0% found this document useful (0 votes)
391 views151 pages

Introduction To The Pentateuch

Uploaded by

akko0453
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
391 views151 pages

Introduction To The Pentateuch

Uploaded by

akko0453
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 151

INTRODUCTION

to the
PENTATEUCH

R. N. Whybray .

WILLIAM B. EERDMANS PUBLISHING COMPANY


GRAND RAPms, MICHIGAN ·
© 1995 Wm. B. Eerdmans Publishing Co.
255 Jefferson Ave. S.E., Grand Rapids, Michigan 49503
All rights reserved

Printed in the United States of America

00 99 98 97 96 95 7 6 5 4 3 2 1

ISBN 0-8028-0837-9

Scripture quotations are from the New Revised Standard Version of the Bible ©
1989 by the Division of Christian Education, National Council of the Churches
of r'nrist in the United States of America, used by permission.

Note: Biblical references are given as in modern English Bibles rather than as in
Hebrew Bibles, where the verse numbers sometimes differ slightly.
Contents

Abbreviations Vl

Preface Vll

1. What Is the Pentateuch? 1

2. Who Wrote It? Problems of Composition 12

3. The "Primeval History'' (Genesis 1-11) 29

4. The History of the Patriarchs (Genesis 12-50) 49

5. Exodus, Leviticus, Numbers: Narratives 63

6. The Book of Deuteronomy 85

7. The Laws 107

8. Reading the Pentateuch 133

Index 144

v
Abbreviations

AB Anchor Bible
ANET Ancient Near Eastern Texts Relating to the Old Testament, ed.
J.B. Pritchard (Princeton: Princeton University Press, 1 950,
31969)
BA Biblical Archaeologist
BZAW Beihefte zur Zeitschrift for die Alttestamentliche Wissenschaft
CBQ Catholic Biblical Quarterly .
CBSC Cambridge Bible for Schools and Colleges
ICC International Critical Commentary
]BL Journal ofBiblical Literature
]SOT journal for the Study of the Old Testament
NCBC New Century Bible Commentary
NRSV New Revised Standard Version
OTG Old Testament Guides
OTL Old Testament Library
RB Revue Biblique
REB Revised English Bible
SBT Studies in Biblical T heology
SNTS Studiorum Novi Testamenti Societas
VT Vetus Testamentum
VTS Supplements to Vetus Testamentum
WBC Word Biblical Commentary

Vl
Preface

N
I
THIS TEXTBOOK, distinguished biblical scholar R. Norman Whybray
provides a straightforward and insightful introduction to the back­
ground, content, and themes of one of the major portions of the biblical
corpus. T his entry-level resource for colleges and seminaries strives to make
sense of major critical issues in a field where new and conflicting theories
abound, by not only surveying recent studies but also introducing students
to the major contributions of earlier scholars that have brought biblical
studies to this point. Boldy delineating and analyzing the cutting edge of
current literary, historical, and sociological approaches to biblical criticism,
the author stresses the intention and meaning of the biblical text as a whole
in its final ("canonical") form, remaining sensitive to its literary merit,
theological import, and compelling power as the word of God.
Recognizing the needs of contemporary students, many of whom
come to theological studies as a second career with no previous biblical or
even humanities training, this introduction responds to the demand for a
clear, comprehensive presentation of pertinent issues and data, set forth
succinctly and with helpful explanations of technical material.

T HE PUBLISHERS

vii
CHAPTER 1

What ls the Pentateuch? -_

T
HE TERM "Pentateuch'' is used by scholars to designate the first five
hooks of the Old Testament {Genesis to Deuteronomy), which have
been regarded since early times as the first of its three major divisions. T he
other two divisions are the Prophets (the "Former Prophets" - Joshua,
Judges, Samuel, and Kings - and the "Latter Prophets" - Isaiah,
Jeremiah, Ezekiel, and the "Book of the Twelve, " the so-called Minor
Prophets from Hosea to Malachi) and the Writings, comprising the other
hooks.
T he Pentateuch has always been an essential part of Holy Scripture,
recognized as such by Jews and Christians alike. For the Jews, whose name
for it is "the Torah," it holds the first and most authoritative place in their
Scriptures, being traditionally regarded as the work of Moses. As the only
person who spoke with God face to face {Exod. 33: 1 1 ; Deut. 34: 1 0) ,
Moses was God's most authoritative spokesman, communicating the will
of God to his people. For Christians also the Pentateuch is, together with
the rest of the Old Testament, an essential part of the Holy Scriptures. In
the Gospels, Jesus is represented as quoting or alluding to the authoritative
teaching of "Moses" {i.e. , the Pentateuch) more frequently than to any
other Old Testament hook; and references to it by the other New Testament
.
writers are even more numerous.
In modern times some scholars have questioned the appropriateness
of this traditional way of dividing the Old Testament hooks. It has been
argued, on the one hand, that it is in the hook of Joshua, with the account
of Israel's settlement in the Promised Land, that the true conclusion of
the Pentateuchal story is to he found. (T hese scholars speak of a "Hexa­
teuch, " meaning a group of six hooks. ) Others contend that the real

1
INTRODUCTION TO THE PENTATEUCH

conclusion of the story is to be found even later, with the history of the
monarchy, in the books of Samuel or Kings. On the other hand, some
scholars (e.g., Martin Noth) speak of a "Tetrateuch'' (four books) , on the
grounds that Deuteronomy does not properly belong with the previous
books but marks the beginning of another major historical work, the
so-called Deuteronomistic History, comprising Deuteronomy, Joshua,
Judges, Samuel, and Kings. Each of these proposals has merit. It is true
that God's promise to Abraham that his descendants will possess and
occupy the land of Canaan - a theme that in one way or another dom­
inates the whole of the subsequent events - remains unfulfilled in the
Pentateuchal narrative and only reaches its completion in Joshua. However,
it is also a fact that Deuteronomy, which looks forward to the occupation
of the land as the fulfillment of the original promise, makes an appropriate
starting point for the events narrated in the books that follow. In addition,
the language, style, and theological ideas of Deuteronomy have a greater
affinity with those later books than with the preceding ones.
However, there are good reasons for retaining the notion of a Pen­
tateuch and the traditional division of the text. The ancient title of "the
Five Books of Moses," although unacceptable to modern scholarship as a
statement about authorship, is not entirely inappropriate as a statement
about the content of the story; Admittedly, the book of Genesis is entirely
concerned with .persons who lived and events that took place before Moses
was born. However, the following four books are entirely dominated, from
the human point of view, by the figure of Moses, whose birth is recorded
at the very beginning of those books (Exod. 2:2) and his death at their
very end (Deut. 34:5) . That his death marked the "end of an era" is
emphasized by a final passage (Deut. 34: 1 0 - 1 2), which asserts the unique
importance of the figure of Moses. The book ofJoshua, which begins with
a reference to Moses' death and proceeds immediately to the divine com�
missioning of his successor Joshua to lead the people into the land, clearly
marks the start of an entirely new age. Later Judaism was to look back to
Moses as the person who, under God, had not only laid the foundations
of the subsequent life of Israel but furnished it so completely with its'
religious institutions. that it needed nothing more to guide and sustain it
as the. uniquely cho�en people of God. Viewed in this way, Genesis may
be seen as an introduction to, or preparation for, that unique era.

2
What Is the Pentateuch?

The Contents of the Pentateuch

Genesis 1 - 1 1 (the so-called Primeval History) has a peculiar character of


it� own. It depicts a world in which the supernatural is commonplace -
where God (and the "sons of God," Gen. 6: 1 -2) can converse and have
relationships with human beings as a matter of course, and where snakes
<;:an have conversations with men and women. It begins with the creation
of the world (heaven and earth) together with the heavenly bodies, the
flora and fauna, and human beings, and then passes immediately to
describe a series of human attempts to frustrate God's purpose for human­
ity and to claim for human beings powers and privileges which . God had
not intended them to have: the disobedience of the first man and woman
(ch. 3), the destruction of a fellow human being (ch. 4); a semi-divine
status (6: 1 -4) . The total corruption of the human race with the exception
of Noah leads God to determine its destructipn (ch. 6) . This is carried out
by means of a devastating flood; but Noah and his family, together with
representative specimens of the animal species, are saved by God's direction
to build a boat (or "ark") on which they remain until the flood has subsided
(chs. 7-8) . So God plans, through Noah and his sons, to build a new
humanity and a new human society. He promises never again to send such
a flood on the earth to destroy the human race, establishing an "everlasting
covenant" with it and giving the rainbow as a sign of this promise (ch. 9) .
But the Primeval History does not end on this hopeful note. On the
CO?-trary, ch. 1 1 recounts that a later generation once more attempted to
defy God by building "a city and a tower with its top in the heavens" (the
To . wer of Babel) , so making themselves as powerful as God himself ( 1 1 : 6) .
God responded by confusing the speech of the rebels so that they no longer
understood one another, and scattering them over the face of the earth
( 1 1 :6-9) . This part of Genesis thus ends on a negative and menacing note
which bodes ill for humanity's future well-being, as far as the narrative
thread is concerned.
It is important to pay attention to the genealogical tables in these
chapters (4: 1 7-22; 5 : 1 -32; 1 0 : 1 -32; 1 1 : 1 0-29) . One of the functions of
. these is to link the narratives together in a chronological series; but "they
also provide various kinds of additional information. Gen. 4: 1 7-22 intro­
duces the reader to those people who invented certain arts and ways of
life beneficial to mankind; ch. 10 is an attempt to account for the origins
and geographical locations of the nations of the world. More important
for the structure and continuity of the Pentateuchal story is the genealogy

3
INTRODUCTION TO THE PENTATEUCH

in 1 1 : 1 0-29, which gives the ancestry of Abram (Abraham in the later


chapters) . This introduces the significant statement with which this section
of Genesis ends ( 1 1 :3 1-32) that Abram's father Terah migrated with his
family, including Abram and his wife Sarai (later Sarah) from Ur of the
Chaldeans with the intention of going to the land of Canaan, but in fact
settled in Haran. With this short narrative the link is made between the
Primeval History and the main narratives of Genesis, the stories of the
"patriarchs."
Genesis 12-50 is the story of a family. It concentrates on four
generations of that family, first on three generations of individuals -
father, son, and grandson - and then on the twelve sons of the head of
the third generation, Jacob, who after wrestling with a mysterious adversary
at the ford of the Jabbok, is renamed "Israel" (32:28) . It is then clearly
stated near the end of the book that Jacob's twelve sons "are the twelve
tribes of lsrael" (49:28) . These chapters, then, take the story from the first
mention of Abram in 1 1 :26 to the first mention of Israel as a people, a
people blessed by God with a special blessing. It is here that the subject
of the Pentateuch first becomes clear: it is to be about Israel and about
God's gracious nurture and protection of Israel in the remote past. The
selective genealogies of chs. 1 - 1 1 have progressively narrowed down their
scope and interest to one individual - Abram. Now the genealogy is
widened out again to that of a larger group - the sons ofJacob/Israel and
their families, the ancestors of what is to become a great nation.
Direct communication by God to individual persons, which we have
seen to characterize chs. 1 - 1 1 , continues with Abraham, Isaac, and Jacob.
God's first words to Abraham, in which he commands him to leave Haran
and proceed to an unknown destination, initiate a theme which recurs
several times and dominates these "patriarchal" stories: the promise, in this
case ( 1 2 : 1 -3) that Abraham's descendants will become a great nation and
that he will be especially blessed. In subsequent passages (especially 17:4-8;
22: 17- 1 8; 26:3-5, 24; 28 : 1 3-15; 35: 1 0- 12) the promise is repeated, first
to Abraham himself and then to Isaac and Jacob, in somewhat different
words, and with one notable addition: a promise of possession of the land
of Canaan.
A further crucial theme which runs through these chapters is that of
God's constant care in the protection of the lives of Abraham and his
family from dangers and other circumstances which threaten them and so
threatel'l the fulfillment of the promises. Twice Abraham and Sarah in their
wanderings undertaken in obedience to God's initial command are

4
What Is the Pentateuch?

miraculously saved from the danger of death at the hands of foreign kings
- from Pharaoh ( 1 2 : 1 0-20) and from the king of Gerar (20: 1 - 1 8) - and
materially rewarded instead. Isaac is similarly preserved (26: 1 - 1 6). Sarah's
ip.abilicy to bear children because of her advanced age is miraculously
overco � e, with the result that Isaac, Abraham's heir, is born ( 1 7: 1 5- 1 9;
21: 1 -7). The life of Jacob, threatened by his brother Esau, is saved (ch.
27), and he is also preserved from other dangers (e.g., chs. 3 1 , 33). Joseph
escapes from his brothers, who intend to kill him (ch. 37) , and becomes
a great man in Egypt, second only to Pharaoh himself (4 1 :37-45) , and is
thus enabled to save his father and brothers from starvation (chs. 42-46) .
.
The theme of divine providential care is put into words by Joseph himself
(45 :7-8; 50:20), summing up the whole patriarchal story.
In the book of Exodus it is already a people rather than a family
whose fortunes are recounted, as indicated in Exod. 1 : 1 -7. The expression
"sons oflsrael" (Hebrew beney yisrael), which in v. 1 refers quite literally to
Jacob's sons - the individuals who _entered Egypt with their families to
live there at Joseph's suggestion and at Pharaoh's invitation, seventy persons
all told - has already acquired in v. 7 the meaning "Israelites," which it
is to retain throughout Israel's history (so translated, e.g., in NRSV).
With Exod. 1 :8 there begins a new era. Joseph is dead, and a new
pharaoh knows nothing about him. Pharaoh is, however, very aware of
Israel as a people, whose numbers have now become so great that they
constitute a threat to Egypt's security. The first chapters of Exodus recount
yet another threat to God's plan, this time on a massive scale: an attempt
to exterminate, or at least to enslave, an entire people (chs. 1 -2). But again,
as in his previous dealings with the ancestors, God intervenes to save his
people from the danger which threatens them. The instrument chosen for
this purpose is Moses, who is himself destined with other male Israelite
babies to be killed at birth, but whose life is saved in an incident which
is ostensibly fortuitous (2: 1 -10) . .& God's emissary (chs. 3-4) the adult
Moses secures the release of the Israelites from Egypt by means of a
devastating series of plagues which demonstrate God's overwhelming
power (chs. 7-1 2), and they depart from Egypt.
Pharaoh's change of mind about letting the Israelites go ( 14:5) creates
a new danger: they are now pursued with hostile intent by the Egyptian
army. But again the threat is averted. The Israelites escape by means of a
miraculous crossing of the "sea'' (Hebrew yam suph, probably "Sea of
Reeds") on dry land, while the pursuing Egyptian troops are drowned (ch.
1 4).

5
INTRODUCTION TO THE PENTATEUCH

The Israelites' release from bondage in Egypt and the subsequent


crossing of the "sea" are represented as the supreme demonstration of the
power of the God of Israel as well as his love for Israel, and "the people
feared the Lord and believed in the Lord and in his servant Moses." Yet
the book of Exodus also introduces a quite different theme: Israel's constant
ingratitude and lack of faith. This already appears in a reluctance to accept
Moses as God's emissary (5:2 1 ) , and then becomes a constant motif in the
subsequent forty years' journey in the wilderness in search of the Promised
Land ( 1 6:3ff.; 1 7: 1 -7; 32) , continuing into the book of Numbers. More
than once (e.g., Exod. 32:7- 1 4) the continuation of the special relationship
which God had established with Israel is endangered. Only when Moses
pleads with God for the people's forgiveness is its continuance promised
- but not without due punishment (32:3 1 -35).
It is at Sinai, "the mountain of the Lord,'' that Israel encounters God
and that he binds them with a covenant (Exod. 24) and they promise to
obey his commands. These are expressed in the Ten Commandments
(Exod. 20: 1 - 1 7) which Moses brought down from the mountain, and in
a longer series of laws which God conveys to the people through Moses
(20:22-23: 1 9) . Israel remains at the mountain for a long time, and many
further laws are added: Exod. 25-40, the whole of the book of Leviticus,
and parts of the book of Numbers. Finally (Num. 1 0 :33) , the people of
Israel set out from the mountain. The remainder of Numbers describes
their further journeys and attempts, through many vicissitudes, still
frustrated by further disobedience, to reach the Promised Land, having
been condemned to remain in the wilderness for a whole generation -
forty years (Num. 1 4:26-35). Even Moses is not permitted to enter the
Promised Land on account of the people's sins (Num. 20: 12; 27: 1 3-14;
Deur. 1 :37, etc.) .
The book of Deuteronomy consists almost entirely of words spoken
by Moses (the longest of his speeches runs from 5 : 1 to 26: 1 9 and includes
an entire code of laws, chs. 1 2-26) . The scene is the plain of Moab on the
east side of the river Jordan; Israel is poised to cross the river and to take
possession at last of the Promised Land on the other side. Unlike the other
books of the Pentateuch, Deuteronomy begins with a recapitulation of
past history. In his first speech ( 1 : 1-4:40), Moses reminds the people of
the events that have taken place since their departure from Horeb (clearly
identica1 with Sinai, although not called that here). The dominant notes
of these chapters, as also of other parts of the book, are their homiletic
character and an insistence on God's care for Israel in overcoming the

6
What Is the Pentateuch?

hostility of the nations on the east side of the Jordan and the conquest of
territory there - the first of Israel's territorial conquests. Another promi­
nent themf is the need to beware of the anger of a God who, though a
loving Gld and the giver of all good gifts, is also a terrible God, to be
feared.
A second speech (5 : 1-26: 1 9) is also hortatory. The first part (5:1-
1 1:32) includes a reminder of the promulgation of the Ten Command­
ments at Horeb/Sinai (5 :6-2 1), recited in full for a second time. Here also
is the famous "Shema'': "Hear, 0 Israel: The Lord your God, the Lord is
one [or "the Lord alone'']. You shall love the Lord your God with all your
heart, and with 'all your soul, and with all your might" (6:4-5), later to be
treasured by the Jews as expressing the essence of their faith. The Shema
speaks of God's great gifts to Israel (including the Law given on
Horeb/Sinai) , but also of lsrael's history of sin and God's forbearance, and
Israel's unworthiness and inability to achieve anything for themselves.
Prosperity is promised if they are faithful and obedient, but disobedience
will bring on them curses rather than blessings.
The laws which follow (chs. 12-26) are referred to in 29: 1 as "the
words of the covenant that the Lord commanded Mosest o make with the
Israelites in the land of Moab, in addition to the covenant that he had
made with them at Horeb.'' This second covenant was then executed (ch.
29), and its "words" - that is, the preceding code of laws - were written
by Moses in a ',' book," the book of the law (3 1 :24), which was to be a
"witness" against the people if they disobeyed (vv. 26-29) . The laws them­
selves are in part a repetition of the laws promulgated at Sinai according
to Exod. 20:22-23: 1 9, but many of them are expressed somewhat differ­
ently and in a somewhat different spirit, while others do not appear in
Exodus at all. They begin with an important new requirement that Israel
when it settles in Canaan shall no longer offer its sacrifices in a variety of
places but only at an (unidentified) "place that the Lord will choose" (Deut�
1 2 : 1 3-14).
Deuteronomy ends with a number of shorter pericopes. There is a
more detailed warning about the respective effects of future obedience and
disobedience in terms of curses and blessings (chs. 27-28), including a
warning that disobedience will lead to . military defeat and captivity to a
foreign nation and to exile from the land (28 :25-44). Then follow the
account of the making of the second covenant (ch. 29) , Moses' farewell
speech (ch. 3 1 ) , the Song of Moses (ch. 32) , and Moses' blessing of the
tribes of Israel (ch. 33). In 32:48:-52 Moses receives instructions concerning

7
INTRODUCTION TO THE PENTATEUCH

the place where he is to die; he will be able to see from the summit of
Mount Nebo the land which God is to give to Israel, but is again told
that he will not enter it himself. Deut. 34: 1 -8 records Moses' death. It is
stated in 34:9 that Joshua, who had already been commissioned by God
as Moses' successor (3 1 :23) , had also been consecrated by Moses for his
task by the imposition of Moses' hands, and that the Israelites obeyed him,
as God had commanded. Finally in 34: 1 0- 1 2 comes the assessment of
Moses previously mentioned (p. 2) .
In these final chapters, although the appointment of Joshua to lead
the people into the Promised Land clearly looks to the future, there is an
unmistakable sense of a definite ending, not found in any other book of
the Pentateuch. The era of Moses, which began at the beginning of the
book of Exodus, is at an end. It is made clear that this had been the decisive
era for Israel, in which Israel had been rescued from bondage in Egypt
and, at Sinai/Horeb, had received from God at the hands of Moses its
definitive character as God's chosen people, furnished with everything
necessary for its future life.

The Pentateuch as "Story''

The above outline is an attempt - necessarily somewhat selective - to


present the main features of the "story line" of the Pentateuch as it would
appear to someone reading it for the first time as a "book," unaware of
critical notions of multiple authorship, sources, gradual growth over a long
period of time, and the like. It is most important to read it in this way
before beginning to study those other ways of reading it, in order to be
able to answer the question which is the title of this chapter, "What is the
Pentateuch?" as distinct from the question, "How did the Pentateuch
become what it is?" This second question will be considered in the fol­
lowing chapters of this book. That there are minor inconsistencies in the
story would no doubt occur to a perceptive reader. But unless he or she
is a committed deconstructionist, these will not be enough to disturb an
appreciation of the Pentateuch as "story." That is the way in which it has
been read for many centuries by the great majority of both Jews and
Christians.
The Pentateuch presents itself as a history. That is, it is a narrative
or "story" in which the events which it narrates, from the creation of the
world to the death of Moses, are arranged in chronological sequence. Even

8
What Is the Pentateuch?

the recapitulation of past events in the early chapters of Deuteronomy is


assigned its place in the narrative as part of a speech of Moses spoken at
a later ti � ej it is not represented as the composition of the narrator of the
Pentateuc;h'.. Again, although about one-half
. of the text of the Pentateuch
consists of laws rather than being a narrative of events, these are stated to
have been promulgated by God or through Moses at particular moments,
and so are formally part of the ongoing story. The same applies to the
poems which occur from time to time in the text. They are stated to have
been recited or sung by particular persons at particular momet?-ts in the
course of the story. At every point the reader is told what happ.ened and
when. The impression given to the reader is thus of a grand sweep of
history; It would be wrong to think of the Pentateuch as just a miscel­
laneous collection of stories, laws, and poems lacking an overall theme
and purpose.
The theme of the Pentateuch is not difficult to discern. It is the story
of the birth and adolescence of a nation. Such awork is not in itself unique
in the literatures of the ancient world, particularly in the world of late
antiquity, nor is the fact that it begins with an account of the origin of
the world itself unique. Its notion of a benevolent deity presiding over the
nation's destiny is not by any means unusual. However, in some crucial
respects the Pentateuch is unique as a national history. This is true of its
portrayal of the divine. The Pentateuch as it presents itself to the reader
knows of one, supreme God. Although this God is known, especially in
Genesis, by a variety of names - the Lord (Yahweh), the Lord God, God
Almighty (El Shadday) , God Most High (El Elohim), or simply .as God
(El or Elohim) - for the Pentateuch these are all names of the same God
who is the creator of the world and supreme over it. There are no rival
gods.
But equally unique is the portrayal of God's chosen people, Israel.
The Pentateuch is no boastful, triumphalist work lauding Israel over all
other nations for its virtues, its merits, or its valor. On the contrary, Israel
is represented as constantly rebellious, ungrateful, disobedient to God. Yet
it is also portrayed as completely without any ability to achieve anything
by its own efforts, totally dependent on God for any successes it may have.
The story of Israel in the Pentateuch is not one of which the nation can
be proud. If it is to become a great nation (Gen. 1 2 :2), this will be through
no merit of its own. In contrast with other national histories, the Penta­
teuch does not present to its readers the picture of a glorious past.
In fact, it is God rather than Israel who has the leading role in the

9
INTRODUCTION TO THE PENTATEUCH

Pentateuch. From the very beginning (Gen. 1 - 11) God's intentions and
actions with regard to his human creatures are represented as just and
severe, yet merciful. The first man and woman are expelled from the garden
for disobedience, but their lives are preserved. Even the first murderer,
though banished from the presence of God, is given a chance to live out
his life. Corrupt humanity is drowned, yet a new beginning is made with
Noah. The presumption of the builders of the Tower of Babel is punished
by their being divided and scattered throughout the earth and deprived
of their common language, but the human race is nevertheless permitted
to survive and multiply.
With God's choice and commissioning of Abraham, the theme of
divine gift, human shortcoming, and divine forbearance - seen first as
concentrated upon the fortunes of a single family and then on the children
of Israel grown into a nation - becomes a constant and consistent feature
of the Pentateuchal story. The threefold promises of blessing, numerous
progeny, and possession of the land dominate much of the story. Although
the last of these remains largely unfulfilled, the other two promises still
hold, despite massive and frequent disobedience both before and after the
encounter with God at Sinai. In an important sense, therefore, the Pen­
tateuch as a whole teaches a moral and religious lesson which the reader is
intended to heed and to take at the same time as both an encouragement
to his own generation to trust God's gracious purpose and a warning to
live a life of obedience in the future. This lesson is taught most unmis­
takably in the final book, Deuteronomy; but it is implicit in all the previous
books. In some parts of the story this religious and theological note is
more evident than in others; the story of Gen. 2-3, for example, has always
been recognized as a profound parable of the human condition.
Finally, it is important that the Pentateuch should be appreciated as an
outstanding literary achievement. Some parts ofit, of course, are justly famous
as "Bible stories." Others, however, hardly qualify by themselves as literary
masterpieces. But taken as a whole, the Pentateuch is a kind of epic on a
massive scale. The unknown person - called by modern scholars the "final
redactor" - who was responsible for its final shape has taken a mass of
material, some of it rather unpromising, and forged it into a compelling
"story." Like Homer and other ancient writers who also told heroic tales of
the remote past, usually in poetry, the redactor has introduced us, in prose
rather than poetry, into a "world" unfamiliar to us but which has its own
logic and its own rules. How this was achieved, and what were the materials
which he used will be the subject of the next chapter of this book.

10
What Is the Pentateuch?

· For Further Reading

Alter, Robert, and Kermode, Frank, eds. The Literary Guide to the Bible.
Cambridge, Mass. : Harvard University Press and London: Collins,
1 987, 36-1 0 1 .
Blenkinsopp, Joseph. The Pentateuch: An Introduction to the First Five Books
ofthe Bible. New York: Doubleday and London: SCM, 1 992, ch. 2.
Clines, David J. A. The Theme of the Pentateuch JSOT Supplement 1 0 .
..

Sheffield: ]SOT Press, 1 978.

11
CHAPTER 2

Who Wrote It?


Problems of Composition

T
I
HAS LONG BEEN recognized that the traditional view - not stated in
the Pentateuch itself, but already assumed elsewhere in the Old Testa­
ment - that Moses was the author of the Pentateuch cannot be correct.
This conclusion was not derived from the fact that Moses' death and burial
are recorded in the Pentateuch itself (Deut. 34:5-8) . It would after all be
possible to regard this final chapter, which also refers to the appointment
of Joshua as Moses' successor and concludes with a general assessment of
Moses' achievements, as simply a postscript added to a work (Genesis to
Deuteronomy) which was, with this single exception, the work of Moses
himself. This view was actually held in later Judaism. A passage in the
Jewish Talmud attributes the verses in question to Joshua. The rejection
of Mosaic authorship rests, as will be seen below, on other criteria.
It should perhaps be stressed at this point that, despite the vast
amount of scholarly work which has been published - especially during
the past century - concerning the authorship, date, and history of com­
position of the Pentateuch, these are basically side issues. The real interest
for readers of the Bible does not lie here. If it did, the present generation
of readers would experience only frustration. For although it may be true
that recent scholars have succeeded in exposing many of the errors of earlier
critics, it must be admitted that as far as assured results are concerned we
are no nearer to certainty than when critical study of the Pentateuch began.
There s at the present moment no consensus whatever about when, why,
·

how, and through whom the Pentateuch reached its present form, and
opinions about the dates of composition of its various parts differ by more

12
Who Wrote It? Problems of Composition

than five hundred years. This chapter, therefore, may be regarded by many
as an irrelevancy as regards a serious understanding of the meaning and
purpose of the Pentateuch, though the questions with which it is concerned
r�main - and are likely to remain - a major item on the agenda of
academic Old Testament study. The important question is not one of the
sources available to the compiler but what the Pentateuch was intended
to mean in its present form.
Doubts about - and even denials of - Mosaic authorship were
voiced sporadically over the centuries by individual writers both Jewish
and Christian and (later) both Catholic and Protestant, some of whom
proposed specific dates and events in Israel's history for the P�ntateuch's
composition or for parts of it. One of the most important of such scholars
was the twelfth-century Jewish scholar Ibn Ezra, who (though in an allusive
way in order to avoid hostile criticism) pointed to a number of passages
in Genesis and Deuteronomy which could only be understood as written
from a standpoint much later than that of Moses. Here was an early
example of the historical-critical method which was to play a crucial part
in later, especially nineteenth-century, critical discussion. The seventeenth­
century French Catholic priest Richard Simon may also be mentioned
here. He regarded the Pentateuch as a compilation of numerous written
sources of different dates. But no comprehensive investigation of the
composition of the Pentateuch as a whole had as yet been undertaken.
A significant pioneer in this respect was another Frenchman, the
physician Jean Astruc, who in 1 753 published a study of Genesis, which
he claimed had been constructed out of earlier written "memoirs." Astruc
did not, however, reject Mosaic authorship, or at least editorship. Rather,
his purpose was to defend it: it was Moses himself who had made the
compilation. A feature of particular importance in Astruc's analysis was
the discovery of a difference of terminology in different passages. In
particular, Astruc distinguished two documents which differ in the ways
in which they refer to God: one calls him by the name Jehovah ("the Lord"
in most English versions) , the other by the word Elohim ("God"). This
early source (or documentary) theory marked the beginning of what was
to be the dominant method of Pentateuchal criticism in the nineteenth
and early twentieth centuries. In 1953 the bicentenary of the publication
of Astruc's book was celebrated in recognition of his importance as the
real initiator of modern Pentateuchal criticism. (See Roland de Vaux's paper
read at the international Old Testament congress held at Copenhagen and
subsequently published in vrs 1 [ 1 953] : 1 82- 1 98.)

.13
INTRODUCTION TO THE PENTATEUCH

Johann Gottfried Eichhorn in his introduction to the Old Testament


( 1 780- 1 783) took up and developed Astruc's approach. He pointed to
further stylistic distinctions between the two sources (or "strands") in
Genesis, the "Yahwistic" and the "Elohistic" (subsequently dubbed "J, "
from the German spelling Jahve, and "E," respectively) . A further refine­
ment to the theory was made by Karl David Ilgen, who in 1 798 anticipated
later critical study by arguing that there are in fact not one but two distinct
Elohistic sources, so making three sources in all.
Meanwhile two quite different approaches to the problem began to
be pursued, leading to the development of the so-called fragment and
supplement theories. In order to understand the rationale of these theories
it may be useful to consider the general question of the main ways in
which a literary work which is composite in character - that is, one which
was not created in a single act of composition, but developed gradually
over a period of years - may have reached its final form. First, the work
may be the result of the combination of two or more older complete
written works (the "documentary" method) . Second, it may be the result
of the combination of a number of isolated shorter units either written or
oral (the "fragment" method) . Third, it may have come about by the
gradual expansion or supplementation of an original single work by addi­
tional material with the intention of modifying its character or extending
its scope (the "supplement" method) . In the case of a lengthy and complex
work such as the Pentateuch, these three approaches are not necessarily
mutually exclusive. One method may have been employed by a redactor
in one part, another by a different redactor in another. For example, the
formation of the legal parts of the Pentateuch may have occurred in a way
quite different from the way in which narratives were combined.
The Fragment Hypothesis (dealing now not only with Genesis but
with the whole Pentateuch) was espoused by another Catholic priest,
Alexander Geddes ( 1 792 and 1 800), and by Johann Severin Yater ( 1 802-
1 805) . Yater postulated a series of unrelated fragmentary sources which
continued to be accumulated over the years, resulting eventually in the
completed Pentateuch at the time of the Babylonian Exile in the sixth
century B.C. A weakness of this theory if applied wholesale to the Penta­
teuch is that if fails to identify a credible motive for the process. W. M. L.
de Wette ( 1 806, 1807) supported the Fragment Hypothesis, but only in
part. Ha found it necessary to combine it with the Documentary Hy­
pothesis. But de Wette also put forward a quite novel theory which was
to become one of the principal bases of all subsequent Pentateuchal criti-

14
Who Wrote It? Problems of Composition

cism until recent years. Comparing the details of King Josiah of Judah's
late seventh-century reform of the Jerusalem cult as described in 2 Kgs.
22-23 (and in another version in 2 Chr. 34-35) with the book of Deuter­
onomy, he argued that this reform was based on the laws of Deuteronomy,
which must therefore be identical with the "book of the law" found in
the temple and used to initiate the reform. On the assumption that that
"book" was a recent composition, this thesis - which was widely, and
eventually almost unanimously, accepted - was a breakthrough in Pen­
tateuchal study, in that now for the first time a substantial part of the
Pentateuch could be precisely dated.
This dating of Deuteronomy was to become the cornerstone of the
"new documentary hypothesis" generally associated with the name ofJulius
Wellhausen, but which owed much to his predecessors, especially Eduard
Reuss, Hermann Hupfeld, Abraham Kuenen, and Karl Heinrich Gra£ It
now became possible to attempt the dating of the other Pentateuchal
sources J, E, and the "second Elohist," later known as P. Since they reflected
earlier stages of the Israelite religion, J, E, and P were all at first thought
to have preceded D (Deuteronomy) in point of time. Doubts, however,
came to be expressed about the early dating of P, and it was Graf ( 1 866)
who finally established that the laws of P (in Leviticus and the latter part
of Exodus) belong t,o the latest strand in the Pentateuch, and that P is
consequently its latest source, later than D. The four sources had been
combined by a series of redactors, referred to as RJE, RD and RP, whose
participation in the process of composition, however, is never very clearly
defined.
The Supplement Hypothesis, according to which the Pentateuch
consists of a single basic source subsequently supplemented by later writ­
ing, had a number of advocates during the nineteenth century, including
Friedrich Bleek, de Wette (toward the end of his life), and Franz Delitzsch.
Some of its advocates, however, later abandoned this hypothesis. And
although it has recently been revived in a somewhat different form, the
future clearly lay with the Documentary Hypothesis.
The Documentary Hypothesis in its new form, finally presented in
masterly fashion byWellhausen in a series of publications from 1 876 to
1 884, dominated Pentateuchal study for almost a century. It was ad­
mittedly, and not surprisingly, vigorously opposed by conservative scholars
such as Ernst Wilhelm Hengstenberg in Germany (in the earlier stages of
the hypothesis) and Edward B. Pusey in England, who were fundamentally
opposed to all biblical criticism, and by Catholic scholars. But in academic

15
INTRODUCTION TO THE PENTATEUCH

circles in general, the Documentary Hypothesis came to be regarded as


unassailable. Until very recent times it has been taken for granted by
biblical scholars, who have treated its hypothetical sources or "documents"
as though they actually existed as independent books providing reliable
information about particular definable stages in the history of the religion
of Israel.
As has already been suggested, Wellhausen was a historian, concerned
not only with a literary question but with the history of Israel and more
particularly with the history of its religion. He saw the sources or "strata''
(Schichten} of the Pentateuch as reflecting different stages of this historical
development, especially of the development of the cult. Although he
recognized E as a distinct source (though one which was often difficult to
distinguish from J - the criterion of the two divine names is only appli­
cable to Genesis and the first few chapters of Exodus; after that this
distinction breaks down) , Wellhausen worked mainly with three strata,
the "Jehovist" (a combination ofJ and E),, D, and P. He argued that both
JE and the stories in Judges, Samuel, and Kings (before their later editing)
present a picture of Israel and its worship which is clearly earlier than that
presented by D and P, and indeed earlier than the time of the eighth­
century prophets. He further contended that the influence of those proph­
ets is discernible in D, however, while the elaborate system of laws in P
(which is presupposed in the books of Chronicles) is postexilic and rep­
resents a degeneration from the spirit of the great prophets. Also, JE
belongs to the early part of the period ofthe monarchy, before the eighth
century. (It is precisely this notion, that Israel at such aii early stage of
cultural development could have been the first people ever to conceive and
produce such extensive "historical" works, that is one of the main draw­
backs to the hypothesis.)
It is unnecessary here to pursue the later history of the new Docu­
mentary Hypothesis in detail. One of its notable fo�tures was the further
multiplication of sources. Attempts were made to distinguish E more
clearly from J. J itself was, on the basis of supposed linguistic and theo­
logical distinctions, separated into two sources, J1 and J2, and even into
three (Otto Eissfeldt proposed a "lay source" L, Robert H. Pfeiffer an
Edomite or "southern" source S, Georg Fohrer a "nomadiC" source N),
and E also into E1 and E2. A distinct "Holiness Code" (Lev. 1 7-26) was
"discoY �red" embedded in P, and P itself was fragmented into several
successive strands. This process of dissection was to a large extent based
on an unrealistic notion that authors are always completely consistent, so

16
Who Wrote It? Problems of Composition

that differences in statements of fact, point of view, or language infallibly


indicate the work of different authors. In some cases too the supposed
inconsistencies are minute, too inconsiderable to be significant.
This kind of criticism, though hardly recognized at the time, repre­
sented a partial disintegration of the Documentary Hypothesis and a
tendency toward a revival of the Fragment Hypothesis. It also served to
raise questions, hardly before considered, about the nature of the supposed
sources: were they unitary works, or were they themselves collections of
heterogeneous material?
It was a younger contemporary of Wellhausen, Hermann Gunkel,
who addressed himself to questions of this kind in relation to the book
of Genesis. In his commentary on that book (first edition, 1 90 1 ), he
attempted to go back behind J and E to discover how its various stories
had originated and taken shape in an e�lier, preliterary period. Gunkel
made use of the investigation of European folklore that had begun in the
late eighteenth century and had been further developed in the early
nineteenth century by Jacob and Wilhelm Grimm; he especially employed
the work of his own contemporary, the Danish folklorist Axel Olrik on
the early Icelandic Sagen or traditional, orally transmitted, tales of the
preliterary past. Following a set of "epic laws" enunciated by Olrik, Gunkel
classified the stories of Genesis (which he believed to have originated
independently of one another) according to their formal characteristics
with a view to uncovering the various situations in which they had arisen.
In this, he was the pioneer of the method of "form criticism." In the third
edition of his commentary Gunkel stated clearly that "Genesis is a collec­
tion of Sagen," which had originated at an early stage in Israel's history
and had been handed down orally over a long period of time before being
committed to writing.
Gunkel was thus not only the first "form critic"; he was also the first
"tradition critic." He did not abandon the Documentary Hypothesis.
Rather, Gunkel attempted to trace the gradual combination of the in­
dividual stories - especially those concerning the patriarchs Abraham and
Jacob - into "circles of Sagen' and their eventual collection in written
form by the authors of J and E. This "history of traditions" method was
subsequently extended and applied to the whole narrative tradition of the
Pentateuch by other scholars, notably Hugo Gressmann.
Gerhard von Rad ( 1 938) built upon Gunkel's work on the oral
tradition and used it as a foundation for a comprehensive theory of the
development of the Pentateuchal traditions. He held that the main themes

17
INTRODUCTION TO THE PENTATEUCH

of those traditions - the entry into Egypt of Israel's ancestors, their op­
pression in Egypt, their deliverance in the Exodus, their guidance to their
final destination, and the gift of the land of Canaan - were brought
together to become the "common heritage" of all the tribes at the celebra­
tion of the Feast of Weeks held at Gilgal in the time of the Judges. The
evidence for this he found in a number of biblical passages, most clearly
in Deut. 26:5-9, which he called a "Little Credo," recited at the annual
offering of the firstfruits and containing all these elements in a liturgical
prayer. The only main element of tradition lacking was the giving of the
Law at Sinai; this was an entirely separate tradition celebrated at the Feast
of Tabernacles at Shechem. The two sets of tradition were expanded and
continued to develop separately until they were combined and committed
to writing by the Yahwist (J), a designation which for von Rad seems to
represent the entire pre-Deuteronomistic narrative tradition, including the
older parts of Gen. 1 -1 1, which the Yahwist was himself responsible for
incorporating.
Von Rad's dating of the Yahwist is even earlier than that proposed
by Wellhausen: the reign of Solomon (tenth century) , the age of Israel's
imperial expansion. This dating was connected with von Rad's view that
at that time Israel - or its ruling class - experienced a strong cultural
impulse, the consequence of international contacts especially with the
superior culture of Egypt, resulting in an unusually swift development of
literary as well .as other . skills (the so-called Solomonic enlightenment).
The Yahwist's history and other narrative works were a product of this
cultural revolution; and in an age of political confidence this history not
unnaturally took the form of an account of the origins of the nation, which
was now for the first time enabled to take pride in its past.
The influence of von Rad's work was at first immense. However,
both his notion of a "Solomonic enlightenment" and his theory of the
"Little Credo" have now lost much of their credibility. The former is now
regarded as greatly exaggerated, partly because historical writing com­
parable with von Rad's Yahwist's history was not an art practiced in Egypt,
while the Credo in Deut. 26:5-9 (which is part of the Deuteronomic laws)
and similar passages are now believed to have been composed at a much
later date, being late summaries of the earlier traditions rather than the
starting point of their development.
l'.1 :min Noth's A History ofPentateuchal Traditions (1948) is in many
respects a continuatio.n of the work of von Rad. It is a much larger work
than that of von Rad, and ,attempts to describe the entire process of the

