Isaiah XXI, A Palimpset

Isaiah xxi
A palimpsest
Isaiah xxi
A palimpsest
A. A. MACINTOSH
Fellow, Tutor and Dean of St John's College, Cambridge
CAMBRIDGE UNIVERSITY PRESS

London
Cambridge
New York New Rochelle
Melbourne Sydney
CAMBRIDGE UNIVERSITY PRESS

Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sao Paulo, Delhi
Cambridge University Press
The Edinburgh Building, Cambridge CB2 8RU, UK
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521229432
Cambridge University Press 1980
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 1980
This digitally printed version 2008
A catalogue record for this publication is available from the British Library
ISBN 978-0-521-22943-2 hardback
ISBN 978-0-521-09462-7 paperback
To my wife and to all others

who have cheered on the tow-path
Contents
Acknowledgements
page ix
List ofprincipal abbreviations
Introduction
1 The text and its meaning: a preliminary survey

Verses 1-10
Verses 11 & 12
Verses 13-15
Verses 16 & 17
2 Exegesis and historical background
Verses 1-10
Verses 11 & 12
Verses 13-15
Verses 16 & 17
4
4
39
53
60
63
63
75
91
99
3 Verses 1-10
History and exegesis
Translation and commentary
103
103
116
4 Verses 11-17
131
131
137
Principal works consulted

Indexes
144
151
Vll
Acknowledgements
I wish to record my gratitude to the trustees of the BethuneBaker Fund of the University of Cambridge for a generous grant
to assist in the publication of this work; to J. A. Emerton,
Regius Professor of Hebrew in the University of Cambridge
and a colleague in the fellowship of St John's College, who read
earlier drafts and made many helpful suggestions; to Valerie
Collis for her generous and enthusiastic help with typing and
indexing; to N. C. Buck and M. B. Pratt of St John's College
Library for their unfailing helpfulness; to Cambridge University
Press (and particularly to R. Coleman and Jane Van Tassel) for
their patient care of my work.
A.A.M.
St John's College
2j December igyg
Principal abbreviations
A.N.E. T.
B.A.S.O.R.
B.D.B.
B.H.(3)
B.H.S.
C.A.H.
D.K.
E.T.
G.K.
H.U.C.A.
J.A.O.S.
J.B.L.
J.N.E.S.
J.R.A.S.
J.S.S.
J. T.S.
K.B.
K.B.(3)
LXX
J. B. Pritchard, Ancient Near Eastern Texts Relating to

the Old Testament (Princeton, N.J., 1950)
Bulletin of the American Schools of Oriental Research
F. Brown, S. R. Driver and G. A. Briggs, A Hebrew
and English Lexicon of the Old Testament (Oxford,
1907)
Biblia Hebraica, 3rd edn, ed. R. Kittel (Stuttgart,
1937)
Biblia Hebraica Stuttgartensia, ed. D. W. Thomas
(Stuttgart, 1968)
The Cambridge Ancient History. See Principal Works
Consulted, sect. 7
C. F. A. Dillmann and R. Kittel, Der Prophet
Jesaja, 6th edn (Leipzig, 1898)
English translation
Gesenius-Kautzsch, Gesenius* Hebrew Grammar, 2nd
English edn, rev. A. E. Cowley (Oxford, 1910)
Hebrew Union College Annual
Journal of the American Oriental Society
Journal of Biblical Literature
Journal of Near Eastern Studies
Journal of the Royal Asiatic Society
Journal of Semitic Studies
Journal of Theological Studies
L. Koehler and W. Baumgartner, Lexicon in Veteris
Testamenti Libros (Leiden, 1953)
L. Koehler and W. Baumgartner, Hebraisches und
Aramdisches Lexikon zum Alien Testament, 3rd edn, ed.
W. Baumgartner, B. Hartmann and E. Y.
Kutscher, vols. 1-11 (Leiden, 1967-74)
The Septuagint. See Principal Works Consulted,
sect. 2
ABBREVIATIONS
M.T.
N.E.B.
P.E.Q.
P.T.
1 QIs a
R.B.
R.E.J.
R.S.V.
R. V.
Th.St.Kr.
V.T.
W.O.
Z*A.W.
Z*D. M. G.
Z- Th.K.
XI
The Massoretic Text. See Principal Works Consulted, sect, i

The New English Bible
Palestine Exploration Quarterly
The Palestinian Talmud. See Principal Works
Consulted, sect. 7 (Talmud)
The (complete) Isaiah Scroll from Qumran cave 1.
For the text, see M. Burrows (ed.), The Dead Sea
Scrolls of St. Mark'*s Monastery, vol. 1 (New Haven,
Conn., 1950)
Revue Biblique
Revue des Etudes Juives
The Revised Standard Version of the Bible
The Revised Version of the Bible
Theologische Studien und Kritiken
Vetus Testamentum
Die Welt des Orients
Zeitschriftfur die Alttestamentliche Wissenschaft
Zetiscnrtft dw Deutschen Morgenldndischen Gesellschaft
Ze^sc^riftfur Theologie und Kirche
Introduction
Any attempt to interpret and to understand prophetical texts of

the O.T. must take into account a number of factors which are
so closely related to each other that it is difficult to determine
their order of priority. The factors include: an estimate of the
likely historical background of the text, the recognition of
the literary forms which the prophet uses, the establishing of
the exact meaning of his words and expressions, and the sifting
of those words and expressions so as to recognize any secondary interpretations and additions either by the prophet or by
those who transmitted his words.
It may seem correct to attempt first to establish the text and
the meaning of its words and expressions. Yet considerable
difficulties arise in the case of texts in which many words and
phrases are ambiguous or obscure. For the attempt to resolve
problems of meaning depends to some extent upon an estimate
of the attitude of the prophet and of what he is likely to have
said or to have written in a particular historical situation. Thus,
if the prophet of Isa. xxi is, like that of Isa. xiii, concerned with
the imminent fall of the Babylonian empire in 539 B.C., then it
is possible that he gave expression in xxi 2 to the sentiment:
* Cause all her [sc. Babylon's] pride to cease'; 1 for such sentiments were indeed expressed by prophets as they contemplated
that fall (e.g. Isa. xiii 19). But from the philological point of
view, ' pride' is but one possibility amongst several for the word
nhmx2 and a different understanding of the historical background of the prophecy may at once render ' pride' unlikely
and another possibility (on that assumption) probable.
Alternatively, it may seem appropriate from a preliminary
1 So Eitan; see below. (For all citations by author alone, see Principal Works
Consulted, pp. 144-50.)
2 And some of these are equally suitable to the circumstances of the fall of Babylon
in 539 B.C.
ISAIAH XXI
study of the text to estimate what is its historical background

and thereafter to interpret all the words and phrases of the text
in the light of that estimate. Yet this procedure too may lead
the interpreter astray: first, it may blind him to the possibility
that the words of the text have been reinterpreted and redirected
by those who transmitted them to situations and circumstances
different from those in which originally they were uttered by
the prophet himself; and, secondly, it may lead the commentator to overinterpret the text, to find detailed historical references
where none exists. For, as it is generally agreed, prophets were
not interested primarily in the events of history but in the
interpretation of those events, and what they have given us are
not purely historical records but religious pronouncements (cf.
Otzen, pp. 36f).
What I have said of the relationship between estimates of the
likely historical background and the meaning of the text is true
also of the relationship between the meaning of the text and the
literary forms employed by the prophets. For from the understanding of the words of the text is derived recognition of
literary forms, and yet recognition of literary forms effects the
illumination of words and phrases of the text. Thus, the obscure
words and dialectal forms of the Dumah oracle (Isa. xxi i if), by
virtue of the words and forms that are apparently not obscure
and dialectal, may be interpreted as appropriate to a watchman's dialogue with those who by night await entry to his city
gate. In the light of that supposition an estimate is given of the
meaning of the obscure words and phrases (so Lohmann; see
below).
So to determine the context in which words are given precise
meanings is, however, far from certainly reliable, as is clear
from the existence of other treatments of the easier words of the
Dumah oracle; for such alternative treatments in turn give rise
to other estimates of the context and consequently to different
meanings being given to the difficult words.
The recognition of such difficulties and the fact that they
attend all attempts, at interpreting prophetical texts serve as a
warning to interpreters that guesswork and supposition form an
integral part of their task and that consequently no interpretation is likely to be finally correct. Rather the grounds upon
which our guesses and estimates are made must be continually
INTRODUCTION
reviewed in the light of the increase in our knowledge of ancient

Near Eastern history as well as of the language and literature of
the Old Testament.
The purpose of the present study is to conduct such a review
in regard to Isa. xxi. To that end the text and language of the
chapter will first be examined; the meaning of its words and
phrases will be discussed in the light of interpretations of them
made by the ancient versions, by medieval Jewish scholars and
by modern commentators. In the light of this examination,
consideration will be given to the historical background of the
oracles of the chapter, their literary forms and their interpretation in the tradition of the bible itself as well as in later times.
The Text and Its Meaning: A Preliminary Survey
Verses I-IO
Verse ia, D;-1?7
The words are generally taken to be the title of the prophecy
contained in verses I - I O . As is clearly the case in Isa. xxi 13 and
in Isa. xxii 1 (cf. xxx 6), the title is derived from a keyword
contained within the oracle itself. The main verb *o of this
verse is followed immediately by the word i r a a and it is this
that was regarded as the keyword. All witnesses to the M.T.
substantiate the reading o** *m& except 1 QIs a which reads " m
for laiD and the LXX which, with TO opajjoc TT\S eprmou,
appears to have read simply "Uifc.1
By analogy with the other texts cited and on the evidence of
the LXX, it may be presumed that iîafe) NtPfc is an older
reading 2 and that the later reading & 131D (of which iQIs a 's
Dn *)T7 is perhaps a corruption) arose either as a deliberate
expansion of the original title or from a corruption of a word
originally occurring as the first word of the main text of the
oracle; 3 in either event the change gave the title a different
significance from that of its earlier form.
If *D7fc() XtPfc is indeed the original form of the title, then
being a mere repetition of the word inafe) contained in the
1 For the view that the text of the LXX originally included the word OocA&crcrns
and that the word was subsequently omitted by a scribe who did not understand
its meaning, see Rosenmiiller.
2 It is possible that the preposition (]) was originally prefixed to the word-iaiE
in the title (as it is in the text of the oracle) in the same way as 3 occurs prefixed
to the word my in the title of verse 13 as well as in the text. The Targum, quoted
below, employs the preposition p . On this view the first a may have dropped
out by haplography and the resulting ima was thought to be in need of an
explanatory gloss; cf. Kaiser. On the other hand, the 3 prefixed to p-m xî
(xxii 5) is not found in the title of xxii 1.
3 So Cobb, Marti, Scott and G. R. Driver; for their views, see pp. 8ff.
THE TEXT AND ITS MEANING
text, no further interpretation of the words is necessary. But the

reading D** *m KtPfc, however it arose, is clearly early and must
be intended to furnish, or was seen to furnish, a title suitable to
the contents of the prophecy over which it stands.
As the prophecy with which we are concerned culminates
(verse 9) with the dramatic cry 'Babylon is fallen', it has long
been customary to see in the title & 13113 a reference to that
city. Thus, Theodoret (c. 393^. 458, Bishop of Cyrus in Syria),
comments: eprmov OaAocacrav TTJV (3a(3uAcova KOCAET.1 Similarly,
ibn Ezra and Qimhi, for whom the prophecy as a whole concerns the fall of Babylon in 539 B.C. at the hands of the Medes
and Persians, suggest that Babylon is called D** imD because it
lies to the west of the Medes and Persians and because the word
D"' is capable of that meaning (cf. B.D.B., p. 411a). For ibn
Ezra it is called 12173 by prolepsis (i.e. Babylon is soon to become
a desert), though in Qimhi's opinion mifc is used in order to
indicate that the territory which lies between Persia and
Babylon, and over which the imminent invasion is to come, is
desert.
In fact, these comments are less appropriate to the title QT&
UH than to the phrase *o l a m a within the prophecy itself, for
they are concerned not so much with Babylon as with the mode
and direction of the attack upon it. It may be suggested that
Qimhi's comment at least has been influenced by the interpretation of the Targum (which he quotes): p pn*n fTWto fitoti
Xillft ' The oracle (or march) of the armies which come from
the wilderness', and that this interpretation in turn is ultimately derived from the words *o 12HDD within the oracle
itself, whether or not the Targum actually read in the title (as
is possible) m a a (i.e. with preposition).2
Amongst modern scholars, Dillmann-Kittel and Fohrer
follow these rabbinic commentators in the sense that they regard
the phrase as indicating not Babylon but the desert of South
Babylonia which lies on the Persian Gulf either south-east of
Babylon between the city and Elam (D.K.), or west of the
lower Euphrates (Fohrer); consequently the phrase indicates
the direction of the attack on the city. This interpretation of the
title (which both scholars regard as a later addition) may be
1 For the text, see Mohle.
2 That the Targum also read D*1 is clear from the next phrase in it: *D.
ISAIAH XXI
regarded as plausible only if the contents of verse i do in fact

refer to armies advancing on Babylon from the desert. But that
view rests upon an interpretation of verse i which is far from
certain.
Other geographical allusions to Babylon or its environs are
seen by modern scholars. Thus Delitzsch1 considers that 131a
denotes the great plain on which Babylon stood and which in
the south was so intersected by marshes and lakes, as well as by
the Euphrates, that it floated, as it were, in the sea and indeed
was called 8dAacr<7OC by Abydenus (apud Eusebius, Praeparatio
Evangelica ix 41). Further, such natural features of Babylon are
taken by the prophet, fond as he is of symbolical names, to
refer to its ultimate fate. Thus, Jeremiah (li 13), who according
to Delitzsch (1) knew this oracle, 'was acquainted with its true
significance5.
Secondly, attention is drawn to the Akkadian phrase mat
tdmti 'the land of the sea' (see Schrader, p. 353) used by the
Assyrians to describe South Babylonia (i.e. the land nearest the
Persian Gulf) and it is suggested that on i m a is the Hebrew
equivalent of it.2 Against this view, however, may be set the
arguments advanced by Dillmann-Kittel: if South Babylonia
is called dAaaaa/m^ tdmti then it is unlikely that upper
Babylonia and the city itself should be so called. Further, that
D** 13TD is a Hebrew equivalent of mat tdmti is far from clear; for
why should mat 'land' be translated 'desert'? Dhorme's argument that the presumably accurate equivalent o^ *pK would
have given rise to a bald contradiction in terms and that it was
therefore modified to D** 121ft ' steppe of the sea' is scarcely convincing, for even if it does not merely replace one contradiction
in terms by another, the word -QTO is not elsewhere used to
denote tracts of land covered with lakes and marshes.3
1 His views appear to be dependent upon a number of earlier scholars; see
Michaelis and Vitringa.
2 So Cheyne, P.L, and more recently P. Dhorme in R.B. 31 (1922), pp. 4036,
followed by Erlandsson.
3 Dhorme, followed tentatively by Erlandsson, further suggests that there is in the
title D"113173 a deliberate reference to mat tdmti, Merodach-baladan's abode in
South Babylonia which was also called Bit Takin (Merodach-baladan was called
'son of Yakin'). Apart from the fact that the suggestion rests on a number of
questionable textual assumptions (as has been shown above), it is also unlikely
on the grounds that no further reference to South Babylonia is discernible in the
prophecy.
In conclusion, I accept the view that the original title of the

oracle read imafe) KtPfc1 and that the later reading i m a TO
& has been subsequently interpreted in one way or another as
being an appropriate reference to some aspect of the contents
of the prophecy.
Verse i b , naoia f}K5 *a ni^jpT? *]V?n!? aga n t a o s
The sentence appears to consist of a main verb (xa), apparently

without an expressed subject. Whoever or whatever 'comes5 (or
'has come') 2 does so from the desert (laTOfc) which is further
defined as a 'terrible land' (n*m2 <pN)-3 The manner of his or
its coming is described by a simile, *]V?nV 3U3 niDOD 'like
whirlwinds sweeping along in the Negeb'.
The ancient versions witness to a text which differs little (if
at all) from the M.T., though the LXX and Peshitta appear to
have understood the whirlwind as a singular (cos Kcrrcciyls/Â;
HCV) and to have divided the phrases of the sentence rather
differently. The LXX's feminine epxoiisvrj (cf. Pesh.: wff)>
which corresponds to the M.T.'s an, may be held to indicate
that this version (so the Pesh.)4 took the whirlwind to be the
subject of xa as well as of the gerundial infinitive construct
*f?nV. The Targum, with p a n KEP *D ('like the waters of the
sea which sweep along'), appears to have interpreted the D* of
the M.T.'s title as part of the simile;5 its main verb ]n*0 which
answers to the M.T.'s xa is predicated of the armies (ptpfc)
which it has supplied as a subject.
The Targum's interpretation of the main verb xa, viz. that
an invading army must be understood as its subject, is followed
by ibn Ezra and Qimhi (cf. Rashi) who again refer to the
expedition of Medes and Persians against Babylon which arrives
from the desert between the city and Persian Sea. The inter1
2
3
4
So, essentially, Duhm, Marti, Gray, Procksch, Kaiser, Wildberger, etc.

So Wildberger, on the grounds that -ran in the following verse is perfect.
On the 'terrible land', see further below.
In the case of the Peshitta, it is not immediately clear what is the subject of the
feminine verb(s). However, the noun C/C/D is the only feminine noun in the verse,
and for this reason the judgement is made that, as in the LXX, the whirlwind
is the subject.
5 Though not to the exclusion of the whirlwinds which are also included in the
strange phrase jnx pVw^ya p i n xa* **&>
: * like the waters of the sea which sweep
along in whirlwinds do they (sc. the armies) come'.
ISAIAH XXI
pretation is adopted by a number of modern commentators,

and, like the Targum, they seem to assume it from the context
(see Rosenmuller; Delitzsch; Cheyne, P.L; D.K., and Kaiser).
Amongst the rabbinic commentators, Saadya alone does not
follow the interpretation of the Targum. The words *
are rendered by him (in Arabic) SnVtf p <nirMl!Tl?KD
('Un ev&iement, semblable aux tourbillons. . . arrivera du
desert').
It may be assumed that he wished to reproduce more literally
the sense of the M.T. with its unexpressed subject, though it is
interesting to note that in his translation he appears to feel the
need to supply some indication of how the verb is to be understood. From what follows in verse 2 of his translation, viz. that
the inhabitants of the terrible land say that they have been told
a hard prophecy, it appears that he understands (the news of)
the terrible event (*IK) to be the subject of the verb K3. In any
case this is an alternative explanation of the verb *o which is
adopted by those moderns who, like Saadya, do not follow the
Targum's understanding of it (so e.g. Duhm, Gray, Marti and
Procksch). For them the (unexpressed) subject of *o is understood as the terrible import of the vision described in verse 2,
later (verse 9) precisely defined as the fall of Babylon. For these
commentators the desert from which the news comes is the
Syrian desert which lies between Babylon and the (Palestinian)
dwelling place of the prophet.
As I have observed in my comments on the (M.T.) title of the
oracle, some modern scholars have seen in a*1 a corruption
of the initial word of this prophecy. In part, attempts to restore
this supposed first word are designed also to alleviate the
problem of the unexpressed subject of the verb N3. Marti 1
suggests that o** is a corruption of an original T\V2T} ' a roaring as
of storms which sweep on in the Negeb! It comes etc' By
making the simile clearly attach to the word rp&n he frees *o
to be read more closely with what follows in verse 2. The
construction, however, remains somewhat awkward and the
emendation purely conjectural.
Secondly, R. B. Y. Scott,2 on the basis of the reading of
1 He rightly rejects Cobb's earlier suggestion that Ds is the perfect of a lost on*
with the same sense.
2 In V.T. 2 (1952), 278ff.
1 QIs , suggests that the beginning of verse i is corrupted by

haplography from an original niDIODff13112112 Km etc. and
he suggests that the main verb *o was originally Q^|. The
meaning of the text of the oracle so restored is: ' Words are
coming like storm winds which sweep on in the Negeb.5 Against
such a restoration and consequent emendation of the text are
the telling arguments of Kaiser, who urges that K3 when
predicated of 131 means 'fulfil, come to pass5 and never denotes
the initial reception of a revelation.
More plausible is the solution proposed by G. R. Driver 1 and
adopted by the N.E.B. For Driver D*1 is read DS (DV) and taken
to mean ' (stormy) weather' after the meaning ' stormy weather,
storm demon' attested for the Accadian cognate word umu.
Driver suggests that the word nv bears this meaning also in
Judg. xix I I , and it may be noted that K.B.(3) (p. 384) suggest
the similar meaning ' wind' for nv in Song of Songs ii 17 and
iv 6 2 and 'storms' in Zeph. ii 2.3 The translation offered by
Driver runs: ' (Stormy) weather, like tempests in the South in
their passing, comes from the wilderness, from a terrible
land. 54
Driver's treatment and interpretation of the text have the
great merit that they furnish xa with a subject and afford a
clear and straightforward construction for the verse which as a
whole becomes a simile or parable. For Driver the parable
alludes to the westward sweep of the Elamites and Medes whose
armies come across the intervening desert to overrun Babylon.
However, it is more natural to suppose that the storm is a
parable of (or is coincident with) the coming to the prophet of
his 'harsh vision' 5 (verse 2) rather than of invading armies.
For while the desert is not further defined in the prophecy, the
simile within the verse itself ('like whirlwinds') refers to the
1 P. 46.
2 For a different view, see W. Rudolph, K.A.T. xvn, following D. W. Thomas in
Expository Times 47 (1935-6), 43if.
3 It is not possible here to examine in detail the theory of Driver and Baumgartner
other than in relation to Isa. xxi 1. It is possible to argue, however, that the usual
translation ' day' makes good sense in Judg. xix 11 and Zeph. ii 2 and that it is
possible in the passages from Song of Songs.
4 Procksch, commenting on xa, similarly suggested that what comes is the (ghostly)
storm; for him, however, that meaning is discernible by carrying over the sense
of an initial nTprp. (sic, after Marti).
5 So, with a different understanding of the text, Marti, Procksch, Gray, Duhm.
10
ISAIAH XXI
Negeb, i.e. the desert of southern Palestine,1 a place of sudden

swift-moving storms (cf. Ps. xxix) which are alluded to elsewhere in the O.T. as a symbol of Yahweh's impending intervention in the affairs of his people (Judg. v 4, Hab. iii 3, Zech.
ix 14). On this view, the imagery of the verse indicates that
Yahweh has revealed to the prophet his 'harsh vision' (verse 2),
and that he is the author of the awesome events which are
proclaimed by it.2
HNTll *pK 'a terrible land'. The phrase is most likely to be
construed as in apposition to *m& and as referring to the same
desert; cf. Deut. i 19 and viii 15 (so Rashi).
In iQIs a n*m2 is found superimposed upon the word npim
which has been deleted. The Peshitta, too, implies the reading
npim (V rhyqf) and these facts suggest a textual tradition at
variance with the M.T. If this is the case then memory of it may
have survived also to the time of Qimhi who, commenting on
Hani! f*1N, states that the land of the Medes and Persians is
meant 'for they are far from Babylon'. He continues: pimn
pimn KT OINIW ^ *rm *nj7 'the far-off is terrible because
men fear it'.
On the other hand, the evidence of iQIs a and of the Peshitta
may be taken to indicate a very early interpretation of nNTU
which subsequently found its way into some textual traditions.
If the form-critical observations on the prophecy set out below
(chapter 3) are correct, then the word npim is a later interpretation of the word nXTti.
The Targum interprets the phrase pôn na KTMIwr Ksnttd
'from the land where terrible things are wrought' and this too
may reflect the interpretation of the ' terrible land' as that of
the Medes and Persians. Certainly ibn Ezra and Qimhi (the
latter quoting the Targum) take this view.3
rjVn1?. For an infinitive construct with V to describe attendant
circumstances or to define more precisely, see G.K. 1140. The
word depicts 'the swift passage of the storms' (Driver).
1 The argument that an here means the (Arabian) desert south of Babylon
(Delitzsch followed by D.K.) is most unlikely, for the word su is not elsewhere
attested as referring to any other specific geographical area than that to the
south of Judah.
2 Wildberger similarly finds such an allusion in the phrase.
3 They are followed by Delitzsch, and tentatively by D.K. For Duhm, Babylon,
the scene of terrifying events, is meant.
T H E T E X T AND ITS MEANING
II
Verse 2a, ***? Tjn ntfj? mm

The verse begins with the phrase 'A grievous (or hard) vision is
declared to me' (R.V.).
The fact that the passive verb is third person masculine
singular and that mm is feminine suggests that the const uction
is that described in G.K. i2ia-b; the passive is usedrimpersonally with the object of the active construction still subordinated in the accusative, i.e. * There is/was told to me a harsh
vision' (cf. Gen. xxvii 42). Saadya's M S M12 KrmK lp 'we
have been informed of a harsh vision' is grammatically
accurate. 1
H^pT 'harsh' or 'calamitous'; cf. with Cheyne 1 Kings xiv 6.
The problems of the passage in which the phrase occurs are
largely exegetical and must be considered later in regard to the
whole form of the oracle. Here it may be observed that, on the
whole, rabbinic commentators considered that the vision was
harsh specifically for Babylon (ibn Ezra, Rashi, Qimhi), though
they follow the Targum in interpreting *? as referring to the
prophet himself. Saadya, however, understands ^ as referring
to the inhabitants of the terrible land (Vlp^ riVriN) 'its inhabitants say. . . ' , which suggests that he understood the phrase as a
quotation. Dillmann-Kittel and Gray consider it significant
that the vision is heard rather than seen by the prophet, and
they refer to verse 6 where a watchman relates to the prophet
what he has seen (i.e. the watchman sees, and the prophet
hears).
Verse 2b, "nitf i^Wgi ilia
The four words evidently characterize in an epigrammatical
way the contents of the harsh vision.
*M. The word is generally taken to denote treacherous conduct,2 specifically in matters of property and rights (e.g. Ex. xxi
8), in regard to adultery (e.g. Jer. iii 20), to disobedience (1
Sam. xiv 33), to broken trust (Job vi 15) and to rebellion (Judg.
ix 23).3 Rosenmliller suggests the meaning 'plunder' for this
word because the parallelism demands it and because, he says,
1 Qimhi assumes that the passive verb implies the agency of God.
2 See B.D.B., K.B.(3> and Gesenius-Buhl.
3 Michaelis's view that 'contempt' is an older meaning of the word does not affect
its usage in the present text.
12
ISAIAH XXI
the verb is very often translated by Aramaic m ' to despoil' in

the Targum. Yet such a meaning cannot be said to be clearly
attested for the verb, and it is not indicated in this verse either
by the ancient versions or by the rabbinic commentators. The
Targum and Peshitta suggest 'oppression' (Targum: 03N,
which often answers to the Hebrew pt??; Pesh.: tlm). The LXX's
dOeTeiv suggests 'breaking faith5.1 Amongst the rabbinic commentators, Saadya renders TTKJIVN which implies treachery,
deceit, and ibn Ezra uses as a synonym the verb THE1? 'to
rebel'. Finally, it should be noted that the word *rn occurs
elsewhere in Isaiah, in xxiv 16 and xxxiii i; but these verses are
generally and most plausibly regarded as derived from the
present text (see below).
The verb ntf is well attested with the meanings ' deal violently
with, despoil, devastate' (see B.D.B., K.B. and K.B.(3)), and
translations of it in this verse, both ancient and modern, show
no divergence from this view.
The main problems attaching to the saying are, however,
syntactical and exegetical. In the M.T. we are presented with
two pairs of qal participles and in each case the first occurrence
of the participle is prefixed by the definite article. These defined
participles are clearly designed to indicate a specific (enemy ?)
force. Further, the participles (without article) which in each
case follow denote either a present continuous action on which
the (enemy?) force is engaged (so the LXX, Vulg. and Pesh.),
or an action on which he is about to embark (so Bright in
Peahens Commentary, p. 504). A third possibility is that suggested
by ibn Ezra who gives to each pair of participles respectively a
relative and permissive sense: b'DV T)3QVI YH&1? nSTttf & 'He
who wishes to be rebellious and treacherous will be able to be
so.'2
The Targum alone of the ancient versions translates the
second word in each pair as a passive (e.g. poatnafcTOUK):'The
oppressors are oppressed and the spoilers are despoiled', and
this interpretation is followed precisely by Saadya. Rashi and
1 See Liddell and Scott, p. 31. The first part of the Vulgate's qui incredulus est
infideliter agit defies (my) explanation.
2 Galling suggests a variant interpretation of the syntax: for him the defined
participles are designations of vocation, the undefined having permissive force:
'Permitted is it for the robber to rob. . .etc' The words then form part of the
divine commissioning of Elam and Media.
13
Qimhi, too, seek to understand the phrase in this way, though

they attempt to harmonize the M.T. with the interpretation of
the Targum; thus Rashi states that TTttPH is in the accusative,
and 'another comes and plunders him [the plunderer] 5 . 1 For
both commentators the reference is to the king of Babylon once the plunderer, now the plundered.
Such an understanding of the phrase before us may be
detected in the Hebrew text of Isa. xxxiii i which is generally
taken to be a later expansion and reworking of the words of
Isa. xxi 2.2 It is here suggested that the Targum's translation of
Isa. xxi 2 is influenced by the M.T. of Isa. xxxiii i and that it
attempts to harmonize the two.
In attempting to determine who is referred to by the participles of the phrase before us we have noted the view of Rashi
and Qimhi that it is the king of Babylon.3 In the O.T. itself
Jeremiah refers to the invading Babylonian foe under the term
nW (vi 26, xii 12, xv 8), as he does in his oracle against Moab
(xlviii 8, 18, 32). On the other hand it may be noted that the
word "nW is used in Isa. xvi 4 of an enemy who attacks Moab,
thereby causing fugitives to flee to Jerusalem. The Moab
oracles of Isa. xv and xvi are notoriously difficult, but one view
is that at least part of them may go back to the time of the
Assyrian conquests of Palestine in the latter part of the eighth
century. If this is the case it is possible that the verb TIW constitutes a reference to the Assyrian4 whose coming is elsewhere
in Isaiah associated with the words V?? 'booty', T? 'plunder'
(viii 3f, x 5f). It should also be noted that in Jewish tradition the
word TTO was sometimes associated specifically with Sennacherib ; thus Qimhi, notwithstanding his reference to the king of
Babylon (see above), states in his comments on Isa. xvi 4, and
on xxxiii 1, that the participle TTW is a reference to that Assyrian
king.
Finally, it should be noted that amongst modern commentators (e.g.) Dillmann-Kittel, Marti, Procksch and Galling take
1 Thereby giving to the second participle an impersonal subject.
2 The same interpretation is also found in the Targum of Isa. xxiv 16.
3 Amongst modern commentators, this view is adopted by e.g. Delitzsch, Ewald,
Duhm and Gray.
4 So e.g. Cheyne, P.I.; Procksch, and more recently Erlandsson. That Jer. xlviii is
clearly dependent on Isa. xvf suggests a pre-exilic date for the latter. See further
below.
14
ISAIAH XXI
the view that the words characterize the action or imminent

action of the Elamites and Medes and they should be taken
closely with the words that follow.
"HD njx D^57 ^B. These words constitute perhaps the most
important piece of evidence within the text as to its historical
background and date. As happens elsewhere (see e.g. vii 2) 1 the
feminine singular verbs (here in the imperative) are used in
addressing collectively the Elamites and Medes. The verb nVtt
is used in vii 1 to denote an attack on Jerusalem 2 but it may be
taken to mean attack in a general sense rather than ascent.3
Apart from the LXX and the Peshitta, the ancient versions
and later commentators are unanimous in understanding "HIS
as an imperative of the well-known verb Tis cto besiege' (B.D.B.,
p. 848b). The rendering of the Peshitta wtwry (and the hills of
(Media)' clearly indicates that the same Hebrew text was read,
though the translators understood the word as the construct
plural of the cognate noun *ns 'rock' (B.D.B., p. 849b). The
resulting sense 'Go up Elam and hills of Media' is hardly
satisfactory and it is therefore probable that the Peshitta was
mistaken in interpreting the Hebrew word as a noun rather than
as a verb.4
The LXX's oi TTpea^sis (TCOV TrepcrGov) suggests that it too
understood the word as a noun, but here as the construct plural
of TB ;envoy, messenger' (B.D.B., p. 851b). iQIs a has *TE and
this may constitute (manuscript) evidence of the same tradition.
Again, however, the LXX's rendering is unlikely to indicate a
better way of understanding the text (cf. Seeligmann). No
intelligible meaning can be given to their rendering of the preceding words (DV57 *bv) eir* sjioi oi AiAaiaiTcci where ^57 is read
as the preposition *?S7 with first person suffix. It remains more
likely that the two words ^x and m s are verbal forms.
1 Cf. G.K. 145k.
2 Cf. Bach, p. 63.
3 Gray's argument - that, because the word is geographically inappropriate to an
advance on Babylon, it has been deliberately transferred by the prophet to that
use from its proper use in regard to an attack on Jerusalem - is overingenious.
4 It should be noted that Diettrich records the alternative reading wtwr, which he
interprets as an imperative of the verb twr 'to fly'. If the reading was original, it
would not indicate a different Hebrew text. In any case, wtwry (or wtwr in the
singular?) 'and hills of is likely to be the original reading because it can be
explained in relation to the M.T.'s ms, whereas, because no Hebrew word "lix
is attested with the meaning 'to fly', wtwr meaning 'fly' cannot.
15
Ibn Ezra compares the verb m s with Deut. xx 12 where the

verb (???) is used of besieging a city. In this text, however, the
verb is followed by an indirect object rp!?? 'against it' (sc. the
city). Now, if m s in Isa. xxi 2 means 'besiege', the problem
arises that only here is it used absolutely (i.e. without object).
The problem is apparently eased by reference to verse 9 where
notice of the fall of Babylon is recorded, and from this point of
view the absence of an object is deliberate and heightens the
tension of the oracle (cf. Wildberger).
The verb nVtf is well attested in the sense 'to go up (toward)'
and according to this usage has virtually the sense of English 'to
attack' (see B.D.B., p. 748b (2c)). While *m 'to besiege' may
be regarded as an appropriate parallel expression to n*?S7 'to
attack', the same is also true of the verb T12 'to show hostility,
to vex' (see B.D.B., p. 865b). For this verb too is used of attack
and of military harassment (see Num. x 9, xxv 17, xxxiii 55). It
is then possible that the consonants m s constitute an imperative
of the verb m s 01* > ^ ) .
Such an understanding of the consonants has the merit that
it diminishes the difficulty occasioned by the absolute use of the
verb. For while the verb TIE also is not used elsewhere in the
bible absolutely (i.e. without object), its meanings may be held
to be more capable of such absolute use than are those of the
verb IIS.
Ibn Ezra, commenting on this verse, considers that the whole
oracle, from DV57 ^V onwards, is put by the prophet into the
mouth of the king of Babylon on the grounds that, as is well
known, joy rather than terror and despair (verse 3) was the
feeling of the prophet as he contemplated the fall of Babylon.
He views the phrase before us as a further quotation within the
king's speech. Thus the king, when he heard the shouts of the
attacking soldiers: 'Go up Elam, etc.', soothed Babylon's sighs
by the use of musical instruments. The fact that ibn Ezra has
interpreted the oracle in the light of Belshazzar's feast in Dan. v
will be considered below, but his view that the phrase before us
is a battle-cry of the Elamite and Median soldiers should here
be noted.1 That such quotations are likely to be included in the
1 It is possible that Saadya too considered that the phrase was a quotation, for he
introduces verse 2 with the phrase 'Its [sc. of the terrible land] inhabitants
say...'
l6
ISAIAH XXI
text of Isaiah may be confirmed by reference to e.g. Isa. vii 2

where the report of an alliance is quoted verbatim.
By contrast, modern commentators are virtually unanimous 1
in seeing the command as on the lips of Yahweh, who thereby
commissions the Elamites and Medes to do his will in bringing
about the fall of Babylon.2 For such commentators the Elamites
and Medes signify the united Persian kingdom of Cyrus and the
attack is that of 539 B.C.3 To those like Kleinert and Cheyne
(P./.) who consider that the prophecy concerns an attack on
Babylon at the end of the eighth century, the Elamites and
Medes are auxiliary forces in the Assyrian army (as the Elamites are alleged to be in xxii 6), drawn from those parts of
Sargon's empire that he had annexed. For further discussion of
the historical background, see below.
Verse 2c, ^ n
nnmg VD
The Massoretes (so Vulg. and Pesh.) understand the verb as a

hiphil first person singular of the verb mtP. The Vulgate and
Peshitta take nnmN as the noun nn3K 'sighing' with the third
feminine singular suffix (omnem gemitum eius cessarefeci; cf. Pesh.).
The LXX witnesses to a different text or to a very different
understanding of the text: vuv orevd^co Kai TrapaKaAeaco ejjauTOV.4 In the M.T. the final n of nnniN is pointed with raphe,
though in some MSS. the reading nnniX is attested.5 According to
Rashi there is no mappiq because the word is emphatic and
indicates all the (world-wide) sighing caused by Babylon (cf.
Qimhi). Ibn Ezra, however, is content to refer the word nnniX
(with suffix) to Babylon's anxious sighing which is to be stillfed
by music at the command of her king.
Rashi's interpretation of the pointing of nnmN on the one
hand, and that of ibn Ezra on the other, is reflected in modern
1 An exception is Erlandsson, who, like ibn Ezra, sees the phrase as ' the battle cry
of the allies in their attacks against Assyria'.
2 For a thorough exposition of this summons to battle as a literary category, see
Bach.
3 Cf. Qimhi who states that the whole prophecy concerns the Persians' attack on
Babylon.
4 For the view that LXX has probably understood the radicals m both as a form
of ma (OTEV6CCO) and of om (TrapotKaAeaco), see Goshen-Gottstein.
5 See Goshen-Gottstein (apparatus m).
17
commentators' views of it. For some the word is understood to

be formed with poetic double feminine ending (see e.g. Ewald);
for others (so e.g. G.K. ()ie, D.K. and Wildberger) the third
feminine singular suffix is affixed to the word but it is pronounced softly by reason of the n which follows it.
On the view that the text means * I have made all her sighing
to cease', a number of exegetical problems present themselves
to modern commentators, who differ in their solutions of them.
First, it is not clear whose sighings are made to cease; are they
those (last sighs) of Babylon (Marti) or of those who sigh over
her or of those who have been made to sigh because of her
oppression (see e.g. D.K.) ? Secondly, who is the subject of
TQtfn? Thirdly, is that word used appropriately with nnmN
'sighing' as an object when it is normally used of the ending of
e.g. mirth or pride (so Cheyne, P./., who compares xiii 11 and
xvi 10, and Duhm) ? The recognition of these difficulties has
led many scholars to the conclusion that the text is corrupt and
consequently they have sought to emend or to delete it. It is not
possible or desirable to list here all such proposals, but the most
commonly accepted for nnniN are those which seek meanings
such as 'jubilation' or 'arrogance' (e.g. Duhm p*o, Gray
niYKO; cf. Wildberger). More plausible is Eitan's treatment of
the word; he obtains for it the meaning 'pride, arrogance' by
supposing the word is here cognate to Arabic nhwt which bears
this meaning, and by supposing that the initial x is prosthetic.
Against his view, however, are the considerations that a Hebrew
nm(tf) cognate to Arabic nhwt is not elsewhere attested and
neither the versions nor the rabbinic commentators indicate
such a tradition.
With regard to TQtPn, if that reading is to be retained, the
subject can, as Marti maintains, hardly be other than Yahweh;
but as he also observes, the sudden introduction of Yahweh as a
subject in the first person is harsh.1 As a consequence many have
accepted his initial suggestion2 that the word be read as a second
feminine imperative W?^n parallel to the foregoing imperatives
1 Wildberger, on the other hand, argues that the sudden introduction of Yahweh
('I') as the instigator of the events of history is entirely consistent with O.T.
views of history.
2 His final conclusion, however, is that the words are a secondary addition which
gives expression to Jewish hatred for Babylon and threatens its end as a world
power.
l8
ISAIAH XXI
(cf. Gray and Procksch). More recently G. R. Driver,1 following

this view of VDttfn (though he emends it to the plural form
WWn) has understood the Hebrew nnmx to be capable of
bearing the meaning 'weariness' as well as 'groaning'. This he
does by appeal to Accadian anahtu 'toil, weariness', and he finds
such a meaning appropriate to the Hebrew root nix in Ex, ii
23, Joel i 18 and Lam. i n . On this view the exhortation to the
Elamites and Medes concludes with the words 'Lay aside all
weariness' ('No time for weariness!' - N.E.B.).
Driver's solution is certainly a possibility; from the philological point of view it is unexceptionable, and, provided that
the text (iMffn) is emended it can be said to fit the context
tolerably well. On the other hand there is again no evidence to
support it in the ancient versions nor amongst the medieval
Jewish commentators and his statement that 'exhaustion' fits
the other three texts better than the usual meaning 'sigh' or
'groan' is highly questionable. On balance and unless further
evidence becomes available, Driver's view should be treated
with caution.
Amongst rabbinic commentators Saadya alone appears not
to have taken the usual view of nnmtf. He renders the phrase in
which it occurs: nnV&B lp ^*rn SPfci, which Derenbourg
renders: 'Tous (les cris de): Gare! je les ai supprim^s.'2 It is
unfortunately not possible to be certain why Saadya considered
the Arabic word hdr to be a suitable translation of Hebrew
nnmN, though one possibility is that he saw (or knew a tradition
which saw) in nnnaK a form of the root ma.3
If this assumption is correct, appeal may be made to (the
late) Isa. xxx 15 and to the noun nna in the phrase final T\2W2
pBgjP 'in returning and rest shall ye be saved' (R.V.). The
words form part of Yahweh's recollection of his appeal to his
people to adopt an attitude of quiet, trustful waiting, of its
rejection by them in favour of military might, and of Yahweh's
subsequent resolution to punish them by force of arms. The
oracle (though late) reflects accurately the teachings of Isaiah
ofJerusalem (cf. vii 4, 9; viii 6, etc.). In Isa. xxi 2, therefore, the
1 P. 47. He is followed by the N.E.B.
2 Derenbourg's translation does not appear to take into account the first person
suffix in "Hin. Does Saadya mean 'I have discontinued all my watchfulness', and
intend the phrase to refer to the careless Babylonians ?
3 The suggestion was made to me privately and tentatively by S. Morag.
ig
word nnmK may be a corruption or later adaptation of the

