So under what conditions, if any, is paternalism morally justifiable? In the Western philosophical tradition justified paternalistic intervention, has focused on detailed philosophical analysis of the interwoven concepts of harm and rationality. First of all, since John Stuart Mill, the Western liberal tradition has distinguished between “harm to self” and “harm to others” and that paternalism, by definition, must target “harm to self.” Justified governmental intervention in regard to “harm to others” involves different principles and arguments. As stated in my previous essay, most human actions and paternalistic interventions affect others. So, with all this in mind, for now let’s focus on “harm to self.”
Paternalistic intervention is, by definition, an act of beneficence, where an individual (or group) coercively violates the liberty of another individual (or group) in order to prevent and/or remove “harm.” Well, what is harm? I don’t think I can improve much on Joel Feinberg’s definition, as the “invasion of an interest.” What is an interest? Well, it’s “anything you have a stake in.” If you have a “stake” in something, it means that, ultimately, your pleasure or pain is at stake. We all have an interest in a lot of different things, but obviously, we value some interests more than others. Let’s call these individual and collective rankings “interest hierarchies.”
Are these interest hierarchies “objective” or are they “relative” to individuals, small groups, or large groups? One might argue that interest invading harms are objective to the extent that harm can be measured in terms of magnitude and probability. Magnitude is the sheer degree of harm: "Death has greater magnitude than embarassment." Probability is the sheer likelihood of that harm. Thus, some harms are of great magnitude (death), but improbable (while flying); others are lesser in magnitude (embarrassment) but highly probable (farting in public). Moreover, the magnitude and probability of harm can be assessed over both the short-run and the long-run. Hence, most of us agree that Russian roulette is a high-magnitude, high-probability harm over the short run; smoking is a high-magnitude, high probability harm over the long run. (For the sake of argument, let’s assume that our beliefs about the causal relationship between smoking and cancer are true. My only point here is that smoking and Russian roulette are different.)
But in the real world, magnitude cannot be measured in a vacuum. It is always relative to other harms and benefits, and therefore subject to trade-offs. Therefore, rationality and irrationality are about those "interest hierarchies," risk-taking and trade-offs. We take risks when we either “do things” and when we “don’t do things,” therefore paternalistic interventions can either force us to “do things,” or prevent us from “doing things.” J.S. Mill suggested that irrational persons take irrational risks; that is they do things (or don’t do things) where the potential (or actual) costs of those actions exceed the potential (or actual) benefits. In short, irrational risks are not worth taking relative to an interest hierarchy. Unfortunately, he didn’t provide much insight into what this might mean in the real world. So what kinds of risks are rational?
Bernard Gert argues there harm is more or less objective. He observes that, unless a person has a good reason to do otherwise, rational individuals tend to avoid: death, pain, disability, and loss of opportunity. So what’s a “good reason” to choose death? Well, most Americans living in the twenty-first century agree that it is not, necessarily, irrational to choose death over suffering high levels of intractable pain or disability. But what happens if suicide is sometimes a rational act, relative to a large group, but irrational relative to a small group or an individual? If a teenager wants to jump off the bridge because of social embarrassment (farting in math class), then that it seems objectively irrational. After all, over the long run, embarrassment is, at most, a low magnitude short-term harm. But note that our large group, small group, and individual judgments are based the assumption that it is irrational to kill one's self out of embarrassment. Objectively speaking, embarassment is a mow magnitude harm, therefore suicide, a major harm, can never be justified. But in many parts of the world, embarrassed fathers routinely kill their daughters in the name of “honor” (embarrassment) and many daughters commit suicide after dishonoring (embarrassing) their fathers. Although most American believe that honor-based suicides are irrational (and immoral), how does one go about establishing an objective, universal standard?
So here lies the basic paternalistic dilemma. If our individual interest hierarchies are at least influenced by small groups and large groups, what does personal liberty mean in this context? (I'll address this in the next essay.) And, when there’s a conflict between individual and collective interest hierarchies, which interest hierarchy prevails over the others? Whose risk-assessment trumps whose? Large-group communitarians tend to empower large group hierarchies (American's), and small-group communitarians tend to empower small group hierarchies (Roman Catholic's), and libertarians empower individual hierarchies (Ron White's). But the question for libertarians is whether paternalistic intervention can ever be morally justified? As stated at the outset, traditionally that question has involved the analysis of rationality and harm. But another approach to this question is to take a closer look at how human knowledge underlies specific paternalistic interventions.