Abstract
This article traces the development of the concept of the Hexateuch in five major stages: (1) its beginnings in the late eighteenth and nineteenth centuries, (2) its floruit in the late nineteenth and early twentieth centuries, (3) the challenge to the Hexateuch hypothesis by the Deuteronomistic History hypothesis, (4) the partial decline of the Hexateuch hypothesis during the second half of the twentieth century, and (5) its recent revival and reinterpretation, particularly since the turn of the millennium. Within the current discussion, the most decisive question is whether one should conceive of the Hexateuch as an early (i.e., pre-Priestly and/or pre-Dtr) narrative work, as a late redactional construct, or both.
Keywords
Early Propositions for Regarding Genesis–Joshua as a Literary Unit, 1792–1876
In 1792, the Scottish theologian Alexander Geddes argued that the Pentateuch and the book of Joshua were ‘compiled by the same author’ and that the book of Joshua ‘is a necessary appendix to the history contained in the former books’ (1792: 1.xxi). Geddes’s observations mark the starting point for a shift in critical scholarship that was to become widespread beginning around a century later: the analysis of the books of Genesis through Joshua as a coherent unit, or a ‘Hexateuch’. Although the term ‘Hexateuch’ was first used by the Church Fathers to designate the books of Genesis through Joshua as a literary unit (Auld 1980: 3 n. 8), it would not be employed again until the late nineteenth century (see below). Nevertheless, as Geddes’s statement indicates, scholars of the Hebrew Bible began to include the book of Joshua in their discussions of the formation of the Pentateuch long before the term ‘Hexateuch’ gained currency.
Discussion of the connections between the Pentateuch and book of Joshua continued during the early nineteenth century (see already Bleek 1822: 44). In 1831, Ewald concluded that an older literary work spanning from the creation of the world up to the conquest of the land underlies the Pentateuch (1831: 602). Four years later, Stähelin (1835: 472) advocated a similar view. Other early nineteenth-century works that adopted a Hexateuchal perspective include the fifth edition of de Wette’s Lehrbuch der historisch-kritischen Einleitung in die kanonischen und apokryphischen Bücher des Alten Testaments (1840: 221) and Ewald’s Geschichte des Volkes Israel bis Christus (1843–52: 1.75-164; 2.225-70).
During the third quarter of the nineteenth century, both Bleek (1860: 312) and Knobel (1861: 357-488, 547-59) reckoned with narrative sources that spanned from Genesis to Joshua. Similarly, Kuenen argued that the simplest solution to explaining the similarities between the Pentateuch and book of Joshua is to posit that the same documents can be found in both works and that the redactor of the book of Joshua is ‘none other than that of the Pentateuch’ (1861: 183). Shortly after the publication of Bleek’s and Knobel’s studies in Germany and the first edition of Kuenen’s Historisch-kritisch onderzoek in the Netherlands, Colenso adopted a similar approach in his multivolume work The Pentateuch and Book of Joshua Critically Examined, explaining the formation of the book of Joshua in terms of Pentateuchal documents: the ‘Original Story’ (= JE), the Deuteronomist, and ‘Later Legislation’ (= P) (1862–70: 6.112-29).
Wellhausen and the Flourishing of Hexateuchal Criticism, 1876–1938
The term ‘Hexateuch’ appears to have first been used in a series of articles by Wellhausen entitled ‘Die Composition des Hexateuchs’ (1876–77) (for this observation see also Rake 2006: 8). Here, Wellhausen followed earlier scholars in suggesting that in the book of Joshua ‘some of the sources, particularly E in its Jehovistic reworking, are likely continued here’ (1876: 585). In his analysis of the composition of the book of Joshua, Wellhausen divided the material primarily between JE and Q (= P), although he also attributed certain passages to the Deuteronomist, including all of Joshua 1 (1876: 585-602). Surprisingly, given his novel use of the term ‘Hexateuch’, Wellhausen provides no explicit justification for the shift from ‘Pentateuch’ to ‘Hexateuch’. In fact, apart from a brief discussion of his use of sigla at the beginning of the article, the remainder of the study consists entirely of close textual analysis, without any discussion of broader theoretical issues. Nevertheless, Wellhausen’s use of the term ‘Hexateuch’ was quickly adopted by other scholars, including in the second edition of Kuenen’s Historisch-kritisch onderzoek naar het ontstaan en de verzameling van de boeken des Ouden Verbonds (1885) and its translations into English and German. Wellhausen himself also used this term in two later studies that elaborated on his article from 1876–77 (1885; 1889). The latter of these, Die Composition des Hexateuchs und der historischen Bücher des Alten Testaments, set the standard for treating the Hexateuch as a literary work—and for subjecting the book of Joshua to classical source-critical analysis—well into the twentieth century (cf. Dillmann 1886; Albers 1891; Addis 1892–98; Carpenter 1902; Steuernagel 1900; Holzinger 1901; Cooke 1918; Eissfeldt 1922; Bertholet 1927–31; and Noth 1930).
