Libertarianism, Compatibilism and Moral Responsiblity: A Philosophical Introduction to Available Models. Part 4

Part Four: Hierarchial and Semi-Compatibilism

1 Hierarchical Compatibilism

Hierarchical compatibilism is the classical style of compatibilism held by free will such theorist as Jonathan Edwards,[1] Suzan Wolf,[2] Gary Watson,[3] and Harry Frankfurt.[4] Frankfurt’s analysis has probably received in the most attention in the contemporary literature so I will focus my attention on his account. Frankfurt proposes that “one essential difference between person’s and other creatures is to be found in the structure of a person’s will.”[5] Frankfurt suggests that while humans are not alone in possessing the capacity for motives and desires, they are alone in their ability to have what he calls second-order desires.

Besides having first-order desires, desires or motives to do this or that, persons also have desires to have or not have desires. These are called second-order desires. “No animal other than man appears…to have the capacity for reflective self-evaluation that is manifested in the formation of second-order desires.”[6] A person, according to Frankfurt, has desires as to whether he wants or does not want to be moved by his first-order desires. He argues further that it is in virtue of this capacity of humans to form second order desires that they are able to have another significant feature often considered the most distinguishing mark of the human condition—free will.[7] He writes:


According to a familiar philosophical tradition, being free is fundamentally a matter of doing what one wants to do…although its focus needs to be sharpened and its formulation refined, I believe that this notion does capture at least part of what is implicit in the idea of an agent who acts freely….Analogously, then, the statement that a person enjoys freedom of the will means (also roughly) that he is free to want what wants to want. More precisely, it means that he is free to will what he wants to will, or have the will he wants. Just as the question about the freedom of an agent’s action has to do with whether it is the action he wants to perform, so the question about the freedom of his will has to do with whether it is the action he wants to perform. It is in the securing of the conformity of his will to his second order volitions, then, that a person exercises freedom of the will.[8]

It is important to note here that Frankfurt makes a distinction between acting freely and freedom of the will. One has freedom of the will if and only if (1) he has the capacity for and actually posses second-order volitions, (2) his first-order desires are not discordant with his second-order desires, and (3) his reason for having second-order desires is in virtue of his having first order desires.[9] However, for Frankfurt, a person acts freely if she did what she wanted to do when she wanted to do it and if the act issued from her own free will.


2 Semi-compatibilism

In my estimation, the most intuitively plausible and philosophically sophisticated account of compatibilism to date is that of John Martin Fischer and Mark Ravizza. Fischer calls this view semi-compatibilism. He summarizes his position as follows:


The actual-sequence reasons-responsiveness account of guidance control (and moral responsibility) thus yields “semi-compatibilism:” moral responsibility is compatible with causal determinism, even if causal determinism is incompatible with freedom to do otherwise. Compatibilism about determinism and responsibility is compatible with both compatibilism and incompatibilism (as well as agnosticism) about determinism and freedom to do otherwise.[10]

This is significantly different from many classical accounts of compatibilism which have embraced PAP and sought to show that an agent could have done otherwise if they wanted to.[11] As the basis for his semi-compatibilist model, Fischer employs an actual-sequence reasons responsive mechanism which is fully possessed by the agent (i.e. it is his own mechanism) and issues in the agent’s guidance control over his own action. Each of these concepts deserve separate consideration. I will treat them in reverse order.


2.1 Guidance Control

At the center of Fischer and Ravizza’s account[12] is the thesis that a certain type of control over one’s actions is a necessary condition for grounding moral responsibility. They suggest two types of control as likely candidates: regulative control and guidance control. Regulative control involves alternative possibilities and consists of a kind of dual power over one’s actions. Now the agent, according to Fischer and Ravizza, has regulative control if she has “the power freely to do some act A, and the power freely to do something else instead (where “doing something else” may be simply refraining from acting at all, or “doing nothing”).”[13] Given the success of the Frankfurt cases, however, it is clear that this is not the kind of control needed to ground moral responsibility. They conclude on the basis of such examples that we can safely affirm that “moral responsibility does not require the sort of control that involves the existence of genuinely open alternative possibilities.”[14] In order make this control-distinction explicit, they ask us to imagine a scenario involving a driver named Sally.


