Overview of the Problems of Free Will

Free Will

• The basic problem: how can we reconcile our status as agents – creatures who make rational decisions and are able to settle how our bodily actions occur – with the picture of the world given to us by physics, a world in which the mechanical processes governing sub-agential constituents are a sufficient basis for everything else?
  • A perennial philosophical problem, though looms larger with the rise of mechanistic, deterministic science in the seventeenth century.
• This basic problem can be refined and articulated in various ways. We’ll look in depth at one: Peter van Inwagen’s treatment in An Essay on Free Will (1983).
  • Other treatments worth looking at, in addition to those on the syllabus: a number of SEP survey articles (Vihvelin 2018; McKenna and Coates 2021; O’Connor and Franklin 2022); the collections edited by Watson (2003) and Pereboom (2009); the collaborative effort by Fischer, Kane, et al. (2007); and relatively recent books by Steward (2012) and Vihvelin (2013).
  • A couple of fictional treatments by Ted Chiang (2005, 2019) may be of interest; we’ll come back to them at the very end.

Problems of Free Will

• Van Inwagen identifies two (van Inwagen 1983: 2):
  1. The compatibility problem, concerning whether free will is or is not incompatible with determinism.
  • This will be our focus next time.
  2. The traditional problem, concerning whether we have free will or whether instead determinism is true.
  • The existence of this problem presupposes that the answer to the compatibility problem is no, they are not compatible (the view known as incompatibilism) – otherwise we aren’t forced to make the choice the traditional problem is pressing on us.
  • We look at this problem at length at the end of the course.
• Van Inwagen’s own position is that incompatibilism is correct, and that we do have free will (1983: 222–23).

The Structure of the Space of Positions (van Inwagen 1983: 13–14)

• Those who say that the answer to the compatibility problem is that free will and determinism are compatible are known as compatibilists.
  • Compatibilists who accept that free will exists and determinism is true are known as soft determinists.
  • There could also be compatibilists who deny free will; some who think that free will requires determinism (Hobart 1934; van Inwagen 1983: 126ff) might also think indeterminism is true. We’ll return to this strand of compatibilism later on.
• Incompatibilists – those who face the traditional problem – also divide:
  • Hard determinists accept determinism, and therefore deny the existence of free will;
  • Libertarians accept that free will exists, and deny determinism.
  • The traditional problem may also be answered neither; this might be the position of some incompatibilist who thought, for example, that the laws of nature made all events essentially random, our actions included.

Determinism Defined

Preliminary Definition of Determinism

• Determinism is ‘the thesis that the past determines a unique future’; but that’s pretty unhelpfully circular. More precisely:

  Determinism may now be defined: it is the thesis that there is at any instant exactly one physically possible future. There must, of course, be at least one physically possible future; if there is more than one, if at some instant there are two or more ways in which the world could go on, then indeterminism is true. (van Inwagen 1983: 3)

• Two things we need to get clear on:

  1. What exactly is physical possibility?
  2. How should we represent the multiple possibilities involved in indeterminism?

Kinds of Possibility

• Possibly and necessarily and cognates (might, have to, must, could, …) are context-dependent: they expresses different sorts of modality in different contexts (recall Lewis (1976) on can from the first lecture):

  • It has to be raining. (epistemic modality)
    
  • Visitors have to leave by six pm. (deontic)
  • I have to sneeze. (circumstantial)
    
  • To get home in time, you have to take a taxi. (teleological)

  (von Fintel 2006)

• Account due to Kratzer (1981): conversational context contributes a modal base, the kind of possibility involved; and the modals are existential (possibility) or universal (necessity) quantifiers over the modal base.

• We can make explicit the modal base: In view of \(F\), possibly \(p\), e.g., according to the hospital regulations, she can have up to 4 visitors.

In View of What is Something Physically Possible?

• So van Inwagen says: \(p\) is physically possible at \(t\) iff in view of the laws of nature and the past as of \(t\), possibly \(p\) (van Inwagen 1983: 3).
  • A kind of combination of deontic modality (the laws of nature do regulate the behaviour of objects, though not in the same way as moral laws) and circumstantial (the contingent facts of the initial configuration constrain what is possible).
• Which in turn is to say: there is a possible world which shares our laws and our past and in which \(p\); the laws and past do not suffice to entail \(¬p\), and so do not rule out \(p\). (This picturesque metaphor need not be taken literally, but is too helpful to avoid.)
• A law of nature is a certain sort of objectively true proposition governing the behaviour of scientific entities; but specifying precisely which sort is tricky.
• We need to assume that the laws are dynamical: they concern the evolution of the world over time. The classic example might be Newton’s laws of motion.
• We’ll assume that the laws are given by some correct final scientific theory; to investigate the compatibility problem, we needn’t suppose we know what that theory is, or that it is among any of the current candidates.

