Abstract
Alfred Mele’s original Zygote Argument is invalid. At most, its premises entail the negative thesis that free action is incompossible with deterministic laws, but its conclusion asserts the positive thesis that deterministic laws preclude (make impossible, undermine) free action. The original, explanatory conclusion of the Zygote Argument can be defended only by supplementing it with a best-explanation argument that identifies deterministic laws as menacing. (By the same reasoning, it follows that every manipulation argument pinpointing a specific threat to free will requires a best-explanation argument). Arguably, though, the best explanation for the manipulation victim’s lack of freedom and responsibility is his constitutive luck, which is a problem irrespective of the natural laws that obtain. This proposed explanation leads to a new “diagnostic” version of the Zygote Argument which concludes that free action is impossible even though deterministic laws pose no threat whatsoever to free will.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
I borrow the convenient term "menacing" from Ishtiyaque Haji.
In early discussions of the Zygote Argument, Mele allows that a proponent of ZA could also give a positive argument in defense of premise 1. Citing a passage from Thomas Kapitan (2000: 90), Mele supplies one "predictable" defense of premise 1, namely that Ernie is “deliberately caused to behave in a certain way in much the same way that designers of robots program the responses of their machines to various stimuli” (Mele 2006: 189, 2008: 280). Notably, this argument (if it deserves to be called one) mentions but does not identify deterministic laws or deterministic causation as a threat to free will. (It is unclear, for example, whether the menacing feature of Ernie's story is that he was programmed, that he was deliberately programmed, that he was deliberately programmed and then subjected to deterministic laws, or some other combination of the features that Mele mentions). However, in more recent (e.g. Mele 2013) formulations of the zygote argument, Mele seems to have adopted the more standard line according to which the basis for one's judgment of the truth of premise 1 rests entirely on one's intuitive judgment of Ernie's status as a free and responsible agent.
Notably, the meaning of Mele’s term ‘incompatibilism’ is unclear in this context. He states that incompatibilism is “The thesis that neither free action nor moral responsibility is compatible with the truth of determinism” (2006; my emphasis), but does not adequately specify the notion of “(in)compatibility” at issue.
The creation stories grounding ZA and ZAM are slightly different. These differences may be relevant to the soundness of the argument, but the differences are not relevant to the present discussion of the logical properties shared by these arguments. As such, I will not address their differences here.
On the downside, adding such restrictions even when it is unclear that they are relevant to the existence of freedom or responsibility may obscure the logic and lesson of an argument (e.g., see Campbell's (2007) "No-past Objection" to the Consequence Argument). Mele's repetition of the fact that Ernie lives in a deterministic universe gives rise to the misimpression that the argument pinpoints deterministic laws as a threat to free will (at least for beings of a certain sort), which may partially explain why the invalidity of ZA was overlooked for so long. The arguments in Sect. 5 of this essay suggest that the restrictions that Mele’s builds into the conclusion of ZAM—including "in a deterministic universe"—are not freedom- or responsibility-relevant ones.
In my view, the formal structure of this template (as well as McKenna's preferred statement of the general template for manipulation arguments (e.g., McKenna 2012: 151)) is slightly misleading. Non-diagnostic manipulation arguments of this sort have a simple modus ponens form: The Non-diagnostic Manipulation Argument Template:
-
1.
Victim Premise: Due to some feature of the (apparent) manipulation scenario, the (apparent) manipulation victim S is not free or responsible for performing an action A.
-
2.
Generalization Premise: If S is not free or responsible for performing A, then then no one in a (normal) determination scenario is free or responsible.
-
3.
Conclusion: No one is free or responsible in the determination scenario.
Premise 2 of Mele's SMA is really a supporting claim for the Generalization Premise of the above template. In support of the truth of the Generalization Premise, a philosopher might point to such things as a positive diagnosis the menacing feature that is common to both scenarios and the methodological principle that like cases must be judged alike.
-
1.
Mele (2008) does not consider the Zygote Argument to be an instance of SMA. However, Mele does not explicitly deny that the Zygote Argument has the logical form outlined by SMA. Rather, Mele argues that because the "original design" story he tells does not involve any genuine manipulation, the Zygote Argument is not technically a manipulation argument at all. In my view, Mele's proposed individuation principle gives rise to an overly narrow conception of the manipulation-argument strategy. There certainly seems to be some sort of manipulation taking place when Diana creates Ernie as she does, and this appearance of manipulation seems to be essential to the common intuitive reaction to the case (but the genuineness of the manipulation does not). Moreover, some (e.g., Barnes 2013) have argued that there is genuine manipulation in Mele's zygote case. If there is genuine manipulation in the case, then Mele has categorized the Zygote Argument wrongly according to his own individuation principle. For reasons such as these, I contend that the Zygote Argument is best classified as a manipulation argument. That said, manipulation arguments that rest on "original-design" stories are surely worth distinguishing from others on the grounds that they are the most compelling instances of the manipulation-argument strategy.
