1. See generally Anthony Boardman, David Greenberg, Aidan Vining & David Weimer, Cost-Benefit Analysis: Concepts and Practice (1996) (discussing normative economic concepts).

2. See generally Shelly Kagan, Normative Ethics 70–78 (1998) (discussing deontological constraints and options).

3. Eyal Zamir & Barak Medina, Law, Economics, and Morality (2010).

4. See Alexander, Larry, Deontology at the Threshold, 37 San Diego L. Rev. 893 (2000)Google Scholar (discussing challenges facing theories of deontological obligations that can be trumped by good enough consequences).

5. See Zamir & Medina, supra note 3, at 127–347.

6. See id. at 79–104.

7. Indeed, it does seem likely that linearity is a severe straightjacket on intuitions. We might feel that we should kill ten innocents only when this would save the lives of ten thousand. But the ratio of 1000:1 might seem quite inappropriate if this required us to refrain from using an atomic bomb that would kill 10 million innocents but that would thereby save the entire human race of 7 billion people. The issues seem morally separate and not susceptible to any linear solution.

8. The concepts of incommensurability and incomparability (distinct concepts but both required for adding) are addressed in Ruth Chang, Incommensurability, Incomparability, and Practical Reason (1997).

9. See Charles Fried, Right and Wrong 10 (1978) (arguing that wrongs such as promise violations might be justifiable by avoiding catastrophic consequences but that these should be seen as extreme situations in which normal categories of right and wrong do not apply).

10. See Zamir & Medina, supra note 3, at 86 (lexical priority of lives over pecuniary values).

11. Of course, in some way this function would itself face an issue of comparability. Perhaps the function would involve maximizing total as opposed to average utility, since the total would be affected both by the number of lives and by the average happiness of whatever lives there were. Total utility is in some ways not very attractive, however, and there might be better ways of converting these two consequential values into a common denominator.

12. Zamir & Medina, supra note 3, at 87–91.

13. Id. at 87. The issue is discussed in Samantha Brennan, Thresholds for Rights, 33 S. J. Phil. 143, 160 (1995).

14. See Andreu Mas-Colell, Michael D. Whinston & Jerry R. Green, Microeconomic Theory 46–47 (1995).

15. See John Broome, Weighing Goods 66 (1991) (“By ‘the utility of A’ I mean the number assigned to A by a function that represents an ordering, and nothing else.”).

16. A proof of this proposition appears in Mas-Colell, Whinston & Green, supra note 14, at 46.

17. The idea that normative orderings of goods can be treated on an analogy to descriptive orderings of preferences is stated in Broome, supra note 15, at 165 (1991) (principle of personal good instead of Pareto-based theory of preferences).

18. Zamir & Medina, supra note 3, at 57–78.

19. Sunstein, Cass R. & Vermeule, Adrian, Is Capital Punishment Morally Required? Acts, Omissions, and Life-Life Tradeoffs, 58 Stan. L. Rev. 703 (2005)Google Scholar.

20. See Zamir & Medina, supra note 3, at 60–61 (“It is logically possible and normatively plausible that governments are morally required to actively promote the well-being of all of their people and at the same time are subject to the moral duty to refrain from actively harming them.”) (emphasis in original).

21. See Christopher Kutz, Torture, Necessity, and Existential Politics, 95 Cal. L. Rev. 235, 256 (2007) (moderate deontology at governmental level creates risk of overuse of torture).

22. See Milton Friedman, Capitalism and Freedom 192–194 (1962) (defending the concept of the top 90% imposing a tax on themselves for the benefit of the bottom 10%). This seems to imply that the rich are doing something that no one else has the right to make them do, which may be true for some categories of relief and not for others.

23. Poor relief that is optional is somewhat analogous to contributions for public goods. A person may have a moral obligation to contribute his fair share toward a public good such as national defense, which is hard to enforce against individuals in the absence of a coordinating state apparatus.

24. The distinction between retaliation and preemptive self-defense is stressed in Alexander, Larry, A Unified Excuse of Preemptive Self-Protection, 74 Notre Dame L. Rev. 1475, 1476 (1999)Google Scholar.

25. See Sunstein and Vermeule, supra note 19, at 721 (“When Homeowner B sues Factory A over air pollution, a decision not to rule for Homeowner B is not a form of inaction: it is the allocation to Factory A of a property right to pollute. In such cases, an apparent government omission is an action simply because it is an allocation of legal rights.”).

26. See Robert Nozick, Anarchy, State, and Utopia 6 (1974) (“What persons may and may not do to one another limits what they may do through the apparatus of a state, or do to establish such an apparatus.”).