Laws of Nature

Philosophers have on occasion proposed necessary conditions for a proposition’s being a law of nature. A law, for example, is supposed to be true, to be contingent, to entail the existence of no particular (contingent) individual, and to ‘support its counterfactuals’…. But even if these conditions are jointly necessary, they are certainly not jointly sufficient. (van Inwagen 1983: 6)

• This highlights the problem with giving an analysis of the concept, ‘law of nature’ (van Fraassen 1989).
• But still, law of nature ‘seems to be an intelligible concept and one we can’t along without if we wish to give a complete description of the world’ (van Inwagen 1983: 7).

Branching Worlds

• One way to think about ‘more than one’ physically possible future is this: an indeterministic world is temporally asymmetric with respect to what is settled by the laws (Lewis 1983: 359).
• To the past, there is the settled history. But to the future, there are many ‘branches’, differing from each other in what events they contain. Each represents a different lawful continuation of the past.
  • If some branches feature \(p\) being true, and others feature it being false, then the future is unsettled with respect to \(p\).
• If determinism is true, there is only one lawful continuation of the past.
  • If there is only one branch, nothing is unsettled. On this view, a world is indeterministic iff there are unsettled propositions.
• As time passes, if indeterminism is true, the branches that are not ‘taken’ – those corresponding to unactualised possibilities – drop off. (That’s why the past is a settled singular ‘trunk’.)
  • Thus this is a version of the A-theory: the present has a special significance as the last settled moment before the branching begins, and the overall branching structure of spacetime changes over time.

Parallel Worlds

• A B-theoretic alternative: there are many parallel worlds (ways things might have been), each with a linear time-series ordered by B-relations (Lewis 1983: 360).
• Determinism is when there is only one lawful future: if possibilities \(w\) and \(w'\) agree on some initial temporal segment and their laws, then \(w=w'\).
  • This is roughly to say that the physically possible (lawful) futures supervene on the past – no difference in what will happen without some difference in what already did.
  • Compare Earman (1986: 13): for him, the laws are deterministic iff for any possible worlds \(w\) and \(w'\) which share those laws, if ever \(w\) is in the same global state as \(w'\), then \(w=w'\). A world is deterministic iff its laws are.
• Under indeterminism, among the worlds with our laws, there are many that have an initial segment just like our past, but which diverge from ours in the future; these are the different physically possible continuations of our past.
• Accordingly, though we inhabit just one of these worlds, its future isn’t fixed by its past and laws; it is physically contingent how it is, hence we won’t know what it will be like (though there is something it will be like).
• B-theoretic indeterminism doesn’t require the non-existence of a unique future; just as a purported failure of the mind to supervene on the brain doesn’t entail the non-existence of mental states.

Interlude: Universal Causation and Prediction

• Historically, free will wasn’t contrasted with the doctrine of determinism; Hume’s discussion in the Enquiry for example is concerned with the threat that universal causation poses to ‘liberty’ (Hume 1777: §7).

• But determinism and universal causation (‘everything has a cause’) are not equivalent. In particular, causation would have to involves necessitation to support determinism: that when \(c\) causes \(e\), then it is physically necessary that if \(c\) then \(e\).

  • This is doubtful: coin tosses seem to cause their outcomes without necessitating them (Lewis 1983: 360; Eagle 2007).

• Determinism is also historically linked with predictability, e.g., Laplace’s demon:

  An intelligence that, at a given instant, could comprehend all the forces by which nature is animated and the respective situation of the beings that make it up, if moreover it were vast enough to submit these data to analysis, would encompass in the same formula the movements of the greatest bodies of the universe and those of the lightest atoms. For such an intelligence nothing would be uncertain, and the future, like the past, would be open to its eyes. (Laplace 1825: 2)

  • Yet determinism might be true even if no intelligence could use it to predict.

• Don’t get hung up on these related ideas – determinism is the main foil to freedom.