This is not to say that those who took the Hexateuch as a legitimate unit of analysis necessarily shared all of Wellhausen’s assumptions. Indeed, during the early twentieth century, one of the consequences of searching for Pentateuchal sources in Joshua was the impulse to fit (or force) texts in the book of Joshua into the ever-more complex source divisions worked out primarily on the basis of Pentateuchal texts. Thus, Smend (1912) divided J into two independent narrative threads (J1 and J2), and this resulted in an attempt to locate each of these threads in the book of Joshua. Similarly, von Rad’s division of the Priestly narrative into two sources (PA and PB) led him to differentiate between these two sources in the book of Joshua as well (1934). Perhaps the most extreme example of such source-critical analysis of the Hexateuch is Eissfeldt’s Hexateuch-Synopse, in which Eissfeldt argued that the Hexateuch (or, more precisely, the narrative portions of Gen. 1.1–Judg. 2.9) is composed of four internally coherent and originally independent sources: J, E, P, and L (for ‘Laienquelle’ or ‘lay source’) (1922: xii).
In sharp contrast to Smend and Eissfeldt, Mowinckel and Rudolph argued that once P is removed from the Hexateuchal narratives, only one continuous source remains: J. Indeed, for both of these scholars, the conquest narratives and related texts are important evidence for this conclusion. In Mowinckel’s view, the passages that other scholars typically ascribed to E are, upon closer analysis, consistently dependent on J and represent a later development based on the J-tradition (Jv for ‘J variatus’) (1930: 270; 1964a: 59-118). Similarly, Rudolph (1938: 255) argued that E is an illusory source and that much of what is often ascribed to E should be attributed to J or to various additions to J. Interestingly, although Mowinckel’s and Rudolph’s ideas on the formation of the Hexateuch were soon overshadowed by Noth’s theory of a Deuteronomistic History, many of their arguments have resurfaced in the renewed discussion of the Hexateuch that has taken place since the late 1990s (see ‘New Theories of the Hexateuch, 1976–2017’, below).
Noth’s Challenge to the Theory of a Hexateuch
The attempt to identify Pentateuchal sources in the book of Joshua that had been practiced for over a century received a (near-)fatal blow with the publication of Noth’s 1938 commentary on the book of Joshua, in which Noth made two basic arguments against the continuation of the classical Pentateuchal sources in the book of Joshua: (1) the P-like material in Josh. 13.1–21.42 had its own literary prehistory that is independent of both the other parts of Joshua and the Pentateuchal narratives; and (2) even in the other parts of the Joshua narrative, the literary evidence differs from that found in Genesis (the classical case study for source-critical analyses) (1938: viii). Notably, however, Noth’s denial of narrative continuity between the book of Joshua and the preceding books is based on only two examples: (1) differences in the representation of the miracle at the sea in Exodus 14 and in Josh. 2.10; 4.23 and (2) divergences between the description of certain events as narrated in the Pentateuch and the review of those events in Josh. 24.2b-13 (p. xiii). Noth echoed this skepticism about the existence of a Hexateuch as an independent literary work in his Überlieferungsgeschichtliche Studien (1943: 253), yet he also found it difficult to abandon the notion that the ‘old sources’ of the Pentateuch originally contained a conquest narrative (1943: 210; 1948: 16, 54-58, 77-79). Nevertheless, Noth denied that such a conquest narrative is preserved in the book of Joshua.
That Noth would base such a major claim as the absence of an overarching, pre-Deuteronomistic narrative thread connecting the conquest with the events that preceded it on only two textual examples is striking. Even more striking is Noth’s increasing reliance on his Deuteronomistic History hypothesis in the second edition of his Joshua commentary from 1953: The question of the appearance of one of the old Pentateuchal ‘sources’ in Joshua must therefore be answered in the negative, namely, on the basis of the literary evidence in Joshua. That this is the case is all the more understandable considering that the book of Joshua belongs in the larger literary context of the Deuteronomistic History, which developed completely independently of the large traditional work of the Pentateuch. (1953: 16, emphasis added)
In other words, by 1953 Noth had concluded that the place of the book of Joshua within the Deuteronomistic History is practically evidence enough that it does not fit within the context of a Hexateuch. Yet Noth’s reliance upon his own Deuteronomistic History hypothesis in challenging the existence of a Hexateuch sets his argument on unstable ground, for if this hypothesis is questioned, then Noth’s denial of a Hexateuch is left without a firm foundation (cf. Frevel 2004: 80-86).
Nevertheless, in the wake of Noth’s Deuteronomistic History hypothesis, it became customary to situate the book of Joshua within the framework of the Deuteronomistic History rather than the Hexateuch. For example, in a discussion of the issue in the Interpreter’s Dictionary of the Bible, Good concluded that ‘it seems justifiable to doubt that the Pentateuchal documents continue into Joshua… The only literary contact on which everyone agrees is with Deuteronomy and the rest of the Deuteronomic History (Judges–Kings)’ (1962: 990). Fourteen years later, Miller came to a similar conclusion in a supplementary volume to the same work: ‘[A]s far as their literary history is concerned, Deuteronomy and Joshua are not to be understood as the conclusion to the Hexateuch but as the beginning of the Deuteronomic History’ (1976: 493).
