Abstract
Archaeological evidence of certain cultural traits supports the witness of some biblical passages that the (northern) kingdom of Israel was ethnically diverse, with non-Israelite populations in the northern lowlands. Texts in Judges 1 and 1 Kings 9 stating that ‘Canaanites’, etc., were subjected to forced labour, the corvée, either by Israelite tribes or by Solomon, should be regarded as texts of cultural memory, justifying the actual practice of the kingdom of Israel in the ninth and eighth centuries BCE in restricting the corvée to non-Israelites. This article proposes that these texts should be placed alongside the story of the rebellion against the house of David in 1 Kings 12, which, although in its present form is written from the point of view of Judah, carries clear traces of its origin as cultural memory in Israel. Taken together, the texts suggest that the kingdom avoided imposing the corvée on ethnic Israelites.
Introduction
States in the ancient Near East used levies of forced labour from their subjects, the corvée, to work on all their large-scale public works. There are a number of references to this in the Hebrew Bible, and there is a technical vocabulary for it. The present study discusses such references in the account of Solomon's reign and the division of the kingdom, and in Judges 1. Apart from these, and apart also from the obscure reference to Issachar's corvée service in the blessing of Jacob (Gen. 49.15), the context of the references is the kingdom of Judah. They are: 1 Kgs 15.22, where king Asa ‘called together all Judah, with no exemptions—יקנןיא’ to use the stones from Baasha's abandoned fortress of Ramah to build defences at Geba and Mizpah; Micah's accusation (Mic. 3.10) against the rulers that they had built or were ‘building Zion with blood and Jerusalem with wickedness’; and Jeremiah's condemnation of Jehoiakim (Jer. 22.13) for ‘building his house without right and his upper chambers with injustice, making his neighbour work for nothing and not giving him his wages’. The prophetic accusations should probably both be taken as pejorative references to the corvée rather than to the abusive use of wage labour. 1
For Micah, see Kessler 1999: 163–65, against e.g. McKane 1998: 112. For Jeremiah, Bright (1965: 145) speaks of ‘conscripted labour’; but Jones (1992: 289) implies it is a matter of failure to pay wages due, and Carroll (1986: 427) denies vv. 13–15a were originally about a king at all, necessarily therefore not a reference to the corvée. But in the light of the Micah text, from an oracle which was known to the editors of Jeremiah (Jer. 26.18-19), the verse should be seen as criticizing the use of the corvée for an essentially private rather than public purpose.
However, as far as I am aware there is no direct reference to the use of the corvée in the kingdom of Israel from Jeroboam I on, as distinct from the largely or entirely legendary account of Solomon. There is of course less material on Israel than on Judah in the Hebrew Bible, but it does include a very hostile view of the Omride dynasty in Kings. None of the numerous denunciations of injustice in Amos—from a later period— appear to have the corvée in view. Yet it is certain that the Omrides at least must have used it. They were responsible for extensive building works all over the kingdom. There are the palaces and defence works of Samaria and Jezreel and a number of fortresses on the borders; and the recent lowering of the radiocarbon dates for Iron IIA—which I do not intend to discuss here—brings even more of the archaeological record into their period (see Finkelstein 2013: 85–105). Writers who are interested in the subject have normally assumed that Omri and his successors must have conscripted the labour of their Israelite peasant subjects to erect these works. Thus, for example, Chaney (1986: 71–72) says that Omri ‘follow[ed] in the footsteps of Solomon’, and refers to ‘increased corvée’ exacted from the highland peasantry as a result of his wars.
