What's the hook? smoking is more than a chemical bond.

During the presidential election campaign, Bill Clinton successfully cast Big Tobacco as a national enemy, with Bob Dole playing the role of collaborator by downplaying the addictiveness of nicotine. Meanwhile, the Food and Drug Administration has been asserting jurisdiction over cigarettes as "nicotine delivery devices," arguing that tobacco companies intend to hook their customers, just like schoolyard drug pushers. Hundreds of pending lawsuits, including class actions and cases filed by state governments, similarly allege a conspiracy to addict smokers. These developments represent important changes in our attitudes toward cigarettes. Though justified in the name of public health, the increasing emphasis on the enslaving power of nicotine may only make matters worse.

Understanding why requires careful consideration of the conventional wisdom about tobacco addiction, which recycles mistaken assumptions about illicit drugs. During the latter half of this century, the classical model of addiction, derived from observations of narcotic abuse, increasingly has been used to describe the cigarette habit. The classical model states that consumption of certain chemicals causes a physical dependence, either immediately or after prolonged use, characterized by withdrawal symptoms - symptoms that can be avoided or escaped only by further drug use. As Steven Hyman, director of the National Institute of Mental Health, opined recently in Science, "Repeated doses of addictive drugs - opiates, cocaine, and amphetamine - cause drug dependence and, afterward, withdrawal."

This cyclical model, in which the drug serves as both problem and solution, offers a simple, easy-to-grasp account of the addiction process, giving the concept great staying power in the public imagination. In the case of smoking, this view of addiction is central to the rationale for regulating tobacco and the concern that the cigarette companies have been doping their products with extra nicotine. But the classical model tends to conceal rather than elucidate the ultimate sources of addiction, and it is just as ill-suited to the cigarette habit as it has always been for understanding illicit drug use.

If a chemical compound can be addictive in the manner described by NIMH Director Hyman, we would expect anyone who regularly uses such a substance to become addicted. Yet only a minority of those who use illicit drugs - whether marijuana, cocaine, or heroin - ever develop a dependence on them. The prevalence of addiction, as defined by the American Psychiatric Association's Diagnostic and Statistical Manual, among users of alcohol and cocaine runs about 15 percent and 17 percent, respectively. Even in a sample of 79 regular crack users, Patricia Erickson and her colleagues at Toronto's Addiction Research Foundation found that only about 37 percent used the drug heavily (more than 100 times in their lives), and 67 percent had not used in the past month. A similar pattern holds for tobacco. In the 1994 National Household Survey on Drug Abuse, 73 percent of respondents reported smoking cigarettes at some time, but only about 29 percent reported smoking in the previous month, and not necessarily on a daily basis. Writing in the May/June Mother Jones, Jeffrey Klein manages to argue that nicotine enslaves its users and, at the same time, that Tobacco Inc. seeks to recruit young smokers to replace the 1.3 million Americans who quit each year. If nicotine is so relentlessly addictive, how can it be that 50 percent of all Americans who have ever smoked no longer do?

The classical model also suggests that the cigarette habit should be highly amenable to nicotine replacement therapy, such as the nicotine patch. Yet few of the tens of thousands of patch users have actually broken the habit (only about 10 percent to 15 percent succeed). In direct conflict with the classical model, most keep smoking while on the patch, continuing to consume the carcinogens in cigarette smoke while obtaining considerably higher blood levels of nicotine. A 1992 study of nicotine replacement therapy reported in the journal Psychopharmacology concluded that the "overall lack of effect [of the patch] on cigarette consumption is perhaps surprising and suggests that in regular smokers the lighting up of a cigarette is generally triggered by cues other than low plasma nicotine levels."

Most people who successfully quit smoking do so only after several failed attempts. If addiction is driven by physical dependence on a chemical - in this case, nicotine - relapse should occur during withdrawal, which for nicotine typically lasts a few weeks. Yet a sizable proportion of relapses occur long after the smoker has suffered through nicotine withdrawal. In fact, studies do not even show a relationship between the severity of withdrawal and the likelihood of relapse. As any former smoker could tell you, ex-smokers crave cigarettes at certain times and in certain situations for months, even years, after quitting. In these cases, the desire to smoke is triggered by environmental cues, not by withdrawal symptoms. This is one reason why people who overcome addiction to illicit substances such as heroin or cocaine often say they had more difficulty breaking the cigarette habit. Because regular tobacco users smoke in a wide array of circumstances (when bored, after eating, when driving) and settings (home, work, car), the cues that elicit the urge are more ubiquitous than for illicit drug use.

These failures of the classical model illustrate how conventional wisdom over-simplifies the dynamics of cigarette smoking. This reductionist view is dangerous because it ignores the psychosocial factors that underlie addiction. In coming to terms with cigarette addiction as a psychosocial process, rather than a simple pharmacological one, we need to distinguish between cigarette addiction and nicotine addiction. Certainly no one (except perhaps the tobacco companies) denies that cigarette smoking can be addictive, if by addiction one means a stubborn urge to keep smoking. But it is quite a different matter to say that nicotine accounts for the addictiveness of smoking. Nicotine withdrawal notwithstanding, nicotine alone is insufficient, and may even be unnecessary, to create cigarette addiction.

