« Questions: Their Raising and Their Begging | Main | Saturday Night at the Oldies: A Rarely Played Crystals Number »

Friday, May 29, 2009

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

I first want to say how much I’ve enjoyed this discussion of Frankfurt-style examples with Peter and Bill and all of you. I think the comments to the previous post on this topic would already fill a small volume. Peter in an exemplary effort has driven home some valid criticisms against what I’m prejudicially going to call the crude, early version of a Frankfurt-style example. I agree that where Evil Manipulator, as Peter calls him, is set up to come in and takes control of the agent who is trying to change his mind, we have coerced behaviour, not action, and no counterexample to any form of the PAP.

Rather than abandoning ship because one crude of Frankfurt-style example blatantly doesn’t work, I’ve been trying to float a more carefully crafted example that I believe may avoid the difficulties that Peter raises. The protagonist in my new & improved Frankfurt-style example is a thoroughly rotten character called Bad Bob. Bad Bob loves money, especially other people’s money, and he decides to embezzle a big chunk of money from his company. He lays a careful plan to do so. Unbeknownst to Bob, during a recent surgery, some electrodes where implanted in his brain and he is now the subject of intense study by a CIA operative we shall call the Evil Manipulator (EM). The EM discovers Bob’s plan to embezzle and is delighted. He decides to HELP Bob if necessary. Bob, he knows, is greedy and larcenous to the core, and he is already deeply committed to his embezzlement plan. The only thing that can stop Bob at this point is some last minutes fears about being caught. Bad Bob, you see, is also a bit of a coward and sometimes abandons his criminal schemes because of a last minute case of nerves. But now, courtesy of some well placed electrodes, the EM can help Bad Bob do what he wishes to do and not be stopped by some fears he can’t control. The EM monitors the parts of Bad Bob’s brain where fears will arise. At any sign of a nascent fearful arousal, he will trigger a big dopamine release from Bob’s nucleus accumbens, and Bob will feel wonderful, not anxious, about the embezzlement. The EM believes that Bad Bob will likely accomplish the embezzlement with no help from him, but just in case Bob needs a shot of “feel good” quiet-his-nerves medicine, the EM stands ready to deliver.

Before I speak to Bad Bob’s moral responsibility, I want to emphasize some set-up features of this example. First, Bob will embezzle the money, with or without help from the EM. Second, Bob will not change his mind. His motivation for committing the embezzlement is already in place and strongly driving him. The only thing that could stop is that he prone to last minutes fears. He has poor self-control, if you wish. The EM can now help Bad Bob carry through with his plans by neurological overwriting any nascent fears with feeling of joy and confidence. (He could probably do the same thing by giving Bob a stiff shot of whiskey and telling him “go for it”, but we want a surer way to help Bob.) Peter’s premise (III) is therefore false in this example. Third, the EM never “takes control of Bad Bob” or “causes” him to do anything. Either way, with or without help from the EM in suppressing his fears, Bob is doing exactly what he wants to do from reasons and motives that come right out of his rotten character. The EM implants no thoughts in Bob’s head—he can’t–and the only thing he can do is make Bad Bob feels a little happier about carrying out his plan.

I suggest that he would hold Bad Bob responsible for his embezzlement in this example, both if does it with no help, and also if the EM helps him. If the EM helps Bob, the EM is clearly an accomplice to the crime, but his helping Bob to do exactly what Bob wants to do cannot excuse or exonerate Bob. Thus we have a clear counterexample to any version of PAP that requires the ability to do otherwise as a condition of responsibilty.

The case of Bad Bob raises some interesting questions about the springs of action. Bob wants to act as he desires to act and has planned to act, but there is another part of the brain that could interfere with that because has not learned to overmaster his fears. The EM allies with Bob’s desires and the conscious planning parts of Bob’s brains to prevent fear from intruding and wrecking everything. Far from “taking control” of Bob, I would say the EM helps Bob STAY IN CONTROL.

Can you clarify if Bob is ever "in control" in the world he lives in? When he decides to embezzle, could he have decided otherwise? When his fear prompts him to back out of the embezzlement scheme, could he have resisted it and gone ahead with the plan? The Evil Manipulator's manipulations certainly seem irresistable. From a libertarian point of view that affirms the principle of alternative possibilities, Bob would be unfree and hence not morally responsible even if there were no Evil Manipulator, as long as his actions could not have been otherwise. Put differently, if Bob's fear makes it impossible for him to go ahead with the embezzlement, then from a libertarian point of view he cannot be morally praised for backing out. The libertarian will not follow you to the point of concluding that Bob is responsible for anything at all if he can at no point do otherwise than he does, whether or not the Evil Manipulator is involved.

A separate point has to do with whether, Evil Manipulator aside, Bob can be responsible for what he does under the compulsion of his own character. Clearly for the libertarian if Bob isn't free in his decision to embezzle because he is subject to physical laws, then he can't be morally responsible according to PAP. But if Bob isn't free because his vicious character compels him to embezzle, then his character, which compels him, developed either through things which could not have been otherwise (both things that happen to him and "choices" he made in his past) or through things which could have been otherwise (choices he made which could have been otherwise). If the former, then Bob can't be responsible for what he does under the compulsion of his character, since in no way could his character be other than it is, or do otherwise than compel him to embezzle. But if the latter, then Bob would bear the degree of responsibility borne by those who form their character through free choices and whose character then compels them to act in some way, even though once their character is formed they cannot but obey it for good or ill.