Noth’s model of the Deuteronomistic History was also adopted in many of the standard commentaries on Joshua during the last three decades of the twentieth century. Thus, Soggin wrote in his 1972 commentary on Joshua that ‘the thesis of the Deuteronomic history work is in the course of becoming “a commonplace of Old Testament scholarship”’ (1972: 6-7). Despite Soggin’s adoption of this ‘commonplace’ view, it is to his credit that he also provides a thorough list of works that challenge Noth’s theory (p. 7 n. 3). Ten years later, in Boling and Wright’s Joshua commentary, such thorough reference to opposing views is strikingly absent. In the introduction to the commentary, Wright simply presumed the existence of a Deuteronomistic History. In a discussion of the literary criticism of the book of Joshua, Wright concluded that the ‘attempt to analyze the basic epic narrative into J and E in Joshua cannot be said to be successful’ (Boling and Wright 1982: 57). Wright adopted Noth’s view that the pre-Deuteronomic materials in Joshua 13–21 cannot be related to any of the Pentateuchal sources, while for Joshua 2–11 he concluded that ‘the question as to whether they are a continuation of Pentateuchal JE can only be raised; it cannot be answered. Indeed, the mixture of elements crucial to the analysis, where they exist, is so different as to lead one to be skeptical’ (p. 66). The influence of Noth’s model of the composition of the conquest narratives in Joshua is also reflected in Fritz’s 1994 Joshua commentary, which succeeded that of Noth in the series Handbuch zum Alten Testament. In Fritz’s view, apart from a few narrative threads in Joshua 1–12 that seem to reflect older oral traditions (i.e., parts of Joshua 2; 8; and 10), the remaining narratives within these chapters can be attributed to the author of the base layer (Grundschicht) of the book of Joshua, namely, the Deuteronomistic Historian (1994: 5-7). For Fritz, ‘This thesis has become widely accepted in scholarship; it does not need to be substantiated again here’ (p. 7).
Source-critical Models for a Hexateuch in the Era of the Deuteronomistic History, 1948–86
In 1950, von Rad remarked that advances in Hexateuchal criticism had become confirmed to the extent that ‘it had become something academically uninteresting; a certain weariness of scholarship was unmistakable in the area of Hexateuchal criticism’ (1949–50: 52). Amidst this ‘weariness of scholarship’, von Rad regarded the implications of Noth’s hypothesis as ‘cataclysmic’ (‘umwälzend’), particularly insofar as it implied a radical independence of the book of Joshua from the sources JE and P (p. 53). Despite Noth’s ‘cataclysmic’ hypothesis, however, the theory of a Hexateuch did not disappear completely from scholarly discussions. In fact, a steady stream of studies, particularly in German-language scholarship, continued to deal with the notion of a Hexateuch from various perspectives (for this observation see also Otto 2015: 333).
During the 1940s and 1950s, scholars who dealt with the Hexateuch question tended either to disregard Noth’s work or to point to weaknesses in Noth’s argumentation. For example, Simpson (1948) continued the type of source-critical analyses of the Hexateuch produced in the late nineteenth and early twentieth centuries, postulating sources such as J1, J2, E, and P throughout the books of Genesis–Joshua. Likewise, Hölscher (1952: 271-409) followed a long tradition of attempting to identify the classical Pentateuchal sources (especially J and E) throughout Genesis–Kings (cf., e.g., Meyer 1881; Stade 1881; and Budde 1890; for further references see Langlamet 1971: 7-8 nn. 20-46). Von Rad also continued to defend the concept of a Hexateuch, remarking that Noth’s analysis was complicated by the place of P in Joshua (1949–50: 55).
A number of German-language introductions to the Old Testament also continued to adopt a source-critical model for the composition of the book of Joshua. Thus, in the fifth edition to his Einleitung in das Alte Testament, Weiser maintained that the book of Joshua belongs with the Pentateuch and even that the various Pentateuchal sources find their continuation in Joshua (1963: 131). Likewise, Eissfeldt argued that once the Priestly and Deuteronom(ist)ic elements are removed from the books of Genesis through Kings, two (or three) parallel narrative strands can be identified in Genesis–Joshua, each spanning from the creation of humanity (or from Abraham) to the conquest of Canaan (1964: 178-80; see also 1966: 39). Thus, Eissfeldt’s conclusions regarding the existence of a Hexateuch had changed little from those laid out over 40 years earlier in his Hexateuch-Synopse.
An interesting attempt to harmonize Noth’s views on the distinctive nature of the books of Deuteronomy and Joshua with the long-standing tradition of identifying the classical Pentateuchal sources in the book of Joshua is Fohrer and Sellin’s Einleitung in das Alte Testament from 1969. In their view, the original traditions about the ‘Moses cohort’ (‘Moseschar’) formed part of a larger narrative that included an account of the conquest of the land. Implicitly following Noth, however, Fohrer and Sellin argued that the tradition of the Moses cohort’s settlement in Canaan has not been preserved. Rather, the traditions preserved in the book of Joshua reflect an ‘additional conquest narrative’ (‘weitere Landnahmeerzählung’), which has replaced the original settlement tradition that belonged with the narratives of the Moses cohort. This ‘overwriting’ of the conquest traditions associated with Moses by those associated with Joshua occurred in the formation of a narrative that served as the Vorlage to J and E (1969: 136-40). This reconstruction allowed Fohrer and Sellin to reconcile the contradictory views that (1) the original settlement narratives that concluded the Moses cohort-traditions have been lost and (2) the ‘old sources’ of the Pentateuch continue up to Judg. 2.5 (pp. 214-19).