This claim can be clarified by two dramatic case studies reported in the British Journal of Addiction in 1973 and 1989. The earlier article described a 47-year-old woman with a two-and-a-half-year-long dependence on water, one of several such cases noted by the author. The woman reported a nagging withdrawal symptom - a dry, salty taste in her mouth - that was alleviated by the persistent drinking of water (up to 60 glasses per day). This case of dependence on a nonpsychoactive substance contrasts sharply with the second account, which described an 80-year-old woman who used cocaine without incident for 55 years. The authors reported that "she denies any feelings of euphoria or in-creased energy after [snorting] the cocaine nor any depression or craving for cocaine when her supplies run out.... She appears to have suffered no ill effects from the prolonged use of cocaine in physical, psychological or social terms." So we see that not every addiction involves drug use and not every instance of drug use involves an addiction.

To say that cigarette addiction is a psychosocial process means that social, cultural, and economic factors play a crucial role in acquiring and keeping a cigarette habit. In fact, the tendency to reduce the cigarette experience to chemical servitude may be one of the most powerful cultural factors driving addiction. Cigarette lore wrongly teaches smokers (and smokers-to-be) that they will suffer badly if they attempt to quit, while at the same time freeing them of responsibility for their drug use once they begin. Such beliefs also help romanticize cigarette smoking, elevating nicotine to a sublime abstraction. This not only reinforces the forbidden fruit effect, it helps transform the habit of smoking into a cult behavior. Smoking thus acquires the kind of meaning that the youth of America are most in search of: social meaning. As Richard Klein writes in Cigarettes Are Sublime, "smoking cigarettes is not only a physical act but a discursive one - a wordless but eloquent form of expression."

To counteract the forces that give momentum to drug use, the public meaning of addiction needs to be broadened to include the many, changing facets of the psychosocial realm in which we develop. "Putting people back in charge" of their addictions, as John Leo puts it in U.S. News & World Report, will not work if we focus only on the naked individual. Rather than pushing the pendulum of public policy between scapegoating the substance and scapegoating the individual, we should seek a middle ground. Realizing that the addiction process has at least three levels of complexity is a good place to start.

First, at the basic and most immediate level, are the short- and long-term biological processes that underlie the psychological experiences of drug use and drug abstinence. Even with the same drug, these experiences vary greatly across individuals. Scientists and journalists too easily forget that every psychological process is built on biology. Discoveries of biological mechanisms and processes underlying addiction are not proof that the problem is biological rather than social and psychological. Eating rich foods has powerful biological effects in both the short and long run, but we should not therefore conclude that the rise in obesity in the United States is a biological problem. Indeed, attempts to alter the addiction process that emphasize biochemistry (such as the nicotine patch) have met with little success.

At the next level are psychological processes (social, motivational, learning) that, although rooted in biology, are shaped by personal experience. Because each of us has unique life experiences, we do not necessarily interpret the same events in the same way. The reasons for one individual's addiction may be altogether different from the reasons for another's. As the recent Scottish film Trainspotting makes clear, stories of addiction are no less complex than any other personal stories. Still, intervention at this level has had some success with users of alcohol or illicit drugs, and several research and treatment institutions are examining methods for "matching" addicts with different treatment strategies based on their social and psychological characteristics.

Drug effects and drug addiction also vary greatly across time and place, implicating cultural factors as the third and most general aspect of drug addiction. These factors are rooted in but not reducible to psychological processes, just as psychological processes are not reducible to biology. Patterns of alcohol use around the world, which show that the prevalence of drinking problems cannot be predicted by consumption alone, illustrate the importance of culture. Italians, for example, historically have consumed large quantities of alcohol with relatively low rates of drunkenness and alcoholism. The effects of alcohol on human behavior - violence, boorishness, gregariousness - also have been shown to vary dramatically across cultures.

Given the cultural role in addiction and the radical changes that have occurred in attitudes about smoking, it is quite possible that the young smokers of today are not at all like the smokers of 50 years ago. Those who begin smoking now do so with the belief that it is addictive, causes poor health (and wrinkles!), and can be deadly. If individuals are willing to start smoking despite such knowledge, it is likely that they will acquire and keep the habit, seeming to confirm the current, politically correct image of addiction. And if this self-fulfilling prophecy is realized, chances are that interventions aimed at the social realm will continue to miss their target and fail to curtail addiction.

Richard J. DeGrandpre ([e-mail]), an independent scholar of drugs and culture, is a visiting assistant professor of psychology at Saint Michael's College in Vermont. He is co-editor, with Warren Bickel, of Drug Policy and Human Nature (Plenum) and coauthor, with Stanton Peele, of The Changeable Self (forthcoming from Simon & Schuster).