This is a quick afterthought on the piece I posted the other day about religiously-based exemptions, and also a response to some of the comments made on that piece. My central objective in that argument was simply to point out that many rules in our society already have exemptions built into them, in order to handle circumstances in which there is significant heterogeneity in the underlying interests. In some cases, just averaging them all together, or taking the vector sum, is not the best way of responding. This is particularly true in cases where a small minority has an interest (or assigns weight to an interest) quite divergent from that of the majority. In that case, it may make more sense to have a rule that applies in most cases, but then to grant an exemption for those in the minority. The example I gave was of speed limits, which have selective exemptions for emergency vehicles.

In the case of Sikh turbans, however, there is an additional wrinkle to the argument, which I was intentionally glossing over, but which came up in the comments (here, and a lot more on reddit). Rules that require hardhats on construction sites, or helmets while riding motorcycles, also happen to be paternalistic. What this means, roughly, is that if the person being coerced by the law (i.e. forced to wear a helmet) where to ask why, the justification offered would make reference that that person’s own interests. Paternalistic laws, in other words, are ones that are “for your own good.” Now there is a long-standing tradition in liberal societies of skepticism about such laws. After all, who is the best judge of where an individual’s interests lie, the actual individual concerned, or the government? These concerns were articulated canonically by John Stuart Mill in his On Liberty, and they have generated an enormous presumption in our society against the legitimacy of paternalistic laws. This is true even in Canada, where the libertarian streak is not nearly as strong as in the United States – even if we do not categorically prohibit such laws, we tend to view them as rather dubious, and seek to limit their spread.

So how does this affect the debate over turbans, with respect to, say, mandatory motorcycle helmet laws? Generally speaking, the fact that such laws are paternalistic tends to bolster the case for an exemption for Sikhs. Why? Because when pressed to justify the law, the government must basically say to motorcycle riders, “You don’t know what’s best for you. We, the government, have a better understanding of where your interest lie than you do, as a result of which we are going to force you to do something that you don’t want to do.” This is semi-plausible when the interests in question seem to be capricious, or part of the folly of youth – such as wanting to “look cool” on a motorcycle. In such cases, we might be disposed toward sympathy with the paternalistic legislation. But when the interests in question arise from a deeply held religious conviction, central to the individual’s personal identity, the government’s claim that its “knows best” seems more likely to be specious. This speaks heavily in favour of an exemption.

As a result, people who are opposed to Sikh turban exemptions face a double hurdle. Not only do they have to argue against the granting of exemptions from general rules, they must do so even when the general rules in question are paternalistic. Because this is a difficult case to make, a very common “move” made by opponents is to deny that motorcycle helmet laws are paternalistic. And the way that many people do this is by making an argument that Mill complained about a great deal in On Liberty. The best way to show that a law is not paternalistic is to show that the behaviour being prohibited is not just bad for the person doing it, but that it also creates a harm for others. (This is why so much anti-tobacco legislation appeals to the effects of second-hand smoke, rather than the ill effects on the smoker.) In cases where the harm to others is not apparent – e.g. motorcycle rider runs into pole, goes flying, hits head, dies – then it is always possible to invent one, by appealing to a downstream consequence. To pick an example that Mill discusses, an alcoholic, by spending all his money on booze, might fail to support his wife and children, and thus his behaviour harms others. And in cases where specific harms like this cannot be found, it is always possible to come up with the all-purpose harms, such as “setting a bad example for others” or “eroding the moral fabric of society.” So the alcoholic, by staggering about inebriated in the middle of the afternoon, might be seen by young children with impressionable minds, and thus diminish their resolve to remain upstanding members of society.

The problem with these “harms,” in Mill’s view, is not that they are unreal, or that they cannot be regarded, in some sense, as harms. The problem, he argues, is that if you understand “harm” so expansively, then any time anyone does anything wrong at all, it automatically produces a harm for others. As a result, the distinction between law and morality, or even law and etiquette, collapses, and every obligation that anyone could ever be under becomes legally enforceable. No one could want to live in such a state (Mill claims). I refer to this as Mill’s “death of liberty” reductio. It goes roughly like this: Suppose that x is categorized as a “harm,” such that if an action produces an effect of type x, it is subject to state prohibition. If this categorization results, in principle, in the complete eradication of the sphere of individual liberty, then one should not categorize x as a “harm.” (A milder version of the principle would substitute “intolerable interferences” for “complete eradication.)