word hrtt, and the phrase TQtfn (n)nni (tf)1?3 indicate that, as a
result of the disobedience of his people in trusting in military
alliances (and such may be the significance of the cry DVS? ^
etc.; see below), Yahweh has brought an end to (the possibility
of) the quiet, trustful waiting 1 he had previously enjoined.2
Verse 3, WJ81 nffr 'T?? "pTpg D*T?
The meaning of the words contained in this verse is reasonably
clear and their usage to describe terror is well attested (see e.g.
Cheyne, P.I.; Kaiser). There is some discussion concerning the
force of p in the second part of the verse; thus the LXX,
Peshitta and Targum (TO [xr\ dcKoOcrai, dV lsmc, r&tpVfc) followed
by ibn Ezra and Qimhi 3 understand the p to be privative
while the Vulgate (cum audirem) understands it to have consequential or causative force (i.e. in consequence of what I hear,
I am bent with pain - so e.g. Duhm). There is some force in
Marti's argument that p Vtt at the beginning of the verse conveys adequately the consequential meaning and that, therefore,
the p is best regarded as privative. The verse may be taken to
mean: 'in consequence of this (p *?*?, i.e. in consequence of the
revelation of verse 2) pangs have taken hold of m e . . .so that I
cannot hear, e t c '
The exegetical problem of this verse (and of verse 4) concerns
the identity of the speaker, and it turns on the view taken of the
prophecy as a whole. As has been noted above, ibn Ezra poses
the problem forcefully:4 ' I think that the whole of this passage
1 Such a meaning has been suggested for the verb mix in Hab. iii 16; for references
see B.D.B., p. 628a. That the semantic range of the root ma extends much further
than 'rest, repose' is clear from e.g. O. Eissfeldt's study of the meaning of the
word in Isa. vii 2; see Kleine Schriften, vol. in (Tubingen, 1966), pp. I24ff. For him
nm in Isa. xxx 15 means specifically 'covenant, fidelity' ('Vertragstreue') and
naw (following Duhm) 'withdrawal from war' ('Abgewandheit vom Kampfe').
2 Obermann suggests that nnms concealed nnm, (i.e. nm with third feminine
singular suffix; for him the word denotes ' ease' and he takes the phrase nnm vCiD
(KVD for ^3) to mean 'destroy (Babylon's confident) ease'. The difficulty arises,
however, that nm in Isa. xxx 15 is used in a good sense, and nm * ease' is better
attested in rabbinical Hebrew than in biblical.
3 So e.g. Rosenmiiller, D.K., Marti.
4 Friedlander's translation here is not followed precisely, as he has failed to understand ibn Ezra.
20
ISAIAH XXI
from nnniK VD to the end of the chapter is the speech which the
prophet put into the mouth of the king Belshazzar - for it is well
known that the prophet rejoiced in the fall of Babylon. How
then could he have said "Therefore are my loins filled with pain
. . . " ? 5 1 So with minor variations Rashi and Qimhi; thus
Qimhi states that the prophet speaks in the name of every
Babylonian or in the name of Belshazzar. Rashi, however,
proposes as an alternative explanation that it is the prophet
who laments sympathetically over the punishment of the
peoples.
Modern commentators are unanimous in seeing the prophet
as the speaker, and have resolved the problem presented by
the unexpected horror of a Jewish prophet at the fall of Babylon
by following Rashi's second explanation (so e.g. Ehrlich) or by
seeing in the words a reflection of the psychological state of the
prophet who in ecstasy saw his vision, but did not in that state
perceive the salutary implications for his people (see particularly Duhm, and cf. Fohrer).
On a simpler level, Kaiser's observation does much to alleviate the problem. He shows, by reference to xv 5; xvi 9, 11, and
Ezek. xxi 11 that the language of personal involvement on the
part of a prophet ' may serve simply to emphasize the severity
of the events foretold5.2 But if his observation alleviates the
problem, it does not entirely remove it. No other Jewish prophet
was struck by horror at the (coming) fall of Babylon in 539 B.C.
or regarded that fall as involving anything other than redemption and liberation for his contemporaries.
Verse 4a, "inns? rart ^
nrr
The word MV may be taken either to refer to the mental faculties of the prophet (so e.g. Saadya, Duhm, Gray, Fohrer) or, as
I think, more physically to the heart. For both verses manifestly
refer to physical symptoms;3 further, the word nVT) denotes
physically 'to wander about' (see B.D.B.) and, predicated of
the heart, may be held to describe its fearful commotion; cf. Ps.
1 That the problem was felt also by the Targum can be inferred from the fact that
it replaces the first person singular of the M.T. by the third person plural. It is
not, however, clear who ' they' are.
2 Cf. Wildberger's similar (though different) observations.
3 Cf. Wildberger's (general) comments, p. 777.
21
xxxviii i i , and the similar phrase imno ^V lit. 'goes around,

is in commotion5.1
2nrW2 JYlS^D. There are considerable variations in the
versional translations of the phrase, though the general import
is clear. LXX: f) dcvouioc |ie (3cnrri6i;2 Vulgate: tenebrae stupefecerunt me; Targum: junrm pmsrni XpS7 'distress and terrors
have taken hold of them'; Peshitta: wswrri* zj> cwny 'dizziness
has terrified me'. For niS^D ibn Ezra compares nnsV& 'her
pride' in i Kings xv 13 and argues that on this view 'the
boasting of the Persians terrified me' (sc. King Belshazzar). But
he admits the meaning 'trembling' as an alternative, comparing
Job ix 6. It is this latter view which is generally accepted by
modern lexicographers and commentators.3 For Qimhi, the
phrase indicates 'trembling' and echoes specifically Dan. v 6,
Jtfpl *oV *T7 nrODIKI 'his (i.e. Belshazzar's) knees smote each
other'. The phrase may be rendered 'trembling, convulsions
have seized, overwhelmed me'.
Verse 4b, n^q 1 ? *? *% ?.ttfJ 1 $ m
From ancient times this phrase has been translated and interpreted in a number of different ways. First, the LXX f) <p\JxA
uou ecpeorriKev eis 9o(3ov apparently read WD2 for r\vn and took
the phrase as a further definition of (the prophet's?) terror.4
scpeorriKev construed with f) yvxr) as subject may be taken as a
way of rendering an impersonal rmn1? ^b Dtp; cf. the Vulgate's
posita est mihi, the Peshitta's smw ly and Saadya's *? *IKS 'est
devenue pour moi'. This interpretation of the phrase fits the
context admirably (cf. Hitzig). The Targum renders by rra *inK
-inn*? pnV mn Jimxrm 'their place of refuge has become a
destruction (or terror) for them', rj^j is rendered OKOTOS
by Aquila and Theodotion and this denotation is further
interpreted to indicate Babylon by Jerome who compares Isa.
xiii 2, Hgttfa ^D bv5 which he renders in montem caliginosum. He
1 See further my remarks in V.T. 23 (1973), 59.
2 On this see Goshen-Gottstein.
3 See e.g. B.D.B., p. 814; nns^DD, used of an idol in 1 Kings xv 13, is there explained as 'thing to be shuddered at'.
4 The word -p^n is not apparently reproduced in the LXX.
5 See his commentary on Isaiah (Migne, P.L., xxiv, pp. 159, 270). now is not,
however, to be connected with *\vi but with HDP ; see B.D.B., p. 1045. The error
probably arises from the phrase ]tfj vin in Jer. xiii 16, 'dark mountains' (R.V.).
22
ISAIAH XXI
translates the phrase as a whole: Babylon dilecta mea posita est

mihi in miraculum.
Ibn Janah shares the view of Aquila and Theodotion and

states that *]t2tt strictly denotes darkness and twilight (l^lm
wHgbs); he records that some give it here the meaning 'night 5
as in Job iii 9 and xxiv 15, while others say that it is the twilight
of day (i.e. dawn) or of the night (i.e. dusk; for that of dawn he
compares Job vii 4). Saadya renders the phrase *DW ino 'the
dawn of my ardent zeal', which suggests that he was one of the
authorities referred to by ibn Janah. That this was also the view
of the Peshitta may be inferred from swpf dsbyny lit. 'the beauty
of my desire' where swpf is probably an internal corruption of
spf 'dawn'. 1 In his comments on p& (verse 5) ibn Janah, however, takes the view that the phrase denotes Belshazzar's night
of rejoicing (Dan. v). This view is adopted by ibn Ezra ('the
night of my pleasure - the night he drank out of the holy
vessels'; cf. Dan. v 3), and by Rashi and Qimhi who refer to
the first and last verses of Dan. v (Rashi),2 and to Dan. v 6
(Qimhi). For Rashi an initial defeat of the Persians is the reason
for his desiring (nptpn) joy and feasting, though for Qimhi the
king was happy because his desire (ipwn) to be made king was
accomplished.3
Modern commentators are more restrained in their interpretation of the phrase. Almost universally *]tw is taken to mean
'twilight' and so 'evening'; yvm *|tPl means 'the twilight that
I have longed for' and reference is seen in the phrase either to
the evening of rest and recreation expected by the prophet after
his harsh experiences of the day 4 or to the evening time when he
delighted to have visionary experiences.5
An exception to the usual view of the word f|tr?2 is that advanced by G. R. Driver.6 For him, the traditional rendering ' the
1 So Payne Smith, col. 4274; cf. G. H. Bernstein in Z.D.M.G. 3 (1849), 393.
2 The verses describe the feast and the death of the kings as happening on the
same night.
3 Such views are quoted by Delitzsch (1) in the name of Umbreit, but they are
dismissed out of hand as 'quite impossible'.
4 Marti, Procksch, Fohrer, Wildberger.
5 Kleinert. Others take the view that the harsh vision took place in the evening
which the prophet had looked forward to; so e.g. Duhm and Kaiser; the latter
freely admits that it is not certain.
6 See Von Ugarit nach Qumran (Eissfeldt Festschrift; . 4 . W. Bei. 77), ed. J. Hempel
and L. Rost (Berlin, 1958).
23
twilight of my desire etc' is 'a strange mixture of phrases', and

consequently he seeks to find a somewhat different meaning for
the word f|tP2 by appeal to the Arabic cognate nsf. Because the
verb nasufa can mean (vmth theme) ' whispered (words)' and
the noun nasif'truce, secret conversation' he posits for the noun
*]W3 the meaning 'faintest suspicion, trace' and for the phrase
ywfi *]tP2 the meaning 'my faintest (i.e. scarcely breathed)
wish', rendering the sentence as a whole 'my faintest wish has
been turned into anxiety for me'. Driver continues by appealing
to Saadya's use of the word 'sihru "anything, of which the
chance of obtaining it is slight and slender"' which, says Driver,
'brings out the sense of the Hebrew word quite well'.
Two criticisms may be directed against Driver's theory. First,
his rendering 'faintest (i.e. scarcely breathed) wish' is speculative in that he gives to rpa a particular meaning akin to that
which he claims for wnV in Isa. xxvi 16 (i.e. lit. 'whisper') 'a
very small amount'. Now this is not certainly the denotation of
wnV in that verse (for another view, see B.D.B.) nor is it clearly
attested for the Arabic root nsf to which Driver appeals.
Secondly, his appeal to Saadya's rendering is ill-founded.
The word nno is clearly Arabic shr ' dawn' x which answers to
the well-attested Hebrew word *]&* ' twilight' (see on ibn Janah
above).
To return for a moment to Saadya's view that rpi denotes
'dawn', it is possible, perhaps, to see in the phrase a reference
to the morning of salvation which is probably alluded to in the
Dumah oracle of Isa. xxi i if (cf. Ps. xlvi 6) and in Isa. xvii 14.
On this view the dawn of peace that the prophet had hoped
for is, in consequence of the dire contents of his vision, changed
to an occasion for horror and trembling. Thus Saadya understands the whole phrase as the cause or occasion of the prophet's
alarm (aiNtfTK ^7 INS *tm "ino iK 'puisque l'aurore de mon
amour est devenue pour moi une dpouvante').
Verse 5, Vto$ n-rag nbx
:]$ inw? onfen vnp nhtf
The verse as a whole is generally taken to depict a third detailed
scene of the prophet's vision because what is set forth clearly
1 And so it is translated by Derenbourg.
24
ISAIAH XXI
differs from that of the opening scene (verses i ff) and from that
which follows (verses 6 ff).
The infinitive absolutes which characterize the first half of the
verse are either taken as imperatives * or as denoting description
in a lively narration. 2
jnVtPn "p57. At first sight no problems are presented by the
phrase, for the meaning 'prepare, set in order the tables5 (i.e.
the dishes upon it - so B.D.B.) is clearly attested elsewhere (see
e.g. Prov. ix 2; Ezek. xxiii 41; Pss. xxiii 5, lxxviii 19) and in this
way the phrase is almost universally translated. Reference is
consequently seen in the words to the feasting Babylonians soon
to be roused from their revels by the writing on the wall or by
the death of Belshazzar (cf. the rabbinic comments on verse 4).
Some scholars3 take the phrase to indicate the nocturnal feasting in the centre of Babylon mentioned by Herodotus and by
Xenophon as having taken place on the eve of its fall to Cyrus
(for references etc., see below).
lYDSn HDE. Considerable difficulties attach to these words (the
second of which is a hapax legomenon) and the problem of their
interpretation turns largely on their being understood closely
with the preceding phrase. The ancient versions4 clearly connect both words with the root HDS ' to look out or about' which
also occurs in verse 6. The reference is then to the setting up of a
watch simultaneously with preparing for a meal (so e.g. Saadya
and Qimhi; the latter states that some Babylonians, fearful of
the Persians, urged that a watch be set).
On the other hand, reference at this point to a watch may be
said to be inconsistent with the ease and relaxation which mark
the beginning of the feast; it is likely that this consideration lies
behind the variety of alternative explanations advanced by the
rabbinic commentators. For ibn Janah rPDSn HDS means sfff
'arrange, set in order the rows' 5 and lsfwf is either
1 G.K. 1130b; so the ancient versions and e.g. Rashi, ibn Ezra, Procksch.
2 G.K. H3ff; so e.g. Saadya, D.K., Marti, Kaiser, Fohrer, Wildberger.
3 So e.g. Rosenmiiller and Delitzsch; others (e.g. Gheyne, P.I.; Marti, and Kaiser)
are cautious about the identification.
4 The LXX excepted; the phrase is not apparently rendered there. Whether or not
it was in the text before the translator is not certainly known. Ziegler (Untersuchungen) takes the view that it was omitted in the LXX because the translator
did not understand it.
5 Ibn Janah clearly uses the verb sff because of its similarity in form to Hebrew
HDS. But he lists the latter word under "n's's and not under "*]'D"S.
25
rows of the food on the table or the ranks of the diners at

the table. 1
Ibn Janah goes on to state that the verb nDS is similar in
meaning to "py 'to set in order' and he compares the juxtaposition of the two words in this verse with that in Ps. v 5.
Ibn Ezra notes that the word rPDS occurs only here and says
that, in his opinion, the phrase means 'make melody' (pa) but
he gives no indication how he achieves this meaning. For Rashi
and Qimhi (as one suggestion) rrDS refers again to an aspect of
Belshazzar's feast, viz. to lighting the lights; and for this Qimhi
refers to Genesis R. 63 (on Gen. xxxv 34) which, referring to
Isa. xxi 5, states, in the name of R. Abba bar Kahana, 'there
are places where they call a lamp KrPDS9.
Modern commentators who do not accept the meaning 'set
the watch' usually accept Hitzig's theory that the phrase means
the spreading of rugs or mats upon which diners reclined.2
Cheyne (I.B.I.) takes a similar view and compares Jewish
Aramaic NriD1'?3 'matting, mat'. With this explanation, as
Delitzsch observes, there is the difficulty that it is without
support in biblical usage and is not attested by any old tradition.
nntP VOK. The meaning of these words is clear, though Duhm
(so also e.g. Marti) raises the important question whether they
should be deleted as a superfluous later addition. There is no
manuscript or versional evidence to support this view, though
it may be noted that the words occur in the following chapter of
Isaiah (xxii 13), and it is possible that they found their way
thence to this chapter as an interpretative gloss on the difficult
phrase rvDXn HDS.
pa intPfc onwn laip. The phrase as a whole is usually taken to
indicate the sudden call to arms directed at the feasting nobles.
pa is understood to denote specifically ' shield (s)' by the
ancient versions other than the Targum, which renders the
word more generally by a n 'weapons', inwa indicates the
preparation (LXX: Tbi|jacFaTe) or taking up of (Vulg.:
arripite)4 the shield, though the Targum furnishes a more
1 Eitan accepts ibn Janah's explanation of the phrase.
2 Hitzig's argument that the root nos here is a form of nos and may be compared
to Arabic sfh is most unlikely.
3 For this word, see Jastrow, p. 1279.
4 For Dathius' view that the Vulgate read "own (with s), see Rosenmiiller.
26
ISAIAH XXI
general meaning with insnsi lp*HD 'polish and make bright

(your weapons) V Rashi considers that intPB denotes oiling the
shield and for this practice he compares 2 Sam. i 21.
Modern commentators generally follow such interpretations
though different views are held of the reason for oiling shields.
For some (e.g. Delitzsch; cf. Kaiser) the purpose was to make
the surface slippery and thereby to facilitate the deflection from
it of missiles and blows. Others (e.g. Duhm, Gray) suppose that
the straps of the shields are oiled so that they should not cut
into the arms of the soldiers.2
A very different view of the phrase is taken by other rabbinic
commentators. Saadya understands the verse to depict a people
who feast while their leaders arise and anoint a king from their
number (onilD iV&^K inotti DnxOTi DKp Ipi). It is not possible
to be certain what Saadya in general thought the words depicted, but it is likely that his views are reflected by ibn Ezra.
He, accepting the view that JJID means king, comments that
Darius was anointed king in the same night that Belshazzar
was killed.3 Apart from the question whether the phrase has to
do with Belshazzar's death and its consequence, Saadya's view
of the meaning of pa mtPD is possible and deserves consideration.4
Ibn Janah presents us with yet a third view of pa ITOO,
though for him too the phrase indicates the circumstances after
Belshazzar's death. When the pleasant evening (^ptpn *]$!)
becomes terror (rmnV) after the death of the king, the tables
are left by the diners who wipe them down, pa, then, for ibn
Janah denotes 'tables' 5 and tnw means ^mshw0 (wipe off) or
1 Qimhi quotes the Targum with approval.
2 Fohrer's view (cf. also Wildberger) that the shields are consecrated by anointing
for use in a holy war is fanciful. There is no clear evidence to support his assertion.
3 Similarly ben Bilam and Qimhi. The former thinks that the words pa imp
are a quotation of the princes' resolution to appoint a successor to Belshazzar.
4 Ibn Ezra and ben Bilam think that the word can denote 'king' because a king
protects his people like a shield. For discussion of pa in a number of texts having
the meaning 'ruler', see J. D. Michaelis, Neue Orientalische und Exegetische
Bibliothek, vol. vn (1790), pp. I78ff, and 205; G. R. Driver in J.T.S. 33 (1932),
44, and 34 (1933), 383^ M. Dahood, The Psalms (Anchor Bible, 3 vols., Garden
City, N.Y., 1966-70), indices; J. Barr in Bibliotheca Ephemeridum Lovaniensium 33
(1974), 45ff.
5 He gives no explanation of this view of pa other than using the synonym ]vk>w
and the translation nPydt.
27
*nz u? (strip bare). In general the reference is to the removal of

food from the table, though ntPfc specifically denotes here the
polishing movement of the hand upon the table.1
Ibn Janah's view may rest upon his giving to ntP a meaning
attested rather for Arabic msh than for Hebrew ntPD. No other
instance of the verb with the denotation ' wipe down' or ' clear
away (food)' is attested in the O.T. A further difficulty is that
ibn Janah gives no etymological grounds for his view that pa
here means 'table'. 2
It was observed above that the interpretation of rrssn HDS is
often influenced by its juxtaposition to jnVtPH "pS7. If the latter
phrase means 'prepare the tables' then it is likely that rrwn HDE
denotes some aspect of the feast. The argument is, however,
entirely reversed by one modern scholar, K. Galling. He takes
the view that rVDSH HSS means 'keep watch' and, as in verse 6,
the words are to be derived from the root nDS (see above).
jnVtPn *\*M is a corruption of an original nV$n *?p3 'men prepare their weapons, and keep watch...' These preparations
are the Persians' response to Yahweh's call to take up arms
against Babylon (verse 2b), the signal itself being given by
the words nntPn 1&1p which follow; nnt2n VDX are understood
as a later explanatory gloss defining the significance of 'the
table'.
Galling's view of the historical background of the oracle is
discussed elsewhere. However, it may be suggested that his
emendation, which involves the alteration of only one letter of
the consonantal text, and the deletion of two words (the authenticity of which may be doubted on other grounds; see above) is
worth consideration. For, first, it enables us to retain the bestattested meanings for the words nD2 and n^D^n and, secondly,
it avoids the necessity of making a number of suppositions as to
the sequence of events (e.g. feasting followed by call to arms).
It may be added that -py is clearly attested elsewhere of preparing weapons for battle (Jer. xlvi 3, msi p&; 1 Chron. xii 9,
mm ms).
1 gry Hyd Hyh - the same movement that is employed in the anointing of
persons.
2 Cf. ben Bilam's similar criticisms of ibn Janah.
3
h
b
28
ISAIAH XXI
Verse 6, T?5?n rf? s m ^K "iK HD

: T r HKT -wto
i
v :
v -:
The verse seems to depict a third and different scene. No difficulties are presented with regard to the meaning of the words,
though differing views are set forth concerning their exegesis.
^K is usually taken to indicate the prophet himself. For.
Qimhi the prophet is, as it were, in Babylon, and Yahweh
speaks to him as a representative of the princes of Babylon,
who, concerned about the advance of the Medes and Persians,
set a watch.1 That Qimhi reflects Saadya's view is probable; the
latter translates ^X by vSo 'concerning us' and Tarn by iDpiK
mb 'set for yourselves (a watchman)'. The pronouns in Saadya's
translation are defined by his interpretation of the beginning
of verse 5 ('It is a people whose tables are prepared etc.').
n&sa suggests 'watchman, lookout' to all the ancient versions
and to the rabbinic commentators. For Rashi the reference is
specifically to the prophet Habakkuk and he compares Hab. ii
1 TllhVX ^ 8 $ ? V? I will stand upon my watch' (R.V.). Thus,
Isaiah is promised by God that Habakkuk will prophesy in the
future the fall of Babylon. For Rashi the identification is made
certain by the reference in verse 8 to rtHK which, by glmatriyd
(i.e. the numerical value of the letters), is equivalent to Habakkuk. While it is impossible to accept this far-fetched and strained
exegesis of HDSfc, Rashi's comparison with Hab. ii 1 (which also
contains the verb HDSN) is valuable as indicating that another
prophet uses the imagery of the watchman and of looking out.2
On the basis of such comparative evidence modern scholars
(following Ewald, Duhm 3 and Marti) have seen in the watchman a reference rather to the prophet's alter ego which is the
subject of visionary experiences than to an actual separate
personality appointed to make the observations which the
prophet then interprets. It is possible that ibn Ezra held a
similar view (though of course without a specifically psychological account), for he notes that the prophet did not appoint
1 Obermann adopts this view enthusiastically.
2 But cf. Wildberger's (implied) caution against too-facile attempts to harmonize
Isa. xxi with Habakkuk, whether it is done by textual emendation or by resort
to psychological theories.
3 His account from the psychological point of view is particularly impressive.
29
1
a watchman in reality but only rWDl "p"n 'in his vision'. Buhl
and B. Stade 2 find this sort of explanation quite unacceptable
as being 'unnatural and quite without analogy' (Buhl). Consequently, Buhl, referring to the LXX's reading 6 ocv i'Sris (i.e.
second masculine singular) for the M.T.'s nXT (third masculine
singular), proposes the emendation: n$?fi *)#K n?? rnS7 te1?
T?n ' Come take your stand, O watchman; relate what you see.' 3
Such alteration of the text, however, is unjustified. The LXX's
rendering is likely to be merely interpretative and does not
suggest a different text (cf. Goshen-Gottstein). Further, that
the prophet should speak figuratively4 of his function is, as has
been shown, quite reasonable and, in view of the evidence of
Hab. ii i (cf. also Isa. xxi n ) , likely.5
The remaining problem of the verse concerns the precise
force of the opening o. For the majority (e.g. D.K., Duhm) the
particle denotes the introduction to the details of the harsh
vision of verse 2, i.e. the news of Babylon's fall. For others (e.g.
Marti) it introduces the reason for the prophet's terror. Galling,
who emends the text of the previous verse (see above), considers
that the particle affords a connexion between verses 5 and 6
which both concern the opening of the war by the Persians.
The force given to "O, however, depends entirely upon the view
taken of the shape of the prophecy as a whole and it is not possible here to resolve the question finally.6
Verse 7, itoq M ? o"t?hs> iM 333 n*ni
If waw consecutive is prefixed to third person singular masculine verbs, then the first half of the verse may be a subordinate
1 Cf. Wildberger (who follows J. Hanel); he repudiates the alter ego theory,
preferring to think of the prophet as appointing an imaginary watchman in his
trances and visions.
2 See Z-A.W. 8 (1888), 1576, 1656.
3 Stade suggests the modification njân lbs? ns1? 'Go, set yourself on the watch',
etc.
4 At least figuratively; I do not regard myself as competent to judge whether a
psychological account is justified on the evidence. Cf. Kaiser's comments and
those of Wildberger.
5 Another objection to the proposal to emend is that mentioned by Wildberger;
emendation of this verse requires consequential emendation of verse 8.
6 Cf. Wildberger's observations. For him the usual ' for' is out of place as there is
no logical connexion between the verse and what precedes it, and to claim here
an asseverative force for it is mere expediency.
30
ISAIAH XXI
temporal clause. Thus, ibn Ezra renders ywp*. . .run DN 'When

he sees. . . then shall he pay attention.' 1 On this view the phrase
constitutes a continuation of the divine command of verse 6.
Saadya, by contrast, takes the perfects to be prefixed by simple
waw and consequently translates the verse as an account of the
(Babylonians') fulfilment of the order 2 (. . .'am HNlDpXD 'So
we set (the watchman) in his place and he saw. . .').
The LXX's KOU eTSov (i.e. first person) may be regarded as
an accommodation to its (free) translation of the previous verse
rather than as implying that it read 0*^1. Buhl, however, takes
the view that an original rp*ni was read rPNTt by the LXX. He
repoints ywpTV\ as an imperative and takes the verse to mean
'When you see...then pay attention.' The interpretation
differs little in substance from that of ibn Ezra (see above) and,
as the latter is entirely appropriate, Buhl's emendations may be
judged superfluous.
Considerable uncertainty attaches to the precise meaning of
the terms used of what is to be (or was) seen. In particular the
words 201 and t2HD have been variously understood.
(1) 1D1 (first occurrence); the LXX renders &va(3oTocs
'riders' while Symmachus and Jerome render 'chariot' (apucc
currum). Targum's *]m is capable of either meaning 3 as is the
Peshitta's rkwb\* Saadya renders NVTD 'horses' but ibn Barun
compares for this verse Arabic rkb 5 'a party of riders above ten
(on camels)'. Ibn Barun's view is adopted by a number of
modern commentators (e.g. D.K., Marti, Gray, Kissane), while
others insist that the word is better attested in the bible with the
meaning ' chariot(ry)' (e.g. Stade, Procksch, Galling, S.
Mowinckel, Fohrer, Wildberger). Ibn Janah appears to agree
with this view and states that 201 can properly denote chariotry
collectively and he illustrates this view by quoting 2 Kings vii
14, D*010 M l TO inpn 'they took therefore two chariots with
horses' (so R.V.), and this verse.
(2) BhD. Symmachus (ITTTTGOV) and Jerome (equitum) take the
1 See G.K. 164b (4); so e.g. Marti, Procksch.
2 Similarly, Stade thinks the verse relates what the prophet saw and is not part
of the command; so also Delitzsch.
3 Cf. Stenning: * cavalcade (or chariot)'.
4 Cf. Brockelmann and Payne Smith.
5 Wechter, p. 119. For rkb meaning riders, horses, cavalcade, caravan, troop, etc.,
see e.g. Wehr.
31
word to denote horses and the three words D^tZHD T&E 231
together to indicate two-horse chariot(s).1 The LXX, Targum
and the Peshitta on the other hand take D^ttHD to mean pairs of
* riders' (ITTTTEIS; pBHD; prsyn);2 the Targum further indicates
for the pair of' riders' that one rode upon an ass, the other upon
a camel.3 iQIs a may indicate the same tradition, for the reading
2DT1 is there found for the second 4 and third occurrences of the
word 2D1 in the verse. Rashi appears to follow the Targum and
he suggests that the two riders on their respective mounts constitute a symbol of the Medes and Persians.5
Saadya renders D^ttHD by JKO^DVN 'horsemen' and takes Tfcs
BWito to denote pKHD^K p *CN1T*O 'some pairs of horsemen'
amongst other riders who were mounted on asses and camels.6
This view is adopted by a number of modern scholars who see
in the phrase 'couples of horsemen' i.e. horsemen riding in a
column in pairs (e.g. Hitzig, D.K., Marti, Gray). 7
This second interpretation of oÊDD has been vigorously
challenged by Galling (.Th.K. 53 (1956), i29ff) and Mowinckel ( F T . 12 (1962), 278ff) whose conclusions are virtually
identical. Galling argues from the evidence of usage that DID
used together denotes horses and war chariots and that
are specifically war horses yoked to chariots. He states
that there was no comprehensive term to denote chariot riders,
who, in any case, comprised archers etc. In Isa. xxi 9 the
unique expression tPK 201 must be explained by reference to the
contrast it affords to n&n 3D").9 *]n M*i indicates the two1 So e.g. Galling, Mowinckel and Wildberger; for their exposition of the words
331 and one, see below. Stade's view that a (dispatch) rider is mounted on a pair
of chariot horses, unhitched for the purpose of speedily bringing the news of
Babylon's fall, is absurd. As Duhm observes, one horse would be unhitched in
such circumstances.
2 In Aramaic and Syriac the word does not carry the meaning 'horse'.
3 So perhaps the Peshitta: rkwb0 dtryn prlyn drkyb hmr0 wdrkyb gml.
4 The waw is here added secundus manus; so Goshen-Gottstein. The reading was
earlier proposed conjecturally by Buhl and Stade.
5 So amongst moderns Procksch, who sees in them representatives of Elam and
Media.
6 Does 3*0-1 indicate that he read 331 (second occurrence) as 351?
7 Hitzig and Marti observe that even if D^BHD here denotes * horses' then the riders
are included in that definition; but their observation depends on their taking 331
(first occurrence) not as chariots but as a cavalcade.
8 tmnti is likely to be a North Syrian-Aramaic technical term (? denoting a
particular breed of horse) adopted by Hebrew.
9 So, mutatis mutandis, *?nj 3D1.
32
ISAIAH XXI
wheeled baggage wagon pulled by asses, by the side of which the

driver walked. The terms 201 (verse 7) and WN 201 (verse 9)
are both further defined by the term owito TftX which means
(chariot with a) 'pair of war horses'. The phrase D^ttHD TE 201
means two-horse chariots, and for this 2 Kings ix 25 is quoted as
evidence.
BPK M l in verse 9 denotes manned chariots as opposed to
wagons. Thus, as Galling understands the terms, the beasts
indicate the nature of the vehicle which they pull (cf. English
'horse-carriage', 'motor-carriage'). Galling's account of these
terms, while it is soundly based on the evidence of usage in the
O.T., is not wholly convincing in regard to two phrases in Isa.
xxi. First, BPN Ml (verse 9) as a manned chariot; by analogy
with his account of man 2D1 etc., rK ID") ought strictly to mean
a cart pulled by men and not a manned chariot. Secondly, it is
doubtful whether the word Ml was ever used to denote baggage
wagons as opposed to chariots, and certainly, as Wildberger
observes (p. 782), there is no evidence in the O.T. that asses or
camels pulled such wagons. That they should have pulled
chariots is a fortiori most unlikely.
According to Galling, the correct denotation for a^&ns was
obscured because cavalry, and horses bred specifically for the
purpose, later became the norm in the Orient. When the twowheel war chariot, still prevalent in the seventh century B.C.,
began to fade into the background, eventually to vanish, the
distinction between paras ( = chariot horse) and parras (with
geminated resh 1 ), to which the Massoretic O^ttHB bears witness,
was coined {parras with the meaning 'trooper').
The secondary meaning, 'cavalry', probably belongs to the
time of cavalry par excellence, i.e. to the time of the Persians.
If, as seems probable, Galling's views2 are substantially
correct the use of the terms in this verse may be important as
evidence for the dating of the prophecy.
1 On this further Gesenius-Buhl. Galling compares the form asn.
2 MowinckePs views are virtually identical. His treatment of Isa. xxi 7 and 9,
however, lacks precision; thus e.g. he adopts L. Koehler's 'train' for 331 and
thereby seems to imply that the asses and camels themselves bore the baggage,
all of which contradicts his thesis that 33*1 means ' chariotry'; further, he quite
fails to give an adequate account of BTN 33*1. For these reasons I have preferred
to follow Galling's account. For an older argument to the effect that avnv
means 'horses' and never 'riders', see W. R. Arnold in J.B.L. 24 (1905), 45ff.
For further discussion of the etymology of the word BHD, see D. R. Ap-Thomas.
33
TtfpTV). For the emphatic expression 'he shall

pay very great attention', cf. with Marti Isa. xxviii 21 and
lxiii 7.
Verse 8, ^
**?& ^
. The Vulgate, Aquila, Symmachus, Theodotion and

the Targum appear to confirm the M.T.'s mK 'lion'. The lion
is taken to be the subject of the verb by Saadya and Rashi, the
latter explaining that by gimatriyd the prophet Habakkuk is
meant. 1
Theodotion's 'ApifjA may indicate a tradition that the word is
to be explained by reference to Isa. xv 9 where the remnant of
the Moabites are destined for the lion. Here the LXX renders
mK by 'ApifjA (cf. LXX, Isa. xxix 1). On this latter verse ibn
Ezra states that rVHK is a reference to the king of Assyria.2 In a
somewhat similar way, Qimhi takes the view that rVHK in Isa.
xxi 8 contains a reference to the Medes and Persians.
Ibn Ezra and ben Bilam interpret JTHK as if a kaph were prefixed to it and for the construction they compare (respectively)
Deut. iv 24 and Prov. xi 22, 'And he cried like a lion.' This view
of the phrase is adopted by a number of modern scholars,3
amongst them Rosenmuller who compares Rev. x 3.
iQIs a records the variant reading n*nn 'the seer' for mN
and this is widely adopted as the most likely solution as it
furnishes a subject for the verb Nip**!.4 It is interesting to record
that it was proposed as an emendation before the discovery of
iQIs a by R. Lowth in 1778, and adopted as such by Procksch.5
The remainder of the words of this verse are tolerably clear,
and the divergences in the versions etc. are minor. It is worth
noting, however, that the Peshitta interprets the speaker as God
who addresses (presumably) the prophet and informs him that
he (God) is continually standing upon his watchtower day and
night.
1 Cf. Ibn Ezra, who records the tradition as from the Midrash.
2 Cf. Qimhi, notwithstanding the fact that he and Rashi think primarily of
Nebucha dnezzar.
3 E.g. Delitzsch, D.K.
4 So e.g. B.H.S. and N.E.B., Fohrer, Wildberger. The verb n*o occurs in verses 6f.
5 Though he attributes it to Sievers. Wildberger wrongly mentions Lohmann as
the originator of the suggestion.
34
ISAIAH XXI
The most natural interpretation, however, is that which

assumes that the prophet (or his alter ego) is speaking and that
he is expressing his impatient suspense as he waits for fulfilment
of the promised vision. Marti argues that the words between
H*nK anfri (thus he emends) and mn (verse 9) are an interpolation inspired by Isa. lxii 6 where the watchmen continually
look out for the dawn at the end of the time of judgement. He
regards the words as impairing the structure and parallelism of
this part of the oracle and as contradicting the meaning of verse
4 which implies that the prophet received his dreadful vision in
the same evening in which he received his order (verse 7).
On the other hand, it is possible to argue that Isa. lxii 6 is
dependent upon this prophecy for some of its phrases and it is
likely that the words here convey the protestation of the prophet
(or his watchman) that he has faithfully obeyed the command
to pay very great attention.
A number of the words and expressions of the verse are found
in Hab. ii 1, as has been noted above. For discussion of this, see
below, pp. I23ff.
Verse 9a,
The verse may be regarded as the climax of the prophecy and
the words 'fallen is Babylon5 as the burden of its contents.
BPK 251 is taken to refer to a single rider by the LXX, Vulgate
and Peshitta. For the Targum there comes a cavalcade of men
(ttttK *]T\i) and with them a pair of horsemen (pttHD 3M fT&sn).
This interpretation is adopted by ibn Ezra who states that 0? is
to be supplied before D^ttHD TfcS, though he says nothing of the
sort in regard to verse 7.
Saadya, like the LXX etc., thinks that the phrase denotes
the advent of a lone horseman behind whom are a pair of
riders. The former addresses the prophet with the words
'Babylon is fallen.' For Qimhi, on the other hand, the phrase
denotes the arrival of the Persian army in Babylon.
The LXX's ava(3arr|s ovvcopiSos suggests to Procksch that it
translated D^ttHD *T&2 M l (as in verse 7) and that t2PN was not
present in the text that it translated. He concludes that ( M I )
ETK is an explanatory gloss which attempts to define twit as
35
'men' rather than 'horses'. Procksch's view is attractive on the

grounds that the word t2PK does not qualify MH in verse 7 and
because no satisfactory account of the meaning of &rK M l has
been given. It should be noted that the phrase ontnfi) DTK M l (a)
occurs in xxii 6. It is possible that the text of xxii 6, if indeed it is
authentic, 1 has influenced the text of the verse before us just as
xxi 5 may have been influenced by xxii 13 (see on verse 5 above).
For the first part of the verse modern commentators direct
themselves to the question why the vision, by comparison with
verse 7, contains an apparently abbreviated description of what
comes. For some, what the prophet sees in verse 9 is not the
same as what is described in verse 7. For example, Delitzsch
thinks of a smaller cavalcade whose function is to carry the
news of Babylon's fall, whereas verse 7 refers to the whole
Persian army approaching Babylon. Other commentators,
however (e.g. D.K.), think that the reference in both verses is
the same and that the abbreviation of the account of a previously described and expected vision is entirely natural (cf.
Wildberger).
On the view that verse 7 contains a sentence with protasis
and apodosis (see above), the use in verse 9 of the demonstrative
H? may indicate what in the event the prophet in his vision
actually saw, i.e. D*tnD "rS M*i as opposed to i^n M l etc.
On this supposition, it is further possible that verse 7 lists a
number of alternative objects of sight for which the prophet is
to wait, i.e. either twnt TBB 221 or *nn M l or V&l M*1. On the
other hand nt is probably used here merely as an enclitic (cf.
B.D.B., p. 261 4g), and in this case it has no demonstrative
force.
ifciOl ]SP1. The subject of the verbs is not expressed (so the
ancient versions). For ibn Ezra it is all the persons of the
cavalcade mentioned in the earlier part of the verse, and mutatis
mutandis for Saadya it is the lone horseman.2 Qimhi, on the
other hand, regards the watchman (i.e. the prophet) as the
subject of the verb (so e.g. D.K., Duhm, Marti), and Kaiser,
comparing e.g. Amos vii 8, Jer. i 12, Zech. i i4f,3 thinks that
1 Marti, following Gratz, believes that the phrase D*BnD (sic) i&s 3D"> in xxii 6 is a
gloss derived from xxi 7.
2 So, amongst moderns, e.g. Delitzsch, Buhl, Stade, Procksch.
3 For the full list, see his commentary.
36
ISAIAH XXI
God is the subject. Wildberger, too, argues that Yahweh (or,

as in Zech. iv iff, his angel) is the subject of the verbs; he reveals
to the prophet the meaning of the vision (cf. Zech. ii 4). For
Wildberger the use of first person suffixes (Ttrp/WTB) in
the following verse tells in favour of this explanation. On
balance, however, it seems most natural to see the watchman as
the subject of the verbs.1
Van n*?D2. The repetition of the all-important cry is, as Qimhi
observes, emphatic. The Targum interprets the repetition:
'Babylon has fallen and is also about to fall (xTns; *]K1 n*?D3
V^fc1?)', and this interpretation is further explained by Rashi
who states that the first fall is that at the hands of the Persians
and that the second is at God's hand and to this Isa. xiii 19
alludes ('And Babylon... shall be as when God overthrew
Sodom and Gomorrah' - R.V.). While this interpretation is
clearly contrived and fanciful, it represents two possible interpretations of the verb nVsi; it is either an historic perfect (with
the messenger (s) as the speaker) or a prophetic perfect (with
God or the prophet as the speaker).2
Verse 9 b ,
The verb izw is best regarded as having an impersonal third
masculine singular subject, for which the Vulgate, Targum and
Peshitta (so Saadya) make use of the passive voice. Some
modern commentators, however, take God as the subject of "Dtp
(so e.g. Rosenmiiller, Delitzsch and Procksch; the last, however,
prefers to emend to Viatf on the evidence of the LXX). 3
Amongst the versions, the LXX alone appears to diverge
widely from the M.T.: ml TICO/TOC TCC dydAjjiaTa aurfis mi TOC
Xeipo7roiT|Ta aurfjs for JTTIVK ^DD VD. The rendering may be
a doublet of an earlier text which read JT'V'OS (i.e. of Babylon)
or ^fl^S only (cf. Gray).
1 Wildberger's argument that it would be strange if x-ipî in verse 8 were expanded
by "Win ]V^ in this verse is not convincing. The verb niy is used not only to
denote 'answering, replying', but also 'speaking in view of circumstances' and
'testifying'; see B.D.B.
2 The LXX does not repeat the verb and on this evidence Procksch proposes to
delete the second n^DJ. Such emendation is hardly justified; cf. Wildberger.
3 M. Dahood's suggestion (in Biblica 40 (1959), 165) that pN here denotes the
underworld may be rejected; he adduces no arguments in its favour other than
to say that it is 'probable'. Dahood follows Cross and Freedman (in J.N.E.S. 14
(1955), 247ff), but they say even less than Dahood in support of the theory.
37
For the rest, the closing phrase of the verse is of interest as

possible evidence for the dating of the prophecy. For, as is
widely observed among modern commentators, Cyrus did not
shatter idols, but, on the contrary, was sympathetic to the gods
of the peoples whom he vanquished. Further, Babylon fell into
his hands without fighting. Those who take the view that the
oracle concerns the fall of Babylon in 539 B.C. emphasize that
we have here no vaticinium ex eventu and that, in that he foretold
what did not happen, the prophet was mistaken or merely
represented a motif which belongs to the prophetic Gattung - the
fall of a world power (so e.g. D.K., Duhm, Gray and Wildberger). For other scholars, who on various grounds regard the
prophecy as indicating a fall of Babylon in Assyrian times, such
considerations constitute further evidence pointing to that
conclusion (so e.g. Cheyne, P./., p. 126, and Erlandsson, p. 91).
Verse 10, njn? naa