Identifying Free Will

The Meaning of ‘Free Will’

When I say of a man [sic] that he ‘has free will’ I mean that very often, if not always, when he has to choose between two or more mutually incompatible courses of action—that is, courses of action that it is impossible for him to carry out more than one of—each of these courses of action is such that he can, or is able to, or has it within his power to carry it out. A man has free will if he is often in positions like these: he must now speak or now be silent, and he can now speak and can remain silent; he must attempt to rescue a drowning child or else go for help, and he is able to attempt to rescue the child and able to go for help; he must now resign his chairmanship or else lie to the members, and he has it within his power to resign and he has it within his power to lie. (van Inwagen 1983: 8)

Free Will Defined?

• Sometimes people say that we need to define our terms before we start doing philosophy. This seems generally bad advice, though it’s useful in specific cases where there is evidence of prior equivocation.

• Van Inwagen treats free will as a technical term, and stipulates its definition: having free will involves being able to do either of mutually incompatible acts; i.e., being such that one can do either.

• The stipulation makes use of the circumstantial ability-attributing modal can (van Inwagen 1983: 12–13), and that is not given a definition.

  I do not know how to define ‘can’…. Nevertheless, I think that the concept expressed by ‘can’…—the concept of the power or ability of an agent to act—is as clear as any philosophically interesting concept is likely to be. In fact, I doubt very much whether there are any simpler or better understood concepts in terms of which this concept might be explained. (van Inwagen 1983: 8–9)

• We’ve already noted that can is context-sensitive, and can have a number of different kinds of meaning (circumstantial, deontic, etc.). Van Inwagen tries to pin down his preferred concept by distinguishing it from other nearby concepts (1983: 9–13) – not a definition, but an indication of which uses of can are his focus.

Ability-Can distinguished from nearby notions

• Van Inwagen distinguishes ability from physical possibility.

  • Examples might include: it is physically possible for me to speak French, but I’m not able to; it is physically possible for the prisoner in the indeterministically locked room to leave, but they cannot (suppose the unlocking is very low probability).
  • Consider Hume’s ‘prisoner in chains’ (Hume 1777: §8.23): they could have an ability without any physical possibility of acting on it.

• If ability was just physical possibility, incompatibilism would be trivial (we cannot have a unique physically possible future and two or more distinct future courses of physically possible action). Van Inwagen does not think the compatibility problem is trivial.

• Van Inwagen also distinguishes ability from capacity, contrasting ‘being able to speak French’ and ‘being capable of understanding French’:

  The concept of a … capacity would seem to be the concept of an invariable disposition to react to certain determinate changes in the environment…, whereas the concept of an agent’s power to act would seem … the concept of a power to originate changes in the environment. (van Inwagen 1983: 11)

Worry: can we fix on a particular meaning for can?

• Worry: why think free will is linked to some particular use of can, rather than to can itself? Consider this proposal:

  (FWC)

  Someone has free will with respect to \(A\) iff they can do \(A\) and can refrain from doing \(A\).

• In (FWC) the expression on the right is context-sensitive due to the presence of can – accordingly, so is the expression on the left.

  • Some restriction to the modal base is needed; we wouldn’t be free just in case we are permitted to either do or refrain from doing \(A\). But there is still context-sensitivity in the conversational background, even if we fix on, e.g., a circumstantial modal base.
  • Van Inwagen’s stipulation superficially resembles this, but he considers free will to be introduced by a specific use of the right hand side.

• Being truly ascribed free will is, on (FWC), rather like being truly ascribed tallness.

  • A 1.5m tall person X might be tall for a child and not tall for a basketball player; the sentence X is tall expresses a truth in some contexts and a falsehood in others, depending on which comparison class is conversationally salient.

Contextualism about free will

• According to (FWC) we might happily in some contexts say that an agent has free will, and in others deny it, with no change in the agent or their circumstances – only a change in which aspects of the circumstances are conversationally salient.
  • This, remember, is something like what Lewis says about can in the grandfather paradox (Lewis 1976).
• This is a kind of contextualism about free will, based on the context-sensitivity of can (Jaster 2020; see also Eagle 2011).
  • Other contextualisms about free will have been defended (Hawthorne 2001; Rieber 2006; Grim 2007: 195–97), though the mechanism involved in those cases is different.
• Contextualism might be relevant to the free will debate; it may turn out that, in some contexts, free will is incompatible with determinism expresses a truth; and in others, a falsehood. It may be that some arguments for (in)compatibilism are undermined by this covert context-sensitivity.
  • Eagle (2011: 285–88) offers a parallel discussion for compatibilism about chance and determinism.