Phil,

You maintain that my III is false. Let us assume so. Then the following counterfactual is true:

(C) It is not the case that Bad Bob could have changed his mind and intentionally refrained from embezzling the bank at t.

Of course, the second conjunct of (C) is also needed in order to have a counterexample to PAP.
Now I have this crucial question:

Question: From which individual component of your "new and improved" Frankfurt-style example or even from any combination thereof does (C) logically follow?

Surely, (C) must somehow follow from your Frankfurt-style example. My challenge is: "Show me the money!"

It certainly does not follow from:

(i)"First, Bob will embezzle the money, with or without help from the EM."
Why? Assume that (i) is true. Still, from the fact that Bob in fact will embezzle the money a counterfactual to the effect that he could have intetnionally done otherwise cannot be deduced.

It does not follow from:

(ii)"Second, Bob will not change his mind. His motivation for committing the embezzlement is already in place and strongly driving him."
Why? For essentially the same reason as in the case of (i). From the actual fact that Bob will not, or does not, change his mind we cannot logically infer a counterfactual proposition to the effect that he could not have changed his mind. Even if we assume that the fact that he will not change his mind is supported by the most intense motivation etc., it still would not logically follow that Bob could not intentionally change his mind.

So, I cannot see any way to logically infer the counterfactual (C) from any of the elements you listed.

But, perhaps, you have another strategy in mind:

(iii) "The only thing that could stop [Bob] is that he [is] prone to last minutes fears. He has poor self-control, if you wish."
Note: I have edited in brackets some missing words in the quotation so as to make the sentence complete.

So, if I understand you correctly, your argument is based upon the following assumption:

(P) Once Bob formed his original intention, the only way he could change this intention is due to fear.

Notice that (P) has a modal component, which is why it might sanction an inference to (C). Once we have (P) in place, you seem to maintain, we rule out at once all other possible ways in which Bob could intentionally change his mind, except one --fear. And fear is then taken care of by EM.

But, first, what reason do we have to believe (P)?
--Why would it be impossible for Bob to have a change of heart?
--Why do we rule out the possibility that subsequent to Bob forming his original intention, he begins exploring his moral stand in life and comes to the conclusion that perhaps embezzling a bank is not something he wishes to do?
--Or why can't Bob come to the conclusion that being ruled by greed is being a slave to money and he decides to free himself from such slavery which in turn has the consequence of changing his intentions about embezzling the bank?
etc.

Notice, that all of these possibilities that might lead Bad Bob to change his original intentions are exactly the sort of considerations that underwrite the intuitions behind a LFW conception of moral responsibility: i.e., PAP. By ruling them out via an assumption such as (P) you basically beg the question against PAP from the get go. You simply assume a different conception of agency according to which the character (or whatever) of a person is such that once an intention flows from this character, nothing other than fear of retribution could allow him to change his intentions. But this "psychological determinism" just is antithetical to PAP; it cannot be presupposed as part of an example that is supposed to provide a counterexample to PAP.

Therefore, I maintain that the new and improved version of Frankfurt-style examples either fail to logically entail (C) or they must include a premise; namely, (P) from which we can infer (C), but such a premise simply begs the question against PAP.
Therefore, such examples also cannot be genuine counterexamples to PAP.

peter


Interesting!
I take Casey second point above to be in line with my point that Phil assumes something like (P) which begs the question against PAP.

peter

Phil,

Thanks for 'forcing' me to get clear about these matters. The case described above is supposed to be a counterexample to something, but what EXACTLY is it a counterexample to? In other words, how exactly do you formulate PAP? It is clear that a CE to one version of PAP might not be a CE to a different version. Forgive me if you have already provided a rigorous statement of PAP in an earlier thread. But I don't see one above, and so perhaps you will provide a statement in this thread.

To illustrate what I am getting at, compare Peter's formulation of PAP above with mine. My PAP*, I claim, captures what Frankfurt actually says in his seminal 1969 article.

Phil writes, >>Thus we have a clear counterexample to any version of PAP that requires the ability to do otherwise as a condition of responsibilty.<<

So Phil's PAP appears to be this: A person P is morally responsible for an action A only if (i) P did A, and (ii) P had the ability at the time he did A to do other than A (either by doing nothing or by doing an action inconsistent with A).

Phil's claim is that Bad Bob is morally responsible for having embezzled funds despite his inability to do other than embezzle funds. Hence, by Phil's lights, PAP is false.

Have I understood you, Phil?

This is similar to the Desert Flame case. Phil claimed that I lack the ability -- due to my supposedly virtuous character -- to quit the philosophy discussion and split for the strip joint, there to be inflamed by nekkid Sonoran hotties cavorting about poles. My intuition was that I lacked not the ability in question.

In the Bad Bob case, I doubt that he lacks the ability to refrain from embezzling funds. But I suppose it depends on exactly what 'ability' means. And the same goes for 'could have done otherwise.'

Note the difference: (a) I could have done other than A: I have the ability to do other than A but forces external to me forced me to do A or prevented me from doing other than A; (b) I could have done other than A: I lack the ability to do other than A.

This being tough stuff, we need to lighten up with a joke. And this being my site, I will be permitted a bad joke. "What do you call a putative counterexample to the Principle of Alternate Possibilities?" A PAP smear!

