As someone who has spent nearly 60 years as a professional writer, I am inevitably set in my ways, though capable of changing them radically in a crisis. But I recognise that my ways are not typical, that there is no such thing as a typical writer. Starting early is for me axiomatic (it is 6.45 a.m. as I write these words). It was for Trollope too, who paid his groom an extra sum annually for bringing him a scalding cup of coffee as dawn was breaking. And I, like Trollope, start writing immediately. By contrast, J.B. Priestley told me that he needed anything up to an hour fiddling with the objects on his desk before actually beginning the process of putting words on to paper, though once begun he wrote steadily.
A morning start, however, is beyond many writers. Tom Stoppard, I know, rarely gets going before midday, or even later. But then he can write in the evening, which I hate, and often into the small hours. I read that Honoré de Balzac habitually wrote from midnight until noon the next day. Can one believe this? Twelve hours’ continual writing is an immense physical effort. I certainly cannot do more than eight, and normally, when writing a big book, six hours’ productive writing is as much as I can manage, and the effort leaves me weary. But so much about Balzac would be incredible did not hard evidence exist to prove it true. His record in producing, I think, 92 novels in little over 20 years, was only possible by putting in 12-hour days.
I do not have actual figures for Balzac’s regular output of words, though they could easily be worked out. Trollope knew exactly how many he could do in an hour, and had his watch placed on his desk to check, every quarter, how he was progressing and whether his speed was being kept up. That kind of industrial efficiency makes many writers shudder, and Trollope’s Autobiography in which his working methods were revealed, did his reputation harm when it came out. But the alternative is much more painful. Flaubert had no difficulty in working regular, sensible hours but his output was often so meagre as to be scarcely measurable — a word an hour on average during a dark spell. You get this feeling of agonising costiveness if you examine his manuscripts in the Bibliothèque Nationale. He was envious of his friend George Sand, from whom, he said in a letter to her, ‘words flow steadily and copiously like a great river’. She was almost in the Balzac class. Not so fluent, because more of a perfectionist, was Dickens. There is, in New York’s Pierpont Morgan Library, the original manuscript of his A Christmas Carol, his most successful shorter tale, over which he took much trouble. They produce a beautifully bound facsimile of it, which I recommend as a generous Christmas present, and worth studying. This reveals a number of crossings-out, some major, at the rate of about 15 a page. But the underlying writing is smooth and steady, giving no evidence of compositional agonies. It is clear that Dickens could, and often did, write at speed, when in the throes of a tale that tested his genius to the utmost. He writes in a letter of being ‘in a frenzy of Copperfield’. And he would jump up from his chair to pull faces in the looking-glass, then sit down again quickly to describe them in words, to breathe life into one of his characters. I sometimes think of these strange authorial activities (often unconscious, for we owe the face-pulling description not to Dickens himself but to one of his daughters, who observed it secretly) when I am reading one of his novels.
Graham Greene said he might produce 300 words in a working day, and that 500 words was a fair day’s output. That is slow going in my opinion. Evelyn Waugh would do 2,000 words in a day when properly launched, and perhaps even a little more. That seems to have been his typical speed when writing Brideshead Revisited. But of course he did not write every day. There were gaps, for all sorts of reasons. That is why he needed to go to his special boarding-house in Devon, to ‘get away’. It is important, in a major work, for a writer to keep up a regular rhythm, to enjoy consistency and continuity. It is usually possible for an experienced writer to spot a major interruption in the work of a fellow-writer. There were some big ones, as we know from other evidence, in the sad life of Charlotte Brontë, caused by illness and death in her family. One occurred after she was a third of the way through Jane Eyre — perhaps the finest work of sustained writing in the whole of English fiction — and it is never quite the same afterwards. I suspect something happened in Jane Austen’s life after she had written the first 3,000 words or so of Mansfield Park, a brilliant beginning, which I strongly recommend to all aspirant writers of fiction, as an exemplary way of opening up a story. There was then a pause, and when resumed there is a faint but perceptible difference in tone.
There is no virtue in sheer quantity of words, as the case of Edgar Wallace suggests. Another prodigy in this respect was Jean-Paul Sartre, who, when stirred by a political event, and egged on by his acolytes, was capable of trotting out 20,000 words, a kind of cavalry-division verbal charge, in a single day and night, for one of his ephemeral pamphlets. Even the most productive writers, having poured out words in a torrent, tended to have second thoughts at a later stage in the process of book-making. Balzac was a terror to compositors, and to publishers, for he worked like a demon on his proofs, cutting, condensing, substituting and rewriting. Once the copy is first set in type, the author is charged for such changes (beyond correcting ‘literals’, as they are called). Balzac’s income from his books was severely reduced by the gigantic size of the printer’s bills he was thus charged.
It says a lot about the vagaries of genius that the greatest writers have totally differing views about revision. Shakespeare, it is said, ‘never blotted a line’. It poured forth from his copious fountain, and that is the unaltered nectar we taste today, four centuries later. I suspect the same is true for Chaucer (though not for Dante). On the other hand, Kipling, once a writer who wrote effortlessly to order, to fill holes in the Indian newspapers, gradually allowed celebrity and greater discernment to take over his pristine output. He became a relentless cutter of his texts. Whether his cuts were more substantial at the manuscript or the proof stages I do not know — a study of this topic would be welcome — but the result was a reduction by up to a half, or even more, of the original text. This is why some of the later short stories are so hard to understand at first reading, though some might, and do, call them miracles of compression. Tennyson had a different method. He would write a poem, fiddle with it, send it to the printer, and then keep the proof by him, often for years, reading it to friends perhaps but without publishing it. This is the kind of behaviour which enrages everyone else in the business of producing literature. But then writing is a difficult trade. And writers are bound to be difficult people.