-reference in these examples is not to any specific event, but rather to the
preceding section of the narrative as a whole. The QATAL which follows the temporal
281
expression indicates that either the event took place or the state of affairs existed prior to
the event of the first WAYYIQTOL of the narrative which follows.
9.3.1.1.3
Followed by WE-X-QATAL
In one instance, the verbal element following the temporal clause is WE-X-QATAL.
After these things, God tested
Abraham, and said to him,
Abraham! And he said, Here
I am.
Pc-vqw3msXa Pd Pa-ncmp Pa-acp
Pc-Pa-ncmp vpp3ms Po-np Pcvqw3ms PpX3ms np Pc-vqw3ms
PiX1cs
hL#ah
@ * <yr]bD
* h
+ ^ rj^a^ yh!y+w~
Gen 22:1
<h*rb
` a
= -^ ta# hS*n] <yh!Oa$h*w+
.yn]Nh
} ! rm#aY{w~ <h*rb
` a
= ^ wyl*a@ rm#aY{w~
282
Wenham also comments regarding the use of <yh!Oa$ that [i]t is unusual that this
story begins with this generic form rather than with his personal name, the Lord
(Wenham 1994, 103). It is surprising, however, that there is no mention of the word
order. Regarding this aspect of Gen 22:1, Hamilton comments that
[n]ormal Hebrew syntax calls for the verb to precede the subject; hence we
would expect: tested Elohim Abraham. But the placing of the subject
first, as here, draws special attention to it: the Elohimhe tested
Abraham! Since the he is already contained in the verb, the Elohim
must be taken as a casus pendens. (Hamilton 1995, 101)
Comments like these show the important role that grammar and syntax play in
how commentators interpret aspects of the text. Their grammatical tradition and
analytical perspective play a fundamental role in what is perceived as significant and
noteworthy. For example, Speisers comments on Gen 22:1 are as follows:
God put Abraham to the test. Heb. is inverted for emphasis, and the effect
is heightened by the definite article with Elohim. The idea is thus
conveyed that this was no ordinary procedure, but that God had a
particularly important objective in mind. But the precise shading is
difficult to determine. It might be that God chose to do so, or that it was an
exceptional test. (Speiser 1964, 162)
It is clear from these comments that how these syntactic features are perceived
can greatly influence the importance that is attributed to them. Full discussion of the role
and function of the WE-X-QATAL is not possible here, but it would be necessary for a
complete analysis of why this part of Gen 22:1 is <h*rb
` a
= -^ ta# hS*n] <yh!Oa$h*w+ rather than
<h*rb
` a
= -^ ta# <yh!Oa$h* hS@n~yw+ .~ The text needs to be accepted as it is, but comprehensive
analysis involves understanding the factors which motivate the different syntactic options
available in the language. Neither analysis above of the WE-X-QATALas casus pendens
283
or as emphasisprovides a satisfactory answer to the question of its function here. This
is an issue which requires more detailed analysis and research.
9.3.1.1.4
First of all, these occurrences provide excellent evidence for why all the data
needs to be considered in the study of an item like yh!yw+ .~ As mentioned in 2.4.3, on the
basis of the occurrences in Genesis, it is not only reasonable but methodologically valid
to formulate a hypothesis regarding the possible temporal-structuring-function of this
phrase. When all the data is compiled, however, the apparent importance based on
@ * <yr]bD
* +h^ rj^a^ only occurs three other times in
Genesis is mitigated by the fact that hL#ah
what do not appear to be narratively strategic uses of the expression. This does not
necessarily invalidate the expressions potentially significant role in Genesis, but it does
caution against making generalizations based only on its occurrences there.
There is a sense in which the narrative preceding the yh!yw+ ~ leads up to or is
logically prior, but the connection is not very precise. A comparison of English versions
of Gen 39:7 and 40:1 shows the following differences in how after these things has
been rendered:
Gen 39:7
NASB
NIV
NRSV
NLT
NASB
NIV
NRSV
NLT
284
rj^a^ does not specify the exact amount of time that has passed, and these things
cannot be read as a specific reference to certain events that were the direct cause of what
follows. It is certainly true that a certain amount of time has to transpire for the
hL#ah
@ * <yr]bD
* h
+ ^ to take place, but this is not literally a temporal reference. For example,
referring to Gen 39:7, Hamilton comments that [t]he these things with which the verse
begins refers back to Potiphars entrusting Joseph with the supervision of his household
(Hamilton 1995, 463). In contrast, explicit temporal reference is seen in the following
examples:
After a considerable time Shua's
daughter, the wife of Judah,
died;
Gen
38:12
Josh 23:1
byb!Sm
* ! <h#yb@ya
+ -) lK*m! la@rc
` y= l
] = hw`hy+
.<ym!YB
` ^ aB* /q@z` u~vo% hyw]
WhY`la
! -@ la# hy`h* hw`hy+-rb^dW+ <yB!r~ <ym!y` yh!yw+ ~
1 Kgs
rm)al@ tyv!yl!Vh
= ^ hn`VB
* ^
18:1
dq@PT
* ! <yB!r~ <ym!Ym
` !
Ezek
38:8
Since explicit temporal reference is possible by these and other means, the
translation of hL#ah
@ * <yr]bD
* h
+ ^ examples should be expressed in a way that maintains
reference to the preceding events. It should be noted, but will not be discussed here, that
285
the temporal reference in Ezek 38:8 is the same as Josh 23:1, except for the yh!yw+ .~ The
issue of the occurrence of temporal expressions like these without yh!yw+ ~ is discussed in
Chapter 10.
The full analysis of these occurrences needs to also take into consideration the
@ * <yr]bD
* h
+ ^ rj^a^ without yh!yw+ .~ Previous comments regarding
following instances of hL#ah
the relatively low frequency of occurrence of this expression remain essentially valid,
since there are only five occurrences without yh!yw+ ,~ making a grand total of thirteen.
& ,^ and rj^aw^ + are all found. The in the Ezra example indicates the
Notice that rj^a,^ yr}ja
intervening text that is not displayed.
After these things the word of
the LORD came to Abram in a
vision, saying, Do not fear,
Abram, I am a shield to you;
Your reward shall be very
great.
Gen 15:1
<r`ba
= ^ ar`yT!-la^ rm)al@ hz\jM
& B
^ ^ <r`ba
= -^ la#
.da)m= hB@rh
+ ^ ;r+kc
* = El* /g}m* yk!na
{ *
Pd Pa-ncmp Pa-acp vqp3ms ncmscnp Pp-np Pp+Pa-ncms Pp-vqc Pdvqi2ms{1}Jm np pi1cs ncms
PpX2fs ncmscX2ms vha Pd
tWkl=mB
^ = hL#ah
@ * <yr]bD
* h
+ ^ rj^aw^ +
hy`rc
` -= /B# ar`zu
+ # sr`P-* El#m# aT=sv
= j
^ T
= r
^ a
+ ^
.hY`ql
! j
= -! /B# hy`rz+ u
~ -& /B#
lb#Bm
* ! hl*u* ar`zu
+ # aWh
hL#ah
@ * <yr]bD
* h
+ ^ rj^a^
yT!vw= -~ ta# rk^z` vorw}vj
= a
^ & El#Mh
# ^ tm^j& Ev)K=
.h*yl#u* rz~gn+ -] rv#a& ta@w+ ht*cu
* -* rv#a& ta@w+
Ezra 7:1
Ezra 7:6a
Esth 2:1
286
After these events King
Ahasuerus promoted Haman, the
son of Hammedatha the Agagite,
and advanced him and
established his authority over all
the princes who were with him.
hL#ah
@ * <yr]bD
* h
+ ^ rj^a^
at*dM
` h
= -^ /B# /m*h-* ta# vorw}vj
= a
^ & El#Mh
# ^ lD~G]
w{as=K-! ta# <cY# w` ~ Wha@Cn= y~ w+ ~ yg]ga
` h
& *
.w{Ta! rv#a& <yr]Ch
* -^ lK* lu^m@
Pd Pa-ncmp Pa-acp vpp3ms Pancms np Po-np ncmsc-np Pa-np Pcvpw3msX3ms Pc-vqw3msXa PoncmscX3ms Pp-Pp ncmsc-Pa-ncmp
Pr PpX3ms
Esth 3:1
hL#ah
@ * tm#ah
$ w* + <yr]bD
* h
+ ^ yr}ja
& ^
/j^Yw] ~ hd`Whyb! ab)Yw` ~ rWVa^-El#m# byr]jn@ s
+ ^ aB*
2 Chr
32:1
.wyl*a@ <u*qb
= l
! = rm#aY{w~ torx%Bh
= ^ <yr]uh
* -# lu^
Ezra 7:1
Esth 2:1
Esth 3:1
2 Chr 32:1
At first glance, there appears to be no difference between these examples and the
preceding ones with yh!yw+ .~ The issue of the occurrence of examples like these without
yh!yw+ ~ will be discussed in Chapter 10.
9.3.1.2
@ * <yr]bD
* h
+ ^
the following examples, a nominal or verbal element takes the place of the hL#ah
287
these things. Notice that the verbal element immediately following the temporal
expression in these examples is a WAYYIQTOL.
9.3.1.2.1
Gen
25:11
hp*GM
} h
^ ^ yr}ja
& ^ yh!yw+ ~
Num
25:1926:1
Josh 1:1
288
After the death of Joshua the
sons of Israel inquired of the
LORD, saying, Who shall go
up first for us against the
Canaanites, to fight against
them?
Judg 1:1
ql@mu
* h
& -* ta# toKh^m@ bv* dw]dw` +
2 Sam
1:1
.<y]nv
` = <ym!y` gl*qx
= B
! = dw]D` bv#Yw} ~
Et@ur
* -` lK* yr}ja
& ^ yh!yw+ ~
.hw]hy+ yn`da
{ & <a%n+ El* yoa yoa
Ezek
16:23-24
Pc-vqw3msXa Pd ncmsc-ncfscX2fs
Pi Pi PpX2fs ncmsc np np Pcvqw2fs-PpX2fs ncms Pc-vqw2fsPpX2fs ncfs Pp-ncmsc-ncfs
9.3.1.2.2
After: yr}ja
& ^ yh!yw+ ~ Followed by a VERB
The difference in the following set of occurrences is that the temporal expression
is composed of yr}ja
& ^ yh!y+w~ + VERB, rather than the nominal element found in the
preceding set.
9.3.1.2.2.1 Infinitive Construct
In this set, the verbal element that is part of the temporal clause is an infinitive
construct: yrja yh!yw+ ~ + INFC. In these occurrences as well, the verbal element following
the temporal expression is a WAYYIQTOL.
289
after they had departed, they
came up out of the well and went
and told King David; and they
said to David, Arise and cross
over the water quickly for thus
Ahithophel has counseled
against you.
<T*kl
= # yr}ja
& ^ yh!yw+ ~
2 Sam
dw]D` El#Ml
# ^ WdG]Yw~ ~ Wkl=Yw} ~ ra@Bh
= m
^ @ Wlu&Yw~ ~
17:21
w{totv= yr}ja
& w^ + <j#l# w{lk=a* yr}ja
& ^ yh!yw+ ~
1 Kgs
13:23
1 Kgs
13:31
rb#QB
# ^ yt!a) <T#rb
+ q
^ W= yt!omB=
Pc-vqw3msXa Pd vqcX3ms
PoX3ms Pc-vqw3ms PpncmpcX3ms Pp-vqc Pp-vqcX1cs
Pc-vqp2mp{2} PoX1cs Pp+Pancms Pr ncms Pa-ncmp vqsms
PpX3ms Pp ncfpcX3ms vhvmp PoncfpcX1cs
.yt*mx
) u
= -^ ta# WjyN]h^ wyt*mx
) u
= ^ lx#a@
2 Chr
25:14
Temporal Expression
1
290
In particular, notice:
1) The coordinate structure of temporal expressions produced by
yr}ja
& w^ + yr}ja
& ^ yh!yw+ ~
2) The parallel use of the infinitive constructs + 3ms pronominal suffix
In addition to these observations of the syntactic components of 1 Kgs 13:23,
even more intriguing is the interplay here between yh!yw+ ~ and w+. This is an important area
for further research.
9.3.1.2.2.2 QATAL
In the following examples, the verbal element of the temporal expression with
yh!yw+ ~ is a QATAL, which is then followed by a WAYYIQTOL.
After they had brought it
around, the hand of the LORD
was against the city with very
great confusion; and He struck
the men of the city, both young
and old, so that tumors broke
out on them.
1 Sam
5:9
lodG`-du^w+ /f)Qm
* ! ryu!h* yv@na
+ -^ ta# EY~w~
.<yl!pu
* ( <h#l* Wrt=CY* w] ~
hL#ah
@ * <yr]bD
* h
+ -^ ta# hw`hy+ rB#D] rj^a^ yh!yw+ ~
boYa!-la#
yn]my* T@h^ zp^yl!a-$ la# hw`hy+ rm#aY{w~
;yu#r} yn}vb
= W! ;b= yP!a^ hr`j*
.boYa! yD]bu
= K
^ = hn`okn+ yl^a@ <T#rB
+ d
^ ] aO yK!
Job 42:7
291
The following features of the context need to be considered:
1 Sam 5:9 In 1 Sam 5:8 the event of moving the ark was already narrated
by WBS@Yw~ .~ WBs^h@ resumes the narrative by stepping back and picking up the
previous event. Since this is not normal narrative succession, the QATAL is
used.
Job 42:7 The temporal expression in Job 42:7 does not occur in normal
narrative succession with what precedes it. rB#D] rj^a^ yh!yw+ ~ encompasses
more than just the previous clause. The Lord has not just spoken. In fact,
Job is the previous speaker which is further evidence that rB#D] refers back
farther than the typical conversational interchange.
With /k@-yr\
\ja^
& ^
a
yrj&
9.3.1.3
Followed by WAYYIQTOL
In the following set of examples, the verbal element which immediately follows
/k@-yr\ja
& ^ is a WAYYIQTOL.
After this he loved a woman in
the valley of Sorek, whose name
was Delilah.
/k@-yr}ja
& ^ yh!yw+ ~
Judg 16:4
/k@-yr}ja
& ^ yh!yw+ ~
w{ta) dw]D-` bl@ EY~w~
.lWav*l= rv#a& [n`K-* ta# tr~K* rv#a& lu^
1 Sam
24:6
292
After that David inquired of the
LORD, saying, Shall I go up to
one of the cities of Judah? And
the LORD said to him, Go up.
So David said, Where shall I
go up? And He said, To
Hebron.
/k@-yr}ja
& ^ yh!yw+ ~
rm)al@ hw`hyB^ dw]D` la^vY= w] ~
2 Sam
2:1
/k@-yr}ja
& ^ yh!yw+ ~
<u@yn]kY= w~ ~ <yT!vl
= P
! -= ta# dw]D` EY~w~
2 Sam
8:1
.<yT!vl
= P
! = dY~m! hM*ah
^ * gt#m-# ta# dw]D` jQ^Yw] ~
/k@-yr}ja
& ^ yh!yw+ ~
/k@ yr}ja
& m
^ @ yh!yw+ ~
<ys!sw% + hb*Kr
* m
+ # <olv*ba
= ^ w{l cu^Yw~ ~
Pc-vqw3msXa Pp-Pd Pd Pcvqw3msXa PpX3ms np ncfs Pcncmp Pc-abp ncms vqPmp PpncbpcX3ms
15:1
.wyn`pl
* = <yx!r` vya! <yV!m!jw& ~
2 Sam
<yT!vl
= P
! -= <u! bogB= hm*jl
* =Mh
! ^ dou-yh!Tw= ~
21:18
[s^-ta# yt!vj
* h
% ^ yk^Bs
= ! hK*h! za*
.hp*rh
` * yd}ly! B! rv#a&
/k@-yr}ja
& ^ yh!yw+ ~
Whn}jm
& -^ lK*-ta# <r`a-& El#m# dd~h-& /B# JB)qY= w] ~
2 Sam
/k@-yr}ja
& ^ yh!yw+ ~
10:1
.wyT*jT
= ^ w{nB= /Wnj* EOm=Yw] ~
2 Sam
2 Kgs
6:24
293
After this David defeated the
Philistines and subdued them
and took Gath and its towns
from the hand of the Philistines.
Pc-vqw3msXa Pd-Pd Pc-vhw3ms
np Po-np Pc-vhw3msX3mp Pcvqw3ms Po-np Pc-ncfpcX3fs Ppncfsc np
/k@-yr}ja
& ^ yh!yw+ ~
<yT!vl
= P
! -= ta# dyw]D` EY~w~
1 Chr
18:1
.<yT!vl
= P
! = dY~m! h*yt#nb
{ W= tG~-ta# jQ^Yw] ~ <u@yn]kY= w~ ~
/k@-yr}ja
& ^ yh!yw+ ~
/oMu^-yn}B= El#m# vj*n` tm*Y`w~
1 Chr
19:1
.wyT*jT
= ^ w{nB= EOm=Yw] ~
/k@yr}ja
& ^ yh!yw+ ~
<yT!vl
= P
! -= <u! rz\gB
\ = hm*jl
* =m! dm)uT
& w^ ~
1 Chr
20:4
1 Sam 24:6
2 Sam 2:1
2 Sam 8:1
2 Sam 10:1
2 Sam 15:1
2 Sam 21:18
2 Kgs 6:24
1 Chr 18:1
1 Chr 19:1
1 Chr 20:4
294
After the capture of Rabbah
Followed by QATAL
In only two instances, the verbal element which immediately follows /k@-yr\ja
& ^ is
a QATAL.
after this the sons of Moab and
the sons of Ammon, together
with some of the Meunites, came
to make war against
Jehoshaphat.
2 Chr
.hm*jl
* M
= l
! ^ fp*vo* hy+-lu^ <yn]oMu^hm
* @ <h#Mu
* w! +
20:1
2 Chr 24:4
2 Chr
24:4
295
In 2 Chr 20:1, the QATAL WaB* indicates the anteriority of the kings coming
together: they had come together and then the report was sent.
Immediately prior to the occurrence in 2 Chr 24:4, 24:1-3 is an introductory
paragraph. The QATAL in 24:4 indicates that it had been on Joashs mind to restore the
house of the Lord, then he called them together.
In one additional case, 2 Sam 13:1, the expression with /k@-yr\ja
& ^ comes after
David captured Rabbah. The temporal expression indicates a move forward to a new
temporal setting. This is then followed by a w+ + complex NOUN PHRASE which provides
important setting information for the following narrative. It is difficult to reflect this well
in English, as seen in the NASB below:
Now it was after this that
Absalom the son of David had a
beautiful sister whose name was
Tamar, and Amnon the son of
David loved her.
/k@-yr}ja
& ^ yh!yw+ ~
rm*T* Hm*vW= hp*y` toja* dw]D-` /B# <olv*ba
= l
^ W=
2 Sam
13:1
Absalom had a beautiful sister reads as if this is the next event following the
passage of time, but the point is the introduction of a new participant into the narrative.
This awkwardness was likely a decisive factor in the NLTs rendering, Davids son
Absalom had a beautiful sister. The NLT properly introduces the sister, but
unfortunately the temporal link with the preceding narrative is lost. The precise amount
of time that has transpired is not in focus, but it is important to maintain reference to the
temporal transition as well as introduce the new participant into the narrative.
296
9.3.1.4
as the base. The role of yh!yw+ ~ with rja can only be discerned if it is seen in relief with the
other expressions. For example,
Afterward she bore a daughter and
named her Dinah.
wc*u@ bq@uB
& ^ tz\ja
# ) w{dy`w+ wyj!a* ax*y` /k@-yr}ja
& w^ +
Gen
30:21
Gen
25:26
.<t*a) td\lB
# = hn`v* <yV!v!-/B#
297
there is a close nexus between the spatial and temporal. As BHRG comments regarding
B=, it indicates a time frame in which an event or state of affairs needs to be positioned
(van der Merwe, Naud, Kroeze 1999, 281). The first occurrence below, Gen 4:8,
provides a clear example:
Gen 4:8a
hd\CB
* ^
<t*oyh=B!
in the field
in their being
yh!yw+ ~
(i.e. while they were in the field)
298
With B=
9.3.2.1
The temporal occurrences of the preposition B= with yh!yw+ ~ are divided into two main
sections:
9.3.2.1.1 Occurrences of Infinitive Construct INFC-B= + yh!yw+ ~ (pages 293-327)
9.3.2.1.2 Occurrences of Specific Temporal Reference -B= + yh!yw+ ~ (pages 328-61)
9.3.2.1.1
There are three sub-sections here, separating the occurrences on the basis of the
syntax of what follows the dependent clause with yh!yw+ ~ and the infinitive construct. In the
first set of examples, the infinitive construct is followed directly by the next verb which is
a WAYYIQTOL. In the second set, the verb following the infinitive construct is also a
WAYYIQTOL,
but with the important difference that there are other clauses which
intervene. The presence of these intervening clauses requires careful analysis to ensure
the proper connection of the dependent temporal clause with its immediate context. The
third section has one occurrence where the next verb after the dependent clause is a
QATAL.
It is also significant to observe that two SUBJECT patterns obtain between the
temporal clause and the main clause. The temporal clause and the main clause may have
the same SUBJECT or they may be different. Genesis 11:2 is an example of SAME SUBJECT
and Num 10:35 of DIFFERENT SUBJECT.
When they journeyed east, they
found a plain in the land of
Shinar
<d\Qm
# ! <u*sn= B
` = yh!yw+ ~
ru*nv
+ ! Jr\aB
# = hu*qb
= ! Wax=mY= w] ~
Gen 11:2
299
When the ark set out, Moses
said
/r{ah
* * u~sn) B
+ ! yh!yw+ ~
Num
hv#m) rm#aY{w~
10:35
The significance of being attentive to whether the subject changes or remains the
same is that it helps anticipate certain components of the syntactic structure and the
typical patterns of participant reference. For example, Gen 19:29 below follows the
pattern of same subject, but the explicit subject <yh!Oa$ is repeated in both the temporal
and the main clause. The repetition of the subject is not syntactically required in the
main clause, but it eliminates any possible ambiguity in the third person participant
reference. Lot and Abraham have been the main third person participants in this part of
the narrative. Now, all of a sudden, <yh!Oa$ is introduced as the subject, so the repetition
makes it clear that <yh!Oa$, not Lot, remembered Abraham.
In the following examples, there are certain syntactic features that require further
comment, but this will follow the display of the data.
9.3.2.1.1.1 Infinitive Construct Followed by WAYYIQTOL
The following occurrences of the yh!yw+ ~ + -B= + INFC are followed directly by a
WAYYIQTOL.
hd\CB
* ^ <t*oyh=B! yh!yw+ ~
Gen 4:8
.Whg}rh
+ Y^ w~ ~ wyj!a* lb#h-# la# /y]q^ <q*Yw` ~
<d\Qm
# ! <u*sn= B
` = yh!yw+ ~
.<v* Wbv=Yw} ~ ru*nv
+ ! Jr\aB
# = hu*qb
= ! Wax=mY= w] ~
Gen 11:2
300
When God destroyed the cities of
the valley, God remembered
Abraham, and sent Lot out of the
midst of the overthrow, when He
overthrew the cities in which Lot
lived.
Pc-vqw3msXa Pp-vpc ncmp Poncfpc Pa-ncfs Pc-vqw3ms ncmp Ponp Pc-vpw3ms Po-np Pp-ncmsc Pancfs Pp-vqc Po-Pa-ncfp Pr-vqp3ms
PpX3fp np
rK*Kh
! ^ yr}u-* ta# <yh!Oa$ tj@vB
^ = yh!yw+ ~
jL^vy^ w+ ~ <h*rb
` a
= -^ ta# <yh!Oa$ rK)zY+ w] ~
Gen
19:29
<yr]uh
* -# ta# Ep)hB
& ^ hk*ph
@ &h^ EoTm! fol-ta#
.fol /h@B* bv^y-` rv#a&
HT*dl
+ B
! = Ht*vq
) h
= b
^ = yh!yw+ ~
td\Ly# m
~ h
= ^ Hl* rm#aT)w~
Gen
35:17
Gen
35:18
awh!h^ Jr\aB
* * la@rc
` y= ] /K)vB
= ! yh!yw+ ~
wyb!a* vg\ly# P! hh*lB
= -! ta# bK^vY= w] ~ /b@War+ El#Yw} ~
Gen
35:22
td\Ly# m
~ h
= ^ jQ^Tw! ~ dy`-/T#Yw] ~ HT*dl
+ b
! = yh!yw+ ~
.hn`va
) r] ax*y` hz\ rm)al@ yn]v* w{dy`-lu^ rv)qT
= w! ~
Gen
38:28
301
When the ark set out, Moses
said,Rise up, O LORD! And let
Your enemies be scattered, And
let those who hate You flee
before You.
/r{ah
* * u~sn) B
+ ! yh!yw+ ~
Num
hv#m) rm#aY{w~
10:35
;yb#ya
+ ) Wxp%yw` + hw`hy+ hm*Wq
Pc-vqw3msXa Pp-vqc Pa-ncbs Pcvqw3ms np vqvmsXaXh np Pcvqi3mpXa{1}Jm vqPmpcX2ms Pcvqi3mp{1}Jm vpPmpcX2ms PpncbpcX2ms
.;yn\Pm
* ! ;ya#nc
+ m
^ = Wsn|yw` +
/r{ha
& -^ lu^w+ hv#m-) lu^ hd`uh
@ * lh@Qh
* B
! = yh!yw+ ~
Num
/n`uh
* # WhS*k! hN}hw! + du@om lh#a-) la# Wnp=Yw] ~
17:7
(16:42)
Josh 5:13
Ha*obB= yh!yw+ ~
hd\c* h*yb!a-* ta@m@ loav=l! Wht@ys!Tw= ~
romj&h^ lu^m@ jn~xT
= w! ~
.EL*-hm^ bl@K* Hl*-rm#aY{w~
Josh
15:18
302
When she came to him, she
persuaded him to ask her father
for a field. Then she alighted
from her donkey, and Caleb said
to her, What do you want?
Pc-vqw3msXa Pp-vqcX3fs Pcvhw3fsX3ms Pp-vqc Pp-PpncmscX3fs Pa-ncms Pc-vqw3fs PpPp Pa-ncms Pc-vqw3ms-PpX3fs np
pii-PpX2fs
Ha*obB= yh!yw+ ~
Judg 1:14
hd\Ch
* ^ h*yb!a-* ta@m@ loav=l! Wht@ys!Tw= ~
romj&h^ lu^m@ jn~xT
= w! ~
.EL*-hm^ bl@K* Hl*-rm#aY{w~
w{aobB= yh!yw+ ~
Judg 3:27
<y]rp
` a
= # rh^B= rp*oVB^ uq^t=Yw] ~
hm*ym
+ V
^ h
* ^ j~Bz@ M
+ h
! ^ lu^m@ bh^Lh
^ ^ tolu&b^ yh!yw+ ~
j~Bz@ M
+ h
! ^ bh^lB
^ = hw`hy+-Ea^lm
= ^ lu^Yw~ ~
Judg
13:20
.hx*ra
+ * <h#yn}P-= lu^ WlP=Y]w~ <ya!r{ w{Tv=aw! + j~onm*W
Pc-vqw3msXa Pp-vqc Pa-ncms PpPp Pa-ncms Pa-ncmpXd Pcvqw3msXa ncmsc-np Pp-ncms Pancms Pc-np Pc-ncfscX3ms vqPmp
Pc-vqw3mp Pp-ncmpcX3mp
ncfsXd
1 Sam
16:6
yT!vl
= P
! h
= -^ ta# toKh^m@ dw]D` bWvB= <a*obB= yh!yw+ ~
rovl* la@rc
` y= ] yr}u-* lK*m! <yv!Nh
` ^ hn`ax#Tw@ ~
tar~ql
= ! tolj)Mh
= w^ +
.<yv!lv
! b
* W= hj*mc
= B
! = <yP!tB
% = El#Mh
# ^ lWav*
1 Sam
18:6
303
Now Jonathan, Saul's son, had a
son crippled in his feet. He was
five years old when the report of
Saul and Jonathan came from
Jezreel, and his nurse took him
up and fled. When she hurried to
flee, he fell and became lame.
And his name was
Mephibosheth.
<y]lg* r
+ ~ hk@n+ /B@ lWav*-/B# /t*no` hyl!w+
lWav* tu^mv
% = ab)B= hy`h* <yn]v* vm@j-* /B#
2 Sam
4:4
sn{Tw* ~ w{Tn+ma
^ ) Wha@CT
* w! ~ lau#rz+ Y+ m
] ! /t*no` hyw]
sWnl* Hz`pj
= B
* = yh!yw+ ~
2 Sam
11:16
Pc-vqw3msXa Pp-vqc np Po-np Ppvqc np ncms Pa-ncbs Pp-vpc Po-Pancmp Pc-vhw3ms ncmsc-ncms Ppnp Pp amsc ncmp vqp3ms-Pd np
Pc-ncmsc-np Pp-vhp3ms ncmscncms Pp-np Pc-vqw3ms np pi3ms
Pc-ncmp np Pp-ncmpc ncmscX3ms
PpX3ms Pp-vqc np Pc-np ncms ams
<yl!lj
* h
& -^ ta# rB@ql
^ = ab*X*h^ rc^ ba*oy tolu&B^
.<oda$B# rk*z-` lK* EY~w~
ba*oy <v*-bv^y` <yv!dj
` ( tv#v@ yK!
.<oda$B# rk*z-` lK* tyr]k=h-! du^ la@rc
` y= -] lk*w+
yd}bu
= m
^ @ <yY]md
! a
{ & <yv!na
` w& ~ aWh dd~a& jr~bY= w] ~
./f*q* ru^n~ dd~hw& ~ <y]rx
` m
= ! aobl* w{Ta! wyba
! *
1 Kgs
11:15-17
304
So Elijah went to show himself
to Ahab. Now the famine was
severe in Samaria. Ahab called
Obadiah who was over the
household. (Now Obadiah
feared the LORD greatly; when
Jezebel destroyed the prophets
of the LORD, Obadiah took a
hundred prophets and hid them
by fifties in a cave, and provided
them with bread and water.)
Pc-vqw3msXa np Pp-vnc Pp-np PcPa-ncms ams Pp-np Pc-vqw3ms np
Pp-np Pr Pp-Pa-ncms Pc-np vqp3ms
ams Po-np Pd Pc-vqw3msXa Ppvhc np Po ncmpc np Pc-vqw3ms np
afs ncmp Pc-vhw3msX3mp abp
ncms Pp+Pa-ncfs Pcvdp3msX3mp{2} ncms Pc-ncmp
ba*ja
= -^ la# toar`hl
@ = WhY`la
! @ El#Yw} ~
1 Kgs
./orm=vB
) = qz`j* bu*rh
` w* +
18:2-4
ty]Bh
* -^ lu^ rv#a& Why`db
+ u
^ -) la# ba*ja
= ^ ar`qY= w] ~
.da)m= hw`hy+-ta# ar}y` hy`h* Why`db
+ u
^ w) +
hw`hy+ ya@yb!n+ ta@ lb#zy\ a! tyr]kh
= B
^ = yh!yw+ ~
<ya!bn! + ha*m@ Why`db
+ u
^ ) jQ^Y]w~
hr`uM
* B
= ^ vya! <yV!mj
! & <a@yB!jY= w~ ~
.<y]mw* ` <j#l# <l*Kl
= k
= w! +
ayb!Nh
` ^ WhY`la
! @ vG~Yw] ~ hj*nM
+ h
! ^ tolu&B^ yh!yw+ ~
1 Kgs
qj*xy= ] <h*rb
` a
= ^ yh@Oa$ hw`hy+ rm^aY{w~
18:36
hr`us
* B
= ^ WhY`la
! -@ ta# hw`hy+ tolu&hB
^ = yh!yw+ ~
2 Kgs 2:1
.lG`lG= h
] -^ /m uv*yl!aw$ \ WhY`la
! @ El#Yw} ~ <y]mV
* h
* ^
<Y]wl
] h
= -^ ta# <yh!Oa$h* rz{u=B# yh!yw+ ~
1 Chr
15:26
.<yl!ya@ hu*bv
= w! + <yr]p-* hu*bv
= !
305
When the command and decree
of the king were heard and many
young ladies were gathered to
the citadel of Susa into the
custody of Hegai, Esther was
taken to the king's palace into
the custody of Hegai, who was in
charge of the women.
Jb@Qh
* b
! W= w{td`w+ El#Mh
# -^ rb^D+ um^Vh
* B
! = yh!yw+ ~
Esth 2:8
Jr\ah
* -* la# lb#B-* El#m# rX^ar\dk
+ W^ bn+ tolu&B^ yh!yw+ ~
Jer 35:11
<yD]cK
= h
^ ^ lyj@ yn}Pm
= ! <]lv
^ W* ry+ aobn`w+ WaB) rm#aN{w~
.<*l
! v*WryB! bv#Nw} ~ <r`a& lyj@ yn}Pm
= W!
rm)al@ <yD]Bh
^ -^ vb%l= vya!h-* ta# w{tW{xB
^ = yh!yw+ ~
Ezek
10:6
Dan 8:15
306
As he was talking with him, the
king said to him, Have we
appointed you a royal
counselor? Stop! Why should
you be struck down? Then the
prophet stopped and said, I
know that God has planned to
destroy you, because you have
done this and have not listened
to my counsel.
2 Chr
25:16
Pc-vqw3msXa Pp-vpcX3ms
PpX3ms Pc-vqw3ms PpX3ms PaPp-vqPms Pp+Pa-ncms
vqp1cpX2ms vqvms-PpX2ms Pg
vhi3mpX2ms Pc-vqw3ms Pa-ncms
Pc-vqw3ms vqp1cs Pp-vqp3ms
ncmp Pp-vhcX2ms Pp-vqp2ms afs
Pc-Pn vqp2ms Pp-ncfscX1cs
Three occurrences merit special comment because of the syntactic features of the
dependent clause with yh!yw+ .~ Even though these occurrences appear to have intervening
clauses, they are, in fact, examples of complex temporal expressions.
1 Sam 18:6 Immediately following the dependent clause formed by
INFC-B= yh!yw+ ,~ there is another occurrence of INFC-B=. This second temporal
clause provides further specification for the first. Both occurrences of the
INFC-B= are governed by the same yh!yw+ .~
1 Kgs 11:15 This occurrence is very similar to the preceding example.
Both occurrences of the INFC-B= are governed by the initial yh!yw+ .~ The
second dependent clause provides further information about the setting in
which the event of the next main verb, EY~w,~ takes place.
Esth 2:8 The third example is similar, but adds another element. The
second INFC-B= is prefixed by w+, which underscores the coordinate nature of
the two events expressed by the infinitive constructs. The w+ also
introduces an element of sequentiality.
In all of the preceding examples, the INFC-B= yh!yw+ ~ establishes the temporal frame
within which the action of the WAYYIQTOL occurs.
307
9.3.2.1.1.2 Infinitive Construct Followed by WAYYIQTOL with Intervening Clauses
In the following set of examples, close reading of the syntactic shape of the
surrounding context is imperative. This set is very closely related to the previous one,
since the main verb which follows the dependent clause is a WAYYIQTOL, but in each case
there is at least one intervening clause. These clauses need to be analyzed in order to
determine the proper connection of the dependent clause formed with yh!yw+ ~ and the
independent clause to which it is syntactically joined. Further discussion follows the
display of the data.
When Pharaoh let the people go,
God did not lead them by the
way of the land of the
Philistines, even though it was
near; for God said, The people
might change their minds when
they see war, and return to
Egypt. Hence God led the
people around by the way of the
wilderness to the Red Sea; and
the sons of Israel went up in
martial array from the land of
Egypt.
