The Bechdel test is named after Alison Bechdel, who introduced it in a comic strip in 1985, but Bechdel herself attributes the idea to her friend Liz Wallace. A work of fiction passes the test if it contains two female characters who have a conversation about something other than a man. It sounds like a pretty low bar to jump over, but apparently it's alarming how few books or movies pass this minimal test. (I'm reluctant to start checking through my favorite sci-fi and fantasy books — and the majority of my favorite authors are women.) This should be surprising. It's not asking the moon in political correctness. Have a couple of female characters talk about something other than a man, even very briefly, once in three hundred pages of novel or two hours of movie. How hard can that be?

Well, it's not actually quite so minimal a test as it may at first sound. If there are equal numbers of male and female characters, and they pair off at random to talk, then only about one in four conversations should be between two females. And then a fair amount of any story's dialog is likely to be about other characters in the story, so even without any gender bias at all, there would be a a fair number of conversations between female characters that did at least mention some man. Some stories may not have so much dialog in the first place. It doesn't take an absurd amount of bias to fail the Wallace-Bechdel test. It's easy to not notice that it's being failed.

And that is really the point. Finding books and movies in which no two men ever discuss anything besides women is hard. A book that failed that test would stick out like a sore thumb, for having either very little dialog or unbelievable characters. Even if it weren't necessarily unbelievable that the men were constantly thinking about women, men only talking about women really is unbelievable. Even soldiers and convicts in prison talk about lots of other things. Men in fiction always talk about other things. A book or movie that failed the sex-reversed Wallace-Bechdel test would be ridiculous. Yet it's easy to get through a book or movie that fails the Wallace-Bechdel test, and not realize anything amiss. This is the point.

I stopped posting installments of my story here, but I have kept on writing it, and I recently finished a major section of it, comprising what should be at least a third of the total text. It's just over 70 000 words. I only considered it a first draft of that initial section, and expected to revise it a fair amount. Well, I happened to read about the Wallace-Bechdel test.

My story passed, but only just, with squinting. So I fixed it and now it passes quite fair and square. It didn't actually take much re-writing at all, but it made the story much better.

The problem was that my story had far too little dialog. It has plenty of female characters. The protagonist-narrator is female, and I have carefully balanced genders among all the other characters, including roughly balancing for prominence in the story and coolness. The break in the story I've reached now is to switch to a male narrator. I don't think this comes off as fussily PC. There's no mirror symmetry, but there are lots of rough parallels, and I don't think you can avoid that if you have even rough gender balance in a story. Anyway, I didn't have trouble with Wallace-Bechdel for lack of women.

But the story has first-person narration, and I had far too many passages in which the narrator just told the reader things. Several of these had some character saying something to the narrator, and the narrator telling the reader what she thought about it. (The squinting that made my first version pass was to call a two-sentence response to a long monologue a conversation, and to note that the response was to the short last part of the monologue, the bulk of which had been about the narrator's father. All the ingredients for a much better conversation were being wasted in asides to the reader.)

By just re-writing a few of these scenes into dialog, though, I had three quite substantial female-female conversations in just the first few chapters. One of these does mention a man, but is not mainly about him, and the others have nothing to do with any male. (Well, they do mention people in general, but this is a very gender-balanced setting — people in general clearly does not mean men in particular, either to the characters or, at this point, to the reader. To me this meets the W-B criterion. And just to be sure, there are two more passages I still have to dialogue-ify that will be about strictly inanimate subjects.)

I'm not completely off the hook for gender bias, because my first draft did pass the sex-reversed version of Wallace-Bechdel quite straightforwardly, if still only minimally. It had one brief conversation between two male characters, entirely about something other than a woman. It had two conversations between males about the female protagonist. All the remaining conversations were male-with-female, some of them with several of each talking together. With the narrator being female, though, the number of female-female conversations should have been closer to half, and there are plenty of things for these women to talk about besides men. My problem was that I just didn't turn all these opportunities into interesting dialogue. I left them as monologues of one kind or another.

Now I'm thinking that I need to add more conversations among my male characters. It really livens up the story, and makes characters other than the narrator seem a lot more real. Dialogue is also fun to write. I have a thing for stories with multiple narrators. This one has two, and most of the other stories I've started over the years have had at least two. Well, even with only one actual narrator, a dialogue is a bit like having two narrators. You get to juxtapose two points of view. It's fun to shift back and forth.

Wallace-Bechdel is a good tool for revealing gender bias, just because it is such a minimal test. And yet I think that a work that passes W-B is probably not going to do too badly for gender bias. One conversation isn't much in itself, but if you write even one conversation between two characters, other than as a way of writing about a third character, then you're likely to write more than one conversation between those two, and you might even want to see what a third would have to say to them.

There is no reason to be afraid of the Wallace-Bechdel test as a PC shibboleth. If you're willing to have even just two female characters in your story, then all it asks you to do is to give them something to talk about, and write it as dialogue. The way I'm thinking now, that's never a bad idea for any two characters in any story.

The middle section of my story, with the male narrator, will not be scrupulously gender-balanced. It's set in a pre-modern society, and its narrator is an amateur actor-playwright, on the side from other things, to whom the Wallace-Bechdel test would never occur. This section may not pass it, but we'll see.