LIBERTARIANS, SMOKING, AND INSANITY: How Ideology Affects Ideas About Freedom and Health

Related articles

This "e-monograph" was reprinted in Liberty magazine. Todd Seavey is Director of Publications at the American Council on Science and Health ( and Editor of His views do not necessarily represent those of ACSH.Table of contents








The American Council on Science and Health, for which I work, is sometimes described as "libertarian except on smoking," a label that I will argue is inaccurate. (I should note, first, that we are a scientific organization and that we have advisors from all parts of the political spectrum, but we are frequently characterized as libertarian, and I'll happily accept that characterization for purposes of the current argument.)

The implication of the "libertarian" part of the label is that we want almost everything (except assault, theft, and fraud) to be legal. While ACSH's only official goal is to promote sound science, it is true that, as a practical matter, this often puts us at odds with environmental activists and others who want every imagined or hypothetical risk to be met with a new regulation. If you are presumed innocent until proven guilty, you should not be jailed, fined, or regulated simply because, say, some activist theorizes, without offering any good evidence, that you are elevating cancer levels or causing rashes among your neighbors. For simplicity, then, I'll accept the libertarian label.

What, though, causes some people to see our position on smoking as an exception to the libertarian rule? (And when I say "our position," I should note that ACSH has writers and advisors of varying political bents, united by their desire for sound science, but it is primarily our president, Dr. Elizabeth Whelan, whose vocal opposition both to nonsensical regulations and to smoking has earned us the "libertarian except" label, so I'll treat her position as the consensus ACSH view for purposes of the current argument. To get an idea what most libertarians think of anti-smoking efforts, see Jacob Sullum's article "What the Doctor Ordered" at and then see ACSH's response, as well as responses from Jeff Bucholtz and Joseph Olson, at

Does ACSH call for outlawing smoking? No. On the face of it, then, there must not be anything unlibertarian about ACSH's smoking position. Ah, but we do call attention to the terrible dangers of smoking: to the fact that it diminishes the health of all users and kills hundreds of thousands of Americans every year (or, if you insist, it is the means by which hundreds of thousands of Americans choose to kill themselves each year but I will use the two ways of phrasing the matter interchangeably).

This, it appears, is enough to make us suspect in the eyes of many libertarians and many conservatives who fear we are no different than left-leaning, pro-regulation anti-smoking groups; Ralph Nader-style safety fanatics; and other promoters of the nanny state.

This suspicion is not only unfair but is evidence of some problems, if not with libertarianism per se, then at least with libertarians as they attempt to apply the philosophy. As we'll see, these misapplications of libertarian principles can lead to extreme and implausible libertarian statements on matters ranging from smoking to suicide to schizophrenia.


Libertarians justly fear misguided government regulation, but their fear of regulation should not cause them to pooh-pooh all worries about the troubling social trends regulations are intended to combat. For the most part, libertarians understand this distinction, but they often become knee-jerk defenders of disapproved activity when it comes to smoking, dismissing its dangers. Thus, the American Council on Science and Health, despite its strong libertarian bent, has time and again found itself called unlibertarian for its constant warnings about the very real dangers of smoking.

Libertarians should be the last people to think that all criticism of a thing (such as smoking) logically implies a desire to outlaw that thing or, for that matter, that all praise logically implies a desire to subsidize. The whole point of libertarianism, I thought, was to remove most of life's myriad debates from the legislative arena. If, in the libertarian view, humans are not supposed to settle matters of right and wrong with laws (except for matters involving assault, theft, and fraud), humans should instead use private social pressure and that is all that ACSH's advice amounts to in the end, though that advice carries the imprimatur of science.

If libertarianism really meant "you cannot criticize anyone's private actions" (rather than simply "you shouldn't outlaw them"), libertarians would logically have to oppose not only anti-smoking commentary but weight-loss advice, movie criticism, etiquette, religious sermons, and baseball fans shouting "batter can't hit."

Furthermore, as I've noted in an earlier article, even if selling cigarettes ought to be legally permitted, we can still call the practice evil, as surely as we can call someone evil for tempting his friend to get drunk before walking out onto thin ice. Undoubtedly, the drunken friend bears final responsibility for his own actions, but it would be nihilistic of us to approve of the tempter's words. And moral nihilism is precisely the (unfair) charge, often leveled by conservatives, that libertarians are always denying. Just because libertarians want most things to be legal does not mean they are alienated, cocaine-snorting, porn-watching, prostitute-hiring, anti-social loner-misfits. Nor does it mean they must approve of others being that way.

In Defense of Busybodies

No one likes a busybody, but there are times when being a busybody is the appropriate course of action. If your cousin, who tends to take your advice, has become anorexic, it would be foolish to say that this was just a choice on her part, no better or worse than any other, and therefore behavior that you ought to respect and keep silent about. If you can persuade her to eat, do so, and I'm confident no one will call you unlibertarian for it. And if that sort of intervention is acceptable, then why not a strenuous effort to warn people about smoking, which kills hundreds of thousands of Americans a year and makes millions sick? If mass death is not reason to speak up, what is?

Would an organization be considered unlibertarian if, for instance, it consistently opposed regulation yet loudly urged people to wear bike helmets? Is your next-door neighbor being unlibertarian when she lectures you for not wearing boots when you go out in the snow? For that matter, would someone be unlibertarian if he adamantly opposed laws against pornography while just as adamantly urging people not to buy the stuff? I don't think so. Or, if such an individual is to be considered unlibertarian, the term "libertarian" does not mean what I thought it did when I adopted it, and I will have to drop the label after all. Nineteenth-century philosopher John Stuart Mill, often regarded as a precursor to modern libertarians, was himself fuzzy on this distinction, sometimes describing both Victorian social pressure and law as undesirable constraints, other times lauding the power of social norms as superior to the blunt instrument of law.

So under what circumstances should we expect libertarians to object to social pressure, and under what circumstances should we expect libertarians to accept social pressure?

