Monday, June 23, 2008

Do Frankfurt Counterexamples Presuppose Materialism?

According to this paper, they do.

12 comments:

Anonymous said...

not going to convince many christians since it sems to deny that God could know the future free actions of human agents.

IlĂ­on said...

Victor Reppert: "Do Frankfurt Counterexamples Presuppose Materialism?"

You may (or may not) have noticed that I've never commented on any threads about "Frankfurt counterexamples" Well, I do try to make it a habit to not express an opinion on that of which I'm wholly ignorant. After all, how can one actually have an opinion without a basis?

For me, "Frankfurt counterexamples" still signifies nothing -- I haven't yet seen a need to find out what you all are referring to by the term. And then, when I see how the discussions about them seem always to go, I conclude (only part jesting here): "Oh! This is one of those pointless and meaningless things professional philosophers like to yammer on about from time to time."


Nevertheless, my answer to the question is "Yes. Obviously." Though, in truth, that "obviously" is qualified as I'll explain.

A couple of weeks ago I was Googling something or other. One of the hits I read was a page on the "Stanford Encyclopedia of Philosophy" which includes a small discussion of "Frankfurt cases:"


Stanford Encyclopedia of Philosophy: Foreknowledge and Free Will - The Frankfurtian/Augustinian solution:

"... Here is an example of a typical Frankfurt case intended to show that an agent can act freely even when she lacks alternate possibilities:

Black, an evil
neurosurgeon, wishes to see White dead but is unwilling to do the deed himself. Knowing that Mary Jones also despises White and will have a single good opportunity to kill him, Black inserts a mechanism into Jones's brain that enables Black to monitor and to control Jones's neurological activity. If the activity in Jones's brain suggests that she is on the verge of deciding not to kill White when the opportunity arises, Black's mechanism will intervene and cause Jones to decide to commit the murder. On the other hand, if Jones decides to murder White on her own, the mechanism will not intervene. It will merely monitor but will not affect her neurological function. Now suppose that when the occasion arises, Jones decides to kill White without any “help” from Black's mechanism. In the judgment of Frankfurt and most others, Jones is morally responsible for her act. Nonetheless, it appears that she is unable to do otherwise since if she had attempted to do so, she would have been thwarted by Black's device. (Adapted from an example by John Fischer, 1982).

Most commentators on examples like this agree that the agent is both morally responsible for her act and acts freely in whatever sense of freedom they endorse. They differ on whether ...
"


I'm assuming that a "Frankfurt case" is the same thing as a "Frankfurt counterexample" and my answer of "Yes. Obviously." follows from that assumption and the nature of the above "example of a typical Frankfurt case," which is so obviously constructed on a presupposition of materialism.

Anonymous said...

ilion,

God could do the same with the mind. He knows if you were to try to choose other than he wanted you to. He could then, if he saw you were going to choose other than he wanted you to, make you stay on course.

Say libertarian free will is true.

Say Judas had it.

Say God wanted Judas to betray Jesus to fulfill prophecy.

Say that if Judas backed out at the last second, God would control his cartesian mind (or emergant mind, or whatever) through his omnipotence and "force" Judas to betray Jesus. But, it just so happens that he sees Judas going through with it and so doesn't need to do anything.

These three seem true:

A. Judas is responsible.

B. Judas couldn't have done otherwise.

C. Judas had an immaterial mind.

Saint and Sinner said...

"Do Frankfurt Counterexamples Presuppose Materialism?"

Only if you assume that the spirit/soul is by definition causally-indeterminate.

[No, I didn't read the paper. I have to go back to work now.]

Clayton said...

I just tried to post this but the blogger ate the comment.

I've read the paper and at a crucial point in the argument for the claim that Frankfurt examples presuppose the falsity of substance dualism, the author seems to be assuming:

(*) On every sensible version of substance dualism, when an agent S performs an action A at t that we'd identify as one for which S is properly held morally responsible, God could not have known just prior to t that S would perform this action A.

