The detective story, with its persistent emphasis on the one correct solution of a crime, is the most resolutely end-oriented of narrative modes. Given the financial pressures of writing, however, it is hardly surprising that authors of detective stories should occasionally seek to postpone that end by turning their short stories into novels. What is most interesting in this process is not questioning the author's motivation nor judging relative degrees of success but rather identifying what is added to a detective story when it becomes a detective novel. Structural analysis has shown that detective stories, long and short, employ a common narrative pattern: a crime, usually a murder, is committed by some person unidentified to the reader; a detective, amateur or professional, becomes interested in the crime; he examines the evidence, interviews the suspects, and finally announces his conclusions, his explanation forming the climax of the narrative. Because even short detective stories normally include all these elements, detective novels must logically include something more, something not dictated by the common structure.

Structural analysis shows what makes a given novel a detective novel without showing what makes it a novel, what makes it long. It may seem captious to raise such a question: most readers, after all, think of stories as just long enough to compass their material and to reach their conclusion. Our ability to summarize novels and short stories in a few sentences or paragraphs suggests that a given narrative can always be described briefly in terms of its structure, but narratives seem to have an optimal length which exceeds that of their summaries. Boccaccio begins the Decameron with a summary, in a sentence or two, of each of his hundred tales; the summaries, despite their accuracy, are not themselves satisfactory tales. Henry James's forty-page “Project” for The Ambassadors is of great interest for a student of James without being an adequate substitute for the novel. From the reader's point of view, a plot summary of this form—“a crime is committed by an unknown person; a detective investigates and discovers the criminal”—is not worth reading; from the writer's point of view, it is not worth writing; it is not tellable, to use Mary Louise Pratt's term,¹ unless something is added.

Narratives, then, tend to have an ideally tellable length. Plot summaries are too short to be tellable (even though James's summary of The Ambassadors is longer than most short stories), and everyone can remember stories, oral or written, which were too long to be properly tellable. The nature of an author's material evidently dictates, if not an exact number of pages, at least a certain scale, ten pages or two hundred or a thousand. The events of Proust's masterpiece could be compressed into a single volume, but not without changing it into a different kind of work, a work more centrally concerned with events as events. The transformation of detective stories into detective novels is peculiar precisely because it belies all the intuitive rules about tellability. A detective story is tellable at a length of twenty pages; after its expansion, it is tellable at a length of two hundred pages; and in both cases it seems to be the same kind of narrative. Granted, not every reader accepts the longer version as tellable—detective novels, whether or not they have been fashioned from short stories, are notoriously subject to the charge of padding—but many readers do; and determining the exact degree of success or failure is only peripheral to analyzing the process of expansion as such. What does an author do to a tellable detective story to make it a tellable detective novel?²

The most obvious way to make a novel out of a detective story is to build it out of several stories, in the manner of Raymond Chandler. Three of Chandler's novels—The Big Sleep (1939), Farewell, My Lovely (1940), and The Lady in the Lake (1944)—each combine plots from two or three shorter stories. In The Big Sleep, for example, one plot, involving pornography and the blackmailing of Carmen Sternwood, leads directly to the second, involving the disappearances of Rusty Regan and Mrs. Eddie Mars. In Farewell, My Lovely, two plots, concerning the murder of Lindsay Marriott and the quest for Moose Malloy's old flame Velma, are made into a search for the same character, so that the solutions are identical, and a third plot, involving police corruption in Bay City, provides a background and a pivotal episode.³ Chandler is able to combine several stories into a novel because his hard-boiled formula employs such a limited number of situations and characters—the shadowy underworld kingpin, the eccentric old patriarch, the beautiful but treacherous heroine, the incorruptible wise-cracking detective, the hired gunman waiting outside the door—that each story, like the commedia dell 'arte, presents further adventures of the same cast. By combining stories, Chandler achieves wider social reference: what appear, in the beginning of The Big Sleep and Farewell, My Lovely, to be crimes committed against private citizens by habitual criminals are later seen to have roots in the social structure itself, in General Sternwood's family or Mr. Merwin Lockridge Grayle's marriage or the corruption of legal institutions and their officers. More generally, by including several stories in a single novel, Chandler changes his focus from the sequential narrative of crime and detection to the evocation of a world in which blackmail, kidnapping, theft, and violent death are normal occurrences.

