Jun 17, 2013

On the bystander effect and how it perversely protects victims of male violence

In the wake of the terrible news about Charles Saatchi seemingly assaulting his wife in public last Saturday, one of the most persistent bits of outrage I’ve seen is directed not at him, but at the other patrons of the restaurant in which the alleged assault took place. Namely: how could they have not intervened? How could they have just gone on with their lunch? Why did the photographer not do something? They are all terrible, callous, selfish human beings & should be ashamed of themselves.

It’s a question that comes up a lot: how come nobody did anything? In all sorts of crimes, but especially crimes of male violence against women and children, when the crime becomes public and the full facts of it emerge (or as full as can ever be the case for events that are normally private and hidden), it becomes obvious that there was in fact a web of people who had known all along that something was wrong and chose not to intervene. The recent revelations of systemic rape and sexual exploitation of vulnerable girls in Oxford is a good example.

The culprit, though, is not people’s inherent callousness and lack of empathy: it’s a psychological phenomenon called “the bystander effect”, and it’s an incredibly powerful instinct. It works like this: when there is more than one person witnessing some kind of event (say, a man throttling a woman), the likelihood of any one of them responding/intervening is in inverse proportion to their number. In other words, the more people witness an attack, the less likely any one of them is to offer help to the victim.

There are all kinds of reasons hypothesized for this phenomenon (read the Wiki article), but I think at least in part it has to have an adaptive element to it. Being the first person to respond to an abnormal situation can be dangerous, especially if one is not sure of the social rules governing the situation. Be the case as it may, the point I’m making is that the bystander effect doesn't make us bad people, just people who have a tendency to react with a remarkable degree of similarity to certain stimuli.

The good news about the bystander effect is that it’s relatively easy to disrupt. As soon as one person moves to intervene – it can literally be something as simple as getting up and turning to walk in the direction of the incident – the charm is usually broken and more people will offer help. The question, specifically in the case of what looks like a domestic violence incident, is: should they?

Male violence in the context of intimate relationships is not just a series of unconnected incidents; it has its own cycle and follows an internal logic that can sometimes be counter-intuitive. It’s possible that the bystander effect unwittingly protects women who are victims of violence from worse violence. Confronting an abuser might feel good to us, and might stop the attack that is happening in front of our eyes, but it is highly likely to lead to further retribution for the victim behind closed doors. The perpetrator may, and in fact probably will, punish the victim for “embarrassing him” by “making him” hit her in public, exposing him to what he sees as undeserved shaming.

The most dangerous times for victims are times at which the perpetrator feels like his power over them is being questioned or shaken. Most people now know that a key turning point can be when a victim tries to leave (the majority of women who are murdered by partners are killed during or just after leaving), but the challenge can come from outside the relationship as well as inside. For example, when family members try to stage an intervention or openly get the victim away from the perpetrator, she often ends up getting  worse than usual beating, to reinforce the man’s power over her.

So while it’s tempting (and certainly common) to sneer with contempt at the over-privileged restaurant patrons who failed to lift a finger to help Nigella Lawson, we must remember that 1) it’s not really their individual moral failing and 2) it might have put her in greater danger. As it happens, she has already been put in greater danger, by the exposure of the attack in the press and the media circus around it. Charles Saatchi has not, so far, paid any kind of price for allegedly attacking her; in fact, he’s been given a bigger than usual platform by the Evening Standard, to pooh-pooh the whole thing as a “playful tiff”. Meanwhile we don’t know how his wife is doing, or what she is thinking and doing – or whether she is even OK.


  1. thank you so much for this. There is so much anger towards the bystanders of this incident, when really they did nothing wrong. Very few people would have intervened, and i keep reading commenters saying 'well I would have jumped straight in and saved the day' with very little insight on how naive they sound.

  2. Ok, firstly, you’re wrong, because it is a zero-sum game.

    Where to start...
    Your blog name, “It's not a zero-sum game” implies at least a basic knowledge of game theory. As I'm sure you are aware, a number of games in game theory assume that all players are rational actors.

    The public, in this instance, are rational actors, as are the man and the woman.
    You assume that Saatchi (or indeed ‘the man’) in this situation, will continue to beat the woman once they are home.

    A zero-sum game, in basic terms, is where the benefit for one player is directly equivalent the loss to another player. On a hypothetical scale of 10 to -10, if one player gets a benefit of 6, the other player would get a loss of -6, etc.

    Now, for it to be a zero-sum game, we have this situation:

    The man beats the woman, and gains a benefit (gratification, pride, self-worth, whatever you like). The public intervenes, causing a loss. Then, when they go home, as you assume, the woman is beaten again, the man has to beat her more to make up for the loss he suffered by the public. Here, we have the benefit to the man, being directly equivalent to the loss of woman.
    So, let’s say that the public don’t intervene, as you assume they won’t. The man beats the woman, and beats her at home.

    Thus, the outcome is the same, thus zero-sum, because her loss (pain) is directly equivalent to his gain (beating).

    Thus, as you can see your argument doesn’t tie with theory. This is because the theory (and a Wikipedia reference, really?) states that it’s not in people’s interest to intervene. In this situation, it’s only not in people’s interest to intervene if you assume the man will beat the woman at home. Which you don’t know and the public doesn’t know either, so they will only see ‘gains’ in their actions, and thus it is in their interest to intervene.

    Also, separately from the ‘game theory’ you used, the bystander effect is not applicable to the Oxford case, as it was only one neighbour who intervened from the safety of their home without social pressure. Moreover, from the articles I have read on this case, the Oxford gang did not carry out their abuse in public where the bystander phenomenon would be valid.
    I do not condone these acts of violence, especially as a woman. If you are going to use theories such as the Bystander effect ect., and terms such as ‘zero-sum game’, you would be well advised to do a little background reading on them, as it degrades your argument and makes feminists look weak.

  3. Read your article, very stimulatiing! Debate.

  4. To the girl who knows her Game Theory ... DING DONG!

  5. Very good point about the bystander effect. Essentially, it is a restatement of the herd effect in animals. It takes an individual to intervene, and that breaks the effect causing more to come forward. As to whether the effect protects the victims (by not escalating the violence), is debatable, but you make a fair point.