The Borges Test

Note: To celebrate the World Science Fiction Convention this week in San Jose, I’m republishing a few of my favorite pieces on various aspects of the genre. This post originally appeared, in a slightly different form, on June 21, 2017.

In his introduction to The Garden of Forking Paths, Jorge Luis Borges, who was arguably the greatest writer of speculative fiction of the twentieth century, offers a useful piece of advice:

It is a laborious madness and an impoverishing one, the madness of composing vast books—setting out in five hundred pages an idea that can be perfectly related orally in five minutes. The better way to go about it is to pretend that these books already exist, and offer a summary, a commentary on them. That was Carlyle’s procedure in Sartor Resartus, Butler’s in The Fair Haven—though those works suffer under the imperfection that they themselves are books, and not a whit less tautological than the others. A more reasonable, more inept, and more lazy man, I have chosen to write notes on imaginary books. Those notes are “Tlön, Uqbar, Orbis Tertius” and “A Survey of the Works of Herbert Quain.”

His later stories in the same vein include “Three Versions of Judas” and “The Theme of the Traitor and the Hero,” one of my favorites, in which Borges writes: “In my spare evenings I have conceived this plot—which I will perhaps commit to paper but which already somehow justifies me.” It’s a considerate way of saving time for both the author and the reader—and it’s unfortunate that it’s become so associated with Borges that it’s hard for other writers to utilize it. And it only works for stories in which an idea, rather than characterization or style, constitutes the primary attraction.

It’s also no accident that Borges arrived at this method after years as a great reader of mystery fiction and, to a lesser extent, of science fiction and fantasy, which are the genres most vulnerable to the charge that they have nothing to offer but an idea. The most damning case against the hard science fiction epitomized by John W. Campbell’s Astounding is that many of these stories could be reduced to a paragraph of plot summary with minimal loss. Most fans, I think, can relate to the experience of being halfway through a story and impatiently skipping to the end, since the writing and characters don’t provide nearly enough incidental pleasure to justify wading through the rest. At its worst, you get the kind of scientific problem story published by Analog at its least inviting, with the reader forced to stare at names on the page and incomprehensible jargon for twenty minutes, only to be rewarded with the narrative equivalent of a word problem in a physics textbook. And this doesn’t extend to bad stories alone, but to some of the important works ever published in the genre. I don’t think I’m wrong in saying that almost all of Asimov’s robot stories could be condensed to a few sentences that lay out the situation and the solution without losing much of the experience. (A trickier example is Philip K. Dick’s The Man in the High Castle, which I suspect would work better as a five-page Borges story. The idea of an alternate World War II novel in which the characters are reading an alternate World War II novel about our own world, filled with plausible inaccuracies, is one that Borges would have loved. Ursula K. LeGuin famously referred to Dick as “our own homegrown Borges,” and it’s noteworthy that Dick, as an American novelist, just went ahead and wrote the whole book.)

You could say much the same of detective fiction of the locked-room variety, which exists entirely to deliver the twist, and which might work better as one of the one-minute mysteries that children consume in grade school. (“What made Encyclopedia Brown so sure? Turn to page 61 for the solution to ‘The Case of the Giant Mousetrap.’”) This frequent inability of the mystery to rise above its origins as a puzzle is part of the reason that they irritated the critic Edmund Wilson, who wrote in his famous essay “Why Do People Read Detective Stories?”:

I finally got to feel that I had to unpack large crates by swallowing the excelsior in order to find at the bottom a few bent and rusty nails…It is not difficult to create suspense by making people await a revelation, but it does demand a certain talent to come through with a criminal device which is ingenious or picturesque or amusing enough to make the reader feel that the waiting has been worth while…You cannot read such a book, you run through it to see the problem worked out.

Under such circumstances, it can be a courtesy for one reader to summarize the contents of such a story for another. Many of Borges’s best essays consist of little more than a condensed version of another book, from William Beckford’s Vathek to Farid ud-Din Attar’s The Conference of the Birds, as filtered through his unique sensibilities. And you see a similar impulse, at much lower level, when we go online to read the spoilers for a bad movie that we have no intention of ever seeing.