Pc-vqw3msXa Pp-vpc np Po-Pancms Pc-Pn-vqp3msX3mp ncmp
ncbs ncbs np Pp ams pi3ms Pp
vqp3ms ncmp Pc-vni3ms Pa-ncms
Pp-vqcX3mp ncfs Pc-vqp3cp{2}
npXd Pc-vhw3ms ncmp Po-Pancms ncbs Pa-ncms ncmsc-ncms Pcvqsmp vqp3cp ncmpc-np Pp-ncbs
np
Exod
13:17-18
308
When Moses came down from
Mount Sinaiwith the two
tablets of the testimony in his
hands as he was coming down
from the mountain), he didnt
know that the skin of his face
shone because of his speaking
with Him. So when Aaron and
all the sons of Israel saw Moses,
behold, the skin of his face
shone, and they were afraid to
come near him.
td|uh
@ * tj)l% yn}vW= yn~ys! rh^m@ hv#m) td\rB
\ = yh!yw+ ~
ud~y-` aO hv#mW) rh*h-* /m! w{Td+rB
] = hv#m-) dy~B=
Exod
34:29-30
.w{Ta! w{rB=dB
~ = wyn`P* rou /r~q* yK!
/D}rY+ h
~ -^ ta# rb)ul
& ^ <h#yl@h(am
* @ <u*h* u~sn) B
+ ! yh!yw+ ~
.<u*h* yn}pl
= ! tyr]Bh
= ^ /ora*h* ya@cn= { <yn]hK
& h
) w^ +
/D}rY+ h
~ -^ du^ /ora*h* ya@cn= { aobk=W
hx@qB
= ! WlB=fn= ] /ora*h* ya@cn= { <yn]hK
& h
) ^ yl@gr
+ w~ +
Josh
3:14-16a
309
When they fled from before
Israel, while they were at the
descent of Beth-horon, the
LORD threw large stones from
heaven on them as far as
Azekah. More died from the
hailstones than those whom the
sons of Israel killed with the
sword.
Josh
10:11
hq*zu
} -& du^ <y]mV
^ h
* -^ /m!
dr`Bh
* ^ yn}ba
= B
^ = Wtm@-rv#a& <yB!r~ Wtm%Yw` ~
.br\jB
* # la@rc
` y= ] yn}B= Wgr+h* rv#am
& @
1 Sam
23:6-7
.j~yr]bW= <y]tl
^ D
* + ryu!B= aobl* rG~sn= ]
310
Now there was a man in Maon
whose business was in Carmel;
and the man was very rich, and
he had three thousand sheep and
a thousand goats. While he was
shearing his sheep in Carmel
(now the man's name was Nabal,
and his wife's name was Abigail.
And the woman was intelligent
and beautiful in appearance, but
the man was harsh and evil in
his dealings, and he was a
Calebite), David heard in the
wilderness that Nabal was
shearing his sheep.
lm#rK
+ b
^ ^ Whc@um
& W^ /oum*B= vya!w+
1 Sam
<yp!la
* -& tv#Ov= /ax) w{lw+ da)m= lodG` vya!hw* +
25:2-4
<yZ]u! [l#aw# +
.lm#rK
+ B
^ ^ w{nax)-ta# zz{gB
+ ! yh!yw+ ~
ly]gb
` a
! & w{Tv=a! <v@w+ lb*n` vya!h* <v@w+
ra^T) tp^yw] lk#c-# tb^of hV*ah
! w* +
.wB!lk
! * aWhw+ <yl!lu
* m
& ^ ur~w+ hv#q* vya!hw* +
.w{nax)-ta# lb*n` zz}g-{ yK! rB*dM
+ B
! ^ dw]D` um^vY= w] ~
yv!yl!Vh
= ^ <oYB^ gl^qx
= ! wyv*na
` w& ~ dw]d` ab)B= yh!yw+ ~
1 Sam
gl^qx
= -! la#w+ bg\n-\ la# Wfv=p* yq!lm
@ u
* w& ~
30:1-2
hm*jl
* M
= h
! ^ toyh=B! yh!yw+ ~
dw]D` tyB@ /yb@W lWav* tyB@ /yB@
2 Sam
3:6
311
It happened that when the
priests came from the holy
place, the cloud filled the house
of the LORD, so that the priests
could not stand to minister
because of the cloud, for the
glory of the LORD filled the
house of the LORD. Then
Solomon said,
The LORD has said that He
would dwell in the thick cloud. I
have surely built You a lofty
house, A place for Your dwelling
forever. Then the king faced
about and blessed all the
assembly of Israel, while all the
assembly of Israel was standing.
vd\Qh
) -^ /m! <yn]hK
& h
) ^ tax@B= yh!yw+ ~
.hw`hy+ tyB@-ta# al@m* /n`uh
* w# +
1 Kgs
8:10-14
/n`uh
* # yn}Pm
= ! tr}vl
* = dm)ul
& ^ <yn]hK
& h
) ^ Wlk=y-` aOw+
.hw`hy+ tyB@-ta# hw`hy+-dobk= al@m-* yK!
.lp#ru
` B
& * /K)vl
= ! rm^a* hw`hy+ hm)Ov= rm^a* za*
El* lb%z+ tyB@ yt!yn]b* hn{B*
.<ym!lo* u ;T=bv
= l
! = /okm*
1 Kgs
w{l rya!vh
= -! aO av*uB
= ^ tyB@-lK*-ta# hK*h!
16:11
Esth 3:4
312
I looked in the vision, and while
I was looking I was in the citadel
of Susa, which is in the province
of Elam; and I looked in the
vision and I myself was beside
the Ulai Canal.
Pc-vqw1cs Pp+Pa-ncms Pcvqw3msXa Pp-vqcX1cs Pc-pi1cs
Pp-np Pa-ncfs Pr Pp-np Pa-ncfs Pcvqw1cs Pp+Pa-ncms Pc-pi1cs
vqp1cs Pp-ncmsc np
/ozj*B# ha#ra
+ w# `
Dan 8:2
yt!ar
) B
+ ! yh!yw+ ~
hn`yd]Mh
= ^ <l*yu@B= rv#a& hr`yB!h^ /v^WvB= yn]aw& ~
vd\Qh
) -^ /m! <yn]hK
& h
) ^ tax@B= yh!yw+ ~
/ya@ WvD`qt
^ h
= ! <ya!xm
= N= h
] ^ <yn]hK
& h
) -^ lK* yK!
.toql=jm
= l
^ = romv=l!
/m*yh@l= [s*al
* = <L*kl
% = <yr]rv
& m
) h
= ^ <Y]wl
] h
= w^ +
<yv!Bl
* m
% = <h#yj@al
& w^ + <h#yn}bl
= w! + /Wtd|yl!
<yd]mu
= ) torN{kw! + <yl!bn* b
+ W! <y]Tl
^ x
= m
! B
= ! JWB
ha*ml
@ = <yn]hK
& ) <h#Mu
* w! + j~Bz@ M
+ l
! ^ jr`zm
+ !
.torx=xj
) B
& ^ <yr]rx
] j
= =m^ <yr]cu
= w# +
u~ym!vh
= l
^ = <yr]rv
& ) ml
= w^ + <yr]xx
= j
) M
& l
^ ^ dj*ak
# = yh!yw+ ~
todh)lW= lL@hl
^ = dj*a-# loq
<y]Tl
^ x
= m
! b
= W! torx=xj
) B
& ^ loq <yr]hk
* W= hw`hyl^
bof yK! hw`hyl^ lL@hb
^ W= ryV!h^ yl@kb
= W!
.hw`hy+ tyB@ /n`u* al@m* ty]Bh
^ w^ + w{Ds=j^ <l*oul= yK!
2 Chr
5:11-13
313
The men of Judah raised a war
cry. When the men of Judah
raised the war cry, God routed
Jeroboam and all Israel before
Abijah and Judah. The sons of
Israel fled before Judah, and
God gave them into their hand.
Pc-vhw3mp ncms np Pc-vqw3msXa
Pp-vhc ncms np Pc-Pa-ncmp
vqp3ms Po-np Pc-ncmsc-np Ppncbpc np Pc-np Pc-vqw3mp
ncmpc-np Pp-ncbpc np Pcvqw3msX3mp ncmp Pp-ncfscX3mp
2 Chr
13:15-16
<u*br
= y` -` ta# [g~n` <yh!Oa$h*w+
.hd`Whyw] hY`ba
! & yn}pl
= ! la@rc
` y= -] lk*w+
hd`Why+ yn}Pm
= ! la@rc
` y= -] yn}b= WsWnY`w~
.<d`yB
` = <yh!Oa$ <n}TY= w] ~
One of the main issues in all of these examples is the status of the WE-al-QATAL
and WE-X-QATAL. Additionally, there are some relatively surprising cases which require
discussion. The main syntactic features are presented below:
Exod 13:17-18 The question here is whether the WE-al-QATAL qualifies
as the main verb of an independent clause. This depends on the analysis
of verbs which are said to interrupt the sequences or chains of
WAYYIQTOL in narrative. Faced with Exod 13:17 <yh!Oa$ <j*n` aOw+, this is
no longer a theoretical matter. In order to properly read or translate this
verse, the relationship of the WE-al-QATAL to the preceding yh!yw+ ~ must be
understood. One of the issues here is the pattern of negation in narrative.
One of the first considerations is that a negated WAYYIQTOL is
morphosyntactically impossible. Since there is no way to negate a
WAYYIQTOL, WE-al-QATAL clauses in this type of syntactic relationship
are considered to be the negated counterpart of the WAYYIQTOL clause in
narrative. Further examples of WE-aO-QATAL clauses are found in Gen
26:22; 27:23; 30:40; 31:33, 34, 35; 34:19; 35:5; 37:4; 39:6, 10; 40:23; and
45:3.
Exod 34:29-30 In Exod 34:29-30, there is a case of w+ + NOUN followed
by B= + infinitive construct, then a WE-X-al-QATAL; Again, the issue here
is what the relationship is between the yh!yw+ ~ at the beginning of 34:29 and
the two w+ + nominal structures which follow. In the NASB, for example,
ud~y-` aO hv#mW) is translated as the main verb.
Josh 10:11 In Josh 10:11, there is an explicit pronoun <h@, then WE-XQATAL. The use of the pronoun <h@ in this way allows the statement of
further information regarding the same subject of the preceding verb. In
this case, <h@ agrees with the pronominal suffix on the infinitive construct
314
following yh!yw+ ,~ allowing further specification. The other intervening
clause begins with WE-X-QATAL, raising questions as to its status as main
verb and relationship to the initial yh!yw+ .~ Many describe this type of clause
as disjunctive or as background information. The NASB, however, has the
LORD threw heavy stones from heaven giving this WE-X-QATAL the status
of the next main verb. In fact, the type of translation in the NASB is
exactly what would be expected if the Hebrew verb were a WAYYIQTOL:
El@vY= w~ .~ The goal is not to translate so that the underlying Hebrew word
order comes through, but is there no difference in narrative between a
WAYYIQTOL clause and one with WE-X-QATAL? Further discussion of this
matter is found in Chapter 10.
1 Sam 23:6-7 The occurrence of X-QATAL rather than WE-X-QATAL is
quite unusual because the noun which precedes the QATAL is the object of
the verb, whereas in the typical occurrence of WE-X-QATAL, the noun is the
subject. In 1 Sam 23:6-7, the subject of the X-QATAL clause is the same as
that of the preceding dependent clause. If the options for subject and
object are considered, to simply maintain the same subject in this clause,
there would be no need for it to be WE-X-QATAL starting with rt*yb
` a
= w# ,+
unless there were some particular reason to call attention to Abiathar.
Similarly, to start this clause with dopa@w+ would shift the ephod into a
different role, most likely making it the subject of the clause. In the clause
w{dy`B= dr~y` dopa@, the subject is elided and dopa@ before the verb brings it
into heightened prominence.
1 Sam 30:1-2 What connection does the WE-X-QATAL have to the
temporal expression with yh!yw+ ?~ The temporal sequencing of the events in
1 Sam 30:1-2 requires careful analysis. First of all, in 29:11, David and
his men rose early to set out on their journey, and according to 30:1 they
arrive in Ziklag three days later. At first glance, the clause in 30:3a,
ryu!h-* la# wyv*na
` w& ~ dw]d` ab)Yw` ,~ appears to restate their arrival. The
question, then, has to do with the temporal relationship of their arrival and
the Amalekites raid. A second question has to do with the temporal
relationship of the QATAL in Wfv=p* yq!lm
@ u
* w& ~ to the following series of
WAYYIQTOL verbs. A common sense approach to how these
WAYYIQTOL verbs should be translated would say that, on the basis of
30:3, it is obvious that the raid had taken place before Davids arrival.
This may result in a translation that makes sense, but does not help explain
the pattern of usage of the verb forms here.
2 Sam 3:6 The WE-X-QATAL in 2 Sam 3:6 states what was happening
while the war was taking place between Saul and David. The concurrent
nature is reinforced by both the use of -B= yh!yw+ ~ and the WE-X-QATAL. In
315
addition to the concurrent temporal nature, the WE-X-QATAL also gives a
nuance of the narrative parity of the war and Abners actions. The WE-XQATAL here is not, however, simply a case of WE-X-QATAL; this WE-XQATAL has the QATAL of hyh + PARTICIPLE, bringing out the ongoing
nature of Abners actions. The setting function of this part of the narrative
is continued in 3:7 with the comments about Sauls concubine named
Rizpah. The narrative proper begins with the WAYYIQTOL rm#aY{w~ in the
middle of 3:7.
1 Kgs 8:10-11 In 8:12, there is WE-X-QATAL, WE-al-QATAL, za, then
WAYYIQTOL in 8:14. A comparison of the NASB and JPS shows two
different strategies for dealing with the WE-X-QATAL in 8:10:
NASB
JPS
The NASB reads as if the WE-X-QATAL were the next event in narrative
succession. The JPS, however, by using em dashes and had filled, renders
the WE-X-QATAL as ANTERIOR. According to the analysis proposed here,
the JPS reading better captures the nuance of the WE-X-QATAL.
1 Kgs 16:11 Note both B= and K=. The B= yh!yw+ ~ governs the temporal frame
within which the temporal expression with K= is embedded.
Esth 3:4 The temporal overlap of the expression with B= yh!yw+ ~ is
reinforced contextually by <oyw` <oy. Following this dependent temporal
expression, the WE-al-QATAL is the nuclear clause. As stated above, the
WE-al-QATAL in this type of syntactic relationship is interpreted as the
next clause in narrative succession.
Dan 8:2 This verbless clause suspends the narrative succession, further
specifying the setting established by the preceding temporal expression
with B=.
2 Chr 5:11-13 Taking the broader context into account is important in
the analysis of this occurrence. After the DEICTIC occurrence of yh!yw+ ~ in
316
5:9, 5:10 continues with a description of the contents of the Ark. 5:7
narrates the event of the priests entering the sanctuary, and the temporal
expression in 5:11 relates when they exit. The temporal expression is
followed in 5:11 by <Y]wl
] =hw^ ,+ which begins the complex noun phrase,
followed by <yd]mu
= .)
2 Chr 13:15 In 2 Chr 13:15, the clause with yh!yw+ ~ restates the event of
the previous clause. This, then, is followed by the WE-X-QATAL under
consideration here, which brings <yh!Oa$ into prominence.
The main issue involved in the analysis of these occurrences is the status of the
different types of verbal clauses in narrative. The WAYYIQTOL, the WE-X-QATAL, and
other clauses need to be analyzed not as autonomous entities, but rather as interactive
components within the same narrative system. Analysis of the role of yh!yw+ ~ in narrative
requires understanding the function of other clauses in the context.
9.3.2.1.1.3 Infinitive Construct Followed by QATAL
In only one occurrence, the verb which directly follows the dependent clause with
yh!yw+ ~ is a QATAL.
When the priests who carried the
ark of the covenant of the LORD
had come up from the middle of
the Jordan, and the soles of the
priests' feet were lifted up to the
dry ground, the waters of the
Jordan returned to their place,
and went over all its banks as
before.
Pc-vqw3msXa Pp-vqc Pp-vqc Pancmp vqPmpc ncbsc ncfs-np Ppncmsc Pa-np vnp3cp ncfp ncfdc Pancmp Pp Pa-ncfs Pc-vqw3mp
ncmpc-Pa-np Pp-ncmscX3mp Pcvqw3mp Pp-Pd-Pd Pp-ncmscncfpcX3ms
hb*rj
` h
* # la# <yn]hK
& h
) ^ yl@g+r~ toPK^ WqT=n]
<m*oqm=l! /D}rY+ h
~ -^ ym@ Wbv%Y`w~
.wyt*odG+-lK*-lu^ <ovl=v-! lomt=k! Wkl=Yw} ~
Josh 4:18
317
Careful reading of the preceding context reveals that the use of the QATAL WqT=n] in
Josh 4:18 is the result of it not being normal narrative succession. The actual narration of
the event of the priests coming up out of the river carrying the ark is in 4:11. In Josh
4:14, the events of that day are reiterated and it is in this context that the QATAL occurs
in 4:18 to express ANTERIORITY.
9.3.2.1.1.4 Summary of Infinitive Constructs Used with -B=
B= yh!yw+ ~
In the preceding examples, the verbs which occur with B= yh!yw+ ~ are: hyh, ucn,
tjv, hvq, dly, dry, axy, /kv, ilv, lhq, hlu, awb, zzg, zpj, rmv, trk, rzu, umv, hwx,
har, rbd. In 9.3.2.2.1.5 below, these verbs will be compared with those which occur
with K= yh!yw+ ~ to explore the parameters that might influence the choice of preposition.
9.3.2.1.2
There are two sets of occurrences in this section, divided into those which are
followed by a WAYYIQTOL and those with a QATAL. There is an additional difference
within each set of temporal expressions with B= yh!yw+ ,~ namely that in the expressions with
aWhh^ or awh!h^ as a component, the temporal reference is backward-looking, establishing
that the next narrated event took place at the time of a previous temporal frame of
reference. In contrast, the temporal expressions such as br\ub
# * yh!yw+ ~ establish a new
temporal point of reference for the narrative which follows.
318
9.3.2.1.2.1 Specific Temporal Reference Followed by WAYYIQTOL
At first glance, there appears to be a wide variety of ways to make temporal
reference following B= yh!yw+ ,~ but analysis shows that there is essentially one pattern, even
though the lexical items used in it vary significantly. The basic pattern is:
+ NOUN PHRASE
MODIFIER
9.3.2.1.2.1.1
9.3.2.1.2.1.1
9.3.2.1.2.1.2
9.3.2.1.2.1.3
9.3.2.1.2.1.4
9.3.2.1.2.1.5
9.3.2.1.2.1.6
9.3.2.1.2.1.7
9.3.2.1.2.1.8
9.3.2.1.2.1.9
9.3.2.1.2.1.10
9.3.2.1.2.1.11
9.3.2.1.2.1.12
9.3.2.1.2.1.12
9.3.2.1.2.1.12
9.3.2.1.2.1.13
9.3.2.1.2.1.14
awh!h^
INFC
aWhh^
aWhh^
<h@h*
<h@h* <yB!rh
~ *
NUMBER-h^
NUMBER-h^
INFC
/s*yn]
______
+ NOUN
rq#B)
br\u#
tu@
tu@
<oY^
hl*yL
+ ^
<ym!Y`
<ym!Y`
<oY
hn`V*
<oy
tyu!bV
! =
hl*yL
+ h
^ ^ yx!j&
rq#Bh
) ^ tr\mv
) a
= ^
<y]rh
~ X
( *
vd\j)
ym@y
+ PREPOSITION
yh!yw+ ~
b=, b*, or b^
In some of the examples, the NOUN PHRASE is realized by nothing more than an
obligatory noun, such as rq#B) or br\u.# In the others, the NOUN PHRASE has other optional
components, with the lengthiest example being 1 Kgs 6:1 with an extended description of
the time period after Israel left Egypt.
319
Since the structure of the temporal expressions considered in 9.3.2.1.2.1.1-14 is so
similar, there will be minimal comments throughout this section of B= yh!yw+ ~ with specific
temporal reference.
) ^ / br\
\ub
# *
9.3.2.1.2.1.1 With rq#Bb
bru#
With yh!yw+ ,~ br\ub
# * occurs twice and rq#Bb
) ^ eight times. The words evening and
morning occur many other times, of course, but only a total of ten times with yh!yw+ .~ The
main function is to specify the period of the day during which the event of the main
clause takes place.
In the evening he took his
daughter Leah, and brought her
to him; and Jacob went in to
her.
Gen
29:23
Pc-vqw3msXa Pp+Pa-ncms Pcvqw3ms Po-np ncfscX3ms Pcvhw3msXa PoX3fs PpX3ms Pcvqw3msXa PpX3fs
Gen
29:25
w{jWr <u#PT
* w! ~ rq#Bb
) ^ yh!yw+ ~
<y]rx
~ m
= ! yM@fr
% j
+ -^ lK*-ta# ar`qY= w] ~ jl^vY= w] ~
hu)rP
+ ^ rP@sy^ w+ ~ h*ym#kj
* -& lK*-ta#w+
.hu)rp
+ l
^ = <t*oa rt@oP-/ya@w+ w{mOj&-ta# <h#l*
Gen 41:8
320
In the evening the quails came
up and covered the camp, and in
the morning there was a layer of
dew around the camp.
Pc-vqw3msXa Pp+Pa-ncms Pcvqw3fsXa Pa-ncfs Pc-vpw3fsXa
Po-Pa-ncbs Pc-Pp+Pa-ncms vqp3fs
ncfsc Pa-ncms Pd Pp+Pa-ncbs
br\ub
# * yh!yw+ ~
Exod
hn\jM
& h
^ -^ ta# sk^Tw= ~ wl*Ch
= ^ lu^Tw^ ~
16:13
.hn\jM
& l
^ ^ byb!s* lF^h^ tb^k=v! ht*yh
+ * rq#Bb
) W^
<u*lB
= -! ta# ql*B* jQ^Yw] ~ rq#Bb
) ^ yh!yw+ ~
Num
22:41
rq#Bb
) ^ yh!yw+ ~
1 Sam
20:35
lb*Nm
` ! /y]Yh
~ ^ tax@B= rq#Bb
) ^ yh!yw+ ~
1 Sam
hL#ah
@ * <yr]bD
* h
+ -^ ta# w{Tv=a! w{l-dG\Tw^ ~
25:37
./b#al
* = hy`h* aWhw+ w{Br+q!B= w{Bl! tm*Yw` ~
rq#Bb
) ^ yh!yw+ ~
2 Sam
11:14
hj*nM
+ h
! ^ tolu&K^ rq#Bb
) ^ yh!y+w~
Jr\ah
* * al@MT
* w! ~ <oda$ Er\Dm
\ ! <ya!B* <y]m-^ hN}hw! +
.<y]Mh
* -^ ta#
2 Kgs
3:20
321
In the morning he went out and
stood and said to all the people,
You are innocent; behold, I
conspired against my master
and killed him, but who killed all
these?
Pc-vqw3msXa Pp+Pa-ncms Pcvqw3ms Pc-vqw3ms Pc-vqw3ms
Pp-ncmsc-Pa-ncms amp pi2mp Pi
pi1cs vqp1cs Pp-ncmscX1cs Pcvqw1csX3ms Pc-pii vhp3ms Poncmsc-acp
rq#Bb
) ^ yh!yw+ ~
rm#aY{w~ dm)uY& w~ ~ ax@Yw} ~
2 Kgs
10:9
yT!rv
+ q
^ * yn]a& hN}h! <T#a^ <yq!Dx
] ^ <u*h-* lK*-la#
.hL#a-@ lK*-ta# hK*h! ym!W Whg}rh
+ a
= w# ` yn]da
{ -& lu^
322
At that time, Abimelech and
Phicol, the commander of his
army, spoke to Abraham, saying,
God is with you in all that you
do;
21:22
Gen
Gen 38:1
1 Kgs
<*l
! v*Wrym! ax*y` <u*br
= y` w` +
11:29
Er\D\B^ ayb!Nh
` ^ yn]OyV!h^ hY`ja
! & w{ta) ax*mY= w] ~
hv*dj
` & hm*lc
= B
^ = hS#Kt
^ m
= ! aWhw+
.hd\CB
* ^ <D`bl
^ = <h#yn}vW=
323
was enjoying the good favor of Solomon. The WE-X-QATAL here
expresses ANTERIORITY relative to the WAYYIQTOL which follows.
9.3.2.1.2.1.3 Infinitive Construct + tu@B= yh!y+yw+ ~
There are three occurrences of tu@B= yh!yw+ ~ with an infinitive construct and each one
has a slightly different form. These variations will be discussed after the display of the
data.
At the time when the flock was
mating, I lifted up my eyes and
saw in a dream, and behold, the
male goats which were mating
were striped, speckled, and
mottled.
Pc-vqw3msXa Pp-ncbs vpc Pa-ncbs
Pc-vqw1cs ncfdcX1cs Pcvqw1csXa Pp+Pa-ncms Pc-Pi Pancmp Pa-vqPmp Pp-Pa-ncbs amp
amp Pc-amp
Gen
31:10
<yl!uh
) * <yd]Tu
% h
^ * hN}hw! +
.<yD]rb
| W= <yD]qn% + <yD]qu
% & /aX)h-^ lu^
HT*dl
+ ! tu@B= yh!yw+ ~
.Hn`fb
= B
! = <ym!oat= hN}hw! +
td\Ly# m
~ h
= ^ jQ^Tw! ~ dy`-/T#Yw] ~ HT*dl
+ b
! = yh!yw+ ~
.hn`va
) r] ax*y` hz\ rm)al@ yn]v* w{dy`-lu^ rv)qT
= w! ~
Gen
38:27-28
324
Whenever the chest was brought
in to the king's officer by the
Levites, and when they saw that
there was much money, then the
king's scribe and the chief
priest's officer would come,
empty the chest, take it, and
return it to its place. Thus they
did daily and collected much
money. The king and Jehoiada
gave it to those who did the work
of the service of the house of the
LORD;
2 Chr
24:1112a
Gen 31:10 The occurrence of /aX)h^ <j@y~ tu@B= yh!yw+ ~ in Gen 31:10 is a
complex temporal expression in which the infinitive construct gives a lessanchored temporal reference than a finite verb form would indicate. This
^ ^ tax@ tu@l= in 2 Sam 11:1 listed in
is similar to the occurrence of <yk!al=Mh
9.3.2.3.2.1 below, where tax@ tu@l= is a general reference to the customary
time when kings go out to battle. If this were a reference to when a
specific king went to battle, then the infinitive construct would not be
used. The difference, however, between tu@l= and tu@B= is consistent with
the usage of l= expressing APPROXIMATION and b= expressing inclusion
within the time specified. Presumably, if Gen 31:10 were <j@y~ tu@l,= it
would express about the time for mating rather than during that time and
likewise, if 2 Sam 11:1 were tax@ tu@B,= the sense would be during the
time when kings go to battle.
Gen 38:27-28 The occurrence of the infinitive construct HT*dl
+ ! following
tu@B= yh!yw+ ~ indicates that at the time she was to give birth, it was discovered
that she had twins. This contrasts with the temporal expression later in the
verse, HT*dl
+ b
! = yh!yw+ ,~ while she was giving birth.
2 Chr 24:11-12a The occurrence of the YIQTOL in the following,
/ora*h-* ta# ayb!y` tu@B= yh!yw+ ,~ is motivated by the aspectual contours of the
narrative where the description is of what happened day after day
325
(<oyB= <oyl= Wcu* hK)) rather than a description of what happened on one
particular day. This is a good example of how aspectual distinctions are
contextual, being expressed by linguistic items other than the verbs.
9.3.2.1.2.1.4 With aWhh^ <oYB^
The expression aWhh^ <oYB^ establishes that the event which precedes the temporal
clause occurred on the same day as the event which follows it. It is remarkable that this
specific expression only occurs twice in the Hebrew Bible.
On the same day, Isaac's
servants came in and told him
about the well which they had
dug, and said to him, We have
found water.
Pc-vqw3msXa Pp+Pa-ncms Papi3ms Pc-vqw3mpXa ncmpc np Pcvhw3mpXa PpX3ms Pp-ncfp Pancfs Pr vqp3cp Pc-vqw3mp
PpX3ms vqp1cp ncmp
Gen
26:32
rv#a& ra@Bh
= ^ toda)-lu^ w{l WdG]Yw~ ~
.<y]m* Wnax*m* w{l Wrm=aY{w~ Wrp*j*
1 Sam
3:2
The occurrence of aWhh^ <oYB^ yh!yw+ ~ in Gen 26:32 follows a sequence of four
WAYYIQTOL
verbs in 26:31, establishing that the arrival of Isaacs servants took place on
the same day. In 1 Sam 3:2, however, the temporal connection is not as immediately
apparent. Looking at the broader context, in 1 Sam 2:27 a man of God comes to Eli and
326
2:27-36 reports the content of his speech to Eli. 2:36 brings the man of Gods speech to a
close and chapter 3 begins as follows:
The boy Samuel was ministering
to the LORD before Eli. And
word from the LORD was rare
in those days, visions were
infrequent.
yl!u@ yn}pl
= ! hw`hy+-ta# tr}v*m= la@Wmv= ru^Nh
~ w^ +
<ym!YB
` ^ rq*y` hy`h* hw`hy+-rb^dW+
1 Sam
3:1
w{tyB@-lu^ ht*mr
* h
` * hn`ql
* a
= # El#Yw} ~
yl!u@ yn}P-= ta# hw`hy+-ta# tr}vm
* = hy`h* ru^Nh
~ w^ +
1 Sam
2:11
./h@Kh
) ^
1 Sam
2:18
1 Sam
2:22
327
The boy Samuel was growing in
stature and in favor both with
the LORD and with men.
1 Sam
2:26
What is the function of these WE-X-QATAL clauses? How do they contribute to the
overall narrative structure? These are important questions to address in the attempt to
understand the nuances and movements of the biblical Hebrew text. Further research is
needed to define the parameters of the interplay between w+ and yh!yw+ .~
9.3.2.1.2.1.5 With aWhh^ hl*yL
+ B
^ ^
Following the same pattern as aWhh^ <oYB^, this expression establishes that the
same night is the temporal setting for the events which it links together.
On the same night the LORD
said to him, Take your father's
bull and a second bull seven
years old, and pull down the
altar of Baal which belongs to
your father, and cut down the
Asherah that is beside it;
Pc-vqw3msXa Pp+Pa-ncms Papi3ms Pc-vqw3ms PpX3ms np
vqvms Po-ncms-Pa-ncms Pr PpncmscX2ms Pc-ncms Pa-ams afs
ncfp Pc-vqp2ms{2} Po-ncmsc Panp Pr Pp-ncmscX2ms Pc-Po-Pa-ncfs
Pr-PpX3ms vqi2ms
aWhh^ hl*yL
+ B
^ ^ yh!yw+ ~
Judg 6:25
aWhh^ hl*yL
+ B
^ ^ yh!yw+ ~
.rm)al@ /t*n-` la# hw`hy+-rb^D+ yh!yw+ ~
2 Sam
7:4
328
That same night the angel of the
LORD went out and struck
185,000 in the camp of the
Assyrians; and when men rose
early in the morning, behold, all
of them were dead.
Pc-vqw3msXa Pp+Pa-ncms Papi3ms Pc-vqw3ms ncmsc np Pcvhw3ms Pp-ncbsc np afs amp Pcams ams Pc-vhw3mp Pp+Pa-ncms
Pc-Pi ncmscX3mp ncmp vqPmp
hw`hy+ Ea^lm
= ^ ax@Yw} ~ aWhh^ hl*yL
+ B
^ ^ yh!yw+ ~
2 Kgs
19:35
rq#Bb
) ^ WmyK!vY= w~ ~ [l#a* hV*mj
! w& ~
.<yt!m@ <yr]gP
` = <L*k% hN}h!w+
aWhh^ hl*yL
+ B
^ ^ yh!yw+ ~
.rm)al@ /t*n-` la# <yh!Oa$-rb^D+ yh!yw+ ~
1 Chr
17:3
<h@h* <ym!YB
` ^ yh!yw+ ~
hv#m) lD~gY+ w] ~
Exod
2:11
Judg 19:1
329
In those days the Philistines
gathered their armed camps for
war, to fight against Israel.
<yT!vl
= p
! = WxB=qY= w] ~ <h@h* <ym!YB
` ^ yh!yw+ ~
la@rc
` y= B
] = <j@Lh
* l
! = ab*Xl
* ^ <h#yn}jm
& -^ ta#
1 Sam
28:1
<h@h* <yB!rh
~ * <ym!Yb
` ^ yh!yw+ ~
<y]rx
~ m
= ! El#m# tm*Yw` ~
Exod
2:23
hd`bu
) h
& -* /m! la@rc
` y= -] yn}b= Wjn+aY* w} ~
<t*uw* v
+ ^ lu^Tw^ ~ Wqu*zY+ w] ~
.hd`bu
) h
& -* /m! <yh!Oa$h-* la#
330
On the third day, when they
were in pain, two of Jacob's
sons, Simeon and Levi, Dinah's
brothers, each took his sword
and came upon the city
unawares, and killed every male.
Pc-vqw3msXa Pp+Pa-ncms Pa-ams
Pp-vqcX3mp vqPmp Pc-vqw3mp
amdc-ncmpc-np np Pc-np ncmpc np
ncms ncfscX3ms Pc-vqw3mpXa
Pp-Pa-ncfs ncms Pc-vqw3mp
ncmsc-ncms
<yb!aK
& ) <t*oyh=B! yv!yl!Vh
= ^ <oYb^ yh!yw+ ~
yw]lw@ + /oum=v! bq)uy& -~ yn}b-= yn}v= Wjq=Yw] ~
Gen
34:25
hu)rP
+ -^ ta# td\Lh
# % <oy yv!yl!Vh
= ^ <oYB^ yh!yw+ ~
wyd`bu
* -& lk*l= hT#vm
= ! cu^Yw~ ~
Gen
40:20
<yq!vM
= h
^ ^ rc^ var{-ta# aC*Yw] ~
.wyd`bu
* & EotB= <yp!ah
) * rc^ var{-ta#w+
rq#Bh
) ^ ty{hB
= ! yv!yl!Vh
= ^ <oYb^ yh!yw+ ~
Exod
19:16
yu!yb!Vh
= ^ <oYB^ yh!yw+ ~
rj^Vh
^ ^ tolu&K^ WmK!vY= w~ ~
<ym!uP
* = ub^v# hZ\h^ fP*vM
= !K^ ryu!h-* ta# WBs)Yw` ~
ryu!h-* ta# Wbb=s* aWhh^ <oYB^ qr~
.<ym!uP
* = ub^v#
Josh 6:15
331
On the fourth day they said to
Samson's wife, Entice your
husband, so that he will tell us
the riddle,
Pc-vqw3msXa Pp+Pa-ncms Pa-ams
Pc-vqw3mp Pp-ncfsc-np vpvfs PoncmscX2fs Pc-vhi3msXa{1}JtPpX1cp Po-Pa-ncfs
yu!yb!Vh
= ^ <oYB^ yh!yw+ ~
Ev@ya!-ta# yT!P^ /ovm=v-! tv#al
@ = Wrm=aY{w~
Judg
14:15
<ym!Yh
` ^ tu^bv
= ! wyl*u* Eb=Tw@ ~
hT#vM
= h
! ^ <h#l* hy`h-* rv#a&
Judg
14:17
yu!yb!Vh
= ^ <oYB^ yh!yw+ ~
Wht=qy^ x!h$ yK! Hl*-dG\Yw~ ~
.HM*u^ yn}bl
= ! hd`yj!h^ dG}Tw^ ~
yu!yb!rh
+ * <oYB^ yh!yw+ ~
Judg 19:5
rq#Bb
) ^ WmyK!vY= w~ ~
hr`uN& h
~ ^ yb!a& rm#aY{w~ tk#ll
# * <q*Yw` ~
yv!yl!Vh
= ^ <oYB^ yh!yw+ ~
aB* vya! hN}hw! +
<yu!rq
| = wyd`gb
` W= lWav* <u!m@ hn\jM
& h
^ -^ /m!
w{var{-lu^ hm*da
` w& ~
.WjT*vY= w] ~ hx*ra
+ ^ lP)Yw] ~ dw]D-` la# w{ab)B= yh!yw+ ~
2 Sam
1:2
332
On the seventh day the child
died. And the servants of David
were afraid to tell him that the
child was dead, for they said,
Behold, while the child was
still alive, we spoke to him and
he did not listen to our voice.
