The Science of Stigma

This essay applies the naturalistic conclusion about the non-existence of free will to the problem of addiction and our attitudes towards drug abusers. The initial choice to use drugs is often voluntary, but voluntary choices are just as caused as compelled choices, an insight which should lead us to re-evaluate punitive social policies brought to bear against addicts.

For a more complete exploration of these issues, see Materialism and Morality: The Problem with Pinker. The Science of Stigma was published in The Scientist in the August 17, 1998 edition, under the title "To help addicts, look beyond the fiction of free will." See the comments on this essay. See also Michael Massing's New York Times article, Seeing Drugs as a Choice or as a Brain Anomaly.

Ordinarily, we don’t suppose that people are to blame for their illnesses. That is, prior to discovery, many diseases develop independently of what the sick person does or thinks. This is why the disease model of addiction, widely espoused in the therapeutic community, is so controversial. Commonsense suggests that a person’s choice to start using an addictive substance is often voluntary, and often made with the knowledge, either vague or specific, of the risks of getting hooked. Of course, some people may have a genetic predisposition to dependence, and some are placed in environments in which turning down that first drink or cigarette is nigh impossible, but nevertheless it’s hard to eliminate voluntary choice in our explanations of why individuals become addicts.

This question is now very much central to the current debate on how we should respond to the problem of addiction. Alan Leshner, director of the National Institute on Drug Abuse, urges that dependence is a chronic, relapsing disease and that, therefore, we should treat addicts as sick and not punish them for their illness. Leshner points out that the addict's brain is radically changed by drugs--"hijacked," as he puts it--so that the ability to resist drugs is severely compromised, if not altogether eradicated. Punishment won't help an addict to get clean, while treatment just might. But Leshner also recognizes that the physical, compelling nature of full-blown dependence does not exonerate an addict from having made the choice to start using drugs.

In a recent interview in The Scientist (12[3]:1, Feb. 2, 1998), he remarked: "What do you say to people who suggest that there's a difference between addiction and other sorts of brain disorders, like Alzheimer's disease or schizophrenia, because the latter conditions don't arise from a voluntary act of will? But lung cancer [caused by smoking] does occur from a voluntary act of will, and we still pay to treat people for it. The question is whether you want to fix it or not. Whether you think the person is evil and you hate them is not relevant. ... You need to deal with it as a health issue, even if you hate them while you're doing it."

Is Suffering Deserved?

Leshner says the issue of responsibility should be outweighed by our obligation to cure. But in reality many feel that the disabilities individuals bring on themselves are their just comeuppance. If an addict is to blame for his addiction, doesn't he deserve to suffer for it? Whether or not someone is "evil" and hated is indeed relevant to whether we feel motivated to help them or punish them. So, we can't duck the issue of moral responsibility when talking about addiction, and the disease model alone won't be enough to deflect the opprobrium many feel toward addicts.

But what accounts for "voluntary acts of will"? Most would say that there is something irreducibly personal driving the initial choice to use drugs, something that derives neither from an individual's environment nor his biological endowment. According to what might be called the "standard model" of free will, we all possess the capacity to act, in some important respect, independently of internal and external conditioning factors. It's that capacity that justifies praise and blame, reward and punishment. If the situation were replayed with all factors the same, the addict could have made a different choice, so he bears an inescapable, originative responsibility for his condition.

The problem, of course, is that the standard model is false. Although it is not yet widely appreciated, this sort of free will seems increasingly implausible given the rapidly growing scientific understanding of human biology and behavior. Acts may be voluntary in the sense that they are not compelled (most nicotine addicts don't start smoking with a gun to their heads), but they are nonetheless fully caused, a function of personality, motives, predispositions, and situations, none of which springs full blown from an uncaused agent within the person. Even if we presently lack the skill to track its causal antecedents, the will--the net vector sum of desire--is just as determined as anything else in nature, says science. And you won't find blameworthy free will in the residuum of what's undetermined, since, after all, we are usually only blamed and praised for what we intend to do. Intentional behavior of the sort we can be held responsible for certainly doesn't include random acts that surprise even the actor.

