Volume 15 - Issue 1

The origins of Israel in Canaan: an examination of recent theories

By John J. Bimson


‘There have been only two important views of the conquest of Palestine by ancient Israel’, wrote G. E. Mendenhall in 1962, in an article in which he offered a third.1 Since then hypotheses have proliferated, and the question of Israel’s origins has become a vastly complex one. The purpose of this article is twofold: to provide some account of the development and current standing of the main theories on offer, and to assess their relative merits. There are, of course, more theories than can be discussed even in this overlong article, but most of the omitted ones are variants of those included, so that many of my comments will be applicable beyond the scope of this discussion.

The theories discussed here fall into two main groups: those which assume that Israel entered the land of Canaan from outside, and those (now the majority) which assume that Israel was to a great extent indigenous to Canaan. Those in the latter category all post-date 1960. I will give a brief account of each before offering an assessment of it.

1. Hypotheses in which Israel enters Canaan from outside

1a. A 13th-century conquest

We begin with the view which adheres most closely to the biblical traditions: that the Israelite tribes entered Canaan by force, acting more or less in concert. There are, of course, a number of variants on this view in terms of the date of Israel’s entry into the land. The most noteworthy alternative to the view outlined here is that the exodus and conquest occurred in the 15th century bc, as implied by a straightforward reading of 1 Kings 6:1. Although this continues to be favoured by a number of American evangelical scholars,2 and by the present writer,3 it has not been influential in recent decades and will not be treated in detail here. We will see below, however, that it deserves renewed attention.

The majority of those scholars who wish to retain the biblical picture of a more or less unified and violent conquest have long favoured the theory that this event occurred in the 13th century bc. Its classic form took shape at the hands of the so-called Baltimore School, consisting primarily of W. F. Albright and his pupils J. Bright and G. E. Wright. Albright’s own excavations in the 1920s and 1930s at Beitin and Tell Beit Mirsim (which he believed to be the sites of biblical Bethel and Debir respectively) unearthed destruction levels which he associated with the traditions of the conquest. A date at the end of the LBA (Late Bronze Age), in the second half of the 13th century bc, seemed to be indicated by the pottery evidence.4 British excavations (1932–8) at the site of Lachish (Tell ed-Duweir), under the direction of J. L. Starkey, also produced a destruction layer, which Albright dated c. 1230/1220 bc5 (though Olga Tufnell, one of the British excavators, preferred a date in the 12th century bc6). Y. Yadin’s excavations at Hazor (1955–8) also produced evidence of a violent destruction, in this case clearly datable to c. 1220 bc, thus adding a fourth major city to the list.7Hence Bright was able to write in his A History of Israel: ‘… It may be regarded as certain that a violent irruption into the land took place late in the thirteenth century!’8

These destruction levels were merely part of an impressive web of evidence which seemed to point to a 13th-century setting for the exodus and conquest. With the conquest set at around 1220 bc, the exodus would have occurred (according to Nu. 14:33, etc.) forty years earlier, around 1260 bc. This places it in the reign of the pharaoh Ramesses II, in keeping with the reference in Exodus 1:11 to the enslaved Israelites building the store-cities Pithom and Raamses. Excavations at the likely site of Raamses revealed no evidence of occupation during Egypt’s Eighteenth Dynasty, effectively excluding a date before the 13th century bc for the exodus.9 N. Glueck’s surface-surveys of Transjordan, undertaken chiefly in the 1930s, led to the conclusion that the region lacked any settled population between the 19th and 13th centuries bc, which ruled out a date before the 13th century for the events of Numbers 20:14–21 and 21:21–24:25.10

This 13th-century scenario has been adopted by several evangelical scholars. Among British evangelicals K. A. Kitchen has long been its foremost exponent.11 Some American evangelicals have also adopted it,12though others have baulked at a non-literal treatment of 1 Kings 6:1 (discussed below) and have continued to argue for a 15th-century date.13

Problems and methodology

The 13th-century date has never been without its problems. To set against Debir, Bethel, Lachish and Hazor, with their clear destruction levels, there are the troublesome cases of Jericho, Ai and Gibeon.

Since Kathleen Kenyon’s excavations at Jericho (1952–8) it has been widely accepted that no traces of the city attacked by Joshua are to be found. Kenyon’s own view was that, following the destruction of MBA (Middle Bronze Age) Jericho, around 1560 bc, the mound was deserted until a LBA town began to grow up around 1400 bc. This occupation lasted only about a century, and was followed by another long period of abandonment.14 No evidence was found which could support the existence of a town in the second half of the 13th century bc. The site of Ai (Khirbet et-Tell, or simply Et-Tell) is another parade example of the failure of excavations to support the biblical tradition. Here the gap in occupation lasts at least a thousand years, c.2200–1200 bc. Once again there is no city for Joshua’s forces to attack in the late 13th century. This problem was already known after the excavations by Mme. J. Marquet-Krause in 1933–5,15 and it was hoped that the renewed excavations of 1964–72, led by J. A. Callaway, would produce some kind of solution, either by uncovering hitherto neglected evidence from the tell itself, or by finding an alternative site for the biblical city nearby. These hopes came to nothing; the gap at Et-Tell was confirmed and the search for an alternative site proved fruitless.16 J. B. Pritchard’s excavations (1956–62) at El-Jib, the site of Gibeon, produced no LBA material except some burials from the 14th century bc, leading Pritchard to the conclusion that no town had existed there during the LBA.17 As early as 1965 Pritchard remarked that the problems encountered at Jericho, Ai and Gibeon (ironically, the three cities to which Joshua 2–9 give extended treatment) ‘suggest that we have reached an impasse on the question of supporting the traditional view of the conquest with archaeological undergirding’.18

Supporters of the 13th-century conquest have not accepted this verdict, emphasizing the positive evidence and offering explanations for the negative finds from Jericho, Ai and Gibeon. For example, Bright has stressed that extensive erosion makes the situation at Jericho unclear;19 Wright and Yadin, following the same line, have explained the lack of LBA fortifications there in terms of the re-use of the MBA walls—though Wright admitted that there was no shred of evidence to support this theory.20 Albright explained the problem of Ai by suggesting that the narrative in Joshua 8 originally referred to the taking of Bethel21 (though it bears no similarity to the account of Bethel’s capture in Jdg. 1, beyond the fact that in both cases the Israelites take a Canaanite city!). He assumed there had been a very small settlement at Gibeon at the time of Joshua, thus explaining the virtual absence of LBA material there, and suggested that Joshua 10:2 (which speaks of Gibeon as a major city in that period) is an erroneous scribal gloss.22

There are patent weaknesses in such explanations, especially in the latter two, where the biblical tradition has to be adapted to some extent before the archaeological evidence can be said to support it (and in the case of Gibeon there has to be a further assumption that some archaeological evidence has been missed). Criticisms recently levelled at the methodology of the Baltimore School, that it involves inconsistencies, circular arguments and overinterpretations of the evidence,23 have been justified more often than not.

Those evangelical scholars who have adopted the Baltimore School’s scenario have tried to be more rigorously scientific in their defence of its weak points. Kitchen has emphasized that evidence of occupation can be eroded away during periods of abandonment, or simply missed by the excavator when (as is usually the case) limited areas of a mound are explored. He illustrates these points with examples from the archaeology of Egypt, in which excavation has failed to produce remains from sites which are known, on the basis of textual evidence, to have been occupied at the relevant time. Hence he has suggested that 13th-century bc Jericho has been entirely eroded away, that the LBA burials at Gibeon may indicate an occupation which has been missed by the excavator, and that the site of LBA Ai still awaits discovery.24 Indeed, Kitchen has repeatedly stressed that ‘absence of evidence is not evidence of absence’, i.e. that lack of occupation cannot be assumed simply because no trace of it is found. Is this approach on firmer ground than that of the Baltimore School?

Kitchen is certainly right to caution against arguments from silence. However, we must ask how profound the silence is allowed to become before some notice has to be taken of it. To the old chestnuts of Jericho, Ai and Gibeon, we must also add Hebron, Arad and Zephath/Hormah, where the gaps are equally problematical.25 Has the evidence been eroded or missed in every one of these six cases? Perhaps it has, but as the argument is extended to include more sites it inevitably becomes less convincing. It has also been noted that Kitchen does not consistently apply his own dictum.26 He has used the lack of appropriately-dated evidence for activity at the site of Raamses, and for occupation in Transjordan, in order to argue against dating the exodus before the 13th century bc.27 This line of argument has proved distinctly unwise in the case of Transjordan, where more recent surveys and excavations have overturned Glueck’s theory of a gap in occupation;28 and it may yet prove erroneous in the case of Raamses as well.29 That, however, is not my main point here; the point is that ‘absence of evidence’ has not been allowed any significance in the cases of Jericho, Ai and Gibeon, but has been treated as ‘evidence of absence’ in the other cases mentioned above.