This observation about SMA is noteworthy because SMA is (as far as I can tell) formally equivalent to Michael McKenna's generic template for manipulation arguments, "The Manipulation Argument" (e.g. 2012:151). As such, McKenna's template, too, fails to outline an argument for incompatibilism. Thus, recasting the Zygote Argument as a (valid) instance of McKenna's template would not be relevantly different from recasting it as an instance of SMA.
It seems that the terms 'compatibilist' and 'incompatibilist' were introduced by Keith Lehrer, and were first used in print in Leher's (1960) dissertation; he introduced the more standard characterizations of these terms in print eight years later (Cornman and Lehrer 1968: 130). It seems that the corresponding terms 'compatibilism' and 'incompatibilism' were first used in print by van Inwagen in his (1969) dissertation, but van Inwagen (in correspondence) credits Lehrer--who was Second Reader on van Inwagen's dissertation defense committee--with the coining of these terms as well.
Technically, incompatibilism entails only the qualified incompossibilist thesis that there is no possible universe in which deterministic laws obtain and someone who is subject to the laws performs a free action. According to unqualified incompossibilism, it is even impossible for a being who is not subject to the natural laws (e.g., some god-like being who could change or violate the laws of nature) to act freely in a universe with deterministic laws. Since the differences between incompossibilism and more qualified versions of incompossibilism are not pressing in the present context, I will ignore these subtleties for the purposes of this essay.
I am oversimplifying the dialectic here. I do not mean to imply that a manipulation argument presents a challenge only to those philosophers who have a (strong) victim intuition in response to the argument’s manipulation story. It seems that a manipulation argument will have at least some purchase so long as there is a difference between one's intuitive judgments of the manipulation victim and the naturally determined agent, whether in degree (e.g. one has a strong intuition that the naturally determined agent is free but only a weak intuition that the manipulation victim is free) or in kind (e.g., one has a victim intuition in response to the manipulation case but also the intuition that the merely determined agent is free). However interesting, these complicated dialectical issues are not pressing in the present context, where the primary focus is exposing the logical structures of manipulation arguments and not identifying the full range of their target audience.
Notably, as van Inwagen defines 'incompatibilism', incompatibilism is the mere denial of the thesis that determinism and the free-will thesis could both be true (e.g. 2008: 330). Standard formulations of the Consequence Argument defend only this negative, inconsistency thesis and not the positive, explanatory thesis that (necessarily) if determinism is true, then the free-will thesis is false because determinism is true. As such, there is room to argue--as I do in "No Past? No Problem" (unpublished manuscript)--that the Consequence Argument itself would need to be supplemented with a best-explanation argument in order to pinpoint a specific threat to free will, and there is even room to deny that the lesson of the Consequence Argument (if sound) is that being subject to deterministic laws undermines free will.
One might argue that diagnostic manipulation arguments are not as purely “persuasive” as non-diagnostic manipulation arguments. Because the best-explanation argument defense of the Diagnostic Premise picks out a feature of the manipulation story which would make a victim intuition a rational response to the story, it also implies that anyone who fails to have the victim intuition is suffering from some sort of problem, e.g., irrationality or what Mele (2013: 182) has called "intuition deficit disorder." As such, rejecting the truth of the Victim Premise of a diagnostic manipulation argument (i.e. defending a "hard-line" reply) may be dialectically more complicated than rejecting the same claim in the context of a non-diagnostic manipulation argument. Such differences between the dialectics surrounding diagnostic and non-diagnostic manipulation arguments are worth exploring, but I gloss over these differences here because they are not relevant to my argument that a valid manipulation argument for incompatibilism must include a best-explanation argument.
