Greg Forster has written to challenge Alister McGrath’s insistence that we correct the date of C. S. Lewis’s conversion. McGrath bases this correction on the lack of evidence for it in Lewis’s life and writings from that period. If Lewis had really converted in 1929, McGrath wonders, wouldn’t he have said something about a deity when his father died later that year? Wouldn’t his behavior have changed immediately, as it did the following year? Wouldn’t the proof, in other words, be in the pudding?
Forster dismisses this reading, arguing that belief takes time to turn into meaningful action. He suggests Lewis might have been ashamed, or that his belief had not yet moved fully to the “God of the Scriptures,” but remained an intermediate “intellectual conversion” that was not yet “heart-changing.” At the end of his article, however, Forster concedes: “McGrath has done a great deal of historical study that I haven’t done, so I’m perfectly ready to be convinced by him. But he’ll have to come up with better evidence than he has so far presented.” Thankfully, we have some.
In an unpublished manuscript that Walter Hooper refers to as “Early Prose Joy”—a draft that would later become Surprised by Joy, and on which Andrew Lazo has done work at the Marion E. Wade Center to be published in the forthcoming issue of VII: An Anglo-American Literary Review—Lewis correctly dates his conversion to a specific three-week period, not in “Trinity Term of 1929,” as he wrote in Surprised by Joy, but in 1930. He had not misremembered in “Early Prose Joy”; he had simply made a transcription error when copying from it.
“That God encountered him” is the main thing, McGrath says, and I am inclined to think he is right. But in scholarship, dates do matter. The Victorian poet Alexander Smith (1830–1867) was pilloried for plagiarizing Tennyson’s Idylls of the King with his Edwin of Deira when both were released in 1861, though evidence now shows that Smith’s work had actually been written earlier. Auden’s poems turn on the hinge of his own conversion; that story is upset if we slide the hinge.
The problem with much Lewis scholarship is that we trust the author too much. I am not suggesting, with Barthes, that we kill him, but that we trust our own acumen and readerly good sense more readily, even when they threaten to melt that most holy of sacred cows: authorial intent. I suspect that, casting about for firm ground on which to stand in the absence of a reliable canonicity of taste, scholars hold up as a final source the one source they take to be originary and treat it as the end of discussion.
This philosophy lies at the root of most textual studies in our time, with many negative consequences. As I have had occasion to say in a recent article, when an author makes what is obviously a sloppy mistake, we do not owe it to posterity to preserve the error in what will become the authoritative edition of her works. Not every misplaced capital is fraught with meaning, nor every spelling (or dating) error an intentional subversion of critical norms.
One critical debate in which this disciplinary weakness appears most egregiously is Charles Robinson’s much-vaunted new parallel edition of The Original Frankenstein, which purports to solve the dilemma of Frankenstein’s collaborative authorship—raised bravely, if a bit quixotically, by John Lauritsen in The Man Who Wrote Frankenstein. Robinson’s method is to print in one version “the words [Percy Shelley] wrote into [Mary Shelley’s] draft of the novel”; in the second, he “removes all of these PBS words and restores MWS’s more colloquial words that he had canceled.”
The trouble with this approach is that we know for sure Mary was a paid copyist—she had worked transcribing Byron’s poems before—and we know that Percy used stenographers to whom he would dictate, including Mary, frequently. There is no reason to think that the words in Mary’s handwriting are “hers,” or “authorial,” any more than we imagine the Byron poems in her handwriting to be. There is no reason to suppose that she was not simply making a fair copy of her husband’s scrawl, which draft she then consigned to the fire and which proofs he then corrected in his own hand. The point is not to claim that Percy composed the novel (a suggestion that earns one execution without a trial), but that a volume predicated on a notion of authorial intent based on manuscript copy imagines that it settles a matter when in fact it leaves the issue no clearer than it was before.
A second Lewisian debate muddied by exactly this sort of practice is the ordering of the Narnia books. HarperCollins reordered the series for its 1980 edition, moving from the order in which the books were written and published to a kind of “internal order” in which The Magician’s Nephew, book six of the seven, somehow comes first, since it concerns the creation of Narnia and acts as a prequel to the events in the other stories.
Paul Ford has already described in Companion to Narnia how several scholars have weighed in against this decision. There are very good reasons for doing so, not least that The Magician’s Nephew says, “You may have read a book called The Lion, the Witch and the Wardrobe”—which, if this is the first book in the series, the reader will not have done, leaving her with the feeling that she has already missed part of the story (as, if she reads the reordered version, she has). In fact, The Magician’s Nephew assumes we know all kinds of things about Narnia, Aslan, and how magic and animals work, assumptions that change the reading experience quite drastically.
The reason for this intrusion—for upsetting the internal balance, points of reference, and creation of suspense, and for divorcing modern readers from the experience of the generations who encountered the stories in their proper order—is authorial intent. Lewis tosses off in a letter to a child that he supposes, following the child’s suggestion, that The Magician’s Nephew really ought to come first, which gives a kind of blessing to the procedure. It does so, however, only if we believe that letters occupy a more original, more authorial voice than, say, a publication history, or the readerly and critical good sense that McGrath follows in making his conversion estimates.
That the facts unearthed by Lazo support McGrath’s position is convenient, but not necessary to a judgment. We should, sometimes, know better than the author what he or she means. That is part of the artistic conversation, and it is a position with which Lewis agreed in that same letter to the same child, when he wrote that an author “is not necessarily the best, and is never a perfect, judge” of such things, and when he said—echoed too by Alister McGrath—“perhaps it does not matter very much.”
This essay was originally published in First Things.
