I’m currently reading a rather fascinating book called Influence, The Psychology of Persuasion, by Robert B. Cialdini, PH.D. and last week I read in it something both astonishing and disturbing. Every time there is a well-publicised case of suicide in the media, there invariably follows a dramatic increase, not only in copycat suicides, but in reported car and plane accidents as well.
Why might it be more dangerous to travel by road or air immediately following a well-publicised suicide?
Could it be that the same socio-economic conditions that contributed to the motivation for suicide, led to an increase in carelessness at the wheel or joystick? It appears not, because the spikes in the number of accidents correlate, not with the number of suicides, but with the amount of publicity these suicides receive, and it appears highest in the areas where publicity is greatest.
OK then, could it be that reading about suicide makes people more careless and therefore more prone to accidents? Again, another detail hidden in the statistics suggests that it’s actually far more disturbing than this. When the suicide reported in the media is a simple one-person act of destruction, then the type of car and plane accidents increasing the most, also tend to involve only single victims. However, when the publicised suicide victim has taken others with him, there tends to be a marked increase in accidents with multiple fatalities. What could possibly explain this?
It was the sociologist, David P Phillips, who in 1974 hit upon a possible, somewhat worrying explanation – that these accidents were not in fact accidents at all. After trawling through decades of statistics from both the UK and US, Phillips proposed that the accident fatalities responsible for the spikes following publicised suicides, were in fact copycat suicides.
The Werther Effect
Ever since 1774 when the publication of Goethe’s The sorrows of Young Werther led to a wave of young male suicides across Europe (as well as an increase in young men dressing in the fashion of the protagonist, Werther), the phenomenon of copycat suicide, otherwise known as the Werther Effect, has been well documented. The difference here was that the cases examined by Phillips had been made to look like accidents – presumably to save friends and relatives the added distress of discovering that death had occurred by the victim’s own hand.
If this was true, Phillips argued, we should further find that the accidents following a well-publicised suicide would contain both a higher percentage of fatalities, and less evidence of evasive action having been taken. His rationale was that those intent on self-destruction would presumably prefer to ensure a quick death rather than risk being maimed or left severely handicapped for the rest of their painful lives. Sure enough, when he delved deeper into the details, this prediction was also borne out – both for car and aeroplane accidents . Indeed, this explanation as suicide-by-accident appears to fit the data better than any other proposal before or since, although it is not without criticism.
In his book, Influence, Cialdini uses this disturbing phenomenon as an example of the effect of Social Proof. This refers to one of the evolutionary short-cuts our brains use to determine what action to take in given situations. Whether aware of it or not, we all take cues from those around us. It is why, even though most of us find the canned laughter of sitcoms both annoying and obviously fake, it has been proven in numerous studies that we still find them funnier with it than without.
As a side note, the principal of social proof also explains why so few people in busy public places will stop to help those in need; we look around and see that nobody else appears to be worrying about the man lying on the pavement, groaning, and assume him to be another homeless drunk. Counter-intuitively, such a man is far more likely to be helped when the number of bystanders is very small (3 or fewer).
One property of social proof which also seems to validate Phillips’ explanation is that the effect is greatest among those of a similar background to the subject of the proof. In other words, if the publicised suicide victim is a young, white male from a middle-class background, the spike in accidents show a disproportionately large number of young, white, middle-class males.
I don’t know about you, but I find the implications of this, if true, extremely disturbing. It means that enough people out there are suicidal enough, that when given a suitable cue, their copycat suicide-by-accident will influence accident statistics in a noticeable way. Knowing this, I will, from now on, almost certainly try to avoid travelling in the immediate aftermath of any well-publicised suicides.
It also makes me wonder whether my own book, Connected, having now found its way onto nearly 40,000 Kindles and itself centred on two suicides, may one day be taken by some poor unfortunate as the cue for his own desperate exit from this world.