"According to My Private Canons It Is Not a Pure Detective Story"

THOMAS STEARNS ELIOT (1888-1965) is usually acknowledged as having a profound influence on poetry in the first half of the 20th century, but like many other intellectuals he had his guilty pleasures, including a genuine fondness for the detective fiction genre:

   That T. S. Eliot, of all people, was a devoted fan of the genre must have rankled [snobbish super-highbrow critic Edmund] Wilson in particular. Eliot, the author of famously difficult and formidably learned poems, whose every critical pronouncement was seized upon by dons and converted into doctrine, was an unimpeachable authority in matters of literary judgment. Wilson, indeed, had played a part in establishing Eliot’s reputation as such, having gushed, in his era-defining study “Axel’s Castle” (1931), that the poet-critic had an “infinitely sensitive apparatus for aesthetic appreciation”—a sensitivity presumably not worth squandering on something as puerile and formulaic as mysteries.
   But, as scholars like David Chinitz have pointed out, Eliot’s attitude toward popular art forms was more capacious and ambivalent than he’s often given credit for. — Paul Grimstad (see below)

~ ~ ~ ~ ~ ~ ~ ~ ~ ~

After middlebrow literary critic Gilbert Seldes dared to publish a detective novel, The Victory Murders (1927), albeit under the pseudonym of "Foster Johns," T. S. Eliot had a chance to let Seldes know what he believed constituted a "pure detective story":

   12 April 1927
   The New Criterion

   My dear Seldes,
     I was glad to get your letter of the 31st March, and particularly because it clears up the mystery of your mystery stories. It is needless to say that I had read 'The Victory Murders' within 24 hours of receipt, and had been speculating on the identity of the author without success. I have enjoyed the book very much; my only criticism is that according to my private canons it is not a pure detective story, but a mixed detective and adventure story. That is to say, in a pure detective story there are no adventures after the first chapter; the book is entirely concerned with the accumulation, selection and construction of evidence about something which has already happened. In your story things keep happening. This is by no means a disadvantage; it is merely a nice point of definition. I am perfectly willing to admit that the pure detective story is extremely rare; the most austere example of the type is of course The Case [sic] of Marie Roget.

~ ~ ~ ~ ~ ~ ~ ~ ~ ~

By the way, a contemporary ad for The Victory Murders describes it this way:

   A mystery story of International interest set in France, England, and America. The motive for the deaths of famous actresses which always occur or are threatened on Armistice Night is carefully concealed until the end of the story.

About this time last year an article appeared highlighting Eliot's admiration for mysteries:

"What Makes Great Detective Fiction, According to T. S. Eliot."
By Paul Grimstad.
First appearance: The New Yorker, February 2, 2016.
Article (6 pages).
Online (HERE).

Possibly the most important litterateur of the last century owed more to Sherlock Holmes than some would like to admit:

   A key tenet of Golden Age detection was “fair play”—the idea that an attentive reader must in theory have as good a shot at solving the mystery as the story’s detective. To establish parameters of fairness, Eliot suggests that “the character and motives of the criminal should be normal” and that “elaborate and incredible disguises” should be banned; he writes that a good detective story must not “rely either upon occult phenomena or . . . discoveries made by lonely scientists,” and that “elaborate and bizarre machinery is an irrelevance.” The latter rule would seem to exclude masterpieces like Doyle’s “The Adventure of the Speckled Band,” which involves a murder carried out by a snake trained to shimmy through a heating duct, then down a bell rope whose tassel extends to the victim’s pillow. But Eliot admitted that most great works broke at least one of his rules. He in fact adored Arthur Conan Doyle, and was given to quoting long passages from the Holmes tales verbatim at parties, and to borrowing bits and ideas for his poems.