The tradition which we find in 1 Kings 12 suggests, on the contrary, that Israelite peasants did not expect to have to provide corvée labour. But is there any historical value to this tradition? Where, if that is so, did the kings of Israel find such labour? Na'aman points to Mesha of Moab's reference in his inscription to the use of prisoners of war for public works, and suggests that Omri's extensive conquests would have gained him large numbers of prisoners of war for use in the building of Jezreel (Na'aman 1997: 122–24). This, however, would have been a wasting asset. A more enduring source of labour would have been the non-Israelite portion of the Israelite kings' subjects, those who had been subject to Canaanite city states particularly in the Jezreel and upper Jordan valleys, and had later become subject to Israelite rulers. In an earlier publication (Houston 2008: 39), I suggested briefly that the kingdom may have drawn on this source in order to avoid using the labour of their Israelite subjects. It is the object of this article to demonstrate the probability of this hypothesis in detail. It will explore the Israelite tradition on the subject, understood as cultural memory, in relation to the historical reality of ethnic diversity in the kingdom of Israel, as demonstrated archaeologically.
It will be convenient to begin by showing that there were distinct ethnicities in ninth- to eighth-century Israel, and that it was likely that the state discriminated in favour of one and against others.
Finkelstein (2013: 110), noting the ‘settlement and cultural continuity’ in the northern valleys between Late Bronze and Iron I, which was maintained in the rural areas even after the destruction of the Iron I cities, infers that there was a population in the valleys ethnically distinct from the highland areas where the kingdom was based.
Faust aims to demonstrate such a distinction by highlighting contrasting cultural traits in the two respective areas that appear in the archaeological record of the Iron II period (Faust 2012: 230-54, esp. 241-47; cf. earlier Faust 2000). He refers to three excavated sites: one in the foothills of Carmel, one in the Beth-Shean Valley, and one on the eastern shore of the Sea of Galilee. It could be argued that, because none of them are properly speaking in the Jezreel Valley, they are hardly characteristic of the area. But what is important is that none of them are in the highlands.
Faust lays most stress on the plan of the dwelling house, arguing that ‘almost all the houses known in Israelite cities and villages’ were of the so-called four-room house type, including as a sub-type the ‘three-room house’ (Faust 2012: 215), without denying that this type is also exemplified sporadically elsewhere (2012: 218-19). His use of the word ‘Israelite’ here is ethnic not political, and it assumes what needs to be proved. I take it that by ‘Israelite’ he means ‘highland’. But in the three valley sites referred to the houses are not of this type. A study of the plans of the houses in these villages (Faust 2012: 236, 238) shows that they are clearly different from those characteristic of villages in the highlands (Faust 2012: 214). Some have only one room; others have two or more rooms en suite, which is a feature excluded by the so-called four-room plan. It is not relevant for our purposes that the four-room plan may be used elsewhere or in Iron I, especially in the Transjordan (see, e.g., Routledge 2000), for we are only concerned with the contrast between highland and lowland within the territory of the kingdom of Israel in Iron Age II.
Faust draws attention to numerous other features (2012: 241-47), including the lack of walls round these rural settlements of the valleys, which are regular in highland settlements; the presence of cultic buildings or rooms at two of the sites, invariably absent in Iron Age highland villages (Faust 2010); the smaller size of the houses, suggesting nuclear rather than extended family occupation; the presence of pig bones at one site; and the presence of imported pottery, also absent in the highlands, and not only at rural sites. Even though we are able to identify only such cultural elements as are capable of leaving archaeological traces, these are sufficient to delineate a culture different from that of the highlands.
But are we constrained to define the difference as ethnic? (See the discussions in McGuire 1982; Routledge 2000: 63–65; Faust 2012: 230-33.) This is not a matter that can be decided by a checklist of distinctions. Distinctions such as the ones just listed may characterize groups defined in other ways, for example by class, caste, or occupation. Ethnic distinction is subjective, something defined by the groups themselves, who perceive themselves as belonging to different peoples or nations, and take certain cultural markers as recognized signs of the difference. But as Routledge points out (2000: 64), the observer may note other differences, resulting from differences in behaviour, which are not understood by the actors themselves as ethnic markers. It is only if we have the words of the actors that we can be sure that a particular cultural boundary is regarded as ethnic, and for an ancient people that means texts (McGuire 1982: 163; Faust 2012: 233); in this case biblical texts, as no relevant epigraphic evidence survives from the kingdom of Israel. Of the above list, the only one referred to in the Bible is abstinence from pork (Lev. 11.7: ‘the pig…shall be unclean to you’), and that text, certainly in its present form, comes from a later period. 2 Other texts, however, confirm that the geographic boundary between highland and lowland was seen to be correlated with an ethnic one, notably those in Joshua 16 and 17 and Judges 1, which are discussed below. Ethnic distinction is often, though by no means always, correlated with political boundaries; in this case it would presumably be related to a former political boundary, now erased by conquest.
