A Literary Perspective on the Biblical ‘Canon’

[First posted in 2012.  This is from:  The Literary Guide to the Bible, eds. Robert Alter and Frank Kermode. Reformatting and highlights added.]

 

——————

 

The Canon

Frank Kermode

 

This chapter  offers some explanation of the processes by which the Bible came to include the books it does—insofar as that can be done in reasonable space, if indeed at all—and to venture some remarks on the consequences of their transmission to us as a single book. But it is necessary to begin by saying why we have chosen this particular version of the Bible; for there are many differently constituted Bibles, each with its own version of the canon, and it might be thought that our choice is quite arbitrary.

 

• Most obviously, the Jewish Bible lacks the twenty-seven books of the New Testament.
• The Jewish Bible in Greek—a collection of great antiquity and authority—differs as to contents, and frequently as to text, from the Hebrew Bible.
• The Latin Bible of the Roman Catholic tradition contains in its Old Testament books dismissed by the Bibles of the Reformed churches as apocryphal.
• Those churches include as their Old Testament the books of the Hebrew Bible and the twenty-seven New Testament books.
  • This is the ”Bible” treated in the present book;
  • it is what most people think of when they think of the Bible;
  • it is the collection to which modern literatures mostly refer;
  • and the fact that all Bibles have them, no matter what else they include, gives them an importance greater than that of the disputed elements.

This does not imply a literary judgment on the works excluded, nor does it reflect a belief that all the canonical books are of superior merit.

 

We do not understand all the criteria of canonicity, but we know enough to be sure that modern criteria of literary quality have no relevance to them.  Even the most learned explanations of how the constituent books found themselves together in a canon are highly speculative and have to deal with an intractable mixture of myth and history.

 

Once a sacred book is fully formed, deemed to be unalterable and wholly inspired, it acquires a prehistory suitable to its status and related only very loosely to historical fact or probability.

 

The real history involves all manner of external influences:

 

• for example, the closing of the Jewish canon must be in some sense consequent upon the waning of Hebrew as a spoken language,
• and upon the destruction of the Temple in 70 C.E., when the book rather than theTemple cult became central to religion.
• Already there were more Jews in the Diaspora than in Palestine, so the time for such a change was ripe, and the Bible already holy, acquired an extra cultic sanctity.

In the case of the New Testament it seems possible that—

 

• the lack of an appropriate technology prevented its achieving definitive shape until the fourth century.
• The Christians preferred the codex or leaf-book to the scroll,only in the fourth century did it become possible to produce a codex that would hold all the accepted Christian Scriptures.
  • and during the earlier period these newfangled codices could not contain texts of any great extent;
• Thus canon formation is affected by what seem on the face of it to be political, economic, and technological forces without immediate religious or literary relevance.

The legendary account of the growth of the Bible tells of—

• the destruction of the sacred books during the Babylonian Captivity
• and their reconstruction by the divinely inspired memory of Ezra.

By this time (fifth century B.C.E.) the canon was virtually complete, though Daniel, traditionally ascribed to the sixth century, was added in the second.

At the end of the first century C.E. a final list was established at the Council of Jamnia.

 

A more scholarly account would say that—-

 

• the importance of the Law after the return from Babylon speeded the process by which all the disparate material in the Pentateuch acquired final form and authority;
• the other two sections, the Prophets and the Writings, developed at a different pace,
• and in some instances, notably that of the Song of Songs, there was dispute about a book’s status well into the second century C.E., tradition has it that the Song of Songs was saved by the advocacy of Aquiba, as a religious allegory.
• Although the proceedings at Jamnia are not nowadays thought to have been concerned with the canon, the learned still appear to accept the date, ca. 100 C.E., as about right for the closure of the canon.
• It was of course necessary to leave things out as well as let things in, and a distinction was drawn between.  Such was the practice as early as Ezra, who, according to legend, set aside for the use of the wise seventy books apart from the Scripture proper.
  • books which “defiled the hands” because of their sacred quality,
  • and “outside” books which presumably failed this test, though they might still be granted a certain extra-canonical utility.
• Books thus set aside or hidden away would be apocrypha in the original sense; the word later acquired dyslogistic overtones, and the apocryphal came to mean the false or inauthentic.

It would be wrong to suppose that all the constituent books were submitted to the same impartial examination.

