• Autonomy
• Decision-making
• Ethics
• Paternalism
• Public Health Ethics

I am grateful to the Journal of Medical Ethics for asking these critics to discuss my book, and am grateful to each of the critics themselves for raising interesting and often difficult issues for me to think about.

Alan Wertheimer makes a number of good points. One of the most significant, to me, is how paternalism might function at what I will call an institutional level. In my book, I endorse paternalistic actions by the state, when the cost benefit analysis justifies that. I have not supported paternalistic interventions by private individuals, though. For one thing, private individuals will make decisions without public input, and without their justification being examined by experts in the field, and for these reasons they are too likely to make mistakes as to when intervention is appropriate. For another, the interventions by random individuals haranguing us about our failure to eat our broccoli is likely to drive us crazy—unexpected interference from people with no particular authority will give us just that sense of harassment I say paternalists need to avoid.

However, between the public, authorised actions of a democratic state and the private actions of individuals with no authority lies the significant area of non-state institutions. Such institutions develop regulations and standards their members are expected to act in accordance with, and these in turn can have a significant impact on private persons. Of course, there have also been legal actions, notably court decisions, that have played a role here, but much of the specific interpretation of things like consent has been determined by the medical community developing what it believes to be an appropriate ethic for itself. And we know, here, that in recent years there has been a movement away from the paternalism that was once common, and a reorientation towards respect for patient autonomy. Do the empirical findings of behavioural economics suggest that the move towards patient autonomy has been mistaken?

Perhaps. As will be the case with a number of these good questions, I find myself torn. Institutional rules, unlike personal decisions, are subject to the kind of vetting we want: to examination by relatively disinterested parties, who are experts in the field. If standards are revised through conscious effort, with adequate input, that is a far cry from your bossy sister-in-law smacking your hand when you reach for a second piece of pie. At the same time, though, physicians or, in Wertheimer's example, members of an Institutional Review Board, are not democratically elected, and not subject to recall if their regulation is, in terms of those who suffer from it, a failure. We don't have the control over them that we would like to have over those who regulate us.

So, there are pros and cons to allowing paternalistic standards by medicine as an institution. On the whole, I am inclined to think it can be acceptable in certain areas. As with paternalism generally, I think institutions should aim for transparency: that these are the standards of the institution should be something individuals can know easily. This doesn't allow for the same responsiveness to public input as do elections, but it does allow for some, and obviously allows the potential patients, participants in research and so on some sense of control. This is a really important question that deserves a great deal more discussion, but so far my thoughts are that in some aspects of medicine, more paternalism than is now accepted would indeed be justified.

I do, however, think this should be done only to further the patient's own goals, not to substitute more ‘rational’ goals for those the subject of the rules endorses. While I understand Wertheimer's contention (obviously, one he shares with others) that some goals are more rational than others, it's not something I've ever felt competent to judge. If the means you choose are manifestly at odds with the ends you want to achieve, I have a standard by which I can criticise your choice: it simply won't get you where you want to go. When it comes to ends, I have no such standard. Of course, certain ends seem extremely peculiar to me, and in a non-philosophical mode I'd be perfectly willing to say it's crazy to e.g. climb Mt. Everest—so tiring! so cold! so slow!—whereas it's perfectly reasonable to spend an afternoon watching videos of cute puppies, but philosophically I know this makes no sense—I can't even pretend there is some objective basis for my preferences that theirs lack.

Mr Pugh thinks that I nonetheless may be suggesting we should prioritise some ends over others, so that my account does allow for a kind of back-door perfectionism. Thus, he suggests that I require that we prioritise health over the ends met by the unhealthy consumption of ‘whiskey, cigarettes and fried food’, whose protection Joel Feinberg worried about.

This isn't true, though. First, for those who know me and who will be astonished at such a suggestion from someone whose life includes considerable consumption of Lagavulin, nowhere in the book do I suggest that government intervention should extend to general alcohol use, beyond the concern for third party harm we show in preventing intoxication in the exercise of dangerous activities like driving. (Mr Pugh doesn't say that I do this, but his emphasis on the Feinberg quote might suggest it.) I don't recommend greater control over alcohol precisely because many people greatly value the experience of alcohol, and seem to care more about it than the (real, even if usually not considerable) bad effect it can have on your health. Nor, for that matter, do I say we should outlaw fried food, because people enjoy it enough, as far as I can tell, to outweigh some costs to health. What I advocate as to junk food is portion control, precisely because there is a lot of evidence that the amount of food we eat is not based on desire, but on availability. (If we are given the big portion we will consume it, but we are equally happy with a small portion.) The same is true of cigarettes: my argument is not that we should be healthy even if we'd prefer to smoke, but that there is a lot of evidence (including the huge amount we spend trying to quit smoking) that people would prefer not to smoke, because they would rather be healthy. If this were proven wrong, then I would withdraw my opposition to smoking.

Nir Eyal also raises difficult questions. He says, correctly, that it is not a cognitive bias if one simply values at one time something that a later time one won't value, or won't value as much. This is clearly true, and it gives rise to difficult philosophical problems. Any picture of human good that rests on the satisfaction of desires has to deal with the fact that desires change. Certainly, the ‘satisfaction’ of desires previously, but no longer, felt, can be useless—like giving me now the patent leather shoes I longed for as a 6-year-old. Whether it's just as senseless to take steps now to satisfy future desires that I will feel, but don't now feel, is more of a question. Most people seem to think attention to future desires not yet felt is more sensible than attention to past desire no longer felt, since in future I will get some benefit out of the satisfaction of those desires. How much the satisfaction of a desire I will feel only in future should constrain present action is certainly difficult to determine, though, and can involve us in controversial discussions of what constitutes a good life.

However, these thorny problems are not so problematic in the determination of paternalistic policies as Eyal seems to think. Say that I feel irritation at having to put on my seatbelt. Does this mean that I am being forced to ignore my present desire for comfort in favour of a future desire not to fly through the windshield, that, at present, I'm indifferent to? It doesn't seem likely. It's reasonable to think that even at the time I'm irritated at the sensation of the seatbelt I have an even stronger desire to reach home safely. I'm just not calculating the costs and benefits correctly.

So with fatty foods. As I've said, I haven't advocated doing away with them, precisely because people really enjoy them, apparently sufficiently to outweigh some costs to health. I also argued that people's (present) love for soda appears to outweigh the (future) costs to health (although, in light of present information about sugar, I may have been wrong in reaching that conclusion). But, there is strong evidence that even when they are eating it, people don't care much about having a large portion rather than a smaller one (and if they do, they can go to the counter and get more—but as with soda laws, companies oppose portion size laws precisely because they foresee that patrons don't want more fries enough to get up and get more). So, even granting that people care less about the satisfaction of future desires than present desires, it seems unlikely that they now desire huge portions of fries more than they now desire future health. (One may note, too, that there are present costs to obesity, both in terms of activity and conventional attractiveness, things people do seem to care about.) Similarly with cigarettes, as discussed above: we know that the majority of cigarette smokers try to quit, which suggests they don't find the present rewards to outweigh their present desires for a healthy future.

This does leave difficult questions. The issue of the Jehovah's Witness who asks for medical help after years of opposing it is such a one: the patient seems to be acting in a way that is inconsistent with his overall values, and yet it seems perverse to deny him the medical help that can easily restore his health. In this case, I agree, such a denial is perverse. However, the situation is ambiguous—is the patient acting in a way inconsistent with his true values, or have his values simply changed in light of experience and reflection? He might, after all, decide that he cares more about living than about God's will or he might decide that his previous interpretation of God's will was mistaken. In any given moment it's hard to tell if a person's thinking is distorted by cognitive bias or weakness of the will, or if he has simply changed his values in light of experience or reflection. However, for purposes of public policy, I don't think we need to determine this, because in policy we want to err on the side of caution. If a patient chooses a non-risky procedure that brings about a healthy future life, including chances to rectify what may have been an irrational choice, we should allow that. As I've argued in the book, there is nothing about accepting the principles that justify paternalism that means we then must interfere in each and every circumstance in which someone acts irrationally. It will depend on the costs and benefits of the policy's application.

Since Gerald Dworkin agrees with my position that paternalism need be neither insulting nor disrespectful, I haven't a great deal to add to his account. As I say in Against Autonomy, paternalistic laws do not pick out particular groups as being less rational than others (say, motorcycle riders). It picks out some tendencies we all share, and that manifest themselves differently in different circumstances. The idea is not that some of us, who are superior, should boss around others, who are inferior. Rather, we should all help one another out. While this does amount to giving up certain pretensions about our abilities, it seems more of an insult to think we can only be worthy of respect if we claim attributes we don't really have. Surely, biases and all, we are valuable as we are!