How then can we tell him that
the child is dead, since he might
do himself harm!
yu!yb!Vh
= ^ <oYB^ yh!yw+ ~
2 Sam
dl#Yh
` ^ tm*Yw` ~
12:18
yT!dl
+ l
! = yv!yl!Vh
= ^ <oYB^ yh!yw+ ~
rz`-/ya@ wD`jy= ~ Wnj=na
~ w& ~ taZ{h^ hV*ah
! -* <G~ dl#Tw@ ~
1 Kgs
3:18
.ty]BB
* ^ Wnj=na
~ -& <y]Tv
^ = yt!lW* z ty]BB
^ ^ WnT*a!
<ym!y` tu^bv
= ! hL#a@ jk^n{ hL#a@ Wnj&Yw~ ~
1 Kgs
yu!yb!Vh
= ^ <oYB^ yh!yw+ ~
20:29
hm*jl
* M
= h
! ^ br~qT
= w! ~
<r`a-& ta# la@rc
` y= -] yn}b= WKY~w~
.dj*a# <oyB= yl!gr
+ ~ [l#a-# ha*m@
333
yv!yl!Vh
= ^ <oYB^ yh!yw+ ~
Esth 5:1
tWkl=m^ rT@sa
= # vB^lT
= w! ~
El#Mh
# -^ tyB@ rx^jB
& ^ dm)uT
& w^ ~
El#Mh
# w^ + El#Mh
# ^ tyB@ jk^n{ tym!yn]Ph
= ^
tWkl=Mh
^ ^ tyb@B= w{tWkl=m^ aS@K-! lu^ bv@oy
.ty]Bh
* ^ jt^P# jk^n{
Why`ld
= G~ -+ ta# tym!hl
* = yn]Vh
@ ^ <oYB^ yh!yw+ ~
.ud`y` aO vya!w+
Jer 41:
4-5
olV!m! <k#Vm
= ! <yv!na
` & Wab)Yw` ~
yu@rq
| W= /q*z` yj@Lg= m
| = vya! <yn]mv
) = /orm=Vm
) W!
hn`obl=W hj*nm
+ W! <yd]dG+ t
{ m
= W! <yd]gb
` =
.hw`hy+ tyB@ ayb!hl
* = <d`y`B=
In five of the preceding examples, the time established by <oYB^ + -h^ + NUMBER is
further specified:
Gen 34:25
Gen 40:20
Pharaohs birthday
Exod 19:16
1 Kgs 3:18
of my giving birth
Jer 41:4-5
This further specification of the time is included within the temporal frame
established by <oYB^ + -h^ + NUMBER.
334
9.3.2.1.2.1.9 With hn`VB
* ^ + -h^
h^ + NUMBER
The following three examples are essentially identical to the preceding set, except
for the temporal specification being changed from day to year. The first example,
however, has different word order because of the numeric expression.
In the four hundred and
eightieth year after the sons of
Israel came out of the land of
Egypt, in the fourth year of
Solomon's reign over Israel, in
the month of Ziv which is the
second month, he began to build
the house of the LORD.
1 Kgs 6:1
<y]rx
~ m
= -! Jr\am
# @ la@rc
` y= -] yn}B= tax@l=
yn]Vh
@ ^ vd\jh
) ^ aWh wz] vd\jB
) = tyu!yb!rh
+ * hn`VB
* ^
la@rc
` y= -] lu^ hm)Ov= EOm=l!
.hw`hyl^ ty]Bh
^ ^ /b#Yw] ~
tyv!yl!Vh
= ^ hn`VB
* ^ yh!yw+ ~
.la@rc
` y= ] El#m-# la# hd`Why+-El#m# fp*vo* hy+ dr\Yw} ~
1 Kgs
22:2
vd\jl
) ^ hV*mj
! B
& ^ yV!VB
! ^ tyV!Vh
! ^ hn`VB
* ^ yh!yw+ ~
Ezek 8:1
yn`pl
* = <yb!vo= y hd`Why+ yn}q=zw] + yt!yb@B= bv@oy yn]a&
.ho]hy+ yn`da
{ & dy~ <v* yl^u* lP)Tw! ~
The example in Ezek 8:1 has a clause bv@oy yn]a,& included within the temporal
frame, that provides further setting for the event of the main clause. The concurrent
nature of the event of this additional clause motivates the use of the participle here.
335
9.3.2.1.2.1.10 With <oyB= + INFC
In one case, immediately following <oyB= yh!yw+ ~ is an infinitive construct as the
verbal element of the dependent clause.
On the day Moses finished
setting up the tabernacle, he
anointed it and consecrated it
with all its furnishings and the
altar and all its utensils; he
anointed them and consecrated
them also.
/K*vM
= h
! -^ ta# <yq!hl
* = hv#m) toLK^ <oyB= yh!yw+ ~
Num 7:1
tyu!bV
! B
= ^ yh!yw+ ~
1 Kgs
18:44
.<v#Gh
` ^ hk*rx
+ u
* y^ ~ aOw+ dr}w` rs)a$ ba*ja
= -^ la#
336
9.3.2.1.2.1.12 At a Specific Time
Five occurrences are attested of B= indicating a specific time at which the next
event occurred. The first four specify a specific time of day such as midnight, the
morning watch, or noon. The fifth uses an expression with an infinitive construct to
= ! tL^j!tB
= ! yh!yw+ ~ at the beginning of their living
specify the time of the next event: <v* <T*bv
there.
At midnight the LORD struck all
the firstborn in the land of
Egypt, from the firstborn of
Pharaoh who sat on his throne
to the firstborn of the captive
who was in the dungeon, and all
the firstborn of cattle. Pharaoh
arose in the night, he and all his
servants and all the Egyptians,
and there was a great cry in
Egypt, for there was no home
where there was not someone
dead.
Exod
12:29-30
yb!Vh
= ^ rokB= du^ w{as=K-! lu^
.hm*hB
@ = rokB= lk)w+ roBh^ tyb@B= rv#a&
wyd`bu
* -& lk*w+ aWh hl*yl
+ ^ hu)rP
+ ^ <q*Yw` ~
<y]rx
` m
= B
! = hl*dg{ + hq*ux
* = yh!Tw= ~ <y]rx
~ m
= -! lk*w+
.tm@ <v*-/ya@ rv#a& ty]B^ /ya@-yK!
Ruth 3:8
337
At the morning watch, the LORD
looked down on the army of the
Egyptians through the pillar of
fire and cloud and brought the
army of the Egyptians into
confusion.
Exod
14:24
.<y]rx
` m
= ! hn}jm
& ^ ta@
<y]rh
~ X
( b
* ^ yh!yw+ ~
1 Kgs
WhY`la
! @ <h#B* lT@hy^ w+ ~
18:27
<v* <T*bv
= ! tL^jt
! B
= ! yh!yw+ ~
2 Kgs
hw`hy+-ta# War+y` aO
17:25
toyr`ah
& -* ta# <h#B* hw`hy+ jL^vy^ w+ ~
.<h#B* <yg]rh
+ ) Wyh=Yw] ~
338
The words of Nehemiah the son
of Hacaliah:
In the month Chislev, in the
twentieth year, while I was in
Susa the capitol, Hanani, one of
my brothers, and some men from
Judah came;
ncmpc np ncmsc-np Pc-vqw3msXa
Pp-ncms-np np ncfsc amp Pc-pi1cs
vqp1cs Pp-np Pa-ncfs Pcvqw3msXa np ams Pp-ncmpcX1cs
pi3ms Pc-ncmp Pp-np
hy`lk
= j
^ -& /B# hy`mj
= n# + yr}bD
= ]
<yr]cu
= # tn~v= wl@sK
= -! vd\jb
) = yh!yw+ ~
Neh 1:12a
<yr]cu
= # tn~v= /s*yn] vd\jB
) = yh!yw+ ~
Neh 2:1
aC*aw# ` wyn`pl
* = /y]y~ El#Mh
# ^ aT=sv
= j
^ T
= r
^ a
+ l
^ =
El#Ml
# ^ hn`Ta
= w# ` /y]Yh
~ -^ ta#
.wyn`pl
* = ur~ yt!yy]h-* aOw+
<yf!pV
= h
) ^ fp)v= ym@yB! yh!yw+ ~
hd`Why+ <j#l# tyB@m! vya! El#Yw} ~ Jr\aB
* * bu*r` yh!yw+ ~
.wyn`b* yn}vW= w{Tv=aw! + aWh ba*om yd}cB
= ! rWgl*
Ruth 1:1
339
In the days of Ahasuerus, the
Ahasuerus who reigned from
India to Ethiopia over 127
provinces,
Pc-vqw3msXa Pp-ncmpc np pi3ms
np Pa-vqPms Pp-np Pc-Pp-np afs
Pc-amp Pc-afs ncfs
vorw}vj
= a
^ & aWh vorw}vj
= a
^ & ym@yB! yh!yw+ ~
Esth 1:1
<yr]cu
= w# + ub^v# vWK-du^w+ WDh)m@ El@Mh
) ^
.hn`yd]m= ha*mW@
El#Mh
# ^ rb^dB
+ ! aobl* yT!vw= ~ hK*lM
= h
^ ^ /a@mT
* w= ~
Esth 1:12
El#Mh
# ^ [x)qY= w] ~ <ys!yr]Sh
* ^ dy~B= rv#a&
.w{b hr`uB
& * w{tm*jw& ~ da)m=
Pc-vpw3fs Pa-ncfs np Pp-vqc Ppncmsc Pa-ncms Pr Pp-ncfsc Pancmp Pc-vqw3ms Pa-ncms Pd PcncfscX3ms vqp3fs PpX3ms
WhY`va
! y{-/B# <yq!yo` hy+ ym@yB! yh!yw+ ~
Jer 1:3-4
hn`v* hr}cu
= # yT@vu
= ^ <T)-du^ hd`Why+ El#m#
hd`Why+ El#m# WhY`va
! y{-/b# WhY`qd
! x
+ l
! =
.yv!ym!jh
& ^ vd\jB
) ^ <]lv
^ W* ry+ tolG+-du^
.rm)al@ yl^a@ hw`hy+-rb^d+ yh!yw+ ~
Pc-vqw3msXa Pp-ncmpc np ncmscnp ncms np Pp-vqc afs afs ncfs Ppnp ncmsc-np ncms np Pp-vqc np
Pp+Pa-ncms Pa-ams Pcvqw3msXa ncmsc-np PpX1cs Ppvqc
The Esther example is noteworthy for the lengthy span of clauses between the
temporal expression in 1:1 and the WAYYIQTOL of 1:12, indicated above by the . Esther
1:1-11 is the complex setting for the narrative which follows and as such does not have
the typical characteristics of sequential narrative.
9.3.2.1.2.2 Specific Temporal Reference Followed by QATAL
The following occurrences differ from the preceding set in 9.3.2.1.2.1 in that the
verb which follows the temporal expression is a QATAL. The use of the QATAL indicates
that the event or state of affairs expressed by that verb had already taken place or were
340
already in effect. Since the syntactic shape of these occurrences is so similar to those in
9.3.2.1.2.1, minimal comments are made throughout the following data display.
* ^
9.3.2.1.2.2.1 With ayh!h^ hn`VB
In the same year, in the
beginning of the reign of
Zedekiah king of Judah, in the
fourth year, in the fifth month,
Hananiah the son of Azzur, the
prophet, who was from Gibeon,
had spoken to me in the house of
the LORD in the presence of the
priests and all the people,
saying,
Jer 28:1
tyu!br
! h
+ * tn~vB
= ! hd`Why+-El#m# hY`qd
! x
+ ! tk#lm
# m
= ^
hy`nn+ j
~ & yl^a@ rm^a* yv!ym!jh
& ^ vd\jB
) ^
hw`hy+ tyb@B= /oub=Gm
] ! rv#a& ayb!Nh
` ^ rWZu^-/b#
.rm)al@ <u*h-* lk*w+ <yn]hK
& h
) ^ yn}yu@l=
One further instance has both the specific temporal reference, <oyB= on the day, but
this is immediately followed by the QATAL rB#D] he spoke. The temporal reference is not
merely on that day, the Lord spoke, but on the day the Lord spoke. One might expect an
infinitive construct with B=, but this is the regular QATAL rB#D.]
On the day when the LORD
spoke to Moses in the land of
Egypt, the LORD spoke to
Moses, saying, I am the LORD;
speak to Pharaoh king of Egypt
all that I speak to you.
Pc-vqw3msXa Pp-ncms vpp3ms np
Pp-np Pp-ncbs np Pc-vpw3ms np
Pp-np Pp-vqc pi1cs np vpvms Pp-np
ncms np Po ncmsc-Pr pi1cs vqPms
PpX2ms
Exod
6:28-29
341
9.3.2.1.2.2.2 With hZ\
\h^ <oYh^ <x#uB
# =
hZh^
The occurrence of hZ\h^ <oYh^ <x#uB
# = specifically expresses that the following
action took place on that very day itself as an extension of its meaning as substance, self
(BDB 1996, 782-83).
At the end of four hundred and
thirty years, to the very day, all
the hosts of the LORD went out
from the land of Egypt.
Pc-vqw3msXa Pp-ncms amp ncfs
Pc-afs afp ncfs Pc-vqw3msXa Ppncfsc Pa-ncms Pa-ams vqp3cp
ncmsc-ncbpc np Pp-ncbs np
Exod
12:41
.<y]rx
` m
= ! Jr\am
# @ hw`hy+ toab=x-! lK* Wax=y`
Exod
la@rc
` y= ] yn}B-= ta# hw`hy+ ayx!oh
12:51
.<t*ab
) x
= -! lu^ <y]rx
~ m
= ! Jr\am
# @
dw]dl
` = lWav*-tB^ br~m-@ ta# tT@ tu@B= yh!yw+ ~
1 Sam
.hV*al
! = yt!lj
* M
) h
= ^ la@yr]du
+ ^l= hn`Tn= ] ayh!w+
18:19
342
On the sixth day they gathered
twice as much bread, two omers
for each one.
Exod
dj*al
# * rm#uh
) * yn}v= hn\v=m!
16:22
yu!yb!Vh
= ^ <oYB^ yh!yw+ ~
Exod
16:27
yn]ym!Vh
= ^ <oYB^ yh!yw+ ~
Lev 9:1
.la@rc
` y= ] yn}qz= l
] W= wyn`bl
* W= /r{ha
& l
^ = hv#m) ar`q*
In the following two sections, the same type of temporal reference to months and
years is expressed.
9.3.2.1.2.2.5 With vd\
\jB
) ^ + -h^
h^ + NUMBER
vdj)
In the first month of the second
year, on the first day of the
month, the tabernacle was
erected.
tyn]Vh
@ ^ hn`VB
* ^ /ovar]h* vd\jB
) ^ yh!yw+ ~
Exod
./K*vM
= h
! ^ <q^Wh vd\jl
) ^ dj*aB
# =
40:17
lau@mv
* y= ] aB* yu!yb!Vh
= ^ vd\jB
) ^ yh!yw+ ~
2 Kgs
hk*WlM=h^ ur~Zm
\ ! um*vy* l!a$-/B# hy`nt
+ n^ -+ /B#
25:25
Why`ld
= G~ -+ ta# WKY~w~ w{Ta! <yv!na
` & hr`cu
* w& ~
<yD]cK
= h
^ -^ ta#w+ <yd]WhY+h-^ ta#w+ tm)Yw` ~
.hP*xM
= B
! ^ w{Ta! Wyh*-rv#a&
343
lau@mv
* y= ] aB* yu!yb!Vh
= ^ vd\jB
) ^ yh!yw+ ~
Jer 41:1
hk*WlM=h^ ur~Zm
\ ! um*vy* l!a$-/b# hy`nt
+ n^ -+ /B#
<yv!na
` & hr`cu
* w& ~ El#Mh
# ^ yB@rw~ +
ht*Px
* M
= h
! ^ <q*yj!a-& /b# Why`ld
= G~ -+ la# w{Ta!
.hP*xM
= B
! ^ wD`jy= ~ <j#l# <v* Wlk=aY{w~
dj*aB
# = /ovar]B* hn`v* toam@-vv@w+ tj^aB
^ = yh!yw+ ~
Gen 8:13
Jr\ah
* * lu^m@ <y]Mh
^ ^ Wbr+j* vd\jl
) ^
hb*Th
@ ^ hs@km
= -! ta# j~n{ rs^Y`w~
.hm*da
` h
& * yn}P= Wbr+j* hN}hw! + ar+Yw~ ~
Deut 1:3
344
In the eighteenth year of King
Josiah, the king sent Shaphan,
the son of Azaliah the son of
Meshullam the scribe, to the
house of the LORD saying,
WhY`va
! y{ El#Ml
# ^ hn`v* hr}c=u# hn\mv
) B
= ! yh!yw+ ~
Why`lx
= a
^ -& /B# /p*v-* ta# El#Mh
# ^ jl^v*
2 Kgs
22:3
2 Kgs
25:27
Ed~rm
{ = lyw]a$ ac*n` vd\jl
) ^ hu*bv
= w! + <yr]cu
= B
# =
var{-ta# w{kl=m* tn~v=B! lb#B* El#m#
.al#K# tyB@m! hd`Why+-El#m# /yk!yo` hy+
WhY`qz! j
+ ! El#Ml
# ^ hn`v* hr}c=u# uB^ra
+ B
^ = yh!yw+ ~
Isa 36:1
Jer 52:31
345
In the thirtieth year, on the fifth
day of the fourth month, while I
was by the river Chebar among
the exiles, the heavens were
opened and I saw visions of
God.
hV*mj
! B
& ^ yu!yb!rB
+ * hn`v* <yv!Ov=B! yh!yw+ ~
Ezek 1:1
Pc-vqw3msXa Pp-amp ncfs Pp+Paams Pp+Pa-ams Pp+Pa-ncms Pcpi1cs Pp-ncmsc-Pa-ncfs Pp-ncmscnp vnp3cp Pa-ncmp Pc-vqw1cs
ncfpc ncmp
vd\jl
) ^ dj*aB
# = hn`v* hr}cu
= #-yT@vu
= B
^ = yh!yw+ ~
Ezek
26:1
Ezek
29:17
Ezek
30:20
yv!yl!VB
= ^ hn`v* hr}cu
= # tj^aB
^ = yh!yw+ ~
Ezek
31:1
rc*u-* yn}vB
= ! hn`v* hr}cu
= # yT@vB
= ! yh!yw+ ~
Ezek
32:1
346
In the twelfth year, on the
fifteenth of the month, the word
of the LORD came to me saying,
hV*mj
! B
& ^ hn`v* hr}cu
= # yT@v=B! yh!yw+ ~
Ezek
32:17
yr]cu
! B
& * hn`v* hr}cu
= # yT@v=B! yh!yw+ ~
Ezek
fyl!Ph
* ^ yl^a-@ aB* Wnt@Wlg`l= vd\jl
) ^ hV*mj
! B
& ^
33:21
.ryu!h* ht*Kh
= % rm)al@ <]l^vW* rym!
El#Mh
# ^ vw\yr
` d
+ l
` = uB^ra
+ ^ tn~vB
= ! yh!yw+ ~
Zech 7:1
vd\jl
) ^ hu*Br
* a
+ B
^ = hy`rk
+ z^ +-la# hw`hy+-rb^d+ hy`h*
.wl@sk
= B
! = yu!vT
! h
= ^
<yr]cu
= B
# = yn]Vh
@ ^ vd\jB
) ^ tyn]Vh
@ ^ hn`VB
* ^ yh!yw+ ~
Num
.td|uh
@ * /K^vm
= ! lu^m@ /n`uh
* # hl*un& ~ vd\jB
) ^
10:11
<u*bj
= r
^ + El#Ml
# ^ tyv!ym!jh
& ^ hn`VB
* ^ yh!yw+ ~
1 Kgs
.<*l
! v*Wry+-lu^ <y]rx
~ m
= -! El#m# qv^wv! hl*u*
14:25
va*ohy+ El#Ml
# ^ hn`v* vOv*w+ <yr]cu
= # tn~vB
= ! yh!yw+ ~
2 Kgs
.ty]Bh
* ^ qd\B-# ta# <yn]hK
& h
) ^ WqZ+j-! aO
12:7
347
In the third year of Hoshea, the
son of Elah king of Israel,
Hezekiah the son of Ahaz king of
Judah became king.
2 Kgs
18:1
WhY`qz! j
+ ! El#Ml
# ^ tyu!yb!rh
+ * hn`VB
* ^ yh!yw+ ~
hl*a-@ /B# u~vo@ hl= tyu!yb!Vh
= ^ hn`Vh
* ^ ayh!
2 Kgs
18:9
rs#an# m
+ l
^ v
= ^ hl*u* la@rc
` y= ] El#m#
.h*yl#u* rx^Yw` ~ /orm=v-) lu^ rWVa^-El#m#
w{kl=ml
* = tyu!yv!Th
= ^ tn~vb
= ! yh!yw+ ~
rX^an\dk
+ b
^ n% + aB* vd\jl
) ^ rocu*B# yr]yc!uh
& * vd\jB
) ^
2 Kgs
25:1
<]lv
^ W* ry+-lu^ w{lyj@-lk*w+ aWh lb#B-* El#m#
.byb!s* qy}D` h*yl#u* Wnb=Yw] ~ h*yl#u* /j^Yw] ~
<u*bj
= r
^ + El#Ml
# ^ tyv!ym!jh
& ^ hn`VB
* ^ yh!yw+ ~
yK! <*l
! v*Wry+-lu^ <y]rx
~ m
= -! El#m# qv^yv! hl*u*
2 Chr
12:2
.hw`hyB^ Wlu&m*
Jer 36:1
348
In the fifth year of Jehoiakim the
son of Josiah, king of Judah, in
the ninth month, all the people in
Jerusalem and all the people
who came from the cities of
Judah to Jerusalem proclaimed
a fast before the LORD.
vd\jB
) ^ w{kl=ml
* = tyu!vT
! h
= ^ hn`Vb
* ^ yh!yw+ ~
Jer 52:4
rX^ar\dk
+ W^ bn+ aB* vd\jl
) ^ rocu*B# yr]yc!uh
& *
w{lyj@-lk*w+ aWh lb#B-* El#m#
qy}D` h*yl#u* Wnb=Yw] ~ h*yl#u* Wnj&Yw~ ~ <]lv
^ W* ry+-lu^
.byb!s*
Jer 36:9
yv!mj
! B
& ^ tyu!yb!Vh
= ^ hn`VB
* ^ yh!yw+ ~
Ezek
la@rc
` y= ] yn}qZ= m
] ! <yv!na
` & WaB* vd\jl
) ^ rocu*B#
20:1
.yn`pl
* = Wbv=Yw} ~ hw`hy+-ta# vr{dl
+ !
One of the questions not addressed here is the possible narrative structuring
function that temporal expressions like the preceding sets of data may have. This would
require extensive narrative analysis of each of the books from which these examples
come and is beyond the scope of the present study.
349
9.3.2.1.2.2.7 With <u^PB
^ ^ + -h^
h^ + NUMBER
This expression parallels the structure of the previous three, but with <u^P^ time
rather than day, month or year.
At the seventh time, when the
priests blew the trumpets,
Joshua said to the people,
Shout! For the LORD has
given you the city.
Wuq=T* tyu!yb!Vh
= ^ <u^PB
^ ^ yh!yw+ ~
Josh 6:16
Ru*nv
+ -! El#m# lp#rm
` a
= ^ ym@yB! yh!yw+ ~
<l*yu@ El#m# rm#ul
) r
* d
+ K
` = rs*La
* # El#m# Eoyr+a^
.<y]oG El#m# lu*dt
+ w! +
<d{s= El#m# ur~B-# ta# hm*j*lm
= ! Wcu*
Gen
14:1-2a
350
In the days of Ahaz, the son of
Jotham, the son of Uzziah, king
of Judah, Rezin the king of Aram
and Pekah the son of Remaliah,
king of Israel, went up to
Jerusalem to wage war against
it, but could not conquer it.
When it was reported to the
house of David, saying, The
Arameans have camped in
Ephraim, his heart and the
hearts of his people shook as the
trees of the forest shake with the
wind.
Isa 7:1-2
9.3.2.2
With K=
The first question that arises regarding the occurrences of K= yh!yw+ ~ is how they
differ from the similar use of B= yh!yw+ ~ with infinitive constructs. For example, the
infinitive construct of awb occurs with B= yh!yw+ ~ in Judg 3:27 and K= yh!yw+ ~ in 1 Sam 4:5.
This provides good contrast for exploring the different nuances with these prepositions.
As he arrived, he blew the
trumpet in the hill country of
Ephraim; and the sons of Israel
went down with him from the hill
country, and he was in front of
them.
Pc-vqw3msXa Pp-vqcX3ms Pcvqw3ms Pp+Pa-ncms Pp-ncms np
Pc-vqw3mp PpX3ms ncmpc-np PpPa-ncms Pc-pi3ms Pp-ncmpcX3mp
w{aobB= yh!yw+ ~
<y]rp
` a
= # rh^B= rp*oVB^ uq^t=Yw] ~
Judg 3:27
351
When the ark of the covenant of
the LORD came into the camp,
all Israel shouted with a great
shout, so that the earth
resounded.
hn\jM
& h
^ -^ la# hw`hy+-tyr]B= /ora& aobK= yh!yw+ ~
hl*odg+ hu*WrT= la@rc
` y= -] lk* Wur]Yw` ~
1 Sam
4:5
.Jr\ah
* * <h)Tw@ ~
Pc-vqw3msXa Pp-vqc ncbsc ncfsnp Pp-Pa-ncbs Pc-vhw3mp ncmscnp ncfs afs Pc-vnw3fsXa Pa-ncbs
As stated in 9.3.2, B= yh!yw+ ~ draws the temporal circle within which the action of
the next independent clause takes place. This is not simultaneity, but rather a statement
that during the action of the verb in the dependent clause, the action of the verb in the
main clause occurs. In Judg 3:27, then, the occurrence of w{aobB= yh!yw+ ~ would indicate that
during Ehuds arrival, he blew the trumpet. In contrast, in 1 Sam 4:5 aobK= yh!yw+ ~ indicates
that the ark of the Lord entered the camp and then all Israel shouted. As Gropp states,
concerning the use of K= compared with B=, [t]he basic distinction is between temporal
succession and temporal overlap (Gropp 1995, 205). This distinction should be reflected
in the translation of the occurrences of the infinitive construct with K= yh!y+w~ and B= yh!yw+ .~
The following sections of data with K= yh!yw+ ~ are quite extensive, being divided as
follows:
9.3.2.2.1 With Infinitive Constructs (pages 350-76)
9.3.2.2.2 With Specific Temporal Reference (page 379)
9.3.2.2.1
The occurrences with K= yh!yw+ ~ parallel those with B= yh!yw+ ~ discussed above in
9.3.2.1. Here, however, there are four sections of data:
9.3.2.2.1.1 Infinitive Construct Followed by WAYYIQTOL
352
9.3.2.2.1.2 Infinitive Construct Followed by WAYYIQTOL with Intervening Clauses
9.3.2.2.1.3 Infinitive Construct Followed by QATAL
9.3.2.2.1.4 Infinitive Construct Followed by YIQTOL
In the following examples, the object or complement of the infinitive construct,
no matter how complex, does not qualify as an intervening clause.
9.3.2.2.1.1 Infinitive Construct Followed by WAYYIQTOL
The instances of INFC K= yh!yw+ ~ follow this pattern very consistently, but there are
certain syntactic features which vary slightly. For example, in Num 16:31; Deut 31:24;
Josh 8:24, 10:20; Judg 15:17; 1 Sam 13:10, 18:1, 24:17; 2 Sam 13:36; 1 Kgs 9:1; 2 Kgs
10:25; Jer 26:8, and 43:1 the infinitive construct with K= is followed by another infinitive
construct with l=. In every instance listed here, the first infinitive construct is of the verb
hlk to finish, as in Num 16:31,
rB@dl
~ = w{tOk^K= yh!yw+ ~ when he finished speaking
The subject of the infinitive construct is realized either as a pronominal suffix
attached to the infinitive construct itself, or as an explicit subject as in Deut 31:24,
bT)kl
= ! hv#m) toLk^K= yh!yw+ ~ when Moses finished writing
The comments in 9.3.2.1.1 regarding SAME and DIFFERENT SUBJECT also apply
here.
Also, in Gen 24:30 and 2 Chr 12:1 the parallel structure is noteworthy:
K=W
...
K= yh!yw+ ~
w{um=vk
* W=
ta)rK
+ ! yh!yw+ ~
w{tq*zj
+ k
# W= <u*bj
= r
^ + tWkl=m^ /yk!hK
* = yh!yw+ ~
Gen 24:30
2 Chr 12:1
... K= yh!yw+ ~
353
No instances are found of K= K= yh!yw+ ,~ which is congruent with K= yh!yw+ ~ indicating
temporal succession in contrast to B= yh!yw+ ~ which indicates temporal overlap.
Before proceeding with the examples, three points of clause syntax should be
addressed. Gen 39:13 provides the first point of departure:
When she saw that he had left
his garment in her hand and had
fled outside,
Ht*oar+K! yh!yw+ ~
.hx*Wjh^ sn`Yw` ~ Hd`yB
` = w{dg+B! bz~u-* yK!
Gen
39:13
The yK! clause is not analyzed here as an intervening clause because of its function
as a complement of har. It functions as the object of the transitive verb har just like
<z\Nh
\ -^ ta# in Gen 24:30.
Object
bz~u-* yK!
yh!yw+ ~ +
INFC of transitive verb
Ht*oar+K! yh!yw+ ~
<z\Nh
\ -^ ta#
ta)rK
+ ! yh!yw+ ~
Figure 27: Object of Transitive Verb
A second issue in the Gen 39:13 example is the occurrence of hx*Wjh^ sn`Yw` ~, a
WAYYIQTOL
within the yK! clause. This should not be construed as an example of the
consecutive use of the WAYYIQTOL after a QATAL. There are very specific syntactic
parameters that govern the occurrence of the WAYYIQTOL here. It does indeed occur after
the QATAL bz~u,* but this is not just random narrative. The occurrence of bz~u* is tightly
constrained by yK! (See also dk^l-* yK! in Josh 10:1.). To describe the occurrence of sn`Yw` ~
after bz~u,* it is preferable here to say that the WAYYIQTOL is the default form for past
narrative. This may very well be what consecutive is intended to communicate, but the
preference here is to avoid the term consecutive because of its association with the waw-
354
consecutive theory of the WAYYIQTOL. The use of WAYYIQTOL is governed by syntactic
and pragmatic parameters that are more complex than what is typically contemplated by
the waw-consecutive theory.
The rv#a& clause in 1 Kgs 13:4 is similar to the yK! clause in that it functions within
the temporal clause initiated by K= yh!yw+ .~ In the 1 Kgs 13:4 example, ar`q* rv#a& is part of
a relative clause used descriptively to modify <yh!Oa$h-* vya! rb^D-+ ta#. For this reason, it
is not considered to be an intervening clause between the yh!yw+ ~ and jl^vY= w] .~
Other semantic and syntactic features will be discussed after the data display.
When Abram came into Egypt,
the Egyptians saw that the
woman was very beautiful.
Pc-vqw3msXa Pp-vqc np npXd Pcvqw3mp Pa-np Po-Pa-ncfs Pp-afs
pi3fs Pd
hm*yr
+ x
` m
= ! <r`ba
= ^ aobK= yh!yw+ ~
Gen
hV*ah
! -* ta# <yr]xM
= h
! ^ War+Y]w~
12:14
Gen
19:17
hr`hh
* * rK*Kh
! -^ lk*B= dm)uT
& ^-la^w+
.hp#ST
* -! /P# fl@Mh
* !
<yd]mX
! h
= -^ ta#w+ <z\Nh
\ -^ ta# ta)rK
+ ! yh!yw+ ~
hq*br
= ] yr}bD
= -] ta# w{um=vk
* =W w{tj)a& yd}y-+ lu^
vya!h* yl^a@ rB#d-] hK) rm)al@ w{tj)a&
<yL!mG^ h
+ -^ lu^ dm@u) hN}hw! + vya!h-* la# ab)Yw` ~
./y]uh
* -* lu^
Gen
24:30
355
When Laban heard the news of
Jacob his sister's son, he ran to
meet him, and embraced him
and kissed him and brought him
to his house. Then he related to
Laban all these things.
Pc-vqw3msXa Pp-vqc np Po-ncms
np ncmsc-ncfscX3ms Pc-vqw3ms
Pp-vqcX3ms Pc-vpw3ms-PpX3ms
Pc-vpw3ms-PpX3ms Pcvhw3msX3ms Pp-ncmscX3ms Pcvpw3ms Pp-np Po ncmsc-Pa-ncmp
Pa-acp
Gen
29:13
Ht*oar+K! yh!yw+ ~
.hx*Wjh^ sn`Yw` ~ Hd`yB
` = w{dg+B! bz~u-* yK!
Gen
39:13-14
ar`qa
= w# ` yl!oq yt!my) r]h-& yK! w{um=vk
* = yh!yw+ ~
Gen
39:15
Pc-vqw3msXa Pp-vqcX3ms Ppvhp1cs ncmscX1cs Pc-vqw1cs Pcvqw3ms ncmscX3ms PpX1cs Pcvqw3ms Pc-vqw3ms Pa-ncmscXd
ar`qa
= w# ` yl!oq ym!yr]hK
& ^ yh!yw+ ~
.hx*Wjh^ sn`Yw` ~ yl!xa
= # w{dg+B! bz{uY& w~ ~
Gen
39:18
356
When his master heard his
wifes words, which she spoke to
him, saying, This is what your
slave did to me, his anger
burned.
w{Tv=a! yr}bD
= -] ta# wyn`da
{ & u~mv
) k
= ! yh!yw+ ~
rm)al@ wyl*a@ hr`BD
= ] rv#a&
Gen
39:19
;D\bu
= ^ yl! Hc*u* hL#ah
@ * <yr]bD
* K
+ ^
.w{Pa^ rj^Yw] ~
la@rc
` y= -] yn}B= td~u-& lK*-la# /r{ha
& ^ rB@dK
~ = yh!yw+ ~
Exod
16:10
./n`uB
* # ha*rn+ ] hw`hy+
Num
11:25
<yn]qZ@ h
+ ^ vya! <yu!bv
= -! lu^ /T@Yw] ~
WaB=nt
~ Y= w] ~ j~Wrh* <h#yl@u& j~onK= yh!yw+ ~
.Wps*y` aOw+
Pc-vqw3msXa np Pp+Pa-ncms Pcvpw3ms PpX3ms Pc-vhw3ms PpPa-ncbs Pr PpX3ms Pc-vqw3ms Ppamp ncms Pa-amp Pc-vqw3msXa
Pp-vqc PpX3mp Pa-ncbs Pcvtw3mp Pc-Pn vqp3cp
hL#ah
@ * <yr]bD
* h
+ -^ lK* ta@ rB@dl
~ = w{tOk^K= yh!yw+ ~
Num
.<h#yT@jT
= ^ rv#a& hm*da
` h
& * uq^BT
* w! ~
16:31
yr}bD
= -] ta# bT)kl
= ! hv#m) toLk^K= yh!yw+ ~
.<M*T% du^ rp#s-@ lu^ taZ{h^-hr`oTh^
<Y]wl
] h
= -^ ta# hv#m) wx^yw+ ~
.rm)al@ hw`hy+-tyr]B= /ora& ya@cn= {
Deut
31:24
357
When all the kings of the
Amorites who were beyond the
Jordan to the west, and all the
kings of the Canaanites who
were by the sea, heard how the
LORD had dried up the waters
of the Jordan before the sons of
Israel until they had crossed,
their hearts melted, and there
was no spirit in them any longer
because of the sons of Israel.
yr]ma
) h
$ * yk@lm
= -^ lK* u~mv
) k
= ! yh!yw+ ~
Josh 5:1
yn]un& K
~ h
= ^ yk@lm
= -^ lk*w+ hM*y` /D}rY+ h
~ ^ rb#uB
@ = rv#a&
hw`hy+ vyb!oh-rv#a& ta@ <Y`h-^ lu^ rv#a&
la@rc
` y= -] yn}b= yn}Pm
= ! /D}rY+ ~h^ ym@-ta#
<b*bl
* = sM^Yw] ~ Wnr`bu
= -* du^
.la@rc
` y= -] yn}B= yn}Pm
= ! j~Wr dou <b* hy`h-* aOw+
torp*VB
) ^ Wuq=tY= w] ~ <u*h* ur~Yw` ~
Josh 6:20
Josh 8:14
358
When Israel had finished killing
all the inhabitants of Ai in the
field in the wilderness where
they pursued them, and all of
them were fallen by the edge of
the sword until they were
destroyed, all Israel returned to
Ai and struck it with the edge of
the sword.