If the standard model of free will is false, it seems as if the addict is completely off the hook and morally blameless; given the governing conditions, things couldn't have turned out otherwise. Indeed, it seems that by challenging the assumption of free will, science threatens the very foundations of moral judgment. In his recent book How the Mind Works (New York, W.W. Norton, 1997, p. 56), cognitive neuroscientist Stephen Pinker confronts this dilemma and reaches a rather startling conclusion, especially for a scientist: We must idealize ourselves as uncaused creatures in order to have morality. In our ethical deliberations we must pretend, contrary to science, that human behavior is in some crucial respect independent of circumstances. For this to work, we must see that, "Science and morality are separate spheres of reasoning. Only by recognizing them as separate can we have them both. ... A human being is simultaneously a machine and a sentient free agent, depending on the purposes of the discussion. ... The mechanistic stance allows us to understand what makes us tick and how we fit into the physical universe. When those discussions wind down for the day, we go back to talking about each other as free and dignified human beings."

Science and Morality Linked

Since Pinker thinks that the "machine" view of ourselves undermines moral agency, naturally he is forced to keep science and morality in separate spheres. But is this dramatic and artificial disconnection of our explanatory powers and our ethical intuitions really necessary? The example of addiction will show that it is not.

First, it's clear that knowing that the voluntary choice to start using drugs is not a matter of uncaused free will doesn't invalidate our judgment that such a choice is bad. After all, addiction destroys lives, rendering addicts unproductive and unhappy. We have excellent reasons for expressing disapproval of drug abuse, since such disapproval can help dissuade those who aren't in the grip of nicotine or alcohol or heroin from taking that first, second, or nth hit. Precisely because the decision to drink or smoke or shoot up is, at first, voluntary--not internally compelled by addiction--stigmatizing that decision can reinforce the resolve to abstain or cut back.

Second, seeing that the traditional notion of free will is false may have an effect on just how much stigma we want to dispense, as well as its timing. Knowing that voluntary choices are just as determined as knee jerks undercuts the moralistic self-righteousness of those who suppose that they would never have succumbed to the temptations of crack or speed: there but for the contingency of life go I. Our anger at moral failure might be tempered, and seeing the causal necessity behind choices to use drugs might lead us to de-emphasize punitive measures with little or no utility (e.g., cutting welfare and health benefits for addicts) and increase our commitment to prevention, treatment, and outreach programs that actually reduce addiction and its harms. At the very least, seeing through the fiction of free will should lead us to re-evaluate the effectiveness of punishment, since it isn't deserved in the traditional sense.

We all know the power of stigma; just ask any youngster if he or she wants to grow up to be a junkie or a crackhead. But, as investigations of what works in treatment have shown, direct shaming of addicts is usually counterproductive: they need to learn that, despite the mistakes they have made, they are worthy of rehabilitation. What brings them and keeps them in treatment is knowing that someone cares, that they possess some kernel of dignity that merits praise and support, not continued condemnation. Stigma may play a useful corrective role in the early stages of drug abuse, but not after addiction has taken hold.

It's a Matter of Degrees

Although a thorough understanding of what causes addiction may lead us to reconsider the utility of stigma, it doesn't change our basic moral stance toward addicts and potential addicts: your choices were bad; do better next time or suffer the consequences. But such an understanding can significantly fine-tune the practical ethics of blame. To repeat: the vehemence of stigmatization will be kept in check by understanding that the choice to use drugs isn't a matter of uncaused free will, and its timing will be informed by how far along in the addictive process the individual has progressed. Such are the virtues of not compartmentalizing science and morality.

The growing scientific consensus that all human behavior lies within the natural realm has implications far beyond the study of addiction, of course. Whenever individuals are given all the blame or take all the credit for who they are and what they do, we can defeat such claims by pointing out the causal antecedents of their faults and virtues. Responsibility for good and bad doesn't inhere within some special, uncaused internal agent, but is distributed over the influences that create character and motive. Nevertheless, we must still praise or condemn an individual's behavior, because that, in part, is how people are shaped. Even though we can't any longer suppose that rewards and sanctions are deserved, in the pre-scientific sense of that word, they must still be applied, judiciously and compassionately. People are moral beings that should be held accountable--that is, be subject to rewards and sanctions--just to the extent that they have the capacity to anticipate and be influenced by them.

