Book Review Draft

Four View on Free Will, by John Martin Fischer, Robert Kane, Derk Pereboom and Manuel Vargas.  Blackwell, 2007. 232 pages. hardback $74.95/ paperback $31.95.

Kevin Timpe, University of San Diego

             Four Views on Free Will is an excellent addition to Blackwell's Great Debates in Philosophy series.  According to the publisher, the volumes in this series attempt to "capture the flavour of philosophical argument and to convey the excitement generated by the exchange of ideas. Each author contributes a major, original essay stating his or her own position. When these essays have been answered, the authors are each given the opportunity to respond to the opposing view."  Because of this format, the volume is composed of eight chapters.  In the first four chapters, each author elaborates and defends a central and influential position in the contemporary debates about free will and moral responsibility:  Kane on libertarianism, Fischer on compatibilism (or, more accurately, semi-compatibilism), Pereboom on hard incompatibilism, and Vargas on revisionism.  The last four chapters, which are the authors' responses to the other three initial essays, help clarify and expand their initial presentations.  The first three of the four positions are defended in greater detail elsewhere, as each of their proponents have monographs devoted to more complete expositions of their views;[1] the present volume could also be used as a jumping off point for these more thorough (though earlier) treatments.  While revisionism is not as prominent a position as are the other three, its inclusion is beneficial in that Vargas's contributions not only help sharpen the disagreements between the other three positions but also offer much in their own right.  Many of the arguments and maneuvers will be familiar to those already versed in the existing literature.  But the present volume does more than merely represent earlier work, as all of the authors also venture into new territory (though some more than others) and respond to recent objections.

            Elsewhere, Kane distinguishes four questions that are at the center of the debates about free will:

The Compatibility Question: Is free will compatible with determinism?

The Significance Question:  Why do we, or should we, want to possess a free will that is incompatible with determinism?...

The Intelligibility Question:  Can we make sense of a freedom or free will that is incompatible with determinism?  Is such a freedom coherent or intelligible?...

The Existence Question:  Does such a freedom actually exist in the natural order, and if so, where?[2]

In the opening chapter of the present volume, Kane primarily focuses on the Compatibility and Intelligibility questions, though he also addresses the other two in the course of his essay.  He begins by noting that the exercise of free will requires an individual to believe that multiple options are open to her--that is, to believe that the future is a garden of forking paths.   He then presents a version of the Consequence Argument, which holds that if determinism is true, then there are not multiple options open to us and we cannot do otherwise than we actually do.  In other words, there is nothing we can now do to change the fact that any of our actions occur.  Kane thinks that the Consequence Argument is insufficient to establish the truth of incompatibilism because "focusing on 'alternative possibilities' (or 'forking paths' into the future) or the 'power to do otherwise alone, as the Consequence argument does, is too thin a basis on which to rest the case for the incompatibility of free will and determinism" (13).  The having of alternative possibilities plus indeterminism are thus jointly insufficient for free will (even if, as Kane thinks, they are both necessary for it).  Instead, Kane thinks the case for incompatibilism should be made on the basis of ultimate responsibility: "to be ultimately responsible for an action, an agent must be responsible for anything that is a sufficient cause or motive for the action's occurring" (14).  It is for this reason that Kane's view is a version of what is known as source incompatibilism.  According to Kane, what ultimate responsibility shows is that free will is primarily a function of the source or origin of one's will in that at least some of an agent's willings must lack sufficient causes if her will is to be free.  These undetermined actions are what Kane calls self-forming actions.  Kane then argues that while ultimate responsibility is more fundamental than the having of alternative possibilities, the two are fundamentally linked, for by their very nature self-forming actions involve multiple open options.  Thus, Kane's argument for incompatibilism is in a sense indirect.  Free will requires ultimate responsibility.  Ultimate responsibility requires at least some alternative possibilities.  The having of alternative possibilities requires the falsity of determinism.  Therefore, free will and determinism are incompatible. 

