In this series, I examine the Mystery Decalogue of Fr. Ronald Knox.

The fifth commandment of mystery fiction is:

No Chinaman must figure in the story.

In his 1939 commentary on his decalogue, Fr. Knox said:

Why this should be so I do not know, unless we can find a reason for it in our western habit of assuming that the Celestial is over-equipped in the matter of brains, and under-equipped in the matter of morals. I only offer it as a fact of observation that, if you are turning over the pages of a book and come across some mention of ‘the slit-like eyes of Chin Loo’, you had best put it down at once; it is bad. The only exception which occurs to my mind—there are probably others—is Lord Ernest Hamilton’s Four Tragedies of Memworth.

This is a rule that, in its specifics, is really only about a time and place in which most of us do not live (England in the 1920s and 1930s). But I think we can both generalize it and answer the question of why it was so at the same time.

Let’s start by looking at the phrase which Fr. Knox suggests is a sufficient warning-sign of a bad book: “the slit-like eyes of Chin Loo”.

The first thing to notice about this phrase is that it’s simply wrong. The epicanthic fold typical of the Chinese (and others) looks different than the eyes of those who don’t have them, but it really does not make the eye look like a slit. Eyes look like slits when the eyelid is mostly closed, and this is true of all human beings. Saying that the Chinese were “slit-eyed” was a sort of cant, not an actual description.

One of the curious things about literary traditions (whether in the printed word or in movies and television) is how much it is possible for storytelling to reference other stories, rather than real life. It can be a valuable sort of short-hand, but it can also perpetuate entirely fake atmospheres and backstories.

And the slit-eyed Chin Loo is, I suspect, exactly this sort of reference to an evocative but bad story. I have no idea where the unrealistic Chinaman first appeared in English fiction, but I strongly suspect that it was in a very vividly told story, all the more vivid for being new. This will have impressed people, who borrow elements of the story for themselves, and among those whose experience of the world—or at least of the Chinese—comes primarily from books rather than from people, this becomes its own sort of reality.

So we have the first element of why a book which references “the slit-like eyes of Chin Loo” is a bad book—it is a book which is fictionalizing, not real life, but another book. This is fine for satires, of course, since that’s what a satire is—but a satire is all about the distance between another book and reality. As we have a copy of a copy of a copy, the distance to reality will get further and further without the author realizing it.

(There is, also, the simple correlation that lazy authors rarely write good books, but I pass that observation as uninteresting.)

The other thing we can see in the phrase “the slit-like eyes of Chin Loo” is a cheap form of exoticism which, in detective stories, is often a means of obscurantism. By obscurantism, I mean making the story appear mysterious, not by creating a tangle, but simply by referring to knowledge not commonly held, but if known, makes the entire thing clear from the beginning. For example, if you knew at the outset of a story that the Mexican mocking tarantula leaves a bite that looks exactly like the byte of the (east Asian) king cobra and moreover that it is driven into a fury whenever a red-headed woman sings “hush little baby” at night, the death of a red-headed woman whose window was open and who bears what looks like the bite of a king cobra, and who moreover was in the habit of singing “hush little baby” each night for some reason, would not be a mystery at all. No more than a person who died of a dog bite where there is a vicious dog nearby would be.

A mystery requires an apparent contradiction, at least of the evidence to a probably innocent person, but better yet an apparent contradiction to telling us who the murderer is. In extreme cases it can be an apparent contradiction to there being a murder at all. It should not be, simply, the ignorance of the reader to the obvious solution.

The worst form of obscurantism, it should be noted, is the obscurantism of purely fictitious knowledge. It is bad enough when a man looks to have died of natural causes but was actually poised by a rare but real poison that, had the reader known of it, would have been obvious from the start. It’s simply intolerable when the poison doesn’t really exist. (But about poisons specifically, that’s rule #4.) And this goes equally well for motives. The specifics of the motive will, necessarily, be fictitious. This is owing to the fact that all the people in the story are fictitious. But the type of motive must be real. If the victim transgressed a point of honor which is not a point of honor among any real people, this is simply cheating.

And here we come to Mr. Loo, who, in being so exotic, can be made to be anything the author needs in the moment. He can kill because he’s a member of a sect whose dark god demands it, or because of some arcane business deal back in China, or because of some strange rule in the triad in which Mr. Loo is an agent. He has easy access to poisons we don’t know about, ways of killing use hair-thin needles or even just properly applied pressure, and all sorts of other means of killing about which the reader can know nothing because they are made up, but which the author does not necessarily feel is cheating because they were first made up in other books. He may interfere with evidence for arcane reasons of Chinese superstition. In short, being completely fake, he may be anything and the reader has no way of guessing what.

So I think that the other generalization of this rule is also a generalization of rule #4—the people must be fake but the motivations, tools, reactions, etc. of those fake people must all be real.