Preview of the Argument

The Consequence Argument, first pass

• Grant that free will concerns ability, not physical possibility. If free will doesn’t concern physical possibility, while determinism is a thesis about physical possibility, at first glance we may not see any conflict.

• But van Inwagen offers a connection:

  If determinism is true, then our acts are the consequences of the laws of nature and events in the remote past. But it is not up to us what went on before we were born, and neither is it up to us what the laws of nature are. Therefore, the consequences of these things (including our present acts) are not up to us. (van Inwagen 1983: 16)

  • The link here is the idea of something’s being ‘up to us’: intuitively, free actions are up to us, but determined actions are ‘up to’ their determining prior circumstances.

• Compatibilists, van Inwagen suggests, have often ‘diagnosed’ incompatibilism as stemming from one or another conceptual mistake. The consequence argument is a direct argument for incompatibilism, and rests on no such mistake – or so van Inwagen maintains.

• This argument will be our principal focus next time.

Compulsion and Free Will I

• Many compatibilists claim that the concept opposing liberty is not necessity, but constraint:

  For what is meant by liberty, when applied to voluntary actions? We cannot surely mean that actions have so little connexion with motives, inclinations, and circumstances, that one does not follow with a certain degree of uniformity from the other, and that one affords no inference by which we can conclude the existence of the other. For these are plain and acknowledged matters of fact. By liberty, then, we can only mean a power of acting or not acting, according to the determinations of the will; this is, if we choose to remain at rest, we may; if we choose to move, we also may. Now this hypothetical liberty is universally allowed to belong to every one who is not a prisoner and in chains. Here, then, is no subject of dispute. (Hume 1777: §8.23)

Compulsion and Free Will II

[Some] argue, to do something of one’s own free will is to do that thing without being compelled to do it, and to behave in accordance with a deterministic … law is not to be compelled. I reply that this argument confuses doing things of one’s own free will with having free will about what one does. … Suppose a certain man is in a certain room and is quite content to remain there… Then, I should think, he remains in the room of his own free will. But … it does not follow from a person’s doing something ‘of his own free will’ that he can do otherwise. … the Compatibility Problem is not going to be solved by jejune reflections on compulsion. (van Inwagen 1983: 17)

• The agent in the room does what he wants; but is he free to do otherwise?

Does Free Will Matter?

• Hume accepts Incompatibilism about free choice: ‘liberty, when opposed to necessity, not to constraint, is … universally allowed to have no existence’.
• What Hume denies is that free choice is anything valuable for human agents. What’s important is the freedom to act on your intentions, without constraint.
  • For the ability to freely choose to do otherwise will, in many cases, involve the ability to intend things out of keeping with your character and prior history? And why would that be a sort of freedom worth having?
• On this view, the person content to remain in a locked room is doing what he wants, and thus has all the freedom that is desirable.
  • But suppose you lack freedom; I subsequently arrange matters, by brain surgery or the like, so that you want to do only what you will do regardless.
  • Now you do only what you want, because I have crippled your desires to match your abilities. How could that double loss be a success in restoring your free will?

Why Free Will Matters: Moral Responsibility

• For van Inwagen, free choice is tied to humanly valuable things:

  1. Deliberation: ‘one cannot deliberate without believing that the things about which one is deliberating are things it is possible for one to do’ (van Inwagen 1983: 19);
  2. Moral responsibility: ‘moral responsibility is possible only if we have free will’ (van Inwagen 1983: 20).

• Deliberation would be possible even if we lacked free will, so long as we continue to believe in it. Hence van Inwagen’s principle focus is responsibility:

  If we do not have free will, then there is no such thing as moral responsibility; therefore, since there is such a thing as moral responsibility, there is such a thing as free will. (van Inwagen 1983: 22–23)

• Accordingly, determinism is false.

• Is an argument based on these premises really capable of overturning the deliverances of science?

• Deciding what to believe in any philosophical dispute is often a matter of accounting; and if the costs shift (if science tells us that determinism is true, rather than suggesting it to be false, for example), then which further theses we accept may shift too.