            Kane then turns his attention to the Intelligibility Question.  He is acutely aware that the indeterminism required by the previous argument can seem to undermine, rather than support, free will.  He aims to meet this challenge of how indeterminism can support the existence of free will without appeal to any 'extra factors' such as agent causation.  In addressing the Intelligibility Question, Kane begins by showing how indeterminism per se need not undermine control and responsibility.  He then argues that just as the presence of indeterminism need not undermine control, neither need it diminish it.  He then argues that in self-forming actions, an agent can be responsible for her choice even if that choice was undetermined insofar as she succeeded in doing what she was trying and wanting to do all along.  So long as the indeterminism present is the result of conflict in the individual's larger motivational system, its presence is necessary for free will:

[B]y being a hindrance to the realization of some of our purposes, indeterminism paradoxically opens up the genuine possibility of pursuing other purposes--of choosing or doing otherwise in accordance with, rather than against, our wills (voluntarily) and reasons (rationally)....  [Such] hindrances and obstacles and resistance in the will are precisely what are needed for free will, which, like life itself, exists near the ends of chaos (39).

In this way, indeterminism helps support the existence of free will.

Fischer's particular version of compatibilism, semicompatibilism, is well known.  Incompatibilist arguments which aim to show that determinism would mean that no agent is able to do otherwise than she actually does are "highly plausible" (56) and likely sound; thus, traditional versions of compatibilism resting on conditional analyses of freedom to do otherwise fail.  However, there is another kind of freedom--which Fischer calls 'guidance control'--which does not require alternative possibilities and is sufficient to fulfill the freedom-relevant condition on moral responsibility.  Like Kane, Fischer thinks that what matters for the kind of freedom that matters is how an action or choice is generated.  But, unlike Kane, he does not think that the source of a free action or choice needs to involve indeterminism; his view can thus be understood as a version of source compatibilism.  According to Fischer, the kind of freedom at issue--guidance control--requires two things:  "the mechanism that issues in action must be the 'agent's own,' and it must be appropriately 'reasons-responsive'" (78).  Both of these conditions can be satisfied even if determinism is true.  

While the above account, and further details of that account, are well-established through Fischer's other writings, Fischer's portrayal of semicompatibilism in the present volume is unique because of the considerable attention he gives for preferring the semicompatibilist position to source incompatibilism.  Fischer elaborates two motivations for semicompatibilism.  The first is what has been called the "resiliency intuition"[3]--namely that our view of ourselves as free and responsible agents should be resilient to whether or not theoretical physicists discover if determinism is true.  Fischer grants that there are some empirical discoveries that would threaten our view of ourselves as free and responsible agents--but not the mere truth of indeterminism.  A second motivation is that, on the semicompatibilist picture, moral responsibility does not depend on things which it shouldn't; more specifically, an agent's responsibility only depends on those things which the agent can reasonably be said to control.  To make this point in a negative way, an agent's responsibility cannot simply be a matter of luck.  According to Fischer, "once one recognizes the pervasiveness of a certain sort of luck, one will find an incompatibilist source condition less attractive" (68f).   But there is a tension between these two motivations in the following sense:  the more motivational strength Fischer assigns to one of these two reasons for preferring semicompatibilism, the less motivational strength he can assign to the other.  In order for his view to be resilient, Fischer's semicompatibilism needs to be compatible not only with the truth of causal determinism, but with indeterminism as well--and in fact compatible with as many different kinds of indeterminism as possible given the other constraints inherent in his view.   As a result of holding that such a range of indeterminism would not undermine moral responsibility, considerations of luck ought to carry less motivational weight for semicompatibilism.

Convinced that considerations of Frankfurt-inspired scenarios show that sourcehood approaches to free will are preferable to those approaches based primarily on alternative possibilities, Fischer goes on to argue against source incompatibilism as follows:  "Perhaps the incompatibilistic sourcehood requirement comes from, or is suggested by, a certain picture of agency.  On this picture, the locus of control must be entirely within us, if we are to be morally responsible....  It is as if the proponent of the incompatibilistic sourcehood constraint thinks of agents who are morally responsible as having 'total control'" (67).  Such total control, Fischer argues, is "a chimera," "a total fantasy," "metaphysical megalomania," and "manifestly ludicrous"--for our existence is dependent on any number of factors that are beyond our control (e.g., the sun's continuing to shine, a meteor not hitting the United States).   While I think that Fischer is aware that no source incompatibilist requires what he refers to as total control, he thinks that seeing the inflated and illusory nature of total control will undermine the appeal of source incompatibilism:  "my suggestion (and it is only a suggestion) is that, once one recognizes the pervasiveness of a certain sort of luck, one will find an incompatibilistic source condition less attractive" (68f).  While I have doubts that source incompatibilism is motivated by anything nearly as robust as total control, Fischer's discussion in this volume of the motivations for various positions regarding free will is quite illuminating.

            Like Kane and Fischer, Pereboom advances a source-based approach to free will--that is, an account of free will according to which the agent's having of alternative possibilities for action is not the primary condition which needs to be satisfied for free will.  Instead, the "more significant role" is ascribed to the action's causal history, or source (86f).  Furthermore, like Kane, Pereboom thinks that no agent could satisfy the sourcehood requirement if causal determinism were true:  "an action's being produced by a deterministic process that traces back to factors beyond the agent's control, even when she satisfies all the conditions on moral responsibility specified by the prominent compatibilist theories, presents in principle no less of a threat to moral responsibility than does deterministic manipulation" (93).  Pereboom elaborates a version of the four-case manipulation argument he also defends elsewhere.[4]  The four-case argument is built on the following two claims:

(i)                 manipulation undermining claim:  in cases of overt manipulation, the agent fails to be the proper source of his action in the way required for free will and moral responsibility, and

(ii)               similarity claim: causally deterministic cases are similar, with respect to the agent's failure to be the proper source of his actions, to cases of overt manipulation.

The conclusion of the four-case argument is that compatibilism is false.  After arguing against compatibilism, Pereboom then argues against libertarianism.  First, he argues against even-causal libertarian views via the luck objection:  "if ... causal determinism rules out moral responsibility, then it is no remedy simply to provide slack in the causal net by making the causal history of actions indeterministic....  In particular, it would not provide the capacity for an agent to be the source of her decisions and actions" (103).  Pereboom also argues that Kane's own version of event-causal libertarianism fails to provide the "control required for moral responsibility, despite the proposed complexity of the decision's underlying structure" (110).   Such additional control could be provided by agent-causation since agent-causes have the fundamental power to cause decisions without being causally determined to do so.  And while Pereboom is open to the possibility that scientific advance may show that we do in fact have such agent-causal powers, there is no extant evidence for such powers.  Pereboom thus advances a position he calls hard incompatibilism, which denies that human agents are free and morally responsibility. 

Pereboom ends by showing how hard incompatibilism can account for punishment, meaning, and the reactive attitudes despite the fact that we are not morally responsible agents.  While the truth of hard incompatibilism would undermine retributive theories of punishment, are while he rejects the two most prominent deterrence theories on independent grounds, Pereboom argues for a quarantine model of punishment based on the analogy between the appropriate treatment of criminals and the appropriate treatment of the carriers of dangerous diseases (116).  While this discussion about punishment is perhaps not as developed as one may like, it does show how Pereboom's view is "strongly revisionist about the key notion of moral responsibly" (122). 

The topic of revisionism brings us to the fourth contributor.  Vargas' revisionism is a relative new-comer compared with the positions defended by Kane, Fischer, and Pereboom.  Nevertheless, its inclusion in the present volume is welcome and warranted.  According to Vargas, "a theory counts as revisionist in the sense I am interested in if it offers a different prescriptive theory of responsibility (an account of what we ought to believe) than it offers for a diagnostic theory of responsibility (an account of what we tend to believe)" (151).[5]  Vargas' revisionism is split on these two elements as follows:  "In the diagnostic part I will argue that our ordinary thinking about free will has elements that are incompatibilist.  In the prescriptive part I will argue that we should revise away from these commitments" (129).  At least some of the disagreement between these two aspects, Vargas argues, is a result of the fact that different aspects of work on free will draw from different philosophical methodologies.  Those aspects of the free will debates that draw primarily from metaphysics will tend to be more descriptive, as "contemporary metaphysics has, at least methodologically, hewn more closely to approaches that less readily forsake the constraints of commonsense" (130).  In contrast, those aspects of the debates that draw more heavily from ethics are more likely to tolerate revision aware from commonsense beliefs.

As support for the incompatibilist diagnosis, Vargas marshals three primary sets of considerations.  The first of these is the set of traditional arguments for incompatibilism (e.g., the Basic Argument and the Consequence Argument).  Vargas announces that he is "somewhat more optimistic than Robert Kane ... [that] what makes these arguments powerful is not so much that they rule out the possibility of compatibilism but rather that they show how easily incompatibilism seems to capture ordinary ways of thinking about our own agency" (132).  This transitions into the second support for an incompatibilist diagnosis, where Vargas quickly summarizes experimental work which provides evidence for incompatibilist folk-thinking (even if one can also elicit compatibilist intuitions in different contexts).  These studies are "evidence that our ordinary understanding of free will and moral responsibility have incompatibilist elements, and that any theory that fails to acknowledge this will fail as a diagnostic theory of our folk concepts" (138).  Third, citing the influence of both dualism and Christianity, Vargas suggests that incompatibilism is reinforced by, and perhaps even rooted in, the cultural history of the West. 

Having argued in this way that our ordinary thinking is at least partially incompatibilist in nature, Vargas rejects both libertarianism and hard incompatibilism.  The problem with libertarianism is two-fold.  First, since libertarian theories require indeterminism to be present in particular places in the causal history of actions at particular times, libertarianism will be "comparatively less empirically plausible than the alternatives" (141) which do not have these requirements.  Second, libertarian accounts of agency are not supported by the empirical data we have from the relevant cognitive sciences.  Vargas then argues against hard incompatibilism (and other forms of free will skepticism) by undermining the assumption that our intuitions about free will tell us anything about the actual nature of free will.  Until someone provides an argument to show that there can be no disparity here, "we are not entitled to conclude that the implausibility of our self-conception is evidence that we are not free and responsible, for we might have free will but it might be different than we tend to suppose" (146).  Vargas then argues that there is justification for what he calls the "responsibility system," which is constituted by the whole of our "responsibility norms and their attendant social practices, characteristic attitudes, and paradigmatic judgments" (154).  The primary aim of such a system is "to get creatures like us to better attend to what moral consideration[s] they are and to appropriately govern our conduct in light of what moral reasons those considerations generate" (155), and the system as a whole is justified by its fostering of an increased ability to detect and respond to moral considerations.  And here Vargas thinks that a compatibilist account will be able to undergird a responsibility system even if it departs from our commonsense incompatibilist intuitions.[6] 

Four Views on Free Will serves as an excellent introduction to many of the central issues in the contemporary free will and moral responsibility literature.[7]  It is very readable; the chapters are approachable for advanced undergraduates or graduate students (although Pereboom's article is a little more technical than the others), and it has a very helpful and thorough index for cross-referencing topics between the various chapters.  As such, it will be particularly attractive as a text for a course on free will.  While not as wide-ranging in its coverage as its most likely competitor in the regard, Kane's excellent A Contemporary Introduction to Free Will, Four Views on Free Will is able to go into greater depth on those issues that it does treat.