Let us suppose that Sally is driving here car. It is functioning well, and Sally wishes to make a right turn. As a result of her intention to turn right, she signals, turns the steering wheel, and carefully guides the car to the right. Further, we here assume that Sally was able to form the intention not to turn the car to the right but to turn the car to the left instead. (We are thus not making any special assumption, such as that causal determinism is true.) Also, we assume that, had she formed such an intention, she would have turned the steering wheel to the left and the car would car would have gone to the left. She controls the car, and also she has a certain sort of control over the car’s movements. Insofar as Sally actually guides the car in a certain way, we shall say that she has “guidance control.” Further, in so far as Sally has the power to guide the car in a different way, we shall say that she has “regulative control.”[15]

In the Frankfurt cases considered earlier, the prima facie assumption was that Jones was morally responsible for his actions but yet lacked regulative control over his actions. But if Jones does not have some type of control over his action, how could we hold him responsible for those actions? Fischer and Ravizza contend that while Jones does not have regulative control over his actions, he does have the type of control necessary for responsibility conferral—guidance control. I summarize Fischer and Ravizza’s two conditions for this type of control in the following way:


(GC) An agent has guidance control over an act if and only if (1) he has ownership of the mechanism which issues in the act and (2) the mechanism that issues in the act is moderately reasons-responsive.



2.2 Ownership

It seems fairly obvious that if an action does not issues from one’s own mechanism (e.g. volition, free will, etc.), that action is not one for which one may be held responsible. In the Frankfurt-cases, for example, it is clear that in the alternative sequence where Black, the counterfactual intervener, does play a role, that the Jones is not morally responsible. This is because the action issues from Black’s mechanism, not Jones’. Perhaps part of what is embodied in this concept is the common notion that an agent does not exercise guidance control over actions which are internally or externally coerced. It seems that what Fischer and Ravizza have in mind, however, is a slightly more specified idea which involves taking ownership and thus responsibility for one’s actions. By taking responsibility for one’s action an agent takes ownership of the mechanism from which they issued.[16] This act of “taking responsibility,” according to Fischer and Ravizza, is something that happens at a particular moment in one’s life in which a person take a certain stance toward their own mechanism. This stance involves acknowledging one’s self as a “fair” recipient of reactive moral attitudes. In other words, when a child is not surprised or confused when she is punished for lying to her parents but expresses remorse, this is a sign that she has taken responsibility for the mechanism which issued in the act of “lying.” She recognizes her ownership of that mechanism (although she is probably not so philosophically reflective about that ownership!). This aspect of the Fischer-Ravizza account renders it drastically different from many classical compatibilist accounts because it takes into consideration the history of agent’s taking ownership of her own mechanism.



2.3 Actual-Sequence Reasons-Responsiveness

It is not, however, sufficient that the agent have mere possession of the action-issuing mechanism. In order to be responsible, the mechanism itself must also be moderately reasons-responsive in the actual sequence. For Fischer, the alternative sequence is irrelevant since it is never the alternative sequence in which we are interested in holding people morally responsible. There doesn’t seem to be any cases in which we hold people morally responsible in the actual sequence (what they actually do) for actions they could have done in a variety of alternative sequences. While a wife may rightly hold her husband accountable for not going to the store to get the necessary items for dinner that evening, this is something which he did in the actual sequence. It would be laughable, in other words, if the husband rejoinded to the wife: ‘Yes, honey, I did not go get the groceries in the actual sequence, but you should be happy that I got them in several alternative sequences.’ We are simply not interested in ascribing moral praiseworthiness and blameworthiness to actions performed in the alternative sequences. This is why Fischer insists that as long as the action-issuing mechanism is moderately reasons-responsive in the actual sequence, an agent can be held morally responsible for those actions. But what does it mean for a mechanism to be reasons-responsive?

Reasons-responsiveness, according to Fischer and Ravizza, means that the relevant mechanism is able to weigh and respond appropriately to reasons. They suggest three ways in which a mechanism may be reasons-responsive: strong, weak, and moderate. They argue that strong reasons-responsiveness, where there is a very close fit between sufficient reasons and action, does not allow for responsibility conferral regarding irrational behavior. Say, for example, that Fred plans to go the movies on Friday evening. As it turns out, Fred is an architect who has project due on Monday mourning which will not get done if Fred chooses to go to the movies on Friday. But Molly convinces Fred to go with her any way. Here we agree that Fred is responsible for his action even though it does not issue from a strong reasons-responsive mechanism. So this kind of mechanism cannot be what is required for guidance control. But what about a weekly reasons responsive mechanism? Initially, when Fischer introduced his view, he held that this was the kind of mechanism that was needed for guidance control.[17] On this kind of mechanism there is only a loose fit between reasons and actions. In other words, there must be some possible world w1 where reasons R are sufficient for person p to do otherwise than A and in w1 p acts on the same mechanism as she acts on in the actual world and does otherwise than A because of R. So this time we imagine that Fred will go to the movies with Molly even if it means not finishing his project which is due on Monday. But if going to the movies with Molly meant, say, that Fred’s wife would make him sleep on the couch for a week (w1), then Fred would decide not to go to the movies with Molly.


