Most likely at least one adult — probably a coal miner from Poland, Czechoslovakia, or Italy — brought the virus to Limburg. This one adult could have died from AIDS with little notice… He could have transmitted the virus to his wife and offspring. His infected wife (or girlfriend) could have given birth in a Swedish barrack to a child who was HIV infected but seemingly healthy. Unsterilized needles and syringes could have spread the virus from child to child.

The truly strange thing about this story, of course, is that not all of the children died. Only a third did. The others did what today would seem almost impossible. They defeated HIV, purged it from their bodies, and went on to live healthy lives. In other words, the strains of HIV that were circulating back in the 1950s were a lot different from the strains of HIV that circulate today. They were every bit as contagious. But they were weak enough that most people — even small children — were able to fight them off and survive them. The HIV epidemic tipped in the early 1980s, in short, not just because of the enormous changes in sexual behavior in the gay communities that made it possible for the virus to spread rapidly. It also tipped because HIV itself changed. For one reason or another, the virus became a lot deadlier. Once it infected you, you stayed infected. It stuck.

This idea of the importance of stickiness in tipping has enormous implications for the way we regard social epidemics as well. We tend to spend a lot of time thinking about how to make messages more contagious — how to reach as many people as possible with our products or ideas. But the hard part of communication is often figuring out how to make sure a message doesn't go in one ear and out the other. Stickiness means that a message makes an impact. You can't get it out of your head. It sticks in your memory. When Winston filter-tip cigarettes were introduced in the spring of 1954, for example, the company came up with the slogan "Winston tastes good like a cigarette should." At the time, the ungrammatical and somehow provocative use of "like" instead of "as" created a minor sensation. It was the kind of phrase that people talked about, like the famous Wendy's tag line from 1984 "Where's the beef?" In his history of the cigarette industry, Richard Kluger writes that the marketers at R. J. Reynolds, which sells Winston, were "delighted with the attention" and "made the offending slogan the lyric of a bouncy little jingle on television and radio, and wryly defended their syntax as a colloquialism rather than bad grammar." Within months of its introduction, on the strength of that catchy phrase, Winston tipped, racing past Parliament, Kent, and L amp;M into second place, behind Viceroy, in the American cigarette market. Within a few years, it was the bestselling brand in the country. To this day, if you say to most Americans "Winston tastes good," they can finish the phrase, "like a cigarette should." That's a classically sticky advertising line, and stickiness is a critical component in tipping. Unless you remember what I tell you, why would you ever change your behavior or buy my product or go to see my movie?

The Stickiness Factor says that there are specific ways of making a contagious message memorable; there are relatively simple changes in the presentation and structuring of information that can make a big difference in how much of an impact it makes.

Every time someone in Baltimore comes to a public clinic for treatment of syphilis or gonorrhea, John Zenilman plugs his or her address into his computer, so that the case shows up as a little black star on a map of the city. It's rather like a medical version of the maps police departments put up on their walls, with pins marking where crimes have occurred. On Zenilman's map the neighborhoods of East and West Baltimore, on either side of the downtown core, tend to be thick with black stars. From those two spots, the cases radiate outward along the two central roadways that happen to cut through both neighborhoods. In the summer, when the incidence of sexually transmitted disease is highest, the clusters of black stars on the roads leading out of East and West Baltimore become thick with cases. The disease is on the move but in the winter months, the map changes. When the weather turns cold, and the people of East and West Baltimore are much more likely to stay at home, away from the bars and clubs and street corners where sexual transactions are made, the stars in each neighborhood fade away.

The seasonal effect on the number of cases is so strong that it is not hard to imagine that a long, hard winter in Baltimore could be enough to slow or lessen substantially — at least for the season — the growth of the syphilis epidemic.

Epidemics, Zenilman's map demonstrates, are strongly influenced by their situation — by the circumstances and conditions and particulars of the environments in which they operate. This much is obvious. What is interesting, though, is how far this principle can be extended. It isn't just prosaic factors like the weather that influence behavior. Even the smallest and subtlest and most unexpected of factors can affect the way we act. One of the most infamous incidents in New York City history, for example, was the 1964 stabbing death of a young Queens woman by the name of Kitty Genovese. Genovese was chased by her assailant and attacked three times on the street, over the course of half an hour, as thirty-eight of her neighbors watched from their windows. During that time, however, none of the thirty eight witnesses called the police. The case provoked rounds of self-recrimination. It became symbolic of the cold and dehumanizing effects of urban life. Abe Rosenthal, who would later become editor of the New York Times, wrote in a book about the case:

Nobody can say why the thirty-eight did not lift the phone while Miss Genovese was being attacked, since they cannot say themselves. It can be assumed, however, that their apathy was indeed one of the big-city variety. It is almost a matter of psychological survival, if one is surrounded and pressed by millions of people, to prevent them from constantly impinging on you, and the only way to do this is to ignore them as often as possible. Indifference to one's neighbor and his troubles is a conditioned reflex in life in New York as it is in other big cities.

This is the kind of environmental explanation that makes intuitive sense to us. The anonymity and alienation of big-city life makes people hard and unfeeling. The truth about Genovese, however, turns out to be a little more complicated — and more interesting. Two New York City psychologists — Bibb Latane of Columbia University and John Darley of New York University — subsequently conducted a series of studies to try to understand what they dubbed the "bystander problem." They staged emergencies of one kind or another in different situations in order to see who would come and help. What they found, surprisingly, was that the one factor above all else that predicted helping behavior was how many witnesses there were to the event.

In one experiment, for example, Latane and Darley had a student alone in a room stage an epileptic fit. When there was just one person nest door, listening, that person rushed to the student's aid 85 percent of the time. But when subjects thought that there were four others also overhearing the seizure, they came to the student's aid only 31 percent of the time. In another experiment, people who saw smoke seeping out from under a doorway would report it 75 percent of the time when they were on their own, but the incident would be reported only 38 percent of the time when they were in a group. When people are in a group, in other words, responsibility for acting is diffused. They assume that someone else will make the call, or they assume that because no one else is acting, the apparent problem — the seizure-like sounds from the other room, the smoke from the door — isn't really a problem. In the case of Kitty Genovese, then, social psychologists like Latane and Darley argue, the lesson is not that no one called despite the fact that thirty-eight people heard her scream; it's that no one called because thirty-eight people heard her scream. Ironically, had she been attacked on a lonely street with just one witness, she might have lived.