More on Haecceities

The venerable blogger Ocham (curator of the Logic Museum) has put up some good posts on my questions about haecceities:

http://ocham.blogspot.com/2010/09/paasch-on-haecceity.html
http://ocham.blogspot.com/2010/09/is-haecceity-repeatable.html
http://ocham.blogspot.com/2010/09/indivisibility-and-unrepeatability.html
http://ocham.blogspot.com/2010/09/why-haecceity-is-not-repeatable_24.html

I thought I could do otherwise, but I was mistaken

Let's assume that an action is free if it could have been otherwise: 

(T1*)  For any person x who performs an action A,
x performs A freely = iff
x could have done other than A,
given the same history. 

Now imagine the following scenario. Suppose you live in a big mansion. You're sitting in your study, reading an excellent philosophy book. I sneak up outside the door and lock you in, but I do this so stealthily that you don't even notice that I've done it. Soon it hits your bedtime (which is probably 9 o'clock), but after looking at your watch, you decide to stay and read a bit longer.

Was that a free choice? You couldn't have left, even though you don't know it. Some people might say that it is a free choice, even though you couldn't have done otherwise. If so, then this would be a counter-example to the idea that an action is free if it could have been otherwise.

We could go even further here and introduce moral responsibility. Let's assume that a person is morally responsible for what she does if and only if she could have done otherwise:


(T3)  For any person x who performs an action A,
x can be held morally responsible for A = iff
x could have done other than A,
given the same history.

Now suppose that we are back in the days of the cold war, and you decide to kill President Reagen. I find out about your scheme, and being a KGB spy, I want to ensure that you go ahead with it. While you're sleeping, I implant a little device in your head that will make you kill Reagen when activated. You know, in case you loose your nerve, I can activate the device, and you'll kill Reagen. But again, you don't know this. In the end, I don't need to activate the device, because you go ahead and kill Reagen anyway.

Would you be morally responsible for this? This is very nearly parallel to the last example. You couldn't have done otherwise, but you don't know it. Some people might think that you would be morally responsible, even though you couldn't have done otherwise. If so, then this is a counter-example to the idea that moral responsibility requires that you could have done otherwise.

(Notice a difference between these two examples: in both cases you do not know that you could not have done otherwise, but in the one case the question is whether you're free, and in the other case the question is whether you're morally responsible.)

Moral responsibility and free will

In my last post, I mentioned that some people say that a person can be held morally responsible for their actions if and only if they perform those actions freely, by which we mean they could have done otherwise than they did:

(T3)  For any person x who performs an action A,
x can be held morally responsible for A = iff
x could have done other than A,
given the same history.

This might be too strong. I'm not sure, but maybe. I'm thinking about cases where we have a very strong impulse to do something bad --- e.g., if someone held a gun to my head and made me smash your favorite porcelain plates, or if I stole your favorite porcelain plates so I could sell them and feed my crack addiction.

In these cases, someone could say:

'Look, you're still free there, because you could have done otherwise. You could have opted to be shot instead of smashing those plates, and you could have opted to suffer withdrawal instead of stealing those plates'.

Okay, sure. I'll grant that. After all, I'm assuming for the moment that an agent acts freely if they could have done otherwise. So these would be cases of 'free acts' in that sense. 

Still, would I be morally responsible for these actions? Probably not in the first case. (Unless I said something to you like this: 'I did it at gunpoint, man! You can't hold me responsible for destroying those plates that I hate so much'.)


I guess I'm saying that there might be cases where the accused had the choice to do otherwise than they did (and so were 'free' in that sense), but yet are not morally responsible for doing what they did. So I'm not sure I'd buy T3, which says that 'moral responsibility' and 'free action' extend to exactly the same cases.  

Moral responsibility and could-have-done-otherwise

Sometimes people say that moral responsibility requires free will, i.e., that we can act freely.

(T2)  For any person x who performs an action A,
x can be held morally responsible for A = iff
x performs A freely. 

But we can go further. We might assume that moral responsibility requires the sort of free will where we could have done otherwise. After all, when we say 'you should not have done what you did do', or 'you should have done what you did not do', we seem to be implying that you could have done otherwise. So if we assume that free will is this:

(T1*)  For any person x who performs an action A,
x performs A freely = iff
x could have done other than A,
given the same history.

Then one could say:

(T3)  For any person x who performs an action A,
x can be held morally responsible for A = iff
x could have done other than A,
given the same history.

Free will, could-have-done otherwise, and causal determinism

One common notion of free will is the idea of choice: you have a number of options in front of you, and you pick one. Or, perhaps a better way to talk about this is like so: you have free will if you could have done otherwise than you did. A provisional definition could be this:

(T1)  For any person x who performs an action A,
x performs A freely = iff
x could have done other than A. 

But that's not really good enough, because a determinist who believes that our actions are causally determined by our prior histories could agree with it. Such a determinist might say: 'of course you could have done otherwise, if you had a different history.'

Well, sure, I could have done otherwise if I had a different history. But that's still deterministic: one history determines me to do certain things, and another history determines me to do other things. And that's not really what I mean when I say that I could have done otherwise. What I really mean is that I could have done otherwise, even with the same history. So we could revise T1 something like this:

(T1*)  For any person x who performs an action A,
x performs A freely = iff
x could have done other than A,
given the same history.

That seems a little better.

Powers and impossibility

One might think that powers necessarily involve some relation to the activities for which they are powers. After all, we do not say that 'Jane has the power' without stating (or at least implying) what Jane has the power for. Thus, it would seem that a connection with the relevant sort of activity is built into the very nature of a power.

Still, I wonder what that actually means, ontologically speaking. One theory would be this: something has the power to do an activity if it is possible for it do that sort of activity. And if that were the case, then something x would have the power to perform an activity A if there were some world W where x (or a sufficiently similar counterpart of x) actually performs A.

I don't know if I buy this particular view, but it seems to me rather intuitive. After all, if there were no worlds at all where x performs A, then it would be impossible that x performs A. That's just the definition of 'possible' and 'impossible': something is possible if it happens in some world, and it is impossible if it happens in no world. So also here: if x does A in some world, then it's possible, but if x doesn't do A in any world, then it's impossible. And if it's impossible, then surely x does not have the power to perform A, plain and simple.

On this view, one could say that a power (or at least a type of power) is essentially connected to its corresponding activity, for there will always be at least one world where the agent actually performs the activity in question.