Scriptural Interpretation in the Writings of Josephus Steve Mason Associate Professor Division of Humanities York University North York Ontario Canada M3J 1P3 Tel. (416) 736-5158 E-Mail: shlomo@vm1.yorku.ca December, 1993 Scriptural Interpretation in the Writings of Josephus For the study of scripture in early Judaism, as for almost any question in this period, the writings of Josephus are a first reference point. Indeed, his extensive biblical paraphrase in the Judean Antiquities has invited a massive amount of careful study, both in general and in numerous particulars. Since the main results of that research are readily available in recent surveys, another survey would be superfluous here. I propose rather to focus on one passage and one issue. The issue is the shape of Josephus's Bible, or "canon," and the passage is his most programmatic statement, Against Apion 1.37-43. In the current debate over the shape of the first-century Jewish canon(s), this passage must be dealt with by those who argue for all positions_open or closed, tripartite or bipartite. Not surprisingly, perhaps, he tends to support whomever is making the argument. Sid Z. Leiman avers: "From Josephus' statement, it is evident that he recognized a tripartite canon. . ." John Barton, however, contends: "At all events the primary idea to which Josephus is a witness is not that the books of Scripture were organized into a tripartite form, but that they derived from either of two sources: Moses and the prophets. . . . Such evidence as we have from Hellenistic Judaism thus confirms the essentially bipartite character of Scripture. . . ." This debate provides a telling example of W. C. van Unnik's observation that Josephus is extensively used but rarely understood on his own terms as an intelligent author. A conspicuous example of the problem is provided by J. Alberto Soggin's discussion of canon in his influential textbook. Soggin reports that, "Flavius Josephus, Contra Apionem I.8 [=38-46], lists the following qualifications needed by a book for it to become part of the canon as conceived by the Pharisaic movement." The three qualifications are: that the book have been written between Moses and Ezra; that it "have a certain objective sacred quality which differentiated it from all other non-sacred books"; and that it be included among the "twenty-two books listed by Josephus." As we shall see, however, very little of this language_ "qualifications," "canon," "become part of," "Pharisaic movement," "Ezra," "objective sacred quality," "twenty-two books listed"_is warranted by the text itself. Soggin's assessment reflects the traditional tendency of scholars, even those who handle their own (often biblical) texts with finesse, to exploit Josephus without much regard for his literary purposes. Things are changing. Leiman and Barton are at least somewhat sensitive to the particularity of Josephus's language and worldview. But even they have only noted some inconveniences in using Josephus; they still end up drawing more or less direct support from him. I propose to contribute to the discussion by beginning at the other end, with Josephus's writings themselves. We shall see that his celebrated passage is even less useful than has been feared for the scholarly canon quest, though it throws much light on his general outlook. If we are to understand Josephus, we need in the first instance to lay aside our questions, and to ask what his questions are. We shall examine first, therefore, the role that AgAp 1.37-43 plays in its literary context, and what the key terms of the passage might reveal about Josephus's world of thought. A discussion of his use of the Bible in Ant and War will then provide a check on the degree to which his statement correlates to his actual practice. Only in the conclusion, with an interpretive proposal in hand, can we consider the implications of this passage for the question of the first-century Jewish canon(s). Context and Purpose of Against Apion 1.37-43 Josephus's last extant composition has a clear plan. He writes out of a perception that his magnum opus, the Ant, has provoked a chorus of detractors, who disbelieve what he says about Judean antiquity because of the lack of reference to Judeans in Greek literature (AgAp 1.1-5). Josephus's rebuttal has the following structure, which he highlights with pointed transitional remarks. INTRODUCTION (1.1-59) 1.1-5 General Introduction: recapitulation of Ant; reasons for writing; scope; general thesis_the antiquity of the Judeans. 1.6-59 Digression: Greek and Oriental ("Barbarian") Historians. Thesis: Greek historians are not the last word; Orientals are better. BODY (1.60-2.286) I. Reasons for Greek Silence about Judeans (1.60-68) II. Oriental Evidence for Judean Antiquity: Egyptian, Phoenician, Chaldean (1.69-160) III. Overlooked Greek Evidence for Judean Antiquity (1.161-218) IV. Refutation of Anti-Judean Slanders (1.219-2.144) V. Preemptive Positive Portrayal of Judeans' Laws (2.145-286) CONCLUSION (2.287-296) Thus the body of the work examines the evidence for Judean antiquity, both Oriental and Greek, and then argues the nobility of these ancient Judean traditions. Before he reaches the body, however, Josephus interjects a note of alarm that, in principle, Greek sources should be considered the final authority. The argument within this lengthy digression is also clearly structured. It has three main points: the lateness of Greek culture; the many contradictions among Greek accounts, which he attributes to a lack of official records (ajnagrafaiv) and a preoccupation with rhetorical competitiveness; and the superiority of Oriental (bavrbaroi, 1.58) record-keeping. A. 1.6-14 Relative Lateness of Greeks; their culture is derivative B. 1.15-43 Contradictions among Greek Writers 1.15-18: Examples of Greek Contradiction 1.19-22: Reason 1: lack of official records (ajnagrafaiv) 1.23-27 Reason 2: rhetorical competitiveness C. 1.28-59 Orientals' Care for Official Records (ajnagrafaiv): 1.28 Well-Known Examples: Egyptians and Babylonians entrust records to priests and Chaldeans 1.29-59: Lesser-Known Example: the Judeans 1.29: thesis: Judeans entrust their official records (ajnagrafaiv) to chief priests and prophets 1.30-36: on the priests (precautions for pure lineage) 1.37-43: on the prophets (writing of records) 1.44-56: modern example of Judeans' precision: Josephus the eyewitness, priest, and lover of truth. 1.57-59: summary and prospectus: Greeks, Orientals, Judeans. Within the sub-argument on Oriental record-keeping, Josephus intends to show that the Judeans' measures for maintaining official records are even more rigorous than those of their more famous Eastern neighbours, which are in turn qualitatively superior to Greek practices. 1.28 Since everyone concedes, first, the care shown by the Egyptians and Babylonians concerning their official records [th;n peri; ta;" ajnagrafa;" ejpimevleian]_whereby in the one case the priests had been entrusted and would philosophize concerning them, and the Chaldeans among the Babylonians_and, second, that certainly, of those mixed in among the Greeks, the Phoenicians used letters to the highest degree_with respect to the whole administration of life and also for the tradition of the communal accomplishments [pro;" th;n tw'n koinw'n e[rgwn paravdosin]_I think it best to leave (these matters) alone. 29 But concerning our forebears, that they practiced the same care_I think it permissible to say: even greater (care) than those mentioned_with respect to the official records [peri; ta;" ajnagrafa;" ejpimevleian], having assigned this matter to the chief priests and prophets, so that until our own times this charge has been cherished with all precision_and if it is not too bold to say, it will be (so) cherished (in the future also)_I shall try concisely to demonstrate. Whereas the Egyptians and Babylonians entrusted their records_the precise mechanisms are not explored_to priests and Chaldeans, the Judeans assigned theirs to chief priests and (corresponding to Chaldeans) prophets. Josephus is least clear about the role that Judean priests play in the keeping of records, since he devotes most of their space to their stringent measures for ensuring genealogical purity. The section ends: 36 But the surest proof of our precision is this: the chief priests among us for 2000 years are named, sons from fathers, in the archives. And it is forbidden to those who violate the aforementioned (rules) either to offer the sacrifices or to participate in the rest of the ritual. It may be inferred from his emphasis on the continuation of the priestly succession in contrast to that of the prophets, from earlier remarks in War and Ant, and from later statements in this work, that he sees the chief priests and their subordinates as preservers, executors, and philosophical expositors of the records, somewhat like the aforementioned Egyptian priests. This conclusion is confirmed by his discussion of the prophets, which is our main interest, for he leaves no doubt that it was the prophets who actually wrote the records: 37 Accordingly_or rather necessarily_then, seeing that the writing (of the records) is not the personal prerogative of everyone, nor is there actual disagreement among any of the things written, but the prophets alone learned the highest and oldest matters [ta; me;n ajnwtavtw kai; palaiovtata] by the inspiration of the God, and by themselves plainly recorded events as they occurred, 38 so among us there are not tens of thousands of discordant and competing volumes, but only twenty-two volumes containing the record of all time [tou' panto;" e[conta crovnou th;n ajnagrafhvn], which are rightly trusted [ta; dikaivw" pepisteumevna]. 