The overall purpose of this paper is to suggest that the various factors involved in an adequate description of the relationships, say, between the Temple Scroll and the Torah or between the Pesharim and the Prophets, are foundation stones which should be used for building an understanding of how any composition in the Second Temple period moved from authority to canon. It may be inappropriate at the end of the symposium to say that the title of our three days work is off the mark, but to speak of the reworking of the Bible at Qumran, to my mind, puts too much emphasis on the closed demarcation of one set of writings and the entirely secondary or derivative nature of the other. It will come as no surprise to learn that, for the late second temple period, I side with those who would rather speak of authoritative scriptures than books of the Bible. This distinction to some may sound like splitting hairs, but it goes some way to taking into account the diversity of Second Temple textual and artefactual evidence for each scriptural book. Nevertheless I am inclined to think that even this distinction does not go far enough for the obvious reason that it still gives an authoritative primacy to what later became canonical and the thrust of my presentation is that some, if not all, texts moved authority to canon in the second half of the Second Temple period not least because such texts attracted and provoked the very reworkings with which we have been concerned.

A wealth of insightful material has been written on exegetical traditions of all kinds, and increasingly such work has involved detailed consideration of the so-called reworked or rewritten Bible compositions. However, with some noteworthy exceptions, little of this scholarly description of exegetical works gives much attention to the status or authority that works containing exegetical traditions may have had. My argument from the reworked scriptural materials found in the Qumran library is that they show in a significant way various features of how the transformation of authoritative scriptures into canonical Bibles took place. From the post-canonical perspective these reworked compositions seem to fall into two groups, revisions of biblical books and more through-going rewritings of such books. In this essay all these materials will be taken together for the sake of the discussion; 4QpaleoExodm, the Books of Chronicles, the varieties of collections of Psalms, the Book of Jubilees, and the Temple Scroll all share various features. Considering all these compositions together shows that the transformation from authority to canon took 400 years and had certain very intriguing characteristics. In the rest of this presentation I wish to draw out in a preliminary way some of the implications of the Qumran evidence for the wider issue of the making and handling of sacred texts.

2. Implicit and explicit politics
All declarations of authority concerning texts have a political dimension and it seems that any acceptance of such declarations are political affirmations as well.

It is widely thought that every book of the Bible has been found at Qumran except Esther. The absence of Esther has been accounted for in various ways, not least being the way in which the many and various calendrical compositions from Qumran fail to make any mention of Purim, just as they fail to mention Hanukkah. In fact, there are good reasons to suppose that Esther in some form was known at Qumran, since some of its distinctive phraseology may be discernible even in some of the more narrowly sectarian compositions and the copies of so-called proto-Esther indicate that the stable from which Esther may also have derived was leaving an ongoing imprint in the Qumran library. It is worth noting on the way that the Book of Esther is reworked more than once, and almost certainly not solely in the Greek tradition; by itself it could form a study in how reworkings belong to the transformation of texts from positions of authority to canonical status.

But let us also consider the case of the Books of Chronicles. The claim is that Chronicles is represented in a manuscript from cave 4 (4Q118). However, it is clear from what survives of the manuscript that although part of 2 Chron 28:27-29:3 survives in the extant part of column 2, the remains in column 1 cannot be identified suitably. J. Trebolle Barrera comments that the one clear "reading is found neither in Chronicles nor in the Vorlage of Paralipomena, nor does it correspond to any parallel passage that precedes this material in Kings. The small size of the fragment prevents estimating the verse or verses of text to which this reading could correspond."

This evidence, or lack of it can be made sense of by contrasting the use made of Samuel-Kings and Chronicles in the Qumran community. As has sometimes been pointed out Samuel-Kings functions ideologically as a religio-political statement which puts David in his place; the chosen one slips up and the royal territory disintegrates after Solomon. Pride of place is given to those with prophetic insight. The more overtly Davidic ideology of Chronicles expresses post-exilic aspirations from a priestly point of view; many would date the work to the Ptolemaic period. What is the politics of this preference for Samuel-Kings at Qumran? There seem to be two sides to a possible answer to this question. On the one hand, from within the community the answer may well concern its antipathy to the monarchy and to the Hasmonean claims to be heirs to the Davidic tradition. On the other hand, if the politics of canonisation are promoted by the Hasmoneans in the late second and early first centuries BCE, then Chronicles could have supported their Davidic aspirations. Such Davidic aspirations are notably absent from explicit Qumran messianism until the Herodian period or later. There seems to be a fit between the Qumran evidence and the politics of the canonical process in the Hasmonean period.