This raises serious methodological issues, especially when placed alongside Kitchen’s treatment of 1 Kings 6:1. He correctly stresses that the OT’s chronological statements must be treated in accord with ‘Ancient Oriental principles’, but this does not ultimately help his suggestion that the figure of 480 years in that verse results from a totalling of selected figures which actually represented overlapping periods.30 As I have pointed out elsewhere, an alleged Egyptian parallel to which Kitchen appeals does not actually illustrate the process which he envisages, and 1 Kings 6:1 itself contains no hint that the figure results from totalling lesser periods.31 Furthermore, the 15th-century date produced by this verse is supported by the figure of 300 years in Judges 11:26, and to dispense with this fact by treating the latter figure in the same way32 is completely unconvincing. What has happened is that Kitchen and others have found the archaeological evidence for a 13th-century date so compelling that they have sought to reinterpret 1 Kings 6:1 and Judges 11:26 in the light of it. But at the same time they have employed the archaeological evidence inconsistently in support of the 13th-century date. In other words, the standard evangelical argument for the 13th-century date has been no less flawed with circularity than the methodology of the Baltimore School.

To be fair, however, it must be mentioned that Kitchen has offered other evidence in favour of the 13th-century date, namely the clearly attested parallels between the form of the Sinai covenant and treaty texts of the late second millenium bc.33 But this alone is not a sufficient counterbalance to the problems facing the 13th-century date.34 These have recently multiplied, as we will now see.

Further difficulties raised by recent archaeology

Three recent developments now make any defence of a 13th-century bc conquest extremely difficult. One concerns the nature of the early Iron Age (Iron I) settlements which sprang up after the collapse of the LBA cities. Supporters of the 13th-century conquest have interpreted these as the work of the newly-arrived Israelites, but it is now acknowledged that their remains attest considerable continuity with the material culture of the LBA. If (as is virtually certain) these settlements are indeed the work of the Israelites, it seems unlikely that they were newly arrived in the late 13th century bc. This matter will be discussed more fully below in considering the infiltration theory of Israel’s origins.

A second development has been the growing realization that very few cities of the LBA had any defensive walls. In the light of a thorough study of the evidence by R. Gonen, it is now clear that ‘Only a handful of settlements were surrounded by a wall during the entire period, or even during part of it.’35 The theory that LBA inhabitants reused the old MBA defences is disproved in a number of cases by the fact that LBA houses were built across the remains of the MBA walls. The biblical tradition that the Israelites were faced by towns which were ‘fortified and very large’, ‘great and fortified up to heaven’ (Nu. 13:28; Dt. 1:28; see Jos. 2:15; 6:1, 5; 7:5; 8:29; 10:20; 14:12; etc. for further references to walls and gates) simply does not fit the picture of relatively small, unwalled towns which has now emerged for the 13th century bc. Shortly before his death in 1985, Yadin tried to incorporate this discovery into his (essentially Albrightian) defence of the biblical conquest. In common with other supporters of the 13th-century date, Yadin had previously assumed that the culture of LBA Canaan was based on fortified city-states. Compelled by Gonen’s evidence to abandon this view, he subsequently suggested that the weakness of Canaan’s unwalled cities lent plausibility to the OT’s picture of a swift conquest of the land: ‘Decaying Canaan of the Late Bronze Age … was relatively easy prey for the highly motivated and daring tribes penetrating the country’.36 As a method of salvaging the 13th-century conquest scenario this is not likely to be congenial to evangelical scholars, since it supports one aspect of the biblical tradition by surrendering another, namely that the Israelites encountered towns with walls and gates.

The third discovery concerns those towns which previously seemed to provide evidence for a ‘Violent irruption into the land’ around 1230/1220 bc. B. G. Wood’s thorough comparative study of the pottery from a wide range of sites reveals that the destructions of Bethel, Debir, Lachish and Hazor all require redating to different degrees.37 They did not all fall at the same time, but have to be distributed among no less than threewaves of destruction spanning roughly a century. The first wave occurred at the end of the subdivision of the LBA known as LB IIB1, and should now be dated c. 1210 bc. Of the places mentioned in the Bible as taken by Israel, it included only Hazor. The second wave occurred c. 1170/1160 bc, at the end of LB IIB2 (the final phase of the LBA). This included Tell Beit Mirsim (Albright’s candidate for Debir) and Beitin (traditionally Bethel). However, it is now very widely agreed that the true site for Debir is Khirbet Rabûd,38 which was not destroyed in any of these three waves of destruction. The number of biblical sites involved in this second wave is therefore no more than one (Bethel), and even this should probably be excluded, as there are now serious problems involved in locating Bethel at Beitin.39 The third wave of destruction fell early in the Iron Age, at the end of Iron IA1, c. 1125 bc. Of the places Israel is said to have taken, this included only Lachish.

It is clear, therefore, that either Israel’s conquest of Canaan is not to be linked with these destructions (for which there are, in any case, alternative explanations), or it was a long, drawn-out affair, spanning about a century. The latter view has actually been proposed by D. Ussishkin, the current excavator of Lachish.40Again, this latter approach to the problem is not likely to appeal to evangelical scholars, surrendering as it does important aspects of the biblical tradition.

In the light of these serious difficulties it is very unlikely that the theory of a 13th-century conquest can be rescued. Since the theory has for a long time been the favourite defence of the historicity of the biblical traditions among evangelicals, its demise, once accepted, will require a major rethinking exercise. We will return to this in the final section of this article.

1b. The infiltration theory

That Israel’s constituent tribes came into Canaan from elsewhere is a presupposition shared by the infiltration and conquest theories of Israel’s origins. In their classic forms both theories have also focused on the 13th century bc as the time when Israel’s entry took place.41 But they differ in two important respects.

While the 13th-century conquest theory, at least in its most widely held form, posits a more or less united movement by all or most of the Israelite tribes, the infiltration theory envisages groups with diverse origins settling at different times in different areas. Only after the settlement of these disparate groups did they coalesce into the entity Israel. Secondly, while the conquest theory involves the violent overthrow of Canaanite cities as an initial move by the newcomers, the infiltration theory relegates clashes with the Canaanites to a later stage in the process of Israel’s formation. In the view of A. Alt, who originated the theory in 1925,42 the first phase was totally peaceful. It involved semi-nomadic pastoralists, who spent their winters in the desert fringes beyond Canaan, gradually making the transition to a settled agricultural existence in the hills where they were accustomed to graze their flocks each summer. The central highlands of Canaan were thinly populated, so their settlement involved little or no conflict with the existing inhabitants of the land. Only when these settlers had become somewhat established, united and more numerous, did they attempt to wrest new land from the Canaanites of the plains and valleys. Thus began the phase of territorial expansion, which involved extensive armed conflict, but this did not come until the early days of the monarchy.

Alt’s initial theory was supported and developed by M. Noth, who used literary-critical approaches to the text in an attempt to reconstruct the complex process of tribal settlement.43 More recently M. Weippert has championed the theory and offered some refinements of it,44 and J. M. Miller’s reconstruction of the occupation of the land leans heavily on the Alt-Noth model.45 Among Israeli archaeologists, the late Y. Aharoni was a staunch proponent of the theory,46 and it is currently supported by M. Kochavi, A. Zertal and others.47

Presuppositions and methodology

Although the infiltration theory does make use of specific biblical traditions, it clearly rejects the overall picture of Israel’s origins found in the Pentateuch and the book of Joshua. It replaces a united, military conquest with a peaceful and piecemeal process. Any theory which proposes a picture so different from the biblical one must provide a plausible explanation of how the biblical picture arose. Does the infiltration theory succeed in this respect?

The stories in Joshua 2–9 of the destruction of Canaanite cities by the Israelites are explained by Alt and Noth as folk aetiologies (identified as such by the formula ‘to this day’, e.g. Jos. 8:28–29). The purpose of such stories was to explain contemporary phenomena in terms of past events. Specifically, the conquest narratives explain the existence in the land of ruined ancient cities. These aetiologies were originally Benjaminite traditions, later given a pan-Israelite colouring through the role of Gilgal (in Benjamin) as the central sanctuary of an amphictyony of twelve tribes. The war stories of Joshua 10:1–15 and 11:1–15 are viewed as belonging originally to the tribes of Benjamin and Naphtali respectively, likewise acquiring their ‘all Israel’ flavour through the formation of the amphictyony focused on a central shrine. Similarly Joshua, whose historical role was supposedly limited to the traditions of Joshua 17:14–18 and 24:1–28, acquired the status of leader of all Israel.48

This reconstruction of the formation of the tradition obviously depends heavily on the theory (developed chiefly by Noth) of a twelve-tribe Israelite amphictyony, and this has come under heavy criticism in recent years.49 Without it the ‘all Israel’ traditions (and, indeed, the fact of Israelite unity itself) are difficult to account for if the tribes had diverse origins.50

Secondly, Alt and Noth have been accused of assuming too readily that a narrative is unhistorical simply because it contains an aetiological element. Bright argued long ago that literary form alone cannot determine the historicity or otherwise of a story.51 B. S. Childs, C. Westermann and others have argued for a tighter definition of aetiology than the one Alt and Noth operated with, and Childs has shown that aetiological elements are secondary in the conquest narratives.52 In the light of such objections, even Weippert admits: ‘Any solution of the problems raised by Alt and Noth would then be possible only on the basis of external evidence.’53 That is, he admits that archaeology could theoretically arbitrate in matters of historicity. Noth himself was, in fact, more prepared to listen to archaeological evidence than his critics have sometimes allowed. Although he originally classified the story of the destruction of Hazor in Joshua 11:1–15 as an aetiology, he revised his opinion in the light of Yadin’s excavations and admitted that some historical reminiscence lay behind it.54 However, since archaeological evidence has generally been ambiguous at best, Weippert affirms that ultimately the question of the nature of Israel’s arrival in Canaan ‘still depends on the evaluation of the literary character of the texts’.55 Nevertheless, he (and implicitly Noth) has made an important methodological concession, for if another chronological setting could be found for the conquest, in which the relevant cities were destroyed, the aetiological explanation for the narratives would presumably be abandoned.

Optimism is not to be recommended, however, for Weippert’s use of archaeological evidence is subjective and inconsistent. Thus he is happy to state that Et-Tell is ‘definitely to be identified’ with Ai56 while rejecting the identification of El-Jib with Gibeon on the grounds that ‘… the results of Pritchard’s excavations … do not agree with the history of the town of Gibeon as this can be deduced from the written sources’57—a reference to the lack of LBA occupation there. This statement is truly astonishing, because the case for identifying El-Jib with Gibeon is actually much stronger (involving inscriptional evidence) than that for identifying Et-Tell with Ai. What lies behind Weippert’s inconsistency is the fact that the gap in occupation at Et-Tell is in keeping with the supposed aetiological nature of Joshua 8:1–19, whereas the gap at El-Jib is inconvenient because Weippert wants to make historical use of the tradition in Joshua 9; the narrative of Israel’s treaty with the Gibeonites in that chapter points, in his view, ‘to peaceful relations between the Canaanite cities and the “Israelite” groups which came into the country’.58 A more rigorous approach would demand either that Joshua 9 be divested of historical significance, or that Et-Tell be rejected as the site of Ai, because the results of excavations ‘do not agree with the history of the town … as this can be deduced from the written sources’!

In this instance (as, unfortunately, in others) the presuppositions of the infiltration theory are seen to be paramount. External evidence in practice has little or no chance of influencing ‘the evaluation of the literary character of the texts’; that evaluation, and the evaluation of the external evidence itself, are both actually determined in advance by the theory.

Archaeology and the nature of the Israelite settlement

In the 1950s, during a survey of southern Upper Galilee, Y. Aharoni discovered ‘many small Iron Age settlements in close proximity … sometimes only one or two kilometres apart’.59 The density of Iron Age settlement in this region exceeded that of any later period. On the basis of a pottery assemblage excavated during a trial dig at Khirbet et-Tuleil (Horvat Harashim), Aharoni assigned these settlements to the beginning of the Iron Age and dated them to the 13th–12th centuries bc. He stated: ‘There is no doubt, in my opinion, that this wave of settlement from the beginning of the Iron Age is Israelite.…’60

Subsequent surveys conducted in other parts of the hill country by a younger generation of Israeli archaeologists have produced similar results. It is now widely recognized that the early Iron Age saw a proliferation of small settlements in the highland regions. The total number discovered now exceeds 300. Dozens more occur on the Transjordanian plateau.

The majority of these sites are villages of only 1–2 acres, though a few are considerably larger. Most are unwalled, though some (e.g. Giloh, south of Jerusalem, and several Upper Galilee sites) had fortifications. The majority of the early Iron Age villages were new foundations; for example, in the centre of the country 97 of 114 Iron I sites showed no trace of Late Bronze occupation.61 The material culture of the Iron I villages is uniformly poor. They display a limited variety of pottery types, indicative of a subsistence economy.

Because the central highlands were the region initially colonized by the Israelite tribes, and because the highland settlements display cultural continuity with the Iron II period, which is the time of the Israelite monarchy, the majority of archaeologists have agreed with Aharoni that the Iron I settlements should be associated with the Israelites. For this reason the Iron I settlements have recently assumed a centre-stage position in the debate over the nature of lsrael’s emergence in Canaan. There have (predictably) been major differences of opinion over the interpretation of these settlements.

Aharoni held firmly to the view that the Galilee settlements predated the fall of Hazor and thus supported the infiltration theory; they showed, in his view, that the conquest of Canaanite cities had been preceded by a time of peaceful settlement. Yadin consistently opposed him,62 and recent studies have shown conclusively that Aharoni was wrong; the destruction of LBA Hazor preceded by some decades the earliest Iron I villages.63However, these studies cannot be said to have vindicated the conquest theory, since (as we have seen above) it no longer seems likely that the fall of LBA Hazor can be associated with the conquest.

Another aspect of the Iron I settlements which has emerged through more recent research is the high degree to which their material culture is in continuity with that of the preceding LBA, notwithstanding its relative poverty. In fact, continuity with the LBA has been observed in every individual aspect of Iron I culture: pottery, axes, daggers, knives, clothing-pins, artistic traditions, cultic objects, scripts and architecture.64 It is also evident (as will be discussed below) that the Iron I people were familiar with aspects of settled life and agriculture. These facts are difficult to square with the idea that they were pastoral nomads from the desert fringes. (As noted above, they are equally difficult to square with the notion that the settlers were Israelites newly arrived.)

The evidence indicates that the Iron I settlers were people who had had prolonged contact with the urban culture of LBA Canaan, not people who had initially settled well away from the city-states and later had only hostile relations with them. In short, the most recent assessment of the Iron I settlers does not sit comfortably with the infiltration theory, especially as formulated by Noth.

It is appropriate to mention here another objection which has been raised to the theory: that it depends on an outmoded view of nomadism.65 Noth characterized the ancestors of Israel as ‘land-hungry semi-nomads’ hankering after ‘a more settled life in the coveted agricultural countryside’.66 Recent studies of nomadism have made clear that the pastoral nomads of the ancient Near East were not ‘land-hungry’, nor did they live isolated from or hostile to settled societies. Their relationship with urban culture was generally close and symbiotic; the economic and social relationships between the semi-nomadic and sedentary populations were reciprocal, pastoral nomadism complementing an agricultural economy to maximize the potential of the land’s resources.67

These findings do not totally rule out the infiltration theory, though they do require its drastic modification. This has been recognized by its most recent supporters, who have begun the modification process.68 The scholar who has carried this furthest is V. Fritz, who has proposed what he calls a ‘symbiosis hypothesis’.69This takes account of the latest nomadism studies, and of the continuity between LBA material culture and that of the Iron I settlers. Whereas Noth favoured the 13th century bc as the time when Israel’s ancestors entered the land, Fritz pushes their arrival back to the 14th or even the 15th century bc.70 This allows for a long period of contact between Israel’s ancestors and Canaanite society of the LBA before She semi-nomadic groups eventually became sedentary at the beginning of Iron I. Only then did they become ‘visible’ archaeologically, since semi-nomads leave few traces of their existence. Their sedentarization, according to Fritz, was a response to changed economic conditions which affected the whole of Canaanite society at the end of the LBA.

While Fritz’s symbiosis hypothesis saves the infiltration theory with respect to the archaeological evidence and nomadism studies, it does not, of course, avoid the problems raised above with respect to the biblical traditions.

2. Hypotheses in which Israel is indigenous to Canaan

2a. The Peasant Revolt theory

G. E. Mendenhall’s attempt to construct a third alternative to the military conquest and peaceful infiltration theories emerged from a sense of frustration with current methodologies. The fact that such irreconcilable accounts of Israel’s origins could be developed from the same evidence led him to question fundamental assumptions underlying them both. He listed these assumptions as follows: ‘a. That the Twelve Tribes entered Palestine from some other area just prior to or simultaneously with the “conquest”. b. That the Israelite tribes were … “semi-nomads” who seized land and settled upon it during and after the conquest, c. That the solidarity of the Twelve Tribes was an “ethnic” one, and that kinship was the basis of the contrast between Israelite and Canaanite.’71

After debunking widely-held misconceptions concerning semi-nomads and their relationship with settled agriculturalists (as mentioned above), and the belief that tribal organization indicated a (semi-)nomadic background, Mendenhall discussed the relationship between the terms ‘Hebrew’ and ʿApiru (the latter occurring most notably in the ‘Amama letters’, correspondence between Canaanite city-state rulers and their Egyptian overlords in the 14th century bc). By assuming these terms to be synonyms, and defining them to mean someone who ‘has renounced any obligation to the society in which he formerly had some standing … and has in turn deprived himself of its protection’,72 Mendenhall produced a radically new insight into Israelite origins: ‘For if the early Israelites were called “Hebrews”, they could be termed so only from the point of view of some existing, legitimate political society from which they had withdrawn.’73 The early Israelites were therefore to be seen as the native peasantry of Canaan withdrawing from the oppressive regime of the city-state system. ‘There was no radical displacement of population, there was no genocide, there was no large-scale driving out of population, only of royal administrators (of necessity!).’74

Mendenhall did not, however, reject entirely the historicity of the Exodus-Sinai traditions. Indeed, they play a vital role in his hypothesis. In Mendenhall’s view, the Yahwistic faith, and the concept of a community related to Yahweh by covenant, were brought to Canaan by a group of former slave-labour captives who had escaped from oppression in Egypt. When this group reached Canaan, the loyalty and obligations of the covenant community proved ‘attractive to all persons suffering under the burden of subjection to a monopoly of power …’, who therefore identified themselves with the former slaves and rapidly swelled their community.75 But the original nucleus of Israel was ‘only a small group’; ‘there was no statistically important invasion of Palestine’.76

Like the theories which it was intended to replace, Mendenhall’s assumed that Israel’s origins were to be sought in the late 13th century bc. The reasons for fixing on this date are not at all clear, and Mendenhall seems to have inherited it somewhat uncritically from the older theories.

