What Josephus Says about the Essenes in his Judean War

Steve Mason
York University
Toronto, ON M3J 1P3

From: http://orion.mscc.huji.ac.il/orion/programs/Mason00-1.shtml

Part I

Part II

Probably the most famous text-artifact connection made by twentieth-century scholarship in our field has been the marriage of Qumraner with Essene--a figure known for two millennia in Greek and Latin texts. Parallels between Josephus's Essenes, particularly those of War 2.119-61, and the people of the Dead Sea Scrolls have been crucial in forging this happy union, along with the elder Pliny's notice about Esseni who lived in the Judean desert (N. H. 5.73). In the past decade an apparently growing cadre of scholars has objected to the Qumran-Essene theory from various perspectives, while the orthodox have responded by making the theory unfalsifiable, along such lines as these (Rajak 1994: 124):

Josephus was describing Jewish ascetics who were, at the very least, part of the same tradition as those who, over the generations, wrote the sectarian scrolls from Qumran, even if the correlation is not exact; ... the portrait [of the Essenes in Josephus] is based upon those who occupied, in one fashion or another, at one stage or another, the installations at Qumran.

With similar flexibility, the "Groningen hypothesis" postulates a rift within Essenism to explain the differences between Qumran and the classical sources (García Martínez 1996: lii-liii). And Hartmut Stegemann tries to reconcile all of the evidence by understanding Qumran as a mere outpost, a quiet research retreat (1992: 161), of a much larger Essene community--understood as the "main Jewish Union" of the time (165). This is a long way from the original and still popular Sukenik/de Vaux understanding of Qumran as the Essene home or base, and it shows the difficulty one faces in trying to engage the "Qumran-Essene theory": the theory will accommodate almost any problem by resorting to different times, places, and factions of an exquisitely malleable "Essene" phenomenon.

My principal aim in this essay is not to meddle in such a famous, though by no means blissfully secure, marriage. I want to ask, very simply, what Josephus says about the Essenes in the Judean War. I want to ask this chiefly because the question has not yet been answered, or even been unambiguously posed to my knowledge. We need to read Josephus's Essenes in the contexts he provides. As a corollary (only), I shall ask how any fair reading of Josephus's Essenes in the War bears on the Qumran-Essene question.

Issues and Stakes

Whatever sins Josephus may have committed during his lifetime, he has atoned for them many times over at the hands of his users, religious and academic. Whether he has been exploited for the purposes of Roman politics or Christian self-justification and anti-Judaism, for his potential as tour guide to Crusaders, pilgrims, and archaeologists, or for the marvellous and ubiquitous sources that scholars have liked to imagine him borrowing intact, he has until recently met with every fate except a patient reading.

This odd situation, which would not have been tolerated so long for even the most distasteful classical or biblical author, has begun to change in Josephan studies too. A burgeoning library of analyses, since the appearance of Rengstorf's Concordance (1973-83) and the ground-breaking work of André Pelletier (1962: 252), Helgo Lindner (1972: 40-45), Heinz Schreckenberg (1977: 173), Harold Attridge (1976: 17), and Louis Feldman (1998a and 1998b--representing three decades of his work), has begun to explore in depth the structures and themes of Josephus's thirty-volume corpus.

Yet the trend-setting scholarship on the Essenes, following upon the Qumran discoveries in the late 1940s and 1950s, was mainly completed under an old paradigm that did not encourage serious interpretation of Josephus. He was assumed either to be spinning some opportunistic and ephemeral yarn or, if one were lucky, to be plagiarizing a truly useful source. Either way, one was entitled to wrench his comments out of their present narrative contexts, considered a hopeless mish-mash at any rate, in order to salvage them for comparison with the Scrolls. Then, through an impressive transubstantiation, the "meaning" of Josephus became not what his writings might convey by themselves, but what this new mental amalgam of "Josephus-and-the-Scrolls" should mean. Posthumously and without his consent, he has achieved that extinction of the ego recommended by other oriental philosophers.

Moreover, this was a time--before Martin Hengel's Judaism and Hellenism (ET: 1974) had made its impact--in which scholars still tended to assume a sharp division between the categories Jewish (i.e., rabbinic) and Hellenistic. Josephus was regarded as a Hellenizer of some concealed, more genuinely Jewish conceptions.

Let me illustrate such readings of Josephus on the Essenes with a few salient examples.

Frank Moore Cross was one of the first to undertake this kind of interpretation by conflation, in his famous Haskell lectures of 1956-57, subsequently published as The Ancient Library of Qumran (1961). In his chapter on the "The Old Sources and the New Scrolls" (1961: 70-106), which will ostensibly explain how the new discoveries match what was known before (including Josephus), he entirely ignores the context provided by each classical source. He looks instead for one-to-one correspondences between them and the Scrolls, dismissing inconsistencies by a vaguely conceived "exterior view or Hellenizing tendency" of the Greek writers (1961: 70). Cross homogenizes the Essene and Qumran texts to claim that: they all stress the physical separation of Essenes from other Jews; they support the picture of Essenes as "an apocalyptic community" (1961: 78), though this is "partially obscured" by the Hellenizing Philo and Josephus (76); the fatalism that Josephus attributes to the Essenes should be understood by the two-spirits teaching of 1QS 3-4 (92); and the group's interest in predicting the future, in Josephus's description, "points directly to its apocalyptic structure" (93).