The verse appears to contain a reflection of the prophet on the
import of his message, and the first person suffix attached to
*TW7to 'my threshed one' and Ttt p refers to the prophet or to
God (so Qimhi) in whose name the prophet speaks. Ibn Ezra
mentions both interpretations as possibilities, adding a third:
that it refers to the tTN 3D1.1
For Saadya Y>tzn and T U p are references to Babylon
which is so described in its overthrow ('OKTBD mN2> 'She
[Babylon] has become like threshed corn etc.'). Similarly, ibn
Ezra and Qimhi regard Babylon as the object of the threshing,
though the former mentions as a possibility that TiJi p contains a reference to Israel as the pure corn kept in the threshing
floor.
The LXX clearly interprets the terms of those who are
abandoned to suffering (oi KonraAeAeimJiEvoi KOCI oi oSuvco^voi),
and it is likely that the choice of words enables the translator
to think of the diaspora of his own time.2 The Peshitta understands verse 10 as a part of verse 9: 'The graven gods are
1 It is not clear (to me) how ibn Ezra takes this latter interpretation.
2 See Seeligmann, pp. 109, 113.
38
ISAIAH XXI
shattered'; it continues brco mn hsd wmn bly dr (?) 'in the land
for lack of harvest and threshing floor'.
The Targum expands the verse so as to present a simile:
' Kings who are skilful in waging war shall come against her to
plunder her, even as the husbandman who is skilful at threshing
grain', 1 by which presumably it seeks to portray Babylon as
attacked (cf. rabbinic commentators above).
Rosenmuller and a number of modern scholars refer to Jer.
li 33 where very similar phraseology is used of Babylon: ' The
daughter of Babylon is like a threshing floor at the time when it
is trodden' (R.V.). Delitzsch, however, urges caution in regard
to the natural assumption that the comparison suggests Babylon
as the object of the threshing in the present verse. For, as he
says, 'Jeremiah has given a different turn to Isaiah's figure.'
Rather Israel is alluded to (so e.g. Delitzsch, Procksch) as long
oppressed and ill-treated (so e.g. Cheyne, P./., and Marti but
with different oppressors in mind).
That the term Tft p lit. ' son of my threshing-floor'2 constitutes a synonym for Tittna seems probable, and for this
figurative use of the term p , the phrases Hf? ^2 'sparks' (Job
v 7) and rMftT'T? 'arrow' (Job xli 20) may be compared (cf.
B.D.B., p. 121a 6). The phrase then denotes 'corn of my
threshing-floor', ^rittfttt is to be regarded with Procksch as an
abstractum pro concreto and consequently its juxtaposition in the
singular to DD1? in the plural is intelligible and ODV need not be
deleted as a secondary amplification of a shorter original (so
Marti). The two synonyms then denote ' my threshed corn' and
constitute a metaphor of a people subjected to oppression.3
The words V N W TIVK mrp nK& are taken by some modern
commentators to be an editorial addition which has the function of making clear the source of the revelation of verses 9b and
10. There is something to be said for this view, for, as Wildberger observes, the parallelism of the verse is marred by its
presence.
The sympathetic tone of the verse and of its phraseology most
naturally indicates a reference to Israel or to the prophet's
1 For the Aramaic text, see Stenning.
2 Wildberger's insistence that the phrase means 'mein Tennensohn' rather than
* Sohn rneiner Tenne' is otiose. No commentator, so far as I am aware, has wished
to claim the opposite.
3 For which metaphor, cf. Mic. iv i2f and Isa. xli 15.
39
contemporaries rather than to Babylon (cf. Rosenmuller, D.K.,

Duhm, Marti, Procksch, Gray and Wildberger). Whether Israel
was regarded as threshed by reason of the Babylonian exile or
because she was caught up in the turmoils of the Assyrian
domination of the east (so e.g. Cheyne, P./., and Erlandsson)
depends on the view taken of the prophecy as a whole.
Verses n & 12
Verse 11 a, Hn Kfrfc
As in Isa. xxi 1 and 13, and in xxii 1, the two words constitute
the title of the oracle contained in verses 1 if. All witnesses to the
M.T. substantiate the reading except possibly the LXX which
reads TO opociicc TT\S ISOUIJIOCIOCS.1 As has been shown (see on
verse 1 above), the titles of the other oracles cited are derived
from a word in the text. In this case, however, no word approximating in form to ran is discernible in the text. The only
word in the oracle with geographical import is "T1???? 'from
Seir' and as Seir denotes the land of Edom in Gen. xxxii 14
and Judg. v 4, it is reasonable to suppose that ran denotes that
same country. With such considerations in mind, and on the
evidence of the LXX, some modern scholars have suggested
that ran is a corruption of an original (n)fcnK.2 On the other
hand, the LXX's rendering may equally well be regarded as an
interpretation of ran. Thus, in P.T. Tcfanith 1:1 (p. 5) it is
recorded in the name of R. Hanina ben R. Abahu that in R.
Meir's book they found written ' the burden of Dumah' - ' the
burden of Rome', w i HtPfc ran KtPfc.3 It is possible that R. Meir
wrote Dnx as a marginal annotation, though it is more likely
that he wrote w on the basis of the simple resemblance between *i and i. 4 In any case, the tradition affords evidence of
exegetical interpretation of ran as D"TN (or Rome) at a very
early date.
Ibn Janah thinks ran denotes 'the evil kingdom of Edom'.
Under the root 0"?lT ibn Janah compares the title of Isa. xxi 11
1
2
3
4
The word is simply transliterated by the Targum, Peshitta, Vulgate and Saadya.
So e.g. R. Lowth, Marti and, supposing 'ambivalence' (on Jwa*), I. O. Lehman.
The equation Edom = Rome was made by R. Aqiba; see Bereshith Rabba 65,21.
See J. P. Seigen, The Severus Scroll and iQIs*, (Missoula, Mont., 1975), pp. 47f.
40
ISAIAH XXI
with Ezek. xxvii 32, &<! tfria TOIS lis? 'Who is there like
Tyre, like her that is brought to silence in the midst of the sea ?'
(R.V.), and Ps. xciv 17, *$*! T\mi T\l?ti 'My soul had soon
dwelt in silence' (R.V.). n&V7 then denotes 'destruction' (hlkt)
and the title signifies 'the people doomed to destruction'
By contrast, ibn Ezra, followed by Qimhi, takes the view that
denotes an Arabian people and he refers to Gen. xxv 14
where HDH is listed as an offspring of Ishmael. Ibn Janah also
notes this view and states that it may be intended by the
Targum in that it leaves nan untranslated. He is, however, dissatisfied with the suggestion because it does not accord with the
reference to Seir (Edom) which so closely follows. Ibn Ezra
attempts to resolve the problem by suggesting that the whole
prophecy depicts a spy sent from Seir to search out Dumah. His
suggestion is, however, unconvincing because it is fanciful and
the details of it so involved that they defy comprehension.
Amongst modern commentators some follow ibn Janah's
view that n&n is a figure of speech denoting Edom (so e.g.
Rosenmuller, D.K., Duhm, Gray and Procksch), while others
follow ibn Ezra in seeing in the word a reference to a people or
city of that name (so e.g. Michaelis, Rabin, Wildberger). In
connexion with this view two places with the name Dumah are
attested, of which one only has been advocated as the Dumah of
Isaiah xxi. First, there was a city of this name in the mountains
of Judah mentioned in Jos. xv 52 and identified with modern
Daume, south-west of Hebron. 2 It is unlikely, however, that a
prophecy concerning a small Judaean town would find its place
in the oracles of Isaiah (cf. D.K. and Wildberger). Secondly,
there was the important city of North Africa named Dumat-alJandal identified with the modern oasis el-J6f. J. D. Michaelis
was the first to argue that Isa. xxi 1 if referred to this city, which
is usually identified also with the offspring of Ishmael of Gen.
1 Similarly ben Bilam and Rashi. Ibn Janah thinks that Jerome may have left the
word untranslated because he understood its 'secret' (i.e. that it signified 'destruction') and could not reproduce it in Latin.
2 See C. W. M. van der Velde, Memoir (Gotha, 1858), p. 308; cf. F.-M. Abel,
Geographie de Palestine, 11 (3rd edn, Paris, 1967), pp. 3o8f. It is likely that
Jerome's mention (commentary - see Migne) of a Dumah in Idumaea represents
a confusion with this (Judaean) Dumah; at any rate, he is alone in mentioning
such a place.
41
xxv 14. The city is mentioned in Assyrian texts as Adummujutu2

and is not to be confused with Udumu (= Edom) which is also
mentioned in such texts.
Verse n b , Wwp-ng ">?# nV^Va-na igtf Tvfra *np ^ K
*np ''Vx. The words are rendered by the LXX: 3 Trpos [xe
KocAei Ttapd TOU Irjip 'One is calling to me from Seir' with
which the Vulgate agrees. The Targum is similar except that
fcCfttf p 'from heaven' appears as an interpretation of the
M.T.'s T M (fcTfctP p ^DK ^V - 'Unto me did one cry from
heaven'). It is possible that *?*< (interpreted secondarily as **?*?
'my God') may have influenced the Targum, for in P.T.
Tcfanith i: i (p. 5) 4 R. Simon ben Yohai is said to have understood T5W& KTip *h* as 'My God calls to me from Seir' on the
grounds that God was always present with Israel wherever they
were exiled.
Rashi evidently follows this tradition with its double interpretation of 'Vx - ' The Holy One has said to me: the prophet or
angel is calling to me because of the yoke of the kingdom of
Edom.' It is possible too that it has influenced Qimhi who
regards the voice of prophecy as the subject of the verb.
The Peshitta takes the watchman as the subject of the verb
*np and the later words V>VD n& nV'Vfc n as an indication of the
time at which the watchman was on duty: ly qf mn scyr ntwf
dbllf ' to me is calling from Seir the watchman that is (on duty)
in the night'. The version is likely to represent a free and
exegetical handling of the M.T. rather than as indicating a
different text (cf. Goshen-Gottstein).
A different view is adopted by Saadya who translates ''Vx'by
vfiy 'concerning us', taking (the people of) Dumah to be the
subject of Kip 'Men ask about us from Mount Seir.'
The p prefixed to TW is understood as p of origin by all
the ancient versions and rabbinic commentators except ibn
1 See B.D.B. For a more recent statement of this view, see Rabin.
2 For the identification, see M. Weippert in W.O. 7 (1973), 441124. Adumma/utu
was taken by Sennacherib in c. 691-689 B.C.
3 For the minor Greek versions, see below. There is some MS. evidence to suggest
that the LXX (like the minor Greek versions) takes Nip as an imperative (e.g.
KaAeiTe in A). Ottley, however, argues for the superiority of KOcAe? (indicative) the reading of N, B and Q.
4 Cf. Schwab, p. 144. Jerome in his commentary mentions the possibility.
42
ISAIAH XXI
Janah and Rashi. TSWfc denotes the place of origin of the call,
though for the minor Greek versions it denotes irpos s[xk K&Aei
TOUS 96uyovTccs Trapa TOU S/qeip ' call to me those fleeing from
Seir'. 1
Galling suggests that this reading presupposes that ^ 7 7 ^
(or collectively "H^n) originally stood in the Hebrew text. Such
a view is, however, extremely unlikely; rather the rendering of
the minor Greek versions bears all the marks of an attempt to
clarify and elucidate the rendering of the LXX with which
substantially it is verbally identical (NB especially the use in
both of the preposition Trapa).
Ibn Janah regards p here as having the same force as the
preposition V 2 and as meaning ' I heard one calling concerning
(cri) Edom.' Rashi takes the p as causative and renders 'The
prophet is calling to me [sc. God] because of the yoke of Edom
(DTK rvo^n Viya).'
Modern translators and commentators are, however, virtually unanimous in giving to p its usual sense of origin. The *VN
is taken to refer naturally to the prophet who hears in imagination the Seirites calling to him. As the words denote a visionary
or contrived image no difficulty need be seen in a Judaean
prophet hearing a voice from Edom or in a Judaean prophet
being consulted by foreigners3 (even by Edomites!).
n&tP. The massoretic pointing of the word as a participle with
the meaning ' watchman' is reflected in the ancient versions and
rabbinic commentators other than the LXX. 4 The latter
renders <puAd(j(7STe eiraAêis 'guard the battlements' apparently understanding n&tP as an imperative. The version continues
<puAaacrco TO -nrpcbi ml TT]V vuicra which differs widely from the
M.T. 5 as does the Targum with K...KnKm JV pnV ttme
W&V Tnsn 'O prophet, interpret unto them the prophecy,
what is to come.'
1 See Field, p. 645. Rabin adopts a similar view of TWO: TwraixntP "0 'he who
comes from Seir' (i.e. to Dumah).
2 See Neubauer, pp. I55f; ibn Janah cites Ex. xiv 3 and Ps. xxxi 12 as parallel
instances.
3 So Delitzsch, D.K., Fohrer, Kaiser and Wildberger. Marti and Gray apparently
think of an envoy of Edomite individuals travelling to Judah.
4 The Targum iraa 'O prophet' is interpretative.
5 It is possible that they read ipnn nKinff Ow). See Goshen-Gottstein. For a
discussion of the absence in the LXX of words corresponding to the last three
words of the M.T. of this verse, see Ziegler, Untersuchungen, p. 48.
43
x
The view of the Targum that the watchman is a figure for

the prophet is generally assumed, though for Rashi the term
denotes the guardian of Israel (i.e. God). The term w is
usually compared with HDS& in verse 6, where the watchman
denotes the prophet or the prophet's alter ego,2 To this view
P. Lohmann objects on the grounds that the word *19# (as
opposed to nD2&/nD2) is not an O.T. designation of a prophet; 3
he concludes that the word means 'watchman' simply, and
that the poem as a whole is a modification of an old (secular)
watchman's song. Lohmann's view of the oracle as a whole
must be considered later; here it is sufficient to note that his
statement that n&tP is not an O.T. designation for a prophet does
not amount to grounds for denying that the specific denotation
'watchman' may be used as a figure for 'prophet'. For, as he
himself admits, nDEfc is widely so used. It should be noted also
that the noun mDlPtt is used in verse 8 in close connexion with
the n&S (cf. Wildberger).
V^fc n& nV^D na. The words are taken to be an urgent
question4 addressed to the watchman. The p prefixed to nV?
(and V'V) is thought to mean 'concerning' by the Vulgate
{quid de node) and by ibn Ezra, Rashi and Qimhi. For ibn Ezra
the night is a particular night of the week and he refers to the
night in which Moab was destroyed according to Isa. xv i. For
Rashi the night signifies the darkness and oppression suffered by
Israel by reason of the yoke of Edom.
Saadya takes p as partitive, and renders: DD1 W?N p 'to DD
Hlfc yi' How much of the night has elapsed and how much of it
is still to come ?' This latter interpretation of p is adopted by
virtually all modern commentators, though it should be noted
here that it is rejected by ibn Ezra.5
The word VV in the second half of the repeated question
1 For 173^ with the meaning 'city watchman', see Song of Songs v 7.
2 So most modern commentators; see e.g. Marti and Procksch.
3 He discounts the late Isa. lxii 6 where onnff denotes superna ural (angelic)
beings.
4 Hence, according to e.g. Qimhi, Delitzsch, D.K., Wildberger, the repetition.
5 Rabin's view of the phrase is substantially the same as Saadya's, though he
explicitly regards it as an Arabism. He observes that the phrase nV^n no is
unparalleled in biblical or later Hebrew literature and that the preposition p
in classical Arabic is commonly used to distinguish parts of the day or month.
V^
means ' what time is it ?' * what watch of the night is it?'
44
ISAIAH XXI
appears to be pointed as a construct. For this usage ibn Ezra

refers to Isa. xv i (where the identical form occurs); he goes
on to compare Song of Songs vii 10 with Ezek. xxvii 18; in the
former text Sitpn jwp omits the place of origin (in the absolute)
of the wine, while in the latter (P^VO H?) the place is recorded.
He deduces that a particular night - 'such and such a night of
the week' - is intended (as in Isa. xv i). Qimhi also notes that
the form is a construct and he compares the phrase *T5? ^ 0 as it
occurs in Isa. xxxvi 2. In fact these forms are best regarded as
absolutes and as contractions of the normal forms *??!?, ^n. 1
As was indicated above, most modern scholars accept
Saadya's view of the meaning of the question. On this view the
image of the watchman is continued, and under the outward
form of the normal question addressed to city (night) watchmen
concerning the time of night,2 the prophet is consulted by his
questioners about their fate. Rashi's view that night here signifies the hard times of oppression is further supported by appeal
to Mic. iii 6 and Isa. ix i. 3
One very different tradition concerning the words HV'VB and
V'Vfc should be mentioned. According to P.T. Tcfanith 1:1 (p.
5),4 V'Vfc n& lav is interpreted as if it read Vrfc n&5 'What did
the watchman [sc. God] say?' It is probable that a similar
tradition lies behind the addition in certain LXX manuscripts
listed by Field (p. 465).6 They read TI oAoAuêis (puA&o-acov (TI
daro VUKTOS;) 'What do you cry, O watchman? (What of the
night?)' Here, however, the consonants were possibly read as a
piel participle of the root V77 'to howl'.
These divergent traditions are best seen as arising from
exegetical considerations rather than as indicating a variant
Hebrew text. Thus, the P.T. account of the oracle clearly
understood at least one set of the letters V^&TO in the sense given
1 Cf. G.K. 9of, 93W. For the view that the form is pausal, see e.g. Procksch.
2 Cf. e.g. D.K. and Lohmann, who cite Song of Songs iii 3; Pss. cxxi 3, cxxvii 1,
cxxx 6.
3 So e.g. D.K. and Erlandsson; cf. Wildberger.
4 See Schwab, p. 144.
5 I.e. as third masculine singular piel perfect of the (Aramaic) word y?Q (V1???) 'to
speak'. Cf. Rabin, Hebrew, p. 242.
6 Cf. Ziegler's edn, p. 195. In Untersuchungen (p. 48) Ziegler raises the question
whether the phrase is the original LXX reading at this point.
7 So Rabin, English, p. 3 0 3 ^ and Goshen-Gottstein; cf. Wildberger.
45
to it by the Massoretes. Further, the repetition itself was

(mis) understood as conveying the sense that the questioners
approached on two successive occasions. On the first occasion
they asked what was to come from the night, and on the second,
when Isaiah had had a chance to make enquiry, they asked
what the guardian of the world had said.
Verse 12, r\bf? mi i j nna
K. The words are deleted by Procksch metri causa? and he
cites in support the absence of any word in the LXX which
corresponds to IK and the fact that the prophet speaks of himself in the first person in verse 11. The LXX alone renders *i&w
by a finite verb. Its cpvAaaoxo, however, hardly answers to an
Aramaism RMJ 19W (so Procksch); it is more probable that it
read *fotf& (< n ^ ( i a ) x ) . Procksch's argument that *itP *lK
should be deleted may be rejected; for considerations of metre
in so short an oracle are very far from certain, and his insistence
on consistency in the use of first or second person is out of place.
riV1*? Dn 1p2 nn. With the exception of the LXX (on which,
see above), the words are taken as the watchman's answer to the
question of verse 11. The Vulgate and the Peshitta appear to
give a literal rendering: venit mane, et nox/t spf ^p llf 'Morning
has come and night.' The Targum interprets day and night as
symbols of reward and punishment: * There is a reward (ux)
for the righteous and punishment or vengeance (iismfc) for the
wicked.' This interpretation is followed by the P.T. 3 which
states ' Morning for Israel, night for idolaters' and it recurs in
a number of modern commentators, e.g. Vitringa, Rosenmuller,
Delitzsch, Dillmann-Kittel.
Saadya gives to nntf a somewhat different meaning; he
renders: pyniK Spn DD^Vl DD1KH3 lp 'Your day and your
night are both at an end.' In his notes on the text he paraphrases with nV*1? KVI K&V DDV pT DV 'YOU have no longer day
or night.' It is likely that Saadya invests nnN with meanings
1 Thus nV^n -pn?3 Mb xw no (p. 5). Cf. the LXX MSS. quoted above.
2 The line as a consequence has seven beats.
3 Taanith i: i (p. 5); cf. Schwab, p. 144.
46
ISAIAH
XXI
attested rather for the Arabic cognate verb ty1 than for Hebrew/
Aramaic nnK/KfiK which means 'to come, to arrive' (see B.D.B.,
p. 86; Jastrow, p. 132).
Qimhi appears to know Saadya's interpretation, for he says
that some interpreters give the words the meaning: ifcVtPl ISD
mV^m m m * Already (Dumah's) days and nights are over'
(i.e. O2Pp pT snn 'The time of their end has arrived'). On the
other hand he himself is of the opinion that day and night are in
the future and indicate a period of respite (for Dumah) to be
followed by renewed attack against that town.
Modern commentators usually follow the more literal translations of the Vulgate and Peshitta, though there is some discrepancy about how the verb nnK (in the perfect tense) is
translated and some discussion about the force of 03fl. The R. V.
renders the verb 'cometh' but notes in the margin as one
alternative 'is come'. N.E.B. has 'the morning comes'. Amongst
the commentators (e.g.) Ewald, Delitzsch, Cheyne (P.L), Buhl,
Duhm, Kaiser and Fohrer assent to a present tense (in some
cases without referring to the problem) while Dillmann-Kittel,
Marti, Gray, Procksch and Wildberger indicate a perfect tense
(i.e. 'morning has come'). Dillmann-Kittel explicitly repudiate the view that the perfect nnK can properly be rendered by a
present tense, as does Procksch, though the latter scholar raises
the question whether the perfect is a prophetic perfect and
finally concludes with the suggestion that the verb is best read
as an (Aramaic) participle (**?).
DM is rendered simply et by the Vulgate (cf. Pesh.) and this is
followed by most modern commentators. F. Buhl,2 however,
gives to the expression a concessive force: ' Morning comes even
if it is also night (wenn es auch Nacht ist)' on the grounds that
this alone makes sense of the watchman's answer. Marti and
Procksch both reject this translation on the grounds that it is
doubtful, the latter suggesting that the absence of a verb makes
it particularly so. On the other hand s ? 0? is attested with
concessive force3 and, as m alone is so used in Isa. xlix 15, it is
1 'To complete conclude, terminate (active)' Wehr, p. 3. Lane, p. 16, quotes
the phrase J%JS>. *JLc ^ 1 ' a year passed over him'.
2 Geschichte der Edomiter (Leipzig, 1893), pp. 68ff. More recently he has been
followed by Kaiser.
3 See B.D.B., p. 169b 6, and G.K. 160.
47
not unreasonable to suppose that mi was capable of bearing

this meaning. It seems to me that the real difficulty with Buhl's
view is that he renders nnK by a present tense for which there is
no justification (so Wildberger). If a past tense is substituted 'morning has come, even if it still be night' - the result is nonsensical even if interpreted symbolically of salvation and
oppression.
Procksch believes that a verb is missing from the phrase
nV^H Qtt. On this assumption, and comparing Isa. xviii 5, he
proposes to read nVV V&M 'the night is complete', assuming
further that the V of V&JI has dropped out by haplography. The
proposal is, however, to be rejected; first, there is no evidence
for such an emendation and, secondly, *?&! is not elsewhere
predicated of times or seasons in the sense of 'be full, complete'.
In Isa. xviii 5 it denotes grapes that are ripe.
Two other modern scholars also consider that a verb should
be predicated of nV1*? as nnx is of npa. A. Scheiberx and Rabin,
apparently independently, see such a verb in oa with which
they compare Arabic gmm 'to be full'. Rabin notes that the
Arabic word is used of wells, measures etc. and suggests that in
this verse the same verb might indicate that a period of time is
full (i.e. complete) in the way attested for the verb x*?D.
Scheiber admits that such a verb oJi does not occur elsewhere
in the bible, though Rabin (in V.T. 5 (1955), 154) finds another
instance in the obscure Hab. i g,2 ' their faces are sappy with burning wind (i.e. full, as with liquid)'. The translations of Isa. xxi
12 offered by both scholars are the same:' The day has come, the
night is full (i.e. completed)', i.e. the period of oppression is over.
The suggestion is attractive 3 in that it gives, without emendation, the more obvious meaning previously sought by Procksch.
In Rabin's case it is also consistent with his view that the oracle
is replete with Arabisms. On the other hand, according to Lane,
Arabic gmm properly denotes (of water) 'it became much or
abundant or it collected' and (of a measure) 'it becamefull orfilled'*
1 In V.T. 11 (1961), 455. He approves the sense gained by Procksch's emendation
though not the emendation itself.
2 B.D.B., p. 169.
3 It is mentioned as a conjecture by K.B. (3) , p. 190, and recorded as an alternative
in the margin of the N.E.B.
4 Wehr gives 'to gather'. B.D.B. compare the Hebrew particle with this Arabic
word.
48
ISAIAH XXI
There is, it seems to me, no evidence that the word could

properly be predicated of periods of time, and to assume for the
word, as Rabin does, the same usage as is attested for Hebrew
tfVfc involves a considerable degree of speculation including the
supposition (to which Rabin freely admits) that Isaiah is
'employing an unusually high degree of stylistic artificiality'.
That the author of the oracle should have used so unusual a
verb when NVD and (Arabic) mV would more readily have conveyed in Hebrew and Arabic the meaning 'to be full' is most
unlikely. There is, moreover, no support for such a view of Dil
either in the ancient versions or in the rabbinic commentators
(including significantly those who wrote in Arabic).
VS73 JVSnn DK. The LXX, Vulgate and Peshitta all follow the
M.T. closely, though the LXX has singular verbs (eocv nTfjs
r|Tei). The Targum reproduces the conditional sentence but
its verb am clearly answers to the M.T.'s law rather than to the
phrase TM p ^ i n which is not reproduced; inn pnK prrn DN
'If ye will repent, repent.' The Peshitta renders the verb nS72
by the Syriac cognate 6 and the Vulgate by quaero 'to seek,
ask, enquire'. Ibn Ezra observes that the word rwa is of Aramaic
origin and that the third radical yodh ( for n) is included in
both forms.1 He gives to the words the meaning to seek' or
' desire' (ttpa). It is not entirely clear (to me) what he means by
the paraphrase wp2 KiaV Wp2T\ DN but he probably intends ' If
you desire to come, (you may) desire' (i.e. it is permitted). 2
Saadya renders: m&raVtt 3*?DD 'ND ' No matter what may be
the quest you have entered upon (turn back from it and
approach).' In his notes he gives the following paraphrase:
Kmamsp oniD vbH mnVK p WinKD 'Turn back from the war
which you were intending (to prosecute).' He notes that the two
words vsn rsnn are Aramaic and he compares with them Dan.
vi 8, W? fljn? ^-Vs ('Whosoever shall ask a petition' (R.V.)).
By his comparison with this phrase, Saadya probably intends to
indicate that he takes the word TM in Isa. xxi 12 as the cognate
noun W3 which follows the verb (jvsnn) in the accusative, for
his translation is consistent with this supposition.
The Aramaic word NS72 properly denotes 'to seek, ask for,
1 iQIs a has the forms without the yodh.
2 Friedlander renders: 'If you wish to come, do so.'
49
pray' and with this Saadya's choice of the word tlb as a translation appears to agree. On the other hand, in his paraphrase he
gives to the words the particular meaning ' to seek (to prosecute)
a war'. 2
Another military use of n*to is attested in Obad. 6, and ibn
Janah, in his comments on the root nsn, compares Isa. xxi 12
with this verse. Here the niphal of nsn is found in parallelism
with the niphal offo&H:T3&2& W21 Ifr? titerjl f g 'See how
Esau's treasure is ransacked, his secret wealth hunted out!'
(N.E.B.; cf. R.V.: 'How are his hidden treasures sought up!').
The word nsn here plainly indicates that Edom's wealth was
' exposed and plundered' at the hands of Edom's enemies (see
B.D.B. under foDri).
Reference to the Arabic cognate word bgy may be held to
substantiate such hostile meanings for the root nsn. 3 As well as
' he sought diligently \ Lane gives for the word 'he acted injuriously
or wrongfully or tyranically' and for the phrase fcuJLc jr 4 'he
exalted himself against him9 'overpowered or oppressed him\
Dozy,
moreover, notes the meaning 'demander avec arrogance' for

the same word.4
Rabin, 5 who notes Saadya's interpretation of the verse, is
inclined to see behind ]V*nn (but not in rsn) the rare Arabic
word bcw which he thinks means 'to attack' as well as 'to sin'. 6
The word bcw is attested in Old South Arabian, and given the
meaning 'to commit a crime or treason' by A. Jamme (p. 85).
Rabin, in his English article, questions whether such a meaning
fits the contexts of the relevant South Arabian texts and then,
because the word is attested with the meaning 'to borrow a
1 See B.D.B., p. 1085; cf. Jastrow, p. 181. Syriac bc0 has similar meanings.
2 Ibn Ezra in his paraphrase also seems to allude to a military expedition; it is
clear from the context that 'to come' in the phrase wpa xia1? wpan DK has a
military connotation.
3 A hostile sense for Aramaic xya in the early Aramaic inscriptions is also listed
by Hoftijzer, p. 40.
4 Ibn Barun explicitly compares nva in Isa. xxi 12 with Arabic bgy,^ though
Wechter offers the English translations 'to request, desire*. It is doubtful, however, whether bgy can mean * request*. Gesenius-Buhl hold the theory that
' overpass the mark, swell up' is the fundamental meaning of the root rwn/bgy.
Cf. Donner and Rollig, 11, glossary.
5 For Rabin's complete translation of the Dumah oracle, see below, p. 80.
6 Belot gives for bc0 'commettre un de*lit'; the word is not listed by Lane, Dozy or
Wehr.
50
ISAIAH XXI
riding animal' (which, Rabin supposes, denotes part of the

preparations for war), he posits for it the extended meaning
'attack'. 1 Rabin's argument is unconvincing by reason of the
considerable degree of supposition that is necessary to achieve
such a meaning for what is (as he admits) a rare word. I am not
competent to enter into discussions of the Old South Arabian
evidence, but I may be permitted to comment that Rabin not
only claims for bcw the meaning 'attack', but also for bg in the
Thamudic phrase slm bg2 in which he thinks the second word is
a form of bgy.3 If, then, he thinks Arabic bgy (i.e. with g) is
capable of the meaning 'attack', it is not necessary to posit for
Saadya's understanding of Isa. xxi 12 that jrsnn is cognate with
a rare Arabic word bcw; it is enough to suggest that it reflects the
well-attested root bgy (see above). With this conclusion also the
M.T., which has a yodh between the y and the 1 of JPMD, is
consistent.
It is convenient here to consider also Rabin's view of the
word VM. In common with the Targum, Vulgate and Peshitta,
he takes the word to be an imperative and he gives to it the
meaning 'to pray' which is well attested for Aramaic NSn. He
considers that the word is here cognate with Arabic bgy, which,
in his English article, he says means 'to desire, to ask for'. In
his Hebrew article, however, he appears to give to the Arabic
word the meaning 'to pray'. He renders the phrase as a whole:
'If you have sinned/attacked, pray.' 4 The conditional structure
of the phrase matches the conditional phrase (compounded of
two imperatives) which then follows: ' If you repent, you will
go home' (for details of this, see below).
Again difficulties are presented by Rabin's analysis. First,
that the two words of the phrase should represent two different
roots, while possible, is surely unlikely.5 Secondly, it seems to be
necessary to Rabin's argument that the words should be couched
at least in 'pseudo-Arabicized Hebrew'. 6 ran with the meaning
that Rabin seeks for it might be characterized as Aramaicized
1 In his Hebrew article he states more boldly (but without any references) that it
is used many times of a raid or a sudden attack. See Hebrew, p. 249.
2 Branden, p. 418. He gives it the meaning 'vaincre'.
3 Rabin, English, p. 3o8n2ia.
4 As frequently in the Talmud, 'pray* means 'implore mercy'; cf. Jastrow, p. 181.
5 Cf. also Isaiah's UTasn xb *D "irnxn x1? ON where die same root is used. Isa. vii 9.
6 His actual phrase, used in an analogy, is 'pseudo-Italianized English'.
51
Hebrew with talmudic overtones, but it cannot be Arabicized

Hebrew. For bgy> while certainly capable of the meaning ' to
desire', is not attested, so far as I am aware, with the meaning
'to pray' and certainly not with the extended meaning 'to
implore mercy'.
One other modern interpretation of the phrase in question
should be noted. H. Winckler1 considers that in Job iii 5 a form
of the verb ns?a is paralleled by the word *?X3i and he consequently posits for the verb nsn the meaning c to help'. 2 The whole
phrase of Isa. xxi 12 vntf "nit? nsn VV2T) OK (sic) then means ' If
you really wish to help, then occupy your place on my side.'
Winckler's view may be rejected. He adduces no philological
evidence to support his theory; nor does he show why the verb
in Job iii 5 should not be taken as a form of the root nsn as is
customary.
Among the rabbinic scholars Qimhi paraphrases the forms of
nsn by using the verb *?Ktf 'to ask, enquire'. He interprets the
phrase (as a conditional imperfect followed by an imperative)
to mean that the enquirer will wish to ask again by reason of the
continued threat of the enemy which is implied by the symbol
of night following dawn.
Qimhi's view that the meaning of TW2 is *?KW 'ask, enquire' is
almost universally adopted by modern commentators and
translators.3 In conjunction with the words m x MW which
follow, the words are taken to indicate that because an ambiguous answer has been given by the prophet, the enquirers are free
to return to pose again their question.
However, as Rabin properly remarks, it is very doubtful
whether nsto can bear the meaning 'enquire, ask'. For, as he
says, in biblical Aramaic NSn never means 'to enquire, ask a
question' and in Targum Onkelos to the Pentateuch KSn
answers to Hebrew tfpn 'request, seek' and never to *?*?# (for
which *?$# is reserved). Further, on the view that this oracle is
early, no instance is cited by Hoftijzer for KSn with the sense
'ask, enquire' in early Aramaic inscriptions; rather he lists
1 Altorientalische Forschungen, vol. m (Leipzig, 1902), pp. 2i8ff.
2 He emends (by reference to the LXX) Job xxx 24 to find another instance of
the verb in this sense.
3 E.g. R.V., N.E.B., Hitzig, Delitzsch, Cheyne (P./.), D.K., Duhm, Marti, Gray,
Kaiser, Fohrer, Wildberger.
52
ISAIAH XXI
'vouloir, d^sirer, chercher'. We may add that B.D.B., K.B. and

K.B.(3) cite Isa. xxi 12 alone as an instance for ntf-t 'enquire,
fragen'. 1
YTIN *ntP. The two words are taken as imperatives by the
M.T., Vulgate (convertimini, venite) and by the rabbinic and
modern commentators. The LXX's -nap' ejioi OIKEI appears to
reflect a reading of the verb aw rather than of mtP and Trap' eôi
answers, perhaps, to **&$ rather than to TDK (see GoshenGottstein). The Peshitta renders MW by the adverb twb ' again'
and rr>N by a participial form (with pronoun) Dtytwn.2 The
Targum has '(If you will repent, repent) while ye can repent'
(anV yb^ pn*n where pnx may indicate that it read v>r>X in
this way).
Ibn Ezra interprets the words more physically: ' If you wish
to come, you may - return to those that sent you and come all
together' (sc. as a military expedition).3 For Saadya the phrase
means iVnpKl Htt? TrinN ' desist from (your military quest) and
approach', though unfortunately his notes do not make clear
what he intended by 'approach'.
As has been indicated above most modern commentators
interpret the words (taken with rsn VV2D DK) to indicate an
assurance to the prophet's questioners that they may return
later for a more definite answer if they so wish. 'If ye will
inquire, inquire, come again' (so e.g. R.V., margin).
Delitzsch is inclined to see a secondary meaning in 'mtp, i.e.
'to repent', and to this extent he follows the Targum and the
Vulgate (cf. Rashi). Dillmann-Kittel and Duhm, however, repudiate this view on syntactical grounds (D.K.) 4 or as being inconsistent with the function of the prophet of the oracle (Duhm).
Rabin has revived the idea that law here denotes 'repent'
in a religious sense. For TDK, however, he compares Ethiopic
atawa with the particular sense ' to return home' (either to one's
house or country) and he posits precisely this meaning for rnx in
1 It is likely that the Vulgate's si quaeritis, quaerite has considerably influenced
Western translations and commentators. For quaero in Latin means 'to ask,
enquire' as well as 'to seek'.
2 I.e. Si quaesituri estis, quaerite, turn demum venietis. So Walton, in, ad loc. Cf. the
grammatically identical but variant tyn ntwn cited by Diettrich.
3 Qimhi's alternative interpretation.
4 I.e. because 'repent' and 'come', even if asyndetically coupled, do not make
sense when juxtaposed.
53
the present verse. The two imperatives vnK "DIP constitute a conditional clause (cf. G.K. i iof) with the meaning 'If you repent,
then you will return' (i.e. home) or ' repent, that you may return \
The difficulty with Rabin's view is that he invests YnK with a
meaning which is attested only in Ethiopic and (possibly) in
Old South Arabian, but not in Hebrew, Aramaic or Arabic (cf.
the standard Arabic-English lexica). Even on Rabin's view that
the author is making use of Arabicized Hebrew, it is surely
implausible that he should make use of a word widely attested in
Semitic speech but with a special meaning attested only in
Ethiopic and (possibly) in South Arabian and for which there
is no supporting evidence in the ancient versions or rabbinic
commentators.
Another possibility for the word vnx may be posited by
reference to its two other occurrences in the book of Isaiah. In
lvi 9 and 12 the identical imperative plural form is found complete with the radical yodh.2 In lvi 12 the plural imperative is
immediately followed by a first person singular verb: VpK nnp>N
]?! 'Come ye, say they, I will fetch wine' (R.V.). It is probable
that the word is used here merely as a stereotyped interjection
just as the imperatives rD1?, T\27] etc. are also used without concord in this way.3 That in lvi 9 the same imperative does still
express an independent idea (viz. 'Come to eat') does not
militate against this view as e.g. TVD^/teb are also clearly used
sometimes in this way. On this view rnK in Isa. xxi 12 expresses merely encouragement like English ' Come now.. .'
Verses 13-15
Verse 13a, 3nr? Nfe&
Except for the LXX, which has no corresponding words,4 all
witnesses to the M.T. substantiate its reading.5 The words are
1 He quotes also as evidence Old South Arabian tw ' to return' and, tentatively,
Thamudic ty in the same sense. (He notes that Branden gives to the word the
meaning 'est venu (ici)', i.e. to arrive.) It is not clear to me how these words
substantiate the specific meaning 'come home*.
2 For this form, see G.K. 2o,t and 75U. Ibn Ezra notes the irregular tsere for
hateph-pathah beneath the aleph and suggests that the pointing arises from the
guttural character of aleph.
3 Cf. G.K. 105b.
4 On this, see Ziegler, Untersuchungen, p. 48.
5 The Peshitta does not reproduce the beth, having simply: m$ql dcrby. (The
readings dcrb and dcrb are attested; for which, see Diettrich.)
54
ISAIAH XXI
taken as the title to the prophecy contained in verses 13-17

and the title is derived from the keyword nnsn as it occurs in
the text of verse 13 (cf. on xxi 1). In the text of verse 13, the
M.T. reads 2*1^5, though the ancient versions understand the
consonants here to denote 2?? 'in the evening'. The versions
and commentators are, however, unanimous in understanding
S"isn in the title to refer to the Arabs or Arabia rather than to
* evening'. 1 As the title merely reproduces a keyword from the
text and so labels the oracle (i.e. The Oracle Tisn), Qimhi's
attempt to give to a the same meaning as Vtt 'concerning' is
otiose.
Verse 13b, lUpfl rrirrjfc vVn :ns?

The word ti^n is taken as an imperfect or jussive by all the
ancient versions and commentators,2 though the LXX and
Targum have for the M.T.'s second person plural the third
person singular and third person plural respectively. The LXX,
by so doing, appears to link verse 13 with verse 12 and to regard
the verb (KoiiJir)0f|OT|) as an amplification of the phrase Trap' ejioi
OIKEI (for which, see above).
DTH mmK is generally understood as a vocative following
the second person plural verb (e.g. 'you caravans of Dedan' N.E.B.), though in the Targum it is the subject of the third
person plural verb and in the LXX (ev TTJ 66<2> Ae8ccv) (m)niK
is taken as a form of rnfc ' path' rather than as the plural of the
noun ^0*?^ 'caravan'. 3 The phrase OTH mmK refers to the
caravans of the well-known merchant tribe who, according to
Ezek. xxv 13, lived in north-west Arabia on the southern
boundary of Edom, though according to Gen. x 7 they were
further to the south.4
A different view of the words OTH mmx M*br\ is taken by
Saadya. For him the word DTH alone is vocative and mmK is
1 The versions and Massoretes probably intended 'Against Arabia or the Arabs'.
2 Amongst moderns, e.g. Delitzsch, Duhm, Gray, Kaiser, Fohrer see the word as
a jussive.
3 B.D.B., p. 73b, mentions the possibility that the word mmx 'caravans' may be
derived from ma 'path' by metonymy; cf. ibn Janah's view: ^ U 3 (_
J A W I 'caravans which traverse the path'.
4 See further W. F. Albright, in G. Ebeling (ed.), Geschichte und Altes Testament (Alt
Festschrift, Tubingen, 1953), pp. iff.
THE T E X T A N D ITS MEANING
55
to be rendered DDDK^SK 'your guests', whereby he understands

the word as a qal participle form of the verb elsewhere used in
this sense in Jer. ix i. 1 ti^n is rendered as if it were a hiphil
(prpan pzr) and the phrase as a whole is taken to mean 'is it a
good 2 thing that you should make your guests pass the night in
the scrub ?'
lira. The word is rendered 'in the forest' or 'thicket' by the
ancient versions (6pu|ico, saltu, Nttnin, cb). The translation
'forest' is, of course, quite unsuited to the context (cf. Delitzsch).
Saadya uses the word HW (scry) which denotes 'scrub country'.
Delitzsch draws attention to the Arabic cognate root wcr which
is used properly in the sense 'rocky terrain', though such terrain
was also sometimes planted with trees which furnished firewood. Wehr gives for the noun the meaning 'covered with rock,
debris, rugged, wild, rough, roadless' and we may confidently
assert that the word nsp in this text denotes the harsh desert in
which the fugitives are to spend the night.
As has been indicated above, ms?3 is read by all the ancient
versions as the noun 2*1$ 'evening'. This interpretation is noted
by Qimhi, though he seems to prefer the alternative 'Arabia'.
Saadya renders by jonK 'is it good?', presumably deriving my
from the root ms? m 'to be sweet, pleasing' (see B.D.B., p. 787),
and interpreting msn adverbially.
Some commentators (e.g. Lowth, Michaelis, Rosenmuller,
Hitzig) adopt the reading 3??? 'in the evening' on the grounds
that it fits the context admirably and avoids the otherwise
harsh asyndetical 'in the forest, in Arabia'. 3 More recently,
however, it has been usual to retain the pointing of the M.T.
and to reject the tradition of the versions.4 Thus, D.K., Duhm,
Marti and Wildberger all reject the reading 33?? 'in the evening' on the grounds that it is superfluous with the verb irVn.
The argument is hardly an impressive one. Admittedly the
English phrase 'in the evening pass the night' (cf. German
' iibernachten/die Nacht zubringen') appears tautologous, but
yh includes in its semantic range the meaning ' lodge, encamp'
1 He explains the feminine ending as a mere variant like
2 For this translation, see under 3"wn below.
3 The same harshness led Marti to the view that nya was a corrupt dittograph of
4 Delitzsch thinks both traditions are correct and that as Dumah in the preceding
oracle has two senses, so here does any.
56
ISAIAH
XXI
and this conjoined with 'in the evening' gives a meaning which
is entirely appropriate.
Because the word ms? is not attested elsewhere in the O.T.
with the sense 'Arabia', and also because a does not elsewhere
follow m ('oracle') in the sense 'against', a number of scholars
have followed Ewald in seeing in m m the meaning 'in the
steppe, desert'. Ewald asserts that this is the original meaning
of the word which only later came to be used to characterize
the steppe-dwellers or Arabs.1 The word, however, does not
occur elsewhere in this sense,2 which is conveyed rather by the
feminine noun HT157 (B.D.B., p. 787b (4 and 5)). Further, there
is no versional or early Jewish tradition to support such a
meaning for the word in this verse. The noun 21V in the bible
properly denotes the Arabs or steppe-dwellers of North Arabia
(cf. Wiseman, Peoples, p. 289) and consequently the N.E.B. (so,
hesitantly, Gray) is probably accurate when, taking the
massoretic view of the pointing, it renders: ' (You caravans of
Dedan, that camp in the scrub) with the Arabs'. 3
For Saadya's taking the verse as a question, see on verse 15
below.
Verse 14,
The interpretation of this verse turns largely on the way in
which the verbs vnn and ia*rp are taken. The massoretic pointing of the former verb is ambiguous in that it may indicate
either an imperative or a perfect.4 The latter verb, however, is
pointed as a perfect and consequently it is natural to assume
that vnn was taken as a perfect by the Massoretes (cf. G.K.
7 6d).
The ancient versions, on the other hand, are unanimous in
taking both verbs as imperatives, with the natural consequence
that NttY) "pK i|-Wl is understood as a vocative. A number of
1 The shift in meaning took place, according to Ewald, not before the seventh
century B.C. and is first attested in Jer. iii 2 and Ezek. xxvii 21.
2 D.K. and Wildberger, who adopt Ewald's view, admit as much.
3 Cf. Rashi and ibn Ezra's paraphrases of the title oriiv Vy (Rashi), Drn,3-wi;?y
-lip " (ibn Ezra). Similarly Saadya renders the title: ax-waVx D nxj? 'The
prophecy about the Arabs'.
4 Jer. xii 9 contains the same form where the context clearly indicates an imperative; cf. G.K. 68i.
57
modern commentators adopt this view of the verse on the

grounds that it can, as a consequence, be brought into relation
with verse 13, and because the imperfect n^Vn in verse 13
cannot be harmonized with perfects in verse 14. Accordingly
they emend the pointing of the second verb to read ^7
Amongst the rabbinic commentators, Saadya also regards the
two verses as connected. Verse 14 continues the question posed
in verse 13: ' (Is it good).. . that you offer food to the fugitive
without coming to meet the thirsty with water ?' In his notes on
this passage Saadya states that he understands the prophecy as a
whole to refer by allusion (fnKtPKV*o) to the incident mentioned
in P.T. Tcfanith iv.2 The passage, which in turn refers to Isa.
xxi 13^ relates that 80,000 young men of the priesthood fled
through Nebuchadrezzar's lines to the Ishmaelites; the latter,
when asked for refreshment, gave the refugees only a species of
salt plant and leather skins full of air with the unfortunate result
that one of them choked and died (for Saadya the number who
died has swollen to 80,000).3
It is likely that this particular interpretation of the prophecy
has led Saadya to adapt his translation of the Hebrew text
accordingly. Apart from the question of the meanings he gives
to the words (and some are at least possible), there is nothing
in the Hebrew text to justify his rendering the first half of verse
14 in a negative or privative sense, i.e. 'without coming to meet
the fugitive with water.. .' ()K jri p ) .
It is possible that the tradition to which Saadya refers has
influenced the massoretic pointing of *&7R (i-e a s a perfect), for
thereby reference to the incident mentioned is facilitated.
Another view of the relationship between verses 13 and 14 is
presented by Gray, who adopts the pointing of the M.T. for the
verbs of verse 14. He regards verse 13 as fragmentary and the
words in it OTH mmN as parallel to KYI -pK ia&r and as
marking the beginning of the couplet:
The caravans of the Dedanites to meet the thirsty
brought water,
1 So e.g. Rosenmuller, Cheyne (P./.), D.K., Duhm, Marti, Procksch, Kaiser,
Wildberger.
2 P. 43; cf. Schwab, pp. ig2f. An expanded form of the same story occurs in
Midrasch Echa Rabbati, ed. S. Buber (Wilna, 1899), 11, 4, p. 22a.
3 Rashi in his comments also refers to this story.
58
ISAIAH XXI
The inhabitants of the land of TemaD met the

fugitives with bread.
The reconstruction is attractive but it should be noted that
Gray caji make virtually nothing of the remainder of the words
of verse 13.
Kissane follows Gray in understanding ' caravans of Dedanites' as parallel to 'inhabitants of the land of Tema D \ For him,
though, the first words of verse 13 constitute a relative clause in
apposition to DTH mmK which follows: 'Ye that abide in a
forest in the steppe, ye caravans of Dedanites...' The verbs
which follow are taken as imperatives and the fugitives are the
tribe of Kedar.
Kissane's treatment of the verse is an improvement on Gray's
and his understanding of the syntax, though perhaps somewhat
strained,1 is possible. On the other hand, it is strange on his
view that no indication is given of the identity of the fugitives
until verses i6f.
KTi is the important place in the Hejaz in north-west Arabia
(modern Teyma) which was famous for its water supply (for
literature, see B.D.B., p. 1066). In Gen. xxv 15 Teima is listed
as the ninth son of Ishmael, and in Jer. xxv 23 KJDTI and p i are
again found juxtaposed.
The LXX renders by Gaipiocv and the other ancient versions
by 'austri', K&m and tymri3. It is most unlikely that the LXX
read pTi in this verse, as Gaijjiav occurs also in the LXX of Gen.
xxv 15 and Jer. xxv 23. Kocherus's2 analysis of the treatment of
both K&TI and pT) by the ancient versions suggests that the
LXX used 0ai|aav indifferently to render Njrn and pTi. While
the other versions generally understood NfcT) to be a proper
name they occasionally confuse it (as here) with p*n; for this
latter is translated by them in a number of places as ' South' (i.e.
from the root p*).3
1nVa. The third person singular suffix is usually understood
to refer to the fugitive, i.e. the bread that he needs.4 The suffix
1 He cites no other instance of a relative clause preceding a subject to which it
refers.
2 See Rosenmiiller, pp. 487^ I have not had access to Kocherus's work.
3 T. Noldeke, in Encyclopaedia Biblica, 1, ed. J. S. Black and T. K. Cheyne, under the
entry 'Ishmael' thinks that xwr\ and JOT are synonyms; the former is derived
from a root x?r with the meaning 'South'.
4 Cf. Rosenmuller, Delitzsch, Wildberger, etc.
59
is not translated by the LXX and Vulgate, but this does not
amount to evidence that it was not before them. The Peshitta
has a second masculine plural suffix (Ihmkwn) with which the
Targum appears to agree: "p^DK pn*n X&& ' (prepare) what you
are eating5. These again are best regarded as legitimate (if free)
renderings of the M.T. rather than evidence for a different text.
Verse 15, nw&i r^n - ? VTJJ rrirnn

V
mm v l ' v
T
The verse seems to indicate the cause of the flight of the refugees.
The repetition of 'JBB together with the listed weapons depicts
sharply the overwhelming nature of the danger from which they
flee (so e.g. Delitzsch).
D at the beginning of the verse is best taken as 'for' with the
Targum and Peshitta and Vulgate rather than as asseverative.
The LXX renders ^Dfc each time by 5id TO TrAf|0os which may
represent a misreading of 3in as :n (multitude) or 2V31&
(abundance).1
Amongst modern scholars Kissane has sought to avoid repetition of the word 'sword' by repointing the M.T.'s ^ 7 H
'swords' as fl^ri 'desolations'. There is, however, no evidence
to support his proposal. Further, as the word ^??D denotes
ruins (e.g. of a city) it hardly fits the context in which men are
said to flee from the weapons and press of war.
Wlttt Tin. The passive participle n&nttt is rendered ND^tP
'drawn' by the Targum. Rosenmiiller approves this meaning,
citing the Talmudic word wttt with the meaning 'to skin (an
animal) '. 2 The Peshitta renders ltys 'sharpened, polished' with
which Saadya's V)pS ^ 0 'sharpened sword' may be compared.
Ibn Ezra, probably referring to Saadya's version, states that
some authorities regard ntmttt as a by-form of mnB1?, the nun
and the lamedh being interchangeable.3 A number of rhodern
commentators adopt this view of the word, or mention it but
prefer to think that nt2n&2 is in fact a scribal mistake for n
1 See Ottley, p. 209. iQIs a reads rva-i for main which may (but GoshenGottstein thinks not) reflect the same tradition.
2 For this word, see B.D.B., p. 643. On this view it would be more appropriate to
cite Isa. xxxiii 23 where the word is used of slack (i.e. loosed) rigging.
3 Cf. Rashi (who cites the parallel roip? for ro^V in Neh. xiii 7) and ben Bilam,
P- 69.
4 Cf. e.g. Marti, Gray, Procksch, Kissane.
60
ISAIAH XXI
A different view of the word is suggested by ibn Janah, who

is followed by ibn Ezra, Rashi and Qimhi. He compares the
phrase predicated of the Amalekites in i Sam. xxx 16, D^tmDl
f *)Kn VD ^D'VB, and renders the word DÎDI by mnbstwn ' they
were spread out (over all the ground) \ x The meaning of n n a
n&nt?2 is, as Qimhi makes plain, that the fugitives flee from the
swords which are spread abroad (n&tPD) in their land.
w m nttfp lit. 'the trodden bow5, the bow which has been
strung ready for battle. It does not indicate the bow bent by
the archer's drawing back the string prior to shooting.2
n&nV&n T3D. The noun is used of a storm in Isa. xxx 27. Here
it seems to indicate vehemence, force.
Verses 16 & 17
The verses are similar in form and language to Isa. xvi 14.
iQIs a reads ww WlVtP for the M.T.'s nitf and this may constitute evidence that yfov was lost by homoioarkton. The reading
is adopted by Kaiser who notes that the figure has 'long been
noticed as missing'. Earlier, Duhm, followed by Marti, compared Isa. xvi 14 where the number three does occur and consequently they supposed that the prophet here left a space for
the number until he could be sure what it should be. In support
of their supposition they draw attention to 1 Sam. xiii 1 where
a number has dropped out and to the plural phrase TDtP "WD
which, they argue, follows awkwardly in the M.T. upon a
single year.
Fundamentally, the argument of Duhm and Marti rests upon
the supposition that, because the phrase TDP 'WD is awkward
if it follows a single year, a higher number ought to be present
in the text. Their fundamental assumption is, however, far from
axiomatic. The phrase from TI57D to TDW is rendered by Saadya
'In a year counted as the years of a hireling' (i.e. exactly; cf.
N.E.B.) which indicates the possibility that TW nBD was a
1 Rashi cites also 1 Sam. v 18, and Qimlji, Judg. xv 9.
2 See J. A. Emerton in J.T.S. n.s. 27 (1976), 39if.
6l
stereotyped phrase with this particular meaning. That the

phrase occurs in identical form but after 'three years' in Isa.
xvi 14 is also consistent with this view of it. Similarly, the renderings of the LXX and Vulgate 1 are consistent with this view, for
they make a singular year of the hireling match precisely the
singular year designated TI EVIOCUTOS COS IVIOUTOS liiaOcoToO/
Adhuc in uno anno quasi anno mercenarii.
If the argument of Duhm and Marti fails at this point, then

there is no force in an appeal to the parallel in Isa. xvi 14.
Indeed, it is equally possible that the reading o*W BnVtP of
iQIs a arises from an attempt to harmonize the two texts.2
YTp TOD TOD may denote either Kedar's (military) dominance (cf. with D.K. viii 7, x 18, xvi 14, xvii 3f), or its illustrious
leaders (so Saadya, xVix); either interpretation could be predicated of the verb n*?D.3
Up. The word denotes a particular Arabian tribe in Gen.
xxv 13 but is used elsewhere,4 as here, of all the desert people
east and south of Palestine.5
TWp 1D0D i w . The versions all follow the M.T. in understanding ixm as a noun in the construct which is predicated of
the verb iro\ For the LXX and Vulgate up *n mm is an
absolute following rwp 1D0 irw6 but for the Targum and
Peshitta the phrase is understood to be in apposition to mm
lip *12 and a second subject of the verb IBS?^.
The word iKtm is read by Saadya as a waw consecutive with
perfect and he renders the initial phrases: 'There will remain
('paPl) a small number of their archers and their heroes will
decrease...' 7 In his notes he observes that IDO& is capable of
the meaning 'small number' and for this usage he compares
1 The Peshitta follows the M.T. more closely and the Targum - 'at the end of the
years' is no help.
2 Gf. Wildberger. That no reading .OTO (as opposed to nw) is attested further
weakens the case for adopting the reading of iQIs ft . Here the old maxim
difficilior lectio potior is appropriate.
3 For a personal subject of n^s, cf. Isa. i 28, xxix 20, xxxi 3; for an abstract
subject, cf. xvi 4.
4 Cf. with Gray Isa. xlii 11, lx 7; Jer. ii 10.
5 So Rosenmiiller. Saadya appears to take this view: he notes that the oracle opens
with a rebuke against the Dedanites but closes with notice of the punishment of
the Sons of Kedar, * for the many Arabian tribes all had dealings with each other
and were as one people'.
6 I.e. ivm is followed by five genitives (so Delitzsch, who compares Isa. x 12).
7 So, apparently independently, Eitan. N.E.B. adopts this view.
62
ISAIAH XXI
Gen. xxxiv 30 (*IBOB TUD ^N1 'My numbers are few' - N.E.B.)
and Jer. xliv 28. The interpretation has the merit that it makes
two parallel clauses of the verse, and avoids understanding the
verse as composed of five consecutive constructs.
Lowth (so Cheyne and Marti), referring to the Vulgate's
sagittariorum fortium, proposes to read ntPj? "H2? ' mighty bowmen' for the M.T.'s H'Dl Wp. This conjectural restoration is
probably occasioned1 by the view that nwp properly denotes
'bows' rather than 'bowmen'. A similar view of ntPp is taken by
Kissane who proposes the emendation ntPp Wl after the phrase
ntPp 7]fcM 'bowmen' in Jer. iv 29.
On the other hand, the Vulgate, Peshitta and Saadya translate the word rwp 'archers' (Vulg.: sagittariorum, Pesh.: qst\
Saadya: Dnn*Wi) and with Rosenmiiller we may suppose that
the word rwp 'bow' can by metonymy indicate 'bowmen'
collectively.2
That the Arabs were noted as bowmen is clear from Gen. xxi
20 where Ishmael is recorded as having appropriated the art (so
Rashi). According to Qimhi the prophecy against the sons of
Kedar was occasioned because of their oppressive behaviour
towards Israel at the time, perhaps that they attacked Israel as
allies of the king of Assyria.
The concluding authentication of the prophecy as a word of
Yahweh, the God of Israel, matches the similar conclusion of
verses 1-10. No important textual or exegetical questions are
raised by the words.
1 Lowth himself offers no explanation other than that his emendation 'seems
right'.
2 Hitzig, who compares i Kings xix 18 where 'knees' (o-oia) stands for the number of individual men.
Exegesis and Historical Background

Verses I-IO
As has been indicated in the textual commentary above, the
rabbinic commentators ibn Janah, ibn Ezra and Rashi regard
this prophecy in the light of Daniel v, and by reference to that
text they interpret a number of the words and phrases of Isa.
xxi I - I O . For these authorities, therefore, the prophecy as a
whole concerns the fall of Belshazzar and the Babylonian
empire at the hands of Darius the Mede. However, the legendary character of the book of Daniel together with the secondcentury date to which it is generally ascribed render such a
comparison with the detailed deductions drawn from it highly
questionable. It remains of value, however, as an indication that
Jewish tradition from comparatively early times considered
that the prophecy concerned the fall of the Neo-Babylonian
empire, which we know to have taken place at the hands of
Cyrus II of Persia in 539 B.C.1
In general, this view of the prophecy is adopted by subsequent commentators. Thus, for example, Rosenmuller (1793)
states that the prophecy was obviously delivered at the same
time as that of Isa. xiii and concerns (see verse 9) the fall of
Babylon at the hands of Cyrus. The contents of the prophecy,
to the smallest detail, match exactly the accounts of that fall
preserved in other ancient authors. It is for Rosenmuller
beyond question that the poet (not, of course, Isaiah) was
himself present in Babylon and witnessed in person the events
that he describes.
The ancient authors to whom Rosenmuller refers are the
Greek historians Herodotus and Xenophon.2 In their accounts
of the fall of Babylon at the hands of Cyrus both authors recount
that the Babylonians in the centre of the city were dancing
and making merry at a festival (Herodotus) which lasted all
1 See C.A.H., in, p. 224.
2 See Histories 1 191 and Cyropaedia vn 15*
63
64
ISAIAH
XXI
night long (Xenophon) and consequently were ill-prepared to

defend the city. These accounts together with the story of
Belshazzar in Daniel v are held by Rosenmuller to match precisely the scene of the feasting princes depicted in Isa. xxi 5 as
well as the reference in verse 4 to the evening of pleasure
('pwn *]tPl) which was turned to trembling. Further correspondence of details is seen by him in the reference to asses and
camels in verse 7, for Xenophon mentions the important part
played by camels in Cyrus's defeat of Croesus (Cyropaedia vn 27),
and Herodotus that played by asses in Darius Hystaspis's
battles against the Scythian horsemen.1 For Rosenmuller,
therefore, Isa. xxi 1-10 reflects accurately the military characteristics of the Persian armies.
Rosenmuller's understanding of the historical background
of the prophecy may be regarded as the orthodox critical view
of the matter. It is followed, for example, by Delitzsch (1),
though he initially differed from Rosenmuller in considering
that there was nothing inherently improbable in Isaiah's having
composed it.2 In his treatment of Isa. xiii, which he also initially
attributed to Isaiah, he concluded that Isaiah had clearly
perceived the pattern of Yahweh's activity in history and seen
that the overthrow of the Babylonian empire would follow
proteron-hysteron the overthrow of the Assyrian. 'The fact that so
far-reaching an insight was granted to him into the counsels of
God, was not merely founded on his own personality, but
rested chiefly on the position which he occupied in the midst of
the first beginnings of the age of the great empires.' 3 More
particularly Isa. xxxix indicates that in Hezekiah's time Isaiah
had seen that ultimate judgement would be inflicted by the
Babylonian empire rather than by the Assyrian. These earlier
comments of Delitzsch are of considerable importance in
regard to the transmission of the text of Isa. xxi and to them we
shall have occasion to return.
P. Kleinert was the first to question what I have described
as critical orthodoxy. In fact Kleinert was anticipated by
1 Histories iv 129; according to Rosenmuller, the asses were probably ridden by
(Persian) Carmanians.
2 In the 4th edn of his commentary Delitzsch adopts the orthodox critical view,
having explicitly changed his mind; for his comments, see 4th edn, ad loc.
3 E.T. (Martin), p. 296.
EXEGESIS AND HISTORICAL BACKGROUND
65
G. Smith, but the latter scholar did not provide a critical and
exegetical justification for his suggestion. To him the language
of the prophecy, its terseness, its weighty expression and its
formative power2 tell decisively against a date in the sixthcentury Babylonian period. Such characteristics distinguish it
very markedly from the allegedly comparable Isa. xiii f and
xxxiv, as well as from the Babylonian oracles of Jeremiah and
Ezekiel.
Those prophecies from the book of Isaiah which concern the
downfall of Babylon in the sixth century B.C. (Isa. xiii f and
xlvff) depict that event as an occasion of comfort and joy for
Judah, which had hitherto experienced the overpowering might
of her tyranny. Isa. xxi, on the other hand, offers a message
(including the phrase 'Babylon is fallen5, verse 9) which, so far
from being an occasion of triumph and relief, is one of terror
and anguish (see verses 3f) for God's people, for whom 'as
threshed5 (verse 10) the prophet feels painful and resigned
sympathy. Finally, while the Medes (verse 2) were known as
destroyers of Babylon in the sixth century (cf. xiii 17; Jer. li 11,
28), the Elamites (verse 2) were not. It was the Aryan Persians,
not the Semitic Elamites to whom hopes for freedom from
Babylon were directed. Elamites are indeed mentioned by
prophetic writers of the time but simply as one of the peoples
who, like Assyria, were to be brought low before Nebuchadrezzar's onslaught (Jer. xlix 34ff, Ezek. xxxii 24).
For Kleinert such difficulties prompt the question whether a
fall of Babylon in Isaiah's own time might not fit better the
language and expression of the prophecy. Because the prophecy
follows immediately that of chapter xx, and because chapter xx
concerns Sargon5s capture of Ashdod in 711 B.C., he posits that
Assyrian king's capture of Babylon in 710 {C.A.H., in, p. 50) as
the true background to Isa. xxi 1-10. Kleinert finds confirmation of his hypothesis in a number of references in the prophecy.
First, verse 2 designates the attacker as boged and soded; the
retrospective notice of Isa. xxxiii 1, ' Woe to thee that spoilest,
and thou wast not spoiled (sdd); and dealest treacherously and
they dealt not treacherously (bgd) with thee', clearly indicates
1 In Transactions of the Society of Biblical Archaeology (London, 1873), P* 329
2 All of it is in vocabulary and thought demonstrably comparable to other
Isaianic material.
66
ISAIAH XXI
the Assyrians. Secondly, the reference to Elamites and Medes

(verse 2) as attackers may be understood in the light of the fact
that Sargon conquered Elam in 721 and Media in 715. Isa. xi 11
and 2 Kings xvii 6, xviii 11 mention Elam and Media as places
to which exiles from the Northern Kingdom were transported,
and Isa. xxii 6 suggests that Elamites served as loyal vassals in
the Assyrian army. Thirdly, the caravan of asses and camels
indicates not a campaign march, but a column of booty wagons
conducted by cavalry; mention of such columns is typical of
Assyrian records of the time. As an example Kleinert quotes
Sennacherib's description (Bellino cylinder; cf. Luckenbill, 11,
no. 270) of his spoils following the battle of Kish (702 B.C.):
'The chariots, wagons, horses, mules, asses, camels and
(Bactrian) camels which (the enemy) had forsaken at the onset
of battle, my hands seized.5
Kleinert's second main point is that the fall of Babylon at the
hands of Sargon was one in which Judah and its prophet
must have had a burning interest. The sickness and recovery of
Hezekiah, which Kleinert dates in 712, afforded Merodachbaladan of Babylon an occasion to send an embassy to Jerusalem
with the purpose of seeking an alliance with Hezekiah against
the common Assyrian foe (2 Kings xx). Hezekiah's enthusiastic
treatment of that embassy suggests that Judah was inclined to
see in Babylon and Merodach-baladan a hope of deliverance;
to Isaiah, however, the terrible truth was revealed that Babylon
would fall, that Judah would continue on the threshing-floor
(verse 10) and that night would continue to darken Edom
(verses 1 if). The prophecy, following that of chapter xx
(711 B.C.), is to be dated in 710 when the decreed fate of
Babylon was set in motion.
Kleinert's treatment of the prophecy was substantially
accepted by T. K. Cheyne in his Prophecies of Isaiah first
published in 1880. Cheyne noted, however, that the evidence
in favour of Kleinert's dating was largely exegetical and that as
such it would not command universal assent. He further gladly
admitted that ' a fuller knowledge of the circumstances of the
Jews might conceivably enable us to reconcile the prophecy
with a date at the close of the exile'.
By 1895,
his publication Introduction to the Book of Isaiah,
Cheyne had changed his mind, and he then advanced a number
67
of exegetical considerations to meet the more important points

raised by Kleinert.1 It is clear that to a considerable extent
Cheyne was influenced in his change of heart by the arguments
advanced against Kleinert and in favour of the 'orthodox'
view in two important commentaries published in Germany in
the intervening years, viz. that of C. F. A. Dillmann (1890)
and that of B. Duhm (1892).
For Dillmann the fatal objection to Kleinert's understanding
of the prophecy was the fact that Elam and Media, represented
in the prophecy as attacking Babylon (verse 2), were enemies
of Sargon and indeed were allied with Babylon against him.2
Furthermore, it is most unlikely that Isaiah with his indefatigable dislike of all alliances would have regarded with grief and
dismay the (even enforced) ending of an alliance with Babylon.
On the contrary, the prophet is depicted (verse 8) as long
awaiting with anticipation the fall of Babylon together with its
gods.
While he is prepared to agree with Kleinert that some of the
language of the prophecy is Isaianic, Dillmann is concerned to
point out that some of it is not. Apart from individual words and
phrases,3 Dillmann suggests that the general lack of clarity
(verses 2, 4, 7, 9), the emphasis on personal feelings (verse 3)
and the distinction between prophet and watchman (verse 6)
tell against Isaianic authorship. On the contrary, they suggest
an exilic prophet who awaited long and anxiously Babylon's
end and who, being unknown to his contemporaries, properly
described himself as a watchman. He concludes that the
prophecy is to be dated just before Cyrus's campaign of 539 B.C.
which ended in Babylon's fall and is as old as that of Isaiah xiii
but from a different hand.
Dillmann and Duhm both comment further on the mention
in verse 2 of Elam and Media. For them it is significant that
Cyrus, before he conquered the Medes in 550-549, and
constituted their district part of his empire, was king of Anshan,
a place which, with its capital Susa, so far from being Persian,
1 It is not necessary to enter into the details of Cheyne's somewhat agonized reappraisal of exegetical considerations.
2 In regard to xxii 6 Dillmann subscribes to the view that Elamite mercenaries
served in Sennacherib's army.
3 For which, see Dillmann, ad loc.
68
ISAIAH XXI
was situated in the eastern or northern district of Elam. 1 It is

consequently reasonable to suppose that from the Babylonian
point of view Cyrus's armies should be designated Elamites and
Medes.
These criticisms of Kleinert's views (particularly in regard
to his treatment of Elam and Media), together with Dillmann's
argument that Elamites and Medes (verse 2) denote the united
forces of Cyrus, are reproduced in the majority of later commentators 2 who are thereby prompted to champion anew the
orthodox critical view of the prophecy. It is convenient here to
record that Galling, who accepts this view, attempts to define
more precisely the date of the prophecy on the basis of our
increased knowledge of the movements of the last Babylonian
king, Nabonidus, derived from the newly discovered Harran
inscriptions.3 For Galling, 539 (the date of Babylon's fall) is the
terminus ad quern of the prophecy, which can be no vaticinium ex
eventu, as Cyrus did not destroy the images of the Babylonian

gods (cf. verse 9) but rather was concerned to foster the cults of
his subject peoples. The conquest and annexation of Media is,
in the light of verse 2, the initial terminus a quo. This can,
however, be brought down to 546, the date of Nabonidus's
return to Babylon from his sojourn in the Arabian desert. For
the fact that Isa. xxi 1 if and i3ff are concerned with upheaval
in the deserts to the south of Palestine answers to what is likely
to have been the situation there after Nabonidus returned to
Babylon.4 That these oracles are closely connected with Isa.
xxi 1-10 (even if they were added by a redactor), suggests that
all three oracles were concerned with the same political situation. The situation involved on the one hand unrest amongst
the Arab tribes following Nabonidus's departure and on the
other hand the united Persian empire being then in a position
to mount a pincer attack on Babylon, the Medes attacking from
the north, the Elamites from the south. Such a situation would
1 See, with Dillmann, particularly HaleVy in R.E.J. (1889), i62ff.
2 So e.g. Cheyne (I.B.I.), Marti, Gray, Procksch, Fohrer, Kaiser, Galling and
Wildberger. Kaiser and Galling follow C.A.H., iv, in giving 553 B.C. as the date
for Cyrus's defeat and annexation of the Medes. Wildberger argues that the use
of the term * Medes' in xiii 17 and of' Medes and Elamites' in xxi 2 may suggest
that the prophecies were compiled before Cyrus the Persian gained complete
mastery of these peoples; see further below.
3 For which, see Gadd.
4 For details, see on these prophecies below.
69
be fraught with anxiety and uncertainty for the Jews and yet it
would fill them with hope that their time of exile was nearly at
an end. In short, the situation was similar to that nearly two
centuries earlier when the words of Isaiah were proclaimed in
the tension-filled time of the Syro-Ephraimite war (Isa. vii).
Galling's account of this prophecy cannot be evaluated
without reference to the plausibility of his treatment of the
other prophecies of the chapter on which substantially it is
dependent. It is nonetheless based, as are others of the orthodox
critical school, upon a particular interpretation of the allimportant reference to Elam and Media in verse 2, viz. that it
contains a reference to the united forces of Cyrus.
In 1898, W. H. Cobb published another attempt to attribute
the prophecy to the eighth century B.C. and to the hand of
Isaiah of Jerusalem. Cobb believed that Jerusalem rather than
Babylon was the beleaguered city of verses 1-8, and that consequently the prophet's distress is of the same sort as that
portrayed in chapter xxii 1-14. Verse 9 refers to a different
situation, namely, the fall of the Assyrian empire, for which the
name Babylon is used. For Cobb the similarity of words and
expressions in chapters xxi 2-7 and xxii 4-7, 12-14 is so striking
that when placed side by side they amount almost to 'a continuous narration'. For example, the enemies of verse 2 (the
Elamites and Medes) are matched by the enemies of xxii 6
(Elam and Kir). The eating and drinking of the princes
(verse 5) is matched in xxii 12 by the feasting of the inhabitants
of Jerusalem. Such similarities indicate for Cobb that xxi 1-8,
like xxii 1-14, is concerned with the Assyrian pressure on
Jerusalem at the end of the eighth century B.C. (705-700) and
the reference to Elam in verse 2, like that of xxii 6, indicates
that Elamites were serving in Sennacherib's army.
In verse 9 Cobb argues that Babylon stands for the united
world power Assyria-Babylon which, with its gods, is itself to
fall when Yahweh's purposes for Jerusalem are fulfilled. 'As the
representative city of Jahve is now threatened by heathen
Asshur, so the representative city of that kingdom shall then
be destroyed.' But in order that his dating of the prophecy
should not be entirely dependent upon the identification of
Babylon with Assyria, Cobb proposes another possibility for
verse 9, viz. that it concerns Sennacherib's capture of Babylon
70
ISAIAH XXI
in 704 (cf. C.A.H., in, pp. 63f), which the prophet sees as the
only possible outcome of an attempt by Hezekiah to enlist
the support of Merodach-baladan. Such an attempt on the part
of Hezekiah is assumed by Cobb as natural in view of his
having sent an embassy to Egypt (Isa. xxx 6). Cobb sees in
Isa. xxi 7 and 9 references to the dispatch of presents in a
guarded convoy to Merodach-baladan in Babylon (verse 7)
and, after a period of suspense, its return (verse 9) with the
accompanying revelation that Babylon is fallen and therewith
Judah's hopes for freedom.
Cobb's views have attracted little or no support, and with
justice they are described by Marti as 'geistreich'. Apart from
the fact that the unity of thought in the prophecy is quite
destroyed and the text reduced to a number of somewhat
unrelated sections, his arguments are tortuous and poorly based.
For example, while his comparison of chapters xxi and xxii
indicates some points of contact, it obscures many of the
differences by the dubious device of suggesting that the verses,
juxtaposed, form a continuous narration. Secondly, his argument that Babylon stands for Assyria is based largely upon
observations of C. P. Tiele1 who, justifying the title of his work,
was concerned to show with a wide perspective the essential
unity of the history of these two nations. To appropriate such
observations for the purpose of supporting the allegation that
Hebrew authors did not always distinguish the two powers is
little short of absurd.
In 1900 and 1913 two British scholars, W. E. Barnes and
C. Boutflower, both published articles urging that Isa. xxi 1-10
be attributed to Isaiah and dated in Assyrian times.
Barnes urges that the exegetical difficulties presented by the
oracle are fewer and less important on the assumption that it
belongs to the reign of Hezekiah than on the assumption that
it belongs to the time of Cyrus's capture of Babylon in 539.
Amongst the difficulties he adduces for the orthodox critical
view are, e.g.: (verse 2) Cyrus never styled himself an Elamite,
nor was he known as such either in the bible or in Persian
monuments; (verses 3f) the distress of the prophet at the fall
of Babylon alluded to in verse 2 is very difficult to understand
1 Babylonische-assyrische Geschichte (8 vols., Gotha, 1886-8).
71
if it marks the end of the exile and, in any case, is at odds with
the apparent satisfaction of verse 9; (verse 5) while it matches
the ' scandalous' accounts of Herodotus and Xenophon, it does
not match the Nabonidus-Cyrus Chronicle (cf. A.N.E.T.,
pp. 3i4ff) which states that Cyrus entered Babylon without
fighting; the latter notice is usually regarded as more reliable
than the former.
Barnes himself assumes that verses 1-5 concern not Babylon,
but Jerusalem threatened from the south-west (the area of the
Egyptian border) by Sargon II's forces (including Elamite and
Median auxiliaries). In this situation the Judaeans mistakenly
put their hopes in and rejoice over (verse 5; cf. xxii 13) an
alliance with Merodach-baladan, de facto king of Babylon. In
this connexion verses 6-9 contain Isaiah's prophecy of Merodach-baladan's fall and the end of Babylon in 710. Among the
advantages of such a view Barnes notes that the prophet's
display of lively distress (verses 3f) is better explained if the
object of the threat (verse 2) is Jerusalem rather than Babylon.
That he displays (if not satisfaction) cold impartiality at the
fall of Babylon (verse 9) is also better explained on the assumption that Babylon is an ally of Judah - one more broken reed
upon which the people were tempted to lean. With this
assumption the sympathetic tone of verse 10 (Israel as crushed)
is in accord.
Barnes's account of the prophecy has much to commend it.
His criticisms of the orthodox critical view are well directed and
telling, and his explanation of the reactions of the prophet far
less strained. His exegetical arguments (and such primarily
they are) constitute what is (necessarily) a somewhat speculative
hypothesis, but it is a hypothesis that meets the major problems
confronting exegetes. On the other hand, the very fact that he
is able to lay all such problems on the basis of a hypothesis for
parts of which (at least) there is no support from tradition1
itself suggests some doubts about its correctness. Further, it
should be noted that one part of Barnes's hypothesis rests upon
ready and uncritical acceptance of the view that the Elamites and
Medes of verse 2 (cf. xxii 6) were auxiliaries in the Assyrian army.
Boutflower's account of the oracle is not dissimilar to that of
1 E.g. his supposition that verses 1-5 concern a threat to Jerusalem from the
south-west.
72
ISAIAH XXI
Barnes, though there are some important differences. For

Boutflower, the views of Smith and Kleinert were substantially
correct but they were wrong to suggest that the fall of Babylon
mentioned was that of 710. For Sargon, like Cyrus, entered the
city peacefully and, so far from according its gods sacrilegious
treatment (see verse 9), he outdid his predecessors in the costly
offerings that he made at their shrines (see Luckenbill, 11,
no. 184). If then Cyrus's capture of Babylon cannot be the
point of the oracle, neither can Sargon's. Boutflower solves this
problem by supposing that verses 1-5 (verse 2b excepted)
refer to Sargon's Babylonian campaign of 710 while verses 6-10
constitute a prophecy of the destruction of Babylon and its
gods by Sennacherib in 689. With this latter view of verses
6~io Sennacherib's Bavian inscription (cf. Luckenbill, 11,
no. 340) is in agreement: 'The gods dwelling therein . . . my
men took and they broke them in pieces.' Boutflower regards it
as significant that the Assyrian word used for 'broke them in
pieces' (ushabbiruma) is 'radically identical to the mtP of verse 9'
(cf. Erlandsson, apparently independently).
For Boutflower, the all-important half verse 'Go up,
Elam . . J is the voice of God and it is parenthetical. The
prophet is afforded en passant & vision of the fall of Assyria, the
devastator of Babylon. Again the prophecy was an accurate
one, and Boutflower expends considerable effort to show that it
was exactly fulfilled in the circumstances of the fall of Nineveh
in 612 B.C. Thus, it was Cyaxares at the head of a Median
army who dealt the final blow to Nineveh (see Herodotus,
Histories 1 106). Secondly, Boutflower recalls that Nineveh was
threatened by a 'very great multitude of barbarians . . . (that)
had come against (the king of Assyria) from the sea'. 1 In all
probability it is the Elamites attacking Assyria from the Persian
Gulf who are described in this report. Nabonidus's notice (see
A.N.E.T., pp. 3o8f) concerning the fall of Nineveh is cited by
Boutflower as evidence to suggest that, while the Median
forces actually captured Nineveh, Elamite and Babylonian
forces were detained at Assyria's southern border.2 Thus
1 So, according to Boutflower, Abydenus (a disciple of Berosus).
2 No reference is in fact made to the southern border of Assyria in this text, which
states that the Babylonian Nabopolassar declined to take part in the destruction
of Assyrian temples.
73
Isaiah's prophecy was exactly fulfilled. 'Go up, Elam . . .' Elam is to lead the way (by attacking Assyria from the sea);
'besiege, O Media5 - Cyaxares the Median was to carry out the
siege and capture of Nineveh.
Boutflower's treatment of the oracle is consistent with what he
says of it: 'No prophecy was ever better interpreted by its
fulfilment.' His view that Isaiah initially spoke (verse 1-5) of a
campaign against Babylon by Sargon in 710 and then waited
until approximately 689 to announce its fulfilment (verses 6-10)
is just conceivable; indeed it is even possible to agree that it is
consistent with a 'long' (Boutflower's interpretation) wait by
Isaiah on his watchtower; but that Isaiah en passant should
have been vouchsafed a vision (correct to the smallest degree)
of an event (viz. the fall of Nineveh) which took place a hundred
years later is not consistent with what we know of prophecy
nor is it a priori likely. Further, in order to make a case for his
view, Boutflower is obliged to make a number of historical
assumptions for which he has little or no evidence. Boutflower's
treatment of the oracle is thus implausible, and implausible
simply because he insists on finding for every phrase of it a
detailed historical fulfilment over a period of at least one
hundred years.
Amongst more recent commentators on Isaiah, the interpretation of the prophecy offered by E. J. Kissane (i960) agrees in
part with that of Cobb (and it is apparently independent of it).
For Kissane the orthodox critical view is open to question on a
number of grounds; for example, the prophet speaks from the
standpoint of Palestine rather than from that of Babylonia, and
his emotion (verses 3 ff) is quite inexplicable if Babylon alone
is to fall. The apparently contradictory data of the text can be
reconciled by taking verses 1-2 to refer to the oppression of
Israel, the end of the prophecy (as it stands) to the fall of
Babylon at the hands of Cyrus, and the intervening verses to the
long period of waiting for the end of oppression. For Kissane,
however, the prophecy, like that of Isa. xiii, may originally
have concerned not the oppression and subsequent fall of
Babylon, but that of Assyria or its capital Nineveh. The
substitution of Babylon for the original reference probably took
place during the exile, thereby giving further significance and
point to the original prophecy of Isaiah.
74
ISAIAH XXI
In support of his suggestion that Babylon was substituted for

an original reference to Assyria, Kissane draws attention
particularly to Isa. xxiii 13 where 'Assyria' has been supplanted by 'the land of the Chaldeans', and where a scribal note
to this effect is incorporated in the text itself. He cites further
2 Chron. xxxiii 11 and Mic. iv 10 where 'Babylon' may have
been added by a later hand. 1
Kissane's detailed interpretation of Isa. xxi, being very
similar to that of Cobb, is open to similar criticisms. Yet his
more general theory that a prophecy of the eighth century B.C.
may have been adapted in the period of the exile to the
circumstances then obtaining is worthy of serious consideration
as it is compatible with evidence derived from a comparison
between certain of Jeremiah's (sixth-century) foreign oracles
and certain of Isaiah's foreign oracles (see chapters 3 and 4
below).
A radically different view of Isa. xxi 1-10 is taken by the
recent German commentator O. Kaiser (1973; E.T. 1974). For
Kaiser, the poem is best seen on form-critical grounds as a late
work of art composed by a devout Jew 'in the study or temple
cell'. Such a view of the poem accounts for the alleged 'artificiality of its prophetic features, the transitions which can be
observed in the use of traditional themes and its deliberate
but by no means naive dramatic construction'. For Kaiser,
then, the poem is essentially a late apocalyptic work which
portrays 'the imminent fall of the world city'. This fall is
eschatological and with other 'woes of the final age' heralds
the dawning of the age of salvation. The fabric of the oracle,
then, is 'drawn from' the events leading up to the fall of Babylon
in 539, and by reference to them the prophet portrays the
eschatological scheme which he wished to reveal. For Kaiser
such an understanding alone explains why the conclusion of the
poem does not describe jubilation on the part of the prophet at
the proclamation of liberation but rather consists of a ' compassionate address to Zion as a child constantly flayed on the
threshing floor of history'.
Kaiser's assessment of Isa. xxi 1-10 is, on his own admission,
based upon personal impressions. Moreover, it may be said to
1 See Kissane, p. 147.
75
lack precision. He gives no indication of the date of the activity

of his 'Jew in the study or temple cell', and he does not compare
the language and form of the verses of this oracle with those
of the writings widely accepted as apocalyptic in Isa. xxivxxvii (commonly called the Isaiah Apocalypse). Such a comparison may be held to indicate that the prophetic features of
Isa. xxi 1-10 are, by comparison with those of Isa. xxiv-xxvii,
far from artificial. Indeed, the particular references in the
former text (e.g. Elam, Media, Babylon) as well as the firstperson account of the reception of visions and of their effect are
radically different from the generalized and (certainly) artificial
statements of the Isaiah Apocalypse. Further, no comparison of
Isa. xxi 1-10 with other late apocalyptic writings (such as the
book of Daniel) is drawn by Kaiser. The manifestly contrived
style of such works with their future reference is entirely different from the vigorous and dramatic style of Isa. xxi I - I O with its
historical allusions and past reference.
If Kaiser is right in sensing a certain 'artificiality' in Isa.
xxi 1-10, that artificiality, so far from being of a sort with that of
admittedly apocalyptic writings, is capable of the general
interpretation of the passage advanced by Kissane, viz. that the
text has been reworked in order to make it relevant to a later
situation. This explanation has the merit that it is consistent
with the evidence furnished from a consideration of Jeremiah's
foreign oracles (see below, chapters 3 and 4).
Verses 11 & 12
The rabbinic commentators, whether they incline to the view
that verses 11-12 concern Edom or (Arabian) Dumah, do not
appear to record specific traditions or opinions concerning the
historical circumstances to which the oracle refers. It is true that
ibn Janah characterizes Edom as an 'evil kingdom', but that is
typical of the traditional Jewish attitude to Edom from 587 B.C.
onwards. For by their gloating over and taking advantage of
the downfall of Judah at the hands of the Babylonians, the
Edomites earned the undying hatred of the Jews. 1 With
this attitude the same commentator's explanation of Dumah is
1 Cf. Obadiah, Jer. xl 7flfand Ezek. xxv 12-14, Mai. i 2-5, Ps. cxxxvii 7 and Sir.
125-
76
ISAIAH XXI
consistent, viz. that the evil kingdom of Edom is called ' Dumah'
by the prophet because they are a people '(justly) doomed to
(the silence of) destruction5.
By contrast, as a number of modern commentators have
observed, the oracle itself is apparently free from the marks of
hatred toward Edom characteristic of exilic and post-exilic
biblical references to that country. On the basis of this observation some are inclined to the view that the oracle cannot be
dated after 587.
Vitringa considers that the prophecy denotes a time of
calamity experienced by the Edomites in common with the
Jews. While the prophet is able to announce the dawn of relief
for Israel, he proclaims that the night of calamity is to continue
for Edom.1 The time of this common calamity was for Vitringa
when Nebuchadrezzar invaded Palestine and took many
captives from Israel as well as from Edom and other Palestinian
peoples.2
Cheyne (7.5./.), 3 apparently independently, adopts the same
point of view. Referring to the notices in Jeremiah (xxvii 3 and
xlix 7f, 28) which advert to the rebellion of Zedekiah together
with other Palestinian peoples against Babylon and to Babylon's
revenge upon Edom, he concludes that the Dumah oracle
belongs to 589, the date when Nebuchadrezzar moved into
Syria and sent detachments to deal with Palestine and its
neighbours. That Jeremiah's oracle against Edom (xlix) also
mentions Dedan suggests to Cheyne that Isa. xxi 13-15 may
also belong to this time. And with this comparatively late date
for the oracle, the 'Aramaic colouring' of Isa. xxi n f is consistent.4 But for Cheyne it is the equilibrium of feeling in the
prophet that constitutes the all-important indication that the
prophecy predates 587 (the year of the beginning of Judah's
hatred for Edom).
For other modern scholars the relationship of the oracle with
that of verses 1-10 provides a more important criterion for the
1 The motif' Dawn for Israel, night for the gentiles' can be traced as far back as
the Targum and Palestinian Talmud (see ch. i above).
2 Vitringa does not indicate whether he has in mind the invasion of 597 B.C. or
that of 589-586.
3 In P.I. he dates the prophecy in Sargon's reign.
4 For Procksch's criticism of the argument, see below.
77
dating of it. Thus, for example, Marti regards the two oracles
as unmistakably from the same prophet because the questionand-answer form is common to both, as is the distinction
between prophet and watchman; in both oracles the watchman's answer is indecisive; and the prophet's knowledge of the
storms of the Negeb (verse i) is consistent with his dwelling in a
place to which the Edomites could readily be invited to return
with further questions. Since, therefore, the oracle of verses
1 ~io concerns the fall of Babylon at the hands of Cyrus in 539,
the Dumah oracle must belong to the same period (i.e. between
549 and 538). The objection that the note of bitterness and
hostility towards Edom is absent and that therefore the oracle
must predate 586 is contested by Marti with the argument that
the prophet merely displays in both oracles the same 'remarkable objectivity and neutrality'. 2
Marti does not provide further indications for regarding the
prophecy as belonging to the period 549-538. He is content to
follow Duhm's similar supposition and to assert that the news
of the fall of Babylon had prompted the Edomites to ask the
prophet whether their own circumstances would now improve.
For they were torn between the hope of freedom on the one
hand and, on the other, fear of losing business interests with the
demise of their trading partner Babylon. The prophet's answer
is equivocal: just as in verses 1-10 he was unable to see clearly
the results of the fall of Babylon for the Jewish people, so in the
Dumah oracle the fate of Edom is regarded as uncertain. The
morning of hope may appear to be dawning, but it is not certain
that another night will not descend. In such circumstances the
Edomites are encouraged to return at a later time to the
prophet's dwelling place in southern Palestine for further
information.
If Marti is inclined to regard the prophet's somewhat
unsatisfactory answer to his enquiry as plausible, F. Buhl is not.
For him it is not admissible that the prophecy should conclude
with a note of unresolved ambiguity ('unaufgelosten Disharmonie') and consequently the words of the prophet's
1 Cf. the more recent presentation of such arguments by Wildberger, though for
him Duma(t al Jandal), not Edom, is the burden of the prophecy.
2 The phrase is quoted from Duhm. Marti's arguments rather than Duhm's are
discussed here because they are fuller.
78
ISAIAH XXI
answer must be taken to mean ' morning comes even though it is