But when you’re a writer, particularly of mystery or science fiction, you need to constantly ask yourself why your story is better than its own summary. (If anything, this is especially true of science fiction mysteries, which is the category in which I tend to write.) One obvious answer is to make it as short as possible. There’s a grand tradition of short science fiction—one of the first anthologies I ever owned was One Hundred Great Science Fiction Short Short Stories, which I still love—and the platonic ideal is a story that takes the same amount of time to read as it would for someone to tell us the premise. The other approach is to emphasize qualities that can’t be summarized, like character, style, atmosphere, and suspense. In science fiction, my favorite example is A.E. van Vogt, whose plots defy summarization, and who justifies his existence only by making readers feel as if they’ve lived through an experience that they can’t explain. On the mystery side, Wilson hints at this when he describes the Sherlock Holmes stories as “fairy tales,” and in his consideration of Raymond Chandler, he also gets at one of the risks:

It is not simply a question here of a puzzle which has been put together but of a malaise conveyed to the reader, the horror of a hidden conspiracy that is continually turning up in the most varied and unlikely forms…It was only when I got to the end that I felt my old crime-story depression descending upon me again—because here again, as is so often the case, the explanation of the mysteries, when it comes, is neither interesting nor plausible enough.

If you can’t do either of the above, then the idea probably isn’t ready yet. That’s the Borges test. And if you decide that it would work better as a short story by Borges, you can console yourself with the fact that it’s far from alone.

The Borges Test

In his introduction to The Garden of Forking Paths, Jorge Luis Borges, who was arguably the greatest writer of speculative fiction of the twentieth century, offers a useful piece of advice:

It is a laborious madness and an impoverishing one, the madness of composing vast books—setting out in five hundred pages an idea that can be perfectly related orally in five minutes. The better way to go about it is to pretend that these books already exist, and offer a summary, a commentary on them. That was Carlyle’s procedure in Sartor Resartus, Butler’s in The Fair Haven—though those works suffer under the imperfection that they themselves are books, and not a whit less tautological than the others. A more reasonable, more inept, and more lazy man, I have chosen to write notes on imaginary books. Those notes are “Tlön, Uqbar, Orbis Tertius” and “A Survey of the Works of Herbert Quain.”

Later stories in the same vein include “Three Versions of Judas” and “The Theme of the Traitor and the Hero,” one of my favorites, in which Borges writes: “In my spare evenings I have conceived this plot—which I will perhaps commit to paper but which already somehow justifies me.” It’s a considerate way of saving time for both the author and the reader, and it’s unfortunate that it’s become so associated with Borges that it’s hard for other writers to utilize it without turning it into an homage. And it only works for stories in which an idea, rather than characterization or style, constitutes the primary attraction.

It’s also no accident that Borges arrived at this method after years as a great reader of mystery fiction and, to a lesser extent, of science fiction and fantasy, which are the genres most vulnerable to the charge that they have nothing to offer but an idea. The most damning case against the hard science fiction epitomized by John W. Campbell’s Astounding is that many of these stories could be reduced to a paragraph of plot summary with minimal loss. Most fans, I think, can relate to the experience of being halfway through a story and impatiently skipping to the end, since the writing and characters don’t provide nearly enough incidental pleasure to justify wading through the rest. At its worst, you get the kind of scientific problem story published by Analog at its least inviting, with the reader forced to stare at names on the page and incomprehensible jargon for twenty minutes, only to be rewarded with the narrative equivalent of a word problem in a physics textbook. And this doesn’t extend to bad stories alone, but to some of the important works ever published in the genre. I don’t think I’m wrong in saying that almost all of Asimov’s robot stories could be condensed to a few sentences that lay out the situation and the solution without losing much of the experience. (A trickier example is Philip K. Dick’s The Man in the High Castle, which I suspect would work better as a five-page Borges story. The idea of an alternate World War II novel in which the characters are reading an alternate World War II novel about our own world, filled with plausible inaccuracies, is one that Borges would have loved. Ursula K. LeGuin famously referred to Dick as “our own homegrown Borges,” and it’s noteworthy that Dick, as an American novelist, just went ahead and wrote the whole book.)

You could say much the same of detective fiction of the locked-room variety, which exists entirely to deliver the twist, and which might work better as one of the one-minute mysteries that children consume in grade school. (“What made Encyclopedia Brown so sure? Turn to page 61 for the solution to ‘The Case of the Giant Mousetrap.’”) This frequent inability of the mystery to rise above its origins as a puzzle is part of the reason that they irritated the critic Edmund Wilson, who wrote in his famous essay “Why Do People Read Detective Stories?”:

I finally got to feel that I had to unpack large crates by swallowing the excelsior in order to find at the bottom a few bent and rusty nails…It is not difficult to create suspense by making people await a revelation, but it does demand a certain talent to come through with a criminal device which is ingenious or picturesque or amusing enough to make the reader feel that the waiting has been worth while…You cannot read such a book, you run through it to see the problem worked out.