But the story is no doubt far more complicated. In large part Chronicles is the reworking of earlier traditions to meet a particular set of needs. H.P. Mathys, for one, has recently endorsed the view that Chronicles is a counter to Manetho, Hekataios and Berossos, an authoritative presentation of a conviction that Israel has belonged in the land since the beginning of time, and a description of the temple as the means to salvation. Mathys, somewhat clumsily, describes the Chronicler's reworking of sources as "a midrash, Targum, or 'the rewritten Bible'." If promoted or adopted in particular by the Hasmoneans as reflecting their Davidic self-understanding and their role as cultic reformers, then Chronicles demonstrates that authority in a text is in part the result of it being a satisfactory reworking of the tradition. In other words, the reworking of earlier tradition may have been viewed by some Jews in the latter half of the Second Temple period as a standard way through which any composition might be able to lay some claim to authority. In the case of Chronicles this eventually led to its acquiring canonical status; and the same happened to the Book of Jubilees in some religious communities. Over against modern views of rewritings as evidently secondarily and plagiaristic, in early Judaism such imitation with its own form of exegetical innovation was entirely justifiable as a claim to the authoritative voice of the tradition. Although evidently not a major player in the life of the Qumran community as far as the clearly sectarian compositions in the library are concerned, the place of Chronicles should not be entirely dismissed. D.D. Swanson has shown that in several places in the Temple Scroll, or at least in the sources on which the scroll depended, Chronicles seems to play a major role.

Although such evidence could be read, probably correctly, as an indicator of the pre-sectarian and even pre-Hasmonean character of the Temple Scroll, it should also be remembered that the composition was certainly copied during the lifetime of the occupancy of Qumran, probably at Qumran itself, 11QTa being in an Herodian script with plene spelling as is characteristic of the so-called Qumran scribal school.

The point to be noted here is that attention to the politics of canon has seldom taken adequate account of the ways in which what may be labelled as ideological pluralism or multiple narration actually works. It is thus seldom a matter of one group claiming a composition as authoritative which then inhibits another group from so doing. If the Hasmoneans laid claim to the Books of Chronicles, such a claim does not prevent other groups from using or appealing to the same text for their own reasons. In fact, the logic of the processes of canonisation is that one group's assertions about the authoritative status of a composition requires other groups to acknowledge that authoritative status as well. Although the process of canonisation may lead to a politics of the exclusion of various literary compositions, it leads to the inclusion of an ever-widening readership or audience.

Put the other way round, an implicitly exegetical composition such as the Rule of the Congregation could not achieve canonical status without a wider audience than that intended by its sectarian authors. If S.Pfann's interpretation of many of the 4Q Cryptic fragments establishes itself, then it is clear that the Rule's reworking of scripture was presented as authoritative, and the sources of the authoritative reading made explicit in terms of the Sons of Zadok, but the restriction, even encoded restriction of the composition in the second century prevents the composition from belonging to the same process of the transmission of authoritative texts which see such reworkings of tradition as Chronicles or the Book of Jubilees eventually attaining canonical status.

3. Language and script.
Mention of the Rule of the Congregation in its Cryptic forms introduces two matters which might need to be considered together as part of how texts, especially reworked texts might move from authority to canon. It is clear that the great majority of reworked 'Bible' compositions are composed and transmitted in Hebrew, the language of the models which are the subject of the reworking. Language is part of the political dimension of the process of moving from authority to canon, though it is also more than that, since the language, that is the vocabulary and phraseology of authoritative texts provides the vocabulary in which it is possible to say anything at all that resonates with others.