This is not to say that Chandler's novels are more realistic than his short stories: in fact, the opposite is the case. When George Grella supports his contention that “although the hard-boiled writers set out to write tough, contemporary mysteries in modern colloquial language, they ultimately wrote romantic rather than realistic fiction” by tracing the stylistic “evolution of the American detective thriller from the tough action story of the pulps to the very literary novel of the American romance tradition,”⁴ he significantly opposes the realism of hard-boiled stories to the romantic stylization of hard-boiled novels. When Philip Marlowe first comes to the Sternwood house in The Big Sleep, he sees over the entrance doors “a broad stained-glass panel showing a knight in dark armour rescuing a lady who was tied to a tree and didn't have any clothes on. … I stood there and thought that if I lived in the house, I would sooner or later have to climb up there and help him.”⁵ This image is more appropriate to the novel than its component stories for several reasons. First and most obviously, Chandler clearly brings a portentousness, almost a pretentiousness, to a work whose audience is not limited to the readership of Black Mask and other pulp magazines. Second, the analogy to courtly romance—an ironic analogy, as Grella points out—is far more appropriate to Marlowe's involvement with the Sternwood sisters than to the adventures of his predecessors among the crooked cops and small-time hoods of the short stories. Edward A. Nickerson obliquely makes this point when he observes that “Chandler gives us a host of secondary crooks who seem right out of everyday life, but supplies a primary murderer who is larger than life.”⁶ The stylized romantic analogues, together with the elaborate similes and metaphors and the meditative introspection characteristic of Chandler's novels, seldom appear in his short stories. Chandler's adoption of a more consciously literary style in his novels may have had an element of snobbery, but the multiple plots of The Big Sleep justify the image of knight and lady as an integrative principle. The short stories, which move toward a sequential resolution of events, do not require the thematic integration necessary for a novel incorporating several plots. The world of Chandler's novels is organized in terms of violence, retribution, discovery, and justice as themes rather than actions.

Agatha Christie never uses violence to define a world, as Chandler does: even when she expands her short stories, she does not combine or significantly elaborate her original plots but simply displays more fully their possibilities for suspicion and intrigue. In The Mystery of the Blue Train (1928), based on the short story “The Plymouth Express,” and in Dead Man's Mirror (1931), based on “The Second Gong,” she retains the murderer's plot, involving in both cases a fundamental deception or red herring, and the clues, and adds a prologue which gives further background for the crime, expands the number of innocent suspects and takes more pains to multiply the reader's suspicions, adds further romantic complications, and describes in more detail an upper-class milieu which the murder disrupts. These additions may seem extraneous to the original plot, but their purpose is simply to exploit more fully the dramatic revelations implicit in the plot, the consequences of those revelations, and the reader's uncertainty about the suspects. In The Mystery of the Blue Train Christie involves the heroine romantically with two men, one of whom is the murderer, in order to display her plot's potential for suspicion and suspense to greatest advantage. The puzzle is the same as in the shorter version, but the reader's sympathies are more fully engaged by its implications. In “The Second Gong,” which contains an unusually elaborate plot for a short story, the characters are defined almost entirely in terms of their respective situations: there is no room for any further development. Dead Man's Mirror, which is three times as long as the earlier story, follows it closely in narrative plan—both versions consist mainly of a series of interrogations followed by a revelation—but allows more room for the characters. Christie continues to use the same stock figures the shorter story implies: the elderly, eccentric, imperious victim, his plausible nephew, his willful adopted daughter, and so on. In the longer version, however, these characters are displayed more fully so that they make a stronger, although not a fundamentally different, impression. The difference is not between simple and complex or between flat and round characters, for Christie's characters are flat at any length. It is a difference which might more aptly be described by noting the potential of characters in Dickens and P. G. Wodehouse (whose manner Christie has surely copied in drawing her minor figures) for histrionic display, which allows us to spend more time with them and so to become more aware of the distinctive qualities of their presence without significantly modifying our original perceptions of them. I might indicate what Mr. Pickwick was like by describing him as an unusually guileless man whose innocence kept getting him into trouble, but his story would never be worth telling or reading unless it offered Dickens numerous opportunities for putting Pickwick on stage, for letting him act himself out, for displaying his amusing possibilities. It is in this sense, the sense of animating a known and unchangeable quantity, that Dead Man's Mirror allows Christie to display her characters more fully.

Anthony Berkeley, working within the same conventions as Christie, expands his story “The Avenging Chance” into a novel, The Poisoned Chocolates Case (1929), by exploding the conventions through a reductio ad absurdum. The Poisoned Chocolates Case presents a club, obviously modeled on Berkeley's own Detective Club, whose members act as armchair detectives in attempting to solve a murder whose facts are set forth in “The Avenging Chance.” Each member refutes the previous member's solution and offers his own instead. The seven solutions offered include Berkeley's original solution, but this time it is not finally correct. The Poisoned Chocolates Case is thus at once an exploration or anatomy of all the logical possibilities of the original situation and a parody of the conventions—the disinterested detective, the power of logic, the unsuspected solution—on which it depends.

Despite the much greater length of the novel, which is ten times as long as the story, Berkeley follows his own original closely with regard to facts and phrases. His principal additions are in his roster of detectives (his original detective becoming now one of six), in the multiple twists given earlier plot devices, and in the Crimes Circle Club's extended discussions of motives, relations, and alibis. Each detective presents not only a new theory of the crime but new revelations in evidence: Sir Eustace Pennefather was in fact about to be divorced; he was carrying on an affair with Joan Bendix; he hoped to marry Mrs. Bendix as soon as possible; and so on. Although each successive theory incorporates material from previous theories, the theories as a group tend less to provide deeper insight into a given set of events and relations (as do the narratives, for example, of The Sound and the Fury) than to cancel each other out. When the final theory is confirmed in a way that rules out the possibility of retribution, its authority is simultaneously vindicated and undermined, as befits the conclusion to a novel which, by presenting the conventions of formal or Golden Age detective fiction in their most extreme and schematic form, calls into question not only their mimetic value but their formal potency. If “The Avenging Chance” is a classic detective story, The Poisoned Chocolates Case is a novel about detection, a novel whose focus is less the problem to be solved than the interminable comedy of problem-solving.

One final example is less easy to classify: Helen McCloy's novel Through a Glass, Darkly (1950), based on her story of the same title. Again the plot and clues are identical. Reports of a young woman's eerie doppelgänger drive her away from her job and friends and finally mystify investigators of her fatal heart attack. The mystery in both versions is solved by Dr. Basil Willing, McCloy's didactic and super-rational psychologist-detective, and his presence dominates the shorter version, most of which describes his investigation into the mystery. In the longer version, however, McCloy is far more concerned to establish sympathy for her victim, who does not die until the book is three-quarters over (although McCloy adds an earlier murder to raise the level of suspense). The leading qualities of the novel—a sense of supernatural foreboding and dread, pathos for the victim, outrage over the manner of her death—are nearly absent from the short story.