This issue came to a head in the 1950s and 60s, in debates over the legal prohibition of homosexual conduct. The general current of legal opinion ran against prohibition, on the grounds that the regulation of consensual activity among adults is paternalistic. Thus defenders of prohibition sought to find some “harm” that homosexuals were inflicting upon society. What they wound up pointing to were rather diffuse harms, such as “eroding the moral fabric of society” or “weakening the family.” This turned out to be a losing argument, since a concept of harm so expansive could be used to regulate practically anything.

There is, however, a more up-to-date version of the “moral fabric” argument, which is particularly popular in Canada. It is “costs to the health-care system.” Whenever a person engages in self-destructive behaviour, at some point it is almost guaranteed to produce a doctor’s bill. And since the individual does not pay for his or her own health care, this “cost to society” constitutes a harm to the taxpayer. People trot this argument out with depressing regularity, in order to justify all sorts of paternalistic laws (such as tobacco regulation, mandatory seat-belts, and yes, motorcycle helmets). So we saw another round of that with respect to the turban exemption: they have no right to ride around without helmets because, if they have an accident, it’s the rest of us who wind up footing the bill.

Now it is worth pointing out that this “costs to the health-care system” is often put forward in bad faith. The first clue is that the person making the argument almost never stops to think whether the behaviour in question is actually producing costs to the health-care system. For example, smoking is often cited as a form of behaviour that produces costs to the health-care system, when in fact the opposite is true. To the extent that people making this argument put any effort into it at all, they usually just add up the costs of smoking-related illnesses and call that the cost. The mistake they make is ignoring all the benefits that smokers produce for the health-care system, first and foremost, by dying young, and by dying suddenly (heart attacks) or from untreatable conditions (lung cancer). If you actually do a proper cost-benefit analysis, it turns out that smoking saves everyone money – in other words, it produces “benefits to the health care system,” because of the premature deaths it causes.

As far as motorcycle accidents are concerned, anyone who stopped to think about it for more than a second would realize that people who ride without a helmet are undoubtedly producing benefits as well. Cranial injuries sustained in motorcycle accidents are almost always lethal. In fact, within the medical culture of transplant surgery, motorcycles are commonly referred to as “donor-mobiles.” Indeed, motorcycle accidents and gunshot wounds are the two major sources of high-quality transplant organs – i.e. young men dead from massive head trauma, but perfectly intact and healthy bodies. And this is with exceptionless motorcycle helmet laws. Not to belabour the obvious, but killing yourself in a very sudden, untreatable fashion saves the health care system a great deal of money. (The expensive way to die is to malinger for a decade or more with diabetes and high blood pressure, before finally succumbing to cancer.) Thus it seems rather likely that Sikh motorcycle riders going about without helmets would save the health-care system some money, and might even produce more general benefits by expanding the pool of organs for transplant.

In any case, none of this matters all that much, because even if it were the case that motorcycle helmet law exemptions for Sikhs created benefits, rather than costs, treating “costs to the health care system” as a harm is a classic “death of liberty” move. Think of all the things that the government could regulate, if this argument were to be taken seriously! Obviously children should not be playing hockey, nor should adults be skiing. Or how about obesity, and the fact that individuals are allowed to choose their own diet? Obviously there should be state regulation of what food we eat. Oh, and getting pregnant obviously creates costs to the health care system, so the state should be able to regulate that… in fact, since having sex usually creates a risk of pregnancy, that should be regulated too… and so on.

Come to think of it, who could have imagined that the creation of a public health care system would require that individuals surrender all of their freedom to the state? If it is true, that the state is entitled to legally prohibit any behaviour that generates “costs to the health care system,” because this harms the taxpayer, then that is in fact a powerful argument against having a public health care system.

The upshot is, I hope, pretty clear. Please, please, stop claiming that paternalistic laws can be justified through appeal to health care costs. I’m very far from being a libertarian, but even I balk at the thought that the busybodies in the public health department could be legally entitled to make me eat bean sprouts and tempeh instead of hamburger – because lord knows, if we let them, they would.

P.S. One other thing that I’ve noticed, tangentially related to this. People in the exciting new academic discipline of “health studies” appear to have never heard any of these arguments, and indeed, to have no idea whatsoever that there is any sort of constraint or taboo on the enactment of paternalistic laws. These people seem willing to regulate the temperature of your bath water, if they could. Someone needs to put “paternalism” on the curriculum of every introductory health studies course.