Mendenhall’s thesis initially had a cool reception. Its unfamiliar sociological approach made it less attractive than the older theories, which still retained their popularity. (The current fascination of biblical scholars with matters sociological was relatively undeveloped in 1962.) Also, the thesis was baldly stated and inadequately documented, which left it wide open to criticism. For example, Mendenhall had assumed, rather than argued, that ‘Hebrew’ was synonymous with ʿApiru, that the latter had the meaning he attributed to it, and that the political upheaval of the Amarna period provided a model for what happened in Canaan in the late 13th century bc. Critics raised objections to ail these points, and for at least a decade after its publication Mendenhall’s thesis was widely rejected, often after only the briefest discussion.77 A notable exception was provided by the second edition of Blight’s A History of Israel,78 which made significant concessions in Mendenhall’s direction. However, during the 1970s, as criticism of the methodology of the Baltimore School mounted, Mendenhall’s thesis received increasing attention in scholarly debate.79 It also attracted proponents who gave it a much stronger basis than Mendenhall had achieved for it in his original article.

The most important new proponent of the theory is N. K. Gottwald, who has developed it along distinctive lines.80 He departs from Mendenhall by rejecting the Mosaic covenant and its religious ideology as a significant factor in Israel’s emergence from Canaanite society, and, indeed, has expressed strong doubts about the historical value of the Exodus-Sinai traditions.81 Also, while Mendenhall depicts the polarization of Canaanite society and the emergence of Israel as happening virtually overnight, Gottwald extends the process, supplementing the ‘peasant rebellion’ model with one of ‘social revolution’, the latter ‘ebbing and flowing over two centuries’.82 Thirdly, Gottwald regards the Iron I settlements in the hill country as the work of rebel groups withdrawing to regions of lowest resistance;83 Mendenhall, on the other hand, believes that the rebels withdrew ‘not physically and geographically, but politically and subjectively’ from the existing regimes.84


As noted above, it has been easy for critics to pick holes in Mendenhall’s original statement of the peasant revolt theory. The defences and modifications produced by newer proponents such as Gottwaid and M. L. Chaney85 have removed some of its serious weaknesses, so that criticisms which could be levelled at Mendenhall do not apply to more recent versions of the theory. In the interests of economy, I will mention only those weaknesses which the main versions have in common.

The objection raised earlier to the infiltration theory, that it fails to account for the biblical traditions, is applicable also to the peasant revolt theory. Mendenhall offers the best explanation of how the biblical tradition arose, since he is able to point to a group which actually did come out of Egypt, become a covenant community at Sinai and subsequently migrate to Canaan. In his view, the larger groups which later joined these arrivals identified themselves so fully with the deliverance from Egypt that ‘the original historic events with which all groups identified themselves took precedence over and eventually excluded the detailed historical traditions of particular groups who had joined later’.86 It must remain a matter of opinion whether Israel’s Canaanite origins could have been so totally eclipsed by the notion that it was foreign to the land, but it seems highly unlikely. We must remember that this idea is found not only in the exodus tradition, but also in the stories of the patriarchs.87

Gottwald has tried to discover clues within the biblical tradition which indicate that Israel’s origins were at least in part indigenous, and he makes ingenious use of such snippets as Joshua 6:25 and 9:3–15.88 But this kind of evidence cannot bear the weight which he tries to place on it. All it shows is that, once it had arrived in the land, Israel assimilated certain Canaanite groups. It may, indeed, have absorbed more Canaanite groups than the traditions imply. But that is still a long way short of saying that Israel was in essence autochthonous and originated in a peasant revolt.89

The traditions concerning conquered cities have been explained as arising from the rebels’ victory over the city-state rulers. For example, Joshua 12 is viewed as a list of those kings whose holdings were successfully wrested from them.90 But the theory does not adequately account for those narratives in which entire cities are destroyed. It would clearly have been against the rebels’ own interests to destroy their own homes once they had liberated them from the city-state rulers.91

One fact which is extremely damaging for the theory is that there is no scrap of external evidence that a peasants’ revolt took place in the late 13th century bc. There are no texts such as the Amarna letters of the previous century which refer to any such event, and none is attested archaeologically. Indeed, the archaeology of 13th-century Canaan does not support the distinction between a rural peasantry and an urban aristocracy which supporters of the theory presuppose.92 The collapse of LBA cities and growth of Iron I villages have been interpreted as evidence of the revolt, but this view faces several serious problems.

First, L. E. Stager has noted that population figures deduced from LBA and Iron I settlement patterns seem to tell against the theory: ‘Given the low aggregate of the Late Bronze Age population throughout Canaan, it appears unlikely that the peasantry, even if they had all “revolted”, could have been large enough to account for the total Iron Age I village population …’.93 Secondly, if the Iron I highland societies were born out of conflict, it is surprising that so many of them were unfortified. This implies that prevailing conditions were peaceful and that no-one contested the occupation of the highlands.94 Thirdly, if the Iron I villages mark an attempt to break away from the city-state system, they should all lie some distance from the surviving centres of power. In fact this is not the case. Izbet Sartah lay close to the coastal plain,95 and a number of sites (Giloh, Tell et-Ful and Mevasseret Yerushalayim) clustered near Jerusalem, which, according to biblical tradition, remained outside Israelite hands until the time of David (i.e. the start of Iron IIA). Fourthly, as we will see below, there are strong reasons to believe that the Iron I villagers had formerly been pastoral nomads rather than farmers from the plains and valleys.

In short, the archaeological evidence suggests that the peasants’ revolt, theory is not an accurate account of the processes which were taking place in Canaan at the end of the LBA and the start of the Iron Age. Indeed, one of the strongest criticisms levelled at proponents of the theory is that they ignore or misapply archaeological evidence.96

2b. Other theories in which Israel is indigenous to Canaan

While Mendenhall’s theory, even as modified by others, has not won majority support, his criticisms of traditional assumptions (especially those concerning nomadism) have proved extremely influential. The peasant revolt theory has spawned several other views in which Israel originated within the land of Canaan, so that there is now a general consensus that Israel was autochthonous.97

An objection which can be made to all these views is that they share with the peasant revolt theory the same difficulties with respect to the biblical traditions. To avoid becoming repetitious, this will not be pointed out in each of the following cases. They also have another major difficulty in common, which will be discussed after the various views have been outlined.

2bi. Finkelstein’s view of Israel’s nomadic origins

I. Finkelstein has recently developed a theory,98 on the basis of archaeological studies of the Iron I settlement process, which shares certain features with Fritz’s ‘symbiosis hypothesis’ discussed above. Finkelstein agrees with Fritz that the Iron I settlers were formerly pastoral nomads who had had prolonged contact with LBA Canaanite culture. However, while Fritz retains the notion that these semi-nomads had originally entered the land from elsewhere, Finkelstein rejects it. He admits that early Israel may have had diverse origins, and concedes that there is probably a historical kernel to the tradition of an exodus from Egypt, but he believes that the vast majority of Iron. I settlers were indigenous.

His argument in favour of their semi-nomadic status before the beginning of the Iron Age is much more rigorously developed than Fritz’s and is worth summarizing briefly. Essentially Finkelstein makes four points in favour of viewing the Iron I settlements as evidence of a sedentarization process.

i. The earliest Iron I villages were established on the central ridge and in the small intermontane valleys—the best areas for grazing and dry farming. Orchard agriculture only came later with the settlement of the western slopes of the Ephraimite hill country. This pattern of settlement indicates that the settlers had a pastoral rather than an urban or rural background.

ii. The layout of many Iron I sites—typically an elliptical plan in which dwellings surround a central space—resembles the layout of the duwwar tent encampments of nomadic groups, and is to be seen as a development from them.

iii. Extensive use of subterranean storage silos by the Iron I villagers is a characteristic of nomadic societies settling down, because the most urgent requirement of such societies is storage space for silage.

iv. Simple broadroom houses and pillared, four-room houses are developments from the bedouin-style tent, and thus indicate the nomadic antecedents of the Iron I settlers.