Peter,

Perhaps I misunderstand, but you seem to be objecting to the premises of my counterexample, and you can’t really do that. These are stipulations (and I think you will shortly agree they entail C). To avoid the distractions of ambiguous modal terms, I have been using “will”, but perhaps I need to put it this way: Bad Bob MUST embezzle. Bad Bob CANNOT change his mind. GIVEN that Bad Bob must embezzle and cannot change his mind, I claim we still hold him responsible because the EV only helps him controls himself and accomplish what he wants to accomplish. Your substantive point is that if Bob cannot change his mind, he is not in control of himself and is being compelled to BEHAVE in a certain way. I believe I’ve made a very credible case that Bad Bob is very much in control of himself and ACTING exactly as he wants to act, with a little help from the EV in controlling himself. Bad Bob is guilty despite any help he may or actually does receive from the EV. So Bad Bob is a counterexample to at least the standard version of the PAP.

Your objection to P raises an interesting issue about the springs of action (which is one of my main interests in Frankfurt-style counterexamples). I could just say P is a stipulation of the example, without, I think, offending anyone’s sense of psychological reality. Bad Bob’s “mission in life” is to scam people and steal money, that’s what drives him, and nothing else can possibly stop him except his occasional incontinent bout of fearfulness. But let’s grant the version of Bob you want and have him subject to other psychological weaknesses that impair his criminal will. It doesn’t matter. The EV is able to help him overcome any random aversive reaction that pops up and threaten to compromise Bob’s plan of embezzlement. The EV stands ready and able to immediately help Bob overcome any nascent aversive reaction that he begins to experience. In other words, Bob will never have any emotional motivation to do anything other than the thing that he wants to do most in his life (steal money). Bob CANNOT change his mind, yet we hold him responsible. The EV is a steadfast ally of Bob’s conscious will and desire to steal, but, I repeat, Bob only ACTS as he really wants to ACT.
.


Bill,

You are well within your rights to demand a clear & canonical formulation of the PAP. I count at least five non-trivial variants that have been invoked in this discussion, but I doubt the Mavericks have an appetite for this kind of technical taxonomy. A dissertation could be won just sorting these varieties out. May I suggest we instead look at the general motivation for propounding a PAP. The intuition is that our responsibility for a deed is contingent upon our being to do something else than this deed. Or perhaps, at least try to do something different. The view I’m suggesting is that this intuition does not stand up to a series of counterexamples including Frankfurt-style counterexamples.

Strawson made the point throughout the 60’s and 70’s that our actual moral practices cannot be reduced to a set of (reforming?) rules like the PAP. His main point was that these rules render judgmenst of moral responsibility undecidable in a great number of important (non-fictional) cases, especially those that involve horrific criminal wrong-doing. As neuroscience makes advances in studying the deviant brains of psychopaths, we are less and less sure of ourselves on issues of what someone could or couldn’t do. We really don’t know whether people like John Wayne Gacy could have done otherwise, given their neuro-pathology. Strawson’s point is that, thank God, both common sense and the law don’t require us to puzzle out whether Gacy could have acted differently. A jury that had to answer this question probably would have hung and let Gacy walk. A good result, do you think? The criminal standard in the US is that the agent knew what he was doing and knew it was wrong. That was very decidable in the Gacy and similar cases. I think many of us absorb this legal standard in our ordinary judgments of responsibility. I for one have no interest in trying to puzzle out whether someone’s neurology made him unable or virtually unable to do otherwise. I find the PAP totally impracticable and unappealing.

Frankfurt counterexamples point to another problem with PAP. We are really interested in what the agent actually planned to do and did, when this was fully consonant with his character. Whether he received some dopamine or alcohol to help him persevere in his evil plans, we don’t really care. Help excuses nothing.

Phil,

>>You are well within your rights to demand a clear & canonical formulation of the PAP. I count at least five non-trivial variants that have been invoked in this discussion, but I doubt the Mavericks have an appetite for this kind of technical taxonomy.<<

I find this remark puzzling. It sounds as if you are begging off a serious discussion. Don't worry about being technical, without technicality we won't make any progress. PAP is more a principle-schema than a principle. If you have five variants in mind, let us hear which of the five you think your Bad Bob case refutes.

The fact that I post on a wide variety of topics in different ways does not mean that I shy away from technical discussions. That's exactly what I want if I can find the people ready, willing and able to engage in them.

Phil, Bill

Phil, let me address a couple of issues your recent exchange with Bill raised. First, regarding the formulation of PAP, there is Frankfurt’s original version of PAP which Bill cited. It is inadequate because it is ambiguous between action and behavior. Bill recently proposed the following formulation, where ‘S’ is a person; ‘X’ is an action or omission; and ‘t’ a time:

PAP*: S is morally responsible for intentionally doing X at t only if (i) S intentionally does X at time t, and (ii) S could have intentionally refrained from doing X at t.

I accept this formulation. Remember: intentionally doing something is equivalent in meaning to an action. PAP* involves the following constituent propositions:
(A) S is morally responsible for intentionally doing X at t;
(B) S intentionally does X at time t;
(C) S could have intentionally refrained from doing X at t.

The question is this: What is the logical form of a legitimate counterexample to PAP*? I think everyone agrees that a legitimate CE-set is the following: {A, B, ~C}. For the purposes of evaluating the legitimacy of Frankfurt-style examples, we may grant that (A) and (B) hold. (B) holds because according to Frankfurt-style examples S does in fact intentionally do X at t. (A) holds because in such examples we do consider S morally responsible for doing X at t. The contested proposition, then, is (C). The proponents of PAP* maintain that in such examples, when (A) and (B) hold, so does (C). Therefore, no counterexample has been produced. Those who view Frankfurt-style examples as legitimate counterexamples to PAP* maintain that (~C) holds. The burden of proof is on those who claim that Frankfurt-style examples provide a counterexample to PAP*. Therefore, their task is to demonstrate that from (A), (B), the details of a particular Frankfurt-style example, and any other non-contested and non-question begging principles they can logically derive (~C). I have asked Phil in a previous post to demonstrate this fact. Let us state explicitly what needs to be derived:
(~C) It is not the case that S could have intentionally refrained from doing X at t.
Here is my question once again: How does (~C) logically follow from any Frankfurt-style example? Phil seems to answer this question as follows:

“These are stipulations (and I think you will shortly agree they entail C). To avoid the distractions of ambiguous modal terms, I have been using “will”, but perhaps I need to put it this way: Bad Bob MUST embezzle. Bad Bob CANNOT change his mind. GIVEN that Bad Bob must embezzle and cannot change his mind, I claim we still hold him responsible because the EV only helps him controls himself and accomplish what he wants to accomplish.”

But, what does this response really mean? It seems to be saying that we simply stipulate that Bad Bob *must* embezzle; he *cannot* “change his mind. But, these stipulations are equivalent to (~C). They do not logically follow from the premises of a Frankfurt-style example: they are presupposed by it. If so, then your version of Frankfurt-style examples is not a legitimate counterexample to PAP*: it is not even a counter-example. It simply stipulates that PAP* is false because it maintains outright that (~C) holds. But since we already granted that (A) and (B) hold, it follows that Phil’s argument is simply that the CE-set = {A, B, ~C} all hold by mere stipulation. I do not think that this sort of reasoning is going to convince anyone that Phil has provided a legitimate counterexample to PAP*.

Note: Bill the following is my revised formulation of the argument you graciously presented in this post.

Let me construct what I believe to be the argument which the proponents of Frankfurt-style examples have in mind as to how to derive (~C) from such examples. I will assume Phi’s Bad Bob example.

Argument for (~C).
The proponents of Frankfurt-style examples have the following principle or *analytical postulate* in mind:
(i) Analytical Postulate I: for each time t, if x can prevent y from intentionally doing X at t, then it is not the case that y could have done X at t.
An instance of this analytical postulate is the following:
(ii) For each time t, if EM can prevent Bad Bob from intentionally refraining from embezzling the bank at t, then it is not the case that Bad Bob could have intentionally refrained from embezzling the bank at t.
(iii) For each time t, EM can prevent Bad Bob from intentionally refraining from embezzling the bank at t. [details of the Frankfurt-style example]
Therefore,
(iv) For each time t, it is not the case that Bad Bob could have intentionally refrained from embezzling the bank at t. [from (ii) and (iii)]
Therefore,
(v) For every time t, it is not the case that Bad Bob could have intentionally refrained from embezzling the bank at t. [from (iv)]
Since (v) is identical to (~C) except for the temporal quantifier, we can take the above proof as a proof of (~C).
Therefore,
(vi) ~C.

There are several major problems with the above proof:

Problem (a): Let us focus upon the step above that goes from (iv) to (v). This step is not a valid step. From the fact that something holds at *each* time, it simply does not follow that it holds *every* time. Consider the following example: while it may be true for a woman that at each moment of time prior to t she can be impregnated, it does not follow from this that for every moment prior to t she can be impregnated. This is so for the obvious reason that if she is impregnated at a given time prior to t, she cannot be impregnated again between that time and t.
Now, someone might wonder whether an inference from *each* to *every* is required for the purpose of proving (~C): i.e., whether the inference from (iv) to (v) is really indispensable. Why not drop (v) and go directly from (iv) to (~C)? The trouble is that (iv) is not equivalent to (~C). In order to see this point, you need to realize that PAP* is going to be true provided Bad Bob could have decided to refrain from embezzling the bank at any point in time between forming his original intention and t. In order to render (C) true, and thus (~C) false, all we need is one instance were Bad Bob takes such a decision. Therefore, in order to establish (~C), we must rule this possibility out at every single moment from the time Bad Bob formed his intention to embezzle the bank and the time he actually did it. That is, one must prove that at *no time* Bad Bob could have decided to intentionally refrain from embezzling the bank at t. But this later proposition is equivalent to saying that at every moment t, it is not the case that Bad Bob could have decided to refrain from embezzling the bank at t. Hence, the inference from (iv) to (v) is required in order to prove (~C). But this inference is not valid.

Problem (b): The Analytical Postulate I used in the above argument is incorrect. It simply does not follow that y could not have done X at t from the mere fact that x can prevent y from doing X at t, unless x actually prevents y from doing X at t. From the mere fact that I can prevent a robber from holding up a grocery store it does not follow that the robber cannot hold up the grocery store, for I might decide not to do what I can, in which case the robber succeeds to hold up the grocery store. The correct Analytical Postulate regarding ‘prevent’ is this:

Analytical Postulate I*: for each time t, if x can prevent y from intentionally doing X at t and actually does so, then it is not the case that y could have done X at t.

Notice that this analytical postulate is true even if one of the conjuncts in the antecedent is false. Particularly it is going to be true even if x fails to actually prevent y from intentionally doing X at t. However, under such circumstances, one cannot construct a *sound* argument that detaches the consequent, because one will have to include a false premise in order to do so. And this is precisely the case in Frankfurt-style examples. Since EM never actually prevents Bad Bob from intentionally refraining from embezzling the bank at t, a sound argument cannot be constructed which features the conclusion that it is not the case that Bad Bob could have intentionally refrained from embezzling the bank at t.