Like Sellin and Fohrer, Mowinckel attempted to defend the existence of Hexateuchal sources while also upholding Noth’s view that the conquest narratives in Joshua are distinct from the preceding Pentateuchal narratives. Thus, on the one hand, Mowinckel seemed to accept the theory of a Deuteronomistic History as a scholarly consensus (1964b: 13). Indeed, because of his assumption that Joshua 1–11 represent Dtr’s version of the conquest, Mowinckel found only sparse traces of J in these chapters (such as in Josh. 11.13) and looked elsewhere for a conquest account that might be attributed to J, such as in Judges 1 (a text which more recent studies have tended to identify as late; cf. Auld 1975; Van Seters 1983: 337-42; Becker 1990: 21-62; and Rake 2006). Mowinckel also followed Noth in attributing a pre-Dtr version of Joshua 2–11 to a ‘collector’ (‘Sammler’) who was active around 900
Like Mowinckel, Langlamet (1971; 1972) found it difficult to deny the Deuteronomistic History hypothesis completely and thus proposed a solution that attempted to take into account both the classical notion of a Hexateuch and the notion of a Deuteronomistic History. Langlamet’s solution was to posit two different versions of the Hexateuch, one of which preceded the formation of the Deuteronomistic History and another which followed it. Thus, in Langlamet’s view, DtrH used materials from an early JE Hexateuchal narrative (including parts of Joshua 2–4) but did not deem it necessary to include the full history of the people, which resulted in the historical summary in Deuteronomy 1–3 (1971: 337-45). Later, however, a Priestly redactor regretted that DtrH had omitted an entire block of Israel’s history and thus reconstituted a (post-Priestly) Hexateuch as a literary work (Langlamet 1971: 17). Interestingly, Langlamet’s distinction between a pre-Deuteronomistic Hexateuch and a Hexateuch that includes Deuteronomistic and Priestly material—which he terms ‘the Hexateuch proper’ (‘l’Hexateuque proprement dit’)—foreshadows the divergence in the use of the term ‘Hexateuch’ that has occurred in more recent scholarship (see ‘New Theories of the Hexateuch, 1976–2017’, below).
In the 1970s and 1980s, a number of scholars continued to identify classical Pentateuchal sources in the book of Joshua, even if they did not address the Hexateuch theory in a systematic fashion. For example, Otto argued that the same Yahwist that is present in Genesis through Numbers can be found in Joshua 2–11 (1975: 95-103), while Lohfink (1978; 1983) and Seebass (1985) argued for a continuation of P in Joshua. Similarly, in a monograph on Joshua 24, Mölle divided that chapter into layers resembling the classical Hexateuchal sources, including an Elohistic layer, a Jehovistic layer, as well as preexilic and exilic Deuteronomistic layers. Indeed, for Mölle, the literary-critical analysis of Joshua 24 confirms ‘the correctness of the Hexateuch theory’ (1980: 282).
By the mid-1980s, however, scholars had become reluctant to assign texts in Joshua to the classical ‘Hexateuchal’ sources with the same confidence as their predecessors. For example, Schwienhorst (1986) analyzed the narrative of the conquest of Jericho in Joshua 6 into a basic layer, a ‘Jehovistic’ expansion, three Deuteronomistic expansions, a Priestly layer, and other post-Priestly expansions. Significantly, although Schwienhorst assigned a particular layer of texts in Joshua 6 to the ‘Jehovist’ and pointed to connections between this layer and other ‘Jehovistic’ texts such as Exod. 3.1-6; 4.24-26a*; and 34.12-26 (pp. 82-84), he avoided the question of whether a coherent Jehovistic ‘source’ can be identified elsewhere in the book of Joshua. For Schwienhorst, both the basic layer and the pre-Deuteronomistic expansions of Joshua 6 were ‘probably part of the conquest narrative of a larger literary work that extends into the Pentateuch’, although he left this as an open question that ‘deserves closer investigation’ (p. 143 n. 1). Schwienhorst’s reluctance to apply his analysis of Joshua 6 to broader theories for the composition of the Hexateuch is symptomatic of a widespread malaise—what Hayes called ‘a state of anxious disarray’ (1979: 195)—that had entered Pentateuchal studies beginning in the late 1970s, as scholars began to question the classical Documentary Hypothesis yet were unsure about how to proceed in constructing alternatives to it.
New Theories of the Hexateuch, 1976–2017
In 1977, Rendtorff’s study Das überlieferungsgeschichtliche Problem des Pentateuch set off a scholarly discussion that has influenced the course of Pentateuchal criticism up to the present. One of Rendtorff’s main methodological contributions in this study is his critique of the fact that the classical Documentary Hypothesis (Die neure Urkundenhypothese) had become identified with the literary-critical method as such (1977: 149). As an alternative to taking the Documentary Hypothesis as a starting point, Rendtorff advocated investigating the growth of the Pentateuchal narratives from smaller cycles into ‘larger units’ (‘größere Einheiten’) without assuming that every Pentateuchal text necessarily formed part of an independent larger unit (pp. 154-58). In the wake of Rendtorff’s work, a significant number of scholars, particularly in Europe, abandoned the classical Documentary Hypothesis and began developing a variety of alternative models for understanding the formation of the Pentateuch (for discussions of this process see, e.g., Nicholson 1998: 95-221; Zenger 2008: 74-123; and Römer 2013). These changes in Pentateuchal criticism—combined with an increasing questioning of Noth’s Deuteronomistic History hypothesis—provoked significant changes in the Hexateuch debate as well. Following a handful of studies in the 1970s and 1980s, a renewed willingness to speak of a Hexateuch began to gain momentum in the 1990s. Since 2000, the Hexateuch question has once again become a central aspect of Pentateuchal criticism (for an early status quaestionis see Frevel 2004: 80-91). A central point of debate in this more recent discussion of the Hexateuch is whether the Hexateuch should be defined as a pre-Priestly and/or pre-Deuteronomistic narrative (a ‘primitive Hexateuch’) or as the books of Genesis through Joshua in their near-final form (a ‘redactional Hexateuch’ or a ‘Hexateuch proper’).