This model of morality ensures that, in the event a naturalistic, scientific self- understanding takes hold, excuse-mongering will not flourish. It will be generally conceded that we are all, in all respects, products of environment and heredity, but for that very reason social sanctions (and some hurt feelings, unfortunately) will be deemed essential to bring out the best in us, and inhibit the worst. On the other hand, it is likely that this same understanding will moderate the excesses of our all-too-punitive culture, transmuting our zeal for punishment into a concern for changing the conditions that produce destructive behavior in the first place. In a culture more fully informed by science, it will be a good deal more difficult to become an addict, and a good deal easier to find help, not rejection, should the unfortunate choice to abuse drugs arise.


The following are comments on "The Science of Stigma", an opinion piece published in the August 17, 1998 edition of The Scientist under the title "To help addicts, look beyond the fiction of free will".  Many thanks to these contributors for their input, received in August and September, 1998.

Free will: supposing one is exempt from statistical probabilities

From Don Zimmerman, British Columbia:

....In the past, I have discussed the relation between a deterministic point of view and treatment for addiction with my wife Doreen, who worked in the addiction field back in Ontario in the 1980’s and early 1990’s.

Here is a notion that might possibly be worth exploring. It occurs to me that acceptance of a naturalistic orientation is pertinent not only to the point of view of treatment providers, but also to the point of view of the addict. I mean: Many people who experiment with drugs and end up becoming addicted begin with a non-scientific, free-will orientation throughout their lives and reason, "I am a free agent. I have choice. I will never become addicted. I can stop whenever I choose," and so on. At the same time, they have a perception of their own uniqueness—that is, they tell themselves something like "I am special. I deserve more than my share of good things and have always had my way. I am not weak-willed like some people. I am immune from all these statistics about drug addiction. It will never happen to me," and so on. And then they are surprised to find themselves unable to escape from the scientifically lawful progression of events leading inevitably to addiction.

A scientific education and adoption of a naturalistic point of view might help people realize that freedom and uniqueness are illusory and do not confer exemption from statistical probabilities. Then, they would be more inclined to say to themselves, "If I experiment with drugs, I, like everyone else, will be governed by scientific principles that make it highly probable that things will keep getting worse," and so on. Having this realization, maybe the individual would be less likely to take the "first step" toward addiction.

So much for my rambling. Maybe this has all been said before. But I emphatically agree that a naturalistic point of view is beneficial in regard to this (and many other) social problems.

Intervene before drugs cause structural changes in brain

From David Rowden, Caliber Associates:

I have recently drafted an essay I intend to submit for publication that may fit well with your thinking. The focus is on the relationship of prevention and treatment within the disease model where recreational use (read acute phase of use) is addressable with prevention and treatment, but that once the disease process attains chronicity then a chronic disease model takes over and treatment/maintenance becomes the best model management. This thinking is based in part on the research reported in the September 1997 issue of Science, and on the Institute of Medicine publication on Dispelling the the Myths of Addiction. The underlying premise in the research reported in these sources is that continued use leads to structural changes in the brain, and that once these changes occur a chronic disease state exists. The thesis of my paper is that prevention efforts and early onset treatment effort are where the emphasis should be place with regard to resources. The logic (following your premise) is that the only effective time to intervene is when free will is still operating, i.e., before a chronic condition is established.

Philosophers have actually made progress on the free will question (!?)

