Chances are that if people know anything about that towering twentieth-century literary figure, T. S. Eliot, as a detective fiction critic, it is the fact that the great poet, playwright and essayist praised Victorian writer Wilkie Collins’ stolen gem mystery The Moonstone as “the first, the longest and the best of modern English detective novels.” Yet T. S. Eliot had much more to say about detective fiction than those admiring words he afforded Wilkie Collins’ esteemed novel, even going so far as to devise a set of rules for the writing of it. Although he regarded The Moonstone as the pinnacle of English mystery fiction, he nevertheless appreciated as well the work of more modern crime writers from the “Golden Age” of the detective novel (roughly 1920 to 1939), including the “Queen of Crime” herself, Agatha Christie. He also highly praised the Sherlock Holmes tales of Sir Arthur Conan Doyle, who published the last of his tales about the Great Detective in 1927. Like many a mystery reader in the Golden Age, T. S. Eliot had catholic taste in crime fiction and enjoyed both character-driven mysteries like The Moonstone and those which have been dismissed as “mere puzzles:” detective stories depending primarily on the mechanical cleverness of their plots to grab readers.
In “Wilkie Collins and Dickens,” a 1927 essay that was central in the modern Wilkie Collins literary revival, T. S. Eliot made his original critique of The Moonstone. Although he lauded Collins’ mystery as the first and greatest English detective novel, Eliot pointed out immediately after making this assertion that he had deliberately used the adjective English as a qualifier. In the field of mystery fiction, noted Eliot significantly, there also were to be found the stories of the American Edgar Allan Poe, with what he termed their “pure detective interest.” In Poe’s deft hands, the detective story became “something as specialized and intellectual as a chess problem.” Puzzle-oriented detective fiction of this sort relied for its appeal less on “the intangible human element” of The Moonstone and more on “the beauty of the mathematical problem.” For his part Eliot was drawn both by the compelling human drama of The Moonstone and the enticing logic puzzles found in much of Twenties mystery fiction.
In The Criterion, the prestigious literary journal he founded and edited, Eliot had a forum where he could share his fascination with detective fiction and its aesthetics. Between 1927 and 1929 Eliot in its pages reviewed thirty-four mystery novels and short story collections, as well as two works on true crime. Like a kind of highbrow pope he lent detective fiction, at a crucial time in its development as an art form, the considerable cachet of his intellectual benediction.
T. S. Eliot’s mystery criticism debuted in the January 1927 issue of The Criterion, with a review essay concerning no less than nine mystery novels and short story collections—what he called “a small, but I dare say representative, selection from the season’s product.” Eliot used the occasion of this review essay primarily to offer his formulation of five rules “of detective conduct” in mystery stories. Surprisingly, although Eliot’s enjoys great prominence in literary history and the rules he promulgated preceded into print those by the detective novelists Willard Huntingdon Wright (an American who wrote mystery bestsellers as “S. S. Van Dine”) and Ronald Knox (an English Catholic priest and founding member of England’s Detection Club), Eliot’s rules are far less well known than those by Wright and Knox (just as his mystery criticism is far less well known than that of Dorothy L. Sayers).
Here are Eliot’s rules in brief:
“The story must not rely upon elaborate and incredible disguises….Disguises must be only occasional and incidental.”
“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.”
“The story must not rely either upon occult phenomena or, which comes to the same thing, upon mysterious and preposterous discoveries made by lonely scientists.”
“Elaborate and bizarre machinery is an irrelevance….Writers who delight in treasures hid in strange places, cyphers and codes, runes and rituals, should not be encouraged.”
“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.”
Like the later rules of S. S. Van Dine and Ronald Knox, Eliot’s strictures emphasize the importance of maintaining “fair play” (“we should feel that we have a sporting chance of solving the mystery ourselves”). To do this, Eliot advocates a prohibition on outré devices: incredible disguises; insanity; occult phenomena and fantastic science; and elaborate and bizarre machinery, such as cyphers and codes, runes and rituals. Testing the nine mystery works in his review essay against these rules, Eliot concluded that of them R. Austin Freeman’s detective novel The D’Arblay Mystery was “the most perfect in form” (despite one violation).