Indeterminism and Free Will

• Yet the threat to free will that derives from the basic problem we started with remains live:

  However physics turns out, there will be a complete description of physical reality that never mentions agents, or agency, at all; and hence agents can never play the sort of explanatory role in understanding the origination of various courses of action that the postulation of free will seems to strive for. Agency falls out of the picture, unless it is reducible to the collective behaviour of non-agential entities. The consequence argument would then show that since we lack control over those things, we are cannot be free, simpliciter. So something’s wrong with the consequence argument.

• Van Inwagen’s response comes when he discusses what he calls the ‘Mind argument’; we look at this, and moral responsibility, in lecture 11.

Fatalism

Fatalism Defined

Fatalism is … the thesis that it is a logical or conceptual truth that no one is able to act otherwise than he in fact does; that the very idea of an agent to whom alternative courses of action are open is self-contradictory. (van Inwagen 1983: 22)

• Fatalism is the thesis that there is an conceptual incoherence in the idea of being able to do otherwise, which means that free will is internally inconsistent.
  • If true, we don’t need to do any empirical investigation into whether the laws of nature are deterministic to find out that we lack free will. We could find it out merely by armchair reflection on the concept of being able to do otherwise than we in fact do.
  • As van Inwagen defines free will, fatalism entails that we lack free will, but not vice versa. We could lack free will even if fatalism were false – suppose, for example, that it is an empirical and contingent truth that everything we do, we must do. PvI’s view is that, if determinism is true, it would be true only contingently, so it would rule out free will only contingently.

Inevitability

• Fatalism is sometimes used for the thesis that everything is inevitable, that Fate has certain outcomes in store for us, no matter what.
• Inevitabilism, in this sense, is not fatalism as van Inwagen defines it.
  • In the vignette Appointment in Samarra (van Inwagen 1983: 24), there are multiple courses of action open to the servant: he can stay in Baghdad, or he can go, as he does, to Samarra. The fact that each course of action leads to the same inevitable outcome doesn’t mean there aren’t multiple courses of action.
  • By contrast, someone’s action can be fated in van Inwagen’s sense without the consequences of that action being inevitable. Suppose that I will go to the SA Museum tomorrow; fatalism entails I cannot do anything else, and one consequence is that I will be in Adelaide tomorrow. Yet had I got on a plane to Melbourne I could quite easily have not been in Adelaide tomorrow. So while my being in Adelaide is fated, it is not inevitable. (It is not inevitable because there is a true counterfactual had I got on the plane to Melbourne, I wouldn’t have been in Adelaide, and the antecedent is possible, even though I can’t do it because I won’t.)
    
• There are no serious arguments for inevitabilism.

A Sophistical Argument for Inevitabilism

Let (E) be any event that might happen. Consider the theorem of logic ‘\((q \to p) \vee (r \to ¬p)\)’. In virtue of this theorem, one of the other of the following two propositions must be true:

If I try, by any means whatever, to prevent \(E\), \(E\) will happen;

If I try, by any means whatever, to bring about \(E\), \(E\) will not happen.

Therefore, either \(E\) or the non-occurrence of \(E\) is strongly inevitable for me. (van Inwagen 1983: 28)

• The truth of these conditional propositions is insufficient for inevitability: they fail to establish the required counterfactual claim if I had tried to prevent \(E\), \(E\) would still happen.
  • At best, these give: if \(p\) will happen, then it will happen together with whatever I do. But this is not to say that it would happen no matter what.
  • This argument really tells us more about the material conditional \((\phi \to \psi)\) than about inevitability – the argument rests on the fact that if \(\psi\) is true, so is \((\phi \to \psi)\), regardless of any connection between them. Consider: I will go cycling tomorrow; therefore, if I die tonight, I’ll go cycling tomorrow.

The Consequences of Fatalism

Anyone who accepts fatalism must regard all ascriptions of moral responsibility as incorrect, and must, on pain of self-contradiction, refrain from deliberating about future courses of action. (van Inwagen 1983: 30)

• This is derived from van Inwagen’s ideas about the consequences of lacking free will, which we will come to discuss later.
• But in any case, it is clearly important in ascriptions of responsibility, to know what the abilities of the agents involved are; and certainly the fact that I was unable to do otherwise than I did would be a mitigating circumstance. So fatalism would give everyone an excuse about everything.