The reader should not take this conclusion as implying that all manipulation arguments for incompatibilism are instances of Mele's "Best-explanation Manipulation Argument" template (Mele 2008: 276). Mele's Best-explanation Manipulation Argument template mischaracterizes the role played by best-explanation arguments in manipulation arguments for incompatibilism. Mele describes a best-explanation manipulation argument as one that differs from an instance of SMA only in its second premise: the former has a "Best-explanation Premise" identifying the menacing feature of the manipulation scenario instead of the latter's No-difference Premise. As such, Mele's Best-explanation template does not make the case that the same menacing feature is present in both the manipulation scenario and the normal deterministic scenario. This is important because even when the menacing feature is correctly identified as a type of determination, it remains an open question whether the same menacing type of determination is present in the normal scenario because not all deterministic manipulation perfectly mimics natural determination cf. (Demetriou 2010). In order to bridge that gap, a manipulation argument needs something like what I (above) call a "Same-feature Premise," and Mele provides no such bridge premise in his template.
When G. Strawson makes claims such as “According to the Basic Argument, it makes no difference whether determinism is true or false” (2008: 289; my emphasis), it is not entirely clear whether he means that both sorts of laws pose a threat, neither do, or some alternative. One might think that Strawson means by this that his Basic Argument is equivalent to what is often called the "Standard Argument," an argument for impossibilism that roughly consists of the conjunction of the Consequence Argument and the Mind Argument (for a description of the Mind Argument, see van Inwagen 1983:126–152). However, I deny that the Basic Argument is equivalent to the Standard Argument. As noted above (fn. 12), it is unclear whether the Consequence Argument is best understood as an argument for incompossibilism or incompatibilism. However, the Mind Argument does identify indeterministic causation as a positive threat to free will. As such, the conclusions of the Basic Argument and the Standard Argument are not equivalent. Whether the explanatory conclusions of these two arguments are consistent seems to depend on the modal semantics being assumed. Assuming impossible worlds semantics, one might meaningfully claim that no possible being acts freely because no possible being satisfies the "starting point" sourcehood condition—but if, counterpossibly, someone did satisfy it, indeterministic causation would (if properly situated in the causal chain leading to action) undermine that person's freedom and moral responsibility. In that case, the conclusions of the two arguments are consistent. However, assuming standard possible worlds semantics, it seems that the conclusion of the Basic Argument and the conclusion of the Standard Argument are contrary claims: if the explanation of the impossibility of free will defended by the Basic Argument is correct, then there is no possible world at which indeterministic causation (of any sort) undermines free will, in which case the explanatory conclusion of the Standard Argument is false; on the other hand, if there is some possible world at which indetermnistic causation (of some sort) undermines free will, then the explanatory conclusion defended by the Basic Argument is false.
A remote past is a time "before there were any human beings" (van Inwagen 1989: 224).
The interesting dialectic between incompatibilist-impossibilists and non-incompatibilist-impossibilists has not been given much attention in the contemporary literature—presumably because the distinction between incompatibilism and incompossibilism has not been widely recognized. Kadri Vihvelin is the most significant exception. Among other things, Vihvelin (e.g. 2008, 2013) argues that impossibilism and incompatibilism are logically inconsistent views. If Vihvelin is right, then a philosopher could not, for example, argue for impossibilism by using (1) the Zygote Argument to defend incompatibilism with respect to free will and deterministic laws and (2) a second argument–e.g. the Mind Argument (van Inwagen 1983:126–152)—to defend the incompatibility of free will and indeterministic laws. While I have argued elsewhere (Mickelson forthcoming) that Vihvelin’s arguments for the purported inconsistency of incompatibilism and impossibilism fail, I am sympathetic to the nearby view that any compelling best-explanation argument for incompatibilism must assume (or argue for) the metaphysical possibility of free action. In order to get clear on the lesson of the Zygote Argument, it seems that philosophers will have to consider such taxonomical and dialectical issues in greater detail.
I borrow the notion of a “disappointed compatibilist” from Levy (2011: 2). Levy says that he is “disappointed” insofar as he was hoping that compossibilism (rather than incompossibilism) would turn out to be true, but a “compatibilist” insofar as he denies the incompatibilist’s claim that deterministic laws preclude free will.
References
Berto, F. (2013). Impossible worlds. In E. N. Zalta (Ed.), The stanford encyclopedia of philosophy (Winter 2013 Edition). http://plato.stanford.edu/archives/win2013/entries/impossible-worlds/
Campbell, J. (2007). Free will and the necessity of the past. Analysis, 67, 105–111.
Cornman J. and Lehrer K. (1968). Philosophical Problems and Arguments: An Introduction. New York: MacMillan Company.
Demetriou, K. (see also Mickelson, K.) (2010). The Soft-Line Solution to Pereboom’s Four-Case Argument. Australasian Journal of Philosophy, 88(4), 595–617.