also night5. (For this translation, see the textual commentary
above.) Buhl is of the opinion that the oracle indicates the
period of Hezekiah's accession to the throne of Judah (shortly
before 722 B.G. -C.A.H., in, p. 388) when a movement towards
emancipation from Assyria was developing among the vassal
states of Palestine. Edom's king Kausmalak was among those
who had paid tribute to Tiglath-pileser (745-728 B.G. - cf.
A.N.E.T., p. 282), and, with Philistia, Judah and Moab, she
was involved shortly after 715 (C.A.H., in, p. 388) in an
abortive rebellion against Sargon, a rebellion which was
crushed when Sargon invaded Ashdod and Gath in 712. The
prophet's advice to the enquiring Edomites at an early stage
in this period, viz. that they should remain in quiet anticipation
of their freedom and that they should ask the prophet further
about it at a later stage, is consistent with Isaiah's fundamental
policy of caution. That Edom had a part in the abortive
rebellion of 715-712, referred to above, indicates that the
Edomites did not in fact follow the prophet's wise advice.1
Procksch is also of the opinion that the Dumah oracle
belongs to this period (721-710 B.C.). The absence of hatred for
Edom rules out a date in exilic or post-exilic times; and the
fact that the oracle is parallel at some points to that concerning Moab in Isa. xv and xvi 1-5 serves to point to 715
when the Assyrian army subdued certain Arab tribes (cf.
C.A.H., in, p. 58, and Luckenbill, 11, no. 17), arriving, so
Procksch assumes, by way of Moab. Thus the Dumah oracle,
like that concerning Moab, reveals one of Judah's neighbours
turning to Jerusalem and receiving a prophetic message for the
future (cf. Isa. xvi). Since Edom and Moab were close neighbours, it is reasonable to suppose that the circumstances reflected
in both oracles are those of the same historical situation. On the
assumption that Moab was subdued by the Assyrians as they
advanced to impose tribute on the Arabian tribes, and further
that Edom managed to avoid the same fate, it is entirely
likely that Edomites enquired urgently of Jerusalem concerning
their own future. For Procksch, the presence of Aramaisms in
the oracle, so far from being evidence of a late date, is a further
1 For a similar view of the date of the prophecy, cf. Gheyne, P.I. In his later work
(I.B.I.) Cheyne dates the oracle in the sixth century (see above).
79
indication of this period. For such dialectal forms are to be

regarded as Edomite1 or, more probably, as reflecting the
international and business Aramaic known to have been
employed at the time from e.g. Isa. xxxvi n .
The arguments concerning the dating of the Dumah oracle
which have so far been outlined may be said to turn upon one of
two considerations, viz. either (a) the absence of anti-Edomite
sentiment in the oracle or (b) the unity of the oracle with that
preceding it. If the first consideration is adopted it follows that
the period following c. 588 is excluded; if the second, the period
immediately before the fall of Babylon in 539 may be positively
indicated. Those further considerations which are advanced in
order to suggest more precisely the date of the oracle (e.g. its
dialectal colouring) are to an extent secondary and dependent
upon the adoption of one of the foregoing.
Of the two considerations, (a) is the more likely, at least
initially, to prove convincing. For while both considerations
are to some extent the product of the personal impressions of the
critic, it is easier to recognize the objectivity of the former than
of the latter.2
A radically different approach to the Dumah oracle is that
adopted by C. Rabin. In Rabin's opinion the language of the
oracle is dialectal and reflects Isaiah's predilection for ' creating
atmosphere' in a foreign oracle by making use of a word or
words in the language of the nation concerned. The foreign
phrase is not necessarily a correct expression in the relevant
language but is coined as a ' stage' expression to indicate to the
listener the particular nation and its language. As an illustration of what he has in mind Rabin cites the 'pseudo-Italianized
English' of ' I no go' which can be readily understood by an
Englishman but also suggests to him that the speaker is Italian.
Secondly, the cryptic brevity and apparent triviality of the
1 This Procksch characterizes as close to 'Aramaic-Arabic'. More recently
Edomite has been described by Naveh as a Ganaanite dialect akin to Hebrew,
Moabite and Ammonite. It is likely that as early as the sixth century B.C. the
script employed in writing it was Aramaic (see Naveh, pp. 27f). Bartlett ('From
Edomites to Nabateans') cites evidence for the use of the Aramaic language and
script as far south as Teima from the period between the sixth and fourth
centuries B.C.
2 For arguments against the unity of xxi 1-12, see Lohmann. He also considers
that the 'objectivity and neutrality' of the Dumah oracle are more apparent
than real.
80
ISAIAH XXI
Dumah oracle suggest to Rabin that it was an intricate piece of

rhetoric which said much more than meets the eye and that, in
the manner of Greek oracles, it revealed its meaning only after
much pondering.1
Rabin considers that the language of the oracle reflects
Arabic and that its real concern is with the Arabian oasis of
Dumat al Jandal to which, he supposes, Edomites fled as
refugees (cf. the similar flight in the Arabian oracle which
follows). In their panic and trepidation they are pictured as
having recourse to the oracle of Dumah, asking its god Wadd
and his priests2 what the future has in store for them. The
prophecy as a whole is translated thus:
Prophecy of Dumah. Someone from Seir calls out to me:
'Watchman/Shammarite, what watch of the night is it,
watchman, what watch of the night is it ?' The
watchman/Shammarite replied: 'The morning has come,
and the night is fulfilled. If ye have sinned/attacked,
pray; if ye (truly) repent, ye will return home.'
Question and answer are ostensibly directed to and given by a
city watchman, and, together with references to day and night,
constitute a highly rhetorical formulation of the Edomites5
concern. The question expresses their anxiety apparently in
terms of the sleepless sufferer impatiently awaiting the morning
watch. The answer 'If ye have sinned (or attacked), pray,
etc.' 3 refers 'with all necessary clarity' to the attacks of Edom
on Israel, or to the injustice and harm done by Edom to Israel
and further indicates that these crimes are the reason for
Edom's disaster.4
In his English article, Rabin makes no reference to the
historical background of the oracle (as understood by him)
other than by referring to Edom's crimes against Israel. The
same reference to Edom's crimes reappears in his Hebrew
article (p. 249) but here he makes further observations about
1 English, pp. 304^ 307.
2 Rabin considers that the use of the word law * watchman' may be an allusion on
Isaiah's part to the Arabian tribe of Shamar who (Rabin conjectures) may have
originally inhabited Al-Jauf (English, pp. 305^ Hebrew, p. 247).
3 For a discussion of the linguistic and philological issues involved in this translation, see ch. 1 above.
4 So English, pp. 3o8ff.
8l
the historical background of the oracle (pp. 25off). First, he

draws attention to Kaufmann's finding that in chapters i-xxxiii
of Isaiah there is no trace of an attitude of vengeance or retribution against the nations, as there is, by contrast, in the chapters
which follow chapter xxxiii. The positive doctrine of these
earlier chapters is that repentance on the part of sinful nations
will effect their forgiveness and this is clearly further exemplified
in the Dumah oracle as Rabin understands it. Consequently,
the oracle can be attributed to the early period of Isaiah.1
Secondly, Rabin observes that the Edomite refugees must
have consulted the oracle of Dumah at a time when it was
functioning normally as an important sanctuary. In this regard
Rabin recalls Esarhaddon's notice (Luckenbill, n, nos. 318a,
536) that his father Sennacherib captured Dumah and brought
its gods to Nineveh. In the same text Esarhaddon goes on to
record that in answer to the request of Hazael, king of the
Arabs, he restored the gods of the Arabs to their sanctuaries
(probably in 676; see C.A.H., in pp. 836). Since, then,
Sennacherib silenced the oracle of Dumah, Rabin argues that
the prophecy cannot be attributed to the years of his reign. Nor
can the oracle belong to the time before Sennacherib, because
that would require it to be dated before Isaiah's time (Hebrew,
p. 250). Consequently Rabin concludes that the oracle belongs
to the reign of Esarhaddon and of course to a period after he
had restored the Arabian gods to their shrines (i.e. after 676;
see C.A.H., m, pp. 836). Rabin reaches this conclusion tentatively and with the explicit admission that Esarhaddon has left
no record of an expedition to Edom (Hebrew, p. 251).
There are many points in Rabin's argument that are open to
challenge - even on the assumption that his translation of the
text is correct and that the prophecy refers to the shrine of Wadd
at Dumat al Jandal. For example, why should not the prophet
refer to Edomite refugees passing through Dumah in Sennacherib's time before he invaded the place ? All that is required
of the prophecy is that it should depict Edomite refugees as
having recourse to the oracle of Dumah; a prophecy which (in
Rabin's own words) is 'highly rhetorical' does not in the nature
of the case have to be a vaticinium ex eventu, an accurate chronicle
1 Hebrew, p. 250.
82
ISAIAH XXI
of precise historical events. And why, for that matter, is the

time before Sennacherib excluded? Rabin suggests that it is
impossible on the grounds that it is 'before the time of Isaiah'
to whom, on the grounds adduced by Kaufmann, he is inclined
to attribute the prophecy. But Isaianic oracles clearly predate
the reign of Sennacherib (e.g. chapter xx explicitly mentions
Sargon), and the year of Isaiah's call (Isaiah vi i) is probably
to be dated some thirty-five years before the accession of
Sennacherib. Indeed such considerations render it very unlikely
that Isaiah was active in Esarhaddon's reign (681-669 B.C.), as
Rabin suggests; for then he would most probably have been
over eighty years old. At any rate an oracle referring to events
in Esarhaddon's reign is unlikely to belong, in Rabin's words,
'to the early period of Isaiah'.
Such matters, however, are trivial compared with one particular major discrepancy in Rabin's argument. How can an
oracle whose purpose is (in Rabin's opinion) to invite Edomites
to repent of their conspiracy against and attacks upon Israel
be attributed 'to the early period of Isaiah', and, even tentatively, be dated in the time of Esarhaddon ? If Rabin refers
to Edomite treachery other than that of 590-586,x he does not
make this plain; on the contrary, he appeals explicitly to the
general biblical view that Edom incurred guilt by reason of
her attacks upon and conspiracy against Israel (see Hebrew,
P. 249)If the recognition of this discrepancy occasions serious doubts
about Rabin's historical observations, it has repercussions also
for the author's philological arguments. Rabin suggests, for
example, that nsn (first occurrence) has the sense 'to sin or
attack' because such a meaning fits the context. But if the
oracle does not contain a reference to Edomite treachery
against Israel, the meanings 'sin' for nS71 (first occurrence),
'pray' for nSD (second occurrence), and 'repent' for mtP find
no place in it.
It is not here necessary to enter into the details of Rabin's
interesting account of the history of the shrine at Dumat al
Jandal. What does concern us is the question of the antecedent
likelihood of an Israelite prophet (whether Isaiah or a later
1 Such as, for example, that alluded to in 2 Chron. xxviii 17.
83
author) making use of an oracular answer of an obscure foreign

deity in order to convey to Israel the word of Yahweh concerning Edom. The question is equally pertinent whether the
prophet's allusion to that oracle relates to an historical situation
known to him or whether it arises solely from his prophetic
imagination. The answer to the question must surely be that it
is unlikely in the extreme. For no instance even remotely
comparable can be cited from the prophetic literature of the
O.T. And it is not surprising that none is in fact cited by
Rabin.
What may be characterized as a form-critical account of the
Dumah oracle is given by P. Lohmann and by means of this
approach he posits a date for the oracle at a time towards the
end of the exile.
Lohmann argues that the oracle should not be attributed to
the author of verses I - I O . Indeed, so far from being properly
an oracle, it is a secular poem which has later been adapted
for use as a foreign oracle and, as such, it has been tacked on to
the Babylonian oracle of verses I - I O . The poem underlying
verses i if is of a sort with the vintage songs alluded to in Isa.
xvi 10 and Jer. xxv 30 and with the harlot's song quoted in
Isa. xxiii 16. The watchman's song constitutes a joke ('Scherzliedchen') current in the ancient Near East and known to the
prophet and his hearers. The song may, indeed, be very old
and, as is the case with the Song of Songs, its Aramaisms need
not tell against this view. Such poems are by their very nature
anonymous and timeless.
Lohmann conjectures that, in its original form, the song
began with the word 21???? which was subsequently displaced
and wrongly affixed to the beginning of verse 13.1 Lohmann
further conjectures that ^S$? ('from the gates') stood where
now TSNWD ('from Seir') is read. The term *ip denotes a city
watchman2 who is questioned by those waiting impatiently
outside the city gates as to how late in the night it is and how
long there is to wait until dawn and the opening of the gates.
The joke is occasioned by the mischievous and short-tempered
1 He notes that the LXX at verse 13 does not have a word corresponding to the
M.T.'s m3 and he further observes that the word 3"un in the M.T. is superfluous.
2 Lohmann compares Song of Songs iii 3, Pss. cxxi 3 and cxxx 6.
84
ISAIAH XXI
reply of the watchman, who tells them in effect to go home and

not to bother him further.1
The change from 15^13 to TWfi was, however, no accident
but was effected by the creative talent of a prophet. Thereby
the poem became an oracle expressing hatred for the treacherous Edomites, and, when appended to Isa. xxi 1-10, was
understood as a heathen oracle indicating that at the end of the
exile day would dawn for Israel but night would descend upon
Edom. What is intended by the last words of verse 12 (fpsnn DK
etc.) within the heathen oracle is, to Lohmann, obscure and he
is content to regard them simply as a survival of the original
watchman's song.
To Lohmann it seems probable that the prophet who incorporated the watchman's song and made of it a foreign oracle
was none other than the composer of verses 1-10. For both
oracles are concerned with the same historical situation and
they are juxtaposed in the literary tradition of Isaiah. In the
Babylonian oracle there are expressed hatred and threat against
Babylon and hope for the Jews; in the Edom oracle, threat and
mockery are directed against Edom but hope is expressed for
Judah. The difference of expression (e.g. the use of the unparalleled *IB 'watchman' for the prophet in verses 11, 12)
can be accounted for by the recognition that verses 1-10
represent a prophetic composition but that verses 1 if are a reworking of an existing secular composition.
Lohmann's treatment of the Dumah oracle has the considerable merit that it is able to give an account not merely of
the meaning of the words of verses 1 if but also of the form of the
composition. His comparison with the use made of the harlot's
song in Isa. xxiii 16 is particularly telling, and militates against
Galling's particular criticisms of his views. For Galling asserts
that it is antecedently improbable that a prophet should make
use of such a drinking song to convey a message on so important
a question as the dawning of salvation for Israel. Yet Galling
appears to ignore the evidence of Isa. xxiii 16 where the harlot's
song is used by the prophet to depict the fate of Tyre.
1 Lohmann is not inclined to decide between Buhl's translation of the reply
('morning comes while it is still night') and that of Duhm ('morning comes and
also [i.e. at the same time] night'); whichever is adopted, the effect, for him, is
the same.
85
Lohmann's account, moreover, has the merit that the

'allegorical' meaning which he detects (i.e. 'dawn for Israel,
night for Edom') is consistent with the likely feelings of the
Jews in exilic times, and, as has been mentioned above (p. 45),
it is actually found in Jewish tradition as far back as the
Targum and the Palestinian Talmud.
On the other hand, the rigidity of Lohmann's distinction
between secular poem and prophetic oracle has been pressed
further than is warranted by the evidence.1 For example, his
argument that the words must have been originally a secular
poem because there is no explicit mention of divine inspiration
is hardly convincing, for a number of heathen oracles lack such
notices (see e.g. Isa. xvf). Again his insistence that n&tP is not an
O.T. designation for a prophet is not convincing; for, as has
been observed above, to assert that "lfctp is not an O.T. designation for a prophet does not amount to grounds for denying that
a prophet may have so described himself in a particular oracle.
A further weakness of Lohmann's case lies in the fact that he is
unable to give an account of the force of the last words of the
oracle qua oracle; he has to content himself with giving a
translation of the words appropriate to the poem's original
situation: 'If you wish to ask, ask; go now and come back.'
If he is unable to determine the sense of these words as part of
the foreign oracle, his case for a distinction between secular
song and foreign oracle is further weakened.
It is, however, the rigidity of Lohmann's distinction that
enables him to find in verses 11 f mockery of Edom and consequently to date the oracle (or adapted song) at a time after
587 and towards the end of the exile. But it may be urged
equally well that the words of verses 11 f are more simply (and
therefore more plausibly) to be interpreted as a poetical expression of the prophet's teaching concerning Edom and that he
has had recourse to the metaphor of the city watchman just as
he has depicted his function in the previous oracle in terms of a
look-out (nDXfc). That the words of the oracle do not apparently
give expression to anti-Edomite feeling remains prima facie an
indication that the prophecy is pre-exilic.
Lohmann's further conclusion that the oracle is to be dated
1 His account (all-important to his theory) of the word -pywa rests purely upon
conjecture.
86
ISAIAH XXI
in late exilic times because the prophet who adapted and made
use of it is the author of verses I - I O depends, of course, on his
acceptance of the view that the latter prophecy is also to be
dated in the same period. That, however, is not evident, and
has not been conclusively demonstrated.
If for Lohmann form-critical methods illuminate the Dumah
oracle, for K. Galling it is Babylonian historical records,
and more specifically the recently published Harran inscriptions of the last Babylonian king, Nabonidus. 1 Noting that the
Dumah oracle, like that which follows it, is concerned with the
inhabitants of the deserts to the south-east of Judah he supposes
that the whole of Isa. xxi reflects the historical situation following Nabonidus's return to Babylon after his campaign and
subsequent ten-year sojourn in the oases of North Arabia. For
Galling, Nabonidus's return to Babylon may be dated precisely
in the years 546-545 B.C.2
There has been considerable discussion about the reasons for
Nabonidus's protracted stay in North Arabia, but recent
commentators favour the view that he went there on a military
expedition with the ultimate intention of securing the trade
route which ran along the eastern shore of the Red Sea.3
Galling is content to recall, however, that Nabonidus himself
mentions as the reason for his withdrawal from Babylon the
avowed hostility of the great cities of Babylon and their hierarchies towards him and his religious policies.4 At any rate,
accompanied by his troops from 'Akkad and the Hatti-Land' 5
Nabonidus set forth 'on the road to Tema', Dadanu, Padakku
. . . and as far as Iatribu', subduing the oases of the Hijaz as
far as Iatrib (Medina) and investing them with garrisons. At
Teima he appears to have made his headquarters, building
there a residence for himself and a shrine for his beloved god
1 For these texts, see Gadd.
2 The latter date is computed by excluding the king's year of accession. Gadd (p.
75) considers that it is not possible to be precise about the dates of the Arabian
sojourn.
3 See S. Smith, Isaiah, pp. 38ff, quoted by Lindsay, p. 37; W. F. Albright in
B.A.S.O.R. 82 (1941), pp. 1 if, and Gadd, pp. 88f. For an earlier discussion, see
R. P. Dougherty in J.A.O.S., 1922, pp. 3O5ff.
4 Cf. Gadd, p. 88. For Nabonidus's actual words, see ibid. pp. 568*.
5 The former term indicates native Babylonian troops; the latter, contingents from
Syro-Palestine. Gadd supposes that in fact the Babylonian troops were sent back
to Babylon under Belshazzar, while it was Syro-Palestinian troops (including
Jews) who accompanied Nabonidus to Arabia (pp. 850).
87
Sin. Thereafter for ten years he moved from place to place,

reviewing his garrisons, putting down rebellions on the part of
the Arabs, and enjoying the benefits of'wealth and abundance'.
In regard to his suzerainty over particular Arab peoples, two
recently discovered Thamudic inscriptions which mention ' the
spear of the king of Babylon' and 'the War of Dedan', indicate
' the vivid impression made upon the inhabitants of the area
already literate enough to record their recollections of so stirring
an episode' (Gadd, p. 78). For Galling, mention of the War of
Dedan may indicate either the original subjugation of that
oasis, or the battles following a rebellion on its part.
It is, however, the triumphal return of Nabonidus to Babylon
which is, in Galling's opinion, of such great importance to
Isaiah xxi. Apparently, circumstances in Babylon in 545 were
more favourable to Nabonidus and he was able to return to
his capital and to resume work on his cherished plan of restoring the Sin temple at Harran. When he left Teima and as
he travelled the desert road, he records that he received the
homage of neighbouring kings (Gadd, p. 63). Yet, although the
texts are naturally silent on the subject, Galling considers it
reasonable to suppose that Nabonidus's departure must have
raised in the hearts of the peoples of the west both turmoil and
hope. While his departure did not necessarily entail the
immediate decline of Babylonian dominance, it may have been
thought to involve eventually substantial change for the area.
Would the remaining Babylonian troops be able to contain the
situation? Would all the various Arab tribes remain loyal?
Would fighting break out among them? Such questions must
have achieved new importance and significance as the king
left the deserts of Arabia for his Babylonian homeland.1
Galling's understanding of Isa. xxi 1-10 has been considered
above (pp. 68ff) and it is there shown to be essentially a
variant of the orthodox critical view of the oracle. In Galling's
opinion its concern is with the imminent fall of Babylon at the
hands of the Persians - the Elamites and Medes of verse 2.
Because, however, the oracle is followed by two prophecies2
concerned ex hypothesi with the turbulence of desert peoples
1 Galling's views are tentatively accepted by Wildberger.
2 The theory is unaffected by the consideration that it was a redactor rather than
a common author who combined the oracles.
88
ISAIAH
XXI
south-east of Judah following Nabonidus's departure in 546/5,

Galling believes it likely that verses 1-10 reflect the same
situation and should therefore be dated between 545 and 540.
As has been stated above, the orthodox critical view of the
oracle stands or falls with the identification of Elam and Media
in verse 2 with the united empire of Cyrus. In regard to Galling's
particular further deduction, viz. that the oracle reflects the
general disquiet in the west following Nabonidus's departure
from Arabia, a particular question immediately presents itself:
why does the oracle contain no mention of nor allusion to that
departure of Nabonidus if it was the all-important cue to
peoples of the west ?
Where the Dumah oracle is concerned, Galling believes that
the title 'Dumah' must refer to a specific place (as does e.g.
Isa. xxii 1) and that that place is Dumat al Jandal in Arabia.1
He further suspects that the reading of the minor Greek
versions (for which see above) presupposes the original sense:
'summon (or someone is summoning)2 to me those who have
fled from Seir'. Just as following 545 Dedanites found shelter in
the oasis of Teima (verses 136), so here refugees from Dumah
are pictured as fleeing via Seir to West Jordanian Edom and as
having recourse to a Jewish prophet in order to ascertain their
destiny. That they should do so is rendered less surprising by the
further assumption that the refugees were in fact Jews, and that
assumption in turn is rendered possible for Galling by the
consideration that Jews are likely to have been amongst the
soldiers and settlers introduced to the North Arabian oases by
Nabonidus (see Gadd, pp. 856). At any rate, the likely tension
arising after 545 between Babylonian troops and settlers on the
one hand, and freedom-seeking Arabs on the other, may have
furnished the occasion for such a flight of Jews from Dumah
and for their anxious questioning of a Judaean prophet. His
reply according to Galling could, in the circumstances, be
nothing but enigmatic. The night of misery remained, but dawn
would follow at some time in the future known only to Yahweh.
Return for further questioning might find the prophet better
able to furnish clearer answers.
1 He does not believe that the LXX's 'Edom' reflects a text preferable to the
M.T.
2 Reading KOcAei
89
Any account of so short and terse an oracle as the Dumah

oracle must make assumptions. But the number of assumptions
made by Galling are surely too numerous and tenuously based
to command ready agreement. Above all, his exegesis of the
oracle is substantially dependent upon his suggestion that the
minor Greek versions indicate the true text and meaning of the
oracle. It is very unlikely, however, that those versions bear
witness to a Hebrew text divergent from the M.T. On the
contrary, as has been argued above, their rendering bears all
the marks of an attempt to clarify and elucidate the rendering
of the LXX, with the wording of which it is substantially
identical.
Another attempt to link the Dumah oracle with Nabonidus's
Arabian campaign is that of J. Lindsay. He is not concerned to
emend or alter the M.T., and he is content to reproduce without
further comment the translation of the R.S.V. What is significant to him, however, is the word ' Dumah' in the title of the
oracle, and the fact that it is 'in parallelism' with 'Seir'. The
two words juxtaposed indicate that Edom is meant; and by
Edom, Lindsay argues, the author seeks to depict a greater
Edom, i.e. an area comprising both Edom itself and the area of
its trade influence, extending as far east as Dumah and as far
south-east as Dedan. That such a greater Edom was recognized
by the Jews seems to be confirmed by such texts as Ezek. xxv 13
(which mentions Teman and Dedan as distant places in Edom
against which Yahweh would stretch forth his hand). Similarly
in Jer. xlix 7 Dedan is addressed by name within the context of
an oracle addressed to Edom.
That Nabonidus's Arabian campaign of 552 affected Edom
as well as the Arabian oases themselves may be inferred from
the Nabonidus Chronicle (A.N.E.T., p. 305) which states that
in the course of the campaign it was necessary to reduce a
certain city whose name ends in ' dummu5; this is now generally
taken to denote Edom rather than Dumah. 1 For Lindsay, then,
the evidence of the Babylonian texts concerning Nabonidus's
itinerary and campaign reflect, as do the biblical texts cited
1 See S. Smith against W. F. Albright in J.R.A.S., 1925, pp. 2 9 3 ^ 5o8ff; A. K.
Grayson, Assyrian and Babylonian Chronicles (New York, 1975), p. 282; W. G.
Lambert in Proceedings of the 5th Seminar for Arabian Studies, Oxford 1971 (London,
1972), p. 55-
go
ISAIAH
xxi
above, the reality of a greater Edom. This, he submits, is the

subject of the oracle of Isaiah xxi i if.
While the historical and geographical considerations advanced by Lindsay seem to be unexceptionable, his treatment
of the Dumah oracle is far from satisfactory. On the one hand
he accepts Lehman's view (for which, see p. 39 n. 2 above) that
the original reading of the text of the title was ' Edom' and not
'Dumah'. On the other, the drift of his argument requires that
the oracle contain in parallelism (his word) Seir and Dumah
(not Edom); for in that way, rather as the Edom oracle of
Jer. xlix makes mention of Dedan, Isaiah's oracle can be held to
reflect the greater Edom whose probable existence is attested
on other grounds. But either 'Edom' or 'Dumah' was the
original reading of the title. Lindsay must decide; he cannot
have it both ways.
Similar inconsistency in Lindsay's argument may be detected
in his assertion that Lehman's finding (viz. that 'Edom' was the
original reading of the title) 'reinforces the opinion long held by
a majority of scholars that the M.T. reflects the close association
in the minds of Isaiah's contemporaries between Edom and
Dumah'. It is hard to see how this could be the case if Dumah
were not in fact mentioned either in the oracle or its title.
Furthermore, if the oracle refers to Nabonidus's overthrow of
Adummu/Edom, as Lindsay suggests, it is hard to see how the
M.T. can be held to reflect any association whatsoever in the
minds of Isaiah's contemporaries.1
Kaiser's view of the Dumah oracle, though expressed very
tentatively, is of a sort with his view of the oracle of verses 1-10
and merits similar criticisms. He supposes that the poem,
originally without title and introductory words (' One is calling
to me from Seir'), was appended to verses 1-10 as a word of
comfort to the faithful concerning the dawn of salvation following the fall of the world capital (verse 9); ' Morning is coming
though it is still night.' This view, he says, is ' the least artificial
because it avoids all speculation as to why a prophet or writer
of Judah should suddenly be interrogated by Edomites or be
concerned with what happens to them'. It is difficult to be sure
what Kaiser means by his phrase 'least artificial' ('ungekun1 If by the use of the name Isaiah he means to suggest the later author of this
oracle, he does not make his intention clear.
91
steltste'). It seems to me, however, that his own treatment is in

fact very highly artificial and that one aspect of that artificiality
is to be seen in his arbitrary and speculative deletion of the
introductory words of the oracle.
Kaiser's alternative suggestion that the oracle is a late
redactional addition made in order to complete a full complement of foreign oracles seems to rest ultimately upon the view
that its Aramaisms indicate at least an exilic (and probably a
post-exilic) date. Procksch's observations on the early use of
Aramaic have been noted above and in the light of them we
may discount the most important element in Kaiser's argument.
What remains is purely speculative.
Further observations about the historical background to the
Dumah oracle are made in connexion with the Arabian oracle;
see below.
Verses 13-15
Apart from the Talmudic tradition1 (which is repeated by
Saadya and Rashi) that the oracle alludes to an occasion
following the fall of Jerusalem in 587 B.C. when Arabs failed to
offer proper hospitality to refugees of the Jewish priesthood,2
the rabbinic commentators do not offer much in the way of
traditions concerning its date and circumstances. Ibn Ezra,
commenting on the mention of the sword in verse 15, states
simply that it is the king of Assyria who has precipitated the
flight of the refugees.
Modern commentators are divided on the question of the
dating of the oracle. Broadly speaking three periods may
be said to have attracted support amongst them; 3 first,
Nebuchadrezzar's campaign, usually dated in 589* which,
according to Jer. xlix, terrified the inhabitants of Dedan
and Kedar as well as those of Edom; secondly, the period of
1 See above, p. 57.
2 Rashi (on verse 13) also records a similar failure on the part of Arabs to show
pity to Jews - this time 'when Assyria exiled my people'.
3 Kaiser is an exception in that he considers that the oracle refers to ' the eschatalogical threat from unnamed enemies from the North'. This represents an extension of his views of the other oracles in the chapter. As no further arguments in
their favour are adduced in connexion with this oracle, no further arguments
against them are here required or given.
4 E.g. Rosenmiiller and Gheyne, I.B.I.
92
ISAIAH XXI
the fall of Babylon in 539;x thirdly, the Assyrian period,2 and

more particularly the Arabian campaigns of Sargon (715)
and Sennacherib (some time before 691).3
The criteria by which any one of the periods is selected are
fundamentally literary; that is to say, the view taken by critics
of the relationship of the Arabian oracle to other oracles in the
chapter or elsewhere in the bible largely determines their view
of its date (so e.g. Wildberger).
Nebuchadrezzar's Palestinian campaign of 589 B.C.
Cheyne (I.B.I.) is inclined to emphasize the fact that the oracle
(verses 13-17) is juxtaposed to a prophecy apparently concerned with Edom. In this connexion, like Rosenmiiller, he
notes that Jer. xlix also mentions, along with Edom, the tribes
of Dedan and Kedar as fearing the chastening hand of Nebuchadrezzar. He concludes that Isa. xxi n - 1 7 , like Jer. xlix, refers
to the situation following Zedekiah's revolt against Babylon,
when, in 589 B.C., Nebuchadrezzar advanced into Syria/
Palestine.4 To the possible objection that there is in the Dumah
oracle no trace of the anti-Edomite feeling which one would
expect at this time, Cheyne is content to emphasize that Edom
was certainly an ally of Zedekiah of Judah shortly before 587,5
and that the Dumah oracle is likely to be pre- rather than
post-exilic because it displays an * equilibrium of feeling . . .
which, if it prevents cordiality, not less forbids positive hostility'.
Cheyne's arguments are open to two objections. First, the
1 E.g. Duhm, Marti, Gray and Galling.
2 E.g. Cheyne, P.I.; Delitzsch; D.K.; Procksch; Kissane.
3 Erlandsson (pp. 94ff) views the oracle in the light of the Arabian campaigns.
However, he adduces no exegetical or literary critical arguments in favour of
doing so. He is content to quote with favour Procksch's statement to the effect
that in the book of Isaiah the Assyrians are the foe par excellence (' bieten sich
im Jesajabuch wider die Assyrer an5).
4 Cf. C.A.H., in, pp. 2i3f. We have no indication from Babylonian records of an
attack by Nebuchadrezzar upon Edom, though Josephus, citing Berosus, refers
to a campaign undertaken by that king against the Arabs (Contra Apion 1 19).
The general reliability of this notice seems now to be confirmed by Babylonian
records; see Wiseman, Chronicles, p. 70; cf. pp. 3iff. Here it is stated that a
campaign of plunder against the Arabs was undertaken in 599-598 - in the
sixth year of Nebuchadrezzar. It is not essential to Cheyne's argument that
Nebuchadrezzar should have invaded Edom and Arabia in 589, for he is not
committed to the view that the oracle is a vaticinium ex eventu; it is enough that
these countries should have been threatened at this time.
5 See, with C.A.H., m, Jer. xxviif.
93
5
weakness of the argument that the 'Aramaic colouring of Isa.

xxi 11-17 precludes a date earlier than Nebuchadrezzar's time
has been exposed by Procksch.1 Secondly, where Jer. xlix is concerned, the balance of scholarly opinion may be held to favour
the view that this chapter, containing as it does material found
also in e.g. Obadiah, is a composite work2 (as is Jer. xlviii)
which contains material found in Isa. xvf. On the further
question whether these oracles of Jeremiah are dependent upon
Obadiah and Isaiah, opinion is divided. It is possible to argue
that the authors of Jer. xlviii f on the one hand, and of the
relevant chapters of Isaiah and of Obadiah on the other, have
all made use of older prophetic material.3 At any rate, it is clear
that Jer. xlix cannot be cited as an independent and contemporary witness in support of the supposition that Isa. xxi 13-17
is concerned with a threat by Nebuchadrezzar to Judah's
southern neighbours.4 For, apart from the question of the
contrived character of Jer. xlix, Isa. xxi 13-17 makes no
mention of Nebuchadrezzar, whereas Jer. xlix (at least in the
case of Kedar) explicitly does so.
To return for a moment to the question of the relationship
between Isaiah's foreign oracles and those of Jer. xlviii f, certain
considerations may suggest that the balance of probability
favours the literary dependence of the latter upon the former.
Thus, in Jer. xlix there are verses (8a and 30a) similar in
sentiment if not identical in vocabulary to Isa. xxi 13 ; 5 there is
also the close association of Dedan with Edom in both oracles.
Further, there are verses in Jer. xlix which are word for word
common to verses in Obadiah. 6 These considerations suggest
1 See pp. 78f. above. D.K. also discount such arguments.
2 Whether or not there is a kernel of authentic words of Jeremiah is disputed; cf.
the commentaries of e.g. Volz and Rudolph.
3 See e.g. J. Bright, Jeremiah. Cf. on Isaiah xvf esp. D.K.
4 Cheyne, in fact, introduces his argument with an acknowledgement that he is
assuming Jeremiah's authorship.
5 The argument is not affected by Bach's attempt to attribute all such statements
to the formal genre 'Aufforderung zum Flucht'. For both authors have used such
statements (whatever their form) specifically and strikingly in connexion with
the Dedanites.
6 For recent discussion of the relationship between Obadiah and Jer. xlix, see
Rudolph in K.A.T. xm 2, p. 297, and H. W. Wolff, pp. 2off. Rudolph takes the
view that Obadiah is earlier than Jer. xlix; but both he and Wolff consider it
likely that the two are dependent upon an earlier common source (for Wolff
that source is specifically oral tradition).
94
ISAIAH
XXI
strongly that Jer. xlix is a composite work, that it is dependent

upon Obadiah and that it alludes to Isa. xxi (cf. Delitzsch (i)).
With this conclusion two other considerations appear to agree;
first, Isa. xxi i3f is manifestly a coherent unity and consequently
Jer. xlix, which on other grounds is judged a composite work, is
best seen as a conscious reflexion of it. Secondly, scholars are
generally inclined to accept that Jer. xlviii is dependent upon
Isa. xvf.1 If, then, Jer. xlviii is dependent upon Isa. xvf, it is not
unlikely that Jer. xlix reflects Isa. xxi i3f.
Because the argument of Rosenmuller and Cheyne rests upon
a comparison of Isaiah's Arabian oracle with Jer. xlix as an
independent and contemporary witness to reactions amongst
Judah's southern neighbours c. 589 B.C., it must be judged
implausible. On the contrary, examination of their argument
and of the nature of Jer. xlix indicates that Isa. xxi antedates
this work and provides elements which its author was inclined
to use (see chapter 3 below).
The fall of Babylon in 539 B.C.
Duhm, Marti and Gray may be regarded as typical of those
who adopt the theory that the Arabian oracle (i.e. verses 13-15
alone) belongs to the period of the fall of Babylon in 539.
For all three scholars considerable importance attaches to the
judgement that the oracle was written by the author of verses
1-10. For Duhm this is undoubtedly the case; both oracles
present vivid pictures, and here the prophet relates a vision of
the Dedanites pursued by Persian horsemen. For Marti, metre
and structure are the same in verses 1-10 and 13-15; consequently the two oracles are to be attributed to the same
author, though possibly the Arabian oracle is slightly later.
The oracle follows admirably upon one which recounts the
fall of Babylon and it indicates that the seer did not expect that
the Persians would deal favourably with those peoples who had
hitherto been subject to Babylon.2 Marti also approves of
Duhm's suggestion that the author is likely to have dwelt in
1 See the commentaries of Duhm, Rudolph and Weiser. For a judicious statement
of this case and for further references, see Wildberger, pp. 6o5ff.
2 Wildberger (p. 801) attributes to Marti the view that the Persian army sought
to secure the caravan routes of Arabia for the empire. This, however, Marti does
not actually say.
95
southern Palestine and as a consequence was naturally sympathetic to Judah's Arab neighbours.
Gray's approach is similar to that of Duhm and Marti,
though it is very much more tentative. He is inclined to suppose
that the piece is from the same hand as verses I - I O and that it
probably relates the contents of a vision rather than an actual
historical flight. For Gray the Dedanites are likely to be fleeing
from Babylon. Arriving with the expectation that they would
conduct their usual business affairs they find instead Babylon
threatened by the Persians and consequently they take to the
caravan routes in flight.
Two modifications of this general approach are proposed by
Galling and Wildberger. Both scholars connect the Arabian
oracle closely with the preceding Babylonian one, but Galling
considers that the very great distance (800 km) between Babylon and Teima precludes the generally accepted notion that
Persian mounted detachments1 should have pursued the
Dedanites over such a distance through the desert immediately following the fall of Babylon. The danger zone for the
Dedanites, if they are pictured as originally in and around
Babylon, is Babylon and not the far-off region of Teima.
Further, even if so long a flight is in the prophet's mind, why is
his advice concerned only with the later stage of it (i.e. that
around Teima) ? For Galling, it is much more likely that the
prophecy (like those that precede it) is to be dated soon after
Nabonidus's departure from Teima in 545. In this situation, the
Babylonian garrisons left behind by Nabonidus are likely to
have been involved in skirmishes with the native Arabs, who,
following that monarch's departure, may be presumed to have
made bids for freedom. For Galling, then, the threat from which
the Dedanites are pictured as fleeing consists of Babylonians
(or their allies) rather than Persians, and their flight is
confined from beginning to end to the deserts around Teima.
Galling's criticisms of what he regards as the generally
accepted view of the Arabian oracle (i.e. what has been
described above as the view of Duhm, Marti and Gray) are
cogent, and, in their own right, worthy of acceptance. On the
other hand, his own understanding of the background to the
1 He notes that nothing in the text suggests such mounted troops.
96
ISAIAH XXI
oracle rests upon various inferences and above all upon the
assumption that the oracles of the chapter, like that of verses
I - I O , concern the (imminent) fall of Babylon in 539. While,
then, his criticisms of Duhm, Marti and Gray's view of the
oracle are in their own right convincing, his view, like theirs, is
ultimately dependent upon a particular understanding of the
historical background to Isa. xxi 1-10. If doubts attach to that
understanding, Galling's subsequent inferences, however plausible in themselves, are likely to prove erroneous.
Wildberger is content merely to quote the views of Marti and
Galling on the Arabian oracle without attempting to judge
between them. However, he suggests as a further possibility that
the oracle may reflect internal conflicts between the Arabian
tribes, and more particularly an attempt by the tribe of Kedar
to control and to dominate the trade routes of the area; such an
assumption would explain why verses i6f (which concern
Kedar) were considered a necessary addition by the editor of
the chapter. At any rate, as Wildberger inclines to the view that
both the Dumah and the Arabian oracles are best understood
as belonging to the period after the fall of Babylon, the criticisms
of Galling's views mentioned above apply mutatis mutandis with
equal force to his.
The Assyrian period
Those commentators1 who understand the Arabian oracle to
have as its background the period of the Assyrian attacks upon
North Arabia do so because they believe that it is closely related
to the Dumah (Edom) oracle which precedes it and with the
Kedar oracle (verses i6f) which follows it. Where the latter
text is concerned, comparison is made with Isa. xvi 13f where
virtually identical words are used as the substance of an
additional revelation about Moab. The words as they occur in
Isa. xvi i3f are regarded as genuinely Isaianic by all these
commentators, and consequently they incline to the view that
the words of xxi i6f are also genuine. In both cases the texts
have the particular function of elucidating and determining
more accurately the contents of the oracles which precede
them, whether those oracles are appropriated by Isaiah from
1 E.g. Cheyne, P.I.; Delitzsch; D.K.; Procksch; Kissane.
97
older prophetic material (e.g. Hitzig and D.K.) or whether they

are merely earlier poems of the prophet subsequently in need
of such elucidation (e.g. Delitzsch and Procksch).
For commentators of this school, therefore, the Moab oracles
of Isa. xvf and the Dumah/Arabian oracles of Isa. xxi i iff have
as their background the years 720-710 B.C. and Sargon IPs
campaigns against South Palestine and North Arabia which can
be dated in this period (C.A.H. m, p. 58). Where the words of
xxi i6f are concerned, two of the commentators provide
particular and differing accounts. First, Dillmann-Kittel
regard them as the contents of a revelation to Isaiah himself in
which the contents of the Arabian oracle (earlier appropriated
and reproduced by Isaiah) were confirmed, the only difference
being that Isaiah's own revelation spoke of Kedar (i.e. the main
tribe of North Arabia beyond Edom), whereas it was the
Dedanites who were mentioned in the oracle which he had
earlier appropriated. Secondly, Procksch regards as significant
the fact that the Dedanites are the object of Isaiah's sympathy,
while the tribe of Kedar is the object of his threat. He suggests,
therefore, that the tribe of Kedar were allies of the Assyrians1
and that it was they in particular who threatened the Dedanites. Isaiah's prophecies indicate his hope that the Dedanites
would shortly be freed from the particular menace facing them,
as so many small nations would be freed from that of the
Assyrians themselves.
Commentators of this school, who, by contrast, believe that
Isa. xxi 1-10 has as its background the period of the fall of
Babylon in 539, are inclined to emphasize what they regard as
the considerable differences between that oracle and those
which follow it. Procksch, for example, is at pains to discount
the arguments presented by those who see a unity between
verses 1-10 and the oracles that follow it. However, the
arguments that he in turn produces are unconvincing in
character. For example, he notes that revelation comes directly
to the prophet in verse 11 (^S) but through the prophet's
alter ego in verses 1-10; or again verses 1-10 use the word ri
while the prophet is called *)&tP in the Dumah oracle.
1 He compares Isa. xxii 6 where, he claims, other allies of Assyria are mentioned
by name. Some doubts, however, attach to this interpretation of the verse; see
below.
98
ISAIAH XXI
Some considerable advantages attach to the view that the