Arguments for Fatalism: Permanent Truth

The Argument for Fatalism from Permanent Truth

Let us consider some proposition about the future [\(S\)]. … [Suppose] \(S\) is true. That is, \(S\) is true now. But it would seem that if a proposition is true at some particular moment then it must be true at every moment.… If \(S\) is true, \(S\) is unchangeably true. … But if \(S\) is unchangeably true, then I cannot render \(S\) false, for the same reason that, if a certain king is unchangeably powerful, I cannot render him helpless. But if I cannot render \(S\) false, then I cannot refrain from shaving tomorrow morning; for if I could refrain from shaving tomorrow morning, then I could render \(S\) false. (van Inwagen 1983: 34)

• What is an unchangeable truth? It is a proposition that is true at some moment, and not possible that there be two moments where that proposition differs in truth value (i.e., it is not possible to change the truth of the proposition, whatever it is; and it is in fact true).

Unchangeable Truth?

• Van Inwagen is a propositional eternalist (recall our discussion of temporalism). For any proposition \(S\), it is not true only relative to some times:

  there are possible worlds in which \(S\) is always true, and possible worlds in which \(S\) is always false, but there are no possible worlds in which \(S\) is at one time true and at another false. (van Inwagen 1983: 34, cf. §2.4)

• Van Inwagen thus accepts the characteristic B-theory claim that all truths are unchangeably true. This is because, according to the B-theory, necessarily, truths are permanently true.

  • In light of our earlier discussions of change: a changeable truth is one that literally could vary in truth value over time in a single world, and necessarily, no proposition does that.

• But how is fatalism supposed to follow from this? Only if I cannot render \(S\) false follows from \(S\) is unchangeably true.

  • To accept that is to think that every truth has its actual truth value necessarily follows from every truth has its current truth value always. On the face of it, that doesn’t seem to follow.

A Fallacious Argument

1. \(S\) is true, and permanently true.

2. It is not possible that both (1) and that I render \(S\) false through my actions.

3. It is not possible that I render \(S\) false through my actions. (1, 2)

4. I cannot render \(S\) false through my actions, i.e., fatalism is true. (from 3, given that impossibility \(\to\) inability.)

• But this argument embodies a modal fallacy at step (3) (van Inwagen 1983: 40): \[p, ¬\lozenge(p\wedge q) \nvDash \neg\lozenge q.\]
  • This argument form is exemplified in this obvious howler: I am in my office; it is not possible that (I am in my office and not in my office); therefore, it is not possible that I am not in my office.

Why Does This Argument Tempt Us?

• One reason why this fallacy is tempting relates to the context-sensitivity of modal expressions.
• The most general meaning we can assign to the word possible is this: \(p\) is truly said to be possible in context \(c\) iff \(p\) is consistent with the facts that are relevant in context \(c\).
• This makes possible susceptible to shifts in meaning as context changes. In particular, as a conversation goes on, and claims are added to the accepted conversational background, what can be said to be possible is more and more restricted.
• So there is a reading on which (3) is true: Given the facts in the conversational background, including \(S\), it is not consistent that \(S\) be false. But this is a pragmatic phenomenon, derived from the dynamics of conversation.
  • It’s ‘conversationally necessary’ that \(p\), but that’s not the kind of necessity that would undermine my abilities.
• If we don’t equivocate, and continue to use the word possible in the sense of premise (2) – something like logically possible – then (3) does not follow from the premises.

The Time of Truthmaking

You admit that \(S\) was true yesterday. But it’s not now up to you what was the case yesterday: it is not within your power to change the fact that \(S\) was true yesterday. And it is a logical consequence of \(S\)’s having been true yesterday that you will shave tomorrow. Therefore, it is not within your power to change the fact that you will shave tomorrow. (van Inwagen 1983: 40)

• But, on the face of it, part of what was the case yesterday is up to me: those parts of what was the case yesterday that concern yesterday’s future!
• Given the permanence of truth, it cannot be that we change what was true to being false. Rather, instead ‘what follows is that it is within my power to make it the case that \(S\) was always false’ (van Inwagen 1983: 42).
• If \(S\) concerns the events of tomorrow, then the permanent truth of \(S\) is made true by what happens tomorrow. The ‘past truths’ are just the truths; some of them are grounded in the past, some in the future. If I can influence the future, I can influence what’s true, and hence influence the past truths.
• This isn’t any kind of magical interference with the past, nor any sort of mysterious determination of the future. The thing I will do ‘is an irrelevant fact about the future masquerading as a relevant fact about the past’ (Lewis 1976: 151).