From Steve George, Amherst College:

I just wanted to comment on one aspect of your views on free will. In The Scientist piece, you laid out what you called the "standard model" of free will, which allows for uncaused human actions. I wonder whether you are aware of the long history of thoughtful contributions to this subject by philosophers. I notice that you referred in the article and in your web site only to very recent writings on this subject, as if the discussion was entirely a modern one. However, here’s a bit of what David Hume wrote in 1746:

"Thus, for instance, in the human body, ...when irregular events follow from any particular cause, the philosopher and physician are not surprised at the matter, nor are ever tempted to deny, in general, the necessity and uniformity of those principles by which the animal economy is conducted. They know that a human body is a mighty complicated machine: That many secret powers lurk in it, which are altogether beyond our comprehension: That to us it must often appear very uncertain in its operations: And that therefore the irregular events...can be no proof that the laws of nature are not observed with the greatest regularity in its internal operations and government. The philosopher, if he be consistent, must apply the same reasoning to the actions and volitions of intelligent agents..."

I think scientists sometimes assume that our own discipline is characterized by steady progress, in contrast to philosophy which goes in circles, never making definite progress. However, philosophy has succeeded in organizing puzzlement once and for all about many things, one of which is free will. ("Organizing puzzlement" doesn’t mean giving an absolutely final answer, but scientific answers are not absolutely final either.) Probably not a single contemporary philosopher accepts the "standard model" as described in your article, and most would not acknowledge any contradiction or opposition between free will and determinism. Free will indeed requires determinism, because the absence of determinism is randomness, which is not compatible with free will.

TC replied:

I think you’re correct to say that most modern philosophers don’t defend what I call the standard model of free will, and correct again, therefore, that progress has indeed been made on this question. But outside of academia, I think many ordinary folks still subscribe to this model, which is why I take such pains to correct it. There probably isn’t much new in what I’m saying, but it still needs to be said. 

Thanks for the quote from Hume. If I had to pick just one favorite philosopher, it would be him. It’s also nice to hear something complimentary said about philosophy by a scientist, i.e., that it occasionally makes progress!

Free will is first of all a neurobiological problem

From Werner T. Schlapfer:

As a former neurobiologist (now a research administrator) I have known for a long time that Free Will must be an illusion. But it almost appears that there is a conspiracy of silence among scientists about this most important topic. While many philosophers seem to be concerned with Free Will (and most of there writings are totally useless, in my opinion), very few neurobiologists have approached the subject. The problem of Free Will is NOT A PHILOSOPHICAL PROBLEM, but first of all a NEUROBIOLOGICAL one. After all, the decision process happens in our brain and I know of no neuronal or nerve net mechanism that would be able to generate something akin to free will. The concept does not even make sense. I have been looking for books or articles by biologists (more specifically neurobiologists) that discuss this issue, but have only found a few (J.-P. Changeux, "Neuronal Man"; F. Crick: "The Astonishing Hypothesis"; R. Wright: "The Moral Animal). If you know of any recent biological treatment of free will, I would appreciate a e-mail note.

TC replied:

Have you read Benjamin Libet’s work on measuring cortical activity during voluntary action? He showed that neural impulses controlling movement begin 500 milliseconds before the conscious intention to act arises. However, he goes on to argue that free will still exists in the ability of the subject to "veto" the actual movement. But of course that begs the question of the causal antecedents of the desire to veto. Libet wrote an article in The Sciences several years back describing this research (to which I wrote a highly critical letter in reply), and you might be able to dig it up on line or in a library. I don’t know of any more recent work on this, although the Journal of Consciousness Studies will be having an issue on volition coming out soon (see their web site) and I bet they’ll have some neuroscience contribution in it.

The pragmatic utility of blame

From Reed Richter, Chapel Hill, NC

Naturally, I have my own take on the free will issue (as does every professional philosopher). I think our ordinary common sense notion of "free will" is vague and probably inconsistent with causal determinism (CD) as you claim. But as you recognize, what is important is not whether we have "free will," rather it is whether or not we can make sense of "moral responsibility and choice" given that the world is for all practical purposes determined. You correctly point out that even if CD is true, praise and blame (rewards and sanctions) are appropriate "just to the extent that [an agent has] the capacity to anticipate and be influenced by them." Our system of morality has a pragmatic utility in shaping behavior that we value.

Thus far I agree, but I’m not sure that it establishes the conclusions that you want: viz., … We can no longer justify the claim that rewards and sanctions are "deserved" (as opposed to being pragmatically useful or justified.) and … There is no good reason to stigmatize an addict’s behavior after the addiction has taken hold.

