Here's Strawson's argument (quoting from the NYT piece):
(1) You do what you do ... because of the way you then are.
(2) So if you’re going to be ultimately responsible for what you do, you’re going to have to be ultimately responsible for the way you are ... .
(3) But you can’t be ultimately responsible for the way you are ... .
(4) So you can’t be ultimately responsible for what you do.
I take it that (2) is supposed to follow from (1) (since there's a 'So' at the beginning of (2)). Now, what implicit premise does Strawson need to make that inference valid? Something like the following:
(*) If X because of Y, then if you are ultimately [morally] responsible for X, then you are ultimately [morally] responsible for Y.
In other words, you can only be morally responsible for the effect if you are also morally responsible for the cause (=you can't be morally responsible for an effect unless you are also morally responsible for the cause).
I think this principle (*) may have counter-examples; I'm curious to hear about other people's reactions. Here's one such example: A person is driving along a very infrequently traveled mountain road, when some giant falling rocks hit him and his car. His car careens off the road, and he is badly injured by the rocks and subsequent crash. If he does not get to a hospital soon (and there's one 30 minutes away by car), he will die.
Now suppose someone drives by a few minutes later, and she sees that the first person is badly injured and will probably die soon if he doesn't make it to the hospital (imagine the second driver is a physician). She decides that picking him up, putting him in the car, and taking him to the hospital would be an annoyance, and she leaves him there (let's assume that taking him to the hospital wouldn't really be any significant cost/ burden for her).
Now, I think this may be a counterexample to Strawson's argument. Why? Let X be the first driver's death, and Y be the falling rocks hitting the car. It seems reasonable to say that (i) the first driver died because of the avalanche of falling rocks, but (ii) the second driver is at least partially morally responsible for the guy's death, even though (iii) she is certainly not morally responsible for the rocks hitting his car (she wasn't standing at the top of the mountain, pushing down rocks to harm passersby). If (i)-(iii) are correct, then we've got a counter-example to (*).
Someone who wanted to defend Strawson has at least one reasonable response to this: the second driver's leaving the first one to die is (at least part of) the cause of the guy's death; at the very least, we have the right sort of counterfactual dependence that is a necessary condition for causation (If she had not driven right past, then he would not have died). But now we are in the difficult and murky issue of causation by omission -- a contentious and vexed topic, since if we allow omissions to be causes in general, then a LOT of things become causes that we do not ordinarily think of as causes. (Sarah McGrath's "Causation by Omission" is an excellent treatment of this issue.)
So I guess my final conclusion is that, in order to accept (*), you have to accept a contentious metaphysical thesis about omissions being causes as well. This is of course not a refutation, but it certainly provides a respectable 'way out' of Strawson's basic argument.
8 comments:
While it seems about right to me that 'causation by omission' will make things a little awkward here, I also believe that there might be different degrees to that.
Your example for instance is designed in such a fashion that most legal systems in the world will hold the second driver responsible for the first driver's death. I don't know about the US, but e.g. in German law you can actually commit (the equivalent of) first degree murder by omission. So the example might be actually be underlining Strawson's point about moral responsibility comes down to control over causes.
That said I'm not sure the causation by omission thing is quite as clear cut as you indicate. To a certain degree it is a linguistic artifact: I can say the second driver drives on, not helping (omission), but I can also say he withholds treatment (action), depending on which path of action I accept as the 'default'.
I want to think a little more about the principle and the counter-example, but briefly, just to follow up on Firionel, the semantics for "cause" and its close relatives is very tricky. One thing that I think is especially misleading is that metaphysicians, who have worries about causation by absence, are imagining causation as a relation between events. Whereas, ordinary use of "cause" typically takes causation to be a relation beteen agents (or their actions) and outcomes. Moreover, ordinary use of "cause" is pretty thoroughly infected by normative concepts anyway. So, except maybe for billiard-ball collisions, when the man on the street says that x caused y, he or she is probably already saying something normative.
I think your example points to an even bigger problem. Even if the omission by the second driver is a cause or part of the cause, there is still a straightforward sense in which the first driver died because of the rocks falling. So it seems (by *) that the second driver still ends up being responsible for the rock slide. What's needed is some fancy clarification about X being because of Y in the right way. The trick then is that the 'because' in (1) has to be shown to be the same kind of because.
First, do you think that principle (*) is equivalent to:
(**) In order to count as [morally] responsible for Y, one must be [morally] responsible for all of the causes of Y.
I'm not quite sure how we're supposed to understand the "because" in Strawson's discussion.
If (*) and (**) are equivalent, then (*) just seems clearly nutty. As P.D. says, lots of things that we (apparently) aren't [morally] responsible for cause things that we (apparently) are [morally] responsible for.
But that makes me wonder. Is Strawson endorsing (*) or is he pointing out the consequences of (*) as part of a reductio?
@Jonanthan's 2nd (and sorta PD):
I certainly share your feeling that (**) is nutty, and the way I interpret Strawson's vocabulary, (*) and (**) are equivalent. (Perhaps this is a reductio of the claim that I understand Strawson.)
And this argument of is not presented in the context of a larger reductio (well, perhaps it's a reductio for you and me, but not for Strawson).
The fact that Strawson is really committed to something very close to (*) can be seen from his argument for the premise labeled (3) in the NYT article:
"But you can’t be ultimately responsible for the way you are." What he explicitly says is:
(i) You are not ultimately morally responsible for your genes or your upbringing.
(ii) You are the way you are because of your genes and upbringing.
Therefore, you are not ultimately morally responsible for the way you are.
And (it seems to me) he needs exactly the same premise (namely, (*)) to reach that conclusion from (i) and (ii). (Though here he's using the contrapositive: if you're not ultimately morally responsible for the cause, then you are not ultimately morally responsible for the effect.)
Final note: I wonder whether it matters that Strawson usually doesn't talk about moral responsibility period, but rather "deep moral responsibility" or "ultimate moral responsibility." Perhaps he would try to evade the nuttiness charge by drawing a distinction between (ordinary?) MR and deep/ultimate MR.
Having read over Strawson's post again, I think that the principle behind his Basic Argument -- at least in its second formulation -- is not so nutty as (*). I think it's closer to
(#) In order to count as [morally] responsible for some state of affairs Y, one must be [morally] responsible for at least some cause of Y.
Strawson then points out that there is an infinite regress of causes. Since we are never [morally] responsible for an infinite chain of causes, one is never responsible for any state of affairs.
I think this is both a stronger principle and a stronger argument than what we've been considering so far. But maybe I'm missing something. What do you guys think?
If every cause in your scenario has a reason and those reasons have their reasons then the person’s decision is but one reason in a complex network of causal events. Attributing responsibility is a survival mechanism that allows us to isolate a key component in this network of interacting events without which the problem event is less likely to occur or occur again. Isolating one element in the chain allows us to quickly and effectively deal with the problem. As people are usually a prime candidate for neutralising a chain of events we attribute what we call moral responsibility to them. All you effectively do in these thought experiments are measure the causal role a person is likely to play in a given scenario taking the knowledge and understanding he or she has at a particular time. In practise the event of moral responsibility becomes a mechanism for preventing other event from occur or re-occurring. So although moral responsibility plays a key role in decision making it is in fact an injustice as all past events are ultimately responsible for whatever action a person might take.
You can read more about my views in my book Get Me Adjusted (getmeadjusted.com)
Hi Philippe --
Thanks for stopping by. I think your response is more or less exactly the one Strawson (and his fellow moral-responsibility-deniers) would give for why practices of praise and blame, and accompanying carrots and sticks, are good.
Post a Comment