Many of the reworkings of scripture for which the scrolls from Qumran now give us examples have been composed in pre-Hasmonean times, but the Hasmoneans in giving a boost to the process of canonisation seem also to have adopted and expressed what may have been latent in the scriptural reworkings. Hebrew was evidently a symbol of national independence. For the Hasmoneans it was Hebrew in which their court propaganda, now known as 1 Maccabees, was composed, and John Hyrcanus I set a trend of Hebrew being used on coins. The Hasmonean use of Hebrew capitalised on the very process which can be observed in the reworked and rewritten scriptural texts, that authoritative status had a language preference, and that preference was for Hebrew.

The use of paleo-Hebrew script on coins and in some manuscripts of the books of the Torah and Job is probably also to be taken as an indication of authoritative status, as the Samaritan Jewish community have long perceived. Are there any reworked scriptural compositions preserved in manuscripts written in paleo-Hebrew? Some might think the question easy to answer in the negative, but I would suggest that a more suitable starting point would be to consider the character of each manuscript in turn and come to some understanding of its textual character. The likes of 4QpaleoExodm with its harmonisations and supplementary scriptural material and 11QpaleoLeviticus with its independent textual character, in evidence not least for example in its combination of the sexual prescriptions of Leviticus 18 and 20, might both be suitably categorised as reworked scripture, and their presentation in paleo-Hebrew might be conceived primarily as an additional device to assert the authority of the reworkings which they contain.

But overall in the matter of language (and to a lesser extent script) it is not easy to be definitively assertive. It is widely agreed that Aramaic was the lingua franca in Judaea from the Maccabees to Bar Kokbah, so the survival of Hebrew needs some explanation. Often this is expressed either in terms of the nationalistic significance of the language or in terms of the dominant place of the Torah and the temple in early Jewish cultural and religious life. It seems to me that before the nationalistic use of the Hebrew language in the second century BCE, the reworked scriptural compositions indicate in a very significant way how the teaching of the Torah and the language of the cult was updated and presented to each generation anew. To some extent, perhaps to a great extent, the reworked scriptural compositions carry the authority of the tradition forward in ways in which the primary texts on which they depend could not. Subsequently at the time of the listing of authoritative and canonical works, such reworked scriptural compositions generally found themselves excluded because it was sufficient to retain those books to which the reworked compositions pointed, but until such time the reworked scriptural compositions can be understood as the principal vehicle through which interest in the texts which became canonical was maintained.

But, what of a work like the Genesis Apocryphon? Beyond most others, this composition has been assigned in many discussions of reworked scripture to the category of rewritten Bible. And yet it is in Aramaic. The matter should not be pressed overmuch, though it is worth noticing that most of the other compositions in the Qumran collection which are both close to reworkings of scripture and also in Aramaic tend to be associated more or less easily with the great figures of old, such as Enoch or one of the twelve patriarchs (in the testamentary genre). Aramaic reworkings tend to be pseudepigraphs and the same may also be the case for the Genesis Apocryphon. The whole topic of pseudonymity and canon needs to be revisited in light of all the evidence that now exists in the Qumran library.

The Qumran community's own rule books are exclusively in Hebrew. This is often characterised as an indication of how the community understood itself to be in continuity with ancient Israel. It was not a post-biblical community but part of a movement which could claim to be the true ongoing heirs of Israel of pre-exilic times, heirs with a renewed covenant. The presence of so many compositions of reworked scripture within the library of a community with such a self-understanding should also be taken into account. But this is not just an issue of language, but a matter which also has to do with the nature of religious continuities and this requires us to turn to the topic of institutionalisation.

4. The adequacy and inadequacy of institutions
Is it too broad a generalisation to suggest that canons are declared at times of institutional uncertainty? One mark of such uncertainty might be changing institutional structures or even their demise. We cannot be certain that the process of the move from authority to canon for the Hebrew scriptures was a Hasmonean project, but the need of the Hasmoneans to reinforce and promulgate their legitimacy can be seen as part and parcel of their overall institutional frailty.

For the Qumran community and its library of scrolls, the withdrawal from the Temple may have resulted in the search for alternative institutional supports and the promotion of authoritative texts correctly interpreted within the community. In fact it may be that it is not solely a matter of withdrawal from the temple. If the occupation of the land is also considered to be one of the factors of institutional stability in second temple times, then the Qumran community's (and the broader movement's) identification with the Israels of the Exodus and Exile may indicate that they rejected the status of the rightful occupation of the land as well as the suitability of contemporary priestly practices in the Temple. How the contents of the Qumran library are adequately described is immensely important for the better understanding of what takes place within Judaism more broadly in subsequent generations when the Temple is destroyed and the occupation of the land curtailed.

In several ways that are akin to some aspects of the literary activity of second temple Jews in the dispersion, the contents of the Qumran library indicate in an anticipatory way the processes which the wider Jewish community was subsequently to undergo. It is noticeable that in the Qumran literary collection there is a mixture of explicit and implicit commentary on authoritative scriptures. I am inclined to think that the explicit commentary such as is found in the pesharim is generally to be considered later than those compositions which contain implicit exegesis in their reworkings of authoritative texts. This means that one of the initial methods used by the covenanters to assert the authority of certain texts and to enhance that authority in various ways in the movement was through the promulgation of reworkings of the text. It could be argued that even the multiple copying of scriptural texts themselves with minor stylistic and other improvements belong to the spectrum of activity which is exemplified by the reworkings of scripture to be found in such texts as the Genesis Apocryphon or the opening columns of the Commentary on Genesis A, the Temple Scroll, the Apocryphon of Joshua, the Apocrypha of Jeremiah or Pseudo-Ezekiel, and various collections of Psalms. As mentioned already these reworkings have authority both in themselves as the effective means of maintaining and enhancing the ongoing significance of primary sources and inasmuch as they are derived from texts which are known to have or to have had authority in themselves.

The scribal activity evident in all the reworkings of scripture, some of which is evident in what are now classed by Qumranologists as biblical manuscripts and some of which are in the category of reworked or rewritten scripture indicate one of the means through which the covenanters compensated for the lack of institutional support from the Temple and for their sense of exile.

5. Closure and creativity.
As exercises in delimitation and exclusion the formation of canons are commonly supposed to be straightjackets of doctrinal conformity. But, as has been argued by many folk, canons are as much a starting point as an end point. In other words, rather than being the final word on what may be taken as authoritative in any religious tradition, canons of scripture tend to provoke extensive, elaborate and creative exegesis. This applies, even in those religious traditions which claim a doctrine of sola scriptura.

This model of canons as provoking creative exegesis needs to be applied to the pre-canonical period in early Judaism in which the process of canonisation or the establishment of scriptural norms is under way. The exegetical creativity to be found in the reworked scriptures amongst the Qumran library thus not only speaks of the authority of the base texts which are so richly reworked, but also suggests that the processes of establishing canonical norms are already under way. This is clear in what can be seen from the prologue to the Greek translation of Ben Sira, the so-called canon note in 4QMMT, and in the designations in some Qumran texts that reference should be made to the Law (or Moses) and the Prophets, however Prophets might be construed.

The very creativity in the exegesis of the community and the parent movement from which it sprang or the other parts of early Judaism whose literary works the covenanters chose to preserve is an indicator that closure has begun. The discovery of explicit commentary in the Qumran library shows that the process with regard to a certain selection of literary traditions is nearly complete.

It should be noted, moreover, that creativity is inhibited through its insistence on the relationship to what it interprets. Most overtly it is possible to recognise that in the reworked scriptural compositions found in the Qumran collection, there is little if any generic novelty. Although the absence of beginnings and ends forces scholars to label various compositions very vaguely, often as Apocrypha, most scriptural reworkings seem to be consciously modelled on scriptural antecedents and for the most part generic categories already in use suffice for describing the genres of what survives.

6. The presence or absence of the divine voice.
This sense of closure causes a tension which is reflected in interpretative activity. The status and authority of scriptural texts rests in no small measure in how such texts are viewed in relation to the sense of the divine voice in the community. Closed canons can readily be associated with the cessation of prophecy in the community, the absence of the divine voice. There may be occasional exceptions when a bath qol is recognised or asserted, or when religious subgroups attempt to construct their own direct routes to the divine, but most commonly closed canons have their counterpart in divine silence, at least until some future ideal time when God's presence will be restored.

However, as an indication that the process has begun but that there is no formal closure to the canon, it is worth noting how closely both implicit and explicit exegesis in the Qumran materials are to be associated with the ongoing activity of God or his agents. In the Book of Jubilees the angel of the presence delivers the content of at least some of the heavenly tablets, in the Temple Scroll the composition is constructed as a divine speech, in the Damascus Document there is mention of the nistarot which must be understood alongside what has already been revealed and in the pesharim the explicit claim is made that it is the Teacher of Righteousness to whom God has made known all the mysteries of the words of his servants the prophets. H. Najman has aptly termed this activity in relation to the Book of Jubilees "interpretation as primordial writing." In one sense the reworking of scripture is presented as prior to scripture itself and therefore as of great or even greater authority. In addition, however, it is clear that the reworked composition never attempts to replace or displace the scripture it reworks. Thus the reworkings are presented as a complementary aspect of the divine voice: perhaps where scripture is indirect and available to all, the reworked composition is direct and available to a select few. This may be likened, then, to the relationship between the oral and written Torah, except that in this case both forms are written. Whatever the case, the reworked scriptural composition is presented as the continuation of the divine voice in the community.

This observation leads naturally to the question: are all reworked scriptural compositions to be understood as presenting authentic divine speech in ways in which their scriptural counterparts might never claim? What works for the Book of Jubilees and the Temple Scroll might also work for the Apocryphon of Jeremiah or Pseudo-Ezekiel. It might not be necessary to insist that all reworked scripture that is worth the name carries this quality of interpretation as primordial, but the tendency for almost all extant scriptural traditions to be matched by interpretative reworkings in the pre-canonical period may indicate that rewriting and reworking were normally or usually carried out on compositions that carried some authority. Any text worth its salt would naturally be accompanied by a tradition of reworkings which presented its ongoing authority through a more explicit use of the divine voice.

7. The role of the community
It has long been noted that scriptural canons reflect, generally or specifically, the ideologies and religious practices of the communities that promote them. The scribal practices of the latter half of the second temple period indicate that scribes were often more than straightforward copyists. As they copied, they worked to improve the text in minor ways, through harmonisations, small exegetical additions, stylistic improvements, and so on. As stated at the outset of this presentation even the more elaborate reworked scriptural compositions belong together with this scribal activity. The variety of textual forms for the various scriptural books, together with the number of the reworked compositions shows how seriously the Qumran community and its forebears took their inherited authoritative traditions. These works were not to be left untouched on the shelf.

The place of scripture in the community's own sectarian compositions may be read in one way as an insistence on their looking at themselves as continuous with Israel's earlier experiences, but from another angle it is closer to an appropriation of language from another time so that the community's contemporary experiences are biblicized. The reworkings of scripture in the Qumran library reflect both the character of the identity of the community and the process through which it wished to project itself.

8. Cultural coherence
The association to some degree of the community with the reworked compositions is a reflection of another aspect of the process of the shift from authority to canon. Inasmuch as canons suggest the ideologies of particular communities, they also imply a measure of cultural coherence by which such communities can express their self-identity and regulate their day-to-day practices. The emerging canon of the Law and the Prophets in the second temple period.

Within the reworked scriptural materials in the Qumran library, it is clear that there is a move towards just such a coherent outlook. Most obviously this concerns the calendar, which has affected the rewriting of Genesis in the narrative of the flood in Commentary on Genesis A and of Genesis-Exodus in the Book of Jubilees, it lies behind the presentation of adapted festival legislation in the Temple Scroll, and it may have influenced the collection of Psalms in the 11QPsa manuscript.

The absence of explicit legislation in the Torah for the construction of the present or future temple, for all that such may be transferred from the description of the wilderness Tabernacle, permits and even encourages the composition of works such as the Temple Scroll in which the language and phraseology of scripture is reused to fill the gap.

9. "Literary power and sheer popularity"
Although the place of the community in determining authority in texts and in encouraging a particular kind of cultural coherence should not be underestimated, there are other factors involved in the way in which certain compositions achieve authoritative status. Two of these may well lie beyond the control of small religious communities such as the Essene movement or the Qumran community.

