Response to David Lambert, ‘What is Scripture?’

David Lambert’s stimulating and imaginative chapter draft, pointing forward to a forthcoming book, is a significant contribution to our understanding of the growth of the biblical canon. In this short piece I shall try to draw out some of its implications.

In some ways I think that Lambert’s ideas, though highly original, could also be seen as the next logical step in understanding the formation and reception of the Hebrew Bible. There seem to me to have been three broad periods in the study of the biblical canon so far. In the days, which I still remember, when no one was much interested in the canon at all and it formed the last chapter tagged on at the end of Introductions to the Old Testament, the consensus (such as it was) was that the canon had been fixed at the council of Jamnia, as it was called, at the end of the first century CE. But most of the Bible had been recognized as authoritative scripture well before this, and probably in the order Torah, Prophets, and Writings, as the traditional Jewish arrangement would probably lead one to expect.

A new phase was opened through the work of Albert C. Sundberg in the mid-1960s, who argued that the canon was not fixed at all in the first century but consisted of what he called “a wide religious literature without definite bounds,” which only got delimited and set in stone some time into the Common Era.^[1] Jews and Christians made different selections from this wide literature, hence the divergence between the Hebrew and Greek canons, which was not a distinction between a shorter Palestinian and a longer Alexandrian canon, as had previously been thought, but represented rather the canons of what were by then two different religions, Judaism and Christianity.

A third phase came about with the distinction between canon and scripture, a distinction drawn by James Barr^[2] and then developed by Eugene Ulrich.^[3] I also wrote a bit on this.^[4] Scripture, in this distinction, means books that are accepted as having authority; canon means a list of the works that alone are to be reckoned as scripture. This is a temporal distinction. Books first become scripture, and this may in some cases be an early development—Deuteronomy was perhaps scriptural, in the sense of authoritative, almost from the moment of its composition. Only later do people come to think that there is a limit to the number of books that can properly be regarded as scripture, and once that has happened, we have a canon. I would put it this way: to describe books as scriptural is to say that at least those books have religious authority. To describe them as a canon is to say that at most those books have authority, those and no others. The canon of the Hebrew Bible is then a work of the very late first century CE or maybe even a bit later, as Timothy Lim argues,^[5] but the idea that the books now in it are authoritative scripture goes back considerably further. If we look, for example, at the New Testament evidence, it is clear that more or less all the books now in the canon were already treated as scriptural, but there is little to point to any limitation of which books count as scripture. The quotation of 1 Enoch 1:9 in Jude 14-15, for instance, suggests that that book was then still seen as scripture rather than being outside the canon, which for that period is still an anachronistic concept.

That seems to me where we have got to so far in most discussions of scripture and canon. What Lambert has done—developing ideas already adumbrated in the work of Hindy Najman^[6] and Eva Mroczek^[7]—is to help to move the discussion on to a fourth phase in which even the idea of scripture is seen to be more problematic than has traditionally been thought. This has two aspects. One is that the idea that books have authority emerges as much less clear than it may at first appear. Scriptures are texts important enough for Jews and, in due course, Christians, to enter into dialogue with—they are texts that are regarded as non-trivial. But to say that they exercise authority is to oversimplify the way texts and readers interact. A scriptural text is one actor in a complex of relationships and its authority is negotiated as people react to it and appeal to it in various different ways, alongside other traditions, customs, or networks. Lambert in his chapter draft deploys Actor Network Theory as developed by Bruno Latour and others to describe how all this works, and provides what we might call a phenomenological rather than an authority-based or deontic account of how scriptures operate in Judaism in practice—as opposed to how some supposed canonizing body might have thought they ought to operate, if there had been such a body. In practice, scriptures are part of an interconnected set of agents which also includes rabbinic authorities, popular piety and practices, and rational or non-rational beliefs. Recently I have been working with the idea of the Bible and religious faith and practice as intersecting but only partially overlapping circles: much that is in scripture is in practice ignored by religious practitioners, whether Jewish or Christian, and conversely much that is done and believed does not really derive from scripture, even though there may be a theory that it does or should.^[8]

The second aspect of Lambert’s chapter draft that I would highlight is the idea that the supposedly fixed canon never did really get fixed. There were groups that espoused written works that were never in any “official” canon at all, Jubilees being a salient example. Jubilees is by its own claims Mosaic, but revealed to Moses by the angel of the presence and hence in a sense written by God; and therefore presumably trumps even the Pentateuch in many ways, since the Pentateuch does not claim divine inspiration except for a narrow strand of laws. We treat Jubilees as one of the so-called Pseudepigrapha, outside the canon, but there are historically speaking no rulings that so exclude it and many people in antiquity treated it, clearly, with great respect. To call it non-canonical is probably anachronistic not only for the first century CE, which might be widely agreed, but down to a much later period. There are analogies in Christianity. The Protevangelium of James is outside the New Testament canon, and does not appear even in manuscripts such as Codex Sinaiticus that contain books we regard as non-canonical (Barnabas and The Shepherd of Hermas). Yet features of it occur in almost all popular presentations of the nativity story: Joseph as an old man with children from a previous marriage, Mary riding on a donkey, the nativity happening in a cave used as a stable. The average Christian would say that such things are in the New Testament, but they’re not. They are however in practical terms “authoritative” just as much as the details that are in the New Testament, as you will discover if you try to tell people that they are apocryphal. The effective canon includes them even though the theoretical one doesn’t. Similar things could be said about Jubilees. Molly Zahn similarly points out that the category “Rewritten Bible” does not do justice to “reworkings” of Pentateuchal material at Qumran, since they may have had an authority, if we are to use that word, as great as or even greater than the texts of which they are a partial rewrite.^[9] Once again we should try to avoid anachronism. Scripture is a fluid category, including texts that were never “canonized” at all but which were held in reverence as ancient, and oral traditions that were never even fixed in writing. Later stories about scriptural figures can be as important as what is actually in the Bible, and some may be regarded as actually predating biblical texts—what Lambert calls “Prewritten Bible.”

I found this passage from his chapter draft particularly helpful in summing up his argument:

Our ancient sources rarely themselves speak of texts (and, when they do, not necessarily biblical texts); they certainly do not speak of readers, at least, not in our customary sense. Instead when their associations with the biblical are allowed to flow unimpeded, we encounter a much broader range of agents. Far from a study hall where the text is the sole point of origin, the source, the authority (I know of none actually like this), we find a carnival of actors: human and divine, animate and inanimate, biblical and non-biblical, past and present, as well as future. “Powers” or “capacities” are not localized in a single unit, like a text or reader, but are “overflowing in all directions.”

The whole chapter draft shows convincingly how oral traditions, texts, recipients of both (whether readers or hearers), and whole communities attached to a source of teaching or argument, process and interiorize the ideas that emerge from a complex interplay of voices. This does indeed move the discussion of “canon” to a new level, and marks Lambert out as one of the significant players in it.