~ ~ ~ ~ ~ ~ ~ ~ ~ ~

Eliot reviewed detective stories fairly often; here's a famous one from his own publication in which he lays out his own, ahem, criteria for what should comprise a good detective story:

   ‘Homage to Wilkie Collins: An omnibus review of nine mystery novels’
   New Criterion, January 1927

   The D’Arblay Mystery, by R. Austin Freeman; The Footsteps that Stopped, by A. Fielding; The House of Sin, by Allen Upward; The Diamond in the Hoof, by Traill Stevenson; The Dangerfield Talisman, by J. J. Connington; The Mysterious Disappearances, by G. McLeod Winsor; Footsteps in the Night, by C. Fraser-Simson; The Bishops Park Mystery, by Donald Dike; The Massingham Butterfly, by J. S. Fletcher

   During the last year or two the output of detective fiction has increased rapidly. I presume that detective fiction is successful, with a rising demand; otherwise one or two such thrillers would not appear on nearly every publisher’s list. It might be interesting to speculate on the reasons for this increased demand, but our conclusions would be undemonstrable. What can be shown, and is of interest in itself, is that the increased demand and competition is producing a different, and as I think a superior type of detective story; that some general rules of detective technique may be laid down; and that, as detective fiction observes the rules of the game, so it tends to return and approximate to the practice of Wilkie Collins. For the great book which contains the whole of English detective fiction in embryo is The Moonstone; every detective story, so far as it is a good detective story, observes the detective laws to be drawn from this book. The typical English detective story is free from the influence of Poe; Sherlock Holmes himself, and in spite of his numerous progeny, is in some important respects a sport. I say the ‘typical’ English detective story, because I believe that the crime fiction of every country has its own national character: it would be interesting, in this connexion, to show how French crime stories – notably Arsène Lupin and Jacques Rouletabille – may be derived from The Count of Monte-Cristo in the same way that English fiction is derived from The Moonstone; but that would lead us too far.

   A detective story cannot be analysed like other fiction: the reviewer must not reveal the plot, or the reader will be robbed of his pleasure. I have therefore arranged the fiction here ‘reviewed’ – a small, but I dare say representative selection from the season’s product – as nearly as possible in what I think the order of merit. The Massingham Butterfly must be considered hors de concours, as it proved to be merely a collection of unrelated short stories of detective type; they are too slight to deserve reprinting, but suggest that Mr Fletcher’s longer detective stories are probably very good. The two preceding (Footsteps in the Night and The Bishops Park Mystery) are not properly detective stories either, because they have no detectives; therefore they are technically of little interest. All of the rest have some merit: all of them violate, as Wilkie Collins never violates, some obvious rule of detective conduct.

". . . technically of little interest."

   I do not know how many of these rules can be formulated; the following are drawn up from my study of the stories above, and other recent stories, and the list ‘does not pretend to completeness’. Every one of these stories commits one of these faults; they are, between one story and another, more or less heinous or excusable:

   (1) The story must not rely upon elaborate and incredible disguises. We accepted them from so engaging a character as Holmes, as we accept them from the more farcical Lupin: but we consider them to be trick work. Disguises must be only occasional and incidental: here Wilkie Collins is impeccable. Elaborate double lives, in disguise, are an exaggeration of this vice: Arsène Lupin disguised for four years as the head of the Paris police, and actually being the head of the police, is admirable fooling. But in general it is reprehensible. But for a device of this sort, The Footsteps that Stopped would be the best of our list.

   (2) The character and motives of the criminal should be normal. In the ideal detective story we should feel that we have a sporting chance to solve the mystery ourselves; if the criminal is highly abnormal an irrational element is introduced which offends us. If the crime is not to have a natural motive, or is without motive altogether, we feel again that we have been tricked. But for this fault, another story on my list would have been placed higher than The D’Arblay Mystery. No theft, for instance, should be due to kleptomania (even if there is such a thing).