The Theory of a ‘Primitive Hexateuch’
One year before the publication of Rendtorff’s influential study, Tengström (1976) published a monograph on the Hexateuch question that foreshadowed some of Rendtorff’s critiques of the classical Documentary Hypothesis. Tengström’s arguments for a Hexateuch began with a critique of one of the logical conclusions of Noth’s Deuteronomistic History hypothesis: if the continuation of the Pentateuchal narratives is not to be found in the present book of Joshua, then one is forced to reckon with a conquest narrative that is now lost (see Noth 1943: 252). For Tengström, such a hypothesis seemed extremely unlikely and was in any case unverifiable. Rather, Tengström argued that the problem of the relationship between the Pentateuch and the Former Prophets cannot be solved without positing a ‘Hexateuch’ as an original narrative unit and thus concluded that the Deuteronomistic History should be viewed as a continuation of this older narrative (1976: 10-12). From a methodological point of view, Tengström’s study differs from most earlier studies dealing with the Hexateuch question and in certain ways foreshadows some of the changes that would occur in Pentateuchal criticism beginning in the late 1970s. Tengström consciously departed from the Documentary Hypothesis, critiquing the traditional method of extrapolating the source analysis of individual texts (such as the narratives in Genesis) to the Hexateuch as a whole (p. 18). Rather than positing continuous, parallel Hexateuchal sources, Tengström adopted a supplementary model that posited an early ‘Israel-saga’ and various additions to this narrative (p. 16).
Even if Tengström’s approach in theory made advances over prior methods, his application of this approach produced rather surprising results: in contrast to many literary critics who situated J in the time of Solomon, Tengström in fact situated the Hexateuch’s earliest ‘epic work’ in premonarchic Shechem (pp. 13-14). Tengström’s attribution of texts to his ‘Israel-saga’ is based primarily on two criteria: (1) the importance of Shechem, which allows one to create a profile for the contemporary context of the work, and (2) the promise to the ancestors, which reappears as a Leitmotiv throughout the Hexateuch (p. 102). Tengström’s dating of the ‘Israel-saga’ and his delimitation of its scope is open to criticism in several respects. First, many of Tengström’s arguments for a pre-monarchic date of composition are either inconclusive or could also be used in support of a post-monarchic dating (see, e.g., Achenbach 2005: 146 n. 115) and reflect a conflation of the narrated time of a text and its time of composition. Another shortcoming in Tengström’s method is his assignment of material to the ‘Israel-saga’ based on thematic considerations (such as the promise to the ancestors) rather than the isolation of a more basic narrative thread using purely literary-critical means. Nevertheless, Tengström’s study is significant insofar as it represents an attempt to reckon with a ‘primitive Hexateuch’ without relying on the classical source-critical model.
Coats also advocated a sort of ‘primitive Hexateuch’, albeit on traditio-historical grounds (1987: 16). Although Coats was loath to search for Pentateuchal sources in the book of Joshua, he attempted to demonstrate the continuity between Joshua and the preceding books on the basis of form-critical observations. For Coats, the fact that the book of Joshua narrates both the ‘conquest theme’ and a ‘heroic saga’ about Joshua ‘sets the book of Joshua alongside the longer tradition about Moses as heroic man and man of God’ (p. 25). Yet Coats’s reluctance to identify the conclusion to a Pentateuchal narrative thread in the book of Joshua—influenced by his acceptance of the Deuteronomistic History hypothesis—forced him to resort to a rather vague explanation for why the book of Joshua fits so well with the Pentateuch: ‘It is a product of the tradition’s history at a level of growth in the tradition rather than at the level of a combination of literary sources’ (p. 28).
Bieberstein’s 1995 study Josua – Jordan – Jericho marked a significant turning point in the discussion of a ‘primitive Hexateuch’. According to Bieberstein, the narrative that ends with the stories of conquest in the first part of the book of Joshua does not begin in Genesis but rather in Exodus 2 (1995: 336-41). Although Bieberstein was not the first to note the disjunction between the books of Genesis and Exodus (for a review of literature on this subject see Schmid 1999: 56-102), his study is significant for proposing (albeit tentatively) the existence of an early ‘exodus-conquest narrative’ spanning from Exodus 2 to Joshua 24 on the basis of this disjunction (see also more recently Bieberstein 2011). This isolation of an early exodus-conquest narrative has come to play a major role in the discussion of a ‘primitive Hexateuch’.
For example, in his 1999 study Erzväter und Exodus, Schmid noted the spatial and temporal coherence of the narratives in Exodus–Joshua and concluded that ‘a clear division of the narrative material in Exodus–Joshua is not possible’ (1999: 131). According to Schmid, if one analyzes the latest layer of the book of Exodus that does not presuppose a connection with Genesis, a literary horizon spanning from Exodus to 1 Kings 12 (and perhaps beyond) seems likely (pp. 140-41). Schmid identified the authors behind this (pre-Priestly) ‘Exodus-Moses Story’ as ‘Deuteronomistic’ and suggested that this narrative arc is a better alternative to Noth’s Deuteronomistic History beginning in Deuteronomy 1 (p. 158). Thus, Schmid argued for an Enneateuch (or more precisely: an Octateuch) as the best framework for understanding the pre-Priestly narratives in Exodus through Kings, although he did not address in detail the question of identifying a potentially pre-Deuteronomistic narrative within these books (but see now Schmid 2008: 89 [ET 79]; 2014: 45-46).
Like Schmid, Kratz (2000a) has argued for a Deuteronomistic Enneateuch (more precisely: Octateuch) as one of the prior stages in the formation of the books of Genesis–Kings but also seeks to isolate earlier literary works that underlie this ‘Enneateuch’. In Kratz’s view, one of these works consisted of the history of the people from the exodus up to the entry into the land (2000a: 210 [ET 202]). This ‘exodus-conquest narrative’, in turn, was formed on the basis of several originally independent narrative traditions, such as early versions of Joshua 6 and 8 (p. 292 [ET 284]). In Kratz’s view, these and similar isolated narratives were formed into the base text (Grundschrift) of the exodus narrative (= EG, with ‘E’ now signifying ‘exodus’ rather than ‘Elohist’) through unifying elements such as the figure of Moses, Miriam and the sea, the wilderness itinerary, the figure of Joshua, and the crossing of the Jordan. This base text was later supplemented, particularly within the wilderness itinerary (= ES) (for the delineation of EG and ES see pp. 293-94 [ET 292]). From this point, the pre-Deuteronomistic ‘Hexateuch’, which Kratz dates to the period between 720 and 587
Several other scholars have also adopted the theory of an exodus-conquest narrative as a literary precursor to the Pentateuch and book of Joshua. Müller has proposed that such a narrative may have ended with a direct transition from Josh. 11.23a to Josh. 24.28-30 (2004: 77, 231-32; cf. also Konkel 2008: 260, who argues that Joshua 24* may have been part of a pre-Deuteronomistic Hexateuch). Likewise, in his 2008 commentary on the book of Joshua, Knauf argues for the existence of a pre-Deuteronomistic exodus-conquest narrative that originated in the area of the former northern kingdom sometime around 600
The Theory of a ‘Redactional Hexateuch’
This ‘new consensus’, however, is hardly unanimous. Indeed, parallel to the recent Hexateuchal theories that focus on isolating a pre-Priestly and pre-Deuteronomistic narrative in Exodus–Joshua, another group of scholars has argued that the Hexateuch first emerged through the redactional joining of a Tritoteuch (Genesis–Leviticus*) or Tetrateuch (Genesis–Numbers*) with some sort of Deuteronomistic literary work, whether an independent Deuteronomistic conquest narrative (Deuteronomy–Joshua* = DtrL) or the Deuteronomistic History (Deuteronomy–Kings). As noted above, the notion of a ‘redactional Hexateuch’ was already proposed by Langlamet (1971: 17) (although Langlamet also posited an earlier ‘primitive Hexateuch’). Such a view was further developed by Blum, who concluded that the Hexateuch is a relatively late literary work created by a small number of redactional texts linking the Deuteronomistic History with the materials that precede it (1984: 45-61, esp. 60). For Blum, the notion of a ‘redactional Hexateuch’ is supported by the reference to a ‘book of the Torah of God’ (ספר תורת אלהים) in Josh. 24.26, which, in Blum’s view, seeks to establish a Hexateuch following the formation of the Pentateuch, advocating the authority of a ‘book of the Torah of God’ rather than a ‘book of the Torah of Moses’ (2006: 97-100; 2007: 95-97 [ET 69-71]; cf. Albertz 2015: 71).
Another proponent of a redactional Hexateuch is Römer, who has expressed his views on the Hexateuch in a number of publications (Römer and Brettler 2000; Römer 2005: 178-83; 2006; 2010; 2011). Similarly to Blum, Römer (and Brettler) have called attention to the links that Joshua 24 forms with texts such as Gen. 35.1-7 and Deut. 34.7-9, which, like Joshua 24, contain a mixture of Deuteronomistic and Priestly language (Römer and Brettler 2000: 411). Based on this mixture of language in Joshua 24 (as well as the presence of Priestly language elsewhere in Joshua but rarely in Judges–Kings), Römer and Brettler conclude that the Hexateuch was the product of a minority ‘deuteronomistic-priestly group’ in the Persian period (pp. 408-16; cf. Römer 2005: 178-83; 2007: 444). Like Blum, Römer concludes that the earliest Hexateuch post-dates the formation of the Pentateuch, which in Römer’s view itself arose through the redactional joining of a Priestly Tritoteuch in Genesis–Leviticus* with the book of Deuteronomy, now detached from the Deuteronomistic History (2002: 222-23; 2007: 426). This post-Deuteronomic and post-Priestly Hexateuch briefly competed with the Pentateuch for canonical status but was ultimately overshadowed by the latter (2006: 545-47).
Otto and Achenbach have also adopted the notion of a redactional Hexateuch, although unlike Blum and Römer, they regard the formation of a Hexateuch as earlier—not later—than the delineation of the Pentateuch as a discrete literary work. For example, in his more recent work, Otto has argued that the Hexateuch is the product of the redactional combination of DtrL (Deuteronomy 1–Joshua 23*) and P (Genesis 1–Leviticus 16*) during the fifth century
Albertz has recently proposed a reconstruction of the formation of the Pentateuch and Hexateuch that combines the approach taken by Blum and Römer on the one hand and that taken by Otto and Achenbach on the other. Like Römer, Albertz argues that an early form of the Pentateuch was created by a ‘late-Deuteronomistic redaction’ (‘D’) joining the book of Deuteronomy with a non-Priestly and Priestly Tritoteuch (2012: 23). This early Pentateuch was then expanded through a ‘Hexateuch redaction’, which Albertz regards as more extensive than the Hexateuch redactions proposed by Blum and Römer but less radical than those proposed by Otto and Achenbach (for a precise delineation see Albertz 2015: 56-67). Like Otto and Achenbach, however, Albertz also proposes a separate and later Pentateuch redaction that added further materials in the process of (once again) delineating the Pentateuch as a distinct literary unit (2015: 71-72).
Intermediate Positions
Several scholars have taken an intermediate position between regarding the earliest Hexateuchal literary work as an early exodus-conquest narrative and defining the Hexateuch as a post-Deuteronomic and post-Priestly redactional construct. Beginning in the 1970s, Zenger and Weimar (both together and separately) developed a model for the formation of the Pentateuch that Zenger has more recently termed the ‘Münster Pentateuch model’ (Zenger 1971; Weimar and Zenger 1975; Weimar 1977; for an overview see Zenger 2008: 101-103). Within this model, Zenger and Weimar posited the development of two separate stories of origins sometime after the fall of the northern kingdom in 722 bce: a narrative cycle binding together a series of smaller narratives about Abraham and Jacob (Gen. 13–50*) and a narrative binding ‘the old exodus story’ and disparate ‘Joshua stories’ together into a ‘conquest narrative’ (‘Landnahmegeschichte’) (Zenger 2008: 101). Not long after these two larger narrative cycles coalesced, they were combined in a normative historical narrative spanning from the creation of humanity to the conquest of Canaan (Genesis 2–Joshua 24*), which included several important ‘bridge texts’ such as Genesis 15*; Exodus 34*; and Joshua 24*. Zenger compared the scope of this narrative with Wellhausen’s ‘Jehovistic history’ but noted that his (and Weimar’s) understanding of its literary pre-history differs radically from that of Wellhausen, since Wellhausen reckoned with the parallel sources J and E, while Zenger and Weimar reckon with the joining of independent narrative cycles (Zenger 2008: 102). In short, this stage of the ‘Munster Pentateuch model’ posits the existence of a pre-P and pre-D Hexateuch spanning from Genesis 2 to Joshua 24 by the late seventh century
Like Weimar and Zenger, Carr has inquired into ‘the existence of a pre-Priestly composition extending from Genesis through the Moses story and beyond’, although Carr’s primary interest is to reconstruct a ‘post-D Hexateuchal composition layer’, that is, a Hexateuch that included the book of Deuteronomy (2011a: 255). For Carr, a significant feature of this ‘post-D Hexateuchal composition layer’ is the distribution of the conquest motif to other texts in Genesis–Joshua (2011a: 272). Similarly to Otto and Achenbach, Carr has argued that three distinct compositions preceded the formation of a post-D Hexateuch: ‘non-P Genesis materials…, a pre-D Moses story, and a composition starting with Deuteronomy and including (at least) Joshua’ (2011a: 278; cf. 2011b: 75). For Carr, the argument that Deuteronomy and Joshua originally formed a distinct ‘document’ (reminiscent of Lohfink’s DtrL, although Carr does not use this siglum) prior to being incorporated into a post-D Hexateuch is based primarily on the premise that the historical summaries in Deuteronomy 1–11 must have formed the beginning of an independent literary work (2011b: 75-76). Thus, Carr’s ‘post-D, pre-P “Hexcomp” layer’ (2011b: 82) represents an intermediate position between the notion of a ‘primitive’ (i.e., pre-Priestly and pre-Deuteronomic) Hexateuch and the notion of the Hexateuch as a late (i.e., post-Deuteronomic and post-Priestly) literary work. From a broader perspective, however, Carr’s model is a variant of the ‘redactional Hexateuch’ approach, since Carr regards Deuteronomy–Joshua* as originally independent from the narratives in Exodus–Numbers*.
Synthesis and Evaluation
The foregoing survey of Hexateuchal theories from the mid-nineteenth century to the early twenty-first century has shown that interest in a ‘Hexateuch’ as a literary work has remained strikingly constant despite the vicissitudes of the various theories for the composition of the narrative books of the Hebrew Bible. During the second half of the twentieth century, although Noth’s hypothesis of a Deuteronomistic History caused many scholars to view the books of Deuteronomy and Joshua as more closely linked to Judges–Kings than to Genesis–Numbers, a significant minority of scholars continued to speak of a Hexateuch as a discrete literary work. Somewhat ironically, however, the literary-critical models that some of these scholars employed may have contributed to others’ skepticism of the notion of a Hexateuch in the face of the Deuteronomistic History hypothesis. For example, scholars such as Simpson, Hölscher, and Eissfeldt continued to argue rather dogmatically for a source-critical division of Genesis–Joshua without addressing the Deuteronomistic History hypothesis at all. Meanwhile, others such as Sellin, Fohrer, and Mowinckel attempted to reconcile a source-critical approach to the book of Joshua with Noth’s arguments that Joshua contains independent narrative traditions that have been reworked by the Deuteronomistic Historian. Such attempts, however, either required these scholars to posit hypothetical compositional stages for which there is no evidence (such as the shared source for J and E proposed by Sellin and Fohrer) or to maintain blatant contradictions in their argumentation (such as Mowinckel’s attribution of Joshua 2–11 to both Jv and Dtr). Even Tengström’s methodologically innovative study, which departed from earlier attempts to analyze Hexateuchal narratives in terms of classical source criticism, is open to criticism from other fronts, such as Tengström’s dating of the earliest Hexateuchal narrative to the pre-monarchic period and his delineation of the basic narrative in part on thematic (rather than literary-critical) grounds.