Some of the prophetic denunciations of ‘idolatrous’ cults may be related to the presence or absence of shrines in residential areas: see perhaps especially Jer. 7.16-19. But they are not connected in such passages with ethnic distinctions—that the prophets were attacking ‘Canaanite’ cults is a scholarly conceit extrapolating from Deuteronomic texts (e.g. Deut. 7.1-6; 12.29-31). The reference to walled and unwalled settlements in Lev. 25.29-31, brought forward by Faust at 2012: 241, is likewise not concerned with ethnic distinctions.
It is no objection to the identification of ethnicity as a factor in the diversity of populations in Israel that ethnic Israelites may have been of local origin, contrary to their myths of origin. Those myths themselves define Israel as of foreign origin; and there are other examples where people of similar origins have begun to define themselves as belonging to separate peoples (e.g. Northumbrian Angles living on either side of the border between England and Scotland established in the eleventh century CE).
But does it follow from this that the people of the lowlands were discriminated against? An older scholarly position, identified for example with Alt, held that the establishment of the monarchy led to the rapid integration and indeed promotion of the ‘Canaanite’ element in the population. It was not absorbed into the Israelite tribes, but into the territories of David's administration, with the same rights and duties as Israelites (Alt 1953: II, 52). Indeed, with Solomon's establishment of a chariot force, the Schwergewicht in the kingdom began to shift towards the Canaanites, and this ultimately led to the restoration of the old cultural and political dominance of the plains over the hills (Alt 1953: II, 53).
But Alt does not clearly identify the signs of this dominance, and there is much evidence against it. The supposedly Solomonic chariot force first appears in history at the battle of Qarqar in the reign of Ahab. But it is just at this period, in the mid-ninth century, that contemporary evidence begins to appear that the god of Israel is YHWH: witness, for example, the reference to ןרמשׁ הוהי at Kuntillet ‘Ajrud, or the Yahwistic names that begin to appear in the Israelite dynasties. Ahab's three known children all have Yahwistic names. Now, there is an interesting passage in 1 Kgs 20.23, 28, where the Aramaeans before the battle of Aphek are represented as arguing that Israel's god is a god of the hills, and therefore could be defeated in a battle on level ground. This surely reflects a consciousness in the author that the heartland of YHWH's worshippers was the highlands. In what is most important in culture, the advantage remains with the hills.
The more recent archaeological evidence surveyed by Faust (2012: 258-68) suggests a more complex reality. The kingdom deepened its power and extended its reach in the country in large part by building and developing cities (cf. Fritz 1995: 13–14; Houston 2008: 38), and these were mainly in the lowlands. Such places as Megiddo and Hazor were old Bronze Age cities rebuilt and made regional centres of influence. Among these, Hazor's residential quarters are sufficiently exposed to make it clear that it was a multi-ethnic settlement, in which the elite, those with the largest and best-built houses, were highlanders, to judge from their house plans. Thus it would seem that Israelites had moved into the conquered areas to take control of them on behalf of the state. It is in this sense that the cultural and political dominance of the plains was restored, being exercised by people of highland origin living in the plains. Certainly, some of the old lowland elite could have been accepted into official and other leading positions, but in doing so they would have adopted aspects of Israelite culture.
Alongside this, we find in the non-Israelite villages of the same areas public buildings (Faust 2012: 242-43), a feature normally absent in highland villages, apart from grain silos and the like, and Faust suggests that ‘the villages…served as outposts of the city or palace, and the public buildings belonged to it’ (2012: 242). Villages forming parts of crown estates are common enough elsewhere in the Near East, as Faust points out. Alternatively, they could be seen as being under a type of feudal or prebendal control. This is in marked contrast to the highland villages, which show no signs of elite dominance, and do show signs that they retained surpluses.
