Praise and prevention

Abstract

I argue that it is possible to prevent (and to be praiseworthy for preventing) an unwelcome outcome that had no chance of occurring. I motivate this position by constructing examples in which it makes sense to explain the non-occurrence of a certain outcome by referring to a particular agent's intentional and willing behavior, and yet the non-occurrence of the outcome in question was ensured by factors external to the agent. I conclude that even if the non-occurrence of an unwelcome outcome is ensured, the agent whose action explains this non-occurrence is praiseworthy for preventing the outcome. My examples are similar in important respects to Frankfurt-type cases. In the second half of the paper, I discuss the relationship between my examples and Frankfurt-type cases involving both actions and omissions. I conclude that while I may be responsible for the consequences of an action even if those consequences are unavoidable, this is not necessarily so for the consequences of omissions.

Keywords:

• responsibility
• prevention
• Frankfurt-type cases
• consequences
• omissions

Acknowledgements

I thank Joseph Baltimore, John Martin Fischer, Dana Nelkin, David Shoemaker, Dan Speak and Christopher Yeomans for sharing their thoughts on earlier versions of this paper, and for discussing various cases of prevention with me. I also thank two anonymous referees for Philosophical Explorations for their very helpful comments and advice.

Notes

Some might insist that moral praise and blame should respond solely to our intentions and motivations, and not to whether these happen to lead to good or bad consequences. On this view, the fact that the statue turned out to be plastic would not affect the praise to which I am entitled. However, I shall assume in this paper that it makes sense to hold people morally responsible for, and to praise or blame them on account of, the consequences of their actions.

I am grateful to a referee for Philosophical Explorations for suggesting this way of expressing the worry.

Some with whom I have discussed the permanent force field case have been reluctant to agree that I do not prevent a basket in this case. It can help to consider the following variation. Suppose there is a basketball court inside a brick gymnasium and that outside the gym a child is bouncing a basketball against its walls. Sometimes the child throws the ball in such a way that it would go into the basket inside the gym, if the gym walls were miraculously to vanish. Now suppose a second child sometimes intercepts the ball before it hits the wall. Even in those interception cases in which the ball would have gone into the basket if it had not been intercepted and if the walls of the gym had vanished, we would not say that the second child prevented the ball from going into the basket: he merely prevents the ball from hitting the wall. We should say something similar in the permanent force field case – namely, that I merely prevent the ball from encountering the force field – for it would be no less miraculous for a ball to pass through the impenetrable force field than it would be for the gym walls to disappear.

A referee for Philosophical Explorations suggests that we might reject my interpretation of the temporary force field case on the following grounds. Since the force field is triggered automatically, no one needs to do anything for the force field to prevent the ball from entering the basket, so the case is too similar to the permanent force field case to justify drawing different conclusions in the two cases. However, if the gambler had to watch the progress of the ball, and press a button to trigger the force field, then – the referee suggests, and I concur – I would prevent a basket if I kept a ball from reaching the point at which the gambler would trigger the force field. And this is so even if it is certain that the gambler will always press the button at the right moment. If a reader shares this reservation about the temporary force field example, I think she can substitute this new ‘watchful gambler’ case without threatening the argument in the rest of the paper. However, I think the cases of the temporary and permanent force fields are sufficiently different to justify different conclusions about whether I prevent a basket. Moreover, it seems to me that the temporary force field case is relevantly similar to the watchful gambler case. While the temporary force field activates automatically, this does not mean that it will necessarily activate; it will do so only if nothing has already kept the ball from getting close to the basket. Presumably, the mechanism that generates the force field has some means of registering a ball's trajectory, and reliably correlating this with the activation of the force field. But this is very similar to the role that the watchful gambler (or Superman) is supposed to play – at least if we are certain that the gambler will stop the ball from reaching the basket (if it has not already been stopped). Of course, the gambler triggers the force field intentionally and on the basis of certain motivations, and this does not happen if the force field is triggered automatically, but I do not see that this is a relevant difference in this context. Since the temporary force field will not necessarily activate, the possibility of its activating is one of several changes that might occur to stop the ball; of course, the force field is a ‘fail safe’, but the watchful gambler (or Superman) plays the same sort of role. The permanent force field, on the other hand, is already activated before I do anything, so its activation is not one of several possible changes in the world that might explain why a ball does not make it to the basket. The permanent force field is, rather, a background feature of the causal landscape against which events unfold. In the temporary force field case, one of the relevant background features is that the force field might activate (and is guaranteed to do so in certain circumstances), just as in the watchful gambler case it is a background feature that the gambler might stop the ball (and is guaranteed to do so in certain circumstances). This difference between the two force field cases seems to me to account for why I prevent a basket in one case, but not the other.