Problem (c): Premise (iii) above is false. It is simply not the case that EM can prevent Bad Bob from intentionally refraining from embezzling the bank at t. At each time in between the time Bad Bob formed his intention to embezzle the bank and the time he actually does it, EM can cause Bad Bob to behave in a manner that embezzles the bank at t; therefore, at each such time, EM can prevent Bad Bob from behaving in a manner that refrains from embezzling the bank at t. Nevertheless, at each moment Bad Bob can form a new intention to refrain from embezzling the bank at t. While EM could then cause him to behave in a manner that embezzles the bank anyway, Bad Bob is no longer an agent with respect to this behavior and, therefore, is not responsible for behaving in the manner stated. The following substitute for (iii) is true:

(iii*) For each time t, EM can prevent Bad Bob from behaving in a manner that refrains from embezzling the bank at t.

The problem is that if we substitute (iii*) for (iii) in the above Argument-for-(~C), we are not going to get a conclusion that is identical to (~C). Instead we are going to get the following conclusion:

(~D) It is not the case that S could have behaved in a manner that refrains from doing X at t.

But, (~D) is not the negation of (C). Therefore, by replacing premise (iii) with (iii*), we do not get a case that is a legitimate counterexample to PAP*. It is important to remember this fact.

Let us suppose that we waive all of the above difficulties with the Argument-for-(~C). Let us suppose that there is some argument which validly entails (~C). It is imperative to realize that any such argument, if there is one, will have to include at least premise (iii); without this premise, no argument based upon a Frankfurt-style example has a chance of deriving (~C). I hope this point is clear. Since such an argument must include (iii), I will present an alternative argument also employing (iii) which validly entails (C):

Argument-for-(C).
I have an analytical postulate of my own pertaining to ‘prevent’:

(i)’ Analytical Postulate II: For each time t, if y is in a state of mind capable of forming intentions and x can prevent y from intentionally refraining from doing X at t, but x does not do so, then y could have intentionally refrained from doing X at t.
(ii)’ For each time t, if Bad Bob is in a state of mind capable of forming intentions and EM can prevent Bad Bob from intentionally refraining from embezzling the bank at t, but does not do so, then Bad Bob could have intentionally refrained from embezzling the bank at t. [instance of (i)’]
(iii) For each time t, EM can prevent Bad Bob from intentionally refraining from embezzling the bank at t. [facts of the Frankfurt-style example]
(iv)’ For each time t, Bad Bob is in a state of mind capable of forming intentions. [facts of the case].
(v)’ For each time t, EM does not prevent Bad Bob from forming the intention to refrain from embezzling the bank at t. [facts of the case]
Therefore,
(vi)’ For each time t, Bad Bob could have intentionally refrained from embezzling the bank at t.
[from (ii)’, (iii), (iv)’, and (v)’]
(v)’ Conclusion (vi)’ is equivalent to (C).
Therefore,
(vii)’ C.

The above Argument-for-C is valid. So if there is a valid Argument-for-(~C) involving premise (iii) based upon a Frankfurt-style example, then there is a corresponding valid Argument-for-(C) involving premise (iii) that is based upon the very same facts of a Frankfurt-style example. But, clearly the two arguments entail a contradiction. Therefore, they cannot both be sound: some of the premises must be false. I suggest that premise (iii) is the false premise. However, as noted above, without premise (iii) there cannot be an Argument-for-(~C). Therefore, without premise (iii) there cannot be a Frankfurt-style counterexample to PAP*. Hence,

(P1) If a Frankfurt-style example entails (~C) and satisfies this requirement to be a counterexample to PAP*, then it is inconsistent.

(P2) If a Frankfurt-style example is consistent, then it cannot include premise (iii); it cannot entail (~C) and, therefore, it is not a legitimate counterexample to PAP*.

Or else we can simply assert:
(P3) Either a Frankfurt-style example is inconsistent or it is not a counterexample to PAP*.

Therefore, no Frankfurt-style example can be a consistent counterexample to PAP*. This is what I aimed to prove in the post above.

peter


Peter,

Thanks for that excellent presentation. There is a lot to respond to. I agree with you that 'doing' is ambiguous as between 'behaving' and 'acting.' And I agree that actions are intentional by definition. I would add this: there are mental actions and physical actions. Intentions are mental actions. Actions which are not intentions are 'animated' by intentions.

You also make the excellent point that the burden of proof is on the denier of PAP. I could explain why, but I don't think I need to inasmuch as we will all agree, including Phil. Or at least I hope so.

But there is a distinction that may need to be drawn more sharply, and that is between Lockean examples and Frankfurt examples. In a Lockean case, someone is (unbeknownst to him) prevented from doing something, while in a Frankfurt case, someone is assisted in doing something. Correspondingly there are two PAPs:

PAP 1: A person is morally responsible for having done A only if he has done A and had at the time he did A the ability to refrain from doing A.

PAP 2: A person is morally responsible for failing to do A only if he has failed to do A and at the time of the failure he had the ability to do A.

To expand upon what I just wrote:

>>The proponents of Frankfurt-style examples have the following principle or *analytical postulate* in mind:
(i) Analytical Postulate I: for each time t, if x can prevent y from intentionally doing X at t, then it is not the case that y could have done X at t.<<

Phil might object: what does this have to do with my Bad Bob case? In the Bad Bob case, the agent is not prevented from doing something but rather assisted in doing something.

Bill,

Thanks. As I noted in the post, I consider omissions as a species of actions; thus a failure to do something is an omission which is itself an action. So PAP1 and PAP2 collapse under one heading of actions/omissions. Of course, if for some principled reason we are inclined to distinguish between actions and omissions, then we will have to make the appropriate distinctions along the lines you propose. The other question is regarding your current formulation of PAP in terms of ability vs. could have done otherwise.
As for Analytical Postulate I. So far as I understand, an essential feature of Frankfurt-style examples is that the control mechanism *prevents* the agent from acting or behaving otherwise. That is the reason Frankfurt-style examples are alleged to refute the relevant clause of PAP. There is no need to *assist* the agent, as opposed to prevent an alternative course of action, because in such cases the agent already formed the intention to do such-and-such and never really alters this intention. The point of such examples, then, is to rule out an alternative course of action by means of some control mechanism that functions as a preventive causal agent that is on stand-by just in case. If this is the correct understanding of Frankfurt-style examples, then Analytical Postulate I, or some version thereof, is essential for Frankfurt-style examples.

peter

In the early sixties my favorite TV series was I DREAM OF GENIE. If Bill will post a picture of the gorgeous Barbara Eden, you will understand why immediately. Barbara played an out-of-the –bottle genie always trying to do what’s best for her master, poor Captain Nelson. An interesting sub-plot in the series invented an evil black-haired twin sister, who would occasionally show up, masquerade as Genie, and wreak havoc in Captain Nelson’s life. (Barbara of course played both roles. I fancied her more in the black wig, but here I digress.)

Imagine that you have acquired an Evil Genie who shows up whenever you plan to do something bad. Instead of trying to save you from yourself like a good Genie would, Evil Genie takes it as her mission to see that you succeed in your dastardly plot. She has magical powers to remove all obstacles in your path, and she can also sense your thoughts, so that if she even suspects you might be about to get cold feet, she immediately gives the cortical equivalent of a stiff drink. You feel joy and confidence instead of anxiety. With all obstacles removed, within and without, you cannot fail in doing the dastardly deed.

This is another version of Bad Bob. What this version may make a little clearer is that we have two people, or one person and a genie, acting in concert. Evil Genie stands ready to facilitate, and if necessary, actually facilitates your plan to do something bad. Normally we have no difficulty in holding an instigator and a facilitator jointly responsible for what they bring about. What’s abnormal in this case is that the facilitator has super-human powers to ensure the instigator succeeds. This matters, but it does not matter enough for us to give the instigator—and he is the one who actually does the evil deed--- a pass. “An Evil Genie helped me do this” is not an excuse we will accept even if it’s true!

On a deeper level, what this sort of Frankfurt-style story is trying to do is get us to reject the excuse or defense of “I could not have done otherwise.” (cf “I was just following orders.”) What these stories argue is that it doesn’t matter if you could not have done otherwise, if in fact you did exactly what you wanted to do. You were not coerced in any way, you were only assisted if necessary. You knew what you doing, you knew it was wrong, you didn’t know you had a secret helper, but wouldn’t you also have wished for Evil Genie to help you if you knew she was available?

I think we should call the “I could not have done otherwise” defense the Psychopath’s Defense, because this is where it has been tried in earnest. The Psychopath’s Defense is not a claim of coercion or duress, but the claim that brain pathology makes monsters like JWC (and his many lesser imitators) not responsible for their horrific deeds. If we make it a condition of responsibility that Gacy could have resisted torturing and killing boys who fell into his hands, and in particular, if we must be certain beyond a reasonable doubt that he could have resisted his vile impulses, then we cannot find him guilty. Is that an acceptable outcome? I don’t know and I don’t care whether Gacy could have resisted his impulses. I hold him responsible for becoming the monster he was and for the deeds that flowed from his bestial nature. Gacy didn’t need a Evil Genie helping him, but if someone says his pathological brain was the near equivalent of an Evil Genie, I reject this excuse utterly.

I see you are up early, Phil. I must be having a good influence on you.

I agree with Peter that none of your examples count as genuine counterexamples to PAP. In effect what you do is beg the question against it. You say: here is a case (the Gacy case for example) in which an agent is morally responsible for action A, and yet the agent was not able to do otherwise (due to brain malfunction, demonic influence, whatever.) My counterresponse is that your agent is not morally responsible, and that therefore the CE is not genuine.

I think the problem here is that you are not really interested in the freedom of the will, but in the rationale for incarceration and execution. You want to be able to justify the execution of a Ted Bundy. Well, one could justify it on grounds other than the ground that he is morally responsible for his deeds. You could say this: he is an utterly worthless individual and a grave danger to others and we need to protect ourselves; incarceration is not sufficent protection in the case of someone as clever as he is (especially if he is under demonic influence, a possibility not easily excluded!) and so execution is justified. There are two sorts of jusdgment we should keep separate:

AXIOLOGICAL JUDGMENT: So-and-so is worthless, has a thoroughly rotten character, etc.
DEONTOLOGICAL JUDGMENT: So-and-so is morally blameworthy.

The first can be true while the second is false. The first could give you what you need as far as a rationale for incarceration/execution goes.

Good morning Bill,

I don’t see that the Frankfurt-style Bad Bob examples beg any questions. You can of course disagree with the judgment that Bad Bob is morally responsible for his dastardly deeds, because the Evil Genie is there to ensure he doesn’t fail. I don’t see how you can defend that position consistent with the view that both instigator and facilitator are jointly responsible for their malefactions, but maybe you can try some special pleading. I think the view that the presence of the Evil Genie makes Bad Bob not an agent or “actor” is dead wrong. I was tempted to trot out something like a Davidsonian analysis of action to set all this out in detail, but that’s a long road and I just don’t see ANY reason to believe bad Bob only behaves and doesn’t act. He does, after, exactly what he wants and plans to do, and he acts precisely as his character and present desires direct him. I can say no more there.

Gacy is different. Gacy is like a Frankfurt-style example in some ways, but not the same. I premise that we assume Gacy is Aristotle’s bestial man, his moral faculties incurable corrupted by all sorts of factors. All of this is reflected in the “deviant” condition of the relevant parts of his brain. Gacy is nowise cognitively impaired—to the contrary, he is a very clever boy—and he knows what he is doing is evil, but he is irresistibly drawn to it and really enjoys torturing & killing. "It more than a hobby for him." I conclude that we ought to hold Gacy responsible for his actions regardless of his inability to restrain himself. I accept that you do not share this opinion about Gacy. How can I persuade you here beyond re-iterating that our criminal law also rejects Gacy’s “I couldn’t do otherwise” defense. The law rejects it as irrelevant to Gacy’s guilt or innocence. The law wisely rejects it, as Strawson tells us, because it is for the most part undecidable beyond a reasonable doubt. You can say that you disagree, that in your opinion it should excuses him, and that’s fine. You can believe as you wish to believe here, just as long we keep you off capital juries.

You offer me the option of condemning Gacy & Bundy and friends on “axiological” grounds: they are worthless pieces of human trash and danger to us all and need to be flushed. I appreciate the offer but I can’t take it. Justice forbids it. If we are to condemn Gacy, as I think we must, justice demands we condemn him for evil deeds that he has been tried and found guilty of. If we find that he didn’t know what he was doing, was coerced, etc, we cannot convict him of a capital offense. To be sure, we may need to permanently instutionalize him for the general good, but that is different. To condemn him we must first find him responsible for what he has done.
.

I seem to recall, Bill, that the Soviets had capital charge that they used to go after first class good-for-nothings. These people were charged with being “wreckers”. Wreckers were people habitually engaged in worthless or prohibited activities that were deemed harmful to the public good. No particular capital offense had to be proved for wreckers, but only that they were chronically up to no good. Wreckers also did not enjoy a right of appeal of their death sentence, and were usually taken out and shot pretty soon after there trial. The Chinese, I’m told, still have and vigorous apply a similar statute. This is why your axiological approach scares me. Here we see it at work in two legal codes, and the result both frightens me and offends my sense of justice. If you want to execute some bad guy, at least give him the right to trial and the due process of being found guilty (responsible) of a capital offense.

Phil writes, >>I don’t see that the Frankfurt-style Bad Bob examples beg any questions. You can of course disagree with the judgment that Bad Bob is morally responsible for his dastardly deeds, because the Evil Genie is there to ensure he doesn’t fail. I don’t see how you can defend that position consistent with the view that both instigator and facilitator are jointly responsible for their malefactions, but maybe you can try some special pleading.<<

Under dialectical pressure from Peter, you modified your Bad Bob example to the point where he HAD TO embezzle the funds. Your view is that the malefactor's rotten character deterministically drives his behavior so that, given an opportunity to embezzle, he does so. You think this a CE to PAP because Bob is both morally responsible and not able to do otherwise. But surely this does not refute PAP. If you stipulate that Bob is a deterministic system, then (given that you are an incompatibilist) of course Bob is not morally responsible. But you need an example in which the agent is both morally responsible and not able to do othewise.

Peter: Do you agree with what I just wrote?

Phil,

I think we have here a severe case of disagreement on what is the logic of counter-examples. You have provided several variations on the Bad Bob case and according to my understanding of counterexamples, none of them qualifies as a legitimate counterexample to PAP*. So let me focus exclusively on this point.

A counterexample to any version of PAP must show that the consequent of PAP is false, while its antecedent is true. Since we all agree that Bab Bob intentionally embezzled the bank and we are willing to concede that he is morally responsible, your burden of proof is to show that in any of these cases the following counterfactual is false:

(C) Bad Bob could have intentionally refrained from embezzling the bank at t.

1) It is important to see that the fact that Bab Bob did in fact intentionally embezzle the bank is already GRANTED: this fact has nothing, absolutely nothing, to do with whether he could have intentionally refrained from doing so.

2) You cannot simply *stipulate* that Bad Bob MUST intentionally embezzle the bank, because you cannot stipulate that (C) is false: you must demonstrate it based upon the example you give.

3) In your original Bad Bob case where EM implanted a chip in Bad Bob's head, you have only showed that the chip would cause Bad Bob to behave in a manner that embezzles the bank in the event Bad Bob changes his mind. This at most would show that Bad Bob could not have behaved otherwise; not that he could not have acted otherwise.

4) All the rest of the examples are cases where some external entity helps, instigates, facilitates, or whatever Bad Bob's decision to embezzle the bank and his actions to do so. Here is how you put it:

"Imagine that you have acquired an Evil Genie who shows up whenever you plan to do something bad. Instead of trying to save you from yourself like a good Genie would, Evil Genie takes it as her mission to see that you succeed in your dastardly plot. She has magical powers to remove all obstacles in your path, and she can also sense your thoughts, so that if she even suspects you might be about to get cold feet, she immediately gives the cortical equivalent of a stiff drink. You feel joy and confidence instead of anxiety. With all obstacles removed, within and without, you cannot fail in doing the dastardly deed."

How does any of this prove that in any circumstance you envision Bad Bob could not have done otherwise, as long as he can form intentions? There is nothing here that proves THIS counterfactual. As long as Bad Bob can form intentions of his own and the Genie, EM, or whatever does not control this faculty, if he is responsible, then he could have on his own form the intention to do otherwise.