39 Now of these, five are those of Moses, which comprise both the laws and the tradition from human origins [a} touv" te novmou" perievcei kai; th;n ajp j ajnqrwpogoniva" paravdosin] until his passing; this period falls little short of 3000 years. 40 From Moses' passing until the Artaxerxes who was king of the Persians after Xerxes, the prophets after Moses wrote up what happened in their times [or, as they saw it; ta; kat j aujtou;" pracqevnta] in thirteen volumes. The remaining four (volumes) comprise hymns toward the God and advice for living among men [u{mnou" eij" to;n qeo;n kai; toi'" ajnqrwvpoi" uJpoqhvka" tou' bivou]. 41 From Artaxerxes until our own time all sorts of things have been written [gevgraptai mevn e{kasta], but they have not been considered of the same trustworthiness [pivstew" dV oujk oJmoiva" hjxivwtai toi'" pro; aujtw'n] as those before them, because the exact succession of the prophets failed [dia; to; mh; genevsqai th;n tw'n profhtw'n ajkribh' diadochvn]. 42 Now it is clear in practice how we approach our special texts: for although such an age has already passed [sc. since Artaxerxes], no one has dared either to add anything or to take away from them or to alter them. But it is innate among all Judeans from their very first moment of existence to consider them decrees of God, to stand by them, and for their sake, if necessary, cheerfully to die. 43 Thus already many of (our) prisoners of war have on many occasions been seen patiently enduring tortures and the ways of all sorts of deaths in theatres, without letting slip a single word against the laws and the related official records [para; tou;" novmou" kai; ta;" meta; touvtwn ajnagrafav"]. The main contribution of this section to the larger argument comes in the first sentence (37-38). Unlike the generalized Greek situation (Josephus has in mind writers from Homer to at least Thucydides; 1.12-18), among the Judeans only prophets could write official records: they were enabled by inspiration to learn of things beyond the limits of human knowledge; for the rest, they recorded affairs of their own times. The result is a collection of twenty-two harmonious and wholly reliable volumes of national records, which are so consistent that Josephus can designate them all as a single record (ajnagrafhv; 1.38). The practical corollary of having such an established and carefully preserved tradition (42-43) is that Judeans know it well and are wholly committed to it. The middle section (39-41) supports the conclusion that prophets wrote a unified national record by analyzing this material according to several criteria. In terms of authorship, there are the books of Moses and those by later prophets. But the authorial distinction is not paramount here. Though Ant (1.18-26) and the rest of AgAp (2.173 et passim) feature Moses as the Judean legislator, the distinction between Moses and the others in this passage has mainly to do with the different periods of history they covered. Yet the chronological distinction is not fully developed either: Josephus does not bother to say when the remaining four books were written. Those non-historical books are mentioned only for the sake of completeness. Josephus is concerned to show that the records cover the history of "all time." His most comprehensive criterion for distinguishing among the records here is that of genre. Moses' writings themselves include both laws and tradition for a period of 3000 years. The subsequent prophets continued the records up until Artaxerxes by writing about "what happened" in their days, and so also wrote history or tradition. Finally, the collection includes a small amount of hymnic and sapiential material. Of the four genres_laws, tradition/history, hymns, and advice_Josephus's argument requires his preoccupation with history. Contrary to almost universal opinion, Josephus does not offer the slightest hint that these genres correspond to "divisions" or "sections": he does not suggest that Moses composed two sections, though he wrote in two different genres, or that the other four books represent two (small!) sections. Still less is there anything remotely like a division called "Prophets" in this passage, for all of the authors are prophets. The distinction of genres, along with the two other criteria, simply help the Gentile reader understand the various kinds of material within the twenty-two volumes of official Judean records. His main point, then, seems clear. In contrast to the many Greek authors who aggressively contradict each other concerning a brief and recent period of time (1.6-18), the Judeans' few records span the whole period from creation to Artaxerxes in a linear fashion. Moses had no competitors in recording the laws and the first 3000 years of history. Nor did the prophets after him, who successively took up the task of recording "what happened in their own times." And this whole process is so ancient that it was already completed by the time of the Persian Artaxerxes, successor of the famous Xerxes who had captured Athens_thus, at a time in which Greek historiography had not yet begun. Judeans do not even credit their own writings from the subsequent period, in which Greek history first appeared. I. Key Terms Josephus's digression on Judean traditions is highly stylized, exhibiting a marked preference for formal balance over rigorous logic. His language is consciously generic: each nation is assumed to have its own official records (ajnagrafaiv) or tradition of communal lore (paravdosi"), and Josephus can impose these universal categories to the Judeans' advantage over against "the Greeks." His use of paravdosi", for example, to describe what Moses taught in addition to the laws (1.39), does not appear as a native Judean category. In context, it parallels th;n tw'n koinw'n e[rgwn paravdosin of the Phoenicians, just mentioned (1.28), and the "very ancient record (paravdosin) of the past" possessed by Orientals in general (1.8; cf. also Ant 20.259). Similarly, Josephus takes over the ajkrivbeia word group from commonplace Hellenistic historiography, and applies it with vigour to all aspects of Judean culture: here, to the preservation of the official records; elsewhere to his own writings (War 1.9, 17; 7.454; Ant 20.263), or to the failure of others to achieve precision (AgAp 1.18). Judeans, he claims, offer the precision that is so widely sought. Or again, when he says that the Judean records are rightly trusted (dikaivw" pepisteumevna) he is not reflecting the Judean community's language of "faith" or some such thing, but only providing a contrast to his earlier notice that the Greeks are loathe to trust (movli" aujta; pisteuvousin) even their oldest works (1.14; cf. 1.161). For us who wish to learn from Josephus's statement about intramural "Jewish" phenomena, his argument is largely opaque. The "Official Records" The most important word in this section of the argument is ajnagrafhv. Josephus begins with the observation that the Greeks lack ancient and stable ajnagrafaiv (1.7, 11, 20, 21, 23); he contrasts Egyptian, Babylonian, and Judean ajnagrafaiv (1.9, 28, 38, 43); and he concludes that non-Greeks are far superior in maintaining ajnagrafaiv (1.58). This noun occurs forty-two times in his writings, but more than half of those instances (twenty-four) are in AgAp 1. It is not that he learned the phrase or thought of applying it to the Judean texts late in life, for his own speech in War 6, written a couple of decades earlier, already refers to "the official records (ajnagrafaiv) of the ancient prophets" (6.109). And both the introduction and conclusion to Ant designate the Bible by this term (1.12; 20.261). But ordinarily, when Josephus wishes to refer generically to a nation's traditional laws and customs he will use phrases like oiJ pavtrioi novmoi, ta; pavtria e[qh, ta; pavtria, or oiJ novmoi (e.g., AgAp 2.164; 2.237) His preference for ajnagrafaiv in AgAp 1 evidently reflects his desire here to stress that the Greeks, though they do have national traditions, possess only late and lacunose written records. He makes the point early, as we have seen, and later returns to an explicit contrast between Greeks and Judeans (2.155-156): Among the Greeks the very word "law" (novmo") was not even known in ancient times: Homer, who nowhere made use of it in his poetry, is a witness. For the populace was governed not by law, but by imprecise maxims and directives from kings; for a long time afterwards they continued to use unwritten customs (e[qh a[graga), and many of these they routinely altered according to circumstance. But our lawgiver . . ., when he had received a complete prescription for the regulation of life, persuaded them to welcome it and secured its preservation forever. Although it serves his purposes elsewhere to place the Judeans' pavtrioi novmoi on a par with those of other nations, here he wants to press his advantage by referring to what only Judeans and other Easterners allegedly have: ancient official records. Accordingly, most of the remaining occurrences of ajnagrafhv in this work refer to the "official records" of the Oriental nations, which plainly mention the Judeans. Even the Eastern authors on the Judeans, such as Manetho, erred when they departed from their own official records (1.287). "The Succession of the Prophets" Given his emphasis on the long-standing perfection of the Judean ajnagrafaiv, it is noteworthy that Josephus uses the same term of his own major compositions (War 7.455; Ant. 1.18). Does he, then, think of his writings as an authoritative (and prophetic) continuation of the ancient records? That question brings us to Josephus's famous observation that things written after Artaxerxes are not credited by Judeans in the same way as the ajnagrafaiv because "the exact succession of the prophets failed" (1.41). This statement is a lynch-pin in discussions of the cessation of prophecy in Israel and the closure of the "canon." What does it mean? We have already seen that as early as War 6.109 Josephus had connected the prophets with the writing of ajnagrafaiv. In the preface to that work, further, he proposes to begin his account in the time of Antiochus IV (1.19) because: Many Judeans before me have written up the affairs of the ancestors with precision, and some Greeks have recast those (accounts) into their own language without missing much of the truth. So, where the historians of these affairs finished, and our own prophets, I shall make the beginning of my account. (1.17-18) In this scheme it is not immediately clear how the second pair, historians and prophets, relates to the first, Judeans and Greeks. Barton infers that the historians (suggrafei'") wrote only until the Exile, since that is the last item mentioned in Josephus's list of things already documented (1.17). But that list is representative, not exhaustive, for Josephus plainly establishes the beginning point of his narrative at the time of Antiochus (1.19); this, not the Exile, must be where the "historians" finished. I would propose rather that the prophets and historians correspond, chiastically, to the Judeans and Greeks of the first sentence. This is indicated by the phrase "our own prophets" (oiJ hJmevteroi profh'tai), which implicitly distinguishes the historians as aliens, and by the close parallel in AgAp 1.218, where the identical Greek phrase, "without missing much of the truth" is used of Demetrius Phalereus, Philo the Elder, and Eupolemus. All of these Josephus regards as Greek historians: they came close to the truth because they consulted the Judean records, and their minor flaws can be excused on the basis of misunderstanding. If this reasoning is sound, then Josephus's later conception of prophets (in AgAp) as ancient authors of the national records would be represented early in his development. We can only conclude that his connection of the Judean ajnagrafaiv with long-gone profh'tai played a fundamental role in his outlook. This hypothesis is confirmed by Josephus's instinctive lexical choices, for it is now well known that throughout his writings he reserves the profht- word group for those who lived long before his own time. Almost all such occurrences_36/38 for profhteiva, 57/58 for profhteuvw, 281/287 for profhvth", and 3/3 for profh'ti"_refer to the characters of the Hebrew Bible who play a role in Ant. Josephus often introduces prophetic vocabulary where it is absent in the extant Hebrew and OG, thus enhancing the role of prophets in the biblical narrative, and demonstrating his interest in the subject. His prophets include those who wrote books, like Moses and Joshua (Ant 2.327; 3.60; 4.311, 420), along with many who did not, like Jacob, Aaron, Phineas, and Nathan (Ant 2.194; 3.192; 5.120; 6.57; 7.214). The only non-biblical referents of profht-language are as follows. (a) Contemporaries like Theudas called themselves prophets (War 2.261; 6.286; Ant 20.97, 169), but they are dismissed by Josephus as "false prophets" and deceivers. (b) Josephus's observation that Lysimachus's account of Judean origins has the Egyptian king consulting the oracle of Ammon, rather than the "Egyptian prophet" cited by Chaeremon (AgAp 1.312; called a iJerogrammateuv" in 1. 289), is sarcastic. (c) Josephus claims to quote verbatim the second-century bce Alexander Polyhistor, who refers to a Judean historian as "Cleodemus the prophet" (Ant 1.240); the language is explicitly attributed to Alexander. (d) Josephus makes quite a bit of his assertion that the Hasmonean John Hyrcanus (ruled 135-104 bce) enjoyed the gift of prophecy along with the high priesthood and rule of the nation (War 1.68-69; Ant 13.299). Indeed, Hyrcanus is the last prophet recognized by Josephus, and his exceptionally late date may therefore account for the phrase "exact succession of the prophets": he was a prophet long after prophecy had otherwise ceased. (e) Finally, Rebecca Gray has cogently argued that Josephus understands the shining of the high priest's "breastplate" as a prophetic phenomenon, and that when he dates its cessation to 200 years before his own time of writing (Ant 3.218) he also has in mind the tenure of John Hyrcanus. This consistency of usage is the more conspicuous when we realize that Josephus refrains from using profht- language even of his most admired contemporaries, who nevertheless receive authentic revelations and make accurate predictions, viz.: the Essenes and himself. Certainly, he is not reluctant to claim perfect accuracy for his works (War 1.1-16; 7.455; Ant 1.17; 20.260-263); he claims to have experienced genuine divine revelations (War 3.350-354, 406); he explicitly and implicitly parallels his own career to Jeremiah's and Daniel's (War 5.391-393; Ant 10.119); he everywhere stresses his priestly qualifications (War 1.3; 3.352; Life 1-6; AgAp 1.54); and he models Ant on the much earlier Greek translation by the Seventy (Ant 1.10-13). Moreover, after discussing the roles of priests and prophets in maintaining the Judean ajnagrafaiv, he continues our passage by offering his own writings as an example of the Judeans' concern for historical truth: "But I myself have composed a truthful record" (ajnagrafhv; 1.47). Given the context, in which ajnagrafhv is a key word, it is hard to avoid the conclusion that he means to insinuate himself into the company of the ancient prophets. Yet for all his insinuation he does not actually use profhvth" of himself, much less of any contemporary (see below), and this must be significant in view of his interest in the subject of prophecy. Further, his whole point that the Judean records had been long completed, and that no one would countenance an addition, would be invalidated if he seriously placed his own work on the same level. And even if he had done so, few other Judeans would have granted his records any national authority, so there is no question of his works being "trusted" in this way (cf. 1.41). It seems, rather, that he simply takes advantage of his own rhetoric about the ancient Judeans to present himself as an embodiment of the best Judean traditions: if there were prophecy in his day, he would be a prophet, but of course there is not. As important as this chronological distinction, but seldom discussed, is the ethnic one, viz.: although profhvth" and cognates were well established in the Greek world, and although Josephus elsewhere uses generic vocabulary without hesitation, he follows the Bible in reserving this language for the ancient Judean tradition; Gentile and contemporary Judean phenomena are almost always designated by the mant- word group, along with terms like crhsmolovgo", provgnwsi" and proginwvskw. Thus the Gentile Balaam, the Witch of Endor, and Egyptian seers are not prophets for Josephus. Where an ancient Gentile seer is called a prophet and the term is not directly traceable to Josephus's source, his emphasis is on the ineffectiveness of these "prophets" (Ant 8.339; AgAp 1.312); this sounds like sarcasm. His refusal even to use the customary profhvth" for the renowned oracle at Delphi confirms his Tendenz. That he considers prophecy the preserve of the Judeans perhaps explains why he compares Judean prophets to Chaldeans in our passage. It has been suggested that Josephus's concern in AgAp 1 with the prophets' role as record-writers does not match his preoccupation with the predictive aspect of prophecy in Ant. This perceived contrast has even been used to argue that Josephus regarded "only one very limited type of prophecy" as having ceased. But on this score, too, his language is consistent. To be sure, in Ant he seems genuinely impressed that Judean prophets hundreds of years earlier had predicted the conquests of Alexander and the Romans, along with the fall of Jerusalem (10.142, 276; 11.1-3, 331-335). Innocent of Porphyry's later dating of Daniel, he celebrates this author's detailed description of future events (Ant 10.266). Still, even in Ant, where he tries so hard to impress his readers with the accuracy of these predictions, he adheres rigorously to his goal of providing a historical account. He omits all of the biblical wisdom material along with most of the major and minor prophets. From Isaiah and Ezekiel, and even from Jeremiah and Daniel_his chief exemplars of predictive prophecy, who play such a large role in his self-understanding_he excerpts the historical material almost exclusively. In his discussion of Daniel he pointedly excuses himself from elaborating on the predictions, in view of his task as a historian (Ant 10. 