I think there are two possibilities. The first can be found in the correlation between smoking and depression, a link discovered only recently. In 1986, a study of psychiatric outpatients in Minnesota found that half of them smoked a figure well above the national average. Two years later, Columbia University psychologist Alexander Glassman discovered that 60 percent of the heavy smokers he was studying as part of an entirely different research project had a history of major depression. He followed that up with a major study published in the Journal of theAmerican Medical Association in 1990 of 3,200 randomly selected adults. Of those who had at some time in their lives been diagnosed with a major psychiatric disorder, 74 percent had smoked at some point, and 14 percent had quit smoking of those who had never been diagnosed with a psychiatric problem, 53 percent had smoked at some point in their life and 31 percent had managed to quit smoking. As psychiatric problems increase, the correlation with smoking grows stronger. About 80 percent of alcoholics smoke. Close to 90 percent of schizophrenics smoke. In one particularly chilling study, a group of British psychiatrists compared the smoking behavior of a group of twelve- to fifteen-year-olds with emotional and behavioral problems with a group of children of the same age in mainstream schools. Half of the troubled kids were already smoking more than 21 cigarettes a week, even at that young age, versus 10 percent of the kids in the mainstream schools. As overall smoking rates decline, in other words, the habit is becoming concentrated among the most troubled and marginal members of society.

There are a number of theories as to why smoking matches up so strongly with emotional problems. The first is that the same kinds of things that would make someone susceptible to the contagious effects of smoking — low self-esteem, say, or an unhealthy and unhappy home life — are also the kinds of things that contribute to depression. More tantalizing, though, is some preliminary evidence that the two problems might have the same genetic root. For example, depression is believed to be the result, at least in part, of a problem in the production of certain key brain chemicals, in particular the neurotransmitters known as serotonin, dopamine, and norepinephrine. These are the chemicals that regulate mood, that contribute to feelings of confidence and mastery and pleasure. Drugs like Zoloft and Prozac work because they prompt the brain to produce more serotonin: they compensate, in other words, for the deficit of serotonin that some depressed people suffer from. Nicotine appears to do exactly the same thing with the other two key neurotransmitters — dopamine and norepinephrine. Those smokers who are depressed, in short, are essentially using tobacco as a cheap way of treating their own depression, of boosting the level of brain chemicals they need to function normally. This effect is strong enough that when smokers with a history of psychiatric problems give up cigarettes, they run a sizable risk of relapsing into depression. Here is stickiness with a vengeance: not only do some smokers find it hard to quit because they are addicted to nicotine, but also because without nicotine they run the risk of a debilitating psychiatric illness.

This is a sobering fact. But it also suggests that tobacco may have a critical vulnerability: if you can treat smokers for depression, you may be able to make their habit an awful lot easier to break. Sure enough, this turns out to be the case. In the mid-1980s, researchers at what is now the Glaxo Wellcome pharmaceutical firm were doing a big national trial of a new antidepressant called bupropion when, much to their surprise, they began getting reports about smoking from the field. "I started hearing that patients were saying things like, 'I no longer have the desire to smoke,' or 'I've cut down on the number of cigarettes I'm smoking,' or 'Cigarettes don't taste as good anymore,"' said Andrew Johnston, who heads the psychiatry division for the company. "You can imagine that someone in my position gets reports about everything, so I didn't put much stock in them. But I kept getting them. It was very unusual." This was in 1986, before the depression — smoking link was well understood, so the company was initially puzzled. But what they soon realized was that bupropion was functioning as a kind of nicotine substitute. "The dopamine that nicotine releases goes to the prefrontal cortex of the brain," explains Johnston. "That's the pleasure center of the brain. It's what people believe is responsible for the pleasure, the sense of wellbeing, associated with smoking, and that's one of the reasons it's so hard to quit. Nicotine also increases norepinephrine, and that's the reason that when you try to quit smoking and you no longer get so much norepinephrine, you get agitation and irritability. Bupropion does two things. It increases your dopamine, so smokers don't have the desire to smoke, then it replaces some of the norepinephrine, so they don't have the agitation, the withdrawal symptoms."

Glaxo Wellcome has tested the drug — now marketed under the name Zyban — in heavily addicted smokers (more than 15 cigarettes a day) and found remarkable effects. In the study, 23 percent of smokers given a course of anti-smoking counseling and a placebo quit after four weeks. Of those given counseling and the nicotine patch, 36 percent had quit after four weeks. The same figure for Zyban, though, was 49 percent, and of those heavily addicted smokers given both Zyban and the patch, 58 percent had quit after a month. Interestingly, Zoloft and Prozac — the serotonin drugs — don't seem to help smokers to quit. It's not enough to lift mood, in other words; you have to lift mood in precisely the same way that nicotine does, and only Zyban does that. This is not to say that it is a perfect drug. As with all smoking cessation aids, it has the least success with the heaviest smokers. But what the drug's initial success has proven is that it is possible to find a sticky Tipping Point with smoking: that by zeroing in on depression, you can exploit a critical vulnerability in the addiction process.

There is a second potential Tipping Point on the stickiness question that becomes apparent if you go back and look again at what happens to teens when they start smoking. In the beginning, when teens first experiment with cigarettes, they are all chippers. They smoke only occasionally. Most of those teens soon quit and never smoke again. A few continue to chip for many years afterward, without becoming addicted. About a third end up as regular smokers. What's interesting about this period, however, is that it takes about three years for the teens in that last group to go from casual to regular smoking — roughly from fifteen to eighteen years of age — and then for the next five to seven years there is a gradual escalation of their habit. "When someone in high school is smoking on a regular basis, he or she isn't smoking a pack a day," Neal Benowiu, an addiction expert at the University of California at San Francisco, says. "It takes until their twenties to get to that level."

Nicotine addiction, then, is far from an instant development. It takes time for most people to get hooked on cigarettes, and just because teens are smoking at fifteen doesn't mean that they will inevitably become addicted. You've got about three years to stop them. The second, even more intriguing implication of this, is that nicotine addiction isn't a linear phenomenon. It's not that if you need one cigarette a day you are a little bit addicted, and if you need two cigarettes a day you are a little bit more addicted, and if you need ten cigarettes you are ten times as addicted as when you needed one cigarette. It suggests, instead, that there is an addiction Tipping Point, a threshold — that if you smoke below a certain number of cigarettes you aren't addicted at all, but once you go above that magic number you suddenly are. This is another, more complete way of making sense of chippers: they are people who simply never smoke enough to hit that addiction threshold. A hardened smoker, on the other hand, is someone who, at some point, crosses that line.