What justifies that assumption? It seems to assume (at the very least) that:

(**) If S's action A at t is not causally determined by events just prior to t, God could not know just prior to t that S would A.

That claim is controversial (to say the least). It suggests that God could have extremely limited knowledge of future events in worlds that are not strictly deterministic. Moreover, it seems that we can distinguish between having 'knows' in a typical loose sense and having absolutely certain Cartesian knowledge. While (**) might be true if we have that Cartesian conception of knowledge in mind, it isn't clear why that sort of knowledge is required for a Frankfurtian intervention story.

Anonymous said...

Hi Clayton, yes, that's the point I made above. If Victor uses this to argue against Christian compatibalists, or to provide non-compatibilist Christians with a defeater for Frankfurt, he'll only convince a handful.

I'd also like to point out:

1) If one views the libertarian as claiming that PAP is *necessary* for responsibility, then it doesn't much matter if the Frankfurt examples presuppose physicalism (which they don't need to) since it's a possible counter example. I mean, how would Victor feel if an atheist wrote a paper:

"Does the Answer to the Logical Problem of Evil Presuppose Libertarianism?"

And then argued, well, since libertarianism is *false*, he didn't answer the *logical problem* of evil.

It doesn't matter that libertarianism is false for the argument to work (if it even does), and it shouldn't matter if physicalism is presupposed. The point has nothing to do with physicalism. It has to do with whether one can be responsible if they couldn't have done otherwise.

Victor Reppert said...

If a paper on the problem of evil were to argue that effective replies to the argument from evil presuppose libertarianism (which, pace the Calvinist crowd, I think they do), and that we have good reason to reject libertarianism, that would be a substantial argument that one would have to deal with.

However, I am less than satisfied with this line of defense.

Clayton said...

Anon,

Agreed. Just wanted to flesh it out a bit.

I'd like to get the technical details just right because I think it's actually a very interesting issue. From what I can tell, what we need for a Frankfurt style intervener is some agent or mechanism that can monitor an agent's deliberation in such a way that we say:
(1) In a range of choice situations {S1, S2, S3, ..., Sn} an agent's choices {c1, c2, c3, ..., cn} are choices for which the agent is properly held responsible;
(2) In at least some of these choice situations an intervener who intends to see to it that the agent does not do other than {c1, c2, c3, ..., cn} will remain idle _BUT_ had the agent intended to do other than {c1, c2, c3, ..., cn} the agent would have intervened to ensure that {c1, c2, c3, ..., cn} occurred.

From what I can tell, even if we add an element of indeterminacy _and_ deny that, say, God or some other intervener could have knowledge of the agent's choices in the strict sense I don't see why we should assume that the relevant counterfactuals would all come out false. I don't see that there's an inference from (3) to (4):
(3) If the agent had intended to do other than {c1, c2, c3, ..., cn} the agent would have intervened to ensure that {c1, c2, c3, ..., cn} occurred.

(4) The conditions necessary for knowing that the agent would respond to {S1, S2, S3, ..., Sn} by way of choices {c1, c2, c3, ..., cn} obtain.

But, if (3) could be true even if (4) is false, merely denying knowledge to the intervener will not be enough to show that the Frankfurt cases cannot be formulated. It might require tinkering to formulate it, but maybe it could be done.

So, we have two possible lines of response.

R1: The indeterminacy you get with substance dualism is compatible with divine foreknowledge and God can still serve as an intervener.
R2: That God could not know what choices someone would make would not by itself show that the relevant counterfactuals needed for formulating Frankfurt cases are false.

Anonymous said...

Clayton wrote:

“From what I can tell, what we need for a Frankfurt style intervener is some agent or mechanism that can monitor an agent's deliberation in such a way that we say:”

There is something that has always bothered me about Frankfurt cases. They basically present a situation where an “intervener” or better stated “possible intervener”, foreknows what an agent would **freely** chose to do, say mow the lawn next Saturday, if allowed to do so. I say **freely** because if the action of his mowing the lawn was say predetermined so that he could not do otherwise then the “possible intervener” and his ‘possible intervention” would not be necessary to ensure that he mows the lawn. He is predetermined to mow the lawn and so he will mow the lawn and it is IMPOSSIBLE THAT HE DO OTHERWISE. If it is impossible that he do otherwise than he definitely does not have libertarian free will. So it is not predetermined that he will mow the lawn next Saturday in order for the Frankfurt case to work.