These arguments are not all of equal value, and the last two are particularly open to criticism.99Nevertheless Finkelstein has succeeded in greatly strengthening the case for viewing the Iron I settlements as evidence for the sedentarization of pastoral nomads. He has also made an important contribution by presenting archaeological evidence for the existence of a significant population of pastoral nomads in LBA Palestine. Nomadic groups are notoriously difficult to detect archaeologically, but Finkelstein argues that sanctuaries and cemeteries away from the centres of settled population point to the existence of such groups during the LBA, and he tentatively identifies them with the shasu/sutu referred to in LBA texts.100

Like Fritz, Finkelstein explains points of continuity between the material cultures of LB and Iron I in terms of close contact between these pastoral nomads and Canaanite society during the LBA.101 Also like Fritz, he interprets the sedentarization of pastoral nomads at the start of Iron I as a response to changed economic conditions which attended the collapse of the LBA city-states. He departs from Fritz, however, in arguing that the semi-nomadic groups were indigenous to Canaan.

Finkelstein suggests that their origin should be sought in the deterioration of Palestine’s rural and urban systems at the end of the Middle Bronze Age, i.e. in the 16th century bc. As permanent settlements were destroyed or abandoned, a large proportion of the sedentary population became nomadized, and remained so until the process was reversed at the beginning of the Iron Age. While proof of this thesis is lacking, it does explain what became of Palestine’s population during the LBA; the sedentary population was drastically reduced at the end of the MBA, but increased again at the start of Iron I, and the suggestion that the ‘missing’ proportion lived as semi-nomads, sparsely attested by archaeological evidence, makes sense of this situation.102 It is not, however, the only possible explanation, as will be pointed out below.

Although Finkelstein uses the term ‘Israelite’ to describe the Iron I settlers, he stresses that in doing so he is simply using it as ‘a terminus technicus for “hill country people in a process of settling down” ’.103 In fact he does not believe there was such a thing as Israelite identity as early as the sedentarization stage: ‘The formation of the Israelite identity was a long, intricate and complex process which, in our opinion, was completed only at the beginning of the monarchy.’104 We will see below that this evaluation is unacceptable, even if one shares Finkelstein’s disregard for the biblical traditions as a historical source.

2bii. Callaway’s theory of displaced populations from the coastal plain

The late J. A. Callaway105 has put forward arguments against semi-nomadic origins for the Iron I settlers, preferring to view them as Canaanite villagers displaced from the coastal plain and the Shephelah.

Referring to the use of household cisterns at highland sites such as Tell en-Nasbeh and Khirbet et-Tell (as many as three to each house at the latter), he points out that their location beneath walls and floors suggests that they were excavated before the houses were built, which in turn suggests that the settlers already possessed a working knowledge of household cisterns, and the technology to create them, when they arrived in the highlands. He also points to the ability of the Iron I settlers to create agricultural terraces, making possible the cultivation of the steeply sloping hillsides. He sees this ability as a prerequisite for the colonization of the highlands, and evidence for an agricultural background for the settlers. He also argues that the material remains of the Iron I villages point to a cultural background in the Shephelah and the coastal plain. Ceramic and metal artefacts and inscribed objects are cited in support of these cultural links. In Callaway’s view the Iron I settlers in the hills were refugees from population pressure and conflict in the coastal plain and Shephelah, caused by the arrival of the Philistines and other ‘Sea Peoples’. These highland settlers eventually emerged as Israel, so that Israel’s origins must ultimately be sought in the Canaanite villages of the plains and lowlands. Callaway’s view therefore has something in common with the peasant revolt theory, in that the Iron I settlers are presented by both theories as refugees from the lowland districts. But whereas the peasant revolt theory depicts them as escaping from oppressive economic exploitation, Callaway would see them as refugees from aggressive newcomers.

Quite apart from the fact that Callaway’s theory runs into many of the same difficulties as the peasant revolt theory, he has not succeeded in establishing an agricultural background for the Iron I settlers. He has not produced any evidence which cannot be accounted for in terms of close contact over a long period between semi-nomads and the settled Canaanites of the LBA. Cistern technology could have been acquired through such contact; cisterns (even waterproofed with plaster) were used at, e.g., Hazor and Taanach in the LBA. The artefacts and inscribed objects which Callaway cites as evidence of cultural roots in the lowlands could indicate continuing contact with the surviving cities of the coastal plain (through limited trade or transhumance) after the semi-nomads had begun to settle down. The implications of terracing technology for the origin of the highland settlers are at best uncertain. Callaway assumes that the settlers brought the technology with them ready-made,106 but D. C. Hopkins, in a major study of highland agriculture, takes a different view. He argues that terracing techniques were not an essential prerequisite for cultivation in the highlands, and that it is ‘much more cogent’ to assume that the techniques were developed by the settlers as they sought to cope with the exigencies of their highland environment.107

The economy of the Iron I villages at Khirbet et-Tell and Khirbet Raddana, to which Callaway chiefly refers, does not help his case. Animal enclosures and domestic stables clearly indicate the importance of sheep and goats in that economy, while the areas available for cultivation were very limited. Callaway admits this, and estimates that perhaps ‘up to one-half of a family’s subsistence came from its animals’.108 Without wishing to play down the importance of dry farming in the economy, we must question Callaway’s insistence that these villages ‘were established from the beginning on a subsistence base of agriculture primarily and small cattle secondarily’.109 In short, his view that the settlers were originally agriculturalists rather than shepherds is an assumption, not a conclusion forced upon us by the evidence.

Finally, if the Iron I villagers were refugees from the coastal plain and the Shephelah, moving inland under pressure from the Philistine invasion of those regions, we would expect them to have primarily colonized the highlands of Judah, immediately to the east. In fact, however, early Iron Age settlements were very sparse in that area; only about ten are known, compared with 115 in Ephraim and ninety-five in Manasseh (i.e. between Jerusalem and the Jezreel Valley).110 In short, Callaway’s theory accounts for only a small part of the archaeological data (which is just as well accounted for in other ways), and is contradicted by the bulk of it.

2biii. Lemche’s ‘evolutionary Israel’

In the context of a major critique of the peasant revolt theory, N. P. Lemche has put forward his own alternative view of Israelite origins. Rejecting Israel’s revolutionary beginnings, he prefers to speak of an ‘evolutionary Israel’.111

Lemche is even more radical than Mendenhall and Gottwald in breaking with the biblical traditions. He considers the traditions of Israel’s early history to be so late in origin as to be useless for historical reconstruction: ‘… I propose that we decline to be led by the Biblical account and instead regard it, like other legendary materials, as essentially ahistorical, that is, as a source which only exceptionally can be verified by other information.’ His alternative reconstruction is based entirely on what we can deduce from archaeological materials ‘of the social, economic, cultural and political developments in Palestine towards the close of the second millennium’.112 Lemche does not wish, however, to limit his investigations to the late 13th century bc, for the Israelite state may have been formed ‘as the result of a development which may actually have covered the whole of the Late Bronze Age (or, for that matter, an even longer period of time)’.113 After surveying the conditions and institutions of the LBA, and examining possible reasons for the economic and cultural decline which marked the end of the period, Lemche offers a ‘working hypothesis’, which may be summarized as follows:

From at least as early as the first half of the 14th century bc the central highlands were the habitation of the ʿApiru, whom he defines as ‘a para-social element … [consisting] of runaway former non-free peasants or copyholders from the small city-states in the plains and valleys of Palestine’.114 In other words, they had once been settled agriculturalists. These groups are attested by the Amarna correspondence, but do not otherwise feature in archaeological evidence for that period, because they were not sedentary at that time; they lived as ‘outlaw groups of freebooters’.115 However, when new settlements appear in the highlands over a century later, at the start of the Iron Age, they are evidence of new political structures emerging among those same groups. The Iron I settlements attest a return by those groups to a settled, agricultural lifestyle, and the beginning of a (re)tribalization process. Israel was the end-product of that process.

Lemche’s view has much in common with Finkelstein’s. Both see Israel developing from non-sedentary groups in the highlands, and both believe those groups had formerly been settled. However, they clearly hold different views on the origins and lifestyles of those groups. Lemche also offers a different explanation for their sedentarization. He believes that their return to an agricultural existence was not possible before the late 13th century bc, because agriculture could not be carried out in the highlands without terracing techniques and lined water-cisterns, neither of which, according to Lemche, was developed Until that time.116 This argument is completely fallacious, Since (in common with Callaway’s argument discussed above) it makes wrong assumptions about both these technological innovations. However, the major weaknesses of Lemche’s view will be brought out below, after I have outlined one more theory of indigenous origins.

2biv. The synthesis of Coote and Whitelam

The reconstruction put forward by R. B. Coote and K. W. Whitelam is more complex and nuanced than that of either Finkelstein or Lemche. Nevertheless, it has some features in common with both.

Like Lemche, Coote and Whitelam explicitly reject the biblical narratives as a source for the reconstruction of Israeli’s early history. Rather, the historian’s task is ‘to explain the archaeological record in the context of comparative history and anthropology’.117 They review an impressive range of evidence in an attempt to fulfil this task, and are to be commended for producing a challenging new synthesis, only one small part of which can be commented on here—that which bears most directly on the nature of Israel’s origins.118

After examining the nature and location of the Iron I settlements, they conclude that the archaeological evidence is not compatible with the conquest, infiltration or revolt theories. Rather, the origin of those settlements is to be set in the context of an economic decline which occurred at the end of the LBA, probably resulting from a breakdown of the interregional trade on which Canaan’s urban economy ultimately depended. This urban economic collapse triggered a number of processes.