   (3) The story must not rely either upon occult phenomena, or, what comes to the same thing, upon mysterious and preposterous discoveries made by lonely scientists. This, again, is the introduction of an irrational element: ghosts, influences, strange elements with terrifying properties (‘the destruction of the atom’ will probably flourish for several years in bad detective stories) are all in the same category. Writers of this sort of hocus-pocus may think that they are fortified by the prestige of H. G. Wells. But observe that Wells triumphs with his scientific fiction just because he keeps within the limits of a genre which is different from the detective genre. The reality is on another plane. In detective fiction there is no place for this sort of thing. Two of our list fall through this sin.

   (4) Elaborate and bizarre machinery is an irrelevance. Detective writers of austere and classical tendencies will abhor it. Some of the Sherlock Holmes stories make far too much of stage properties. Writers who delight in treasures hid in strange places, cyphers and codes, runes and rituals, should not be encouraged. But we must distinguish carefully. In Poe’s Gold Bug the cypher is good, because we are given the legitimate intellectual exercise of its explication; there is nothing sham or meretricious about it, but the gold bug itself, and the skull, are unnecessary and childish trappings. But in The Moonstone, the Indian business (though I fear it has led to a great deal of bogus Indianism, fakirs and swamis, in crime fiction) is perfectly within the bounds of reason. Collins’s Indians are intelligent and resourceful human beings with perfectly legitimate and comprehensible motives.

". . . perfectly within the bounds of reason."

   (5) The detective should be highly intelligent but not superhuman. We should be able to follow his inferences and almost, but not quite, make them with him. It is perhaps in the Detective that the contemporary story has made the greatest progress – progress, that is to say, back to Sergeant Cuff. I am impressed by the number of competent, but not infallible professionals in recent fiction: Scotland Yard, or as it is now called, the C. I. D., has been rehabilitated. The amateur detective no longer has everything his own way. Besides the C. I. D. Inspector, another type is successful: the medical scientist whose particular work brings him into touch with crime. But Mr Freeman’s Dr Thorndyke and Mr Upward’s Dr Tarleton are also unpretentious professionals; they have personality, but are without extraneous trappings. One of the most brilliant touches in the whole of detective fiction is the way in which Sergeant Cuff, in The Moonstone, is introduced to the reader (in the narrative of the butler Betteredge). He is unimpressive, and dreary. But suddenly, while he is talking to the gardener about roses, ‘Hello,’ he says, ‘here’s a lady coming. Might it be Lady Verinder.’ Now Betteredge and the gardener had reason to expect Lady Verinder, and from that direction; Cuff had not. Betteredge begins to think better of Cuff. It is not that Cuff has superhuman powers; he has a trained mind and trained senses.

". . . progress, that is to say, back to Sergeant Cuff."

   One of the most promising of the younger detectives is Inspector Gilmour in The Mysterious Disappearances. He is all the better for being rather a disagreeable person, and his peculiar talents and limitations are very lifelike. If the author will abandon his scientific trickery (error 3 above) there is a brilliant future for Inspector Gilmour. Of all of the stories named above, The D’Arblay Mystery is the most perfect in form. The second is the most remarkable in its complication of plot; and only towards the end becomes improbable; the third is also first-rate work. The rest are inferior to these three.

~ ~ ~ ~ ~ ~ ~ ~ ~ ~

Resources:
- Of course Wikipedia (HERE) has a comprehensive article about T. S. Eliot, but you can find still more discussions about Eliot and mysteries in:
   ~ Curtis Evans, "Eliot Elucidates: T. S. Eliot's Detective Fiction Rules," The Passing Tramp (June 18, 2015) (HERE)
   ~ C. V. Weaver, "T. S. Eliot’s Rules for Detective Fiction" (August 19, 2014) (HERE)
- He may have been a critic with an elevated brow, but Gilbert Seldes was also a fan of detective fiction; go to ONTOS (HERE) and (HERE).
- And several excerpts from Eliot's article "Sherlock Holmes and his Times" (1929) are 
online (HERE).