So assume that his action of mowing the lawn next Saturday is not predetermined, that he has libertarian free will (he could choose to mow the lawn next Saturday or he could choose not to mow the lawn next Saturday). Next assume that the Intervener wants him to mow the lawn next Saturday. And how does the Intervener know that he will mow the lawn next Saturday (if the outcome of him mowing the lawn is not foreknown)? It seems to me that the Intervener must have the ability to foreknow freely chosen actions before they occur (or put another way, the Intervener must have the ability to foreknow freely chosen actions of the libertarian free will sort). So the Intervener has two possibilities then: (1) he foreknows that the agent will not choose to mow the lawn, unless the intervener intervenes to make sure the agent chooses to mow the lawn; or (2) he foreknows that the agent will choose to mow the lawn and so the intervener can allow the event to occur without any intervention on his part. These two possibilities both presume that the Intervener has the capacity of foreknowing freely chosen events before they occur. In one instance if he knows the agent would not have mowed the lawn, he has to intervene to ensure that he ends up mowing the lawn; in the other he knows the agent if not interfered with, will in fact end up mowing the lawn.

The thing that bothers me is that if it requires the Intervener intervening and ensuring that he mows the lawn, to make sure that he does not do otherwise and not mow the lawn, doesn’t that mean that LFW must exist in order for the Frankfurt case to work? I mean if the agents action of mowing the lawn next Saturday is predetermined and so it is impossible for him to do otherwise then there is absolutely no need for an intervention by the intervener. The intervention is only necessary if in fact the agent can both mow the lawn next Saturday OR DO OTHERWISE AND NOT MOW THE LAWN NEXT SATURDAY. But if the agent has the ability and opportunity to both (though not simultaneously) mow the lawn or choose not to mow the lawn, then he has LFW.

Another problem that I have is that say the agent can both mow the lawn or not mow the lawn, and that the Intervener must foreknow what he would actually do absent intervention (i.e., if no intervention then he will freely choose not to mow the lawn). So the Intervener knows that he must intervene so that the agent instead of choosing not to mow the lawn (which he foreknows that the agent would have done absent the intervention if left to freely choose on his own), will instead choose to mow the lawn. This ‘intervention”, wouldn’t it amount to coercion? I mean the Intervener foreknows that absent the intervention the agent will not choose to mow the lawn (not do what the Intervener wants to be the outcome) so he has to intervene and force him to mow the lawn. How is this not coercion?

And if it is coercion, isn’t it true that even compatibilists admit that coerced actions are not freely performed actions?

Another problem I have is that Frankfurt cases are supposed to (according to some) show that PAP does not exist. But PAP must exist if the Intervener knows/foreknows what the agent will do absent the intervention as well as what the agent will do only by intervention. PAP is predicated on there being two actual possibilities. And if the agent can both mow the lawn and not mow the lawn, then PAP is present for the agent. The Intervener stands ready to intervene (and eliminate one possibility through his intervention) if the agent had decided to not mow the lawn, but via his foreknowledge the Intervener knows that the agent will on his own mow the lawn so he will be left alone to mow the lawn (and note the agent had the ability and opportunity to do otherwise but via his foreknowledge of what he would end up doing the Intervener knows that he will mow the lawn if left alone). The possibility that he may at the “last minute” change his mind and decide not to mow the lawn is precluded by the foreknowledge of the Intervener (since he foreknows that he will in fact mow the lawn no intervention is necessary; if he had in fact changed his mind and decided to not mow the lawn then the Intervener by means of his foreknowledge would have foreseen that possibility and then known that an intervention/coercion would then be necessary.