Enough has been said to suggest that the lowland villages, as distinct from the cities, and as distinct also from the ethnically different highland villages, show clear signs of a subordinate position in the state of Israel. It does not require arguing that these ethnically subordinate villagers were subject to corvée demands; what I need to show now is that the cultural memory of Israel, as embodied in the biblical texts on Solomon and the succession to Solomon, serves to justify, and therefore for us to demonstrate, Israelites' exemption from it.
I am here using ‘cultural memory’ in the sense developed by Jan Assmann (2011), of a commonly accepted narrative of the past that determines and confirms customs and attitudes of the present.
Solomon in Kings
The principal narrative concerned is 1 Kgs 12.1-20, the account of Israel's rejection of the rule of Rehoboam, to be discussed below. But texts in the account of Solomon also require examination. 1 Kings 5.27 (EVV 5.13) states that Solomon ‘raised a levy (סמ) from all Israel of 30,000 men’ to work on supplies for his building projects, and this must be the presupposition for the later account. But a much interpolated passage in 1 Kings 9 contradicts this. ‘This is the account of the levy that king Solomon raised…’ begins v. 15, and after a meandering account of his building projects and Pharaoh's conquest of Gezer, the passage returns to the matter of the levy in w. 20–21: ‘All the people that were left of the Amorites, the Hittites, the Perizzites, the Hivites and the Jebusites, who were not of the Israelites, their descendants who were left after them, whom the Israelites were not able to put to the ḥerem, Solomon conscripted as forced labourers, as remains the case today (הזה םויה דע דבע־סמל המלשׁ םלעיו)’. (As Rainey has shown [1970: 200], the term סמ refers to the personnel, ‘forced labourers’, ‘levy’, not to the custom or institution of the corvée.)
Many commentators regard the statement in ch. 9 as a late apologetic attempt to clear Solomon's name of the odium of being responsible for the practice of imposing forced labour on the people of Israel; the language and ideology are clearly Deuteronomistic (e.g. Montgomery 1951: 137, 205, 209; Cogan 2000: 309). 1 Kings 5.27, on the other hand, has generally been regarded as a more reliable statement, particularly because it is apparently presupposed by the narrative of the rejection of Rehoboam in ch. 12. But it is not always observed that to regard the corvée as odious is in contrast with its normality among ancient Near Eastern kingdoms, including Judah. Even if 9.20-21 is late, it must therefore reflect a view traditional in Israel, as Würthwein sees (1977: 113).
There has been an attempt, now it seems abandoned, to make a distinction between סמ simple, as at 5.27, and דבע־סמ as at 9.21, thus removing the contradiction. This goes back to Bachmann in 1868 (141–42), who defined סמ as the responsibility of free citizens to supply labour from time to time, while דבע־סמ defined a permanent state of slavery resulting from conquest and the loss of independence. All Israel had to perform corvée when required, but the Canaanites were subjected to a permanent servile state. Mendelsohn (1942: esp. 17) championed this distinction, arguing for an institution of state slavery, and in one form or another such a distinction is accepted by Noth (1968: 217) and Gray in his second edition (1970: 155). But in view of the variation in the vocabulary the distinction cannot be sustained. The דבע in 1 Kgs 9.21 is omitted in 2 Chron. 8.8; and while it is stated in 1 Kings 9 that the non-Israelites were levied as דבע־סמ, the statements about the same peoples in Judges 1, which we shall look at shortly, use סמ alone, while a late addition (not in the Old Greek) in Josh. 16.10 uses דבע־סמ.
The only leading commentator to take a view diverging from the consensus in favour of 5.27 is Noth (1968: 216-18). He argues that in 9.20-21 it is possible to distinguish a Dtr redactional stratum from an underlying text (Grundschrift), which read ‘All people who did not belong to the Israelites Solomon conscripted for servile forced labour, and this is still the case today’. Noth views the underlying text throughout 9.15-23 as having an official character, shared by this reconstructed text of vv. 20–21, whereas 5.27 is part of a Dtr or post-Dtr passage. Thus Noth reverses the usual view, and asserts that there are no adequate grounds to question the historicity of 9.15-23 (1968: 218).