There are a couple of issues you need to address in order to satisfy my doubts.

1. You don’t address the issue that there is a big difference between addiction and (say) Alzheimer’s. Addicts can and often do take responsibility for their condition and get clean. The Alzheimer’s patient literally have no control over their condition. I can understand that from the point of view of the rehabilitation therapist, HIS continuing to stigmatize the addict is counterproductive. But that doesn’t mean that there is no pragmatic value for the rest of us to continue to stigmatize. For example, perhaps the recognition that society as a whole continues to express moral disapproval of the addict is a motivating background factor in therapy. You obviously are in a better position to address this issue than I.

2. What is your position on the behavior of a serial killer like Ted Bundy? Can we not justifiably be disgusted with his behavior both before and after he became a committed killer? Can’t this disgust play a useful role in discouraging other potential killers, and also in shaping society’s response to serial killing in general. Maybe societal stigma is the main force motivating society to deal with addiction. Even legalization may be seen as a result of stigmatization if it is seen a means to lower addiction rates and minimize its effects.

3. It may be pragmatically useful to pursue a strategy of "deserved" blame and punishment. We know from game theory that in many common situations (e.g., iterated prisoner dilemmas) a tit for tat strategy leads to the most beneficial outcome. In past contexts it is plausible to suppose that it was a useful (Darwinian) strategy for a family or individual to seek revenge and strike back if injured. The family (or individual) reputation would then be a useful deterrent. If this strategy has survival value it is reasonable to suppose that humans who have a proclivity to value revenge for its own sake will be selected for. Hard-wiring this preference or value is a lot more efficient than relying on some reasoned calculation of the consequences; i.e. human groups that were thus hard-wired in general would do better. I am not being original here. This issue is much discussed cf. Passions within Reason. Obviously there are circumstances where this proclivity for revenge leads to annihilation ; e.g. the Mideast, Ireland, the Hatfields and McCoys, etc.. In such cases survival depends on reason overruling passion. Nevertheless, even in our society pursuing revenge may be useful to some significant degree, e.g. don’t mess with litigious people. The law becomes another means of retaliation. The point is that desiring punishment as a means of revenge may have some substantial pragmatic utility. Thus it is not in the least bit clear that causal determinism undermines the reasonableness of viewing justice as deserved punishment.

TC replied:

Re 1:  Ongoing social disapproval of addiction is obviously going to continue, and should. The question is, how is disapproval of addiction expressed in social policy? If it’s by the punitive withdrawing of the supports necessary for recovery, then it’s counterproductive. If it’s expressed by attacking the root, structural causes of addiction (e.g., employment, education), as well as in public health warnings about cigarettes, booze, and drugs, then it’s productive. Shunning and otherwise punishing addicts and other deviants comes quite naturally to all of us, so there’s little danger that it will soon disappear as an inducement to stay clean. The problem, though, is that it often doesn’t work particularly well, given the strong incentives arising out of impoverished living conditions (both economic and cultural) to use substances.

Re 2:  We can’t help but be disgusted with serial killers, both before and after they kill, and that motivates us to deter and restrain, but should also motivate us to create conditions which reduce the likelihood of serial killers being produced again. But we don’t focus on the last option since many suppose people can simply rise above their circumstances, and thus don't need our help.

Re 3:  I don’t doubt that the retributive impulse has been hard wired into us, for "good" evolutionary reasons, so the notion that someone deserves to suffer for having themselves incurred suffering has had pragmatic value in ensuring survival. But getting rid of the metaphysical confusion about the causally disconnected self that has originative free will might help mitigate the excesses of tit-for-tat revenge so characteristic of human conflict. A naturalized notion of desert, as part of a naturalized understanding of morality, might help to keep punitive sanctions at a merely useful, not horrific, level, and encourage attention to those factors that give rise to aggression in the first place. Once we have good alternatives to punishment, then the pragmatic value of desiring retribution will diminish, and will count less as a justification for imposing it.

Other Categories: