| 
  • If you are citizen of an European Union member nation, you may not use this service unless you are at least 16 years old.

  • You already know Dokkio is an AI-powered assistant to organize & manage your digital files & messages. Very soon, Dokkio will support Outlook as well as One Drive. Check it out today!

View
 

Notes on the Detective Story by Raymond Chandler

Page history last edited by Jon 14 years, 5 months ago

Notes on the Detective Story by Raymond Chandler

 

1. It must be credibly motivated, both as to the original situation and denouement; it must consist of the plausible actions of plausible people in plausible circumstances, it being remembered that plausibility is largely a matter of style. This requirement rules out most trick endings and a great many "closed circle" stories in which the least likely character is forcibly made over into the criminal, without convincing anybody. It also rules out such elaborate mises-en-scène as Christie's Murder in a Calais Coach, where the whole setup for the crime requires such a fluky set of happenings that it could never seem real.

 

2. It must be technically sound as to the methods of murder and detection. No fantastic poisons or improper effects from poison such as death from nonfatal doses, etc. No use of silencers on revolvers (they won't work) or snakes climbing bellropes ("The Speckled Band"). Such things at once destroy the foundation of the story. If the detective is a trained policeman, he must act like one, and have the mental and physical equipment that go with the job. If he is a private investigator or amateur, he must at least know enough about police methods not to make an ass of himself. When a policeman is made out to be a fool, as he always was in the Sherlock Holmes stories, this not only deprecates the accomplishment of the detective but it makes the reader doubt the author's knowledge of his own field. Conan Doyle and Poe were primitives in this art and stand in relation to the best modern writers as Giotto does to da Vinci. They did things which are no longer permissible and exposed ignorances that are no longer tolerated. Also, police art, itself, was rudimentary in their time. "The Purloined Letter" would not fool a modern cop for four minutes. Conan Doyle showed no knowledge whatever of the organization of Scotland Yard's men. Christie commits the same stupidities in our time, but that doesn't make them right. Contrast Austin Freeman, who wrote a story about a forged fingerprint ten years before police method realized such things could be done.

 

3. It must be honest with the reader. This is always said, but the implications are not realized. Important facts not only must not be concealed, they must not be distorted by false emphasis. Unimportant facts must not be projected in such a way as to make them portentous. (This creation of red herrings and false menace out of trick camera work and mood shots is the typical Hollywood mystery picture cheat.) Inferences from the facts are the detective's stock in trade; but he should disclose enough to keep the reader's mind working. It is arguable, although not certain, that inferences arising from special knowledge (e.g., Dr. Thorndyke) are a bit of a cheat, because the basic theory of all good mystery writing is that at some stage not too late in the story the reader did have the materials to solve the problem. If special scientific knowledge was necessary to interpret the facts, the reader did not have the solution unless he had the special knowledge. It may have been Austin Freeman's feeling about this that led him to the invention of the inverted detective story, in which the reader knows the solution from the beginning and takes his pleasure from watching the detective trace it out a step at a time.

 

4. It must be realistic as to character, setting, and atmosphere. It must be about real people in a real world. Very few mystery writers have any talent for character work, but that doesn't mean it is not necessary. It makes the difference between the story you reread and remember and the one you skim through and almost instantly forget. Those like Valentine Williams who say the problem overrides everything are merely trying to cover up their own inability to create character.

 

5. It must have a sound story value apart from the mystery element; i.e., the investigation itself must be an adventure worth reading.

 

6. To achieve this it must have some form of suspense, even if only intellectual. This does not mean menace and especially it does not mean that the detective must be menaced by grave personal danger. This last is a trend and like all trends will exhaust itself by overimitation. Nor need the reader be kept hanging onto the edge of his chair. The overplotted story can be dull too; too much shock may result in numbness to shock. But there must be conflict, physical, ethical or emotional, and there must be some element of danger in the broadest sense of the word.