18
c Who Wrote It? Problems of Composition

compilation of the Pentateuch (though in fact Noth concerned himself


here only with the first four books - the "Tetrateuch," excluding Deuter­
onomy) "from beginning to end" in · all its details. Each of the "themes"
of von Rad's "Little Credo" was originally a separate tradition, the property
of a particular tribe or group of tribes preserving its own recollection of
its past experiences. These were pooled when the tribes united to form the
tribal league or "amphictyony'' which constituted the earliest Israel, and
became the common tradition of the whole nation. This single narrative,
which may or may not have been a written work, Noth called G (for
Grund/age, "foundation'') . The later J and E, which were written works,
were separate histories which drew on the common material of G but
added new material of their own. Eventually they were combined, E being
preserved only in a partial form as a series of "enrichments" of the hence­
forth dominant J. Another prominent feature of Noth's work was his
skepticism about the historicity of the stories about Moses. He did not
doubt the actual existence of Moses, but believed that hardly any authentic
material had been preserved about him.
Noth's reconstruction of the formation of the Pentateuch seemed at
first very impressive. Comparatively few of his theses, however, have sur­
vived unscathed. His postulation of G has met with little acceptance; and
his notion of an "amphictyony'' is now completely discredited as is also,
in many circles, his radical skepticism about the figure of Moses. Many
of his particular arguments about the combination of specific traditions
have been dismissed as unduly speculative and lacking in proof. Most of
all, however, Noth's views about oral traditions have been severely ques­
tioned.
Thus Gunkel, von Rad, and Noth all retained the Documentary
Hypothesis in their own ways. It was on their views about the gradual
development of orally transmitted traditions before their committal to
writing that they principally differed from Wellhausen. Subsequent study
by anthropologists of the traditions of modern preliterate peoples has,
however, revealed that the notion that such traditions can be orally pre­
served over many generations is a false one. After two or three generations
these traditions tend to become so seriously distorted as to be unrecogniz'­
able or to disappear altogether. This fact has led to a growing tendency to
reject the idea that the stories concerning such figures as Abraham, Jacob,
and Moses can be much older than the written sources. Together with
this, there is a growing tendency to doubt the antiquity of the written
sources themselves, and to suggest that very little of the Pentateuch is older

19
INTRODUCTION TO THE PENTATEUCH

then the sixth century B.C. or even later. Nevertheless, the Documentary
Hypothesis continues to be accepted by some scholars (see below) .
The Swedish scholar Ivan Engnell, whose early death in 1 964 left
his work uncompleted and many of his ideas unpublished, was a thorough­
going advocate of the tradition-historical method. He completely rejected
the existence of the sources postulated by the Documentary Hypothesis.
Together with other Swedish scholars, he held a strong belief in the
reliability of the transmission of oral traditions, and, although convinced
that the Pentateuch is to all intents and purposes a postexilic work, he
held that its narratives had been transmitted orally over a period of many
centuries. Engnell thus succeeded in combining a theory of late written
composition with a belief in the antiquity of much of the material which
it contains. This ancient material was finally edited by P in what Engnell
called the "P work,'' distinguishing this from the "D work'' of Deuter­
onomy and the following "historical" books. Engnell's theories have re­
ceived comparatively little subsequent attention, no doubt partly because
he did not live to expound them fully.
Ever since the rediscovery of the Egyptian and Mesopotamian civi­
lizations in the nineteenth century, attempts have been made to relate the
Pentateuchal data to this international background - a matter of obvious
importance for determining the antiquity ofthe Pentateuchal narratives
and laws. Here it is possible only to refer to two of these investigations.
A matter of perennial debate has been the dating of what came to be
known as the "patriarchal age,'' in other words, the attempt to discover what
period of ancient Near Eastern history best fits the stories of Abraham and
his family. Several possibilities in the second millennium B.C. were discussed
by historians, archaeologists, and biblical scholars. One theory which for a
time attracted particular attention associated the patriarchal stories with data
provided by a collection of family archives from the Mesopotamian city of
Nuzi, dating from the fifteenth and fourteenth centuries B.C. and mainly
published from the 1 920s on. Analogies were drawn ·between certain prac­
tices of the patriarchs not attested in the Old Testament outside Genesis and
certain family and legal customs at Nuzi which appeared to be peculiar to
the Hurrian-speaking peoples of these texts. Subsequent improved knowl­
edge both of Hurrian and of other Mesopotamian legal and family customs
has, however, revealed that these practices have either been misunderstood
or VI tre in fact not confined either to Nuzi or to the second millennium, but
are also attested more generally and in the first millennium as well.
Thomas L. Thompson in The Historicity ofthe Patriarchal Narratives ( 1 974),

20
Who Wrote It? Problems of Composition

whose conclusions have been confirmed by other scholars, showed that not
only the Nuzi hypothesis but also other attempts to locate the patriarchal age
in history were without substance and that there is in fact no way of
demonstrating the antiquity of the patriarchal stories of Genesis. John Van
Seters (Abraham in History and Tradition, 1 975) and others have argued with
some plausibility that these reflect rather the concerns of the middle of the
first millennium or even later. The search for the "historical Moses" has been
no more successful.
A second theory based on Near Eastern parallels was put forward in
1 954 by George E. Mendenhall. This was concerned not with Genesis but
with the idea of a covenant (Hebrew berith) between Yahweh and Israel
found especially in the Exodus accounts of the covenant at Sinai (which were
generally supposed to be early) and with the Pentateuchal laws which were
considered as constituting the conditions of obedience then imposed by
Yahweh on Israel. Mendenhall linked this notion of a divinely imposed
covenant both in contents and form with extant Hittite treaties imposed by
human suzerains on their vassals, dating from the fourteenth and thirteenth
centuries, suggesting that it was on a generally current familiarity with that
kind of treaty that the concept of a divine covenant with Israel was based.
This theory was more fully developed by Klaus Baltzer ( 1 97 1 ) .
However, like the Nuzi hypothesis, this theory, although at first
widely accepted, was destined to have only a temporary success. It was
pointed out on the basis of later discoveries . that the international treaty
form continued to be in use during the first millennium, many centuries
later than the Hittite treaties, and that if Israel did in fact borrow the
notion from elsewhere this borrowing could have taken place as late as the
seventh century B.C. In 1 969 Lothar Perlitt argued that not only the treaty
form but the very idea of a covenant between Yahweh and Israel was an
invention of the Deuteronomists and thus could not be earlier than the
seventh century. This late date for the covenant idea continues to be
disputed; but the theory of a direct analogy with Hittite treaties made in
the time of Moses is no longer regarded as plausible.
The tendency to regard the Pentateuch as essentially a late composi­
tion was strongly reinforced by Van Seters, who retained the term Yahwist
0) but, while leaving the question of · date theoretically open, offered
evidence which suggested that it best fits the late monarchic or exilic
period. Of the stories of Abraham and his family he wrote: "There is
nothing in this presentation of the 'nomadic' patriarchs which is inappro­
priate to the portrayal of pastoral life in the period of the late Judean

21
INTRODUCTION TO THE PENTATEUCH

monarchy or exilic periods, but there is much that speaks against the choice
of any earlier period" ( p. 38). Thus, while he retained the notion of written
"sources," Van Seters abandoned the Wellhausian scheme of successive
"histories" stretching over a period of several centuries. Hans Heinrich
Schmid, writing about " T he So-Called Yahwist" ( 1 976), expressed some­
what similar views, though for him the Pentateuch is a work closely
associated with the Deuteronomists, who were also responsible for the
"Deuteronomistic History" Qoshua, Judges, Samuel, Kings) , so recounting
the entire " history'' from the creation of the world to the fall of the Judean
monarchy in the early sixth century.
Rolf Rendtorff in The Problem of the Process of Transmission in the
Pentateuch ( 1 977; English translation, 1 990) discussed traditional source
criticism and tradition history as methods of approaching the question of
composition. He saw the two as totally incompatible, arguing that it was
a mistaken loyalty to source criticism which had prevented such scholars
as von Rad and Noth from carrying their tradition-historical work to its
proper conclusion. Like Schmid, Rendtorff regarded the Pentateuch as
basically a Deuteronomistic composition. While virtually all earlier
scholars had left room for P as an independent continuous source, Rend­
torff regarded the "priestly" contribution as limited in scope and lacking
in homogeneity. (More will be said about D and P later in this chapter. )
Rendtorff's method was to begin by considering the smallest elements of
tradition and, abandoning the notion of continuous sources, to endeavor
to show how these had been built up through stages first into intermediate
complexes and · finally into larger blocks of material each with its own
theme. T hese larger blocks ( e.g., the Exodus story) had remained entirely
independent of one another until they were combined at a late stage to
form a comprehensive "history."
Rendtorff's pupil Erhard Blum in two works on the composition of
the patriarchal stories ( 1 984) and of the Pentateuch as a whole (1990)
developed Rendtorff's work further, tracing the process of composition in
greater detail than Rendtorff had thought possible. He also assigned a
greater role to the work of individual authors of complexes of intermediate
size such as the Jacob-Laban stories. Blum was particularly skeptical about
the possibility that the traditions originated at a time before the period of
the monarchy. In his second book he saw the priestly material as an attempt
to t:orrect certain elements of the Deuteronomistic theology; the Penta­
teuch is then a postexilic compromise beriveen two schools of thought,
made under the impulse of the Persian demand for a "Jewish law."

22
Who Wrote It? Problems of Composition

This chapter has so far been concerned mainly with the material
which has for many years been known as the work of the Yahwist CT) and
which has been generally at the center of the d�bate. But what of the
re.�naining material, known to the Documentary Hypothesis under the
symbols E, D, and P?
As has already been seen, the Elohist has always been a somewhat
shadowy figure, in that eve� those most committ�d to the "four-document
theory'' foup.d it difficult to distinguish E from J, especially outside Gene­
sis. The strongest evidence for a . parallel source to J was probably the
existence of what were taken to be duplicate accounts of identical events
found in different chapters (e.g., the expulsion of Hagar and Ishmael, Gen.
1 6 and 2 1 ) and passages in which it appeared that two accounts had been
woven together, causing internal duplication of details (e.g. , the story of
the Flood, Gen. 6-9, and parts of the story of Joseph, notably Gen. 37) .
Differences of point of view (theology) and to .some extent of language
were taken to indicate the existence of a separate continuous source (E)
rather than of a number of fragmentary additions, though it wru; admitted
that E had been only partly preserved, having been inserted into the now
more continuous J narrative.
In 1 933 Paul Volz and Wilhelm Rudolph made a detailed study of
the supposed E source in Genesis, concluding that there was no evidence
for its existence: the passages .in question could be adequately accounted
for i,n a variety of other ways. In 1 938 Rudolph extended this investigation
to include . the remainder of the Pentateuch (and also. the book of Joshua,
to which also the Documentary Hypothesis was commonly applied) . This
theory caused something of a scandal in Old Testament scholarship, and
gained little acceptance at the time. However, despite attempts such . as
that of Hans Walter Wolff ( 1 972) to rehabilitate the Elohist as a distinct
source with its own theological point of view, it has ceased to be a
significant element in Pentateuchal research. Modern approaches . to Old
Testament narratives based on contemporary literary theory, such as that
of Robert Alter's The Art of Biblical Narrati.ve ( 1 9 8 1 ) , have shown that
duplications in narratives - which are by no means absent from modern
fiction! - may be explained as a deliberate feature of literary technique
erriployed by single authors for purposes of emphasis and to give artistic
smi�·ture to their works.
The book of Deuteronomy raises distinct questions of its own, which
will be discussed in chapter 6 below. Although the dating of the Code of
Deuteronomy (Deut. 1 2-26) tg_Jhe- time of Josiah in the seventh century
_
_

23
INTRODUCTION TO THE PENTATEUCH

was a cornerstone on which the chronological succession of all four sources


of the Pentateuch according to the Documentary Hypothesis was based,
D was believed until very recently to be a source quite distinct from the
other three, of which virtually no traces were to be found in the other four
books. But this view has now been challenged. Both Schmid ( 1 976) and
Rendtorff ( 1 977, 1 9 83) have pointed to evidence of Deuteronomic the­
ology in Genesis-Numbers. Rendtorff spoke of a redaction or reworking
of the whole Pentateuch which "in its ideas and language is closely related
to Deuteronomy" ( The Problem ofthe Process of Transmission in the Penta­
teuch, 99) . He also argued that this was the final redaction, so rejecting
the general view that it is P which is the latest strand in the Pentateuch
(see below) .
Schmid, writing after Rendtorff's 1 975 article on the question, differs
from him in that he retains the concept of a Yahwist but dates this work
(with Van Seters) after the conclusion of the monarchic period. He sees
it as a complex literary and theological work rather than as the work of a
single author, which "belongs in the proximity of the Deuteronomic­
Deuteronomistic . . . tradition-forming and literary work." Schmid, how­
ever, does not define the relationship between this "Yahwist" and this
"Deuteronomist" more closely. He does not deal with the question of P.
Whatever may be the final judgment on the general views of Schmid and
Rendtorff, their views about some kind of "Deuteronomic" influence on
Genesis-Numbers constitute a new development which will certainly need
to be taken into account in future Pentateuchal study.
The bulk of the material attributed to P consists of laws mainly
concerned with the details of sacrificial worship and with the appointment
and fu�ctions of the priesthood. But there are also a number of narratives
associated with it, including the creation story in Gen. 1 and various
incidents connected with the leadership of Moses, scattered throughout
the books from Genesis to Numbers. The view that P is the latest strand
in the Pentateuch dates from the recognition that there are two "Elohistic"
strands in the Pentateuch, first proposed by Graf in 1 866 (see above) . This
view was generally accepted until recently. The main points which re­
mained in dispute were as follows:

1 . The date of P (exilic or postexilic?) .


2 The question whether P is not only the final strand, but also con­
stitutes the final redaction of the Pentateuch: in other words, whether
it was the author(s) of P who created the Pentateuch in its final form,

24
Who Wrote It? Problems of Composition

incorporating all the earlier material, or whether the final act of


composition was the work of a later, final redactor who put together
all four strands to make the completed work.
3. Whether the author(s) of P incorporated earlier, preexilic priestly
lore or were interested only in presenting a new legislation solely
concerned with the practice of their own time - or, if they worked
during the Exile, with what they intended should become actual
practice (compare the anticipatory legislation of Ezek. 40-48).
4. Whether the legislation of P is homogeneous or whether it contains
a series of supplements to an original core reflecting a need for
modification of the original.
5. The relationship between the narrative and the legal parts of P.

The legislation of P, or of the material commonly attributed to P,


will be considered in chapter 7 below. The relative consensus regarding P
as a whole has been challenged in recent years. AB already stated, for
Rendtorff P is not the final strand of the Pentateuch, and indeed is not a
continuous source at all. He has argued, partly but not wholly on the
grounds of style, that P's supposed narratives are not genuine narratives,
but are rather theological interpretations of other material. Further, al­
though P contains a series of brief chronological notices scattered
throughout the Pentateuch, these lack homogeneity of style. They do not,
as has been supposed, constitute a framework for the narrative corpus, but
are merely sporadic. In short, P is not a connected narrative source - a
point made already in 1 933 by Volz in an appendix to his and Rudolph's
work, entitled "P is no narrator"!
The conventional view of an exilic or postexilic date for P has now
also been challenged by two Israeli scholars, Menahem Haran ( 1 978, 1 9 8 1 )
and Avi Hurvitz ( 1 982), and by the American scholar Jacob Milgrom
( 1 9 9 1 ) . These reject the view that P's legislation reflects the cultic practice
of the postexilic temple. Haran connects this legislation with the cultic
reform carried out by Hezekiah. Hurvitz has carried out a thorough
investigation of its language and style, especially comparing it with the
laws of Ezekiel, and has concluded that it precedes Ezekiel rather than
postdates it, as had been generally supposed. Thus P, like the other
"sources" of the Pentateuch, has been thrown back into the melting pot.
The existence of P as a Pentateuchal source is the main aspect of the
Documentary Hypothesis retained by Van Seters. This scholar remains a
"documentary" critic in as far as he still speaks of a Yahwist. But for him

25
INTRODUCTION TO THE PENTATEUCH

the Yahwist is a late writer who is virtually the author of the Pentateuch, .
a historian who out of a mass of both early and contemporary traditions
composed a history of the origins of the nation - to which, however, a
later P made some further contributions. (However, if P is to be dated
earlier [Haran and Hurvitz] , the Pentateuch in its final form may be to
all intents and purposes the. work of a single writer.) Van Seters's main
contribution to the debate is his conception of the Pentateuch as the work
of a historian.
Earlier views of J or JE as preexilic and the work of "the world's first
historian'' suffered from the improbability that Israel alone at such an early
date should have been capable of conceiving and producing a work cover­
ing such an extensive period and manifesting such a highly developed
concept ofhistorical purpose and of causation when the far more culturally
advanced civilizations of Egypt and Mesopotamia were apparently in­
capable of doing so. Van Seters, by moving the "Yahwist" down the
·centuries, brought this achievement within the chronological range of the
earliest Greek historians of the sixth and fifth centuries B . C, notably
Herodotus, who did in fact achieve works of this kind. Without stressing
any possible direct relationship between the Pentateuch and these Greek
historians, Van Seters demonstrated that the Pentateuch strikingly re­
sembles the Greek authors in various ways, both in style and also in its
aims. The Pentateuch was an attempt to give the Jewish nation, in a late
) stage of its history, a sense of identity and of its past. In my The Making
of the Pentateuch ( 1 987) I have attempted to support Van Seters's thesis,
while further suggesting that the hypothesis of a priestly author later than
the (late) Yahwist_ may ht; unnecessary, and that the Pentateuch. may be
regarded as to all intents and purposes the work of a single author.
Readers previously unacquainted with the problem of the composi­
tion of the Pentateuch will no doubt find the above historical outline of
conflicting theories - sketchy and incomplete though it is - bewilder­
ing. But at least it will be clear, as stated at the beginning of this chapter,
that although certain tendencies can perhaps · be discerned - one of these
may well be the use of literary theory to reveal more clearly the character
of the completed Pentateuch as a work ofliterature - no unanimity about
its provenance exists at the present time. The debate is likely to continue
indefinitely, and whether a new consensus will eventually emerge is far
from certain. It would therefore be premature to attempt either an assess­
ment of the present situation or a prognostication of the future. But it is
important to realize that in such a matter as this we are dealing entirely

26
Who Wrote It? Problems of Composition

with hypotheses and not with facts. Proof, either in the mathematical or
in the logical meaning of that word, will never be attainable. The only fact
available to us is the text of the Pentateuch itself in all its complexity.

For Further Reading

Alter, Robert. The Art of Biblical Narrative. New York: Basic Books and
London: Allen and Unwin, 1 98 1 .
Blenkinsopp, Joseph. The Pentateuch: An Introduction to the First Five Books
ofthe Bible. New York: Doubleday and London: SCM, 1 992, ch. 1 .
Childs, Brevard S . Introduction to the Old Testament as Scripture. Philadel­
phia: Fortress and London: SCM, 1 979, 1 1 2- 1 35 .
Driver, Samuel R. Introduction to the Literature of the Old Testament. 8th
ed. Edinburgh: T. & T. Clark, 1 909.
Eissfeldt, Otto. The Old Testament: An Introduction. New York: Harper
and Oxford: Blackwell, 1 965, 1 5 8- 1 82. (First published in German,
1 934; this translation is based on the 3rd German ed., 1 964.)
Gunkel, Hermann. The Legends of Genesis. New York: Schocken, 1 964.
(Translated from the introduction to his first [German] commentary
on Genesis, 1 9 0 1 .)
Haran, Menahem. "Behind the Scenes of History: Determining the Date
of the Priestly Source," ]BL 100 ( 1 9 8 1 ) : 32 1 -333.
Hurvitz, Avi. A Linguistic Study of the Relationship Between the Priestly
Source and the Book ofEzekiel Paris: Gabalda, 1 982.
Kaiser, Otto. Introduction to the Old Testament. Minneapolis: Augsburg
and Oxford: Blackwell, 1 975, 33-45 . (First published in German,
1 969.)
Mendenhall, George E. "Covenant Forms in Israelite Tradition," BA 1 7
( 1 954) : 50-76. Repr. i n The BA Reader 3, ed. E . F. Campbell, Jr.,
and David Noel Freedman. Garden City: Doubleday, 1 970, 25-53.
Noth, Martin. A History ofPentateuchal Traditions. Englewood Cliffs, N.J . :
Prentice-Hall, 1 972. Repr. Atlanta: Scholars Press, 1 989. (First pub­
lished in German, 1 948.)
von Rad, Gerhard. "The Form-Critical Problem of the Hexateuch," in
The Problem of the Hexateuch and Other Essays. Edinburgh: Oliver
& Boyd, 1 966, 1 -78. Repr. Philadelphia: Fortress and London:
SCM, 1 984. (First published in German, 1 938 .)
Rendtorff, Rolf. The Old Testament: An Introduction. London: SCM, 1 98 5 ,

27
INTRODUCTION TO THE PENTATEUCH

and Philadelphia: Fortress, 1 986, 1 3 1 - 1 64. (First published in Ger­


man, 1 983.)
--- . The Problem ofthe Process of Transmission in the Pentateuch. ]SOT
Supplement 89. Sheffield: JSOT Press, 1 990. (First published in
German, 1 977.)
Schmidt, Werner H. Old Testament Introduction. London: SCM, 1 984,
44-6 1 . (Translated from the 2nd German ed., 1 982.)
Soggin, J. Alberto. Introduction to the Old Testament. OTL. Rev. ed.
Philadelphia: Westminster and London: SCM, 1 980, 83-98. (Trans­
lated from the Italian ed. , 1 974.)
Thompson, Thomas L. The Historicity ofthe Patriarchal Narratives. BZAW
1 33. Berlin: de Gruyter, 1 974.
Van Seters, John. Abraham in History and Tradition. New Haven: Yale
University Press, 1 975.
---. In Search of History: Historiography in the Ancient World and the
Origins ofBiblical History. New Haven: Yale University Press, 1 983.
Wellhausen, Julius. Prolegomena to the History ofAncient Israel. Edinburgh:
A. & C. Black, 1 88 5 . Repr. Magnolia, Mass. : Peter Smith, 1 977.
(First published in German, 1 878.)
Whybray, R. Norman. The Making of the Pentateuch. ]SOT Supplement
5 3 . Sheffield: JSOT Press, 1 987.
Wolff, Hans Walter. "The Elohistic Fragments in the Pentateuch," Inter­
pretation 26 (1 972) : 1 58- 1 73 .

See also Rolf Rendtorff, "The 'Yahwist' as Theologian? The Dilemma of


Pentateuchal Criticism," with responses and reviews by various scholars
in ]SOT 3 ( 1 977) : 2-60.

28
CHAPTER 3

The ''Prim eval History"


(Gen esis 1-1 1)

Trather than
HESE CHAPTERS must be regarded as a "prologue" to the Pentateuch
as part of the main body of the work. If the aim of the
Pentateuch as a whole is to narrate a connected human "history'' (even
though in a sense different from that of modern historians), these chapters
- which begin with the action of God even before the universe existed
(Gen. 1 : 1 -2) - can obviously not be based on any record of what actually
"occurred." And although the later parts of these chapters refer to many
names of human persons and in some cases (Cain, Abel, Enoch, Noah)
recount their actions and even mention the number of years that they
lived (ch. 5; 9:29; 1 1 : 1 0-32) , the very longevity of these persons alone is
enough to show that we are here dealing with a very different "world"
even from that of the chapters which follow in Genesis - the stories about
Abraham and his family.
These stories do not constitute a single narrative sequence. They have
been linked together only in a very artificial way with long genealogies
(4: 1 7-22; 5 : 1 -32; 1 0: 1 -32; 1 1 : 1 0-32) . They are, in fact, "universal" stories.
They deal not with human beings as we know them but with "giants" or
"heroes" in something like the legendary sense of those words. They tell
us how their authors, or their authors' contemporaries, imagined that it
"might all have begun." However, as we shall see, they also had a much
deeper purpose than that.
Many peoples have, or have had at some stage of their development,
comparable stories about the origins of the world and the early history of
the human race. Moreover, many of the stories in Gen. 1 - 1 1 have a

29
INTRODUCTION TO THE PENTATEUCH

"family" likeness to various "origin stories" current in the ancient Near


Eastern milieu to which ancient Israel belonged (see ANET, 3- 1 5 5 , and
the works listed at the end of this chapter) . These all bear the marks of
the _ particular civilizations - Egyptian, Mesopotamian, Hittite, etc. -
which produced them, . and Gen. 1 - 1 1 is no exception. The most obvious
difference between the Israelite "universal history'' and the others is that
the Israelite stories - in their present form at least - are monotheistic.
All the others are polytheistic. That is, only in the Israelite stories is all
the divine action, from the creation of the world on, attributed to the
unaided work of a single supreme God. Indeed, there is not the slightest
suggestion that there are other gods, while in the non-Israelite stories there
is a multitude of "gods." These ancient Near Eastern deities sometimes
collaborate as a kind of "committee" but at other times violently or subtly
oppose one another; none possesses universal power, though one may be
more powerful than the others. The term "myth'' is often applied to these
stories; but since there is no agreement about the meaning of this term it
is probably best to avoid it (see John Rogerson, Genesis 1-11, 55) .
Genesis 1 - 1 1 is, as a literary work, more than the sum of its parts.
In order to understand its nature and purpose it is first necessary to take
a preliminary look at its contents. These may be summarized as follows
(slightly different distribution of the topics is possible) :

l : l-2:4a: An account of the creation by God (Elohim) of the


heavens and the earth and all the species of flora and
fauna including humankind
2:4b-2 5 : An account of the creation by the Lord God (Yahweh
Elohim) of the first man and woman and his placing
them in a "garden in Eden" to till it and manage it
3 : 1 -24: Story of the eating of a forbidden fruit by the man and
woman at the instigation of a cunning snake, and their
consequent expulsion from the garden
4: 1 - 1 6: Story of Cain's murder of his brother Abel and his banish­
ment by Yahweh
4: 1 7-22: List of descendants of Cain
4:23-24: Lamech's song
4:25-5 :32: Lists of descendants of Seth
6: 1 -4: Story of the union of the "sons of God" with human
women and the appearance of the giants (Nephilim)
6:5-9 : 1 7: Story of God's decision to destroy a corrupted humanity

30
The "Primeval History " (Genesis 1-1 1)

by a flood, the sending of that flood, the rescue of Noah


and his family in an ark, and the institution of a new
beginning for humankind
9: 1 8-29: The sons of Noah, his drunkenness, and his death
1 0 : 1 -32: List of descendants of Noah's sons (the "Table of the
Nations")
1 1 : 1 -9 : Story of an arrogant attempt by mankind (literally, "the
whole earth") to build a city and a tower "with its top in
the heavens" and Yahweh's frustration of their project by
scattering them throughout the earth · and "confusing
their language," so rendering mutual communication im­
possible for them (the story of the "Tower of Babel")
1 1 : 1 0-30: List of descendants of Shern
1 1 :3 1 -32: Migration of Terah, Abram, Lot, and Sarai from Ur of
the Chaldeans to Haran

It has been suggested that the final author or compiler of these


chapters has left an indication of the structure of the work in his use of
the Hebrew word toledoth, especially in the phrase "These are the toledoth
of. . . . " This, or comparable phrases, occurs in Gen. 2:4 ("of the heavens
and the earth'') ; 5 : 1 ("of Adam'') ; 6:9 ("of Noah'') ; 1 0 : 1 ("of the sons of
Noah"); 1 1 : 1 0 ("of Shern"); 1 1 :27 ("of Terah") . Its use was not, however,
confined to chs. 1 - 1 1 ; the phrase occurs also several times in the later
chapters of Genesis with regard to Ishmael (25 : 1 2), Isaac (25 : 1 9) , Esau
(36: 1 , 9) , Jacob (37:2, where RSV reads "This is the history of the family
of Jacob") . It cannot, however, be regarded as a basic structural marker
either of chs. 1 - 1 1 or of the whole book of Genesis. One reason is that
the word toledoth is used, in these chapters as elsewhere, in more than
one distinct sense, and with more than one distinct function. Its most
frequent sense is that of "descendants" or "list of descendants"
(genealogy); here its function is to introduce a list of names (Adam, Cain,
Lamech, Seth, Noah, Shern) . But elsewhere it appears to mean "history''
or "narrative." For example, "These anuhe toledoth of the heavens and
the earth when they were created" (the only occurrence of the phrase
where it is not associated with personal names) in 2:4 appears to be a
reference to events (i.e. , to God's creative acts) . Moreover, its function
here is not introductory but summarizing; it forms the conclusion of a
narrative. In 37:2 "These are the toledoth of Jacob" also refers to a
narrative rather than to a list of names; but in that case it precedes the

31
INTRODUCTION TO THE PENTATEUCH

narrative of the chapters which follow, which recount the history of the
family of Jacob-Joseph and his brothers.
Clearly the " toledoth series" cannot be regarded as a reliable indication
of the structure of Gen. 1 - 1 1 . It is not, in fact, a coherent system; the
term toledoth has been used here in several quite different ways. But this
does not mean that these chapters have no comprehensive structure.
One way of discerning the structure of Gen. 1 - 1 1 is to note the
fluctuation in the relationship - if one may so call it - between the
· principal characters, God and his human creatures. This fluctuation can
perhaps be most clearly seen in some of the words spoken by these
characters and the comments of the narrator. For example,

Narrator ( 1 :27) : So God created humankind (Hebrew adham) in his own


image.
Narrator ( 1 :3 1 ) : God saw everything that he had made, and behold, it
was very good.
God to the man · (2: 1 6� 1 7) : You may freely eat of every tree in the garden;
but of the tree of the knowledge of good and evil you must not eat,
for in the day that you eat of it you will surely die.
Snake to the woman (3:4-5) : You will not die; for God knows that when
you eat of it your eyes will be opened, and you will be like God,
knowing good and evil.
God to the .man (3 : 1 7- 1 9) : Because you . . . have eaten of the tree which
I commanded you that you should not eat of it, cursed is the ground
because of you.
God to himself (3:22) : Behold, the man has become like one of us, know­
ing good and evil; and now, lest he put forth his hand and take also
from the tree of life, and eat, and live for ever. . . .
Narrator (3:23) : Therefore the Lord God sent him forth from the garden
of Eden to till the ground from which he had been taken.
Narrator (4:8) : Cain rose up against his brother Abel, and killed him.
God to Cain (4: 1 2) : You will be a fugitive and a wanderer on the earth.
Narrator (4: 1 5) : But the Lord put a mark on Cain, so that no one who
came upon him would kill him.
Narrator (4:26) : At that time people began to call upon the name of the
Lord.
Narrau:ir (5:24) : Enoch walked with God; then he was no more, for God
took him.
Narrator (6:5) : The Lord saw that human wickedness was great . . . and

32
The ''Primeval History " (Genesis 1-1 1)

that every inclination of the thoughts of their hearts was only evil
continually.
God to himself (6:7) : I shall blot out from the earth the human beings
that I have created . . . , for I regret that I made them.
Narrator (6:8) : But Noah found favor in the eyes of the Lord.
Narrator (6:9) : Noah walked with God.
God to himself (8:2 1 ) : I will never again curse the ground . . . nor will I
ever again destroy every living creature as I have done.
God to Noah (9: 1 1- 1 2) : I establish my covenant with you . . . for all
generations.
Narrator (9:20-2,1 ) : Noah . . . became drunk, and he lay uncovered in
his tent.
"The whole earth'' ( all mankind) (1 1 :4) : Let us build ourselves a city,
=

and a tower with its top in the heavens, and let us make a name for
ourselves.
God to himself ( 1 1 :6) : Behold, . . . this is only the beginning of what
they will do; nothing that they plan to do will be impossible for
them.
Narrator (1 1 :8-9) : The Lord scattered them abroad . . . over the face of
all the earth, and they left off building the city. Therefore it was
called Babel, because the Lord confused the language of all the earth.

Looking back on the above data, one may say that the author
presents the reader with a rather unexpected portrait of God. Despite
· the assurance at the beginning ( 1 : 3 1 ) that when God surveyed his creative
actions he concluded that the result was wholly good, God appears quite
soon to be somewhat nervous. about what he has done and, in particular,
concerned about his own supremacy over his creatures. He sets conditions ·

to mankind's freedom (2: 1 7) and threatens immediate death as a con­


sequence of disobedience, a threat that the snake - quite correctly, as it
turns out - flatly rejects, calling God a liar. When the act of disobe­
dience has been committed, God betrays nervousness about the possi­
bility that mankind may now go further and seize the immortality which
properly belongs only to God (3 :22) . The same anxiety betrays itself in
the final episode of these chapters ( 1 1 :6) , when God fears that mankind
may succeed in wresting unlimited (i.e. , divine) power for themselves.
In. both cases God takes steps to forestall these ambitions. Meanwhile
the mysterious incident of the union of the "sons of God" with the .
daughters of men (6: 1 -2) also seems to have been regarded by God as

33
INTRODUCTION TO THE PENTATEUCH

containing a threat of immortality human or semi-divine (compare 3:22) ,


which needed to be suppressed.
It is in keeping with this presentation of the divine character that
God is credited with a change of mind, and one of crucial importance for
mankind. He regrets that he created them and determines to destroy them
(6: 7) because they have become irredeemably wicked (6:5). Creating them,
then, was a mistake; the God who had at first assessed his own creative
work as perfect had been unable to foresee what it would become.
Throughout most of this narrative God appears to be reacting at every
turn to what mankind was doing rather than taking the initiative. For­
tunately for mankind (and God?) , Noah turned out to be an exception to
the general depravity (6:8, 9) , and a new beginning became possible.
An alternative way of viewing the purpose and structure of chs. 1 - 1 1 ,
however, is that it presents a picture of "the growing power o f sin in the
world," together with a parallel picture of "a hidden growth of grace" {so
Gerhard von Rad, "The Form-Critical Problem of the Hexateuch," 64,
65) . This view has some plausibility as regards chs. 2-9: the initial act of
disobedience is followed by the first murder, then by the boast of Lamech
that he will wreak vengeance not merely seven but seventy-seven times
(4:24) , and finally by the total and irredeemable depravity of mankind in
ch. 6. It might also be argued that although there is a corresponding
increase in the severity of the punishments inflicted on the sinners, culmi­
nating in the virtually total destruction of mankind, the extent of the
divine grace and forgiveness also expands, culminating in ch. 9 with the
blessing given to Noah and his descendants and the covenant made by
. God not only with these but also with all living creatures (9: 1 2) , coupled
with God's promise never again to destroy mankind or to disrupt the
normal course of a beneficent nature (8:2 1 -22) .
But if such a pattern is in fact to be found here, it surely comes to an
end in 9: 17. The story of the Flood and the (partial) restoration of (what is
left of) humanity to something resembling, the dose relationship with God
portrayed in ch. 2 marks a satisfactory co nclusion to the whole "universal
history." The view of von Rad ("The Form-Critical Problem of the Hexa­
teuch," 66, n. 1 07) that it is the story of the Tower of Babel ( 1 1 : 1-9) which
marks "the climax of the whole section" is not well founded. Rather than a
climax, the scattering of humanity and the confusion of their speech, though
hardly -propitious for mankind's future, came rather as an anticlimax after
the story of their almost total destruction in the Flood. Indeed, one might
suppose that the story of the Tower of Babel is a: kind ofauthor's afterthought.

34
The ''Primeval History " (Genesis 1-1 1)

However, it is probably best to regard it as an appendix to chs. 1 -9, included


as a final parable of human sin and God's response. It is significant that the
Tower of Babel account contains no proper names and mentions no in­
dividuals at all; after the plethora of proper names (especially in ch. 1 0) this
story seems to move in a different world. The builders of the city and tower
are curiously called "the whole earth" (v. 1 ) and "they'' (v. 2) . The story is a
powerful parable about the fate that awaits human rebelliousness, ambition,
and arrogance - a timeless narrative akin to the fairy story. But in its present
position it also balances - and corrects - ch. 1 with its insistence that what
God created was "very good."
Genesis 1 - 1 1 , then, is not merely concerned with speculating about
the remote past - about "how it might have been'' when the world began.
These chapters have clearly been composed with an eye to what was to follow:
the som�hat more recent past and the history of God's chosen people. They
do not, it is true, mention the people of lsrael, which had not yet come into
existence; they are about mankind in general - a universal history - but
they end with an unmistakable presage of what is to come. With 1 1 : 1 0 there
is a resumption of the genealogical lists, with a list of the descendants of
Shem. Toward the end of that chapter we encounter Terah, the father of
Abram. Finally in v. 3 1 we are told that Terah, together with his son Abram,
his grandson Lot (the son of Haran) , and Abram's wife Sarai migrated from
his home in Ur of the Chaldeans to go into the land of Canaan, but that they
went only part of the way and settled in Haran. With this information we
have been introduced to the principal characters of the chapters which
immediately follow, in particular, to the ancestors of Israel - to ''.Abraham
our father" and to Sarah, who was to become the mother of the promised
heir Isaac and the ancestress of the nation. With this, the prologue to the
Pentateuch is at an end.
What is the point of this prologue? Why did the author of the
Pentateuch begin his history of Israel's ancestors with these chapters?
In attempting to answer t.hese questions, it may first be pointed out
that histories of particular nations or peoples which began with an account
of the origins of the world and then, by genealogical or other means, linked
this to "historical" times were not entirely unusual in ancient literature.
Israel was by no means unique in this respect. This is especially true of
some of the early Greek historians of the sixth and fifth centuries B.C.
(see John Van Seters, In Search ofHistory, 8ff.). But the very early Sumerian
King List (ca. 1 800-1 600; ANET, 265-66), which begins with the words
"When kingship was lowered from heaven," immediately proceeds with a

35
INTRODUCTION TO THE PENTATEUCH

list of dearly legendary kings there stated to have reigned "before the
Flood" for an incredible number of years (e.g., Alalgar, 36,000 years; cf.,
though in less extravagant style, the lifespans in Gen. 5 and 1 1 : 1 0-25)
before descending to more "historical" kings with reigns of "normal"
lengths. The so-called Creation Epic (Enuma Blish) from Babylonia
(ANET, 60-72) represents the creator-god Marduk as having subsequently
ordered and carried out the building of the undoubtedly historical city of
Babylon so that he might be worshipped there. Similar "histories" that
proceed from the mythical origins of (he world to accounts of actual
historical persons and events can also be cited from other parts of the
world - for example, the Japanese Kojiki, "Chronicle of Ancient Times."
Evidently an interest in the way in which the world and humankind
came into existence and in the history of the earliest times was charac­
teristic of the ancient civilized world. At any rate, various "origin stories"
or "creation myths" about the activities of a variety of creator-gods are still
extant in what remains of the literatures of ancient Egypt and ancient
Mesopotamia. But the combination of such accounts with narratives about
more recent times testifies to an additional motivation. The aim of such
works was to give their readers - or to strengthen - a sense of national
or ethnic identity, particularly at a time when there was for some reason a
degree of uncertainty or hesitation about this. Such works are basically
national or ethnic histories rather than universal ones: the center of interest
is a particular ethnic group. In order to foster a sense of identity it was
necessary to create an understanding of the place of that nation or people
in the world - that is, among the other nations in the world whose
existence was acknowledged and with which some kind of relations. were
necessary, but which were felt to be alien. It was therefore important to
know how a humanity that was believed to have had a single origin had
become divided into separate nations and had developed different customs
and languages. In the case of Gen. 1 - 1 1 , this theme - already hinted at
in the various genealogies - is explicitly dealt with in ch. 1 0, evidently
intended as a comprehensive list of the peoples of the world and the
locations to which they had migrated, and in 1 1 : 1 -9, which accounts for
their failure to remain united and gives the reason for their dispersal and
their subsequent mutual alienation through inability to communicate with
one another.
The placing of Gen. 1 - 1 1 as a prologue to the main body of the
work :rlso afforded the opportunity to express certain distinctively Israelite
articles of faith which it would have been more difficult to introduce into

36
The "Primeval History " (Genesis 1-11)

the later narratives, particularly with regard to the doctrine of God. Specifi­
cally, although the creation narratives of chs. 1 and 2 have unmistakable
affinities with those of other Near Eastern peoples (see below), the mon­
otheistic character of all these chapters is quite striking and even polemical,
despite the isolated "Let us make mankind in our image" of 1 :26 and "like
one of us" in 3 :22. Ch. 1 contains further examples of a polemical or
anti-polytheistic stance. For example, whereas in the Babylonian Enuma
Blish the stars are divine beings (Tablet V, lines lf£; ANET, 67) , in Gen.
1 : 1 4- 1 8 the heavenly bodies are merely created objects set in the sky with
specific functions to perform. Again, the "great sea monsters" . (Hebrew
tanninim, 1 :2 1 ) , elsewhere in the Old Testament (Ps. 74: 1 3- 1 4; Isa. 5 1 :9)
presented as adversaries that God had to fight and kill (a tradition found
also in the non-Israelite literature, e.g., Enuma Blish, Tablet IY, lines
1 O l f£), simply appear as his creatures. Here we have a "demythologization''
of polytheistic beliefs held both in Israel and elsewhere. There are other
polemical traits in these chapters - for example, in the story of the Flood
- which point to a time of composition when Israel's theology, already
monotheistic, had been again subject to insidious polytheistic notions,
especially those of Babylonian origin.
Some scholars have gone further and seen the narratives of Gen. 1 - 1 1
as a whole to be reflecting the experiences of the Babylonian Exile or the
early postexilic period. Thus the themes of punishment for sin, especially
banishment from God's presence and/or dispersal or destruction (3:23-24;
4: 1 2, 1 6; 6-8; 1 1 :4, 9), have been taken as symbolic of Israel's richly
deserved banishment from the land of Canaan. The signs of grace and
forgiveness, especially God's acceptance of Noah's sacrifice and the
covenant which God made with him (8:20-9 : 1 7) , would, it has been
supposed, suggest to the exilic/postexilic reader that God has even now
not cast off his people but is a God of infinite patience and forgiveness
who may yet again rescue them from their folly and their guilt. Such an
allegorical interpretation of these chapters, however, perhaps reads more
into them than their author intended.
These stories also betray an interest in etiology - in seeking the
origin of various phenomena of universal human experience which appear
to defy explanation. This kind of questioning ("Why?") is a further in­
dication of the breadth of approach of this "universal history." The ques­
tions raised (by implication) here are almost all questions concerning the
basic aspects of human existence, and have been taken to reflect a kind of
universal "wisdom" or seeking after truth not confined to any one people.