As the lowland urban centres declined in material prosperity, a settlement shift occurred among rural groups, including agriculturalists, pastoral nomads and even bandits, all of whom were economically dependent on the lowland cities to some extent. The development of agriculture in the highlands, which had not been economically viable during the LBA, now became an attractive option for such groups. Settlements were founded, based on a mixture of agriculture and pastoralism. Subsequently ‘The settlement into villages in the hinterland was given political and incipient ethnic form in the loosely federated people calling themselves Israel’.119

Thus in the reconstruction offered by Coote and Whitelam the ancestors of Israel were not exclusively peasants, bandits or pastoral nomads, but a mixture of all three, thrown together by the seismic effect of the decline in inter-regional trade. The attractiveness of such a broad synthesis is that it avoids the weaknesses of narrower approaches. It seems able to accommodate virtually all the available data (except, of course, for the biblical traditions, which are explicitly rejected on methodological grounds). We will now see, however, that there is at least one piece of evidence which overturns all four of the theories reviewed in this section.

A major criticism of the above theories

All four of the theories outlined above, either explicitly or implicitly, give a relatively late date to the formation of an entity called Israel. Israel did not even begin to take shape until a group or groups embarked on the process of settlement in the highlands, as evidenced by archaeology. In other words, there was no ‘Israel’ before the beginning of Iron I, and perhaps not for some time afterwards.

It is well known that the earliest reference to Israel outside the Bible comes from the late 13th century bc. This occurs in the final stanza of a hymn celebrating the victories of the pharaoh Merenptah (or Merneptah), inscribed on what has become known as the ‘Israel stela’, dating from Merenptah’s fifth year, i.e. 1207 bc in the low chronology now preferred for Egypt’s 19th Dynasty. It is the relative dating of this reference to Israel and the earliest Iron I settlements which vitiates the above theories of Israel’s origins.

Until recently the beginning of the Iron Age has been very loosely defined. It has commonly been assigned a date of around 1200 bc, in recognition that this is merely a handy, round number, and that some Iron I settlements may have appeared earlier and some later. Hence Lemche, discussing the appearance of Iron I villages, is able to say: ‘The date of all this activity perhaps falls towards the close of the 13th century; many are in any case from the beginning of the 12th century’.120 Coote and Whitelam remark: ‘Israel originated during the third and fourth quarters of the thirteenth century with the shift in land use and settlement patterns of the Palestinian highland and dry land margin.’121 Finkelstein also assigns the beginning of sedentarization in the highlands to the late 13th century bc, though he admits that the data for this are ‘few and inconclusive’.122However, it has emerged from B. G. Wood’s recent, magisterial study of the LBA/Iron I transition that the beginning of Iron I should actually be dated close to 1170 bc, and not loosely to around 1200 bc as previously supposed.123 This means that we have irrefutable evidence for the existence of Israel some 30–40 years before the earliest Iron I settlements which supposedly mark the beginning of Israel’s formation. Coote and Whitelam may be hoping to head off such a criticism when they suggest that Merenptah’s inscription ‘may not refer to … any social group directly ancestral to monarchic Israel’.124 This fantastic statement is clearly special pleading. The evidence of Merenptah’s stela cannot be disposed of simply because it does not fit a particular theory of Israelite origins; instead, the theory must be adapted to fit the evidence.125

In summary, it is clear from Merenptah’s inscription that Israel’s origins must be sought before the beginning of Iron I. Other evidence from within the OT itself points to the same conclusion. Kitchen has repeatedly put forward arguments demonstrating that the form of the Sinai covenant must go back to the second millennium bc,126 and several scholars have argued for the pre-monarchic origins of much early Hebrew poetry, effectively demonstrating the great antiquity of the historical traditions which it embodies.127The traditions of Israel’s early history cannot be cavalierly relegated to an exilic or other late date in the way that Lemche, Coote and Whitelam and others have stated.

Conclusion: scope for an alternative view

Both internal and external evidence requires that we treat with greater respect Israel’s traditions concerning her origins and early history. This in turn requires that we look for a historical, archaeological and cultural context in which the traditions concerning the exodus, wilderness wanderings and conquest can retain their integrity. Many evangelical scholars have long believed that the 13th century bc provides such a context. As we noted earlier in this article, it now seems very unlikely that it does. Is there an alternative?

I have argued in detail elsewhere128 for a return to the 15th-century date implied by the OT’s internal chronology (1 Ki. 6:1; Jdg. 11:26). Some of the reasons for rejecting this date (such as the alleged gap in occupation in Transjordan) have disappeared and others are not so strong as has sometimes been supposed. One major difficulty for the 15th-century date has been the apparent absence of evidence for a violent conquest at the end of that century. In response, I have tried to Show that the missing evidence is probably provided by the fall of Canaan’s fortified cities at the close of the MBA.129 This event has traditionally been dated between 1550 and 1500 bc and attributed to Egyptian campaigns. Arguments against the traditional view are now emerging, lending plausibility to my suggestion that these cities were actually destroyed about a hundred years later, and that their destroyers were the incoming Israelites.130 It will be a good while before enough evidence is available for a final verdict (insofar as final verdicts are ever reached in such matters!), but this approach currently seems to be a promising one.

Several of the arguments put forward by the proponents of alternative views actually complement this theory. As noted above, Finkelstein draws attention to the collapse of Canaan’s urban culture at the end of the MBA, and to evidence for a significant semi-nomadic population during the LBA. He suggests that the change should be explained in terms of the nomadization of a large proportion of the previously urban population. An alternative hypothesis would be that invaders slaughtered or dispersed a large part of the MBA urban population, and subsequently lived in the land as semi-nomads. For a variety of political and economic reasons, settlement may not have been an attractive option for the newcomers, until, that is, the socio-economic complexion of the country was changed once again by the decline of Canaan’s remaining city-states at the end of the LBA. I am suggesting, of course, that the invaders were the Israelites, who certainly arrived in Canaan as tent-dwellers and remained such for an uncertain period after entering the land.131 This view allows them to be a well-established part of the Canaanite scene by the time of Merenptah, though not archaeologically ‘visible’ until their sedentarization a few decades later, at the beginning of Iron I.

Obviously, much more would need to be said to fill out and defend this picture, and this is not the place to do it. I am simply trying to show that the biblical traditions are not incompatible with some of the new understandings of Canaan’s archaeological, social and economic history.

Whether more evangelical scholars will find this approach attractive remains to be seen. What is certain is that, a major rethink is required among defenders of the biblical traditions concerning Israel’s origins. The old synthesis of the Baltimore School cannot be kept alive by tinkering with a few details. The real need is for a completely new synthesis which takes account of the best of recent analyses of archaeological and other evidence while refuting the worst.

1 G. E. Mendenhall, ‘The Hebrew Conquest of Palestine’, BA XXV/3 (1962), pp. 66–87.

2 E.g. L. T. Wood, ‘The Date of the Exodus’, in J. Barton Payne (ed.), New Perspectives on the Old Testament(Waco: Word Books, 1970), pp. 67–86; B. K. Waltke, ‘Palestinian Artifactual Evidence Supporting the Early Date for the Exodus’, Bibliotheca Sacra 129 (1972), pp. 33–47; C. F. Aling, Egypt and Bible History from Earliest Times to 1000 bc (Grand Rapids: Baker Book House, 1981), pp. 77–96; C. H. Dyer, ‘The Date of the Exodus Reexamined’, Bibliotheca Sacra 140 (1983), pp. 225–243; E. H. Merrill, Kingdom of Priests (Grand Rapids: Baker Book House, 1987), pp. 57–75. It has to be said that a number of these defences of the 15th-century date have been seriously flawed (see Bimson, as in following note, pp. 114–119, 250, and my review of Merrill forthcoming in Christian Arena).

3 J. J. Bimson, Redating the Exodus and Conquest (Sheffield: Almond Press, 21981).

4 For detailed references see ibid., pp. 44–46.

5 Ibid.

6 O. Tufnell, ‘Lachish’, in D. Winton Thomas (ed.), Archaeology and Old Testament Study (Oxford: OUP, 1967), p. 302. (Tufnell’s view has been vindicated by more recent excavations; see D. Ussishkin cited in n. 40.)

7 See conveniently Y. Yadin, ‘Hazor’, in ibid., p. 260.

8 J. Bright, A History of Israel (London: SCM, 1960), p. 120.

9 G. E. Wright, Biblical Archaeology (London: Duckworth, 21962), p. 60. The fact that a different site is now favoured for Raamses has not changed this conclusion; see Kitchen as cited in n. 27.

10 See conveniently N. Glueck, The Other Side of the Jordan (Cambridge MA: American Schools of Oriental Research, 1940), pp. 114, 125–147.

11 E.g. K. A. Kitchen and T. C. Mitchell, ‘Chronology of the Old Testament’, in J. D. Douglas (ed.), New Bible Dictionary (London: Tyndale Press, 1962), pp. 214–216; revd. edn. (Leicester: IVP, 1982), pp. 191–192; K. A. Kitchen, Ancient Orient and Old Testament (London: Tyndale Press, 1966), pp. 57–75; The Bible in its World(Exeter: Paternoster Press, 1977), pp. 75–91.