 

7. It must have color, lift, and a reasonable amount of dash. It takes an awful lot of technical adroitness to compensate for a dull style, although it has been done, especially in England.

 

8. It must have enough essential simplicity to be explained easily when the time comes. (This is possibly the most often violated of all the rules). The ideal denouement is one in which everything is revealed in a flash of action. This is rare because ideas that good are always rare. The explanation need not be very short (except on the screen), and often it cannot be short; but it must be interesting in itself, it must be something the reader is anxious to hear, and not a new story with a new set of characters, dragged in to justify an overcomplicated plot. Above all the explanation must not be merely a long-winded assembling of minute circumstances which no ordinary reader could possibly be expected to remember. To make the solution dependent on this is a kind of unfairness, since here again the reader did not have the solution within his grasp, in any practical sense. To expect him to remember a thousand trivialities and from them to select that three that are decisive is as unfair as to expect him to have a profound knowledge of chemistry, metallurgy, or the mating habits of the Patagonian anteater.

 

9. It must baffle a reasonably intelligent reader. This opens up a very difficult question. Some of the best detective stories ever written (those of Austin Freeman, for example) seldom baffle an intelligent reader to the end. But the reader does not guess the complete solution and could not himself have made a logical demonstration of it. Since readers are of many minds, some will guess a cleverly hidden murder and some will be fooled by the most transparent plot. (Could "The Red Headed League" ever really fool a modern reader?) It is not necessary or even possible to fool to the hilt the real aficionado of mystery fiction. A mystery story that consistently did that and was honest would be unintelligible to the average fan; he simply would not know what the story was all about. But there must be some important elements of the story that elude the most penetrating reader.

 

10. The solution must seem inevitable once revealed. This is the least often emphasized element of a good mystery, but it is one of the important elements of all fiction. It is not enough merely to fool or elude or sidestep the reader; you must make him feel that he ought not to have been fooled and that the fooling was honorable.

 

11. It must not try to do everything at once. If it is a puzzle story operating in a rather cool, reasonable atmosphere, it cannot also be a violent adventure or a passionate romance. An atmosphere of terror destroys logical thinking; if the story is about the intricate psychological pressures that lead apparently ordinary people to commit murder, it cannot then switch to the cool analysis of the police investigator. The detective cannot be hero and menace at the same time; the murderer cannot be a tormented victim of circumstance and also a heavy.

 

12. It must punish the criminal in one way or another, not necessarily by operation of the law. Contrary to popular (and Johnston Office) belief, this requirement has nothing much to do with morality. It is a part of the logic of detection. If the detective fails to resolve the consequences of the crime, the story is an unresolved chord and leaves irritation behind it.

 

Addenda

 

1. 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. It would be nice to have Dashiell Hammett and Austin Freeman in the same book, but it just isn't possible. Hammett couldn't have the plodding patience and Freeman couldn't have the verve for narrative. They don't go together. Even a fair compromise such as Dorothy Sayers is less satisfying than the two types taken separately.

 

 2. The most effective way to conceal a simple mystery is behind another mystery. This is literary legerdemain. You do not fool the reader by hiding clues or faking character à la Christie but by making him solve the wrong problem.

 

3. It has been said that "nobody cares about the corpse." This is bunk. It is throwing away a valuable element. It is like saying the murder of your aunt means no more to you than the murder of an unknown man in an unknown part of a city you never visited.

 

4. Flip dialogue is not wit.

 

5. A mystery serial does not make a good mystery novel. The "curtains" depend for their effect on your not having the next chapter to read at once. In book form these curtains give the effect of a false suspense and tend to be merely irritating. The magazines have begun to find that out.

 

6. Love interest nearly always weakens a mystery story because it creates a type of suspense that is antagonistic and not complementary to the detective's struggle to solve the problem. The kind of love interest that works is the one that complicates the problem by adding to the detective's troubles but which at the same time you instinctively feel will not survive the story. A really good detective never gets married. He would lose his detachment, and this detachment is part of his charm.

 

7. The fact that love interest is played up in the big magazines and on the screen doesn't make it artistic. Women are supposed to be the targets of magazine fiction and movies. The magazines are not interested in mystery writing as an art. They are not interested in any kind of writing as an art.

 

8. The hero of the mystery story is the detective. Everything hangs on his personality. If he hasn't one, you have very little. And you have very few really good mystery stories. Naturally.

 

9. The criminal cannot be the detective. This is an old rule and has once in a while been violated successfully, but it is sound as it ever was. For this reason: the detective by tradition and definition is the seeker after truth. He can't be that if he already knows the truth. There is an implied guarantee to the reader that the detective is on the level.

 

10. The same remark applies to the story where the first-person narrator is the criminal. I should personally have to qualify this by saying that for me the first-person narration can always be accused of subtle dishonesty because of its appearance of candor and its ability to suppress the detective's ratiocination while giving a clear account of his words and acts. Which opens up the much larger question of what honesty really is in this context; is it not a matter of degree rather than an absolute? I think it is and always will be. Regardless of the candor of the first-person narrative there comes a time when the detective has made up his mind and yet does not communicate this to the reader. He holds some of his thinking out for the denouement or explanation. He tells the facts but not the reaction in his mind to those facts. Is this a permissible convention of deceit? It must be; otherwise the detective telling his own story could not have solved the problem in advance of the technical denouement. Once in a lifetime a story such as The Big Sleep holds almost nothing back; the denouement is an action which the reader meets as soon as the detective. The theorizing from that action follows immediately. There is only a momentary concealment of the fact that Marlowe loaded the gun with blanks when he gave it to Carmen down by the oil sump. But even this is tipped off to the reader when he says, "If she missed the can, which she was certain to do, she would probably hit the wheel. That would stop a small slug completely. However she wasn't going to hit even that." He doesn't say why, but the action follows so quickly that you don't feel any real concealment.

 

11. The murderer must not be a loony. The murderer is not a murderer unless he commits murder in the legal sense.

 

12. There is, as has been said, no real possibility of absolute perfection in writing a mystery story. Why? For two main reasons, one of which has been stated above in Addenda Note 1. The second is the attitude of the reader himself. Readers are of too many kinds and too many levels of culture. The puzzle addict, for instance, regards the story as a contest of wits between himself and the writer; if he guesses the solution, he has won, even though he could not document his guess or justify it by solid reasoning. There is something of this competitive spirit in all readers, but the reader in whom it predominates sees no value beyond the game of guessing the solution. There is the reader, again, whose whole interest is in sensation, sadism, cruelty, blood, and the element of death. Again there is some in all of us, but the reader in whom it predominates will care nothing for the so-called deductive story, however meticulous. A third class of reader is the worrier-about-the-characters; this reader doesn't care so much about the solution; what really gets her upset is the chance that the silly little heroine will get her neck twisted on the spiral staircase. Fourth, and most important, there is the intellectual literate reader who reads mysteries because they are almost the only kind of fiction that does not get too big for its boots. This reader savors style, characterization, plot twists, all the virtuosities of the writing much more than he bothers about the solution. You cannot satisfy all these readers completely. To do so involves contradictory elements. I, in the role of reader, almost never try to guess the solution to a mystery. I simply don't regard the contest between the writer and myself as important. To be frank I regard it as the amusement of an inferior type of mind.

 

13. As has been suggested above, all fiction depends on some form of suspense. But the study of the mechanics of that extreme type called menace reveals the curious psychological duality of the mind of a reader or audience which makes it possible on the one hand to be terrified about what is hiding behind the door and at the same time to know that the heroine or leading lady is not going to be murdered once she is established as the heroine or leading lady. If the character played by Claudette Colbert is in awful danger, we also know absolutely that Miss Colbert is not going to be hurt for the simple reason that she is Miss Colbert. How does the audience's mind get upset by menace in view of this clear knowledge? Of the many possible reasons I suggest two. The intelligence and the emotions function on different levels. The emotional reaction to visual images and sounds, or their evocation in descriptive writing, is independent of reasonableness. The primitive element of fear is never far from the surface of our thoughts; anything that calls to it can defeat reason for the time being. Hence menace makes its appeal to a very ancient and very irrational emotion. Few men are beyond its influence. The other reason I suggest is that in any intense kind of literary or other projection the part is greater than the whole. The scene before the eyes dominates the thought of the audience; the normal individual makes no attempt to reconcile it with the pattern of the story. He is swayed by what is in the actual scene. When you have finished the book, it may, not necessarily will, fall into focus as a whole and be remembered by its merit so considered; but for the time of reading, the chapter is the dominating factor. The vision of the emotional imagination is very short but also very intense.

 

Raymond Chandler.

 

Also:

 

Dear Frederic Dannay:

No, I would not care to nominate the ten best living detective-story writers. I don't mind sticking my neck out, but the point is, one has to agree on a few fundamentals before one starts picking lists of ten bests. For instance, does the category include writers of suspense stories in which there is little mystery, or none at all? If it does not, you eliminate some of the best performers, such as Elisabeth Sanxay Holding, certainly one of my favorites. And if it does, why call them detective stories? Charlotte Armstrong's Mischief contains no puzzle element whatever. On the other hand, some puzzle merchants, the people who have timetables and ground plans and pay the most meticulous attention to details, can't write a lick. There is a saying that a good plot will make a good detective story, but I personally question whether you can have a good plot if you can't create any believable characters or situations. My list, if I made it, would probably leave out some of those names which will inevitably appear on your ten best list. I just don't think they're any good, because by my standards they can't write. And it may also happen that a single book, such as The 31st of February by Julian Symons, or Walk the Dark Streets by William Krasner, or the aforesaid Mischief, or Mr. Bowling Buys a Newspaper by Donald Henderson, will immediately put the writer above and beyond a whole host of writers who have written twenty or thirty books and are extremely well known and successful, and from a literary point of view entirely negligible.

 

I don't particularly care for the hard-boiled babies, because most of them are traveling on borrowed gas, and I don't think you have any right to do that unless you can travel a little farther than the man from whom you borrowed the gas. I don't care about the had-I-but-known girls, because I don't care whether dear little Lucille gets her neck stretched or not. But that's not quite honest. If I had a choice, I'd prefer that she did get it stretched. I don't care for the week-end chichi either here or in England. I don't seem to care who conked Sir Mortimer with the poker, nor why, nor who set the grandfather's clock twenty minutes slow. How did Frank Fustian come to be eating fly agaric in the locked room? I couldn't care less.

 

It isn't that I don't like puzzles, because I like, for example, Austin Freeman. I like him very much. There is probably not one of his books that I haven't read at least twice, yet he bores a lot of people stiff. I even like his Victorian love scenes. And I have liked some very pedestrian stories, because they were unpretentious and because their mysteries were rooted in hard facts and not in false motivations cooked up for the purpose of mystifying a reader. I suppose the attraction of the pedestrian books is their documentary quality and this, if at all authentic, is pretty rare, and any attempt to dish it up with chichi and glamour turns my stomach immediately.

 

I think you are up against a difficult problem, because I think we may take it as granted that a mystery fan would rather read a bad mystery than none at all. You are bound to give some weight to volume of production, and strictly speaking volume of production means absolutely nothing. A writer discloses himself on a single page, sometimes in a single paragraph. An un-writer may fill a whole shelf, he may achieve fame of a sort and fortune of a sort, he may occasionally concoct a plot which will make him seem to be a little better than he really is, but in the end he fades away and is nothing. All good writers have a touch of magic. And unless we are to agree with Edmund Wilson that detective fiction is on the sub-literary level, and I personally do not agree with this, we demand that touch of magic; at least I do, although I am well aware that the public does not.

 

Yours very truly,

 

Raymond Chandler.

 

Comments (0)

You don't have permission to comment on this page.