Josh 8:24
/D}rY+ h
~ ^ rb#uB
@ = rv#a& <yk!l*Mh
= -^ lK* u~mv
) k
= ! yh!yw+ ~
<Y`h^ [oj lk)bW= hl*pV
@ b
= W^ rh*B*
Josh 9:
1-2
yT!jh
! ^ /onb*Lh
= ^ lWm-la# lodG`h^
.ys!Wby+hw^ + yW]jh
! ^ yZ]rP
] h
= ^ yn]un& K
~ h
= ^ yr]ma
) h
$ w* +
<]lv
^ W* ry+ El#m# qd\x-# yn]da
{ & u~mv
) k
= ! yh!yw+ ~
Hm*yr]jY& w~ ~ yu^h-* ta# u~vo% hy+ dk^l-* yK!
yu^l* hc*u-* /K@ HK*lm
= l
^ W= ojyr]yl! hc*u* rv#aK
& ^
yb@vy= { Wmyl!vh
= ! yk!w+ HK*l=ml
^ W=
.<B*rq
+ B
! = Wyh=Yw] ~ la@rc
` y= -] ta# /oub=g]
da)m= War+yY]w~
Josh
10:1-2a
359
When they brought these kings
out to Joshua, Joshua called for
all the men of Israel, and said to
the chiefs of the men of war who
had gone with him, Come near,
put your feet on the necks of
these kings. So they came near
and put their feet on their necks.
Josh
10:24
hm*jl
* M
= h
! ^ yv@na
+ ^ yn}yx!q-= la# rm#aY{w~ la@rc
` y= ]
Wmyc! Wbr+q! w{Ta! aWkl=hh
* #
hL#ah
@ * <yk!lM
* h
= ^ yr}aW+x-^ lu^ <k#yl@gr
+ -~ ta#
Josh 11:1
<yr]bD
* h
+ -^ ta# hw`hy+ Ea^lm
= ^ rB@dK
~ = yh!yw+ ~
Judg 2:4
<olj&h^ rP^sm
= -! ta# /oud+G] u~mv
) k
= ! yh!yw+ ~
la@rc
` y= ] hn}jm
& -^ la# bv*Yw` ~ WjT*vY= w] ~ w{rb=v-! ta#w+
WmWq rm#aY{w~
./y`dm
+ ! hn}jm
& -^ ta# <k#dy+ B
\ = hw`hy+ /t^n-` yK!
Judg 7:15
360
When he saw her, he tore his
clothes and said, Alas, my
daughter! You have brought me
very low, and you are among
those who trouble me; for I have
given my word to the LORD, and
I cannot take it back.
wyd`gB
` -= ta# ur~qY= w] ~ Ht*oa w{toar+k! yh!yw+ ~
yn]Tu
! r
= k
~ h
= ! u~rk
} h
= ^ yT!B! Hh*a& rm#aY{w~
Judg
11:35
yp!-yt!yx!P* yk!na
{ w* + yr`ku
= B
) = t=yy]h* T=aw^ +
.bWvl* lk^Wa aOw+ hw`hy+-la#
Judg
14:11
Pc-vqw3msXa Pp-vqcX3mp
PoX3ms Pc-vqw3mp amp ncmp Pcvqw3mp PpX3ms
w{dY`m! yj!Lh
= ^ El@vY= w~ ~ rB@dl
~ = w{tOk^K= yh!yw+ ~
Judg
15:17
hn\jM
& h
^ -^ la# hw`hy+-tyr]B= /ora& aobK= yh!yw+ ~
1 Sam
4:5
.Jr\ah
* * <h)Tw@ ~
Pc-vqw3msXa Pp-vqc ncbsc ncfsnp Pp-Pa-ncbs Pc-vhw3mp ncmscnp ncfs afs Pc-vnw3fsXa Pa-ncbs
1 Sam
4:18
361
So they sent the ark of God to
Ekron. And when the ark of God
came to Ekron the Ekronites
cried out, saying, They have
brought the ark of the God of
Israel around to us, to kill us
and our people.
1 Sam
5:10
WmK!vY= w~ ~
rj^Vh
^ ^ tolu&K^ yh!yw+ ~
1 Sam
9:26
hL#ah
@ * <yr]bD
* h
+ -^ ta# rB@d~l= dw]D` toLk^K= yh!yw+ ~
1 Sam
;l=qh
) & lWav* rm#aY{w~ lWav*-la#
24:17
(16)
hm)Ov= yr}bD
= -] ta# <r`yj! u~mv
) K
= ! yh!yw+ ~
<oYh^ hw`hy+ EWrB* rm#aY{w~ da)m= jm^cY= w] ~
1 Kgs
5:21
(5:7)
362
When Solomon finished building
the house of the LORD, and the
king's house, and all that
Solomon desired to do, the
LORD appeared to Solomon a
second time, as He had
appeared to him at Gibeon.
Pc-vqw3msXa Pp-vpc np Pp-vqc
Po-ncmsc-np Pc-Po-ncmsc Pa-ncms
Pc-Po ncmsc-ncmsc np Pr vqp3ms
Pp-vqc Pc-vnw3msXa np Pp-np afs
Pp-Pr vnp3ms PpX3ms Pp-np
1 Kgs 9:1
<u*br
= y` ` bv*-yK! la@rc
` y= -] lK* u~mv
) K
= ! yh!yw+ ~
1 Kgs
hd`uh
@ -* la# w{ta) War+qY= w] ~ Wjl=vY= w] ~
12:20
yr}ja
& ^ hy`h* aO la@rc
` y= -] lK*-lu^ w{ta) Wkyl!mY= w~ ~
.w{Db^l= hd`Why+-fb#v@ yt!lW* z dw]D-` tyb@
1 Kgs
13:4
rm)al@ j~Bz@ M
+ h
! ^ lu^m@ w{dy`-ta# <u*br
= y` ` jl^vY= w] ~
jl^v* rv#a& w{dy` vb^yT!w~ Whc%pT
= !
.wyl*a@ Hb*yv!hl
& ^ lk)y` aOw+ wyl*u*
h*yl#gr
+ ~ loq-ta# WhY`ja
! & u~mv
) k
= ! yh!yw+ ~
jt^Pb
# ^ ha*B*
hM*l* <u*br
= y` ` tv#a@ ya!B) rm#aY{w~
.hv*q* Ey]la
^ @ j~Wlv* yk!na
{ w* + hr`Kn@ t
~ m
= ! T=a^ hZ\
1 Kgs
14:6
363
When Baasha heard of it, he
ceased fortifying Ramah and
remained in Tirzah.
av*uB
= ^ u~mv
) K
= ! yh!yw+ ~
1 Kgs
.hx*rt
+ B
! = bv#Yw} ~ hm*rh
` -* ta# tonB=m! lD~jY= w\ ~
15:21
ryu!h* hd`Kl
= n= -] yK! yr]mz= ] toar+K! yh!yw+ ~
1 Kgs
[r{cY= w] ~ El#Mh
# -^ tyB@ /omr+a-^ la# ab)Yw` ~
16:18
WhY`la
! -@ ta# ba*ja
= ^ toar+K! yh!yw+ ~
1 Kgs
.la@rc
` y= ] rk@u) hz\ hT*ah
^ ^ wyl*a@ ba*ja
= ^ rm#aY{w~
18:17
<y]rh
~ X
( h
* ^ rb)uK
& ^ yh!yw+ ~
1 Kgs
hj*nM
+ h
! ^ tolu&l^ du^ WaB=nt
~ Y= w] ~
18:29
Pc-vqw3msXa Pp-vqc Pa-ncmp Pcvtw3mp Pp Pp-vqc Pa-ncfs Pc-Pdncms Pc-Pd-vqPms Pc-Pd ncms
WhY`la
! @ u~mv
) K
= ! yh!yw+ ~
1 Kgs
ax@Yw} ~ w{Tr+Da
~ B
^ = wyn`P* fl#Yw` ~
19:13
hN}hw! + hr`uM
* h
= ^ jt^P# dm)u&Yw~ ~
.WhY`la
! @ hp) ;L=-hm^ rm#aY{w~ loq wyl*a@
364
When Jezebel heard that Naboth
had been stoned and was dead,
Jezebel said to Ahab, Arise,
take possession of the vineyard
of Naboth, the Jezreelite, which
he refused to give you for
money; for Naboth is not alive,
but dead.
1 Kgs
21:15
yl!au@rz+ Y+ h
] ^ tobn` <r\K-# ta#
yj^ tobn` /ya@ yK! [s#kb
# = ;l=-tt#l* /a@m@ rv#a&
.tm@-yK!
Pc-vqw3msXa Pp-vqc np PpvPp3ms np Pc-vqw3msXa Pcvqw3fs np Pp-np vqvms vqvms Poncms np Pa-np Pr vpp3ms Pp-vqcPpX2ms Pp-ncms Pp Pd np ams Ppvqp3ms
1 Kgs
21:16
.w{Tv=rl
] = yl!au@rz+ Y+ h
] ^
Pc-vqw3msXa Pp-vqc np Pp
vqp3ms np Pc-vqw3ms np Pp-vqc
Pp-ncms np Pa-np Pp-vqcX3ms
hL#ah
@ * <yr]bD
* h
+ -^ ta# ba*j=a^ u~mv
) k
= ! yh!yw+ ~
1 Kgs
21:27
bk#rh
\ * yr}c* toar+K! yh!yw+ ~
1 Kgs
.wyr`ja
& m
^ @ WbWvY`w~ aWh la@rc
` y= ] El#m-# aO-yK!
22:33
ba*ja
= ^ tomK= yh!yw+ ~
.la@rc
` y= ] El#mB
# = ba*om-El#m# uv^pY= w] ~
2 Kgs 3:5
365
When the vessels were full, she
said to her son, Bring me
another vessel. And he said to
her, There is not one vessel
more. And the oil stopped.
<yl!Kh
@ ^ taOm=K! yh!yw+ ~
yl!K# dou yl^a@ hv*yG]h^ Hn`B-= la# rm#aT)w~
h*yl#a@ rm#aY{w~
./m#Vh
* ^ dm)uY& w~ ~ yl!K# dou /ya@
2 Kgs 4:6
lm#rK
+ h
^ ^ rh^-la#
2 Kgs
4:25
dg\Nm
\ ! Ht*a) <yh!Oa$h-* vya! toar+K! yh!yw+ ~
.zL*h^ tyM!nW~ Vh^ hN}h! w{ru&n~ yz]jy& G}-la# rm#aY{w~
rp#Sh
@ -^ ta# la@rc
` y= -] El#m# ar{qK
= ! yh!yw+ ~
2 Kgs 5:7
yl^a@ j~lv
@ ) hz\-yK! toyj&hl
^ =W tym!hl
* =
an`-WuD+-Ea^ yK! w{Tu=rX
~ m
* ! vya! [s)al
$ #
.yl! aWh hN\at
^ m
= -! yK! War+W
2 Kgs 5:8
366
When they had come into
Samaria, Elisha said, O
LORD, open the eyes of these
men, that they may see. So the
LORD opened their eyes and
they saw; and behold, they were
in the midst of Samaria.
/orm=v) <a*bK
) = yh!yw+ ~
hL#a-@ yn}yu@-ta# jq^P= hw`hy+ uv*yl!a$ rm#aY{w~
2 Kgs
6:20
hV*ah
! * yr}bD
= -] ta# El#Mh
# ^ u~mv
) k
= ! yh!yw+ ~
hm*jh
) -^ lu^ rb@u) aWhw+ wyd`gB
` -= ta# ur~qY= w] ~
2 Kgs
6:30
.ty]Bm
* ! w{rc*B-= lu^ qC^h^ hN}hw! + <u*h* ar+Yw~ ~
Pc-vqw3msXa Pp-vqc Pa-ncms Poncmpc Pa-ncfs Pc-vqw3ms PoncmpcX3ms Pc-pi3ms vqPms PpPa-ncfs Pc-vqw3msXa Pa-ncms PcPi Pa-ncms Pp-ncmscX3ms Ppncms
El#Mh
# -^ la# <yh!Oa$h* vya! rB@dK
~ = yh!yw+ ~
tl#s-) ha*sW= lq#vB
# = <yr]u)c= <y]ta
^ s* rm)al@
2 Kgs
7:18-19
./orm=v) ru^vB
^ = rj*m* tu@K* hy\hy= ] lq#vB
# =
rm^aY{w~ <yh!Oa$h* vya!-ta# vyl!Vh
* ^ /u^Yw~ ~
.<yB!rh
~ * h*yp#vk
* W= ;M=a! lb#zy\ a! yn}Wnz+-du^
2 Kgs
9:22
367
When the letter came to them,
they took the king's sons and
slaughtered them, seventy
persons, and put their heads in
baskets, and sent them to him at
Jezreel.
Pc-vqw3msXa Pp-vqc Pa-ncms
PpX3mp Pc-vqw3mp Po-ncmpc Pancms Pc-vqw3mp amp ncms Pcvqw3mp Po-ncmpcX3mp Pp+Pancmp Pc-vqw3mp PpX3ms npXd
<h#yl@a& rp#Sh
@ ^ ab)K= yh!yw+ ~
vya! <yu!bv
= ! Wfj&vY= w] ~ El#Mh
# ^ yn}B-= ta# Wjq=Yw] ~
2 Kgs
10:7
hl*uh
) * tocu&l^ w{tOk^K= yh!yw+ ~
2 Kgs
<yv!lV
! l
* w^ + <yx!rl
` * aWhy} rm#aY{w~
10:25
WhY`qz! j
+ ! El#Mh
# ^ u~mv
) K
= ! yh!y+w~
tyB@ ab)Yw` ~ qC*B^ sK^tY= w] ~ wyd`gB
` -= ta# ur~qY= w] ~
2 Kgs
19:1
.hw`hy+
2 Kgs
.wyd`gB
` -= ta# ur~qY= w] ~
22:11
WhY`qz! j
+ ! El#Mh
# ^ u~mv
) K
= ! yh!y+w~
tyB@ ab)Yw` ~ qC*B^ sK^tY= w] ~ wyd`gB
` -= ta# ur~qY= w] ~
.hw`hy+
Isa 37:1
368
When Jeremiah finished
speaking all that the LORD had
commanded him to speak to all
the people, the priests and the
prophets and all the people
seized him, saying, You must
die!
rB@dl
~ = Why`mr
= y+ ] toLk^K= yh!yw+ ~
Jer 26:8
<t*ar`ql
= ! hy`nt
+ n^ -+ /B# lau@mv
* y= ] ax@Yw} ~
Jer 41:6
Jer 41:7
lau@mv
* y= -] ta# rv#a& <u*h-* lK* toar+K! yh!yw+ ~
yr}c-* lK* ta@w+ j~rq
} -* /B# /n`jo* y-ta#
.Wjm*cY= w] ~ w{Ta! rv#a& <yl!yj
` h
& ^
Jer 41:13
369
When Jeremiah, whom the
LORD their God had sent,
finished telling all the people all
the words of the LORD their
Godthat is, all these words
Azariah the son of Hoshaiah,
and Johanan the son of Kareah,
and all the arrogant men said to
Jeremiah.
Pc-vqw3msXa Pp-vpc np Pp-vpc
Pp-ncmsc-Pa-ncms Po-ncmscncmpc np ncmpcX3mp Pr
vqp3msX3ms np ncmpcX3mp
PpX3mp Po ncmsc-Pa-ncmp Pa-acp
Pc-vqw3ms np ncmsc-np Pc-np
ncmsc-np Pc-ncmsc-Pa-ncmp Paamp
Jer 43:12
vm#Vh
# ^ j~rz{ K
+ ! yh!yw+ ~
Jonah 4:8
Ruth
1:19
hm*rh
` -* ta# tonB=m! lD~jY= w\ ~ av*uB
= ^ u~mv
) K
= ! yh!yw+ ~
.w{Tk=al^m-= ta# tB@vY= w~ ~
2 Chr
16:5
370
When the captains of the
chariots saw that it was not the
king of Israel, they turned back
from pursuing him.
bk#rh
\ * yr}c* toar+K! yh!yw+ ~
2 Chr
.wyr`ja
& m
^ @ Wbv%Yw` ~ la@rc
` y= ] El#m# hy`h-* aO yK!
18:32
ba*ja
= ^ tyB@-<u! aWhy} fp@Vh
* K
! = yh!yw+ ~
2 Chr
22:8
.<g}rh
+ Y^ w~ ~ Why`zj
+ a
^ l
& ^ <yt!r+vm
* = Why`zj
+ a
^ &
hr`oTh^ yr}bD
= ] ta@ El#Mh
# ^ u~mv
) K
= ! yh!yw+ ~
2 Chr
.wyd`gB
` -= ta# ur~qY= w] ~
34:19
Comments regarding the verbs which occur in these temporal expressions are
found in the summary in 9.3.2.2.1.5.
9.3.2.2.1.2 Infinitive Construct Followed by WAYYIQTOL with Intervening Clauses
In sixteen cases, an intervening clause with w+ occurs after the yh!yw+ .~ Certain
syntactic features of some of these intervening clauses will be discussed at the end of this
section.
When he drew back his hand,
behold, his brother came out.
Then she said, What a breach
you have made for yourself! So
he was named Perez.
Pc-vqw3msXa Pp-vhPms
ncfscX3ms Pc-Pi vqp3ms
ncmscX3ms Pc-vqw3fs pii-vqp2ms
PpX2ms ncms Pc-vqw3ms
ncmscX3ms ncms
w{dy` byv!mK
@ = yh!yw+ ~
wyj!a* ax*y` hN}hw! +
Jr\P* ;yl#u* T*xr
= P
~ -* hm^ rm#aT)w~
.Jr\P* w{mv= ar`qY= w] ~
Gen
38:29
371
When she spoke to Joseph day
after day, he did not listen to her
to sleep with her or be with her.
One such day, he went into the
house to do his work, and none
of the men of the household was
there inside.
Pc-vqw3msXa Pp-vpcX3fs Pp-np
ncms ncms Pc-Pn-vqp3ms PpX3fs
Pp-vqc PpX3fs Pp-vqc PpX3fs Pcvqw3msXa Pp-Pa-ncms Pa-ams Pcvqw3msXa Pa-ncmsXd Pp-vqc
ncfscX3ms Pc-Pd ncms Pp-ncmpc
Pa-ncms Pd Pp+Pa-ncms
Gen
39:10-11
ht*yB
+ h
^ ^ ab)Yw` ~ hZ\h^ <oYh^K= yh!yw+ ~
ty]Bh
^ ^ yv@na
+ m
^ @ vya! /ya@w+ w{Tk=al^m= tocu&l^
.ty]BB
* ^ <v*
Ev#jh
) ^ EoTm! loQh^-ta# <k#um
& v
= K
* = yh!yw+ ~
yl^a@ /Wbr+qT
= w! ~ va@B* ru@B) rh*hw* +
.<k#yn}qz= w] + <k#yf@bv
= ! yv@ar`-lK*
Wnyh@Oa$ hw`hy+ Wna*rh
+ # /h@ Wrm=aT)w~
Wnu=mv
^ * w{lq)-ta#w+ w{ld+G-` ta#w+ w{db)K-= ta#
Wnya!r` hZ\h^ <oYh^ va@h* EoTm!
.yj*w` <d`ah
* -* ta# <yh!Oa$ rB@dy~ -+ yK!
Deut
5:20-21
(23-24)
372
And it was so, that when Joshua
had spoken to the people, the
seven priests carrying the seven
trumpets of rams' horns before
the LORD went forward and
blew the trumpets; and the ark
of the covenant of the LORD
followed them.
Pc-vqw3msXa Pp-vqc np Pp-Pancms Pc-ams Pa-ncmp vqPmp ams
ncmpc Pa-ncmp Pp-ncbpc np
vqp3cp Pc-vqp3cp Pp+Pa-ncmp Pcncbsc ncfs np vqPms PdX3mp
Josh 6:8
torp=ov hu*bv
= ! <ya!cn= { <yn]hK
& h
) ^ hu*bv
= w! +
torp*oVB^ Wuq=tw* + Wrb=u* hw`hy+ yn}pl
= ! <yl!bo= Yh^
.<h#yr}ja
& ^ El@h) hwh
` y+ tyr]B= /ora&w~
tj*a# <u^P^ [Q@h^ ryu!h-* ta# hw`hy+-/ora& bS@Yw~ ~
6:11
.hn\jM
& B
^ ^ Wnyl!Yw` ~ hn\jM
& h
^ ^ Wab)Yw` ~
la@rc
` y= ] yn}bW= u~vo% hy+ toLk^K= yh!yw+ ~
<M*T-% du^ da)m-= hl*odg+ hK*m^ <t*oKh^l=
Josh
10:20
hl*uh
) * tolu&hl
^ = w{tOk^K= yh!yw+ ~
1 Sam
13:10
.w{kr&bl
* = w{tar`ql
= ! lWav* ax@Yw} ~
lWav*-la# rB@dl
~ = w{tOk^K= yh!yw+ ~
dw]D` vp#nB
\ = hr`vq
= n= ] /t*no` hy+ vp#nw\ +
.w{vp=nK
~ = /t*no` hy+ w{bh*aY$ w\ ~
1 Sam
18:1
373
As soon as he had finished
speaking, behold, the king's sons
came and lifted their voices and
wept; and also the king and all
his servants wept very bitterly.
WaB* El#Mh
# -^ yn}b= hN}hw! + rB@dl
~ = w{tOk^K= yh!yw+ ~
2 Sam
El#Mh
# -^ <g~w+ WKb=Yw] ~ <l*oq Wac=Yw] ~
13:36
1 Kgs
12:2
.<y]rx
` m
= B
! = <u*br
= y` ` bv#Yw} ~ hm)Ov=
hZ\h^ rb*Dh
` -^ ta# u~mv
) K
= ! yh!yw+ ~
1 Kgs
toKS%B^ <yk!lM
* h
= w^ + aWh ht#v) aWhw+
20:12
1 Kgs
aWh la@rc
` y= -] El#m# Ea^ Wrm=a* hM*hw@ +
22:32
374
When they had crossed over,
Elijah said to Elisha, Ask what
I shall do for you before I am
taken from you. And Elisha
said, Please, let a double
portion of your spirit be upon
me.
2 Kgs 2:9
EM*um
! @ jq^La
* # <r\fB
# = EL*-hc#ua
$ # hm* la^v=
uv*yl!a$ rm#aY{w~
.yl*a@ ;j&WrB= <y]nv
~ -= yP! an`-yh!yw]
loka$l# <yv!na
` l
& ^ Wqx=Yw] ~
Wqu*x* hM*hw@ + dyz]Nh
` m
^ @ <l*k=aK
* = yh!yw+ ~
2 Kgs
4:40
hl*Pa
= w# ` yn]a* ra^va
& n}w+ <t*oKh^K= yh!yw+ ~
Ezek 9:8
hw]hy+ yn`da
{ & Hh*a& rm^aw) ` qu^za
+ w# ` yn~P-* lu^
tyr]av
@ -= lK* ta@ hT*a^ tyj!vm
= h
^ &
.<*l
! v*Wry+-lu^ ;t=mj
* -& ta# ;K=pv
= B
* = la@r`cy= ]
tm@ hy`nB
` -= /b# Why`fl
= p
^ W= ya!bN= h
` K
! = yh!yw+ ~
Ezek
lodG`-loq qu^za
+ w# ` yn~P-* lu^ lP)aw# `
11:13
hw]hy+ yn`da
{ & Hh*a& rm^aw) `
.la@rc
` y= ] tyr]av
@ = ta@ hc#u) hT*a^ hl*K*
375
When Jeroboam the son of
Nebat heard of it (for he was in
Egypt where he had fled from
the presence of King Solomon),
Jeroboam returned from Egypt.
2 Chr
10:2
.<y]rx
` M
= m
! ! <u*br
= y` ` bv*Yw` ~
2 Chr
aWh la@rc
` y= ] El#m# Wrm=a* hM*hw@ +
18:31
The following chart displays the type of intervening clause which occurs in this
set of data. Detailed analysis and discussion of each of the intervening clauses in the
previous set of data is beyond the scope of the present study, but one representative
occurrence from each category will be briefly discussed here.
hnhw QATAL
Gen 38:29
1 Sam 13:10
2 Sam 13:36
awhw
|
WE-X-QATAL WE-al-QATAL
Josh 6:8
1 Sam 18:1
1 Kgs 22:32
2 Kgs 2:9
2 Kgs 4:40
Ezek 11:13
2 Chr 18:31
Gen 39:10
LOC
1 Kgs 12:2
2 Chr 10:2
w-PTC
Ezek 9:8
PTC
1 Kgs 20:12
376
had not been used after this temporal expression, the following clause
would most likely be wyj!a* ax@Yw} .~
In the NASB, 2 Kgs 2:9 is rendered as follows: When they
had crossed over, Elijah said to Elisha In this version, rm^a* WhY`la
! w@ +
! @ rm#aY{w.~ The previous set of seventy
reads just as if it were WhY`la
examples in 9.3.2.2.1.1 is ample evidence that a WAYYIQTOL can follow
this type of temporal expression, so the logical question is why the QATAL
occurs in examples like 2 Kgs 2:9. One of the motivations for this question
is that differences between QATAL and WAYYIQTOL appear to be
significant elsewhere. There is, of course, much more to the WE-X-QATAL
than just the occurrence of the QATAL; it involves the matter of clause
positions and the order and type of elements which occur in them.
Analysis of the WE-X-QATAL involves describing the function of w+ prefixed
to a non-verbal item in clause initial position.
WE-X-QATAL:
WE-al
alal QATAL:
377
The issue in each case is how to describe the function that the clause performs.
These examples underscore the necessity of clearly understanding the syntactic patterns
and narrative functions of the various biblical Hebrew clauses.
9.3.2.2.1.3 Infinitive Construct Followed by QATAL
In seven cases, the yh!yw+ ~ + INFC is followed by a QATAL instead of the more typical
WAYYIQTOL.
The use of the QATAL in each of these examples indicates that the event or
state of affairs expressed in the clause with QATAL is anterior to the temporal expression.
Note that 2 Chr 12:1 is another example of the parallel -K=W -K= yh!yw+ .~
When David came to Mahanaim,
Shobi the son of Nahash from
Rabbah of the sons of Ammon,
Machir the son of Ammiel from
Lo-debar, and Barzillai the
Gileadite from Rogelim, had
brought beds, basins, pottery,
wheat, barley, flour, parched
grain, beans, lentils, parched
seeds, honey, curds, sheep, and
cheese of the herd, for David
and for the people who were
with him, to eat; for they said,
The people are hungry and
weary and thirsty in the
wilderness.
Pc-vqw3msXa Pp-vqc np npXd Pcnp ncmsc-np Pp-np ncmpc-np Pc-np
ncmsc-np Pp np Pc-np Pa-np Pp-np
ncms Pc-ncfp Pc-ncms vqPms Pcncfp Pc-ncfp Pc-ncms Pc-ncms Pcncms Pc-ncmp Pc-ncms Pc-ncms
Pc-ncfs Pc-ncbs Pc-ncfsc ncms
vhp3cp Pp-np Pc-Pp+Pa-ncms PrPpX3ms Pp-vqc Pp vqp3cp Pa-ncms
ams Pc-ams Pc-ams Pp+Pa-ncms
hm*yn+ j
` m
& ^ dw]d` aobK= yh!yw+ ~
/oMu^-yn}B= tB^rm
~ @ vj*n-` /b# yb!vw) +
rb*d+ aOm! la@yM!u-^ /B# ryk!mW*
.<yl!gr
+ m
{ @ yd]ul
* G= h
] ^ yL^zr
] +bW^
<yr]uc
) W= <yF!jw! + rx@oy yl!kW= toPs^w+ bK*vm
= !
.yl!qw* + <yv!du
` w& ~ lopW yl!qw* + jm^qw# +
2 Sam
17:27-29
378
When Solomon finished praying
this entire prayer and
supplication to the LORD, he
had arisen from before the altar
of the LORD, from kneeling on
his knees with his hands spread
toward heaven.
hw`hy+-la# lL@Pt
^ h
= l
! = hm)Ov= toLk^K= yh!yw+ ~
<q* taZ{h^ hN`jT
! h
= w^ + hL*p!Th
= -^ lK* ta@
1 Kgs
8:54
wyK*rB
+ -! lu^ u~rK
{ m
= ! hw`hy+ jB^zm
+ ! yn}pL
= m
! !
.<y]mV
* h
* ^ tocr|P= wyP*kw^ +
Pc-vqw3msXa Pp-vpc np Pp-vtc Ppnp Po ncmsc-Pa-ncfs Pc-Pa-ncfs Paafs vqp3ms Pp-Pp-ncbpc ncmsc np
Pp-vqc Pp-ncfdcX3ms PcncfdcX3ms vqsfp Pa-ncmp
<u*br
= y` ` tyB@-lK*-ta# hK*h! w{kl=mk
* = yh!yw+ ~
1 Kgs
<u*br
= y` l
` = hm*vn* -+ lK* rya!vh
= -! aO
15:29
hw`hy+ rb^dK
+ ! w{dm!vh
= -! du^
.yn]OyV!h^ hY`ja
! & w{Db=u-^ dyB
~ = rB#D] rv#a&
Pc-vqw3msXa Pp-vqcX3ms
vhp3ms Po-ncmsc-ncmsc np Pnvhp3ms ncmsc-ncfs Pp-np Ppvhp3msX3ms Pp-ncmsc np Pr
vpp3ms Pp-ncfsc-ncmscX3ms np
Pa-np
hL#ah
@ * <yr]bD
* h
+ -^ ta# yu!m=vK
* = yh!yw+ ~
Neh 1:4
.<y]mV
* h
* ^ yh@Oa$ yn}pl
= ! lL@P^tm
= W!
Pc-vqw3msXa Pp-vqcX1cs Po-Pancmp Pa-acp vqp1cs Pc-vqw1cs Pcvtw1cs ncmp Pc-vqw1csXa vqPms
Pc-vtPms Pp-ncbpc ncmpc Pa-ncmp
hK*lM
= h
^ ^ rT@sa
= -# ta# El#Mh
# ^ toar+k! yh!yw+ ~
wyn`yu@B= /j@ ha*cn= ` rx@jB
* # td\mu
# )
bh*Zh
` ^ fyb!rv
+ -^ ta# rT@sa
= #l= El#Mh
# ^ fv#oYw~
var{B= uG~Tw! ~ rT@sa
= # br~q=Tw! ~ w{dy`B= rv#a&
.fyb!rV
+ h
^ ^
Esth 5:2
379
When they had heard all the
words, they turned in fear one to
another and said to Baruch,
We will surely report all these
words to the king.
<yr]bD
* h
+ -^ lK*-ta# <u*mv
= K
* = yh!yw+ ~
Jer 36:16
.hL#ah
@ * <yr]bD
* h
+ -^ lK* ta@ El#Ml
# ^ dyG]n~
w{tq*zj
+ k
# W= <u*bj
= r
^ + tWkl=m^ /yk!hK
* = yh!yw+ ~
.w{Mu! la@rc
` y= -] lk*w+ hw`hy+ tr~oT-ta# bz~u*
2 Chr
12:1
hu*Br
* a
+ w^ + totl*D+ vOv* yd]Why+ aorq=K! yh!yw+ ~
Jer 36:23
El@vh
= w^ + rp@Sh
) ^ ru^tB
^ = h*u#rq
` y= ]
<T)-du^ ja*h-* la# rv#a& va@h-* la#
.ja*h-* lu^ rv#a& va@h-* lu^ hL*gM
] h
= -^ lK*
380
not take place one time, but rather each time three or four columns were read, the king
would cut the scroll. It is the peculiar narrative depiction of these events that motivates
the use of the YIQTOL.
9.3.2.2.1.5 Summary of Infinitive Constructs Used with K= yh!yw+ ~
The verbs occurring with K= yh!yw+ ~ are: hlk, axy, har, umv, bWv, rbd, <Wr, jWn,
rma, awb, rkz, hlu, rbu, tWm, alm, lka, arq, vgp, hkn, abn, and jrz. As stated
previously, the verbs occurring with B= and K= will be compared here. The following chart
lists the verbs to facilitate comparison. It is important to keep in mind that this chart only
lists the occurrences attested with yh!yw+ ;~ there are other occurrences without yh!yw+ ~ which are
not contemplated here.
Only B=
Hyh
ucn
tjv
hvq
dly
dry
/kv
ilv
lhq
zzg
zpj
rmv
trk
rzu
hwx
Both B= and K=
awb
axy
har
hlu
rbd
umv
Only K=
hlk
bwv
<wr
jwn
rma
rkz
rbu
twm
alm
lka
arq
vgp
hkn
abn
jrz
381
This is an important claim to test. To state that the choice between these
prepositions is motivated by the aspectual nature of the verbs, assumes that the verbs
have an inherent aspect. Even though there are only six verbs listed above that occur
with both B= and K=, this is sufficient evidence to question the claim by Lambdin. Also, to
state that K= is by far the preferred preposition over B= is not decisive. An important
principle with a limited corpus like that of biblical Hebrew is that what is attested does
not define what is possible or grammatical. The most decisive situation is if a certain
verb always occurs with only one preposition, but if two different prepositions can
potentially occur with the same verb, then the factors which motivate the choice of
preposition are not aspectual features inherent to the verb, but rather are part of the
textual context. As Bybee comments:
the function of aspect is to allow the temporal dimensions of a situation
to be described from different points of view depending on how the
situation is intended to fit into the discourse. (Bybee 1985, 142)
This is based on a definition of aspect as the internal temporal shape of states or
events (Payne 1997, 238), with the most basic distinction being between perfective and
imperfective. There are other semantic features of certain verbs, such as verbs of
sensation or mental perception which can also influence their usage. For example, Givn
claims that four major groups of verbs exist: compact, accomplishment, activity, and
stative (Givn 2001, 287-88). According to Givn, compact verbs depict temporally
compact events of extremely short duration and are at one extreme of the perfectivity
scale (Givn 2001, 287-88), but this does not necessarily prevent all compact verbs from
being used in imperfective situations in narrative. For example, Lambdins claim that K=
382
occurs more frequently than B= with umv, may seem to fit a presumed notion of hearing
being a compact event. u~mv
) K
= ! when he heard in 1 Kgs 5:21 is an example of umv in
what seems like a compact event. The sense of K= as temporal succession seems more
appropriate than the temporal overlap indicated by B=.
When Hiram heard the words of
Solomon, he rejoiced greatly
and said, Blessed be the LORD
today, who has given to David a
wise son over this great people.
hm)Ov= yr}bD
= -] ta# <r`yj! u~mv
) K
= ! yh!yw+ ~
<oYh^ hw`hy+ EWrB* rm#aY{w~ da)m= jm^cY= w] ~
1 Kgs
5:21
(5:7)
1 Sam
11:6
383
discourse (Bybee 1985, 142). The aspectual nuance of the verb and the choice of
preposition to accompany it are both context sensitive. Therefore, rather than claim that a
particular preposition is used because of what seems to be the aspectual nature of the
verb, it is preferable to describe the context-sensitive features of the verb within the
temporal and aspectual dimensions of the narrative.
It is not possible in this study to analyze every occurrence of these prepositions
with the infinitive constructs, but comprehensive analysis of the temporal organization of
the biblical Hebrew text would require close attention to these details.
9.3.2.2.2
The following set has four examples. The first two are of K= used temporally,
indicating APPROXIMATION. This is an extension into the temporal realm of the usage of
K= as in <yr]uc
) = hp*ya@K= yh!yw+ ~ about an ephah of barley (Ruth 2:17), which is listed in
8.2.1.8.5.
About three months later Judah
was informed, Your daughterin-law Tamar has played the
harlot, and behold, she is also
with child by harlotry. Then
Judah said, Bring her out and
let her be burned!
Gen
38:24
<ym!Yh
` ^ tr\cu
# K
& ^ yh!yw+ ~
1 Sam
25:38
384
On one day like this, he went to
the house to do his work, and
none of the men of the household
was there in the house.
Pc-vqw3msXa Pp-Pa-ncms Pa-ams
Pc-vqw3msXa Pa-ncmsXd Pp-vqc
ncfscX3ms Pc-Pd ncms Pp-ncmpc
Pa-ncms Pd Pp+Pa-ncms
Gen
39:11
.ty]BB
* ^ <v* ty]Bh
^ ^ yv@na
+ m
^ @ vya! /ya@w+
The expression hZ\h^ <oYh^K= yh!yw+ ~ occurs only here in Gen 39:11, although hZ\h^
<oYh^K= without yh!yw+ ~ is found in Ezra 9:15, where the sense is as this day.
O LORD God of Israel, You are
righteous, for we have been left
an escaped remnant, as this day;
here we are before You in our
guilt, for no one can stand
before You because of this.
np ncmpc np ams pi2ms Pp-vnp1cp
ncfs Pp-Pa-ncms Pa-ams PiX1cp
Pp-ncbpcX2ms Pp-ncfscX1cp Pp Pd
Pp-vqc Pp-ncbpcX2ms Pp-afs
Ezra 9:15
In Gen 39:11, however, the use of hZ\h^ <oYh^K= yh!y+w~ gives the sense on one day like
this, referring to the type of day described in the context. Hamilton comments that [t]he
thrust of One such day (kehayym hazzeh) is something like as his custom was or as
usual (Hamilton 1995, 464). This prepares the reader to expect something out of the
ordinary on Josephs otherwise ordinary day.
9.3.2.3
With l=
Two temporal expressions with l= occur in connection with yh!yw+ :~
9.3.2.3.1 l= + Infinitive Construct
9.3.2.3.2 l= + Temporal Phrase
In GKC 102c, the meanings for l= are given as towards, (belonging) to, for
(GKC 1910, 298). In some of the following examples, towards seems to be the sense of
385
the TEMPORAL APPROXIMATION of l=. According to BHRG, this preposition has a very
unspecialized meaning. It is a preposition that indicates a very general relationship
between two entities (van der Merwe, Naud, Kroeze 1999, 284). Unfortunately,
BHRG does not discuss l= used temporally, so there are no examples to compare with the
following set of data. Williams 268, however, gives towards and by as possible
temporal senses of l= (Williams 1976, 48).
qj*xy= -] ta# <yq!a* yt!yr]B-= ta#w+
hr`c* ;l= dl@T@ rv#a&
Gen
17:21
.tr\ja
# h
^ * hn`VB
* ^ hZ\h^ du@oMl^
yv!yl!Vh
= ^ <oYl^ <yn]kn) + Wyh*w+
Exod
19:11
.yn`ys! rh^-lu^
The Exod 19:11 example provides good contrast between B= and l= used
temporally. The temporal use of l=, then, seems to be the indication of about, near, or by
the time specified. This indicates, contrary to BHRG, that l= does have a fairly
specialized meaning, at least when used temporally.
9.3.2.3.1
or in other words, that the action is about to take place: 1) when the sun
was about to go down and 2) when the gate was about to close. Both occurrences start
with the same structure:
yh!yw+ ~
NOUN-DEF
INFC-l=
386
aobl* vm#Vh
# ^ yh!yw+ ~
<r`ba
= -^ lu^ hl*pn= ` hm*Dr
} t
+ w^ +
Ev#jB
) ^ roGs=l! ru^Vh
^ ^ yh!y+w~
15:12
Josh 2:5
Wax*y` <yv!na
` h
& w* +
<yv!na
` h
& * Wkl=h* hn`a* yT!u=dy~ ` aO
9.3.2.3.2
Gen
1 Sam
1:20
387
About the turn of the year,
around the time when kings go
out to battle, David sent Joab
and his servants with him and
all Israel, and they destroyed the
sons of Ammon and besieged
Rabbah. But David stayed at
Jerusalem.
<yk!al=Mh
^ ^ tax@ tu@l= hn`Vh
* ^ tb^Wvt=l! yh!yw+ ~
w{Mu! wyd`bu
* -& ta#w+ ba*oy-ta# dw]D` jl^vY= w] ~
2 Sam
11:1
br\uh
# * tu@l= yh!yw+ ~
w{bK*vm
= ! lu^m@ dw]D` <q*Yw` ~
2 Sam
11:2
2 Sam
<y]rp
` a
= -# <u! rv#a& roxj*lu^bB
^ = <olv*ba
= l
^ =
13:23
.El#Mh
# ^ yn}B-= lk*l= <olv*b=a^ ar`qY= w] ~
hn`Vh
* ^ tb^Wvt=l! yh!yw+ ~
1 Kgs
20:26
.la@rc
` y= -] <u! hm*jl
* =Ml
! ^ hq*pa
@ & lu^Yw~ ~
388
About spring time, at the time
when kings go out to battle, Joab
led out the army and ravaged
the land of the sons of Ammon,
and came and besieged Rabbah.
But David stayed at Jerusalem.
And Joab struck Rabbah and
overthrew it.
<yk!lM
* h
= ^ tax@ tu@l= hn`Vh
* ^ tb^WvT= tu@l= yh!yw+ ~
ab*Xh
* ^ lyj@-ta# ba*oy gh^nY+ w] ~
1 Chr
20:1
Pc-vqw3msXa Pp-ncbs ncfsc Pancfs Pp-ncbs vqc Pa-ncmp Pcvqw3ms np Po-ncmsc Pa-ncbs Pcvhw3msXa Po-ncbs ncmpc-np Pcvqw3msXa Pc-vqw3msXa Po-np
Pc-np vqPms Pp-np Pc-vhw3ms np
Po-np Pc-vqw3msX3fs
Josh
10:27
389
By the time Solomon was old, his
wives had turned his heart away
after other gods; and his heart
was not wholly devoted to the
LORD his God, as the heart of
David his father had been.
1 Kgs
11:4
2 Chr
<]lv
^ W* ryw] hd`Why+-la# Wab)Yw` ~
24:23
2 Chr
21:19
tu@KW= + INFC
390
The more general temporal expression with -l= yh!y+w~ is further specified by the
second part of the complex construction. Compare 2 Kgs 3:20 in where further
specification takes place within a temporal expression initiated by B=.
In this set of examples, the QATAL indicates that the action or event expressed by
that verb had already occurred by the time of the temporal setting established by l= yh!yw+ .~
9.3.2.3.2.3 Followed by WE-X-QATAL
In only one case, the next verbal element is a WE-X-QATAL.
<ym!Yh
` ^ tu^bv
= l
! = yh!yw+ ~
Gen 7:10
.Jr\ah
* -* lu^ Wyh* lWBM^h^ ym@W
Analysis of this example requires an answer to the question of how the WE-XQATAL
functions in narrative. Based on the previous two sets of data, this example could
have been:
WAYYIQTOL
QATAL
.Jr\ah
* -* lu^ lWBM^h^ ym@ Wyh=Yw] ~
.Jr\ah
* -* lu^ lWBM^h^ ym@ Wyh*
<ym!Yh
` ^ tu^bv
= l
! = yh!yw+ ~
<ym!Yh
` ^ tu^bv
= l
! = yh!yw+ ~
The issue of narrative strategy involved in the choice between these options is an
important area for further research and is discussed in more detail in the next chapter.
9.3.2.4
With /m!
The final preposition to be considered here is /m!, which occurs in five different
types of occurrences:
9.3.2.4.1 With JQ@m!
( m
* !
9.3.2.4.2 With tr`jM
391
9.3.2.4.3 With <ym!Ym
` !
9.3.2.4.4 With <oYh^-/m!
9.3.2.4.5 Event-Referenced Use of /m!
The essential structure of all of these expressions is the same:
+ Noun Phrase
JQ@
tr`jm
( *
<ym!y`
<oYh^
+ /m!
yh!yw+ ~
In its spatial sense, /m! indicates movement from a particular location. Similarly, in
its temporal sense, /m! establishes a point in time which is the starting point for the next
event.
9.3.2.4.1
With JQ@m!
392
After some time, Cain brought
an offering to the LORD of the
fruit of the ground.
Gen 4:3
.hw`hyl^ hj*nm
+ ! hm*da
` h
& * yr]Pm
= ! /y]q^ ab@Yw` ~
<oy <yu!Br
* a
+ ^ JQ@m! yh!yw+ ~
Gen 8:6
Josh 3:2
.hn\jM
& h
^ ^ br\qB
# = <yr]fV
= h
) ^ Wrb=uY^ w~ ~
Josh 9:16
hn`v* <yu!Br
* a
+ ^ JQ@m! yh!yw+ ~
<L@va
^ w& ~ aN` hk*la
& @ El#Mh
# -^ la# <olv*ba
= ^ rm#aY{w~
2 Sam
15:7
./orb=jB
# = hw`hyl^ yT!rd
+ n~ -` rv#a& yr]dn+ -] ta#
1 Kgs
2:39
393
After a while the brook dried up,
because there was no rain in the
land.
1 Kgs
17:7
2 Kgs 8:3
<yT!vl
= P
! = Jr\am
# @ hV*ah
! * bv*Tw* ~
.Hd`c-* la#w+ Ht*yB@-la# El#Mh
# -^ la# qu)xl
= ! ax@Tw@ ~
Jer 13:6
.<v*-w{nm=fl
* = ;yt!yW]x! rv#a& roza@h-* ta#
<ym!y` tr\cu
# & JQ@m! yh!yw+ ~
Jer 42:7
.Why`mr
= y+ -] la# hw`hy+-rb^d+ yh!yw+ ~
<ym!y` tu^bv
= ! hx@qm
= ! yh!yw+ ~
Ezek
3:16
394
At the end of four hundred and
thirty years, to the very day, all
the hosts of the LORD had left
the land of Egypt.
Pc-vqw3msXa Pp-ncms amp ncfs
Pc-afs afp ncfs Pc-vqw3msXa Ppncfsc Pa-ncms Pa-ams vqp3cp
ncmsc-ncbpc np Pp-ncbs np
Exod
12:41
.<y]rx
` m
= ! Jr\am
# @ hw`hy+ toab=x-! lK* Wax=y`
hl*yl
+ * <yu!Br
* a
+ w^ + <oy <yu!Br
* a
+ ^ JQ@m! yh!yw+ ~
Deut
9:11
.tyr]Bh
= ^ tojl% <yn]ba
* h
& *
.lyl!Gh
` ^ Jr\aB
# = ryu! <yr]cu
= # <r`yj!l=
1 Kgs
9:10-11
395
In 2 Chr 8:1-2, the element which precedes the QATAL is more complex and
deserves more detailed analysis than is possible here. Briefly, the element in focus here
is: <t*a) hm)Ov= hn`B* hm)Ov=l! <r`Wj /t^n` rv#a& <yr]uh
* w# ,+ literally, the cities which Huram
* # rather than the
gave to Solomon, Solomon built them. The fronting of the object <yr]uh
subject <r`Wj invites further scrutiny, but the use of the QATAL indicates the inclusion of
its action within the time period established by the temporal expression with JQ@m! yh!yw+ .~
At the end of the twenty years in
which Solomon built the house
of the LORD and his own house,
he built the cities which Huram
had given to him, and settled the
sons of Israel there.
2 Chr
8:1-2
.la@rc
` y= ] yn}B-= ta# <v* bv#oYw~
396
After two years passed, Pharaoh
was dreaming, and behold, he
was standing by the Nile.
<ym!y` <y]tn^ v
` = JQ@m! yh!yw+ ~
Gen 41:1
.ra)yh
+ -^ lu^ dm@u) hN}hw! + <l@j) hu)rp
+ W^
9.3.2.4.2
With tr`jM
( m
* !
Notice that in all of these examples, the next verb is WAYYIQTOL, indicating that
the action of that verb is the next narrated event on the following day. The expression
with tr`jM
( m
* ! establishes the next day as the starting point for the WAYYIQTOL.
On the following day, the
firstborn said to the younger,
Behold, I lay last night with my
father; let us make him drink
wine tonight also; then you go in
and lie with him, that we may
preserve our family through our
father.
rm#aT)w~ tr`jM
( m
* ! yh!yw+ ~
vm#a# yT!bk
= v
^ -* /h@ hr`yu!Xh
= ^-la# hr`yk!Bh
= ^
Gen
19:34
hl*yL
+ h
^ -^ <G~ /y]y~ WNq#vn= ~ yb!a-* ta#
.ur~z` Wnyb!am
* @ hY\jn^ W+ w{Mu! yb!kv
= ! ya!bW)
tr`jM
( m
* ! yh!yw+ ~
Exod
18:13
.br\uh
* -* du^ rq#Bh
) -^ /m! hv#m-) lu^ <u*h* dm)uY& w~ ~
tr`jM
( m
* ! yh!yw+ ~
Exod
32:30
hl#ua
$ # hT*uw^ + hl*dg{ + ha*fj
* & <t#af*j& <T#a^
.<k#ta
= F^j^ du^B= hr`Pk
= a
^ & yl^Wa hw`hy+-la#
397
Now on the next day Moses went
into the tent of the testimony;
and behold, the rod of Aaron for
the house of Levi had sprouted
and put forth buds and produced
blossoms, and it bore ripe
almonds.
Num
17:23
.<yd]qv
@ = lm)gY+ w] ~ Jyx! Jx@Yw` ~ jr~p# ax@Yw{ ~
(17:8)
tr`jM
( m
* ! yh!yw+ ~
Judg 9:42
.El#my# b!al
& ^ WdG]Yw~ ~ hd\Ch
* ^ <u*h* ax@Yw} ~
j~Bz@ m
+ ! <v*-Wnb=Yw] ~ <u*h* WmyK!vY= w~ ~ tr`jM
( m
* ! yh!yw+ ~
Judg 21:4
.<ym!lv
* W= tolu) Wlu&Yw~ ~
tr`jM
( m
* ! yh!yw+ ~
1 Sam
11:11
rq#Bh
) ^ tr\mv
) a
= B
^ = hn\jM
& h
^ -^ Eotb= Wab)Yw` ~
<oYh^ <j)-du^ /oMu^-ta# WKY~w~
Wxp%Yw` ~ <yr]av
* N= h
] ^ yh!yw+ ~
.dj^y` <y]nv
~ = <b*-Wra&vn= ] aOw+
1 Sam
dw]dw` + ty]Bh
^ -^ Eotb= aB@nt
~ Y= w] ~ lWav*-la#
18:10
.lWav*-dy~B= tyn]jh
& w^ + <oyB= <oyK= w{dy`B= /G}nm
~ =
398
The next day, the second day of
the new moon, that David's
place was empty; so Saul said to
Jonathan his son, Why has the
son of Jesse not come to the
meal, either yesterday or
today?
yn]Vh
@ ^ vd\jh
) ^ tr~jM
( m
* ! yh!yw+ ~
1 Sam
20:27
<yT!vl
= p
! = Wab)Yw` ~ tr`jM
( m
* ! yh!yw+ ~
lWav*-ta# Wax=mY= w] ~ <yl!lj
* h
& -^ ta# fV@pl
^ =
1 Sam
31:8
.u~Bl
) G= h
] ^ rh^B= <yl!pn= { wyn`B* tv#Ov=-ta#w+
rB@kM
= h
^ ^ jQ^Yw] ~ tr`jM
( m
* ! yh!y+w~
wyn`P-* lu^ cr{pY= w] ~ <y]MB
^ ^ lB)fY= w] ~
2 Kgs
8:15
.wyT*jT
= ^ la@hz`j& EOm=Yw] ~ tm)Yw` ~
tr`jM
( m
* ! yh!yw+ ~
<yl!lj
* h
& -^ ta# fV@pl
^ = <yT!vl
= p
! = Wab)Yw` ~
1 Chr
10:8
lWav*-ta# Wax=mY= w] ~
.u~Bl
) G= ] rh^B= <yl!pn= { wyn`B*-ta#w+
tr`jM
( m
* ! yh!yw+ ~
tk#Ph
* M
= h
^ -^ /m! Why`mr
= y+ -] ta# rWjv=p^ ax@Yw{ ~
Why`mr
= y+ ] wyl*a@ rm#aY{w~
Jer 20:3
399
9.3.2.4.3
<yB!r~ <ym!Ym
` ! yh!yw+ ~
hw`hy+ j~yn]h-@ rv#a& yr}ja
& ^
byb!Sm
* ! <h#yb@ya
+ -) lK*m! la@rc
` y= l
] =
.<ym!YB
` ^ aB* /q@z` u~vo% hyw]
wyn`qz@ l
+ ! la@rc
` y= -] lk*l= u~vo% hy+ ar`qY= w] ~
rm#aY{w~ wyr`fv
= l
) W= wyf*pv
= )lW= wyv*ar`lW=
.<ym!YB
` ^ yt!aB* yT!nq
+ z^ ` yn]a& <h#la
@ &
<ym!Ym
` ! yh!yw+ ~
Josh 23:1
Judg 11:4
.la@rc
` y= -] <u! /oMu^-yn}b= Wmj&LY* w] ~
<yF!j-! ryx!q= ym@yB! <ym!Ym
` ! yh!yw+ ~
<yZ]u! yd]gB
+ ! w{Tv=a-! ta# /ovm=v! dq)pY= w] ~
hr`dj
+ h
* # yT!va
= -! la# ha*ba
) * rm#aY{w~
.aobl* h*yb!a* w{nt*n-+ aOw+
Judg 15:1
400
9.3.2.4.4
There are only two occurrences of yh!yw+ ~ with this temporal expression which
occurs in two forms: <oYh^m@ and <oYh^-/m!. In the first example, the temporal expression is
followed by a WAYYIQTOL.
From that day forward, he made
it a statute and an ordinance for
Israel to this day.
Pc-vqw3msXa Pp-Pa-ncms Papi3ms Pc-PdXd Pc-vqw3msX3fs
Pp-ncms Pc-Pp-ncms Pp-np Pp Pancms Pa-ams
hl*um
= w* ` aWhh^ <oYh^m@ yh!y+w~
1 Sam
qj)l= h*mc
# y! w+ ~
30:25
In the second example from Nehemiah, there is a complex description of the work
on the wall that precedes the WAYYIQTOL in 4:13a. The use of the PARTICIPLES and w+ with
NOUN
heightens the vivid description that leads into the action expressed by the
WAYYIQTOL.
401
From that day on, half of my
servants carried on the work
while half of them held the
spears, the shields, the bows and
the breastplates; and the
captains were behind the whole
house of Judah. Those who
were rebuilding the wall and
those who carried burdens took
their load with one hand doing
the work and the other holding a
weapon. As for the builders,
each wore his sword girded at
his side as he built, while the
trumpeter stood near me. I said
to the nobles, the officials and
the rest of the people, The work
is great and extensive, and we
are separated on the wall far
from one another.
<yr]Ch
* w^ + <yn]yr
{ V
+ h
! w^ + totv*Qh
= w^ + <yN]gM
] h
* ^
Neh
4:10-13
(16-19)
hm*ojh^-lu^ <yd]rp
` n= ] Wnj=n~aw& ~
.wyj!am
* @ vya! <yq!ojr+
9.3.2.4.5
In some of the temporal constructions with /m! yh!yw+ ~ , the reference is to an event
rather than a period of time. For example, in Gen 39:5, which is the only occurrence of
za*m@ yh!yw+ ,~ the reference is to when Joseph being put in charge of matters in the
Egyptians house. Likewise, in 1 Sam 7:2, the reference is to the event of placing the ark
402
in Kiriath-jearim. These expressions establish that event as the starting point from which
the action of the next verb takes place.
From the time he made him
overseer in his house and over
all that he owned, the LORD
blessed the Egyptian's house on
account of Joseph; thus the
LORD'S blessing was upon all
that he owned, in the house and
in the field.
Gen 39:5
<yr]uy* + ty~rq
+ B
! = /ora*h* tb#v# <oYm! yh!yw+ ~
hn`v* <yr]cu
= # Wyh=Yw] ~ <ym!Yh
` ^ WBr+Yw] ~
1 Sam
7:2
.hw`hy+ yr}ja
& ^ la@rc
` y= ] tyB@-lK* WhN`Yw] ~
The 1 Sam 7:2 example is curious in that the next event is the passing of many
days.
9.3.2.5
complex area of Hebrew syntax and deserves close attention. Throughout the analysis,
the focus sometimes seemed to be on yh!yw+ ,~ but at other times it would shift to the
preposition being considered. This is normal in analysis which is exploring contextual
factors that motivate linguistic choice. After repeated review of the data and further
reading of the biblical Hebrew text, it became apparent that full analysis of the role yh!yw+ ~
403
plays in these temporal expressions will necessarily involve analysis of the same
temporal expressions where they occur without yh!yw+ .~ This is a very important aspect of a
comprehensive analysis of yh!yw+ ,~ but exhaustive analysis is not feasible there. There is
further discussion of this issue, however, in Chapter 10.
With rv#aK
& ^ as Temporal
In 35 instances, rv#aK
& ^ follows yh!yw+ ~ with a temporal function. In all of these
examples, rv#aK
& ^ yh!yw+ ~ is consistently followed by a QATAL. rv#aK
& ^ yh!yw+ ~ has the sense of
just as, as soon as. In the following 34 occurrences, the verb which follows the temporal
expression is a WAYYIQTOL.
Just as he was coming to Egypt,
he said to Sarai his wife, See
now, I know that you are a
beautiful woman;
Pc-vqw3msXa Pp-Pr vhp3ms Ppvqc npXd Pc-vqw3ms Pp-np
ncfscX3ms Pi-Pi vqp1cs Pp ncfs
afsc-ncms pi2fs
hm*yr
+ x
` m
= ! aobl* byr]qh
= ! rv#aK
& ^ yh!yw+ ~
yT!ud
= y~ ` an`-hN}h! w{Tv=a! yr~c-* la# rm#aY{w~
Gen
12:11
.T=a* ha#rm
+ -^ tp^y+ hV*a! yK!
Gen
20:13
404
As soon as the camels finished
drinking, the man took a gold
ring weighing a half-shekel and
two bracelets for her wrists
weighing ten shekels in gold,
toTv=l! <yL!mG^ h
+ ^ WLK! rv#aK
& ^ yh!yw+ ~
yn}vW= w{lq*vm
= ! uq^B# bh*z` <z\n\ vya!h* jQ^Yw] ~
Gen
24:22
.<l*qv
* m
= ! bh*z` hr`cu
* & h*yd\y-` lu^ <yd]ym!x=
<h#yr}bD
= -] ta# <h*rb
` a
= ^ db#u# um^v* rv#aK
& ^ yh!yw+ ~
.hw`hyl^ hx*ra
+ ^ WjT^vY= w] ~
Gen
24:52
Gen
29:10
ra@Bh
= ^ yP! lu^m@ /b#ah
# -* ta# lg\Yw` ~ bq)uy& ~ vG~Yw] ~
.w{Ma! yj!a& /b*l* /ax)-ta# q=vY= w~ ~
Gen
30:25
.yx!ra
+ l
^ W= ym!oqm=-la#
Gen
37:23
405
As soon as they finished eating
the grain which they had
brought from Egypt, their father
said to them, Go back, buy us a
little food.
Pc-vqw3msXa Pp-Pr vpp3cp Ppvqc Po-Pa-ncms Pr vhp3cp Pp-np
Pc-vqw3ms PpX3mp ncmscX3mp
vqvmp vqvmp-PpX1cp amsc-ncms
rb#Vh
# -^ ta# lk)al
$ # WLK! rv#aK
& ^ yh!yw+ ~
Gen 43:2
<y]rx
` M
= m
! ! Wayb!h@ rv#a&
hn\jM
& h
^ -^ la# br~q* rv#aK
& ^ yh!yw+ ~
Exod
32:19
tj)Lh
% -^ ta# w{dY`m! El@vY= w~ ~
.rh*h* tjT
^ ^ <t*a) rB@vy^ w+ ~
hm*jl
* M
= h
! ^ yv@na
+ -^ lK* WMT^-rv#ak
& ^ yh!yw+ ~
.<u*h* br\Qm
# ! tWml*
Deut
2:16-17
yoGh^-lk* WMT^-rv#aK
& ^ yh!y+w~
Josh 4:1
/D}rY+ h
~ -^ ta# robu&l^
.rm)al@ u~vo% hy+-la# hw`hy+ rm#aY{w~
robu&l^ <u*h-* lK* <T^-rv#aK
& ^ yh!yw+ ~
Josh 4:11
.<u*h* yn}pl
= ! <yn]hK
& h
) w^ + hw`hy+-/ora& rb)uY& w~ ~
Pc-vqw3msXa Pp-Pr-vqp3ms
ncmsc-Pa-ncms Pp-vqc Pc-vqw3ms
ncbsc-np Pc-Pa-ncmp Pp-ncbpc Pancms
Josh 5:8
406
As soon as he finished
presenting the tribute, he sent
away the people who had
carried the tribute.
hj*nM
+ h
! -^ ta# byr]qh
= l
^ = hL*K! rv#aK
& ^ yh!yw+ ~
Judg 3:18
.hj*nM
+ h
! ^ ya@cn= { <u*h-* ta# jL^vy^ w+ ~
Judg 8:33
la@rc
` y= ] yn}B= WbWvY`w~
<yl!uB
* h
= ^ yr}ja
& ^ Wnz+Yw] ~
.<yh!Oal@ tyr]B= lu^B^ <h#l* Wmyc!Yw` ~
la@rc
` y= -] <u! /oMu^-yn}b= Wmj&ln= -] rv#aK
& ^ yh!yw+ ~
Judg 11:5
tj^ql
^ * du*lg= ] yn}qz= ] Wkl=Y}w~
.bof Jr\am
# @ jT*py= -] ta#
1 Sam
8:1
<yT!vl
= P
! = yr}ja
& m
^ @ lWav* bv* rv#aK
& ^ yh!yw+ ~
rB^dm
+ B
! = dw]d` hN}h! rm)al@ w{l WdG]Yw~ ~
1 Sam
24:1 (2)
.yd]G\ /yu@
2 Sam
<olv*ba
= -^ la#
16:16
<Ov*ba
= -^ la# yv^Wj rm#aY{w~
.El#Mh
# ^ yj!y+ El#Mh
# ^ yjy! +
407
As soon as the kingdom was
firmly in his hand, that he killed
his servants who had slain the
king his father.
w{dy`B= hk*lm
* M
= h
^ ^ hq*zj
+ * rv#aK
& ^ yh!yw+ ~
.wyb!a* El#Mh
# -^ ta# <yK!Mh
^ ^ wyd`bu
* -& ta# EY~w~
2 Kgs
14:5
1 Chr
17:1
<yz]ra
` h
& * tyb@B= bv@oy
.touyr]y+ tj^T^ hw`hy+-tyr]B= /ora&w~
wyl*u* hk*lm
* M
= h
^ ^ hq*zj
+ * rv#aK
& ^ yh!yw+ ~
.wyb!a* El#Mh
# -^ ta# <yK!Mh
^ ^ wyd`bu
* -& ta# gr{hY& w~ ~
2 Chr
25:3
fL^bn^ s
+ ^ um^v* rv#aK
& ^ yh!yw+ ~
Neh 3:33
(4:1)
Neh 4:1
(4:7)
408
As soon as the Jews who lived
near them came and told us ten
times, They will come up
against us from every place
where you may turn,I stationed
men in the lowest parts of the
space behind the wall, the
exposed places, and I stationed
the people in families with their
swords, spears, and bows.
<l*xa
= # <yb!vY= h
{ ^ <yd]WhY+h^ WaB*-rv#aK
& ^ yh!yw+ ~
<ym!uP
* = rc#u# Wnl* Wrm=aY{w~
.Wnyl@u* WbWvT*-rv#a& tomq)Mh
= -^ lK*m!
<oqM*l^ toYT!jT
= m
^ ! dym!ua
& w^ `
<u*h-* ta# dym!ua
& w^ ` <yY]jj
! X
! B
= ^ hm*ojl^ yr}ja
& m
^ @
.<h#yt@tV
) q
= w^ + <h#yj@mr
= ` <h#yt@br
) j
+ -^ <u! tojP*vm
= l
! =
Neh 4:9
(4:15)
Pc-vqw3msXa Pp-Pr-vqp3cp
vqPmpcX1cp Pp-vnp3ms PpX1cp
Pc-vhw3msXa Pa-ncmp PoncfscX3mp Pc-vqw1cp Pc-vqw1cp
ncmscX1cp Pp-Pa-ncfs ncms PpncfscX3ms
fL^bn^ s
+ l
^ = um^vn= ] rv#ak
& ^ yh!yw+ ~
yK! Wnyb@ya
+ ) rt#yl
\ W= yb!ru
+ ^h* <v#gl
\ W= hY`bo! fw+
HB* rt^on-aOw+ hm*ojh^-ta# yt!yn]b*
totl*D+ ayh!h^ tu@h-* du^ <G~ Jr\P*
.<yr]uV
* b
= ^ yT!dm
+ u
^ h
$ -# aO
rm)al@ yl^a@ <v#gw\ + fL^bn^ s
+ ^ jl^vY= w] ~
onoa tu^qb
= B
! = <yr]yp!KB
= ^ wD`jy= ~ hd`uW& n` w] + hk*l=
Neh 6:12
409
As soon as all our enemies
heard of it, and all the nations
surrounding us saw it, they lost
their confidence; for they
realized that this work had been
accomplished with the help of
our God.
Neh 6:16
totl*Dh
+ ^ dym!ua
& w^ ` hm*ojh^ ht*nb
+ n= ] rv#aK
& ^ yh!yw+ ~
.<Y]wl
] h
= w^ + <yr]rv
+ m
) h
= w^ + <yr]uo& Vh^ Wdq=PY* w] ~
Neh 7:12
Pc-vqw3msXa Pp-Pr vnp3fs Pancfs Pc-vhw1cs Pa-ncfp Pcvnw3mp Pa-ncmp Pc-Pa-vePmp PcPa-np Pc-vpw1cs Po-np ncmscX1cs
Pc-Po-np ncms Pa-ncfs Pp-np Pppi3ms Pp-ncms ncfs Pc-vqp3ms PoPa-ncmp Pp-amp
tB*Vh
^ ^ yn}pl
= ! <]lv
^ W* ry+ yr}u&v^ Wll&x* rv#aK
& ^ yh!yw+ ~
totl*Dh
+ ^ WrgS
+ Y* w] ~ hr`ma
= w) `
tB*Vh
^ ^ rj^a^ du^ <WjT*py= ] aO rv#a& hr`ma
= w) `
<yr]uV
* h
= -^ lu^ yT!dm
+ u
^ h
$ # yr~uN* m
+ W!
.tB*Vh
^ ^ <oyB= aC*m^ aoby`-aO
Neh
13:19
410
As soon as Zedekiah the king of
Judah and all the men of war
saw them, they fled and went out
of the city at night by way of the
king's garden through the gate
between the two walls; and he
went out toward the Arabah.
Pc-vqw3msXa Pp-Pr vqp3msX3mp
np ncms-np Pc-ncms ncmpc Pa-ncfs
Pc-vqw3mp Pc-vqw3mp ncms PpPa-ncfs ncbs ncbsc Pa-ncms Ppncms Pp Pa-ncfd Pc-vqw3ms ncbs
Pa-ncfs
hd`Why+-El#m# WhY`qd
! x
+ ! <a*r` rv#aK
& ^ yh!yw+ ~
Jer 39:4
hm*jl
* M
= h
! ^ yv@na
+ ^ lk)w+
Wjr+bY= w] ~
El#Mh
# ^ /G~ Er\D\ ryu!h-* /m! hl*yl
+ ^ Wax=Yw} ~
.hb*ru
` h
& * Er\D\ ax@Yw} ~ <y]t*mj
) h
) ^ /yB@ ru^vB
^ =
Gen
27:30
411
example, however, the next clause is a WE-X-QATAL, which is then followed by a
WAYYIQTOL
in 27:31. Two main narrative functions are operative here: 1) the interplay of
the preceding yh!yw+ ~ and the w+ of wc*uw@ + highlights the temporal parity of Jacob leaving and
Esau arriving; 2) the fronting of wc*u@ establishes him as the focal participant at this stage
in the narrative.
9.3.3.2
& ^ as Manner
With rv#aK
In the following three examples, the combination rv#aK
& ^ yh!yw+ ~ indicates manner
Wnl*-rt^P* rv#aK
& ^ yh!yw+ ~
.hl*t* w{ta)w+ yN]K-^ lu^ byv!h@ yt!a) hy`h* /K@
Gen
41:13
wyd`bu
* m
& @ <yv!na
` & hr`cu
* & /oud+G] jQ^Yw] ~
Judg 6:27
Zech
7:13
412
hm*da
` h
& * yn}P-= lu^ br{l* <d`ah
* * lj@h-@ yK! yh!yw+ ~
Gen 6:1
<d`ah
* * tonB=-ta# <yh!Oa$h-* yn}b= War+Yw] ~
lK)m! <yv!n` <h#l* Wjq=Yw] ~ hN`h@ tb)f) yK!
.Wrj*B* rv#a&
<ym!Yh
` ^ <v* w{l-Wkr+a* yK! yh!yw+ ~
Gen 26:8
Pc-vqw3msXa Pp vqp3cp-PpX3ms
Pd Pa-ncmp Pc-vhw3msXa np
ncms np Pp Pa-ncbs Pc-vqw3msXa
Pc-Pi np vpPms Pp np ncfscX3ms
ta)rm
+ @ wyn`yu@ /`yh#kT
= w! ~ qj*xy= ] /q@z-` yK! yh!yw+ ~
ld{Gh
` ^ w{nB= wc*u-@ ta# ar`qY= w] ~
.yn]Nh
} ! wyl*a@ rm#aY{w~ yn]B= wyl*a@ rm#aY{w~
Gen 27:1
413
When we came to the lodging
place, that we opened our sacks,
and behold, each man's money
was in the mouth of his sack, our
money in full. So we have
brought it back in our hand.
Gen
43:21
.Wnd}yB
` = w{ta) bv#Nw` ~ w{lq*v=mB
! = WnP@sK
= ^ w{Tj=Tm
^ a
= ^
yb!a* ;D+bu
= -^ la# Wnyl!u* yK! yh!yw+ ~
Gen
.yn]da
{ & yr}bD
= ] ta@ w{l-dG\Nw~ ~
44:24
Wnj@Lv
= l
^ = hu)rp
+ ^ hv*qh
= -! yK! yh!yw+ ~
Exod
rk)Bm
= ! <y]rx
~ m
= ! Jr\aB
# = rokB=-lK* ho`hy+ gr{hY& w~ ~
13:15
/K@-lu^ hm*hB
@ = rokB=-du^w+ <d`a*
<yr]kZ* h
+ ^ <j#r\ rf#P-# lK* hw`hyl^ j~bz@ { yn]a&
.hD\pa
= # yn~B* rokB=-lk*w+
Pc-vqw3msXa Pp-vhp3ms np PpvpcX1cp Pc-vqw3ms np ncmscncms Pp-ncbs np Pp-ncms ncms PcPp-ncms ncfs Pp-Pd pi1cs vqPms
Pp-np ncmsc-ncmsc ncms Pa-ncmp
Pc-ncmsc-ncms ncmpcX1cs vqi1cs
la@rc
` y= ] yn}B= Wqz+j* yK! yh!y+w~
sm^l* yn]un& K
~ h
= -^ ta# WnT=Yw] ~
Josh
17:13
.w{vyr]oh aO vr}ohw+
<c#Yw` ~ la@rc
` y= ] qz~j-* yK! yh!yw+ ~
.w{vyr]oh aO vyr}ohw+ sm^l* yn]un& K
~ h
= -^ ta#
Judg 1:28
414
when they cried out
Pc-vqw3msXa Pp-vqp3cp ncmpcnp Pp-np Pp ncfp np Pc-vqw3ms
np ncms ncms Pp-ncmpc np Pcvqw3ms PpX3mp Pd-vqp3ms np
ncmpc np pi1cs vhp1cs PoX2mp
Pp-np Pc-vhw1cs PoX2mp Ppncmsc ncmp
hw`hy+-la# la@rc
` y= -] yn}b= Wqu&z-` yK! yh!yw+ ~
./y`dm
+ ! toda) lu^
Judg 6:78
la@rc
` y= ] yn}B-= la# ayb!n` vya! hw`hy+ jl^vY= w] ~
yh@Oa$ hw`hy+ rm^a-* hK) <h#l* rm#aY{w~
<y]rx
~ M
= m
! ! <k#ta
= # yt!yl@uh
$ # yk!na
{ * la@rc
` y= ]
.<yd]bu
* & tyB@m! <k#ta
= # ayx!aw) `
<ym!Yh
` -^ lK* h*yr\bd
* b
+ ! w{L hq*yx!h-@ yK! yh!yw+ ~
.tWml* w{vp=n~ rx^qT
= w! ~ Whx@la
& T
^ w= ~
Judg
16:16
Judg
16:25
<yr]ys!ah
& * tyB@m! /ovm=vl
! = War+qY= w] ~
w{toa Wdym!uY& w~ ~ <h#yn}pl
= ! qj@xy^ w+ ~
.<yd]WMu^h* /yB@
2 Sam
6:13
2 Sam
7:1-2
415
Since he came from Jerusalem to
meet the king, the king said to
him, Why did you not go with
me, Mephibosheth?
El#Mh
# ^ tar~ql
= ! <]lv
^ W* ry+ ab*-yK! yh!yw+ ~
2 Sam
yM!u! T*kl
= h
^ -* aO hM*l* El#Mh
# ^ w{l rm#aY{w~
19:26
.tv#by) p!m=
(19:25)
2 Kgs
17:7
.<yr]ja
@ & <yh!Oa$ War+yY]w~ <y]rx
` m
= -! El#m#
hT#vM
= h
! ^ ym@y+ WpyQ!h! yK! yh!yw+ ~
Job 1:5
rq#BB
) ^ <yK!vh
= w! + <v@Dq
+ y^ w+ ~ boYa! jl^vY= w] ~
.<ym!Yh
` -^ lK* boYa! hc#uy& ~ hk*K*
9.3.4.2
the full discussion of this analysis, see Deictic Viewpoint in Biblical Hebrew Text: A
Syntagmatic and Paradigmatic Analysis of the Particle yk (ki) (2001). The concept of
metarepresentation is central to the following analysis:
Though the semantic content of the yk clause may be construed as a
reason for their banishing him, yk does not explicitly mark the reason per
416
se, but rather that the clause content is mentioned or metarepresented as a
correcting proposition relative to the speakers estimation of Jepthahs
thoughts/knowledge state. (Follingstad 2001, 266)
The most salient phrase here is the clause content is mentioned or
metarepresented. The unique contribution of yK! is that the contents of the clause
following yK! are communicated as a representation of the reason. From a slightly
different perspective, Arnold and Choi discuss the function of yK! as EVIDENTIAL:
Although translated similarly to the causal, the evidential use of yK!
presents the evidence or motivation that lies behind a statement, rather
than presenting the cause of an action or situation. Thus, the causal link is
with the action of speech, not the contents of speech; the focus is not on
what is spoken but on the reason the speaker is saying something. (Arnold
and Choi 2003, 149)
One of the motivations for looking for a meaning for yK! other than temporal when
is the fundamental principle of choice. The rule-governed, context-sensitive nature of
language indicates that the use of linguistic items like yK! is motivated by contextual
factors. The alternative is to say that yK! means when, but the same has been said in many
& .^ This is perhaps considered satisfactory when the text is
analyses of K=, B=, and rv#aK
approached from an atomistic, lexical equivalent approach, but if a set of examples with
K= are compared with another set with B=, another set with yK!, and yet another set with
rv#aK
& ^ and they all are translated when, it is only logical to ask what difference there
might be between them. The principle of choice, mentioned above, compels the analysis
to consider contextual motivations for the use of linguistic items like yK!. Rather than go
into an in-depth analysis of these occurrences here, the reader is referred to the volumes
by Follingstad and Arnold and Choi.
417
Followed by WAYYIQTOL
In the following examples, <oYh^ yh!yw+ ~ establishes a temporal point of reference
that is not directly linked chronologically as in the expressions like the next day or after
four days. The temporal reference is more general in these occurrences.
There was the day when Elkanah
offered sacrifices. He would give
portions to Peninnah his wife
and to all her sons and her
daughters;
<oYh^ yh!yw+ ~
w{Tv=a! hN`np
] l
= ! /t^nw` + hn`ql
* =a# jB^zY+ w] ~
1 Sam
1:4
<oYh^ yh!yw+ ~
lWav*-/B# /t*no` y rm#aY{w~
hr`Bu
= n= w~ + hk*l= wyl*k@ ac@n{ ru^Nh
~ -^ la#
zL*h^ rb#um
@ @ rv#a& <yT!vl
= !P= bX^m-^ la#
.dyG]h! aO wyb!al
* W=
1 Sam
14:1
418
There was the day when Elisha
passed over to Shunem, where
there was a prominent woman,
and she persuaded him to eat
food. As often as he passed by,
he turned in there to eat food.
Pc-vqw3msXa Pa-ncms Pc-vqw3ms
np Pp-np Pc-Pd ncfs afs Pcvhw3fsXa-PpX3ms Pp-vqc-ncms
Pc-vqw3msXa Pp-ncmsc vqcX3ms
vqi3ms PdXd Pp-vqc-ncms
<oYh^ yh!yw+ ~
2 Kgs 4:8
<oYh^ yh!yw+ ~
2 Kgs
hM*v* ab)Yw` ~
4:11
dl#Yh
` ^ lD~gY+ w] ~
.<yr]xQ
= h
) -^ la# wyb!a-* la# ax@Yw} ~
<oYh^ yh!yw+ ~
<oYh^ yh!yw+ ~
2 Kgs
4:18
Job 1:6
bX@yt
~ h
= l
! = <yh!Oa$h* yn}B= Wab)Yw` ~
.<k*otB= /f*Ch
* -^ <g~ aobY`w~ hw`hy+-lu^
<oYh^ yh!yw+ ~
Job 2:1
hw`hy+-lu^ bX@yt
~ h
= l
! = <yh!Oa$h* yn}B= Wab)Yw` ~
<k*tB
) = /f*Ch
* -^ <g~ aobY`w~
.hw`hy+-lu^ bX@yt
~ h
= l
! =
In all of these examples, the translation reads there was the day when This
reflects the use of <oYh^ yh!yw+ ~ to establish a new point of temporal reference, but as stated
previously, the reference is general in nature.
419
9.3.5.2
Followed by WE-X-QATAL
In only two instances, the temporal NOUN PHRASE is followed by WE-X-QATAL. In
the first case, <oYh^ yh!yw+ ~ is like the examples above, but the next constituent immediately
following it are two participial phrases initiated by w+s.
There was the day when his sons
and his daughters were eating
and drinking wine in their oldest
brother's house, a messenger
came to Job and said, The oxen
were plowing and the donkeys
feeding beside them,
Pc-vqw3msXa Pa-ncms PcncmpcX3ms Pc-ncfpcX3ms vqPmp
Pc-vqPmp ncms Pp-ncmsc
ncmscX3mp Pa-ncms Pc-ncms
vqp3ms Pp-np Pc-vqw3ms Pa-ncms
vqp3cp vqPfp Pc-Pa-ncfp vqPfp PpncfdcX3mp
<oYh^ yh!yw+ ~
Job 1:13
The two cases of the WE-X-QATAL here provide setting for the action of the
narrative which continues with rm^aY{w.~
In the second example, the temporal clause is <yB!r~ <ym!y` yh!yw+ .~ Based on the
present analysis, the WE-X-QATAL in this example indicates that the word of the LORD had
come prior to the end of the time period delimited by the temporal expression.
After many days, the word of the
LORD came to Elijah in the
third year, saying, Go, show
yourself to Ahab, and I will send
rain on the face of the earth.
Pc-vqw3msXa ncmp amp Pcncmsc-np vqp3ms Pp-np Pp+Pancfs Pa-afs Pp-vqc vqvms vnvms
Pp-np Pc-vqi1cs{1}Ct ncms Ppncbpc Pa-ncfs
1 Kgs
18:1
rf*m* hn`Ta
= w# + ba*ja
= -^ la# ha@rh
` @
.hm*da
` h
& * yn}P-= lu^
420
9.3.6.1 While: du^ + QATAL
9.3.6.2 Meanwhile: hK)-du^w+ hK)-du^
9.3.6.3 As often as: INFC + yD}m!
9.3.6.3 Only, just: Ea^
This makes it appear as if these temporal adverbs are not very frequent, but it
needs to be kept in mind that these occurrences are only when they co-occur with yh!yw+ .~
9.3.6.1
(1996, 723-25). These meanings are not randomly interchangeable since there are
contextual features that affect the sense that is appropriate. BDB states that the while
meaning for du^ is rare and unfortunately, 1 Sam 14:19 is the only case attested. BDBs
entry for 1 Sam 14:19 does not, however, include yh!yw+ ~ with du^. This makes it difficult to
determine the possible role of yh!yw+ ~ with du^. In contrast, there are numerous occurrences
of du^w+ used both spatially and temporally.
While Saul talked to the priest,
the commotion in the camp of
the Philistines continued and
increased; so Saul said to the
priest, Withdraw your hand.
Pc-vqw3msXa Pp vpp3ms np PpPa-ncms Pc-Pa-ncms Pr Pp-ncbsc
np Pc-vqw3msXa vqa Pc-ams Pcvqw3ms np Pp-Pa-ncms vqvms
ncfscX2ms
/h@Kh
) -^ la# lWav* rB#D] du^ yh!yw+ ~
1 Sam
14:19
421
9.3.6.2
as BDB states till now and till then (BDB 1996, 462).
In a little while the sky grew
black with clouds and wind, and
there was a heavy shower. And
Ahab rode and went to Jezreel.
1 Kgs
18:45
Pc-vqw3msXa Pp-Pd Pc-Pp-Pd PcPa-ncmp vtp3cp ncmp Pc-ncbs Pcvqw3msXa ncms ams Pc-vqw3ms
np Pc-vqw3msXa npXd
9.3.6.3
<yT!vl
= p
! = yr}c* Wax=Yw} ~
1 Sam
lWav* yd}bu
= ^ lK)m! dw]D` lk^c* <t*ax@ yD}m! yh!yw+ ~
18:30
<yx!rh
` * <WaC*y] hw`hy+ tyB@ El#Mh
# ^ ab)-yD}m! yh!yw+ ~
1 Kgs
.<yx!rh
` * aT*-la# <Wbyv!hw$ \
14:28
422
Now there came a day when
Elisha passed over to Shunem,
where there was a prominent
woman, and she persuaded him
to eat food. As often as he
passed by, he turned in there to
eat food.
<oYh^ yh!yw+ ~
2 Kgs 4:8
<yx!rh
` * WaB* hw`hy+ tyB@ El#Mh
# ^ aob-yD}m! yh!yw+ ~
2 Chr
.<yx!rh
` * aT*-la# <Wbv!hw$ \ <Wac*nW+
12:11
yD}m! is also used temporally in 1 Sam 1:7; 7:16; Isa 28:19; 66:23; Zech 14:16, but
these occurrences do not have yh!yw+ .~ Clause syntax and clause position are crucial factors
in the analysis of these examples. Further discussion of these factors is found in Chapter
10.
9.3.6.4
infinitive absolute then QATAL, giving the sense of the action having just taken place. As
stated in GKC 164b, footnote 1, the immediate succession is especially emphasized by
Ea^ and the infinitive absolute (GKC 1910, 501). See 9.3.3.1 for discussion of the first
yh!yw+ ~ in this example.
423
As soon as Isaac had finished
blessing Jacob, and Jacob had
just left the presence of Isaac his
father, Esau his brother came in
from his hunting.
Gen
27:30
CHAPTER 10
EXPLORING THE DISCOURSE-PRAGMATIC
USES OF yh!y+w~
10.1 Introduction
The preceding categorization of yh!yw+ ~ s verbal and temporal uses is the foundation
for exploring the discourse-pragmatic uses of yh!yw+ ~. This moves the analysis to another
dimension.
The analysis of yh!yw+ ~ in chapters eight and nine required excluding certain more
global considerations that need to be dealt with after the more basic syntactic patterns and
uses are clearly established. Within the analytical perspective that is proposed and
implemented here, the analysis of yh!yw+ ~ as an isolatable linguistic item is carried out with
constant awareness of the fact that full analysis eventually requires consideration of the
whole system of which it is a part. This chapter now explores other dimensions of the
uses and functions of yh!y+w.~
Before proceeding with the topic of this chapter, a brief restatement of the
significance of understanding the function of yh!yw+ ~ is in order. With regard to the verbal
uses, it is important to see the close syntactic connection it has as main verb in the clauses
424
425
where it occurs. yh!yw+ ~ s functions as a WAYYIQTOL are also important to keep in mind in
order to properly read it within its various contexts. Specifically with reference to its use
in temporal clauses, yh!yw+ ~ represents a class of linguistic item that, if translated literally,
gives essentially no indication of why it is in the Hebrew text and what its function is.
This gives the impression that it can be left untranslated with seemingly little loss in the
new translation.
Another example of this type of item would be hN}h!. This comparison, however, is
not meant to imply that hN}h! and yh!yw+ ~ perform similar linguistic functions. The
comparison is rather of the translation strategy for these items. If it is merely translated
directly as behold or even sometimes left untranslated, the end result for the reader of the
translation is essentially the same. Whether it is translated as behold or not represented in
the text, the reader is given very little help in understanding why hN}h! is in the text.
One of the most basic underlying principles of the analysis presented in this study,
however, is that linguistic items like yh!yw+ ~ and hN}h! are not just in the text with no purpose.
Research into the function of these linguistic items has the objective of probing the effect
they have on the understanding of the text. If analysis can demonstrate that these
linguistic items do indeed affect the way in which the text is understood, then the
implications for translation must be considered. As a matter of principle, then, an item
like yh!yw+ ~ is in the text because it has a function to peform there; consequently, translation
must take it into account.
426
The categorization of the verbal and temporal occurrences of yh!yw+ ~ has made it
extremely clear that there are definite syntactic differences between these two major uses.
In the verbal uses, yh!yw+ ,~ as the main verb of the clause, is obligatory, unless as mentioned
in Psa 76:3 (listed in 8.2.1.8.3), there is elision of the verb. Chapter nine lists the
temporal uses of yh!yw+ ,~ but there are many instances of the same or similar temporal
expressions without yh!yw+ .~ In simple terms, yh!yw+ ~ must be there when it is the nuclear verb,
but it is not a required part of every temporal expression in every instance. Full analysis
of the role of yh!yw+ ~ in temporal expressions ultimately must consider the structure,
context, and function of temporal expressions without yh!yw+ .~
At numerous points throughout the preceding chapters, the promise was made to
deal with issues and questions regarding the uses and functions of yh!yw+ ~ here in this
chapter. These issues, as well as the questions raised by the claims in the literature
review (4.1.5), are summarized in the following points:
1) The narrative and discourse-pragmatic functions of yh!yw+ ~
2) The possible role of yh!yw+ ~ in indicating tense
3) The implications for temporal clauses that do not include yh!yw+ ~
4) The impact of the concept of Reference Time on the analysis of yh!yw+ ~
5) The benefits of a cognitive analysis of yh!yw+ ~
Ultimately, however, this study is not just about yh!yw+
~ it is a study of how the
temporal structuring of Hebrew text is understood and the role yh!yw+ ~ plays in that
427
important aspect of narrative. Full analysis of the temporal organization of the biblical
Hebrew text is, however, beyond the scope of the current study.
10.2.1 Macrosyntactic
First of all, the notion macrosyntactic needs further clarification. Typically,
macrosyntactic is used to refer to the function of certain items or markers in a text, but its
scope needs to be clearly defined. The term macrosyntactic undoubtedly came into
existence to make a distinction between smaller levels of analysis, typically referred to
as just syntaxthe analysis of phrase, clause, or sentence patternsand larger areas of
analysis beyond the sentence. This type of distinction is fine, but the problem resides in
the failure to clearly define exactly what macrosyntactic refers to. Some authors appear
to use macrosyntactic to refer to anything interclausal, but others use the term to refer to
indicators or markers of more global textual organization.
The traditional and descriptive grammars typically recognize that yh!yw+ ~ plays some
type of interclausal role. The discussion of whether yh!yw+ ~ is more closely linked to what
precedes or to what follows demonstrates that its analysis is a matter of interclausal
syntax. The term macrosyntactic, however, was brought into use in the midst of this
awareness of yh!yw+ ~ playing some type of role in signaling connections across clause
428
boundaries. In fact, most of the references to a macrosyntactic function of yh!yw+ ~ deal with
issues of text-segmentation or narrative organization.
In the model implemented here, the Morpho-Syntactic Orientation discussed in
5.2.2 encompasses all levels of morphosyntax up to and including interclausal
phenomena. At the level of morphology, linguistic systems are much more tightly rulegoverned than they are farther away from the nuclear constructions. As one moves into
the higher levels of textual organization, there is greater flexibility. The threshold in
the model implemented here is at the interclausal level. This does not imply, however,
that textual organization is random beyond the clause, but there is less predictability due
to the increased variables to which communication is context-sensitive. Longacres
notion of a text profile (1996, 2) is important to consider here. A profile is not intended
to predict every detail of a text and should, therefore, not be interpreted as a rigid
template. The context-sensitive nature of language and communication means that text is
emergent. A profile, then, is what develops as one travels through the text, encountering
its contours. Journeys through similar texts of comparable genre will reveal that they
share certain features and also differ in certain respects.
In summary, the main issue with macrosyntactic is its lack of precision.
Therefore, the label macrosyntactic marker is deemed too imprecise to describe the
functions of yh!yw+ .~ The preference here is to speak of the discourse-pragmatic functions of
yh!yw+ .~ The terms morpho-syntactic and discourse-pragmatic make a clear distinction
between the traditional realm of syntax and the realm of textual function and
organization. Also, because a linguistic item may function both morphosyntactically in
429
its local context and discourse-pragmatically in the global textual context, the term
discourse-pragmatic allows direct reference to these functions.
hd\Ch
* ^ ax@w+ ;T#vq
= w^ + ;y+lT
= # ;yl#k@ an`-ac* hT*uw^ +
Gen 27:3
The function of hT*uw^ + here is not to establish the temporal reference for the action,
but rather it indicates the urgency, the importance of that action, marking pragmatically
salient information within reported speech. Perhaps one of the greatest liabilities of the
term discourse marker is the tendency to read text-segmentational or structural
significance into marker. For example, there are 26 occurrences of hT*uw^ + in Genesis, but
430
a mere list of the occurrences of hT*uw^ + does not provide a structural outline of Genesis.
This may seem patently obvious, but the term markerwhether of the macrosyntactic or
the discourse varietytends to imply a segmentational function. It may be true that hT*uw^ +
sometimes occurs at a significant transition point in a narrative, but its function does not
necessarily extend to the global textual level.
The concern here is to consistently apply the context-sensitive model
recommended in this study to every level of linguistic analysis. Even though there may
seem to be some similarities between discourse markers and how yh!yw+ ~ functions, the
preference here is to not use the term discourse marker with reference to yh!yw+ .~ It is
preferable to speak of yh!y+w~ having certain discourse-pragmatic functions, rather than
giving it the label of discourse marker.
431
432
to this question is found by considering the implications of the WAYYIQTOL form of hy`h.*
In the analysis implemented in this study, the WAYYIQTOL form has a variety of uses as
described in 6.4.1. Consider, for example, 1 Sam 7:14:
There was peace between Israel
and the Amorites.
.yr]ma
) h
$ * /yb@W la@rc
` y= ] /yB@ <olv* yh!yw+ ~
1 Sam
7:14
Before assuming that this clause with yh!yw+ ~ functions as a type of summary or
conclusion, three factors need to be recognized:
1) The EXISTENTIAL sense of hy*h* plays a significant role in this clause
2) One of the functions of the WAYYIQTOL is CONCLUSION (See 6.4.1.7) and,
3) The position of the clause in the narrative is also an important factor
A very similar case could be made for <yh!Oa$ ud~Y}w~ in Exod 2:25. There is a
definite dimension of closure operative in this occurrence of ud~Yw} ~ because of its position
in the narrative.
These factors demonstrate the interactive nature of semantic, pragmatic, and
narrative features in communication. The yh!yw+ ~ in 1 Sam 7:14 should not be assigned
discourse-pragmatic significance merely because it is yh!yw+ .~ This places too much
functional load on yh!yw+ ~ when, in fact, a variety of factors contribute to its textual role.
Stated in another way, the attribution of any narrative significance to the clause in 1 Sam
7:14 is the product of multiple factors. This may indeed be a significant concluding
remark, but that significance is not the result of the occurrence of yh!yw+ ~ alone.
433
The possibility of yh!yw+ ~ having some function outside the clause in which it occurs
is not affected by the close syntactic connection between yh!yw+ ~ and the clausal elements
with which it occurs. This is in direct contrast to the following statement by Niccacci:
Only when grammatical analysis shows that a given element has no role in
the sentence does one have to look for a semantic or pragmatic role on the
higher level, the text level. (Niccacci 1994, 117)
Niccacci appears to be saying that only when an element has no clausal or
sentential role does one look for a pragmatic role in the text. The analysis in this study
has shown, to the contrary, that yh!yw+ ~ often has a role at the clause or sentence level as
well as on the level of the text. It is important, however, to distinguish between
semantics and pragmatics. In other words, the meaning that yh!yw+ ~ may have should not be
confused with the possible functions that it may perform. For example, in the /k@-yr\ja
& ^
examples in 9.3.1.3, the semantics of the expression should not be confused with a
pragmatic function of something like episode closure. If the proper narrative parameters
& ^ would not occur. This reflects the symbiotic
were not operative, the expression /k@-yr\ja
nature of language that is part of the context-sensitivity of communication. Linguistic
items like yh!yw+ ~ are not like isolated beads dropped into a string of other items. An
expression occurs where contextual features produce the motivating factors for its use. At
the same time, the expression contributes semantically to the environment in which it
occurs. This requires that analysis implement the notion of context-sensitivity in every
decision. Analysis can greatly benefit from computer searches, but they extract items
from their contextualized, interactive stream, making it easy for the analyst to overlook
the contextual factors that motivated the use of the item being studied.
434
In addition to the use of yh!yw+ ~ as CONCLUSION discussed above, its use in setting or
ORIENTATION
Gen 39:2
j~yl!xm
= ^ vya! yh!yw+ ~
.yr]xM
= h
! ^ wyn`da
{ & tyb@B= yh!y+w~
Before assuming that there is any particular pragmatic significance to the mere
repetition of yh!yw+ ~ here, the principle of choice must be considered. yh!yw+ ~ is the normal,
default verb in these three verbal occurrences in Gen 39:2. The concurrence of three
instances in consecutive clauses may be unusual, but the verbal use of yh!yw+ ~ in each of
these cases is not at all out of the ordinary. Therefore, just the mere repetition of yh!yw+ ~ in
Gen 39:2 is not deemed pragmatically significant. What is pragmatically significant is its
use as ORIENTATION or in stage-setting clauses.
ORIENTATION
covers several
informational categories necessary for the subsequent development of the text. These
categories include information such as temporal and locational reference, introduction of
participants and other elements germane to setting up the communicative frame of
reference between the author and reader (Grimes 1974). ORIENTATION is closely related to
the EQUATIVE, EXISTENTIAL, and DEICTIC uses considered in chapter eight.
There are no rules that can predict the precise form ORIENTATION will take in any
given text, although certain common characteristics can be defined. The most basic
characteristic of ORIENTATION is that in narrative text, for example, it occurs in the
margins of the narration itself. The margin may come before the main body of the
narrative or at its close or in both positions. There are also other margins within the
435
narrative, typically realized as temporal expressions. The ORIENTATION function of yh!yw+ ~
in temporal expressions is discussed below.
It is important to note that the Gen 39:2 example is surrounded by many other
occurrences of yh!yw+ .~ Chapter 39 alone has 11.54% of all the occurrences in Genesis, not
to mention that chapter 38 has another 4.62%, giving these two chapters more than 16%
of all the occurrences in the whole book. Again, just the mere statistics do not tell the
whole story. The various uses must be considered, separating the verbal occurrences
from the temporal ones, before determining the significance of the high frequency of
occurrence.
From the perspective of the ORIENTATION function of yh!yw+ ,~ the discoursepragmatic significance of this cluster of occurrences is in the setting being established for
the upcoming climactic narrative which ends the book. This is, however, not merely a
result of occurrences of yh!yw+ ,~ but rather of yh!yw+ ~ working together with other narrative
features.
436
pre-nuclear margin or the post-nuclear margin? Is there a difference in function between
the pre-nuclear and the post-nuclear temporal expression? Regarding yh!yw+ ,~ is there a
positional difference in its occurrence?
CLAUSES:
POST-NUCLEAR
MARGIN
PRE-NUCLEAR
MARGIN
NUCLEUS
SENTENCE
Both the pre- and post-nuclear margins are dependent clauses, whereas the
nucleus is the independent or main clause of the sentence. Examples:
Nuclear clause only:
He went down to Joppa.
opy` dr\Yw} ~
Jonah 1:3
hm*yr
+ x
` m
= ! <r`ba
= ^ aobK= yh!yw+ ~
Gen 12:14
hV*ah
! -* ta# <yr]xM
= h
! ^ War+Y]w~
.da)m= awh! hp*y-` yK!
2 Sam
4:4
Some temporal expressions occur in the post-nuclear margin as in 2 Sam 4:4, but
yh!yw+ ~ never occurs here. In temporal expressions, yh!yw+ ~ only occurs in sentence-initial, prenuclear margins. The use of yh!yw+ ~ in temporal expressions contrasts only in the prenuclear position. Grammar and syntax textbooks should be careful to take this into
account when illustrative examples are selected. For example, in Arnold and Choi (2003,
437
103), examples are given of B= used temporally, but the examples come from very
different textual and clausal environments. At a certain level, the strict syntax of the
components from different clause environments may for all practical purposes be the
same. This is problematic, however, when the question of use arises. What may appear to
be identical syntactic structures may have very different functions and contexts of use.
Regarding the role of yh!y+w~ in indicating tense, Ross is a good example of the view
that identifies yh!yw+ ~ as a macrosyntactic marker, which indicates simply that the narrated
events occurred in the past (Ross 2001, 139-40). In response, first of all, the indication
of the temporal reference is not dependent upon the presence of yh!yw+ ~ alone. There are
many other contextual indicators that the events occurred in the past. The uneven
distribution of yh!yw+ ~ raises serious issues for this claim. If yh!yw+ ~ were the indicator of tense
for the narrative, one would expect a much more even distribution without the repeated
cases which sometimes occur. If the function were that of indicating past tense, the use
of yh!yw+ ~ in successive clauses would be very difficult to explain.
The frequent temporal use of yh!yw+ ~ with infinitive constructs is, however, one of
the contexts in which yh!yw+ ~ does play a role in establishing temporal reference for the
atemporal infinitive construct. Especially when the use of yh!yw+ ~ and hy*hw* + are compared in
this syntactic relationship, it seems clear that yh!yw+ ~ does indeed indicate past tense. This
temporal reference is limited to the clause within which the expression with yh!yw+ ~ and the
infinitive construct occur and does not establish the tense for the whole narrative. The
results of the current research indicate, then, that yh!yw+ ~ is not a tense marker, but rather it
438
is to be read as one of the components of temporal reference in the text. For instance, in
the example discussed above from Gen 22:4, there is no yh!yw+ ~ to indicate the tense, but the
narrative continues to refer unambiguously to the past.
In 9.2.2.1, the section on Speaker Deixis, it was stated that each WAYYIQTOL
moves along the temporal dimension of the narrative, with each event establishing a new
Reference Time. One of the ways that yh!yw+ ~ s behavior as a WAYYIQTOL is seen is that
yh!yw+ ~ also makes a move along the temporal dimension of the narrative. The particular
temporal expression with which yh!yw+ ~ occurs may be backreferencing, but yh!yw+ ~ establishes
a new moment or place along the temporal dimension. In the discussion of van der
Merwes analysis above, reference was made to the statement that yh!yw+ ~ signals that the
reference time of a state of affairs is that of the current reference time of a preceding
temporally anchored event or events (van der Merwe 1999, 113-14). The use of yh!yw+ ~ in
temporal expressions, however, indicates that the expression occurring with yh!yw+ ~ has
moved along the temporal axis of the narrative. Just as each WAYYIQTOL establishes its
own Reference Time, the temporal expression with yh!yw+ ~ also establishes a new point of
temporal reference from which the ensuing events are viewed. The temporal expression
itself may be backreferencing, but yh!yw+ ~ establishes a new temporal reference point.
Throughout the data display in Chapter nine, the occurrences of yh!yw+ ~ are
categorized according to the form of the verb which follows. The ORIENTATION function
of yh!yw+ ~ is fundamental to the reading of these examples. The temporal expression with
yh!yw+ ~ sets the stage for the narrative it precedes. If the following verb form is a
439
WAYYIQTOL,
the temporal reference of the expression with yh!yw+ ~ sets the temporal
reference for that event as the first in the upcoming narrative succession. If the following
verb is a QATAL, the temporal expression still establishes a new point of reference, but the
QATAL
440
441
simultaneity is communicated by language-specific mechanisms that represent that
temporal relationship. Flashback, then, is a linguistic means of expressing events in a
way that cognitively represents experience in the event world.
Just as certain linguistic items will be employed to properly signal events in a
flashback, there are a variety of items which function to establish and maintain temporal
reference throughout a text. In Lambrechts Information Structure model, there are
different states of IDENTIFIABILITY in communication (Lambrecht 1994). Proper
communication depends on the IDENTIFIABILITY of referents and entities. Lambrecht
does not discuss the temporal organization of text, but the same principles apply. The
proper interpretation of narrative depends on the hearer or reader accessing the same
temporal organization with which the text was communicated. One of the cognitive
functions of yh!yw+ ~ involves accessing proper temporal reference. This is closely related to
its DEICTIC function discussed in 9.2.2.1. One of the potential misconceptions of an item
like yh!yw+ ~ perhaps perpetuated by labels like marker or signis that a text can be
segmented on the basis of its occurrences. yh!yw+ ~ is, however, only one element in the
intricate network of temporal and deictic reference.
Some of the central concepts have already been addressed in the discussion of the
grammaticalization of yh!yw+ ,~ but it is important to reiterate the fact that communicators do
not intend all linguistic items they use to perform the same referential function. The
function of some items is not strictly referential, but rather is related to the proper
cognitive processing of the text. At a cognitive level, yh!yw+ ~ aids the proper temporal
interpretation of the text and contributes to its proper segmentation. The propositional
442
content of the temporal expressions is provided by the expressions themselves: after these
things, three days later, at the end of ten days, etc. yh!yw+ ~ signals the way in which the
temporal expression is intended to connect to its context. Temporal expressions without
yh!yw+ ~ are still involved in the temporal organization of the narrative, but the discoursepragmatic connection differs.
10.5.1.2 Episode Initiator?
The term episode refers to a unit within a narrative, and the question regarding
yh!yw+ ~ is whether it plays some role in episodic structure. The analysis of yh!y+w~ s role in
episodic structure is complex, to say the least. Gen 25:19 begins the qj*xy= ] td{lo= T the
generations of Isaac, going through 35:29. There are twenty-five occurrences of yh!yw+ ~ in
this section of Genesis that provide a good case study for considering episode structure.
However, only the first few occurrences will be discussed here.
The qj*xy= ] td{lo= T beginning in 25:19 is introductory to the whole Isaac narrative,
then the occurrence of the verbal yh!yw+ ~ in 25:20, expressing Isaacs age is at the beginning
of this part of the narrative. It should be noted that the occurrence of an item at the start
of a narrative is not necessarily to be equated with that item having the function of
initiating that narrative.
Isaac was forty years old when
he took Rebekah, the daughter of
Bethuel the Aramean of Paddanaram, the sister of Laban the
Aramean, to be his wife.
Pc-vqw3msXa np ncmsc-abp ncfs
Pp-vqcX3ms Po-np ncfs-np Pa-np
Pp np ncfsc np Pa-np PpX3ms Ppncfs
hn`v* <yu!Br
* a
+ -^ /B# qj*xy= ] yh!yw+ ~
yM!ra
~ h
& * la@WtB=-tB^ hq*b=r-] ta# w{Tj=qB
^ =
.hV*al
! = w{l yM!ra
~ h
& * /b*l* toja& <r`a& /D~Pm
^ !
Gen
25:20
443
When the boys grew up, Esau
became a skillful hunter, a man
of the field, but Jacob was a
peaceful man, living in tents.
Pc-vqw3mp Pa-ncmp Pcvqw3msXa np ncms vqPms ncms
ncms ncms Pc-np ncms ams vqPms
ncmp
<yr]uN* h
+ ^ WlD+gY+ w] ~
hd\c* vya! dy]x^ u~dy} { vya! wc*u@ yh!yw+ ~
Gen
25:27
.<yl!ha
* ) bv@y{ <T* vya! bq)uy& w~ +
/ovar]h* bu*rh
` * db^Lm
= ! Jr\aB
* * bu*r` yh!yw+ ~
Gen 26:1
.hr`rG` + <yT!vl
= P
! -= El#m# El#My# b!a-& la#
These three examples are the type of occurrence that seem to fit the profile of yh!yw+ ~
initiating an episode. The clauses where Isaac is introduced, Esau is a hunter when the
boys grow up, and a famine comes, seem to indicate transition points in the narrative.
Upon closer examination, however, it seems that 25:24 is a significant moment in the
textwhen Rebekah gave birth to Jacob and Esau, but there is no yh!yw+ .~ Also, it should
be noted that in 25:27, the occurrence of yh!yw+ ~ is not in a temporal expression: when the
boys grew up, but rather is in the clause that describes Esau as a hunter. At first glance,
these cases might appear to be instances of yh!yw+ ~ indicating episodic transitions, but upon
further examination, the common feature of these examples is that they have a DEICTIC
function within the narrative. It is also crucial to acknowledge that each of these
examples are typical uses of yh!yw+ :~
Gen 25:20: Expressions of Age, discussed in 8.2.1.6.1
Gen 25:27: The EQUATIVE Use of yh!yw+ ,~ discussed in 8.2.1
444
Gen 26:1: The EXISTENTIAL Use of yh!yw+ ,~ discussed in 8.2.2
The textlinguistic significance of these examples cannot be automatically
assumed without taking into consideration the fact that yh!yw+ ~ is typically the default verb
for these expressions. This does not preclude some type of discourse-pragmatic role, but
rather is a plea for careful consideration of the syntactic and textual context. One of the
DEICTIC
functions of occurrences of this type is that they provide SETTING for the
narrative in which they occur. Some statements of setting may only be one clause and in
other places there may be extensive stage-setting information for the narrative.
The occurrence in 26:8 is yet another example of the need to consider the context.
There is a famine and Isaac and Rebekah are in Gerar, as the LORD instructed them.
When he had been there a long
time, Abimelech king of the
Philistines looked out through a
window, and saw, and behold,
Isaac was caressing his wife
Rebekah.
<ym!Yh
` ^ <v* w{l-Wkr+a* yK! yh!yw+ ~
/oLj^h^ du^B= <yT!vl
= P
! = El#m# El#my# b!a& [q@vY= w~ ~
.w{Tv=a! hq*br
= ] ta@ qj@x^m= qj*xy= ] hN}hw! + ar+Yw~ ~
Gen 26:8
Pc-vqw3msXa Pp vqp3cp-PpX3ms
Pd Pa-ncmp Pc-vhw3msXa np ncms
np Pp Pa-ncbs Pc-vqw3msXa Pc-Pi
np vpPms Pp np ncfscX3ms
If yh!yw+ ~ does perform some type of episode initiating function, then it seems that
the episodic level indicated by the occurrence in Gen 26:8 is different from the previous
ones considered in the qj*xy= ] td{lo= T. Isaac and Rebekah are still in Gerar and the
temporal expression indicates the passage of days when the king sees Isaac and Rebekah
together. Rather than including levels of sub-episodes, it is preferable to also see this
occurrence as DEICTIC or as providing stage-setting information for the narrative. A very
important point here is that w+ also performs stage-setting functions in narrative, which is
445
the motivation for the mention of the interplay of yh!yw+ ~ and w+ in narrative as an important
area for further research.
Hatav makes the observation that the non-verbal occurrences of yh!yw+ ~ function as a
segmentational particle, marking mainly temporal segmentation (Hatav 1997, 70), but
one of the liabilities of a statement like this is that it can easily give the impression that a
narrative can be temporally divided into segments or episodes based solely on the
occurrences of yh!yw+ .~ It is especially difficult to argue for the segmentational function in
27:30, where there are two occurrences in consecutive clauses:
As soon as Isaac had finished
blessing Jacob, and Jacob had
just left the presence of Isaac his
father, Esau his brother came in
from his hunting.
Gen
27:30
446
of the subsequent one. Gen 21:34 prepares the reader for a transition, leading into Gen
22:1:
And Abraham stayed in the land
of the Philistines a long time.
<yT!vl
= P
! = Jr\aB
# = <h*rb
` a
= ^ rg`Yw` ~
.<yB!r~ <ym!y`
Gen
21:34
As mentioned above, Gen 22:1 starts with a temporal expression with yh!yw+ .~ The
next occurrence of a temporal expression is in Gen 22:4:
On the third day, Abraham
looked up and saw the place
from a distance.
wyn`yu@-ta# <h*rb
` a
= ^ aC*Yw] ~ yv!yl!Vh
= ^ <oYB^
.qj)rm
` @ <oqM*h-^ ta# ar+Yw~ ~
Gen 22:4
This expression is different in that there is no yh!yw+ ~ before <oYB^. This is a good
example of the use of a pre-nuclear temporal expression without yh!yw+ .~ The function of the
temporal expression itself remains essentially the same in that it establishes the temporal
reference within which the following event takes place. The absence of yh!yw+ ,~ however,
changes the relationship or connection of the temporal expression to the surrounding
= ^ <oYB^ still signals a temporal transition in the narrative,
context. The temporal yv!yl!Vh
but without yh!yw+ ~ there is not the same DEICTIC function of establishing this as a new point
of reference on the temporal axis. Any attempts to determine the episodic structure of the
narrative must consider these factors. It is not possible at this point to claim that the
temporal expressions with yh!yw+ ~ indicate major episode breaks. Further research and
careful attention to the full array of narrative strategies and features are needed.
Because of these functions, translation must account for yh!yw+ ~ in some functionally
appropriate way. Translation strategies and decisions must take the contextual factors into
447
consideration so that the DEICTIC and ORIENTATION functions are represented in the new
version.
10.5.1.3 Focus and Foreground
First of all, one of the major difficulties in this area of research is that terms like
foreground and background imply a binary distinction that does not accurately reflect the
multidimensional nature of language. There are multiple levels of saliency as all
linguistic components perform their various functions, working together to make
effective communication possible. Background is just as important as foreground in the
overall communicative process.
So, as discussed in 3.4.8, does yh!yw+ ~ mark the main segments of a narrative
(Schneider, p. 265) and distinguish the main story from the embedded stories (Talstra
1978, 173)? A proper answer to this question would involve detailed analysis of the
narrative contours of each book in the Hebrew Bible. This is beyond the scope of the
present study, but some preliminary observations can be offered here.
In 4.1.3, the question was asked whether there is some significance marked in
Jonah by the occurrences of yh!yw+ ~ with the storm, Jonah being in the fishs belly, and the
sunrise. Do these occurrences help distinguish the main story from the embedded
stories? Based on the occurrences in Jonah, the distinction between main and embedded
stories does not appear to apply. The occurrences in 1:1 and 3:1 are both in FORMULAIC
expressions, but yh!yw+ ~ alone is not responsible for the division of the book into two halves.
The verbal occurrence in 1:4, <Y`B^ lodG`-ru^s^ yh!yw+ ,~ is part of the setting for the ship
episode, performing its function as ORIENTATION, which also includes the w+ + noun
448
structures: <Y`h-^ la# hl*odG+-j~Wr lyf!h@ hw`hyw~ and .rb@Vh
* l
! = hb*Vj
= ! hY`na
] h
( w* .+ To say that
lodG`-ru^s^ has some peculiar role merely because yh!yw+ ~ occurs with it is artificial.
Longacre claims that yh!y+w~ does not function on the storyline of a narrative
(1989, 66), yet in BHRG, the claim is made that yh!yw+ ~ may indicate a state of affairs that
plays a pivotal role in the subsequent narrative or it may indicate an event that is not mere
background (van der Merwe, Naud, Kroeze 1999, 333). These claims reveal the
complexity of the systems of reference in narrative and demonstrate the need for a more
nuanced view of narrative strategy than is possible within a binary approach to
foreground and background. This is an area deserving much more research. The
classification of the examples of yh!yw+ ~ presented in the preceding chapters has been done
with the hope of contributing to this ongoing research. The syntactic connections of yh!yw+ ~
and its uses as a WAYYIQTOL argue, however, for a role in the narrative at the same level
as all other WAYYIQTOLs. At this point, integral role seems more appropriate and
judicious than pivotal role, but more research is needed in this important area.
449
connection of yh!yw+ ~ with the clause in which it occurs. In order for the occurrence to be
verbal, it must be determined that yh!yw+ ~ is the main verb of the clause.
yh!yw+ ~
|
Verbal?
Temporal?
Temporal yh!yw+ ~
With rj^a/^ yr\ja
& ?^
With a Preposition?
With rv#aK
& ?^
With yK!?
With Specific Temporal Reference?
With a Participle?
With an Adverb?
Figure 33: Category of Use?
450
10.6.4 Summary
One of the main things to avoid when encountering occurrences of yh!yw+ ~ in text is
to immediately assume that it means discourse marker without paying attention to the
type of use and the possible function(s) it might have in the text. There is no substitute
for repeated close readings of the text, taking note of the different systems of temporal
and deictic reference in order to see how they play out in the narrative itself.
451
and EXISTENTIAL, closely paralleling the uses of yh!yw+ .~ One of the main
differences between the uses of yh!yw+ ~ and hy*hw* + is their characteristic temporal reference of
past and future, respectively. In character with the IRREALIS tendencies of future
reference, there are many MODAL occurrences of hy*hw* + that are not found with yh!yw+ .~
452
Further research is needed to fully categorize and analyze the occurrences of hy*hw* ,+
which would add another dimension to the analysis of yh!yw+ ~ presented here.
/k@-yr}ja
& ^ yh!yw+ ~
2 Sam
<yT!vl
= P
! -= <u! bogB= hm*jl
* =Mh
! ^ dou-yh!Tw= ~
21:18
[s^-ta# yt!vj
* h
% ^ yk^Bs
= ! hK*h! za*
.hp*rh
` * yd}ly! B! rv#a&
453
each other, but the analysis of one aspect of a language will always have some connection
to other aspects. For this reason, it is incumbent upon the language analyst to consider
the implications of claims and statements regarding the function and use of parts of the
language being studied. For example, Pratico and Van Pelt comment regarding yh!yw+ ~ that
it is called a temporal modifier because it marks the beginning of a past tense narrative
sequence (Pratico and Van Pelt 2001, 196). The implications and consequences of this
statement are enormous. One of the main problems with this claim is that it is stated too
categorically, seemingly without adequate consideration for the possibility of narrative
sequences beginning without yh!yw+ .~ This claim makes it appear as if all narrative
sequences begin with yh!y+w~ and that the function of every yh!yw+ ~ is to begin a narrative
sequence. Even though the consideration of the global textual context can make it seem
that no analysis can ever be definitive, this should be seen as a result of the nature of
language rather than a product of the model of analysis implemented here. Statements of
use and function should reflect context-sensitivity in the way they are framed.
The approach implemented in this study was referred to in 5.2 as a Functional,
Discourse-Pragmatic view of language. As previously discussed and as demonstrated
throughout this study, there are many components to this view and approach. Beyond the
focus on the analysis of yh!yw+ ,~ one of the goals of this research has been to demonstrate the
benefits of this model of language analysis.
Another goal has been to insist on the need for thorough analysis of biblical
Hebrew syntax. As Arnold and Choi comment in the opening line of their A Guide to
454
Biblical Hebrew Syntax, [a]t the heart of biblical interpretation is the need to read the
Bibles syntax (Arnold and Choi 2003, 1)
Finally, this study is about more than yh!yw+ .~ It is about using the most appropriate
tools and models to better understand the biblical Hebrew text. As Silva comments in the
foreword to the 2004 revision of Machens New Testament Greek for Beginners:
It is not the primary purpose of language study to provide the means for
reaching astounding exegetical conclusions, although sound linguistic
training can at least prevent students from adopting inadmissible
interpretations. The true goal of learning New Testament Greek is rather to
build a much broader base of knowledge and understanding than the
student would otherwise have. Occasionally, this knowledge may indeed
supply fairly direct answers to exegetical questions. But what matters most
is the newly acquired ability to interpret texts responsibly on the basis of
comprehensive rather than fragmented (and therefore distorted)
information. (Silva 2004, 11)
The goal of this model is to enable the reader or interpreter to develop greater
sensitivity to the transitions and contours of the biblical Hebrew text.
WORKS CITED
Alter, R. 1981. The Art of Biblical Narrative. New York: Basic Books.
Andersen, F. I. 1974. The Sentence in Biblical Hebrew. The Hague: Mouton.
Anderson, S. R. and E. L. Keenan. 1985. Deixis. In Language Typology and Syntactic
Description: Grammatical Categories and the Lexicon, ed. T. Shopen, 259-308.
Cambridge: Cambridge University Press.
Arnold, B. and J. Choi. 2003. A Guide to Biblical Hebrew Syntax. Cambridge:
Cambridge University Press.
Barr, J. 1983. The Semantics of Biblical Language. London: SCM Press.
Bartelmus, R. 1982. HYH. Bedeutung und Funktion eines hebrischen
Allerweltswortes. Arbeiten zu Text und Sprache im Alten Testament 17. St.
Ottilien: EOS Verlag.
Bayly, C. 1782. An Entrance into the Sacred Language Containing the Necessary Rules
of Hebrew Grammar in English. London: n.p. Quoted in L. McFall, The Enigma
of the Hebrew Verbal System: Solutions from Ewald to the Present Day.
(Sheffield: Almond Press, 1982), 15.
Beaugrande, R. de and W. U. Dressler. 1981. Introduction to Text Linguistics. London:
Longman.
Berlin, A. 1989. Lexical Cohesion and Biblical Interpretation. Hebrew Studies 30: 29-39.
. 1994. Poetics and Interpretation of Biblical Narrative. Winona Lake:
Eisenbrauns.
Block, D. I. 1997. The Book of Ezekiel, Chapters 1-24. New International Commentary
on the Old Testament. Grand Rapids: Eerdmans.
Bloomfield, L. 1933. Language. New York: Henry Holt and Company.
Boas, F., ed. 1911. Handbook of American Indian Languages. U.S. Bureau of American
Ethnology. Bulletin No. 40. Washington: Government Printing Office.
456
457
Brown, F., S. R. Driver, and C. A. Briggs. 1996. The Brown-Driver-Briggs Hebrew and
English Lexicon: With an Appendix Containing the Biblical Aramaic. Peabody,
MA: Hendrickson.
Brown, G. and G. Yule. 1983. Discourse Analysis. Cambridge: Cambridge University
Press.
Bybee, J. L. 1985. Morphology: A Study of the Relation between Meaning and Form.
Amsterdam: John Benjamins Publishing Company.
Carroll, R. P. 1986. Jeremiah: A Commentary. Philadelphia: Westminster Press
Chisholm, R. B. 1998. From Exegesis to Exposition: A Practical Guide to Using Biblical
Hebrew. Grand Rapids: Baker.
Churchland, P. M. 1996. The Engine of Reason, the Seat of the Soul. A Philosophical
Journey into the Brain. Cambridge: MIT Press.
Dik, S. C. 1978. Functional Grammar. Amsterdam: North-Holland Publishing Company.
Endo, Y. 1996. The Verbal System of Classical Hebrew in the Joseph Story: An
Approach from Discourse Analysis. Assen: Van Gorcum.
Eskhult, M. 1990. Studies in Verbal Aspect and Narrative Technique in Biblical Hebrew
Prose. Stockholm: Almqvist and Wiksell.
Exter Blokland, A. F. den. 1995. In search of text syntax: towards a syntactic textsegmentation model for biblical Hebrew. Amsterdam: VU University Press.
Fauconnier, G. 1985. Mental Spaces : Aspects of Meaning Construction in Natural
Language. Cambridge: Cambridge University Press.
Fee, G. D. 2002. New Testament Exegesis: A Handbook for Students and Pastors. 3d ed.
Louisville: Westminster John Knox.
Fokkelmann, J. P. 1999. Reading Biblical Narrative: An Introductory Guide. Louisville,
KY: Westminster John Knox Press.
Foley, W. A. and R. D. Van Valin Jr. 1984. Functional Syntax and Universal Grammar.
Cambridge: Cambridge University Press.
Follingstad, C. M. 2001. Deictic Viewpoint in Biblical Hebrew Text: A Syntagmatic and
Paradigmatic Analysis of the Particle yk. Dallas: SIL International.
Gesenius, W. and E. Kautsch. 1910. Gesenius Hebrew Grammar. Trans. and rev. by A.
E. Cowley. 2d ed. Oxford: Clarendon Press.
458
Gibson, J. C. L. 1994. Davidsons Introductory Hebrew Grammar ~ Syntax. Edinburgh:
T. & T. Clark.
Givn, Talmy. 2001. Syntax: An Introduction. 2d ed. Amsterdam: John Benjamins
Publishing Company.
Goldfajn, T. 1998. Word Order and Time in Biblical Hebrew Narrative. Oxford:
Clarendon Press.
Grimes, J. E. 1974. The Thread of Discourse. The Hague: Mouton.
Gropp, D. M. 1995. Progress and Cohesion in Biblical Hebrew Narrative: the Function of
k- / b- + the Infinitive Construct. In Discourse Analysis of Biblical Literature:
What it is and what it offers, ed. W. R. Bodine, 183-212. Atlanta: Scholars Press.
Halliday, M. A. K. and R. Hasan. 1976. Cohesion in English. London: Longman.
Hamilton, V. 1995. The Book of Genesis: Chapters 18-50. New International
Commentary on the Old Testament. Grand Rapids: Eerdmans.
Harper, W. R. 1892. Elements of Hebrew Syntax by an Inductive Method. New York:
Charles Scribner's Sons.
Hatav, G. 1997. The Semantics of Aspect and Modality: Evidence from English and
Biblical Hebrew. Amsterdam: John Benjamins.
Hopper, P. J. and E. C. Traugott. 1993. Grammaticalization. Cambridge: Cambridge
University Press.
Hymes, Dell. 1974. Foundations in Sociolinguistics: An Ethnographic Approach.
Philadelphia: University of Pennsylvania Press.
Jackendoff, R. 1994. Patterns in the Mind: Language and Human Nature. New York:
Basic Books.
Joon, Paul. 1996. A Grammar of Biblical Hebrew. Translated and Revised by Takamitsu
Muraoka. 2 vols. Rome: Editrice Pontificio Instituto Biblico.
JPS Hebrew-English TANAKH. 2000. Philadelphia: The Jewish Publication Society.
Kelley, P. H. 1992. Biblical Hebrew: An Introductory Grammar. Grand Rapids:
Eerdmans.
Keown, G. L., P. J. Scalise, and T. G. Smothers. 1995. Jeremiah 26-52. Word Biblical
Commentary. Waco: Word Books.
459
Khan, G. 1999. The Karaite Tradition of Hebrew Grammatical Thought. In Hebrew Study
from Ezra to Ben-Yehuda, ed. W. Horbury, 186-203. Edinburgh: T. & T. Clark.
Koehler, L. and Baumgartner, W. 1995. The Hebrew and Aramaic Lexicon of the Old
Testament. Leiden: Brill.
Labov, W. 1972. Sociolinguistic Patterns. Philadelphia: University of Pennsylvania
Press.
Lakoff, G. and M. Johnson. 1980. Metaphors We Live By. Chicago: Chicago University
Press.
Lambdin, T. O. 1971. Introduction to Biblical Hebrew. New York: Charles Scribners
Sons.
Lambrecht, K. 1994. Information Structure and Sentence Form: Topic, Focus and the
Mental Representations of Discourse Referents. Cambridge: Cambridge
University Press.
Longacre, R. E. 1989. Joseph: A Story of Divine Providence. A Text Theoretical and
Textlinguistic Analysis of Genesis 37 and 39-48. Winona Lake, Ind.: Eisenbrauns.
. 1996. The Grammar of Discourse. 2d ed. New York: Plenum.
Longacre, R. E. and S. J. J. Hwang. 1994. A Textlinguistic Approach to the Biblical
Hebrew Narrative of Jonah. In Biblical Hebrew and Discourse Linguistics, ed. R.
D. Bergen, 336-58. Dallas: Summer Institute of Linguistics.
Longman, T. 1996. Literary Approaches to Biblical Interpretation. In Foundations of
Contemporary Interpretation, ed. M. Silva, 91-192. Grand Rapids: Zondervan.
Louw, J. P. 1986. Sociolinguistics and its Role in Text Analysis. In Sociolinguistics and
Communication, ed. J. P. Louw, 103-15. London: United Bible Societies.
Lowery, K. E. 1995. The Theoretical Foundations of Hebrew Discourse Grammar. In
Discourse Analysis of Biblical Literature: What it is and what it offers, ed. W. R.
Bodine, 103-30. Atlanta: Scholars Press.
Lyons, J. 1969. Introduction to Theoretical Linguistics. Cambridge: Cambridge
University Press.
Machen, G. 2004. New Testament Greek for Beginners. 2d ed. Revised by D. G.
McCartney. Upper Saddle River, NJ: Pearson Prentice Hall.
460
Mann, W. C. and S. A. Thompson. 1987. Rhetorical Structure Theory: A Theory of Text
Organization. Technical Report ISI/RS-87-190. Marina del Ray: Information
Sciences Institute.
McFall, L. 1982. The Enigma of the Hebrew Verbal System: Solutions from Ewald to the
Present Day. Sheffield: Almond Press.
Meek, T. J. 1945. The Syntax of the Sentence in Hebrew. JBL 64, 1-13.
Miller, C. L. 1996. The Representation of Speech in Biblical Hebrew Narrative. Atlanta:
Scholars Press.
. 1999. The Pragmatics of waw as a Discourse Marker in Biblical Hebrew
Dialogue. Zeitschrift fr Althebristik 12: 165-91.
Mitchel, L. A. 1984. A Students Vocabulary for Biblical Hebrew and Aramaic. Grand
Rapids: Zondervan.
Mller, A. 1883. Outlines of Hebrew Syntax. Glasgow: Maclehose.
Muraoka, T. 1983. Emphatic Words and Structures in Biblical Hebrew. Jerusalem:
Magnes Press.
New American Standard Bible. 1995. Updated ed. Anaheim, CA: Lockman Foundation.
New International Version. 1984. International Bible Society. Grand Rapids: Zondervan.
New Revised Standard Version. 1989. Division of Christian Education of the National
Council of the Churches of Christ in the U.S.A.
Newmeyer, F. J. 1980. Linguistic Theory in America: The First Quarter-Century of
Transformational Generative Grammar. New York: Academic Press,
Niccacci, A. 1990. The Syntax of the Verb in Classical Hebrew Prose. Sheffield:
Sheffield Academic Press
. 1991. Lettura Sintattica della Prosa Ebraico-Biblica: Principi e Applicazioni.
Jerusalem: Franciscan Printing Press.
. 1994. On the Hebrew Verbal System. In Biblical Hebrew and Discourse
Linguistics, ed. R. D. Bergen, 117-37. Dallas: Summer Institute of Linguistics.
. 1997. Basic Facts and Theory of the Biblical Hebrew Verb System in Prose. In
Narrative Syntax and the Hebrew Bible: Papers of the Tilburg Conference 1996,
ed. E. van Wolde, 167-202. Leiden: Brill.
461
Payne, T. 1997. Describing Morphosyntax: A Guide for Field Linguists. Cambridge:
Cambridge University Press.
Pike, K. L. 1967. Language in Relation to a Unified Theory of the Structure of Human
Behavior. 2d ed. The Hague: Mouton.
. 1982. Linguistic Concepts: An Introduction to Tagmemics. Lincoln and London:
The University of Nebraska Press.
Pike, K. L. and E. G. Pike. 1991. Grammatical Analysis. 2d ed. Dallas: Summer Institute
of Linguistics.
Pratico, G. D. and M. V. Van Pelt. 2001. Basics of Biblical Hebrew Grammar. Grand
Rapids: Zondervan.
Putnam, F. C. 2003. An Introduction to the Grammar and Syntax of Biblical Hebrew.
Unpublished Manuscript.
Reuchlin, J. 1506. Rudimenta linguae hebraicae. n.p.
Richter, W. 1980. Grundlagen einer althebrischen Grammatik. Arbeiten zu Text und
Sprache im Alten Testament 13. St. Ottilien: EOS Verlag.
Risselada, R. and W. Spooren. 1998. Introduction: Discourse Markers and Coherence
Relations. Journal of Pragmatics 30/2: 131-33.
Robins, R. H. 1968. General Linguistics: An Introductory Survey. Bloomington: Indiana
University Press.
Rocine, B. M. 2000. Learning Biblical Hebrew: A New Approach Using Discourse
Analysis. Macon, GA: Smyth and Helwys.
Ross, A. P. 2001. Introducing Biblical Hebrew. Grand Rapids: Baker.
Sampson, G. 1980. Schools of Linguistics. Stanford: Stanford University Press.
Sapir, E. 1921. Language: An Introduction to the Study of Speech. New York: Harcourt,
Brace.
Saussure, F. de. 1916. Cours de linguistique gnrale. Paris: Payot.
Schertz, M. H. and P. B. Yoder. 2001. Seeing the Text: Exegesis for Students of Greek
and Hebrew. Nashville: Abingdon Press.
Schiffrin, Deborah. 1988. Discourse Markers. Cambridge: Cambridge University Press.
462
Schneider, W. 1974. Grammatik des biblischen Hebrisch. Ein Lehrbuch. Munich:
Claudius Verlag.
Seow, C. L. 1987. A Grammar for Biblical Hebrew. Nashville: Abingdon.
Shopen, T. 1985. Language Typology and Syntactic Description. Cambridge: Cambridge
University Press.
Silva, M. 2004. Foreword to New Testament Greek for Beginners, by G. Machen. 2d ed.
Revised by D. G. McCartney. Upper Saddle River, NJ: Pearson Prentice Hall.
Speiser, E. A. 1964. Genesis: Introduction, Translation, and Notes. Garden City:
Doubleday and Company.
Sperber, D. and D. Wilson.1995. Relevance: Communication and Cognition. London:
Blackwell.
Stuart, D. 1987. Hosea-Jonah. Word Biblical Commentary. Waco: Word Books.
. 2001. Old Testament Exegesis: A Handbook for Students and Pastors. 3d. ed.
Louisville: Westminster John Knox.
Talstra, E. 1978. Text Grammar and Hebrew Bible. I: Elements of a Theory. Bibliotheca
Orientalis 35: 169-74.
. 1997a. A Hierarchy of Clauses in Biblical Hebrew Narrative. In Narrative
Syntax and the Hebrew Bible: Papers of the Tilburg Conference 1996, ed. E. van
Wolde, 85-118. Leiden: Brill.
. 1997b. Workshop: Clause Types, Textual Hierarchy, Translation in Exodus 19,
20, and 24. In Narrative Syntax and the Hebrew Bible: Papers of the Tilburg
Conference 1996, ed. E. van Wolde, 119-32. Leiden: Brill.
Tene, D. 1971. Hebrew Linguistic Literature. In Encyclopedia Judaica, vol. 16, cols
1352-90. Jerusalem: Keter.
Thompson, S. 1992. Functional Grammar. In International Encyclopedia of Linguistics.
ed. W. Bright, 37. Oxford: Oxford University Press.
Translators Workplace 4.0. 2002. Dallas: SIL International.
van der Merwe, C. H. J. 1994. Discourse Linguistics and Biblical Hebrew Grammar. In
Biblical Hebrew and Discourse Linguistics, ed. R. D. Bergen, 13-49. Dallas:
Summer Institute of Linguistics.
463
. 1997. Reconsidering Biblical Hebrew Temporal Expressions. Zeitschrift fr
Althebraistik 10: 42-62.
. 1999. The Elusive Biblical Hebrew Term yh!yw+ :~ A Perspective in Terms of Its
Syntax, Semantics, and Pragmatics in 1 Samuel. Hebrew Studies 40: 83-114.
van der Merwe, C. H. J., J. A. Naud, and J. H. Kroeze. 1999. A Biblical Hebrew
Reference Grammar. Sheffield: Sheffield Academic Press.
Van Pelt, M. V. and G. D. Pratico. 2003. The Vocabulary Guide to Biblical Hebrew.
Grand Rapids: Zondervan.
van Wolde, E., ed. 1997. Narrative Syntax and the Hebrew Bible: Papers of the Tilburg
Conference 1996. Leiden: Brill.
Waltke, B. K. and M. OConnor. 1990. An Introduction to Biblical Hebrew Syntax.
Winona Lake, IN: Eisenbrauns.
Waltke, B. K. and C. J. Fredericks. 2001. Genesis: A Commentary. Grand Rapids:
Zondervan.
Weingreen, J. 1959. A Practical Grammar for Classical Hebrew. 2d ed. Oxford:
Clarendon Press.
Wenham, G. 1994. Genesis 16-50. Word Biblical Commentary. Waco: Word Books.
Wildberger, H. 1991. Isaiah 28-39. The Continental Commentary. Minneapolis: Fortress
Press.
Williams, R. J. 1976. Hebrew Syntax: An Outline. 2d. ed. Toronto: University of Toronto
Press.
Winther-Nielsen, N. 1995. A Functional Discourse Grammar of Joshua. A Computerassisted Rhetorical Structure Analysis. Stockholm: Almqvist and Wiksell
International.
APPENDICES
465
Pc
conjunction
Pd
adverb
Pg
interrogative
Pi
interjection
Pn
negative
Po
Pp
preposition
Pr
relative
Pronoun (pi-)
pii
pi[pg#]*
interrogative
=
independent
proper name
ng
nc[g#s]*
common noun
466
Adjective/Numeral (a-)
a[g#s]*=
adjective
paragogic heh
[tag]Xd
directional heh
[tag]Xn
paragogic nun
[tag]X[pg#]* =
(e.g. 'ncfpcX2mp' = a construct feminine plural noun with a 2nd masculine plural
suffix)
Verb (v-)
vsapg#s**
vqi3ms
vqi3msj
vqw3ms
vqvms
vqa
infinitive absolute
vqc
infinitive construct
vqPms
vqPmsc
467
participle masculine singular construct
vqsms
vqsmsc
first person
m =
masculine
second person
feminine
= plural
construct
third person
both (noun) d
= dual
determined
other
qal
perfect
apocopated form
nifal
imperfect
cohortative form
piel
waw consecutive
P =
pual
imperative
hitpael
infinitive construct
hifil
infinitive absolute
H =
hofal
P =
particple
Q =
468
RESIDUE
There are nine occurrences listed here. The title of this section, residue,
acknowledges that these occurrences do not fit into the categories used in chapters eight
and nine. They are, however, important cases to consider, so the characteristics of each
occurrence will be briefly discussed here.
The first three cases share certain features and will be described together. In these
three examples, the item which immediately follows yh!yw+ ~ is lK*. These examples do not
fit the categories in the preceding chapters.
However, he refused to turn
aside; therefore Abner struck
him in the belly with the butt end
of the spear, so that the spear
came out at his back. And he fell
there and died on the spot. All
who came to the place where
Asahel had fallen and died,
stood still.
Pc-vpw3ms Pp-vqc Pcvhw3msX3ms np Pp-Pd Pa-ncfs PpPa-ncms Pc-vqw3fs Pa-ncfs PpPdX3fs Pc-vqw3ms-Pd Pc-vqw3ms
PpX3ms PpX3ms Pc-vqw3msXa
ncmsc-Pa-vqPms Pp-Pa-ncms Prvqp3ms Pd np Pc-vqw3msXa Pcvqw3mp
yr}ja
& B
^ = rn}ba
= ^ WhK@Yw~ ~ rWsl* /a@my* w+ ~
wyr`ja
& m
^ @ tyn]jh
& ^ ax@Tw@ ~ vm#jh
) -^ la# tyn]jh
& ^
w{tj^T^ tm*Yw` ~ <v*-lP*Yw] ~
<oqM*h-^ la# aB*h-^ lK* yh!yw+ ~
tm)Yw` ~ la@hc*u& <v* lp^n-` rv#a&
.Wdm)uY& w~ ~
2 Sam
2:23
469
Absalom used to rise early and
stand beside the way to the gate;
and when any man had a suit to
come to the king for judgment,
Absalom would call to him and
say, From what city are you?
And he would say, Your
servant is from one of the tribes
of Israel.
Pc-vhp3ms{2} np Pc-vqp3ms{2}
Pp-ncfsc ncbs Pa-ncms Pcvqw3msXa ncmsc-Pa-ncms Prvqi3ms-PpX3ms-ncms Pp-vqc PpPa-ncms Pp+Pa-ncms Pc-vqw3ms
np PpX3ms Pc-vqw3ms Pg-Pp-ams
ncfs pi2ms Pc-vqw3ms Pp-ams
ncmpc-np ncmscX2ms
ru^Vh
* ^ Er\D\ dy~-lu^ dm^uw* + <olv*ba
= ^ <yK!vh
= w! +
byr]-w{L-hy\hy= -] rv#a& vya!h-* lK* yh!yw+ ~
2 Sam
15:2
fP*vM
= l
! ^ El#Mh
# -^ la# aobl*
wyl*a@ <olv*ba
= ^ ar`qY= w] ~
<ovl=v! lomT=am
! @ w{ud+oy-lK* yh!yw+ ~
1 Sam
10:11
Pc-vqw3msXa ncmscvqPmscX3ms Pp-Pd Pd Pcvqw3mp Pc-Pi Pp-ncmp vnPms Pcvqw3ms Pa-ncms ncms PpncmscX3ms pii-ams vqp3ms Ppncmsc-np Pg-Pc np Pp+Pa-ncmp
1 Kgs
El#m# lu^Bt
^ a
= -# tB^ lb#zy\ a!-ta# hV*a! jQ^Yw] ~
16:31
470
The 2 Kgs 20:4 is anomalous in the occurrence of the noun Why`uv
= y^ + immediately
following yh!yw+ ,~ which is then followed by ax*y` aO.
Before Isaiah had gone out of
the middle court, the word of the
LORD came to him, saying,
2 Kgs
20:4
yr}cw* + la@rc
` y= ] yn}qz= w] + dyw]d` yh!yw+ ~
1 Chr
/ora&-ta# tolu&hl
^ = <yk!lh
= h
) ^ <yp!la
* h
& *
15:25
.hj*mc
= B
! = <d{a-$ db@u) tyB@-/m! hw`hy+-tyr]B=
tr`jM
( m
* ! yh!yw+ ~
1 Sam
11:11
rq#Bh
) ^ tr\mv
) a
= B
^ = hn\jM
& h
^ -^ Eotb= Wab)Yw` ~
<oYh^ <j)-du^ /oMu^-ta# WKY~w~
aOw+ Wxp%Yw` ~ <yr]av
* N= h
] ^ yh!yw+ ~
.dj^y` <y]nv
~ = <b*-Wra&vn= ]
471
The final two cases are a Kings-Chronicles parallel, but with slight differences:
2 Kgs 8:21 has hl*yl
+ ^ <q* aWh-yh!yw+ ,~ whereas 2 Chr 21:9 reads hl*yl
+ ^ <q* yh!yw+ .~ The
unusual feature here is that yh!yw+ ~ is not verbal, but it is not followed by any temporal
expression. In a certain sense, the readings are more similar than they appear at first
glance since the pronoun aWh in 2 Kgs 8:21 is not strictly required syntactically. In both
cases, the convergence of yh!yw+ ~ and <q* is unusual. It would be more in line with the
+ B
^ ^ yh!yw+ .~
temporal expressions in chapter nine if the clause started with hl*yL
Then Joram crossed over to
Zair, and all his chariots with
him. And he arose by night and
struck the Edomites who had
surrounded him and the captains
of the chariots; but his army fled
to their tents.
Pc-vqw3ms np npXd Pc-ncmsc-Pancms PpX3ms Pc-vqw3msXa-pi3ms
vqp3ms ncms Pc-vhw3msXa Po-np
Pa-vqPms PpX3ms Pc-Po ncmpc
Pa-ncms Pc-vqw3ms Pa-ncms PpncmpcX3ms
w{Mu! bk#rh
\ -* lk*w+ hr`yu!x* <r`oy rb)uY& w~ ~
hl*yl
+ ^ <q* aWh-yh!yw+ ~
2 Kgs
8:21
wyl*a@ byb@Sh
) ^ <oda$-ta# hK#Yw~ ~
bk#rh
\ * yr}c* ta@w+
.wyl*ha
* l
) = <u*h* sn`Yw` ~
w{Mu! bk#rh
\ -* lk*w+ wyr`c-* <u! <r`ohy+ rb)uY& w~ ~
hl*yl
+ ^ <q* yh!yw+ ~
wyl*a@ bb@oSh^ <oda$-ta# EY~w~
.bk#rh
` * yr}c* ta@w+
2 Chr
21:9
BIBLIOGRAPHY
Alter, Robert. 1981. The Art of Biblical Narrative. New York: Basic Books.
. 1985. The Art of Biblical Poetry. New York: Basic Books.
Amit, Yairah. 2001. Reading Biblical Narratives: Literary Criticism and the Hebrew
Bible. Minneapolis: Fortress Press.
Andersen, F. I. 1970. The Hebrew Verbless Clause in the Pentateuch. Nashville:
Abingdon.
.1974. The Sentence in Biblical Hebrew. The Hague: Mouton.
Andersen, T. D. 2000. The Evolution of the Hebrew Verbal System. Zeitschrift fr
Althebristik 13: 1-66.
Anderson, S. R. and E. L. Keenan. 1985. Deixis. In Language Typology and Syntactic
Description: Grammatical Categories and the Lexicon, ed. T. Shopen, 259-308.
Cambridge: Cambridge University Press.
Arnold, B. T. and J. H. Choi. 2003. A Guide to Biblical Hebrew Syntax. Cambridge:
Cambridge University Press.
Bandstra, B. L. 1992. Word Order and Emphasis in Biblical Hebrew Narrative: Syntactic
Observations on Genesis 22 from a Discourse Perspective. In Linguistics and
Biblical Hebrew, ed. W. R. Bodine, 109-123. Winona Lake, IN: Eisenbrauns.
Bar-Efrat, S. 1989. Narrative Art in the Bible. Sheffield: Almond Press.
Barr, J. 1983. The Semantics of Biblical Language. London: Oxford University Press.
Bartelmus, R. 1982. HYH. Bedeutung und Funktion eines hebrischen
Allerweltswortes. Arbeiten zu Text und Sprache im Alten Testament 17. St.
Ottilien: EOS Verlag.
Barton, J. 1996. Reading the Old Testament: Method in Biblical Study. Louisville:
Westminster John Knox Press.
472
473
Bayly, C. 1782. An Entrance into the Sacred Language Containing the Necessary Rules
of Hebrew Grammar in English. London: n.p. Quoted in L. McFall, The Enigma
of the Hebrew Verbal System: Solutions from Ewald to the Present Day.
(Sheffield: Almond Press, 1982), 15.
Beaugrande, R. de and W. U. Dressler. 1981. Introduction to Text Linguistics. London:
Longman.
Benigni, A. 1999. The Biblical Hebrew Particle yk From a Discourse Analysis
Perspective. Zeitschrift fr Althebristik 12: 126-45.
Bergen, R. D., ed. 1994. Biblical Hebrew and Discourse Linguistics. Dallas: Summer
Institute of Linguistics.
Berlin, A. 1983. Poetics and Interpretation of Biblical Narrative. Sheffield: Almond
Press.
. 1989. Lexical Cohesion and Biblical Interpretation. Hebrew Studies 30: 29-39.
Berman, R. A. 1994. Relating Events in Narrative: A Crosslinguistic Developmental
Study. Hillsdale, NJ: Lawrence Erlbaum Associates.
Berns, M. S. 1984. Functional Approaches to Language and Language Teaching: Another
Look. In Communicative Competence, ed. S. J. Savignon and M. S. Berns, 3-22.
Reading, Mass.: Addison-Wesley.
Binnick, R. I. 1991. Time and the Verb: A Guide to Tense and Aspect. New York: Oxford
University Press.
Blake, F. R. 1951. A Resurvey of Hebrew Tenses, with an Appendix: Hebrew Influences
on Biblical Aramaic. Rome: Pontificium Institutum Biblicum.
Blakemore, D. 2002. Relevance and Linguistic Meaning: the semantics and pragmatics
of discourse markers. Cambridge: Cambridge University Press.
Blass, R. 1990. Relevance Relations in Discourse: A Study with Special Reference to
Sissala. Cambridge: Cambridge University Press.
Block, D. I. 1997. The Book of Ezekiel, Chapters 1-24. New International Commentary
on the Old Testament. Grand Rapids: Eerdmans.
Bloomfield, L. 1933. Language. New York: Henry Holt and Company.
Boas, F., ed. 1911. Handbook of American Indian Languages. U.S. Bureau of American
Ethnology. Bulletin No. 40. Washington: Government Printing Office.
474
Bodine, W. R., ed. 1995. Discourse Analysis of Biblical Literature: What it is and What it
Offers. Atlanta: Scholars Press.
Brockelmann, C. 1956. Hebrische Syntax. Neukirchen: Neukirchener Verlag.
Brown, F., S. R. Driver, and C. A. Briggs. 1996. The Brown-Driver-Briggs Hebrew and
English Lexicon: With an Appendix Containing the Biblical Aramaic. Peabody,
MA: Hendrickson.
Brown, G. and G. Yule. 1983. Discourse Analysis. Cambridge: Cambridge University
Press.
Bublitz, W., U. Lenk, and E. Ventola, eds. 1999. Coherence in Spoken and Written
Discourse: How to Create it and How to Describe it: Selected Papers from the
International Workshop on Coherence, Augsburg, 24-27 April, 1997. Amsterdam:
John Benjamins.
Buth, R. 1992a. Functional Grammar, Hebrew and Aramaic: An Integrated Textlinguistic
Approach to Syntax. In Linguistics and Biblical Hebrew, ed. W. Bodine, 77-102.
Winona Lake, Ind.: Eisenbrauns.
. 1992b. The Hebrew Verb in Current Discussions. Journal of Translation and
Textlinguistics 5: 91-105.
Bybee, J. L. 1985. Morphology: A Study of the Relation between Meaning and Form.
Amsterdam: John Benjamins Publishing Company.
Canale, M. 1983. From Communicative Competence to Communicative Language
Pedagogy. In Language and Communication, ed. J. C. Richards and R. W.
Schmidt, 2-27. London: Longman.
Carroll, R. P. 1986. Jeremiah: A Commentary. Philadelphia: Westminster Press
Carson, D. A. 1984. Exegetical Fallacies. Grand Rapids: Baker.
Chisholm, R. B. 1998. From Exegesis to Exposition: A Practical Guide to Using Biblical
Hebrew. Grand Rapids: Baker.
Chomsky, Noam. 1977. Essays on Form and Interpretation. New York: North Holland.
Churchland, P. M. 1996. The Engine of Reason, the Seat of the Soul. A Philosophical
Journey into the Brain. Cambridge: MIT Press.
Comrie, B. 1985. Tense. Cambridge: Cambridge University Press.
. 1987. Aspect. Cambridge: Cambridge University Press.
475
. 1989. Language Universals and Linguistic Typology. 2d ed. Chicago: University
of Chicago Press.
Cooke, G. A. 1903. A Text-Book of North-Semitic Inscriptions: Moabite, Hebrew,
Phoenician, Aramaic, Nabataean, Palmyrene, Jewish. Oxford: The Clarendon
Press.
Costermans, J. and M. Fayol, eds. 1996. Processing Interclausal Relationships: Studies
in the Production and Comprehension of Text. Hillsdale, NJ: Lawrence Erlbaum
Associates.
Coulthard, M. 1997. An Introduction to Discourse Analysis. London: Longman.
Crystal, D. 1997. Dictionary of Linguistics and Phonetics. 4th ed. London: Blackwell.
Culicover, P. W. 1982. Syntax. 2d ed. New York: Academic Press.
DeCaen, V. 1996. Ewald and Driver on Biblical Hebrew Aspect: Anteriority and the
Orientalist Framework. Zeitschrift fr Althebristik 9:129-49.
De Regt, L. J. 1999. Participants in Old Testament Texts and the Translator. Amsterdam:
Van Gorcum.
Dik, S. C. 1978. Functional Grammar. Amsterdam: North-Holland Publishing Company.
Dooley, R. A., and S. H. Levinsohn. 2001. Analyzing Discourse: A Manual of Basic
Concepts. Dallas: SIL International.
Driver, S. R. 1998. A Treatise on the Use of the Tenses in Hebrew and Some other
Syntactical Questions. With an Introductory Essay by W. Randall Garr. Grand
Rapids: Eerdmans.
Endo, Y. 1996. The Verbal System of Classical Hebrew in the Joseph Story: An
Approach from Discourse Analysis. Assen: Van Gorcum.
Eskhult, M. 1990. Studies in Verbal Aspect and Narrative Technique in Biblical Hebrew
Prose. Stockholm: Almqvist and Wiksell.
Even-Shoshan, A. 1981. A New Concordance of the Bible. Jerusalem: Kiryat Sepher
Publishing House.
Exter Blokland, A. F. den. 1995. In search of text syntax: towards a syntactic textsegmentation model for biblical Hebrew. Amsterdam: VU University Press.
Fauconnier, G. 1985. Mental Spaces : Aspects of Meaning Construction in Natural
Language. Cambridge: Cambridge University Press.
476
Fee, G. D. 2002. New Testament Exegesis: A Handbook for Students and Pastors. 3d ed.
Louisville: Westminster John Knox.
Fokkelmann, J. P. 1975. Narrative Art in Genesis. Amsterdam: Van Gorcum.
. 1991. Iterative Forms of the Classical Hebrew Verb: Exploring the Triangle of
Style, Syntax, and Text Grammar. In Studies in Hebrew and Aramaic Syntax
presented to Professor J. Hoftijzer, ed. E. Jongeling et al, 38-55. Leiden: Brill.
. 1999. Reading Biblical Narrative: An Introductory Guide. Louisville, KY:
Westminster John Knox Press.
Foley, W. A. and R. D. Van Valin Jr. 1984. Functional Syntax and Universal Grammar.
Cambridge: Cambridge University Press.
Follingstad, C. M. 2001. Deictic Viewpoint in Biblical Hebrew Text: A Syntagmatic and
Paradigmatic Analysis of the Particle yk. Dallas: SIL International.
Futato, M. D. 2003. Beginning Biblical Hebrew. Winona Lake: Eisenbrauns.
Garrett, D. A. 2002. A Modern Grammar for Classical Hebrew. Nashville: Broadman
and Holman.
Gentry, P. J. 1998. The System of the Finite Verb in Classical Biblical Hebrew. Hebrew
Studies 39:7-39.
Gesenius, W., and E. Kautsch. 1910. Gesenius Hebrew Grammar. Translated and
Revised by A. E. Cowley. 2d ed. Oxford: Clarendon Press.
Gibson, J. C. L. 1994. Davidsons Introductory Hebrew Grammar ~ Syntax. Edinburgh:
T. & T. Clark.
Givn, Talmy, ed. 1979. Syntax and Semantics, Vol 12, Discourse and Syntax. New
York: Academic Press.
. 1983. Topic Continuity in Discourse: A Quantitative Cross-Language Study.
Amsterdam: John Benjamins.
. 2001. Syntax: An Introduction. 2d ed. 2 vols. Amsterdam and Philadelphia: John
Benjamins Publishing Company.
Goldfajn, T. 1998. Word Order and Time in Biblical Hebrew. Oxford: Clarendon Press.
Gordon, C. H. 1983. The Waw Conversive: From Eblaite to Hebrew. Proceedings of
the American Academy for Jewish Research 50:90.
477
Greenberg, J. H., C. A. Ferguson, and E. A. Moravcsik, eds. 1978. Universals of Human
Language. 4 vols. Stanford: Stanford University Press.
Grimes, J. 1975. The Thread of Discourse. The Hague: Mouton.
Groom, S. 2003. Linguistic Analysis of Biblical Hebrew. Carlisle: Paternoster Press.
Gropp, D. M. 1991. The Function of the Finite Verb in Classical Biblical Hebrew.
Hebrew Annual Review 13:45-62.
. 1995. Progress and Cohesion in Biblical Hebrew Narrative: the Function of k- /
b- + the Infinitive Construct. In Discourse Analysis of Biblical Literature: What
it is and what it offers, ed. W. R. Bodine, 183-212. Atlanta: Scholars Press.
Gro, W. 1987. Die Pendenskonstruktion im Biblischen Hebrisch. Arbeiten zu Text und
Sprache im Alten Testament 27. St. Ottilien: EOS Verlag, 1987.
Gumperz, J. J. and D. Hymes, eds. 1986. Directions in Sociolinguistics. Rev. and enl. ed.
New York: Blackwell.
Halliday, M. A. K. 1973. Explorations in the Functions of Language. NY: Elsevier.
Halliday, M. A. K., and Ruqaiya Hasan. 1976. Cohesion in English. London: Longman.
Hamilton, V. 1995. The Book of Genesis: Chapters 18-50. New International
Commentary on the Old Testament. Grand Rapids: Eerdmans.
Harley, B., P. A. J. Cummins, and M. Swain, eds. 1990. The Development of Second
Language Proficiency. Cambridge: Cambridge University Press.
Harper, W. R. 1892. Elements of Hebrew Syntax by an Inductive Method. New York:
Charles Scribners Sons.
Hatav, G. 1997. The Semantics of Aspect and Modality: Evidence from English and
Biblical Hebrew. Amsterdam: John Benjamins.
Hatim, B., and I. Mason. 1990. Discourse and the Translator. London: Longman.
Heimerdinger, J-M. 1999. Topic, Focus and Foreground in Ancient Hebrew Narratives.
Sheffield: Sheffield Academic Press.
Hendel, R. S. 1996. In the Margins of the Hebrew Verbal System: Situation, Tense,
Aspect, Mood. Zeitschrift fr Althebristik 9: 152-81.
Hockett, C. 1942. A System of Descriptive Phonology. Language 18:3-21.
478
Hoey, M. 1991. Patterns of Lexis in Text. Oxford: Oxford University Press.
Hopper, P. J. 1979. Aspect and Foregrounding in Discourse. In Syntax and Semantics,
Volume 12: Discourse and Syntax, ed. Talmy Givn, 213-41. New York:
Academic Press.
Hopper, P. J. and E. C. Traugott. 1993. Grammaticalization. Cambridge: Cambridge
University Press.
Hymes, Dell. 1974. Foundations in Sociolinguistics: An Ethnographic Approach.
Philadelphia: University of Pennsylvania Press.
. 1986. Models of the Interaction of Language and Social Life. In Directions in
Sociolinguistics, ed. J. Gumperz and D. H. Hymes, 35-71. Rev. and enl. ed. New
York: Blackwell, 1986.
Jackendoff, R. 1994. Patterns in the Mind: Language and Human Nature. New York:
Basic Books.
Jongeling, K., Murre-Van den Berg, H. L., Rompay, L. van, and Hoftijzer, J., eds. 1991.
Studies in Hebrew and Aramaic Syntax presented to Professor J. Hoftijzer on the
occasion of his sixty-fifth birthday. Leiden: Brill.
Joosten, J. 1989. The Predicative Participle in Biblical Hebrew. Zeitschrift fr
Althebristik 2:128-59.
. 1992. Biblical Hebrew wqAXal and Syriac hwA qAXel Expressing Repetition in the
Past. Zeitschrift fr Althebristik 5:1-14.
. 1999. The Long Form of the Prefix Conjugation Referring to the Past in Biblical
Hebrew Prose. Hebrew Studies 40:15-26.
Joon, P. 1996. A Grammar of Biblical Hebrew. Translated and Revised by Takamitsu
Muraoka. 2 vols. Rome: Editrice Pontificio Instituto Biblico.
JPS Hebrew-English TANAKH. 2000. Philadelphia: The Jewish Publication Society.
Kaiser, W. C. 1981. Toward an Exegetical Theology: Biblical Exegesis for Preaching
and Teaching. Grand Rapids: Baker.
Kelley, P. H. 1992. Biblical Hebrew: An Introductory Grammar. Grand Rapids:
Eerdmans.
Keown, G. L., P. J. Scalise, and T. G. Smothers. 1995. Jeremiah 26-52. Word Biblical
Commentary. Waco: Word Books.
479
Khan, G. 1999. The Karaite Tradition of Hebrew Grammatical Thought. In Hebrew Study
from Ezra to Ben-Yehuda, ed. W. Horbury, 186-203. Edinburgh: T. & T. Clark.
Koehler, L. and Baumgartner, W. 1995. The Hebrew and Aramaic Lexicon of the Old
Testament. Leiden: Brill.
Kuhn, T. S. 1996. The Structure of Scientific Revolutions. 3d ed. Chicago: University of
Chicago Press.
Kustr, P. 1972. Aspekt im Hebrischen. Basel: Friedrich Reinhardt Kommissionsverlag.
Kutscher, E. Y. 1982. A History of the Hebrew Language. Jerusalem: Magnes Press.
Labov, W. 1972. Sociolinguistic Patterns. Philadelphia: University of Pennsylvania
Press.
Lakoff, G. and M. Johnson. 1980. Metaphors We Live By. Chicago: Chicago University
Press.
. 1999. Philosophy in the Flesh: The Embodied Mind and its Challenge to
Western Thought. New York: Basic Books.
Lambdin, T. O. 1971. Introduction to Biblical Hebrew. New York: Scribner.
Lambrecht, K. 1994. Information Structure and Sentence Form: Topic, Focus and the
Mental Representations of Discourse Referents. Cambridge: Cambridge
University Press.
Leder, A. C. 1999. Reading Exodus to Learn and Learning to Read Exodus. Calvin
Theological Journal 34:11-35.
. 2001. The Coherence of Exodus: Narrative Unity and Meaning. Calvin
Theological Journal 36:251-69.
Lettinga, J. P. 1980. Grammaire de LHbreu Biblique. 2 vols. Leiden: Brill.
Levinsohn, S. H. 1982. Textual Connections in Acts. Atlanta: Scholars Press.
Lode, L. 1984. Postverbal Word Order in Biblical Hebrew: Structure and Function.
Semitics 9:113-64.
. 1985. Postverbal Word Order in Biblical Hebrew: Structure and Function, Part
Two. Semitics 10:24-39.
Long, G. 2002. Grammatical Concepts 101 for Biblical Hebrew. Peabody: Hendrickson.
480
Longacre, R. E. 1986. Who Sold Joseph into Egypt? In Interpretation and History:
Essays in honour of Allan A. MacRae, ed. R. L. Harris, Swee-Hwa Quek, and J.
R. Vannoy, 75-91. Singapore: Christian Life Publishers.
. 1989. Joseph: A Story of Divine Providence. A Text Theoretical and
Textlinguistic Analysis of Genesis 37 and 39-48. Winona Lake, IN: Eisenbrauns.
. 1992a. Discourse Perspective on the Hebrew Verb: Affirmation and
Restatement. In Linguistics and Biblical Hebrew, ed. W. R. Bodine, 177-89.
Winona Lake, IN: Eisenbrauns.
. 1992b. The Analysis of Preverbal Nouns in Biblical Hebrew Narrative: Some
Overriding Concerns. Journal of Translation and Textlinguistics 5: 208-224.
. 1996. The Grammar of Discourse. 2d ed. New York: Plenum, 1996.
Longacre, R. E., and S. J. J. Hwang. 1994. A Textlinguistic Approach to the Biblical
Hebrew Narrative of Jonah. In Biblical Hebrew and Discourse Linguistics, ed. R.
D. Bergen, 336-58. Dallas: Summer Institute of Linguistics.
Longman, T. 1988. How to Read the Psalms. Downers Grove: InterVarsity Press.
. 1996. Literary Approaches to Biblical Interpretation. In Foundations of
Contemporary Interpretation, ed. M. Silva, 91-192. Grand Rapids: Zondervan.
Louw, J. P. 1982. Semantics of New Testament Greek. Atlanta: Scholars Press.
. 1986. Sociolinguistics and Its Role in Text Analysis. In Sociolinguistics and
Communication, ed. Johannes P. Louw, 103-15. London: United Bible Societies.
Lowery, K. E. 1985. Toward a Discourse Grammar of Biblical Hebrew. Ph.D. diss., The
University of California, Los Angeles.
. 1995. The Theoretical Foundations of Hebrew Discourse Grammar. In
Discourse Analysis of Biblical Literature: What it is and What it Offers, ed. W.
Bodine, 103-30. Atlanta: Scholars Press.
Lubuchange, C. J., C. Van Leeuwen, M. J. Mulder, and H. A. Brongers. 1997. Syntax and
Meaning: Studies in Hebrew Syntax and Biblical Exegesis. Leiden: Brill.
Lyons, J. 1969. Introduction to Theoretical Linguistics. Cambridge: Cambridge
University Press.
MacDonald, P. J. 1992. Discourse Analysis and Biblical Interpretation. In Linguistics and
Biblical Hebrew, ed. W. R. Bodine, 153-75. Winona Lake, IN: Eisenbrauns.
481
Machen, G. 2004. New Testament Greek for Beginners. 2d ed. Revised by D. G.
McCartney. Upper Saddle River, NJ: Pearson Prentice Hall.
Malmkjr, K. 1991. The Linguistics Encyclopedia. London: Routledge.
Mann, W. C. and S. A. Thompson. 1987. Rhetorical Structure Theory: A Theory of Text
Organization. Technical Report ISI/RS-87-190. Marina del Ray: Information
Sciences Institute.
Martin, J. D. 1993. Davidsons Introductory Hebrew Grammar. Edinburgh: T. & T.
Clark.
McCarthy, M. and R. A. Carter. 1994. Language as Discourse: Perspectives for
Language Teaching. London: Longman.
McFall, Leslie. 1982. The Enigma of the Hebrew Verbal System: Solutions from Ewald to
the Present Day. Sheffield: Almond Press.
Meek, T. J. 1945. The Syntax of the Sentence in Hebrew. JBL 64, 1-13.
Miller, C. L. 1996. The Representation of Speech in Biblical Hebrew Narrative: A
Linguistic Analysis. Atlanta: Scholars Press.
. 1999. The Pragmatics of waw as a Discourse Marker in Biblical Hebrew
Dialogue. Zeitschrift fr Althebristik 12: 165-91
Mitchel, L. A. 1984. A Students Vocabulary for Biblical Hebrew and Aramaic. Grand
Rapids: Zondervan.
Mller, A. 1883. Outlines of Hebrew Syntax. Glasgow: Maclehose.
Muraoka, Takamitsu. 1983. Emphatic Words and Structures in Biblical Hebrew.
Jerusalem: Magnes Press.
Myhill, John. 1992. Word Order and Temporal Sequencing. In Pragmatics of Word
Order Flexibility, ed. D. L. Payne, 265-78. Amsterdam: John Benjamins.
New American Standard Bible. 1995. Updated ed. Anaheim, CA: Lockman Foundation.
New International Version. 1984. International Bible Society. Grand Rapids: Zondervan.
New Revised Standard Version. 1989. Division of Christian Education of the National
Council of the Churches of Christ in the U.S.A.
Newmeyer, F. J. 1980. Linguistic Theory in America: The First Quarter-Century of
Transformational Generative Grammar. New York: Academic Press,
482
Niccacci, A. 1990. The Syntax of the Verb in Classical Hebrew Prose. Sheffield:
Sheffield Academic Press
. 1991. Lettura Sintattica della Prosa Ebraico-Biblica: Principi e Applicazioni.
Jerusalem: Franciscan Printing Press.
. 1994. On the Hebrew Verbal System. In Biblical Hebrew and Discourse
Linguistics, ed. R. D. Bergen, 117-37. Dallas: Summer Institute of Linguistics.
. 1997. Basic Facts and Theory of the Biblical Hebrew Verb System in Prose. In
Narrative Syntax and the Hebrew Bible: Papers of the Tilburg Conference 1996,
ed. E. van Wolde, 167-202. Leiden: Brill.
Payne, T. E. 1997. Describing Morphosyntax: A Guide for Field Linguists. Cambridge:
Cambridge University Press.
Peckham, B. 1997. Tense and Mood in Biblical Hebrew. Zeitschrift fr Althebristik 10:
139-68.
Pike, K. L. 1967. Language in Relation to a Unified Theory of the Structure of Human
Behavior. 2d ed. The Hague: Mouton.
. 1981. Tagmemics, Discourse, and Verbal Art. Ann Arbor, MI: University of
Michigan.
. 1982. Linguistic Concepts: An Introduction to Tagmemics. Lincoln: The
University of Nebraska Press.
Pike, K. L. and E. G. Pike. 1991. Grammatical Analysis. 2d ed. Dallas: Summer Institute
of Linguistics.
Pratico, G. D. and M. V. Van Pelt. 2001. Basics of Biblical Hebrew Grammar. Grand
Rapids: Zondervan.
Putnam, F. C. 1996. Hebrew Bible Insert: A Students Guide to the Syntax of Biblical
Hebrew. Ridley Park, PA: Stylus.
. 2003. An Introduction to the Grammar and Syntax of Biblical Hebrew.
Unpublished Manuscript.
Regt, L.J. de. 1991. Word Order in Different Clause Types in Deuteronomy 1-30. In
Studies in Hebrew and Aramaic Syntax presented to Professor J. Hoftijzer on the
occasion of his sixty-fifth birthday, ed. K. Jongeling et al, 152-72. Leiden: Brill.
. 1999. Participants in Old Testament Texts and the Translator: Reference
Devices and Their Rhetorical Impact. Assen: Van Gorcum.
483
Rendsburg, G. 1990. Diglossia in Ancient Hebrew. New Haven, CT: American Oriental
Society.
Renkema, J. 1993. Discourse Studies: An Introductory Textbook. Amsterdam: John
Benjamins.
Reuchlin, J. 1506. Rudimenta linguae hebraicae. n.p.
Revell, E. J. 1991. Conditional Sentences in Biblical Hebrew Prose. In Studies in Honor
of Wolf Leslau on the Occasion of his 85th Birthday, 2 vols., ed. A. Kaye, 127890. Wiesbaden: Otto Harrassowitz.
Richter, W. 1978. Grundlagen einer althebrischen Grammatik. Arbeiten zu Text und
Sprache im Alten Testament 8. St. Ottilien: EOS Verlag.
. 1979. Grundlagen einer althebrischen Grammatik. Arbeiten zu Text und
Sprache im Alten Testament 10. St. Ottilien: EOS Verlag.
. 1980. Grundlagen einer althebrischen Grammatik. Arbeiten zu Text und
Sprache im Alten Testament 13. St. Ottilien: EOS Verlag.
Risselada, R. and W. Spooren. 1998. Introduction: Discourse Markers and Coherence
Relations. Journal of Pragmatics 30: 131-33.
Robins, R. H. 1968. General Linguistics: An Introductory Survey. Bloomington: Indiana
University Press.
Rocine, B. M. 2000. Learning Biblical Hebrew: A New Approach Using Discourse
Analysis. Macon, GA: Smyth and Helwys.
Rosenbaum, M. 1997. Word-Order Variation in Isaiah 40-55: A Functional Perspective.
Assen: Van Gorcum.
Ross, A. P. 2001. Introducing Biblical Hebrew. Grand Rapids: Baker.
Rutherford, W. E. 1987. Second Language Grammar: Learning and Teaching. London:
Longman.
Senz-Badillos, A. 1993. A History of the Hebrew Language. Cambridge: Cambridge
University Press.
Sampson, G. 1980. Schools of Linguistics. Stanford: Stanford University Press, 1980.
Sapir, E. 1921. Language: An Introduction to the Study of Speech. New York: Harcourt,
Brace.
484
Saussure, F. de. 1916. Cours de linguistique gnrale. Paris: Payot.
Savignon, S. J. 1983. Communicative Competence: Theory and Classroom Practice.
Texts and Contexts in Second Language Learning. Reading, Mass.: AddisonWesley.
Saville-Troike, M. 1989. The Ethnography of Communication: An Introduction. 2d ed.
New York: Blackwell.
Schertz, M. H. and P. B. Yoder. 2001. Seeing the Text: Exegesis for Students of Greek
and Hebrew. Nashville: Abingdon Press.
Schiffrin, D. 1988. Discourse Markers. Cambridge: Cambridge University Press.
Schneider, W. 1974. Grammatik des biblischen Hebrisch. Ein Lehrbuch. Munich:
Claudius Verlag.
Seow, C. L. 1987. A Grammar for Biblical Hebrew. Nashville: Abingdon.
Shankara Bhat, D. N. 1999. The Prominence of Tense, Aspect, and Mood. Amsterdam;
Philadelphia: John Benjamins.
Shimasaki, Katsuomi. 2002. Focus Structure in Biblical Hebrew: A Study of Word Order
and Information Structure. Bethesda, MD: CDL Press.
Shopen, T., ed. 1985. Language Typology and Syntactic Description. 3 vols. Cambridge:
Cambridge University Press.
Silva, M. 1994. Biblical Words and Their Meaning: An Introduction to Lexical
Semantics. 2d ed. Grand Rapids: Zondervan.
. 1996a. God, Language and Scripture. In Foundations of Contemporary
Interpretation, ed. M. Silva, 193-280. Grand Rapids: Zondervan.
. 1996b. Explorations in Exegetical Method. Grand Rapids: Zondervan.
. 2004. Foreword to New Testament Greek for Beginners, by G. Machen. 2d ed.
Revised by D. G. McCartney. Upper Saddle River, NJ: Pearson Prentice Hall.
Ska, J. L. 1990.Our Fathers Have Told Us Introduction to the Analysis of Hebrew
Narratives Rome: Editrice Pontificio Istituto Biblico.
Smith, M. S. 1991. The Waw-Consecutive at Qumran. Zeitschrift fr Althebristik 4: 16164.
485
Speiser, E. A. 1964. Genesis: Introduction, Translation, and Notes. Garden City:
Doubleday and Company.
Sperber, A. 1996. A Historical Grammar of Biblical Hebrew: A Presentation of Problems
with Suggestions to Their Solution. Leiden: Brill.
Sperber, D. and D. Wilson. 1995. Relevance: Communication and Cognition. Oxford:
Blackwell.
Stabnow, D. K. 2000. A Discourse Analysis Perspective on the Syntax of Clauses
Negated by al in the Primary History. Ph.D. diss., Westminster Theological
Seminary.
Sternberg, M. 1985. The Poetics of Biblical Narrative: Ideological Literature and the
Drama of Reading Bloomington: Indiana University Press.
Stoddard, S. 1990. Text and Texture: Patterns of Cohesion. Norwood, N.J.: Ablex
Publishing Corporation.
Stuart, D. 1987. Hosea-Jonah. Word Biblical Commentary. Waco: Word Books.
. 2001. Old Testament Exegesis: A Handbook for Students and Pastors. 3d. ed.
Louisville: Westminster John Knox.
Stubbs, M. 1983. Discourse Analysis: The Sociolinguistic Analysis of Natural Language
Chicago: University of Chicago Press.
Talstra, Eep. 1978. Text Grammar and Hebrew Bible. I: Elements of a Theory.
Bibliotheca Orientalis 35:169-74.
. 1982. Text Grammar and Hebrew Bible. II: Syntax and Semantics. Bibliotheca
Orientalis 39: 26-38.
______. 1991. Hebrew Syntax: Clause Types and Clause Hierarchy. Pages 180-93 in
Studies in Hebrew and Aramaic Syntax presented to Professor J. Hoftijzer. Edited
by E. Jongeling et al. Leiden: E. J. Brill, 1991.
______. 1992. Text Grammar and Biblical Hebrew: The Viewpoint of Wolfgang
Schneider. Journal of Translation and Textlinguistics 5:269-97.
. 1997a. A Hierarchy of Clauses in Biblical Hebrew Narrative. In Narrative
Syntax and the Hebrew Bible: Papers of the Tilburg Conference 1996, ed. E. van
Wolde, 85-118. Leiden: Brill.
486
. 1997b. Workshop: Clause Types, Textual Hierarchy, Translation in Exodus 19,
20, and 24. In Narrative Syntax and the Hebrew Bible: Papers of the Tilburg
Conference 1996, ed. E. van Wolde, 119-32. Leiden: Brill.
Tene, D. 1971. Hebrew Linguistic Literature. In Encyclopedia Judaica, vol. 16, cols
1352-90. Jerusalem: Keter.
Thompson, S. A. 1992. Functional Grammar. In The International Encyclopedia of
Linguistics, ed. William Bright, 37. Oxford: Oxford University Press.
Tomlin, R. S. 1987. Coherence and Grounding in Discourse. Amsterdam: John
Benjamins.
Translators Workplace 4.0. 2002. Dallas: SIL International.
van der Merwe, C. H. J. 1987. A Short Survey of Major Contributions to the
Grammatical Description of Old Hebrew Since 1800 AD. Journal of Northwest
Semitic Languages 13: 161-90.
. 1989. Recent Trends in the Linguistic Description of Old Hebrew. Journal of
Northwest Semitic Languages 15: 217-41.
______. 1990. The Old Hebrew Particle gam: A Syntactic-Semantic Description of gam
in Gn-2Kg. Arbeiten zu Text und Sprache im Alten Testament 34. St. Ottilien:
EOS Verlag.
______. 1991. The Function of Word Order in Old Hebrewwith Special Reference to
Cases where a Syntagmeme precedes a Verb in Joshua. Journal of Northwest
Semitic Languages 17:129-44.
______. 1991b. The Old Hebrew Particles Ea and qr. In Text, Methode und Grammatik:
Wolfgang Richter zum 65. Geburtstag, ed. W. Gross, H. Irsigler and T. Siedl, 297311. St. Ottilien: EOS Verlag.
______. 1993. Old Hebrew Particles and the Interpretation of Old Testament Texts.
Journal for the Study of the Old Testament 18: 27-44.
. 1994. Discourse Linguistics and Biblical Hebrew Grammar. In Biblical Hebrew
and Discourse Linguistics, ed. R. D. Bergen, 13-49. Dallas: Summer Institute of
Linguistics.
______. 1997a. Reconsidering Biblical Hebrew Temporal Expressions. Zeitschrift fr
Althebraistik 10: 42-62.
487
______. 1997b. Reference Time in Some Biblical Temporal Constructions. Biblica 78:
403-424.
______. 1997c. A Critical Analysis of Narrative Syntactic Approaches, With Special
Attention to Their Relationship to Discourse Analysis. In Narrative Syntax and
the Hebrew Bible: Papers of the Tilburg Conference 1996, ed. Ellen van Wolde,
133-56. Leiden: Brill.
______. 1999a. The Elusive Biblical Hebrew Term yhyw: A Perspective in Terms of Its
Syntax, Semantics, and Pragmatics in 1 Samuel. Hebrew Studies 40: 83-114.
______. 1999b. Towards a Better Understanding of Biblical Hebrew Word Order.
Journal of Northwest Semitic Languages 25/1 (1999): 277-300.
______. 1999c. Explaining Fronting in Biblical Hebrew. Journal of Northwest Semitic
Languages 25/2: 173-86.
van der Merwe, C. H. J., J. A. Naud, and J. H. Kroeze. 1999. A Biblical Hebrew
Reference Grammar. Sheffield: Sheffield Academic Press.
Van Dijk, T. 1977. Text and Context: Explorations in the Semantics and Pragmatics of
Discourse. London and New York: Longman.
Van Pelt, M. V. and G. D. Pratico. 2003. The Vocabulary Guide to Biblical Hebrew.
Grand Rapids: Zondervan.
Van Wolde, E., ed. 1997. Narrative Syntax and the Hebrew Bible: Papers of the Tilburg
Conference 1996. Leiden: Brill.
Wallace, S. 1982. Figure and Ground: The Interrelationships of Linguistic Categories. In
Tense-Aspect: Between Semantics and Pragmatics, ed. Paul J. Hopper, 201-223.
Amsterdam and Philadelphia: John Benjamins Publishing Company.
Waltke, B. and M. OConnor. 1990. An Introduction to Biblical Hebrew Syntax. Winona
Lake, Ind.: Eisenbrauns.
Waltke, B. and C. J. Fredericks. 2001. Genesis: A Commentary. Grand Rapids:
Zondervan, 2001.
Weingreen, J. 1959. A Practical Grammar for Classical Hebrew. 2d ed. Oxford:
Clarendon Press.
Wendland, E. R. 2001. May the Whole World Hush in His Presence! (Habakkuk
2:20b): Communicating Aspects of the Rhetoric of an Ancient Biblical Text
Today. Journal of Northwest Semitic Languages 27/2: 113-33.
488
Wenham, G. 1994. Genesis 16-50. Word Biblical Commentary. Waco: Word Books.
Wildberger, H. 1991. Isaiah 28-39. The Continental Commentary. Minneapolis: Fortress
Press.
Williams, R. J. 1976. Hebrew Syntax: An Outline. 2d. ed. Toronto: University of Toronto
Press.
Winther-Nielsen, N. 1995. A Functional Discourse Grammar of Joshua. A Computerassisted Rhetorical Structure Analysis. Stockholm: Almqvist and Wiksell
International.
Zevit, Z. 1998. The Anterior Construction in Classical Hebrew. Atlanta: Scholars Press.
Zewi, T. 1999. Interrupted Syntactical Structures in Biblical Hebrew. Zeitschrift fr
Althebristik 12: 83-95.
CURRICULUM VITA
Bryan L. Harmelink
2561 Trewigtown Rd.
Colmar, PA 18915
(215) 996-1511
[email protected]
Education:
SIL Experience:
Teaching Experience:
Research Experience:
Research Associate, Toponyms of the Eighth Region with Dr. Mario Bernales,
Universidad de la Frontera, 1991
Research Associate, Mapuche Oral Myths, with Dr. Hugo Carrasco, Universidad
de la Frontera, 1989-90
Research Associate, Rural Education Project, with Patricia de la Pea and Lilian
Gonzlez, Universidad de la Frontera, 1987-89
Research Associate, Toponyms of the Ninth Region with Dr. Mario Bernales,
Universidad de la Frontera, 1987-89
Publications:
1997
1996
La Vida de los Judos en los Tiempos Bblicos. Pucallpa: Misin Suiza en el Per.
Spanish translation of Daily Life in Bible Times, WBT, Dallas, 1988.
1994
1993
1993
1991
1990
1990
1990
1990
1990
1990
1989
1989
1988
1987
1987
1986
1985
1985
1984
1993
1993
1989
1988
1988
1988
1988
1987
1987
1987
1987
1987
1987
Kmeke Dungun Tai Kimael. Glossary of Key Terms in the New Testament.
Temuco: Imprenta y Editorial Kme Dungu.
1999
1995
1994
1993
1992
1990
1988
1987
1986