Arabian oracle, like the Dumah oracle which precedes it, belongs to the Assyrian period. For the judgement, if it can be
sustained by plausible arguments, may in turn have implications for the understanding of the oracle of verses I - I O , and
render possible the view that chapter xxi as a whole (or at least
an earlier form of it), like chapters xx and xxii before and after
it, is to be dated in the last years of the eighth century B.C.
Doubt, however, attaches to one particular link in the
argument presented by those scholars who advocate a date in
Assyrian times for the Arabian oracle. It is the judgement that
verses 16f are genuine (if later) words of Isaiah and the consequent conclusion that the Arabian oracle is to be dated in
Assyrian times. The difficulty arises that the initial judgement
is entirely unsupported (and perhaps unsupportable) by objective criteria. Indeed another and quite different judgment of
the verses concerned has been made, viz. that, being phrased in
prose and having a somewhat stereotyped character, they are
likely to be later editorial additions to earlier prophecies
(whether or not Isaianic).1 Moreover, it is possible to argue that
verses i6f, being in the form of a judgement, are at odds with the
apparently sympathetic attitude displayed towards the Arabs
in the Arabian oracle.
A comparison more likely to yield arguments in favour of an
Assyrian date for the Arabian (and Dumah) oracles is that
which has been made between them and the Moab oracles of
Isa. xvf.2 While it must be admitted that the exegesis of
Isa. xv f is very uncertain and that the conclusions of the
commentators are as tentative as they are varied, there are at
least some indications which may suggest a common background. Thus it is possible to detect a note of sympathy towards
Moab in these chapters ; 3 feelings of solidarity and sympathy
too have been detected in the attitude of the Dumah and
1 E.g. Wildberger; cf. Duhm, Marti and Fohrer. The exegesis and understanding
of Isa. xvi are so uncertain and varied that interpretation of the relationship of
verses 13f to what precedes them is here left out of account.
2 So e.g. (briefly) D.K.
3 See esp. xv 1-5 and xvi 8-11. Wildberger regards these sections as old laments
over Moab. To him it is unthinkable that they should have been on the lips of
Isaiah, but he believes that their sympathetic attitude points to the period
between Tiglath-pileser and the end of Judah when all the Palestinian states
were, so to speak, in the same boat.
99
Arabian oracles (so e.g. Wildberger, pp. 7g8ff). Secondly,

there is the use in both the Moabite and the Dumah oracles of
the unusual form V1!? (Isa. xv i; cf. xxi n ) , a form which, as
ibn Ezra suggested (see above, pp. 43f), may imply that a
particular night is referred to and that the oracles thus indicate
the same historical situation. Thirdly, to both has been appended a prose epilogue1 in wording that is largely identical. As it is
virtually certain that the epilogues derive from the same author,
it is further likely that their author regarded the preceding
oracles as in some sense comparable.
If literary considerations suggest (albeit very tentatively) a
common background for the Moab oracles and those of Isa.
xxi 1 iff the question naturally arises whether Moab, Edom and
North Arabia were ever historically the joint object of an
attack by a foreign power and at a time when we may suppose
that a Judaean prophet might show them at least a modicum
of sympathy. The only period when all such conditions were
fulfilled was that of the Assyrian attacks on southern Palestine
and Arabia at the end of the eighth and the beginning of the
seventh centuries B.C. (see above, pp. 78f, and below,
pp. 131Q.
Verses 16 & 17
The view of those scholars who suppose that this oracle is a
genuine (if later) Isaianic addition have been discussed above in
relation to the Arabian oracle. It remains here to record two
further views of these verses. First, Duhm (followed by Marti)
considers that the oracle is an epilogue belonging to the time of
Alexander Jannaeus (104-73 B.C.) when Aretas and the Arabs
constituted a threat to the Jews.2 In these circumstances the
Jews would have recalled with hope the old prophecy concerning the Arabs (which they attributed to Isaiah - so Marti).
Such views of the verse in question have gained little support
and telling arguments against them have been set forth, e.g. by
Galling.3
1 I.e. on that view of xvi I3f and xxi i6f.
2 Marti suggests the further possibility that the oracle is to be dated in the time of
John Hyrcanus (135-105 B.C).
3 P. 62. In particular the evidence of Sir. xlix (which witnesses to the close of the
prophetic canon in the first decade of the second century B.C) and of iQIs a
(which, dated 100 B.C., contains the whole text of Isaiah) is decisive.
100
ISAIAH
XXI
Secondly, Wildberger believes that the verses concerning

Kedar were added by a later editor to whom Kedar was of
more interest than Duma(t el Gendel),1 Dedan and Teima. It
was not until post-exilic times that Kedar's hegemony was
established as far as Palestine and Egypt2 and their business
enterprises brought to them considerable prosperity and power
(so, according to Wildberger, the word TDD). At any time
during this period the editor concerned could have added such
an oracle in the name of Yahweh (without feeling the need to
specify the agents of Kedar's collapse) in order to give expression to Israel's faith that the apparent superiority of her foes
would be short-lived. Wildberger's treatment of the similar
verses of Isa. xvi 13f is fuller and he there expresses the view
that the words in question are redolent of apocalyptic schematization rather than of prophetic pronouncement. Thus, the
point of the comparison 'like the years of a hireling' 3 is not to
denote the exact computation of a period of time, but rather
the miserable quality of a hireling's (and so Israel's) life. Moab
itself is for the redactor manifestly a paradigm of a heathen
godless nation rather than an historical reality. Even the
reference to a remnant (though here expressly a tiny one) is
stylized, and it is present because such references were thought
to be obligatory.
Much of Wildberger's treatment of these verses (on xvi 13f
and on xxi i6f) is speculative and perhaps it is necessarily so.
On the other hand, his account suffers from the weakness that,
while he inclines to the view that a common author was
responsible for both additions, his treatment of xvi 13f appears
to differ from that of xxi i6f. In the former case he adopts the
view that the words are w<m-apocalyptic; in the latter he
believes that they refer explicitly (if generally) to one of Israel's
post-exilic adversaries. Secondly, his understanding of the
tertium comparationis TDtf 'WD is open to serious objection, for it
appears to rest upon the view that the absence of an article
prefixed to TDtP precludes the translation 'like the years of the
(known class) hireling', but permits the translation 'like hireling
1 Sic ex hypothesi Wildberger.
2 For the evidence for the latter in an Aramaic inscription of the fifth century B.C.,
see Rabinowitz inJM.E.S. 15 (1956), iff.
3 Wildberger is unable to resolve the question whether hireling ('Tagelohner')
or mercenary ('Soldner') is the correct translation of T W .
IOI
years' ('wie Tagelohnerjahre'), i.e. miserable years. In fact,

however, the article (theoretically at least) would equally be
required if TDtP were used to convey Wildberger's adjectival
sense, for if it does convey such an adjectival sense, its grammatical form is still that of a noun and therefore subject to the
usual rules. But theoretical considerations apart, from G.K.
i26p it is clear that indeterminate comparisons do exist in
biblical Hebrew even if they are rare. Wildberger's argument
that TDtP *3BD must mean 'like hireling years' is therefore, on
grammatical grounds, highly dubious. Wildberger, however,
also appeals to Job vii if where, he submits, the comparison is
clearly concerned with the miserable quality of a hireling's
life. With this observation there need be no disagreement.
When, however, TDP *WD follows immediately and explains
references to a specific number of years (in Isa. xxi 16, one; in
xvi 14, three), the 'misery of life' alluded to can hardly imply
anything but that the specified number of years will, for the
reason that they are miserable, be counted with very great
accuracy.
Wildberger's contention that the period of time specified in
Isa. xvi 14 is more akin to apocalyptic schematization than to
prophetic pronouncement raises interesting questions concerning the relationship between (later) prophecy and apocalyptic.
Mention in Isa. xvi 14, however, of so small a period as three
years (and a fortiori in Isa. xxi 16 of one year) is hardly consistent with the broader sweep of apocalyptic seasons and epochs.
The two epilogues to xvi and xxi are better regarded as later
editorial additions designed to give contemporary relevance
to the oracles to which they are appended. It may readily be
agreed that reinterpretation of prophecy and the attempt
to apply older oracles to later situations is an activity which
was to reach its ultimate sophistication in apocalyptic. However, the specific nature and short focus of the epilogues before
us indicate that they belong to an earlier rather than to a later
state in that particular development.
It is difficult to be sure which historical situation the epilogue
of xxi 16f reflects. On the one hand, the period par excellence of
Kedar's dominance in North Arabia was the Persian.1 On the
1 So Wildberger, p. 802. For a recent evaluation of the evidence concerning the
history of Kedar, see Bartlett, 'From Edomites to Nabataeans', pp. 53ff.
102
ISAIAH XXI
other, the name is perfectly well known from Assyrian military

annals and Jer. xlix 28 suggests that Nebuchadrezzar campaigned against the tribe in his campaigns of 589-580.x
Further, from the point of view of the bible, it is possible that
the name c Kedar 5 , originally a tribal name, was used from at
least Jeremiah's time (see ii 10) to denote all the peoples of
Arabia (so Gray; see above).
On the view that Isa. xvi 13f comes from the same hand as
xxi i6f, a date suitable for both epilogues seems to be necessary.
It is possible that Nebuchadrezzar's campaign of 5822 was the
beginning of the end of Moab as a state,3 for thereafter, in the
bible at least, references to that country are literary rather than
historical.4 The period in which earlier oracles concerning
Moab are most likely to have been adapted to express judgement against that country is precisely ' the end of the seventh
and the beginning of the sixth centuries B.C., when Moab
appears to have taunted Judah and even attacked her' (Wiseman, Peoples, p. 242). If such considerations are sound, Isa.
xv f, as redirected by the epilogue to them, may be contemporary with Ezek. xxv 8-11 5 and Jer. xlviii.6
1 C.A.H., m (p. 215) is inclined to accept this evidence as likely. For the evidence
of Josephus and of Babylonian texts, see p. 92 above.
2 So Josephus (Antiquities x 9, 7), who mentions that Nebuchadrezzar's attack on
Moab was in the twenty-third year of his reign.
3 See A. H. van Zyl, The Moabites (Leiden, i960), pp. I57ff.
4 Cf. with Wildberger, Ezra ix 1; Neh. xiii 1, 23; Pss. lx 10, lxxxiii 7, cviii 10; Isa.
xxv 1 of; Dan. xi 41.
5 Zimmerli understands this oracle to reflect the events of 582 B.C.
6 That Jer. xlviii does not contain all the material from Isa. xvf need not tell
against this understanding of the two. For Jer. xlviii is manifestly contrived and
dependent on other biblical texts. That Jer. xlviii should have omitted passages
favourable to Moab from Isa. xvf (e.g. xvi 1, 3-5) is consistent with his way of
announcing doom for Moab; cf. Wildberger, p. 606.
3
Verses i10
History and Exegesis
Consideration of the text and meaning of the oracles of Isa. xxi
and of their exegesis in relation to the historical events of the
ancient Near East in the period in and after the eighth century
B.C. has shown considerable diversity of opinion among scholars.
Indeed it is because uncertainty attends any attempt to determine the meaning and import of so much of the text of this
chapter that recourse is had to particular working hypotheses,
for only by means of them can any reasonably coherent and
satisfactory account of its meaning apparently be given. Thus
the mention in verses I - I O of the fall of Babylon together with
an apparent reference to an attacking force of Elamites and
Medes often constitutes the necessary initial clue for the construction of what I have called the orthodox critical view of this
oracle. On the basis of this clue the orthodox critical school
plump for the fall of Babylon at the hands of Cyrus. That is
their working hypothesis. It is for them the best bet. In support
of it (once it is adopted) various historical arguments are
adduced.1 When a phrase in the text does not apparently
match it, it is emended,2 or a different meaning is postulated
for it,3 or other historical considerations are advanced in
explanation.4 And what is true of the orthodox critical view as a
working hypothesis is mutatis mutandis often true of alternative
accounts of the prophecy, e.g. Cobb's view that the oracle is
an Isaianic composition of the eighth century B.C.
The question now arises whether this approach to the oracles
of Isa. xxi is in fact the one most likely to lead to the correct
understanding of its contents. Fundamentally the approach
1
2
3
4
E.g. Galling on the return of Nabonidus from the Arabian desert.

E.g. Duhm and Wildberger on nnmx.
E.g. Eitan on nnmR.
E.g. D.K., Duhm and Wildberger on the mention of Medes and Elamites rather
than Persians as the agents of Babylon's fall.
103
104
ISAIAH XXI
rests upon the assumption that each of the prophecies1 reflects

one particular historical situation. Is the assumption, however,
a valid one ? Are we right to find in each of the oracles of Isa.
xxi clear and consistent references to any one historical situation ? To what extent, moreover, is the scholastic adage
quidquid recipitur recipitur ad modum recipientis likely to be exem-
plified in the transmission of these oracles ? Thus, for example,

if (at least some of) the words of Isa. xxi I - I O antedate the sixth
century B.C., would they not have attained new significance
precisely in the sixth century - in the eyes of those who received
or read them at that time ? And if the fall of Babylon at the
hands of Cyrus was, after 539, pre-eminently in the minds of
those who received the oracle, is it not reasonable to suppose
that indications of their understanding of the prophecy may
have survived, whether independently of the text (in e.g.
rabbinic traditions) or by the adaptation or even emendation
of its words to this end ?
That prophecies in fact attained new and very particular
significance in the minds of those who later received them may
be illustrated from the present study in connexion with the
Arabian oracle. I recorded (p. 57) the Talmudic tradition that
by allusion the prophecy concerns the plight of Jewish refugees
fleeing to Arab territory after the fall of Jerusalem at the hands
of Nebuchadrezzar in the sixth century B.C. The presence of
this tradition in the Talmud provides us with evidence that
that is how the prophecy was interpreted in Jewish circles at
a comparatively early date. 2 It was further suggested (p. 57
above) that the particular pointing of one word within the
oracle may be an indication that the Massoretes also wished to
refer the prophecy to this incident.
What can be documented for the reinterpretation of the
Arabian oracle from Talmudic and (perhaps) massoretic
traditions may be shown for other Isaianic prophecies within
the (consonantal) text of Isaiah itself. For example, there is the
well-known interpretation of the prophecy of iii if, where
the word 'staff', originally clearly a characterization of the
1 For some, of course, verses 1-10 consist of more than one prophecy.
2 It is conceivable that this linking of the prophecy with events following the fall
of Jerusalem in 587 indicates that the prophecy was already in existence at that
time; if that could be established we would have, of course, a terminus ad quern
for it.
VERSES I - I O
IO5
eighth-century-B.c. pillars of Judah's society (thereafter listed),

is later interpreted as the 'staff of bread' and the 'staff of
water5 (i.e. as the necessities of life) removed in the circumstances attending the siege of Jerusalem in 5871 (Ezek. iv 16,
v 16, xiv 13).
Again, and equally well known, there is the gloss added to
the words of Isa. vi 13, 'the holy seed is the stock thereof
(R.V.), whereby (what was earlier) an oracle of doom has been
modified by a later interpreter so that it should convey a
promise of hope for a remnant. 2
If, then, certain prophecies within the book of Isaiah show
evidence of having been reinterpreted and directed to situations
later than those to which originally they referred, it is possible
that those of chapter xxi reflect a similar process.
If this understanding of the oracles of Isa. xxi is correct, it
follows that the fundamental assumption adopted by most
modern critics, viz. that each prophecy (as we have it) reflects
a single historical situation, is mistaken.
It is readily admitted that much of the argument adduced
above is a priori in character. Yet argument is necessarily of this
sort when basic assumptions are questioned; and particularly
so when those assumptions underlie the difficult quest for the
elucidation of oracles of such antiquity and obscurity as those of
Isa. xxi.
The proof of the pudding, however, is in the eating. Apart
from the question of assumptions, we have to consider which
understanding of the prophecies of Isa. xxi best fits what we
can see of them at present. When it is necessary to judge
between two understandings which are mutually irreconcilable,
then that which leaves unsolved the smaller number of problems
is the more likely to be correct.
An attempt will be made to show that many of the words and
phrases of Isa. xxi 1-10 have their origin in the eighth century
but that the oracle as a whole attained its final form in the
book of Isaiah as a prophecy which portrayed the fall of
Babylon at the hands of Cyrus in 539 B.C. The prophecy may
1 See Gray, p. 63.
2 See e.g. Gray, p. 111. For further discussion of the later reinterpretation of
Isaianic oracles, see D. R. Jones in Z-A-W. 67 (1955), 256ff. See also S.
Mowinckel, Prophecy and Tradition (Oslo, 1946), and E. Nielsen, Oral Tradition
(London, 1954).
106
ISAIAH XXI
then be regarded as a palimpsest. Its reception and interpretation within the Isaianic tradition give to it its superimposed
form and it is this that enables it to speak of that fall of Babylon
which was of such immense importance in the history of the
Jewish people. Beneath the superimposed form, however, we
may catch at least glimpses of earlier prophecies and of the
historical circumstances of an earlier age.
This understanding of the nature of Isa. xxi I - I O has the
advantage over those which see in it a single historical background, that it accounts for much that is otherwise obscure or
apparently archaic. Further, it has the advantage that it
accounts for the unity of thought which has been detected
(especially by Delitzsch) between Isa. xxi I - I O and material in
the book of Isaiah which is likely to be genuine. For Delitzsch
(i) that unity was to be accounted for on the supposition that
Isaiah was himself the author of an oracle which portrayed the
fall of Babylon in the sixth century B.C.1 The unity of thought
which Delitzsch originally detected can, however, more
satisfactorily be accounted for on the supposition that words of
Isaiah were later seen to have their fulfilment in the events of
the sixth century.
As has been almost universally recognized, the most likely
starting point for the interpretation of the oracles of Isa. xxi
is the reference to Elamites and Medes in verse 2. To adherents
of the orthodox critical theory, the force of Elamites and Medes
who are urged to attack must be identified with the forces of the
Persian Cyrus who brought about the end of the Babylonian
empire in 539. When the question is asked why Cyrus's forces
are here characterized as Elamites and Medes, various considerations are advanced to meet the apparent difficulty. Thus
the description is held to reflect the diverse character of Cyrus's
empire which comprised Media (conquered in 550-549) and
Anshan (formerly part of Elam) of which Cyrus was king.2
Again it is urged that, while it is historically accurate to
describe the conquerors of Babylon as Persians, it was customary
in Greece and Egypt as well as elsewhere in the O.T. to refer
1 Delitzsch's original argument (1) concerns primarily the 'Fall of Babylon'
oracle of xiiif; but he claims for Isa. xxi 1-10 the same marks of genuineness.
2 See pp. 67f. For an indication of the uncertain character of this argument, see
Barnes, p. 585^
VERSES I - I O
107
to the Persians as Medes. Where the term 'Elam' is concerned,

it is also urged that there is clear evidence of the continued
existence of Elamite culture and identity within the Persian
empire2 although its separate existence as a nation ceased in
640-639 when Ashurbanipal conquered Susa.
Against such considerations there remain formidable objections. As Barnes observes, Cyrus never styled himself an Elamite
nor was he known as such either in the bible or in Persian
monuments. While it is true that Isa. xiii 17 and Jer. li 11, 28
mention Medes as agents of the fall of Babylon in 539, in none
of these texts is there mention of Elam. It may be added that,
where Jer. li is concerned, this is particularly noticeable in
verse 27 where there is a list of other kingdoms called to attack
Babylon in company with the Medes: 'Ararat, Minni and
Ashkenaz' (R.V.). Barnes further observes that in Isa. xliv 28
and xlv 1, where Cyrus is actually mentioned by name, no
nationality is assigned to him.
We may agree with Wildberger's assertion that the evidence
of the oracle concerning Elam in Jer. xlix, whether that oracle
is to be dated in the first year of Zedekiah (as its title suggests see Rudolph and Weiser) or later (see Fohrer, Einleitung),
seems to corroborate the external indications3 that Elam, after
the fall of Susa in 639, retained some political identity. On the
other hand, as Wildberger also admits, the early exilic prophecy
of Ezek. xxxii 17ff4 tells against the view that Elam was a
significant power in the time of the Babylonian exile,5 for there
Elam (verse 24) is listed with Assyria and Meshech-tubal6
amongst the ghosts of bygone empires lingering in Sheol. In the
1 See Wildberger, p. 507. He follows H. Bengtson in (S.) Fischer Weltgeschichte,
vol. v, p. 37, and Wiseman, Peoples, p. 316. It should be noted, however, that
Persians, not Medes, are mentioned in Ezek. xxvii 10 and xxxviii 5.
2 Wildberger, pp. 772f.
3 E.g. F. W. Konig, Geschichte Elams (Leipzig, 1931) vol. xxix, 4, and Reallexikon
der Assyriologie, vol. 11 (1938), p. 337.
4 For which see e.g. Zimmerli, 11, ad loc.
5 The possibility (and it is no more than that) - see Wiseman, Chronicles, p. 36 that the Babylonian Chronicle refers to a king of Elam as an opponent of
Nebuchadrezzar in 596 hardly contradicts this view. If the reading were to be
verified, it would not amount to evidence that Elam had at the time more than a
residual political identity.
6 I.e. nations of Asia Minor (Phrygia and Cappadocia), known as Muski and
Tabal in Assyrian records (see Luckenbill, 11), who, like Elam, were constant
opponents of the Assyrian kings.
108
ISAIAH
XXI
light of this evidence it would be very strange for a prophetic

writer in exilic times to characterize (even part of) Cyrus's
forces as Elamites.
Although Elam was not a significant political power in the
sixth century B.C. it most certainly was so, as also were the
Medes, in Assyrian times. The latter were formidable opponents
of the Assyrian empire from 745 and the beginning of its
ascendancy under Tiglath-pileser III. This king campaigned
against the Medes (C.A.H., m, pp. 34O, and Sargon II, who
called them 'the mightly Medes' (Luckenbill, 11, no. 11), was
obliged to follow his example. During Sennacherib's reign the
Medes continued to present a challenge to Assyria,1 at that time
particularly threatened by Merodach-baladan of Babylon and
his Elamite allies. In the end it was the Medes under Cyaxares
II who brought about the downfall of Assyria, the Babylonians
under Nabopolassar joining them for the final assault on
Nineveh in 612 (C.A.H. m, pp. 2g6f).
Elam was, perhaps, Assyria's most important and formidable
enemy. From 744 B.C. when Tiglath-pileser III consolidated his
empire east of the Tigris as far as the northern borders of Elam
(C.A.H., in, p. 34), that country, threatened as it was, became
the natural enemy of Assyria. At this time (731) Merodachbaladan, prince of Bit Yakin in South Babylonia, originally
subject to Tiglath-pileser, wishing to become king of an independent Babylon, began his tireless and skilful negotiations with
the Chaldeans in the south and with the Aramaeans in the
north. His purpose was to achieve the necessary prerequisite to
his plan - a united Babylonia {C.A.H., m, pp. 47f). Yet his
greatest achievement was to enlist the support of the Elamites
in his venture, and this apparently he did by offering them
very considerable bribes and gifts (Brinkman, pp. i6iff). At the
accession of Sargon II in 721 he was ready to throw off the
Assyrian yoke and, after the battle of Der in 720, when Elamites
caused the Assyrians to withdraw, he achieved his object. Thereafter he reigned for twelve years until 709 when Sargon ousted
him and caused him to flee to his home in Bit Yakin. In 703,
shortly after the year of Sennacherib's accession to the Assyrian
throne (705), Merodach-baladan regained the Babylonian
1 See Luckenbill, 11, no. 238, and note particularly the phrase * to the yoke of my
rule I made them submit'.
VERSES
I-IO
throne and again sought Elamite military aid to ward off the
expected attack on him by Sennacherib. On this occasion c not
only do the Elamites seem to have provided the majority of the
troops fighting on the Babylonian side, b u t . . . an Elamite
commander actually directed the combined army' (Brinkman,
p. 165). It should be noted here that Merodach-baladan's
embassy to Hezekiah of Jerusalem (2 Kings xx 12-19; Isa.
xxxix) may have taken place at this time, and it is likely that his
intention was to mount a concerted rising against Assyria in the
west (with the aid of such kings as Hezekiah) as well as in
Babylon itself. In the event, his unpopularity in Babylon caused
its inhabitants to try to forestall his seizing the throne, with the
result that he had to act earlier than he had intended (703),
{C.A.H., in, p. 63). At all events, Sennacherib's army won the
ensuing battle and, after a rule of only nine months, Merodachbaladan fled to the southern swamplands. Sennacherib entered
Babylon to a friendly reception.
In 700 B.C. Merodach-baladan fled to Elam before an
Assyrian army (led by one of Sennacherib's sons - C.A.H., m,
p. 65) whose function was to subdue finally Bit Yakin and its
Chaldean inhabitants. He was not heard of again and it is
presumed that he died in exile. Thereafter, continued intrigue
between the Chaldean Mushezib-marduk, a later anti-Assyrian
ruler of Babylon, and the Elamites culminated in the bloody
battle of Khalule (C.A.H., m, p. 68) which marked the beginning of the end for the anti-Assyrian Chaldean party in Babylon.
Though the battle of Khalule was indecisive, Elam was thereafter in a state of weakness because of the illness of her king, and
consequently Sennacherib was able in 689 to deal with Babylon
unhindered. The troops sacked and looted the city and 'the
gods dwelling therein - the hands of my people took them and
they smashed them'. 1
In connexion with this outline of the history of relations
1 Luckenbill, 11, no. 340. Boutflower and Erlandsson note that the word 'smashed'
{uSabbiru) is cognate to that used in Isa. xxi 9; Boutflower regards this as evidence
that this part of the prophecy concerns the fall of Babylon in 689 on the grounds
that a prophecy is 'interpreted by its fulfilment'. Erlandsson merely notes the
correspondence without further comment. It seems to me, however, that little
can be made of this correspondence other than that the Bavian inscription confirms that Assyrians did smash (even) Babylonian gods, whereas Cyrus apparently
smashed no one's.
110
ISAIAH
XXI
between Elam, Babylon and Assyria in the eighth century B.C.,

two further points should be made. First, it is abundantly clear
that various Aramaean peoples living to the east of the Tigris
were throughout this period actively involved on the side of the
anti-Assyrian Chaldeans. Indeed since as early as 818 B.C.
these peoples had sided with Babylon against Assyria (C.A.H.,
HI, p. 27). As has been stated above, Merodach-baladan's first
diplomatic missions were concerned with enlisting their support
(ibid. p. 48), and by the time of the revolt in the years following
705 he could count on them (ibid. p. 62; cf. Brinkman, p. 164).
They were likewise present amongst the forces raised by
Mushezib-marduk in 691 (C.A.H., m, p. 68).
Secondly, as Brinkman has concluded (p. 164) from a survey
of the evidence, it was the Elamites who undertook most of the
actual fighting with Assyria; Merodach-baladan himself always
seemed to avoid direct confrontation with them.
As Erlandsson has said, * It seems obvious that the oracle of
Isa. xxi 1-10 with the names Elam, Media and Babylon
harmonizes very well with the historical actualities around
700 B.C.' It should be noted that Erlandsson's conclusion is a
restatement of what had been suggested for the oracle by other
scholars at an earlier time.1 Now Kleinert's statement of the
case for an eighth-century-B.c. date for the prophecy rested on
the assumption that the Elamites and Medes of verse 2 b were,
like the Elamites and men of Kir in Isa. xxii 6, loyal vassals
serving in the Assyrian army. Where Isa. xxii 6 is concerned,
scholars of the orthodox critical school are largely in agreement
with Kleinert's view of the matter. For Isa. xxi 2, however, they
believe the assumption to be entirely out of place, for, as
Dillmann-Kittel observe (see above, p. 67), Elam and Media,
represented in the prophecy as attacking Babylon (see verses 2
and 9), were at that time enemies of Sargon and allies of Babylon.
With this objection to Kleinert's interpretation there can
be nothing but complete agreement. It is unthinkable that the
author of an oracle in the last decade of the eighth century B.C.
would characterize an Assyrian army as (consisting of) Elamites
and Medes.2
1 On Kleinert, Cheyne etc., see above.
2 The objection is quite as telling when applied to modifications of the original
statement such as that of Barnes; see p. 71 above.
VERSES
I-IO
III
The question, however, remains: why is the Assyrian army

attacking Jerusalem in Isa. xxii 6 1 described as comprising
Elamites and men of Kir (i.e. Aramaeans2) ? The usual
explanation that Elamites and Aramaeans served as vassals
in the Assyrian army is based, so far as I am aware, purely on
speculation for which there is no evidence.3 Further, if it is
unthinkable that Elamites (and Medes) should have served as
loyal vassals of Assyria in her attack on Babylon at the end of
the eighth century (Isa. xxi 2), then it is equally unthinkable
that Elamites and Aramaeans should have done so in her attack
on Hezekiah of Jerusalem (Isa. xxii 6).
There remains an alternative way of understanding both
Isa. xxi 2b and xxii 6 as authentic texts which reflect accurately
the circumstances of the eighth century B.C. They may be
understood as quotations which are reproduced and incorporated in the respective oracles. It was noted above (p. 15)
that ibn Ezra (and, apparently independently, Erlandsson)
regard Isa. xxi 2b as a quotation of the shouts of soldiers urging
each other to battle, and his view, like so much in the rabbinic
commentaries, may depend upon ancient tradition.
If Isa. xxii 6 is indeed a quotation, it is incorporated into the
oracle (presumably by Isaiah) in order vividly to represent the
initial mistaken judgement of Hezekiah's Jerusalem. News of
Merodach-baladan's mobilization of the Aramaeans and his
successful enlisting of the support of Elam had been on the lips
of Jerusalem's inhabitants. On it they had pinned their hopes
for deliverance from Assyria. The actual result of this mistaken
reliance upon foreign powers is depicted in the next verse (7):
'And it came to pass (TPI - i.e. the result of all this is) that thy
choicest valleys were full of chariots, and the horsemen set
themselves in array at the gate' (R.V.). In other words,
Sennacherib's siege of 701-700 was, according to Isaiah, the
inevitable result of Jerusalem's sinful reliance upon foreign
military power.
1 The oracle is usually regarded as substantially Isaianic and reflecting Sennacherib's siege of Jerusalem in 700 B.C.
2 Cf. with most commentaries Amos ix 7.
3 An alternative solution to the problem is to attribute the verse to a later redactor
who mistakenly supposed that Elamites were among the besiegers of Jerusalem;
so e.g. Marti. There is, however, no good reason, apart from the difficulty at
present under discussion, to deny the authenticity of the verse.
112
ISAIAH XXI
This understanding of Isa. xxii 6 has the merit that it fits

perfectly the situation at the end of the eighth century B.C. as
we know it from the Assyrian sources and it is entirely consistent with the thought and style of Isaiah of Jerusalem (see
p. 18 above).
The phrase 'Go up, Elam, etc' in Isa. xxi 2b may be viewed
as a very similar quotation whose incorporation in the oracle
vividly depicts the attitude of the inhabitants of Jerusalem.
Relying on Babylon and its king (see Isa. xxxix 1) they pin
their hopes on his Elamite and Median forces,1 urging them to
victory against the Assyrians. The reaction of Isaiah to this
attitude on the part of his countrymen is related in the following
verse (3). ' Therefore (p bv) are my loins filled with anguish. . .'
His message from Yahweh is that Babylon will fall to the
Assyrians and reliance upon her is as foolish as reliance upon
any other foreign power. Again, this understanding of the
all-important phrase has the merit that it fits the historical circumstances admirably, and is consistent with the thought of
Isaiah of Jerusalem.2
If this interpretation of Isa. xxi 2b is correct, then two assumptions of the orthodox critical view must be abandoned: first,
the assumption that the agents of the fall of Babylon (verse 9)
are named in verse 2b; secondly, the assumption that the
speaker of the command in verse 2b, 'Go up, Elam, etc.', is
Yahweh and that, consequently, the command belongs to the
prophetic form 'summons to war' (Aufforderung zum Kampf).3
The arguments rehearsed above attempt to show that the allimportant phrase 'Go up, Elam, etc' belongs originally to the
earlier level of the palimpsest that constitutes Isa. xxi. It is also
suggested, however, that the words were later seen to have
their fulfilment in the fall of Babylon at the hands of Cyrus in
539. To this interpretation of the prophecy rabbinic tradition
1 The mention of Medes may be a further accurate reflexion of the circumstances,
for the Medes did pose a threat to the Assyrians; see above, pp. 108 ff. On the
other hand, the use of the term may here constitute a parallelistic synonym for
'Elam'.
2 It will be recalled that the distress of the prophet at the news of the fall of
Babylon constituted an important objection to the orthodox critical view; on
the present interpretation it is entirely understandable.
3 So Bach.
VERSES
I-IO
113
bears ample witness (see above, pp. I4ff) and indeed it is in

accord with the teaching of the great exilic writer DeuteroIsaiah. For here, within the Isaianic tradition itself, is set forth
the argument from prophecy. Yahweh, unlike the Babylonian
idols, is God, for he acts in history, and before he acts he tells
his witnesses of his intentions so that they may perceive his
purpose (see e.g. Isa. xli 22ff, xlii 1 iff, xliv 6fF, and esp. xlv 2off).
The prophecies to which Deutero-Isaiah is referring, and on the
fulfilment of which he bases his case, are likely to comprise
material such as Isa. xxi - material that has come down from
Assyrian times.
On the basis of the assumption that Isa. xxi 2b was thus
interpreted by late exilic or early post-exilic readers, two
further points may be made. First, in the new situation the
archaic and anachronistic character of the words will not have
diminished the force of the revelation. The reverse is likely to
have been the case; reference to the Persian Cyrus under the
names of Elam and Media will have served to emphasize the
mysterious consistency and power of Yahweh whose words are
revealed in past history yet find fulfilment in contemporary
events.
Secondly, the reinterpretation of the words will have
invested the phrase as a whole with a new form and significance.
Now, indeed, linked with the news of the fall of Babylon
(verse 9), it becomes the formal summons to war (Aufforderung
zum Kampf), a feature which became so common in exilic
prophecies.1
It has been suggested that Isa. xxi 1-10 is a sort of palimpsest
and that many of its words and phrases may be traced back to
the eighth century B.C. and to Assyrian times. On the other
hand, it is suggested that the oracle in the form in which we
have it was seen to have its fulfilment in the fall of Babylon in
539 B.C. The question arises whether there is any collateral
evidence to corroborate this suggestion. In answer to that
question, attention may be directed to the prophecy of the fall
of Babylon in Jer. If. It is widely believed that this prophecy
is an exilic composition which, like other foreign oracles in
1 Cf. with Bach Jer. xlvi 9 ; xlix 3 1 ; 1 i4f, 2 9 ; li 11, 27.
114
ISAIAH XXI
Jer. xlvi-li, is dependent upon earlier prophecies.1 More

particularly its dependence upon the book of Isaiah, and
especially upon Isa. xiiif,2 is very widely accepted. As one small
but telling illustration, we may cite the use, in Jer. 1 29 and li 5,
of the characteristically Isaianic title of Yahweh, 'the Holy
One of Israel'. Although dependence upon Isa. xiiif has been
widely noticed, attention has not generally been given to the
striking points of contact between these chapters and the oracle
of Isa. xxi 1-10.3 The more obvious of those points of contact
(i.e. where there is at least partial verbal identity) may be
listed as follows:
(1) li 48, 53, 55f. In all these four verses the 'spoiler' (R.V.)
is to ravage Babylon, whether the 'spoiler' is Yahweh himself
(55) or his agents (48, 53, 56). Cf. the use of the word sdd in
Isa. xxi 2.
(2) li 8. In this verse the fall of Babylon is announced in
words identical (*?M n*?D2) to those in Isa. xxi 9.
(3) 1 2; li 47, 52. In these verses God's judgement on the
graven images of Babylon is recorded; the word DÔD is used
in the two latter cases; cf. Isa. xxi 9. It may also be significant
that in these verses (47 and 52) the word *pK occurs after
OÔB as it does in Isa. xxi 9, though of course in a quite
different sense.
(4) li 33. Babylon is described as 'like a threshing-floor at the
time when it is trodden'. Cf. the use of the same word (pa) in
Isa. xxi 10.
Since on other grounds Jer. If is to be regarded as dependent
upon the prophecies of Isaiah, the most likely explanation of
such verbal correspondences is that the author of this material
made use of Isa. xxi 1-10. To that conclusion at least two
further resemblances4 between the texts (but without exact
verbal agreement) may also point. They are:
(1) 1 2, 'Babylon is taken' (R.V.) (*?M mD^l); cf. Isa. xxi 9.
(2) li 1, 'a destroying wind' (R.V.) (rpfWD rm); cf. Isa.
xxi 1.
1 See e.g. the commentaries of Rudolph and Weiser.
2 For a recent treatment, see Erlandsson, pp. 1546.
3 An exception is Delitzsch (i) who initially believed that Jeremiah 'was
acquainted with this oracle'.
4 Other cases where Jer. If possibly alludes to Isa. xxi 1-10 are mentioned in the
commentary below.
VERSES I - I O
115
Jer. If is usually dated at a time just prior to Cyrus's capture

of Babylon,1 and its function seems to have been to set forth
Yahweh's imminent action in destroying Babylon and enabling
the return of the exiles to Judah. That its author has made use
of the oracle Isa. xxi 1-10 is thus an independent indication
that Jews of the late exilic period saw in the history of their times
a fulfilment of (some part of) the words of Isa. xxi 1-10.
It should here be noted that the author of Jer. If has incorporated interpretative changes or amplifications to the prophecies which he has used. Thus, Jeremiah's prophecy (vi 22-4)
depicts a foe from the north who threatens 'the daughter of
Zion': in 1 41-3 it is 'the daughter of Babylon' who becomes
the object of the same threat. In Jer. vi 24 it is 'we' (sc. the
Judaeans) who are terrified by the threat, whereas in 1 43 it is
the king of Babylon. Again, as an example of the author's
treatment of an Isaianic prophecy, Jer. li 53 may be cited. In
Isa. xiv 13*1 will ascend into heaven (jf?S7K D*wn), I will exalt
my throne above the stars of God' (R.V.) becomes, in Jer. li 53
'Though Babylon should mount up to heaven (o*wn *?M nVtfn),
and though she should fortify the height of her strength. . .'
(R.V.).
The same treatment, as one would expect, is to be discerned
in the use that is made of Isa. xxi 1-10. Where Isa. xxi 10
speaks of (?) Israel in the vocative as 'my threshing' and as
'the corn of my threshing-floor' (R.V.), Jer. li 33 explicitly
regards the daughter of Babylon as ' like a threshing-floor at
the time when it is trodden' (R.V.). Isaiah's image is, however,
interpreted further in Jer. li 33 where the statement 'yet a
little while, and the time of harvest shall come for her' (R.V.)
is added.
If, then, there is evidence to the effect that the author of
Jer. If has adapted (at least some of) the material in Isa. xxi
1-10 to the situation of his own times, that constitutes an
external indication in favour of the view suggested on other
grounds above, that the same process took place within the
Isaianic tradition itself and that the text of Isa. xxi 1-10, as we
have received it, reflects that process.
It remains to present a translation and annotation of the text
1 See e.g. Rudolph and Weiser.
Il6
ISAIAH XXI
of Isa. xxi i10 in accordance with this understanding of it.