Six months later, in June 1927, Eliot for a second review essay gathered seventeen more books, fifteen works of fictional mystery and two concerning true crime. In this second essay, Eliot noted that he had “found it necessary to discriminate between books that are detective fiction proper and those which may better be termed mystery stories.” Eliot carefully explained what he saw as the difference between the two sorts of tales: “In the detective story nothing should happen: the crime has already been committed, and the rest of the tale consists of the collection, selection and combination of evidence. In a mystery tale the reader is led from fresh adventure to fresh adventure. In practice, of course, most detective stories contain a few events, but these are subordinate, and the interest lies in the investigation.” Here, Eliot is distinguishing the detective novel from what soon enough came to be most commonly known as the “thriller” or “shocker” story associated with such writers as Edgar Wallace, Sax Rohmer and “Sapper” (H. C. McNeile). Thrillers often made use of the sort of outlandish apparatus Eliot had proscribed in his earlier rules for detective fiction.
Eliot enunciated an important principle: detective novels needed to emphasize either the puzzle element or the human element, because trying to fully treat both at once posed too great a challenge.Of the eleven genuine detective novels and short story collections that Eliot included in his review essay, Eliot singled out for particularly glowing praise S. S. Van Dine’s debut detective novel, The Benson Murder Case, writing: “The Benson Murder Case is extremely well built, the criminal is well concealed, and the author shows great ingenuity in his use of evidence.” Toward the end of the review essay, Eliot enunciated an important principle: detective novels needed to emphasize either the puzzle element or the human element, because trying to fully treat both at once posed too great a challenge. “Without dispraise of any individual writer we may be allowed to complain that modern detective fiction in general is weak in that it fails between two possible tasks,” Eliot explained. “It has neither the austerity, the pure intellectual pleasure of Poe’s Marie Roget, nor has it the fullness and abundance of life of Wilkie Collins. We often wish that the majority of our detective writers would concentrate on the detective interest, or take more trouble and space over the characters as human beings and the atmosphere in which they live.”
This view would be echoed over twenty years later by the great hardboiled mystery writer Raymond Chandler, in his unpublished but percipient essay “Twelve Notes on the Mystery Story,” wherein Chandler wrote: “The perfect detective story cannot be written. The type of mind which can evolve the perfect problem is not the type of mind that can produce the artistic job of writing.”
Eliot’s last major detective fiction review essay in The Criterion, which appeared in April 1929, compared two of the mystery genre’s past masters, Arthur Conan Doyle, world famous creator of Sherlock Holmes, and Anna Katharine Green, American author of the pioneering and bestselling crime novel, The Leavenworth Case (1878). Green’s Leavenworth Case had recently been reprinted in England and Eliot, who as a boy had read the novel (along with Doyle’s Sherlock Holmes stories) around the turn of the century, quickly reread it, only to be disappointed. Why, he wondered in the pages of The Criterion, did the Sherlock Holmes stories, “in spite of their obvious defects from our present high standards of detective fiction, reread so much better than The Leavenworth Case?”
Essentially, Eliot concluded, because in Leavenworth Anna Katharine Green had allowed sentimental melodrama to overwhelm the detective interest. In a mystery novel a writer like Wilkie Collins, possessing “a wider gift for drama and fiction,” could combine a puzzle with plenteous human interest.
***
Eliot’s remaining Criterion pieces on detective fiction are much shorter reviews of specific works that avoid broader pronouncements on the genre. Yet these too have considerable interest, for they reveal that Eliot had a half-dozen favorite authors of modern detective fiction, whom he “recommended to the small, fastidious public which really discriminates between good and bad detective stories.” These authors were R. Austin Freeman, Freeman Wills Crofts, S. S. Van Dine, J. J. Connington, Agatha Christie and Lynn Brock. Eliot had comparatively little to say about Agatha Christie and Lynn Brock, but he went into more detail on the four other authors.
