Rules were made to be broken...

[Note: I am trying out something new today on the blog: instead of reviewing a book, I am discussing my impressions on a topic and trying to put them in a cohesive article. Feedback on this experiment is much appreciated, as I would like to include more posts like this in the future and would like to improve them.]

"This means war!"

—Daffy Duck, Who Framed Roger Rabbit

My reaction to James' book...

In her book Talking About Detective Fiction, P. D. James has a curious fixation for Ronald Knox and his “Ten Commandments” for detective fiction. This is one of the many points about the book which rather annoy me. Knox was not the only one to come up with rules, but James never even mentions this. She seems to believe that everyone took these rules as Law, and that any deviation was frowned upon with nothing but scorn. In fact, she goes on to say that “Rules and restrictions do not produce original, or good, literature, and the rules were not strictly adhered to.” This is a stunningly silly statement—if it were true, we might as well throw every book of poetry into the rubbish bin, because they all adhere to one rule: they must not be written in prose. 

If all goes well, you will hear more about James’ book and my problems with it in a few days. But for now, I’d like to examine Father Ronald Knox’s Decalogue. Just how futile is this attempt to define rules for the genre?

I. The criminal must be someone mentioned in the early part of the story, but must not be anyone whose thoughts the reader has been allowed to follow.

This is a rule that Agatha Christie famously broke in one of her novels, and some readers have never forgiven her for it… even today! I used to be a regular contributor to the Agatha Christie forum, and back then, one member always went into a furious rant over just how atrociously bad this book was. I kept pointing out just how fair it played—every clue was placed there and there were no explicit lies, just carefully sidestepping around the issue of giving information on a point or two.

I think Knox was trying to make a distinction between Christie’s level of ingenuity and that of second-rate hacks who would try clearing suspicion by having X wonder in their thoughts “Oh! This is so horrible! Who could be behind these murders???” only to have X turn out to be the killer. After Christie came up with this twist, many derivative carbon copies were written… including one by Christie herself, late in her career.

II. All supernatural or preternatural agencies are ruled out as a matter of course.

This really depends on what you are trying to write. For instance, when I examined Paul Doherty’s Ghostly Murders, I remarked that the supernatural is required for the mystery. In fact, the mystery revolves around the supernatural— why are ghosts haunting the church and what reparation can be made to get rid of them? Here, the supernatural is perfectly permissible—in fact, I’d love to read more of these hybrids between mystery and ghost story.

However, if your detective is trying to find out just how an assassin got in and out of a sealed room to behead Lord Ragamuffin and then place his head inside a chocolate cake, you can’t just blame it on the ghost of the first Lady Ragamuffin, who returned from the dead and killed her husband because of his second wife’s atrocious taste in ties. It’s “bad form”, but it also makes everything you’ve read seem futile, when such things do not happen in real life and thus, no detective could use their intelligence to get to the truth. After all, isn’t that one of the facets of detective fiction: the use of intelligence, rational thought, and ingenuity to discover the underlying truth about something? Isn’t that thrilling quest for truth one of the reasons we read the genre?

The again, as John Dickson Carr proved in one of his greatest books, you can still include the supernatural in a genuine detective story without that sense of futility. Carr basically takes this rule, smashes it into pieces on the floor, and runs over it with a steamroller before handing it back to you with a smile.

In fact, that’s why I think rules are a rather futile exercise: the best books will do precisely this with almost any rule you can come up with. The problem arises when second-rate hacks decide to write their own derivative versions, without any shred of ingenuity behind the proceedings. This is when stuff like the supernatural or the secret passage (which we’ll get to next) becomes a problem.

III. Not more than one secret room or passage is allowable.

This rule more or less makes sense, especially in terms of the locked-room mystery. How did the assassin kill Peggy Ann in her locked kitchen, with all the appliances except the stove shoved against the door and windows to prevent entry? What’s that? If you turn all the burners on the stove a certain way, you uncover a secret passage? Well… that was silly.

A story can get away with a secret passageway or a hidden room when it demonstrates true ingenuity about it: Sir Arthur Conan Doyle’s The Norwood Builder is a brilliant Sherlock Holmes story that hinges on such a thing. However, you’re able to deduce the existence of such a room and to what use it has been placed: it isn’t sprung on you as the “surprise ending” that wouldn’t have fooled a five-year-old.

What I’d really like to read one of these days is a locked-room mystery with a secret passage, where the passage is a complete blind. The police discover the passage from the get-go, but immediately see a problem—the passage has a thick coating of dust, accumulated over decades, and it’s entirely unbroken.

IV. No hitherto undiscovered poisons may be used, nor any appliance which will need a long scientific explanation at the end.

The first part of the rule makes sense: don’t blame your murder on a mysterious poison unknown to science! Oh, you can create a fictional poison—just as long as you explain how it works. There’s no problem there. That demonstrates true ingenuity. In the hands of a hack, though, “mysterious poisons unknown to science” are examples of a disappointing cop-out ending that just about nobody likes.

The second part of the rule instils more of a mixed reaction in me, though. I like things explained—John Rhode came up with a devilish way to commit a murder by having a tree branch fall on his victim in Death on Sunday. His explanation comes early on, but it could just as easily have been converted into an impossible crime story. Having a long-winded technical explanation might not be the most exciting trick in the book, but it can show ingenuity nonetheless.

V. No Chinaman must figure in the story.

One of the most controversial rules of this Decalogue, it makes sense to me. Hacks would use the “sinister Chinaman” cliché as a cop-out ending: all Chinese people were evil, they were all trying to take over the world, etc. etc. This kind of twist shows no ingenuity whatsoever, and it is very racist to boot. It’s all part of the infamous “Yellow Peril” craze, and this rule should be considered in that light for it to make sense. Otherwise, you might think Father Knox was being a bigoted racist… which in fact is a common mistake, not just about Knox, but about the entire genre!

There’s nothing wrong with having Chinese characters in mysteries, though. Charlie Chan, created by Earl Derr Biggers, was not only a likeable Chinese character, but also a Chinese detective.

VI. No accident must ever help the detective, nor must he ever have an unaccountable intuition which proves to be right.

This is debatable. Sometimes, the detective hears a scrap of conversation from two strangers and exclaims “What a fool I’ve been! Of course!!! Johnny Depp plays Captain Jack Sparrow in four movies! That’s the key to the whole affair!” I think this is a perfectly fair and ingenious way of solving the crime, especially if the detective’s explanation of why that comment helped solve the crime makes sense. You just don’t want the relation to the clues and the problem to be too tangential—that’s one of the problems about John Dickson Carr’s Panic in Box C, where an old baseball player’s name seems to be key to the problem, but in reality, the relationship is unsatisfactory.

Far less contestable are the endings where the detective goes to a party and overhears X confessing to Y that X killed Z out of love for Y. It’s one of the most unsatisfying cop-out endings out there is, and by the way, it's the way the mystery is solved in Sister Carol Anne O’Marie’s Murder at the Monks’ Table. It’s simply not a mystery in my view—it’s just a book where somebody is killed and somebody confesses, with no detection involved whatsoever.

VII. The detective must not himself commit the crime.

It’s best not to dwell on this too often, because it has been carried out before. Sometimes, these twists are brilliant—other times, they’re just derivative “shock” endings. There’s a fine line between being derivative and being ingenious, and unfortunately, ingenuity has not prevailed as often in this category…

VIII. The detective must not light on any clues which are not instantly produced for the inspection of the reader.

Well, yes and no. I want to be given all the clues, but if you’re given every clue to deduce the culprit’s identity, you probably won’t be mad if the off-stage phone call confirms the detective’s guesses that Q lived in Barcelona thirteen years ago and killed K for throwing a gum wrapper on the sidewalk. But again, if you simply hide the clues from the reader, there’s no fun in any of that. The mystery is a great big fencing match between reader and author, where the author tries to outfox the reader all while playing scrupulously fair with the rules. I don’t know about any of you, but I never play chess with someone who knocks your pieces over when you’ve gone to answer the telephone.

IX. The stupid friend of the detective, the Watson, must not conceal any thoughts which pass through his mind; his intelligence must be slightly, but very slightly, below that of the average reader.

This seems to me like more of a guideline to authors who don’t know where to start with mystery writing. The Watson is not even necessary, as has been proven time and time again. Father Brown generally works alone, but his occasional companion Flambeau is a brilliant ex-arch criminal who lacks the outlook on human nature Father Brown has gotten from his years as a priest. Hercule Poirot abandoned Captain Hastings after a while and went on memorable investigations alone, like the tour-de-force Five Little Pigs. So not only is a Watson not required, he doesn’t have to be stupid.

X. Twin brothers, and doubles generally, must not appear unless we have been duly prepared for them.

It’s a good thing Knox included that “unless we have been prepared” clause, because I would’ve whipped out Ellery Queen’s The Siamese Twin Mystery in response. Not only is this a brilliant mystery story, conjoined twins are major characters in there. Like the ban against “Chinamen”, this seems like a ban for the cop-out endings.

“Mary, why didn’t you tell me you saw me strangle Mrs. Featherblossom in the alleyway? I would’ve been able to tell you about my evil twin brother Patrick who escaped from Dartmoor a year ago and has been committing murders posing as me!”

“Oh, John! I didn’t know if I could trust you!”

“Oh, Mary!”

“Oh, John!”

Once again: it’s possible to break this rule with ingenuity, but when the hacks get their hands on this twist, it is tiresome.

In fact, this seems to be the major facet of Ronald Knox’s Decalogue. Despite the impressive “Ten Commandments” title, Golden Age authors probably didn't take them that seriously. Each rule here has been broken in the best of mysteries, which will defy any attempt to create a set of rules. Only one rule stands: fair play with the clues. And I don’t think this is such a bad thing. It’s simply something that defines the genre, just like poetry is defined by verse. If an author can’t work with such a stricture, they’re in the wrong genre and they ought to leave the mystery alone right away.

An incentive
to play fair...

In order to engage the reader’s active participation, there is an unwritten reader-author contract, where both sides agree to play fairly. This isn’t because breaking rules is “bad form to say the least”, but because by sticking to this fair-play standard we can gauge an author’s ingenuity and separate the ones worth reading from the derivative hacks it’s best to forget.

The mystery is not a mere crossword puzzle in prose, but a story about the use and misuse of intelligence and human ingenuity—the murderer will use it for his or her own gain, but the detective will use this in a quest for truth and justice. This doesn’t mean “order must be restored” and any lawbreaker must be captured. In fact, it was fairly common to have sympathetic murderers given an opportunity to escape, and not necessarily via the suicide route. In The Mad Hatter Mystery, the killer is exposed in front of a police officer, Hadley. He decides to forget all about it after hearing the full story. You see that? Social commentary, my friends!!! Social commentary!!! Someone alert the Literary Art brigade!!! What endings like this try to say is that the law is not necessarily equivalent with true justice. Corruption and bureaucracy often get in the way.

Golden Age mysteries are severely underappreciated for these facets, which are often simply ignored. P. D. James ignored the richness of the genre in her highly selective book, and makes gross generalisations of exactly this sort. Mysteries really deserve more respect. All this talk about rules “restricting the genre” is nothing but pure bunkum. For all the discussion about rules, it seems to me they all boil down to personal preferences and guidelines that differentiate between the real geniuses like John Dickson Carr and the literary hacks you will find in any genre.