In either case, the Intervener must have the ability to foreknow what agents will freely choose to do in any situation (sound like the Molinist concept of middle knowledge to anyone?), in order to know whether an intervention would be necessary or not.

If it is not necessary than the person freely chooses to mow the lawn (though he had the ability to not mow the lawn which would be precluded by an intervention, and in having the ability and opportunity to both mow the lawn or not mow the lawn he would have had LFW and PAP would be present). If it is necessary that an intervention occur to ensure that he mows the lawn, then the Intervener had to know that he would have done otherwise absent the intervention but the Intervener intervenes to make sure that he does not do otherwise and not mow the lawn; in which case he also had LFW and PAP was present but the Intervener through coercion eliminates the action by the agent that he does not want to happen. And we have all experienced interventions where a possibility was eliminated by the coercion of another person (the bigger brother takes away the toy from the younger brother who had previous to the intervention been happily playing with his toy, the intervention by the bigger brother coerces the younger brother into not playing with the toy that absent the intervention he would have played with). Do we claim that the younger boy freely chose to not play with the toy when the bigger brother snatched it from his hands? Do we hold the younger brother responsible for not playing with the toy after it had been taken away from him by the intervention of the stronger brother?

I find it a just a bit ironic that Frankfurt cases operate upon facts that include an Intervener with the ability to foreknow what agents will freely (in the libertarian sense) choose to do (the intervener must have the capacity for middle knowledge). And situations that include a person who has both the ability and opportunity to do otherwise with regard to a particular action (where the Intervener allows what he foreknows you will do, or intervenes and coerces you into doing what he wanted you to do if he had foreknown that you would in fact have done otherwise than what he wants to see happen absent his intervention/coercion.

And that is supposed to threaten PAP and LFW?

Robert

Robert said...

Hello Ilion,

So you are just getting acquainted with Frankfurt cases, huh? Calvinists like to appeal to them to give support to their erroneous claim that God holds us responsible even though he predetermined our every move.

“Anonymous” (who I have sneaking suspicions is someone that we already know Ilion, it sure sounds like Paul Manata to me) wrote:

“Say libertarian free will is true.

Say Judas had it.”

OK, assume that Judas had libertarian free will and that God foreknows all events. Ilion, Judas is Manata’s pet example.

“Say God wanted Judas to betray Jesus to fulfill prophecy.”

This is extremely poorly stated and improperly stated (but no surprise coming from a calvinist) as God does not **want us to sin**, though he may **allow us to sin**. So let’s restate it as: “Say God knew that Judas would betray Jesus (via his foreknowledge) and also knew that though He could intervene and prevent it, He would not prevent it from happening. God knowing what would actually occur, what the actual outcome would be, spoke through prophets prophesying these future outcomes.”

“Say that if Judas backed out at the last second, God would control his cartesian mind (or emergant mind, or whatever) through his omnipotence and "force" Judas to betray Jesus.”

In other words God via his foreknowledge knew that absent his intervention that Judas would in fact not choose to betray Christ. If that were true, then God would not have prophesied that Judas was going to betray Christ (again assuming libertarian free will, as we were asked do).

If God knew that Judas would not betray Christ unless God intervened and “controlled” his mind, and forced him into doing what God wanted, and say God knew he would in fact intervene in this way and so Judas would then betray Christ. Then because God is intervening and ***controlling*** Judas’ mind, Judas did not have libertarian free will. Something to remember is that if everything is predetermined by God as calvinists desperately wish to be true, then all of our actions are controlled, we never have a choice, do not have LFW, and of course everything goes exactly the way God wants it to go (no need for an “Intervener” then either). He would be coerced into doing the betrayal of Christ. Even calvinist Compatibilists admit that if we act when being coerced then we are not acting freely and should not be held responsible. So if God has to intervene to cause Judas to betray Christ, then this is simply a case of coercion and Judas was not acting freely nor should he be held responsible. It should also be noted that if everything is predetermined then no interventions would be necessary as the persons would be doing exactly what God wants them to do and predetermined for them to do. You don’t need to intervene when you set up and set in motion all of the dominoes that you had intricately arranged.

