March 28, 2009

Whodunnits and howdunnits

I’m a sucker for detective fiction, even bad detective fiction. When I start a crime novel, I’ll always persist with it, heroically, until the bitter end, suffering abominable writing and gappy plotting, wading through the longueurs and plumbing the depths, taking on the nose insults to my intelligence and assaults on my taste. I wouldn’t bother to do this with any other type of novel, or any other art form come to that. Life’s too short. But with crime novels, giving up feels too much like admitting defeat. I know the rules of the game; the author knows them too; and if either one of us cheats, we make a mockery of the whole thing.

I do occasionally read thrillers, but it’s the classic ‘golden-age’ detective story that really gets me, because the conventions are better defined: the rules are clearer, and I stand a chance of winning.

Apparently it’s generally thought that European detective fiction is enjoying something of a golden age of its own at the moment. It seems otherwise to me, based on my reading of (among others) Henning Mankell’s The Fifth Woman, which was a tedious slog, and even if the leaden style can charitably be blamed on the translator, the author must take all the blame for the clunkiness of the deductions. For instance, in order to bring the detective up to speed with what reader and author already know – that the killer is a woman – Mankell contrives to strike him with the insight that the contents of a suitcase are disposed in a certain way, such that only a woman could have packed it. This makes several errors, not the least of which is that the reader has been allowed to know more than the detective, and so is in a position to see his deduction for the absurd contrivance it is. Second, it places undue demands on the reader, because it does not specify exactly how a woman’s packing technique might universally differ from a man’s (in a way that takes account of the possibility of a particularly fastidious man or a particularly slovenly woman). The best detective fiction must specify every plot point. And thirdly, it plays on gender markers which in classic detective fiction are usually only there to set up a bluff or a double-bluff: if poison is used, the murderer is probably (gender stereotype principle) a woman, and therefore (principle of misdirection), probably a man, and so therefore (double-bluff principle), probably a woman… and so on. Gilbert Adair’s very enjoyable Agatha Christie pastiches get quite a bit of play out of this game of cache-cache between author and reader.

I’ve also recently read a couple of Fred Vargas novels (one in French, one in English translation), which, though they are more lively that the Mankell, are equally irritating in the fudging of plot points. Now, I’m told that the USP of this detective is that he proceeds not by rational deduction but by intuition and flashes of psychological insight; but it is not admissible to have the detective arrive at these pseudo-Freudian insights by the same process as the author presumably did in the first place– that is, by pulling them out of thin air. This is the problem also with the ‘psychological profiling’ sections of serial killer movies: that the process of deduction moves the wrong way, making of a contingent cause a necessary one, so that the procedure resembles less the reconstruction of the motivations of a real person than the supplying of motivations to a fictional one.

Agatha Christie’s Poirot himself liked to tell us that his insights rely not only on the cold rationality of a Sherlock Holmes but on a profound understanding of human psychology; and her plots are satisfying because they do supply plausible motives that always end up having the semblance of necessary and sufficient causes. This is the difference between a psychology that is ‘scientific’, in that it assigns motives to its actors that are ultimately explicable and comprehensible, precisely where they seemed inexplicable and incomprehensible – and a post-Freudian psychology that assigns nebulous, contingent ones. Freudian psychology, whose deductions appear largely subjective and whose conclusions are unfalsifiable, should have no role in detective fiction, whose solutions can only satisfy if the clues can ultimately be shown to point in one direction only.

But my favourite kind of detective fiction in the classic mould is one that generally eschews the problem of motive. It is the ‘howdunnit’ style, best exemplified in the short fiction of G. K. Chesterton, and taken further (some would say exhausted) in the novels of John Dickson Carr. The purest form of this subgenre is the ‘locked room’ mystery. No need to go to Todorov for a typology of this: the best place to look is Chapter 17 of The Hollow Man, where Dickson’s Chestertonian detective Dr Gideon Fell takes a metafictional leap and analyses the conventions of the genre.

I remember reading somewhere the opinion that Chesterton’s stories impatiently dispose of the opening and middle-game phases of the usual type of detective fiction, and present us only with the endgame. (Actually, I don’t think the chess metaphor was used, but I reckon it figures it quite well.) That’s a bit unfair, I think: it’s true that the Father Brown stories don’t really bother much with characterization, or the pacing out of clues, or any of that stuff; but that’s because the form does not require them. Chesterton’s stories don’t even pretend to try to give the reader a chance to catch up with the detective: they simply present an impossible situation, and then a few pages later explain how it happened. It’s clear from his other work (The Man Who Knew Too Much, Manalive) that for Chesterton the appearance of the impossible, the miraculous and the paradoxical had an ethical and religious significance: it is meant to jolt us out of the complacency of reason.

But of course, Chesterton’s solutions are always completely rational, in the how, if not in the who. We don’t expect to beat the detective to the unmasking of the killer, because the motive is rarely a puzzle – it is mere sinfulness, human evil; but we may, if we have read enough of the stories, beat the detective to the solution of how the murder was done (or rather, how the crime scene came to present the appearance it did).

The Chesterton stories, written at the height of modernism (an artistic movement to which Chesterton emphatically did not belong), present some similarities with the games of a Joyce or an Eliot. They are zero-sum games between author and reader. There is little on offer to compensate the inattentive or lazy reader, who wants to be presented with some easy truth about the human condition, and does not delight in ingenious puzzles. But I like puzzles, and paradoxes; and as Schlegel said, ‘all great truths are basically trivial, and so we have to find new ways, preferably paradoxical ways, of expressing them, in order to keep them from falling into oblivion.’


Conrad H. Roth said...

R, what do you make of the various postmodern responses to detective fiction, e.g., Auster's "NY Trilogy", or Eco's pastiche of the closed room problem in "Baudolino"?

As for myself, I can't say I've read much detective fiction. I did enjoy Perec's unfinished "53 Days", but only because it was unfinished: there was something delightfully perverse about ratcheting up the plot twists as much as possible and then stopping before any of them could be unravelled.

Raminagrobis said...

Good question ... the subject of my next post I should think.