In his attempt to place formulaic literature in a wider literary context, John G. Cawelti has identified two opposed psychological needs readers bring to fiction: the needs for the excitement of vicarious experience and for the reassurance of a stable and ordered world. Cawelti concludes that “in reading or viewing a formulaic work, we confront the ultimate excitements of love and death, but in such a way that our basic sense of security and order is intensified rather than disrupted, because, first of all, we know that this is an imaginary rather than a real experience, and, second, because the excitement and uncertainty are ultimately controlled and limited by the familiar world of the formulaic structure.”⁷ More specifically, the formulaic conventions of the detective story provide two kinds of pleasure, opposed but intimately connected: the pleasure of a vicarious exposure to violence, death, and irrationality, and the pleasure of a rational explanation, of knowing the reason for everything. The strength of the second pleasure depends on the strength of the first. McCloy transforms her story into a novel by emphasizing all its irrational elements: hints of the supernatural, a sympathetic presentation of unreasoning fear, and the resistance of the evidence to any logical explanation. The attention to suspects is minimal because suspicion is directed not toward innocent suspects but toward possibilities beyond all reason. The balance between rational and irrational elements is sufficiently altered to make the novel a different kind of narrative from the short story.

This last observation applies in a general sense to all the novels I have been considering. In each case the author turns his story into a novel by emphasizing the provisional suspension of reason. In his novels, Chandler presents a world in which deceit, corruption, and violence are so pervasive that the solution of any one mystery is thematically inconclusive. Christie emphasizes the potential of her original situation for bewilderment and surprising revelation and projects characters whose identities, however superficially established, are clouded by suspicion. Berkeley renders all rational solution absurd by the profusion of solutions he offers. If each novel ends with the triumph of justice and reason, that triumph is always more muted and provisional than in the corresponding story.

This conclusion may seem obvious from the nature of the material: surely, of two versions of the same narrative, the longer one will be more deeply implicated in particular motives and circumstances and hence less completely committed to a definitive resolution. In this case, however, it is surprising how many obvious complications most writers ignore in turning their detective stories into novels. They do not ordinarily further elaborate the plot, either in the sense of making the crime more obscure or of detailing further the process by which the detective discovers the truth. They do not as a rule develop their characters significantly: Philip Marlowe may seem exceptional in this respect, but the gumshoe with the soul of a knight-errant is a figure well within the compass of the short story. Detective story writers do not, with the partial exception of Christie, impute more complex emotional states to their characters or invent incidents that will reveal their unsuspected depths—as, for example, Dostoevsky is always doing, even in his shorter work. What makes most detective novels long, whether or not they have been adapted from short stories, is the addition of something which will fill the space between the posing of the problem and its solution: a series of red herrings (as in the early novels of John Dickson Carr and Ellery Queen and in Christie's The Murder of Roger Ackroyd); digressions in subject or tone (Carr's novel The Blind Barber and Dashiell Hammett's The Maltese Falcon) or atmosphere in the sense of local color (Queen's Calamity Town); an auxiliary plot involving the characters' suspicions of each other (the Had-I-But-Known novels of Mary Roberts Rinehart and others), the consequences of the police's suspicions (Erle Stanley Gardner's Perry Mason novels), the detective's attempts to get information (Rex Stout's Nero Wolfe novels), the detective's interference in the affairs of the suspects (Queen's The Four of Hearts), or the detective's romantic involvements (E. C. Bentley's novel Trent's Last Case and several novels by Dorothy L. Sayers, especially Gaudy Night and Busman's Honeymoon); or, finally, the simple multiplication of corpses (of which the limiting case is Christie's And Then There Were None). The avoidance of psychological complexity or of a series of revelations in themselves expressive and thematically significant (as in Oedipus the King) in favor of a more fully imagined world or a heightening of uncertainty and suspense suggests that insofar as a detective story emphasizes the logical solution of a mystery it is essentially a short form which can be made longer only if some new element or emphasis is added. Ellery Queen's surprisingly incantatory insistence that “the short form … was, is, and forever will be the true form”⁸ of the detective story is justified by the observation that detective stories are made longer primarily by becoming something else.