 

• The Five Books of Moses were naturally of unassailable authority,
• as were the Psalms
• and the Prophets.

The invocation of Old Testament texts in the Gospels is evidence, if such were needed, of the reverence accorded the Scriptures in a time before the canon was finally established. One might say that there was a canonical habit of mind before there was finally a canon, and that it was in evidence during the long centuries that separate Ezra from 100 C.E.

 

There is some question whether it is proper to speak of a Jewish canon at all, and insofar as it has to be accepted as corresponding to real historical developments it may be thought of as a fictional construct concealing the historical truth.

Thus the large redactive enterprises carried out on the Torah are concealed by its canonical form, and scholarship has to break it down again into its original components.

 

It is true that revisions of the Old Testament books were carried out in response to external pressures—for example,

 

• the political needs of post-Exilic Israel,
• and, in the first century C.E., the centrifugal force of heresy and schism.

But the fact that Judaism reacted to these forces by affirming the cohesion of the Scriptures and, ultimately, by effectively closing the canon is sufficient evidence not only of the significance of the individual books, but of the belief that their power was enhanced by membership in a whole greater than the sum of its parts.

 

The evolution of the New Testament is another story, though hardly less complicated and conjectural.

 

The first Christians already had a Bible—the Jewish Bible in various forms, Hebrew and Greek and Aramaic—and saw no need of another.

 

What was central to their beliefs was transmitted by oral tradition; indeed the authority of that tradition survived into the second century, although most of what to become the New Testament already existed.

 

The power of the oral tradition did not reduce the Christian commitment to the Jewish Scriptures; the faithful lived in the end time, history was coming to a close, and events would all occur “according to the scriptures,” as they had in the life of Jesus. In a sense the oral tradition took its place beside the Scriptures, just as the Jewish tradition of oral interpretation filled out the implications of the written Torah. In the end both were written down, but the Christian writings came earlier, partly because as the years passed it must have seemed important to perpetuate the increasingly fragile oral testimony of the works and sayings of Jesus.

 

One consequent of the growth of Christian Scripture was the transformation of the Old Testament into quite a different book, a sort of unintended prologue to the New Testament. Whether it should be retained at all became a serious question; and the reasons for keeping it were of a kind that had nothing to do with Judaism.

 

The gradual replacement of the oral tradition by writing was the necessary prelude to the establishment of a canon, with all the consequences of that development.

• Oral tradition is quite different from written;
  • it is variable,
  • subject to human memory (however aided by mnemonics),
  • discontinuous,
  • selective,
  • and affected by feedback from audiences.
  • It would encourage its transmitters to invent
  • and to add interpretations.

It has been suggested that Mark’s Gospel—which we take to be the first of the canonical four—resulted from a conscious rejection of the oral tradition, which it represents as virtually extinct (the women at the tomb fail to transmit an oral message to the disciples) or as corrupted by the false preachers and prophets Mark assails in chapter 13.

 

Neither Paul nor the evangelists wrote with the object of adding to the existing Bible; indeed the only book of the New Testament that claims such inspired status is Revelation, with its threat of damnation to anybody presuming to add to it.

 

• Paul’s earliest letters belong to about 50 C.E.;
• the Gospels are of uncertain date, the consensus being that they belong to some time between 60 and 90 C.E., though earlier dates have been proposed.
• It seems likely that the contents of the New Testament were written over a span of something close to a century,
• and none of them by writers who supposed they were candidates for entry into a fixed corpus of Scripture.

It is easier to understand why gospels got written (though less easy to see why they took the form they have) than to guess why four, no more and no less, were finally accepted. There must have been many more, and it appears that in the second century there were three versions of Mark available, one public, one reserved for the few, and another used by a Gnostic sect and condemned by the orthodox. Only the public version survives.

 

John was also attractive to Gnostics, and there was accordingly stiff opposition to his inclusion in the canon. Here again we need to remember that “gospel” originally meant not a piece of writing but the good news proclaimed by Jesus; the evangelists wrote down their versions of this news, which were labeled “the Gospel according to X,” and eventually the term came to mean also this new genre.

 

The relation of these new documents to the existing Scriptures was a matter for dispute; the heretic Marcion wanted to do away with the Jewish Scriptures altogether, and to recognize as authoritative only a version of Luke and of some Pauline letters. It was conceivably in response to such ideas that orthodoxy felt it must decide what had authority and what didn’t, settling on four Gospels as part of the New Covenant or Testament.

