11/13/10

Does Determinism Rule out Responsibility?

From the department of fun coincidences--my "Meaning of Life" class read "Why Immortality Is not So Bad" by John Martin Fischer on Monday, and then had the pleasure of chatting with him yesterday--he was on campus to give a talk on free will.

Here's an example from Fischer's talk-- a "Frankfurt case," since Harry Frankfurt discussed a case like this in his 1969 paper "Alternative Possibilities and Moral Responsibility."
Because Black dares to hope that the Democrats finally have a good chance of winning the White House, the benevolent but elderly neurosurgeon, Black, has come out of retirement to participate in yet another philosophical example. (After all, what would these thought-experiments be without the venerable eminence gris—or should it be noir?)   He has secretly inserted a chip in Jones’s brain which enables Black to monitor and control Jones’s activities.  Black can exercise this control through a sophisticated computer that he has programmed so that, among other things, it monitors Jones’s voting behavior.  If Jones were to showany inclination to vote for McCain (or, let us say, anyone other than Obama), then the computer, through the chip in Jones’s brain, would intervene to assure that he actually decides to vote for Obama and does so vote.  But if Jones decides on his own to vote for Obama (as Black, the old progressive would prefer), the computer does nothing but continue to monitor—without affecting—the goings-on in Jones’s head.
In fact Jones (shown in white in the picture) shows no inclination to vote for anyone but Obama, Black doesn't intervene, and Jones does vote for Obama. 

What's the point?  Well, it's often assumed that determinism rules out moral responsibility.  Why?  Because if determinism is true, then whatever you do, you couldn't have done otherwise; whatever you choose, you couldn't have chosen otherwise.  On the face of it, it seems as if a person who couldn't have done or chosen otherwise isn't responsible for what he did do or choose.

But perhaps not!  Jones couldn't have done otherwise than voting for Obama.  But is it really obvious that he's not responsible for doing so?  Black never actually intervenes (he's a "counterfactual intervener"). All unfolds exactly as it would have, had Black not inserted the chip.  Fischer makes a cautious assessment:  if Jones is not responsible, it's not because he couldn't have done otherwise.  That then calls into question the standard reasoning that runs "determinism...so couldn't have done otherwise...so no responsibility."

Once you get to thinking about Black and Jones, you just can't stop. Have fun.

19 comments:

Faust said...

wow. I really don't get this one.

Let me get this straight:

Jones is free as long as he doesn't violate a particular condition. So he is free to choose Obama, but if he chooses not-Obama, he will be overidden.

Therefore "he" is not determined unless he goes outside a particular set of paramaters.

Therefore determinism doesn't take away moral responsiblity?

Is that really the argument? I must be missing something.

Wayne said...

Faust- Right, because it seems like moral responsibility requires the ability to do otherwise. Jones doesn't have the ability to do otherwise, but fully and completely wills to do what he's determined to do. Nothing prevented him from doing what he wanted to do, but he really had no other choice.

So would we say he chose to do what he did? It sure seems like it. So from Frankfurt's position seems to indicate its not about alternate choices available (in hard determinism there is none) but rather whether it is in accordance with the person's desires (specifically their second order desires, which is a desire about a desire. I desire to have some ice cream, and my second order desire about my first order desire (the ice cream desire) is "I want that desire to succeed" or "It's a good desire")

I really like Frankfurt's position, partly because I have trouble arguing against the Hard Determinist position. So metaphysically I might be a hard determinist, but that doesn't make much sense to be concerned about ethics, if I'm a hard determinist. Frankfurt saves me from such a conflict. We can argue about whether or not our second order desire should be this or that.

Faust said...

It seems clear to me that Jones can choose what he wants to do, but he can't choose what he does. If he wants to vote O then he can. If he doesn't want to vote for O then he will.

So his choice of his desire is not determined.

However we can just push the whole thing back. The thought experiment seems like a trick to me: first separate choice and outcome, then say "See! Just because he had no other choice, he's not determined!"

But the real problem with determinism is that the choice ITSELF is what is determined. Was Jones free to choose whether or not he WANTED to vote for Obama. Our DESIIRES are what are determined. Our actions are deriviatve of our desires. Talking about actions is a waste of time. The desires that motivate them are the only thing of interest.

Example: do people choose their sexual orientation? No? Then they have no free will about their sexual orientation. They just have it.

Did Jones choose his "political orientation?" No? then he had no free will about it.

One might say there is a disanalogy here: sexual orientation driven by "the physical" while political orienation is driven by "the mental."

Mental phenomena can be altered by other mental phenomena: this is where we will find free will if such a thing exists.

Generally we are comfortable with the idea that some portion of our being is determined, but there is still some free will wigling about. But the only way to settle the issue is to talk about the origin point of the desires that drive the action, and not get caught up in post-desire actions. It seems to me this thought experiment utterly fails on this point.

Jean Kazez said...

Faust, The thought experiment's purpose is to respond to reasoning that goes "if (A) causal determinism is true, it follows that (B) whatever people choose, they couldn't have chosen otherwise, and from (B) it follows that (C) they're not responsible."

The idea is to challenge the inference from (B) to (C) by describing a case where a person couldn't have chosen otherwise, but they do seem responsible,or at least it's not obvious that they're not responsible. The story of Black and Jones is suppose to be such a case. You're supposed to have the intuition that Jones couldn't have chosen other than the way he did (because Black was on standby to prevent it), yet Jones is responsible for his choice (since Black doesn't actually interfere).

Once you block the inference from (B) to (C) you no longer have any reason to make the inference from (A) to (B) to (C), so you seem to have reconciled determinism with responsibility.

That's the idea anyway!

Faust said...

I get it, but it's just nonsense as near as I can tell.

Here is the problem:

Jones CAN choose otherwise in this example. It's just that WHEN he chooses otherwise then he gets "redirected" by black. That's why we have the "intuition" that he is responsible. Because we know that he COULD have chosen not to vote for Obama (it's just that then he gets "forced" into voting for Obama).

To be exteremely clear there are two possible outcomes here:

Outcome 1: Jones chooses to vote for Obama and then votes for Obama.

Outcome 2: Jones chooses to vote for McCain and then is forced to vote for Obama.

The "blocking" of B is just pure trickery. The "couldn't have chosen otherwise" actually means "could not have had an outcome other than choosing Obama."

Lets replace "voting for Obama" with "murdering an innocent." So Jones is presented with an innocent person. If he murders them, then Mr. Black will not intervene. If he does NOT murder them, then Mr. Black will trigger a "forced murder response." So Mr. Jones is going to murder an innocent. He "cannot do otherwise."

When the murder case is reviewed by a high court armed with cerebrescopes, we intuit that they will find Jones guilty if they discover that he murdered the innocent of his "own free will." This is because he did in fact murder the innocent without any outside intervention.

But what if their cerebroscopes turn up evidence that the nefarious Mr. Black, well known murderer by proxy, hijacked Mr. Jones when he decided not to murder the innocent. Will they then still consider Mr. Jones responsible? Of course not. Mr. Black killed the innocent, and Mr. Jones was only a tool in his hands.

Jean Kazez said...

It seems like you changed your mind half way through.

You say that Jones CAN choose otherwise, even if Black is on standby to make sure he chooses Obama. [My thought-- That's awfully subtle. In what sense can he choose otherwise?]

Then you imagine that Jones chooses to murder someone, and Mr. Black is on standby to make sure he makes that choice. You say he "cannot do otherwise."

Hmm!

I agree that it makes sense for the court to find the second, murderous Mr. Jones responsible, but maybe that's despite the fact that he could not have done otherwise. That's exactly the point of the thought experiment!

Faust said...

Only one of the two murderous Mr. Jones is responsible: the one that was not interfered with by Mr. Black. Similarly in the with the Obama voting.

There is no mind changing here. There is 1. Choosing and 2. Doing.

In both cases Mr. Jones CAN choose otherwise, but he can't DO otherwise.

Responsibility pertains to the choosing not to the doing.

The court will find the muderous Mr. Jones who chose to kill (and then killed) guilty, but find the Mr. Jones who did NOT choose to kill (but who then killed anyway) innocent. The court is interested in INTENT not action.

This intuition pump conflates choosing and doing, it says "cannot DO otherwise" but ignores the fact that it has already given the whole enchilada away by giving Jones CHOICE before the action.

Faust said...

Here is a variant on this:
Let’s say that Mr. Jones is going to vote. He will do this by pushing one of two buttons.

One button says "Obama"
One buttons says "McCain"

Unbeknownst to Mr. Jones, both of these buttons operate the same circuit: they both will vote for Obama.

Can Jones choose not to vote for Obama? Yes. He can choose to vote for McCain.

Will he in fact vote for Obama? Yes he will.

His choice is not predetermined, but the outcome is. If a voting fraud commission were to discover this voting fraud they might solve the fraud by asking each voter their intent. Or perhaps they had a camera watching what button people pushed. The point is that the intent is what is important. It is only ever intent that matters when it comes to responsibility. If we find out that someone did, to the very best of their ability, try to bring something about, but that "outside" circumstances prevented them from doing so then we do not consider them responsible.

Jean Kazez said...

The chip in Jones's brain is on standby to make sure he chooses to vote for Obama (and so does vote for Obama).

"If Jones were to show any inclination to vote for McCain (or, let us say, anyone other than Obama), then the computer, through the chip in Jones’s brain, would intervene to assure that he actually decides to vote for Obama and does so vote."

Faust said...

Even better!

So to reframe:

"If Jones were to show any inclination to refrain from murdering the innocent, then the computer, through the chip in Jones’s brain, would intervene to assure that he actually chooses to murder the innocent."

I'm 100% confident that the court, having discovered the chip in Jones brain that was programed to cause him to murder would find him innocent if the chip had been "activated" by a "inclination to avoid murder." And of course if the chip did NOT so activate, then he would be fully responsible for what he had done.

"Inclination" is too soft here. It should read "should Jones CHOOSE to vote for McCain..." etc. The chip is like a logic gate. (Maybe a NOT gate? As long as it gets a signal that X is going to occur it allows a positive value. If the signal stops (i.e. NOT Obama, NOT murder) then it returns a zero and triggers. I'll have to think about what kind of logic gate this is.)

The essential point here is that Jones is "free" until the chip is activated. Then he is no longer free. So he is responsible up until he has a chip activating choice, at which point he looses his independent choice making capacity (and thus looses responsiblity).

The irony of all this the logic gate metaphor is such that all one would need to do is see if the gate was activated.

If no chip was activated THEN Jones was free (and responsible).

If the chip was activated THEN Jones is not free (and not responsible).

The NEAT TRICK that this thing pulls is to introduce "cannot DO otherwise" because of the chip. But that's only outcome focused. It's just like my button analogy but the buttons are inside Jones's head, right at the beginning of the logic gate.

Wayne said...

Maybe a better example might be Locke's prisoner example. Imagine that you wake up one day in a room that you did not fall asleep in. Its a rather comfortable room. There a lot of interesting things to explore. And inside the room, is a long missed friend happy to explore the room with you. You never once think to yourself, "I would like to leave this room." So you willingly stay in the room. Didn't you choose to stay in the room?

Now on top of everything I said above, the room is locked. You cannot, in fact, leave the room. But isn't that really besides the point? If the room were unlocked, you wouldn't leave.

Frankfurt would say that (so long as it is in accordance with a second order desire) that you're choosing to stay, and that it is of your free will.

Frankfurt's argument is a neat sidestep around the typical objections against compatiblism.

Faust said...

I would say that choosing plays no part in the room bit. Choice has, as it were, not even arisen. It's pre-choice. A bit like distracted driving.

Wayne said...

How is it not a choice? Lets say while you're in this room, you see the door, and think, "gosh I could leave right now..." (which is incorrect) "But I don't want to at all! This place is great! I choose to stay here for a very long time!"

If you want to stick to your guns, the only reason why this is not a choice, is because the door is locked. If the door was unlocked, this would be a choice. But why?

Faust said...

If you ammend it that way I agree it's a choice. In your original description there was no "I could leave right now.." thought presented.

So lets take a look here:

I'm in a room I can't leave (the door opens to a brick wall, it's a false door, a locked door etc etc).

I choose to stay in the room (OR I choose not to leave the room).

But I CAN'T leave the room.

Therefore...so what? The choice swings completely free of the door being a real way out or not.

If I'm in a room and I decide to stay then I've freely chosen to stay.

If I'm in a room and I decide to leave then the fact that I can't leave becomes a discovery about my circumstances. I discover that I can't leave. But I DID CHOOSE to leave. I just can't ACTUALLY leave because I'm constrained.

Same with Jones. He can CHOOSE to vote for McCain (he can "have the inclination" which the chip then alters). He just CAN'T vote for McCain because his inclination will be transmuted by the chip in his head. But the chip IS NOT JONES, just as the locked door has no impact on my CHOICE to leave the room. It impacts my ABILITY to leave the room.

Let me try to make this clear by going back to what this argument is about. As Jean notes above:

"if (A) causal determinism is true, it follows that (B) whatever people choose, they couldn't have chosen otherwise, and from (B) it follows that (C) they're not responsible."

All of the preceeding is about creating a situation where B obtains but C doesn't, thus proving this chain of reasoning false.

The problem with both Mr. Jones and The False Door is that in both cases it is simply false that Mr. Jones "can't not choose Obama" and that the person in the room "can't choose to leave."

Mr. Jones CAN choose McCain, it's just that the chip will activate and change his decision as soon as it dectects it. But the CHIP is making the decision then, not Jones.

Likewise I CAN choose to leave the room but then the door will reveal my inability to leave, and my desire will be thwarted. But my CHOICE was not thwarted, only my ability to leave.

Thus neither example provides a case where someone "couldn't have chosen otherwise" and no headway is made on this problem.

Wayne said...

Right.... but Frankfurt is taking issue with B. He thinks we can be responsible for things that we choose, even if we couldn't choose otherwise.

So I make a choice. X. It seems to me that I could choose Y when I'm choosing, but in fact I can't. We both seem to agree that I've chosen X. (Hard determinists would argue that I didn't choose)

Now let us say that X is a morally wrong thing. If I choose X, then I would be morally culpable for that act, even if I couldn't in fact pick Y.

Frankfurt's drug addicts I think make a reasonable example of what he's trying to say. Imagine an Addict that doesn't want to be an addict (2nd order desire), but still uses drugs. He's clearly not free, he's bound and determined by his addiction.

Now imagine an addict that WANTS to be addicted(2nd order desire). He thinks its great! Would we say that he's determined in the same way as the first addict? Well neither really have the option to do otherwise (lets stipulate for the sake of argument), but the second addict seems far more responsible for his current situation than the first.

Faust said...

I think the second order desire stuff is a bit different from the The False Door example, and the Mr. Jones stuff, and very interesting on its own. But I'm not sure about it. I'm not sure (but open to consider the possibilities surrounding) the following:

1. Whether or not there is any realtionship between a second order desire and the possibility of fulfilling the desire.
2. Whether or not we can escape an infinite regress where we have 3rd and 4th order desires and so on.
3. Whether or not we can always detect if a particular desire is in fact second order e.g. sometimes I simply have two co-equal desires that conflict and I may have different second order desires that switch off depending on which first order desire is in play.

But I do think the second order desire stuff is very interesting. I'm just not convinced it helps us much in the compatability department.

I do think that its disconnected from the other stuff under discussion. I don't see that second order desires have anything to do with Mr. Jones for example.

Wayne said...

Frankfurt says that in order for something to be a second order desire, you have to decisively identify (or something to that effect... I forgot what the exact phrasing is) with it, in order to stop infinite regressions like this.

If Mr. Jones was acting in accordance with his second order desire, then he is free. He gives us this definition of freedom, since traditionally it is having the option to do otherwise, which he is undermining with the Jones thought experiment (an example of a free act that doesn't have any options to do otherwise).

Faust said...

Decisively identify hmmmm? And does one CHOOSE to decisively identify...or is decisive identifcation something that HAPPENS to you?

I think in terms of language with the "X to do otherwise" we need to be very careful. The preceeding dicussion shows very clearly (to my mind anyway) that "couldn't have done otherwise" doesn't cut it as a phrasing. Only "couldn't have chosen otherwise" is of interest when it comes to free WILL, and I reject that the Jones argument pertains to "choosing otherwise" for the reasons given above.

You note in your second post above that

"Frankfurt saves me from such a conflict. We can argue about whether or not our second order desire should be this or that."

But I take issue with this. I don't see how we can get away with avoiding the regress. It seem if we have the ability to choose a second order desire then we just wind up back at the same damn problem.

A general note: I believe that the reason free will presents such a problem is that we have no non-arbitrary way to create a slice in time that decides who the unitary "decision maker" required by our day to day concept of free will IS. WHO is doing this deciding? A humuculous in a Chinese Room? A viewer of a Cartesian Theatre? Is it made after a war of divergent affects a la Nietzsche?

So the problem with the Jones example to me is that people get confused and think that (Jones + Chip) is = Jones. But that's not necessarily right. Jones can choose who to vote for, but (Jones + Chip) can't. But it's easy to get swept away and think that (Jones + Chip) is Jones and that therefore Jones is the one who "can't choose otherwise." But only (Jones + Chip) can't choose otherwise. But who decides at which level of the system we step in and measure the "free will."

And of course as a hard determinist I would say that just plain old Jones is actually (physical bit + physical bit + N bits) each of which are determined (or randomized) and that if we choose a different layer to slice there is no free will left to be found at all.

Jeremy Stangroom said...

This is a strange one. Jones couldn't have chosen any other course of action than he did choose, but he certainly could have *originated* - to borrow the not necessarily coherent language of non-determinism - a choice not to have voted for Obama (which would have been evidenced by the triggering of the chip); and it's that possibility which hard determinists deny, and which lies behind the thought that not being able to act other than one does act does away with moral responsibility.

Origination is still intact in the Fischer scenario. If you take it out of the scenario - which makes the chip an irrelevance (because Jones cannot act other than he does act regardless of the chip) - then you still get the intuition that he isn't responsible for voting for Obama.

In other words, the scenario begs the question.