They must be with him as he lies down to sleep, and as he wakes from his dreams. He must learn to hate them and to love them… He must know of them whether they be cold-blooded or passionate, whether true or false, and how far true and how far false. The depth and the breadth, and the narrowness and the shallowness of each should be clear to him.

And however well I think I know my characters, they reveal themselves more clearly during the writing of the book, so that at the end, however carefully and intricately the work is plotted, I never get exactly the novel I planned. It feels, indeed, as if the characters and everything that happens to them exists in some limbo of the imagination, so that what I am doing is not inventing them but getting in touch with them and putting their story down in black and white, a process of revelation, not of creation. But the process of creation remains mysterious. One writer who has attempted to explain it is E. M. Forster. The well-known passage may be a little high-flown, a little exaggerated in the importance Forster ascribes to the subconscious, but it comes with the authority of the author of A Passage to India, and I think most artists, whatever their medium, feel that it gets close to at least part of the truth.

What about the creative state? In it a man is taken out of himself. He lets down as it were a bucket into his subconscious, and draws up something which is normally beyond his reach. He mixes this thing with his normal experiences, and out of the mixture he makes a work of art… And when the process is over, when the picture or symphony or lyric or novel (or whatever it is) is complete, the artist, looking back on it, will wonder how on earth he did it. And indeed he did not do it on earth. [E. M. Forster, The Raison d’Etre of Criticism in the Arts]

DESPITE PROGNOSTICATIONS that the detective story, particularly in its classical form, is already outworn and doomed to die, it remains obstinately alive, and it is perhaps not surprising that during the decades since the Golden Age those critics not susceptible to its attractions have been vocal in their disparagement, complaining that the educated readers to whom detective fiction appeals-they include some illustrious names-should know better. Some of this aversion has been from readers who dislike detective fiction as others might dislike science fiction, romantic novels or stories in which the protagonist is a child. The field of fiction is rich and remarkably wide and we all have our favourite pastures.

“This is criminal-two wrong spellings and improper use of a semicolon.”

One critic who was impervious to the charms of the genre was Edmund Wilson, who in 1945 published an influential essay entitled “Who Cares Who Killed Roger Ackroyd?” As Mr. Wilson had constantly been exposed to animated discussions on the merits of mystery writers, he enquired of aficionados what author they recommended him to try, and set out conscientiously to justify or modify his prejudice. His correspondents were almost unanimous in recommending Dorothy L. Sayers and placing her novel The Nine Tailors at the top of their reading list. After skipping what he described as “conversations between conventional English village characters,” “boring information on campanology” and “the awful whim sical patter of Lord Peter,” he reached the conclusion that The Nine Tailors was one of the dullest books he had encountered in any field. No doubt, thus filleted, it was.

Mr. Wilson and others of his ilk are certainly entitled to their preferences, and no efforts on the part of their friends are likely to change their minds. And much criticism still relates primarily to the Golden Age: the old argument that the story dominates over any interest in characterisation or setting and is frequently unconvincing; that the basic morality of the genre is strongly right-wing, upholding the right of the privileged against the dispossessed, in which working-class characters are little better than caricatures; and that detective fiction, so far from showing compassion to either victim or murderer, glories in a crude form of communal vengeance. In general these criticisms are so inappropriate to the majority of detective stories being written today that there is little point in refuting them. But a more interesting criticism made during the thirties still echoes in the minds of twenty-first-century critics. Its chief proponent was an influential American critic, Professor Jacques Barzun, who enjoyed detective stories but only those which, like the books of Agatha Christie, confined themselves to the pure puzzle. For him and those who agreed with him, the conventional mystery which relied on logical deduction, and in which the characters solved the plots from observed facts, had an intellectual and literary integrity which was lost if writers attempted to wade through the murky pools of abnormal psychology or to probe the psychological basis of their characters’ actions and personalities. In short, these critics feared that the detective story might be getting above itself.

Somewhat surprisingly, Dorothy L. Sayers, who in Gaudy Night made theme and characterisation dominant over the plot, went some way to justify this view in her essay “Aristotle on Detective Fiction,” published in 1946, taking the great philosopher as her authority.

One may string together a series of characteristic speeches of the utmost finish as regards diction and thought, and yet fail to produce the true dramatic effect; but one will have much better success with a story which, however inferior in these respects, has a plot… The first essential, the life and soul so to speak of the detective story, is the plot and the characters come second.

Very few detective novelists would hold this view today, or hold it so uncompromisingly. Their aim-and it is mine-is to write a good novel with the virtues those words imply, a novel which is at the same time a credible and satisfying mystery. This means that there must be a creative and reconciling correlation between plot, characterisation, setting and theme, and so far from the plot being dominant, it should arise naturally from the characters and the place.

Another ethical criticism of the detective story is that it has at its heart an appalling crime and the suffering of innocent people, and uses them to provide popular entertainment. In Sayers’s novel Gaudy Night, Miss Barton, one of the Shrewsbury College tutors, challenges Harriet Vane about the morality of the books she writes. Surely the sufferings of innocent suspects ought to be taken seriously? To this Harriet replies that she does indeed take them seriously in real life, as must everyone. But was Miss Barton saying that anyone who had tragic experience of sex, for example, should never write an artificial drawing-room comedy? Although there was no comic side to murder, there could be a purely intellectual side to the detection. I myself would argue that it is possible to deal with the intellectual side of the detection while portraying with compassion and realism the emotional trauma of all the characters touched by this ultimate crime, whether as suspect, innocent bystander or indeed the perpetrator. In an Agatha Christie novel the crime is solved, the murderer arrested or dead, and the village returned to its customary calm and order. This does not happen in real life. Murder is a contaminating crime and no life which comes into close touch with it remains unaltered. The detective story is the novel of reason and justice, but it can affirm only the fallible justice of human beings, and the truth it celebrates can never be the whole truth any more than it is in a court of law.