I fear good old-fashioned "reverse psychology," which certainly works well for predicting the behavior of other political factions, may work well for predicting libertarian behavior as well: If the government threatens to outlaw something, libertarians may refuse to criticize that thing, or may even find something admiring to say about it. Such a response is perfectly reasonable when the thing threatened with legal sanctions is truly harmless, but smoking most certainly is not harmless. Libertarians might be on firmer ground if, say, they responded to recent talk of a "fat tax" by praising fattening foods (these can more easily be enjoyed in moderation than smoking), but even in that case, it would be absurd for them to start calling doctors unlibertarian for warning people about the dangers of high cholesterol and obesity. Let us not be knee-jerk defenders of every bad habit that the government threatens to squash, or we will become the libertine caricatures that critics say we are.

In short, we can criticize bad laws and bad behavior at the same time and without fear of inconsistency.

The Moral Responsibilities of Experts

Those who have some familiarity with statistics and medical science arguably have a special obligation to spread the word about dangers such as smoking, since most people are very bad at calculating probability, especially about long-term, far-off risks, such as a gradually-increasing risk of death by lung cancer and the spectrum of other cigarette-related diseases, including many forms of cancer, cardiovascular disease, chronic obstructive lung disease, and more (though there are subtler, more immediate negative effects from smoking).

The nihilistic libertarian temptation is to say that people simply calculate that the joy of smoking outweighs the very high risk of death (about one in three users), but most users likely aren't making that calculation carefully, and many are obviously working very hard to avoid thinking about the odds at all. Avoidance of thoughts about the future is particularly easy for shortsighted, undisciplined teenagers (i.e., most teenagers) which is why, despite all its denials, the tobacco industry thinks of teens as one of its most important markets.

Should public health experts sit back, secure in the knowledge (and happy to take it as an ideological axiom) that smokers "chose" their course of action, thus eliminating any reason for us to regret the consequences? (And even if that line of reasoning sounds plausible to some libertarians, what on Earth explains its appeal to some conservatives, when conservatives are perfectly happy to ban vices far less likely to be fatal, from marijuana to pornography?) Imagine that an engineer knowing that most people have a poor grasp of physics discovered that thousands of his fellow citizens were planning to rush onto a rickety bridge, resulting in injury to all of them and death to many. Would we fault him for screaming about the dangers of walking onto the bridge, perhaps even screaming repeatedly, until people were sick of listening to him?

Some of the people walking onto the bridge might choose to ignore the engineer. Some of them might insist they had gotten the message but show by their behavior that they weren't taking the threat seriously (or weren't rationally calculating the odds, given their apparent concern for their own safety in other settings and their lack of concern in this case). Most frustratingly for the engineer, some people might tell him with looks of weary anguish that they understand the odds but are now very, very attached to the idea of walking onto the bridge and just can't summon the willpower to find another route across the chasm.

Now you have some idea what it's like to be a public health expert and watch millions of people taking up smoking indeed, being urged to take it up, at a young age, by callous, lying tobacco companies while some critics call the public health expert a killjoy.

The Case for Smoking Being Particularly Bad

At the end of the day, the violation of property rights is the one thing to which libertarians are steadfastly opposed, and in a world where property rights are often trampled on, that's a very valuable core principle. There are some other criteria, fuzzier but still valuable, that are commonly used by people of a libertarian bent to spot danger, though:

(1) Libertarianism correctly identifies fraud as one of the most destructive human behaviors.

(2) Libertarianism also often counsels a wariness of so-called junk science, the use of shoddy, phony studies to dupe the public or push a given group's agenda.

(3) Libertarianism usually distinguishes between the rights of rational adults and the more limited rights of minors, presumably on the grounds that there is not as close a correlation between the desires of minors and their long-term happiness as there is between the desires of adults and their long-term happiness.

(4) Libertarianism rightly stresses the importance of individual decision-making, lamenting any loss of individual autonomy.

(5) Libertarianism is, broadly speaking (rights theories aside for the moment), utilitarian. That is, it generally tends to make long-term human happiness an important criterion by which to judge the success or failure of a policy (if millions of people were killed and maimed, libertarians would not be reassured by the assertion that those millions suffered for the greater glory of the emperor or suffered so their attacker could "make a statement" on behalf of animal rights, Islamic fundamentalism, or some abstract conception of social justice).

Libertarians, then, might not want smoking outlawed, but it would be unsurprising if they despised or at least strongly disapproved of the tobacco industry (and discouraged other people from smoking and selling cigarettes), since the industry would appear to violate or at least bend the five principles noted above, in the following ways:

(1) The tobacco industry has long made fraudulent claims about the safety of their products, and not just in order to stave off government regulation. The industry's goal, particularly before the link between smoking and lung cancer was familiar to the general public, was in part to lull consumers into a false sense of safety, in effect tricking them into killing themselves. Call it capitalism, call it fraud, call it what you will it isn't a nice thing to do to people. The familiar mandatory warning labels, seemingly a great blow to the industry, are in fact an industry-approved shield against lawsuits that blends all but unnoticed into the sea of labels and health warnings gracing everything from cyanide to snack foods in our society. The warning label does little, as the industry well knows, to inculcate the full range and relative severity of the risks posed by smoking. Most everyone knows, in the broadest sense, that smoking is unhealthy, but the devil is in the details and few people know them besides health experts and the tobacco industry.

The vague warning labels give no idea what "dose" makes one susceptible to the illnesses hinted at. In a world full of warning labels and minor risks, potential smokers are unlikely to get any sense of the (significant) relative risks of smoking from the brief labels. The tobacco companies, for all their public protestations that the risks of smoking are exaggerated, know that the risks are severe and numerous, many of them little-recognized by the general public and many, such as an increased risk of colon cancer, irreversible after just a few years of smoking, even if the smoker quits (see ACSH's pamphlet on The Irreversible Health Effects of Cigarette Smoking). Half of all lung cancer diagnoses now occur in people who had already given up smoking.