In view of the dubious historical status today of the whole narrative of the United Monarchy, it is meaningless to dispute the historicity of statements about Solomon's rule. But we may ask about the origin and function of such statements in their contemporary context. And that Noth could be right in seeing the original context of 9.20-21* as much earlier than the Deuteronomists is suggested by two points: the clause ‘who did not belong to the Israelites’ is redundant in its present context and not found alongside any other uses of the list of the former nations, which in whatever selection or order is a Dtr cliché; and הזה םויה דע is standard in aetiological statements and implies that this state of affairs is current. This points to an origin in the kingdom of Israel, even if it does not conclusively prove it. A writer aware of current practice in restricting the corvée to non-Israelites traces it back to Solomon.
Judges 1 and Parallels
It coheres with this that in Judges 1, gathering a number of traditions about the tribes, some of which are found in a probably more original form in the tribal allocations in Joshua, pre-Israelite populations which the tribes could not conquer are stated to have been conscripted as סמ. The parallel instance in Josh. 17.11-13 ‖ Judg. 1.27-28 is the most illuminating of these. The tribe of Manasseh is said to have failed to dispossess a list of cities in the northern lowlands, including Beth-Shean, Megiddo, Taanach, Ibleam and Dor, each with ‘its daughters’, that is, its dependent villages; ‘But when the Israelites grew strong, they put them to forced labour (סמל), but they never dispossessed them (ושׁירוה אל שׁרוהו)’ (Josh. 17.13; Judg. 1.28 has לארשׂי in place of לארשׂי ינב). The parallel observations in Josh. 16.10 and Judg. 1.30, 33, 35 all say that the tribes in question themselves put the Canaanites to forced labour (unless ‘the hand of Joseph’ in Judg. 1.35 really means the power of the kingdom of Israel). But this cannot have been the case. Corvée can only be imposed by a state, not by a mere tribe. ‘When (the children of) Israel grew strong’ in effect refers to the foundation of the state. These texts reflect a current or at least recent reality of forced labour in the kingdom being done by non-Israelites, and refer it to tribal action, unrealistically, but in line with Israel's cultural memory of being a tribal people. Members of free Israelite tribes do not do forced labour—they have it done for them. (Compare 1 Sam. 8.11-18, and Crüsemann 1978.)
In the context of the tribal allocations in Joshua, these texts explain in the light of the cultural memory of conquest how it is that much of lowland Israel was not in fact occupied by Israelites. In the Judges context, they serve to indict the tribes other than Judah, from a Dtr point of view, of failure in their God-imposed responsibility of driving out the previous population: ‘could not dispossess’ in Josh. 17.12 is replaced by ‘did not dispossess’ in Judg. 1.28 (O'Connell 1996: 50–72). But neither context requires the reference to the corvée, which I suggest reflects another and earlier context.
1 Kings 12.1-20
In contrast, the narrative of the succession to Solomon in 1 Kgs 12.1-20, as we have seen, implies what is stated by 5.27, that Solomon imposed corvée on ‘all Israel’. The word סמ is not used, but the words in v. 4, ‘your father laid a harsh yoke on us, but if you now lighten the hard service your father imposed (השׁקה דיבא תדבעמ) and the heavy yoke which he laid on us, we shall serve you’, make it virtually certain, as commentators have generally agreed, that the reference is to corvée service. The logic of the narrative implies that after rejecting Rehoboam as king because he refused their modest request for a lightening of the load, Israel would not have accepted anyone else as king who did not agree to at least a lightening of the service, and no doubt preferably a complete remission, even though this is not stated explicitly in connection with their election of Jeroboam (v. 20).
I have stated the hypothesis that this narrative is cultural memory. But the question arises: Whose cultural memory? Is it indeed the memory of Israel, explaining the fact that unlike the non-Israelite population they are exempt from the royal labour service, or is it not rather the memory of Judah or Jerusalem, explaining why the house of David does not rule Israel? Most commentators in fact see the account as written from the point of view of Jerusalem. It begins by assuming that the approval of Rehoboam's kingship by the people is a formality (v. 1; Becker 2000: 216), and ends by asserting ‘so Israel is in rebellion against the house of David down to this day’ (v. 19). In v. 16 the cry appears ‘What part have we in David? And we have no share in the son of Jesse. To your tents, Israel!’, which had almost word for word been uttered previously by that לעילב שׁיא Sheba the son of Bichri (2 Sam. 20.1) to launch his rebellion. Uwe Becker argues, following Wellhausen, that this is a literary reprise showing that the account here is written from the same Jerusalem-based point of view as the account of Sheba's rebellion (Becker 2000: 216-17; Wellhausen 1889: 279). Noth, however, diverging from the consensus as before, does not agree that any literary dependence need be assumed between the two passages, regarding the saying as one emerging among the ‘north Israelites’ and then circulating and becoming well known (Noth 1968: 276).
Becker (2000) performs a redaction-critical dissection of the passage, arguing that the verses dealing with Jeroboam, his rebellion, flight, return and call to the throne—namely, 11.26,40; 12.2*, 20a, 25—are an account readable on its own, and that 12.1, 3b-19, the account of Rehoboam's failure, is a subsequent insertion; but it is not from a pre-Dtr source, for it presupposes the notice of Rehoboam's assumption of power in Jerusalem in 11.43. It is dependent on 5.27-32, of course, the passage that asserts Solomon levied corvée on all Israel, but also, according to Becker, on 9.15-23 that contradicts it, since that passage takes the same ethical position on the corvée. Further, according to Becker, the mise-en-scène of the account, the gathering of ‘all Israel’ to Shechem to confirm Rehoboam's kingship, is modelled on the two elections of David as king in Hebron, by Judah (2 Sam. 2.4) and Israel (2 Sam. 5.1-3), and these also are the work of a late Dtr or post-Dtr redactor, based on the chronological notices in the first edition of DtrH. It takes place at Shechem because there the disastrous story of Abimelech's kingship played itself out. The account is an ‘Ätiologie der Zweistaatlichkeit’. No specific north-Israelite character (Prägung) is detectable (Becker 2000: 216).
Becker's account suffers from defects inherent in Redaktionsgeschichte as a method. It is under-theorized, that is, not supported by any idea of the social environment of the redaction process (contrast the work of David Carr: 2005 and 2011); and it relies on a series of only moderately probable deductions, which cumulatively result in a much less probable conclusion (Carr 2011: 139-40). In particular, Carr notes how insecure readability is as a criterion for the original existence of a reconstructed strand of narrative (such as Becker's Jeroboam text; Carr 2011: 113-14), and that it is precarious to argue for the dependence of one text on another on the basis of isolated characteristics (2011: 138-44). In the present instance, there is no verbal resemblance at all between 1 Kings 12 and the various texts that Becker says it must depend on, except for v. 16. What we can say is that both 1 Kgs 9.20-21* and 1 Kgs 12.1-20* reflect a tradition that corvée service is unacceptable for Israelites. A literary connection is not required. And it is also clearly the case that 1 Kgs 5.27 and 1 Kgs 12.1-20 are parts of the same story: but that should not lead to the conclusion that one depends on the other, only that they belong together.
