Sen, Nozick, and “Breaking Bad”

A moral dilemma from the popular TV show “Breaking Bad” illustrates a critique Amartya Sen made of Robert Nozick’s Anarchy, State, and Utopia and the reason why the refutation fails. In On Ethics and Economics, Sen makes the following critique of Nozick’s libertarian philosophy (heavily paraphrased because the book has yet to be unpacked, and Google Books was no help):

Suppose A knew that C was about to murder D, but needed a car to try to stop the murder. B is nearby in a car. On Nozick’s theory, it would be permissible for A to try to stop the murder without violating anyone else’s rights, but impermissible for A to to try to stop the murder by commandeering B’s car.

Sen seems to think that Nozick’s view is incoherent or at least implausible. Nozick’s theory forbids minor rights violations to prevent major ones. Of course, the theory is incoherent only if one adopts the premise that whatever is morally good must be maximized, a premise that Sen leaves implicit. Sen’s critique suggests a “consequentialism of rights”: always act so as to minimize the number of rights violations.

But the central plot twist of the “Breaking 312px-JesseshootsgaleBad” series shows us why consequentialism of rights is less plausible than a strict deontological view. In this plot twist (writing vaguely to avoid spoilers), the two main characters of the show murder an innocent man because: 1) they know a third person is about to murder them, and 2) they know this third person will not murder them if and only if the innocent man they kill is dead. Both act-utilitarianism and consequentialism of rights ratify their decision: they murdered one innocent in order to save two innocent (let’s suppose) lives from murder.

Intuitively, however, what they did was wrong. Murdering one innocent person to prevent two murders does not seem justifiable. The deontological, Nozickian view is more plausible.

A moderate rights-consequentialist might respond by revising the theory so that rights violations can never be justified to prevent rights violations “of the same kind or lesser.” This revision seems ad hoc and unlikely to work in such a way as both to salvage the theory and to distinguish it from non-absolutist deontology of the sort that Nozick favored. What counts as a “lesser” rights violation? Is rape justified to prevent murder? Armed robbery to prevent murder? Simple assault to prevent murder? As Locke noted in the Second Treatise, any aggression on a person’s body or immediate property poses a threat of murder and legitimate retaliation. After all, once the armed robber (or whatever) has you in his power, what is to prevent him from murdering you? So in the end the theory seems to collapse to something like “petty theft, fraud, embezzlement, and like offenses without risk of a physical encounter may be justified to prevent bodily assaults.”

But this position still doesn’t seem plausible in many cases. E.g., “I’m going to steal thousands of dollars from my employer so that I can hire a bodyguard because I’m afraid that someone, somewhere might assault me.” That doesn’t seem right. And where the position yields conclusions that accord with intuition — “It’s OK to break into a closed gun store to grab a weapon to defend my family who are currently being assaulted” or “It’s OK to steal a megaweapon from a mad scientist to fend off an alien invasion” — non-absolutist deontology can also yield the intuitive conclusion by resort to concepts such as situational monopoly (discussed extensively in Anarchy, State, and Utopia under the rubric of the Lockean Proviso) or hypothetical consent (no owner could reasonably deny emergency access, so long as compensation is provided).

In summary, a strict but flexible “side constraints” morality like Nozick’s provides more intuitive solutions to moral dilemmas across the board than does either utilitarianism or consequentialism-of-rights.

13 thoughts on “Sen, Nozick, and “Breaking Bad”

  1. My intuition is that killing another innocent person to save yourself is justifiable so long as you are innocent. Killing fewer people to save more (e.g., you kill people to maintain your cover in an Al Qaeda cell) makes sense to me – provided the chances are good that you will save more. I don’t know if there’s anything else to be said. That’s why I think intuitions are a poor way of resolving disputes. At what point do intuitions gain validity? When 51% of people have them? 67%? 90%? And why do numbers make them valid?

    1. Unfortunately, we don’t have any alternative to intuition in analytic moral philosophy. We have to get our premises from somewhere, and not from empirical evidence. But the role of intuition is different from what you suggest, not having to do with polling people or making blank assertions but with consulting our most fundamental “priors” and revising them with an eye to consistency. So for instance, I could respond to you that my intuition tells me that people may not be used as means only to others’ ends, and that your intuition ought to suggest the same, because no one can consent to being used as means only herself. Then I could relate my intuition that no one may murder in order to prevent more murders to the more fundamental intuition about how people ought to be treated in general, showing that the former follows logically from the latter.

