Part 3 Alternative Possibilities, Moral Responsibility and Frankfurt Counterexamples:
(Introducing the Philosophical Motivation for Compatibilism)
1. Alternative Possibilities and Moral Responsibility
In each of the libertarian accounts considered above the agent is pictured as the source of his own action. Not only that, great care is taken in these accounts to ensure that the way in which action issues from the agent is internally nondeterministic. Specifically, the agent’s internal psychological states, reasons, desires, etc. are not sufficient conditions for the bringing about of a agent-action. Instead, the springs of action must conceived of as both internally (psychologically) and externally (circumstantially) indeterministic. And these finely tuned, intricate philosophical systems are not merely the sportful exercise of analytical powers; they are rooted in deep seated intuitions regarding the conferral moral responsibility. In particular, the intuition that seems to be driving (at least most) incompatibilist accounts is the notion that for a person to be morally responsible for their action there must be real, sufficiently robust alternative possibilities. This principle has been called the principle of alternative possibilities (PAP).
PAP: A person is morally responsible for what she has done only if she could have done otherwise.
Carl Ginet formulates this driving intuition in the following way:
It is very natural to think that a person deserves blame or credit for a certain thing’s being the case, is morally responsible for it, only if she could have made it otherwise. I deserve blame for the fact that the car is not yet repaired only if I could have made it the case that the car was repaired now. Smith deserves credit for knocking down the opposing quarterback only if he could have avoided knocking down the quarterback.
It is from these intuitions regarding responsibility that PAP is drawn up. The upshot of PAP is that moral responsibility is incompatible with determinism. This notion, in particular, seems to be the central motivation for libertarian construals of human freedom. The argument from alternative possibilities for incompatibilism (and thus libertarianism) may be framed accordingly:
(1) If PAP is valid, determinism (D) and moral responsibility (R) are incompatible
(2) PAP is valid.
(3) Therefore, D and R are in fact incompatible (compatibilism is false).
(4) Therefore, incompatibilism is true.
Lines (1), and (3)-(4) are uncontestable. If PAP is true, moral responsibility and determinism do seem incompatible. (3)-(4) follow logically from the truth of PAP. Clearly (2) is the contentious premise. The compatibilist who desires to block this argument, therefore, will have to offer counterexamples to PAP which show (2) to be incorrect. PAP makes a universal claim about the nature of alternative possibilities and moral responsibility. Hence, a single counterexample to this principle will show its falsity. Harry Frankfurt has focused his work on precisely this point. By employing Frankfurt’s work I hope to criticize the centeral premise governing most libertarian accounts of freedom (PAP) and in so doing, undermine all three varieties of libertarianism mentioned above. This will then pave the way for an alternative, compatibilist account of freedom.
2. Frankfurt Counterexamples
In his ground breaking article “Alternative Possibilities and Moral Responsibility,” Harry Frankfurt proposed a series of counterexamples to PAP which have become the center of much of the contemporary discussion revolving around free will and moral responsibility. A Frankfurt-type example may run as follows. Unbeknownst to Jones, Black, a neurological surgeon, has implanted a device in Jones brain for the purpose of detecting his volitional activity. It is Black’s intention that Jones murder Mayor Mathis. Black knows that Jones has already planed to murder the mayor and at this point he simply wants to ensure that Jones carry out his diabolical plan. If, for any reason, Jones brain activity indicates that he will deliberate against killing the mayor Black will interfere with Jones’ neuronal pathways in such a way as to prevent Jones from carrying out that deliberation. Furthermore, Frankfurt imagines, there is no way for Jones to know that his brain activity is being monitored or that his deliberations against killing Mayor Mathis will ultimately be unsuccessful. As it turns out, there is no need for Black to interfere with Jones’ neuronal activity since Jones (to Black’s satisfaction) carries out his deliberations just as he planned (i.e. he killed the mayor without the inference of Black’s device). Now it seems prima facie clear that Jones, although unable to do otherwise, should be held morally responsible for his actions. It also seems clear, therefore, that a cogent counterexample to PAP can be formulated along Frankfurtian lines. This conclusion, however, has not gone uncontested. Two strategies have generally been offered in defense of PAP: the flicker of freedom strategy and the incoherence charge.
2.1 The Flicker of Freedom Strategy
The first strategy in responding to these types of examples attempts to meet them head on arguing that while Frankfurt’s examples do not—at first glance—seem to contain alternative possibilities, upon closer scrutiny it is realized that there are at least some alternative possibilities open to the agent. For example, it may conceded that the alternative pathway to refraining from killing the mayor was shut down by Black (the counterfactual intervener i.e. an intervener which interferes only with the flow of action if the agent ‘would have done otherwise’). However, this is not to say that the alternative pathway to beginning to deliberate to kill Mayor Mathis was shut down—in fact, it is the very initiation of the deliberation process that signals Black to intervene. It is argued that these may be the factors in Frankfurt’s examples that allow us to hold Jones responsible for his actions. These glimmerings of alternative sequences in the Frankfurt scenarios are called flickers of freedom.
John Fischer identifies four versions of this strategy. Three of these strategies closely investigate the alternative sequence (those courses of action which Jones does not perform) and attempt to locate some alternative possibility which may have gone unnoticed. The fourth strategy focuses its scrutinies on the actual sequence (the course of action Jones actually performed) and asks what exactly the agent is held responsible for. Fischer contends that the argument can be weakened initially by moving the deliberation-sign (Jones’ neuronal activity) to an earlier place in the deliberative process. Now, for instance, the sign that Jones is going to refrain from murdering black is not the initiation of the deliberation process but a blush. If Jones blushes then Black will know, for whatever reason, that Jones will soon deliberate in such a way as to refrain from killing the mayor. Fischer argues that when the examples have been reformulated in this way, the alternative possibilities that remain are not sufficiently robust to confer responsibility. According to Fischer, it hard to see how blushing (an action not directly connected to the deliberative process) in the alternative sequence could confer moral responsibility for Jones murdering the mayor in the actual sequence.
2.2 Incoherence Charge
A second strategy for defending PAP against the Frankfurt attack charges the examples with begging the question. This strategy argues that the Frankfurt-type examples assume causal determinism, the very point which is under debate. Fischer acknowledges that, “Under this assumption, it is unfair and question-begging simply to assert that the relevant agent—say, Jones—is morally responsible for his behavior. But the proponent of Frankfurt-style compatibilism should not—and need not—make such an assertion at this point.” Instead, he suggests, “The argument is in two parts. The first step is to argue—based on Frankfurt-type examples—that intuitively it is plausible that alternative possibilities are irrelevant to ascriptions of moral responsibility. If one agrees with this point, the preliminary conclusion can be stated as follows: if the agent (say, Jones) is not morally responsible for his behavior, this is not in virtue of his lacking alternative possibilities.” For the Frankfurt-style compatibilist, the argument for the actual conferral of responsibility along compatibilistic lines does not come in until the second stage of the argument. It is to this second aspect of the compatibilist argument that we now turn.
To be continued in Part 4…
cf. Harry Frankfurt, “Alternative Possibilities and Moral Responsibility,” Alternative Possibilities and Moral Responsibility.” JP 66 (December 1969): 828-39. Reprinted in John Martin Fischer. Moral Responsibility. Ithaca: Cornell University press, 1986. Reprinted in ACE, 17-25. Reprinted in FW, 167-76. Reprinted in MRAP, 1-16, see pg. 1.
Carl Ginet, “In Defense of the Principle of Alternative Possibilities: Why I Don’t Find Frankfurt’s Argument Convincing,” in MRAP, 75.
Frankfurt, “Alternative Possibilities and Moral Responsibility.”
John Martin Fischer, The Metaphysics of Free: An Essay on Control, ASS Volume 14 (Malden, Mass.: Blackwell Publishers, 1995), 136-40.
Fischer, “Frankfurt-style Compatibilism,” 197.