ALTHOUGH THE HEALTH CONSEQUENCES of smoking have been known for many years, tobacco companies have avoided product-liability judgments for one reason: Smokers have always been held responsible for smoking. But in recent years a new paradigm has emerged. Now smoking is thought to be an addiction, and nicotine is held responsible for smoking. If the old view was that smokers abused nicotine, the new view is that nicotine abuses smokers, which is all that the legal profession needed to bring the tobacco industry to its knees.

This new thinking is said to be based on medical science. In the February 5, 1997, Journal of the American Medical Association, Dr. David Kessler, former director of the Food and Drug Administration, wrote of a "scientific consensus that the nicotine in cigarettes and smokeless tobacco causes and sustains addiction" (emphasis added). A close look at the evidence, however, shows that those who embrace this "consensus" have stepped out on a limb. While research has expanded our understanding of nicotine, it has not provided evidence that nicotine "causes" smoking. In fact, everything we know suggests that smokers cause smoking.

The concept of nicotine addiction springs partly from the phenomenon of nicotine withdrawal. Abruptly stop smoking, and measurable anxiety ensues. Withdrawal symptoms last several days, and every former smoker can tell you how unpleasant they are. Addiction theory has it that smokers are forced to consume more nicotine to alleviate these symptoms. But the theory does not account for the millions who endure withdrawal and quit. Rather than supporting the addiction model, the existence of withdrawal symptoms merely explains why quitting is difficult.

The phenomenon of nicotine craving is similarly used to support the addiction model -- and similarly fails to do so. Recent research has demonstrated that craving is associated with biological activity in the brain -- but of course all thoughts and feelings are associated with biological activity in the brain. Craving, like withdrawal, makes quitting hard.

That nicotine creates its own psychological and physiological incentives is indisputable, but it does not follow that smoking is irresistible. Smoking is complex behavior that requires a series of calculated steps. We have no scientific evidence that nicotine commandeers that process, meaning we have no reason to believe that smoking happens against one's will. We do know, on the other hand, that smokers can quit, as millions have.

Although smoking appears to be an addiction, it is better understood to be what psychologists call contingent behavior. People smoke to obtain rewards. The most common reward is nicotine euphoria (positive reinforcement), but over time smoking brings other rewards, like the suppression of withdrawal and craving (negative reinforcement). Smokers have the ability to quit, but instead they pursue the immediate rewards of smoking at the price of its long- term costs. That most who try to quit end up failing is not so much evidence that smoking is an addiction as much as it is confirmation of what we would expect: Smokers, as a self-selected group, place a relatively high value on short-term rewards.

Pursuing short-term rewards at the expense of long-term costs is understandable. Most people demonstrate some such behavior, if not with nicotine, then with food, sex, money, or power. Moreover, the long-term costs of smoking are not universal. Most smokers get neither lung cancer nor heart disease. Thus, reducing smoking to a "nicotine addiction" misses the much larger issue that smokers like to smoke, even if many have a simultaneous desire to quit.

While the weaknesses of the addiction model are evident enough in theory, real-world problems arise when that model is the basis for social policy. Policies that treat smoking as an addiction actually increase the incentive to smoke. The proposed settlement between the tobacco industry, plaintiff's lawyers, and the state attorneys general is a case in point. If smoking is contingent behavior -- if people smoke to obtain rewards -- then browbeating the tobacco industry (as if smoking were an addiction) absolves smokers of the moral responsibility for smoking, diminishing the personal ownership that is a prerequisite for quitting. Furthermore, if any money from the settlement ever trickles down to smokers, it will create an entirely new enticement to smoke.

For that matter, why have the states been picking up the tab for smoking- related health costs, which is what prompted the tobacco lawsuits in the first place? If smokers overuse health care in the long run (and the point is debatable), that is good reason not to spread the financial risk. Whether government or tobacco companies absorb the costs, the effect is the same: to remove from smokers a consequence of their actions that would otherwise have the potential to limit smoking.

Americans should oppose the pending tobacco settlement, and not simply because lawyers will be the only ones who benefit. If tobacco litigation has any effect on smoking rates, it will probably raise them. Besides, if science can be misused for the sake of routing smoking, it can be made to justify the removal of any liberty. And no one should blow smoke at that.

Michael J. Reznicek is a psychiatrist in Omaha, Nebraska.