5) So here is how I would try once again to get you to answer my original question about (C).

(i) Either Bad Boy can form intentions on his own in between the time he formed the intention to embezzle the bank and the time he actually did it or he cannot.

(ii) If it is the former, then he could have intentionally decided to refrain from embezzling the bank prior of doing so. And if he could have intentionally decided to refrain from embezzling the bank, then there is no counterexample, EM, Genie, and all.

(iii) If Bad Bob cannot form intentions on his own after he formed the intention to embezzle the bank but before doing so, either because of EM or Genie, or what not, then he is not longer an agent with respect to the behavior of embezzling the bank. And for this reason, under such circumstances, he is not morally responsible.

I do not see a third possibility. You have not proved there is one. But it is this third possibility that you need.

6) "You can of course disagree with the judgment that Bad Bob is morally responsible for his dastardly deeds, because the Evil Genie is there to ensure he doesn’t fail."

For the purpose of these examples we have already granted that Bad Bob is morally responsible and that he embezzled the bank. The question is whether he could have done otherwise. PAP entails that he could have. You argue that he could not. But you have not provided one convincing case where Bad Bob can form intentions at the relevant time period, but he cannot form the intention to refrain from embezzling the bank.

peter

Bill,

Yes! I certainly agree.

peter

Peter,

Thank you for persisting. Forgive me, but I just don't understand much of your argument. You like to use some very treacherous philo terms such as “intention” and “intentionally refraining.” Instinctively, I want to avoid them utterly.

For example, I have no idea why you would think Bad Bob, who does exactly (and intentionally) what he wants to do according to his own plans, is a not an agent. Do you have some special theory of human agency you want to deploy here? You seem to want to insist that a human being must be able to change his mind every minute about something he had decided to do or he isn’t a human agent. No, that's obviously wrong. Human agency is not about being able to change your mind every minute. On the contrary, it is about being able to persist in a decison and actually accomplish something. A Woody Allen character is not my paradigm of an effective human agent.

Human psychology sets real limts to how I can change my mind about what I've set out to do. Once I have become fixed on a something I want (e.g., a woman or a job or a pay-off), it is very hard for me to abandon this goal. Something dramatically bad has to happen to me to change my mind. If an Evil Genie guarantees that nothing bad will happen to me, there will be no motivation to change my mind. I cannot change my mind without motivation to change my mind. I will persevere in doing what I really want to do, unthwarted or unobstructed by irrelevant fears that in fact I do not want to stop me. Evil Genie helps me accomplish what I want to accomplish. In this way, Evil Genie actually enhances my ability to act and do what I want to do.

In a way, Bob becomes a kind of paradigm or ideal of human agency, able to do what he wants to do, and, thanks to the Evil Genie, never subject to the akratic breakdown that our weak self-control makes us chronically prone to. What the Evil Genie really gives Bad Bob is reliable self-control. This enhances not reduces human agency.

Peter & Bill,

Why C is false:

Bad Bob is determined to embezzle $10 M USD and is at the point of executing this plan for doing so.
Bob really wants to carry off his big-time scam, but he knows he is sometimes prone to last minute cold feet.
Bob wishes (prays) for the fortitude to persist, and Evil Genie answers his wishes, being able to sense and respond to any nascent fearful reactions as Bob proceeds with his plan. Evil Genie will calm any of Bob’s fears.

These are the set-up stipulations of the counterexample.

Bob cannot change his mind in these circumstances. There is NOTHING that can possibly motivate him to change his mind and abandon what he really wants to do for some worthless unmotivated alternative. Without motivation to do something else, Bob can’t choose to do something else. Evil Genie helps Bob to stay focused on the thing he want to do by acting like his innate faculty of self-control and calming any fears that might threaten his intended actions. Evil Genie gives Bob reliable "self-control". Bob is therefore able persist with his plan and execute it.

Bad Bob cannot change his mind and chicken out because Evil Genie guarantees he will not chicken out. Yet we hold Bad Bob responsible for what he has done whether or not Evil Genie’s actually helps. This is clear counterexample to the PAP.

Phil,

Suppose, unbeknown to Bob, the bank he intends to rob is owned by the mob. They get wind of his plans and deliver the usual dire threats against his loved ones should he go ahead with his scheme. But Bob is still under the influence of Evil Genie. What happens?

Hi David,

The best way to handle "external" obstacles to Bad Bob succeeding, things like the mob intervening, is to let Evil Genie handle them---just as Genie did in the TV series (that you are probably too young to remember). Evil Genie, we stipulate, has to power to protect Bob from people that want to stop or thwart him (if any try).

I was hoping to hear some comments on the fact that the law declines to link criminal responsibility with any finding that the perpetrator “could have done otherwise.” Strawson’s point is that making any kind of judgment of responsibility subject to being convinced that the doer could have done otherwise likely renders many, if not most, such judgments undecidable. This seems to be the thinking underlying the rejection of any sort of PAP condition in the law. As things stand, we really can’t be sure whether psychopaths like Gacy & Bundy could have overcome their brain pathology and acted otherwise. Fortunately, we don’t need to concern ourselves with what these evil men could have done. We judge them on the basis of what they did and intended to do. I share with the law the intuition that a PAP condition doesn’t belong with our judgments of responsibility, criminal or moral. But many philosophers disagree—for reasons I’d like to understand better.

The comments to this entry are closed.

My Photo
Blog powered by Typepad
Member since 10/2008

Categories

Categories

August 2020

Sun Mon Tue Wed Thu Fri Sat
            1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31          
Blog powered by Typepad