Under such circumstances, it can be a courtesy for one reader to summarize the contents of such a story for another. Many of Borges’s best essays consist of little more than a condensed version of another book, from William Beckford’s Vathek to Farid ud-Din Attar’s The Conference of the Birds, as filtered through his unique sensibilities. And you see a similar impulse, at much lower level, when we go online to read the spoilers for a bad movie that we have no intention of ever seeing.

But when you’re a writer, particularly of mystery or science fiction, you need to constantly ask yourself why your story is better than its own summary. (If anything, this is especially true of science fiction mysteries, which is the category in which I tend to write.) One obvious answer is to make it as short as possible. There’s a grand tradition of short science fiction—one of the first anthologies I ever owned was One Hundred Great Science Fiction Short Short Stories, which I still love—and the platonic ideal is a story that takes no longer to read than it would to be orally told the premise. The other approach is to emphasize qualities that can’t be summarized, like character, style, atmosphere, and suspense. In science fiction, my favorite example is A.E. van Vogt, whose plots defy summarization, and who justifies his existence only by making readers feel as if they’ve lived through an experience that they can’t explain. On the mystery side, Edmund Wilson hints at this when he describes the Sherlock Holmes stories as “fairy tales,” and in his consideration of Raymond Chandler, he also gets at one of the risks:

It is not simply a question here of a puzzle which has been put together but of a malaise conveyed to the reader, the horror of a hidden conspiracy that is continually turning up in the most varied and unlikely forms…It was only when I got to the end that I felt my old crime-story depression descending upon me again—because here again, as is so often the case, the explanation of the mysteries, when it comes, is neither interesting nor plausible enough.

If you can’t do either of the above, then the idea probably isn’t ready yet. That’s the Borges test. And if you decide that it would work better as a short story by Borges, you can console yourself with the fact that it’s far from alone.

Raymond Chandler on screenwriting

Most of the boys and girls who write for the screen…devote their sparkling lines and their structural finesse to horse operas, cheap gun-in-the-kidney melodramas, horror items about mad scientists and cliffhangers concerned with screaming blondes and circular saws. The writers of this tripe are licked before they start. Even in a purely technical sense their work is doomed for lack of the time to do it properly. The challenge of screenwriting is to say much in little and then take half of that little out and still preserve an effect of leisure and natural movement. Such a technique requires experiment and elimination. The cheap pictures simply cannot afford it…

There is no correlation of crafts within the studio itself; the average—and far better than average—screenwriter knows hardly anything of the technical problems of the director, and nothing at all of the superlative skill of the trained cutter. He spends his effort in writing shots that cannot be made, or which if made would be thrown away; in writing dialogue that cannot be spoken, sound effects that cannot be heard, and nuances of mood and emotion which the camera cannot reproduce. His idea of an effective scene is something that has to be shot down a stairwell or out of a gopher hole; or a conversation so static that the director, in order to impart a sense of motion to it, is compelled to photograph it from nine different angles.

In fact, no part of the vast body of technical knowledge which Hollywood contains is systematically and as a matter of course made available to the new writer in a studio. They tell him to look at pictures—which is to learn architecture by staring at a house. And then they send him back to his rabbit hutch to write little scenes which his producer, in between telephone calls to his blondes and his booze-companions, will tell him ought to have been written quite differently. The producer is probably correct; the scene ought to have been written differently. It ought to have been written right. But first it had to be written. The producer didn’t do that. He wouldn’t know how.

—Raymond Chandler, in The Atlantic Monthly

“Ilya did not turn away from the window…”

Note: This post is the thirty-third installment in my author’s commentary for Eternal Empire, covering Chapter 32. You can read the previous installments here.

Last year, in a revealing profile of Al Pacino in The New Yorker, the actor’s agent said: “In his halcyon days he made around fourteen million a picture, but the industry’s changed. Nowadays, he gets five million. With a gun—seven million.” I love that extra two million dollars, which tells us so much about how the entertainment industry works, both for better and for worse. It’s safe to say that if the handgun didn’t exist, Hollywood would have been obliged to invent it: just as the cigarette is the ultimate prop for actors, the revolver is the perfect narrative device, and in both cases, unpleasant collateral consequences have arisen in the real world from what originated as a convenient storytelling tool. In my recent post about Friends and the logistics of set design, I pointed out that the homes we see on television shows are implausible large for a practical reason—they’re built to accommodate the blocking of a three-camera sitcom—but they can also affect the inner lives of viewers who come away with unrealistic expectations about how much house they can afford. Cigarettes and guns work in much the same way. They’re so useful to actors and writers that I don’t expect that they’ll ever disappear from our movies or novels, but there’s a troubling sense in which real people continue to make bad decisions because of the image these tools accidentally create.

I don’t mean to reduce the problems of smoking and gun ownership solely to the influence of the media: there are countless other factors at play here, and both issues are more complicated than either side likes to admit. But it’s safe to say that we think about cigarettes and guns more often than we otherwise would because actors and writers like to use them for purely pragmatic ends. For an actor nervous about what to do with his hands, a cigarette is inexhaustibly useful; for writers trying to figure out a story or a scene, a gun is equally valuable. Raymond Chandler’s famous advice—”When in doubt, have two guys come through the door with guns”—may be apocryphal, but it’s amply borne out in practice. Inserting a gun into a scene raises the tension level automatically: an otherwise routine conversation gains a modicum of interest if one character happens to be holding a revolver, even if there’s no explicit threat involved. A handgun can be used to solve narrative problems, and, more helpfully, to create them, as Chekhov knew. It’s violence or action in potential form, a symbol so potent that consumers can learn a lot from book cover or movie poster that displays a girl with a gun, sometimes clumsily photoshopped into place. And a big part of the reason we see so many stories about cops is that they’re among the few members of society who can go about their business with a pistol always at the ready.

And this troubles me, despite or because of the fact that gun violence—or just the use of guns as props—occurs so frequently in my own novels. From the very beginning of The Icon Thief, we’re treated to scenes in which a gun is present simply to send a message from one character to another, and, more subtly, from the author to the reader. And I do it because it works. As no less than Tennessee Williams once said to The Paris Review:

What shouldn’t you do if you’re a young playwright? Don’t bore the audience! I mean, even if you have to resort to totally arbitrary killing on stage, or pointless gunfire, at least it’ll catch their attention and keep them awake. Just keep the thing going any way you can.

Whether or not most suspense novelists have heard this particular admonition, there’s no question that they’ve taken it to heart. I don’t really blame them, any more than I blame myself: the number of useful tricks writers have is so limited that we’re often obliged to take the easy way out, if only to keep the story going in more important ways. But if we used fewer guns, we’d have to think harder. And occasionally, when I look back at my own work, I’m all too aware of the places where I’ve used violence or its threat as an impersonal device, just because my attention was elsewhere at the time.

Take, for instance, Chapter 31 of Eternal Empire. It’s a scene that consists almost entirely of the kind of talky exposition that I find hard to avoid, if only because it’s so central to the conspiracy genre. Maddy is on the phone with Powell and Adam, seated in the garden at Peles Castle, and after updating them on her progress so far, she receives an infodump about Gleb Boky, Alexander Barchenko, and the historical obsession of the Russian secret services with the myth of Shambhala. This is interesting stuff in itself, and it’s going to pay off later on. Still, it’s really just four pages of dialogue, so I heighten the tension in the most straightforward way I can: I establish in the previous chapter that Maddy is being watched through the crosshairs of a sniper’s rifle. It doesn’t require a single line of the scene itself to be changed—the gun, and the threat it presents, is kept completely offstage. Reading it over again now, I can’t quite decide if this represents an admirable act of narrative economy, a huge cheat, or both. (It’s a device that you see frequently in movies, like the Bourne films, since it allows a threat to be introduced from outside without the director having to worry about any particular problems of staging.) Obviously, Maddy survives the scene unscathed, and I don’t think there’s any real fear in the reader’s mind that she’s going to be dispatched here. But the rifle serves its purpose, even if I’d be happier if it didn’t have to be there at all…

The good idea trap

Ideas are poison. The more you reason, the less you create.

—Raymond Chandler

As I’ve noted on this blog many times before, good ideas are cheap. Today, I’d like to make the case that they’re also dangerous, at least when it comes to bringing a story to its full realization. And I say this as someone who has a lot of good ideas. Nearly every novel or short story I’ve written hinges on a clever twist, some of them better than others. (I’m still pleased by the twist in “Kawataro,” and wish I’d done a slightly better job with the one in “The Voices.”) It’s partly for this reason that I tend to focus on suspense and science fiction, which are genres in which conceptual ingenuity is disproportionately rewarded. In some cases, as in many locked-room mysteries and the kind of hard science fiction we find in my beloved Analog, the idea or twist is all there is, and I’m probably not alone in occasionally saving time by skipping ahead to the surprise at once, without having to sit through all the bother of plot or characterization.