As early as June 1927, Eliot speculated that “Mr. Freeman and Mr. Croft [sic]…seem to be our two most accomplished detective writers.” On two other occasions in The Criterion, Eliot bracketed Freeman and Crofts as the finest modern mystery novelists. Austin Freeman, a contemporary of Arthur Conan Doyle who had created Sherlock Holmes’ greatest rival in the form of the brilliant medical jurist Dr. John Thorndyke, had produced his first detective novel as far back as 1907; yet, unlike Doyle, he was an extremely prolific producer of mystery novels and short stories during the Golden Age. Besides singling out Freeman’s The D’Arblay Mystery in his January 1927 review article as the “most perfect in form” of the nine books reviewed, Eliot commented in his June 1927 review essay that he regretted having no Freeman novel on hand to assess, as the author “has more of the Wilkie Collins abundance than any contemporary writer of detective fiction.”
In Freeman Wills Crofts, a railway engineer turned detective novelist who was a meticulous plotter (the acknowledged king of the “unbreakable alibi” story) but rather an indifferent literary stylist, Eliot could glimpse little of that “Wilkie Collins abundance.” In the critic’s view, however, Crofts did not stand in need of that abundance, for he had other gifts valuable to a spinner of mystery tales. In contrasting Crofts with Anna Katharine Green, Eliot noted of the former writer: “Mr. Crofts, at his best, as in The Cask [his celebrated 1920 debut mystery novel], succeeds by his thorough devotion to the detective interest; his characters are just real enough to make the story work; had he tried to make them more human and humorous he might have ruined his story. The love-interest, in The Cask, is a postulate; it does not have to be developed, and puts no strain upon the author.” Eliot indicated in this same 1929 article that Crofts had authored his favorite mystery tale of recent years, Inspector French and The Starvel Tragedy (1927).
The two detective novelists most reviewed by T. S. Eliot in The Criterion, however, were not R. Austin Freeman and Freeman Wills Crofts, but, rather, S. S. Van Dine and J. J. Connington. S. S. Van Dine, the lone American among Eliot’s top-ranked mystery writers, had all three of his earliest detective novels—The Benson Murder Case (1926), The Canary Murder Case (1927) and The Greene Murder Case (1928)—reviewed by Eliot in The Criterion. As mentioned above, Eliot rated The Benson Murder Case quite highly. He was similarly admiring of The Canary Murder Case, singling out “the poker game scene near the end of the book” as “really brilliant.” Eliot’s main criticism was that “one small but vital point” in the tale turned on a mechanical contrivance. He disliked such devices in detective novels, believing them fanciful and unrealistic.
T. S. Eliot’s mystery criticism in The Criterion reveals the great writer as a representative twenties detective fiction fan, one still amused with and mentally stimulated by the ingenuity of the puzzles crime authors were devising.Of The Greene Murder Case, Eliot deemed it “a good detective novel,” though he had greater reservations with it than he had had with either Benson or Canary. The critic again complained of Greene, as he had of Canary, that a vital part of the murderer’s scheme depended on “a piece of ingenious and improbable mechanics.” Eliot speculated that Van Dine’s root problem in this respect was “a tendency to over-elaboration which characterises American detective fiction.” In Eliot’s view the “great weakness” of what he called “the over-elaborate school” of mystery fiction was that when “you build up a very complicated plot, you are likely to descend to a mechanical solution.” Additionally he complained that Van Dine in Greene “comes very near to surrendering to the pathological school of detective fiction,” with the tale’s culprit barely escaping “the American psychopathic ward.” Despite these concerns, however, Eliot maintained that “Mr. Van Dine remains…in the first rank of detective writers, a little lower than Mr. Freeman and Mr. Crofts.”