(2) The tobacco industry, through its Council on Tobacco Research, was for decades one of the most influential purveyors of junk science. They were eager obfuscators on behalf of a deadly product, and they were happy to disparage standard methods of epidemiology and widely-accepted notions of disease causation if it helped confuse potential customers. Libertarians and conservatives wouldn't accept this kind of nonsense from a group promoting fear of ozone depletion. They shouldn't accept it from the tobacco industry. (I should note that at least one prominent libertarian, my old boss John Stossel at ABC News, has consistently pointed out the dangers of smoking and even included the tobacco industry among the culprits in his one-hour special Junk Science.)

Even with the Council on Tobacco Research now defunct, many of the obfuscations it produced linger and are occasionally unwittingly repeated by libertarians and conservatives, such as the idea that smoking might simply be correlated with lung cancer rather than causing it a perversion of an often-sound warning against leaping to epidemiological conclusions, but a warning that has little to do with the strong link between smoking and lung cancer. After careful efforts to rule out confounding factors, it still appears that smoking accounts for a twelve to twentyfold increase in one's likelihood of developing lung cancer, a relatively rare condition among non-smokers. While it is true that "everyone knows cigarettes are bad for you," the industry has done a more artful job than most libertarians recognize of concealing or obscuring facts about smoking's dangers in order to leave the muddled impression that "smoking is bad, but then, so is everything else." Cigarettes marketed as "lite," low-tar, safer alternatives are a reminder that the industry wants people to think of smoking a as a manageable, only moderately dangerous activity.

(3) Despite their claims not to market to minors, tobacco companies' own documents show that they are acutely aware of their need to recruit new customers from the ranks of teenagers, who are notoriously bad judges of long-term consequences. While even children must be held accountable to some degree for their own actions, and even commercial speech featuring cartoon camels may deserve the protection of the First Amendment, surely we cannot be as comfortable with teens' "rational choice" to smoke as we might be with adults'. Tobacco companies aren't particularly subtle about their efforts to recruit teens, either, sometimes sponsoring massive cigarette giveaways in clubs likely to be frequented by youth. As with the classic strategy of the drug pusher, the first sample's free.

(4) But in the end, it's still the individual's choice whether to continue smoking, right? True, but are we to assume that by definition nothing bad ever happens to a person by choice? Some libertarians, as will be discussed below, reject the whole concept of "addiction," saying that it is simply an arbitrary label for what might more accurately be called "lifestyle choices for which some people will give up a great deal." Fair enough. I'm willing to grant that addiction can be described that way, even if addicts began their pattern of self-destructive choices in part because of deceptive safety claims. But again, does calling behavior "choice" place it beyond criticism? Should we not be troubled by any choices?

As Dr. Gilbert Ross, ACSH's medical director, once put it, an autonomous individual may, after repeated efforts to kick a drug habit, technically "choose" to abandon his livelihood, home, and family in favor of slowly dying in a flophouse with a heroin needle in his arm and that's the sort of choice that we mean when we say addiction. If labeling it "addiction" does not prove it is cause for concern, neither does labeling it "choice" prove it is not cause for concern. Frequent nicotine consumption reinforced by physical withdrawal symptoms, repeated unsuccessful efforts to quit, and usually by expressions of regret for having picked up the habit in the first place is a choice, but it would be naive to claim it is a choice as easily resisted as any other, such as whether to buy a particular pair of sunglasses. To pretend it isn't a troubling and special case would be to ignore the empirical facts. Just because smokers choose to smoke rather than having cigarettes forced upon them by armed guards does not mean that all is well.

(5) Some libertarians are strictly concerned with rights-adherence, not with utilitarian consequences (increases or decreases in human happiness), and don't really care (or more likely, claim not to care) how much misery is caused by some social trend so long as no one's property rights are violated in the process. I'll address the fundamental problems with that view at greater length below, though I would imagine most readers don't need my help to see them. Utility, I realize, is impossible to gauge with much accuracy absent telepathy, since we cannot know the intensity of other people's suffering or joy (that is the core insight of the "Austrian school" of economics, whose adherents logically conclude that we ought therefore to leave individuals free to make their own choices, pursuing their own diverse and unknowable preferences in the marketplace). Nonetheless, there are obviously situations in which outward manifestations of suffering are so apparent that we can reasonably extrapolate albeit without certainty about the sufferers' mental states. If that were not true, we would have no rational grounds for choosing who to aid first if confronted by the simultaneous sight of two people, one of them frowning because he can't find his car keys and the other screaming and weeping because his foot is caught in a bear trap.

I would contend that millions of people being ill and scores of thousands of people dying slow, excruciating deaths from lung cancer each year from an avoidable cause (few of them gasping "I have no regrets!" with their final breaths) is probably a non-utility-maximizing outcome. We might, at the very least, want to seek non-coercive means of alleviating that outcome. Most libertarians see the statistic about socialism killing some 100 million people in the twentieth century as pretty damning, and presumably not just because those people were coerced but because...well...they died. Might not a freely-chosen pattern of behavior that killed some 60 million people during the same time period at least be called troubling, even if we are not bothered by it in the same way that we are by political coercion?

Freedom of Choice Remains More Fundamental Than Health

Even with all the reasons to worry about smoking a (1) deceptively-marketed, (2) junk science-backed, (3) youth-tempting, (4) seemingly-addictive practice that (5) leads to mass death some people still say that cigarettes should be legal, since individuals should always, in the end, be allowed to make their own choices. Well, you know what? I agree, and so does ACSH. How's that for libertarian? As Felicia Ackerman, one of my philosophy professors at Brown, rightly noted: health is not necessarily something that everyone values highly, so they should be free to trade off health for other benefits, such as entertainment.

Yet surely, if some practice that has the characteristics described above is not a fit target for regulation, it is at least troubling enough exists in enough ethical grey areas that we ought to take special care, preferably in some non-coercive way, to minimize the negative effects of this particular exercise of freedom. That's all most of us at ACSH want: to non-coercively inform people of smoking's dangers and to allow narrowly-targeted lawsuits against the tobacco industry in those cases where members of an aging, dwindling cohort of smokers can genuinely argue they were fraudulently assured of tobacco's safety a half-century ago and took up the habit as a result. Is that so unlibertarian? (Whether any individuals can legitimately claim to have been deceived by the industry and whether a substance can be addictive enough to take the blame for years of a user's consumption are topics for courtroom debate, but they are not debates that can be ruled out of bounds by basic libertarian principles libertarians still believe in courts, after all.)

