Thinking about Free Will – VAN INWAGEN (M)

VAN INWAGEN, Peter. Thinking about Free Will. Cambridge University Press, 2017. 232p. Resenha de: MERLUSSI, Pedro. Manuscrito, Campinas, v.42 n.1 Jan./Mar. 2019.

The book is chronologically divided into 14 chapters, covering the main issues that have been at the centre of recent concern on free will. For instance, Harry Frankfurt’s influential objection to the principle of alternative possibilities (namely, that being able to do otherwise is a necessary condition for moral responsibility) is discussed in chapters 1 and 6. The author also deals with the Mind and ethics arguments for the claim that indeterminism and free will are incompatible (chapter 2). With respect to the compatibility problem of free will and determinism, there is van Inwagen’s reply to McKay and Johnson’s counterexample (1996) to the no-choice transfer rule (chapter 7). Given the supposed plausibility of the arguments for the incompatibility of free will and determinism and the incompatibility of indeterminism and free will, van Inwagen thinks that free will is a mystery (chapters 7 and 10). Other important topics of interest include a discussion about Daniel Dennett’s book Elbow Room (chapter 4), van Inwagen’s own view that we seldom are able to do otherwise (chapter 5), the notion of ability (chapters 7, 11 and 14) and, perhaps more importantly, a reply to Lewis’ objection to the Consequence Argument, namely, the question of whether we are able to break the laws of nature (chapter 9), among many others.

There are many interesting problems discussed in the book, and most of them are well-known among readers familiar with the discussion. (Naturally, there is also a good deal of overlap between the chapters). It is not my aim to focus on every interesting thesis in the book. What I shall do instead is to focus on van Inwagen’s insights about what is the problem of free will, especially his claim that free will is a mystery given the joint plausibility of the Consequence Argument (which argues for the incompatibility of free will and determinism) and the Mind Argument (which argues for the incompatibility of free will and indeterminism).

As one might expect, there are many philosophical problems about free will, such as the problem of logical determinism (how is free will possible if there are true propositions about our future actions?), the problem of theological omniscience (how is free will possible if God knows beforehand what we are going to do?), etc. But is there a thing which may be properly called the problem of free will? I think van Inwagen provides us with a nice proposal. The idea is that the problem of free will may be presented as a set of propositions for all of which we seem to have good reasons, but which are jointly inconsistent. “Free will” and “determinism” have been used in the literature, to be sure, in different senses. According to van Inwagen, “free will” involves the ability to do otherwise, and “determinism” is nomic determinism, that is, the thesis that the past and the laws of nature determine a unique future. The problem goes as follows (or close enough):

  1. If nomic determinism is true, then there is no free will.
  2. If nomic determinism is not true, then there is no free will.
  3. There is free will

The problem leaps out because we seem to have unanswerable arguments for propositions 1 (namely, the Consequence Argument) and 2 (namely, the Mind argument), and it is no easy task finding out a proper answer to these arguments. And if it turns out that propositions 1 and 2 are true, 3 will be false, and in that case there would not be such a thing as free will. However, van Inwagen also notices that “there are […] seemingly unanswerable arguments that, if they are correct, demonstrate that the existence of moral responsibility entails the existence of free will, and, therefore, if free will does not exist, moral responsibility does not exist either” (149).

It would then be bad news if these arguments were cogent (because it would be difficult to see how moral responsibility is possible). Though the Consequence Argument and the Mind argument have initially seemed unanswerable, there are – I think – promising replies to the main formulations of these arguments in the literature, especially if we make some assumptions about counterfactuals and the laws of nature. The first answer to the Consequence Argument along these lines is Lewis’ `Are we free to break the laws?”, which possibly is – according to van Inwagen – “the finest essay that has ever been written about any aspect of the free will problem”. (I will not, however, present the Consequence Argument here, since it can be found everywhere in the literature).

If we make some assumptions about counterfactuals, then one premise of the Consequence Argument is implausible. Let L be the conjunction of all the correct laws of nature. One premise in the argument tells us that L is true and no one has, or ever had, the ability to do something such that, if one were to do it, L might be false. Consider Lewis’ theory of counterfactuals, where “if p were the case, then q would be the case” is (non-vacuously) true in our world iff q is true in the most similar worlds to ours where p is true. Imagine, for example, a possible world where I am doing otherwise, say, typing different words on this page. If determinism is true, this world cannot have the same past and laws of nature as ours. Now ask yourself: What are the closest worlds to ours where I type different words? Are they worlds where the actual laws of nature are broken or where the past history is different all the way back to the Big Bang? We just have one option, and Lewis tells us that the most similar worlds to ours are those where the laws of nature are slightly different from ours (or broken by what he called a “divergence miracle”), but where the past isn’t different all the way back to the Big Bang. (For objections to Lewis’ theory, see Bennett (1984), Goodman (2015), Dorr (2016)). Lewis points out that the laws need not be broken by our acts, so that the compatibilist need not be committed to the claim that we are able to break the laws of nature. So, if Lewis is right about how we should evaluate counterfactuals on the assumption of determinism, then the premise of the Consequence Argument that no one has any choice about whether L is unjustified.