Arguments for Fatalism: Taylor’s Necessary Condition Argument

Taylor’s Principle (A) and its Consequences

1. ‘No agent is able to perform an act in the absence of a condition necessary for its accomplishment.’ (van Inwagen 1983: 44)

5. Either there will be a sea battle tomorrow or there will not. (premise)

6. If there is a sea battle tomorrow, then the admiral cannot avoid a sea battle. (from A)

7. If there is not a sea battle tomorrow, then the admiral cannot induce a sea battle. (from A)

8. Therefore, either the admiral cannot avoid, or cannot induce, a sea battle. (from 5, 6, 7, given \(p\vee q, p \to r, q \to s \vDash r \vee s\).)

Premise (5): the Law of Excluded Middle

• Perhaps what is wrong is the law of excluded middle, appealed to in premise (5).

  What is of direct relevance to the problem of fatalism is the question whether LEM ‘applies’ to propositions ‘about the future’. (van Inwagen 1983: 52)

• Since we accept that \((p \vee \neg p)\) is a truth of logic, the only way out is to deny that there are any propositions about the future, else logic applies to them. But ‘surely someone may believe or assert or even know’ a sentence about the future, and ‘that is what it is for a sentence to express a proposition’ (van Inwagen 1983: 53).

• Moreover: do not confuse an unwillingness to assert \(p\) and an unwillingness to assert \(\neg p\) with an unwillingness to assert \((p \vee \neg p)\).

An Initial Response

Taylor’s (A) seems dubious … it seems to yield the fatalist conclusion a bit too easily. For supposing that I do not perform act \(S\) (whatever it is), then it follows immediately that there is lacking a necessary condition for my performing \(S\), namely the occurrence of \(S\). So, if (A) is right, it follows immediately that I never have the power to perform any act which I do not actually perform. … one might suggest that not only is this presupposition false, but that it seems true because it is easily confused with another much more plausible proposition:

(A*)

No agent can perform any given act if there is lacking, at the same time or any other time, some condition which is necessary for the occurrence of that act and which it is not in his power to bring about. (Rice 2018: §3.2)

• Maybe we confuse (A) with (A*); but regardless, is there a reason to think (A) is true?

Ambiguous Ability Ascriptions

Ability sentences containing adverbs or adverbial phrases are often ambiguous. Consider:

I can refrain from talking at any time. …

It will be convenient for us to introduce a more explicit device … for the disambiguation of ability-sentences. Let us use round brackets in this fashion:

I can (refrain from talking) at any time;

I can (refrain from talking at any time). (van Inwagen 1983: 45–46)

• That is, do I have at each moment the ability to be quiet at that moment, or the ability to be permanently quiet, quiet at all moments?
  • The latter ability entails the former, but not vice versa.

Principle (A) is ambiguous

• Principle (A) has two readings, in this bracket notation:

A(1)

No agent is able to (perform an act) in the absence of a condition necessary for its accomplishment.

A(2)

No agent is able to (perform an act in the absence of a condition necessary for its accomplishment).

• Clearly A(2) is true, but which is involved in the argument?
• It must be A(1) – for only A(1) would allow us to draw a conclusion at (8) about whether the admiral can avoid a sea battle.
  • Applying A(2) would only get this, from which (8) doesn’t follow:
  9. If there is a sea battle tomorrow, then the admiral cannot (avoid a sea battle in the presence of a sea battle).

Is Principle A(1) Correct?

The only argument Taylor offers in support of (A) … consists simply in exhibiting instances of situations in which he is unable to perform some act…:

I cannot, for example, live without oxygen, or swim five miles without ever having been in water, …, or win a certain election without having been nominated…. (van Inwagen 1983: 49)

• These examples might support A(2). Some might support some instances of A(1). But they don’t seem to support A(1) in general.
  • Take the election case: Trump cannot (win an election without being nominated). But Trump might be able to win an election; many Trumpistas will argue that Trump should become the candidate, precisely because he is able to win – an ability he has even now, before he’s been nominated.
• Important to note: ‘I am not arguing that fatalism is false; rather that a certain argument does not show that it is true’ (van Inwagen 1983: 49).

References