      1. Fundamentalness doesn’t seem especially relevant to intuitionism. If I have a weak fundamental view, say, that moral principles are relative but a strong particular view that everyone ought to think the Holocaust was wrong, I should probably revise the more fundamental assumption.

  2. Sen’s critique of Nozick is potentially a little shallow too. B could be liable for commandeering the car; it doesn’t necessarily mean she shouldn’t take it. She will have to compensate the owner after the fact. Small infringements are easier to compensate than large infringements (like killing or rape).

  3. Your analysis here reflects that simple morality is superior to moral reasoning in many cases. Sen’s conclusion that commandeering the car to save an innocent life is the obviously moral response to the situation you pose. Using moral reasoning to reject an obviously moral action just shows the limits of reasoning.

    To me it doesn’t “seem” wrong that one innocent person should be killed to save two innocents. In fact, without knowing more, it seems right, though certainly morally complicated. If we are talking about one innocent child verses two, sick elderly people, then the situation is different. Context always matters. It is also morally complicated when considering what kind of practical political mechanism could be set up to make such decisions and implement them. It is a good thing that normal life isn’t ruled by these sorts of decisions.

    Is it justified to kill one innocent to save the lives of, say, thousands of similarly situated innocent people? Obviously it is justified (though I think that reasoning from extreme examples is not an effective way to derive sensible rules for moral behavior in everyday life). If your moral intuition tells you that it is not justified to take an innocent life to save (with certainty) many more similarly-situated innocent lives, then your moral intuition has been dulled by reading far too much Kant.

    Before relying on moral reasoning to give us answer to complicated questions, we need to make sure that moral reasoning has the ability to give us answers to the obvious questions. Otherwise we just end up with hubris, not morality.

    1. “Simple morality” seems to be in the eye of the beholder! I disagree that it is obviously the right thing to commandeer the car. What if the driver is armed, believes (reasonably) that the commandeering attempt is armed robbery, possibly to be followed by murder, and justifiably resists with lethal force? At the very least, the commandeer-er is under a moral responsibility to explain the situation & attempt to obtain consent before using force. When intuition doesn’t give us a clear answer, it seems to me we have to resort to moral reasoning: what are morally analogous situations, and what do our intuitions tell us about those? Given that reasonable people have different intuitions, we have to come to a conclusion using reasoning from intuitions that are universally shared.

      To be sure, I assumed in this post that Walt and Jesse’s actions (murdering one innocent to save themselves from being murdered) would be universally condemned by observers, but clearly I was wrong about that!

      1. Yes, it is possible to complicate a simple situation post hoc by adding a bunch of details that you didn’t mention previously that change the context and, thus, the morally right action. But that is just being tricky, not morally sophisticated.

        The question isn’t really whether unknown variables could change the situation into something that it doesn’t appear to be. The question is whether the simple situation (assuming your description describes the morally relevant variables) has an obvious moral response. And it does.

        A better question is whether it is OK to take something of limited value without permission to save someone’s life. In general (though not in every possible context) that isn’t a hard moral question.

      2. Sorry, I should’ve been clearer. I didn’t mean that the situation was materially different from what was originally presented, just that these are possibilities that anyone in that situation would have to consider when deciding what is justified. Bearing in mind Locke’s point about aggressors, someone considering commandeering the vehicle should think twice because it might be the case that the victim is armed & reasonably believes that the attempt is robbery. That consideration applies to any action generally of that type. I didn’t mean to invent post hoc details that materially change the nature of the situation, but to argue that the nature of the situation, taken in general, should always take into account the possible ramifications, and that doing so affects which course of action seems generally justifiable in the first place.

      3. Well, I can agree with this last comment of yours. Even though I can accept the rule that taking things that belong to others is not, in general, morally right, we can concoct specific examples in which taking things definitely is morally right.

        I also agree with your statement that we “should always take into account the possible ramifications.” Taking into account possible ramifications sounds a lot like “weighing the consequences of action” in my mind.

      4. I’ve no problem with taking consequences into account. I just don’t like maximizing maxims. 😉

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s