Since the 1990s, the notion of a Hexateuch has slowly regained wider acceptance, at least in German-language scholarship, although in the process it has been dramatically redefined. This renewed discussion of a Hexateuch can be divided into two distinct approaches. The first approach is the attempt to identify a ‘primitive Hexateuch’ within the books of Exodus through Joshua. The methodological foundations of this approach can be traced back in large part to Rendtorff’s emphasis on the formation of larger literary units (größere Einheiten) through the joining of independent cycles of traditional material. According to such a model, one need not assume that an earlier narrative strand necessarily contained all of the ‘Hexateuchal themes’ (to use von Rad’s term) found in the present text of Genesis through Joshua. This departure from the pattern of Heilsgeschichte has led scholars such as Bieberstein, Kratz, Becker, Knauf, Gertz et al., Schmid, Berner, Frevel, Nihan, and Germany to posit the existence of an early ‘exodus-conquest narrative’ spanning from Exodus 2 to Joshua 10; 11; or 12. Unlike traditional arguments for the existence of a Hexateuch on the basis of both thematic grounds and narrative continuity (cf. Kuenen 1861: 181 and Noort 1998: 26), the guiding criterion for identifying this ‘primitive Hexateuch’ in more recent research is narrative continuity alone, since thematic links—such as the promise of the land to the ancestors—may be the result of later redactional insertions designed to join originally separate materials (cf. Rendtorff 1977: 109-12 and Schmid 1999: 107-11). The second major approach in the renewed discussion of the Hexateuch is the attempt to identify a ‘redactional Hexateuch’ in Genesis–Joshua created by the joining of two earlier literary horizons: (Genesis–)Exodus–Numbers and Deuteronomy–Joshua(–Kings). Despite some differences among individual scholars, this general approach has been taken especially by Blum, Römer, Otto, Achenbach, Carr, and Albertz. Significantly, all of these scholars accept the hypothesis of a Deuteronomistic History (or DtrL), and their understanding of the Hexateuch is strongly influenced by their understanding of Deuteronomy and the books that follow as fundamentally disconnected from Genesis through Numbers.
Notably, both the ‘primitive Hexateuch’ hypothesis and the ‘redactional Hexateuch’ hypothesis depart from von Rad’s notion that all of the major ‘themes’ of the books of Genesis through Joshua (i.e., the primeval history, the ancestral narratives, the exodus from Egypt, the giving of the law at Sinai, the journey through the wilderness, and the conquest of the land) already belonged together at a pre-literary stage of tradition. For the proponents of a ‘primitive Hexateuch’, the narratives in the book of Genesis are regarded as having been joined secondarily to a pre-Priestly (and for some also a pre-Deuteronomic) ‘exodus-conquest narrative’. For the proponents of a ‘redactional Hexateuch’, the process of the formation of the Hexateuch is somewhat more complex and involves the combination of several discrete literary works, including pre-Priestly materials in Genesis and Exodus; a Priestly work in Genesis–Exodus(–Leviticus); and a Deuteronomistic conquest account in Deuteronomy–Joshua*.
Neither of these models is completely free of problems. For the ‘primitive Hexateuch’ model, the most significant problem is the abrupt shift from Moses to Joshua as the main protagonist, which may point to the combination of originally independent traditions revolving around the figures of Moses and Joshua, although any attempt to resolve the problem of the transition from Moses to Joshua is by its nature speculative, since doing so inevitably requires one to make arguments that go beyond the preserved textual evidence. The various theories for a ‘redactional Hexateuch’ are likewise required to reckon with undocumented textual forms (see esp. Otto 2000 and Achenbach 2003a, who conclude that many of the ‘original’ narratives in the book of Numbers have been overwritten by the so-called Hexateuch redactor) but also with additional compositional theories that are open to critique, such as the existence of DtrL and/or DtrH as independent literary works. The most significant problem with such a view is the fact that Deuteronomy 1–3 presupposes a narrative that recounts the people’s journey through the wilderness and thus cannot be independent from the narratives in Exodus–Numbers in any absolute sense (cf. Frevel 2004: 88-89; 2011: 31-34, with further literature). By extension, since the people’s journey through the wilderness in the book of Numbers can hardly constitute the conclusion to the exodus narrative but instead presupposes the people’s entry into the land, then the Vorlage of Deuteronomy 1–3 must have spanned at least from the exodus to the conquest narratives in the book of Joshua (or some other hypothetical account of the entry into the land). This would seem to support the theory of an exodus-conquest narrative as one of the literary precursors to the Pentateuch and the book of Joshua rather than the theory of an originally independent Tritoteuch or Tetrateuch that was only combined with the book of Joshua at a post-Deuteronomic (and post-Priestly) stage of composition.
Although a complete consensus on the formation of the Hexateuch is still not in sight, one step that could benefit the current discussion on all sides is for scholars to clearly lay out how they define the term ‘Hexateuch’ and preferably to avoid using the term to describe a literary work that does not in fact span from the book of Genesis to the book of Joshua. Although I use the term ‘primitive Hexateuch’ above for the sake of comparison, I find the term ‘exodus-conquest narrative’ to be more appropriate with reference to a literary work that does not include the book of Genesis and/or the Deuteronomic law (cf. Germany 2017). Even if the use of more precise terminology may be less elegant, it would allow both current and future scholars to understand and engage more easily with the wide array of theories for the formation of the books of Genesis through Joshua.