“But, it just so happens that he sees Judas going through with it and so doesn't need to do anything.”

So according to these words, God via his foreknowledge knows that Judas will freely choose to betray Christ and that God does not need to intervene for the outcome to be certain. It should be noted that Judas had both the ability and opportunity to do otherwise and choose not to betray Christ, but that God knew via His foreknowledge that IN FACT JUDAS WOULD CHOOSE TO BETRAY CHRIST. So in this scenario Judas had libertarian free will and **could have done otherwise**, but God knew that in fact he **would not do otherwise**.

“These three seem true:

A. Judas is responsible.

B. Judas couldn't have done otherwise.

C. Judas had an immaterial mind.”

The first proposition, A is true, Judas is responsible for his action, as it was his action, and if he freely chose to betray Christ without being coerced then he is responsible.

The third proposition is also true, if you hold to substance dualism and believe that Judas has both a body and spirit.

The second proposition as stated begs the question, Judas couldn’t have done otherwise, only if we know that libertarian free will does not or cannot exist and Judas did not have it. And if we know that God cannot foreknow freely chosen actions (i.e, if divine foreknowledge eliminates LFW). But no where has “Anonymous” shown this proposition to be true, or given any evidence of it being true. He simply assumes it to be true because he is a determinist operating according to certain assumptions (including the calvinist assumption that God cannot foreknow libertarian free will choices; that God only foreknows what He predetermines; and that divine foreknowledge and libertarian free will are incompatible) who rejects libertarian free will and presents it as a truth.

If Judas had LFW and could have done otherwise but God knew in fact that he would not do otherwise, that he would in fact betray Christ, then this proposition in claiming that he could not have done otherwise is false.

If on the other hand, God intervened and forced him, made him betray Christ when God knew he would have done otherwise and not have betrayed Christ (absent God’s intervention) then this is a case of coercion (so proposition B. would be true not because LFW does not exist but because this would be a case of coercion). If it were a case of coercion then it would say nothing about whether or not LFW exists. LFW may exist and yet be temporarily eliminated by an act of coercion (cf., normally I am quite free to walk wherever I want to walk, that is until someone ties me up and then forces me to walk somewhere).

In line with my analysis allow me to restate B. as:

B-1 = Judas could have done otherwise, but God via his foreknowledge knew that he would not do otherwise, that he would in fact freely choose to betray Christ.

So a person who is a dualist and holds to libertarian free will would state it as:

“These three seem true:

A. Judas is responsible.

B. Judas could have done otherwise, but God via his foreknowledge knew that he would not do otherwise, that he would in fact freely choose to betray Christ.

C. Judas had an immaterial mind.”

Proposition B. in the way “anonymous” first stated it, is not very persuasive to someone unless they already hold to determinism.

Robert

Bilbo said...

Robert,

Your analysis sounds pretty good, which makes me wonder why Frankfurt Counterexamples are getting so much attention.

Robert said...

Hello Bilbo,

“Your analysis sounds pretty good, which makes me wonder why Frankfurt Counterexamples are getting so much attention.”

I was only sharing **some** of the problems that I have with Frankfurt cases (there are others). The major reason they get a lot of attention is that many people assume that in order to hold a person responsible for their actions, the person must have been able to have done otherwise (had “free will” or PAP, the principle of alternative possibilities so they could choose otherwise or choose what they actually ended up doing). Frankfurt came up with creative and interesting examples showing that we can imagine situations where the person could not do otherwise (so PAP was not present with respect to a certain choice) and yet still be held responsible for their actions. Some, especially calvinists who want to believe that every event is predetermined by God, believe the Frankfurt cases may help them or support their predeterminism (i.e. God could then have predetermined everything as the calvinists imagine, people could never do otherwise and yet they could still be held responsible for what they did). I think the Frankfurt cases are thought provoking, and even at times humorous and entertaining but both the Frankfurt cases and calvinism have lots of problems.

Robert