J. J. Connington, the pseudonym of Scottish chemistry professor Alfred Walter Stewart, had three novels reviewed by T. S. Eliot in the pages of The Criterion: The Dangerfield Talisman (1926), Murder in the Maze (1927) and Tragedy at Ravensthorpe (1927). Although Eliot wrote that he could not “speak with any enthusiasm” of Connington’s The Dangerfield Talisman, he was surprised and pleased to find the author’s Murder in the Maze “a really first-rate detective story.” Eliot praised Connington’s inspired conception for the setting of the novel’s initial double slayings: “The very idea of murder in a box hedge labyrinth does the author great credit, and he makes full use of its possibilities.” However, he was most delighted of all with Connington’s adept clueing in Murder in the Maze. “What makes the book particularly excellent in its kind is that fact that we are provided early in the story with all of the clues which guide the detective. One ought to know who committed the crime after reading a very few chapters; but one does not.”
The impressed Eliot concluded that with Murder in the Maze J. J. Connington had “made a place for himself in the front row of detective story writers.” However, later that year (1927) he deemed Connington’s newest novel, Tragedy at Ravensthorpe, “not quite up to Murder in the Maze.” Though Eliot thought Ravensthorpe “a very good tale,” its author in the critic’s view had made two key miscalculations in constructing the plot. Nevertheless Eliot continued to deem Connington a formidable maker of mysteries.
T. S. Eliot’s mystery criticism in The Criterion reveals the great writer as a representative twenties detective fiction fan, one still amused with and mentally stimulated by the ingenuity of the puzzles crime authors were devising. Although he believed, as often has been pointed out by partisans of the self-consciously literary “crime novel,” that Wilkie Collins was the ne plus ultra in English mystery, he did not demand that all authors of detective fiction adhere to Collins’ human interest model. In addition to admiring Collins, Eliot also held in great esteem Edgar Allan Poe and his “mathematical,” or “chess-like,” puzzle stories, which emphasized logic rather than human emotions; and he thought it perfectly legitimate for mystery writers to follow Poe’s method more than that of Collins. Far from despising puzzle-oriented detective fiction, as the late crime writer P. D. James claimed in her 2009 book Talking about Detective Fiction was the common view of between-the-wars intellectuals, T. S. Eliot admired it, read it devoutly and desired, like James, to talk about it.
__________________________________
A Fatal Score of T. S. Eliot’s Favorite Mysteries
__________________________________
For “the small, fastidious public which really discriminates between good and bad detective stories”:
“The Mystery of Marie Roget (1842),” Edgar Allan Poe
Le Crime d’Orcival (1867), Emile Gaboriau
The Moonstone (1868), Wilkie Collins
The Adventures of Sherlock Holmes (1892), Sir Arthur Conan Doyle
The Cask (1920), Freeman Wills Crofts
The Deductions of Colonel Gore (1924), Lynn Brock
The Murder of Roger Ackroyd (1926), Agatha Christie
The D’Arblay Mystery (1926), R. Austin Freeman
The Colfax Bookplate (1926), by Agnes Miller
The Cathra Mystery (1926), Adam Gordon Macleod
The Footsteps That Stopped, A. Fielding (Dorothy Feilding) (1926)
The Benson Murder Case (1926), S. S. Van Dine
The House of Sin (1926), Allen Upward
The Canary Murder Case (1927), by S. S. Van Dine
The Crime at Diana’s Pool (1927), Victor L. Whitechurch
Inspector French and the Starvel Tragedy (1927), Freeman Wills Crofts
Murder in the Maze (1927), J. J. Connington
Tragedy at Ravensthorpe (1927), J. J. Connington
The Greene Murder Case (1928), S. S. Van Dine
The Death of Laurence Vining (1928), Alan Thomas
