Whose Argumentative Burden, which Incompatibilist Arguments?—Getting the Dialectic Right

Abstract

Kadri Vihvelin has recently argued that between compatibilists and incompatibilists, the incompatibilists have a greater dialectical burden than compatibilists. According to her, both must show that free will is possible, but beyond this the incompatibilists must also show that no deterministic worlds are free will worlds. Thus, according to Vihvelin, so long as it is established that free will is possible, all the compatibilist must do is show that the incompatibilists' arguments are ineffective. I resist Vihvelin's assessment of the dialectical burdens of compatibilists and incompatibilists, as well as her assessment of the best arguments for incompatibilism.

Notes

¹This presumption permeates the recent history of the literature. See, for example, Campbell [1951], Edwards [1958], James [1897], and Taylor [1974]. As for contemporary theorists, see for example, Fischer [1994].

²Vihvelin also discusses this argument at length in her Stanford Encyclopedia of Philosophy entry, ‘Arguments for Incompatibilism’[2007].

³I offer one in McKenna [2003]. In doing so, I follow numerous others who have advanced compatibilism in this way, such as Fischer [1994], Fischer and Ravizza [1998]; Frankfurt [1971]; Haji [1998]; and Wallace [1994].

⁴Smilansky [2000], like Strawson, also entertains a self-creation conception of freedom. This conception of freedom can be found in Nietzsche's notion of causa sui[1886] and in Sartre's notion of radical freedom [1948]. For replies to Strawson, see Clarke [2005], McKenna [2008d], and Mele [1995].

⁵Vihvelin introduces the useful term, ‘non-godlike’, which is meant to pick out ‘someone who is not omniscient, omnipotent, infallible, infinite, the cause of its own existence, and so on’[2008: 307].

⁶For the threat posed by indeterminism-generating luck to freedom and moral responsibility, see Alfred Mele's Free Will and Luck[2006].

⁷Consider in this connection how close Derk Pereboom comes to an incompatibilist-impossibilist conclusion about the relationship between moral responsibility and determinism. In Living Without Free Will, Pereboom argues that determinism rules out moral responsibility [2001]. But he also considers whether any of the ways that indeterminism could be true would allow for moral responsibility. He argues that no event-indeterministic brands of libertarianism work. (He also assumes that no non-causal libertarians views are successful.) He then turns his attention to agent causal views. In his estimation, agent causation is metaphysically coherent, and so possible; but we have no good empirical reason to believe that at this world it is realized. Thus, Pereboom is an incompatibilist and not an impossibilist. But notice how close he came. Were he to have been swayed by others that agent causation is, after all, not coherent, then he would have landed himself in the camp of the impossibilist. Were he to have done so, given his argument for the relationship between determinism and moral responsibility, he still would have counted as an incompatibilist.

⁸For the original formulations of it, see Ginet [1966], Wiggins [1973], and van Inwagen [1975]. There is a giant secondary literature on this argument. For a thorough discussion of it, see Fischer [1994]. See also the careful treatment of it in Kapitan [2002].

⁹See Fischer on this point [1994: 78].

¹⁰Or perhaps the disagreement will turn upon differences concerning the nature of the laws. I will not pursue this here. For a compatibilist response to the Consequence Argument that makes explicit use of a deflationary, Humean conception of the laws of nature, see Beebee and Mele [2002].

¹¹I am indebted to an anonymous referee for the Australasian Journal of Philosophy for encouraging me to develop the points set out in this last paragraph.

¹²In fact, Vihvelin [2000] has offered a powerful criticism of Frankfurt's argument, and she clearly appreciates how much rides on the soundness of it. For an assessment of the current dialectical status of Frankfurt's argument see my [2008b].

¹³Despite these remarks, I do not mean to claim that Frankfurt's argument is required to establish the thesis that freedom to do otherwise is not necessary for free will or moral responsibility. The thesis might be true even if Frankfurt's argument falls short of establishing that it is, and there might well be other arguments that would succeed. Though I know of none.

¹⁴This incompatibilist proposal can be located in various places. But Mele gives an especially crisp expression of it in, among other places, his Autonomous Agents[1995: 211]. Pereboom advances a similar thought, and also with admirable clarity [2001: 3].

¹⁵I have recently argued against it [McKenna 2008d].

¹⁶Mele's formulation does not work with an already existing agent who is manipulated but instead with a created agent whose creation and existence is targeted to perform some act at a certain point in his life. Thus, his argument is more accurately labelled a creation argument.

¹⁷Recently, I have offered a compatibilist reply to various manipulation arguments in McKenna [2004, 2008a].

¹⁸In fact, I have argued that it is immune to certain compatibilist replies [2001]. More recently [2008c], I have attempted a refutation of it.

¹⁹In McKenna [2001] I considered all three of the arguments for what I call Source Incompatibilism and concluded that the Direct Argument is the best of them. I now think that the best of them is the Manipulation Argument [McKenna 2008c].

²⁰For helpful comments, I would like to thank Randolph Clarke, Ishtiyaque Haji, Alfred Mele, Neal Tognazzini, and two anonymous referees for Australasian Journal of Philosophy. I owe a special debt of gratitude to Kadri Vihvelin for her friendly and spirited disagreement with me.