This monograph is a revised version of Part II of my doctoral dissertation submitted to the University of Cambridge in 1975. During the writing of it I have been aware of my indebtedness to many teachers and friends for their insights and, not least, their questions. I should like to record my thanks in particular to the following: to Dr R. E. Clements (who supervised my research and, as Editor, accepted this work for the Monograph Series), Dr S. P. Brock and Dr C. J. Labuschagne, all of whom read the original manuscript and made many helpful suggestions; to Professor J. A. Emerton, whose rigorous handling of exegetical and other problems has been a source of inspiration throughout my work; to Dr P. S. Alexander and Dr Z. Meshel, who advised me on particular points; to the Society for Old Testament Study and the Managers of the Hort Fund of the University of Cambridge, for generous contributions to the cost of publication; and to the staff of the Cambridge University Press, for their exceptionally careful and courteous handling of the process of production.

The transcription of place-names is a familiar problem to workers in this field, and I have adopted the following policy here. Where there is a well-known English form, as with Biblical names, I have used this; where there is not, I have reproduced the name in an Anglicised form which keeps as close as possible to the original, without, generally, employing the diacritical marks that would be needed for absolute precision.

The Jewish material may for convenience be divided into two parts according to the language in which it was written: Greek in the one case, Hebrew or Aramaic in the other. An important corollary of this linguistic division is the use by the one group of Greek versions of the Old Testament, while the other group works primarily from the Hebrew text. The division according to language also corresponds to a chronological division, as the Greek material dates chiefly from before 100 AD, while the Hebrew and Aramaic texts are later, although they may contain traditions which originated before this date. Although we adopt this division of the material, we do not imply that there is no contact between the two groups of texts. In fact our study confirms that at certain points there is indeed contact between them.

THE SEPTUAGINT

Only occasionally does the LXX itself indicate the direction in which the Israelites were thought to have gone. Most of the Hebrew names are simply transliterated or (occasionally) translated, and variations from MT can be explained by textual corruption in either the Hebrew or the Greek stage of transmission. It is possible that in some cases the LXX preserves the names more accurately than does MT. At one point (Num. 33: 36) the fuller text of most LXX witnesses must be attributed to a secondary harmonisation of the passage to make it agree with a statement earlier in the narrative (Num. 13: 26 (LXX v. 27)).

In surviving Christian literature of the first three centuries AD there is surprisingly little evidence of any acquaintance with a geographical interpretation of the wilderness itineraries, despite the fact that by this time, as we have seen, a number of identifications were current in Jewish circles. The ‘sea’ of Exodus was of course equated with the Red Sea (cf. Acts 7: 36; Heb. 11: 29), following the rendering given in LXX. Other than this only indications of the location of Mount Sinai (and vague ones at that) are found. Paul located it in ‘Arabia’ (Gal. 4: 25), but in view of the wide area covered by this term, which included the Sinai peninsula as well as Arabia proper, we at any rate cannot be sure where Paul thought it was, even if he himself was acquainted with a specific local tradition. A third-century anti-Jewish tract ascribed to Cyprian put Mount Sinai in Palestine, which was at that time a shortened form of the name of the Roman province of Syria Palaestina, whose southern border was in the northern Negeb. The inclusion of the Sinai peninsula in ‘Palestine’ was only possible after the reorganisation of the area at about 300 AD which added most of what had hitherto been Arabia Petraea to Palestine, and seems to be first attested in the sixth century. This could be taken to mean that Pseudo-Cyprian knew a tradition that Mount Sinai was in the hill country in the south of Palestine proper.

Old Testament scholars have generally paid far less attention to Arabic geographical traditions than to those of Jews and Christians in trying to determine the route of the wilderness journeys. This is not surprising when the late date at which they are attested is remembered. Moreover, if the survey that follows (which is based only on what is conveniently available in the selections translated by Le Strange and Marmardji) gives a representative picture, there were remarkably few attempts to give a precise location to the places named in the Bible. Nevertheless, from the point of view of the history of interpretation this branch of the tradition is as important as any other.

The main exception to the general tendency of Old Testament scholars to neglect the Arabic evidence was A. von Gall. In the section on Sinai in his Altisraelitische Kultstätten, von Gall cited a number of passages from the classical Arab geographers which, he claimed, supported his view that Mount Sinai was east rather than west of the Gulf of Akaba. One of the tasks before us is to see whether von Gall has given an accurate picture of what Arab writers of the medieval period thought about the geography of the wanderings.

An appropriate starting-point is the definition of the area associated with the wilderness wanderings as a whole.

Since the discussion of early Jewish, Christian and Arabic tradition about the route of the Israelites has involved so much intricate examination of various literary sources, it may be helpful if, before proceeding further, we summarise our findings.

At the beginning of the history of the interpretation of the wilderness itineraries, so far as we can recover it, the Septuagint already indicates contemporary equivalents for places at the beginning of the journey. It equates Yam Suf with the Red Sea, and this remains basic for most later interpretations. The Antiquities of Josephus show how a more exact conception of the initial stages had emerged and how some important sites in the later part of the journey were beginning to be located, particularly in Transjordan. In Contra Apionem a markedly different tradition, which places the point of departure in the north-east of the Nile delta, appears, but there is nothing to suggest that Josephus took this to imply a divergent route further on. Elements in the Targumim which go back to the Tannaitic period or before presuppose the same northern point of departure and outline a route which passes into the desert east of Egypt and then reaches Mount Sinai in the south of the peninsula. Traditions in the Talmud and Midrash from the same period show that the final encampment in the plains of Moab was being pointed out to travellers, and imply a considerable interest in the location of other events in the wilderness narratives.