210). Similarly, in Ant 9.242 he manages to assert that Jonah's prophecies about Nineveh all came true while omitting the actual predictions, so as not to seem "irksome" to his readers. In both Ant and AgAp, therefore, while insisting that the ancient Judean authors were prophets and were thus granted knowledge of things beyond human ken, Josephus keeps a steady focus on Judean antiquity. There is no conflict between the statement in AgAp and his practice in Ant on this point. He does not subdivide the category "prophecy" for his readers, and efforts to find such distinctions within his own thought have so far proven fruitless. Also consistent with his language elsewhere is the term diadochv ("succession"), which he uses of kings, high priests, and prophets. Although the word group is completely absent from the OG parallels known to us, Josephus often describes a new king as a "successor," with noun or verb, in place of the OG expression "X ruled in place of Y." He has Hezekiah worrying about the prospects for "legitimate succession" (10.25). And he closes Ant with the notice that he has recorded without error "the succession of the kings" (20.261). In the same passage, he claims to have "tried to preserve the official record of the high priests who lived over 2000 years" (20.261). That claim is justified both by his summary excursus on the high-priestly succession from Aaron, at 20.224-251_this summary alone contains eight occurrences of the diadec- word group_and by his efforts within the narrative itself to spell out the succession of high priests from the time at which Moses entrusted the laws to them (4.304; 5.362; 10.152; 11.158). His concern with the high-priestly succession is reflected in our passage too, when he says that the Judean records contain the names of high priests, "sons from their fathers," over 2000 years (AgAp 1.36). The succession of prophets is not nearly as obvious in Ant as the other two. Perhaps it would have been awkward to establish a prophetic succession in the proper sense because the office of prophet was not hereditary. Still, Josephus chooses to summarize Moses' handing on of his role to Joshua (Num 27:15-23) by saying that he appointed Joshua "his successor (diavdoco"), both in the prophetic functions (ejpi; tai'" profhteivai") and as commander" (Ant 4.165). Deuteronomy's emphasis that there was never again a prophet like Moses (Deut 34:9-10) is also reflected by Josephus at 4.329, but he has said enough to hint at the beginnings of a prophetic "succession." This incidental evidence tends to confirm that, although Josephus nowhere articulates for his readers a symmetrical, trilateral succession from Moses' time, he did hold such a concept in his own thought. Of the three lines, only the high-priestly succession has continued uninterrupted through 2000 years to the present, and this guarantees the preservation and proper exposition of the records. That is the line that Josephus most typically celebrates. Royal and prophetic successions lapsed after the return from Exile, although a new royal line emerged with the later Hasmoneans. John Hyrcanus was singularly privileged to revive and embody all three functions within himself (War 1.68; Ant 13.299). Since Hyrcanus, once again, only the high-priestly succession has endured. Josephus, for his part, cannot claim to be a king, a high priest, or a prophet. But his own diadochv gives him a priestly and royal heritage (Life 1-6), while his accurate predictions and record-writing allow him to share a bit of the old prophetic aura. When Josephus writes of the failure of the prophetic succession, therefore, and of the consequent lack of recent Judean ajnagrafaiv, he seems to be tapping a deep and rich vein, which he only fleetingly exposes to our view. It is worth stressing in this regard that the usual English renderings of mh; genevsqai as "lack" or "failure" might give the false impression that Josephus rues the absence of contemporary prophets. In context, on the contrary, we can only conclude that he sees the long ages since official Judean record-writing as a great advantage over the Greeks, who were, as he says, born yesterday (AgAp 1.7). This work is entirely sanguine about Judean culture, and he conveys no sense here of prophet-deprivation. Equally absent is any feeling of "nostalgia" for the prophetic past, for such sentiments would run directly counter to his purpose, which is to boast that the Judean records have been so long in place. In a society in which "old is good," he is quite happy to present the Judean community as thriving on the interest of the ancient deposit, as it were, and he is scandalized by newcomers who claim prophetic gifts. II. The Integrity of Josephus's Biblical "Record" This broad coherence of outlook and language should not blind us to rhetorical flourishes in Josephus's programmatic statement about the national records. Most obviously: his insistence that no Judean has "dared" to alter the records (1.42) seems to conflict with his own practice in Ant 1-11, which is a thoroughly tendentious interpretation of the records rather than a translation. He omits a great deal, adds significant portions, and casts the whole history into a frame that suits his literary purposes. To be sure, his changes follow identifiable and fairly consistent criteria. He sets out to prove the nobility of Judean tradition, that it is highly philosophical, and that its God is active in human affairs, always rewarding virtue and punishing vice (Ant 1.6, 14, 20). Accordingly, he: omits episodes that might be used to support current anti-Judean literary slanders (e.g., about leprous origins or misanthropy); highlights and adds material that features Judean virtue; portrays Abraham, Moses, Solomon, Daniel and others as peerless philosophers; notes the rewards and punishments inevitably received by the virtuous and the impious; stresses divine providence and the fulfilment of prophecy; reflects editorially on universal philosophical issues; relentlessly moralizes; provides as much entertainment (tevryi") as possible; and accommodates the texts to his own times, priestly biases, and career. Although internally consistent, his biblical paraphrase does not consistently coincide with any known version of the text (e.g., MT, LXX A or B, Theodotion, Proto-Lucian, targum) or with rabbinic halakah or haggadah; he parallels all of these from time to time, but often goes his own way. In view of his pointed contrast between the unity of the official Judean records and the competing accounts of the Greeks (AgAp 1.37-38), it is noteworthy that he achieves this unity in practice only with great effort, by tacitly harmonizing biblical documents that do in fact overlap and compete, e.g.: Genesis-Numbers/Deuteronomy, Isaiah/Kings, Jeremiah/Kings, and Kings/Chronicles. While failing to tell the reader about this harmonizing activity, he parades examples of apparent conflict (e.g., between Jeremiah and Ezekiel; Ant 10.106-107, 141) and conspicuously resolves them, in order to reinforce the impression of harmony. He also introduces corroborating testimony from Greek and Oriental writers and quietly corrects the biblical sources (e.g., Ant 10.229; 11.106, 120) to agree with the external evidence. All of this makes it impossible to regard Josephus's Judean history as anything like a translation on the legendary model of the LXX (contra Ant 1.9-13); it is a tour de force in the service of his literary aims. Nevertheless, even here the conflict is not between his thematic statement in AgAp and his practice in Ant, but between his editorial statements and his practice in general. For his strongest assurances about his treatment of the text are those that introduce and punctuate the history itself. For example, he remarks on his carefully crafted summary of selected Mosaic laws, which we know to include non-pentateuchal items: All is here written as he [Moses] left it; nothing have we added for the sake of embellishment [ejpi; kallwpismw/'], nothing which has not been bequeathed by Moses. Our one innovation has been to classify the several subjects; for he left what he wrote in a scattered condition, just as he received each several instruction from God. (Ant 4.196-197, Thackeray; cf. 1.17; 10.218). Although such statements provide ready material for those who see Josephus as an incurable liar, that facile option is excluded by the immediate juxtaposition of these statements with the product itself, and by the sheer energy that was required for him to sustain his argument over such a lengthy history. Note that he makes the preceding statement in the anticipation that Judean readers might accuse him of departing from the texts (4.197). Since he does not so excuse the major alterations that we have noted, we can only conclude that he was largely insensitive to what we post-Enlightenment readers should expect in view of his promises. This insensitivity does not result from his being the "stupid copyist" of old source-critical imagination. His handling of the biblical material is broadly consistent with his handling of non-biblical material in the latter part of Ant and elsewhere. Rather, for Josephus as for every other writer of the Greco-Roman period, rhetorical strategies were so instinctive that he employed them everywhere, even while repudiating rhetoric as the downfall of the "Greeks." III. Scope and Arrangement of Josephus's Scripture We turn now to consider the extent to which Josephus's use of the Bible supports and elucidates his statement about the scope and order of the Judean records. The statement in AgAp 1 leads us to expect, first, that the reign of "the Artaxerxes who was king of the Persians after Xerxes" would mark the end of his Judean history, inasmuch as that history was meant to translate the official records (Ant 1.6, 17). We need first to identify the Artaxerxes in question. Many scholars assume with Barton that the remark "is probably to be understood as meaning that it [an official document] must not postdate the age of Ezra, because it was in that period that prophets existed." The problem is that although the biblical Ezra and 1 Esdras do name Artaxerxes as king of the Persians (e.g. Ezra 4.7, 23), Josephus systematically corrects this to "Cambyses," and he replaces the Artaxerxes of Nehemiah with "Xerxes" (Ant 11.21). The Artaxerxes who succeeded Xerxes appears only in Josephus's rendition of the Esther story (Ant 11.184, 296). This famous Gentile king impresses Josephus because, like the royal family of Adiabene whose story will be told at length near the end of Ant (20.17-96), he conspicuously favoured the Judeans: he married a Judean woman and with her protected the nation from its enemies (11.184-185). These emphases admirably serve the literary goals of Ant. Obviously, however, this story does not mark the end of Josephus's Ant. He does not inform the reader that he has now outrun the twenty-two volumes of AgAp 1, or that he has revised his original plan and will now continue the story up until the eve of the great revolt (20.223). One can easily speculate on his reasons for continuing the story. Thackeray suggested that Josephus wished to imitate the twenty volumes of the Roman Antiquities by Dionysius of Halicarnassus. I would propose further that he became so engrossed in his major themes_divine providence, reward and punishment_that he thought he could fruitfully rework the period already represented by the War along the same lines. His plentiful and in some cases new information about the LXX, the Hasmoneans, the now dastardly Herod, Gaius Caligula, and the rebels against Rome might have invited such a fresh portrayal. Does the continuation of the narrative past Artaxerxes betray Josephus's underlying belief in the open-endedness of scripture? Is the "twenty-two volumes" merely a rhetorical ploy in AgAp? Certainly Josephus is no more averse than any other ancient writer to creating claims about the national records ex nihilo to impress his readers. And he does say in the preface to Ant that "the things presented in the holy writings are innumerable, seeing that they embrace the history of five thousand years" (1.13), but 5000 years would require the 3000 years from Creation to Moses (AgAp 1.39), plus the 2000 years from Moses to his own time (Ant 20.261; AgAp 1.36). We have seen, however, that Josephus's language, even where he is not making a deliberate point of it, consistently places the ancient Judean prophets and their writings in a special category. Like most ancient writers, he was either poor or careless with numbers, so we cannot build a case on such a round figure as his 5000 years. Moreover, the critical reader detects a major editorial seam immediately after Josephus's leisurely retelling of the Esther story. Upon its conclusion (mid to late fifth century bce;11.296), he briefly summarizes the high-priestly succession for the following century (11.297-303), and then jumps to Alexander the Great (ca. 334 bce), who appeared "at about this time" (11.304). From now on, his account will become increasingly uneven as he tries to weave some very detailed sources (e.g., for Herod's life) together with large periods for which he lacks material. Rather than advertise the caesura, however, he is led by his present literary intentions to conceal it from the reader. So he uses his familiar catch-all bridge_ kata; tou'ton to;n kairovn_to cover the gap, as if nothing were really changing. That the seam is discernible in spite of Josephus's present literary intention alerts us to his knowledge that his more or less connected sources, the "records," are exhausted. In any case, Josephus includes much in the latter part of Ant, such as the deathof Gaius, that no one would trace to Judean scripture, so the continuation of Ant cannot be taken as proof of his wide-ranging scripture. Most important, he makes the statement about the twenty-two volumes when Ant has already been published, and in a work that consciously supplements the earlier narrative. He must suppose, then, that Ant does not obviously contradict his general statement in AgAp 1. His claim to speak for all Judeans when he specifies the figure of twenty-two volumes would be vulnerable to immediate disconfirmation if it were merely an ad hoc invention, and it happens to correspond closely to other roughly contemporary means of counting. We conclude that, although Josephus tries to patch over the end of his biblical sources in Ant, because he has now decided to extend the narrative, he is nonetheless aware that the national records have ended with the great and beneficent King Artaxerxes. Unfortunately, Josephus is so committed to the historian's task that he does not permit us to learn much about the internal arrangement of the Judean ajnagrafai. We have seen that he omits almost everything that cannot be accommodated to his continuous narrative and that what remains is treated with a heavy hand. His biblical source material for Ant 1-11 may be sketched roughly as follows: Antiquities Biblical Material Used Book 1 Genesis 1-35 Book 2 Genesis 36-48; Exodus 1-15 Book 3 Exodus 16-40; Exod/Lev/Num conflated for summary of laws Book 4 Numbers 14-36; Deuteronomy, conflated with Exod/Lev/Num Book 5 Joshua, Judges, Ruth, 1 Samuel 1-4 Book 6 1 Samuel 5-31 Book 7 2 Samuel 1-24; 1 Kings 1-2 conflated with 1 Chronicles 1-29; David is a singer and musician (305) Book 8 1 Kings 2-22 conflated with 2 Chronicles 1-18; Solomon composed 1005 volumes of odes/songs and 3000 volumes of parables (44) Book 9 2 Chronicles 19-31 conflated with 2 Kings 1-17, Jonah, Zechariah 14.5, and Nahum 2 Book 10 2 Kings 18-24 conflated with 2 Chronicles 32-36, Isaiah 38-39, Ezekiel 1, 12, and biographical passages (rearranged) from Jeremiah (lament also mentioned; 10.78); Daniel 1-8. Isaiah and Daniel wrote "books" (10.35, 267) Book 11 1 Esdras perhaps conflated with Ezra (generally preferring 1 Esdras); Nehemiah; Haggai and Zechariah mentioned; Esther (including "Greek additions" B-E) By the end of his biblical paraphrase, with the period of Ezra-Nehemiah and Esther, Josephus is relying heavily on Greek biblical texts, in spite of his promise to provide a translation of the Hebrew records (1.17). The nature of his Vorlage is a huge and still unsettled problem: some think that he used primarily Greek texts throughout; others find more evidence of a Semitic source in the early books, and so surmise that he only later opted for the Greek_perhaps through weariness of translation. We cannot tackle the source-critical problem here. We might wonder, though, how Josephus would reconcile his use of the fuller Greek texts of 1 Esdras and Esther with his clear statement about the limited number of official records. Perhaps he knew Hebrew editions of these texts that more closely approximated the Greek than do the ones known to us. In any case, he must have seen little material difference in using the Greek. That these distinctions did not trouble him offends our sense of precision, but we can hardly hold him responsible for that. In view of Josephus's thorough manipulation and reordering of his material, it would be unwise to make deductions about the arrangement of his sources, beyond the obvious: they start with the Pentateuch, after which he follows the best chronology he can. In the main, this agrees with the "trunk" of both the later LXX and the rabbinic canons. If Josephus supports the LXX against the rabbinic canon in placing Ruth between Judges and Samuel, including Daniel with Jeremiah and Ezekiel, and putting Esther after Ezra-Nehemiah, there were historical reasons for doing so: the books date themselves in this fashion. The same considerations lead him to agree with the later rabbinic canon against the LXX in placing Daniel before Ezra-Nehemiah and in omitting texts like ben Sira or 1 Enoch. In the Ant, Josephus's ordering criterion is chronological. No clear sections or divisions are established. To be sure, Moses and his laws have an axiomatic supremacy throughout Josephus's works (e.g., Ant 3.317-322). But notice the role that the laws play. They are the basis of the "constitution" by which Judean communities around the world govern themselves, and so are parallel to the national laws of other peoples. This priority of the laws will be self-evident to his readers: the laws of Lycurgus have a similar function for the Spartans_though Moses' laws are better than theirs (AgAp 2.225-226)_because laws are the foundations of nations. The Judean martyrs of our passage, who die for the sake of their laws (cf. AgAp 2.190, 219, 232, 272), are formally compared to citizens of various Greek nations, who fail to care so deeply about their own "constitutions" (1.44; 2.225-231). Thus Moses' laws do not appear as a peculiar feature of Judean life, much less as a section of Josephus's Bible. When he summarizes the laws (Ant 3.223-286; 4.196-302), he even apologizes for digressing from the narrative, making it clear that they constitute only a small portion of what Moses wrote (3.223; 4.196). Nor can Josephus's designation of Daniel as a "one of the greatest prophets" (Ant 10.266) be admitted as evidence of order within Josephus's Bible. Rather, this designation is consistent with his claim in AgAp 1.37 that all of the record-writers were prophets. Moses too was a prophet without equal (Ant 2.327; 3.60; 4.320, 329). There is no division of Prophets here. A possible exception, which might suggest that Josephus knew a section of the Bible called Prophets, comes in Ant 10.35. Having described there a prediction by Isaiah, Josephus editorializes: Now this prophet, who was acknowledged to be divine and his truth wonderful, confident that he had never spoken a false word, wrote up in volumes everything that he had prophesied, and left them to be recognized from the event. And not only this prophet, but also others, twelve in number, did the same thing, and everything that happens among us, whether good or evil, occurs in accordance with their prophecy. But we shall elaborate on each of these presently. Here we see a possible departure from Josephus's customary use of "prophets"_if, that is, these thirteen authors ought to be linked with the thirteen volumes of "the prophets after Moses" in AgAp 1.40. That premise is often accepted as self-evident. If it were correct, then Josephus's Bible would have a tripartite internal arrangement that he does not otherwise divulge: Moses, Prophets (in a narrower sense), and non-prophets. The narrative of Ant, to be sure, would suggest fifteen prophetic volumes_Joshua, Judges, Ruth, Samuel, Kings, Chronicles, Isaiah, Ezekiel, Jeremiah, Lamentations, Daniel, Ezra (or 1 Esdras), Nehemiah, Esther, and the twelve minor prophets as a group (since he mentions several of them)_but several combinations are plausible to reach the figure of thirteen. On the common identification of "Isaiah plus twelve" with the thirteen "prophets after Moses" of AgAp 1.40, we would have in Josephus an inkling of a tripartite canon, although the second and third parts would differ in content from those of rabbinic tradition. Roger Beckwith has proposed, however, that the twelve prophets in question are more likely the minor prophets, because (i) Josephus stresses their predictive writing, whereas the prophets from Joshua onward engaged mainly in history, and (ii) he is speaking of twelve authors rather than twelve books, whereas the thirteen volumes of AgAp 1.40 could only include the writings of the twelve minor prophets as a single text. If Beckwith's arguments are not conclusive, one might add that Josephus's introduction of these twelve near the end of the biblical paraphrase, as if they were a novelty, and his promise to discuss them in the sequel, would suit the hardly-mentioned minor prophets better than the copiously-discussed major authors. And Beckwith's proposal would maintain the consistency of Josephus's language: he does not say simply that there were twelve prophets in addition to Isaiah, but rather that twelve prophets did the same thing as Isaiah did, namely, they wrote down their predictions. Beckwith's theory is not without its own logical difficulties, for Josephus will go on to parade the fulfilled predictions of Jeremiah, Ezekiel, and Daniel (10.79, 107, 142, 269). This suggests that they should be included among the twelve who are like Isaiah. Perhaps, then, Josephus is not thinking of either the thirteen volumes of AgAp 1.40 minus Isaiah or the twelve minor prophets, but is counting only the predictive prophets. Then, however, he should be left with four or fifteen, depending on whether the minor prophets were reckoned as a group or individually. In so speculating we have long since left the field of interpretation. There are simply too many variables and insufficient evidence to reconstruct Josephus's personal knowledge of the Bible. It is not something that he chooses to divulge. What he tell his readers, consistently, is that all of the Judean records were written by prophets long ago. What he meant by his remark about twelve prophets like Isaiah, we presently lack the resources to determine. IV. The Bible in the Judean War Studies of Josephus's Bible generally ignore the War, for obvious reasons. His first composition, which begins long after the biblical period (according to his exilic dating of Daniel) and deals with national history leading up to the great revolt, has little cause to mention scripture. The only sustained reference to the Bible comes in a part of Josephus's speech before the walls of Jerusalem, in which he adduces examples from Israel's past in favour of the pacifist option (5.379-393). Moreover, an influential stream in Josephan scholarship has found in the War a work of either sheer betrayal and Roman propaganda or post-war politicking among the surviving Judean ‚lites. Scholars of this persuasion tend to argue that Josephus only becomes interested in the religious aspects of Judean culture in Ant. And even this interest is sometimes regarded as a pretence to gain influence with the Yavnean rabbis. Such views leave one hardly inclined to plumb the War for the possible impact of the Bible on Josephus's outlook. Seth Schwartz's recent Josephus and Judaean Politics extends this approach with an examination of allusions and references to the Bible in the War. He concludes: "there is little evidence that [when he wrote the War] he knew the biblical texts at all." Schwartz argues that Josephus's routine contradiction of the Bible in War_even where this does not appear to serve his rhetorical needs_coupled with the priestly bias of his biblical interpretation, suggest that he knew only selections from the Bible, and that he acquired these through the oral culture of the priesthood rather than through first-hand knowledge of the texts. Coming from another perspective altogether, Helgo Lindner has found in War's view of history clear traces of Jeremianic and Danielic influence. These influences appear not so much in direct reference (e.g., Daniel is not mentioned in War) as in Josephus's most basic views: that nations rise and fall under divine supervision, and that Rome is the current choice; God is also using Rome to punish Israel for its transgressions. These views, which constitute the fabric of War, are also important themes in Ant, where they are presented as the main trend of Judean tradition, drawn from Jeremiah and Daniel. As for his many and undeniable departures from the biblical text, I would note: (a) one might gain a similar impression from a proportionate sampling of Ant, where Josephus has biblical texts at hand, and (b) Schwartz may have underestimated the rhetorical force of some changes in War. In any case, Josephus publicly prides himself on the reputation that he had among his compatriots for traditional learning; accordingly, his education between War and Ant consisted mainly of Greek grammar, poetry, and prose (Ant 20.263). It is not clear how this claim to traditional knowledge would have helped him, in view of his many Judean antagonists (cf. Life 416, 424), if it did not reflect a measure of truth. Further, if this period was also devoted to acquiring a basic knowledge of the Bible in Hebrew and Greek, along with Alexandrian and other Judean literature, as Schwartz proposes, then Josephus was most fortunate_to discover so much in the Bible that happened to support his main emphases in War. The question of Josephus's use of the Bible in War is far from settled in the field as a whole; my inclinations will be clear from the foregoing summary. It will doubtless remain a controversial issue for some time to come because it is closely tied to more basic issues of Josephus's literary and intellectual integrity, self-understanding, and Judean identity. Conclusion Our conclusion may be briefly stated. Reading AgAp 1.37-43 in context shows how little amenable it is to our usual questions about the first-century "Jewish canon." In writing to persuade Greek speakers of the nobility and antiquity of Judean culture, Josephus simply means to stress the great age, small number, harmony, and prophetic authorship of the Judean records (ajnagrafaiv). He uses the generic language of his implied readers, not in-house Judean terminology. His actual use of the Bible in Ant agrees by and large with the summary statement in AgAp: he really did believe that prophets wrote the records in a bygone age. Although much is omitted from his biblical paraphrase, what we have represents the heart of both traditional Hebrew and Greek canons; he seems aware, though he does not advertise it, that books like 1 and 2 Maccabees are separate. This effort to engage Josephus's world of thought and language says nothing directly about the shape of the Jewish Bible(s) in the first century. Interpretation of each witness must precede a historical analysis that tries to explain the evidence of all witnesses; my goal has been to clarify what exactly Josephus says that will need to be explained by any broad historical hypothesis. But rather than feign ignorance of the ways in which Josephus is used in historical reconstructions, we may spell out in a preliminary way some direct implications of this study for the historical problem. 1. Josephus boasts about the age of the Judean records and does not convey any sense of either deprivation or nostalgia. In our passage he neither pines for a closed prophetic age nor hopes for its return. Thus he provides a very poor foil for claims that Jesus or Christianity fulfilled a Jewish dream in bringing the return of the "quenched spirit." (Whether such a foil might be found in other literature is another issue.) 2. He presents his positions as the common property of all Judeans_ women, children, prisoners of war_and he would presumably be vulnerable to refutation if he were making this up or presenting idiosyncratic views as common. It would accordingly be hard to argue from Josephus for an open canon or for one that was recently settled_at Yavneh, for example. Barton's argument strains Josephus's words beyond tolerance when he suggests: "In maintaining the small compass of Jewish Scripture he does not, as a matter of fact, say that no other book could conceivably found that would meet the criterion of prophetic authorship, only that no more than twenty-two have until now been found to do so." Those who are convinced by other evidence of the fluidity of scriptural boundaries in the first century do better, perhaps, to isolate Josephus as idiosyncratic or original in spite of his claim to speak on behalf of Judeans. 3. But that step, too, is hard to justify. Perhaps the most significant corollary of this study is its exclusion of the appeal to circumstantial evidence from the argument for an open canon. Rudolf Meyer, for example, argues for an open canon on the grounds that: (a) other sources such as ben Sira's "praise of the fathers" (44.1-50.24) and the DSS make no distinction between biblical, pre-Mosaic, and post-Artaxerxian texts (e.g., 1 Enoch and Tobit); (b) even within the DSS versions of biblical texts like the Psalms, there is much rearrangement and non-biblical material; and (c) and the texts of these documents often differ from the MT. This evident freedom to interpret, add to, and subtract from biblical texts leads Meyer to isolate Josephus's fixed notion of a canon as an inner-Pharisaic view that could only have gradually come to prominence with the emergence of the rabbinic coalition after 70; it cannot reflect a common first-century Jewish view. The problem with this reasoning will now be obvious, for all of the phenomena that Meyer finds in other sources are much more clearly and fully present in Josephus's own use of the Bible in Ant. As we have seen, he continues his narrative to the present, treating books like Pseudo-Aristeas and 1 Maccabees the same way that he treats biblical material. For the biblical period itself, he splices in all sorts of oral and written traditions. He quite thoroughly alters the texts to suit his own needs. And in numerous ways he evokes a prophetic aura for his own accounts. It is fair to say that if we lacked the AgAp, Josephus himself would offer the clearest case for an open canon. But we do have the AgAp, in which this same Josephus most emphatically, not to say matter-of-factly, insists that the Judean records have long since been completed in twenty-two volumes. Plainly, then, the circumstantial evidence of Josephus's own "Bible" in Ant does not mean what it might seem to mean at first: it does not, after all, imply an open canon. Indeed, once we know AgAp, we can go back to Ant and discover that Josephus really does believe that the succession of prophets has ceased, and we can discern a seam after the ajnagrafaiv have been exhausted. AgAp was written as a deliberate sequel to Ant, so it is unlikely that Josephus is aware of any substantial conflict between the two. This means that his willingness to alter the biblical text in manifold ways proves nothing about his formal view of canon. His example removes the force from appeals to circumstantial evidence as proof that the DSS authors or Philo or ben Sira had an open canon. 4. Josephus's remarks in AgAp 1.37-43 cannot be made, no matter how long we gaze at them, to correspond to standard enumerations of divisions within the first-century canon. His language is on a different plane. His most consistent ordering criterion is that of genre, viz.: laws, tradition, hymns, and advice. These genres do not correlate to "sections" of Josephus's Bible. They simply provide a means of elaborating for Gentile readers on the various kinds of material to be found among the twenty-two volumes of records. 5. Because these genres confound all other categories, the phrase "the laws and the related official records" (1.43) should no longer be taken to indicate a major canonical division. Josephus is working with a public world of discourse, according to which a nation's laws are self-evidently basic to its tradition. The phrase "and the related official records" would thus include everything else that Josephus has just mentioned, even Moses' non-legal writings ("tradition"). 6. What we have in Josephus is not inconsistent with the most traditional views of an early and tripartite canon_or for that matter, with modified tripartite and bipartite theories. It is just that he says nothing about any of this. One cannot say either that Josephus's canon differs from traditional canons, and so supports a theory of canonical or scriptural pluralism. We presently have no way of recovering the internal shape of his Bible from AgAp 1 or from his actual use of scripture in Ant 1-11. 7. Paradoxically, the little-noticed Ant 10.35 might say more about Josephus's Bible than the much-discussed AgAp 1.37-43_if its twelve-plus-one prophets correspond to the thirteen prophets of the latter passage. But the meaning of that remark is unclear, and Beckwith's connection of it with the minor prophets faces fewer obstacles than any other theory. Notes From sysadmin Sat Sep 17 12:29:07 1994 Received: from afep.yorku.ca (afep.yorku.ca [130.63.237.102]) by ccat.sas.upenn.edu (8.6.8/CCAT) with ESMTP id MAA42279 for ; Sat, 17 Sep 1994 12:29:06 -0400 Received: (from smason@localhost) by afep.yorku.ca (8.6.9/8.6.9) id MAA23668 for kraft@ccat.sas.upenn.edu; Sat, 17 Sep 1994 12:28:27 -0400 Date: Sat, 17 Sep 1994 12:28:27 -0400 From: "Steve Mason" Message-Id: <9409171228.ZM26738@afep.yorku.ca> In-Reply-To: Robert Kraft "Re: HBOT Footnotes" (Sep 17, 1:19am) References: <199409170519.BAA29133@ccat.sas.upenn.edu> X-Mailer: Z-Mail Lite (3.2.0 26may94) To: Robert Kraft Subject: Re: HBOT Footnotes Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Status: RO G'mornin', Bob: Herewith the notes for my paper. In ASCIfying them, of course, a lot of stuff disappeared (umlauted vowels, open and close quotes or other special characters [apostrophes too], underlining, and small caps). If you would like me to pop another diskette in the mail to you, with both the paper and the notes as real text at the end, I'd be happy to do it. I'm sure there's a convenient way to run global search and replace to preserve special characters, but I don't do it enough to make it second nature. So please let me know if the diskette would be worthwhile, and I'll send it in Monday's mail. Best wishes, Steve *********************************** Notes 1. In general, see H. St. J. Thackeray, Josephus: The Man and the Historian (New York: Ktav, 1967 [1928]) 75-99; S. Rappaport, Agada und Exegese bei Flavius Josephus (Vienna: A. Kohut Memorial Foundation, 1930); M. Braun, Griechischer Roman und hellenistische Geschichtsschreibung (Frankfurt: V. Klostermann, 1934); B. Heller, Grundzge der Aggada des Flavius Josephus MGWJ 80 (1936) 237-246; H. R. Moehring, Novelistic Elements in the Writings of Flavius Josephus (dissertation: University of Chicago, 1957); N. G. Cohen, Josephus and Scripture: Is Josephus Treatment of the Scriptural Narrative Similar Throughout the Antiquities I-XI? JQR 54 (1963-64) 311-332; H. W. Attridge, The Interpretation of Biblical History in the Antiquitates Judaicae of Flavius Josephus (Missoula: Scholars, 1976); and L. H. Feldman, Use, Authority and Exegesis of Mikra in the Writings of Josephus, in Mikra: Text, Translation, Reading and Interpretation of the Hebrew Bible in Ancient Judaism and Early Christianity, ed. M. J. Mulder and H. Sysling (Minneapolis: Fortress, 455-518). Noteworthy studies of particulars include H. Bloch, Die Quellen des Flavius Josephus in seiner Archologie (Leipzig: B. G. Teubner, 1879); M. Olitzki, Flavius Josephus und die Halacha (Berlin: H. Iskowski, 1885); G. Hlscher, Josephus PWRE 18 (1916) 1934-2000; A. Pelletier, Flavius Josephe, adapteur de la lettre dAriste (Paris: Klincksieck, 1962); J. Blenkinsopp, Prophecy and Priesthood in Josephus, JJS 25 (1974) 239-262; D. Goldenberg, The Halakha in Josephus and in Tannaitic Literature: A Comparative Study JQR 67 (1976) 30-43; E. C. Ulrich, The Qumran Text of Samuel and Josephus (Missoula: Scholars, 1978); T. W. Franxman, Genesis and the Jewish Antiquities of Flavius Josephus (Rome: Biblical Institute Press, 1979); H.-F. Weiss, Pharisismus und Hellenismus: zum Darstellung des Judentums im Geschichtswerk des jdischen Historikers Flavius Josephus, Orientalistische Literarzeitung 74 (1979) 421-433; D. Daube, Typology in Josephus, JJS 31 (1980) 18-36; D. Satran, Daniel: Seer, Prophet, Holy Man, in Ideal Figures in Ancient Judaism: Profiles and Paradigms, ed. John J. Collins and George W. E. Nickelsburg (Chico CA: Scholars, 1980), 33-48; B. H. Amaru, Land Theology in Josephus Jewish Antiquities, JQR 71 (1980-81) 201-229; J. L. Bailey, Josephus Portrayal of the Matriarchs, and G. Hata, The Story of Moses Interpreted Within the Context of Anti-Semitism, both in Josephus, Judaism, and Christianity, ed. L. H. Feldman and G. Hata (Detroit: Wayne State UP, 1987), 154-179 and 180-197; L. H. Feldman and G. Hata, Josephus, the Bible, and History (Detroit: Wayne State UP, 1989); C. A. Brown, No Longer Be Silent: First Century Jewish Portraits of Biblical Women (Louisville: Westminster/ John Knox, 1992); G. E. Sterling, Historiography and Self-Definition: Josephos, Luke-Acts and Apologetic Historiography (NovTSup 44; Leiden: E. J. Brill, 1992); and C. T. Begg, Josephus Account of the Early Divided Monarchy (AJ 8, 212-420): Rewriting the Bible (Leuven: University Press/Peeters, 1993). To these must be added the twenty or more studies of major figures in Josephuss biblical paraphrase by L. H. Feldman; references are most conveniently found in the bibliography to his recent Jew and Gentile in Antiquity: Attitudes and Interactions from Alexander to Justinian (Princeton: Princeton UP, 1993). This list is meant only to convey a sense of the scope of research on Josephuss Bible; it is far from exhaustive (see following note for fuller summaries). 2. Cf. H. W. Attridge, Josephus and his Works, in Jewish Writings of the Second Temple Period: Apocrypha, Pseudepigrapha, Qumran Sectarian Writings, Philo, Josephus, ed. Michael E. Stone (Philadelphia: Fortress, 1984) 210-227; Per Bilde, Flavius Josephus Between Jerusalem and Rome: His Life, his Works, and their Importance (Sheffield: JSOT, 1988) 80-104; L. H. Feldman, Josephus and Modern Scholarship (1937-1980) (Berlin: W. de Gruyter, 1984) 121-191; L. H. Feldman, A Selective Critical Bibliography of Josephus, in Josephus, the Bible, and History, ed. L. H. Feldman and G. Hata (Detroit: Wayne State UP, 1989) 352-366; Feldman: Mikra (1990); S. Mason, Josephus and the New Testament (Peabody MA: Hendrickson, 1992) 64-73. 3. The Canonization of Hebrew Scripture: the Talmudic and Midrashic Evidence (Hamden CT: Archon, 1976), 32. In his later article Josephus and the Canon of the Bible (in Josephus, the Bible, and History, ed. L. H. Feldman and G. Hata [Detroit: Wayne State UP, 1989] 55), Leiman qualifies this position in deference to R. Beckwiths proposal that the enumeration of AgAp1.37-41 is ad hoc and does not mirror a canonical arrangement. Still, Leiman seems to maintain (53-54), like Beckwith (below), that the statement reflects an underlying tripartite canon. See R. Beckwith, The Old Testament Canon of the New Testament Church (Grand Rapids: Eerdmans, 1985) 7980 [Beckwith speaks of the thirteen prophetical books and of the final four-book section], 121-127 [125: Josephuss canon has three sections, but he has transferred the narrative books in the Hagiographa to join those in the Prophets]. This was the standard interpretation of the passage, summarized without further ado in J. G. Mller, Des Flavius Josephus Schrift gegen den Apion (Hildesheim: Georg Olms, 1967 [1877]) 99-103. More recently: P. R. Ackroyd and C. F. Evans, The Cambridge History of the Bible, vol. 1: From the Beginnings to Jerome (Cambridge: Cambridge UP, 1970), 117, 136-137; ; E. E. Ellis, The Old Testament in Early Christianity (WUNT 54; Tbingen: J. C. B. Mohr (Paul Siebeck), 1991) 7. 4. Oracles of God: Perceptions of Ancient Prophecy in Israel after the Exile (New York: Oxford UP, 1986), 49. 5. Flavius Josephus als historischer Schriftsteller (Heidelberg: Lambert Schneider, 1978), 18. 6. Introduction to the Old Testament: From its Origins to the Closing of the Alexandrian Canon (third edition; Louisville: Westminster/John Knox, 1989). Though the English version is a translation (by John Bowden), and I lack access to the original, it seems that the major terms would survive the translation. 7. That this outline is obvious is confirmed by the close but independent correspondence between my synopsis and that of Bilde: Josephus 117-118. 8. All of the following translations, unless otherwise noted, are mine. They are deliberately wooden. 9. Although Josephus uses here the same word (ajnagrafaiv) that I elsewhere translate official records, he cannot mean the same thing by it, since the records that are the main subject of this passage were completed by the time of Artaxerxes. Rather, he must be referring to the archives from which he claims to have retrieved his own genealogy (Life 6with the verb ajnagravfw). It is not uncommon for him to play on different senses of the same word within a single passage; cf. S. Mason: Flavius Josephus on the Pharisees: A Composition-Critical Study (SPB 39; Leiden: E. J. Brill, 1991) 285 n. 22. On the 2000 years from Moses (and Aaron) to his own time cf. Ant 20.261. 10. War 2.411; 3.352; Ant 4.304; AgAp 2.185-186, 188. Cf. S. Mason, Priesthood in Josephus and the Pharisaic Revolution, JBL 107 (1988) 657-661. 11. Ackroyd and Evans: History (1970) 136; Leiman: Canonization (1976) 31-34; Beckwith: Old Testament (1985) 125; Barton: Oracles (1986) 38, 49 (in spite of many salutary qualifications); Ellis: Old Testament (1992) 9. 12. Cf. S. J. D. Cohen, History and Historiography in the Against Apion of Josephus in History and Theory 27 (1988): Essays in Jewish Historiography, ed. A. Rapoport-Albert, 1-11. 13. This word provides a fascinating example because Josephus does indeed use it to describe an internal Judean phenomenon: the special Pharisaic tradition (Ant 13.297, 408). But that exception proves the rule, for he must carefully explain to his readers what the word means in that context (Ant 13.297-298). Cf. S. Mason: Pharisees (1991) 233-235, 289-293. 14. Cf. Mason: Pharisees (1991) 64-66, 75-79, 89-96, 108-115. 15. Cf. Mason: Pharisees (1991) 96-106. 16. E.g., J. Jeremias, New Testament Theology I: The Proclamation of Jesus (New York: Charles Scribners Sons, 1971) 80 and n. 6; Ackroyd and Evans: History 1 (1970) 114-135; along with discussions in R. Meyer Profhvth" TDNT 6:817; D. E. Aune, Prophecy in Early Christianity and the Ancient Mediterranean World (Grand Rapids: Eerdmans, 1983) 103-106; Gray: Prophetic Figures (1992) 7-9. 17. Barton: Oracles (1986) 59. 18. Cf. Thackerays note b to vol. 1 of the Loeb Classical Library edition of War 1.18). 19. Cf. J. Blenkinsopp, Prophecy and Priesthood in Josephus JJS 25 (1974) 240-246; L. H. Feldman, Prophets and Prophecy in Josephus, JTS n.s. 41 (1990) 405; R. Gray, Prophetic Figures in Late Second Temple Jewish Palestine: the Evidence from Josephus (Oxford: Oxford UP, 1992) 23-26. 20. Cf. van Unnik: Schriftsteller (1978) 52f.; Blenkinsopp: Prophecy (1974) 239-262; and especially Feldman: Prophecy (1990) 392-393. 21. Gray: Prophetic Figures (1992) 20. 22. R. Mayer and C. Mller, JosephusPolitiker und Prophet in Josephus-Studien: Untersuchungen zu Josephus, dem antiken Judentum and dem Neuen Testament, Otto Michel zum 70. Geburtstag gewidmet, ed. O. Betz, M. Hengel, and K. Haacker (Gttingen: Vandenhoeck & Ruprecht, 1974) 283-284; Blenkinsopp: Prophecy 239-262; van Unnik: Schriftsteller (1978) 42; Feldman: Prophecy (1990) 421-422; Mason: Pharisees (1991) 270; most emphatically Gray: Prophetic Figures (1992) 35-79. 23. Correctly, Feldman: Prophecy (1990) 405; contra all others mentioned in the preceding note (Mason included). 24. Feldman (Prophecy [1990] 416-418), characteristically, discovers it. 25. Cf. also H. Krmer Profhvth" TDNT 6.783-796. 26. Ant 2.241; 4.104, 112, 157; 6.330-331, 338. 27. Cf. Feldman: Prophecy (1990) 417. 28. It might be worth comparing his remarks on exorcism as particularly prominent among the Judeans (Ant 8.44-49), though that power continues to the present. 29. Barton: Oracles (1986) 48. 30. Gray: Prophetic Figures (1992) 9. 31. Cf. Franxman: Genesis (1979) 7. 32. Feldman (Prophecy [1990] 394) plausibly suggests that indulgence in predictions might have provided material for ridicule. 33. Cf. Mason: Pharisees (1991) 235-239. 34. Compare the eschatological hope for a return of anointed prophet, ruler, and priest at Qumran (1QS 9.11, 4Q175). Eschatological urgency and the language of anointing are wanting in Josephus. 35. Contra Barton: Oracles (1986) 60, 115; and Gray: Prophetic Figures (1992) 8. 36. Cf. Cicero On the Laws 2.10.27; On the Nature of the Gods 3.1.5-4.10; Luke 5.39 and Feldman: Jew and Gentile (1993) 177-178. 37. Most conveniently summarized in Attridge: Works (1984) and Feldman: Mikra (1990). 38. S. J. D. Cohen (Josephus in Galilee and Rome: His Vita and his Development as a Historian [CSCT 8; Leiden: E. J. Brill, 1979] 27-29) suggests that this historiographical commonplace was not taken seriously (in our sense) by any of those who used it. Van Unnik (Schriftsteller [1978] 26-40) takes it to mean only that Josephus will not allow flattery or hatred to colour his work; he will recast his sources faithfully. Cf. now Feldman: Mikra (1990) 466-470. 39. The phrase is from R. Laqueur, Der jdische Historiker Flavius Josephus: ein biographischer Versuch auf neuer quellenkritische Grundlage (Darmstadt: Wissenschaftliche Buchgesllschaft, 1970 [1920]), caricaturing (not excessively) the image of Josephus created by such scholars as Bloch (Quellen [1879]) and Hlscher (Josephus [1916]). 40. Barton: Oracles (1986) 60; Ackroyd and Evans: History 1 (1970) 115-116 [who make a great deal of the connection with Ezra]; Soggin: Old Testament (1989) 13. 41. Thackeray: Josephus (1967 [1928]) 56-58. 42. Cf. perhaps War 6.109, 311; Ant 10.210. 43. Cf. Mason: New Testament (1992) 37, 209-210. 44. Cf. Ant 4.226; 5.352; 6.49, 213, 271, 292; 7.21; 9.239 et passim; D. R. Schwartz, KATA TOUTON TON KAIRON : Josephus Source on Agrippa II JQR 72 (1982) 241-268. 45. Cf. Leiman: Canonization (1976) 41-50 (on the Christian evidence) and 51-56 (on the rabbinic count); also Beckwith: Old Testament (1985) 118-127. Granted that the number twenty-two is mainly paralleled in Christian texts, the standard rabbinic figure of twenty-four is not much different. If Josephus knew both ways of counting, as did Jerome, he would have chosen the smaller number for his present argument. 46. Cf. also Appendix B in Seth Schwartz, Josephus and Judaean Politics (CSCT 18; Leiden: E. J. Brill, 1990) 225-226. 47. The problem is conveniently surveyed in Feldman: Mikra (1990) 455-466. 48. Cf. Mason: Pharisees (1991) 96-106. 49. Contra Marcus, note d ad loc. in the Loeb Classical Library edition. Barton: Oracles (1986) 36-38. Curiously, Barton elsewhere recognizes that Josephus considers all biblical writers prophets (p. 19); he seems to have it both ways. 50. So Leiman (Josephus [1989] 55): it can only be understood as a reference to the (total of) thirteen historical books by the prophets. Also Feldman (Prophecy [1990] 409 n. 83) and note c of the LCL translator, R. Marcus, to this passage: there seems to be no other explanation.... 51. The combinations Judges-Ruth; Jeremiah-Lamentations; and Ezra-Nehemiah are all attested in later Christian and Jewish lists. Cf. Leiman: Canonization (1976) 31-34; Beckwith: Old Testament (1985) 235-273. 52. Beckwith: Old Testament (1985) 99-100 n. 80. 53. This stream is represented by Laqueur: Josephus (1920); S. J. D. Cohen: Galilee and Rome (1979); and S. Schwartz: Judaean Politics (1990) among many others. 54. S. Schwartz: Judaean Politics (1990) 25. 55. Helgo Lindner, Die Geschichtsauffassung des Flavius Josephus im Bellum Judaicum (AGAJU 12; Leiden: E. J. Brill, 1972) 42-48. So also Gray: Prophetic Figures (1992) 70-79. 56. Cf. S. Mason, Josephus, Daniel, and the Flavian House, forthcoming in the still untitled proceedings of the International Colloquium on Flavius Josephus (San Miniato, November, 1992), ed. F. Parente and J. Sievers (Leiden: E. J. Brill). 57. See my review of Schwartz in IOUDAIOS REVIEW 2.008 (April, 1992), especially 3.7. This electronic review is available as SCHWARTZ MASON IOUD-REV from listserv@vm1.yorku.ca. 58. Barton: Oracles (1986) 59. 59. So A. C. Sundberg, The Old Testament: A Christian Canon CBQ 30 (1968) 143-155; and R. Meyer, Bemerkungen zur literargeschichtlichen Hintergrund der Kanontheorie des Josephus in Josephus-Studien: Untersuchungen zu Josephus, dem antiken Judentum and dem Neuen Testament, Otto Michel zum 70. Geburtstag gewidmet , ed. O. Betz, K. Haacker, and M. Hengel (Gttingen: Vandenhoeck & Ruprecht, 1974) 285-299. Barton (Oracles [1986] 59): in setting limits to the canon at all Josephus is out of step with his contemporaries. 60. Meyer: Kanontheorie (1974) 290, 298-299. 61. Contra Barton: Oracles (1986) 48; Feldman: Mikra (1990) 470.