Qumran fever was leading a number of other scholars to use Josephus in much the same way at about the same time. Consider the relevant chapters in Dupont-Sommer's influential work, which first quote the classical sources without significant comment (1961: 21-38) and then intertwine Josephus and the others with the Scrolls under thematic rubrics (39-61). In the process, Josephus's narrative is lost. For example, his plain statements about reverence for the sun are flattened into a general modest piety, to accord with the Scrolls (1961: 60).

Under the heading, "The Reports of the Greek Historians," Matthew Black similarly atomizes the narratives of Philo and Josephus for comparison with the Scrolls under convenient thematic headings, ignoring even the most obvious differences (1961: 25-47).

Recent years have seen many products of this same legacy. The survey of Essenes in classical sources edited by Geza Vermes and Martin Goodman (1989) performs a great service by collecting the relevant texts in a fresh translation with notes. Yet their notes to the Essene passages all refer to parallels in other texts. Thus, the very form of this book implies that Josephus's own literary contexts are not critical for understanding what he says. Authors' statements appear as disembodied data. The same could be said of the German equivalent (A. Adam 1971).

Todd Beall does not disguise his goal. Assuming the Qumran-Essene identification as a "working hypothesis," he will illustrate Josephus's accounts of the Essenes by means of the DSS (1988: 3). One is free to choose one's aims, of course. Curiously, however, Beall expects that this procedure will provide "a better understanding of Josephus' description of the Essenes," which he (properly) distinguishes from understanding the historical Essenes (1988: 7). In practice, he moves phrase by phrase, interpreting Josephus's remarks with reference to everything but Josephus himself. He includes no discussion of Josephus's own literary contexts for the Essene passages, and only occasionally notes individual verbal parallels within Josephus (e.g., 1988: 86, 92-93). Josephus's references to marriage, women, organization, and oaths, all of which have important parallels in his own writings (below), are interpreted exclusively by other texts. When Beall comes to the Essenes' apparent sun worship (War 2.128, 148), he must conclude that the apparent meaning is unlikely because Josephus would not endorse such a practice--though Beall adduces no argument from Josephus--and because the Scrolls do not mention sun worship (1988: 52-54). Josephus must be referring to something milder, which the Scrolls would support.

In her essay on Josephus's Essenes, Tessa Rajak perceptively and promisingly remarks: "Apart from any agenda dictated by Qumran scholarship, ... there is also the challenge simply of understanding Josephus on his own terms" (1994: 143). Yet even she considers it necessary to predicate her study upon the Qumran-Essene identification (1994: 141-42), which then enables her to use Qumran literature "as a control on the process" of determining what Josephus was up to (1994: 145).

James VanderKam attempts to provide a balanced and reasonable picture of current research in the The Dead Sea Scrolls Today (1994). Although he largely succeeds, his crucial chapter on identifying the Qumraners with Essenes follows the pattern established above. When he meets inconsistencies between Josephus and the Scrolls, he tellingly remarks: "It is reasonable to interpret the evidence in such a way that the sources do not conflict" (1994: 89). I beg to differ. Although eventual historical reconstruction requires explaining all the sources by a single hypothesis, interpretation requires giving absolute priority to the claims of each individual text. To do otherwise begs the question.

Finally, several general studies of Judaism have adopted the same approach. E. P. Sanders fuses Josephus with Qumran so fully that he can speak of the DSS as "Essene literature" (1992: 342). He treats Josephus and the Scrolls as expressions of the same bifurcated Essene reality, namely: celibates at Qumran (described in 1QS, War 2.120) and married Essenes in the towns (described by CD, War 2.160). Because the DSS are primary sources and Josephus is secondary, on this hypothesis, Sanders can use the Scrolls to accuse Josephus of mistakes, as if he were somehow responsible for explaining the Qumran finds. So: "he did not describe precisely what we have in 1QS and CD combined. Thus on this point [married and unmarried Essenes] his information is both very good ... and inadequate (he does not mention the isolated community on the shores of the Dead Sea)" (1992: 344). Again: although Josephus got some things right, "When it comes to theology we find him a little less trustworthy. Certainly his description does not convey adequately the flavour of the Scrolls" (1992: 379). One finds the same conflation of texts in Lester Grabbe's valuable handbook, even though it is elsewhere concerned to interpret each piece of evidence in its discrete context before attempting historical reconstruction (1992: 2.494-95).

Rebecca Gray subtitles her study of Prophetic Figures "the evidence from Josephus," and insists on reading Josephus's reports in their full literary context (1993: 5). Yet her chapter on the Essenes opens by assuming the Qumran-Essene identification, which immediately leads her into historical rather than interpretive mode. Thus: "the Scrolls sometimes enable us to distinguish those statements in Josephus that refer to the celibate branch of the Essenes from those that refer to the marrying group" (1993: 81). Her analysis of Josephus's Essenes is governed by this "distinction that Josephus does not make" (1993: 81). She continues: "In other ways, too, the evidence from the Scrolls can be used to clarify and fill in the statements that Josephus makes about the Essenes" (1993: 81; emphasis added). So, she explains "the meaning" of the Josephan phrases "sacred books," "various purifications," and "sayings of prophets" (War 2.159) by reference to Qumran finds, not through any sustained discussion of Josephus's own language or larger context. She repeatedly insists, for example, that Josephus's statement about Essene prediction in War 2.159 refers "in the first instance, to the stricter [celibate] group of Essenes who lived at Qumran"--although Josephus says nothing about Qumran or a stricter group--and not to the Essenes of Josephus's narrative who predict the future (1993: 92, 95, 106, 110; emphasis added). In part, perhaps, Gray justifies this non-contextual analysis by the common refuge: "I consider it likely that the general account in War 2.119-61 is from a literary source" (1993: 82).

The problems with all such ostensible readings of Josephus will be obvious. First, they are not readings of Josephus; they are irredeemably contaminated by the intrusion of other texts. It is difficult to see the difference between reading Josephus in light of the DSS and "filling in" Matthew's birth narrative (Matthew 1-2) with material from Luke's (Luke 2), which no critical scholar would countenance. It does not matter for the purpose of interpretation whether one source was indeed primary and the other secondary. Reading any text requires a leap into that text's world of discourse, a willingness to formulate a meaning as its first readers did: in dialogue with that text and with extra-textual resources shared by author and readers.

Second, to the degree that an interpretation of Josephus's Essenes is logically preliminary to the Qumran-Essene hypothesis, any study that both argues the Qumran-Essene theory and begins by reading Josephus's Essenes in the light of Qumran assumes its conclusion. Such question-begging appears regularly in the literature mentioned above. One keeps expecting to find somewhere a reference back to a clean study, which would first seek to understand Josephus's Essenes in a Josephan context and only then explain how the Qumran-Essene hypothesis explains Josephus. One expects this because it is an indispensable condition of an acceptable historical hypothesis: the best hypothesis is quite simply the one that most adequately explains how the relevant evidence came into being. But one expects in vain. No such study exists.

A commendable start was made by André Paul (1992), though he focused only on a couple of issues in the Essene passage of the War, and others have made some initial moves in reading aspects of Josephus on the Essenes. Significantly, many of those who have taken Josephus most seriously doubt or reject the Qumran-Essene hypothesis (Ory 1975: 75-77; A. Baumgarten 1994: 169 n.1; Goodman 1995). So we still need to ask: "How do the Essenes function in Josephus's narratives?"

In his elaborate attempt to mesh Josephus, Philo, and Pliny with the Scrolls, Stegemann seems aware of the logical need for a clean study of Josephus on the Essenes. But he thinks that this need has already been met (1992: 108) by Beall and by Walter Bauer's entry on the Essenes in a Supplementband of 1924 to Pauly-Wissowa. Whereas Beall's limitations have been noted above, Bauer's essay raises another issue much discussed in connection with Josephus's Essenes.

For Bauer on the Essenes, as for Gustav Hölscher on Josephus in the same reference work (1916: 18.1937-2000), to study Josephus is to remember "die Gewohnheit des Josephos, auf Grund von Quellen zu schreiben" (1924: 404). Any perceived inconsistency in a text is attributed ohne weiteres to its underlying sources. In this case, Bauer concluded that War 2.119-61 mainly borrows a lost description of the Essenes by Philo. Bauer's study is fascinating in many ways: for its record of pre-Qumran efforts to link the Essenes with other known texts, for example, and for its understanding of Pliny's remark about the place of the Essenes (1924: 390). But it comes from a time in which scholars imagined that they could recover Josephus's sources more or less intact, because they estimated Josephus's own contribution so miserably. Rather than battling this spectre from generations long past, let us turn to its more recent manifestations.

Josephus's Sources on the Essenes

The most famous case for relieving Josephus of responsibility for the Essene passage in War 2.119-161 was made independently by Matthew Black (1956) and Morton Smith (1958), who proposed that the same source on the Essenes was taken over by Josephus in War (with only small adjustments) and by the early third-century Hippolytus in his Refutation of All Heresies. Although the older assumption had been that Hippolytus used Josephus, Black argued that significant non-Josephan features of Hippolytus could not easily be explained as the latter's elaborations of Josephus, and further that Hippolytus was more Jewish (i.e., closer to rabbinic themes) than Josephus. Smith argued that whereas Hippolytus normally quotes verbatim, he does not have such a relationship with Josephus's text, and there was no other evidence that he knew Josephus.

Their conclusion has since been overturned, however, by Christoph Burchard's fuller analysis of Hippolytus (1977). Burchard shows that this section of the Refutatio is replete with Hippolytus's own language and themes; other passages are plainly Christian (not more purely Jewish than Josephus, as was thought); and in some cases Hippolytus replicates characteristic Josephan phrases in Josephan style. Burchard concludes that War 2.119-161 is the main source for Hippolytus's description of the Pharisees, in which case there is no reason to posit a comprehensive Essene source behind the War. In any event, Smith himself later retracted his proposal under the influence of Shaye Cohen (Bergmeier 1993: 23 n.9).

More recently, Roland Bergmeier (1993) has managed to write an entire book on "the Essene-portrayals in Flavius Josephus" without ever asking how Josephus portrays the Essenes. He does not ask such elementary questions as: "What is this book by Josephus about? How do the Essene passages fit in this book? Why does Josephus describe them? What does his language mean in context?" Instead, he quickly consigns the main Essene passages to Josephus's sources, then spends the balance of his book ingeniously recreating the contours of those sources. He raises the reader's hopes when he observes that the kinds of study I have listed above allowed the DSS to intrude far too much in the interpretation of Josephus. But then, instead of turning to the interpretation of Josephus, he insists on understanding each of Josephus's putative sources in its own right (1993: 51-52). Josephus as an author remains unworthy of attention.

Significantly, Bergmeier embarked upon this study because he grew uneasy about the obvious differences between the DSS (assumed to be Essene documents) and the testimony of one who was supposed to be an eyewitness of the group: Josephus (1993: 9). He resolves this tension by displacing Josephus from the status of eyewitness, reverting entirely to Josephus's sources.

To understand how one could thus discount our only contemporary Judean text that names the Essenes in favour of texts that do not obviously identify themselves as Essene, one needs to recall the history of German Josephan scholarship. Like classical and biblical studies generally, Josephan studies went through a formative period--here, roughly 1870-1920--in which the apparent author was lost in the search for his sources. Josephus, for his part, seemed but a "stupid copyist" (Laqueur 1970 [1920]: viii) who plagiarized large hypothetical, anonymous, intermediate sources.

This approach melted away in cognate disciplines through the middle part of the twentieth century under the gaze of a more alert readership, and it should have done so in Josephan studies too. Richard Laqueur [1920] demonstrated that some of the major inconsistencies in Josephus--between War 1 and Ant. 14, between War 2-3 and Life--could not be traced to Josephus's sources. In the one case, it is a change of narrative colouring (Färbung), not of historical data; in the other, it is Josephus's own career that is differently retold, and thus not something source-dependent. Consequently, we must reckon with deliberate changes on Josephus's part.

In spite of Laqueur and the many studies since that have clarified the art with which Josephus typically uses sources (Feldman et al., above), an arbitrary and reflexive appeal to Josephus's sources as the solution to difficulties in the text is still made with disconcerting frequency.

Bergmeier's examination stakes out its argumentative ground within this same single-minded quest for Josephus's sources. Noting the two different forms of the name "Essene" in Joesphus, he contends that most of the )Essai~oj "anecdotes" have a decidedly non-Jewish colouring and therefore come from Nicolaus of Damascus (1993: 17-18, 21)--the deus ex machina of Josephan source criticism. The )Esshno/j passages and one )Essai~oj story come from three other sources now intermingled in Josephus: a Stoicizing "three-school" source in the doxographical tradition; a Hellenistic-Jewish source representing the themes and speech habits of Alexandrian Jewry, from which Philo and Josephus independently drew; and another source that understood the Essenes as Jewish Pythagoreans, which also influenced Pliny and Philo's description of the Therapeutae (1993: 48). Bergmeier would have us believe that in War 2.119-61 Josephus constantly moved back and forth, meticulously but slavishly combining these sources; that particular phrases, which sound more Stoic or Pythagorean or Alexandrian-Jewish, flag the transition to a new source (1993: e.g. 67, 92-93). As in days of yore, Josephus as author is of negligible concern.

It is of course Bergmeier's method, not his conclusion, that is the problem. No one doubts that Josephus used sources for most of his work, but the question is: How did he use them: as a responsible author or as a slavish compiler? Symptomatic of is Bergmeier's decontextualized interpretation is his reading of War 1.26, where Josephus promises "neither to obscure nor to supplement the things that have come to light." Bergmeier sees this as a promise to treat sources conservatively, even leaving doublets hanging (1993: 79-80)! In context, however, this is only part of Josephus's pledge to be strictly truthful (War 1.2-3, 6, 16); his very similar pledge in the Antiquities (Ant. 1.17; 4.196) obviously does not mean that he will leave his sources unedited.

Lacking the space here for a thorough review of Bergmeier's tour de force, I shall mention only three examples of premise and inference that are both featured in his argument and, as far as I can see, insupportable.

(a) The most important underlying issue for Bergmeier, to which he returns repeatedly, is the difference between )Essai~oj and )Esshno/j as a name for Essenes (1993: 13, 24). Without discussing alternative explanations, he concludes that this variation is a sign of different sources. He further claims that the n-form of the name is connected with the putative "three-school source" (1993: 56), though there is in fact no correlation between the form of the name and the "school" passages (below). He does not examine either Josephus's habits of writing or the general ancient context.

(b) When Bergmeier comes across Josephus's description of the first two Essene oaths--they show piety towards the Deity (eu)sebe/w to\ qei=on) and justice towards humanity (ta\ pro\j a/nqrw/pouj di/kaia fula/cein)--he works out at some length a strained parallel with Philo, and then early Christian literature, in order to help establish his alleged Alexandrian Hellenistic-Jewish source (1993: 36-37). But what if this complementary pair of virtues were characteristic of Josephus, turning up routinely from his portraits of the righteous Judean kings to his presentation of John the Baptist (below)? And what if this felicitous pair were found widely throughout the Greek moral philosophers from Plato onward (Plato, Philebus 39e 10; Euthyphro 12e 6; Gorgias 507b [ta\ di/kaia kai\ o(/sia]; Dionysius Halic., Rom. Ant. 8.2.2; 8.8.1; 8.28.3; 8.62.3; 9.44.8 etc.)? Josephus would then be, as he is, thoughtfully interacting with his environment rather than copying sources on the Essenes, stenographer-like.