To begin with, the literary power of particular texts may encourage further literary activity. Such power may rest upon a variety of factors in addition to literary artistry. The appeal of the Eden narrative in Genesis 2-3 rests in the dynamics of human psychology and the interaction of the sexes as much as it depends upon a particular view about the origins of evil. An alternative view of evil's origins as is present in the Enoch cycle cannot displace Genesis 2-3 from the late second temple library. The literary achievements of the Noah story or the Abram/Abraham cycle of narrative incidents seem to have been sufficient in themselves to provoke ongoing interest and reworking. Where little is said, supplementary material asks to be supplied, as for the testaments of the patriarchs; an unfinished symphony or incomplete requiem yearn for adequate completion for the satisfaction of the hearer.

If many of the ingredients of literary power are not clearly evident, then a composition may nevertheless retain authority through the popular appeal and inherent vision of its narrative. The authoritative place of the Book of Tobit in much Christian tradition no doubt depends to a major extent on the popularity of the storyline rather than on any particular doctrine such as might concern the reward of the righteous or the theology of healing practices. The presence of Tobit in the Qumran collection in both Hebrew and Aramaic attests to the popularity of the work. The multifarious Daniel traditions may have exercised similar popular appeal.

10. Pluralism and subversion
The authoritative scriptures contain two other features which, taken together, may also seem to improve our understanding of the status and function of the reworked scriptural compositions we are considering: intertextuality and transhistoricality. It is evident from the Torah, but also from the Prophets, that the compositions emerging as authoritative from the beginning of the second temple period onwards contain many internal cross-references, an internal intertextuality. This is most overtly evident in the several repetitions to be found in these compositions, some of which are almost certainly recognisable as reworkings. The authoritative texts represent at least some of the tradition in a plural form, and much of the endeavour of reworkings in the late second temple period is the attempt to reduce this pluralism.

Such reduction is done on the one hand to create a single schema through which the authoritative text can be represented with a significance that can be readily grasped by the reader or hearer. But this reduction is not done at the expense of intertextuality. More often intertextuality is made to serve such a project. So in the painstaking task of modern commentary on these reworkings of scripture, the commentator is challenged to rediscover the intertexts and propose a suitable hierarchy of organisation in reading so that some sources can be seen as controlling the use and re-presentation of others. The juxtaposition of material from Exodus and Ezekiel, from Kings and Chronicles in sections of the Temple Scroll, or the interweaving of Ezekiel 1, 10 and Isaiah 6 in Pseudo-Ezekiel's description of the divine throne, carries forward the intertextual adventure of the authoritative texts themselves.

The literary critic R. Alter has recently written some challenging reflections on the nature of the canon of the Hebrew Bible. Provoked in part by those parts of the present canon, particularly in the Writings, which defy neat classification within what may be perceived as a doctrinal strategy in the process of canonisation (which Alter labels the first canonicity), Alter has argued that "a canon is above all a transhistorical textual community." Alter's description is motivated by his observations concerning how several modern authors have appropriated biblical material and used it for their own ends. This leads him to the conclusion that the canonical texts (or indeed any texts) "do not have a single, authoritative meaning, however much the established spokesmen for the canon at any given moment may claim that is the case." If one acknowledges this as basically correct, then the reworkings of scriptural texts preserved in the Qumran library may well be seen as the means through which communities attempt to provide authoritative and possibly singular meaning to texts whose authoritative status means that they cannot be pinned down to a particular time and place. Thus pluralism and the character of transhistoricality thus produce similar results in provoking reworkings of scripture in which localised products are created with singular schemes, products which are themselves unlikely to obtain or retain authority outside the spheres of those who produce them.

11. Conclusion
It might suitably be noted that the Dead Sea Scrolls have come to light and been published in a half century which has seen many kinds of claims about the nature of the Bible. The scrolls provide a means through which many such claims can be suitably assessed and given appropriate historical perspective. This paper has tried to argue that from many perspectives the reworked scriptural compositions, some of which end up as canonical, are a fundamental part of how the processes leading to canonical lists or clear enumerations of authoritative books came about and what was involved in them.