Libertarianism without inequality (4)

by Chris Bertram on November 11, 2003

Below the fold are some more (and slightly belated) reflections on Michael Otsuka’s “Libertarianism Without Inequality”: . Today’s offering concerns chapter four. (Earlier posts concerned “one”: , “two”: and “three”: .) As then, comments are welcome from those who are reading or who have read the book.

Chapter Four of Mike Otsuka’s _Libertarianism Without Inequality_ concerns the right to kill the innocent in self-defence. That probably seems an odd thing to write about, so some clarification is in order. Otsuka takes it that we have a natural right of self-defence which extends to the right to kill others who deliberately set out to kill or injure us. This chapter is concerned to define some of the boundaries of that right. It is one thing to kill an aggressor in self-defence, but here Otsuka wants to clarify how far the right to self-defence might give a
person the right to kill others, such as bystanders, persons he calls “innocent threats” and others he calls “innocent aggressors”.

First, then, some clarification of terms:

“Bystanders” are individuals who are not involved in the chain of events in which another person is a threat to my life.

“Innocent threats” are people who threaten my life, but only do so qua physical object and not because they have formed an intention to do me harm. A person falling out of a building who is about to land on me and crush me to death would be an example.

“Innocent aggressors” are persons who have formed designs against my person, but who cannot be held morally responsible for those designs or the consequent actions. So a person subject to mind control, or whose drink has been spiked with a powerful drug, may form the intention of killing me, but that is not an intention for which we can hold them responsible.

Otsuka argues that none of these people can permissibly be killed by someone in self-defence. That’s a surprising and, I think, counterintuitive, claim. To support it, Otsuka’s strategy is to build outwards from the – as he sees it – uncontroversial case of the bystander, arguing step by step that there are no morally relevant differences between the bystander case, the innocent
threat case and the innocent aggressor case and that therefore we ought to retain the same commitments in the “deeper” cases as in the more straightforward one.

Otsuka’s method is basically casustical: he moves backwards and forwards between examples and principles trying to get a good fit between our intuitions and the general principles we ought to support. I don’t have a problem with that, although it worries me that our intutions in some cases seem to have a privileged status compared to our intuitions in others.

The paradigm of killing a bystander in self-defence would be where a person grabs hold of another and uses them as a shield against an oncoming missile (p. 68). This case or a case where one averts a threat to oneself by initiating a sequence of events that kills and innocent stranger are distinguished by Otsuka from other instances (see. p. 68 n. 7) where one ducks out of the way of an
oncoming threat and someone standing behind you is killed as a consequence. Otsuka therefore defends a strong and clear-cut distinction between doing or making and allowing, where prohibitions attach to the former but less often and less stringently to the latter (and this may trouble some readers).

What is going to bother most readers, I suspect, is Otsuka’s claim that Innocent Threats are morally on a par with Bystanders. The thought here seems to be that neither Innocent Threats nor bystanders are acting with your death as their goal. When a person acts with your death as their objective, they are doing a great moral wrong and violating (or trying to violate) your right not to
be killed. That intentional action seems to be what gives you a permission to kill them, or, to put it another way, removes the prohibition on killing another to save your own life. Innocent Theats are not, _ex hypothesi_ acting in a way that justifies removal of that prohibition.

The extension of this idea to Innocent Aggressors is straightforward enough. True, they are acting with the intent of ending your life. But the conditions surrounding their actions are such that they cannot be reasonably held morally responsible for those very actions. Otsuka distinguished between those who have permanently lost the capacity for moral agency and those who have not. Those who have permanently lost the capacity may indeed be killed. But this is ok because they now have a lower moral status such that preserving my life at their expense does not constitute an impermissible making use of them. Those who are suffering merely temporary loss of moral personhood are, though, in principle, morally equal to myself and may not therefore be treated by me as appropriate objects of killing (even to save my own life).

Otsuka considers the following objection to his view, viz

bq. I have employed the intuition that it is impermissible to kill a Bystander as a fulcrum by means of which to dislodge the intuition that it is permissible to kill a Threat. Reversing this line of argument, one who endorses the oral-equivalence thesis might try to employ the intuition that it is permissible to kill a Threat as a fulcrum by means of which to dislodge the intuition regarding the impermissiblility of killing a Bystander. (p. 79)

But the point in his argument where Otsuka raises this “reversal” possiblility is not the point where it is posed most sharply. The Innocent Aggressor case that he discusses is one where the Aggressor is drugged, but it is easy to imagine other case where the Aggressor lacks moral responsibility but where our intuitions may be different. Consider the case of a child trained from birth as an assassin, or more generally of people who have not grown up in circumstances that have been propitious for their moral developement (perhaps they have been bombarded with anti-semitic propaganda all their lives). When such people come after their victims, gun in hand, Otsuka’s view seem to say that those victims may not kill their attackers in self-defence. That will seem crazy to many people and it is hard to see that their intuitions about such cases are less secure starting points for reflection than their intuitions about Bystander cases.

I’m not sure what to think here, because I want to resist both the temptation to slide back to revise my view on Bystanders and the conclusions of the moral equivalence thesis. There’s a slightly Sorites-like feel to Otsuka’s arguments here, the suggestion that we ought to accept an extension of a moral prohibition to a new class of cases because they are not really significantly different, and then to a further class of cases …



Dick Thompson 11.11.03 at 7:28 pm

Is he able to discuss the case of soldiers in combat? Are the opposition soldiers (who may have recieved strong training and even upbringing to reduce their full scope of moral agency) fair game?


Antoni Jaume 11.11.03 at 11:13 pm

As I see it, it is always bad to kill someone. Now maybe there is some cases in which one has no other choice to avoid that one or another individual is killed. Then what we should seek is if one has taken apropriate mesures to avoid such situation.



Michael Otsuka 11.11.03 at 11:24 pm

Here’s what I say about soldiers: ‘Warfare creates an atmosphere of emergency and catastrophe in which ordinary deontological constraints begin to lose their grip. If we allowed ourselves to be defenceless in the face of a malevolent enemy leadership that operates through the agency of [Innocent] Aggressors or protects itself and manipulates its opponent through the use of innocent shields and hostages, then the constraint against the killing of [Innocent] Aggressors would be observed at too high a price.’ (LWI, p. 82, n. 29)

The last sentence is relevant to Chris’s case of a child trained from birth as an assassin. Just as it might be too costly to adopt a general policy of refraining from killing innocent human shields, it might be too costly to adopt a policy of refraining from killing innocent trained child assassins. If, however, we assume that the adoption of such a policy has no such bad consequences, I think it far from crazy to maintain that the child assassin’s target may not kill this child in self-defence. (Note that we must assume that this child has not permanently lost his capacity for moral agency. He can, in other words, be rehabilitated into a morally decent human being.) What’s the morally significant difference between this case and a case in which the only way to prevent a morally responsible adult assassin from shooting someone is to kill his innocent child? I defy the reader to provide a convincing answer to this question.


john mcgeady 11.12.03 at 6:13 pm

Otsuka’s exception for warfare is a non-sensical cop-out that renders his deontology a self-contradictory mess.

Let’s review: there’s a person and, according to Otsuka, he may not morally kill an innocent agressor. But let’s say there are 2 people, and they call themselves a “nation,” and they call their impending encounter with innocent aggressors “warfare,” then Otsuka says they MAY morally kill innocent aggressors, because the
“price” would be too high otherwise.

So all you have to do to override the “deontological constraints” Otsuka painstakingly presents (i.e. his whole damn moral philosophy), and to start using situational (and of course completely subjective) cost-benefit analysis to determine whether you can morally kill someone, is to move from an “I” to a “we.”

What kind of moral standing does a collective have anyway? It’s an abstraction, as are its interests. It’s moral agency derives from the aggregate or deliberated moral will of the persons in the collective. Yet Otsuka treats its moral interests as more fundamental and worthy of exceptional cost-benefit analysis than the very real moral interests of a very real person.

Otsuka, rethink.

Comments on this entry are closed.