Before that is done, some preliminary observations must be
made:
1. It cannot be emphasized too strongly that, on the view
of the text which has been presented above, any attempt
systematically to distinguish the eighth-century original text
from the sixth-century interpreted text is fraught with very
considerable uncertainty, and consequently speculation must
inevitably be used.
2. If the author of Jer. If made use of Isa. xxi I - I O , he
probably made use of the original text and not of the reinterpreted text. For the reinterpreted text of Isa. xxi I - I O must
have derived its form ex hypothesi at a date not so different from
that of Jer. If.
3. It seems unlikely that Jer. If contains material which can
be regarded as vaticinia ex eventu.1 It is possible (see below on
verse 5) that the reinterpreted text of Isa. xxi 1-10 contains at
least one such vaticinium. On that assumption, the reinterpreted
text of Isa. xxi 1-10 is later than that of Jer. If.
4. At all events, whether Jer. If and the reinterpreted text of
Isa. xxi I - I O are contemporary or not, the former text, being
separate from and fuller than the latter, may in places provide
clues for use in the attempt to recover the original text of
Isa. xxi I - I O .
5. References to pages in the commentary which follows are,
unless otherwise stated, references to preceding pages of this work.
Translation and Commentary
Verse 1
Eighth century
Oracle from the wilderness* A storm,10 sweeping along like whirlwinds
in the Negeb, has come0 from the wilderness, from a terrible land.
a. The title is read *m& xtPfc (p. 7). It is likely to be later in
date than the oracle itself, but probably antedates the sixth
century B.G. (see below).
b. JT is taken to mean 'storm, wind 5 (pp. gf) and to be the
1 See again the commentaries of Rudolph and Weiser.
VERSES I - I O
117
subject of the verb xa. It is possible that the LXX and Peshitta
are witnesses to this way of interpreting the verse (p. 7).
Jer. li 1 (rrnttffc rrn) may also preserve an echo of this motif
(cf. p. 114).
c. *o is probably to be understood as a perfect rather than as a
participle (p. 7n).
The storm is a parable or sign of Yahweh's revelation of his
will;1 the contents of the revelation are described in the verses
that follow (pp. gf).
Sixth century
In Jer. li 42 the sea is said to come up upon Babylon (VM *?y nVs7
DVl). With this motif verse 55 of the same chapter may be
compared. On the assumption that Jer. li is dependent upon
Isa. xxi, Jer. li 42 may constitute evidence that the word D*
(understood as 'sea') was taken to be an agent of Babylon's
fall. With this view the evidence presented by the Targum to
Isa. xxi 1 may be consistent (p. 7n above).
On the other hand, Jer. li 13 (cf. Jer. 1 38), ' O thou that
dwellest upon many waters, (. . .thine end is come)', may be
understood as an indication of an alternative interpretation of
0"S viz. that it was read as an absolute in the expression D^"*D"T&
'desert of the sea'. This then becomes a name for Babylon
which is both descriptive of its geographical features (pp. 5ff)
but also proleptic of its fate (cf. ibn Ezra and Delitzsch; pp. 5-6
above). For while the sea is an agent of Babylon's destruction in
Jer. li 42, verse 43 states that Babylon will eventually become a
dry land and a desert (rDW) ms). The name DÎOTID coined
for Babylon consequently forms a suitable title for the sixthcentury form of the oracle.
On this view of the verse, it would appear at first sight that,
because D* was read as part of the title, there would have
remained no expressed subject of the verb N3. On the other
hand, as has been indicated above, Jer. li 42 may constitute
evidence that the word D**, understood as ' the sea' and perhaps
regarded as a symbol of invading armies,2 was also taken to
be the subject of the verb *o. That is to say, it seems likely that
the word o** in the text of Isa. xxi 1 was interpreted twice: first
1 Cf. perhaps Saadya's understanding of the phrase (p. 8 above).
2 Cf. B.D.B., p. 411, and Targum to Isa. xxi 1 (p. jn above).
Il8
ISAIAH XXI
as a part of the emended title, and secondly as the subject of the

main verb *o.
Oracle, Desert of the Sea. The sea, like whirlwinds sweeping along
the Negeb, has come (sc. upon Babylon) from the wilderness, from a
terrible [distant) land.
The reference is to the threat to Babylon posed by the Medes

and Persians. They are seen as coming from a terrible, distant
land (pp. i of) by way of the desert between Babylon and the
Persian sea (p. 7).
Verse 2 a
Eighth century
A harsh vision is told to me: the treacherous deals treacherously and the
devastator devastates.
The epigrammatic contents of the harsh vision may be paraphrased in order to bring out its meaning: 'Treacherous
rebellion is afoot - but its inevitable result is crushing devastation.' The reference of the term "man (pp. 1 if) is to the parties
of the anti-Assyrian alliance headed by Merodach-baladan, or
perhaps to this king himself. In any case Hezekiah and Judah
were firmly involved in these machinations. The term iiwn
(pp. I2ff) denotes Sennacherib or the Assyrians, whose response
to the rebellion will involve swift and brutal repression.
Sixth century
In Jer. li 55f Yahweh is said to be the devastator of Babylon

{liw) and in verse 53 devastators (DHW) are said to come
upon her (nV IXIT). Further, in the book of Jeremiah (p. 13)
Babylon is characterized as the devastator (*nw) of Jerusalem.
It is likely then that the epigrammatic words were interpreted
in the sixth century to mean that Babylon, once the devastator,
is herself to be devastated.1 With this interpretation the words of
Isa. xxxiii 1 are consistent and, being late, they may be a
reflection of it (cf. also Hab. ii 8). It should be recalled here that
the Targum to Isa. xxi 1 reflects this (re) interpretation of the
verse, having for the second word in each pair a passive verb
(pp. 12f above).
1 Cf. Ps. cxxxvii 8, where niiiwn 0?33) was probably originally rni"ntfn (Van);
soe.g. B.H. 3
VERSES I - I O
119
As there is no manuscript or versional evidence (apart from

the Targum, for which see above, pp. 12f) to indicate that any
of the words of the saying was read as a passive, and in the light
of Jer. li 25 where Yahweh is the subject of the active participle
TTW, it is probable that in the sixth century B.C. the phrase
77W TTWri was understood quasi impersonally: * There is one
who is devastating the (erstwhile) devastator' (cf. Rashi;
p. 13 above). 1
Although such an account of the phrase "mtP TVWT\ is
reasonably satisfactory, it is less easy to offer an estimate as to
the way in which -ma TJ'Dn was interpreted. For the word is
not found in Jer. li and is not predicated elsewhere in the O.T.
of Babylon. On the other hand, if the phrase ITW TIWT\ was
reinterpreted in the way suggested above, it is not implausible
to suggest that the adjoining phrase -ma 1X)27\ was seen to
depict an aspect of the shameful conduct of Babylon as a world
power which now, in her turn, she was to experience. Two of
the ancient versions (p. 12) detected in the word overtones of
oppression and such may have been generally a corollary of the
meaning 'treachery5 that is clearly attested for the word.
A harsh vision is told to me: there is one dealing treacherously with her
who dealt treacherously {with us); there is one devastating her who
devastated (us).
The vision is likely to have been understood to be harsh for

Babylon (cf. the rabbinic commentators, p. 13 above). The
agent of the vengeance exacted on Babylon is thought to be
Yahweh; cf. Jer. li 55.
Verse 2b
Eighth century
' Go up Elam: show hostility, Media'*' Bring an end toh all quietness.0
a. The words are a quotation of the sentiments expressed by

the people of Judah at the prospect of Babylonian-inspired
1 It is unlikely that Isa. xxxiii i can be cited in support of an argument that the
second word in each pair was reinterpreted as a passive. For while a passive
form of nv is found there (-rafan 'thou wilt be devastated'), when the same
construction is sought for the verb un which follows, there is a reversion to the
(equivalent) impersonal third person plural active construction ("-J3 na?' ' men
will deal treacherously with thee').
120
ISAIAH
XXI
rebellion against Assyria. Elam and Media are Merodachbaladan's allies and principal fighting force, and they are urged
to attack Assyria (pp. 14^ 11 iff).
b. watf? is read (pp. 176). It seems unlikely that the sudden
introduction of a first person singular subject (? referring to
Yahweh) belongs to the earlier text. On the other hand, such
an understanding of the text is precisely of a sort with later
reinterpretation. The imperative phrase (in2t2?n) may be taken
to indicate the inevitable result of the policy of rebellion indicated by the quotation; cf. the imperative of Isa. vi gf.1 The sense
of the argument is: 'your reliance upon rebellious intrigue will
have the consequence that you will destroy all possibility of
retaining or adopting that quiet, trustful waiting upon Yahweh
which is proper'; cf. e.g. Isa. vii 9, viii 5ff.
c. A noun derived from the root nil is read. Perhaps rim or
(with prosthetic x) the form nmx2 may be posited.
Sixth century
Go up, Elam; besiege, Media. I have brought an end to all sighing.
Yahweh is the subject of the first person singular verb and is

also understood to be the speaker of the imperative phrase. He
summons the Persian empire (under archaic names) to attack
Babylon (cf. Jer. 1 9; li 11, 27ff) and he announces the final
ending of the oppression (the sighing) of the Babylonian exile
(pp. i6f).
Verse 3
Eighth century
Therefore my loins arefilledwith anguish; pangs have taken hold of me,
as the pangs of a woman giving birth: I am so pained that I cannot
hear, so dismayed that I cannot see.
p bv 'therefore' here denotes consequence (p. 19). The

prophet's terror is induced as a consequence of the attitude
adopted by the Judaeans (p. 112). For the likely privative sense
of p , see p. 19.
1 For the use of the imperative to denote inevitable consequence, G.K. 1 iof.
2 Cf. the form nmn in Est. ii 18.
VERSES I - I O
121
Sixth century
There is no discernible change in the text. It is possible, however, that the speaker was at this time interpreted as the
Babylonians or their king. See the views of the rabbinic commentators, pp. i gf above.
Verse 4
Eighth century
My heart palpitated (wildly), convulsions have overwhelmed me. The
dawn that I longed for has been turned into trembling for me.
For a discussion of the meaning of the words, see pp. 2off.

Dawn, strictly the faint light of dawn, is a symbol for Isaiah
of the possibility of salvation for Judah, for which he had
fervently hoped. Since king and people had rejected reliance
upon Yahweh in favour of alliances with Babylon etc. the bare
hope which could be entertained has been forfeited.
Sixth century
The speaker is probably, as in the previous verse, interpreted as

the Babylonians or their king. It is possible that ywn *]W1 may
have been taken to refer (as a vaticinium ex eventu) to the feasting
in the city of Babylon mentioned as taking place on the night
before its capture. 1 See further on the next verse. In this case the
last phrase would be taken to convey the meaning ' My night of
rejoicing has been turned into trembling' (pp. 2if).
Verse 5
Eighth century
Men prepare weapons,* they set the watch.h *Arise, 0 princes, anoint
(your) shields9 or 'Arise, 0 princes, anoint a king.'0
a. nV#n TpV is read (see p. 27).
b. See pp. 25, 27. nnt? VDK is not part of the eighth-century text.
c. For the alternative translations, see pp. 25ff.
1 For the view of the rabbinic commentators that the reference is to Belshazzar's
feast, see p. 22. It is possible that the reinterpretation of Isa. xxi 5 in the sixth
century was a factor in the creation of the story of Belshazzar's feast in Dan. v.
122
ISAIAH XXI
The first two phrases are understood to be a description of the

anti-Assyrian military preparations taking place in Judah in
Hezekiah's days. The last phrase of the verse is an ironic
exhortation on Isaiah's part to the leaders of the state to take
part in such preparations. 'Anointing the shield' denotes the
preparing of them for use in battle (see pp. 25ff). The alternative translation, presented here as a possibility, might
constitute Isaiah's ironic exhortation to the leaders of the
conspiracy, to effect the enthronement in Babylon of Merodachbaladan.
Sixth century
Men prepare the tables, and light the torches. There is eating and
drinking. ' Arise, 0 princes, anoint {your) shields'
It seems likely that the words of the verse have been adapted
as a vaticinium ex eventu to reflect the circumstances of the fall of
Babylon and particularly the feasting at its centre reported also
by Herodotus and Xenophon (see p. 63 above). The imperative
phrase ' Arise, O princes,. . .' is presumably understood as an
ironic call to the Babylonians to defend themselves (cf. the
more descriptive phrase in Jer. li 30, where Babylon's soldiers
are represented as totally demoralized).
The phrase n w VDK may have found its way into the text
from the nearby Isa. xxii 13 where it occurs in the context of
a description of eighth-century Jerusalem's rejoicing at Sennacherib's failure to take Jerusalem. If this is so, it is possible
that the motif of a city feasting and at ease is derived from Isa.
xxii 12f and that Babylon's fate in Isa. xxi has been interpreted
in the light of it. Such an interpretation might constitute the
sort of exact retribution which was seen in the words of Isa. xxi
2a (see p. 118).1 What Jerusalem suffered, Babylon was to
suffer. Cf. Jer. li 35.
Three verses of Jer. If may possibly reflect the eighthcentury text of Isa. xxi (i.e. 1 42 and li 1 if). In 1 42, Babylon's
attackers are described as fully armed and ready for their work.
The participial form f n ? is there used. In li 1 if Babylon's
enemies are bidden in terms compatible with Aufforderung zum
Kampf to prepare for their attack. There are no exact verbal
i It is unlikely ex hypothesi that Isa. xxii was interpreted in the sixth century B.C.
solely by reference to events of the last decade of the eighth century.
VERSES I - I O
123
parallels (e.g. ^<W&n ipÔH; cf. Isa. xxi 5, ryDXn HDS) but the
sense is very similar.
It is likely that the references in Jer. li 39 and 57 to Yahweh's
imposing upon Babylon's warriors and princes a drunken
stupor are derived from such passages as Jer. xxv 15 where
Yahweh's cup of anger is referred to (cf. e.g. Rudolph). At any
rate the content of these verses is totally different from that of
Isa. xxi 5 - even on the view (which is here repudiated) that the
latter text in its eighth-century form referred to feasting.
It is very difficult to be sure how the phrase w&XTl HDS was
interpreted in the sixth century. On the theory here proposed it
is likely that the words were taken to denote some aspect of the
feast. If they were not interpreted as an indication that the
revellers posted watchmen, then the translation 'light the
torches' may be adopted on the grounds that it seems to have
some support in rabbinic tradition (see p. 25).
Verse 6
Eighth century
Surely thus has the Lord spoken to me:
him recount what he sees.9
Go, station a watchman. Let
The asseverative use o f ? (B.D.B., p. 472) together with the

reference to riDSfc suggests a particular link with the preceding
verse. The Judaeans set their watch in military preparation
(verse 5); but Yahweh too has stationed his watchman (viz. the
prophet) who will recount what is actually to happen. It is not
necessary to regard the watchman as an alter ego of psychological
theory (p. 28); rather the use of imagery is figurative (see the
views of ibn Ezra and Wildberger, pp. 28f above) and the watchman is none other than Isaiah himself, the recipient of the divine
command
Sixth century
There is no discernible change in the text. It is difficult to be

sure of the way in which the verse might have been interpreted
in relation to events of the sixth century. Jer. If provides no
clues, though the sixth-century prophecy of Hab. ii 1-8, which
concerns the imminent fall of Babylon, has striking points of
124
ISAIAH XXI
similarity both with this verse and verse 8 and may therefore
constitute evidence. In particular Hab. ii 1-3 (R.V.):
I will stand upon my watch, and set me upon the tower,
and will look forth to see what he will speak with (or by)
me and what I shall answer concerning my complaint. And
the Lord answered me, and said, Write the vision and
make it plain upon tables, that he may run that readeth
it. For the vision is yet for the appointed time, and it
hasteth towards the end, and shall not lie: though it
tarry, wait for it; because it will surely come, it will not
delay.
Habakkuk's concern was with the moral problem presented by
Yahweh's use of so cruel and inhuman a foe as Babylon for his
purpose in bringing Judah to punishment for her sins. The
prophecy in general has been characterized as 'the fruit of
religious reflection5,1 and there are indications that some verses
of chapter ii (apart from verses iff) are adapted from phrases
in the earlier prophecies of Micah and Isaiah.2 In view of these
considerations, and in accordance with the understanding of Isa.
xxi here presented, it is likely that Hab. ii if consciously reflects
(the eighth-century form of) Isa. xxi 6ff.3
On this view of the relationship between the two texts it
should be noted that Habakkuk is concerned by his use of the
image of the watchman to express his patient expectation of
Yahweh's answer to the problem that he faced. The answer was
to be committed to writing in such a way that the reader would
clearly and readily perceive its import, though its fulfilment
was to be delayed until the appointed time.
In the light of this prophecy, it may be suggested that Isa. xxi
6ff was seen in the sixth century B.C. as a prophecy granted to
Isaiah, the fulfilment of which was delayed until the present times.
That there is no reference in Habakkuk to the detailed contents
of the vision of Isa. xxi 6ff (i.e. to the terms M*l etc.) suggests
1 A. F. Kirkpatrick, The Doctrine of the Prophets, 3rd edn, (London, 1901), p. 273.
2 See S. R. Driver, The Minor Prophets (Century Bible) (Edinburgh, 1906), p. 59.
3 Wildberger explains the striking similarities between Hab. ii 1ffand Isa. xxi by
assuming that the motif of the watchman was a customary way of understanding
the prophetic office in prophetic circles. The possibility of any (literary) dependence is precluded for him because, of course, he attributes Isa. xxi to a date in
the sixth century B.C.
VERSES I - I O
125
that the important element in the original prophecy was

thought in the sixth century to be the motif of the watchman (or
prophet) whose function was patiently to await the fulfilment
of Yahweh's words, however great the apparent delay.1
Whether the apparent distinction between the prophet
('VK) and the watchman (nDXan) was itself the object of sixthcentury interpretation can be only a matter for conjecture.
However, if the essential element was understood to be the
inexorable fulfilment of prophecy, even after long delay (cf.
verse 8), it is probable that the watchman was, in the eyes of
the sixth-century interpreters, either one of themselves or even,
perhaps, the prophet Habakkuk himself. (Cf. Rashi, p. 28
above.)
Verse 7
Eighth century
'And when he sees* two-horse chariotsh>2. . . then he shall pay very
great attention.'
a. It is natural to suppose that verse 7 constitutes a continuation of the command of verse 6 (p. 30).
b. While it is reasonably certain that nDI at this earlier date
denoted c chariots' rather than ' riders' it is not certain whether
a single chariot is meant or whether the word denotes collectively 'chariots'. However, while both usages are clearly
attested (see B.D.B., p. 939), the balance of probability may be
said to favour the latter view.
A number of biblical references refer to the might of Assyria
in terms of her chariots (2 Kings xix 23 = Isa. xxxvii 24;
Nahum ii 4, 5, 14); in particular Isa. xxii 7 makes use of an
expression very similar to that used here in order to depict the
eighth-century invasion of Judah by the Assyrian army. It may
be assumed, therefore, that the prophet (or else the watchman)
1 Where Habakkuk is concerned it is even possible that his emphasis upon a clear
written message represents a desire for elucidation of the obscurer elements in
Isaiah's prophecy.
2 The accentuation of the M.T. is commonly emended so that the athnah is placed
beneath the word O'ETID; see e.g. B.H.(3) and Wildberger. Because the authenticity
of the similar phrase in Isa. xxii 6 (D-SHD DIN 3D"i) is questioned, I have not
considered it as evidence here.
126
ISAIAH
XXI
saw in his vision chariots and that chariots denoted for him the
might of Assyria.1
c. The words *?) M^ *T)n M l are taken to be a later addition
to the text; see below. Here it may be noted that the four words
are not present in verse 9. There is no evidence that the term
M l was used to denote (baggage) wagons as opposed to chariots,
and that camels and asses should have been employed to pull
chariots is surely unthinkable (p. 32).
Sixth century
And he will see* mounted cavalry (Ml), horsemen in pairs (TDS
men riding on asses, men riding on camels (M*I Ti&n M l
) . c And he will pay very great attention.
a. Because now the prophecy is understood to concern the
(contemporary) sixth-century fall of Babylon, the words are
seen as a prediction (uttered by Isaiah). The watchman, now
not Isaiah but a contemporary, will see what Yahweh had
predicted to Isaiah that he (the watchman) would see.
b. It is likely that in the sixth century the word M l was understood to denote mounted cavalry (cf. ibn Barun's philological
explanation, p. 30 above) and that D'WIS) "IS was taken to
mean pairs of horsemen (pp. 3off). Such an interpretation of
the words in question can be traced back at least as far as the
LXX (pp. 3of above).
In this connexion it is suggested that the puzzling phrase
BTK M l in verse 9 may be illuminated. If M l was interpreted
in the sixth century as (human) riders rather than as chariotry,
that interpretation may have thereafter been fixed by the
addition of the word BTK as a gloss to the word Ml. 2
On the view outlined above, readers of the sixth century
would have found in these words a reference to cavalry and
mounted troopers, a feature of warfare conspicuously associated
with the Persians.3
c. The phrase bbl 201 n&n M l cannot, it is submitted, mean
anything other than riders on asses and camels. For this reason,
1 Cf. the reliefs of Tiglath-pileser I I I and Sennacherib displayed in public in the
British Museum. Both are there depicted as occupying two-horse chariots.
2 Cf. D.B.D.'s observations on the phrase mn 331 in Isa. xxii 6; here the L X X
renders: <5cvcc|3<5cTGa ocvOpcoiroi. For Procksch's similar (but not identical) views,
see pp. 34 f above.
3 See p. 32 and Wiseman, Peoples, p. 335.
VERSES I - I O
127
and in the light of the references in Herodotus and Xenophon

(p. 64 above) to the use made by the Persians of asses and
camels, it is here suggested that the phrase was inserted as a
sixth-century vaticinium ex eventu designed to facilitate the use
of the text as referring to the Persians.
Verses 8 and 9a
Eighth century
And he called out 'I was looking*. . . b and, behold, there came
a. The M.T.'s m*iX is taken to be a corruption of an original
T^yfr (see p. 34; so Marti). The verb, used in the imperfect
to express continued action in the past (G.K. 107b), bears this
meaning, as frequently, when used absolutely and followed by
mm (B.D.B., p. 907 6c). As Marti observes, the conjunction of
the verb n*n with mn provides precisely the form in which
Zechariah introduces his visions (Zech. ii 1; iv 2; v 1, 9; vi 1).
b. The words from HD2& bv to the end of verse 8 are understood
to be a later addition to the original text. With Marti two
considerations may be advanced in support of the view that
they form no part of the original text:
(1) The structure of the verse requires two, rather than three,
two-beat parts; and to arrange it in the latter way would have
the effect of spoiling the parallelism.1
(2) The sense of the narrative runs on better without the
words (see above) and their presence in the oracle (with
consequent corruption of the original text) can be explained by
reference to later texts (see below).
Sixth century (verse 8)
And the seer* cried: 'All day long I stand on the Lord's watchtower,
and I am stationed every night upon my watch.'b
a. The reading of iQIs^ is here adopted. On the present view

of the oracle, the original reading n*HK will have been modified
in the sixth century to n*nn2 in order to accommodate the
addition which follows. The reading provides a subject for the
1 Marti's second argument (viz. that the words are at odds with the content of
verse 4) is not here reproduced because I take a different view of verse 4.
2 Goshen-Gottstein inclines to the view that the reading m gave rise to the
reading nxin in iQIs a .
128
ISAIAH XXI
verb Kip*!, and the phrase n ^ n fcnjn ('and the seer cried5)
introduces the direct speech which follows it. The participial
form rttOH depicts the one looking out (i.e. the watchman), but
it also describes his function as the receiver of a revelation from
God (B.D.B., p. go6f (para. ib)).
The reading of the M.T. mN may be regarded as an early
corruption of HK^n1 made either accidentally by scribal error or
deliberately in order to provide a particular interpretation of
the watchman's vision. In connexion with the second of these
possibilities the reference may be made to the avenging wrath
of Yahweh against his enemies (here Babylon), depicted under
the figure of a lion; cf. Isa. xxxi 4; Jer. iv 7, xlix 19 = 1 44.
b. The direct speech of the seer may be illuminated by
reference to Hab. ii iff and to Isa. lxii 6. In the former text
Habakkuk waits patiently at his prophetic station for the time
when Yahweh will vouchsafe an answer to his problem. In Isa.
lxii 6 watchmen, whether (as I think) prophets or angelic
beings, are depicted as waiting constantly on Jerusalem's walls
for the dawn of the day of salvation.
In the light of these two texts (which are verbally very
similar to the direct speech of this verse) it may be assumed that
Isaiah's original and somewhat simple motif of the watchman
has been modified and interpreted in the sixth century by the
addition of the words in direct speech. Thereby, the watchman
becomes for the sixth century the later prophet (nK^n) whose
function is patiently to await and to expound the fulfilment of
Yahweh's word in his own times.
On this understanding of the texts, the detailed relationship
between them will be as follows:
The original text of Isa. xxi 6-8 is the source of the sixthcentury motif (indicated by Hab. ii iff as well as by the later
text of Isa. xxi) of the prophet as a watchman continually
waiting for Yahweh's revelation). If Hab. ii iff is an echo of the
original text of Isa. xxi 6ff then the later text of Isa. xxi 6ff
reflects a similar sixth-century reinterpretation of that earlier
text. Isa. lxii 6 is in turn a post-exilic2 echo of the sixth-century
1 Alternatively, we may suppose that the original nxix was preserved in some
older manuscripts which survived the sixth-century revision and it is this reading
which lies behind the erroneous (or deliberate) reading mx of the M.T.
2 For a date in the mid fifth century B.C., see Marti, pp. 389X
VERSES I - I O
129
texts (with their motif of the prophet as a watchman continually

waiting for Yahweh's revelation). In Isa. lxii 6 the function of
the watchmen (now plural) is to work and pray tirelessly for the
restoration of Jerusalem.
Verse 9
Eighth century
And behold, there came: two-horse chariots,'* And he responded by
saying:10 ' Fallen, fallen is Babylon and all the images of her gods are
shattered to the ground?0
6
a. DÊHD *7&2 M*l is regarded as the eighth-century text, the

wordttrKbeing a sixth-century addition. See above on verse 6.
b. For niS7 with -)&K in the sense * to respond to an occasion, to
speak in view of circumstances', see B.D.B., p. 773 2a. The
prophet, or watchman, himself responds to what he has seen by
uttering his interpretation of it.
c. The reference to the shattering of the Babylonians' idols is
unlikely to be a vaticinium ex eventu. Rather Isaiah saw the
inevitability of Babylon's wholesale destruction at the hands of
the Assyrians. That his estimate was accurate may be inferred
from the events of 689 B.C., when Sennacherib actually invested
Babylon and destroyed her gods (p. 109).
Sixth century
'And behold, there comes mounted cavalry, horsemen in pairs.'* And he
answered and said: ' Fallen, fallen is Babylon and all the images of her
gods are shattered to the ground'
a. See on verse 7, eighth and sixth centuries (pp. I25ff).

As the prophet Isaiah is regarded as appointing a watchman
who will await the fulfilment of his prophecy (cf. on verse 6 sixth century), the subject of the verbs 'answered' and 'said' is
most likely to be Isaiah himself. In other words, the prophecy
is understood to mean 'When you {sc. the watchman of the
future) see the sign of mounted cavalry, my prophecy "Fallen is
Babylon" will be fulfilled.'
Evidence that the phrase c Fallen is Babylon' was interpreted
by reference to the sixth-century fall of Babylon may be deduced
from the presence in Jer. li 8 of very similar words. Again, in
130
ISAIAH XXI
Jer. 1 2 reference is made to the destruction of the Babylonian

gods in language markedly similar to that of Isa. xxi 9 (see
p. 114).
Verse 10
Eighth century
0 my people threshed* and winnowed on my threshing-floor* what I
have heard from the LORD of hosts, the God of Israel, I have declared
to you.
a. Lit. 'my threshing'.
b. Lit. 'son of my threshing-floor, (see p. 38).
The metaphor is used proleptically of Judah's fate.1 By
becoming involved in foreign alliances, Judah was to incur the
inevitable punishment of having to bear the ruthless oppression
of Assyria. This is the message that Isaiah has learned in his
vision and he recounts it to his hearers (cf. p. 112). The note of
sympathy which characterizes the verse is in accord with the
horror experienced by the prophet at the onset of his vision.
Both feelings are experienced by the prophet by reason of his
harsh vision (verse 2).
Sixth century
There is no discernible change to the text. The evidence of

Jer. li 33 suggests that for the sixth century it was Babylon
which was thought of as threshed and winnowed: 'The
daughter of Babylon is like a threshing-floor at the time when it
is trodden. . .' (R.V.). Yet the text with its sympathetic tone
and its emphasis upon its being communicated by Yahweh to
his people (DD1?) suggests that for the sixth century, as for the
eighth, it was Israel who was addressed as threshed and
winnowed (cf. p. 39). We may suppose that for the later
generation recollection of Zion's anguish prompted the
expectation that the agents of that anguish would in their turn
suffer it themselves (see p. 119).
1 If Mic. iv 13 (in which it is said that the daughter of Zion will thresh many
nations) is quoted by Micah from the words of an eighth-century prophet of
national well-being, then Isaiah like Micah may, by use of the metaphor, be
deliberately seeking to reverse the arguments of such contemporary false prophets. For recent views on Mic. iv i iff, see A. S. van der Woude in V.T., 1969,
pp. 2526, Rudolph, K.A.T. xm 3, p. 92, and Lindblom, Prophecy in Ancient
Israel (Oxford, 1962), p. 25 m.
Verses 11-17
The impression gained by a number of scholars that the Dumah
oracle and the Arabian oracles reflect the same general
historical circumstances has been noted (pp. 9iff). By way of
summary, three considerations may be advanced in support of
this view: first, both are concerned with the geographical area
to the south of Judah; secondly, they are both marked by
similar dialectal features; thirdly, they are juxtaposed.1
Again arguments have been advanced above to suggest that
the oracles pre-date the sixth century B.C.; in particular (pp.
93ff) there is evidence that the sixth-century Edom oracle
of Jer. xlix yffknew and made use of Isaiah's (Arabian) oracle.
Further (pp. 98f, 102), it is argued that the epilogues to the
Moab oracles (cf. Isa. xvf) and to the Arabian oracle come
from the same hand and derive from the sixth century; one of
them (Isa. xvi 13) explicitly states that the material to which it
is appended was 'the word that the Lord spake concerning
Moab in time past'.
Finally, literary considerations and in particular comparison
of the Moab oracles (Isa. xvf) with Isa. xxi 1 iff suggested that
these oracles reflect the period of the Assyrian attacks on
southern Palestine and Arabia at the end of the eighth century
B.C. (see pp. g8ff).
In order to gain further insights into the oracles of Isa. xxi
1 iff it is necessary to turn again to the evidence provided by the
sixth-century oracle of Jer. xlix yf[. The evidence that the latter
chapter affords is likely to be of a sort with that which Jer. If
yielded for the elucidation of Isa. xxi 1-10.
1. Jer. xlix 7ff is explicitly and clearly an oracle directed
against Edom. It contains in verse 8a a summons to the
1 The LXX has no title for verses 13fF and consequently appears to regard verses
1 iff as one oracle (see p. 53).
132
ISAIAH XXI
inhabitants of Dedan to flee far off, apparently from Edom, in

order to avoid the calamity which is about to fall upon that
country. Although the verbs are not the same as those used in
relation to the caravans of Dedanites in Isa. xxi 131 the meaning
is clearly so similar that it is natural to assume that Jer. xlix 8a
consciously reflects Isa. xxi 1 iff in exactly the way that Jer.
xlviii reflects Isa. xvf.2
From the considerations so far adduced I conclude that
Jer. xlix 7ff constitutes evidence that the Dumah oracle was
understood in the sixth century B.G. to be an oracle directed
against Edom and not against the Arabian Dumah (Dumat al
Jandal). Further, because there is within the oracle itself
(Isa. xxi 11) a reference to Seir, often in the O.T. a synonym for
Edom, and, because there is no evidence that the reference is a
later addition or correction to an earlier reading {contra
Lohmann; see p. 83 above), I further conclude that the oracle,
as received by the author of Jer. xlix (and therefore originally),
was an oracle directed against Edom (and not against Arabian
Dumah). Secondly, because the Edom oracle of Jer. xlix contains within itself a reference to the Dedanites, it is possible to
conclude that in the sixth century the Edom oracle of Isa. xxi 11
was understood to contain within it a reference to the Dedanites.
In support of that conclusion there is the corroborative evidence
of the LXX (see p. 53 above). Accordingly we may state that
the Edom oracle of Isa. xxi 1 iff was received in the sixth century
as a single oracle which included references to the Dedanites,
rather than as two separate oracles ('Dumah' and * Arabia').
On the other hand, it is possible to argue that to the author of
the Edom oracle of Jer. xlix two oracles of Isaiah ('Dumah' and
'Arabia') were so closely linked that he felt free to modify them
in order to achieve for himself a single oracle against Edom
which included reference to the Dedanites. The balance of
probability may be held to swing in favour of the former
conclusion in view of the further consideration that Jer. xlix 28ff
contains oracles against Arab peoples (Kedar and Hazor); if
1 For the possibility that they do in fact reflect the verbs of Isa. xxi, see p. 139
below.
2 Rudolph (p. 281) has shown that the author of Jer. xlviii at times followed his
older sources very closely and at others treated them with very considerable
freedom. With this conclusion my estimate of the way in which Jer. If has used
Isa. xxi 1-10 seems to be consistent.
VERSES I I - I 7
I33
the author had wished to refer to earlier material concerning the

Arabs, he would have been most likely to include it precisely
here rather than within an oracle against Edom.1
2. The Edom oracle of Jer. xlix 7ff clearly reflects the hatred
felt for Edom by the Jews after 587 B.C.2 Such feelings are not
characteristic of earlier prophecies against that country, and
Jeremiah's original sayings about Edom, for example, do not
display this attitude. 3 In the Edom oracle of Isa. xxi 1 if there
is no apparent trace of hatred; rather a sympathetic note has
usually been detected (see pp. 75ff).
These considerations support the conclusion, reached on
other grounds, that Isaiah's Edom oracle clearly antedates the
sixth century B.C. On the other hand, it is likely that the oracle
would have been the object of the sort of reinterpretation that
has been suggested for the oracle of verses 1-10. It is suggested
below that just such a reinterpretation may be detected, and
that in particular it was achieved by creating out of the single
Edom oracle two separate oracles - one concerning Edom and
the other concerning Arabia.
3. The Edom oracle of Jer. xlix contains material which is
common to that found in the sixth-century (anti-Edomite)
prophecy of Obadiah. The balance of scholarly opinion may be
said to favour the view that the two oracles are dependent upon
a common source - whether that source was written or oral
(seep. 93n).^
Obad. 5f is clearly a variant form of what is found in Jer.
xlix gf. It is likely that Obadiah here preserves a better as well
as an earlier form of the text, for, as Rudolph observes,4 the
text of Obad. 5 fits the general context well, whereas that of
Jer. xlix 9 is in many respects unsatisfactory in the context.
1 That virtually the same verbs are used in relation to the inhabitants of Hazor
(verse 30) as are used in verse 8 of the Dedanites is not here significant. The
important point is that Dedanites are not mentioned within the Arabian oracle
of Jer. xlix 28ff, whereas they are within the Edom oracle.
2 See e.g. Rudolph, pp. 291 ff; a particularly clear example is the use to which
words originally used of Babylon (Jer. 1 44-6) have been put in Jer. xlix 19-21.
Now the fall of Edom rather than that of Babylon causes the whole world to
tremble.
3 See Haller in Z.A.W. Bei. 41 (1925), iogff, and Weiser, p. 407. Jer. xlix 22 may
here be cited as an example of the factual as opposed to the gloating notices of
Edom's discomfiture.
4 K.A.T. XIII 2, p. 297.
134
ISAIAH
XXI
The Hebrew text of Obad. 5 runs:

lair Kibn (nivaii 7*x T\b*b m w ox) i1? IKS D'aa ax
K
(6)
B.H.(3) followed by Wolff is inclined to the view that the words

between the brackets, being absent in Jer. xlix 9, are a gloss.
On the other hand Rudolph does not take this view, and, by
inverting the order of the phrases underlined, he is able to
restore the text adequately. Against the view of B.H.(3) and
Wolff, it should be noted that the word nVV occurs also in the
text of Jer. xlix 9 (though here after the word Dâtt). This,
understood in association with the impression reached by
comparing the two texts, suggests that here as elsewhere we are
likely to be concerned with a variant text rather than with a
gloss.
The Hebrew text may be rendered (after the R.V. but
following Rudolph's rearrangement):
If thieves came to thee, if robbers by night, would they
not steal till they had enough ? How art thou cut off!
If grape-gatherers came to thee, they would not leave
gleaning grapes!1 How are the things of Esau searched
out! How are his hidden treasures sought up!
In this passage there are three Hebrew words (overlined)
which are found in the Dumah oracle of Isaiah, viz. nVV
nrpJina, (cf. nDTT in Isa. xxi 11) and most important, the verb
1M3 from the root run (cf. vsn fraan). Similarly, Obad. 5
and the repeated words of lament in Isa. xv 1 have three words
in common. Isa. xv 1 runs: nT2 aKifc *)$ TlW V*?a *O.
A comparison then of the three texts suggests the following
conclusions: Obad. 5f (cf. Jer. xlix 9f) reflects the Dumah
oracle of Isa. xxi and at least the first verse of the Moab oracle
of Isa. xvf.2 That it reflects specifically the Dumah oracle is
likely in view of the fact that both use the word roa in apparently the same way. Further, this interpretation is consistent with
1 Reading sV for xVn with Jer. xlix; cf. Rudolph's slightly different treatment.
2 The mention of grape-gatherers etc. may be a free reflection of the mention in
Isa. xvi 8ff. of the spoiling of Moab's vines.
VERSES I I - I 7
I35
the fact that Jer. xlix 8, another form of the same Vorlage,
shows knowledge of Isa. xxi 13 (and probably 12b - see below).
Because the Vorlage of Obad. 5f and Jer. xlix appears to
know both Isa. xv 1 and the Dumah oracle and because the
latter texts contain two words in common, viz. (n)V11? and
forms of the root nT, I infer that the Moab oracle and the
Dumah oracle are of a sort and that the former may be used to
illuminate the latter. With this inference the considerations set
out above (pp. g8ff) are consistent: the two texts seem to derive
from Assyrian times when Moab, Edom (and Arabia) were
objects of Assyrian aggression and were shown at least a
modicum of sympathy by the Judaean prophet. There is no
reason to suppose otherwise than that the prophet was Isaiah
ben Amoz.
On this view of the text, Isaiah's Edom oracle is likely to
come from the same time as his oracle concerning Babylon (verses
1-10), i.e. during the last years of the eighth century B.C. In
this connexion it may be recalled that both Edom and Moab
were tributary to Tiglath-pileser III in 724 (A.N.E.T., p. 282).
In 713 Edom and Moab were drawn into the revolt against
Sargon led by Ashdod (Luckenbill, 11, no. 195). The Assyrian
king invested Ashdod and Gath (C.A.H., m, p. 58) and, as a
consequence, Moab and Edom paid tribute and thus avoided
invasion themselves.
The death of Sargon in 705 B.C. was the cause of widespread
revolt in Syria-Palestine, and Hezekiah ofJudah was one of the
principal conspirators. Sennacherib's campaign of 701-700
was directed against this rebellion in which Edom and Moab
clearly played a part. In common with certain other lesser
conspirators, they seem to have paid (heavy) tribute and
consequently yet again they avoided suffering an Assyrian
invasion (Luckenbill, 11, no. 239; cf. C.A.H., in, p. 72).
A Hebrew ostracon recently discovered at Arad 1 and dated
701 B.C. appears to be concerned with communications between
Edom and the king of Judah during the time of Sennacherib's
campaign. Unfortunately the text is too fragmentary to throw
any very particular light on the history of the time or upon the
text of Isaiah. On the other hand, it is interesting that the name
1 See Aharoni.
I36
ISAIAH XXI
'Edom' occurs no less than three times in the text and explicitly alongside mention of the king of Judah. We may
conclude that communication with Edom was of some importance to Judah and to Hezekiah and that for Isaiah to have
delivered an oracle concerned with Edom was in keeping with
his having delivered oracles concerned with the other nations
involved in the rebellion against Assyria.
If Isaiah's Edom oracle has been correctly dated in the
closing years of the eighth century, then it is not to be regarded
as a vaticinium ex eventu. Rather, like the Babylon oracle of
verses 1-10 it sets forth by way of warning the consequences of
involvement in the anti-Assyrian conspiracy. The description
of fugitives fleeing from Edom to Arabia before the vehemence
of war is sufficiently general to imply that the author was able
to imagine the consequences of rebellion. The cipher of the
watchman, on the other hand, seems to reflect a particular
aspect of relations between Edom and Judah, probably their
mutual involvement in the planning of rebellion (see below).
Comparison has been made between the Moab oracles of Isa.
xvf and Isaiah's Edom oracle (see pp. 98ff) and on the basis of
it the suggestion was made, albeit tentatively, that both oracles
belong to the Assyrian period. It has recently been suggested
that the core of the Moab oracles1 is based upon earlier nonIsaianic laments over Moab occasioned by such attacks upon
her as those perpetrated by the 'men of Gidir land'. 2 The latter
are thought to be connected with the semi-nomads east of Moab
in the latter part of the eighth century. It is not possible here to
attempt a full evaluation of the Moab oracles; but if this
assessment of one of them is correct there is no reason why
Isaiah should not have used and redirected them to convey his
message of warning in the period of the revolt following
Sargon's death in 705.3
1 See esp. xv 1-8 and xvi 6-11.
2 For the source of our knowledge of this attack, see H. W. F. Saggs, 'The Nimrud
Letters, 1952 - Part 11', Iraq 17 (1955), 126-60. See also Wildberger, pp. 597,
6o6ff, and Wiseman, Peoples, p. 239.
3 For this sort of approach, see e.g. the admirably cautious and balanced statement of W. Rudolph, H.S.S. Wildberger's contention (p. 604) that no pre-exilic
prophet ever cited and commented upon earlier oracles (and that consequently
Isaiah could not have done so here) is somewhat dogmatic. As Rudolph observes,
if Isaiah preached the imminent fulfilment of older oracles, that does not require
that he identified himself with all their contents, nor does it necessarily detract
from his general originality and greatness.
VERSES
I I-I 7
137
In the Edom oracle the metaphor of the watchman was, as

Lohmann has indicated (pp. 838*above), probably taken over
by Isaiah from everyday life and used (probably as a cipher) to
present his particular message concerning Edom.1 Similar
appropriation may be seen in, for example, the use made by
Isaiah (v 1-7) of the song of the vineyard or (by a later prophet
in xxiii 16) of the harlot's song.

Verse 11
Eighth century
Oracle concerning Dumah* They callh to me from Seir, 6 Watchman',0
either ''what news of the night? Watchman, what news of the night?'
or d 'How much of the night remains? Watchman, how much of the night
remains?'
It is assumed that the image of the city watchman (p. 43)

who replies ostensibly to a question concerning the night
constitutes a cipher by which the prophet refers to the historical
situation. There is no need to assume that the oracle is other
than a contrived or imaginary prophetic vision, though the
anti-Assyrian alliance of Hezekiah's time presumably necessitated communication between Edom and Judah. 2 The speech is
purposely dialectal and the language evoked is likely to be
Edomite or (international) Aramaic. 3
a. nVT W& is read. The term Dumah denotes 'ruin, destruction ' and is used proleptically of the fate which Isaiah believes
will or may befall Edom if she persists in her rebellion against
Assyria (jrsnn DK). The word is to be derived from the root
H*T4 and reflects the use to which the word is put in Isa. xv 1,
viz. that Ar (or Kir or Moab) 5 is destroyed, ruined on the night
of sdd. On the other hand, its form and sound are evocative of
1
2
3
4
I am not committed, of course, to Lohmann's detailed conclusions.

Gf. the contents of the Arad ostracon of 701 B.G. (pp. i35f above).
See above, p. 79.
B.D.B., p. 198; see further G. R. Driver, 'A Confused Hebrew Root' in IDO
TO^TIW (Jerusalem, i960).
5 For the different views taken of the syntax of the line, see the standard comment-
I38
ISAIAH
XXI
the word Edom (on**). The use of the word nV7 to denote
Edom may be compared with Isaiah's use of the term ]VTri *n
in chapter xxii to denote Jerusalem and especially if, as Driver
supposes, the word is here cognate with Arabic hzy and denotes
* valley of calamity'.
b. Lit. ' (Someone) is calling.' The participle is understood to
have an impersonal force. Cf. with Delitzsch xxx 24, xxxiii 4
and Luther's translation: 'man ruft'.
c. The watchman ( i ^ ) is Isaiah and the term may be regarded
as similar to the prophet's description of himself as nosa
(watchman) in the previous oracle. His use here of the term
is an accommodation to the nature of the oracle as a cipher.
d. For these alternative ways of understanding the question,
see pp. 43fF. The question refers to the night, a symbol of
Assyrian oppression. More particularly, it is suggested that the
night is the Tittf V*1? of Isa. xv 1 (cf. Obad. 5), the night of the
*n^ (sic - the word is pointed *??# in Isa. xv 1), and that -n#
denotes, as in Isa. xxi 2, the Assyrian devastator. The further
question whether the night is an image of general Assyrian
oppression or a particular historical example of that oppression
cannot be answered with certainty. The repeated question of
the Edomites portrays (however precisely it is understood)
their longing for the morning of freedom from Assyrian domination.
Verse 12
Eighth century
The watchman said: 'If morning has come, then so has night.* If you
are involved in aggression then you will be the object of aggression 'h
a. For simple perfects (here a perfect and an implied perfect QJ)) in a conditional sentence, G.K. 159I1. This understanding
yields a better sense than the rather dull and straightforward
'Morning has come and with it night.'
The morning is taken to mean the opposite of the night of
oppression referred to in the question. Whether it denotes
1 Cf. e.g. the spelling of the word in the Assyrian texts udumu and in the LXX
I8oupiona.
VERSES I I I 7
139
generally longed-for freedom or whether a particular historical

attempt to achieve freedom cannot now be known.1
b. For the M.T.'s ^V? a pual imperative form is read: T*?a.
The imperative conveys a 'distinct assurance or threat' (G.K.
HOC). For a discussion of the verb nsn, see pp. 48ff. It was
there suggested that the word had a military or hostile connotation and it was noted that in Obad. 6 the niphal of the verb
meant that Edom's wealth was exposed and plundered. The
particular sense required in this context where no object is
provided for the active verb is 'aggressive (military) action'.
The active verb ' If you are involved in aggression' is seen to
match the longed-for 'day' and the passive 'you will be the
object of aggression' matches the menacing nV? DM 'then so has
night'.
Verse 13
Eighth century
Turn back, go into the scrub; you must stay2 amongst the Arabs.
The verbsTONlilt? which in the M.T. form the end of verse

12 are taken with the words msn 15P2 which in the M.T. form
the beginning of the Arabian oracles. In Jer. xlix 8, where the
Dedanites are summoned to flight, the two verbs employed are
ttsn 103. The latter verb is usually regarded as a hophal
imperative of the verb niD with the sense 'be ye turned back'. 3
It is at once apparent that the word is virtually synonymous
with the verb yym used in the phrase vnK lilt? in Isa. xxi 12,
and we may conclude that here as elsewhere Jer. xlix reflects
Isa. xxi. Indeed, the phrase ^DH 102 may constitute a
rendering of the dialectal phrase TnN 'Dlttf. At any rate Jer.
xlix 8 is an indication that the words rr>K 131^ denote flight
and that they belong with what follows them rather than (as in
the M.T.) with what precedes them.
The title msn m is understood to be a sixth-century (or
later) addition to the text (see below). A consequence of this
1 If the latter meaning is the correct one, it is tempting, but probably fanciful, to
see in the Arad ostracon (pp. i35f above) a reference to the Day of Freedom; line
11 contains the phrase rdym and Aharoni, comparing Judg. xix 11, has suggested
that the phrase may denote the passing (lit. 'going down') of a day.
2 For the use of the jussive to convey advice, see G.K. 109 b.
3 B.D.B., pp. 8i5f; cf. G.K. 4 6a.N.
140
ISAIAH
XXI
understanding of the text is that parallelism is restored to

verse 14 (see above, pp. 57^ and below).
The verse as restored and translated above forms the next
part of Isaiah's Edom oracle. As a result of their rebellious
activities, the Edomites will be reduced to fleeing their country
for the safety of Arabia, just as the Moabites in Isa. xv 4ff are
obliged to flee to Edom. Some echo of this form of the text may
be detected in Obad. 7 where Edomites are spoken of as being
* driven out to their border'. 1
Verse 14
Eighth century
0 caravans of Dedanites, bring water to meet the thirsty; 0 inhabitants
of the land of Teima, meet* the fugitives with food.
a. The pointing is emended so that an imperative form is read.

The requirements of parallelism suggest that in the original
text the caravans of Dedanites, like the inhabitants of Teima,
were urged to show compassionate hospitality to the Edomite
refugees. It is suggested below that in the sixth century B.C. the
Dedanites themselves were understood to be the refugees to
whom the inhabitants of Teima (and others) were urged to
bring sustenance, and consequently a separate oracle (TIBS?
'Against the Arabs') was formed. Apart from the argument
concerning parallelism, we may note that in Ezekiel's Edom
oracle (xxv I2ff) it is said that Edom's desolation will stretch
from Teman 'even unto Dedan' (R.V.); it is possible that this
sixth-century text (see Zimmerli, pp. 588ff), in which Dedan is
mentioned as a far-off place where Edomite refugees fall by the
sword, preserves an echo of the original text of Isa. xxi 14, where
Dedanites, so far from fleeing themselves, represented shelter
to which refugees from Edom had recourse.
Against the view that the Dedanites were, in the original
context of Isa. xxi 13ff, pictured as themselves refugees is the
consideration that it would be curious to advise Dedanite
1 It should here be noted that the preceding verse is that which contains a form
of the verb ns73 (see above), and that the second part of Obad. 7 contains a
reference to bread (cf. Isa. xxi 14); that the wording and content of Obad. 7b
differs considerably from Isa. xxi 14 does not necessarily tell against literary
dependence (see above).
VERSES I I - I 7
I4I
caravans to lodge where they were accustomed to travel and

presumably to lodge viz. among the nomadic peoples of North
Arabia
Verse 15
Eighth century
For theyfleefrom the sword, from the strung bow andfrom the vehemence
of war.
The verse states the reason for the flight of the Edomite
refugees from their homeland into Arabia.
Sixth century (verses n-15)
The evidence of Jer. xlix was understood to indicate that the

text of Isaiah which it reflects contained a single oracle against
Edom in which there was mention of the Dedanites (pp. I32ff).
Secondly, the absence from Isa. xxi nff of the customary antiEdomite polemic was noted.
When Isaiah's Edom oracle was received and interpreted in
the sixth century, it is likely that it was understood in the light
of the prevailing attitudes and of more recent historical events.
All the sixth-century Edom prophecies which we have are
marked by uncompromising hatred for Edom on the part of the
Jews. As is well known, this attitude seems to have become
prevalent shortly after 587 and the fall of Jerusalem.1
The evidence of Jer. xlix ff suggests that Nebuchadrezzar's
campaign against Judah's neighbours (599-587) made a very
considerable impression on the Jews. In the course of those
campaigns Nebuchadrezzar's army is known to have marched
into the desert and plundered the Arabs.2 It is possible that
Edom was not left unscathed by such operations. In any case
Edom was besieged by the later Babylonian king Nabonidus in
the course of his march into Arabia in 554-3
1 For recent evaluations by Bartlett of Edom's role in the events leading up to the
fall of Jerusalem, see Wiseman, Peoples, pp. 24 if, and Bartlett, 'From Edomites
to Nabataeans', p. 27n28.
2 In 599-598 according to Wiseman, Chronicles, p. 70; with this notice, Josephus's
mention of an Arabian campaign should be compared; see above, p. 92.
3 C.A.H., in, p. 405. For further references to the debate concerning the identification of 'Adummu', see p. 8gn above. Lindsay, followed by Bartlett ('From
Edomites to Nabataeans'), argues that it was this campaign which brought an
end to the Edomite kingdom as such.
I42
ISAIAH XXI
It is now suggested that in the sixth century B.C. Isaiah's

Edom oracle (received initially as one oracle) was understood
in the light of the Babylonian attacks upon Arabia and Edom
of the first half of that century. In that the prophecy was seen
already to reflect recent history, its words could be taken to
form the basis for new prophecies of the longed-for total
destruction of Edom (e.g. Jer. xlix).
In support of the conclusion that Isaiah's words were taken
to refer both to Edom and to the Arabs, it should be noted that
the sixth-century Jer. xlix 8 contains a summons to the Dedanites to flee from Edom in order to avoid imminent and certain
calamity. Jer. xlix 30 contains virtually the same summons, but
this time it is addressed to the inhabitants of (Arabian) Hazor
and explicitly forms part of an oracle concerned with Nebuchadrezzar's attack upon the Arabs (and in particular upon
Kedar). Thus, Jer. xlix indicates that words derived ultimately
from Isa. xxi 1 iff were later used both in an oracle concerned
with Edom and in one concerned with Arabs.
The mode in which Isaiah's oracle was itself similarly
reinterpreted may be reconstructed as follows: first, the vocative
' caravans of Dedanites' was construed not with the verb that
follows it (D* rnn) but with the jussive and imperatives that
precede it (irVn and vnK yyw - see below). The effect was to
initiate the creation of a separate oracle concerned with the
Arabs. To have understood the text in this way (with Dedanite
rather than Edomite refugees) would have precluded the
detection in the sixth century of the sentiment of sympathy
towards Edomite refugees, a sentiment which would ill accord
with the vengeful spirit concerning Edom typical of the oracles
of the time. Secondly, at this time, the supplementary material
concerning Kedar (verses i6f) was added. Thirdly, and probably at a later stage, the title msn m was added to the text,
but in such a way that it separated the phrase YTIK "aw from
its proper place,1 and caused it to be included in the watchman's
reply of the Dumah oracle.
Only the smallest traces of the Dumah oracle are to be found
in such sixth-century texts as Obadiah and Jer. xlix. In particular there appears to be no reference to the metaphor of the
1 That Jer. xlix 8 preserves traces of the earlier way of construing the phrase has
been noted above.
VERSES I I - I 7
143
watchman and his questioners. It is therefore not possible to

attempt a reconstruction of its detailed interpretation at this
time. Indeed, perhaps its very obscurity and particularity
precluded detailed reinterpretation. On the other hand, it
seems clear that certain of its words and phrases were adopted
in sixth-century oracles concerning Edom. These may be
listed:
1. nDVT. See Obad. 5 where it is said of Edom ^57? TS
' How art thou destroyed!'
2. (n)V11?. See Obad. 5 (cf. Jer. xlix 9) where the phrase HTTO
nV*1? is used in a metaphor whose purpose is to describe the
total destruction of Edom.
3. TOD p n n DK. See Obad. 6. Here Edom's hidden treasures
are ransacked (V3D2
The use of these phrases suggests that the Dumah oracle was,
in the sixth century, thought to be consistent with the familiar
message of Edom's (justified) destruction.
(Oracle concerning the Arabs)
Turn back, go into the scrub;* amongst the Arabs you must stay, 0
caravans of Dedanites. (14) Bring water to meet the thirsty; 0
inhabitants of the land of Teima, meet* the fugitives with food. (15)
For theyfleefrom the sword, from the sharpened sword (or from the
ubiquitous sword), from the strung bow and from the vehemence of war.
(16) For the Lord has spoken to me thus: In a year, measured exactly
(lit. as the years of a hireling),c all the might of Kedar shall be at an
end. (17) And there will remain but a very few of the bowmen of the
warriors of Kedar. The LORD the God of Israel has spoken.
a. Jer. xlix 8 suggests, as we have seen, that rnK nitP were in

the sixth century construed with the Dedanites. At some stage
after the composition of Jer. xlix a redactor added the title
:nsn Ktffc, inserting it after the word rnK and before the
word ispn.
b. ^7E. Piel imperative is read, as in the eighth-century text.
c. It is no longer possible to determine the circumstances which
gave rise to this precise chronological statement; cf. Rudolph,
H.S.S., p. 142.
Principal Works Consulted
(Works which are cited only once are not included)

i Hebrew or massoretic text
Biblia Hebraica, 3rd edn, ed. R. Kittel (Stuttgart, 1937).
Biblia Hebraica Stuttgartensia, ed. D. W. Thomas (Stuttgart, 1968).
The Book of Isaiah (The Hebrew University Bible), ed. M. H.
Goshen-Gottstein (Jerusalem, 1975).
2 The ancient versions
The Peshitta. Vetus Testamentum Syriace, ed. S. Lee (London,
1823). See also Diettrich in sect. 7 below.
The Septuagint. Septuaginta: Vetus Testamentum Graecum, vol. xiv,
Isaias, ed. J. Ziegler (Gottingen, 1939). See also Seeligmann
and Ziegler in sect. 7 below.
The Targum. The Targum of Isaiah, ed. J. F. Stenning (Oxford,
1949);
The versions of Aquila, Synmmachus and Theodotion.
Origenis Hexaplorum quae supersunt, ed. F. Field, vol. 11
(Oxford, 1875).
The Vulgate (sometimes referred to as 'Jerome'). Biblia Sacra
iuxta Vulgatam Versionem,ed. R. Weber et al. (Stuttgart, 1969).
3 Lexica and dictionaries
Belot, J. B. Vocabulaire arab-frangais (Beyrouth, 1896).
Brockelmann, C. Lexicon Syriacum (Edinburgh and Berlin, 1895).
Bf own, F., Driver, S. R. and Briggs, C. A. A Hebrew and English
Lexicon of the Old Testament (Oxford, 1907).
Dozy, R. P. A. Supplement aux dictionaires arabes (Leiden, 1877).
Gesenius, W. and Buhl, F. Handworterbuch iiber das A.T., 16th
edn (Leipzig, 1915).
144
PRINCIPAL WORKS CONSULTED
145
Jastrow, M. Dictionary of the Targumim, the Talmud Babli and

Terushalmi, and the Midrashic Literature (2 vols., London, New
York and Leipzig, 1886-1903).
Hoftijzer, J. Dictionnaire des inscriptions semitiques de VOuest
(Leiden, i960).
Koehler, L. and Baumgartner, W. Lexicon in Veteris Testamenti Libros (Leiden, 1953). Cited as K.B.
Hebraisches und Aramaisches Lexikon zum Alten Testament, 3rd
edn, ed. W. Baumgartner, B. Hartmann and E. Y.
Kutscher, vol. 1 (Leiden, 1967), vol. 11 (Leiden, 1974).
Cited as K.B.<3>
Lane, E. W. Arabic-English Lexicon (London, 1863-85).
Liddell, H. G. and Scott, R. A Greek-English Lexicon, rev. H. S,
Jones (Oxford, 1940).
Payne Smith, R. Thesaurus Syriacus (Oxford, 1879-1901).
Wehr, H. A Dictionary of Modern Written Arabic, rev. J. M.
Cowan (Wiesbaden and London, 1971).
4 Rabbinical commentators and authorities
Ben Bilam ( n t h century). J. Derenbourg (ed.), Gloses d'Abou
Zakariya Tahia Ben Bilam sur Isaie (Paris, 1892).
Ibn Barun ( n t h - i 2 t h centuries). P. Wechter (ed.), Ibn
Barurts Arabic Works on Hebrew Grammar and Lexicography
(Philadelphia, 1964).
Ibn Ezra ( n t h - i 2 t h centuries). M. Friedlander (ed.), The
Commentary of Ibn Ezra on Isaiah (3 vols., London, 1873-7).
Ibn Janah ( i o t h - n t h centuries). A Neubauer (ed.), The Book
of Hebrew Roots (Oxford, 1875).
Qimhi (i2th-i3th centuries). J. Levensohn (ed.), ro*np
mVm, vol. ix (Warsaw, 1866).
Rashi ( n t h - i 2 t h centuries). J. Levensohn (ed.), ibid.
Saadya (gth-ioth centuries). J. Derenbourg (ed.), R. Saadya
ben Josef Al-Fayyoumi, vol. m (Paris, 1896).
5 Commentaries on Isaiah
Bright, J. Peake's Commentary on the Bible, ed. M. Black and H.
H. Rowley (London, 1963).
Cheyne, T. K. The Prophecies of Isaiah, vol. 1 (London, 1889).
Cited as Cheyne, P.L
I46
Introduction to the Book of Isaiah (London, 1895). Cited as
Cheyne, I.B.I.
Delitzsch, F. Biblischer Commentar ilber den Propheten Jesaia
(Leipzig, 1866). Cited as Delitzsch (1) where it is necessary

to distinguish between it and the 4th edn.
The Prophecies of Isaiah, E. T. of Delitzsch (1) by J. Martin,
vol. 1 (Edinburgh, 1877).
Commentar iiber Das Buch Jesaia, 4th edn. (Leipzig, 1889).
Biblical Commentary on the Prophecies of Isaiah, E.T. of the 4th
edn by various translators, vol. 1 (Edinburgh, 1890).

Dillmann, C. F. A. Der Prophet Jesaja, 5th edn (Leipzig, 1890).
Cited as Dillmann.
Dillmann, C. F. A. and Kittel, R. Der Prophet Jesaja, 6th edn
(Leipzig, 1898). Cited as D.K.
Duhm, B. Das Buch Jesaja, 5th edn (Gottingen, 1968).
Ewald, H. Die Propheten des Alien Bundes, vol. 1, 2nd edn.
(Gottingen, 1877).
Fohrer, G. Das Buch Jesaja, vol. 1 (Stuttgart and Zurich,
i960).
Gray, G. B. The Book of Isaiah (I-XXXIX) (Edinburgh, 1912).
Hitzig, F. Der Prophet Jesaja (Heidelberg, 1833).
Kaiser, O. Der Prophet Jesaja {13-39) (Gottingen, 1973).
Isaiah 13-39, E.T. by R. A. Wilson (London, 1974).
Kissane, E. J. The Book of Isaiah, vol. 1 (Dublin, i960).
Lowth, R. Isaiah (London, 1778).
Marti, K. Das Buch Jesaja (Tubingen, 1900).
Migne, J.-P. Patrologiae Latinae, vol. xxiv (Hieronymus iv)
(Paris, 1865).
Procksch, O. Jesaja, vol. 1 (chs. 1-39) (Leipzig, 1930).
Rosenmuller, E. F. C. Scholia in Vetus Testamentum, vol. m, no. 2
(Leipzig, 1793).
Vitringa, C. Commentarius in librum prophetiarum Jesaiae (Leeu-
warden, 1724).
Wildberger, H. Biblischer Kommentar Altes Testament: Jesaja
(13-22), vol. x, fasc. 2 (Neukirchen, 1974-7).

6 Modern commentaries on other O.T. prophets
Bright, J. Jeremiah (Anchor Bible) (Garden City, N.Y., 1965).
Duhm, B. Das Buch Jeremia (Tubingen and Leipzig, 1907).
147
Rudolph, W. Jeremia, 3rd edn. (Tubingen, 1968).

Joel, Amos, Obadja, Jona (Giitersloh, 1971). Cited as K.A.T.
XIII 2 .
Micha, JVahum, Habakuk, Jêphanja (Giitersloh, 1975). Cited
as K.A.T. XIII 3.
Ruth, Hohes Lied, Klagelieder (Giitersloh, 1962). Cited as
K.A.T. XVII.
Volz, P. Der Prophet Jeremia (Leipzig and Erlangen, 1922).
Weiser, A. Das Buch Jeremia (chs. 25, 15-52, 34) (Das Alt
Testament Deutsch, 21) (Gottingen, 1966).
Wolff, H. W. Dodekapropheten, vol. in. Obadja und Jona (Neukir-
chen, 1977).
Zimmerli, W. Ezechiel, vol. 11 (chs. 25-48) (Neukirchen, 1969).
7 Other works cited

Aharoni, Y. 'Three Hebrew Ostraca from Arad\ B.A.S.O.R.
X
97 {l91)> l6 ~42 (esp. 28-32).
Ap-Thomas, D. R. 'All the King's Horses'. In J. I. Durham
and J . R. Porter (eds.), Proclamation and Presence: Essays in
Honour of G. H. Davies (London, 1970).
Bach, R. Die Aufforderungen zur Flucht und zum Kampf in alttestamentlichen Prophetenspruch (Neukirchen, 1962).
Barnes, W. 'A Fresh Interpretation of Isaiah xxi 1-10'. J.T.S. 1

(1900). 583-92.
Bartlett, J. R. 'The Moabites and Edomites'. In Wiseman (ed.),
Peoples, pp. 229-58.
'From Edomites to Nabateans'. P.E.Q^., 1979, 53ff.
Baumgartner. See Koehler and Baumgartner, Hebrdisches, in
sect. 3 above.
Boutflower, C. 'Isaiah xxi in the Light of Assyrian History'.
J.T.S. 14 (1913), 5i-}b;
Branden, A. van den. Inscriptions thamoudeenes (Louvain, 1950).
Brinkman, J. A. 'Elamite Military Aid to Merodach-Baladan'.
J.N.E.S. 24 (1965), 161-6.
Buhl, F. 'Jesaja 21, 6-10'. Z.A.W. 8 (1888), 157-64.
Cambridge Ancient History, The, vol. in. The Assyrian Empire, ed.
J. B. Bury, S. A. Cook and F. E. Adcock, 1st edn (Cambridge, 1925). Cited as C.A.H., in.
I48
Vol. iv. The Persian Empire and the West, ed. J. B. Bury,
S. A. Cook and F. E. Adcock, 1st edn (Cambridge,
1926). Cited as CAM., iv.
Cobb, W. H. 'Isaiah xxi. 1-10 Reexamined'. J.B.L. 17 (1898),
40-61.
Derenbourg, J. See in sect. 4 above (Ben Bilam and Saadya).
Dhorme, P. 'Le desert de la mer (Isaie XXI) \ R.B. 31 (1922),
403-6.
Diettrich, G. Ein Apparatus criticus zur Pesitto zum Propheten
Jesaia {Z-A.W. Bei. 8, Giessen, 1905).
Donner, H. and Rollig, W. Kannaanaische und aramaische
Inschriften, 3 vols. (Wiesbaden, 1962-4).
Driver, G. R. 'Isaiah I-XXXIX: Textual and Linguistic
Problems'. J.S.S. 13 (1968), 46f.
Ehrlich, A.B. Randglossen zur hebraischen Bibel, vol. iv (Leipzig,
1912).
Eitan, I. 'A contribution to Isaiah Exegesis'. H.U.C.A. 12-13

(1937-8), 55-88.
Erlandsson, S. The Burden of Babylon (Lund, 1970).
Fohrer, G. Einleitung in das Alte Testament (Heidelberg, 1965).
Cited as Fohrer, Einleitung.
Friedlander, M. See in sect. 4 above (Ibn Ezra).
Gadd, C. J. 'The Harran Inscriptions of Nabonidus'. Anatolian
Studies. Journal of the British Institute of Archaeology at Ankara
8 (1958), 35792.
Galling, K. 'Jesaja 21 im Lichte der neuen Nabonidtexte'. In
Tradition und Situation: Festschrift A. Weiser (Gottingen,
l
&$)> PP- 49- 6 2.
Gesenius-Kautzsch. Gesenius' Hebrew Grammar, 2nd English edn.,
rev. A. E. Cowley (Oxford, 1910).
Goshen-Gottstein. See in sect. 1 above.
Gratz, H. Emendationes in Plerosque V.T. Libros (Breslau,
1892-4).
Jamme, A. Sabaean Inscriptions from Mahram Bilqis (Baltimore,
Md., 1962).
Kleinert, P. 'Bemerkungen zu Jes. 20-22'. Th.St.Kr. 1 (1877),
174-9Lehman, I. O. 'A Forgotten Principle of Biblical Textual
Tradition Rediscovered'. J.N.E.S. 26 (1967), g6f.
149
Lindsay, J. 'The Babylonian Kings and Edom, 605-550 B.C.'

P.E.Q., 1976, pp. 23-39.
Lohmann, P. 'Das Wachterlied Jes. 21. 11, 12'. .^4.H^. 33
(1913), 20-9.
Luckenbill, D. D. Ancient Records of Assyria and Babylonia, vol. n
(Chicago, 1927).
Michaelis, J . D. Supplementa ad Lexica Hebraica (Gottingen,
1792).
Mohle, A. Mitteilungen des Septuaginta-Unternehmens [Theodoret zu
Jesaia) (Berlin, 1932), p. 88.

Naveh, J. 'Hebrew Texts in Aramaic script in the Persian
Period?' B.A.S.O.R. 203 (1971), 27-32.
Neubauer, A. See in sect. 4 above (Ibn Janah).
Obermann, J. 'Yahweh's Victory over the Babylonian
Pantheon'. J.B.L. 48 (1929), 307-28.
Ottley, R. R. The Book of Isaiah According to the Septuagint
(Cambridge, 1906).
Otzen, B. Studien iiber Deuterosacharja (Copenhagen, 1964).
Pritchard J . B. Ancient Near Eastern Texts Relating to the Old
Testament (Princeton, N.J., 1950). Cited as A.JV.E.T.

Rabin, C. 'An Arabic Phrase in Isaiah'. In Studi sulVOriente e la
Bibbia: Offerti a G. Rinaldi (Genoa, 1967). Cited as Rabin,
English.
K 1JTW ")D03 D"W57, ed. B.-S. Loria (Jerusalem, 1976).
Cited as Rabin, Hebrew.
Rudolph, W. 'Jesaja xv-xvi'. In D. W. Thomas and W. D.
McHardy (eds.), Hebrew and Semitic Studies Presented to G. R.
Driver (Oxford, 1963). Cited as Rudolph, H.S.S.

Schrader, E. Die Keilinschriften und das A. 7~., 2nd edn (Giessen,
1883).
Schwab, M. Le Talmud de Jerusalem traduit, vol. xvi (Paris,
1883).
Scott, R. B. Y. 'Isaiah xxi 1-10: The Inside of a Prophet's
Mind'. V.T. 2 (1952), 278-82.
Seeligmann, I. L. The Septuagint Version of Isaiah: A Discussion of
its Problems (Leiden, 1948).
Stenning, J. F. See in sect. 2 above (Targum).
Talmud, The Palestinian. ^VÎT TiaVn (jlttj? Tsn&. .. KT)
(Zitomir, 1862).
150
Walton, B. Biblia Sacra Polyglotta, vol. m (London, 1657).

Wechter, P. See in sect. 4 above (Ibn Barun).
Wiseman, D. J. (ed.). Chronicles of the Chaldean Kings (London,
1956). Cited as Wiseman, Chronicles.
Wiseman, D. J. (ed.). Peoples of O.T. Times (Oxford, 1973).
Cited as Wiseman, Peoples.
Ziegler, J .
Untersuchungen zur Septuaginta des Buches Isaias
(Minister, 1934). See also Septuagint in sect. 2 above.
Indexes
A. BIBLICAL REFERENCES
(References to Isaiah xxi are not listed)
Genesis
*7
54
62
xxi 20
xxv 13
61
xxv 14
4of
xxv 15
58
11
xxvii 42
xxxii 14
39
62
xxxiv 30
Exodus
ii 23
xiv 3
xxi 8
11
Ezra
ix 1
5
15
15
Deuteronomy
10
i 19
iv 24
33
10
viii 15
XX 12
15
Joshua
xv 52
40
Judges
v4
10, 39, 6on
xv 9
9
139*
xix 11
vi8
1 Kings
xv 13
xiv 6
xix 18
82n
74
iO2n
Nehemiah
xiii 1
iO2n
xiii 7
59
xiii 23
iO2n
Esther
ii 18
120
Job
iii 5
22
iii 9
38
V7
11
vi 15
101
vii if
22
vii 4
22
xxiv 15
xxx 24
5in
xli 20
Psalms
1 Samuel
xiii 1
xiv 33
xxx 16
1 Chronicles
xii 9
27
2 Chronicles
xxviii 17
xxxiii 11
18
42n
Numbers
xxv 17
xxxiii 55
2 Kings
vii 14
30
ix 25
32
xvii 6
66
xviii 11
66
xix 23
125
xx
66
xx 1219
109
6on
60
11
60
2in
11
62n
25
xxiii 5
24
xxix
xxxi 12
xlvi 6
23
io2n
Ix 10
lxxviii 19
24
lxxxiii 7
iO2n
xciv 17
40
cviii 10
iO2n
INDEXES
Psalms, cont.
cxxi 3
44n, 83n
cxxvii 1
44n
cxxx 6
n8n
cxxxvii 8
Proverbs
ix 2
24
xi 22
33
Song of Songs
ii 17
9
44n, 83n
iii 3
9
iv 6
V7
43
vii 10
44
Isaiah
i-xxxii i
81
6in
137
82
i28
v 1-7
vi 1
120
vi gf
105
vi 13
vii
69
vii if
vii 2
vii 4
vii 9
viii 3f
viii 6
viii 7
14
16
18
18, 5on, 120

13, 120
18
61
ix 1
x 5f
x 12
44
13
6in
61
x 18
xi 11
xiii
xiii f
66
1, 63f, 67, 73
65, 114
xiii 11
17
xiii 17
xiii 19
xiv 13
65, 68n, 107

1,36
XV
XV f
XV I
78
13, 85, 93f, 98, 102, 13 l f , 134
43f, 135, 137
xv 1-5
Qon
13^n
140
xv 4fF
20
xv 5
xvi
13,78
xvi 1
iO2n
XV 1 - 8
xvi 1-5
xvi 3ff
xvi 4
78
78, iO2n
13, 6in
xvi 6-11
xvi8ff
xvi 8-11
xvi 9
I36n
134
g8n
20
xvi 10
xvi 11
83
20
96, ggn, 100,102, 131

xvi 13
6of, 101
xvi 14
61
xvii 3f
22
xvii 14
xviii 5
47
xx
65f, 82, 98
xxii
70, 98, I22n, 138
xxii 1
4, 39, 88
xxii 1-14
69
4n
xxii 5
xxii 6
35, 66, 69, 71, noff, i25n
125
xxii 7
xxii 13
25, 35, 7i, J 22
xxiii 13
74
xxiii 16
84,137
75
xxiv-xxvii
xxiv 16
xxv 1 of
xxvi 16
xxviii 21
xxix 1
xxix 20
xxx 6
xxx 15
xxx 27
xxxi 3
xxxiii 1
xxxiii 23
xxxiv
xxxvi 2
xxxvi 11
xxxvi i 24
xxxix
xxxix 1
xli 15
xli 22ff
xlii 11
xlii 1 if
xliv 6ff
xliv 28
xlvff
xlv 1
Xlv 2Off
13
iO2n
23
33,6m
33
6in
4, 63, 70, 128

18
60
6in
I2f, 65, I l 8
59*
65
44
79
125
109
112
38n
113
6in
"3
107
65
107
3
xlix 15
4 6
lvi 9
53
lvi 12
Ix 7
lxii 6
lxiii 7
53
(5 i n
34, 43n, 128
33
INDEXES
Jeremiah
i 12
ii 10
iii 2
iii 20
iv 7
iv 2g
vi 22ff
vi 26
ix 1
xii g
xii 12
xiii 16
xv 8
35
6in, 102
56n
11
128
62
"5
13
55
56n
13
2in
13
123
xxv 15
xxv 23
58
xxv 30
83
92n
xxvii f
xxvii 3
76
xl7ff
75*
xliv 28
62
xlviff
114
xlvi 3
27
xlvi g
ii3n
xlviii
i3n 93^, I 0 2 132
xlviii f
93
l
xlviii 8
3
J
xlviii 18
3
l
xlviii 32
3
xlix
i3g3
76, goff, 107, i34n, 135, 139,
ri
IA.11
4
T A
xlixff
141
xlix 7
8g
xlix 7f
76
xlix 7ff
i3iff
xlix 8
135, 139, 142
xlix 9
134, 143
xlix gf
128
xlix ig
I
xlix igff
33 n
xlix 28
76
xlix 28ff
132, i33 n
xlix 30
142
ii3n
xlix 31
xlix 34ff
65
If
1:i3ff, i22f, i32n
12
114, 130
1g
120
11 n
1 i4f
3
1 2g
ii3n, 114
117
I38
l4iff
"5
I42
122
143
I44
"5
128
153
l44ff
I33n
li
107, 117
i[14, 117
li 1
li 5
1[ 1 4
li 8
i[14, i 2 g
li 11
65, 107, ii3n, 120
li 1 if
122
li 13
6, 117
li 25
"9
li 27
ii3n
Ii27ff
120
H28
65, 107
122
li 30
38, n 4 f
li 33
122
li 35
123
li 39
117
Ii42f
114
li47f
li 5 ^
H57
114, u8f
123
Lamentations
i 11
18
Ezekiel
iv 16
v 16
xiv 13
xxi 11
105
105
105
20
24
xxiii 41
xxv8ff
102
XXV I2ff
75 n , J40
54, 8g
xxv 13
xxvii 10
io7n
xxvii 18
44
xxvii 21
56n
40
xxvii 32
xxxii 17!.T
107
xxxii 24
65
xxxviii 5
io7n
Daniel
Book of
63
v
15 , 22, 63f, 12m
v3
22
v6
22
vi 8
48
i02n
xi 41
Joel
i 18
18
Amos
vii8
ix7
inn
35
154
Obadiah
Book of
75n, 93
1 3 3 ^ l3Q> 143
6
7
49> X39> J 4 3
J
4O
INDEXES
Zephaniah
ii 2
9
Micah
iii 6
4
iv 10
74
I3on
iv 1 iff
iv I2f
38n
iv 13
I3on
Nahum
ii 4f
ii 14
125
125
Habakkuk
19
47
ii
124
iii
28f, 34, 124
ii if
124
ii iff
124, 128
ii 1-8
123
ii 8
118
iii 3
10
iii 16
1911
Zechariah
i Hf
35
127
ii 1
"4
36
iv iff
36, 127
127
v1
vg
127
ix 14
10
Malachi
i 2-5
Sirach
1 25
xlix
75n
75n
ggn
Revelation
X3
33
Talmudic and Midrashic refe
P.T. Ta'anith
1:1
39, 41, 44f, 76n, 85
4:1
57,104
Bereshith Rabba
63
25
65:21
39n
B. GENERAL INDEX
(Authors of commentaries on biblical books are not included)
Abel, F.-M., 4on
Abydenus, 6, 72n
Aharoni, Y., I35n
Akkad, 86
Albright, W. F., 54n, 86n, 8gn
Alexander Jannaeus, 99
al Jauf/el Jof, see Dumat
Alter ego (of Isaiah), 28, 29n, 34, 123
Anshan, 67, 106,
apocalyptic writings, 75, ioof
Ap-Thomas, D. R., 32n
Aqiba, R., 39
Ar, 137
Arabs/Arabia, 40, 53ff, 6if, 68,78, 8off,
86ff, gif, 94ff,ioif, 104,
Arad ostracon, i35
Arameans, 108, 11 off
Aramaic-Arabic, 7gn
Aramaisms/Aramaic, 76, 78f, 83, 91,
93,137
Ararat, 107
Aretas, 99
Ashdod, 65, 78, 135,
Ashkenaz, 107
Ashurbanipal, 107
Assyria, king of, 91
Assyrian empire, 108; fall of, 69, 72
Aufforderung zum Kampf, 93n, ii2f, 122
Babylon, fall of (539 B.C.), 63, 67ff, 72,
77, gsf, iO4ff, ii2f
Babylonian Chronicle, io7n
Bach, R., i4n, i6n, 93n, ii2n, ii3n
Barnes, W. E., 7off, io6n, 107, non
Barr, J., 26n
Bartlett, J. R., 7gn, 10in, 14m
Bavian inscription, 72, iogn
Bellino cylinder, 66
Belshazzar, 15, 20, 22, 24ff, 63f, 86n,
12 in
Bengtson, H.,
INDEXES
Bernstein, G. H., 22n
Berosus, g2n
Bit Yakin, io8f
Black, J. S.,58n
Boutflower, C , 706, iogn
Branden, A. van den, 5on, 53n
Bright, J.,93n
Brinkman, J. A., io8ff
Buber, S., 57n
Buhl, F., 29f, 35n, 46f
Cappadocia, io7n
cavalry, 306
Chaldeans, 74, io8f
chariots, two-horse, war, 306, i26n
Gobb, W. H., 4 n, 69f, 73f, 103
Croesus, 64
Cross, F. M., 36n
Cyaxares, 721*, 108
Cyrus II of Persia, 16, 24, 37, 631*, 676,
77, 88, 1046, iogn, ii2f, 115
Dadanu, 86
Dahood, M., 26n, 36n
Darius Hystaspis, 64
Darius the Mede, 26, 63
Dathius, 25n
Daume (s.w. of Hebron), 40
Dedan/Dedanites, 54, 576, goflf, 95, 97,
100, i32f, i4off
Dedan, war of, 87
Der, battle of, 108
Dhorme, P., 6
Dougherty, R. P.,86n
Driver, G. R., 4n, gf, 18, 22f, 26n, I37n,
138
Driver, S. R., I24n
Dumat al Jandal/Gendel (el Jauf/J6f),
40, 77n, 8ofF, 88, 100, 132
Ebeling, G., 54n
Edom/Edomites, 396, 750, 1316
Egypt, 70, 106
Eissfeldt, O., ign, 22n
Eitan, I., in, 17, 25n, 6in, io3n
Elam/Elamites, i4ff, 18, 65ff, 75, 103,
io6ff, ugf
Emerton, J. A., 6on
Erlandsson, S., 6n, n, i6n, 37, 39,
44n, 72, 92
irof,
Esarhaddon, 8if
Eusebius, 6
Fohrer, G., 107
Freedman, D. N., 36n
155
Gadd, C.J.,68n, 86n, 87f

Galling, K., i2n, 27, 296, 42, 69, 84,
86ff, 92n, 95f, 99, iO3n
Gath, 78, 135
Gidir land, 136
gtmatriyd, 28, 33
Gratz, H., 35n

Grayson, A. K., 8gn
Habakkuk, 28, 33, i24f, 128
HaleVy,J.,68n
Hanel,J.,2gn
Hanina, R., 39
Harran, temple at, 87
Harran inscriptions, 68, 86
Hatti-Land, 86
Hazael, king of the Arabs, 81
Hazor (in Arabia), 132, I33n, 142
Hempel, J., 22
Herodotus, 24, 63f, 7if, 122, 127
Hezekiah, 64,66, 70, 78, 109,111,118,
122, 1356
Hijaz, oases of, 86
Hoftijzer,J.,49n, 51
Iatrib (Medina), 86
Ibn Barun, 30, 49n
Ibn Janah, 22, 24ff, 30, 39f, 42, 49
Isaiah, Deutero-, 113
Isaiah of Jerusalem, 69, 8if, 112, 128,
Isaiah Apocalypse, 75
Ishmael/Ishmaelites, 40, 57, 62
Jamme, A., 49
Jeremiah, 65; foreign oracles of, 746
John Hyrcanus, ggn
Jones, D. R., iosn
Josephus, 92n, iO2n, 14m
Kahana, R. Abba bar, 25
Kaufmann, Y., 8if
Kausmalak, king of Edom, 78
Kedar, 61, 93, 966, 132, 142
Khalule, battle of, 109
Kir (Aramaeans), 69, 11 of
Kir (of Moab), 137
Kirkpatrick, A. F., 124n
Kish, battle of, 66
Kleinert, P., 22n, 646,, 72, n o
Kocherus, 58
Koehler, L., 32n
Kdnig, F. W., i07n
I56
INDEXES
Lambert, W. G., 8911

Lehman, I. O., 3gn, 90
Linblom, J., 13CH1
Lindsay, J., 86n, 89f, 90, 14m
Lohmann, P., 33n, 43, 44*, 7gn,
132, 137
Macintosh, A. A., 2in
Medes, 5, 7, 9f, i6ff, 28,31, 33,65,676,
71, 73, 103, io6ff, ii2f, 1 i8ff
Meir, R.,39
Merodach-baladan, 6n, 66, 7of, io8ff,
118, 120, 122
Meshech-tubal, 107
Micah, i3on
Michaelis, J. D., 6n, n n , 26n, 40
Minni, 107
Moab/Moabites, 13, 43f, 78, 96, 98f,
102, 131, 1346
Mohle, A., 5n
Morag, S., i8n
Mowinckel, S., 3off,
Mushezib-marduk,
Muski,
Nabonidus, 68f, 72, 86ff, 95, iO3n, 141
Nabonidus Chronicle, 71, 89
Nabopolassar, 72n, 108
Naveh,J., ygn
Nebuchadnezzar/Nebuchadrezzar, 57,
5> 7 6 , 9 In j JO2, 104, io7n, i4if
Saggs, H. W. F., 136

Sargon II, 656, 7iff, 76n, 78,82,92,92,
97, 108, n o , 135^ capture of Babylon, 72
Scheiber, A., 47
Schrader, E., 6
Scott, R. B. Y.,4n, 8f
Sennacherib, 13, 4111, 66f, 69, 72, 8if,
92, io8f, i n , 118, 122, i26n, 129,
135
Shamar (Arabian tribe of), 80
Sheol, 107
Seigen,J. P., 39n
Sin (the god), 87
Smith, G., 65, 72
Smith, S., 86n, SQU
Stade, B., 2a.ff, ^5n
Syro-Ephraimite war, 69
Tabal, io7n
Teima/Teyma/Tema', 58, ygn, 86ff, 95,
100,
140
Teman, 58, 89
Thamudic inscriptions, 87
Theodoret, 5
Tiele, C. P., 70
Tiglath-pileser III, 78, 98n, 108, I26n,
135
Tigris, 108, n o
Tyre, 40, 84
Nielson, E., iosn

Nineveh, 72f, 108
Noldeke, T., 58n
Velde, C. W. M. van der, 4on
Wadd (god of Dumat al Jandal), 8ofF

Weippert, M.,4111
Weiser, A. 94n, ii4n, U5n, n6n
Obermann, J., ign, 28n
Winckler, H., 51
Otzen, B., 2
Wiseman, D. J.: Chronicles, 92n, iO7n,
14m; Peoples, 56, 102, i07n, i26n,
Padakku, 86
Persia/Persians (see also under Cyrus and i36n
Medes), 5, 7, 16, 2if, 28, 316, 636, Woude, A. S., van der, i3on
67ff, 70, 87f, 946, ioif, 106, 120, 126
Xenophon, 24, 63f, 71, 122, 127
Philistia, 78
Phrygia,
Yohai, R. Simon ben, 41
Rabin, C , 4off, 476, 496, 796
Zechariah, visions of, 127
Rabinowitz, I., 100
Zedekiah, 76, 92, 107
Rost, L., 22n
Zyl, A. H. van, io2n
Rudolph, W., i36n, 143

Isaiah XXI, A Palimpset

Uploaded by

Document Informationclick to expand document informationcommentary of Isaiah 21

Document Informationclick to expand document information

Copyright:

Available Formats

Isaiah XXI, A Palimpset

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Isaiah XXI, A Palimpset

Uploaded by

Copyright:

Available Formats

Isaiah xxi

CAMBRIDGE UNIVERSITY PRESS

CAMBRIDGE UNIVERSITY PRESS

To my wife and to all others

List ofprincipal abbreviations

1 The text and its meaning: a preliminary survey

Principal works consulted

J. B. Pritchard, Ancient Near Eastern Texts Relating to

The Massoretic Text. See Principal Works Consulted, sect, i

Any attempt to interpret and to understand prophetical texts of

study of the text to estimate what is its historical background

reviewed in the light of the increase in our knowledge of ancient

The Text and Its Meaning: A Preliminary Survey

THE TEXT AND ITS MEANING

text, no further interpretation of the words is necessary. But the

regarded as plausible only if the contents of verse i do in fact

THE TEXT AND ITS MEANING

In conclusion, I accept the view that the original title of the

The sentence appears to consist of a main verb (xa), apparently

So, essentially, Duhm, Marti, Gray, Procksch, Kaiser, Wildberger, etc.

pretation is adopted by a number of modern commentators,

THE TEXT AND ITS MEANING

1 QIs , suggests that the beginning of verse i is corrupted by

Negeb, i.e. the desert of southern Palestine,1 a place of sudden

T H E T E X T AND ITS MEANING

Verse 2a, ***? Tjn ntfj? mm

the verb is very often translated by Aramaic m ' to despoil' in

THE TEXT AND ITS MEANING

Qimhi, too, seek to understand the phrase in this way, though

the view that the words characterize the action or imminent

THE TEXT AND ITS MEANING

Ibn Ezra compares the verb m s with Deut. xx 12 where the

text of Isaiah may be confirmed by reference to e.g. Isa. vii 2

The Massoretes (so Vulg. and Pesh.) understand the verb as a

THE TEXT AND ITS MEANING

commentators' views of it. For some the word is understood to

(cf. Gray and Procksch). More recently G. R. Driver,1 following

THE TEXT AND ITS MEANING

word nnmK may be a corruption or later adaptation of the

THE TEXT AND ITS MEANING

xxxviii i i , and the similar phrase imno ^V lit. 'goes around,

translates the phrase as a whole: Babylon dilecta mea posita est

Ibn Janah shares the view of Aquila and Theodotion and

THE TEXT AND ITS MEANING

twilight of my desire etc' is 'a strange mixture of phrases', and

THE TEXT AND ITS MEANING

rows of the food on the table or the ranks of the diners at

general meaning with insnsi lp*HD 'polish and make bright

THE TEXT AND ITS MEANING

*nz u? (strip bare). In general the reference is to the removal of

Verse 6, T?5?n rf? s m ^K "iK HD

THE TEXT AND ITS MEANING

temporal clause. Thus, ibn Ezra renders ywp*. . .run DN 'When

THE TEXT AND ITS MEANING

wheeled baggage wagon pulled by asses, by the side of which the

THE TEXT AND ITS MEANING

TtfpTV). For the emphatic expression 'he shall

. The Vulgate, Aquila, Symmachus, Theodotion and