Many of us at ACSH are sufficiently libertarian that we favor decriminalizing marijuana, by the way, in part because there is far less evidence for it being addictive or lethal than there is for cigarette smoking.

We do not support special taxes on tobacco nor the familiar mandatory warning label nor the tobacco industry's Master Settlement Agreement, which has been squabbled over in recent years by all the participants. We encourage voluntary social pressure because we don't want people to suffer and die needlessly, and we approve of lawsuits when warranted by demonstrable fraud (and the documents exist to show that the tobacco industry knowingly lied to their customers about the substantial risk of death, most brazenly around the 1950s but throughout the subsequent decades as well, as noted in an article by Dr. Whelan).

Does this concern about smoking, then, make us fascists? Well, at least one libertarian appears to say yes. But then, as we'll see, some libertarians will call you a fascist if you believe that there is a mental state we can label "addiction"...or for that matter, if you actually believe that there is a disease called "schizophrenia." Examining the strange but in some circles very popular views of these radical defenders of eccentric behavior will help explain how ACSH ended up at odds with some of its ideological kin on the issue of smoking.


In the May/June 2002 issue of Society, libertarian psychologist Jeffrey Schaler notes the growing tendency to medicalize all behavioral problems: people who in the past would have been called philanderers are now called sex addicts; irresponsible spenders are now shopaholics; and so forth. Since most people think that the mentally incompetent need treatment or even physical restraint, explains Schaler, this contemporary tendency to medicalize all undesirable behavior becomes a convenient excuse for making everyone either wards of the state or steady sources of income for psychiatrists. Obviously, there is some truth to Schaler's complaint, but he quickly concludes that virtually any expression of concern about "public health" (he uses scare quotes) is thinly-veiled fascism, or at least closely related to fascism philosophically:

In a parallel with Dr. Himmler's "racial hygiene"'s new public health ideology can be characterized as moral hygiene...Both are forms of medical and public health imperialism...Human beings are again being homogenized by the state, in a manner reminiscent of National Socialist Gleichschaltung.

Attacking the concept of alcoholism, Schaler notes:

Cirrhosis of the liver is a disease caused by habitual heavy drinking. Habitual heavy drinking is not a disease. Habitual heavy drinking does not become a disease by being called "alcoholism." Similarly, a broken neck is (in the broadest but still literal sense) an illness, but hang-gliding is not an illness, and hang-gliding would not become a disease by being called suspendere labi.

It will come as no surprise that Schaler is the author of a book entitled Addiction Is a Choice, which attacks the pervasive idea that drug users have no real control over their behavior.

An Ancient Debate in Modern Form: Free Will vs. Determinism

Again, there is some truth to Schaler's argument, but most people would see human behavior as falling along a sort of continuum with thoughtful, informed, carefully chosen behavior on one end, physically-coerced action (such as being grabbed by the neck and forced to kneel) on the other end, and addiction somewhere in between (with those forms of addiction that involve intense physical withdrawal symptoms perhaps falling a bit closer to the coercive end of the spectrum than such pseudo-addictions as "shopaholism"). The continuum model is particularly easy to accept if one is not, in the strict philosophy-class sense of the phrase, a believer in "free will." That is, it is easier to think of ordinary choice, addicted choice, and coercion as differences of degree (rather than utterly dissimilar phenomena) if one is a "determinist": one who believes that the brain is, after all, a physical thing and that each decision it makes could at least in principle be traced back to identifiable prior causes.

This is not to say that the chain of causation necessarily leads back quickly to events external to a person's own skull despite the eagerness of some determinists to point to external historical and economic forces as the immediate causes of every human decision in order to minimize individual responsibility for crime or immoral behavior. One can be a determinist and still think that Person A threw a punch primarily because he's a violent jerk, even if one thinks that ultimately one could trace the genetic and environmental circumstances that led to the creation of this violent jerk. A determinist has as much reason as the free will advocate (if not more) to want criminals imprisoned, kind people praised, and rude people chastised. After all, social incentives are an important cause of behavior, and those of us who (thanks again to our genetic and environmental influences) want the world to be a happy place would not want to see incentives for proper behavior eliminated.

Many libertarians, however, see free will in the philosophy-class sense, not just the legal sense of "being allowed to act without getting put in jail" as a natural complement to their political views and are repelled by any deterministic arguments, since they seem to suggest that some people have more control over their actions than others. Many libertarians are wary of talk of "addiction" or "criminal insanity" not so much because they have dispassionately examined the world and found that the phenomena of addiction and insanity do not exist but because it would be extremely convenient for a philosophy rooted (as they see it) in radical free will if there were no troubling cases such as addiction and insanity.

(Indeed, it would also be convenient if there were no children, since they are generally agreed to be incompetent in many situations to choose in accordance with their own best interests. So it is that libertarians sometimes manage to tie themselves into knots on the topic of the rights of children, with opinions ranging from the view that children have no rights against their parents at least, none enforceable by third parties such as the state to the view that children have exactly the same rights as adults and thus must be bargained with and cajoled but never picked up and hauled screaming out of the playground when dinner time comes.)

For the believer in radical free will, then, addiction and insanity become not just behaviors arranged along a continuum of human action but vexing cases that must be shoehorned into one of two rigid, completely separate categories: chosen and unchosen behavior. And since a free will adherent wouldn't want to coddle people who behave badly by saying that they had no choice but to drink or act crazy, he is left with no choice but to insist until he is blue in the face that even the most addictive or insane behavior is as freely chosen as opting for a red instead of a blue sofa on a shopping excursion. One ardent Schaler fan with whom I am acquainted a talented fiction writer with a theatrical bent likes to caricature the determinist view (and the ACSH position on smoking) as the belief that cigarettes foist themselves on smokers against the smokers' wills, like tiny automatons that leap unbidden into the mouth and murder their innocent victims.

Well, of course, no one at ACSH thinks smokers are that helpless, but my friend is unintentionally parodying her own views, since only the ardent free will advocate such as a Schalerite would see the smoking issue in such cartoonish, black and white terms: Either people rationally, happily opt to continue smoking or else they are violently forced by some external power to do things they don't want to, like slaves or prisoners. The pretence that all human action fits tidily into one of these two categories makes it much easier to be a libertarian and to dismiss any concern about addicts or the insane as authoritarian thinking.

Libertarians have tolerated these convenient but flawed arguments for too long. It is extremism of this sort that renders libertarianism which has many sound observations to offer about economics and law a marginal movement, linked in the minds of many people more to the anarchic excesses of the 60s than to the ideals of the Founders or the principles of a sound economy.

Where Do They Get These Ideas?

Rebellious psychiatry professor Thomas Szasz, now in his eighties (and particularly influential in the 1960s), is the man who has to be given much of the credit or blame for promoting the radically non-judgmental approach to human behavior that has done so much, directly and indirectly, to shape libertarian views on smoking.

However, Szasz does not stop at calling vices such as smoking freely-chosen behavior patterns (that much, we might all agree, is largely a matter of semantics or a matter best left for the aforementioned philosophy classes to debate). In books such as Schizophrenia: Sacred Symbol of Psychiatry, Szasz argues that even schizophrenics are just people who behave badly or in socially unacceptable ways. Mental illness is, one of his book titles tells us, a "myth." He has also written books comparing psychiatry to slavery and the Inquisition not just for poetic effect, mind you, but as examples of similarly oppressive regimes. A brief introduction to his views can be found at Szasz's official website, maintained by Schaler: As Szasz's online manifesto states:

The term "mental illness" refers to the undesirable thoughts, feelings, and behaviors of persons...The classification of (mis)behavior as illness provides an ideological justification for state-sponsored social control as medical treatment...Classifying thoughts, feelings, and behaviors as diseases is a logical and semantic error...

No one who has heard stories of people institutionalized because of their sexual orientation or their dissident political views (as was common in the Soviet Union) can fail to sympathize with Szasz's fear of psychiatry as a means of social control. But to go so far as to call mental illness a myth even to put "schizophrenia" in scare quotes and refer to it as a "supposed illness" is absurd. I do not exaggerate when I say that Szasz and his strictest followers consider schizophrenia little more than a rationalization for irresponsible behavior. Szasz notes, for example, that "schizophrenics" always seem to hear voices that counsel bad behavior providing them with a handy excuse but never good behavior (though it's possible that people are quick to label the good voices "conscience").

No doubt feigned insanity (or the self-deluding belief that one is and must remain mentally incompetent) is sometimes a handy excuse for crimes or other irresponsible behavior. And another oft-repeated argument Szasz makes that no physical abnormalities can be identified as the cause of most familiar mental illnesses, making it impossible to objectively identify such illnesses in the way that one might identify a broken leg may be reason enough, as a practical matter, to abolish the insanity defense in criminal cases and to be extremely cautious about having people committed to asylums. Some people may just be very eccentric, after all. Some people like to water ski and others like to wander the streets unwashed muttering Bible verses to themselves. Just because behavior is odd (or evil, for that matter) doesn't mean it's an illness.

Problems with Szasz: Ignoring Inconvenient Facts?

Naturally, libertarians defend individual freedom of action so long as those actions do not directly harm others, and in saying schizophrenics are fundamentally no different from people with strange hobbies, Szasz is defending their freedom against a society that may not respect them enough. In many legal battles, I would no doubt find myself on Szasz's side (indeed, my aforementioned philosophy professor, Felicia Ackerman, has often said that being institutionalized would be her worst nightmare, since there is no clear way to "prove" that one is sane). But like adherents of all other ideologies, some libertarians perhaps including Szasz may tend to ignore inconvenient facts of life in order to make the world "fit" more smoothly with their ideology. Just as the elimination of the concept of addiction (and the embrace of the radical free will concept) makes it easier to argue against drug prohibition, the elimination of the concept of mental illness makes it easier to argue against involuntary institutionalization or paternalistic psychiatry (which is not to say these political positions cannot be defended by other means on the contrary, it would better for everyone if they _were_ defended by other means).

To some extent, the insanity debate, like the addiction debate, is a semantic one. Szasz often insists that calling mental disorders "illnesses" is a clear-cut definitional error, "like calling a whale a fish," since illness, by definition, must involve observable physical dysfunction. But in insisting that schizophrenics are simply choosing to behave in socially-unacceptable ways, Szasz, Schaler, and their followers seem to resist the common-sense interpretation of some overt manifestations of mental illness, such as seizurelike or hallucinatory conditions.

A young person I know (who, ironically, moves in libertarian circles but is not the Schalerite fiction writer mentioned above) recently developed what most people would regard as classic signs of schizophrenia (though I won't pretend we understand the brain well enough to create a foolproof checklist of symptoms). Previously a confident, outspoken, and very healthy-looking person, my acquaintance became convinced that passersby on the street and fellow passengers in elevators were glaring with undisguised hate, that the staff of a local hotel was putting poison or knock-out drugs in the food they served, and that strangers held mysterious secrets to financial success and should be contacted in the middle of the night. My acquaintance lost the ability to keep track of time and developed the strange habit of getting into cabs without being able to give the driver any instructions on where to go, other than "to someplace safe". My acquaintance, once a lively conversationalist, would repeat the same points obsessively such as the desire to locate specific lost objects all the while looking terribly thin and haunted. Anyone relying on common sense rather than Szaszian stubbornness would have thought "there's something terribly wrong," as I did, long before hearing any official diagnosis.

What is baffling is that Szasz and Schaler have no doubt encountered far more extreme cases of (what most of us call) mental illness without their disbelief in mental illness being shaken. Does someone exhibit high scientific standards or just pig-headed stubbornness if he insists that mental illness is a myth even after meeting, say, a man who constantly convulses and shrieks and insists that he must wear a tinfoil hat to prevent the CIA robots in his stomach from receiving instructions via radio? I don't doubt that there are Szaszians dogmatic enough that they could sit through a conversation with such a delusional individual just as calmly as if speaking to someone with minor misconceptions about geography or someone who mistakenly thinks that President Reagan was elected in 1978.

We may not understand the brain processes responsible for mental illness, but we are justified in strongly suspecting we are dealing with mental illness when something such as schizophrenia produces the same basic symptoms across time and across cultures, in people who sometimes lose so much and suffer so much because of their conditions and have so little to gain from feigning an inability to function or from suddenly adopting self-destructively eccentric lifestyles.

The Real Philosophical Error: When Is Anything an Illness?

There is a more serious problem with the Szaszian insistence on identifying specific physical causes before labeling something mental illness, though.

Szaszians will sometimes concede that a condition marked by both aberrant behavior and a clearly-defined brain abnormality or other physical cause can rightly be labeled mental illness. So, for instance, if overexposure to heavy metals consistently produced delusions, the Szaszian would say the demands of science had been satisfied: both aberrant behavior and a physical cause have been identified.

But wait a second: Why should even this be called mental illness, in the Szaszian view? If we are insistent (whether out of scientific conviction or political expediency) that behavior cannot generally be labeled an illness, why should the fact that it is preceded by a physical event such as heavy metal exposure matter?

I am reminded of a chemist I knew who, after exposure to mercury early in his career, became convinced (whenever he went off his medication) that he was Norwegian royalty. Now, presumably Szaszians wouldn't have called his conviction that he is royalty mental illness if he had manifested the belief in the absence of a physical cause. Why, then, call it madness even in the presence of a physical cause? After all, we don't call every thought that a person has after metals exposure madness. If a man were exposed to heavy metals and then decided to take up cross country skiing, we wouldn't call the skiing insanity simply because it was preceded by an identifiable physical event.

If the Szaszians are willing to call claims of royalty that follow metals exposure mental illness but not willing to call cross country skiing that follows metals exposure mental illness, then they must admit (contrary to their stated principles) that they do judge the sanity or insanity of actions and beliefs by the content of the actions and beliefs, such as the failure of beliefs to jibe with obvious external reality. The Szaszians must, then, either hold that nothing (even in the presence of observable physical causation) can be labeled insane or else abandon their philosophy and join the rest of us in judging people insane however cautiously on the basis of observed behavior.

For Schalerites and Szaszians to imply that they are simply awaiting more detailed descriptions of the physical causation of mental illness is either a lie or a non-sequitur. By their own reasoning, they have no rational grounds for calling anything a mental illness even if we can describe the phenomenon down to the individual neural firing. All we can say, given their linguistic stubbornness, is that, for example, neural firings caused by sexual hormones lead to arousal, while heavy metals lead to believing you're Norwegian royalty. Why call either an illness or dysfunction by Szaszian standards? We should, by their reasoning, just say caffeine leads to heightened alterness, a bullet through the skull leads to altered language use, LSD leads to seeing weird hands coming out of the ground, heavy metals lead to fanciful notions about Scandinavian succession, as-yet-unknown brain events lead to thinking the hotel staff is out to poison you, and "retardation" leads people to "choose" to be very bad at math. Why call any of it a medical condition or illness? These are all just different ways that a brain can be, different lifestyle options, if you will. All is well!

And any Szaszians tempted to bite this particular bullet and just say that no brain state (whether physically understood or not) can ever be labeled dysfunctional or diseased, must contend with a further problem: How is it, then, that we can even label a heart, liver, or lung diseased? If one wanted to be as stubborn about physical illness claims as the Szaszians are about mental illness claims, couldn't one insist that a heart that stops beating is not really "ill" in any meaningful sense but rather is "behaving in a socially disapproved way"? Who's to say that a femur bone ought to remain in one piece instead of being broken up into twelve parts and causing pain? Isn't it just narrow-mindedness on our part to label this femur dysfunctional while labeling the intact ones "healthy"?

That sounds absurd, but then it should also sound absurd to insist that the mind has no identifiable healthy functions, such as perceiving external reality, differentiating between real people and imagined ones, forming basic logical chains of thought, or spotting contradictions and reaching decisions.

Anyone who accepts Szaszianism may someday find himself having the unsettling and upsetting experience of being forced to reject those views due to a direct confrontation with the madness of a relative or friend who was once healthy. I am less concerned, though, about the emotional impact of disillusionment on Szaszians than I am about the fate of their mentally ill friends and patients, who may not get the treatment they need because of Szaszian dogmatism. Anyone who has dealt with schizophrenics knows that some of their most "eccentric" notions are often notions about their medication. The last thing that a potentially self-destructive schizophrenic who perhaps thinks his medication is poison left for him by Martians needs is to be told by an ardent Szaszian that those views about his medication, while mistaken, are within the bounds of sane discourse.


Even if one suspends for the moment all debate over whether schizophrenics are mentally ill and whether smokers are addicted, it would seem reasonable for ACSH to assert that certain activities such as smoking (or attempting to stop a bus by willing one's body to turn into steel) are harmful, even if wholly voluntary.

That is, it would seem reasonable unless one were a particularly dogmatic libertarian and some people are.

Following the Austrian-School economics insight mentioned near the end of Part One (the observation that only an individual can gauge his own happiness), libertarians commonly equate happiness with preference fulfillment. This makes sense. People who want one thing to happen but have another thing forced upon them are usually not happy about it (which is why we fear kidnappers, rapists, and imprisonment). But people can have short-term preferences that are not in accord with the fulfillment of their preferences over the long-term. While drunk, you may want to jet ski, but the resulting paralysis and brain damage may hinder your long-term goal of being a ballerina, for instance. Observation of the real world seems to suggest only that choice and happiness are strongly correlated, not that they are synonymous.

Furthermore, choice and happiness should not be declared synonymous by definition or tautology simply to make a philosophy of individual liberty easier to defend. This point is often made in the pages of the journal Critical Review by its editor, Barnard political science professor Jeffrey Friedman (see Friedman, much like ACSH's Whelan, is sometimes described as unlibertarian (or "postlibertarian," as he insists on putting it) for thinking that libertarians should, in essence, abandon their principles but seek pragmatic, utilitarian arguments for their political program. After all, if your principles do not lead to practical, utilitarian outcomes, rival intellectuals not to mention the rest of humanity are likely to ask why they should bother listening to you. It was partly Friedman's utilitarian consequentialism that inspired me to criticize a libertarian-influenced, pro-tobacco columnist in one of my first columns for ACSH's webzine,

Most libertarians chose their principles because of an empirical (not purely deductive) belief that individual decision-making tends to lead to happiness (after all, why would we feel obliged to respect any principles that caused a net decrease in human happiness?). But even if this tendency (for choice to bring happiness) is so strong that we ought to elevate it to the status of a political principle (let people do what they want with their own bodies and property so long as they do not harm others' bodies or property), it would be intellectually dishonest to pretend that the principle can now be used as proof that no instances of misery-inducing free behavior exist (perhaps even some instances in which individuals would have been happier if legally forbidden to follow their own preferences). To turn the original empirical generalization into a principle and then use the resultant principle to define away all the empirical "tough cases" which is what I fear most libertarians do when confronted with addiction, insanity, or old-fashioned cases of temptation is a bit like noticing that the local theatre almost always shows the best movies on its biggest screen, deciding that one should generally see the movies that play on the big screen, and then, when confronted with a fabulous movie that is being shown on a smaller screen, stubbornly declaring that it cannot be a great film, since it has already been established that the best movies tend to show on the big screen.

Or to put it in more immediately relevant terms: The fact that individual choice works like a charm (that is, increases happiness) 99% of the time may be sufficient reason to adopt an overall anti-government stance or an overall attitude of tolerance, but it cannot tell us whether addiction is an exception to the "choice works" rule, nor for that matter can it tell us whether insanity or a hasty decision to commit suicide while in the throes of depression is an exception to that rule. Once one admits that choice and happiness are merely correlated rather than synonymous, one has to ask whether certain unusual sorts of choices might fall into that hypothetical 1% of cases in which individual choice does not work like a charm and instead leads to misery.

Death and Illness Are Genuine Problems

Any social problem can be looked at either as an opportunity for utilitarian debate or merely as one more opportunity to adhere to libertarian principles. If, for instance, a raging forest fire is about to destroy a town, most people would say the problem is the fire. A sufficiently stubborn libertarian might say that the problem is that the local fire department has not been privatized. While many people might agree that the fire department could be run more efficiently if private, few sane people would say, with flames licking at their homes, that privatization is the primary or immediate concern in this scenario.

Likewise, while it is perfectly legitimate for libertarians to insist that regulations be kept at bay in the smoking debate, it would be narrow-minded in the extreme for them to say that regulations are the primary or only problem in the smoking debate. If everything from death by lung cancer to living as a raving maniac in the streets is simply another choice to be respected, one must ask whether any outcome would ever convince the diehard hyper-Austrian that choices can be wrong wrong in that they lead to misery (by the individuals' own standards of happiness and misery). If humanity were reduced to a sobbing, confused mass of starving wretches, eager to commit suicide after years of self-mutilation and heroin addiction but, hypothetically, might have been spared that fate if forced (only for a moment) to take anti-psychotic pills at a young age and might thus have lived self-reportedly happy and productive lives full of laughter and strong friendships should we, as principled freedom-lovers, be indifferent between the two scenarios?

Libertarians, faced with the question "What if everyone starved under your system?" typically respond, appropriately enough, that they know enough economics to be confident that will not happen. Fine. But implicit in such an answer, as Prof. Friedman likes to point out, is the utilitarian belief that widespread human happiness is the ultimate test of political principles. If the principles produce misery (not just in one isolated case but on balance, over the long haul), there is no sane reason to adhere to them.

The question libertarians who care about utility must imagine themselves being asked, then, before they dismiss the concerns of anti-smoking groups is: "All right, what if virtually no one starved under your system and there was widespread peace and prosperity but some four million people died from smoking around the world each year?" The proper utilitarian solution to this problem may not involve the tiniest bit of government coercion cultural pressure and education may be sufficient to turn people off smoking without any regulations or taxes (or it may indeed be the case that regulating smoking, even if it were the only way to diminish illness and death, would open a Pandora's Box of other social controls and should therefore be rejected) but libertarians should at least be interested in the problem. Libertarians should not dismiss those who worry about smoking as authoritarians (though some anti-smoking activists no doubt are), and they should not play into the hands of their critics by appearing callous about illness or about the painful and premature deaths of many of their fellow beings. Creeping socialism is a problem, but so is death.


When confronted with the fact of widespread cigarette consumption leading to widespread death, a reasonable person wonders if there is some way to retain the benefits of consumer freedom while limiting the disastrous health effects. The ideologue of the left instead says that cigarettes are further proof that capitalism is evil, that it always leads to big companies exploiting and harming the public, and calls without a second thought for massive regulation or even prohibition. (The ideologue of the traditional right opposes the regulations largely, I suspect, to annoy the ideologue of the left.) The libertarian ideologue, meanwhile, says that consumer freedom is sacrosanct and in the end always wise, and he insists that the buying and selling of cigarettes is as ethically neutral an activity as buying and selling Tinker Toys.

What neither the leftist nor the overzealous libertarian is willing to admit is that cigarettes are different, a genuine special case. Nothing else we know of packs the same addictive, popular, legal, deadly, and youth-attracting wallop to public health. If every other product on the market really were as bad for you as cigarettes, a reasonable person probably should abandon free market principles in favor of heavy restrictions on personal liberty.