37
INTRODUCTION TO THE PENTATEUCH

(The genealogical lists in these chapters, of course, also imply that in view
of their common descent all human beings are, in one sense, "brothers.")
The etiologies offered in these chapters are of many kinds. The
question about the reason for human mortality; a common theme in both
Near Eastern and classical thought which sometimes took the form of
narratives in which human beings attempted to wrest immortality from
the gods but failed, is alluded to in 3 :22 - which appears to imply that
mortality is inherent in mankind's status as creature - and in the myste­
rious incident of 6 : 1 -3. It would therefore seem that mortality was not
held to be a punishment for human disobedience, although it is probably
significant that the first human death recorded - inflicted by a fellow
human being - occurred immediately after the expulsion from the gar­
den.
The nature of the relationship between man and woman is discussed
in 2: 1 8 , which explains why both sexes are necessary to a complete human­
ity; and in 2:23-24, which explains the attraction between the sexes and
the forming of permanent relationships between them as due to God's
providence. (Note the word "Therefore" in v. 24 - an explicit introduc­
tion to an etiology, answering the here unspoken question, ''"Why?") . But
in ch. 3 the less ideal realities of the relationship as we know it are attributed
to the disobedience to God's command, in which both the man and the
woman (as well as the snake) are equally involved. V. 1 6, which is the first
reference to childbearing, provides an explanation of women's labor pains,
and also of men's dominance over women.
There is also an etiology of work here. Work in itself is not regarded
as a punishment. Instead, it is a natural and essential (male) activity (2: 1 5) ,
but - i t is implied - a pleasant one. Life in the garden i s pleasant (2: 9) .
The cursing of the ground and the cortsequent harshness of agricultural
labor (3: 1 7- 1 9) are the result of the act of disobedience; the final line of
v. 1 9 ("You are dust, and to dust you shall return"), possibly a common
saying, does not imply that human mortality is the result of disobedience
(see the discussion in Gerhard von Rad, Genesis).
Another matter which evidently needed explanation was the phe­
nomenon of human clothing. The feelings of shame at appearing naked
before others (c£ 9:20-27) and the assumption of the need for clothes are
explained as a consequence of the man and the woman having eaten of
the for-bidden fruit of the tree of knowledge (3:7- 1 2, 2 1 ) ; it is specifically
stated that previously they had not been ashamed of their nakedness (2:25) .
Other etiologies in these chapters include the cause of the general human

38
The "Primeval History " (Genesis 1-1 1)

dislike of snakes and also of snakes' ability to move without legs (3: 14-1 5)
and the reason for the existence of the rainbow (9: 12- 17) .
An etiology which is wholly peculiar to Israel is that of the origin of
the sabbath, unmistakably implied in 2: 1-3 in the statement that God
blessed and declared holy the seventh day, on which he rested from his
creative work. This is an example of a tendency observable elsewhere in
the Pentateuch (e.g. , in many of the laws, and in particular in the descrip­
tion of the tabernacle in the wilderness [Exod. 25ff.] , which was in many
respects an anticipation of Solomon's temple) to ascribe the earliest con­
ceivable time to the establishment of a refigious institution.
To assign a date or historical setting to these chapters is not possible
with any degree of certainty. They are, in a sense, timeless. Source criticism
seems hardly relevant here; in some form these stories could have arisen
at almost any time in the history of Israel. Certain features in the text as
it now stands, however, point to a fairly late date. This applies not only
to the sections which the documentary critics label as P, but also to the
sections attributed to the preexilic sources J and E. While some theories
(e.g., that the theme of banishment from God's presence reflects the exilic
period) are speculative and unprovable, references like that to Ur of the
Chaldeans in Gen. 1 1 :28, 3 1 seem to rule out an early date in view of the
late appearance of the Chaldeans on the international stage. There are
other pointers, not conclusive but nevertheless suggestive, of a late recen­
sion. For example, it is somewhat strange that the garden of Eden is
nowhere mentioned in preexilic Old Testament texts but appears only in
the exilic Isaiah ("Deutero-Isaiah''), Ezekiel (where ch. 28 has a variant
version of the Eden story),. Joel, and Habakkuk. Further, the dependence
on Mesopotamian traditions evident in several chapters raises the question
how these could have been known to an Israelite writer before the period
of Babylonian influence on Israel - from the late seventh century at the
earliest. It is, however, of course possible that much of the material may
have existed in some form some time before its inclusion in these chapters.
If the Pentateuch is to be regarded as a single work, these considerations
are significant for the dating of the work as a whole. John Van Seters's
remark about the patriarchal stories that "there is nothing in this presen­
tation . . . which is inappropriate to . . . the period of the late Judean
monarchy or exilic periods, but there is much that speaks against the choice
of any earlier period" should be borne in mind (Abraham in History and
Tradition, 38) .
It is generally agreed that the stories in Gen. 1 - 1 1 are not a pure

39
INTRODUCTION TO THE PENTATEUCH

invention of the author. However much he may have adapted them to his
own purposes, he clearly made use of traditional themes currently circu­
lating in his own time. Where had these originated? First, it is important
to note that there is no extant ancient Near Eastern text that in any way
covers the same ground as Gen. 1 - 1 1 - that is, that covers all the main
episodes - and no evidence that any other people compiled a comparable
narrative before the Graeco-Roman period, when there was an interest in
such matters and works of this kind may well have become quite common.
Berossus (third century) included such material in a history of Babylon,
Manetho (second century) similarly in a history of Egypt, and Philo of
Byblos (first-second century A.D.) in his Phoenician History. These works,
all written in Greek and unfortunately preserved only in comparatively
brief quotations in other works, attempted to link traditional stories about
the doings of the gods with later historical events. Some earlier Greek
works had already made efforts in this direction. Hesiod's WOrks and Days
(probably eighth century B . C.) was a precursor, and the later Greek his­
torians Hecataeus and Hellanicus, whose works also are preserved only in
fragmentary form, carried the enterprise further (see Van Seters, In Search
ofHistory, 8 - 1 8 ) . But there is no evidence of comparable works from Egypt
or Mesopotamia before the Hellenistic period. There was, however, an
abundance of unconnected stories from those quarters about such matters
as the creation of the world, the creation of mankind, the Flood, and other
"primeval" events.

Genesis l:J-2:4a

This creation story is only one of many current in the ancient Near East.
For example, there are several Egyptian stories extant in which the creation
of the world is attributed to different gods, and the creator-god is not
·
necessarily the principal god - a multiplicity which is due to different
local traditions. Also in Israel itse1£ where there is only one creator, who
has supreme power and no rivals, there are several different versions. In
addition to Gen. 1 and 2, there is another version of which traces appear
in various contexts, in which the creation appears to have followed a
conflict wherein Yahweh defeated or killed a sea monster or monsters
(especi t1Hy Ps. 74: 1 3 - 1 7; Isa. 5 1 : 9) , and yet other versions in Prov. 8 :22-3 1 ,
in parts o f the book of Job, and elsewhere.
The creation story in Gen. l : l-2:4a has long been thought to have

40
The ''Primeval History " (Genesis 1-1 1)

particular affinities with Enuma Elish (ANET, 60-72) ; but a glance at the
Mesopotamian myth shows that the relationship is at most a very remote
one. Apart from the fact that the Genesis story is monotheistic, a crucial
difference between the two accounts is that Enuma Elish belongs to the
category of the conflict tradition, which is entirely absent from the Genesis
account. In Enuma Elish Marduk is able to create the world only by
summoning his allies and killing the sea monster Tiamat and her allies;
heaven and earth are created by the splitting of Tiamat's body into two.
The commonly repeated notion that the Hebrew word translated "the
deep" (tehom, Gen. 1 :2) is a pale reminiscence of Tiamat cannot be
sustained. There is no trace of a conflict here; God is alone, and he is
supreme. There is no explicit statement in the Genesis narrative about
God's purpose in creating the world. However, this purpose is clearly
implied in the great emphasis that is placed on the position of mankind
in God's plan: the creation of mankind, the last of God's creative acts, is
evidently the climax of the whole account, and receives the greatest atten­
tion. That which was created on previous days - light, day and night,
dry land, heavenly bodies, plants and animals - are all by implication
provided for mankind's use and convenience. Mankind is given the plants
for food and power over the. animal creation. Above all, mankind is created
in God's image and likeness. Whatever may be the precise meaning of these
terms (this question has been endlessly debated) , they set mankind apart
from all the other creatures and put them in a unique relationship with
God himself. In none of the other creation stories with which this story
can be compared is such a high status attributed to mankind.
In its cosmology - that is, its understanding of the different parts
of the universe and their relationship to one another - Gen. 1 conforms
to the view generally accepted in the ancient Near East. (In other passages
in the Old Testament this cosmology is described more fully.) The pre­
existent watery waste ( 1 : 1 -2) was divided into two by the creation of a
solid dome or vault (the sky, v. 8) so that there was water both above and
below it. The water below was then confined to a limited area, the sea,
revealing the dry land, which God called "the earth." (According to the
story of the Flood, 7: 1 1 , the sky had "windows" which when opened
allowed the rain to fall.) The heavenly bodies, sun, moon, and stars, moved
across the vault of the sky.
A characteristic feature of Gen. 1 in which it differs from other
creation stories, both Israelite and non-Israelite, is its neatness and the
precision of its presentation of the acts of creation. Using the same phrase-

41
INTRODUCTION TO THE PENTATEUCH

ology repeatedly, it lists these with the dryness of a catalogue. It has, . for
example, nothing of the drama or the imaginative skill of ch. 2. It gives
the impression of an account which has been carefully honed and reduced
systematically to a minimum. Yet, as Claus Westermann and others have
pointed out, there remain certain variations in detail. Thus the creative
acts are introduced in two different ways. In some cases God creates simply
by speaking: ''And God said . . . . " In others we are told that God per­
formed certain actions: he made, separated, named, blessed, placed. A
second "untidy'' feature is that although the entire work of creation was
carried out in six days (to conform with the concept of a week of action
followed by a sabbath rest on the seventh day), there are in fact eight
creative acts: on the third day and again on the sixth two acts of creation
are performed. These features suggest .that the account in its present form
is based on earlier accounts in which the work of creation was originally
performed in different ways, although there is now no way in which these
earlier accounts can be reconstructed.

Genesis 2:4b-3:24

It is evident that this narrative, which could stand by itself as a complete


and independent story, has taken up themes and motifs quite different
from those employed in ch. 1 . It was once generally believed to be older
and more "primitive" in its theology than ch. 1 0 as contrasted with P) .
But more recently it has been doubted whether its "naivete" is in fact more
than apparent. Joseph Blenkinsopp, for example, sees the narrative as
"generated by reflection on the creation account in Genesis I " and as
standing in a wisdom tradition which indulged in "what Plato called
'philosophizing by means of myth' " ( The Pentateuch, 63, 65). Un­
doubtedly some of the motifs employed are themselves considerably older
than the rather late period of composition postulated above; but it is true
that the · telling of tales for edifying or didactic purposes is a characteristic
of a late state of civilization rather than an early one. There is evidence,
too, that some of the vocabulary used in these chapters is distinctively late
rather than early.
This is primarily a story about two people, a man and a woman,
and wh a r happened to them. Although these people are, in the context,
necessarily pictured as the first man and woman, they are clearly symbols
as well as ancestors of the human race itself. Behind his statements that

42
The "Primeval History " (Genesis 1-1 1)

"this is what happened," the author is saying "this is how human beings
behave, and these are the consequences that follow" : the eating of the fruit
is not a single event in the remote past, but something that is repeated
again and again in human history. There is thus here a series of lessons,
applicable to human beings in general, but also in particular to the
history of Israel.. God's intention for the human beings that he has
created is wholly good, but they can be led away by subtle temptation.
Also, disobedience to God, which is self-assertion, may bring greater self­
knowledge, but it leads to disaster. The intimate relationship with God is
broken, and life then becomes harsh and unpleasant. However, even then
God does not entirely abandon his creatures but makes special provisions
for their future life. An Israel which had suffered devastation and exile
from their land as a result of their disobedience to God could hardly fail
to get the message.
It is hardly correct to call ch. 2 a "second creation story'' as is
frequently done, if by that is meant that this is an account of the creation
of the world alternative to that of ch. 1 . Rather, ch. 2 is concerned with
the creation of human beings, and the reference to the creation of the
world (which occupies only vv. 4b-6) simply provides the setting for that.
This account clearly originally belonged to a different tradition from ch.
1 with its Babylonian perspective. The perspective here is that of Palestine,
where rain was crucially important for the fertilization of plant life. But
the author of chs. 2 and 3 has assembled a whole battery of different
traditions to adorn his narrative.
In 2:7 the author chose to describe the creation of humankind in
terms of their formation from the soil (perhaps rather, clay) . It is not
possible to identify the particular tradition that he used: Westermann
pointed out that this .is a common notion among primitive peoples. From
the civilized ancient Near East we know of the Egyptian god Khnum, who
fashions living creatures on a potter's wheel (see ANET, 368, 43 1 , 44 1 ) ,
and from Mesopotamia there i s a reference to the creation o f the wild man
Enkidu from clay ( Gilgamesh, Tablet I, ii, 30ff.; ANET, 74) .
The references to the tree of the knowledge of good and evil (Gen.
2:9, 1 7 and presumably also 3:3, 1 1 , 12) and to the tree of life (2:9; 3 : 22)
constitute a puzzle, in that the latter does not appear in the main story
but only in the two verses mentioned. The problem is usually, and probably
rightly, solved by supposing that the author knowingly combined two
separate traditions and was not much concerned with consistency of detail.
This is not the only inconsistency in these chapters, and it would not be

43
INTRODUCTION TO THE PENTATEUCH

appropriate to speak either of a combination of literary sources or of


subsequent additions made to a completed text.
Both trees have connections with wisdom themes. Knowledge is a
synonym for wisd�m in the book of Proverbs, and in Prov. 3: 1 8 it is stated
that wisdom is a "a tree of life to those who lay hold of her." This might
lead one to suppose that the two trees were the same, were it not for the
statement in Gen. 2:9 and 3:22 that they were distinct.
The themes of knowledge and of immortality have in common the
fact that the attempt . to attain either is an attempt to obtain what God
has not given to mankind and so to encroach on the divine prerogative.
In the Old Testament wisdom is an ambiguous quality, which may be used
either for good purposes or for evil ones. In 3: 1 for example, the wisdom
possessed by the snake (said to be arum, "shrewd") is not presented as
admirable. Indeed, the whole of this story could be interpreted as a warning
that the acquisition of knowledge leads to disaster - a kind of counter­
blast to the optimistic teaching of the book of Proverbs, for which the
acquisition of wisdom is essential to human happiness.
The human desire for immortality (presupposed by God in Gen.
3:22) is a well-known theme in Mesopotamian literature. These texts lay
stress on its inevitable frustration, and sometimes explicitly or implicitly
'
counsel contentment with one's lot in the present life (see especially the
Epic of Gilgamesh, Tablet XI, iii; ANET, 90). In this story the hero Gil­
gamesh actually secures the plant which confers eternal life, but it is then
stolen from him by a snake, so incidentally accounting for the belief that
snakes do possess immortality (the parallels with Gen. 2-3 are easy to see) .
Another story of failure to attain immortality, this time through dissension
among the gods, is that of Adapa, who is offered the bread and water . of
life by the sky-god Anu but refuses it becaµse he has been warned by the
god of knowledge and wisdom not to accept it. In the Genesis story the
supreme god has no rivals, but himself takes decisive measures to exclude
mankind from the possibility of securing immortality.
The snake in this story is an enigmatic figure. It never appears again
in the Old Testament. Westermann, probably rightly, sees this motif of a
talking snake as a "fairy tale trait," probably very ancient, of which the
author made use ( Genesis 1-1 1, 238) . Since it is clearly stated in 3 : 1 that
the snake was simply one of the · creatures that God had made, there is no
justific :ition for seeing it either as a supernatural enemy of God - the
Devil or Satan - or as some kind of "inner voice" within the woman
questioning God's intentions and urging her to do evil. It is best to see it

44
The "Primeval History " (Genesis 1-1 1)

as the text itself appears to do, as a creature, endowed by God with unusual
intelligence and using this to oppose his wish. But it is pointless to expect
precise logic here. If the snake was introduced into the story to account
for the action of the woman, there is still nothing to account for the
behavior of the snake, which was equally God's creature. (The dialogue
between the snake and the woman, however - the first example of such
a conversation in the Bible - is brilliantly achieved.)
However, snakes played a significant part in the mythologies and
religious practices of the . ancient Near East, and it was probably for this
reason that the snake motif was introduced into this story. Snakes were
objects both of fear and of worship, and Israel seems to have been no
exception in this respect. The story in Num. 21 :6-9, in which Moses at
God's command set up a bronze snake in the wilderness to act as an
antidote against the bites of poisonous snakes, illustrates both aspects of
this attitude. The action of King Hezekiah in smashing the bronze snake
called Nehushtan, specifically stated to be the one made by Moses, to
which sacrifices had traditionally been offered in the Jerusalem temple
(2 Kgs. 1 8:4) indicates the hold that this object had had among the
Judeans. The snake in Gen. 3 may be a reflection of the abhorrenee in
which this form of idolatry was held in later times.

Genesis 6:1-9:29

Stories of a great flood sent in primeval times by gods to destroy mankind


followed by some for:m of new creation are so common to so many peoples
in different parts of the world, between whom no kind of historical contact
seems possible, that the notion seems almost to be a universal feature of
the human imagination. The Flood story of Genesis is a clear example of
a type whicl,i was characteristic of Pie Mesopotamian world. The two
extant literary accounts which it most closely resembles are Atra-gasls (see
the translation by Wilfred G. Lambert . and Allan R Millard) and Tablet
..

XI of the Epic of Gilgamesh (ANET, 93-95) . Gilgamesh is extant in more


than one version. The Babylonian text, translated in ANET .and first
announced in 1 872 by George Smith, caused a sensation because of its
astonishing resemblance in detail to the Genesis story. According to Lam­
bert and Millard (p, 1 1) , it was largely derived from the account in
Atra-gasls. Tablet XI of Gilgamesh is not connected with a creation story
and appears to have been borrowed from an earlier text. Atra-gasis, though

45
INTRODUCTION TO THE PENTATEUCH

in its fragmentary state it lacks some of the details (especially the sending
out of birds to discover whether the water has receded) , is closer to Genesis
in that it contains an account of the creation of mankind from clay before
proceeding to the story of the Flood.
It has long been pointed out that Gen. 6-9 contains a number of
details such as the chronology of the Flood and the numbers of animals
taken into the ark which are in contradiction. Attempts to reconcile these,
however ingenious, can hardly be convincing. It is clear that more than
one version have been combined. The question is, at what time and by
whom the combination was carried out. The text cannot be separated into
two complete versions. If the story as we have it is rigidly divided into
two sources, one of these would obviously have been only partially pre­
served; there is, for example, only one account of the embarkation in the
ark, without which there can be no story. Rather than two written sources,
it is then the author himself - who probably knew a number of versions
from which he could choose - who has spliced two of them together
without concerning himself with all the details. We have observed the use
of the same method in his treatment of chs. 2-3 .
As has already been noted, in Genesis the Flood story i s the climax
of a sequence which begins with creation and ends, after total disaster for
mankind, on a positive note with the renewal of mankind through Noah
and his sons. This renewal of mankind is not to be found either in the
Epic of Gilgamesh or Atra-pasis. In Gilgamesh the Flood is only an episode.
The story is told by Utnapishtim, who alone is saved; but far from
becoming the ancestor of new humanity, Utnapishtim is banished to a
place "far away, at the mouth of the rivers, " where he lives alone, and to
which Gilgamesh travels in a vain attempt to obtain for himself that
immortality. Atra-basis, like Gen. 6-9, is also part of a sequence of events,
but one which is quite different from that of Genesis. It begins with a
revolt of the gods, who are overburdened with work; and with the creation
of mankind so that they may be given the work to do in place of the gods.
The occasion of the Flood is the tremendous noise made by mankind,
who have become numerous and disturb the sleep of the gods, who then
attempt unsuccessfully to reduce the human population by sending plague,
famine, and drought on the earth before determining to destroy mankind
altogether by a Flood. The man Atra-pasis alone survives; the end of the
accou11 � is missing. These differences from the Genesis story, together with
the fact that in these nonbiblical versions there is constant quarrelling
among the gods, who attempt to frustrate one another's activities, bring

46
The "Primeval History " (Genesis 1-1 1)

out both the simplicity and the theological distinctiveness of the account
in Genesis.
The genealogical lists in these chapters comprise a very substantial
part of the whole. They are more than simply links in a chain spanning
the period from Adam to Abraham. Like so much of the material, they
belong to a Near Eastern tradition which included theogonies - lists
showing the genealogical relationships between the gods - but also lists
of kings, · including kings said to have reigned "before the Flood" or who
"lived in tents" (for the king lists, see ANET, 265-66, Sumerian; 564-66,
Assyrian; 27 1 -74, Babylonian) . However, the early parts of these lists,
covering the primeval period, are not genealogical but are simply names
of successive kings. The lists go down to historical times. Recent study of
the subject has begun to take into account also oral genealogies transmitted
within modern tribal societies; but much work remains to be done (see
Robert R. Wilson, "The Old Testament Genealogies in Recent Research") .
The lists in Gen. 1 - 1 1 are not entirely consistent. There are, for
example, two different genealogies of Adam, one through Cain (4: 1 7f£)
and one through Seth (5 :3ff.) . Both lists, however, include some of the
same or similar names. Both the non-biblical and modern oral genealogies
show similar inconsistencies. It is clear that such genealogies are fluid; in
the course of time they have come to serve different purposes, often
political or concerned with relationships between tribes, and have been
altered accordingly. In the case of Gen. 1 - 1 1 , the genealogy of Cain seems
to be related to the theme of a general human deterioration which was
the cause of the Flood. It ends with the sinister figure of Lamech (4:23-24) .
On the contrary, the genealogy of Seth, which includes Enoch (who
"walked with God, " 5 :24) , ends with the savior of the human race, Noah,
and his family. The author was of course aware of the inconsistencies, but
used the lists for different purposes.

For Further Reading

Blenkinsopp, Joseph. The Pentateuch: An Introduction to the First Five Books


ofthe Bible. New York: Doubleday and London: SCM, 1 992, ch 3 .
Brandon, S . G. F. Creation Legends of the Ancient Near East. London:
Hodder & Stoughton, 1 963.
Frankfort, Henri, et al. The IntellectualAdventure ofAncient Man. Chicago:
University of Chicago Press, 1 946. Repr. 1 977.

47
INTRODUCTION TO THE PENTATEUCH

Heidel, Alexander. The Babylonian Genesis: The Story ofthe Creation. 2nd
ed. Chicago: University of Chicago Press, 1 963.
Johnson, Marshall D. The Purpose ofthe Biblical Genealogies. SNTS Mon­
ograph 8. Cambridge: Cambridge University Press, 1 969. 2nd ed.,
1 988.
Lambert, Wilfred G., and Millard, Allan R. Atra-baszs: The Babylonian
Story ofthe Flood OJeford: Clarendon, 1 969 .
von Rad, Gerhard. "The Form-Critical Problem of the Hexateuch," in
The Problem of the Hexateuch and Other Essays. Edinburgh: Oliver
& Boyd, 1 966, 1 -78. Repr. Philadelphia: Fortress and London:
SCM, 1 984. (First published in German, 1 938.)
--- . Genesis. 3rd ed. OTL. Philadelphia: Westminster and London:
SCM, 1 972. (9th German ed., 1 972.)
Rendtorff, Rolf The Old Testament: An Introduction. London: SCM, 1 98 5 ,
and Philadelphia: Fortress, 1 986. (First published in German, 1 983.)
Rogerson, John. Genesis 1-1 1. OTG. Sheffield: JSOT Press, 1 99 1 .
Skinner, John. A Critical and Exegetical Commentary on Genesis. 2nd ed.
ICC. Edinburgh: T. & T. Clark, 1 930.
Van Seters, John. Abraham in History and Tradition. New Haven: Yale
University Press, 1 975.
Wenham, Gordon J. Genesis 1-15. WBC 1. Waco: Word, 1 987.
Westermann, Claus. Creation. Philadelphia: Fortress and London: SPCK,
1 974. (First published in German, 1 97 1.)
--- .. Genesis: A Practical Commentary. Grand Rapids: Wm. B. Eerd­
mans, 1 987. (First published in Dutch, 1 986.)
---.. Genesis 1-1 1. Minneapolis: Augsburg and London: SPCK, 1 984.
(Translated from the 2nd German ed. in the Biblischer Kommentar
series, 1 976.)
---. The Genesis Accounts of Creation. Philadelphia: Fortress, 1 964.
(Translated from the 2nd German ed., 1 96 1 .)
Wilson, Robert R. "The Old Testament Genealogies in Recent Research,"
]BL 94 (1 975) : 1 69- 1 8 9 .

48
CHAPTER 4

The History of the Patriarchs


(Genesis 12-50)

· HE TI:-1E SPAN of these chap�ers is com�aratively short. They co �prise


Tthe history of four generations of a smgle family:_ Abraham, his son
Isaac, his grandson Jacob, and his twelve great-grandsons. These were of
vital interest to the later Israelites because they believed them to be their
own ancestors; they are often referred to simply as "the fathers. " Abraham
was the father of the nation, and his great-grandsons the founders of the
twelve tribes, of which the nation was believed to consist. This patriarchal
history is, however, only the first part of the story of the origins of the
nation: it looks forward to the events recounted in the later books. Thus
while the lives of Abraham, Isaac, and Jacob were passed mainly in Canaan,
the story of Joseph with which the book of Genesis comes to an end (chs.
37-50) takes the whole family to live in Egypt. This sets the stage for the
stories of the oppression of their descendants by the Egyptians and of the
Exodus which are the subjects of the first_ chapters of the book of Exodus.
Where did the author obtain the material for his history? It is an
astonishing fact that the preexilic parts of the Old Testament make no
mention at all of the incidents connected with Abraham, Isaac, or Jacob
j with the sole exception of Hos. 1 2:3-4, 1 2, where the prophet shows
familiarity with certain episodes in the life of Jacob. Although the names
Isaac and Jacob were sometimes used in poetical texts (not prose ones) as
designations of northern Israel, even bare references to them as individuals
in texts that are certainly preexilic are extremely infrequent. It is only from
the time of the Exile in the sixth century that we have texts (Isa. 5 1 :2;
Ezek. 33:24) that refer to Abraham as an individual, in connection with

49
INTRODUCTION TO THE PENTATEUCH

problems faced by the exiles or by those left in the land after the Babylonian
conquest. This suggests that the stories about the patriarchs which the
author of Genesis used in his history may for the most part be no older
than that period. There is no evidence that they were current over a long
period before then, and the fact that they are mentioned neither in the
historical books nor by the preexilic prophets suggests the contrary.
At this point it is appropriate to consider the nature of historiography
as it was understood in the ancient world. The ideal (never in fact attained)
of recording, as far as this is possible, the "brute facts" or "what actually
happened" in the past is a very modern one. Not only did an ancient
historian set out with a preconceived aim - political, religious, moral,
educative - in mind, which went beyond the recording of bare facts; the
historian also considered it to be part of his function to arrange, embellish,
and embroider the material to make it more attractive and exciting to the
reader or in some way more palatable or moral, or to use it to make a
religious point. This would necessitate at least a degree of invention, which
in modern terms would be called fiction.
As is generally recognized, a large part of the narrative books of the
Old Testament are literary fiction. This is true of the prologue and epilogue
of the book of Job, the books of Ruth, Jonah, Esther, Daniel 1 -6, and
large parts of the books of Chronicles and probably of other narratives as
well. Some of these stories are expansions of earlier narratives; this is the
case, for example, with the additional material in Chronicles that expands
the story of David. In other cases a whole story appears to be pure fiction.
The book of Jonah takes a name about which nothing at all is known
apart from a single verse in 2 Kgs. 1 4:25, and uses this name as a peg on
which to hang an entire narrative with a religious message - an early
example of what came later to be known as midrash. In the cases of Ruth,
Esther, and Daniel there was, as far as we know, no such known name on
which to hang these completely fictitious tales (though the name Daniel
appears in Ezra 8 :2 in a list of returned exiles) . The parables of Jesus are
also of course fiction. The story of Job is of particular relevance to our
present inquiry because Job, who is universally recognized to be a wholly
fictitious character, is portrayed as a patriarch similar to the Abraham of
Genesis.
The fact that Abraham, Isaac, and Jacob have become such outstand­
ing figures in both Judaism and Christianity may be due more to the skill
of the narrator than to any ancient long-standing traditions that preceded
him. That the figures of these patriarchs were originally unrelated to one

50
The History of the Patriarchs (Genesis 12-50)

another and have been used to create a fictitious "family history" is widely
admitted. There is also a possibility that some of the stories attributed to
one or other of these figures may have originally been told about some
other legendary figures altogether. Such schematization smacks of the
historical novelist rather than of a faithful transmitter of ancient traditions.
If "patriarchal" stories like that of the prose narrative of the book of Job
were circulating in his time, the author would have had material at hand
which he could combine and expand, using them to create an account of
Israel's ancestors with an appropriate religious message - a method
analogous to that employed by the early Greek historians.
The view stated above, however, is not one that has commanded
general assent. A series of scholars has continued to follow Hermann
Gunkel's theory of Sagen or small units of an early period gradually
combined to form the written sources J, (E), and P. This is true of Albrecht
Alt ("The God of the Fathers," 1 929, English translation 1 967) , Gerhard
von Rad, and also of Claus Westermann.
Rolf Rendtorff and John Van Seters, in studies both published in
1 975, rejected the "classical" source theory, but continued to support the
notion of a series of stages leading to the final text of the patriarchal stories.
In his work on the composition of the Pentateuch, Rendtorff did not
clearly indicate at what period he believed the stories to have originated,
though in a later ( 1 982) article on the composition of one story (that of
Jacob at Bethel, Gen. 28: 1 0-22) , he supported an earlier view that the core
of this story is a (presumably rather early) "cult etiology" relating to the
foundation of the sanctuary at Bethel, which subsequently underwent a
series of changes and additions. Rendtorff's pupil, Erhard Blum, in a work
wholly concerned with the patriarchal history ( 1 984) , added further pre­
cision: he saw no evidence to suggest that any of the stories predates the
early monarchy, but he also believed that they had a fairly long and
complex history. They were first combined in two recensions, a "history
of Abraham" composed in Judah, and a "history of Jacob" from the
northern kingdom. These were then combined in Judah in a "patriarchal
history" which underwent two recensions, the first compiled before the
Exile and the second in the late sixth century, before being incorporated
into a "Deuteronomistic Pentateuch."
The view of Van Seters is not dissimilar to this. He postulated an
exilic "J," but this had been preceded by an earlier written work. Van
Seters also saw no evidence that the stories had been preserved from a
"patriarchal age." He held that it is extremely difficult to distinguish

51
INTRODUCTION TO THE PENTATEUCH

between oral and written sources, but also pointed out that oral sources
are not necessarily either early or preliterate. The crucial questions seem
to be whether it is possible for traditions to be preserved orally for several
hundred years, and whether the text as it stands can or cannot be accounted
for as the work of a single author who incorporated his own theologicil
comments and interpretations into material which was in some sense
traditional.
It cannot be said that in combining these stories the author succeeded
in producing an entirely coherent acc<? unt of the lives of the patriarchs. A
number of stories have been · placed in inconsequent positions. This is
particularly true of the Abraham stories (chs. 1 2-25), though the most
incongruous of the dislocations occurs with the story ofJudah and Tamar
(ch. 38) , which interrupts the otherwise well-constructed aecount of the
life of Joseph (chs. 37-50) . Nevertheless, in general an attempt has been
made to present the stories in "chronological " sequence. The patriarchal
history has been arranged in three main, more or less clearly defined, parts:
the stories of Abraham ( 1 2 : 1-25 : 1 1 ) , Jacob (25 :2 1 -35 :29), and Joseph
(chs. 37-50) . The story of Jacob is itself clearly divided into three parts,
corresponding to three stages in his life: the account of his relations with
his immediate family, especially with his brother Esau (chs. 25-27, 32-35),
is bisected by the story ofJacob's j ourneys and his relations with his uncle
Laban (chs. 28-3 1 ) ;
I t can hardly b e said that there i s an independent "story'' o f Isaac.
The events of his life are narrated partly in the story o f Abraham and
partly in that of Jacob. As would be expected in a "family'' history, there
is also some overlap between all these "histories. " Thus the · death of
Abraham (25 :8) takes place only after the account of Isaac's marriage;
Isaac's death (35 :29) comes at the end of the story of Jacob; and Jacob's
death (49: 33) almost at the end of the story ofJoseph. The death ofJoseph
is recorded in the final v�rse of the book (50:26) .
The patriarchal histo ry is interspersed with genealogies: of Nahor
(22:20-24), of Abraham by his second wife (25: 1 -6), of Ishmael (25 : 1 2-
1 6), of Jacob (35:22-26) , of Esau (36: 1 -43, including also a list of kings
of Edom) , and again of Jacob (46:8-27) . The m ain purpose of these
genealogies is to claim Abraham and his family as ancestors of other
peoples: Aramaeans, Moabites, Ammonites, Arabians, Ishmaelites, Hit­
tites. T he final list, however, prepares for the events of the book of Exodus
by naming the descendants of Jacob who left Canaan to reside in Egypt.
The main theme of the patriarchal history is set out from the very

52
The History ofthe Patriarchs (Genesis 12-50)

start ( 1 2: 1 -3) . There Abram, later to be named Abraham, is commanded


by God to leave his country of residence and move to a land to be shown
to him, where he will become the ancestor of a great and famous nation,
specially blessed by God and conferring blessing on other peoples. These
first promises are repeated in fuller detail later to Abraham himself ( 1 5 :4,
7, 1 8-2 1 ; 1 7:4-8; 22: 1 7- 1 8) and then to Isaac (26:2-5, 24) and Jacob
(28 : 1 3- 1 5; 3 5 : 1 1 - 1 2); but the essentials are already present in 1 2 : 1 -3 .
Brevard Childs has pertinently remarked (Introduction to the Old Testament
as Scripture, 1 50) that the promises as a whole "relate, above all, to posterity
and land." The direction to Abraham to move to a particular country may
be taken to imply his future possession of it; the promise that he will
become a great nation and that his name will be great equally necessarily
implies that he will father an heir and that this fertility will be continued
in later generations. The promise of blessing implies their material success;
and it is implied quite clearly that God will guide their fortunes in the
future. There has been much meticulous academic discussion of the var­
iations in content and form of the different passages in question, which
were thought to provide clues regarding the gradual development of the
material, but no consensus has been reached. It was part of the literary
technique of the biblical writers not only to repeat topics but to adapt
their particular expressions to suit their contexts and also to introduce
variations into them to relieve monotony and sustain the interest of the
readers (Robert Alter, The Art ofBiblical Narrative, 88- 1 1 3) .
The promise o f possession of the land of Canaan, to which Abraham
(or his father) had moved from Ur of the Chaldeans ( 1 5 :7), was in fact
never fulfilled in the course of the book of Genesis. The patriarchs are
never pictured as owners of the land. The Canaanites ( 1 2:6; 1 3 :7) and
other peoples (ch. 23) were in possession; Abraham and his family are
described as resident aliens (23:4; 35:2 l), passing through rather than
settling. The promises of possession of the land, then, were not for imme­
diate fulfillment but for the future. This is clear from 3 5 : 12, where in
renewing the promise God says, "the land that I gave to Abraham and
Isaac I will give to you, and I will give the land to your offspring after
you'' - when in fact he had not yet given possession of the land at all,
either to Abraham or to Isaac. The Hebrew verb "give" (nathan), here used
in a past tense, clearly means something more like "promise to give" in
this passage at least.
The promise of the land is expressed in different ways. In 1 5 :7 God
tells Abraham "I am Yahweh .who brought you from Ur of the Chaldeans,

53
INTRODUCTION TO THE PENTATEUCH

to give you this land to possess�" In 1 5 : 1 8, however, the Hebrew phrase


used is "to your seed," that is, to Abraham's descendants - Abraham
himself is not included here. Elsewhere, in the renewals of the promise to
Isaac (26:3) and Jacob (28 : 1 3; 35: 1 2) it is "to you and to your descendants
(after you) ." These differences of phraseology are not, however, really
material, as some scholars have argued. In 17:2 it is clear that the promise
to Abraham includes his descendants by implication, just as "I will make
you (singular) into a great nation'' in 1 2:2 cannot refer just to one man
but must inclu'de several generations of his ancestors.
There is then a constant tension and an atmosphere of suspense
throughout the patriarchal stories, highlighted also by the final words of
the book referring to the burial of Joseph (50:26) : "they e.mbalrried him,
and he was put in a coffin in Egypt' ("in Egypt" being the very last words
of the book) . The experiences of the patriarchs, despite the promises, have
led them merely from one alien land (Ur of the Chaldeans or Haran) to
another (Egypt), after four generations of wandering in the land which
they hoped to possess but did not yet possess. This postponement of the
fulfillment of the promise of the land is further emphasized in God's words
to Abraham in 1 5 : 1 3- 1 6, where he predicts that Abraham's descendants
will be not only aliens but slaves in a foreign land for four hundred years
before settling in Canaan. The author has thus created a dramatic tension
between promise and reality, leading the readers to appreciate the pre­
cariousness of the existence of the patriarchs and to wonder how and by
what means it will come about that their descendants will possess the land
- questions which would have a distinct relevance to later Israelites in
exile or, somewhat later, forced to live once more in a land which·was not
theirs. The Exodus from Egypt recounted in the following book would
perhaps give them new hope.
The other promises were partly but by no means completely fulfilled
within the compass of the book of Genesis. God made a covenant with
Abraham and his descendants ( 1 5 : 1 8; ch. 1 7) which would remain in force
for ever; and he promised to bless them (1 2:2; 22: 1 7; 26:3; 28: 14; 35 : 1 1 )
and to be always with them to guide and prosper them (26: 3) until all the
promises were fulfilled (28 : 1 5) . The covenant, the blessing - which im­
plies both material wealth and progeny, and the continuous divine pro­
tection - which preserved the existence of the family - may be said to
have come into operation immediately. That Abraham and his family
became the ancestors of many nations ( 1 7:4; 3 5 : 1 1) is also hinted at. But
the further promises of becoming a great and famous nation that would

54
The History ofthe Patriarchs (Genesis 12-50)

be victorious over enemies but also be a blessing to other nations ( 1 2 :2-3;


1 7:2-5; 22: 1 7- 1 8; 26:4; 28: 1 4; 35: 1 1 ) remained yet to be fulfilled.
The ultimate fulfillment of the promise to become a great and
numerous nation obviously depended from the start on two things: the
survival of the head of the family and his wife (or wives) in each generation,
and the successful fathering and bearing, in each generation, of a male
heir (Abraham's unenthusiastic proposal to adopt his own servant as his
heir is firmly rejected by God: 1 5:2-4) . It is the uncertainty both about
survival and about the birth of an heir, apparently despite the promises
which God has made, that constitutes the main drama of the patriarchal
history. Time and time again these are placed in danger in one way or
another, creating a dramatic suspense. Each of these situations is then
resolved by God's intervention, often by what we should call a miracle,
only to be succeeded by another; so when the reader reaches the end, with
Jacob's blessing of his twelve sons who are to become the ancestors of the
twelve tribes of Israel, the tension is succeeded by a sense of relief together
with confidence in the future. By his ingenuity in combining and adapting
a multitude of narrative fragments the author succeeded in creating an
epic which held the reader and continues to do so to this day.
The tension begins immediately after Abram's call, with his migration
to Egypt in consequence of a famine in Canaan ( 1 2 : 1 0-20), where the life
of Abram and the role of Sarai as the future mother of the heir are
endangered. Only Yahweh's intervention in afflicting Pharaoh with
plagues, revealing that Sarai is Abram's wife, saves them. The incident
results in the beginning of the fulfillment of the promise of blessing, in
that Abram emerges as a wealthy man. Then in 1 3 : 8 - 1 2 the question of
the.land begins to be settled when Lot chooses the plain ofJordan, leaving
Abram the land of Canaan as his place of residence. In ch. 14 Abram's
military success over the four kings in order to rescue Lot is probably
intended as a foreshadowing of the fulfillment of the promise to become
a great nation.
The chapters which follow are concerned with the promise of an
heir to Abram and its fulfillment despite serious impediments. The promise
is made specifically in 1 5 :4, coupled with the promise of numerous de­
scendants (v. 5). This passage is also significant because it contains one of
the few references (v. 6) to Abram's reaction to the promises: he belie".es
them and trusts God to carry them out - an attitude which. the author
probably wishes to inculcate into a much later Israel in a similar situation
of apparent helplessness. This was a remarkable act of faith in view of

55
INTRODUCTION TO THE PENTATEUCH

Sarai's persistent barrenness ( 1 6: 1 ) and advanced years, and also of Abram's


own old age ( 1 7: 1 7); and God acknowledged this act of faith and gave
Abram credit for it (literally, "counted it to him as righteousness") . Ch.
1 6 tells the story of Hagar and the birth of Ishmael; Abram's ability to
bear children is thus demonstrated, but it is not in this way that he is to
be granted an heir. He is told by God ( 17:20-2 1 ) that Ishmael, whose
mother is Egyptian ( 1 6 : 1 ) , is not to be his heir: the heir is to be the child
of both Abraham and Sarah (both now renamed, 1 7 : 5 , 1 5) . Sarah's infer­
tility is mysteriously overcome, despite her skepticism, by the arrival of
three visitors, who announce the miraculous birth ( 1 8 : 1 - 1 5) . But now
once again an impediment occurs. Once again Abraham and Sarah migrate
to another country and Sarah is taken into the harem of the Philistine
Abimelech, king of Gerar (20 : 1 - 1 8); and once again God intervenes by
telling the king that she is Abraham's wife. Finally all this suspense is
resolved: at last their son Isaac is born, and the continuation of the family
appears to be assured (2 1 : 1 -7) .
Even now, however, a fresh danger appears: Isaac's life is now put in
jeopardy, in the most dramatic of all these incidents, and by God himself.
Abraham is commanded by God to sacrifice his son Isaac. He obeys, and
is at the very point of sacrificing Isaac when the angel of God stops him
in the nick of time. Although it is then explained that God's intention
was only to test Abraham's readiness to sacrifice to God that which he most
dearly loved, the story is told brilliantly in a way that creates the maximum
suspense.
Irt the generations that follow the same theme continues to be central.
The old problem of ensuring that the family succession should be preserved
from foreign taint which had arisen when Ishmael was Abraham's only
. son is still to the fore. The first problem now to be dealt with, then, was
the fear that Isaac might marry a Canaanite wife. Ch. 24 describes how
this danger was averted when Abraham's servant was divinely guided to
the house of Isaac's cousin Bethuel in Aram-naharaim (in Mesopotamia)
and was able to arrange a marriage between Isaac and Bethuel's daughter
Rebekah.
In the next generation the succession to the promises was again in
doubt. The rivalry between Isaac's twin sons Esau and Jacob led to a threat
to kill Jacob (27:4 1 ) , and Jacob had to flee for his life. At Bethel, where
he spent the night while on his journey, it was made plain to Jacob that
he and not his elder twin was to inherit the promises made to Abraham
and Isaac (28 : 1 3- 1 5); and, like his father, he wished to marry his cousin

56
The History ofthe Patriarchs (Genesis 12-50)

(Rachel) . Nevertheless, Jacob had to pass through a series of dangers -


from his father-in-law Laban (ch. 3 1 ), from a mysterious adversary who
wrestled with him (32:22-32), from Esau (ch. 33) , and from the She­
chemites whom his sons had offended (ch. 34) .
1 In the story ofJoseph (chs. 37-50), which differs in many ways from -
those of Abraham and Jacob, the same theme of dire threat to the survival
of the heirs subsequently miraculously averted by God continues unabated.
The central figure in these chapters is Joseph, who as the eleventh son of
Jacob was far from being the expected heir; but, as it becomes clear, Joseph's
survival proves to be esse.ntial for that of his brothers, because it is he who,
·

himself miraculously saved, is to be the savior of their lives.


This story begins (ch. 37) with the young Joseph, hated by his brothers,
who plot his death. He is saved by the persuasion of his brother Judah, but
nevertheless sold to passing Ishmaelites and taken as a slave to Egypt. In chs.
39-4 1 we follow the ups and downs ofJoseph's extraordinary career: sold to
Potiphar, the captain of Pharaoh's guard, as a slave (39: 1); promoted to
overseer of Potiphar's household (v. 6); accused by Potiphar's wife of at­
tempted rape (v. 1 7) and imprisoned (v. 20) ; freed from prison and brought
to Pharaoh's notice by a fellow prisoner who remembered his ability to
interpret dreams (4 1 :9- 14); and forthwith promoted to the highest office in
the Egyptian state, governing the whole country (4 1 :38-45) .
From his position of unlimited power Joseph is instrumental in
saving the lives of his father and brothers and all their children and
grandchildren from certain starvation by inviting them to migrate from
Canaan to Egypt, where alone, thanks to his own efforts, there was plenty
of food (45 :9-46:27) . Later, however (45: 5-8; 50: 1 9-20) , Joseph specifi­
cally attributed his survival and good fortune to God, who was the real
controller of all these events and had sent him to Egypt "in order to
preserve the lives of many people."
In his dying words (50:24) Joseph reminded his brothers of the
promise of the land, not yet achieved: the family had been saved yet again,
this time from famine, but their residence in Egypt was not the final stage,
not what God intended for them. "I am about to die; but God will surely
come to you, and bring you up out of this land to the land that he swore
to Abraham, to Isaac, and to Jacob." This theme, then, is never very far
from the center of the narrative of Gen. 1 2-50.
Of the three main patriarchal "histories" - of Abraham, Jacob, and
Joseph - it has already been remarked that the account of Abraham is the
least integrated. It hardly forms a consecutive "history" at all. Each incident

57
INTRODUCTION TO THE PENTATEUCH

is to a large extent independent and self-contained, and in several - notably


the stories of Hagar and Ishmael (chs. 1 6 and 2 1 ) and of Lot at Sodom (ch. ·
1 9) - Abraham either does not appear at all or is only a peripheral figure.
However, some stories have the character of "set pieces" written to express
the "theology'' of the author. This is true of the theological discussion in
1 8: 1 7-33, which has no bearing on the fortunes of Abraham; ch. 22, though
now integrated into the main theme of the promises by its relevance to the
survival of the heir, is also a theological creation in its present form. A further
feature of these chapters is that some of the incidents described could well
have been originally told of someone other than Abraham, perhaps even of
unnamed persons, in the manner of a legend or even a fairy tale (German
Marchen). All the stories are short - there is no real continuity except what
has been imposed on them by the author.
The history of Jacob is somewhat more integrated. The incidents are
on the whole told at greater length, and the adventures of Jacob form a
· more continuous narrative plot. As has already been noted, two distinct
plots have been combin.ed here: that of Jacob and his relations with his
brother Esau (25 :2 1 -28 :22, resumed in chs. 32-33) and that of Jacob and
his relations with his uncle Laban (chs. 29-3 1 ) ; but the story is told as
part of a single account of Jacob's life. It begins with the births of Jacob
and his twin brother Esau and their rivalry, leading to Esau's threat to
Jacob's life and to the flight of Jacob, and ends with the return of Jacob
and his reconciliation with his brother: a "classic" tale of the unsatisfactory
son who made good. The intervening years of Jacob's life are filled with a
mixture of good and evil fortune: his marriages and the birth of his children
on the one hand and his intolerable servitude to Laban on the other, from
which he eventually escapes with his family. The tale is interspersed with
incidents which confirm Jacob's position as the heir to the promises. There
is thus a kind of jerkiness in the narrative at some points, but Jacob's
personality is depicted in a way which is not the case in the history of
Abraham. The modern reader at leastwill find an interest in the depiction
of the development of that personality, from that of the brash and tricky
young man to that of the responsible patriarch.
When we come to the final section of Genesis, the history ofJoseph,
we are immediately struck by the fundamental difference between it and
the histories of Abraham and Jacob. The Joseph story belongs to a
completely different literary genre. Apart from ch. 38, which is completely
unrelated to its context and has for some reason been inserted into it, this
is a perfectly integrated and fully structured narrative about the life of

58
The History ofthe Patriarchs (Genesis 12-50)

Joseph which at least in its present form is the creation of the Pentateuchal
author. Von Rad characterizes it as "a novel through and through," though
"short story" probably better represents van Rad's meaning ("The Joseph
Narrative,"
I
292) . It has many of the characteristics which we look for in
a novel: unity of plot, suspense, dramatic irony, the depiction of "involved
psychological situations" (van Rad) , characterization of the hero, changes
of tempo to suit particular situations. It is clearly an independent work of
literary artistry, a piece of considerable length and not the result of com­
bining a series of shorter stories. At the same time, in its present form and
in its present position within the Pentateuch it forms part of the wider
history, performing the function of a link between the earlier hu; tories of
Abraham's descendants and the next stage - the Exodus from Egypt -
skillfully contriving a plausible reason for the presence of the tribal ances­
tors in Egypt on which the larger plot depends.
But it also seems evident that the history of Joseph was intended to
serve some didactic purpose beyond this. Simply in order to achieve this
.
result it cannot have been necessary to devote fourteen chapters of the
present text to such an elaborate story -with its descriptions of Egyptian
court life, the administrative problems of the Egyptian economy and their
solution, the long, drawn-out account ofJoseph's treatment of his brothers
when they traveled to Egypt to buy grain, and his ihterpretation of the
dreams of Pharaoh, the butler, and the baker.
Essentially the plot, the story of a young man who through his own
ability rises from obscurity to unheard-of power and wealth, is one com­
monly found in folktales (e.g., Dick Whittington) . But it is also an example
of a particular Jewish variation of this theme: that of the Jewish captive
at the court of the foreign king, who turns the tables on his conquerors
and is awarded the highest position in the kingdom; This is also the theme
of the stories about Daniel in Dan. 1 -6 and of the book of Esther. Joseph
and Daniel also have in common their unsurpassed ability to interpret
dreams, by means of which they achieve their success, and also their
acknowledgment that their success is due not to their own ability but to
God. Joseph, then, is presented in these chapters as a hero, but one who
gives the credit to God. Once more there is a lesson here for later Jews
living in subservience to a foreign conqueror.
Von Rad, who made a special study of these chapters, saw the story
as an example of wisdom literature. Joseph, he argued, was the model of
the accomplished government administrator or court scribe. He possessed
all the virtues which, according to the wisdom teaching both of Egypt and

59
INTRODUCTION TO THE PENTATEUCH

Israel (the latter exemplified in the book of Proverbs) , would lead to success
in that profession, especially ability to give good advice at the right mo­
ment, modesty, learning, courtesy, and self-control, in addition to the
recognition that all success was dependent on the will of God. The Joseph
story was influenced by Egyptian models and was one of the literary
consequences of the cultural "enlightenment" which Israel derived from
Egypt in the reign of Solomon. It was no coincidence, von Rad believed,
that the story has been compared with the Egyptian Tale ofthe. Two Brothers
(ANET, 23-25). Its marked interest in foreign, specifically Egyptian, life
and customs is also readily explicable in terms of this theory.
Von Rad's thesis was at first widely accepted, but it has recently been
subjected to serious criticism. Apart from the widespread rejection of the
theory of a "Solomonic enlightenment" and the opinion of competent
Egyptologists that the story shows little knowledge of Egypt - and cer­
tainly not of the period when von Rad supposed it to have been written
- it has been pointed out that its portrait ofJoseph does not in fact closely
correspond to the scribal ideal. Joseph was not born into a scribal family
as would normally have been the case for scribes. Nothing is said of his
education; rather, his ability seems to have been innate. It is far from being
the case that he was self-controlled: in 45: 1 -2 he broke down completely
and wept so loudly that "the Egyptians heard it, and the household of
Pharaoh heard it." Moreover, his telling of his early dreams to his family
in ch. 37 reads like arrogance rather than modesty, although it may have
been the intention of the author to portray the subsequent development
of Joseph's character. There is little doubt that Joseph was intended as a
model to be in some way imitated; but it must be doubted that he is
represented as a model scribe.
The difference in genre between the story of Joseph and the previous
"histories" suggests that, rather than piecing together and rewriting a mass
of fairly brief but originally independent pieces and adding a few more of
his own, the author has here composed an entirely original story. He has
already mentioned the birth of Joseph in the course of the history of Jacob
(30:22-24), and Joseph's name occurs again several times in the. later part
of that history, although nothing further is recorded of him there. Joseph
appears again, however, in the blessings ofJacob (49 :22-26) , which appear
to be a kind of delayed epilogue to the history of Jacob; but what is said
of hirn there has nothing at all in common with the character and life of
Joseph otherwise depicted in chs. 37-50 except for the statement in 49:26
that he was "set apart from his brothers" (or possibly "was prince over his

60
The History ofthe Patriarchs (Genesis 12-50)

brothers") . It may be this phrase that the author took as his starting point
for his story, though he may have. known of a tradition ab dut the migration
of Jacob and his family to Egypt.
, To conclude, in our discussion of Gen. 1 2-50 we have seen how the
author employed a variety of methods of composition. There is, first of all ,
reason to suppose that it was he who "created" the family history of the
patriarchs. This is made probable by the fact that outside the Pentateuch the
extant preexilic literature shows virtually no knowledge of them as individu­
als or of the events associated with them, although their names sometimes
appear as designations of the nation or of particular segments of it. It is
probable, therefore, that various stories about legendary persons, perhaps
originally attached to particular places or regions of Palestine, have been
linked together by the Pentateuchal author by representing these persons as
successive generations of a single family which branched out in a fourth
generation into the "twelve tribes of Israel," so creating an etiology of the
origin of the nation which later became accepted as the national tradition.
This was a considerable achievement. Admittedly the . stories of
Abraham and Jacob do not read entirely smoothly: in places they remain
episodic, and also some inconsistencies remain. There the rewriting may
have been little more than retouching, although an impression of continu­
ity - of a "biography" - has been achieved. Some of the longer episodes,
however - notably the battle against the kings (ch. 1 4), the announce­
ment of Isaac's birth and the dialogue between Abraham and God (ch.
1 8) , the story of the near-sacrifice of lsaac (ch. 22) , the journey to find a
wife for Isaac (ch. 24) , and the deception of lsaac (ch. 27) - clearly go
beyond the genre of the folktale: they are the work of an accomplished
author. This is pre-eminently true of the story of Joseph, where literary
imagination was given free rein and the literary art displayed to perfection.

For Furth.er Reading

Blenkinsopp, Joseph. The Pentateuch: An Introduction to the First Five Books


ofthe Bible. New York: Doubleday and London: SCM, 1 992, ch. 4.
Davies, Philip R. In Search of '54-ncient -Israel " JSOT Supplement 148.
Sheffield: JSOT Press, 1 992.
Gunkel, Hermann. The Legends of Genesis. New York: Schocken, 1 964.
(Translated from the Introduction to his first [German] commentary
on Genesis, 1 90 1 .)

61
INTRODUCTION TO THE PENTATEUCH

McKane, William. Studies in the Patriarchal Narratives. Edinburgh: Hand­


sel, 1 979.
Moberly, R. W L. Genesis 12-50. OTG. Sheffield: JSOT Press, 1 992.
Noth, Martin. A History ofPentateuchal Traditions. Englewood Cliffs, N.J . :
Prentice-Hall, 1 972. Repr. Atlanta: Scholars Press, 1 989. (First pub­
lished in German, 1 948.)
von Rad, Gerhard. Genesis. 3rd ed. OTL. Philadelphia: Westminster and
London: SCM, 1 972. (9th German ed. , 1 972.)
--- . "The Joseph Narrative and Ancient Wisdom," in The Problem of
the Hexateuch and Other Essays. Edinburgh: Oliver & Boyd, 1 966,
292-300. Repr. Philadelphia: Fortress and London: SCM, 1 984.
(First published in German, 1 953.)
Redford, Donald B . A Study ofthe Biblical Story ofjoseph. VTS 20. Leiden:
Brill, 1 970.
Rendtorff, Rolf. The Old Testament: An Introduction. London: SCM, 1 98 5,
and Philadelphia: Fortress, 1 986. (First published in German, 1 983.)
Van Seters, John. Abraham in History and Tradition. New Haven: Yale
University Press, 1 975 .
--- . Prologue to History: The Ycihwist as Historian in Genesis. Louisville:
Westminster/John Knox, 1 992.
Westermann, Claus. Genesis 12-36. Minneapolis: Augsburg and London:
SPCK, 1 98 5 . (First published in German, 1 98 1 .)
---. Genesis 37-50. Minneapolis: Augsburg and London: SPCK, 1 986.
(First published in German, 1 982.)
Whybray, R. Norman. The Making of the Pentateuch. ]SOT Supplement
5 3 . Sheffield: JSOT Press, 1987.

See also other commentaries listed above.

62
CHAPTER S

Exodus, Leviticus, Num bers:


Narratives

"\VfHEN THE STORY of Abraham's family is resumed at the beginning


W of the book of Exodus they have become a nation. The first seven
verses of the book state that the descendants of the seventy persons who
were living as a family in Egypt in the time of Joseph (Gen. 46:26-27;
Exod. 1 : 5) had grown in number in the course of 430 years (Exod. 1 2 :40;
400 years according to Gen. 1 5 : 1 3) to a numerous people who "filled the
land" of Egypt, in accordance with . the promise made to Abraham of ·
numerous progeny "like the sand of the sea." We are told nothing of what
happened during that period, though it is implied that 'they remained
welcome immigrants, yet they continued to live apart from the Egyptians.
Exodus, the:n , marks an entirely new stage in their history; and the author
makes this quite clear in that the phrase "the sons of lsrael" (i.e., of Jacob)
- which in Exod. 1 : 1 has a literal meaning - has already from v. 7 on
ac;quired the new meaning of "the Israelites. "
Exodus, Leviticus, and Numbers belong together: there are n o nat­
ural divisions between them. The division of the Pentateuch into five
books, though ancient, is not original. It is probably due partly to a
somewhat rough assessment of their co 19ents and partly to practical con­
siderations which limited the quantity of material that could conveniently
be included on a single scroll. Of the five books, only Genesis and Deuter­
onomy have clearly distinct characters of their own.
't_hese thre� books �arry the story - novv the sto ry of the people of
: . . --.-
Israel ___trom-ti1eir s ettled residence ill Egypt to "the plains of Moab by
-
the Jordan at Jerich(;". o n: -t.he e�t side of the �i�er (Num . 3 6 : 1 3) � There

63
INTRODUCTION TO THE PENTATEUCH

they were poised to enter the Promised Land at last, an event delayed only
by a lengthy series of discourses, together with a song and a blessing
pronounced by Moses, and by Moses' death - all of which are the subject
of Deuteronomy.
The distance between Egypt and Palestine, especially from the region
where the Israelites were settled (apparently in the east of the Nile delta
not far from the frontier), is not great. These are adjacent territories, and
there was in ancient times a good military road which passed through
them. Yet we are told that the Israelites took forty years to arrive at their
. destination! It is stated in Exod. 13: 1 7- 1 8 that God deliberately led them
by a roundabout way rather than by way of "the land of the Philistines, "
which would have been nearer. I n Num. 14:26-35; 32: 1 0- 1 3 the reason
for the forty years' journey is given: it was God's intention that this should
be a punishment for disobedience and rebelliousness committed after the
people left Egypt. Forty years is a round figure for a generation, and God
intended that during that period none of the men then aged twenty years
or more (with the two exceptions of Caleb and Joshua) should be permitted
to enter the Promised Land, but that they should all die in the wilderness.
Eventually the Israelites entered the land under Joshua, not from the
Egyptian side but from the east, through Moab and across the Jordan.
These three books may in fact be divided in terms of location into
two main parts. The first fifteen chapters of Exodu5 are located in Egypt
and its environs. They recount the oppression of the Israelites in Egypt,
Moses and the plagues, the Exodus itself, and the miraculous crossing of
the Sea. The second and longer part is entirely concerned with the j ourney
of Israel through the wilderness. This includes the events at Sinai, which
occupy a large section - from Exod. 1 9 to Num. 1 0: 12 - and the entire
book of Leviticus. From the literary point of view, however, these books
differ markedly from Genesis in that only about one-third of these 1 03
chapters can be said to consist of narrative. Most of the rest consists of
laws, instructions, and regulati9ns (and in some cases, especially in the
account of the tabernacle, their execution) laid down by God through
Moses, many of them at Sinai. In many cases, especially in Numbers, law
is so intertwined with narrative that it is hardly possible to distinguish one
from the other:
These laws will be discussed in more detail in chapter 7 below; but
at this point it is relevant to consider the role that they play in their context
in Exodus-Numbers and in the Pentateuch as a whole. The context is a
narrative one: these are not timeless laws unrelated t() the events in the

64
Exodus, Leviticus, Numbers: Narratives

story. They are presented as having been promulgated at particular times


and in particular places, so that they are an essential part of the narrative.
The reader is made aware of the circumstances in which the basic consti­
tution of the nation was given to it. There is a sense in which everything
from Genesis on has been leading up to the Sinai event. It is for this reason
that the whole of the Pentateuch - unlike any other book of the Old
Testament - is known to the Jews as Torah ("law" or "instruction"). Sinai
is the place in the wilderness to which Moses asked and was refused
permission by Pharaoh for the people to go to worship their God (Exod.
5:3; 7: 1 6; 8:27) and to which God is referring when he says, "Tell the
Israelites: 'You have seen what I did to the Egyptians, and how I bore you
on eagles' wings and brought you to myself' " ( 1 9:3-4) . It is here that the
people enter into a covenant with God and promise to obey his laws
( 1 9:3-8; 24:3-8; 34: 1 0, 28) .

The Figure of Moses

Moses dominates the whole of these three books and Deuteronomy as well.
He embodies all the qualities and functions of a leader. Under God's
direction he is the one source of authority as ruler and director of the people.
He is also, on occasion, a victorious commander in war. He is lawgiver and
judge, and performs actions normally associated with the priesthood, al­
though it is his brother Aaron who is said to be a priest. He is also a prophet
and a teacher. It is through Moses that God communicates his will to the
people. He acts as an intercessor when the people have sinned. He is the
mediator of the covenant which God makes with the people, and it is he
who, acting always under God's direction, is their savior and protector. Moses
is, then, what we should call a hero - the unique hero of his time. Yet he is
also supremely a man of God and a servant of God who is himselfthe object
of divine rebukes on several occasions.
It is obvious that this picture of such an all-embracing authority
figure cannot be a homogeneous one, incorporating as it does all the
functions of rulers and of holy men that later Israel was to encounter
during its history. It would seem rather that, whatever historical reality
may lie behind this figure, there has been a legendary development, perhaps
.
of tremendous proportions. Every aspect of greatness and virtue has been
piled on Moses at some later time, making him an ideal person, the fount
and origin of all subsequent nobility and greatness - a development

65
INTRODUCTION TO THE PENTATEUCH

comparable with that concerning David, and to a lesser extent Samuel and
Solomon.
That such a development took place is generally acknowledged; . it
appears to have occurred at a comparatively late date. It is striking that,
as with Abraham, Jacob, and Joseph, there are hardly any references to
Moses in any of the Old Testament books outside the Pentateuch that can
with certainty be regarded as preexilic, and none that shows any knowledge
of the events of his life as recounted in the Pentateuch. Only 2 Kgs. 1 8:4.,
which may have preserved an older tradition, refers to one such incident,
connecting it specifically with Moses' name: his setting up a bronze serpent
in the wilderness (Num. 2 1 :4-9) . This reference is negative rather than
positive; although it does not specifically criticize Moses' action, it regards
the bronze serpent itself- which had apparently been preserved and
placed in the temple at Jerusalem - as something which had become an
object of idolatrous worship and which Hezekiah, in cleansing the temple
and restoring the pure worship of Yahweh there, had praiseworthily de­
stroyed. The earliest reference to Moses in the prophetic books is Jer. 1 5: l ,
where he is named together with Samuel as a famous intercessor. Moses'
name is not so much as mentioned by any of the preexilic prophets.
The view of Moses' importance underwent a complete change in the
postexilic period. Some of the postexilic literature (Mal. 4:4; Neh. 9: 1 4;
1 3: 1 ; Dan. 9: 1 1 - 1 3) cites Moses as lawgiver, or refers to his "book" or
"law" as binding on Israel; in other texts (Isa. 63: 1 1- 1 2; Mic. 6:4; Pss. 77,
1 05, 1 06) there is clearly a familiarity with the Pentateuchal story, with
allusions to Moses in connection with such events as the Exodus and the
miracle at the Sea and his leadership in the wilderness. His reputation,
thus once established, has remained unchanged in Judaism ever since.
It is difficult to account for this transformation. Some scholars have
supposed that the original Moses tradition was confined to only one of
the various Pentateuchal themes, and that the notion of his connection
with the others was a subsequent development. Two themes that have been
frequently suggested as his original "location" are the Exodus (together
with the miracle at the Sea) and the making of the covenant, along with
the giving of the Law at Sinai. Martin Noth saw the origin of the Moses
tradition in the "occupation'' theme, though he was very skeptical even
about this. Even there the original tradition had contained no informat_i on
about Moses apart from the note regarding his burial (Deut. 34:6) - the
only genuine indication that he existed at all! Of the two other proposals,
that the oldest Moses tradition is to be found either in the Exodus story

66
Exodus, Leviticus, Numbers: Narratives

or in the Sinai pericope, the latter is perhaps the more plausible view.
Several references to him in the postexilic books speak simply of the
authoritative "law'' or "book" of Moses, whereas in many of the innumer­
able references to the Exodus or to the miracle at the Sea his name is not
mentioned at all.
To what extent it was the author of the present Pentateuch who
created the "biography'' of Moses is not clear. In vieW of the methods
which he employed in Genesis, it is reasonable to accord him a large part
in this. From the literary point of view it is clear that it is the figure of
Moses which now holds the story together and gives it a focus. · Many of
the stories are quite short, and it seems probable that these could have
already been circulating in the author's time; not all of them need have
been originally connected with Moses himself But if Moses was already
revered as lawgiver, it is not surprising that other great events should have
been attributed to him, notably the Exodus, the miracle at the Sea, and
the journey toward the Promised Land.
That the story of Moses as recounted in the Pentateuch is a late
literary construction is supported by a recent and increasingly accepted
hypothesis put forward on other grounds, that there was no mass immi­
gration of Israel into Canaan from outside at all! George E. Mendenhall
and Norman K. Gottwald were the first to cast doubt on the historical
credibility of the migration from Egypt to Canaan, and on the role of
Moses in . such a movement. Their view has been supported on both
archaeological and sociological grounds. These scholars and those who
have followed them maintain that there was no occupation of the land
from outside. Rather, the later Israelites were actually descendants of part
of the Canaanite poplllation which, whether individually or in a corporate
revolutionary movement, had detached itself from the life of the Canaanite
cities of the plains with their surrounding agricultural territories, and had
gradually established itself in the previously uninhabited, or sparsely in­
habited, hill country. There was no "conquest" of Canaan by immigrants,
nor was there a gradual infiltration by nomads - the main alternative
theories previously dominant. Modern archaeological research has revealed
that there was neither a cultural nor a linguistic break which would suggest
the arrival of a new population. Mendenhall ("The Hebrew Conquest of
Palestine," 73) envisaged

the withdrawal, not physically and geographically, but politically and


subjectively, of large population groups from any obligation to the

67
INTRODUCTION TO THE PENTATEUCH

existing political regimes . . . . In other words, there was no statistically


important invasion of Palestine. . . . There was no real conquest of
Palestine . . . [but] a peasant's revolt against the network of interlocking
Canaanite city states.

Subsequent scholars have differed about the causes of such an internal


movement of population but have held on to the general theory (e.g.,
Gottwald, The Tribes of Yahweh, 36-4 1 ; Niels P. Lemche, Early Israel
4 1 If£ ; Giovanni Garbini, History and Ideology in Ancient Israel 64). But
if the present Moses story is largely a purely literary creation, a question
arises concerning the origin of the religion of Yahweh which the Pentateuch
so closely associates with the figure of Moses. Some proponents of the
"internal development" hypothesis, though rejecting the notion of a mass
immigration, have conceded the possibility that although the origins of
the Israelite people as a whole must be looked for on the soil of Palestine,
there may have been a small group of people (sometimes known as "the
Moses group") who had been slave laborers in Egypt, had made their
escape, and had then arrived in Palestine, having both in Egypt and on
their journey undergone some experiences associated with a god named
Yahweh who had in some way been their deliverer. It was this small group
which, having been absorbed into the newly formed people of Israel, had
introduced the worship of Yahweh to their new compatriots. Such a theory
might account for the fact that some of these traditions seem to have been
known, especially to the eighth-century prophet Hosea and his northern
audience. But this would not account for a Pentateuchal narrative con­
cerning an army of 600,000 men with their families, augmented by a
further "mixed crowd," and accompanied by numerous flocks and herds,
who supposedly walked through the desert under the direction of Moses
for forty years (Exod. 12:37-38).
Why should the Pentateuchal author have concocted such a story,
and why did it eventually come to be so universally accepted by later Jews
as the basis of their beliefs about their origin and their relationship to
God? It must be remembered that the Pentateuch as we have it now is on
any reckoning a document written many centuries after the time when
the events that it describes are purported to have occurred. Consequently,
unless there were strong oral traditions about that remote era which
strongly contradicted the Pentateuchal story, there is no reason why it
should have been disbelieved. There may well have previously been little
or no such "folk memory." It must further be remembered that the Pen-

68
Exodus, Leviticus, Numbers: Narratives

tateuch was probably the first comprehensive account of Israelite origins


to be written, at about the same time as such works became fashionable
in the civilized world. Some , Jewish literature of a later period ­
midrashes, apocalypses, "testaments," and the like - show that many
people were ready to believe much more extravagant stories about Israel's
origins and early history than this.
Moreover, as has already been noted with regard to Genesis, such a
story - of a generation which was so wonderfully saved and protected by
its God, yet because of its sin and rebelliousness was in the end denied
access to the land which had so long been promised - would convey both
a welcome assurance of God's supreme power over the nations of the world
and also a warning message to a postexilic generation which did not yet
repossess that land, that only if it was obedient could it expect further
benefits from its God. If the Pentateuchal story was new to that generation,
it was extremely relevant to its situation. That it has been so completely
accepted as a true account by countless generations of the Jewish people
is a tribute to its literary merit and compelling power.

Moses and the Exodus (Exodus 1 - 1 5)

Johannes Pedersen, writing at a time (1 934) when much of the Old


Testament was thought to be of cultic origin, suggested that these chapters
are a unified story based on the "cultic legend" of the Feast of the Passover
- that is, an account of the origin of that feast which was solemnly recited
at its annual celebration. The core of this theory was, of course, that ch. 12
describes the first celebration o f the Passqver o n the eve o f the departure
of the Israelites from Egypt.
Although Pedersen's theory has not received a general scholarly as­
sent, and these chapters appear to have been pieced together from many
originally independent sources, they have in their present form a strong
element of continuity, stronger than in most of the other parts of the total
story. Not only is the figure of Moses central to them, but they present a
clearly "biographical" appearance. Moses is the principal agent in saving
his people from their distress. His birth and early life are described. He
then must flee for his life to Midian, but receives a commission from God
to return to Egypt and to lead his people away from their oppression there.
After an initial reluctance Moses does return and confronts the pharaoh.
Pharaoh's initial refusal to let Israel go is overcome by the infliction on

69
INTRODUCTION TO THE PENTATEUCH

Egypt of a series of plagues sent by God but announced by Moses as his


agent, ending with a plague so terrible that the terrified Egyptians assist
the Israelites' departure. A further attempt by Pharaoh to retain the service
of the Israelites is overcome in a miraculous crossing of the "Sea'' by the
Israelites, while Pharaoh's army is drowned. We are then told in conclusion
that "the people feared the Lord and believed in the Lord and in his servant
Moses" (Exod. 1 4:3 1), and that a song of victory was then sung.

The Birth ofMoses

The life stories of famous individuals or heroes often began, in the ancient
world and even in more recent times, with an account of mysterious
circumstances that surrounded the birth of the hero and in some way
presaged the hero's future greatness. Such stories frequently involve super­
natural, and even cosmic, happenings. For example, Glendower in Henry
N Part I (3. 1 ) boasted that "the earth did shake when I was born." The
story of the birth of Moses (Exod. 2: 1 - 1 0) is less spectacular, but clearly
serves a similar purpose. Born of obscure parents in the tribe of Levi, he
was placed in a papyrus basket made watertight with bitumen and placed
by his mother in the reeds by the riverbank in an attempt to save him
from Pharaoh's threat to kill all the newborn male Hebrew babies. Moses
was rescued by Pharaoh's daughter and brought up by' her as her son -
clearly destined for greatness.
It is of considerable interest to note that a somewhat similar story is
told of King Sargon of Agade in a legend preserved in Neo-Assyrian copies
(ANET, 1 1 9) . In it the king relates how he was of obscure birth, was placed
in the river in a basket made watertight with bitumen, was rescued, and
eventually became king. There are striking similarities in the details of the
two stories; and although there is no reason to believe that the Pentateuchal
author was familiar with the Sargon legend, it is reasonable to suppose
that in providing Moses with a birth story he was employing a motif which
was well known. The incident presaged the future greatness of one who,
though he did not become a king, was to be the ruler of his people. It is
rather remarkable that, as in the Sargon legend, there is no mention of
divine intervention; but in the context this can be assumed. Like the
patriar,hs of Genesis, the hero was in danger of death, and from the very
moment of his birth; yet he was miraculously preserved.
The account of the birth of Moses should be seen as one of a series

70
Exodus, Leviticus, Numbers: Narratives

of stories about miraculous births. In Genesis, the birth of Abraham's heir


Isaac is due to divine intervention which overcomes Sarah's barrenness.
The story of the birth of Jacob and Esau presages the divine choice of the
ydunger twin as the heir. In the cases of Isaac and Jacob divine guidance
enables them to find their families. With Joseph it is not his birth but his
dreams as a young man that presage his future greatness. But again of
Moses, Samson Qgs. 13), and Samuel ( 1 Sam. 1) there are miraculous
birth stories in which a woman's barrenness is overcome - though it is
no longer the individual heir to the promises who is involved but other
persons whose lives are equally vital to the fate of a nation. Miraculous
births may be said to be part of the basic narrative pattern of the Old
Testament account of Israel's origins.

The Commissioning ofMoses

This account (Exod. 3-6) is prefaced by two stories about Moses as a young
man which account for his appearance as a refugee from Egypt in the land
of Midian, where he married the daughter of the priest ofMidian (2: 1 1 -22).
These chapter� are somewhat confused and contain a nllmber of parallels
which cannot be explained by the Documentary Hypothesis of lllultiple
continuous sources. Joseph Blellfci��opp�-- �omm�nt on . Eioa: -1- ·!s . al.so
,
relevant here: "The compiler . . . doubtless drew on narrative traditions in
·

either written or oral form, but they are not clearly identifiable as segments
of continuous sources" ( The Pentateuch, 145) . This is true even of ch. 6,
which .is usually attributed to P. Despite the inconsistencies in detail, the
author's purpose can be clearly seen. This was to give the reader an impression
of the very great difficulties faced by Moses in his attempt to carry out the
task that God imposed on him, to act as his agent in securing the release of
the Israelites from Egypt (3: 1 - 1 2) . The author presents these difficulti� one
after the other: the_ problem of convincing Pharaoh of the identity and the
power of the God who demanded their release (3: 13-17); the reluctance of
Pharaoh to release the . Israelites and his negative reaction to the request,
demanding even greater productivity from them (5:2-14); and the con­
sequent attack on Moses by the Israelite foremen who regarded Moses as a
troublemaker whose well-meaning interference had made their · situation
worse rather than better (5 : 1 5-2 1 ) . This attack provoked a bitter complaint
to God by Moses, which in return called forth from God a confirmation of
his intention to force Pharaoh's hand. It is note�orthy that this speech by

71
INTRODUCTION TO THE PENTATEUCH

God in ch. 6 is not, as has often been supposed, a "second call" of Moses
parallel with that in ch. 3. It differs significantly from it in that it is a
statement ofwhat Godwill do. The burden is shifted from Moses, whose role
is now simply to inform the Israelites of what God has promised; but they
still remain unconvinced. The scene is now set for the story of the "plagues"
and of Pharaoh's stubborn refusal to release the Israelites until his resistance
is finally broken (12:29-32) .

The Plagues ofEgypt

This story (Exod. 7: 1 4- 1 2:32) is a tour de force. Its intention is to


demonstrate to the reader that there is no limit to the power of the God
of Israel, and that he will exercise that power in whatever way he chooses
in defense of his people. Not even the greatest political power in the world
- and here Egypt is clearly the symbol of all such power - can withstand
God's determination to release Israel from oppression. It is a story that,
with its grim humor, would have delighted those who first read it, and
who had long suffered from oppression by foreign powers. The protracted
series of no less than ten plagues shows that the author had a further
purpose beyond that of accounting for their release. Indeed, it would seem
obvious that the final plague by itself- the sudden death of all the
Egyptian firstborn, in,cluding the heir to the throne ( 1 2:29-30) - would
have been sufficient to bring about Pharaoh's capitulation and make him
willingly expel the Israelites, whom (together with their God) he regarded
as responsible for this calamity. The multiplication of the number of
plagues turns the story into that of a deadly game played against Pharaoh
by God. It also enabled the writer to introduce a number of narrative
elements which enhanced the narrative and gave it a triumphalist character.
It is important to note that, although at first sight it might seem
that it is Pharaoh who plays a game with the Israelites - it is he who time
after time frustrates their hopes by first agreeing to let them go and then
changing his mind ("hardening his heart") - it is several times made clear
(9: 12; 1 0 : 1 -2, 20; 1 1 : 1 0) that it is in fact God who deliberately hardens
Pharaoh's heart, so each time provoking yet another plague. This is made
absolutely clear in 1 0: 1 -2:

Go to Pharaoh; for I have hardened his heart and the heart of his
officials, in order that I may show these signs of mine among them, and

72
Exodus, Leviticus, Numbers: Narratives

that y� u may tell your children and grandchildren how I played a game
[or "toyed"] with the Egyptians, and what signs I have done among
them - so that you may know that I am Yahweh.

A more subtle feature of the story concerns the diminishing role


played by Pharaoh's court magicians. At first, when Moses and Aaron
appear before Pharaoh and Aaron's staff is transformed into a snake, the
magicians are able to do the same with their staffs - though Aaron's snake
swallows theirs (7: 1 1 - 1 2). They are also able to match the first two plagues:
the turning of the waters of the Nile to blood (7:22) and the bringing of
the plague of frogs (8:7) - though ironically this makes matters worse;
they are unable to be really useful and remove the frogs. In 8: 18, however,
the magicians are unable to turn the dust of the earth into gnats, and
exclaim that Moses and Aaron have done this by "the finger of God"
(v. 1 9). Their final appearance is in 9: 1 1 , where they are unable to match
the plague of boils, and are themselves afflicted by them. After this plague
(the sixth) they apparently give up, and disappear from the story.
The ten episodes are recounted with considerable variations within
a common basic pattern. Although much is repeated, there are few of the
incidents of which it could be said that the same precise model has been
adhered to. For example, in a particular case there may or may not be a
warning to Pharaoh by Moses before the infliction of a particular plague;
the instrument used may or may not be Aaron's staff; the role of Aaron
as Moses' companion varies; Yahweh may act directly without reference
to any instrument; Pharaoh may or may not appeal to Moses for the
plague's withdrawal; there may or may not be a statement that the plague
was withdrawl}.; and so on. The variations in details are almost endless and
need not be analyzed in detail. One feature that occurs only in a few cases
is the statement that the Israelites, who lived apart from the Egyptians,
were not affected by certain plagues. The author presumably intended that
this exemption was to be understood as applying to all of them.
Attempts were made by Julius Wellhausen and other documentary
critics to account for these variations by distributing the plagues between
the sources J (or JE) and P on the basis of distinctive form and content.
But in fact they cannot be accounted for by this method; source analysis
does not produce neatly consistent results. Each plague story still retains
its own particular individuality in the way it is told. Further, the notion
that the present text is an amalgam of variant oral sources can hardly be
maintained; it is difficult to see what motive there could have been for

73
INTRODUCTION TO THE PENTATEUCH

such an operation here. More important, any such methods fail to leave
room for the literary imagination of the final author. The action is full of
suspense; and yet it moves forward inexorably, as plague follows plague
and the disastrous effects of the plagues on the very lives of the Egyptians
are intensified, to the final plague and to the climax in the actual departure
of Israel from Egypt, so often demanded and so often frustrated. The
variations in the telling of the story were, in such an extended narrative,
necessary; to describe ten incidents in virtually the same words would have
been intolerably dull. The author chose now one, now another detail,
while leaving a general impression that what was true of one incident was
also true of the others. The result is a masterpiece which would have been
ruined by the imposition of complete uniformity.
The plague narrative is strangely interrupted in 12: 1 -28 by a series
of laws about the observance of the feasts of Passover and Unleavened
Bread, commanded by God to be ma:de known to the people by Moses
and Aaron (v. 1 ends with the phrase "in the -land of Egypt," as if to give
the laws a general rather than a particular setting) . Further such laws are
given in 1 2:43-49 and 1 3 : 1 - 1 6, intermingled with further narrative. The
sequence is as follows:

1 1 :4-1 0: Announcement o f the tenth plague (death o f the Egyp­


tian firstborn)
12: 1 -28: Passover laws
12:29-32: The tenth plague
12:33-36: The Egyptians encourage the departure of the Israelites
and load them with gifts
12:37-39: The Israelites leave Egypt and begin their journey
12:40-42: Notes on the date of the Exodus and its commemoration
by annual keeping of a vigil
12:43-5 1 : Further Passover laws and their execution
1 3 : 1 -16: Further laws on the Feast of Unleavened Bread

The reason for the insertion of these laws into the narrative and the
original connection, if any, between the tenth plague and the laws that
immediately precede it have been the subject of much discussion. The
Pentateuchal author evidently wishes to emphasize the extreme antiquity
of divine laws to the greatest possible extent, at times introducing them
in a somewhat incongruous manner (e.g., the permission to eat meat
provided that the blood was drained from it is traced to the time of Noah;

74
Exodus, Leviticus, Numbers: Narratives

Gen. 9:3-5), and attributes the particular laws of Israel to the time of
Moses; but here he especially presents the laws ofExod. 1 2 as related to
the surrounding narrative (or to some . aspects of it}. The Passover ritual
here prescribed as an annual event (v. 2) is to be a commemoration of
what happened on the eve of Israel's departure from Egypt. The details of
the ritual are presented as determined by the exigencies of the Israelites'
situation: they are to prepare themselves for a hasty flight (v. 1 1) , and to
protect themselves from the imminent slaughter of the Egyptians by smear­
ing the blood of the sacrificial Passover lamb on their doorposts (w. 22-23) .
It is such details as these that are at the basis of Pedersen's theory
that Exod. 1 - 1 5 originated as the "legend" which was recited at the annual
celebration of the Passover (see above) . For Noth (A History ofPentateuchal
Traditions, 66) it was the celebration of the Passover itself that had given
rise to the development of the plague stories. Others, however, have seen
the final plague as an alternative to the story of the crossing of the Sea in
Exod. 1 4- 1 5. They have pointed out that there are inconsistencies in the
present narrative which are exemplified in ch. 1 2 with its emphasis on the
need for haste and secrecy. According to this view the author has (some­
what ineptly) tr�ed . to combine two quite different and mutually incon­
sistent accounts of the Exodus: one which represents it as a secret flight
which Pharaoh, when he became aware of it, tried to thwart by pursuing
the fugitives ( 1 4: 5-9) , and one which represents it as an ordered and public
departure for which Pharaoh had reluctantly given permission after the
slaughter of the Egyptian firstborn ( 12:3 1 -36) . The latter account did not
contain, and had no need for, the miracle at the Sea (see George W. Coats,
Moses, 9 1 f£). The true climax of the plague stories was not a demonstration
of God's power at the Sea but the demonstration of his power in the death
of the firstborn and the consequent release of the Israelites.

The Miracle at the Sea

These chapters (Exod. 1 4 : 1-1 5 : 2 1 ) comprise three main sections. Ch. 14


tells, in prose, of the Egyptians' pursuit of the Israelites, the miraculous
parting of the waters of the · Sea at the stretching out of Moses' hand, the
waters' being driven back by a strong east wind so that the Sea is turned
into dry land, the Israelites' crossing on dry ground, and the drowning of
the Egyptian army by the waters' return to their normal state. Exod.
1 5 : 1 - 1 8 is a victory song then sung by Moses together with the Israelites,

75
INTRODUCTION TO THE PENTATEUCH

praising Yahweh for their miraculous deliverance but ending (vv 1 3- 1 8) .

with references to subsequent events: the continuing guidance and pro­


tection of the people through the wilderness toward the Promised Land;
the terror of the Philistines, Edomites, Moabites, and Canaanites at the
Israelites' approach; Israel's establishment in the land of Canaan; and finally
the building of the temple in Jerusalem. The third, brief component of
the chapter is a short victory song sung by Miriam, Moses' sister, to the
accompaniment of tambourines ( 1 5:20-2 1 ) .
That there are discrepancies between ch. 1 4 and the poem in 1 5 : 1 - 1 8
is hardly surprising, since ch. 1 5 is clearly not intended to b e a literal
account of the events but is a hymn of victory - a type of literature in
which factual precision is not called for or expected. In ch. 14 also there
are some discrepancies. Neither the cause of the drying up of the sea -
a strong east wind or the stretching out of Moses' hand - nor the precise
sequence of events is made dear; it has been supposed that the author has
here combined two separate versions. In essentials, however, the two chap­
ters agree. The Song of Miriam in 1 5 :2 1 is a second, very brief, victory
hymn which is virtually identical with 1 5 : 1 b and which, if it immediately
followed the Song of Moses, would form a convincing conclusion to that
song (compare Pss. 8 and 1 03 for a similar arrangement) . This is the first
mention of Miriam (but see Exod. 2:4) and was no doubt introduced here
in preparation for the numerous references to her in the book of Numbers.
It has been maintained by some scholars (William F. Albright;
Frank M. Cross and David Noel Freedman, Studies in Ancient Yahwistic
Poetry) th.at the Song of Moses or Song of the Sea in Exod . . 1 5 : 1 - 1 8 is very
ancient, perhaps as old as the twelfth century B .C. Its poetic style and
language have been held to be very close to those of Ugaritic poetry. These
arguments, however, have been disputed. It is also dear that the attempts
by these scholars to interpret vv 1 3- 1 8 - which ostensibly refer to later
.

events including the building of Solomon's temple - in quite other ways


must be regarded as tendentious.
What is clear is that the Song has been influenced by a myth found
in early Canaanite (Ugaritic) and, in other forms, in Mesopotamian
literature, namely, what has been called the myth of "God's conflict with
the dragon and the sea" - a myth associated in Mesopotamia (but not,
as far as our evidence goes, in Canaan) with the creation of the world.
This ancient myth has left traces not only in early Israelite literature but
in Old Testament literature of all periods, including the book of Daniel
(second century B.C.); in literature of the intertestamental period, and

76
Exodus, Leviticus, Numbers: Narratives

even quite clearly in the New Testament (Revelation) . Of Exod. 1 5 John


Day writes: "it is very clear that the description has been shaped and
influenced by motifs deriving from the myth of the divine conflict with
the waters. Thus, the motif of the victory at the sea is associated with
Yahweh's eternal kingship" ( God's Conflict with the Dragon and the Sea,
98) . That is to say, the Song is an example of the "historicization of
myth" - a mythical motif has been transformed into a supposedly "his­
torical" event. A similar process is reflected in Isa. 5 1 :9- 1 0. Exod. 1 5,
then, cannot have been written before -the time of Solomon and may be
much later.
What is the location of the "sea'' or "Sea" in question? There has
been much discussion of this problem. It is important to understand what
the question means. Does it mean "Where did these events take place?"
or rather "Where did the author(s) present them as having taken place?"
As has already been noted, these chapters can hardly be considered as
strictly historical accounts; md thus the first of these questions is beside
the point. Nevertheless the author has given the story an apparently precise
geographical setting, and there is therefore some point in the second
question: where did he imagi.ne the events to have occurred? But this
question is also unanswerable. Martin Noth's comment, after many at­
tempts to answer that issue, remains true: "unfortunately we can no longer
make out" where the author located it - "that is if he had any idea at all
about geographical relationships in a neighbourhood far removed from
the later habitations of lsrael" (Exodus, 1 07- 8) . The only things of which
we may be certain are that the author placed it · not on the direct coastal
route (which he anachronistically called "the way of the Philistines") , since
he states that God deliberately guided the people away from this into the
wilderness further south ( 1 3: 1 7- 1 8), and that the sea in question was not
the Red Sea, as the Hebrew yam-suph ( 1 3 : 1 8; 1 5 :4, 22) was rendered in
earlier translations. The Hebrew means "sea of reeds,'' a place whose
location is unknown. Similar questions arise with regard to other places
recorded as located on the route taken by the Israelites, including Sinai.

The Journey through the Wilderness

The account recorded in Exod. 1 2:37-Num. 36 lacks the cohesiveness of


earlier chapters (of Numbers, Rolf Rendtorff wrote, "Of all the books in
the Pentateuch, the book of Numbers is the hardest to survey''; The Old

77
INTRODUCTION TO THE PENTATEUCH

Testament, 146) . These chapters comprise several different kinds of mate­


rial: itineraries, narratives, laws, and institutional matters such as the taking
of a census of the tribes. Although they have a narrative framework that
forms a bridge between Israel's experiences in Egypt and its entry into the
Promised Land, there is no continuous narrative at all. The narrative
sections consist mainly of short incidents most of which, though given
specific (generally not identifiable) locations, are quite isolated in the sense
that they do not form a coherent block but are interspersed with other
types of material. It is not surprising that Noth regarded the theme
"guidance in the wilderness" (excluding the events at Sinai) as "not. a very
important or really independent theme" which "presupposes in every
instance the themes 'guidance out of Egypt' and 'guidance into the prom­
ised land' and depends on both of these," and that "probably this theme
arose simply from the narrative desire to tell something concrete about
the further fortunes of the Israelite tribes after the 'guidance out of Egypt' "
(A History ofPentateuchal Traditions, 58, 59) .
Nevertheless, the author had some quite specific purposes in com­
posing these chapters. This part of his story afforded him a particular
opportunity to make some essential points of a theological nature. Here
was a people in transition - not only literally, between Egypt and Canaan,
but spiritually. Israel was hardly yet a nation at the beginning of Exodus;
by the end of Numbers its character had developed, though scarcely for
the better, despite the events at Sinai.
Both God and the people are represented in a fascinating series of
incidents as though making one another's acquaintance: testing one
another. The word "test, try" (Hebrew nissah) is used in these incidents
in two distinct ways: God tests the people to see whether they will obey
him or not (Exod. 1 5 :25; 20:20) , and the people test God to see whether
or not he will in fact provide· for their material needs in the wilderness
( 1 7:2, 7) and to see to what extent he will suffer their disobedience (Num.
1 4:22) . The many instances of their discontent, lack of faith, and re­
belliousness against Moses or against God himself (Exod. 1 4: 1 0- 14, im­
mediately after their departure from Egypt; Exod. 1 6; 1 7: 1 -7; 32: 1-33:6;
Num. 1 1 ; 1 2; 1 3- 1 4; 1 6; 20: 1 - 1 3) reveal a constant change of mood in
the three participants, God, Moses, and the people. God is shown as
generous and forgiving up to a certain point, but also as angrily punishing
rebels with death. At other moments (Exod. 33:3; Num. 14:20-23) he sets
limits to his promise to lead the people into the Promised Land. In some
cases (Exod. 32:27-34; Num. 1 2: 1 3- 1 5; 1 4 : 1 3-20; 1 6:4 1 -50) God shows

78
Exodus, Leviticus, Numbers: Narratives

himself to be capable of changing his mind, and allows himself to be


persuaded when Moses appeals for mercy for the rebels.
The relations between God and Moses are equally varied. Moses
throughout is God's spokesman, communicating God's words to the
people. He also frequently engages in direct dialogue with God. It is he
to whom God speaks face to face (Exod. 33: 1 1 ) , and to whom God shows
his glory (vv. 1 8-23) . But the texts also speak of Moses' frustration. He
complains to God of the intolerable burden ofleadership which God had
imposed on him, and despairs of his ability to control the people (Exod.
1 7:4) . In a flash of temper he smashes the tablets of the co�enant that
God had given him (Exod. 32: 1 9) . Moses complains of God's treatment
of him, and says that he would rather die than carry on (Num. 1 1 : 1 1 - 1 5) .
In Num. 20: 1 1 - 1 2 he falls under God's displeasure for losing his temper
and is told that he will not be permitted to enter the Promised Land and
attain the goal to which his whole life has been devoted.
The depiction of the people is also complex, but mainly negative.
At Sinai they appear to be awed by the tremendous events there, and repeat
the refrain, "Everything that the Lord has spoken we will do" (Exod. 1 9:8;
24:3). But in fact the story is of almost constant rebellion by all or part
of the people for various reasons. They, like Moses, are condemned to die
without reaching the Promised Land - but large numbers of the rebels
have already been destroyed by plague, the sword, and other divinely
ordained means. The first incidents of rebellion begin immediately after
the deliverance of the people at the Sea; already at Sinai the idolatry of
the Golden Calf occurs at the foot of the mountain itself (Exod. 32) .
Toward the . end of the narrative (Num. 25) the people practice mass
idolatry and sacrifice to other gods at Baal·peor. The author evidently
intended to make it quite clear that the people of Israel were entirely
unworthy to enter the Promised Land - that they totally ignored the laws
which God has imposed on them and that they had promised to keep,
and that they totally failed to appreciate what God had done for them in
saving them from the Egyptians and guiding them through the wilderness.
There could hardly be a clearer lesson for later readers who were living in
exile or deprived of actual possession of the land.
The stories about rebellion in the wilderness are quite varied. Some
have been derived from popular legends about desert life known to the
author. The desert was then, and remains, on the edge of the cultivated
land of Palestine with its towns and villages; and its dangers to travelers,
especially of death from hunger and thirst, would have been generally

79
INTRODUCTION TO THE PENTATEUCH

known to the inhabitants of the land at any period. The author has used
this material, brought it into connection with the person of Moses that
he has constructed, and put his own stamp on it. Much of it is his own
creation, and in some cases he has used a story more than once in slightly
different versions. The main themes on which he lays emphasis are the
desire of the people to return to Egypt (where, as they thought and as
desert travelers would naturally suppose, they would at least be free from
hunger and thirst even though in the immediate past they had lacked
freedom); rebellion against Moses; and, to crown all, apostasy from the
.
God who had delivered them.
The incidents concerned with lack of food and water and with the
miraculous provision of sustenance (water, Exod. 1 5 :22-25; 1 7: 1 -7; Num.
20:2ff.; food, Exod. 1 6: lff.; Num. 1 1) probably have their basis in travelers'
tales and may be compared with that of the miraculous feeding of Elijah
( I Kgs. 1 7: 1 -6; 1 9: 5-8) or with Elisha's purification of the spring water
(2 Kgs. 2: 1 9-22) . The names of the places where these events are stated
to have occurred - especially Meribah, "strife"; Massah, "testing" - may
have given rise to the stories; with regard to the "strife" in particular we
may compare the references to quarrels over the possession of springs of
water in the desert mentioned in Gen. 1 3: 8; 2 1 :25; 26: 1 9-22. The motif
of rebellion against Moses' leadership which occurs in almost all these
stories including Exod. 32 (the Golden Calf) and Num. 1 6 (Korah,
Dathan, and Abiram) probably reflects leadership disputes in the postexilic
period. Num. 1 6; Exod. 32; and also Num. 1 2 (the rebellion of Miriam
and Aaron against Moses) are principally concerned with questions of true
and false worship and false claims to the priesthood which were certainly
living issues at that time.
From Num. 20 on the narrative begins to speak of attempts made
by the Israelites to enter the Promised Land. Since God had previously
ordained that they should not go by the shortest route, through the
(anachronistic) "land of the Philistines" (Exod. 13: 1 7- 1 8), to enter the
country from the southeast, the author sends them round via the south
to be ready to invade from the west, across the Jordan.
The theological lessons of these narratives are already set out in Num.
1 3-14. Moses sends a reconnaissance patrol from the wilderness of Paran
south of Canaan to report on the fertility of the land and to estimate the
military strength of the Canaanites, but they report that an attempt to
invade would result in failure and defeat. This news causes a rebellion
against the leadership of Moses and Aaron: the rebels propose to choose

80
Exodus, Leviticus, Numbers: Narratives

a new leader and to return to Egypt. An attempt by Caleb and Joshua to


persuade the people that victory would be assured if Yahweh were on their
side is met with strong hostility, and this arouses the anger of Yahweh,
who declares his intention to destroy the whole people and to make Moses
the nucleus of an entirely new people. Moses pleads for their forgiveness
on the grounds that such an action would lead the Egyptians and
·

Canaanites to conclude that Yahweh was unable to fulfill his promise to


give the land to the Israelites. His plea for forgiveness succeeds, although
God condemns the whole present generation to die in the wilderness
without entering the Promised Land. Only those who had caused the
rebellion by bringing back an unfavorable report from their reconnaissance
are immediately punished with death by plague. Finally, when the people
attempt to invade and conquer the land in defiance of the divine decision,
they suffer a humiliating defeat, because God is not with them (Num.
1 4:40-45) .
The lessons of these chapters are twofold - on the one hand,
Yahweh's commitment to his promises, the dependence of Israel's success
on his goodwill, and his support of his designated leaders; and on the
other hand, Israel's lack of faith and unworthiness of Yahweh's blessing,
Yahweh's intolerance of disobedience, and his readiness to annihilate the
disobedient (which can, however, sometimes be mitigated by the interces­
sion of those who were possessed of his favor) . These lessons would not
have been lost on the later Israelite readers: only by faith in God's good­
will, by obedience to his chosen leaders, and by rejection of false leaders
could they regain possession of the land that they had lost. These lessons
are exemplified in the narratives that follow, which are a curious mixture
of the positive and the negative: Israel's frustration at the refusal of the
king of Edom to let them pass through his land (Num. 20: 1 4-2 1), an
incident which significantly occurs immediately after Moses' and Aaron's
demonstration of a lack of trust in God (20 : 1 2) ; the victory over the
Canaanite king of Arad (2 1 : 1 -3); the victories over the Amorites and the
king of Bashan, leading to the conquest of their territories (2 1 :2 1-35) ; the
story of the seer Balaam, who was sent for by Balak king of Moab to curse
Israel but found that he could do no other than bless them, since Yahweh
had blessed them (chs. 22-24); and then, quite suddenly, Israel's total fall
from grace: the intercourse with Moabite women and the consequent
worship of their gods, especially Baal of Peor (ch. 25).
The final episodes in this series emphasize the two contrasting yet
not incompatible aspects of Israel's situation in the wilderness. On the one

81
INTRODUCTION TO THE PENTATEUCH

hand, by commissioning Joshua as Moses' successor - though he was to


act only in accordance with the instruction of the priest Eleazar, who would
communicate God's decisions to him (Num. 27: 1 6-23) - God reaffirms
his commitment to his promise to give Israel the land of Canaan. On the
other hand, it is repeated yet again that only the new generation of young
Israelites would be allowed to enter that land and take possession of it
(Num. 26:64-65).
Finally, these books have been given a kind of narrative continuity
by means of the use of names of places where the various incidents
occurred, and by an apparently comprehensive itinerary from Ramses in
Egypt to the plains of Moab in Num. 33: 1 -49 - an itinerary comprising
more than fifty place names. The places named in the main narrative as
the sites of particular incidents are of course far fewer, but almost all of
them appear in the longer list. Many of the names are otherwise unknown,
and some may be �ntirely fictitious. It is also possible that the author used
actual routes. Noth suggested that some of the places mentioned may have
been stations on a well-known pilgrimage route from Canaan to Sinai (see
also Graham I. Davies, The "Wly of the Wilderness; and Philip J. Budd,
Numbers), and there are other possibilities such as caravan routes, not
confined to any particular period.
It is difficult to avoid the impression that these itineraries have been
inserted in order to give an impression of historical reality to "an otherwise
bald and unconvincing narrative" made up of a mass of originally uncon­
nected items. However this may be, the author has succeeded in giving an
impression of Israel as a homeless people constantly on the march - a
continuation, in fact, of the situation of the equally homeless and itinerant
Abraham and his family in Genesis. The need for a settled existence as
promised to the patriarchs and its constant postponement,largely due to the
people's sins and lack of faith, were pertinent themes for a latedsrael.

For Further Reading

Blenkinsopp, Joseph. The Pentateuch: An Introduction to the First Five Books


of the Bible. New York: Doubleday and London: SCM, 1 992, chs.
5 and 6.
Budd Philip J. Numbers. WBC 5. Waco: Word, 1 984.
Childs, Brevard S. The Book of Exodus. OTL. Philadelphia: Westminster
and London: SCM, 1 974.

82
Exodus, Leviticus, Numbers: Narratives

--- . ''A Traditio-historical Study of the Reed Sea Tradition," VT 20


( 1 970) : 406-4 1 8.
Coats, George W "History and Theology in the Sea Tradition," in The
Moses Tradition. ]SOT Supplement 1 6 1 . Sheffield: JSOT Press, 1 993,
45-56.
--- . Moses: Heroic Man, Man ofGod ]SOT Supplement 57. Sheffield:
JSOT Press, 1988.
--- . Rebellion in the Wilderness: The Murmuring Motifin the Wilderness
Traditions of the Old Testament. Nashville: Abingdon, 1 968.
--- . "The Song of the Sea, " CBQ 3 1 (1 969) : 1 - 1 7. .

--- . "The Traditio-historical Character of the Reed Sea Motif," VT


1 7 (1 967) : 253-265.
Cross, Frank M., Jr. Canaanite Myth and Hebrew Epic. Cambridge, Mass. :
Harvard University Press, 1 973, 1 1 2-144.
---, and Freedman, David Noel. Studies in Ancient Yahwistic Poetry.
SBL Dissertation 2 1 . Missoula: Scholars Press, 1 975, 45-65. (Diss.,
Johns Hopkins University, 1950.)
Davies, Graham I. The Wtiy of the Wilderness: A Geographical Study of the
Wilderness Itineraries in the Old Testament. Cambridge: Cambridge
University Press, 1 979.
Day, John. God's Conflict with the Dragon and the Sea'.· Echoes ofa Canaanite
Myth in the Old Testament. Cambridge: Cambridge University Press,
1 985.
Driver, Samuel R . Introduction to the Literature of the Old Testament. 8th
ed. Edinburgh: T. & T. Clark, 1 909.
Garbini, Giovanni. History and Ideology in Ancient Israel. New York:
Crossroad and London: SCM, 1 988. (First published in Italian,
1 986.)
Gottwald, Norman K. The Tribes of Yahweh. Maryknoll, N.Y. : Orbis,
1 979, and London: SCM, 1 980.
Johnstone, William. Exodus. OTG. Sheffield: JSOT Press, 1 990.
Lemche, Niels P. Early Israel. VTS 37. Leiden: Brill, 1 985.
Mendenhall, Goerge E . "The Hebrew Conquest o f Palestine," BA 25
( 1962): 66-87. Repr. i n The BA Reader 3, ed. E. F. Campbell, Jr.,
and David Noel Freedman. Garden City: Doubleday, 1 970, 1 00-
1 20.
Noth, Martin. Exodus. OTL. Philadelphia: Westminster and London:
SCM, 1 962. (First published in German, 1 959.)
---. A History ofPentateuchal Traditions. Englewood Cliffs, N.J . : Pren-

83
INTRODUCTION TO THE PENTATEUCH

tice-Hall, 1 972. Repr. Atlanta: Scholars Press, 1 989. (First published


in German, 1 948.)
--- . Numbers. OTL. Philadelphia: Westminster and London: SCM,
1 968. (First published in German, 1 966.)
Pedersen, Johannes. "The Crossing of the Reed Sea and the Paschal
Legend," in Israel· Its Life and Culture, III-N. London: Oxford
University Press and Copenhagen: Branner, 1 940, 728-737. (First
published in Danish, 1934.)
Rendtorff, Rolf. The Old Testament: An Introduction. London: SCM, 1 98 5 ,
and Philadelphia: Fortress, 1 986. (First published in German, 1 983.)
---. The Problem ofthe Process of Transmission in the Pentateuch. ]SOT
Supplement 89. Sheffield: JSOT Press, 1 990. (First published in
German, 1 977.)

84
CHAPTER 6

The Book of Deuteronomy

A s HAS BEEN indicated earlier in this book, Deuteronomy occupies a


J""\s omewhat equivocal status in the Pentateuch. In the Jewish tradition
it is the fifth and final book of the Torah, which is the first and most
important of the three divisions of the Hebrew Scriptures. In subject
matter it is closely related to Exodus-Numbers, being wholly concerned
with Moses, and its final chapter (Deut. 34) concludes the history of Moses
begun in Exod. 2. However, the view of Martin Noth that Deuteronomy
was not originally connected with Exodus-Numbers but is, on the contrary,
the first part of another work, a "Deuteronomistic History" - a history
of Israel written during the Babylonian Exile and including the books of
Joshua, Judges, Samuel, and Kings - has been very widely accepted.
Since W. M. L. de Wette ( 1 806; see chapter 2 above) it has been
generally accepted that the "book of the law" whose discovery in the
temple during the reign of Josiah (622 B . C.) is recorded in 2 Kgs.
22:8- 1 0 was identical with Deuteronomy or at least with an early version
of it. Despite some recent reservations about the historical reliability of
the account of Josiah's cultic reform in 2 Kgs. 22-23 (and the somewhat
different account in 2 Chr. 34), this view has not been abandoned.
Accordingly, Rolf Rendtorff ( The Old Testament, 1 55) claims that "the
measures taken in Josiah's reform as reported in II Kings 23 show striking
affinity to the demands of Deuteronomy. So the connection is indispu­
table."

85
INTRODUCTION TO THE PENTATEUCH

Questions of Date and Provenance

Both the date(s) of composition of Deuteronomy and its provenance are


disputed. The account in 2 Kgs. 22 gives no indication of the origin or
contents of the "book of the law," except that it was believed to be both
ancient and authoritative and strongly condemnatory of current cultic
practice in Jerusalem. Josiah was appalled by its contents, exclaiming,
"Great is the wrath of Yahweh that is kindled against us, because our
ancestors did not obey the words of this book'' (2 Kgs. 22: 1 3) . · He con­
sulted the prophetess Huldah, who thereupon delivered an oracle from
Yahweh that owing to this disobedience he would bring disaster "on this
place and on its inhabitants" in accordance with the penalties set out in
the book itself (vv. 1 6- 1 7) . The cultic reform was the consequence. That
many of the laws in the book were indeed much older than Josiah's time
was argued by a number of scholars. Adam C. Welch, for example, sug­
gested that the laws in the book had originated in northern Israel in the
time of Samuel. Others, however, such as Gustav Holscher, argued for a
postexilic date. Some recent scholars favor a date shortly before the dis­
covery of the book, possibly the reign of one of Josiah's immediate pred­
ecessors, Hezekiah or Manasseh. ·
However, it is now recognized that, whatever relationship it may
have to Josiah's reform, Deuteronomy is neither the work of a single person
nor of a single period. It contains material of very different kinds. Its maj or
and central section is a collection oflaws (Deut. 1 2-26), many of which
are closely related to the laws in Exod. 20-23, while others are peculiar to
Deuteronomy or differ considerably in detail from those in other collec­
tions. The whole collection thus has its own distinctive character.
The first part of Deuteronomy (chs. 1-1 1) is presented, as is the
whole book, in the form of speeches placed in the mouth of Moses
addressed to "all Israel beyond the Jordan in the wilderness" on the plains
of Moab, and is clearly designed as an introduction to the laws that follow.
It begins with a recapitulation of past events: Israel's journey from Sinai
(here called Horeb) to the plains of Moab, with emphasis on both God's
guidance and the people's apostasy (chs. 1 -3). The following chapters,
which contain among other things the Ten Commandments (5:6-2 1 ) in
a slightly different version from that in Exod. 20: 1 - 17, convey solemn
warnings to the people to recognize God's goodness in the past and to
obey him on their entry into the land, which is imminent. Here also there
are reminders of past rebelliousness and warnings of the disastrous con-

86
, The Book ofDeuteronomy

sequences of further disobedience. These chapters frequently look forward


to the laws that follow from Deut. 12: 1 . Various stylistic features including
a series of new introductory formulae at 4: 1 , 44; 5 : 1 ; 6: 1 , 4; 8 : 1 ; 9 : 1 show
clearly that, although in general both style and language remain constant,
chs. 1 - 1 1 are not a unitary composition.
The third part of the book (chs. 27-34) contains a mixture of
material, including lists of blessings and curses giving warning of the
consequences of future obedience and disobedience respectively (chs. 27-
28), the making of a new and additional covenant (ch. 29) , the writing
down .of the laws of the covenant in a book, together with a command
that they are to be read publicly every seven years at the Feast of Booths
(3 1 :9- 13, 24-26), a song sung by Moses (ch. 32), a blessing pronounced
by Moses on the tribes (ch. 33) , and finally an account of Moses' death
and an assessment of his achievements and character (ch. 34) .
It has been pointed out (Dennis J. McCarthy; George E. Men­
denhall, "Covenant Forms in Israelite Tradition'') that the book as a whole
has structural affinities with ancient Near Eastern treaties or formal agree­
ments defining relationships between vassal kingdoms and their imperial
masters. In these, the vassals undertook to remain faithful to their overlords
. in return for their protection. It may be supposed that similar treaties will
have been made between some Israelite kings and their Assyrian or Baby­
lonian overlords. Vassal treaties differed from one another in details, but
they frequently comprised the following items: a "historical" introduction
which stressed past benefits supposedly conferred on the vassal kingdom
by its overlords; "stipulations" or conditions which the vassal promised to
observe; arrangements for the sealing or ratification of the treaty and its
public proclamation; threats or curses to be carried out in case of disobe­
dience; promises of protection by the overlord; and a list of deities cited
as having witnessed the making of the treaty.
Although Deuteronomy is not to be regarded as a treaty document
in itsel£ there are remarkable similarities between its structure and that of
the extant vassal treaties. Deuteronomy, it has been argued, should be
understood as setting out the relationship of the "vassal" Israel, not with
a human overlord but with its God, Yahweh. Its opening chapters describe
the past benefits conferred by Yahweh on his people. The laws that follow
detail the conditions of the relationship, to which obedience is required.
Blessings and curses and an account . of the making of the covenant,
together with its ratification, · follow. Although there is nothing in the
extant treaties that corresponds to Deut. 30-34, it seems probable that in

87
INTRODUCTION TO THE PENTATEUCH

Deuteronomy we have the earliest comprehensive theological statement


about the relationship between Yahweh and Israel, and that the notion of
the vassal treaty, here called a "covenant" (Hebrew berith) , has played some
part in the development of this theology. A further fact linking the two
kinds of document is that there are clear affinities between the two types
not only in structure but also in style and "technical" language.
Gerhard von Rad ("The Form-Critical Problem of the Hexateuch, "
26ff.) argued that the nucleus of the laws in Deuteronomy could be traced
back to an early covenant ceremony performed at the Feast of Booths at
Shechem. Deut. 27 stipulates that the Israelite tribes, after crossing the
Jordan, are to proceed first to the mountains Ebal and Gerizim, which are
in the vicinity of Shechem, to set up large stones there on which are to
be written down "the words of this law,'' to build an altar, and to offer
sacrifices on it. Other passages such as Deut. 1 1 :29-30; Josh. 8:30f£; Josh.
24 confirm this connection with Shechem.
But whatever may have been the origin of the laws in this book,
Deuteronomy as a whole in its final form bears the marks of a number of
subsequent influences. Various theories have been put forward about the
nature of the circles in which it may have been composed; each of these
has considerable plausibility, yet none of them accounts adequately for all
the maj or features of the book. It must therefore be concluded that the
book of Deuteronomy is the product both of widely different periods and
also of a variety of authors who represented different and not entirely
compatible theological positions. As suggested above, there is a cultic
background: the promulgation of the laws is closely associated with
ceremonies of oath-taking and the like. There are also a number of laws
regulating the membership and status of the priestly class and the conduct
of worship, notably the law prohibiting sacrifice at more that one altar
(Deut. 1 2) . There is no certainty, however, about the degree of antiquity
of this cultic feature.
Another suggested background for Deuteronomy is prophecy.
Ernest W. Nicholson in particular (Deuteronomy and Tradition) pointed
out resemblances between its teaching and that of northern prophets,
notably Hosea. These resemblances are to be regarded as indications of a
northern provenance of parts of the book, though the teaching in question
may have become more widely accepted before its eventual adoption in
Deuteronomy. But it is difficult to relate the book's few specific references
to prophets with the prophets with whom we are familiar from the pro­
phetic books of the Old Testament (this is also true of the Deuteronomistic

88
The Book ofDeuteronomy

History) . There are only three passages which refer specifically to prophets:
Deut. 13:1-5; 1 8: 1 5-22; and 34: 10. In 1 3 : 1 -5 prophets are put on a level
with other kinds of person who falsely claim to predict the future by means
of "dreams and other forms of divination, and who may incite the people
to worship other gods. Deut. 1 8 : 1 5-22, however, distinguishes between
prophets who speak in Yahweh's name and those who either speak in the
name of other gods or pretend falsely to speak in Yahweh's name. But on
a quite different level again it is Moses himself who in 1 8: 1 5 and 34: 1 0
is designated as a prophet - the only prophet who is named in the book.
None of these attitudes toward prophets strikingly recalls the personae of
the "classical" prophets, although there are resemblances to oracles in
Jeremiah about false prophets.
Von Rad (Studies in Deuteronomy, 66f£), noting the homiletic or
sermonlike character of the book, especially chs. 1 - 1 1 , suggested that it is
the work of Levites, living not in Jerusalem but in the country towns, who
addressed the people, attempting to persuade them of the importance of
Josiah's reform. A variation of this theory by C. Johannes Lindblom saw
the authors as northern Levites who had come to live in Jerusalem as a
consequence of the reform (2 Kgs. 23:8-9) . It was they who brought with
them the ancient northern traditions and promulgated them; and they
were also responsible for the fierce nationalistic and militaristic tone which
appears in the book. Although the Levites are represented as instructing
the people in several passages in Deuteronomy (1 7:9-1 1 ; 24:8; 33: 1 0) and
in very late texts (2 Chr. 1 7:9; Neh. 8 :9), the wide homiletic function
ascribed to them by von Rad and others is based on an assumption which
falls considerably short of proo£
Most recently Moshe Weinfeld (Deuteronomy and the Deuteronomic
School) has advanced the theory that Deuteronomy is the work of scribes
in the serviee of the Judean royal establishment, and that · the book is
suggestive of wisdom teaching, which was part of the scribes' stock-in­
trade. This proposal raises difficult questions about the nature of "wisdom"
and its connection with the royal court that cannot be discussed here. The
words "wise" (hakham) and "wisdom'' (hokhmah) occur only rarely in
Deuteronomy. Still, there are three passages that are significant for Wein­
feld's theory. Deut. 4:5-8 extols wisdom as a most desirable and important
human quality, and sees it as exemplified in its highest degree in the laws
that Yahweh is about to teach to Israel. If lsrael obeys those laws, this will
cause the other nations to express admiration of Israel as a supremely wise
people. According to 1 : 1 3 - 1 5 , wisdom is an essential characteristic of those

89
INTRODUCTION TO THE PENTATEUCH

chosen to be leaders in the affairs of the nation; and 34: 9 states that Joshua,
Moses' successor, was "full of the spirit of wisdom, because Moses had laid
his hands on him."
Deuteronomy certainly shares with the wisdom books of the Old
Testament the character of a teaching book. In this ' character it employs
vocabulary and phraseology which correspond to that used in Proverbs;
but both the context and the contents of the two books differ greatly.
Whereas in Proverbs a "father" gives individual instruction to his "son,"
and that teaching is concerned with the importance of adopting "wisdom"
(not with any legal overtones) as a means to personal success in life, what
is to be taught to the children in Deuteronomy - and on a national scale
- is a knowledge of the laws contained in the book (Deut. 6:20-25) . The
consequence of failure to learn and obey these is primarily a national, not
an individual, calamity.
Apart from a common emphasis on the importance of the teaching,
the contents of the two books differ in almost every respect. Deuteronomy
is fiercely nationalistic; Proverbs never mentions Israel at all nor appears
to be concerned with nations as such. Deuteronomy is much concerned
with the elimination of idolatry; Proverbs never mentions any worship but
that of Yahweh. Deuteronomy is concerned with public worship and its
character; Proverbs rarely mentions this. The list of differences could be
extended. It is true that there are laws in Deuteronomy which correspond
closely to passages in the wisdom literature, for example, laws against the
removal of boundary stones (Deut. 1 9 : 1 4; Prov. 22:28; 23: 1 0), against the
use of false weights and measures (Deut. 25 : 1 3- 1 6; Prov. 1 1 : 1 ; 20:23) ,
and on making vows (Deut. 23:21 -23; Prov. 20:25; Eccles. 5 : 1 -6) . But
such matters are not confined to the wisdom literature: they were matters
of general acceptance. More weight should perhaps be attached to Wein­
feld's other maj or contention that the international treaty form which


influenced the composition of Deuteronomy would be particularly well
known to . governme scribes; but in fact the form may have been more
widely known, since y of the extant vassal treaties carry a stipulation
that they should be regul - !ead publicly in the vassal country, a practice
with which Judah would have been only too familiar in the latter half of
its monarchic period.