Another, more interesting problem with the argument is that it presupposes an anti-Humean conception of the laws (see, in particular, Beebee 20002003). According to the Humean conception, the laws of nature are – roughly speaking – the best way to summarise all past, present and future facts. Laws do not govern anything, but merely systematise. If this conception of laws is correct, it is hard to see how the claim that laws are deterministic is a threat to free will. After all, if Humean laws do not govern, they do not place a constraint on our actions. As a result Humeans will not have trouble in saying that the claim that laws are deterministic is consistent with our ability to do otherwise. Humeans like Beebee see no problem in saying that agents are able to break the laws in the sense that the laws are violated or broken by our acts. Lewis, on the other hand, claimed that agents are able to do otherwise than they in fact did even if determinism is true, but denied that agents are able to break the laws of nature in that sense, and his view is known as Local Miracle Compatibilism.

It seems to me that some similar worries apply to the Mind Argument. A toy version of the argument goes more or less as follows: (M) “If what one does does not follow deterministically from one’s previous states, then it is the result of an indeterministic process, and (necessarily) one is unable to determine the outcome of an indeterministic process” (162). The problem here, I think, is that van Inwagen presupposes that (i) the laws of nature cover our actions and (ii) are indeterministic with respect to them. There are two questions about the nature of laws that we should keep separate from one another:

Extent: Is everything that happens covered by the laws of nature? For instance, there may be happenings, or kinds of happenings, or whole domains about which L – the complete set of correct laws – is silent.

Permissiveness: When L speaks about the outcomes that are to occur, what kind of latitude does it admit? For instance, does it always select a single happening? Does it always lay down at least a probability, or can L admit a set of different outcomes, remaining silent about their probabilities?

Clearly there are further implicit conditions if (M) is to follow from the assumption that the laws of nature are indeterministic. I take it that it is presupposed that my action, for instance, is the kind of happening that is governed by laws and that those laws that govern it are indeterministic with respect to it. But the assumption that the laws of nature are universal in extent in the sense that they cover everything that happens in the world, however, is unjustified. There is nothing in the mainstream accounts of the laws of nature that require them to be universal in extent, let alone to cover actional-events.

My suspicion is that the problem with respect to indeterminism and free will arises because it is presupposed that the laws govern or cover our actions, and are indeterministic with respect to it. This is why “indeterminism” seems to rule out control over our actions. But what are the reasons for accepting that the laws govern everything? Perhaps they do not. If so, I find it difficult to accept the core idea behind the Mind (or the luck) Argument.

The problem of free will has indeed momentous philosophical consequences. While many solutions have been offered in the literature, they will involve the acceptance of surprising and not altogether unquestionable philosophical assumptions. As such, even though philosophers have questioned the plausibility of propositions 1 and 2, van Inwagen’s formulation and treatment of the free will problem remains fruitful, and his work continues to shed light in one of the most interesting and intractable problems of philosophy.Thinking about Free Will nicely supplements the existing objections and responses to the theses advocated in An Essay on Free Will and it also covers the main new topics that have been at the centre of recent concern on free will.


BEEBEE, H. “The Nongoverning Conception of Laws of Nature”, Philosophy and Phenomenological Research 61: 571-594, 2000. [ Links ]

_____ “Local Miracle Compatibilism”, Noûs 37 (2): 258-277, 2003. [ Links ]

BENNETT, J. “Counterfactuals and Temporal Direction”, The Philosophical Review 93 (1): 57-91, 1984. [ Links ]

CARTWRIGHT, N. & MERLUSSI, P. “Are laws of nature consistent with contingency?”, in Laws of nature, eds. W. Ott and, 2018. L. PATTO, Oxford: Oxford University Press. [ Links ]

DORR, C. “Against Counterfactual Miracles”, The Philosophical Review, 125 (2): 241-286, 2016. [ Links ]

GOODMAN, J. “Knowledge, Counterfactuals and Determinism”, Philosophical Studies 172 (9): 2275-2278, 2015. [ Links ]

MCKAY, T. J. and Johnson, D. “A Reconsideration of an Argument against Compatibilism”, Philosophical Topics 24 (2): 113-122, 1996. [ Links ]

LEWIS, D. “Are We Free to Break the Laws?”, Theoria 47: 113 – 21, 1981. [ Links ]

Pedro Merlussi – University of Campinas, Center for Logic and Epistemology, Campinas, SP, Brazil. E-mail: [email protected]

Acessar publicação original

Deixe um Comentário

Você precisa fazer login para publicar um comentário.