As will become apparent in Section 5, my approach here is very similar to the one taken by Fischer and Ravizza in their account of responsibility for the consequences of actions.

We would have a prevention example that is also a traditional Frankfurt-type case if we posited a counterfactual intervener who is ready to ensure that Jones (in particular) will prevent a certain outcome. An incompatibilist will have reason to object to this sort of case, but the prevention cases I have considered are not like this.

Fischer (2006) offers a critical discussion of several incompatibilist approaches along these lines – in each case, the incompatibilist aims to find what Fischer has dubbed a ‘flicker’ of freedom.

Of course, we can reinterpret an instance of prevention as a positive causal consequence, and vice versa. Instead of saying that I am responsible for preventing a basket, we could say that I am responsible for the consequence that the basket was not made. This new responsibility claim would be true only if, when we exclude any interference with the path of the ball, the consequence that did in fact occur (i.e. that the basket was not made) would not have occurred. Likewise, ‘Missile 2’ can be described as an instance of prevention. We could say (albeit awkwardly) that Elizabeth is responsible for preventing DC from not being bombed. In this case, there should be an associated counterfactual scenario in which the thing that was actually prevented was not prevented.

We can distinguish between simple (‘bodily’) omissions, and complex omissions (Fischer and Ravizza 1998, 132–4). Simple omissions are constituted entirely by the fact that a certain agent has failed to move his body in a certain way, whereas complex omissions are constituted by simple omissions and consequences in the external world related to those simple omissions.

Here is the relevant passage from Fischer and Ravizza:

Our account implies that we must hold fixed the actually existing pen, when ascertaining whether the event of the child's not being saved by John is appropriately sensitive to John's bodily movements. Thus, we must say that John is indeed morally responsible for not saving the child …. In ‘Penned-In Sharks’, one holds fixed the actualized conditions, and ‘subtracts’ or disregards the conditions that would have obtained in the alternative sequence. (1998, 138)

A referee for Philosophical Explorations notes that we might describe the actual outcome in ‘Penned-In Sharks’ by saying, ‘the child drowns while John decides not to save him’, and we might describe the alternate outcome by saying, ‘the child drowns while John's attempt is thwarted by sharks’. The referee suggests that Fischer and Ravizza might say that this way of characterizing the outcomes makes it clear that John's actual inaction explains the child's drowning in just the way my blocking a shot explains why I prevent a basket (as opposed to what happens when Superman blocks a shot). These re-descriptions of the actual and alternate outcomes in ‘Penned-In Sharks’ are fair, but I believe that what they bring out is that John is open to harsher moral assessment in the actual sequence than in the alternate sequence because in the former case John does not try to save the child. John is (we may think) more blameworthy in the actual sequence because his inaction reveals a defect in his character, but this does not mean that the child's drowning is explained by John's inaction, or by his character defects. Even if John had a better character, and had acted better, the child still would have drowned, so it is hard for me to see how the drowning can be on John's account in a meaningful way. In the Superman case, however, things are different: if I do not stop the ball, then Superman will. Thus, the relevant (unavoidable) consequence in the Superman example is brought about in different ways in the actual and alternate sequences, which makes it intuitive (I have suggested) to explain the actual sequence consequence in terms of my actual–sequence behavior. But in ‘Penned-in-Sharks’, no matter what John does, the child dies in just the same way, so it is not clear that his actual inaction helpfully explains why the child drowned.

In a few passages in this section, I have shifted from consideration of the claim that John is morally responsible for not saving the child in ‘Penned-In Sharks’ to consideration of the claim that John is morally responsible for the fact that the child drowns. This may seem a questionable expansion of Fischer and Ravizza's claim, but I think it is justified. Very many people fail to save the child from drowning, but John alone is supposed to be morally responsible for his failure in this regard. Since the child actually drowns, and John is morally responsible for not saving the child, this is presumably meant to put the drowning on John's account in a way that distinguishes ‘Penned-In Sharks’ from ‘Sharks’; recall that in the latter case, Fischer and Ravizza say that John is responsible merely for not trying to save the child.