27. Louis Kaplow & Steven Shavell, Fairness Versus Welfare 52–58 (2002).

28. Zamir & Medina, supra note 3, at 51 (emphasis in original).

29. Id. at 257–311.

30. See Kronman, Anthony T., Mistake, Disclosure, Information, and the Law of Contracts, 7 J. Legal Stud. 1, 9–18 (1978)CrossRefGoogle Scholar (important to encourage deliberate searches for productive information).

31. See Richard A. Posner, Economic Analysis of Law 119 (7th ed. 2007) (“In many cases it is uneconomical to induce completion of performance of a contract after it has been broken.”).

32. See Polinsky, A. Mitchell, Probabilistic Compensation Criteria, 86 Q.J. Econ. 408, 409 (1972)CrossRefGoogle Scholar (bundles of changes with randomness of distribution).

33. See John D. Graham, Saving Lives through Administrative Law and Economics 395, 414 (2008) (single vs. repeated applications of Kaldor-Hicks).

34. Zamir & Medina, supra note 3, at 225–256.

35. Basing deontological constraint on autonomy or choice-set values is problematic in general. The right to choose one's contractual associates might be regarded as itself a species of autonomy. This applies as well to other examples from the book. Zamir and Medina defend free speech, even when it causes consequential harm, because of the autonomy or choice set of the speakers, but the harm itself, such as reputational harm from defamation, contracts the choice set of the victims. Perhaps one could say that this simply acknowledges that deontological constraint can apply even if violating the constraint would reduce the number of constraint violations, but it still begs the question of which is the constrained act.

36. See Posner, supra note 31, at 357.

37. This is especially likely regarding handicap discrimination, where firms are required to expend real resources in make reasonable accommodations. See, e.g., David C. Stapleton & Richard V. Burkhauser, The Decline in Employment of People with Disabilities: A Policy Puzzle (2003).

38. See Wonnell, Christopher T., Efficiency and Conservatism, 80 Neb. L. Rev. 644, 666–682 (2001)Google Scholar (defending ex ante efficiency ideal against objections).

39. The concept is discussed and criticized in Adler, Matthew D., The Puzzle of “Ex Ante Efficiency”: Does Rational Approvability Have Moral Weight?, 151 U. Pa. L. Rev. 1255 (2003)CrossRefGoogle Scholar. Broadly speaking, Adler criticizes ex ante efficiency because it is not an actual choice that embodies consent-based values nor a decision that would actually be in the person's best interest. The intuition of the ex ante approach, however, is that it makes sense to apply moral criteria to decisions that make the best available use of imperfect information about probabilities.

40. The original trolley hypothetical involved switching tracks so as to kill a smaller number of people, a hypothetical that tends to elicit different attitudes from deliberately tossing an innocent person in the way of the trolley. See Thomson, Judith Jarvis, The Trolley Problem, 94 Yale L.J. 1395 (1985)CrossRefGoogle Scholar (presenting the trolley hypothetical to illustrate deontology).

41. The original Kantian sentiment advises: “Act in such a way that you treat humanity, whether in your own person or in the person of any other, never merely as a means to an end, but always at the same time as an end.” Immanuel Kant, Groundwork of the Metaphysic of Morals 30 (James W. Ellington trans., 3d ed. 1993) (1785).

42. If one thinks of deontology in terms of the agent, A, rather than the victim, B, then the question is whether A should have the power to waive his own deontological objection to participating in the pushing scheme by advanced consent.

43. See note 28supra, and accompanying text.

44. See James M. Buchanan & Gordon Tullock, The Calculus of Consent 63–84 (1962) (describing a constitutional stage where uncertainty is sufficiently great that individuals can rationally support public-regarding policies).

45. This point is of considerable practical importance. One would worry greatly about empowering government officials with the ability to violate deontological constraints based upon theories that people somehow would have waived them at some earlier stage. The idea of “inalienable” rights is probably in many cases a practical expression of this fear.

46. See Zamir & Medina, supra note 3, at 42 (“The central deontological constraint is against harming other people.”).

47. This is not merely a matter of the action's being constrained even in the presence of autonomy benefits nor of reducing the number of constraint violations. The question is: Which action is constrained in the first instance? Is the speaker herself deontologically constrained from speaking in a manner that imposes harm? Is the government deontologically constrained from interfering with private choices of contractual associates? Perhaps there is a right answer to these questions, but the present issue is how easily the norm can be abused to rationalize policy conclusions reached on other grounds.