But cigarettes are unique. So the leftist is wrong. And so is the overzealous libertarian. More reasonable libertarians and I am confident, despite the flack that ACSH gets, that there are many will examine the unique health threat posed by cigarettes in scientific terms first, then consider whether the threat they pose warrants no action, some regulation, or, as ACSH suggests, a combination of concerted voluntary efforts to educate potential and current smokers and, in those cases where it is warranted, suits for fraudulent marketing. If an educated public really cannot be persuaded that the costs of poor health and death outweigh the joys of smoking even if the industry openly and honestly describes the immense risks of their product and their legacy of deceiving customers in order to get them to imperil their lives ACSH is not likely to turn to statist, coercive, regulatory solutions. But saying that is not the same as saying cigarettes pose no vexing problem, and libertarians should acknowledge that.

That willingness on the part of libertarians to engage in messy, practical debate would benefit both public health and the libertarian movement itself, by showing it to be capable of dealing with imperfect reality instead of just the tidy categories constructed by ideology.

Time for Excommunication?

Some will say a libertarian who is less than 100% a priori committed to freedom of choice one who is willing to consider, even for the sake of argument, impinging on individual liberty in the name of utility is no libertarian at all, even if he ultimately counsels voluntary, non-regulatory responses to such crises as smoking and debilitating mental illness.

To that, the proper post-ideological, pragmatic, utilitarian response is probably: Who cares? It's more important to be right than to retain some political label as though it were a badge of tribal membership. Nonetheless, I would call myself and ACSH libertarian and would remind readers who disagree just how flexible such labels have proven in the past and may yet prove in the future: The Libertarian Party has a strict oath forbidding its members to endorse any initiation of physical force but even among the party loyalists there is disagreement about whether this logically implies a tiny government or outright anarchism. The nineteenth-century political writers who first called themselves libertarians, on the other hand, were ardent individualists but, unlike today's Libertarian Party, were not always loyal to the free market. Many dreamt of liberating individuals from both state and capital, like the left-anarchists of more recent times. Some left-anarchists, particularly in Europe, still call themselves libertarians today and resent what they see as the usurpation of the label by capitalists. (But then, what prominent political label hasn't undergone similar transmutations over the past two centuries, with conservative, liberal, moderate, communitarian, communist, and socialist all meaning pro-government at some points in time and anti-government at others, sometimes both at the same time?)

Clear, agreed-upon terminology is convenient, but rigidly fixed political labels may just be a last vestige of humanity's waning tribalism a crutch for those who can't cope with ongoing debate and changing minds. If one were to hew to the strictest interpretation of the Libertarian Party definition of libertarian, for instance, it's not clear that Murray Rothbard, Ayn Rand, Friedrich Hayek, or Milton Friedman would qualify as libertarians, when in fact all are giants of libertarian thought. So let's not waste energy declaring ACSH unlibertarian or "libertarian except on smoking." Better to spend that mental energy trying to figure out how to help millions of smokers break a deadly habit, boring and non-ideological a task though that may seem to some.

Dare I say that ACSH's position on smoking, rather than getting us excommunicated, ought more rightly to be considered the sensible libertarian position? Perhaps the day will come when people simply say ACSH is "libertarian for instance, they want to find ways to get people to stop smoking voluntarily instead of being taxed and regulated into stopping."

I should note that our position has resulted in us being threatened with excommunication by left-wing anti-smoking groups as often as by right-wing pro-tobacco forces, though being attacked by both sides doesn't necessarily prove that one is correct.

In the end, whether called libertarian or not, our primary concerns will be advancing human wellbeing and maintaining the integrity of science. Those goals don't seem to us to be at odds with liberty.


One of the nice things about presenting an argument in an online format, as opposed to the pages of a traditional book, is that the Net has a sort of built-in humility to it: One expects every online pronouncement that attracts attention to be attacked, reformulated, batted around, refuted, debunked, or amended almost immediately.

This state of affairs is perfect for ideologies, such as libertarianism, that can abide evolution and change. It is anathema to the Communist rulers in Beijing and wouldn't have suited Stalin or Hitler, either. Implicit in the argument presented in this e-monograph is the assumption that libertarianism is a philosophy in flux. Mainstream libertarianism may come, in time, to absorb the ACSH position on smoking. Or it may hunker down and cast ACSH into exile (forcing us to call ourselves "postlibertarians," perhaps). Either outcome, though, will be the result of dialogue as opposed to a pronouncement from some central authority.

The 1990s, post-Cold War upsurge in punditry new cable channels, more commentary, shows such as Politically Incorrect was a taste of things to come, with rapid-fire, ephemeral talk replacing (for good or ill) the older pattern of influential books dominating public discourse for years at a time. The next phase in intellectual history may well be modeled on the endlessly-sparring blogs (web logs) that now exist online, where (for good or ill) writers expect their pronouncements to be criticized or amplified by friends and foes alike. Instead of a philosophical manifesto being issued with the expectation that its flawless wisdom will last for all time, ideas can now be tossed into the so-called blogosphere floated like trial balloons so that author and reader alike (and the line between the two is now far from clear) can quickly learn, digest reactions, and revise.

This is not so different from the way science normally works, with constant peer review and conflicting reports taking place. Philosophy in the day to day academic world works this way as well, though few members of the public ever see that face of philosophy, instead making due with a vague knowledge of a few greats that have stood the test of time: Plato, Kant, Nietzsche.

How much healthier it would be if political philosophy now becomes something that evolves and learns too quickly to allow for the formation of rigid, warring ideologies like the ones that nearly destroyed the world in the past century. Instead of a few rigid, doctrinaire, clearly-defined philosophies duking it out like competing tribes, the world may soon become accustomed to a perpetually-bubbling, continually-refined cauldron of political-philosophical ideas, all in conversation with each other.

Surely a freedom-based philosophy such as libertarianism will do well in that environment (while various rigid orthodoxies from religious fundamentalism to the anti-biotech movement will probably fare poorly). There should be room enough in the mix for serious consideration of ACSH's views on smoking. Instead of simply asking in yes-no fashion whether those views are libertarian, perhaps the question should be: Will libertarianism, or some variant of it, adapt in order to incorporate those views?

See also: ACSH's site for teens detailing the specific health effects of smoking,