This conclusion has implications which range beyond the detective story, short or long. Many critics have concurred with Dorothy L. Sayers' assertion that the detective story is of all modern literary genres the most Aristotelian because of the preeminence it assigns to plot and the correspondence of discovery or revelation and reversal to privileged moments in detective fiction.⁹ Aristotle's term for revelation, anagnorisis, indicates a movement from ignorance to knowledge on the part of the tragic hero and, perhaps, of the audience as well, the corresponding moment in the detective story being of course the identification of the culprit by the detective and, perhaps, the reader. But this analogy is valid only up to a point. Aristotelian tragedy, which bases its vision of human experience and human identity on its revelation of the teleology of human action, is essentially a gnostic mode: even an audience who already know the secret of Oedipus' guilt will share his moment of anagnorisis and see him as he really is, a man defined by his own actions. Detective stories, which share a characteristic movement from mystery to a univocal and authoritative revelation and to the retrospective establishment of a rational order, are the most gnostic of contemporary literary genres: their presentation of authoritative wisdom is indeed their most fundamental affinity with Aristotelian tragedy. Even detective novels, however, though on the whole more gnostic than most novels, place less emphasis than detective short stories on the vindication of rationality and greater emphasis on violence, injustice, and the suspension of reason. It is not only detective novels whose teleology is less important, whose wisdom is less authoritative, than detective stories: this pattern holds true more generally for novels and short stories as such.

It would take many pages to illustrate this assertion persuasively; I can give only a few nondetective examples here. Henry James is an author who often makes use of Aristotelian revelations—so often, in fact, that a good deal of his work reads like highly etherealized detective stories in which, as H. G. Wells complained, “people nose out suspicions, hint by hint, link by link.”¹⁰ In such short stories as “Greville Fane” (1892), “The Real Thing” (1892), “Paste” (1899), and “The Tree of Knowledge” (1900), the revelations have both closural and teleological force: James's surrogate consciousness and audience see the world (or the constate dramatic situation) as it really is. In James's longer tales, revelations continue to have a closural force but are thematically equivocal, inconclusive, or ironic. Daisy Miller's early death is rendered more pathetic by Winterbourne's inability to appreciate her value; the narrators of “The Aspern Papers” (1888), “The Liar” (1888), and “The Figure in the Carpet” (1896) are unsuccessful in their attempts to establish and to make public revelations about the dead Jeffrey Aspern, the living Colonel Capadose, and the general pattern in Hugh Vereker's timeless fiction; John Marcher, in “The Beast in the Jungle” (1903), spends his whole life waiting for an experience which turns out to be a revelation of his spiritual barrenness. In his novels, James's revelations are still more attenuated or contingent in their teleology. Christopher Newman's discovery of the Bellegardes' secret in The American (1877) cannot secure his marriage to Claire de Cintré; Isabel Osmond makes a similar discovery about her husband only to return to him at last; the narrator of The Sacred Fount (1901) spends several hours entertaining Mrs. Briss with his climactic speculations about their fellow guests only to have her dismiss them as groundless. When Longmore accidentally recognizes Richard de Mauves with his current mistress in “Madame de Mauves” (1874), the scene tells him everything he needs to know about Madame de Mauves's marriage, but when Strether, in a scene whose design and whose place in the structure of The Ambassadors (1903) recall the earlier scene, recognizes Chad Newsome and Madame de Vionnet as the lovers in a boat he is watching, he has learned far too much to accept this revelation as final: it simply vindicates Mrs. Newsome's view of the facts of the case. The difference here is in part a difference between early and late James; but given the general conclusiveness of the revelations in the later short stories compared to those in the earlier novels, it seems more fundamentally a difference between short stories and long. As in the detective novels I have been considering, revelation in James's longer fiction functions as closural device rather than anagnorisis; its thematic valence often complicates rather than simplifies.

A glance at Ernest Hemingway, an author whose work depends less than James's on revelations, suggests why this might be so. Stories like “Big Two-Hearted River,” “The Killers,” and “The Short Happy Life of Francis Macomber” present themselves first as recitations of events: Nick gets off the train, then hikes up-country, then makes camp. Readers trying to integrate these chains of events into intelligible wholes look for moments of illumination which will provide a rationale for the narrated events or the narrating discourse. Just as detective stories move from a false, misleading, or incomplete perception of the truth to a revelation of the whole truth, readers tend to impute a deferred teleological rationale to Hemingway's deliberately uninflected recitals. In Hemingway, however, the teleology is again contingent, ironic, or epicentral: we are never sure whether Mrs. Macomber killed her husband deliberately or what Ole Anderson had done in Chicago. Hemingway's wisdom, as in “A Clean, Well-Lighted Place” and “The Snows of Kilimanjaro,” is always worldly wisdom: there is no hidden truth or revelation that makes the world intelligible; given the ultimate resistance of the nonhuman world to individual action, Hemingway can only provide information about that world and our possible responses to it. The rhythm of his stories is accordingly marked by disillusionment rather than ripening wisdom; anagnorisis produces not an authoritative teleology but rather a debunking of what has gone before. Hemingway's method, like that of detective novels, is less gnostic than agnostic: the audience is left not in a state of knowledge but of enlightened ignorance.

That short stories, whether or not they are detective stories, should be more gnostic than novels, more fundamentally concerned with the wisdom of authoritative teleology, is not after all surprising. Suzanne Hunter Brown, in reading the same episode from Tess of the d'Urbervilles as a short story and as part of a novel, has recently concluded that “when a reader regards this text as a short story, he tends to prefer repetitive to temporal links in creating it as a meaningful structure. … A reader is likely to order the minimal units of [short stories] synchronically, whereas he will probably order textual elements from [novels] in a diachronic fashion.”¹¹ Given the brevity and concentration of short stories, it is thoroughly logical that readers should integrate the sequences of events they present in terms of a gnostic teleology, a rationale whose authority is not primarily the authority of linear progression. What is surprising is that novels should incorporate a specifically antiteleological principle which makes their closure necessarily contingent and which prevents their events from being integrated except in terms of temporal sequence. Although the short story is customarily treated as a deviation from the unmarked length of narrative (our critical vocabulary, which includes the term “short story” rather than “long novel,” seems to establish novel length as normative), the teleological affinities of short stories with Aristotelian tragedy suggest that the novel is itself the deviant form, the mode of discourse that, by suspending or attenuating the authority of anagnorisis, mystifies, bewilders, and unteaches readers even as it enlightens them. Nowhere is this contradictory double rhythm more oddly apparent than in the detective novels fashioned from short stories, novels which necessarily, by virtue of their length and circumstantial complexity, call into question the foreordained revelations which allow us to read them at all.

Notes

1. Toward a Speech Act Theory of Literary Discourse (Bloomington: Indiana University Press, 1977), pp. 136-138 and passim.

2. To adapt a short story to novel length is more frequent than the reverse. But my argument holds mutatis mutandis for those short stories based on earlier novels—for example, John Dickson Carr's “The Incautious Burglar” (originally “Guest in the House,” 1947), based on Carter Dickson's (Carr's) The Gilded Man (1942).

3. For further details of Chandler's reworking of his short stories in composing his novels, see Philip Durham's Introduction to Killer in the Rain (Boston, MA: Houghton Mifflin, 1964), the posthumous collection of stories Chandler “cannabilized” for his novels.

4. “Murder and the Mean Streets,” Contempora, 1 (March 1970), 15.

5. The Raymond Chandler Omnibus (London: Hamish Hamilton, 1953), p. 3.

6. “‘Realistic’ Crime Fiction: An Anatomy of Evil People,” Centennial Review, 25 (Spring 1981), 109.

7. Adventure, Mystery, and Romance (Chicago, IL: University of Chicago Press, 1976), p. 16.

8. Queen's Quorum (Boston, MA: Little, Brown, 1951), p. 114.

9. See “Aristotle on Detective Fiction,” in Unpopular Opinions (London: Gollancz, 1946), pp. 178-190.

10. Boon (London: Unwin, 1915), p. 108.

11. “‘Tess’ and Tess: An Experiment in Genre,” Modern Fiction Studies, 28 (Spring 1982), 37.