When Mary Sue Failed the Bechdel Test: In Search of A More Nuanced Discussion of Women and Gender in SFF | Tor.com

When Mary Sue Failed the Bechdel Test: In Search of A More Nuanced Discussion of Women and Gender in SFF

Thinking of writing a novel with a female protagonist? Excellent choice! To help you get started, here are just a couple of things you should bear in mind:

First, your heroine should be strong. What does that mean, exactly? Well, we have a slight preference for the action hero model, but we’re flexible. Inner strength is well and good, but should probably be complemented by something a little more badass—like, say, being a brilliant geneticist.

Be careful not to overdo it, though. She should be impressive enough to deserve her place as the main character, but not so impressive that she’s a Mary Sue. We’ll question her agency if she doesn’t solve most of the plot problems on her own—but don’t have her solve all the problems, either, because the line between Chosen One and Mary Sue is, for the female protagonist at least, pretty much invisible. She should rescue her companions from mortal peril as often as possible, but she herself should never, ever need rescuing.

Now that we’re clear about the precise formula for “strong,” let’s talk about the delicate PH balance of “female.”

What you absolutely want to avoid here is a character who’s basically a guy with a thin veneer of femininity—a so-called “man with breasts”. Our heroine should be recognizably female—but not so feminine that she’s stereotypical in some way. That means you’ll want to be careful with those emotional displays. Not too nurturing or needy, and for the love of dog, she should absolutely not demonstrate a desire for babies. We are going to grumble if she’s too pretty, or if she frets about not being pretty. Frankly, the less said about her appearance, the better.

On the other hand, if she drinks and swears and occasionally acts like an arrogant jackhole, we’re probably going to dislike her. Rule of thumb: if she could fairly be described as a “loveable rogue” or “antihero,” you might to want to rethink that. If she’s the female equivalent of a playboy, we’re going to slut shame her. In fact, it’s probably safest to avoid romantic entanglements altogether, lest you inadvertently give the impression that she needs a man. Oh, but don’t make her emotionally unavailable either, because that’s a stereotype.

In sum, when crafting a winning female protagonist, balance is key. Like, say, walking a tightrope. Over a bed of pikes. Writhing with asps.

Wait, where are you going? Not having second thoughts, are you?

If you are having second thoughts, or if you find yourself doing some serious handwringing about how to craft your heroine, you are not alone.

And it’s a damn shame.

Most of us can agree that we’d like to see better representation of women in SFF. That responsibility is typically placed on the shoulders of authors, and to be sure, they’re an important part of the equation. How we tell the story matters.

How we hear the story matters, too. What we say about it afterwards matters.

Authors have their intentions and readers have their experiences, and where those forces collide is where the story takes place. The chemical reaction between what s/he said and what we heard is the story. And a collection of stories becomes a narrative.

That second half of the storytelling experience doesn’t get nearly the scrutiny it deserves. Because as far as we’ve come in terms of demanding better representation of women in fiction, the standards many of us use to judge success or failure in that endeavour are oversimplified at best—and at worst, they’re downright harmful.

Don’t get me wrong: the fact that we’re talking about this at all is a good thing. The fact that something like the Bechdel test exists and is referred to so often is a very good thing. It means the discussion has made its way firmly into the mainstream. But it’s not always a very nuanced discussion. It makes use of some pretty blunt instruments, and it’s littered with its own perverse forms of gender bias. We can do better.

We’re ready to level up. And we need to, because we’re shooting ourselves in the foot.


Mary Sues and Men with Breasts

There are two types of hero in SF/F: the protagonist and the female protagonist.

We approach them differently. Consider, for example, the wealth of articles on how to write believable women. Kate Elliott talked about it here last March. More recently, Mark Lawrence had this to say. Skimming over the titles of those blog posts, one could be forgiven for thinking there’s a trick to it, as compared to writing believable male characters. Look more closely, though, and you’ll see that in both of these examples—and in most other credible ones I’ve come across—the message essentially boils down to this: write a believable character. That is to say, there is nothing special about the process of writing a woman. Which is not the same as writing women.

This is so important that it’s worth repeating. In italics.

Writing a woman is not the same as writing women.

Seems obvious, doesn’t it? Yet as readers, we routinely conflate the two. We draw a straight line between woman and women, and that line ends up being the thin, quivering tightrope I mentioned earlier. The moment a character ceases to be an individual and is taken as a representative of her gender, she is no longer a person but a specimen. Something to be placed under the magnifying glass, dissected and labeled.

But no—that’s not quite right, is it? We don’t approach the female protagonist with the analytical objectivity of an entomologist examining a butterfly. Instead, we scrutinize and vet her qualifications as an ambassador of her sex. We judge her.

We judge (male) protagonists too, of course, but differently. We evaluate the hero’s actions as an individual in the context of his specific circumstances. And when we find him wanting, we’re usually prepared to cut him some slack. We call him an antihero.

The heroine, though, has to be more than just a realistic character we can root for. She’s got to be a shining example of empowered womankind.

So much pressure is put on the female protagonist it’s a wonder she can shoulder it. And yet she does. There are countless examples of kickass heroines who’ve won our hearts. Few, if any, have escaped the rigorous vetting process unscathed, but they’ve survived the scrutiny and endured.

Thing is, they shouldn’t have to.

We’ve been talking about this double-standard for a while now. Liz Bourke tackled it recently in one of her most recent columns for Tor.com, making some of the same arguments I’m making here. We obviously recognize there’s a problem. And yet to my eye, it’s getting worse instead of better.


What do we look for in a hero?

It starts with a well-drawn character. Someone complex and believable, with his own motivations and experiences and flaws. But a hero—one of the story’s main characters—needs to go that extra mile. He needs to be compelling enough to carry significant chunks of the story, and he should play an instrumental role in resolving important plot problems.

So—realistic, interesting, and demonstrating agency.

A female protagonist has to tick these same boxes, but the boundaries are much more tightly drawn.

For starters, she has to be realistic not only as a human being, but as a woman—a narrower subset of humanity with specific characteristics. What exactly those “specific characteristics” look like is a source of much debate, but that doesn’t matter. The character has to resonate with women readers—while at all costs avoiding stereotypes about women. Feminine but not too feminine, even though you and I might have different ideas about femininity.

That’s a very narrow space in which to work, and it’s studded with landmines. Many readers are quick to make the leap from “Character X is indecisive” to “women in Book X are indecisive” to “Author X thinks women are indecisive.” For an author, it can start to look like certain character traits or plot lines are more trouble than they’re worth.

Think this sort of self-censoring is a myth? Think again. I’ve done it myself, to my lasting regret. And I’m not alone.

When that happens, we’ve come full circle and we’re right back to using our preconceived notions of gender to define who a character should be—and who she shouldn’t be. We might be holding up a different model of femininity than the traditionalist ideal, but it’s no more empowering. Empowerment is the freedom not to conform to anybody else’s abstract ideals of womanhood.

We have a gendered view of interesting too.

Female protagonists are generally expected to be likeable, or at least relatable. The antiheroine is a rare creature indeed. Her male counterpart is not only tolerated, he’s never been more popular. Sure, he’s flawed, but he’s compelling and gritty. He might even be a monster, but so long as we give him some plausible backstory to explain why he turned out that way, plenty of readers will happily root for him.

Female antiheroes, when we find them at all, are usually pretty tame by comparison—and they take a lot of heat. We accuse them of representing some kind of “masculine” ideal of power, as if their very existence were somehow a repudiation of femininity.

And then there’s agency.

It’s fair to say that we have a lower tolerance these days for special snowflakes in general, but the Chosen One narrative still enjoys a perfectly legitimate place in SF/F. Unless, of course, you’re a woman writing about a woman.

Enter the Mary Sue.

When it comes to the representation of strong women in SF/F, I have a hard time thinking of a concept more damaging than the Mary Sue. With two small words, we dismiss any female protagonist we deem too capable, too “special,” and make her an object of derision.

The male protagonist, of course, can be Chosen from birth (Harry Potter). He can be awesome enough to beat the machines (Neo) or even a goddess (Raistlin). He can make the sky rain fiery awesome (Pug). He can even be immaculately conceived by particles of awesome (Anakin Skywalker).

But the female protagonist? She should dial her awesome back to a reasonable level if she wants to avoid being labeled a Mary Sue. Speak softly, darling, and take care not to draw too much attention to yourself. It’s not ladylike.