Kapitan, T. (2000). Autonomy and manipulated freedom. Philosophical Perspectives, 14, 81–104.
Kearns, S. (2012). Aborting the Zygote Argument. Philosophical Studies, 160(3), 379–389.
Lehrer, K. (1960). Ifs, cans, and causes. Dissertation, Brown University. Providence: ProQuest/UMI. (Publication No. AAT: 6205755.)
Levy, N. (2011). Hard luck: How luck undermines free will and moral responsibility. New York: Oxford University Press.
McKenna, M. (2010). Whose argumentative burden, which incompatibilist arguments?—Getting the dialectic right. Australasian Journal of Philosophy, 88(3), 429–443.
McKenna, M. (2012). Moral responsibility, manipulation arguments, and history: Assessing the resilience of nonhistorical compatibilism. Journal of Ethics, 16(2), 145–174.
Mele, A. (2006). Free will and luck. New York: Oxford University Press.
Mele, A. (2008). Manipulation, and moral responsibility. Journal of Ethics, 12(3), 263–286.
Mele, A. (2012). Manipulation, moral responsibility, and bullet biting. Presentation at the workshop on manipulation arguments, Central European University. http://humanproject.ceu.hu/events/2012-06-07/workshop-on-the-manipulation-argument
Mele, A. (2013). Manipulation, moral responsibility, and bullet biting. Journal of Ethics, 17(3), 167–184.
Mickelson, K. (see also Demetriou, K.) (forthcoming). A Critique of Vihvelin's Three-fold Classification. Canadian Journal of Philosophy.
Strawson, G. (1986). Freedom and belief. Oxford: Clarendon Press.
Strawson, G. (1994). The impossibility of moral responsibility. Philosophical Studies, 75, 5–24. Reprinted in Watson 2003.
Strawson, G. (1998, 2011). Free will. In E. Craig (Ed.), Routledge Encyclopedia of Philosophy. London: Routledge. http://www.rep.routledge.com/article/V014SECT3
Strawson, G. (2002). The bounds of freedom. In Kane 2002 (pp. 441–460).
Strawson, G. (2008). The impossibility of ultimate moral responsibility. In D. Pereboom (Ed.), Free will second edition 2009 (pp. 289–306); originally from 2008 Real Materialism and Other Essays. Oxford: Oxford University Press.
van Inwagen, P. (1969). An Essay on the Freedom of the Will. Dissertation, The University of Rochester. Rochester: ProQuest/UMI. (Publication No. AAT: 6914798.)
van Inwagen, P. (1983). An essay on free will. Oxford: Oxford University Press.
van Inwagen, P. (1989). When is the will free? Philosophical Perspectives, 3, 399–422. Repr. in Agents, causes, and events, ed. T. O’Connor. Oxford: Oxford University Press, 1995. Page references are to this latter work.
van Inwagen, P. (2008). How to Think About the Problem of Free Will, Journal of Ethics, 12(3/4), 327–341.
Vihvelin, K. (2008). Compatibilism, incompatibilism, and impossibilism. In John Hawthorne, Theodore Sider, & Dean Zimmerman (Eds.), Contemporary debates in metaphysics (pp. 303–318). Malden, MA: Blackwell Publishing.
Vihvelin, K. (2013). Causes, laws, and free will: Why determinism doesn’t matter. New York: Oxford University Press.
Acknowledgments
I presented early drafts of this paper at the Central European University (Workshop on the Manipulation Argument, 2012) and at Southern Methodist University (A Workshop on Free Will, 2013). I am indebted to Michael McKenna, András Szigeti, and Philippe Chuard for organizing these fine events and for inviting me to participate. I received excellent feedback at each, and am especially grateful for helpful comments from Gunnar Björnsson, Andreas Brekke Carlsson, Damir Cicic, Justin C. Fisher, Charles Hermes, Neil Levy, Alfred Mele, Derk Pereboom, Carolina Sartorio, and Patrick Todd. I would also like to thank Joseph Campbell, Michael McKenna, Ish Haji, and Bob Hanna for insightful comments on earlier drafts of this essay. Finally, I would like to thank the University of Minnesota, Morris for two Faculty Research Enhancement Grants that supported my work on this project.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Mickelson, K. The Zygote Argument is invalid: Now what?. Philos Stud 172, 2911–2929 (2015). https://doi.org/10.1007/s11098-015-0449-6
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11098-015-0449-6