(c) In his final chapter, Bergmeier identifies fifty-one hapax legomena in War 2.119-161 as indices of Josephus's use of sources in this passage (1993: 108-9). Although Bergmeier acknowledges that passages with distinctive themes, such as the geographical excursuses, necessarily have an unusually high number of unique words, for some reason he does not think that this principle explains the hapax legomena in War 2.119-166.

Yet if we consider that the War as a whole is a narrative of political and military history, whereas the Essene passage is a lengthy excursus on a tightly-knit philosophical school, we might expect that it would contain many unusual words. And that is just what we find. The unique terms include those describing: sectarian boundary markers (admit, eject, outsiders, condescending to eat [grass, if expelled], neophytes); special philosophical-school traits (dress, being trained, wear white, exchange, return item, school member, refectory, pray beforehand, wrap around [mantle], theologize, wise saying, inculcate); unique implements of this school (small hatchet, loincloth, mattock), and their peculiar ethics and practices, described in a deliberate attempt at sublimity (sobriety, masters [of temper], servants [of peace], medicinal roots, extreme long life). Obviously, Josephus has little occasion to use most of these words elsewhere in a political-military history.

Moreover, arguing from hapax legomena raises the obvious logical problem: What about dis and tris legomena? If the unusual appearance or concentration of vocabulary in one passage suggests that it comes directly from a source, what shall we say when a word appears in the Essene passage and then only once or twice elsewhere in Josephus? Obvious examples: u(peropsi/a (War 2.120; 3.320); e)ntupo/w (War 2.120; Ant. 12.68, 72); katafronhtai/ (War 2.122, 151; Ant. 6.347); a)lei/fw (War 2.123; 5.565; Ant. 6.165); e)/ktopoj (War 2.136; 4.319); and kaqekt- words (-oj, War 2.12; -eon, War 5.20; -ikoj, War 2.135). Shall we say that some source is directly responsible for both the Essene passage and the other occurrence or two of the rare word? Given that the War has an unusually artful style, it makes no sense to conclude that rare words indicate foreign sources.

The problems with using hapax legomena to determine authorship are well known, and they account in part for the rise of stylometric analysis as the preferred tool. In stylometric analysis, one examines the frequency of such incidental function words as particles and conjunctions, against both control texts within the same author and substantial blocks from other authors. Because of natural variations in an author's writing patterns, the critic requires massive cumulative odds against mere coincidence (say, more than 1,000,000:1) to challenge authorship of a given text or section (Williams 1992: 1-22). As it happens, David S. Williams has performed such an analysis on War 2.119-61 and found it thoroughly consistent with Josephus's authorial habits (not to be equated with his deliberate "style") across the broad swathe of his corpus, and in marked contrast to the habits of several contemporary authors, including Nicoalus (Williams 1994). I am not here endorsing stylometry in every respect: I simply note that it is a more refined tool than listing hapax legomena, and that it supports the otherwise probable conclusion that Josephus wrote War 2.119-161, whether or not he also used sources.

No one doubts that thousands of inconsistencies remain in Josephus, but these have many causes. Most prominent is Josephus's rhetorical artifice, whether merely playful or politically motivated, of the kind that allows him to retell his own life story in quite different terms (in War and Life) or dramatically change the motives of his characters from War to Antiquities. Second comes sheer carelessness, of the sort that plagues anyone who writes thirty volumes without benefit of proof-reading tools. (Who will cast the first stone?) Then there is a whole raft of possibilities involving the intrusion of multiple hands, which have been argued successfully for at least a few cases, namely: vestiges of undigested sources; influence from amanuenses and literary synergoi; sporadic imitation of great authors; traces of different editions; and corruption of our manuscripts (the earliest of which date from the ninth century).

The sensible procedure, surely, is first to try to read Josephus coherently. When significant inconsistencies persist, one is entitled to explore the possible causes and to argue for one of them: by showing that it explains the totality of evidence better than the others. One may not, however, simply move from the observation of inconsistency (and that without attempting to read in context) to the conclusion that some source is responsible.

Yet that is what Bergmeier does (e.g., 1993: 5, 17-18, 21-22, 41, 58-59). For example, following Bauer, he imagines that the repetition of "Greeks" in War 2.155 and 2.156, as of Ioudaios within 2.119, constitute doublets, signifying a change of hand (1993: 63). But the passages make perfectly good sense as they stand. Or, finding a difference between Josephus's claims about the unity of Jewish belief in Apion 2.179-81 and the same author's descriptions of differences among the three schools in War and Antiquities, he concludes that the school passages must come from a different source (1993: 57-58). Again, noting that Josephus uses the three "schools" as a set piece, whereas the fourth "school" of the rebels fits awkwardly with the other three, Bergmeier confirms that the three schools must come from a fixed source, which Josephus massages to include the fourth school (1993: 59). But there are simpler explanations, and there are serious problems with such hyper-sensitive distinctions, especially if one were to force them through the entire Josephan corpus. The effort to distinguish Stoic from Pythagorean sources is altogether too knowing for the first-century situation, in which eclecticism was the norm.

Such source hypotheses invariably defer to arbitrary criteria. How is it that the critic can see a difference between the three schools and the fourth, to posit a "three-school source", but not see major differences within the three-school passages--e.g., that one has the Pharisees and Sadducees as polar opposites (War 2.162-66) whereas another puts Essenes and Sadducees on the poles, leaving the Pharisees in between (Ant. 13.171-73)? How can the contrast between an ideal of unity (Apion 2) and a narrative of philosophical disagreement (War 2) be so telling in this case, but not elsewhere in Josephus? Such an interplay between the ideal and the real, for example, characterizes the Antiquities as a whole (contrast the prologue and summaries of the laws in vols. 3-4 with the narrative). How can certain "Greek" conceptions in the Essene passage be attributed to Pythagorean sources when similar ideas appear elsewhere in Josephus, even in the speeches he creates--and indeed motifs from Greek literature and theatre are generously dispersed in this book (see below and Chapman 1998: 1-21)? The tools available to the source critic are not up to the task: they do not work consistently over the entire range of evidence.

A more general problem is this: to the extent that Bergmeier envisions source hypotheses as a solution to perceived inconsistencies in Josephus (1993: 5, 59, 63, 67), he is caught in a logical bind, for invoking sources does nothing to alleviate that problem. Unless, with Hölscher (1916), one conjures up mammoth intermediate anonymous sources to take the blame for entire narratives (but Bergmeier allows that Josephus had some editorial input), Josephus remains under the same cloud of clumsiness as if he had composed the text himself. If an inconsistency is so glaring that it requires a source hypothesis, then it is still a problem for understanding Josephus's method as redactor. To put it the other way: if he could edit a significant problem into existence, then he could also write it into existence--especially if we envision re-writing.

Rather than debating the merits of source criticism further, I turn to the more productive project of trying to read Josephus's Essenes in a coherent way, as part of his narrative.

The Essenes of Josephus: Overview and Correlations

Let us begin our study proper with a synopsis of references to Essenes in Josephus.

This synopsis immediately suggests several points with respect to the names that Josephus uses for the Essenes.

By way of context, we need to recall that group names were extremely fluid in ancient Greek. Also in English we have different ways of rendering such names. For many famous groups (Parisians, Bostonians, Torontonians; Londoners, Berliners, New Yorkers; Vancouverites) the appropriate possessive is established by convention: one simply does not say Londonian. But what should we call someone from Stoney Creek or Sydney?

In ancient Greek there were at least three common possessive forms of names: -aios ( (Rwmai=oj, )Ioudai=oj, )Idoumai=oj, Xaldai=oj, Aqhnai=oj), -?nos (Damaskhno/j, Filadelfhno/j, Pergamhno/j), and -it?s--a form of nomen agentis (Perai5thj, Galaadi/thj). Note that between Greek and Latin even such a famous possessive as "Roman" took two different forms: (Rwmai=oj or Romanus; cf. Latin Athenensis for Greek )Aqhnai=oj. In Greek as in English, some names were fixed by convention, but others had to be developed for each case. Readers of the gospels will recall that "Nazarene" appears in two forms: Nazwrai=oj in Matthew, Nazarhno/j in Mark, but both forms in Luke-Acts (Luke 4:34; 18:37; 24:19; Acts 2:22; 3:6 et passim).

In the writings of Josephus, similarly, we have both Pa/rqoj (e.g., 13.385, 18.96, 317, 325, 340, 355) and Parquai=oj (e.g., 13.384; 18.98, 313, 318, 334, 339, 348) for Parthian, both Perai=oj (War 2.59; Ant. 17.276) and Perai5thj (War 2.520, 566; 3.11) for Perean, and both Galaadi/thj (War 1.89; Ant. 7.230, 232, 272, 387) and Galadhno/j (Ant. 4.173; 5.254; 6.71, 72, 73) for Gileadite. In the Parthian history of Ant. 18, in particular (note the references above), it seems that Josephus deliberately varies the frequently occurring name: the two forms are almost consistently alternated.

Our first conclusion, then, is that Josephus's variation between )Essai~oj and )Esshno/j conforms with his usual style. He places the two most common Greek possessive endings on the essa- root. Such variation of names is entirely compatible with his demonstrable tendencies and does not by itself suggest sources.

Indeed, second, the synopsis above shows that any temptation to attribute these names to discrete sources is ill-advised. It is not only the three "school passages," which have often been attributed to Josephus's sources, that speak of )Esshnoi/. Josephus chooses the same form in passages that are undeniably his: in the editorial comments of Ant. 13.298 (referring back to his War) and in Life 10 (also referring back to the War), as also in his reference to the "Gate of the )Esshnoi/" in Jerusalem (War 5.145). Josephus himself thus uses both forms. As we shall see, he seems entirely conscious of the variation.

Third, although there is no correlation between either form of the name and a source, there is a correlation between each term and grammatical number. All thirteen undisputed occurrences of )Esshno/j are in the plural, though some weaker manuscript evidence also supports )Esshno/j at Ant. 13.311. Of the six occurrences of )Essai~oj, conversely, five are in the singular and refer to named individuals. Josephus prefers to write )Essai~oj of an individual but )Esshnoi/ of the group. Why?

We have a major clue, fourth, in Ant. 15.371-378, the only passage in which Josephus uses )Essai~oj in the plural. We need to pay careful attention to his language there, because he seems to explain what he is doing. After describing how Herod excused the Pharisees from taking an oath of loyalty, he continues:

Also excused from this obligation were those called among us )Essai~oi (oi( par' h(mi=n )Essai=oi kalou/menoi). Now this is a ge/noj that adheres to a way of life laid down among the Greeks by Pythagoras. Although, then, I shall speak about these men more clearly elsewhere, it is fitting to discuss the reason why Herod honoured the )Esshnoi/ thus.... (Ant. 15.371-72). [So also )Esshnoi/ in the balance of the passage: 15.373, 378.]

I submit that the sense of this passage is fairly clear. Josephus has already mentioned the )Esshnoi/ a couple of times in the Antiquities (13.171-73, 298). Now he chooses to note that the group members are actually called )Essai~oi among Judeans, then immediately returns to the established )Esshnoi/ for the remainder of the passage. If Judeans normally call them )Essai~oi, but he uses )Esshnoi/ for his Greek and Roman readers, we may assume that he knew the n-form to be more familiar to his readers.

As it happens, we have confirmation of this understanding in the literature outside Josephus. Philo, a Ioudaios, consistently calls the group )Essai~oi (Free 75-91; Apol. 1-18). The Roman writers Pliny (N.H. 5.73) and Dio of/ Prusa (apud Synesius Dio 3.2), by contrast, use the n-form Esseni or )Esshnoi/. The simple conclusion is that Josephus, with typical sensitivity to his readers, adapted his language to suit them: while noting the Greek spelling that Jews preferred, he nevertheless used the familiar plural for his readers. Why, then, did he revert to )Essai~oj for individuals? We can only surmise that, although he willingly capitulated to the familiar plural, when he came to describe individuals who were not otherwise known, there was nothing to be lost in using the more authentic-sounding )Essai~oj. There was no particular reason to call each of these men an )Esshno/j, which had no established euphony for his readers.

Although Josephus discusses this linguistic principle only in Ant. 15, it is easy to imagine that he unconsciously followed it when he switched from the singular )Essai~oj in War 2.113 to the plural )Esshnoi/ in War 2.119. It is certainly easier to imagine this than to suppose that he was following two different sources within Ant. 15.371-78--where he also speaks about his plans to discuss the Essenes more fully in the future (cf. Ant. 18.18-22)--and two different sources in War 2.113-19.

This understanding of Josephus admittedly requires his awareness that the name of the Essenes was already known to his gentile readers. Since the Qumran-Essene hypothesis has habituated us to viewing the Essenes as a small and isolated community (Stegemann 1992, in critique), this possibility might seem odd. But all of the Essene evidence (i.e., in Greek and Latin texts) points in this direction: the Essenes are the only one of the three Josephan schools mentioned by Philo--at length and repeatedly--and by non-Christian Greek and Roman authors (Pliny and Dio). Josephus agrees with Philo in describing them as widely dispersed throughout Judea. Since Philo and Josephus both choose to make them the shining embodiment of Judean virtue, these authors might well have supposed that at least the name of this group was already known amongst educated Greeks and Romans.

In at least one case, finally, there is good reason to believe that Josephus understands )Essai~oj as an ethnicon, designating someone from Essa. He mentions a place called Essa in the Transjordan (Ant. 13.393 = Gerasa in War 1.104), and a person from Essa would most naturally be called an )Essai~oj. Further, in War 2.567 we are introduced to the commanders chosen for the revolt: Niger the Perean, John the )Essai~oj, Josephus, and others. In War 3.11, similarly, we are told of Niger o( Perai5thj, Silas o( Babulw/nioj, and John o( )Essai~oj. These immediate contexts, taken with the fact that John is never credited with any of the traits otherwise mentioned for Essenes, provide prima facie support for Schalit's proposal (1968: 46 s.v.) that in John's case )Essai~oj means "of Essa."

Josephus's Outlook: in General and in the Judean War

To understand how Josephus uses the Essenes in his narratives, we need to maintain some dialectic between the whole (Why is he writing? What are his narratives about?) and the parts (How do the Essenes contribute to each particular narrative?). Such questions are too large to be treated fully here, but something must be said. I have isolated five propositions about Josephus as an author that I consider both fundamental and easily demonstrable. The first four concern Josephus in general; the fourth focuses on the War. All of his writings presuppose a primarily gentile audience in Rome that is keen to learn about Judean culture.

(a) Josephus postures as a leading aristocrat in a world-wide system of aristocracies, expertly knowledgeable about his nation's constitution and its pa/tria e)/qh or mos maiorum. Directly engaging the long-standing discussion of constitutions among his counterparts in Rome, he wishes to show that the Judean code is the most ancient, the purest, and the noblest in existence. He details its perfect balance of precept and practice, its accessibility to all, and its combination of inexorable justice with humanity (Rajak 1998). In the context of the Flavian house's move toward ever-increasing autocracy, Josephus presents the Judean constitution as one that emphatically rejects monarchy in favour of priestly aristokratia (Ant. 4.223; 5.15, 43, 55, 135; 6.36; 11.111; 12.138, 142). His own ge/noj, paidei/a, and bi/oj provide a brilliant example of the Judean aristocrat in action (Life: Mason 1998a, 2001).

(b) Just as in Rome the republican senatorial movement had some philosophical strings attached, so Josephus's Judean aristocrats view their constitution as the embodiment of the very laws of nature. Many of the nation's founders and great figures were also peerless philosophers (e.g., Abraham, Moses, Solomon, Daniel). Their constitution discusses such basic issues as fate, free will, and the nature of the soul; Judean culture even has its own philosophical schools, which debate these matters. Judeans as a nation live philosophically, which is to say simply and virtuously, despising even death in their determination to follow their excellent laws. They realize in their daily lives the highest aspirations of all peoples. Josephus's philosophical discussions intersect neatly with those current among the popular and moral philosophers of the late first century (Mason 2000).

(c) In the standard fashion of an aristocrat, Josephus views the Judean populace as a rather gullible mass (plh=qoj), needing the direction of the historic ruling class, and all too vulnerable to the pitches of demagogues, who promise them material, economic, and spiritual salvation. The War charts and laments the proper authorities' loss of control to self-seeking would-be tyrants (note close parallels in Roman history from the social revolution through the civil wars of the first century BCE to the alleged demagoguery of emperors such as Gaius and Nero). Josephus despises those who usurp legitimate constitutional rule. In Ant. 4.14-20, the Catiline-like popularis Korah is the paradigm of a demagogue: a nobleman who played to the crowds, pretending to have their interests at heart but really seeking his own aggrandizement (Mason 1998b, 2000). Josephus excoriates such figures, whether militants or would-be Messiahs--i.e., anyone who favours upheaval of any kind.

(d) According to Josephus, the Judeans at their best, and certainly their legitimate leaders, are exemplary world citizens. Because they recognize that their God controls all history--causing now one power, now another, to rise or fall--they easily cooperate with any current power. Josephus has a broad kind of eschatology, but it is world-affirming or healthy-minded, in William James's classic terms (James 1928: 78-123). He already sees the widespread growth of the Judean nation in numbers and in fame, and he looks forward to the day when his nation will have its turn as a world power (Lindner 1972; Mason 1994). This view, similar to Philo's in Rewards and Punishments, lacks any apocalyptic urgency (Hecht 1987). Indeed, Josephus appears markedly anti-apocalyptic, fundamentally opposed to sudden change and messianic personalities. In his biblical interpretation, Josephus decidedly favours the plain sense (peshat): the law's prescriptions and its moral implications. He occasionally ventures into allegory in order to make universal connections with the Judean laws. In his considerable body of biblical interpretation (e.g., Ant. 1-11), there is nothing that qualifies as pesher or Endzeit-driven, though he does look forward to further fulfillments of scripture in a few places. (e)In spite of a long scholarly tradition that has trained us to think of it as Roman propaganda, the War presents itself as a challenge to the various pro-Roman and anti-Judean accounts already in circulation (War 1.1-8). Since we do not possess those accounts, we do not know what was in them. But a "mirror-reading" of the War, complemented by clues from later Roman texts that mention the Judean war, produces a coherent picture: Greek and Roman authors had claimed that the revolt was the expression of an allegedly restive Judean national character, schooled by an anti-social God and set of laws. The Judeans were both impious toward the gods and lacking in appropriate respect for other people. The Roman victory was clear proof of Rome's virtue and fortuna.

Josephus's War attempts to show that, on the contrary, Judeans have an old and noble constitution, that their heroes once ruled a vigorous and independent state with divine aid, and that their most famous king (Herod) was a great friend and ally of Rome. Most important, the recent revolt was not an expression of the national character, but a complete aberration that resulted from the legitimate rulers' tragic loss of control to a gallery of self-serving demagogues--turannoi/, lh|stai/, zhlwtai/, and sicarii--who ultimately elicited God's punishment. Far from abandoning his people, the Judean God actually used the Romans as His instruments, to purge His Temple of the pollution. The Judeans are in fact the most pious of all peoples toward the Deity, and the most concerned about justice toward their fellows. Life was admittedly very difficult in Judea under a series of abominable Roman governors, but even still the Judean leaders would have kept things under control if they had managed to maintain the faith of the populace. Those who fomented revolt, however, have now been duly punished. Therefore, post-war antipathy toward Judeans should cease. On all of this, see Bilde (1979), Rajak (1983), Mason (1992: 53-84).

This general context is, I am arguing, crucial for understanding Josephus's portrait of the Essenes in the War. We have two main options in reading the main Essene passage: to regard it as a coherent part of this work or to choose incoherence. The second option would hold that, lacking sources for the period 6 to 66 CE, Josephus chose to describe the Essenes and the other schools as a sort of space-filler, borrowing a disjointed account from other sources. Although, as we have seen, scholarship on the Essenes has tended strongly toward the latter option, it has precious little to commend it. If Josephus artfully uses the Hasmonean and Herodian histories (War 1.31-2.110) to develop the basic themes of his main story, why should we not look for similar effects in the Essene passage, which forms the principal part of the bridge between King Herod and his own time? The proof is in the language and tone of his presentation.