Daniel Fleming, whose book is aimed at teasing out whatever in the Hebrew Bible originates in text or substance in the kingdom of Israel, agrees with Becker that the account as it stands is written from Judah's point of view, but disagrees that the account has no specifically (north-) Israelite character. It focuses on ‘aspects of a collective or collaborative leadership that would align with the pervasive tradition observed as central to Israel in various biblical writings’ (2012: 111; Fleming refers particularly to the account of the rise of Omri in 1 Kgs 16). Rehoboam negotiates not with Jeroboam as a leader, but with Israel as a collective. This is clear from the Old Greek text of 1 Kgs 12.3, 12, which has no reference to Jeroboam, and should be regarded as original. It should be noted, however, that something similar is said of political practice in Judah in those three places in 2 Kings (14.21; 21.24; 23.30) where ‘all the people of Judah’, or ‘the people of the land’, a collective, however it may be defined, identify and place on the throne the successor to an assassinated king. But I am not convinced by Becker's alternative of a model found in the accounts of the election of David. Here there is no popular assembly: elders and members of the Israelite tribes travel to Hebron to anoint David, rather than requiring him to come to them. And to refer to Judges 9, which concerns Shechem as a city and not Israel, seems perverse in view of the frequent appearance of Shechem as a place of assembly for Israel.
But this is not the most decisive consideration. The decisive point is that while the story is told from the point of view of Judah, or at least Jerusalem, the story itself, that is, the basic plot, without the episode of Rehoboam's seeking counsel, is one which seems far more likely to have originated in Israel. Supposing the story were developed or composed in Jerusalem, that would have been at the expense of the reputation not merely of the inept Rehoboam but of Solomon, a king who is celebrated in DtrH; even in the section devoted to his sin, the sin in question is the typical Dtr theme of religious apostasy rather than the cruel exploitation of his compatriots. It seems unlikely that, after the memory of Solomon's greatness had been established, a later Jerusalem-based contributor would have arbitrarily blackened it. But if the story already existed, it could have been appropriated in an ambiguous way to condemn the Israelites for seizing such a pretext to rebel and to celebrate the memory of Solomon's wise advisers who could have averted the disaster if his foolish son had heeded them (Cogan 2000: 351, following Liver 1967: 96–99).
But the people of Israel had nothing invested in the reputation of the house of David. We have already argued that the odiousness of the corvée reflected in this passage is likely to have been an Israelite tradition, and the odium is here inextricably associated with that house.
There is an important further question to answer. There are two different ways in which cultural memory may be applied in Assmann's theory: as foundational memories, which purport to relate key events of the past that explain, support and justify the present institutions and customs of society, and contra-present memories, which represent past events as enshrining an ideal which is a challenge to the present situation and may justify revolt against it—Assmann offers the example of the story of David treated as the foundation of Messianic belief in the late Second Temple period (Assmann 2011: 62–69). Which of these, for Israelites towards the end of their monarchy, is the story of Israel's revolt against Rehoboam? Is it, as I have suggested, a foundational memory, on which is founded the exemption of Israelites from the corvée; or is it a contra-present memory, encapsulating a grievance against a continuing trespass on the rights and dignity of free Israelite tribesmen?
The evidence marshalled in the earlier part of this article should, I think, incline us to the former alternative. Israelite kings, unlike Judaean ones, had easy access to an alternative source of labour, one which was probably inured to centuries of exploitation in this manner by their own kings (Chaney 1986: 60–61); and I consider that 1 Kgs 9.20-23 (with Noth), along with texts in Judges 1, should be taken as serious evidence that they were treated in a discriminatory manner.
An additional or alternative possibility is that the tradition worked as a contra-present memory for Israelites resident in Judah after 722 BCE and observing, or even being subject to, corvée there. It could be that it was in that situation that the memory of the corvée became associated with the house of David and the story of Rehoboam's rejection arose, or at least was elaborated.
But as a foundational memory the story would have resonated with the dominant foundational memory of the kingdom of Israel, the story of the deliverance of Israel from forced labour in Egypt. This is generally agreed to belong to Israel originally rather than Judah, and van der Toorn has described it as Israel's ‘charter myth’ (van der Toorn 2001; cf. Albertz 1994: I, 141-42).
The sequel bound up with that memory, the conquest of Canaan, is now often seen as ethically and theologically problematic (see, e.g., Prior 1997; Warrior 2006). The answer that the genocide of the Canaanites was not a historical reality, unsatisfactory in any case, must now face what I have here argued was a historical reality: the subjection of non-Israelites to discriminatory treatment, including forced labour on the kingdom's projects.