Nation of Beancounters

Game theory and consent

Posted in Culture, Gender by Navin Kumar on February 17, 2014

A man asks a woman out on a date. She’s interested in a long-term relationship i.e. a boyfriend. She doesn’t want to waste her time or money on someone interested in casual sex i.e. a player . She doesn’t know what her proposer wants. Asking would be silly – a man who’s only interested in sex would lie, enjoy a romp in the hay, and not call the next day. She decides to “play hard to get” – pretend to be more reluctant to go on a date than she is. She believes that a would-be boyfriend would persist in asking her out, while a player wouldn’t.

Now let’s look at things from the man’s perspective. He doesn’t know if the woman is genuinely uninterested or merely playing hard to get. If he wants casual sex, it doesn’t matter – the cost of persisting is too high to bother. If, on the other hand, he wants a long-term relationship and thinks that there is a reasonable chance that she is “just playing”, it’s in his interest to persist. The expected gain from her eventually saying yes may (and here we assume does) exceed the cost of pursuing her. Thus, a player will back off, while a would be boyfriend will not*.

Thus, men and women both end up having their beliefs confirmed. Women who want relationships play hard to get because they know that would-be boyfriends will persist and players won’t. Thus some fraction of women who say “no” don’t mean it. Would-be boyfriends know this and persist. Players don’t persist because sex isn’t worth the trouble. Thus all the individuals in this little game are playing their best response to each others strategies. This is what game theorist call an equilibrium.

The problem is that some women are genuinely not interested in the men asking them out. For them, being repeatedly approached isn’t a positive sign; it’s harassment. It imposes serious psychological costs. It is for such women that sexual harassment laws have been introduced.

However, men’s persistence isn’t the only thing causing their plight. It’s ultimately caused by the equilibrium that all the players find themselves in. An important part of that equilibrium is the strategy other women use to filter out players from potential boyfriends. If they stopped playing hard to get, men would have no incentive to persist, and our victim would have been spared. This is not to let men off the hook. After all, if all men backed off as soon women said no, it would be impossible to tell players and non-players apart from how quickly they backed off. Thus, women would no longer have an incentive use “playing hard to get” as a filter. Both men and women are to blame for this bad equilibrium.

(This is not “blaming the victim”. The victims are the women who are uninterested and so feel harassed; the accused are the women who are interested but play hard to get.)

Yet, whenever the question of consent arises, activists demand changes in male behavior, while ignoring the female behavior that supports it. Worse, they present “No Means No” as a factual statement: all the women who are saying “No” to you are genuinely uninterested in you. This is false. One famous study examined whether women ever say no to sex when they mean yes. 39% of their sample had done so at least once, and more than two thirds had said no when they meant maybe.

Filtering boyfriends from players is just one reason for women to pretend more reluctance that they feel. I’ve given it the spotlight because it’s the easiest one to model. Other reasons include: the desire to filter confident, assertive men from “weak” men; the desire to avoid being labelled a “slut” by others; the desire to prove to oneself that one is not a “slut”; the belief that one will not be “respected” if one “gives it up too easily”; the thrill of the being chased; the desire to enact a rape fantasy; the desire to gain power in a relationship; the desire to extract more resources (attention, dinners etc) before terminating a casual relationship etc. And while I’ve presented the refusal as happening at the point of the date, this logic can apply to any stage of the relationship.

If the consent movement ignores this facet of mating behavior, it will fail. Why? Suppose that men in our model were presented with the proposition that women never play hard to get. If all men believed it and changed their behavior instantly, women would be forced to change theirs as well. The equilibrium would collapse. However, suppose instead that some men believe it and others remained skeptical. Women still have an incentive to play hard to get. By doing so, they would snag the skeptics who are interested in relationships as boyfriends, while avoiding the skeptics and believers who are interested in casual sex. A month after the news that “No Means No” came out, the believers and skeptics gather at the pub and compare notes. All the skeptics who want girlfriends now have them, while the believers don’t. The believers conclude that their new strategy is bust, and go back to persisting in the face of resistance. This is true of any movement which tries to sell “No Means No” as a fact. When skeptics do better than believers, beliefs collapse.

This is unfortunate. I want the “No Means No” movement to succeed. Heck, I want the even more ambitious “Yes Means Yes” movement – which calls for a verbal affirmation before sexual acts – to succeed as a social norm**. These movements started in the West, but I think they will eventually find their way to India. What changes do activists have to make to succeed?

Firstly, they must recognize that female behavior is also responsible for this unhappy state of affairs. They must work to raise awareness about it, and to discourage women from engaging in it. They must recognize that actions that are in the interests of individual women (filtering suitors) are not in the interests of women as a whole (avoiding harassment.)

Secondly, they must present “No Means No” as an ethical norm, rather than a factual statement. The idea should be: a moral man should treat “no” as meaning “no” even if the person is possibly faking it. A moral woman doesn’t say “no” unless she means “no”. As time goes on, and we gradually shift from one equilibrium to another, “no means no” will become a factual statement. In the interim, our best bet is impose the costs for immoral behavior (guilt, shame, censure, ostracism etc) on those who break the rules. Few men will take seriously a movement that claims that “[affirmative] Consent is Sexy” when it’s so far from what experience teaches them and women themselves say. Best stick to Consent is Right.

In closing, there are three additional problems with the current male-centered approach:

Firstly, it denies women agency, to invoke a term that is much in vogue. More correctly, it denies that women have agency. It treats them as playing an entirely passive role in the mating process. It ignores the fact that they are savvy players in a complex game. It contributes to the narrative where women are helpless victims at the mercy of men.

Secondly, it treats the mating process as though it were something simple, made complex because of something called “culture”. Somehow, everyone wants an honest, open, mutually respectful relationship, but “culture” thwarts their attempts to enter one. The reality is different. Most people want the same things from a relationship – sex, companionship, children etc. However, relationships are also the site of much conflict, over sex, money, freedom etc. He wants commitment; she wants a fling. He to splurge; she wants to put the money in a retirement account. He wants to spend the night out with the boys; she wants him home by 10. It is these inescapable conflicts of interest that complicates matters, and lead to dishonesty, not “culture”. In this case, there is a conflict of interest (woman wants commitment; man might want casual sex) that leads to dishonesty (playing hard to get). The man’s persistence isn’t the result of a “culture that devalues consent.”

Finally, it is unfair to men. It forces them to bear the entire costs of shifting from one equilibrium to another. It holds them accountable for a terrible state of affairs while letting equally culpable women off the hook.

UPDATE: To clarify: “Persisting” here is a metaphor or stand-in for all behavior that involves ignoring a woman’s stated preferences. “Playing hard to get” is a metaphor or stand-in for all behavior that involves pretending more reluctance than one feels. The story at the beginning was a metaphor for all situations in which women and men have an incentive to lie and persist respectively. All models are parables of some kind.

Disclaimer: over the course of the post, I’ve used “man” to mean the initiator of a sexual act, and “woman” to mean the non-initiator. I’ve done so for the sake of clarity, and because it it’s a reasonable assumption, given modern gender roles. Nonetheless, it doesn’t always hold: women do initiate, men do respond, these roles aren’t clearly delineated in homosexual relationships, and men too can be the victims of harassment and sexual assault, sometimes at the hands of women.


* To make this example more concrete, let’s assume the following: the cost of persisting is 45 to both men. The payoff to the man seeking sex is 80. The payoff to the man seeking a relationship is 100. 50% of all women are interested in a relationship, while 50% of women are not. The payoff to not persisting is zero.

Thus the payoff to persisting for the man who’s interested in sex = (.5*80) – 45 = -5, which is less that zero, the payoff to not persisting. The payoff to persisting for the man who’s interested in a relationship = (.5*100) – 45 = +5, which is greater than zero, the payoff to persisting. Thus men interested in only sex wouldn’t persist, and others would.

** But not as a legal one.


10 Responses

Subscribe to comments with RSS.

  1. Krishnapriya said, on February 17, 2014 at 11:41 pm

    There can be multiple equilibria. You are ignoring distributions.

    • Navin Kumar said, on February 18, 2014 at 12:07 am

      I know. But the darn post is already at 1500 words, and I don’t think it changes my point substantially.

  2. some guy said, on February 18, 2014 at 4:03 am

    I think your analysis is interesting, but you are conflating between two different versions of “No means No”.

    The weak version of “No means No” implies that any sexual contact should not be initiated without explicit consent.

    There is a second “strong” version of “No means No” perpetuated by feminists (Radical ?) / Social Justice Warriors. In this version any contact, verbal etc. are considered as an abuse.

    Finally there is an even stronger (insane) version, where even expressing a (non-romantic) interest, under certain circumstances is considered as abuse. Case in point Elevatorgate: where a woman was asked for a coffee inside an elevator at some atheist conference.

    Very few people disagree with weak version, its the strong/insane version where we actually test the limits of Freedom of speech / expression. At what point does the person loses rights to express his opinion/desire.

    Most social justice movements nowadays are trapped in puritan mentality, where participants are assumed to posses no agency or capability to make rational distinction. Its a worrying trend.

    Finally as a male in mid twenties, the stronger versions of “No means No” are useless in practice, and persistence is often a better strategy.

    • Navin Kumar said, on February 18, 2014 at 12:56 pm

      I think you should flip the rankings of the second and third entries on the list, but otherwise I agree: they are both not worth engaging with at the moment.

      I don’t think that there is one “weak” version but rather two. There’s the version in which “No Means No” implies that, if a woman says no, you should stop the act. There’s another version – “Yes Means Yes” – which “implies that any sexual contact should not be initiated without explicit consent.” Most people would be agree with the first, but not with the second.

  3. Dushyant said, on February 18, 2014 at 10:25 am

    Your analysis fails to control for what I’d like to call ‘class bias’ – or situations where women don’t feel they have agency, or do not, in fact, have agency. We now take agency as a given – the reality is that it isn’t, not for most of India.

    I suggest that before you conduct a further analysis, you consider empirical data on rape – in particular, data on profiles of victims and criminals. A professor at NLU Delhi is working on building such a data set, and the National Crime Records Bureau has some of it in raw form.

    While your paradigm is of limited use among the fractional section of women with the sort of agency you describe, it is for the most part, across India, irrelevant. While I agree, of course, that there are occasions when women say no and mean yes, before we use those instances to frame laws or policies or even moral codes, we must try and understand the prevalence of this. And one possible way is to look at economic indicators of independence – can the woman travel alone? Can she live alone? Does she have independent income? Is she in a position of financial dependence?

    Failing to account for this will possibly save some men who went ahead with sexual acts (rape) in situations where no actually meant yes. But it will act against thousands, millions of women who are saying no because it means no. Laws and policies atleast, are framed to protect these disadvantaged people against the advantaged. In a perfect world, the law would provide you with a test to differentiate between situations where no means no, and no means yes. In reality, this almost impossible. The (admittedly imperfect) situation is categorically accept, atleast legally, that no means no, and can never mean yes. And personally, if that protects the majority of women, I’m okay with that.

    I do completely agree with you, though, that we need to evolve moral rules of conduct that make it clear that no means no – and that make this clear for all persons, male or female. I do not, however, think that this particular model is at all convincing.

    • Navin Kumar said, on February 18, 2014 at 12:44 pm

      Hi Dushyant. Welcome to the blog.

      The model I presented was theoretical. I agree with you that, to be useful in any particular context, it requires empirical evidence to back up the assumptions. Furthermore, it’s true that the numbers I presented were drawn from an American study, and of a very specific phenomena (token resistance before sex). The numbers could be very different for India, and for situations like agreeing to a date.

      However, the numbers – specifically the proportion of women who say no when they mean yes, which I will refer to with the variable P for brevity – could be higher or lower (or the same). The question we must ask is: which is it ?

      If I understand you correctly, you believe that P is low, because women do not have agency in India. As you say, they aren’t permitted to travel alone, live alone, earn their own money etc.

      However, agency is context-specific. For example, it is not unknown, in countries like Saudi Arabia etc, for female legislators and politicians to be extremely vocal and dominating in the legislature, but submissive towards their husbands at home. They have agency at work, but not at home. Conversely, I have no doubt that Michelle Obama (for example) has an egalitarian personal relationship with Barack Obama. However, as the First Lady, she is expected to conform to certain standards of behavior. She has agency in the private sphere, but not in the public sphere.

      So the relevant question while discussing consent, fake resistance etc isn’t: do women have agency overall? It is: do women have agency in the context of the mating process? For some women, the answer is clearly no – victims of marital rape, for example. Coerced arranged marriages and child marriages are also not unheard of.

      However, India is changing. People have started dating and choosing their own mates. Pre-marital sex is not the taboo it once was. In this context – yes, women do have agency. How will they use it? Will P, in this context, be negligibly low in India? Will no always mean no? Not necessarily.

      One of the reasons a woman may offer token resistance is to convince herself (or her partner, or other people) that she is not a “slut.” Insofar that slut-shaming is higher in India than the US, we might expect to find a higher P in India than the US.

      Another reason I offered was to separate confident, assertive men from men who are seen as “weak” (phattu in India.) In places like Harayan, Delhi etc. where the culture of machismo is strong women may place an even greater weight of filtering out “weak” men, leading to a higher P.

      And so on. Depending on what the reason for false resistance is at a particular stage, we may find a higher or lower P in India relative to the US. Obviously, this is an empirical question, and I don’t know what P is. I merely hope to convince you to not assume that it’s necessarily nil, merely because women are disadvantaged is some spheres.

  4. Clarisse Thorn said, on February 19, 2014 at 1:16 pm

    I see that you linked to my site… did you also read my book about pickup artists? Based on what you’ve said here, I think you might like it a lot.

    • Navin Kumar said, on February 19, 2014 at 5:54 pm

      Hi Clarisse! Welcome to the blog.

      I’m something of a fan of yours, but I haven’t read Confessions yet. It’s on my To Read list. I’m quite certain I’ll enjoy it.

  5. Is game theory normative or descriptive? Is it even a theory? said, on February 26, 2014 at 3:35 pm

    how are you assigning pay-offs? Why is a relationship more valuable to a relationship seeker, than sex is to a player?

    What are the costs of pursuing? If we assume the costs of pursuing are negligible, then the theory fails..

    According to this we would have to assume that sexual harassment is caused necessarily (generally?) by a man who is interested in a relationship, on a woman who is not. Is that not against what we must think is impulsively right, which is the opposite.

    Also, quite often, the payoff to not persisting are negative (assuming again that pay-offs have been assigned in the way that we think the player might perceive them, which doesnt seem to be the case, but should necessarily be)

  6. […] More here. Interesting throughout. My opinion of affirmative consent can be found here. […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: