1 CHAPTER 4, On Liberty. Does Mill Qualify the Liberty Principle to Death? Dick Arneson For PHILOSOPHY 166 FALL, 2006 In chapter 1, Mill proposes "one very simple principle, as entitled to govern absolutely the dealings of society with the individual in the way of compulsion and controL" As described in chapter 1, the proposed Liberty Principle (LP) looks to be substantive and controversial. Question: Do the later chapters 4 and 5 qualify the LP to the point that it becomes toothless-nonsubstantive and uncontroversial? To see the problem, consider the reading of Mill in the "Editor's Introduction" by Elizabeth Rapaport. She says that in chapter 1 Mill asserts Principle I. The only legitimate ground for social coercion is to prevent someone from doing harm to others. But in chapter 4, discussing objections, Mill denies Principle I and in its place asserts Principle II. The only legitimate ground for social coercion is to prevent someone from violating "a distinct and assignable obligation to any other person or persons. In other words, Principle II holds that the only legitimate ground for social coercion is to prevent someone from violating the rights of others. What gives? Why assert in chapter 1 what you are going to retract later? Rapaport says "Mill's procedure is a model of open philosophical inquiry." This does not seem to suffice to excuse Mill from the obligation to explain in chapter 1 that Principle I is only provisional and not his real view. Worse, Principle II does not look like a real principle, it looks like a blank check. One could know all the possibly relevant facts that bear on policy choice and still not know what are the legitimate grounds for coercion according to Mill. In this respect, the supposedly rejected Principle I is superior. Rapaport is untroubled. She writes, "Mill's principle is not designed to settle the question of what are our assignable obligations, what specifically are the rights that society is supposed to protect." But the problem is that as presented by Rapaport, Mill does not tell us how to go about deciding what obligations and rights people should be regarded as having. So far, acceptance of the Liberty principle seems to commit one to nothing. Mill does not write as though for all that he has said, it is an entirely open question, what rights and obligations people have. Discussing the Alliance for temperance reform and its proposals about social rights (pp. 86-88), Mill writes as though anyone who accepts the Liberty Principle is committed to rejecting the expansive understanding of social rights propounded by some temperance reformers and to rejecting banning the recreational drinking of alcoholic beverages as a violation of this principle. If Rapaport's interpretation were correct, it would be hard to understand how Mill could think himself entitled to argue in this way. Here is one possible way to put teeth back into the Liberty Principle, interpreted as Principle II. Add Mill's statement that he regards "utility as the ultimate appeal on all ethical questions" (p. 10), and interpret Mill as implicitly holding that society ought to recognize and enforce just those obligations and rights that are such that recognizing and enforcing them maximizes utility in the long run. Notice that on this way of interpreting Mill, the Liberty Principle itself does not protect broad individual liberty to live as one chooses. Everything turns on the question, recognizing and enforcing WHICH putative rights would be utility-maximizing? There is room for Peter Singer (an advocate of very demanding duties to aid distant needy strangers) to accept the LP but hold that it maximizes utility for society to recognize and enforce a strong duty to aid distant needy strangers of the shape that he proposes. The Liberty2 principle would then NOT rule out enforced charity, contrary to chapter 1 appearances. What liberty the LP protects would depend on whether utilitarianism in particular circumstances implies that an extensive or a minimal set of individual obligations and .corresponding rights should be recognized and enforced. Let's turn to Mill's own elaboration and defense of the LP in chapters 4 and 5. On p. 78 Mill raises the objection, "The distinction here pointed out between the part of a person's life which concerns only himself and that which concerns others, many persons will refuse to admit." The objection is that any action that is self-destructive is also at least indirectly harmful to nonconsenting others. Mill anticipates this objection in a comment on p. 11, lines 8-12 from the bottom of the page, chapter 1. The objection might be put this way. If we try to divide actions into two categories, I, those that affect the interests of persons other than the agent and those who voluntarily consent to be affected, and II, those that affect nobody's interests except those of the agent and others who voluntarily consent to be affected, it turns out that either all actions or all nontrivial actions fall into category I. If the Liberty Principle throws a shield of protection around category II actions, what had looked like a substantive, controversial, and bold principle will turn out to be nonsubstantive and uncontroversial. Only utterly trivial actions such as no one ever bothers to propose restricting, such as whether I will put my sock on my right foot first or second, will fall into the protected category II. Some conduct that is self-harming does not directly harm others, but indirectly, it does harm others. The objector says that since all self-harming action always does or may harm others eventually or indirectly, there will always be legitimate grounds for restricting liberty (namely, harm to others reasons) whenever there are significant harm to self reasons. Notice that we don't want to say that if your action directly causes harm to others (causes harm via a short simple causal process) it may legitimately be restricted but that if your action indirectly causes harm to others (causes harm via a long complex causal process) it may not legitimately be restricted but is protected from restriction by the LP. Whether your act causes harm by a simple or by a complex causal process is surely not per se relevant to whether it should be allowed or not. Mill embarks on an interesting discussion on pp. 78-80. Its conclusion is stated on p. 80: "But with regard to the merely contingent or, as it may be called,