In art history, de-attribution can be frontpage news. (One only has to think of the Rembrandt Project or the purging of the Vermeer canon.) It is otherwise in literature. Here, only attribution really makes the headlines, and many are the column inches granted to the attribution of a new poem (even maybe a terrible new poem) to William Shakespeare. De-attribution, on the other hand, tends to be pictured as a slow and invisible matter, akin to natural selection.
This picture or theory, indeed, probably works well enough with the bulk of literature. But there is a kind of author with whom it falls down completely. We mean such an author as Daniel Defoe, who published mainly anonymously, was enormously prolific, and many of whose writings were never reprinted and exist only in a few scattered copies in major libraries. To form an adequate idea of the canon of such an author, the common reader, and even the scholar, is going to have to depend on somebody's bibliography; and bibliographies are mysterious, not to say all-too-human, things, reflecting their compilers' penchants and obsessions, and - what is even more to the point - being most commonly based on other bibliographies.
For there is a law more fundamental than any other regarding authorial canons. We may call it the law of inertia, and it is to the effect that it is infinitely easier to get a work into an author's canon than to get it out again. It is characteristic of bibliographies (that apparently so inhuman and impersonal genre) that they assign works to an author with magisterial laconism. This means that, when another bibliographer wishes to improve and correct his predecessor's list, he may be in the dark (almost as much so as the uninstructed reader) as to how a work got into the list in the first place. Thus, if he has no positive reason for questioning the attribution, and because life is short, he will be tempted to give it the benefit of the doubt, presuming that his predecessor knew of some good reason for making it. Then, when a third bibliographer comes along, the attribution will appear to have already two supporter. . . . The process, as may be seen, is cumulative.
We have mentioned Defoe, and of no English author is it more true that his oeuvre or canon is the construction of bibliographers. Simple statistics make this plain. He died in 1731, and in the first substantial list of his writings - that by George Chalmers, published in 1790 - he was credited with a little more than a hundred works. Two centuries later, in the second edition of John Robert Moore's Checklist of the Writings of Daniel Defoe (1971) - which has for some time now been the standard authority - the total has swollen to 570! It is a dizzying edifice of attribution, for which we have, essentially, to thank six men: George Chalmers (1742-1825), Walter Wilson (17811847), William Lee (1812-c. 1890), James Crossley (1800-1883), W. P. Trent (18621939), and J. R. Moore (1890-1973).
Some ten years ago we became fascinated by this phenomenon of the burgeoning Defoe canon. It made us want to find out more about those six bibliographers and the way they had gone to work. We also wanted to study the method of transmission between them - especially since, as we soon discovered, they nursed considerable contempt for one another's methods. It has been an absorbing pursuit, full of human as well as other kinds of interest. But it has led us to an alarming, or at any rate awkward, opinion: that is, that almost half of the items in Moore's Checklist may have no business being there at all.
We must not be misunderstood here. Our grounds for casting doubt on these attributions are by no means all of the same kind. Some of the items in Moore's list have, since his day, been convincingly ascribed to another author. In the case of others, there seems to be good evidence, whether external or internal, to suggest they could not be, or are most unlikely to be, by Defoe. But the largest class is of works that, on stylistic or other grounds, could plausibly be thought to be by Defoe, were there anything, any tiny scrap of external evidence, to connect them with him. But since, so far as we know, there is not, we take the view that they ought not to be assigned to him - and above all not given unquestioned status as his. The fact that a work seems as if it might well be by a given author is, in itself, no reason at all for actually assigning it to him or her. Also, to assign a work to an author a century, or two centuries, after his death, and on internal evidence alone - which is the case with the majority of these attributions - seems a rash enterprise anyway.
We have come to the view that Defoe is a rather special case in English literature and that a bibliography of him needs to contain two categories: definite and certain attributions and merely probable ones. If a work is to qualify for the former category, there must be some piece of external evidence connecting it. with Defoe (as well, of course, as convincing internal evidence). For the latter category, on the other hand, it may be sufficient if there is really strong internal evidence - with the all-important proviso, however, that the epithet "probable" should mean precisely what it says and not be a mere synonym for "possibly by Defoe" or "often attributed to Defoe."
But it may be asked, does it greatly matter if a merely plausible attribution is made - made, as it were provisionally, till the question of its authorship can be decided more firmly? Well yes, we think it does. For let us imagine that the work in question is in a genre, or on a topic, not normally associated with Defoe. Its presence in his bibliography will infallibly tempt someone to argue that, if he wrote that work, he was very likely the author of some other work, or works, in the same genre or on the same topic. Thus before long, insensibly, the world will have acquired a new Defoe (as it might be, a Defoe highly knowledgeable about Swedish affairs, or a crime-reporting Defoe); and then the next thing will be that, because of this, his biography will require rewriting. (It should be noted that all the great architects of the Defoe canon, with the exception of James Crossley, also wrote biographies of him.) The danger, of course, is that one will end up with a Defoe of whom it is impossible to make sense - and this, it seems to us, is exactly what has happened.
In the early 1980s, having together become interested in Defoe, we began reading our way through some of his lesser-known writings, and we became puzzled by a pamphlet, reputedly by him, entitled A Vindication of the Press. The author of this rather platitudinous tract, published in 1718, describes himself as a young author, a devout Churchman, and an enthusiast for the theater - all of which Defoe, a Dissenter in his late fifties with a lifetime prejudice against the stage, was not. This, we told ourselves, must no doubt be one of the disguises of which - as one was so often told - the protean Defoe was fond. Still, just out of curiosity, we thought we would find out how and when the attribution to Defoe was first made and reassure ourselves that it was quite solid.
It turned out that the pamphlet was listed - for the first time, apparently - by the American bibliographer W. P. Trent in his Defoe chapter in the Cambridge History of English Literature (1912), subsequently reappearing in the Cambridge Bibliography of English Literature (1940), in J. R. Moore's Checklist (1971), and in the New Cambridge Bibliography of English Literature (1971). A facsimile edition of it had been brought out in 1951 by the Augustan Reprint Society, edited by Otho Clinton Williams. But, so far as we could discover, no bibliographer had ever explained in print why the work was thought to be by Defoe - which, considering that the ascription was so very belated, two hundred years having passed since the work was written, one might surely have expected.
So we studied one or two more slightly questionable-looking pamphlets, to find out which of Defoe's bibliographers had been the first to assign it to him, and to try to establish what grounds (if any) they had offered for doing so. This in turn aroused our interest in these bibliographers themselves. They turned out to be colorful men, remarkably different in what they were looking for in Defoe, and highly idiosyncratic in their methods. But just how idiosyncratic it took...