 

The concept of a new covenant and of its fulfilling or even replacing an older one is immediately indebted to the Eucharist, for Jesus spoke of the cup as the new covenant (he kaine diatheke) (1 Cor. 11:25, Luke 22:20), and ultimately to the covenantal element in Jewish theology.

 

When Paul (2 Cor. 3:14) talks about the Jewish dispensation as the old written covenant now replaced by that of Christ—the letter replaced by the Spirit—he is still thinking of the new testament (this is the Latin translation of diatheke) as unwritten. Indeed the expressions diatheke and testamentum (sometimes instrumentum) were not applied to the new writings until late in the second century, by which time the idea of a body of authoritative Christian writings, including the letters of Paul and the four Gospels, was well established.

 

In the intervening period it is probable that the originals were altered or augmented for the sake of doctrine or inclusiveness; they were not thought of as inspired. Reasons for holding them to be so were provided later. Only when their inspiration became an issue did the discrepancies among the four seem to call for attention.

 

 

Other books were scrutinized according to criteria on the nature of which there is still much dispute, though it is interesting to note that the tests applied were in part philological. It was noticed, for instance,

 

• that the Greek of Revelation is not that of the evangelist John, to whom it was attributed;
• and that the Greek of Hebrews is of a quality sufficient to prove that it was not written by Paul—
• perhaps, it was proposed, Luke wrote it up from notes.
• Doubts were entertained concerning 2 Peter and 2-3 John.

These issues never quite died away and were important at the time of Reformation, since sola scriptura requires one to be sure what scriptura really is. Luther at first rejected Revelation and had grave doubts about James. But all these works have survived in the canon.

 

As time passed Christianity also became to a great extent dependent on a book, and although the authority of the oral tradition survived—and continues to survive in the magisterium of the Roman Church—the written word acquired the greater power.

 

There remained the need to close the canon, and the date given for this is 367 C.E., when Athanasius listed the twenty-seven books as the only canonical ones. He actually used the word, and also gave a list of rejected books, which he called apocrypha. Doubts persisted, and there may be argument as to whether the canon can really be said to be closed; but it is not been added to as yet, nor has anything been taken away from it; and it is hard to see how the Gospel of Thomas, discovered in this century, if added to the canon, could partake of the authority acquired by the others over the years.

 

Kanon, a Greek word originally meaning “rod,” came to signify many other things,

• including an ethical norm or a rule or criterion.
• It could also mean a list of books,
• sometimes—and this is the beginning of the biblical sense—a list of recommended books.
• By 400 C.E. it meant, for Christians, only those books held to be holy and of authority.

The Jewish canon, even though it was not so called, had similar qualities.

• It is characteristic of the Jewish tradition that great care taken over the transmission by copying of the sacred text, which was held to be unalterable and without corruption, though, as bibliographers know well, this is humanly impossible.
• The books contained within the canon or canons are held to be inspired and to be interrelated like the parts of a single book.
• Their relations with “outside” books are of a quite different order. It is important to understand the extraordinary privilege of these inside books.

Religious and political history would have been unimaginably different if the apocalyptic prophecies of Daniel had been excluded from the Old Testament, or those of Revelation from the New. John Barton has some interesting observations on the overwhelming importance of inclusion in the canon: suppose Ecclesiastes had been turned down, lost, and rediscovered recently among the documents at Qumran—would it not be virtually a different book from the one we have? Canonization can thus, as it were, alter the meanings of books.

 

The doctrine that the Bible is its own interpreter was held in different circumstances by both the rabbis and Luther, and the belief that one can best interpret a text by associating it with another text of similar authority clearly presupposes a canon; the idea of explorable correspondences between every part would be absurd if one had no certainty about the extent of the whole.

If the entire text is inspired—a belief deeply held by the Jews, with their scrupulousness about every jot and tittle, and given formal expression for the Christian canon at the Council of Trent in the sixteenth century—then the most fleeting echo, perhaps only of a single word, is significant. And given that everything is inspired, all possible relations among parts of the text are also inspired.

 

The poet George Herbert had these relations in mind when he wrote, in “The Holy Scriptures,II”:

Oh that I knew how all thy lights combine,

   And the configurations of their glorie!

   Seeing not onely how each verse doth shine,

But all the constellations of the storie.

This verse marks that, and both do make a motion

   Unto a third, that ten leaves off doth lie:

   Then as dispersed herbs do watch a potion,

These three make up some Christians destinie.

 

We can now specify certain characteristics of the mythical or magical view of the canon.

Regardless of innumerable historical vicissitudes, redactions, interpolations, and corruptions, the canonical text is held to be eternally fixed, unalterable, and of such immeasurable interpretative potential that it remains, despite its unaltered state, sufficient for all future times.

This perpetual applicability is established by a continuing tradition of interpretation, as the relevance of old texts to new times always is. Interpretation is controlled by changing rules but is remarkably free, for the canonical book, itself fixed in time and probably in a dead language, has to be made relevant to an unforeseen future. It must prefigure history: hence we have typological interpretations. The book becomes a mythical model of the world: the Torah is said to be identical with the Creation, the Christian Bible becomes the twin of the Book of Nature. And the exploration of these world-books requires interpreters who can study the subtle hidden structures just as physicists and chemists (or their ancestors, the alchemists and astrologers and magician) studied the created world.

It is hardly surprising that the assumptions underlying these views collapsed with the onset of modern scientific philology. From the beginning the canon was seen as a late and arbitrary imposition on the books it contained. Those books should be studied like any other ancient texts, understood in their original senses, and valued for what they told us about the past, so that the work of the interpreter becomes primarily archaeological. It is not the book’s membership in a canon that gives it authority, but its report of or allusion to various historical events and persons. And of course the true as opposed to the legendary history of the formation of the canons supports this commonsense view of the matter, for there is little reason to believe that such a series of accidents, unexplained judgments, decisions taken under who knows what political or ecclesiastical duress, should result in a divinely privileged, exclusively sacred, compilation. For the factitious context of the canon the scholars substituted the larger contexts of history. They knew by what methods the sacred texts had been made closely applicable to modern situations; if the New Testament had not already taught them that lesson, the Dead Sea Scrolls, which applied ancient Scriptures exclusively to the concerns of a particular sect at a moment presumed to be just before the End, must have made it plain. And thus the canon, despite its importance in the formation and continuance of the religious institutions which indorsed it, seemed to crumble away. It was no longer a separate cognitive zone, merely a rather randomly assembled batch of historical texts; really, one may say, no longer a Bible so much as a collection of biblia.

 

Such attitudes are as old as “scientific” biblical criticism, from the beginnings of which in the late eighteenth century it was assumed, by Michaelis among others, that the canon was not uniformly inspired, and that by historical analysis one could even assist religion by finding out which books were inspired and which were not. Later the question of inspiration was dropped, or the word acquired a new sense. It might be difficult for some investigators to devote themselves to pure historical truth when it involved the dissolution of the New Testament into a scatter of fortuitously assembled occasional writings; for in most cases these scholars were Christians, and the New Testament is after all the foundation document of their religion. But there were ways out of that dilemma which did not involve their subscribing to obsolete and false ideas about the canon.

 

In recent years the historical-critical tradition, now well over two centuries old, has been under challenge. That tradition also made hermeneutical assumptions of which its practitioners were not fully aware. For example, they were ready to believe that older views on the canon and the status of the separate books could be dismissed as peculiar to their time and as founded on assumptions now evidently false; but they took it granted that they themselves were exempt from historical “situatedness,” that they could, without interference from their own prejudices (of which they were unaware), transport themselves across history in a pure and disinterested way. As Hans-Georg Gadamer has put it, the historical critic is always seeking in the text something that is not the text, something the text of itself, is not seeking to provide; “he will always go back behind [the texts] and the meaning they express [which he will decline to regard as their true meaning] to enquire into the reality of which they are the involuntary expression.” But it is possible to take an interest in the text and its own meaning; that is literary criticism proper, and Gadamer believes that it has for too long (in these circles) been regarded as “an ancillary discipline to history.”

 

The opposition that has lately developed to “scientific” disintegration of the canon is based on the idea that the Bible still ought to be treated as a “collection with parameters.” Brevard Childs, who uses these words, has studied both Testaments from the point of view of a revived but still moderate belief in canonicity. Childs wants to eliminate the tensions between historical criticism and an understanding of the Bible as canonical Scripture; he wants, not a return to precritical notions of the canon, but attention to its historical integrity; for he thinks it important that the canon was the product of historical interactions between the developing corpus and the changing community, not of some belated and extrinsic act of validation. And when fully formed the canon is not just an opaque wrapping that must be removed so that one can get as the contents and see them as they really were. Of course the constituents have their own histories, and it is good to know about them. But their preservation and their authority are owing not primarily to their usefulness as testimony to historical events. It is their capacity to be applied, their applicability to historical circumstances other than those of their origin, that has saved them alive.

 

Whatever one’s view of the controversy now in progress between defenders of the tradition of historical criticism and practitioners of what is now called “canonical criticism,” it is clear that the latter is not a primitive revival of precritical notions of plenary and exclusive inspiration. Since we are still living in an epoch in which the historical or “scientific” approach is normal, and therefore seems commonsensical or natural, we may tend to dismiss the opposition as merely eccentric. Yet its presuppositions are at least as defensible as those of the “normal” practitioners; both sides make large assumptions, the one believing that events and persons can be made available, as if by magic, to the reader, and the other that historical application can form a body of discrete writing into a whole—as if by magic.

 

This, of course, is a different kind of magic from the old one; yet the old one still exerts its attractions. It remains quite difficult to think of the wholeness of a canon without associating the idea with the wholeness of an organism or the wholeness of a world. We observe in the realm of secular literary criticism the powerful effect of canon formation on the kinds of attention paid to the books included, even though it is impossible to think of secular canons as closed with the same definitiveness as ecclesiastical canons. And it is undeniably attractive to be able to think of the canon as forming an intertextual system of great complexity, to be studied, by a weaker magic than was available in the past though it is still a kind of magic, as a fascinating array of occult relations, a world of words.

 

Goethe, commending Hamlet, said it was like a tree, each part of it there for, and by means of, all the others. Five hundred years earlier a Kabbalist said this of the Torah: “Just as a tree consists of branches and leaves, bark, sap and roots, each one of which components can be termed tree, there being no substantial difference between them, you will also find the Torah contains many things, and all form a single Torah and a tree, without difference between them… It is necessary to know that the whole is one unity.” Moses de Leon and Goethe appear to have had the same thought, though we could make the two statements sound very different by examining their contexts: one of them belongs to what we think of as Romantic organicism, the other to Kabbalistic mysticism and a Jewish tradition that has always accommodated change and variety of interpretation but has always thought of the Torah as an entity, coextensive with the created world.

 

A flatter, more rational version of the holisms of Goethe and Moses de Leon might be thought to suit us better in our own time. It is true that both historically and actually we grant a different form of attention to canonical books, and that secular criticism has seriously entertained notions of the literary canon that might well be thought to give it a kind of wholeness and a high degree of intertextual relations. Examples include the canonical element in the criticism of T.S. Eliot and the stronger holistic claims of Wilson Knight. It is surprising, therefore, that the professional biblical critics should feel a renewed obligation to save their canon. Schleiermacher, usually thought to be the founder of modern hermeneutics, was also a major New Testament scholar; he believed that the study of the constituents of the canon must be carried on by exactly the same methods and with the same object as the investigation of secular texts, but he also remarked that “a continuing preoccupation with the New Testament canon which was not motivated by one’s own interest in Christianity could only be directed against the canon.” It was out of such a conflict of interest that new ways of thinking about the interpretation of ancient texts developed, and new ways of thinking about history in general.

 

Whether the canon in question is Christian or Jewish or secular, we can no longer suppose that there is a simple choice between the historical and the canonical approach, since the two are now inextricably intertwined. It is an empirical fact that each book has its own history; it is also true that the association of many books is a canon was the result of a long historical process and owed much to chance and much to the needs and the thinking of people we know little or nothing about. But it is also a fact that works transmitted inside a canon are understood differently from those without, so that, if only in that sense, the canon, however assembled, forms an integral whole, the internal and external relations of which are both proper subjects of disinterested inquiry. Nor need we suppose that we have altogether eliminated from our study of canonical works every scrap of the old organicist assumptions, every concession to a magical view of these worlds and their profound, obscure correspondences. When we have achieved that degree of disinterests we shall have little use for the canon or for its constituents, and we shall have little use either for poetry.