12 E.g. R. K. Harrison, Introduction to the Old Testament (London: Tyndale Press, 1970), pp. 174–176, 315–325.

13 E.g. the works cited in n. 2.

LBA Late Bronze Age

14 K. M. Kenyon, Digging Up Jericho (London: Ernest Benn, 1957), pp. 256–265; Kenyon subsequently modified the dates slightly: Archaeology in the Holy Land (London: Ernest Benn, 41979), p. 208. For the most recent treatment of the relevant Jericho material see P. Bienkowski, Jericho in the Late Bronze Age(Warminister: Aris and Phillips, 1986).

15 J. Marquet-Krause, Les fouilles de ‘Ay (et-Tell) 1933–35 (Paris: Bibliothèque Archéologique et Historique, 1949).

16 E.g. J. A. Callaway, ‘New Evidence on the Conquest of Ai’, JBL 87 (1968), pp. 312–320.

17 J. B. Pritchard, ‘Culture and History’, in J. P. Hyatt (ed.), The Bible in Modern Scholarship (Nashville: Abingdon, 1965), p. 319.

18 Ibid.

19 Bright, History (London: SCM Press, 31981), p. 130.

20 Wright, Biblical Archaeology, p. 80; Yadin, ‘Is the Biblical Account of the Israelite Conquest of Canaan Historically Reliable?’, BAR VIII/2 (1982), p. 22.

21 W. F. Albright, ‘The Kyle Memorial Excavation at Bethel’, BASOR 56 (1934), pp. 3, 11.

22 Albright, The Biblical Period from Abraham to Ezra (New York: Harper Torchbooks, 41963), p. 30.

23 J. M. Miller, ‘Archaeology and the Israelite Conquest of Canaan: Some Methodological Observations’, PEQ109 (1977), pp. 87–93; E. F. Campbell and J. M. Miller, ‘W. F. Albright and Historical Reconstruction’, BA 42/1 (1979), pp. 37–46; J. M. Miller, ‘Old Testament History and Archaeology’, BA 50/1 (1987), pp. 57–58.

24 Kitchen, Ancient Orient, pp. 63–65.

25 See Bimsom, Redating, pp. 188–191.

26 C. H. Dyer, ‘The Date of the Exodus Reexamined’, p. 229.

27 Kitchen, Ancient Orient, pp. 59–62; Kitchen and Mitchell, ‘Old Testament Chronology’, New Bible Dictionary2, p. 191.

28 See Bimson, Redating, pp. 61–68; R. G. Boling, The Early Biblical Community in Transjordan (Sheffield: Almond Press, 1988), pp. 11–35.

29 Details in Bimson, ‘Exodus and Conquest: Myth or Reality?’, Journal of the Ancient Chronology Forum 2(1988), p. 29.

30 Kitchen, Ancient Orient, pp. 73–74.

31 Bimson, Redating, pp. 82–83; ‘Archaeological Data and the Dating of the Patriarchs’, in A. R. Millard and D. J. Wiseman (eds.), Essays on the Patriarchal Narratives (Leicester: IVP, 1980), pp. 82–83.

32 Kitchen, Ancient Orient, p. 74.

33 E.g. ibid., pp. 90–102; The Bible in its World, pp. 79–85; ‘The Fall and Rise of Covenant, Law and Treaty’, Tyndale Bulletin 40 (1989), pp. 118–135.

34 Kitchen’s evidence could theoretically be reconciled with a 15th-century date for the exodus by assuming literary activity on the relevant texts during the 13th century bc. That editorial work continued, e.g. on Deuteronomy, after the time of Moses is evident from Dt. 1:1; 2:12; 34:1–12.

35 R. Gonen, ‘Urban Canaan in the Late Bronze Period’, BASOR 253 (1984), p. 69.

36 Yadin, ‘Biblical Archaeology Today: The Archaeological Aspect’, in J. Aviram et al., Biblical Archaeology Today (Jerusalem: Israel Exploration Society), pp. 23–24.

37 B. G. Wood, Palestinian Pottery of the Late Bronze Age: An Investigation of the Terminal LB IIB Phase (PhD thesis, University of Toronto, 1985), pp. 369–599; ‘The Palestinian Evidence for a Thirteenth Century conquest: An Archaeological Appraisal’, paper presented at the symposium Who Was the Pharaoh of the Exodus? held in Memphis TN, April 23–25, 1987, publication forthcoming. The absolute dates depend on the chronology adopted for Egypt. I have used the low chronology currently favoured by a number of Egyptologists; see K. A. Kitchen, ‘The Basics of Egyptian Chronology in Relation to the Bronze Age’, in P. Åström (ed.), High, Middle or Low? (Gothenburg: Paul Åströms Förlag, 1987), Pt. 1, pp. 37–55.

38 M. Kochavi, ‘Khirbet Rabûd = Debir’, Tel Aviv 1 (1974), pp. 2–33.

39 D. Livingston, ‘Location of Bethel and Ai Reconsidered’, Westminster Theological Journal 33/1 (1970), pp. 20–44; ‘Traditional Site of Bethel Questioned’, WTJ 34/1 (1971), pp. 39–50.

40 D. Ussishkin, ‘Lachish—Key to the Israelite Conquest of Canaan’, BAR XIII/1 (1987), pp. 35–39.

41 M. Noth, The History of Israel (London: Black, 21960), p. 81.

42 A. Alt, Die Landnahme der Israeliten in Palāstina (Leipzig: Reformationsprogramm der Universität Leipzig, 1925). English translation in Alt, Essays on Old Testament History and Religion (Oxford: Blackwell, 1966), pp. 135–169.

43 See M. Weippert, The Settlement of the Israelite Tribes in Palestine (London: SCM Press, 1971), pp. 25–46.

44 Weippert, op. cit.; ‘The Israelite w and the Evidence from Transjordan’, in F. M. Cross (ed.), Symposia Celebrating the 75th Anniversary of the Founding of the American Schools of Oriental Research (1900–1975)(Cambridge MA: ASOR, 1979), pp. 15–34.

45 J. M. Miller, ‘The Israelite Occupation of Canaan’, in J. H. Hayes and J. M. Miller (eds.), Israelite and Judaean History (London: SCM, 1977), pp. 279–284.

46 For his final statement of this see Aharoni, The Archaeology of the Land of Israel (London: SCM, 1982), pp. 177–180.

47 M. Kochavi, ‘The Israelite Settlement in Canaan in the Light of Archaeological Surveys’, in Aviram, Biblical Archaeology Today, pp. 54–60; A. Zertal, The Israelite Settlement in the Hill Country of Manasseh, PhD thesis, Tel Aviv University, 1986 (Hebrew with English abstract).

48 See the summary in Weippert, Settlement, pp. 37–41.

49 E.g. C. H. J. De Geus, The Tribes of Israel (Amsterdam: Van Gorcum, 1976), pp. 193–209; A. D. H. Mayes, Israel in the Period of the Judges (London: SCM, 1974), pp. 15–83, and more briefly in ‘The Period of the Judges and the Rise of the Monarchy’, in Hayes and Miller, Israelite and Judaean History, pp. 299–308; R. de Vaux, The Early History of Israel (London: Darton, Longman & Todd, 1978), vol. 2, pp. 695–715; N. K. Gottwald, The Tribes of Yahweh: A Sociology of the Religion of Liberated Israel (London: SCM, 1980), pp. 345–357.

50 Several scholars have suggested that tribal unity was an achievement of David, but this view raises its own problems; see G. W. Ramsey; The Quest for the Historical Israel (London: SCM, 1982), pp. 88–90, and references supplied there. Cf. P. C. Craigie, ‘The Conquest and Early Hebrew Poetry’, Tyndale Bulletin 20 (1969), pp. 76–94, for evidence of unity in the conquest period.

51 J. Bright, Early Israel in Recent History Writing (London: SCM, 1956), pp. 90–91.

52 B. S. Childs, ‘A Study of the Formula “Until This Day” ’, JBL 82 (1963), pp. 279–292; ‘The etiological tale re-examined’, VT 24 (1974), pp. 387–397; C. Westermann, ‘Arten der Erzählung in der Genesis’, in Forschung am Alten Testament: Gesammelte Studien (Munich: Chr. Kaiser Verlag, 1964), pp. 40–44.

53 Weippert, ‘Canaan, Conquest and Settlement of’, in K. Crim (ed.), IDB Supplement Volume (Nashville: Abingdon, 1976), p. 128.

54 Weippert, Settlement, pp. 36–37.

55 Weippert, ‘Canaan, Conquest and Settlement of’, p. 128.

56 Weippert, Settlement, pp. 24–25.

57 Ibid., p. 13, n. 30.

58 Ibid., p. 18.

59 Y. Aharoni, ‘Problems of the Israelite Conquest in the Light of Archaeological Discoveries’, Antiquity and Survival 2 (1957), p. 146.