But Fischer and Ravizza argue that “strong reasons-responsiveness is too strong: it is to much to ask that an agent be strongly responsive to reasons, in order to be morally responsible for his action. But it turns out that weak reasons-responsiveness is too weak; it is not enough to ask that an agent be weakly responsive to reasons, in order to be morally responsible for his actions.”[18] Instead, they insist that moderate reasons-responsiveness is just right. Moderate reasons-responsiveness has two ingredients: it consist in (1) at least weak-reasons reactivity of the actual sequence mechanism that leads to the action and (2) regular reasons-receptivity.[19] By regular reasons- reactivity, they mean the customary ability “to translate reasons into choices (and then subsequent behavior).”[20] In others words, if a third-party observed one’s action they would be able to recognize a rational, intelligible pattern in that person’s reactivity to reasons. And by receptivity to reasons, they mean “the capacity to recognize reasons that exist.”[21] To illustrate this, Fischer and Ravizza imagine a scenario in which Brown is addicted to a new drug called Plezu.[22] As the drug grows in popularity the price begins to rise. Eventually, Brown predicts, the drug will reach a price that is beyond his means (say, 1000$ per dose). So Brown recognizes that 1000$ is a sufficient reason not to take the drug because a habit that expensive would be beyond his financial means. He recognizes, furthermore, that anything over 1000$ would be too much to pay for the drug (i.e. he is moderately reasons-responsive). But what if Brown informed us that while he would not be willing to pay 1000$ for the drug, he would be willing to pay, say, 1001$ and higher (i.e. he is weakly reasons-responsive). In this case, we would wonder if he really understood the reasons to begin with and whether he could really be held responsible for not buying the drug at 1000$ per dose. We would also wonder whether such a pattern was rational at all (i.e. whether it was reactive to reasons). It would certainly be weakly responsive to reasons but it doesn’t seem that we could hold Brown responsible for such unintelligible behavior—perhaps, he is insane or mentally handicapped in some way.

In summary, then, for an agent to be responsible for his action, he must have guidance control over those actions. Having guidance control, according to Fischer and Ravizza, means that the relevant action issued from one’s own moderately reasons-responsive mechanism. For a mechanism to be moderately reasons-responsive, it must both understand the reasons which are before it and be able to translate those reasons into the appropriate responsive actions.


To be continiued…

[1] Jonathan Edwards, The Freedom of the Will: Which is Supposed to be Essential to Moral Agency, Virtue, and Vice, Reward and Punishment, Praise and Blame (London, Eng.: Thomas Nelson, 1845); Reprint (Morgan, Pa.: Sola De Gloria Publications, 1996).

[2] Suzan Wolf, Freedom Within Reason (Oxford: Oxford University Press, 1990).

[3] Gray Watson, “Free Agency,” in FW, 337-51.

[4] Harry Frankfurt, “Freedom of the Will and the Concept of a Person,” JP 68/1 (1971): 5-20; Reprinted in FW, 322-36.

[5] Frankfurt, “Freedom of the Will and the Concept of a Person,” 323.

[6] Ibid., 324.

[7] Ibid., 330.

[8] Ibid., 331.

[9] For a good summary of Frankfurt’s views as well as sympathetic, though not uncritical, interaction with and revision of them see Eleonore Stump, “Sanctification, Hardening of the Heart, and Frankfurt’s Concept of Free Will,” in PMR, 211-234.

[10] Fischer, The Metaphysics of Free Will, 180.

[11] See for example, Daniel Dennett, Elbow Room: The Varieties of Free Will Worth Wanting (Cambridge, Mass.: MIT Press, 1984), 131-52.

[12] Fischer and Ravizza, Responsibility and Control.

[13] Ibid., 31.

[14] Ibid., 30.

[15] Ibid., 31.

[16] Ibid., 241.

[17] cf. Fischer, The Metaphysics of Free Will, 164-68.

[18] Ibid., 89.

[19] Ibid., 69-89.

[20] Ibid., 69.

[21] Ibid.

[22] Ibid., 75-76.


Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: