Malcolm Gladwell's Email Explains the Tipping Point

Malcolm Gladwell Sent: Sunday, January 20, 2008 1:03 PM To: Matt Elzweig Subject: Re: Questions

In answer to your questions:

The "tipping point" phrase was part of the vernacular of academic and policy circles long before I ever decided to use it as the title of my book. When I was reporter at the Washington Post, covering the AIDS epidemic, I used to hear epidemiologists use it all the time.

Your questions about Schelling and Grodzins are based on a significant misunderstanding. There are two very different academic traditions associated with the idea of tipping points.

Schelling and Grodzins represent the economic tradition. As I understand it (and I might be over-simplifying things here) the way economists think about tipping points goes something like this.

Suppose there are nine white people and one black person living in a neighborhood. Everyone is happy. But the white people vary substantially on the question of how many black people they are willing to live with. White family A will move as soon as there are two blacks. Family B will move once there are three blacks. Family C will move once there are four blacks. And the remaining six are more liberal, and will move only if the number of blacks reaches five. So what happens? One of the white families sells their house to a black family. That triggers white family A's racial tolerance, so they move and their house is bought by a black family. That triggers white family B's racial tolerance, so they sell, which trigger's C's toleranceand so on. Now there are five black families andboomthe rest of the whites families leave. The moment the neighborhood reaches five blacks is the tipping point.

Now, here's the way an epidemiologist thinks. A city has a small, but manageable syphilis epidemic, centered on a number of young, sexually active people in an inner city housing project. Then, simultaneously, one June, the housing project is torn down and the residents dispersed throughout the city, and the city cuts back on the hours of its public health clinics. Two months later, a syphilis epidemic sweeps the city. Why? Because the core of sexually active people who were the carriers have now been spread through the city. The cutback in the clinics means that those carriers the disease stay untreated much longer, maybe them far more infectious. The fact that this all happens at the beginning of the summer is crucial because unsafe sexual activity is far higher in the summer than the winter. To an epidemiologist, the tipping point for that epidemic is the combination of spring, the destruction of the housing project, and the cutback in the clinics.

That's a very different way of thinking about tipping points. The economic model is based on individual preferences, and can be represented precisely and mathematically. The epidemiological model is far more descriptive, and is based on much more than a calculation of the varieties of human preference. In the economic model, the white families don't have to interact at all, and nor do we have to assume that any one family is influenced by the actions of another. The economic model doesn't assume contagion as part of the process of change. The epidemiological model is entirely about contagion. It's highly interactive. It is premised on the notion that a small group of people wield extraordinary social influence, and have a direct impact on their peers. The economic model is just about economics. The epidemiological model combines biology, psychology, sociology and urban geography. Economists and epidemiologists both use the term tipping point. But they're using that term in profoundly different intellectual contexts.

My book is about the epidemiological model. In fact, although it's called the Tipping Point is really a book about epidemics, and how the epidemic model can be applied to social phenemona. The three principlesthe Law of the Few, the Stickiness Factor, and the Power of Contextare all simple restatements of the principles that epidemiologists use in analyzing epidemics. The researchers who are quoted in the book, as a result, are overwhelmingly epidemiologists, psychologists and sociologists. The intellectual bones of the book lie with diffusion theory and Granovetter's pioneering work on networks.

So when Schelling says I didn't ground my work in the tradition of tipping points, what he means is that I didn't write my book from the perspective of his academic tradition. And he's absolutely right. But I didn't intend to, because I don't think the economic model is at all useful in understanding the kinds of contagious social processes (smoking, suicide, crime, bestsellers, social power, fashion etc) that The Tipping Point is focused on. I think the economic model is fascinatingand that's why I credit Schelling in the endnotes and describe his work as an academic classic. But since I'm not building my argument based on the economic model, I'm not sure I need to do more than that.

If you are in any way uncertain about this distinction between economic and epidemiological models, please let me know. It's a pretty critical point.