90
The Book ofDeuteronomy

The Theology of Deuteronomy

Deuteronomy is then probably to be seen as a deliberate amalgam, formed


from many diverse elements of the traditional faith of Israel. It must
therefore, in its final form, be a relatively late book. It has a unique status
not only in the Pentateuch but also in the Old Testament. In the hands
of the final editors it presents a complete, more or less coherent theology
- the only fully conceived theology in the Old Testament, and one which
was to have a profound influence on subsequent thought, as may be seen
especially in the Deuteronomistic History and in the final editions of
several of the prophetic books. The "Deuteronomic" theological move­
ment, which may have begun as early as the religious reform carried out.
by Hezekiah (2 Kgs. 1 8 :3-4) , marked a new beginning in the history of
Israelite religion. The editors of Deuteronomy represented this theology
as if it were the original · true religion inaugurated by Moses. However,
there can be no doubt that the book is the result of an intensive theological
thought superimposed on a much less systematically organized Israelite
religion which - whether Yahwistic or largely polytheistic in character -
had been mainly lacking in systematic reflection.
In its teaching about God, Deuteronomy built on the foundations
of traditional Yahwism, but refined and rationalized it. Although its doc­
trine may not be that of formal monotheism - that is, the denial of the
existence of any deity but one (though see Deut. 4:35, 39) - it constantly
repeats its assertion that there is only one God for Israel. The exact meaning
of 6:4, which is one of the most famous verses in the Old Testament and
is daily recited by Jews in their prayers to this day, is much disputed. It is
not clear whether the passage is saying that Yahweh alone is Israel's God
or that Israel's God is, in an absolute sense, the only God. But the next
verse, which is a command to Israel to devote itself without restraint to
the love of Yahweh, is echoed incessantly throughout the book.
The fierce antagonism expressed in the book to the worship of other
gods clearly presupposes that both in the past and up to the present
polytheism had been widely practiced in Israel. There is ample evidence
of this in the books of Kings and in some of the prophetic books. Ezek.
8, for example, records the practice of idolatrous rites in the temple of
Jerusalem itself on the eve of its destruction in 587 B.C., and Isa. 57:3-8;
65:3-5 probably refer to the continuation of such practices even after the
Exile. There is no agreement at present among scholars about the nature
of Israel's religion (or religions) in preexilic times, although it is generally

91
INTRODUCTION TO THE PENTATEUCH

recognized that Yahweh was not the only deity worshipped. Some maintain
that Israelite religion in that period was essentially the same as Canaanite
religion. Others point to the prevalence, though not the exclusive use, of
proper names compounded with the name of Yahweh at that time, sug­
gesting that although Yahweh may have been one among a number of
other gods worshipped, he was probably recognized as the chief god. Some
(e.g., Morton Smith, Palestinian Parties and Politics; Bernhard Lang, Mon­
otheism and the Prophetic Minority; see also Mark S. Smith, The Early
History of God, for a fuller account) postulate the existence of a "Yahweh­
alone" movement led by prophets and others which originated as early as
the time of Elijah in the ninth century, and which later became generally
accepted. Deuteronomy was clearly a key factor in this development.
Equally abhorrent to the Deuteronomist are magical practices such as
divination and necromancy (Deut. 1 8 : 1 0- 1 4) and objects suggestive of
pagan worship ( 1 6:2 1-22) . But it is the former inhabitants of the land (e.g.,
7: 1 -7, 16; 20: 1 5 - 1 8) who are the principal objects of abhorrence. The people
are commanded to destroy them, not leaving a single person alive. This is
partly be�ause it is to Israel and Israel alone that God has given the land, but
mainly because it is these peoples who worship other gods and may persuade
Israelites to do the same. This commandment is an extraordinary one. Von
Rad saw it as part of an ancient tradition of "holy war" which is found in
other texts. As such it would at least be understandable - ifhardly attractive
to the modern reader - in the context of a speech by Moses to Israel on the
eve of the invasion of the land. But in the context of the book in its present
form, it makes no sense if taken literally.
Whom could the Deuteronomist have meant in referring to these
peoples, who had long ago faded from the scene in his day? Once it is
granted that they stand for people of his own time who were polytheists
or idolaters, it is not difficult to identify them - if not with precision -
as non-Israelite elements of the community with whom it is forbidden to
maintain friendly relations. In a postexilic situation in which Israel was
no longer master of the land - though they still hoped to recover it -
such foreign elements, dangerous to the true faith, were certainly present.
The command to exterminate them, a reminiscence of a former tradition,
is obviously rhetorical and theoretical. But it witnesses to the fierceness of
the Deuteronomic conviction that the land is a gift from Yahweh to Israel
alo ne, and that the true faith must be preserved and protected at all costs
from contamination by foreign idolaters - for that faith is the condition
for the fulfillment of the promises.

92
The Book ofDeuteronomy

Despite this ruthlessness in the Deuteronomist's view of God, his


understanding of God's inner nature is a refined one. For the Deuter­
onomist God is immanent, in that he is active in Israel's history, and far
from remote. In Deut. 4:7, for example, Moses asks, "For what other
great nation has a god so near to it as Yahweh our God is whenever we
call to him?" Yet God is also transcendent. The Deuteronomist's view is
identical with that of Deutero-Isaiah in insisting that Yahweh is essentially
different from the gods of the other nations, which are "made by human
hands, objects of wood and stone that neither see, nor hear, nor eat, nor
smell" (Deut. 4:28; cf. 28: 36, 64; 29: 1 7) . Yahweh is invisible. The people
are told that when he addressed the people at Horeb, "You heard the
sound of words but saw no form: there was only a voice" (4: 1 2) . This
doctrine of the invisible God was probably not an original one. The
principle that Yahweh should not be represented by an image seems to
have been very ancient. But Deuteronomy makes the point with partic­
ular clarity.
Another aspect of Deuteronomy's doctrine of God's transcendence
is supposedly what has been called its "name theology." This term refers
especially to the manner in which God is present in the temple. In other
religions of the ancient Near East a temple was literally the deity's house,
in which he or she resided. In the Ugaritic epic of Baal, for example, Baal's
desire to build a "house" for himself is a major theme. In the Old Testament
too, the usual Hebrew term for the Jerusalem temple is beth yhwh,
"Yahweh's house." It is also sometimes called his "palace" (hekhal). As in
the surrounding religions, it is the place where he "lives" (yashabh). It is
a striking fact that nowhere in Deuteronomy is this expression used of
Yahweh's relationship to the temple, nor is the temple referred to as his
house, except very occasionally in the laws - never in the specifically
"theological" chs. 1 - 1 1 . Rather, the Deuteronomist has his own way of
describing it. For him, the one place in which Israel is to offer its public
worship is regularly referred to simply as "the place that Yahweh your God
will choose out of all your tribes to put his name there,'' or "to cause his
name to dwell there" ( 1 2:5ff.) . (Although it is not named in the book
because the Mosaic fiction requires that it has not yet been chosen by
Yahweh, it is probably to be assumed that the Jerusalem temple is meant.)
There has been much discussion of what is meant in Deuteronomy
by the "name" of Yahweh (see T. N. D. Mettinger, The Dethronement of
Sabaoth) . The notion of a deity's name as somehow distinct from the deity
in question yet still closely related to him or her is found in other ancient

93
INTRODUCTION TO THE PENTATEUCH

Near Eastern religions, and in some cases the name comes close to being
a separate deity itself In the Old Testament it is frequently used simply
as a synonym for God himself; and some scholars deny that the choice of
this word in Deuteronomy has theological significance - that there is no
"name theology" in this book.
Although the "name of Yahweh" which dwells on the earth in
Yahweh's chosen place undoubtedly stands for the presence of Yahweh
with his people, it is in heaven that he permanently resides. It was from
heaven that he spoke to his people in the past to discipline them (4:36) ,
and in ch. 26 the Israelite who has has brought the tithe of his produce
to offer it "before Yahweh" (v. 1 3) prays: "Look down from your holy
habitation, from heaven, and bless your people Israel and the ground that
you have given us" (v. 1 5) . It remains probable that the terminology used
in Deuteronomy marks a movement away from a view that Yahweh, who
is "God in heaven above and on the earth beneath" (4: 39) , might neverthe­
less be limited by a concept of holy space, while yet insisting strongly that
he is active .o n earth in his bounty toward Israel.
Two central aspects of the doctrine of God in Deuteronomy are
represented by the words "love" and "choose." Israel has been singled out
from the nations of the world in that Yahweh has "set his love" upon
Israel, and that he has chosen Israel to be his special people. These themes
run through the whole book, and they are closely related. In 4:37 the
people are told: "Because he loved your fathers and chose their children
after them, he brought you out of Egypt by his great strength." But it is
also made clear that this love and this selection of Israel are not due to
any merit or greatness of theirs. It was not because Israel was a great and
numerous people that God chose it. · Indeed, since that choice was made
Israel had not shown itself worthy of it, but had been continually re­
bellious.
Great emphasis is laid, especially in chs. 4ff., on what God has done,
and on what he will do for his people in the future if they will now be
obedient to his laws. No reason is given for God's choice of them. His
love was antecedent and presumably needed no explanation. It is, however,
traced back to the time of the patriarchs of Genesis - to the promise or
oath that God had made to Abraham to give the land to his descendants
(e.g., 1 :8 ; 6: 1 0; 9:5; 29: 1 3; 30:20; 34:4) . The first example of this love in
action was the Exodus from Egypt, the first demonstration of what God
could do for Israel. This choice of Israel was confirmed and reinforced by
the encounter with him at Horeb, where they became his holy people and

94
The Book ofDeuteronomy

where the relationship was established and the covenant made, with its
obligation to obedience to God's commands. God in turn then demon­
strated his love and care for his people by leading them safely through the
desert for forty years to the place where Moses was now addressing them,
having dealt on the way with those nations that had opposed their progress.
But these historical retrospects do not constitute the central message
of the book. They are essential to it in that they set out the character of
the God who has chosen Israel and the proofs that he has already given
them of his faithfulness despite their own unworthiness to receive these
gifts. However, the emphasis of the book is not on the past but on the
future. The chapter of lsrael's past history is at an end, and they are asking
what they may now expect from this God. The answer is that God will
now extend his bounty to them and give them what has so long been
promised: the possession of the land of Canaan. The richness and material
abundance of that land are described in detail (8 :7-9) ; and it is promised
that "You shall eat your fill and bless Yahweh your God for the good land
that he has given you" (v. 1 0) . The book emphasizes that God has even
prepared the land in advance for them. It contains cities, well-equipped
houses, cisterns, vineyards, and olive groves built and planted by the
Canaanites whom they are to dispossess. But the people are warned of the
danger that in their enjoyment of these things they may forget what God
has done for them (6: 1 0- 1 2) .
Total obedience to God's commands from now on is essential to the
people's well-being, and the commandments which occupy almost half the
book are set out in great detail in the laws of chs. 1 2-26. This loving God
can and will take back all that he has given them if they are disobedient
- and much more than that. While 28 : 1 - 1 4 sets out the benefits that
they will continue to enjoy if they are obedient, the much longer 28: 1 5 -68
describes the horrible fate that disobedience will bring on them. They will
suffer crop failure and famine, pestilence and other diseases, despair, defeat
and conquest by enemies, expulsion from the land and scattering among
the nations, and even a return to Egypt and to renewed slavery there.
One of the most frequently used words in Deuteronomy is "today."
It occurs almost a hundred times, most frequently in the phrase "the
commandment that I am commanding you today." This usage is of great
significance for the theological understanding of the book. Basically it is
used to indicate the crucial nature of the moment at which the covenant
at Horeb is established and the people are summoned to obedience. So in
1 1 :26-28 Moses states, "See, I am setting before you today a blessing and

95
INTRODUCTION TO THE PENTATEUCH

a curse: the blessing, if you obey the commandments of Yahweh your God
that I am commanding you today; and the curse, if you do not obey the
commandments of Yahweh your God, but turn from the way that I am
commanding you today." The people have a choice, and the consequences
of the choice that they make are made clear to them. But the readers of
the book in every subsequent generation are intended to understand the
words as applying to themselves. When in 5 : 3 Moses affirms, "Not with
our ancestors did Yahweh make this covenant, but with us, who are 'all of
us here alive today," the contextual reference is to the preceding generations
as compared with the present one. But the readers of the book would
understand that the "with us" applied to them. They also are the heirs of
the covenant with its obligations - the covenant is for all subsequent
generations. There is an "actualization" here. The reader is to understand
that that moment in the past and the present moment are one. To an exilic
or postexilic readership this was, moreover, true of the whole book. They
too were to live in hope of the (re-) possession of the land and the promises,
the call to obedience, and the warnings about the consequen�es of disobe­
dience applied to them as much as to their ancestors. In much the same
way have both Jewish and Christian readers continued to believe.
The teaching of Deuteronomy about Israel and its relation to God
has to some extent been dealt with already. It follows from Deuteronomy's
understanding of the nature of God, and is expressed mainly in the use .
of two frequently recurring words: "choose" and "covenant."
The notion of choice, with its implication of freedom to determine
one's own actions or mode of life, is one which is characteristic of Deuter­
onomy. God chooses, but human beings also have that freedom. Even a
runaway slave may choose where he wishes to live (23: 1 6) . In 30: 1 9, in
one of the passages which set out the alternatives of obedience and dis­
obedience and their respective consequences, the audience is urged (in an
imperative in the singular, apparently addressed to the individual present
in the crowd) to "choose life." But in the great majority of cases it is God
who, in his sovereign freedom, exercises his choice. He chooses the place
where he is to set his name; it is also God who chooses the kings to rule
over the nation ( 1 7: 1 5); and he chooses, or rather has chosen, Israel "out
of all the peoples" to be a "holy people" and his "treasured possession''
(7:6; 14:2) .
The use o f the verb "to choose" of Yahweh's relationship to the people
as a whole in Deuteronomy is particularly significant. In the Deuter­
onomistic History (e.g., 1 Sam. 1 0:24; 1 6: 8- 1 0; 2 Sam. 6:2 1 ; 1 6: 1 8) it is

96
The Book ofDeuteronomy

used of Yahweh's selection of Saul and of David. The verb is not so used
of later Davidic kings, but the idea appears in Hos. 8:4 with regard to
kings of northern Israel, and in the only passage that refers to kings in
Deuteronomy (Deut. 17: 1 4-20) it is used in the same way. But it is in
Deuteronomy that the verb is first used of Yahweh's choice of the whole
people of Israel, and used frequently and emphatically. This book - in
common with some other late (exilic and postexilic) books such as Second
Isaiah (Isa. 55:3-5) and Ezekiel, in whose blueprint of the future (Ezek.
40-48) there is only a "prince" with very limited functions - has, as it is
often expressed, "democratized" the concept of divine choice. The ancient
Near Eastern belief that divine choice and a special relationship with the
gods was confined to kings, who as sacral figures were regarded as semi­
divine and often as standing in a filial relationship to gods - a belief
which has been to a large extent adopted with regard to the Davidic dynasty
(cf. Ps. 2:7) - was now transferred to the whole nation, which had thus
become a "holy people" (Deut. 7:6; 1 4:2, 2 1 , etc.) . Corresp0ndingly,
although Yahweh is ilot directly called Israel's father except in the poetical
32:6, in 1 : 3 1 and 8 : 5 he is likened to a father; and Israelites are commonly
called "brothers" throughout the book.
The other term which is most frequently used to describe Israel's
relationship with Yahweh is "covenant" (berith) . In almost every case, the
covenant referred to in Deuteronomy (e.g., 5 :2) is that which had been
made at Horeb (Sinai) . In fact the entire book may be described as an
exposition by Moses of the contents and meaning of that covenant. In ch.
29, however, there is reference to a second covenant now made, after the
proclamation of the laws, in the land of Moab, a covenant made " in
addition to the covenant that he had made with them at Horeb" (v. 1 ) .
The significance 0 £ and the necessity for, this second covenant has never
been satisfactorily explained. Deut. 29:9 commands the people to "dil­
igently observe the words of this covenant, in order that you may succeed
in everything that you do."
This second covenant is found nowhere else in the Old Testament.
Various attempts at explaining it have been made, although the commen­
taries have tended to pass over the problem in silence; many scholars (e.g.,
Samuel R. Driver and von Rad) do, however, recognize the special char­
acter of this section of the book. It is widely held that ch. 29, perhaps
together with ch. 30 or with other subsequent chapters as well, constitutes
a kind of supplement or appendix to the laws. Klaus Baltzer ( The Covenant
Formulary, 34ff.) argued that chs. 29-30 are couched in the form of a treaty

97
INTRODUCTION TO THE PENTATEUCH

(see above) , and they may in some sense have constituted an alternative
to the postulate of the rest of the book, that it is the Sinai (Horeb) covenant
which is in question. Others see the second covenant as simply a rein­
forcement or confirmation of the first. A. D. H. Mayes suggested that
since new laws had been promulgated, a new covenant was needed to
embrace them. Norbert Lohfink linked it to the appointment of Joshua
as Moses' successor as noted in ch. 3 1 ; such an appointment would require
a new covenant in which the promise of obedience was renewed. Whatever
may be the solution of the problem, this second covenant detracts from
the "once-and-for-all" character of the Horeb-Sinai covenant as presented
in the rest of the book, and must be regarded as an unexplained anomaly.
Central to the theology of Deuteronomy is the theme of the land,
which was promised to the ancestors and is now to be given to Israel for
immediate occupation - though on condition of obedience to God's laws,
and on the understanding that it will be taken away again in case of
apostasy. As has been pointed out in chapter 4 above, this theme was of
obvious relevance to a postexilic generation which had once possessed, but
then lost, the land. Deuteronomy knows, and frequently alludes to, the
promises to the patriarchs of Genesis. But its obsessive preoccupation with
this theme and the atmosphere of expectancy produced by the imminence
of the possession of the land exceed anything in Genesis. Only in this
book is there a fully developed theology of the land, in which the entire
future of the nation has now been · concentrated.
The question of the antiquity of this tradition of the promise of the
land is a disputed one. Claus Westermann, in a study of the promises in
Genesis, argued that the promise of the land is among the latest develop­
ments of the theme and had its origin at a late period in the history of
the religion of Israel. However that may be, it is only in Deuteronomy
that the theme reached its fullest exposition. It is significant that only here
is it closely tied to the promulgation of the laws. Deuteronomy is a book
which is concerned with a particular present and a particular future. The
laws are not given in a temporal vacuum; it is in the land which Israel is
about to occupy that they are to be obeyed.
What Deuteronomy has to say about worship - or at least the way in
which it expresses it - is revolutionary. The sacrificial cult for Deuteronomy
is an expression of love for God and of gratitude for what he has done and
is do1.ng for his people. Other reasons for offering sacrifice that are found
elsewhere in the Old Testament are passed over in silence here. It may seem
strange that love for God in Deuteronomy is a duty - that the Israelites are

98
The Book ofDeuteronomy

commanded to love him (6: 5). Some scholars have suggested that the word
"love" here is a technical term taken over from the vassal treaties of the
ancient Near East, where it means no more than to remain faithful to an
overlord rather than to have loving feelings toward him; but this is hardly
the case. In Deuteronomy the love is to be "with the heart and inner being
(Hebrew nephesh) and strength" (6:5). Deuteronomy does not see this
command to love God as a contradiction. It is in obeying God's command
that one most truly expresses one's love for him.
Deuteronomy is concerned in its laws about worship with purity
and singlemindedness. Earlier Israel had worshipped at a variety of places
throughout the land, and there was no control over the purity of the
worship offered there, even supposing that it had been offered solely to
Yahweh. Now, in a world teeming with worship offered to other gods and
with superstitious, magical practices, there must be a total ban on such
behavior. All idolatrous worship must be rooted out and its practitioners
destroyed (6: 1 9; 7:5, 16, 22-25; 9:3; 1 2 :2-3) . This language can hardly,
however, be other than symbolic; the readers of Deuteronomy were in any
case in no position to carry out such destruction. But within Israel itself
no idolatrous or magical practices were to be tolerated (4: 1 6; 5 : 8 ; 1 6:2 1 -22;
1 7:2-7; 1 8 : 1 0- 1 4) .
It was in order to secure purity of worship that public sacrificial
worship was now to be restricted to one place (ch. 1 2) . Whether or not
attempts had been made earlier, by reforming kings such as Hezekiah and
Josiah, to make Jerusalem the sole place of public worship, now it was
absolutely commanded that it be offered only in "the place that Yahweh
your God will choose to put his name there." Again, whether or not this
was originally intended to refer to Jerusalem, it came to be so . understood.
In . order to make this law a practicable one, a distinction was made between
animals offered for sacrifice and animals slaughtered for food: the latter
could be slaughtered at home and eaten, provided that the provision
against eating the blood was observed (12: 1 5 - 1 6) . This also was an entirely
new provision. Some have described it as a "secularization" of sacrifice,
but this is a misunderstanding.
Deuteronomy more than any other Old Testament book concerns
itself not only with the obligation to worship and the rules for doing so,
but also with the subjective aspect of worship - with the feelings of the
worshipper and the spirit in which he or she worships. There is often a
personal note in those passages that deal with public worship. It is regularly
stated that these are occasions for rejoicing (e.g., 1 2:7, 1 2, 1 8; 14:26;

99
INTRODUCTION TO THE PENTATEUCH

1 6: 1 1 , 14; 26: 1 1 ; 27:7) . The rejoicing is closely connected with another


feeling, that of gratitude to God.
But there are also times for the individual to draw near to God in
prayerful reflection, not necessarily associated with public worship. An
individual in distress is assured that even when he has seriously transgressed
he can find God in his heart: "You will find him if you search after him
with all your heart and soul" (4:29) . In other passages the nearness of the
worshipper to God is emphasized: "For what other great nation has a god
so near to it as Yahweh our God is whenever we call to him?" (4:7; cf.
30: 1 1- 1 4) . There is nothing automatic or impersonal in the worship of
God as described in Deuteronomy.
Both instruction in the laws and participation in worship were
matters for family activity in Deuteronomy. In 6:6-9 the people are to
teach the laws to their children in the context of family life - at home,
when away from home, when getting up in the morning, and when going
to bed. In 6:20-25 the child who asks about the meaning of the laws is
to be told how God saved Israel from slavery in Egypt and then gave them
the laws for their own well-being. This personal and family note is found
also in ch. 12, in the law about worship: the whole family shares in the
rejoicing and the feasting at the place designated by God for worship.
Further treatment of the laws of Deuteronomy will be found in
chapter 7 below. Here they will be considered in their theological aspect
- that is, as they directly reflect Deuteronomy's main theological con­
cerns. These may be summarized in the phrase "one God, one people, one
land, one place of worship." This insistence on oneness has been called
"centralization''; a better description might be "unification." At the same
time, there is a concern for the individual in Deuteronomy's theology, and
this finds expression also in the laws. (It should be remarked at this point
that many of these provisions were not entirely new, although they often
appear here in a more highly developed form than in earlier legislation.)
As it gives practical expression to the oneness of the deity in its laws
concerning worship, so Deuteronomy gives practical expression to the
oneness of the people in prescribing a unified political and social order. It
lays down rules for the judiciary and the administration of justice and for
the selection, qualifications, and functions of the leading figures in the
united community: king, judges, priests and Levites, and prophets. As has
already been pointed out, the book also makes certain regulations for the
family.
The law of the king is contained in 17: 1 4-20. It is of interest to note

1 00
The Book ofDeuteronomy

that there is no reference to a king oflsrael anywhere else in the Pentateuch.


Evidently the author of this passage took it for ,granted that, like other
nations (v. 1 4), Israel would. again have a monarchy; but this monarchy
would reflect the theological principles of the book Future kings would
have none of either the sacral qualities or the political powers which earlier
kings had claimed for themselves. Indeed, nothing at all is said about either
political or military authority. These kings must be approved or "chosen"
by God (v. 1 5) . As members of the community ("brothers") , they must
not claim superior status (v. 20) and must not emulate their predecessors
in wealth and grandeur or in the possession of numerous wives (vv. 1 6- 1 7) .
They are to possess a written copy of "this law" and study it regularly with
the help of the Levites. Above all, they must scrupulously obey all the laws
in this book (vv. 1 9-20) . In other words, the ultimate authority in Israel
is the law, not the king, who may be presumed to be no more than a kind
of superior civil servant.
Deuteronomy is passionately concerned about justice (Hebrew tsedeq,
mishpat) : "Justice, and only justice, you shall pursue" ( 1 6:20, which makes
this a condition of living and prospering in the land) . This follows from
the doctrine of Israel as a community of "brothers" equal before God. In
1 6: 1 8-20 the administration of justice is put on a systematic basis by the
command to appoint j udges and "officials" "throughout your tribes, in all
your towns" who will be impartial and incorruptible, so that no one shall
be deprived from seeking justice in the courts. (A similar action is claimed
in 2 Chr. 1 9: 5-7 as an innovation by King Jehoshaphat ofJudah.)
But it is not only in the law courts that individual rights are to be
protected, nor does the law stop at legal rights. Love is also required. In
particular, both protection and material help are to be given to individuals
who most need it: orphans, widows, strangers; slaves, poor laborers, and
the poor in general (Deut. 1 5 :4- 1 1 ) . Deut. 1 5 :4 even seems to assert that
owing to such generosity poverty will cease to exist: "There will be no
poor among you, " although v. 1 1 must recognize that this is an unattain­
able ideal.
These laws (some of which are not peculiar to Deuteronomy) have
been described as "humanitarian''; but this modern term is not entirely
appropriate. It is rather a question of gratitude. The Israelites are to put
themselves in the place of the unfortunate and remember how they as a
people were once in a similar helpless plight: "Remember that you were
a slave in Egypt and Yahweh your God redeemed you from there; therefore
I command you to do this" (24: 1 8); ''You shall love the stranger, for you

10 1
INTRODUCTlON TO THE PENTATEUCH

were strangers in the land of Egypt" ( 1 0: 1 9) . The implication is that the


memory of God's love for Israel in the past should color the Israelite's life
in the present and the future.
Deuteronomy's view of the priesthood is not totally revolutionary.
Most of its functions are identical with those prescribed in Exodus­
Numbers; but the emphasis has significantly shifted. It is notable that
Deuteronomy devotes little space to the details of sacrificial worship com­
pared with Exodus-Numbers, although it may be presumed that the priests
continued to preside over it in a similar way. There also seems to have
been a widening of the membership of the priestly class, since several texts
appear to identify the priests with the Levites, who in the legislation of
Exodus-Numbers occupy an inferior position.
In Deuteronomy the non-sacrificial functions of the priesthood are
greatly emphasized, and in fact the priests have become the supreme civil
authority. They are associated with the civil judges in 17 :8- 13 and 1 9: 1 7
i n giving decisions o n difficult cases and i n serious cases such as that of
murder. However, they appear to outrank the judges, since they are men­
tioned first when the two functionaries are named together. In 1 7: 1 2 it is
the priests alone who are mentioned as imposing the death penalty, and
in 2 1 : 5 cases of dispute and assault are dealt with by them alone. It is they
who instruct the people about the law. They are concerned in providing
a copy of the law for the newly appointed king, who seems to be under
their direction ( 1 7: 1 8) . It is they who assemble the people for the public
reading of the law every seven years (3 1 : 10- 1 3) . The priests address the
people before battle with the assurance that God will fight for them
(20:2-4) . They bless the people (2 1 : 5; 27: 1 2), but also proclaim a series
of curses against those who transgress (27: 1 3) .
Prophets, as has been noted above, are mentioned in Deuteronomy
only in two extended passages ( 1 3: 1-5; 1 8 : 1 5-22) , and also in a single
verse (34: 1 0) . In 34: 10 Moses himself is said to be a prophet, the greatest
who has yet arisen in Israel (c£ also 1 8: 1 5 , 1 8) . There is no evidence that
Deuteronomy envisages the establishment of an actual prophetic institu­
tion. The statement by Moses in 1 8 : 1 5 that God "will raise up for you a
prophet like me" probably refers to the future appearance of prophets from
time to time rather than to the expectation of a single, particular prophet.
Clearly prophecy was a current phenomenon at the time when these
passages were written, but it is difficult to say what were the ,particular
historical circumstances that lay behind these particular laws. Von Rad in
his commentary could offer no answer to this question; he remarked that

1 02
The Book ofDeuteronomy

"It is not easy to say what kind of prophets the preacher in Deuteronomy
has in mind" (Deuteronomy, 97) . Both passages have affinities with utter- .
ances of Jeremiah, who had to struggle against the false prophets of his
time, in the last days of the monarchy before its fall in 587. These false
prophets included those who, in opposition to Jeremiah, attempted to
buoy up the confidence of their audience with promises that the Baby­
lonians would not tapture Jerusalem, and those who frankly prophesied
in the name of other gods than Yahweh Qer. 14: 1 3- 1 6; 23:9- 1 7; 27:9- 1 8 ;
28:8-9). In Deuteronomy the warnings against prophets who told lies and
who tried to persuade their audience to turn to the worship of other gods
are no less vehement; but the polemic is more generalized, and directed
more toward an undefined future than to an actual situation. It is, however,
interesting that in both books the crucial question is raised --:- and hardly
satisfactorily answered - how false prophets may be identified and . dis-
tinguished from true prophets.
The functions· of the true prophet in Deuteronomy are to act as a
spokesman of Yahweh, who has "put his words in his mouth," and to
predict the future. But in this book the words of God are primarily the
words of the law; the aim of the false prophet is to turn his hearers away
from "the way in which Yahweh your God commanded you to walk"
(Deut. 1 3:5). It is Moses, the spokesman of the law, who is the greatest
of the prophets (34: 10) . So prophecy is not, as in the prophetic books of
the Old Testament, to be so much an ad hoc message from God but . a
proclamation of the law, which is the supreme religious authority.
A significant innovation of Deuteronomy is that its laws are to be
written in a book (28 : 5 8, 6 1 ; 30: 1 0; 3 1 :24), which is to be kept beside
the ark "as a witness against you'' (3 1 :26) and to be read publicly at
seven-year intervals (3 1 :9- 1 1 ) in an assembly of the whole people. The
only other reference in the Pentateuch to such a "book'' is in Exod. 24,
where it is stated that Moses "wrote down all the words of Yahweh" and
then, in the context of a covenant-making ceremony, "took the book of
the covenant, and read it in the hearing of the people, " who solemnly
promised to obey its laws (Exod. 24:7-8) . Some scholars, including von
Rad, saw here an ancient, premonarchical, tradition; but others (e.g.,
Nicholson, Deuteronomy and Tradition, 70-72) consider this passage to be
an addition to the book inserted by one who followed the tradition of
Deuteronomy.
As has been already noted, Deuteronomy . contains the most com­
prehensive body of laws in the Pentateuch. It is clearly intended to be
I.

103
INTRODUCTION TO THE PENTATEUCH

consulted for guidance on many aspects of daily life, in sharp contrast


with the laws of Leviticus, which are very restricted in scope and mainly
concern the functions of the priesthood. Also, whether the seven-year
public reading ever took place or not, Deuteronomy makes the laws
available to ·the whole people. It marks the beginning of that devotion to
the Torah as the authoritative word of God which has remained a feature
ofJudaism to this day, and has also been a formative influence on Christian
views about the authority of Scripture.
The sryle and language of Deuteronomy are extremely distinctive,
though they are also characteristic of the so-called Deuteronomistic litera­
ture, especially the Deuteronomistic History, and also parts of some of the
prophetic books. So distinctive are they that they are immediately recogniz­
able even to readers who know no Hebrew. Here are some of the most
distinctive phrases: "that your days may be long . . . "; "which I am com­
manding you today"; "so you shall purge the evil from your midst"; "the
place which Yahweh your God will choose"; "do what is right/ evil in the sight
of Yahweh"; "the priests and the Levites"; "with all your heart and with all
your soul"; "the land that you are entering to occupy" (see the more complete
list in S. R Driver, Introduction to the Literature of the Old Testamen� .
Another distinctive feature of the book is of course the homiletic or preach­
ing style, especially in the first eleven chapters; there is a sense in which the
whole book is a sermon placed in the mouth of Moses.

For Further Reading

Baltzer, Klaus. The Covenant Formulary in . Old Testament, Jewish, and Early
Christian Writings. Philadelphia: Fortress and Oxford: Blackwell,
1 971 . (First published in German, 1 964.)
Blenkinsopp, Joseph. The Pentateuch: An Introduction to the First Five Books
·ofthe Bible. New York: Doubleday and London: SCM, 1 992, ch. 6.
Clements, Ronald E. Deuteronomy. OTG. Sheffield: JSOT Press, 1 989.
Driver, Samuel R A Critical and Exegetical Commentary on Deuteronomy.
2nd ed. ICC. Edinburgh: T. & T. Clark and New York: Scribner's,
1 902.
--- . Introduction to the Literature ofthe Old Testament. 8th ed. Edin­
b urgh: T. & T. Clark, 1 909.
Lang, Bernhard. Monotheism and the Prophetic Minoriry. Sheffield: Al­
mond, 1 98 3.

104
The Book ofDeuteronomy

McCarthy, Dennis J. Old Testament Covenant: A Survey of Current Opin­


ions. Richmond: John Knox and Oxford: Blackwell, 1 972.
--- . Treaty and Covenant. Analecta Biblica 2 1 . Rome: Pontifical Bib­
lical Institute Press, 1 963. Rev. ed., 1 978.
Mayes, A. D. H. Deuteronomy. NCBC. London: Marshall, Morgan &
Scott, 1 979, and Grand Rapids: Wm. B. Eerdmans, 1 98 1 .
Mendenhall, George E. "Covenant Forms in Israelite Tradition," BA I 7
( 1 954) : 50-76. Repr, in The BA Reader 3, ed. E. F. Campbell, Jr.,
and David Noel Freedman. Garden City: Doubleday, 1 970, 25-5 3 .
Mettinger, T. N. D . The Dethronement of Sabaoth. Coniectanea Biblica,
Old Testament 1 8. Lund: Gleerup, 1 982.
Nicholson, Ernest W Deuteronomy and Tradition. Philadelphia: Fortress
and Oxford: Blackwell, 1 967.
--- . Exodus and Sinai in History and Tradition. Richmond: John Knox
and Oxford: Blackwell, 1 973 .
Noth, Martin. The Deuteronomistic History. ]SOT Supplement 1 5 . Shef­
field: JSOT Press, 1 98 1 . (Translated from the 2nd German ed.,
1957.)
von Rad, Gerhard. Deuteronomy. OTL. Philadelphia: Westminster and
London: SCM, 1 966. (First published in German, 1 964.)
--- . "The Form-Critical Problem of the Hexateuch," in The Problem
ofthe Hexateuch and Other Essays. Edinburgh: Oliver & Boyd, 1 966,
1 -78. Repr. Philadelphia: Fortress and London: SCM, 1 984. (First
published in German, 1 93 8 .)
---. Studies in Deuteronomy. SBT, 1 st series 9. Chicago: Regnery and
London: SCM, 1 953. (First published in German, 1 948.)
Rendtorff, Rolf. The Old Testament: An Introduction. London: SCM, 1 98 5 ,
and Philadelphia: Fortress, 1 986. (First published in German, 1 983.)
Smith, George Adam. The Book ofDeuteronomy. CBSC. Cambridge: Cam­
bridge University Press, 1 9 1 8 .
Smith, Mark S. The Early History of God: Yahweh and the Other Deities in
Ancient Israel San Francisco: Harper & Row, 1 990.
Smith, Morton. Palestinian Parties and Politics That Shaped the Old Testa­
ment. New York: Columbia University Press, 1 97 1 . 2nd ed., London:
SCM, 1 987.
Weinfeld, Moshe. Deuteronomy and the Deuteronomic School Oxford:
Clarendon Press, 1 972. Repr. Winona Lake: Eisenbrauns, 1 992.
Welch, Adam C. The Code of Deuteronomy. London: Oxford University
Press, 1 924.

105
INTRODUCTION TO THE PENTATEUCH

Deuteronomy: The Framework to the Code. London: Milford,


1 932.
Westermann, Claus. The Promises to the Fathers. Philadelphia: Fortress,
1 980. (First published in German, 1 976.)

1 06
CHAPTER 7

The Laws

A LL THE LAWS in the Pentateuch have been inserted into narrative


.1"\..contexts to which they did not originally belong. (The laws of Deuter­
onomy only appear to be an exception. Although it is true that in this
case the laws came first and the book later, the fact remains that they are
presented in the context of the wider account of Moses' life.) They belong,
in fact, to a quite distinct genre, one which is found in abundance in the
ancient Near Eastern documents that have come down to us.
Martin Noth in his work on the laws of the Pentateuch was well
aware that when used of the ancient Near Eastern world the term "law"
is a somewhat misleading one. (He regularly put the word in quotation
marks to indicate this fact.) None of the several "law codes" in question
(see translations in ANET, 1 59-1 98, and its Supplement, 523-26) is, or
was intended to be, a "statute book'' on the basis of which the king or the
j udges appointed by him were required to make their decisions in the law
courts. In view of the similarities between the laws of the Pentateuch and
these other collections of laws, it is necessary to discuss briefly the nature
and purpose of the ancient Near Eastern laws before considering how far
the Old Testament laws corresponded to or differed from them in this
respect.
These "law codes" - Sumerian, Babylonian, Assyrian, and Hittite
- cover a long period, ranging from the late third millennium B.C. to
the Neo-Babylonian period, which ended in the late sixth century. They
thus attest to a single common, long-standing legal tradition in the ancient
Near East which developed gradually over the centuries. Unfortunately
there is no consensus of opinion about the purposes of these codes. It has
been suggested that they were intended as a kind of royal apologia, praising

1 07
INTRODUCTION TO THE PENTATEUCH

parti eular kings for their concern for j ustice. But not all of them have been .
found --,... as was the Code of Hammurabi � on p ublic monuments, where
they might have been inscribed for perusal by a (literate) public. Others
have been found on small cuneiform clay tablets, in a form more suitable
for deposit in archives or lib raries, and have no prologue or epilogue
explaining their purpose.
Another theory is that the codes were based on actual legal judgments
and were thus available to j udges for consultation in difficult cases. Yet
very few of the extant legal - documents, which are very numerous, ever
refer specifically to the codes. A third view is that the codes are purely
literary documents created by learned scribes, composed out of a love of
systematization and classification in general, and not intended for practical
use. According to some scholars, this may be suggested by the way · in
which the texts sometimes indulge in elaborations of what appear to be
no more than hypothetical cases. In fact, legal judgments were probably
made on an oral basis, though decisions were often recorded in writing.
Whatever may have been the reason for these legal compilations (and
they may not all have been compiled for the same purpose), there can be
no doubt that they refer - whether in reality or hypothetically - to the
laws of the state, that is, of the king; who was the supreme political
authority and the chief judge, responsible for law and order. Although
(e.g., in the Code of Hammurabi) the king may be represented as receiving
the laws from a god, in practical terms the king was the lawgiver.
It is a remarkable fact that -nowhere in the Old Testament is the king
represented as having anything to do with the making of laws. Kings were
the chief judges as elsewhere, and are pictured as making oral decisions,
though not specifically with reference to written laws (e.g., 2 Sam. 14: 1 - 1 1 ;
1 5 :3-5; 1 Kgs. 3 : 1 6-28) . Also, according to 2 Chr. 1 9:8-1 1 Jehoshaphat
of Judah appointed j udges, drawn from the priests, Levites, and heads of
families, to give j udgment on disputed cases which the local courts had
been unable to solve. But according to the Deuteronomistic History, the
kings were themselves in no way above the law but subject to it and judged
by it, and were frequently condemned by prophets for breaches of it. In
Deut. 1 7: 1 9 it is - explicitly stated that the king has to obey the (Mosaic)
law. Admittedly, · one cannot be certain that the Deuteronomistic literature
has correctly represented the state of affairs during the . monarchy; and
specific references to "the law'' as the standard by which kings are judged
occu.t not in the narrative sections of the Deuteronomistic History but in
the editors' comments on and assessments of the reigns of particular kings.

1 08
The Laws

Nevertheless, it is unlikely that the Deuteronomistic historians should have


so altered their narrative sources as to have expunged any references to
royal legislation which might originally have stood there.
A unique characteristic of the Pentateuchal laws is that they are
represented as having been communicated to the whole people directly by
God in connection with the making of the covenant - mostly through
the mediation of Moses, who simply repeated to the people the words
which God had spoken to him, but in some cases with no mediation at
all. In a few cases in Leviticus and Numbers laws are addressed to Moses,
to Moses and Aaron, or to the priests. However, their administration was
necessarily in the hands of particular leaders whose identity varied from
time to time with changes in the political conditions under which the
nation lived. Many of the laws prescribe specific penalties for persons who
breached them, including the death penalty.
Lawmaking in Israel undoubtedly had a long history. Some laws may
have originated before a national state existed, as rules governing certain
aspects of behavior in a tribal society that were or were not allowable in
the interests of community solidarity. It has been suggested (e.g., by
Erhard S. Gerstenberger) that some types of Israelite law share a common
origin with early wisdom, especially those laws prohibiting such things as
theft, murder, and sexual transgressions which would disrupt tribal and
family harmony. Such rules would have been made under the authority
of heads of clans and families.
Later there developed a system of local officials who had authority
to decide cases and, when necessary, to exact penalties for breach of the
laws. Deuteronomy speaks of priests, judges, and officials in this connec­
tion; the complex system of laws about such things as sacrifice and purity
would naturally be in the charge of the priests. During the monarchy and
even later, the main judicial authorities appear to have been the elders or
heads of families who administered justice in the town gate; appeals to a
higher court seem to have been quite exceptional. In Deuteronomy the
elders deal with such cases as those of a · stubborn and rebellious son (Deut.
2 1 : 1 8-2 1 ) , and of a wife accused of not having been a virgin before
marriage (22: 1 3-2 1 ) , both ofwhich carried the penalty of death by stoning.
They also decided "civil" cases such as the refusal of a man to marry his
deceased brother's widow (25 : 5- 1 0) , and also of unsolved murders (2 1 : 1 -
6) . The institution of the elders as judges apparently served to remove such
j urisdiction from the sole hands of the head of a family, making it a
community matter. This continued in a somewhat different form until

109
INTRODUCTION TO THE PENTATEUCH

very late times, as is shown by such incidents in the New Testament as


the stoning of Stephen for blasphemy (Acts 7: 54-60), and of the woman
taken in adultery Qohn 8 : 1 - 1 1 ) , where elements of the population could
incite both elders and people to put supposed offenders to death by
stoning.
The judicial system in operation during the postexilic period is not
clearly understood; but clearly there will have been maj or changes under
Persian rule in Judah and with the breakdown of the whole political system
that had prevailed before the destruction. We are told that Ezra, who was
both a priest and "a scribe skilled in the law of Moses that Yahweh the
God of lsrael had given" (Ezra 7:6) , was sent to Jerusalem by the Persian
king Artaxerxes to "make inquiries about Judah and Jerusalem according
to the law of your God" (v. 14), and that he was given authority to appoint
magistrates and judges with the power to execute those who refused to
obey "the law of your God and the law of the king" (vv. 25-26) . We are
told also that Nehemiah, another emissary sent from Artaxerxes and who
is referred to as "governor," was also assiduous in seeing that the law was
enforced.
Apart from these scraps of information little is known about the
· administration of the law during the rest of the postexilic period. But it
is clear that during this time the attitude toward the law of Moses under­
went considerable change. Eventually, although many of its provisions
were now out of date and no longer literally applicable to the new condi­
tions of life, the law came to be regarded as a timeless entity which stood
at the center of the new Judaism and which was binding, in as far as it
could be observed, o n all Jews. But this obedience had now become an
individual obligation rather than a communal one.

Two Types of Law

One of the most important contributions to the understanding of the


nature and history of the legal portions of the Pentateuch was the obser­
vation made by Albrecht Alt (1 940) of the existence of two quite distinct
types of law that are found together in these collections. The distinction
between the two types, which he called "casuistic" and "apodictic, " is a
formol one; but Alt argued that the difference in form points to a difference
of origin and of (original) purpose. Later scholars have disputed some
aspects of Alt's formal classification, and the details of this discussion

1 10
The Laws

cannot be dealt with in this short survey; but there is still a general
agreement about the existence of these two main types. The clearest and
most frequent examples of the two are those which begin with the words
"If a man . . . " (casuistic) and those which begin "You shall (not) . . . "
( apodictic) .
The simplest kind of casuistic law in the Pentateuch begins by stating
a hypothetical - but entirely conceivable - specific case requiring a legal
decision and then goes on to prescribe the appropriate verdict. For ex­
ample:

If (or "when") a slaveowner strikes the eye of a male or female slave,


destroying it, the owner shall let the slave go, a free person, to compen­
sate for the eye. (Exod. 2 1 :26)
If a man is caught lying with the wife of another man, both of them
shall die (i.e., receive capital punishment) . (Deut. 22:22)

There are variations in the form in which the laws are expressed (e.g.,
some laws which are probably to be classed as casuistic begin with a
participle instead of with a conditional clause) . Further, in some cases (e.g.,
Exod. 2 1 :22-25; Deut. 22:23-27) a casuistic law is extended to cover
particular or extenuating circumstances in which the action is deemed to
have taken place and to vary the penalty accordingly. The similarities in
form and often of matters dealt with, though not usually of penalties
prescribed, between the casuistic laws of the Pentateuch and those of other
ancient Near Eastern, especially Mesopotamian, casuistic laws (e.g., the
Code of Hammurabi) leave no doubt that they belong to the same legal
tradition. This type of law is not unique to Israel. But equally, it is
improbable that direct borrowing from so far afield has taken place.
This legal tradition, stretching back to the third millennium, is of
course much older than any of the Pentateuchal laws; and it has been
suggested that Israelite casuistic law is in fact a Canaanite legacy. This may
at first seem highly improbable, since no law codes have been found in
any of the extant written material (mainly from Ras Shamra-Ugarit) from
that region. Indeed, no evidence has come to light which indicates that
law codes existed in Palestine or its environs before the appearance of
Israelite laws. Nevertheless, the theory of a Canaanite source is plausible.
The lack of relevant finds may be simply due to the chanciness of archae­
ological investigation, and Canaanite law codes may yet be discovered.
Canaan would be a much more probable source of an Israelite casuistic

111
INTRODUCTION TO THE PENTATEUCH

law governing daily life than distant Assyria or Babylonia; and it is widely
agreed that the Canaanite cities of the pre-Israelite period, with their high
culture and position in the Semitic world, should not have lacked a legal
system comparable with that of the surrounding peoples.
This "Canaanite" theory remains a strong possibility, whatever may
be the truth about the origins of the Israelite people - whether they
entered the land from outside or whether they were indigenous, having
once belonged to the civilized life of the Canaanite city-states but having
for some reason detached themselves from it. There can be no doubt that
the culture of the Canaanite city-states which eventually became incor­
porated into the Israelite state profoundly influenced Israel, and that there
were remarkable continuities in more than one cultural, as well as religious,
field. Such borrowing of case law would also account for the existence of
the two types of law postulated by Alt. The origin of the case law would
be quite different from that of apodictic law, which may be seen as having
developed from Israel's own unique tribal traditions.
The origins of these apodictic laws, however, are more obscure than
those of the casuistic ones. There has been much debate also about the
correctness of Alt's classification. The main characteristic of the apodictic
laws is that they are categorical, not dealing with specific hypothetical cases
but making an absolute demand. But, although he distinguished them
totally from the casuistic laws, Alt admitted that they are quite varied in
form. He divided them formally into four types.
1 . The participial form mentioned above, for example:

Whoever strikes [participle] father or mother shall be put to death.


(Exod. 2 1 : 1 5)

This type is now - partly because it prescribes the penalty - considered


by many scholars to be a casuistic type, or perhaps a mixture of the two
categories.
2. A second form is the curse. A prime example is Deut. 27: 1 5-26,
a series beginning:

Cursed be anyone who dishonors (participle) father or mother.

Here the penalty is not prescribed specifically, but is implied; however, the
death sentence is presumed to be carried out not by human hands but
directly by God himself.

1 12
The Laws

3 . A third major category is that which begins with, or contains, the


words "You shall . . . ," for example:

You shall not permit a female sorcerer to live. (Exod. 22: 1 8)


You shall not revile God, or curse a leader of your people. (Exod. 22:28)
Three times in the year you shall hold a festival for me. (Exod. 23 : 14)

4. Alt placed the Decalogue or Ten Commandments (Exod. 20:3-1 7;


Deut. 5 :7-2 1 ) in a separate category.
· It is now recognized that these different types of formation cannot
simply be classed as belonging to a single, "apodictic" category. Any
discussion of original purpose and function must take these differences
into account. Recent study of extant Near Eastern documents has shown
the need for modification of Air's view that there are no parallels at all
between the apodictic laws of the Pentateuch and these texts. For example,
some Near Eastern vassal treaties contain stipulations which correspond
in effect either to the Pentateuchal curses or to the prohibitions, although
their form does not exactly correspond (see, e.g. , George E. Mendenhall,
H.-J. Boecker, Dennis J. McCarthy) .
Alt held that the origin of the apodictic laws is a cultic one. This
view is supported, at least to some extent, by the fact that the series of
curses in Deut. 27: 1 5-26 is represented as taking the form of a cultic
ceremony in which the people respond to each curse with "Amen." This
ceremony, however, is not actually part of the events of Sinai, to which
all the laws have become attached in the process of the composition of
the book. Rather, it is part of Moses' prescriptions for the future. The
people are to perform it, on entering the land, at a prescribed place:
standing on the two mountains of Ebal and Gerizim in the vicinity of
Shechem. The most natural explanation of this scenario is that it goes
back to a ritual action actually performed at some time at that place.
By contrast, the series of laws at Lev. 1 8 :6-23 prohibiting particular
sexual irregularities in the family suggests a quite different origin. It
may represent a code of rules which originally had their setting with­
in a limited family or clan circle, possibly promulgated by the head
of the family. Then again, it may be an example of freely agreed self­
government by a small group, setting limits to what was acceptable
behavior and so intended to maintain cohesion and harmony within
that group. Yet again, the "participial" laws prescribing the death penalty
suggest the more properly legal situation of a larger group in which such

1 13
INTRODUCTION TO THE PENTATEUCH

extreme sanctions were inevitable. In this respect the participial laws do


at least bear a resemblance to the casuistic laws.
The question of the original setting and purpose of these so-called
apodictic laws continues to be a subject of scholarly discussion. Their
origins are probably quite diverse, and the lack of positive evidence makes
it unlikely that agreement will be reached. One feature, however, is quite
evident. In many cases groups of laws, sometimes in groups of ten or
twelve, stand out prominently from their contexts as self-contained entities
- a sign, if one were needed, that these collections of laws are the result
of a complex and probably gradual compilation and development.
Whatever may have been the origins of the various laws and collec­
tions of laws in the Pentateuch, the significant fact is that they have all
been brought into relationship with the making of the covenant at Sinai­
Horeb and the person of Moses. Only in one place in the Old Testament
is there a collection of laws which has no connection with the Mosaic
covenant: Ezek. 40-48. It is this association with Moses and the making
of the covenant at the beginning of Israel's history that makes the Torah
the foundation document of the Bible. The foundations of the traditional
faith of the Jews were laid down once and for all at Sinai, and there is a
sense in which everything that happened subsequently was secondary. The
later history of Israel was seen wholly in terms of obedience and disobe­
dience to the Law, which was the complete revelation of God's will for his
people. Although new circumstances eventually necessitated the creation
of a .massive literature of interpretation of the Law and of adaptation to
it, this was commentary: nothing essential could be added to it or taken
away from it (Deut. 4:2; 12:32) . After the loss of the First Temple and of
the monarchy in the sixth century B. C., and still more after the destruction
of the Second Temple in A.D. 70, the Law became the one stable factor
and the central pillar of Jewish life.
But the context in which the laws have been set is not simply that of
God's demands on his people. It is also that of God's past gifts to his people,
which certainly put Israel under an obligation to present and future obe­
dience, but which are also a token both of God's power and of his willingness
to continue to shower benefits on them. In particular, in all these collections
of laws there are references to God's deliverance of Israel from slavery in
Egypt, so that even laws which may have originally had no specifically
Israe lite character now bear the marks of the Yahwistic faith. In all kinds of
contexts the people are to remember their deliverance from Egypt, and to
frame their present and future conduct in the light of that event.

1 14
The Laws

This connection with the deliverance from Egypt is explicitly made


with regard to the whole body of commandments, for example, in Lev.
1 8�3; Num. 14:4 1 ; and most notably in the opening chapters of Deuter­
onomy. It is also made in the preface to the Decalogue in both of its
versions: "I am Yahweh your God, who brought you out of the land of
Egypt, out of the house of slavery; you shall have no other gods before
me . . . " (Exod. 20:2-3; Deut. 5 : 6-7) . (In Deut. 1 - 1 1 it is not only the
Exodus that is cited, but also God's guidance of the people in their j ourney
through the wilderness.) But the Exodus paradigm is also attached to
various particular laws. Thus Lev. i 1 :45, in a chapter dealing with clean
and unclean animals, relates the Exodus to the requirement to be holy as
God himself is holy. Most frequently the connection is made with regard
to . laws requiring humane treatment of the poor and oppressed, whether
Israelite or foreign residents in Israel; the people are reminded that they
themselves had been poor, oppressed, aliens, and slaves in Egypt until
Yahweh rescued them (Exod. 23:9; Lev. 1 9: 33-34; 25: 39-46; Deut. 1 5 : 1 2-
1 5 ; 24: 17- 1 8}. The Feast of Unleavened Bread is to be observed in the
month Abib, reminding the people that it was in that month that they
came out of Egypt. There are also reminders of the Exodus in connection
with honest trading, a requirement that was not in itself a peculiarly
Israelite requirement (Lev. 1 9 : 35-37), but also. with the food laws, which
were (Lev. 1 1 :44-45) . It is clear that the admonition to remember God's
past acts of redemption applied to the wholelife of Israel - to righteous­
ness and morality as well as to private thoughts, public worship, and the
education of children.

The Decalogue

The Decalogue, or the Ten Commandments .as this short collection of


laws is popularly known, occupies a special place among the Pentateuchal
laws, as is apparent from the fact that it occurs twice: in Exod. 20:2- 1 7
and Deut. 5 :6-2 1 . It has given rise i n modern times t o an immense quantity
of scholarly literature.
The differences between the two versions are small. The main differ­
ence is in the reason given for the Fourth Commandment, on the observance
of the sabbath. According to Exodus, this is to be a commemoration of God's
rest on the seventh day after his works of creation (Exod: 20: l l ; cf. Gen.
2:2-3) . However, in Deuteronomy it is - once again - the memory of

1 15
INTRODUCTION TO THE PENTATEUCH

slavery in Egypt that is invoked, presumably mainly as a reason for giving


freedom to the individual's family, dependents, and domestic animals to have
a regular break from work (Deut. 5 : 1 4- 1 5) . Otherwise the two versions are
practically identical, although there are a few minor differences: the Fifth
Commandment is slightly longer in Deuteronomy (5 : 1 6) than in Exodus
(20: 1 2) , and in the Tenth Commandment Deuteronomy (5:2 1) but not
Exodus (20: 1 7) puts the wife before the house in the law forbidding the
coveting of a neighbor's possessions.
The arrangement of laws in groups of ten or approximately ten
(perhaps twelve may also have been deemed an appropriate number) is
not entirely unusual in the Pentateuch. It is hot this that makes this
Decalogue distinctive. Such groupings were especially characteristic of
apodictic prohibitions of the "You shall not . . . " kind. Such may be the
group of laws regulating sexual relations in Lev. 1 8 :7- 1 8 . Another series
is to be found in Exod. 34: 1 4-26, sometimes referred to as the "Ritual
Decalogue" in distinction from the "Ethical Decalogue"; it is called "the
ten commandments" in v. 28 and was inscribed by Moses at God's dicta­
tion on the second set of tablets that replaced the broken ones. The
collection in Exod. 34 has some laws in common with the "Ethical
Decalogue," but focuses more on cultic matters. The Decalogue in Exod.
20 and Deut. 5 consists mainly of prohibitions, but not entirely so; the
Fourth and Fifth Commandments, on the sabbath and on honoring
parents, are positive laws. The suggestion that originally these also were
expressed in negative terms has nothing to commend it.
Much more, however, is to be said for the view that the Decalogue
as it stands in Exod. 20 . and Deut. 5 is an expansion of an earlier form, .
and many attempts have been made to reconstitute the text of that earlier
form (e.g., M. E. Andrew, Eduard Nielsen) . It is significant that five of
the laws (in Exod. 20, vv. 3, 1 3- 1 6) consist each of a single sentence
containing a brief general command or prohibition. The other five (vv. 4,
7, 8-12, 1 7) have additional clauses that elaborate the prohibition with
further detail, add emphasis to it, or give a reason for it. These may
originally have read as follows:

You shall not make for yourself an idol.


You shall not make wrongful use of Yahweh's name.
Remember the sabbath day, and keep it holy.
Honor your father and your mother.
You shall not covet.

1 16
The Laws

In Exod. 20 the Decalogue has been placed at the most crucial point
in the Sinai story: the making ofthe covenant itsel£ It . is represented not
as commandments given to Moses and subsequently relayed by him to
the people, but as spoken by God directly to the people in the midst of
the thunder and lightning and sou.rid of the trumpet emanating from the
smoking mountain (Exod. 20: 1 8- 1 9) . In Deut. 5 it forms part of a speech
to the people by Moses in . the plains of Moab, in which he reminds them
of what God had said to them at Horeb. Not all modern scholars, however,
believe that the Exodus Decalogue is the older of the. two. Apart from the
question whether the Decalogue in its present forms is derived from an
older original, there is evidence that. the present texts may have influenced
one another in the course of transmission. Deuteronomic influence has
been detected in both.
It is not at all clear what was the original setting of the Decalogue,
its original purpose, or when it was composed. While some scholars (e.g.,
H. H. Rowley, Anthony Phillips) associate it with Moses himself, others
see it as a late composition, assembled as a compendium of general be­
havior formed by excerpts from other laws in the Pentateuch. It has. also
been suggested that the Decalogue's original content was of a general
nature, and that more specific laws such as that concerning sabbath ob­
servance were added later. It is, for example, observable that several of the
laws are not specifically associated with Yahweh, and that others also may
not have had a specific connection with him in their supposedly original
shorter form. This may suggest that Gerstenberger is correct in supposing
that the Decalogue - or part of it - had its origin in family or clan rules
governing what was, or was not, acceptable behavior in such circles. In its
present form, however, it is significant that the list is headed by the laws
which forbid the worship of other gods or their images.
One theory, that of Phillips, that the Decalogue "constituted ancient
Israel's preexilic criminal law code given to her at Sinai" (Ancient Is�aet's
Criminal Law, 1 ) must be judged improbable. Instead, the Decalogµe falls
in the category, already discussed, of apodictic "laws" which are not laws
in the usual sense of the word. No penalties are prescribed as with the
casuistic and some other laws of the Pentateuch, or with such other laws
as could be applied in a court. of law. Als � , .some of the laws are imprecise
(e.g., "Honor your father . and your mother") or appear . to refer not to
indictable offenses but simply to thoughts and desires (e.g., "You shall not
covet"; the view of J. J. Stamm .that this verb has here roughly the same
meaning as "steal" is improbable, if only because this would cause a

1 17
INTRODUCTION TO THE PENTATEUCH

duplication with the Eighth Commandment) . Again, murder, adultery,


theft, and false witness are actions forbidden in most cultures (see also
ancient Near Eastern wisdom books), and would require a stated penalty
to be effective, as is the case elsewhere in both the Old Testament and in
ancient Near Eastern casuistic law.
In brief, the Decalogue should probably be understood - as it has
been throughout subsequent history - as a series of basic principles of
conduct, partly religious and partly ethical, which came to be recognized
in Israel - possibly fairly late, and perhaps in connection with the didactic
"preaching" of the Deuteronomists - as summing up what was essential
to Israel's way of life as commanded by God. It may have had its origins
in the same way as the much more restricted apodictic code of the Re­
chabite sect mentioned in Jer. 35:6-7, who are there said to follow the
commandments of their ancestor Jonadab, to drink no wine, build no
house, sow no seed, and plant no vineyard, but to live in tents. Two
prophetic oracles (Hos. 4:2; Jer. 7:9) show a clear knowledge of a group
of the ethical principles enunciated in the Decalogue, and regard these as
referring to things particularly abhorrent to Yahweh; but there is no way
of proving whether these prophets were familiar with the whole of the
Decalogue in its final form.

The Book of the Covenant (Exodus 20:22-23 :33)

Like the Decalogue, this collection of laws is an originally independent


collection that has been inserted into its present context and so brought
into connection with the making of the covenant on Sinai. The laws
themselves comp rise 20:22-23: 1 9, the final verses (23:20-33) being a
parenetic addition. Casuistic and apodictic laws are to some extent mixed
together, but it should be hoted that, while the casuistic laws constitute
the central section, the apodictic laws are mainly grouped together at the
beginning and the end (20:23-26; 23:9-1 9).
The apodictic laws here, although they include regulations of a more
general kind, are characteristically Yahwistic in that they forbid the making
of divine images and the worship of other gods than Yahweh. They also
include laws concerning the making and use of altars presumably dedicated
to Yahweh and laws concerning dearly Yahwistic religious observances:
sacrifice, the three peculiarly Yahwistic feasts, and the sabbath. If the
"participial" laws carrying the death penalty are counted as apodictic, these

1 18
The Laws

laws repeat no less than four provisions of the Decalogue which occur in
the immediately preceding chapter, concerning the avoidance of the wor­
ship of other gods, murder, behavior toward parents, and the observance
qf the sabbath.
T he casuistic laws, however, have no such specifically Yahwistic
character. Although they are not arranged in strictly logical order, they
may be roughly classified as follows:

2 1 : 1 - 11 Laws regulating the treatment of slaves, particularly


"humanitarian" provisions .
2 1 : 1 2- 1 7 Laws carrying the death penalty ( if these are to be seen
as casuistic) : homicide, behavior toward parents, kidnap­
ping
2 1 : 1 8-36 Laws concerning bodily injuries
22: 1 - 1 6 Laws concerning damage to property
22: 1 7-28 Miscellaneous laws ( including some apodictic laws)
23: 1 -9 Laws concerning the administration of justice and social
obligations (including some apodictic laws)

Many of these laws concern matters also dealt with in ancient Near
Eastern law codes, although none is identical with any extant non-Israelite
law. T he affinities with the Near Eastern legal tradition, however, suggest that
some dependence on the (presumed) laws of pre-Israelite Canaan is a
probability. T he view of some scholars that the laws here, or some of them,
reflect a presettlement nomadic existence is improbable. T hey are largely
concerned with such matters as the ownership or disputed ownership offarm
livestock (which was clearly of great importance) , with the growing of crops,
with viticulture, with theft from private houses, and with the ownership of
domestic slaves - in other words, with a settled mode of agricultural life.
There are few references to religion in the casuistic laws of the Book
of the Covenant. T here is no reason to suppose that the occasional refer­
ences to "God" or "the gods" (Hebrew elohim, ha-elohim in Exod. 2 1 :6,
1 3; 22:9, 28) originally had Yahweh in mind. T he three occurrences of
the name Yahweh are in apodictic, not casuistic, laws, and belong to a
later Yahwistic redaction.
T he final redaction may have taken place when the Book of the
Covenant was placed in its present position; but the combination of the
specifically Yahwistic, apodictic laws with the casuistic ones may well be
of an earlier date, when this was still an independent document.

1 19
INTRODUCTION TO THE PENTATEUCH

The date of its final composition is impossible to determine. All that


can be affirmed is that the Book of the Covenant cannot be earlier than
Israel's first dose contact, as resident in the land, with the Canaanites, nor
obviously earlier than the adoption of the religion of Yahweh by the
Israelites. Also, it must have been earlier than the composition of the laws
of Deuteronomy, which show a clear dependence on it. Since Israelite law,
though perhaps imposed by kings during the time of the monarchy, was
not "royal law" in the sense of having been made by kings, the absence
of any references to kings, or indeed to any centralized government, does
not necessarily imply that it is premonarchic.

The Laws of Deuteronomy

Some aspects of these laws have already been discussed in chapter 6 above.
It is dear that in some respects they are an "improved" version of the Book
of the Covenant; they presuppose, rephrase, and to some extent interpret
and adapt many of the earlier laws. The two collections thus overlap in
subject matter to a considerable extent, and Deuteronomy has retained
the casuistic form in appropriate cases; but the apodictic form is much
more prominent here. The Deuteronomic laws, unlike those of the Book
of the Covenant, were from the outset wholly designed to set forth a series
of obligations imposed unconditionally on the covenant people. In many
cases the older laws have been expanded to give more precise instructions
on particular subjects, and also to take account of the circumstances and
requirements of a more developed society. For example, the laws of Exod.
2 1 : 1 2- 1 4 about homicide have been expanded into a much longer set of
provisions in Deut. 1 9:4- 1 3, which provide among other things for the
legal protection of those who commit unintentional homicides. In Deut.
1 6: 1 8-20:20 there is generally a much more extended treatment of the
administration of justice and of national institutions and officials - king,
priests, prophets - together with laws on the conduct of war. But like the
laws of the Book of the Covenant, the Deuteronomic laws do not provide
for every contingency for which modern state law would be expected to
provide (although to some extent they are intended to limit the discre­
tionary powers of the local courts of elders) . They are not a complete
statu10ry code that would have been consulted by either local or nationally
appointed judges in every case. Nor do the laws follow any dear logical
arrangement, although attempts have been made by some scholars (e.g.,

120
The Laws

Calum M. Carmichael) to discover some kind of scheme or arrangement.


Some laws, however, are roughly grouped together by topic: for example,
Deut. 12: 1 - 1 6: 17 are mainly concerned with worship, and 1 6: 1 8-1 8:22
with official leaders and institutions. From ch. 2 1 on no particular ar­
rangement is evident. The lack of obvious arrangement may be partly due
to additions that have been made to the original collection, which itself
has incorporated a number of groups on particular topics that once had
a separate existence.
The most original, and indeed revolutionary, feature of the Deuter­
onomic legislation is to be found in the laws concerning worship. These
are most clearly set out in ch. 1 2 . This chapter orders the destruction of
all the places at which worship has previously been offered, and commands
that public (sacrificial) worship is to be practiced only in one place, "the
place that Yahweh your God will choose out of all your tribes to put his
name there" (vv. 5 , 1 1 , 14, 1 8) . As has been pointed out, this is a maj or
innovation. The Covenant Code speaks of permitted worship at "every
place where I cause my name to be remembered" (Exod. 20:24) ; and, as
is well known, the preexilic literature testifies to the existence of many
places in the land at which worship was legitimately offered, although
some of the prophets had condemned the kind of worship which was
practiced at some of these. Whatever may have been the connection of
this law with the religious reforms of Josiah (see chapter 6 above), its
purpose was to ensure that all worship should be carried out in accordance
with what the Deuteronomic legislators considered to be the will of God.
The one sanctuary was the outward expression of purity of worship and,
indeed, of purity of life as envisaged by the Deuteronomists. The law of
the single sanctuary marked a crucial and long-lasting change in the
religious practice of Israel.
The moralistic, preaching tone characteristic of the book (see chapter
6 above) is not confined to the more obviously sermonlike chs. 1 - 1 1 but
pervades the laws as well. Although what are known as "motive clauses"
- subordinate clauses attached to particular laws justifying them or ex­
plaining the reason for them - occur already in the Book of the Covenant,
they occur with particular frequency in the laws of Deuteronomy. Berend
Gemser in his seminal study of the question estimated that whereas the
proportion of laws in the Book of the Covenant provided with motive
clauses is 1 7 percent, in Deuteronomy it is 60 percent. The motive clause,
which is apparently a peculiarity of Israelite law - it is not found in any
extant ancient Near Eastern law code according to Gemser - may be of

121
INTRODUCTION TO THE PENTATEUCH

various kinds. In Deuteronomy it is overwhelmingly religious and/or


ethical, mainly intended to remind readers of Yahweh's actions on their
behalf in the past - of the state of misery in Egypt from which he had
rescued his people, of the covenant which he had established with them,
and of his gift to them of the land which they are about to occupy.
As has already been noted, Deuteronomy has a unique literary style
which is unmistakable even when translated into other languages. This is
mainly because it repeats, with slight variations, characteristic phrases
which express its theological point of view (see the list in S. R. Driver,
Introduction to the Literature ofthe Old Testameni) .
References to worship at "the place that Yahweh your God shall
choose to put his name there" are especially frequent. Also frequent are
references to the former slavery of the people in Egypt and Yahweh's rescue
of them. There are fourteen references to Egypt in these laws; several of
them (Deut. 1 5 : 1 5; 24: 1 8 , 22) are attached as motive clauses to laws
commanding j ust and humane treatment of slaves, orphans, and widows.
Others lay particular stress on the necessity to preserve the purity of the
community that Yahweh has redeemed ( 1 3 :5 , 1 0), or express thanksgiving
for that redemption (26:5, 8).
It may be thought strange that the final editor of the Pentateuch
should have preserved two major sets of laws that to a large extent cover
the same ground (the Book of the Covenant and Deuteronomy) , of which
Deuteronomy is partly a revision and expansion and even a correction of
the older collection and has a much more explicit theology. There may
have been several reasons for this.
Clearly the Deuteronomist did not regard the laws of Deuteronomy
as superseding the Book of the Covenant. There are in fact fewer contra­
dictions between the two than has often been supposed. The main apparent
inconsistency concerns the question of whether more than one place of
worship was to be permitted; and even here it is not clear that Exod. 20:24
was understood as contrary to the law of Deut. 1 2. Although it may have
been the intention of Exod. 20:24 that multiple places of sacrifice were to
be tolerated, the text does not actually state that this is the case. This is
in fact suggested only by a single Hebrew word, kol-, presumed to mean
"every," and this word is missing in the Samaritan version of the Penta­
teuch, which reads "in the place."
The duplication of earlier laws in Deuteronomy is not an isolated
phenomenon in the Pentateuch. It has already been pointed out above
that several of the laws in the Book of the Covenant duplicate some of

1 22
The Laws

the commandments of the Decalogue, and there are many other such
duplications elsewhere in the Pentateuch (e.g., be�een the Decalogue, the
"Ritual Decalogue" of Exod. 34: 1 4-26, and the Book of the Covenant) .
Tpis phenomenon may be partly based on the principle that repetition
can be an effective means of drawing attention to what is being said and
of impressing its importance on the persons addressed. But in the case of
the Pentateuch, it may also be due to what Samuel Sandmel called "a
disinclination to expunge" ("The Haggada Within Scripture," 1 20) .
Sandmel's main concern was with the narrative parts of the Penta­
teuch and with putting forward an alternative to the Documentary Hy­
pothesis, but his observations are equally applicable to the laws. He used
an analogy from the practice of the later Jewish haggadah, in which writers
recast biblical stories, embellishing, elaborating, and reinterpreting or cor­
recting them - even after they had been accepted as canonical - without
intending in any way to supersede them or to detract from them, but
rather to provide a commentary on them. This he called "a literature which
grew by accretion'' (p. 1 22) . It may well have been this reverence for earlier
laws which had been presented as having divine and Mosaic authority that
moved these later Pentateuchal writers to preserve them, even though they
regarded them as inadequate or incomplete and needing revision. The
same applies to the priestly legislation of the Pentateuch to be discussed
in the final section of this chapter. This priestly legislation has been
generally recognized since the time of Julius Wellhausi::n to be itself an
example of a "literature of accretion," an original set of laws having been
augmented over a period of time by insertion into the text of new laws
and new clauses which modify, or even sometimes contradict, the original
text in haggadic fashion.

The Priestly Laws

Most of the considerable body of laws in the Pentateuch has been appro­
priately dubbed "priestly." Although often referred to as the "Priestly Code,"
it is not a compact body of laws gathered into one place, but is to be found
� cattered through three of the books: Exodus, Leviticus (where it comprises
almost the entire book), and Numbers. It is attributed to P by those who
accept the Documentary Hypothesis and also by many who have retained
the notion of a P document though not the whole of that hypothesis, and is
therefore generally held to be postexilic in date. Recently, however, both the

123
INTRODUCTION TO THE PENTATEUCH

attribution and the date have been called into question. Although there are
some affinities, both linguistic and theological, between these laws and some ·

of the narratives attributed to P (see chapter 2 above) , th e laws are best


studied, like the other Pentateuchal collections of laws, as having had a
separate origin from the narratives in which they are now set.
The priestly laws are mainly unconnected with the other collections
of laws in the Pentateuch - they are neither revisions of earlier laws now
extant, nor can they be said to have been subject, except internally, to any
inner-Pentateuchal revision. In terms of the topics dealt with, their scope
is infinitely more restricted than that of either the Book of the Covenant
or Deuteronomy. Although like the other collections they are mainly
represented as having been addressed, with some exceptions, to the whole
people through the medium of Moses (and Aaron) at Sinai, irl reality· the
matters with which they deal are matters of particular interest to the
priesthood rather than to the laity: matters concerning priestly functions
such as the design and construction of the tabernacle in the wilderness,
the proper procedures for the offering of the various types of sacrifice, the
anointing and vestments of Aaron and his sons as priests, rules about edible
and forbidden animals, physical impurities and leprosy, the rules for the
observance of the Day of Atonement, and the regulations .for the obser­
vance of the major festivals. Although many of these rules, such as the
actual slaughter of the sacrificial animals and the food laws, were to be
carried out by laypersons, it was the priests whose duty it was to see that
they were properly performed.
Nowhere else in the Pentateuch are the details of sacrificial and other
rites set out. The cultic rules of the Book of the Covenant are restricted
to a few brief references mainly to the altar, the sabbath, and the observance
of the festivals without giving any details . . Deuteronomy also, apart from
its emphasis on the importance of the restriction of sacrificial worship to
a single place, is equally unconcerned with its details. The only part of the

priestly legislation that has substantial affinities with the other collections
of laws in the Pentateuch is the so-called Holiness Code (H), Lev. 1 7-26.
This has marked characteristics of its own (see below) . _

There is a limited sense, then, in which the priestly laws, or a,t any
rate their central body, the book of Leviticus, could be called a handbook
for the use of the priests which was at some later time made public. But
in fa c t it is both less and more than that. Like all Pentateuchal legislation,
despite careful attention to detail the priestly laws fall far short of giving
a complete description of the cultic acts to which they refer. There is no

1 24
The Laws

mention, for example, of texts or fixed forms of words which might have
accompanied the sacrifices. It is not known what words may have been
recited, for example, or whether psalms may have been sung - though
many of the extant psalms in the Psalter appear to be very appropriate for
such occasions. We really have no clear idea of the daily round of worship
that was practiced at the temple. Some later Jewish texts, written after the
final destruction of the Jerusalem temple in A.D. 70, give some additional
details, and these may have preserved some memories of what had taken
place, but it is not possible to be certain of th eir reliability. Much of the
detailed regulations for worship was presumably kept in oral form, handed
down by one generation of priests to the next while the temple(s) lasted.
But the absence of information makes it impossible to construct a h istory
of temple worship, which - though the tradition was probably very con­
servative and only changed slowly - must have undergone considerable
changes over the centuries as a result of the cultic reforms initiated from
time to time by the kings of Judah and above all the reestablishment of
the temple after the Exile. Our lack of knowledge about these things also
makes it extremely difficult to know what period or periods are represented
by the priestly laws which we have in the Pentateuch.
It is easy for modern readers to dismiss the priestly laws as an
unimportant or uncharacteristic side of Israel's religion, as representing a
cruder or more mechanical, legalistic conception of human relations with
the deity than the teaching of the prophets, for example - or simply as
extremely tedious reading! This view has been fostered by the dominant
influence of Wellhausen, who regarded the teaching of the "classical"
prophets with its "ethical monotheism" as the high point of the Israelite
faith, which had been followed by a steep decline especially in the postexilic
period. Wellhausen said of the priestly sacrificial system, "The warm pulse
of life no longer throbbed in it to animate it; it was no longer the blossom
and the fruit of every branch of life. . . . The soul was fled; the shell
remained, upon the shaping out of which every energy was now concen­
trated. A manifoldness of rites took the place of individualising occasions;
technique was the main thing, and strict fidelity to rubric" (Prolegomena
to the History ofAncient Israel 78) . This dismissive view has been perpet­
uated in treatments ofthe subject which until very recent times, like other
aspects of Old Testament study, have been dominated by liberal Protestants
for whom set forms of worship were contrary to their fr�e spirit, and who
had no understanding or appr�ciation of their underlying religious and
theological value.

125
INTRODUCTION TO THE PENTATEUCH

Before considering the validity of this judgment on the priestly laws,


it is important to take into account some recent discussion of the hitherto
dominant consensus about their antiquity. Wellhausen in his Prolegomena
mainly derived his late · dating of these laws from a comparison with the
scattered information in the preexilic literature of the Old Testament about
the practice of sacrificial worship. He did not, however, take sufficient
account of the fact that this literature does not purport to give a precise
account of sacrifice, but refers to it mainly in passing; even the earlier
collections of laws give very few details. The priestly laws are the first -
and indeed the only - formal account of the practice in the Old Testa­
ment. It is important to realize that they are written from a particular
perspective quite different from those of the other books - that is, from
the professional point ofview of the priesthood, which was also the audience
for which they were originally intended.
An even more important point is that Wellhausen did not take into
account - and was in fact not able to take into account at the time when
he wrote - the mass of documents from the ancient Near East which
shows that the priestly laws of sacrifice were not a late Jewish invention
but are in many respects characteristic of ·ancient Near Eastern practice
many centuries earlier than the Old Testament. The complexity and me­
ticulous attention to detail of sacrificial legislation were central features of
the ancient Near East of which Israel was a part. That the significance of
this has only recently been realized, and only by a few scholars, is surpris­
ing. The shadow cast by Wellhausen is a long one.
That these laws may contain some material that is older than the
now conventional exilic or postexilic date has long been surmised by some
scholars. But it was the Jewish scholar Yehezkel Kaufmann ( 1 930) who
first argued against Wellhausen that the whole collection is preexilic and
presents none of the features characteristic of the Jerusalem community
of the postexilic period. More recently other Jewish scholars (Avi Hurvitz,
Menahem Haran, and Jacob Milgrom) have pursued the question further
in general agreement with Kaufmann. Hurvitz confined himself to a study
of the language of the priestly laws, and showed by a comparison of this
with the language of Ezekiel, Chronicles, and Ezra-Nehemiah - books
also much concerned with cultic matters and all written in a Hebrew later
than the "classical" Hebrew of the preexilic period - that the priestly laws
of the Pentateuch show no knowledge of the priestly terminology em­
ployed in those books, but conform linguistically to the Hebrew . of the
preexilic ("First Temple") period.

1 26
The Laws

Haran reviewed the history of the priestly functions and institutions


as far as this is possible, and on the basis of that investigation also came
to the conclusion that the priestly legislation "derives from conditions that
prevailed in the pre-exilic, not the post-exilic, period of ancient Israelite
history" ( Temples and Temple-Service in Ancient Israel v) . He accounts for
the fact that this legislation apparently had no influence on communal
life until the time of Ezra on the grounds that it was an esoteric work
closely guarded by the small circle of priests and not known to the general
public until then. Strangely, however, Haran does not deal with the diffi­
culty, frequently used as an argument for a postexilic date, that two types
of Sacrifice, the "sin offering" (Hebrew hattat) and "guilt offering" (asham)
- which play a prominent part in the priestly sacrificial system and were
public sacrifices - are hardly mentioned, if at all, in the preexilic litera­
ture. (These terms are also not dealt with by Hurvitz; Milgrom uses the
fact that they are virtually confined to the priestly laws to prove that the
laws are early, later superseded by other terms before the Exile; Leviticus
1-16, 1 77.)
Since it has been established by comparison with the practices of
other Near Eastern peoples that a complex system of priestly legislation
could have developed in Israel at a comparatively early date, it remains
to consider the possibilities with regard to the general nature of the
biblical material itself The conventional view is that the legislation is
the work of the Israelite priesthood either during its exile in Babylonia,
when it could not be · put into practice but could make sense as a
"blueprint" for future use in the hoped-for restored Jerusalem temple
(possibly like the "laws" of Ezek. 40-48), or shortly afterwards when the
Second Temple was being, or had just been, built. But the nature and
complexity of the material make this highly improbable. Such complex
systems are not created "from scratch''; old traditions would surely not
have simply been ignored. At the least, it may be assumed that these laws
can only have been based on memories of the system that had actually
been in use just before the destruction of the First Temple. This makes
it extremely probable that what we now have here, however much it may
have been altered and expanded at a later time, is basically the traditional
lore and practice of the temple built by Solomon which had no doubt
developed and been modified over a long period during the following
centuries. However old some of it may have been - Milgrom suggests
that it contains laws which go back beyond Solomon to the ancient
Israelite temple at Shiloh - there seems, now that the question has been

1 27
INTRODUCTION TO THE PENTATEUCH

considered anew, no reason to doubt that it is substantially the priestly


laws of the preexilic period.
Like most legislation, the priestly laws provide little direct informa­
tion about their raison d'etre. Especially in contrast to Deuteronomy,
which is an exception to the rule, there are very few statements here of an
overtly theological nature. Nevertheless, that there is a significant under­
lying theology that gives meaning to the whole cannot be denied.
Various attempts have been made to identify a central theological
theme. Perhaps the concept of divinely established order is that which best
accounts for the multiplicity of matters with which the laws are concerned.
Behind them lies the concept of God as a God of order, and of the world
as an ordered creation which, in as far as it has departed from that order,
needs to be restored to it by means that God himself has provided for its
restoration. Such an essentially static notion of a divinely established norm
to which everything ought to conform in harmony is characteristic of the
priesthoods of other religious systems as well.
This concern with order manifests itself in many ways. One of these
is the great insistence on atonement for sin in the priestly laws. It was the
aim of the sacrificial laws of the sin offering and guilt offering to restore
the sinner and so to reestablish a normal state of affairs in society at large.
This was not automatically achieved simply by the offering of sacrifice,
although that was essential. Most importantly, confession of guilt was
required (Lev. 5 :5-6) . Only in cases of flagrant sin when the sinner "insults
Yahweh" and has "brought the word of Yahweh into contempt and violated
his command" (REB) was forgiveness not possible: "such a person shall
be utterly cut off and bear the guilt" (Num. 1 5: 3 1 ) .
The need fo r redemption from the consequences o f sin was not,
however, simply an individual concern; this was a situation which also
involved the whole people (Num. 25-26) . That is the reason for the setting
aside of one day in the year as a Day of Atonement (Hebrew yom hakkip­
purim, Lev. 23:27) when Aaron and subsequently "the priest who is anointed
and consecrated as priest in his father's place" (Lev. 16:32) was to enter the
holy place to make atonement for the whole people for all their sins
committed in the previous year. Sin was thus a breach of the covenant that
God had made with his people which, if it remained unconfessed and
unatoned for, would result in their total destruction (Lev. 26:23-45) .
The divine ordering of life was not only observable with respect to
Israel; it was cosmic in scope. This is well illustrated by the laws concerning
the classification of animals into two groups: those which were permitted,

128
The Laws

and those which were forbidden, to be eaten (Lev. 1 1 ; a similar list appears
in Deut. 14:3-20) . This clearly implies God's ordering of the whole of the
natural world. Many attempts have been made to . discover a rational, or
even a theological, basis for these distinctions between the kinds of animals;
they have particularly interested anthropologists. No consensus, however,
has been reached on this question. The explanation given in the text is
simply that the people are commanded to "make a distinction between
the unclean and the clean" (Lev. 1 1 :47) . Such distinctions between dean
and unclean, and also between holy and profane, are an extremely prom­
inent feature of the priestly laws (see Lev. 1 0 : 1 0- 1 1 for a general .command
given to Aaron) . They all testify to the concern for the divine order in life,
as also does the insistence on the distinction between the respective func­
tions of the priesthood and the laity, the overstepping of which could be
fatal. Yahweh was to be worshipped only in the prescribed way and in
prescribed places.