Perhaps I’m being a bit unfair. Accusations of Marty/Gary Stu do exist—but they’re relatively rare. Not so for the women. Show me a popular female protagonist and I will show you a heap of one-star reviews specifically citing her supposed Mary Sue-ness, even if she does nothing more impressive than shapeshift now and again. We sling the term around so indiscriminately that it scarcely has meaning anymore. All that’s required is a talented woman written by a woman.

Every time we do this, each and every time, we send a message. We contribute one more drop to the ocean of toxic groupthink telling us that a female hero has to be a certain way.

Mary Sue. Man with breasts. Damsel in distress. By applying these labels, we’re discouraging diverse representations of women in fiction. Because when we take certain characteristics off the table, what’s left behind is just that much more generic. By saying “no” to this feature or that, we’re steadily whittling away at a character until she’s just another faceless wooden doll. In our desire to avoid certain kinds of stereotypes, we’re creating whole new ones.

Nothing should be off the table. In fiction, as in life, women should appear in every permutation and combination imaginable. That necessarily includes some characters we don’t like or approve of, and even some who exhibit traits we consider to be stereotypical. We should be comfortable with that so long as it’s not a pattern among the female characters in a specific work. (The moment a pattern does emerge, we’re perfectly justified in talking about the way women are represented in that work.)

We make the leap from a woman to women so readily in part because women continue to be under-represented, and so the temptation is to make an example of each and every one. Part of the answer, then, is certainly to increase the number of important female characters. But it’s not purely a question of numbers, as the limitations of the Bechdel test make clear.


Beyond the Bechdel Test

The Bechdel test has been hugely influential in advancing the conversation about gender bias in works of fiction, taking it from more rarefied critical circles into the mainstream. A big part of this success owes to its simplicity: the test is a straightforward checklist that asks whether a work of fiction features (a) at least two important female characters who (b) talk to each other about (c) something other than a man.

But in and of itself, the Bechdel test doesn’t actually tell us a whole lot. It’s nice to have a handy scorecard, especially if it produces quantifiable data. But there’s no room on a checklist for nuance. And we need nuance. We need context.

The Bechdel test measures how many important female characters, and gives us an extremely limited insight into how they relate to each other and to important male characters. But it tells us very little, if anything, about how women in general relate to men or society as a whole. In other words, it tells us nothing about gender roles in the story, let alone gender equality or empowerment. It’s not a litmus test of feminism. (And was probably never meant to be.)

Taken on its own terms—as a quick-and-dirty way of measuring tokenistic representation of women—the Bechdel test performs admirably. The problem arises when we use it as a shortcut to assign “pass/fail” on gender. That gets in the way of a more substantive and nuanced conversation about narrative and the extent to which it challenges or upholds traditional gender roles. In other words, the use of this scorecard of tokenism can itself become tokenistic, a way to tick the box of “gender analysis” without actually asking any of the more interesting or challenging questions.

A book or film can “pass” the Bechdel test with flying colours and still send a damaging message about women and gender roles—or it can fail miserably and say something very important indeed. So why is “failing” the Bechdel test so often considered an automatic black mark, a sign that a book has failed feminism writ large? Not only is that unfair, it may discourage authors from telling a certain type of story—one we might very much need to hear. At the very least, it can create its own perverse incentives toward tokenism.

A review that boils down to “Mary Sue fails the Bechdel Test” is not a feminist critique. It’s not moving the conversation forward or even running in place. It’s a step back toward labels and generalizations. We can do better.


How we hear the story matters. What we say about it matters.

Every time we discuss a work of fiction, we contribute to a narrative about what we expect to see in the genre—what we demand. In fiction as in everything else, demand influences supply, and that can be a tremendous force for progress. Or it can inadvertently discourage diversity and stifle certain voices.

Instead of demanding the impossible from female protagonists, we should be demanding more insightful and nuanced analysis of women and gender in fiction. One that doesn’t rely on seductively simple yet ultimately counterproductive tools.

The dialogue between author and reader is a two-way street, and when it works, it’s a beautiful thing. So let’s put away the labels and the checklists and meet in the middle.

Together, we’ll tell a story.

Erin Lindsey writes fantasy and occasionally blogs about it; her latest novel, The Bloodforged, is now available from Ace. You can check out her ramblings on her website: erin-lindsey.com, or on Twitter @etettensor.


Back to the top of the page


This post is closed for comments.

Our Privacy Notice has been updated to explain how we use cookies, which you accept by continuing to use this website. To withdraw your consent, see Your Choices.