Which isn’t to say that a dynamite idea is always a bad thing. A story like Arthur C. Clarke’s “The Star,” for instance, turns almost entirely on the revelation in its final sentence, but that doesn’t make the rest of it any less satisfying—although it also doesn’t hurt that the story itself is relatively short. The real mistake is to assume that the creative process hinges on the idea. As I mentioned in my recent post on Shakespeare, a story’s premise is often the least interesting thing about it: nearly every idea has been done before, and the more it lends itself to being expressed in a single knockout sentence, the more likely someone else has written it already. As a result, an artist who commits himself to the cult of the idea, rather than its execution and elaboration, will eventually start to seem desperate, which goes a long way toward explaining the curious downward arc of a man like M. Night Shyamalan, a director with a sensational eye and considerable talent long since betrayed by his ideas.

It should come as no surprise, then, that good ideas can be the most dangerous, since they’re inherently seductive. A writer with a great original idea is more likely to overlook problems of plot, structure, or language, when a merely decent idea that demands flawless execution may ultimately result in a more satisfying story. I’ve said before that a writer is best advised to start out from a position of neutrality toward his own material, and to allow his passion to flow from the process, and I still think that’s good advice. I’ve learned to be very suspicious of ideas that grab me at once, knowing that it’s going to be hard for me to remain objective. And I’ve found that sustained detachment, which allows me to evaluate each link of the chain on its own merits, is much more valuable than an early rush of excitement. Otherwise, I run the risk of turning into the producer described by David Mamet in On Directing Film, who “sees all ideas as equal and his own as first among them, for no reason other than he has thought of it.”

And the more talented the writer, the greater the risk. All writers have their moments of cleverness and ingenuity; the labor of turning a bad sentence into a good one is the sort of work that encourages the development of all kinds of tricks, and a writer who knows how to get published consistently can only get there with a lot of shrewdness. It’s worth remembering, then, that there are two sides to craft. The word evokes a set of proven tools, but it also carries a negative connotation: when we describe a person as “crafty,” that isn’t necessarily a compliment. The real point of craft is to cultivate the ability to treat all premises as fundamentally equal, and which rise or fall based only on how honestly the author follows through. It treats the best premise in the world as if it were the worst, or at least as if it required the same amount of time and effort to reach its full realization—which it does. It’s the author, not the idea, that makes the difference. And it’s frightening how often a great idea can turn a good writer into a bad one.

Raymond Chandler on exceeding the limits of formula

The emotional basis of the standard detective story was and had always been that murder will out and justice will be done. Its technical basis was the relative insignificance of everything except the final denouement. What led up to that was more or less passage work. The denouement would justify everything. The technical basis of the Black Mask type of story on the other hand was that the scene outranked the plot, in the sense that a good plot was one which made good scenes. The ideal mystery was one you would read if the end was missing. We who tried to write it had the same point of view as the film makers. When I first went to Hollywood a very intelligent producer told me that you couldn’t make a successful motion picture from a mystery story, because the whole point was a disclosure that took a few seconds of screen time while the audience was reaching for its hat. He was wrong, but only because he was thinking of the wrong kind of mystery…

As I look back on my stories it would be absurd if I did not wish they had been better. But if they had been much better they would not have been published. If the formula had been a little less rigid, more of the writing of that time might have survived. Some of us tried pretty hard to break out of the formula, but we usually got caught and sent back. To exceed the limits of a formula without destroying it is the dream of every magazine writer who is not a hopeless hack.

—Raymond Chandler

Quote of the Day

When I split an infinitive, God damn it, I split it so it will stay split.

—Raymond Chandler

Raymond Chandler on society’s constraints

No writer in any age ever got a blank check. He always had to accept some conditions imposed from without, respect certain taboos, try to please certain people. It might have been the church, or a rich patron, or a generally accepted standard of elegance, or the commercial wisdom of a publisher or an editor, or perhaps even a set of political theories. If he did not accept them, he revolted against them. In either case they conditioned his writing. No writer ever wrote exactly what he wanted to write, because there was never anything inside himself, anything purely individual that he did want to write. It’s all reaction of one sort or another.

Oh the hell with it. Ideas are poison. The more you reason, the less you create.

—Raymond Chandler, in a letter to Charles Morton of the Atlantic Monthly

Quote of the Day

When in doubt, have two guys come through the door with guns.

—Raymond Chandler