60 Ibid., p. 149.

61 L. E. Stager, ‘The Archaeology of the Family in Ancient Israel’, BASOR 260 (1985), p. 3.

62 The debate is documented by Yadin, ‘The Transition from a Semi-Nomadic to a Sedentary Society in the Twelfth Century BCE’, in Cross, Symposia, pp. 57–68.

63 I. Finkelstein, The Archaeology of the Israelite Settlement (Jerusalem: Israel Exploration Society, 1988), pp. 97–101, 315–317.

64 Pritchard, ‘Culture and History’, pp. 320–321; A. Mazar, ‘Israelite Settlement in the Light of Excavations’, in Aviram, Biblical Archaeology Today, pp. 61–71; V. Fritz, ‘Conquest or Settlement?’, BA 50/2 (1987), pp. 96–97.

65 Miller in Hayes and Miller, Israelite and Judaean History, p. 270.

66 Noth, History, p. 69.

67 See especially Gottwald, Tribes of Yahweh (see n. 49), pp. 437–448; M. L. Chaney, ‘Ancient Palestinian Peasant Movements and the Formation of Premonarchic Israel’, in D. N. Freedman and D. F. Graf (eds.), Palestine in Transition: The Emergence of Ancient Israel (Sheffield: Almond Press, 1983), pp. 42–43.

68 E.g. Weippert, ‘The Israelite “Conquest” and the Evidence from Transjordan’ (see n. 44).

69 Fritz, ‘Conquest or Settlement?’, p. 98.

70 Fritz, ‘The Israelite “Conquest” in the Light of Recent Excavations at Khirbet el-Meshâsh’, BASOR 241 (1981), p. 71.

71 Mendenhall, ‘The Hebrew Conquest of Palestine’, p. 67.

72 ibid., p. 71.

73 Ibid.

74 ibid., p. 73.

75 ibid., p. 74.

76 ibid., pp. 73, 79.

77 E.g. R. de Vaux, The Early History of Israel, vol. 2, pp. 485–487. (However, note that de Vaux allows social revolt to play a part in his own reconstruction of the settlement of the northern tribes; ibid., pp. 677–678).

78 Bright, History (London: SCM, 21972), pp. 133–139; Mendenhall’s theory is adopted even more confidentially in History (London: SCM, 31981), pp. 137–143.

79 See e.g. the papers by A. J. Hauser and T. L. Thompson, with replies by Mendenhall and Gottwald, in JSOT7 (1978), pp. 2–52, and the literature cited therein; also the essays in Freedman and Graf, Palestine in Transition, all of which were written during the 1970s.

80 Most notably in Tribes of Yahweh (see n. 49), but also in several essays, some of which are referred to below.

81 Gottwald, ‘The Israelite Settlements a Social Revolutionary Movement’, in Aviram, Biblical Archaeology Today, p. 35.

82 Ibid., pp. 37–38.

83 Gottwald, ‘The Hypothesis of the Revolutionary Origins of Ancient Israel’, JSOT 7 (1978), pp. 44, 50.

84 Mendenhall, ‘Hebrew Conquest’, p. 73. Mendenhall has suggested that the Iron I settlements were the work of newcomers from the north (e.g. Mendenhall, ‘Israel’s Hyphenated History’, in Freedman and Graf, Palestine in Transition, pp. 97, 99), a view for which evidence is completely lacking and which currently has no following.

85 M. L. Chaney, ‘Ancient Palestinian Peasant Movements’, in Freedman and Graf, Palestine in Transition, pp. 39–90.

86 Mendenhall, ‘Hebrew Conquest’, p. 74.

87 Cf. Ramsey, Quest, p. 94.

88 Gottwald, ‘The Hypothesis of the Revolutionary Origins of Ancient Israel’, pp. 41–42; Tribes of Yahweh, pp. 215–216.

89 Cf. ‘Hauser, ‘Israel’s Conquest of Palestine: A Peasants’ Rebellion?’, JSOT 7 (1978), p. 10, a point which stands in spite of Gottwald’s reply (previous note).

90 Gottwald, ‘The Hypothesis of the Revolutionary Origins of Ancient Israel’, p. 41.

91 Cf. E. F. Campbell, ‘Moses and the Foundations of Israel’, Interpretation 29 (1975), p. 152.

92 T. L. Thompson, ‘Historical Notes on [Hauser’s] “Israel’s Conquest of Palestine: A Peasants’ Rebellion?” ’, JSOT 7 (1978), pp. 24–25; Finkelstein, Archaeology of the Israelite Settlement, p. 311.

93 L. E. Stager, responding to Gottwald, in Aviram, Biblical Archaeology Today, p. 84.

94 J. A. Callaway, ibid., p. 75; R. B. Coote and K. W. Whitelam, The Emergence of Early Israel in Historical Perspective (Sheffield: Almond Press, 1987), p. 135; Finkelstein, Archaeology of the Israelite Settlement, p. 313.

95 Cf. Coote and Whitelam, Emergence of Early Israel, p. 127.

96 Finkelstein, Archaeology of the Israelite Settlement, p. 308.

97 In addition to those scholars whose views are discussed below, see G. W. Ahlström, Who Were the Israelites? (Winona Lake: Eisenbrauns, 1986).

98 Developed and defended most fully in Finkelstein, Archaeology of the Israelite Settlement.

99 See the review of Finkelstein’s book by D. Esse, BAR XIV/5 (1988), p. 10.

100 Finkelstein, Archaeology of the Israelite Settlement, pp. 343–345.

LBA Late Bronze Age

101 Ibid., p. 344, though Finkelstein greatly plays down the extent of this continuity (cf. Ibid., p. 312).

102 Ibid., pp. 339–348.

103 Ibid., p. 28.

104 Ibid., p. 27.

105 J. A. Callaway, ‘A New Perspective on the Hill Country Settlement of Canaan in Iron Age I’, in J. N. Tubb (ed.), Palestine in the Bronze and Iron Ages: Papers in Honour of Olga Tufnell (London: Institute of Archaeology, 1985), pp. 31–49; also (in slightly different form) in Aviram, Biblical Archaeology Today, pp. 72–77.

106 A view also held by G. W. Ahlström, ‘Where Did the Israelites Live?’, JNES 41 (1982), pp. 133–138.

107 D. C. Hopkins, The Highlands of Canaan (Sheffield: Almond Press, 1985), pp. 179–184; cf. Finkelstein, Archaeology of the Israelite Settlement, p. 309.

108 Callaway in Aviram, Biblical Archaeology Today, p. 75.

109 Ibid., p. 73.

110 Finkelstein, Archaeology of the Israelite Settlement, p. 311.

111 N. P. Lemche, Early Israel (= VT Supp. 37) (Leiden: E. J. Brill, 1985), pp. 411–435; Lemche also puts forward his theory in a more popular form in his Ancient Israel: A New History of Israelite Society (Sheffield: JSOT Press, 1988), pp. 85–90, 100–102.

112 Early Israel, pp. 415–416.

113 Ibid., p. 416.

114 Ibid., p. 427.

115 Ibid., p. 428.

116 Ibid., p. 428–429.

117 K. W. Whitelam, ‘Recreating the History of Israel’, JSOT 35 (1986), p. 60.

118 Coote and Whitelam, The Emergence of Early Israel, pp. 117–138.

119 Ibid., p. 136.

120 Lemche, Early Israel, p. 393.

121 Coote and Whitelam, The Emergence of Early Israel, p. 117.

122 Finkelstein, Archaeology of the Israelite Settlement, p. 321.

123 B. G. Wood, Palestinian Pottery of the Late Bronze Age (above, n. 37), pp. 549–599. This date relies on the low chronology for Egypt which has been employed throughout this article (see n. 37). A higher chronology would raise this date, but would also raise the date of Merenptah’s reference to Israel, so their relative dating would remain the same. See also D. Ussishkin, ‘Level VII and Vl at Tel Lachish and the End of the Late Bronze Age in Canaan’, in the Tufnell Festschrift (see n. 105), pp. 225–226, for 1150 bc as the suggested date for the start of Iron I.

124 Coote and Whitelam, The Emergence of Early Israel, p. 179, n. 3.

125 The claim of Ahlström (Who Were the Israelites?, p. 40), that Merenptah’s stela could refer to a territorycalled Israel, flies in the face of the inscription itself, which defines Israel by means of the determinative for a foreign people.

126 See works cited in n. 33.

127 E.g. Craigie, ‘The Conquest and Early Hebrew Poetry’ (see n. 50); D. N. Freedman, ‘Early Israelite Poetry and Historical Reconstructions’, in Cross, Symposia, pp. 89–96; B. Halpern, The Emergence of Israel in Canaan (Chico: Scholars Press, 1983), pp. 32–43, 117–133, 146–163.

128 Bimson, Redating.

129 Ibid., pp. 106–223.

130 Summarized most recently in Bimson, ‘Exodus and Conquest: Myth or Reality’ (see n. 29), pp. 34–36.

131 Cf. Jos. 3:14; 7:21–24; 22:4–8; etc.; on Israel’s mode of life in the period of the judges, see Halpern, The Emergence of Israel, pp. 210–211.

John J. Bimson

Trinity College, Bristol