The Tabernacle. In Exod. 25-30 Moses is told to make a large structure


called a mishkan (the usual English translation is "tabernacle") and the
various items of furniture for it. Chs. 35-40 report the execution of those
directives. These chapters also contain an instruction to anoint Aaron and
his sons to serve as priests and a corresponding account of its execution.
The tabernacle was to be carried about by the Levites, to accompany the
Israelites during the remainder of their journey through the desert; and
they were to guard it from the approach of "outsiders," the penalty for
which was death (Num. 1 : 50-5 1 ) . When the work was completed it is
recorded that the tabernacle was filled with the appearance of Yahweh's
glory; signifying his approval and presence.
• Much has been written about these chapters. One thing that is agreed
i � that the tabernacle is an ideal rather than a historical phenomenon; that
such a massive structure could have been built and carried for many years
through the desert is an obvious impossibility. These chapters are an
extreme example of a persistent tendency of the Pentateuchal writers to
assert that all later religious institutions were founded at Sinai, at the very
beginning of Israelite history, through the mediation of Moses himself.
The tabernacle as described here is clearly a model for the Jerusalem temple
of the future. It does. not correspond exactly to either Solomon's temple
or the postexilic temple, though it does so in many of its details. An
important article of its furniture is the ark of the covenant, which no
longer existed in postexilic times.

1 29
INTRODUCTION TO THE PENTATEUCH

It is stated in Exod. 25:9, 40 that the model, or pattern (Hebrew


tabhnith), according to which the tabernacle and its furnishings were to
be constructed had been shown to Moses by God when he stood in God's
presence on Mount Sinai. This is another example of the concern of the
priestly laws for the divine ordering of the whole of Israel's life in the
smallest detail. But the chief importance of the tabernacle is theological.
It symbolizes the presence of Yahweh with his people wherever they go
and the importance of the distinction between the holy and 'the profane
(in this case with regard to holy places), as well as the centrality of the
temple in Israel's religious life.

The Holiness Code. Leviticus 1 7-26 has been called the Holiness Code
because of the frequency of the occurrence of the phrase, attributed to
Yahweh: "You shall be holy because I am holy, " which corresponds to the
theological theme of the other priestly laws but here receives a special
emphasis. One other phrase is characteristic of these chapters: "I am
Yahweh" (sometimes "I am Yahweh your God") . It is generally believed
that these chapters originally constituted an independent collection oflaws
that has been subsequently inserted into the main body of Leviticus. There
can be no doubt that they emanate from the same general circles as the
other priestly laws, but their standpoint is slightly different.
The order in which these laws have been arranged is far from obvious.
They are a mixture of laws on a variety of different subjects. Many of them
virtually repeat the substance of laws that occur elsewhere in the Penta­
teuch, but there are a number of minor details in which they differ from
the other priestly laws. The collection also contains a substantial number
of laws about the conduct of ordinary life that resemble the Book of the
Covenant or Deuteronomy rather than Leviticus. This collection con­
cludes, like Deut. 28, with blessings and curses.

For Further Reading

Alt, Albrecht. "The Origins of Israelite Law," in Essays on Old Testament


History and Religion. Oxford: Blackwell, 1 966, and Garden City:
Doubleday, 1 968, 79- 1 32. Repr. Sheffield: JSOT Press, 1 989. (First
p ublished 'in German, 1 940.)
Blenkinsopp, Joseph. The Pentateuch: An Introduction to the First Five Books
ofthe Bible. New York: Doubleday and London: SCM, 1 992, ch. 6 .

1 30
The Laws

Boecker, Hans-Jochen. Law and the Administration ofjustice in the Old


Testament and t&e Ancient East. Minneapolis: Augsburg and London:
SPCK, 1 980. (First published in German, 1 976.)
Carmichael, Calum M. The Laws ofDeuteronomy. Ithaca: Cornell Univer­
sity Press, 1 974.
Cody, Aelred. A History of Old Testament Priesthood. Analecta Biblica 3 5 .
Rome: Pontifical Biblical Institute Press, 1 969.
Douglas, Mary. In the Wilderness: The Doctrine ofDefilement in the Book
ofNumbers; JSOT Supplement 1 5 8. Sheffield: JSOT Press, 1 993.
Gemser, Berend. "The Importance of the Motive Clause in Old Testament
Law," VTS 1 . Leiden: Brill, 1 953, 50-66.
Grabbe, Lester L. Leviticus. OTG. Sheffield: JSOT Press, 1 993.
Haran, Menahem. "Behind the Scenes of History: Determining the Date
of the Priestly Source," ]BL 1 00 ( 1 9 8 1 ) : 32 1-333.
--- .. Temples and Temple:.Service in Ancient Israel. Oxford: Clarendon
Press, 1 978.
Houston, Walter. Purity and Monotheism: Clean and Unclean Animals in
Biblical Law. ]SOT Supplement 1 40. Sheffield: JSOT Press, 1 993.
Hurvitz, Avi. "The Evidence of Language in Dating the Priestly Code,"
RB 8 1 ( 1 974): 24-56.
Levine, Baruch A. Numbers 1-20. AB 4. New York: Doubleday, 1 993.
Mendenhall, George E. ''.Ancient Oriental and Biblical Law," BA l 7
( 1 954) : 26-46. Repr. in The BA Reader 3, ed. E. F. Campbell, Jr.,
and David Noel Freedman. Garden City: Doubleday, 1 970, 3-24.
Mettinger, T. N. D . The Dethronement of Sabaoth. Coniectanea Biblica,
Old Testament 1 8. Lund: Gleerup, 1 9 82.
Milgrom, Jacob. Leviticus 1-16. AB 3 . New York: Doubleday, 1 99 1 .
Nielsen, Eduard. The Ten Commandments in New Perspective. SBT, 2nd
Series 7. Naperville: Allenson and London: SCM, 1 968. (First pub­
lished in German; 1 965 .)
Noth, Martin. "The Laws in the Pentateuch: Their Assumptions and
Meaning," in The Laws in the Pentateuch and Other Studies. Edin­
burgh: London: Oliver & Boyd, 1 966, and Philadephia: Fortress,
1 967, 1 - 1 07. Repr. London: SCM, 1 984. (Translated from the 2nd
German ed., 1 960.)
---. Leviticus. OTL. Philadelphia: Westminster and London: SCM,
1 965 . Rev. ed. , 1 977. (First published in German, 1 962.)
Paul, Shalom M. Studies in the Book of the Covenant in the Light of
Cuneiform and Biblical Law. VTS 1 8. Leiden: Brill, 1 970.

13 1
INTRODUCTION TO THE PENTATEUCH

Phillips, Anthony. Ancient Israel's Criminal Law: A New Approach to the


Decalogue. New York: Schocken and Oxford: Blackwell, 1 970.
Rowley, H. H. "Moses and the Decalogue," in Men of God· Studies in Old
Testament History and Prophecy. London: Nelson, 1 963, 1 -36. (First
published 1 9 5 1 -52.)
Sandmel, Samuel, "The Haggada Within Scripture,''. ]BL 80 ( 1 9 6 1 ) : 1 05-
1 22.
Stamm, J. J., and Andrew, M. E. The Ten Commandments in Recent Re­
search. SBT, 2nd Series 2 . Naperville: Allenson and London: SCM,
. 1 967. (Includes additions from the 2nd German ed., 1 962.)
Wellhausen, Julius. Prolegomena to the History ofAncient Israel. Edinburgh:
A. & C. Binck, 1 88 5 . Repr. Magnolia, Mass. : Peter Smith, 1 977.
(First published in German, 1 878.)
Westbrook, Raymond. "Biblical and Cuneiform Law Codes," RB 92
( 1 985): 247-264.

132
CHAPTER S

Reading the Pentateuch

T
HERE ARE A NUMBER of different ways of reading the Pentateuch.
·
Since it is basically a single continuous narrative into which laws and
poems have been inserted at appropriate points, it can be read simply as
an account of the origin and early history of the people of Israel, preceded
by a more general section dealing with the creation of the world and the
earliest period of human history. It cannot be doubted that, among other
things, this is what the Pentateuch purported to be - what it was in the
eyes of those who produced it in its present form. As such it is similar in
many ways to the accounts that other peopies have given of their origins.
The Pentateuch may also be read simply as literature; and on this
level also it may be compared with other literatures. This approach has
taken several forms. For example, form criticism, initiated by Hermann
Gunkel, has studied the individual stories in Genesis, and to some extent
those of Exodus and Numbers, as examples of "saga," that is, as originally
independent short stories; longer sections such as the story of Joseph have
been taken as more elaborate examples of the short "novel," of which again
other examples are to be found in ancient Near Eastern literatures. Much
attention has also been paid to the structure of the narrative and legal
material and indeed of whole books (e.g., Mary Dougla>) . The narratives
have been investigated as examples of refined and subtle writing (e.g.,
Robert Alter) . These scholars have demonstrated that the Pentateuch as a
whole is indeed an astonishing literary 'achievement.
The Pentateuch has also been studied from the perspective of the
comparative study of religions. The religious customs attributed to the
patriarchs of Genesis and to the age of Moses have been compared with
those of other ancient Near Eastern peoples; and a voluminous literature

1 33
INTRODUCTION TO THE PENTATEUCH

has been produced along these lines, comparing for example the accounts
of the creation of the world and the other narratives of Gen. 1 - 1 1 with
the mythologies of the ancient world.
Th<:! Pentateuchal laws, which . account for about one-half of the
material, have been studied not only by comparing the various groups of
biblical laws with · one another, but also by comparing · them with the
substantial body of extant legal codes from the ancient Near East, so
revealing many affmities with these as well as significant differences.
It has also been supposed that the Pentateuch offers important evi­
dence concerning the development of Israelite religion over the centuries.
This enterprise went hand in hand with the view that the material which
comprises the Pentateuch originated in different histotical periods, and
that it is possible to discover the relative, if not the absolute, ages of
different parts of it and to link this with what is known from elsewhere
in the Old Testament of the developing religion of Israel. This was the
presupposition, and also the fruit, of the source criticism of proponents
of the Documentary Hypothesis which was also in its own right one of
the principal approaches made to the study of the Pentateuch. This attempt
to trace religious development in the Pentateuch has not, however, been
confined to "classical" source criticism.
Archaeology has also had its part to play in the study of the Penta­
teuch. Archaeologists have contributed to the identification of some of the
many places mentioned in these books, for example, stopping places on
the route of the Israelites from Egypt to Palestine. Ancient Near Eastern
texts discovered by archaeologists have also been used in attempts to locate
the historical setting and date of the patriarchs and of Israel's sojourn in
Egypt, though it has since been shown that much of the supposed evidence
in question is in fact invalid and that the quest itself is misconceived (see
Thomas L. Thompson) .
The above list of approaches to the Pentateuch is by no means
exhaustive. Among others that deserve special mention are the geographical
(e.g., in considering the location of the crossing of the Sea), the sociologi­
cal, and the anthropological (e.g. , in the study of the social setting and
family customs of the patriarchs) . Each has its own value, although in
some cases this is a negative one. Two important aspects of reading these
books, however, remain to be considered: the intention and meaning of
the Pentateuch in its final form in the minds of those who were responsible
for its composition, and its meaning and message for today. The second
of these, though obviously of paramount importance, · lies outside the scope

1 34
Reading the Pentateuch

. of this book. It is a theological matter; moreover, it is not a subject for


the study of the Pentateuch alone, but one which pertains to the theological
study of the Old Testament as a whole. The first, however, is that which
l have chosen to follow in this book as my main approach to the Penta­
teuch.
This approach, which characterizes especially chapters 3-5 above, is
what is today known as "synchronic." That is, the Pentateuch in its final
form is treated as a "book'' which exists in its own right as an artifact with
a theme and a message. This approach is distinct from the "diachronic,"
which is concerned primarily with the history of composition, which may
(or may not) have extended over a long period. It has not, however, been
possible to use the synchronic approach consistently in the treatment of
Deuteronomy or of the laws, since in both these cases a consideration of
historical origins and subsequent developments is necessary in order to
explain the motives and intentions of those who composed this material.
Moreover, the indications of different stages of composition are clearer in
Deuteronomy and the laws than in the narrative sections, and the tech­
niques used were different from those used in the narratives. Nevertheless,
the diachronic method has been avoided as much as possible here as well.
The advantage of using the synchronic approach is that in this way
one is dealing with something concrete that actually exists: the text of the
Pentateuch, which lies before us. The diachronic method is necessarily
speculative. It is an attempt to discover earlier "texts" lying below the
surface, whether written or oral, which may have existed at various times
and which are supposed to have been subsequently incorporated into the
present text - often in a greatly changed form and to serve new purposes
- and to postulate historical situations in the history and religious history
of Israel in which they may have arisen. Th.ese earlier "versions" do not
actually exist; they are simply postulated.
For a long time, since the work ofJulius Wellhausen, there was a general
consensus about the identity of these sources - though there has always been
some disagreement over details, and there have always been some scholars who
did not accept the Documentary Hypothesis. But now the consensus has
broken down; and the complete lack of agreement at the present time
about the composition of the Pentateuch should warn the student that
theories about the dates of different parts of it are extremely subjective.
Another circumstance affecting the present-day study of the Penta­
teuch is that there is now great uncertainty about the history of th.e religion
of Israel, especially the earlier stages which were closely linked to theories

135
INTRODUCTION TO THE PENTATEUCH

about sources. A diachronic method is therefore much more difficult to


sustain now than before. It must be admitted that the view taken in this
book about the late date of the Pentateuch - exilic or immediately post­
exilic - is also to some extent subjective. It cannot be proved beyond
possibility of doubt that it was designed to fit such a situation. But dating
it is in fact a secondary issue. What is much more important is to see the
Pentateuch as a (relatively) consistent whole with its own overriding theme.
This primary theme, as has been amply demonstrated above, is
undoubtedly that of the promises made to the patriarchs from Abraham
on. By the end of Deuteronomy these have to a large extent been fulfilled:
here is a great and powerful people, living under the (admittedly condi­
tional) blessing of God. Yet, although Israel is poised to enter the Promised
Land, this final fulfillment of the promises has not yet been achieved. Even
the casual reader who regards the Pentateuch as a self-contained entity and
does not go on to read the book of Joshua will be inclined to ask why.
The answer to this question probably lies in the process by which the
books of the Old Testament were assembled in their present order. Un­
fortunately the mechanics of this process are by no means clear to us.
It is only when the Pentateuch is regarded as a closed entity that this
problem of the non-fulfillment, or partial fulfillment, of the promises
arises. The reader who comes to the Bible without any presuppositions
about the Torah's being a work quite separate from the succeeding books
will find an unbroken story line in the whole complex of Genesis to Kings,
from the creation of the world to the fall of the kingdom of Judah in the
early sixth century B.C. He or she will read straight through the division
that now exists between Deuteronomy and the book of Joshua, noting
that although the anointing of Joshua as Moses' successor and the report
of Moses' death at the end of Deuteronomy clearly mark the end of an
era and the beginning of a new one, Joshua begins quite naturally as a
resumption of the narrative: ''After the death of Moses . . . the Lord spoke
to Joshua'' Qosh. 1 : 1) . The very next verse moves the story on, with
Yahweh's words: "My servant Moses is dead. Now proceed to cross the
Jordan, you and all this people, into the land that I am giving them."
There is no break in the narrative. Even the language used is the same as
in Deuteronomy.
This does not necessarily mean, as some scholars have maintained
in the past, that the first books of the Old Testament originally formed a
"Hexateuch'' or group of six - that Joshua, which completes the fulfill­
ment of the promises, was originally an integral part of this group of books

1 36
Reading the Pentateuch

which subsequently became detached from the others because it does not
belong to the Mosaic era (e.g., S. R. Driver, lntr.oduction to thf -literature
ofthe Old Testament, 5-6, 1 03; contrast Martin Noth, A History ofPenta-
1teuchal Traditions, 6) . The current view that, on .the contrary; the books
"
of Joshua, Judges, Samuel, and Kings together with Deuteronomy form a
distinct work, the so-called Deuteronomistic History, works in the op­
posite direction, reducing the "Pentateuch'1 to a "!etrateuch" (a group of
·

four books) .
The question of the relationship of Deuteronomy to the rest of the
Pentateuch on the one hand and to Joshua and the following books on
the other remains problematic. It is possible that more than one way of
dividing these books was adopted at different times, operating in different
directions. An attractive solution to the problem, offered by A D. H.
Mayes ( The Story ofIsrael between Settlement and Exile, 1 39- 1 49, especially
1 4 1) on the basis of a suggestion by Rolf Rendtorff ( The Problem of the
Process of Transmission in the Pentateuch, 200), is that the Deuteronomistic
History (perhaps in an early edition) was formed first, and that the Pen­
tateuch, which had not previously existed (at any rate as a distinct work)
was composed somewhat later as an introduction to it. This would account
for the absence in the Pentateuch of a full account of the settlement of
the Israelites in Canaan. The Deuteronomistic History, or rather the
narrative part of it, begins with its account of the settlement. There was
therefore no need for the composer of the Pentateuch to repeat this. His
function was to take the story up to the entry into Canaan; and -he j oined
his work to the opening verses of the book of Joshua with his account of
the immediately preceding events, the anointing of Joshua and the death
of Moses. This is probably a better account of the matter than, for example,
Noth's hypothesis that a full account of the settlement originally stood in
the Pentateuch but was subsequently truncated in order to allow room for
the account in Joshua (A History ofPentateuchal Traditions, 7 1 -74) .
The view that the Pentateuch was composed as an introduction to
the Deuteronomistic History is a plausible one. The Deuteronomistic work
was conceived as a history of the Israelite people fro,m their settlement in

Canaan to the fall of the kingdom of Judah. It is claimed that this was
then supplemented by an introductory section which, in the manner of
some other ancient national histories, was concerned to make certain
specific claims about Israel's most remote origins, laying partiq.dar empha­
sis on the nation as especially chosen and guided by God toward the
achievement of a great destiny in the land of Canaan under the divine

137
INTRODUCTION TO THE PENTATEUCH

blessing. This introduction - the Pentateuch - showed how Israel or its


patriarchal ancestors had been delivered by God on a number of occasions
from apparently inevitable destruction, but especially from slavery and the
murderous intentions of the Egyptian ruler and again, after its departure
from Egypt, by the miracle at the Sea - a deliverance that would remain
in the minds of the readers and give them encouragement to persevere
also in later times of oppression and disaster. The further provision in the
opening chapters of Genesis of stories about the creation of the world and
the behavior of the earliest human beings, and the linking of this "primeval
history" to the ancestors of lsrael by a series of genealogies are not uncom­
mon features of such works.
That the Pentateuch bears the marks of the same "school" of writers
and editors as the Deuteronomistic History has now been widely recog­
nized; the lapse of time between the composition of the two may have
been quite short. As has been suggested above, the period into which such
a work most plausibly fits is either that which followed the destruction of
the kingdom of Judah in 587 or the early postexilic period when the
Jerusalem community was attempting, among renewed difficulties, to
reestablish itself Both works, the Pentateuch and the Deuteronomistic
History; are didactic, concerned to inculcate a variety of lessons to their
readers.
The lessons that the Pentateuchal author(s) were concerned to teach
their readers are numerous; they are by no means confined to the theme
of the promises and their fulfillment. The figure of Moses, which domi­
nates the whole work from Exodus on and gives these books their literary
and religious unity, is clearly presented as an example of faithfulness and
obedience to the divine will and to the mission which had been entrusted
to him. It has been said that Moses is essentially a royal figure - a model
of the ideal ruler. He embodies virtually all the qualities of leadership that
are to be found elsewhere in the Old Testament. It is he who creates a
nation out of a confused, helpless, and often rebellious rabble. Moses is
at once its organizer, its guide, its counselor, and its military commander.
But he is also the one who is close to God, God's spokesman, the mediator
between God and the people, and their intercessor. He promulgates God's
laws. Moses is the supreme prophet, and also performs actions usually
associated with the priesthood, though it is Aaron who normally performs
priestly functions.
Yet Moses is not portrayed as a hero in the usual sense of that term.
The principal actor throughout the Pentateuch is always God. Moses is

138
Reading the Pentateuch

,God's servant, but on many occasions the author makes a point of pre­
senting him as a fallible human being. At the outset Moses runs away after
killing the Egyptian (Exod. 2: 1 1 - 1 5) . Later he is full of self-doubt, believ-
1ing himself to be unfit for the task of facing Pharaoh and leading the
people out of Egypt, making various excuses to avoid the task which God
sets him, and pleading for someone. else to go in his place - and, it is
implied, afraid for his life. Even after he has accepted his commission
Moses exhibits human weakness. He is prone to violent outbursts of anger;
he even expresses dissatisfaction with the position in which God has placed
him; and he is capable of direct resistance to God's commands. · In the end,
although Moses is the object of an enthusiastic encomium (Deut. 34: 1 0-
1 2), his disobedience has earned him a divine displeasure so marked that
he is condemned to die without witnessing the fulfillment of his life's
work, entry into the Promised Land. There is clearly a warning here that
even the most outstanding of God's servants ultimately fall short of what
God demands of them. The message that underlies these passages is surely
that in the end it is God alone - not even his appointed leaders - who
can be trusted.
The people of Israel are portrayed in the narratives of Exodus and
· Numbers as almost continually rebellious. Despite their solemn promises
to do all that Yahweh commanded (Exod. 1 9:8; 24:3), they evince a basic
lack of trust in him and in Moses, frequently complaining of hardships
and the danger of starvation in the desert, questioning Moses' competence
to lead them to the Promised Land, and doing as they please in a variety
of ways - but suffering due and severe punishment for their disobedience
and lack of faith. Together with Moses himself they are, as a generation,
punished by being denied entry into the land; and many individuals perish
in great numbers (e.g.; the 14,700 who protested against Moses and Aaron
when the earth swallowed up Korah· and his 250 associates; Num. 16:35,
49) . These incidents and others scattered throughout these books are
clearly intended as warnings to a later generation which had already lost
the land and · suffered great loss of life in consequence of their rebellion
against God, but were waiting and hoping for a new. dispensation in which
God would once again give them unrestricted possession of the land -
that is, a second fulfillment· of the original promises. That generation
undoubtedly believed that God had the power to grant this; but they are
warned that obedience to his laws is first required of them. This point is
made by implication throughout the narratives · of Exodus and Numbers;
and in Deuteronomy it dominates the entire book.

1 39
INTRODUCTION TO THE PENTATEUCH

In the patriarchal stories of Genesis, where it i:s not a whole people


who are portrayed but individuals, the treatment is somewhat different.
The patriarchs, even Abraham and Noah, are indeed presented as fallible
human beings, and · their individual characters, both good and bad, are
brought out in some detail. But these are the early days when the promises
are newly minted; and the emphasis is not so much on rebellion and
punishment as on divine election. One of the leading themes of these
accounts is the unexpectedness of the bestowal of God's favor, and in
particular of his sovereign choice of improbable individuals to be the heirs
of the promises. T his is true not only of Abraham, who is chosen out of
all mankind to receive the promise, but also of his successors: Isaac, Jacob,
and Joseph are all younger brothers whose elders are passed over in their
favor. In Gen. 3- 1 1 , however, the stories about the sins of Adam and Eve,
Cain, the generation of the Flood, and the Tower of Babel emphasize the
universal rebelliousness of humanity, although the Flood story itself stresses
God's unwillingness to destroy it altogether and offers hope for the future.
God is depicted in the Pentateuch in two ways: as both benevolent
and vengeful. T hese two sides of God's nature are clearly expressed in two
passages that stand out from the surrounding text · as formal theological
statements and so draw attention to themselves as such:

I the Lord your . God am a jealous God, punishing children for the
iniquity of parents, to the third and the fourth generation of those who
. reject me, hut showing steadfast love to the thousandth generation of
those who love me and keep my commandments. (Decalogue: Exod.
20:5-6; Deur. 5 :9- 10)
The Lord, the Lord, a God merciful and gracious; slow to anger; and
abounding in steadfast love and faithfulness, keeping steadfast love for
the thousandth generation, forgiving iniquity and .transgression and sin,
yet by no means clearing the guilty, but visiting the iniquity of the
parents upon the children and the children's children, to the third and
the fourth generation. (Exod. 34:6-7; cf. Num. 1 4: 1 8)

T hese passages are obviously intended to act as both an encourage­


ment and a warning to the · readers .;._ and . a very severe warning indeed.
T he readers needed to be reassured that Yahweh's love, so constantly shown
in the past in the events recorded in these books, would not now be
withdrawn but would continue to be shown in the days to come; and
these· theological statements offer that reassurance. But the continuance of

140
Reading the Pentateuch

that love was conditional - it would be given only to those who loved
God and kept his commandments.
We cannot know precisely whom the author had in mind as the
guilty ones in his own day. However, as some episodes in the narratives
show, it must have been difficult to know what was meant by keeping
God's commandments. Moreover, although these passages speak of God's
forgiving sin and of his being "slow to anger," sinners in the narratives are
not always offered the chance to repent and be forgiven. Thus in Exod.
32, as soon as he became aware of the idolatrous making of the Golden
Calf and without allowing any possibility of repentance, God immediately
determined to destroy the whole people by pestilence, and to start afresh
by making Moses, who had taken no part in their sin, into a great nation
(v. 1 0) . This is a very strange incident which reveals more than one
unexpected side of God's nature. God's threat is averted only by an appeal
to him by Moses, who points out the danger to God's reputation if he
turns on his own people, and actually has to remind God of his own
promises - the oath that he had sworn to Abraham, Isaac, and Jacob (vv.
1 1 - 1 3) . There is a similar incident in Num. 14.
Incidents such as these were no doubt intended to send a chill into
the hearts of the readers. At the very least they were reminders that the
worship of Yahweh was no light matter. Yahweh was a potentially danger­
ous God whose motives and actions were unpredictable and beyond
human comprehension, even inconsistent and changeable. While · the
people gratefully recognized God's goodwill to those who faithfully served
him, it behooved them to be on their guard lest they offend him. Also,
. not all the punishments inflicted in these stories appear to have been
commensurate with the. sins committed, as in the case of the "very great
plague" causing many deaths inflicted as a punishment for the people's
greedy behavior when suddenly provided with quail meat after a long
enforced abstinence (Num. 1 1 :3 1 �35).

The Pentateuch and History

In the earlier chapters of this book the question has been raised to what
extent, if at all, the events related in the Pentateuch correspond to what
actually occurred in sober fact. It is obvious that they cannot entirely
correspond. The accounts of the creation of the world in Gen. 1 and 2
belong, if not to the realm of myth, at least to that of cosmological

141
INTRODUCTION TO THE PENTATEUCH

speculation. The creation of the world was observed by no one but God.
Again, in Gen. 2 and 3 the term "the man'' (Hebrew ha-adham), elsewhere
denoting the human race, clearly indicates that this story is a symbolic
one rather than a factual account of something that actually happened.
So mtich will be generally agreed by all except for those who hold
rigidly to a "literal" interpretation of Scripture - a term that is difficult
to define. The matter is different, however, when we come to the other
narratives of the Pentateuch, where the individuals mentioned are named.
In form, as has been pointed out, many of the patriarchal stories have the
characteristics of the folktale. This alone is not sufficient to deny them all
historical value. But it has also been pointed out above that it has proved
impossible to locate these stories within what is known of the ancient Near
East.
The question becomes most acute when we come to the book of
Exodus. A later generation of Israelites came to believe that their ancestors
underwent a period of slavery and oppression in Egypt and were led out
of that land by Moses, were · saved from the pursuing Egyptian army by a
miracle at the Sea, encountered the god Yahweh at Mount Sinai, where
they were given divine laws, and were then guided - more than 600,000 ,

men with their families (Num. 1 :46) - through a desert in which they
were condemned to travel on foot for forty years, to the border of Palestine
in p reparation for the occupation of the land.
Granted that this story has been greatly embellished in the telling,
is it possible to doubt the historicity of its salient features? Were there no
sojourn in Egypt, no Exodus, no miraculous crossing of the Sea, no
lawgiving on Sinai, no journey through the desert? It has been suggested
that there is some evidence that the historical Israel was mainly indigenous
to Palestine and not an immigrant people from outside. However this may
be, it is difficult to account for the subsequent development of belief in
Yahweh which characterized Israel's subsequent history without allowing
for the arrival in Palestine at an early period of a group of people, even if
few in number, who came from outside bringing with them the news of
a god Yahweh who had miraculously saved them from oppression, which
' then formed the nucleus of the Pentateuchal story.
To be preoccupied with the question whether the Pentateuch records
"history," however, is to miss the point of reading it. In this book I have
been - rightly, I believe - concerned not with the Pentateuch's historical
accuracy but with its religious lessons; for it was to teach those lessons
that the Pentateuch was written. In the Old Testament there are other

1 42
Reading the Pentateuch

books which contain profound religious truths expres�ed in story form.


The story of Job is an outstanding example. Like the parables of Jesus, the
book of Job does not invite speculation about whether the central figure
1 is a historical person or not. Rather, it is concerned with the nature of
God and his world and with the problems of human existen ce. This is
also true in its own way of the Pentateuch, whose teaching I have tried to
present in this book.

For Further Reading

Alter, Robert. The Art of Biblical Narrative. New York: Basic Books and
London: Allen and Unwin, 1 98 1 .
Clines, David J . A. The Theme of the Pentateuch. ]SOT Supplement 1 0 .
Sheffield: JSOT Press, 1 978.
Douglas, Mary. In the WiUerness: The Doctrine ofDefilement in the Book
ofNumbers. ]SOT Supplement 1 58 . Sheffield: JSOT Press, 1 993.
Driver, Samuel R. Introduction to the Literature of the OU Testament. 8th
ed. Edinburgh: T. & T. Clark, 1 909.
Mayes, A D. H., The Story of Israel between Settlement and Exile: A Re­
dactional Study ofthe Deuteronomistic History. London: SCM, 1 983.
Noth, Martin. A History ofPentateuchal Traditions. Englewood Cliffs, N.J.:
Prentice-Hall, 1 972. Repr. Atlanta: Scholars Press, 1 989. (First pub­
lished in German, 1 948.)
Rendtorff, Rolf. The Problem ofthe Process ofTransmission in the Pentateuch.
JSOT Supplement 89. Sheffield: JSOT Press, 1 990. (First published
in German, 1 977.)

143
Index

Abraham: history of, 57-5 8; prom- Day, John, 77


ises to,53-57 Death penalty, 1 09- 1 0, 1 1 2
Alt, Albrecht, 1 1 0, 1 1 2- 1 3 Decalogue, 6, 1 1 3, 1 1 5 - 1 8
Alter, Robert, 23 Deuteronomistic History, 2 , 8 5 , 9 1 ,
Amphictyony, 1 9 96-97, 1 04, 108-9, 1 37
Apodictic laws, 1 1 1 , 1 1 2- 14, 1 1 8- 1 9 Deuteronomy: distinctiveness of,
Archaeology, 1 34 1 04; laws of, 1 20-23; literary .
Astruc, Jean, 1 3 style of, 1 22; monotheism in, 9 1 -
Atra-basis, 45-46 92; relationship to other books,
1 37; speeches in, 6-7; structure
Berith. See Covenant of, 86-87; and vassal treaties, 87-
Birth stories, 70-7 1 88
Blenkinsopp, Joseph, 42, 7 1 de Wette, W. M. L., 1 4- 1 5, 85
Blum, Erhard, 22, 5 1 Diachronic approach, 1 35, 136
Book of the Covenant, 1 1 8-20 Dietary laws, 1 28-29
Documentary Hypothesis, 14- 1 7,
Canaan: commands to kill inhabi- 1 9-20, 23, 1 34
tants, 92; conquest of, 67-68 Dragon and the sea, myth of, 76-77
Casuistic laws, 1 1 1 - 1 2, 1 1 9
Childs, Brevard, 53 E source, 23
Choice of Israel, God's, 94-95, 96- Eichhorn, Johann Gottfried, 14 ,

97 Engnell, Ivan, 20
Covenant: second, 7, 97-98; at Enuma Blish, 36, 37, 41
Sinai, 6, 97, 1 1 4; and ,vassal trea­ · Epic of Gilgamesh, 44, 45-46
ties, 2 1 Etiology, 37-39
Creatii> n stories, 40-42 Ezra, lbn, 1 3

D source, 23-24 Flood story, 45-47

144
Index

Form criticism, 1 7, 1 33 Law codes, ancient Near Eastern,


Fragment Hypothesis, 14 1 07-8, 1 1 9
Laws: and the deliverance from
1 Gemser, Berend, 1 2 1 -22 Egypt, 1 1 4- 1 5 ; of Deuteronomy,
Genealogical lists, 3-4, 47 1 20-23; duplication of, 1 22-23;
God: actions of, 1 O; anger of, 7; in enforcement of, 1 09- 1 0; and the
Deuteronomy, 9 1 , 93; grace of, kings, 1 08-9
34-35, 37; love of, 94-95, 1 40-4 1 ; Lindblom, C. Johannes, 89
nature o f, 140-4 1 ; in the Primeval Little Credo, 1 8
History, 33-34 Love o f God, 94-95 , 140-4 1
Grace, 34-35 , 37
Graf, Karl Heinrich, 1 5 Magicians, court, 73
Gunkel, Hermann, 1 7, 5 1 , 133 Mankind: creation of, 4 1 , 43; temp-
tation of, 42-45
Hammurabi, Code o f, 1 08 Mayes, A. D. H . , 137
Haran, Menahem, 25, 1 27 Mendenhall, George E., 2 1, 67-68
Hardening Pharaoh's heart, 72-73 Midrash, 50
Hexateuch, 1 , 1 36-37 Monotheism, 30, 37, 9 1-92
Historiography, ancient, 50 Mosaic authorship, 12, 13
Holiness Code, 1 24, 1 30 Moses: birth of, 70-7 1 ; commission­
Humanitarian laws, 1 0 1-2 ing of, 7 1 -72; fallibility of, 1 39;
Hurvitz, Avi, 25, 1 26 and God, 79; importance of, 2,
65-66, 1 3 8
Ilgen, Karl David, 14 Motive clauses, 1 2 1 -22
Internal development hypothesis, 67-
68 Name theology, 93-94
Israel: God's care of, 4-5, 6-7; in­ Nicholson, Ernest W., 88
gratitude of, 6; rebellion of, 79- Noth, Martin: on Deuteronomy, 85;
80, 80-8 1 , 1 39; release from on law, 1 07; on the Moses tradi­
bondage, 5-6; shortcomings of, 9 tion, 66; on the Passover, 75; on
the Sea of Reeds, 77; and the Tet­
Jacob, history of, 58 rateuch, 2, 1 8- 1 9 ; on the wilder­
Jesus Christ, 1 ness journey, 78, 82
Joseph, history of, 58-6 1 Nuzi, 20-2 1
Josiah's reform, 85-86
Justice, 101 Olrik, Axel, 17
Order, divine, 128-29
Kaufmann, Yehezkel, 126 Origin stories, 29-30, 36
King, law of the, 100- 1 0 1
Kojiki, 36 P source, 24-26, 123-24
Passover, 69, 74-75
Land promise, 53-54, 57, 95, 98 . Patriarchal age, date of, 20-2 1

14 5
INTRODUCTION TO THE PENTATEUCH

.
Patriarchal stories: composition of, Solomonic enlightenment, 18, 60
49-50, 50-52, 6 1 ; as etiology, 6 1 ; Song of Miriam, 76
promise in, 4-5 ; structure of, 52; Song of Moses, 76
treatment of, 140 Song of the Sea, 76
Pedersen, Johannes, 69, 75 Sumerian King List,35-36
Pentateuch: contents of, 3-8; divi­ Supplement Hypothesis, 14, 1 5
sion of, 63; and history, 50-5 1 , Synchronic approach, 1 35
1 4 1 -43; and Israelite origins, 68-
69; literary quality of, 1 0; as litera­ Tabernacle, 1 29-30
ture, 133; moral and religious les­ Tale of the Two Brothers, 60
son of, 10; New Testament Ten Commandments. See Decalogue
references to, 1 ; theme of, 9 Testing, 78
Perlitt, Lothar, 2 1 Thompson, Thomas L., 20-2 1
Phillips, Anthony, 1 17 "Today," 95-96
Plagues, 72-75 Toledoth, 3 1 -32
Priesthood, 1 02 Torah, importance of, 1 14
Priestly laws (Priestly Code) , 1 23-30 Tower of Babel, 34-35
Primeval History: contents of, 3-4; Tradition history, 1 7- 1 9, 20, 22
date of, 39-40; outline of, 30-3 1 ;
purposes of, 35-39; structure of, Universal history, 29-30, 35
3 1 -35 Unleavened Bread, Feast of, 74
Promise: of blessing, 54, 55; and ful­
fillment, 1 36; of land, 53-54, 57, Van Seters, John, 2 1 -22, 25-26, 39,
95, 98; of progeny, 54-57 40, 5 1 -52
Prophets and prophecy, 88-89, 1 02-3 Vassal treaties, 87-88, 99, 1 1 3
Providence, divine, 4-5 Vater, Johann Severin, 14
von Rad, Gerhard: on Deuter­
Rendtorff, Rolf on Deuteronomic onomy, 88, 89, 92, 1 02-3; on the
theology, 24; on Josiah's reform, history of Joseph, 59-60; on theol­
85; on narratives in P, 25; on the ogy of the Primeval History, 34-
patriarchal stories, 5 1 ; and tradi­ . 35; and tradition history, 1 7- 1 8 ,
tion history, 22 19
Ritual Decalogue, 1 1 6
Rudolph, Wilhelm, 23 Weinfeld, Moshe,
89-90
Wellhausen, Julius, 1 5- 1 6, 73, 123,
Sandmel, Samuel, 1 23 125-26, 135
Schmid, Hans Heinrich, 22, 24 Westermann, Claus, 42, 43, 44, 98
Sea of Reeds, 75-77 Whybray, R. N., 26
Shema, 7 Wilderness journey, 6, 77-82
Sin, 3 4-35, 37 Wisdom themes, 44, 89-90
Sinai, 6, 65 Wolff, Hans Walter, 23
Snake motif, 44-45 Worship, 98-1 00, 1 2 1 , 122

146

You might also like