Freedom and the Self: Essays on the Philosophy of David foster wallace

FATALISM AND THE METAPHYSICS OF CONTINGENCY

M. ORESTE FIOCCO

CONTINGENCY AND THE SPECTER OF FATALISM

Contingency is the presence of nonactualized possibility in the world. Given contingency, the world as it actually is is incomplete not in that there are features of reality beyond those that actually exist but in that there could be. If there is contingency, a systematic metaphysics need provide some account of this possibility, including its source and relations to other features of the world. However, fatalism is a view of reality on which there is no contingency. On this view, the world as it actually is is entirely complete: every detail must be just as it is, and there neither is, nor could be, anything beyond this. Hence, there is no need to account for the possibility in the world, for there is none.

It is contingency that permits agency, for if nothing were possible beyond what is actual, every feature of reality would be brute, in that it would have to be just as it is. There would be no distinction, then, between mere happening and intentional behavior. An action, the result of intentional behavior, seems to require—at the very least—being amenable to a certain sort of explanation, one in terms of the mental states of a conscious being. Yet any such explanation is void if every feature of the world simply must be as it is. Hence, in a world in which there is no contingency, there is no place for the direction of an agent or for agency itself. Fatalism is inimical to a natural view of oneself as an active being capable of responding to and freely contributing, even if only slightly, to what the world is like.

In light of this, there has traditionally been much interest in contingency. Such interest has long been embarrassed by the contention that simple and plausible assumptions about the world lead to fatalism. At the outset of Western metaphysical thought, Aristotle presented reasons for thinking that contingency requires the world in time to be a certain way and that this limits what is true of the world. I largely concur with Aristotle—most, however, do not. It is worth examining the Aristotelian considerations in a contemporary context, for what emerges are two incompatible accounts of contingency, two very different ways of situating possibility in the world.

I begin with the Aristotelian argument in the modern form in which it is presented in Richard Taylor’s ruminations on fatalism. Appreciation of this argument has been stultified by a question pertaining to the source of necessity and possibility and a closely related one regarding the nature of metaphysics itself; this can be seen via the criticism of Taylor by his contemporaries. With these questions addressed, the heart of the matter—necessity and possibility in a temporal world—comes to light. This issue is investigated through an important later criticism of Taylor by David Foster Wallace. Wallace’s critique is significant because it brings to the fore the crucial notion for understanding contingency in a temporal world, that of synchronic possibility, the idea that incompatible states of affairs are possible at a single moment. This notion provides the basis of distinguishing two systematic accounts of truth, modality and time: two metaphysics of contingency. On one account, Taylor’s Aristotelian argument is straightforwardly valid and compelling; on the other, it is fallacious. In closing, I present reasons why the former account, supporting the Aristotelian views of time and truth, is correct and make some comments to ameliorate this conclusion.

FATALISM AND TAYLOR’S ARGUMENT

Aristotle’s (1987, 17–19) argument that contingency entails perhaps surprising views of the world in time and of truth is compressed and, largely for this reason, cryptic. Richard Taylor’s modern presentation of essentially the same argument is more elaborate.

Taylor’s Premises

Taylor contends that “presuppositions made almost universally in contemporary philosophy yield a proof that fatalism is true” (Taylor 1962a, 42). This proof rests on six premises. Three of these are merely definitional. Nevertheless, they are controversial, for they introduce the modal notions employed in Taylor’s discussion, and these are vexed (as discussed in “Metaphysics and the Source of Necessity and Possibility,” below).

The first of these definitional premises is that if something, some state of affairs or condition or feature of the world, in itself guarantees the existence of another, then the first is sufficient for the second, that is, it must be the case that if the former exists, so does the latter. Similarly, and this is Taylor’s second premise, if some state of affairs, S₁, requires the existence of another, S₂, that is, if it is impossible that S₁ exist without S₂, then S₂ is necessary for S₁. The third premise follows from these two: if some state of affairs, S₁, is sufficient for another, S₂, then S₂ is necessary for S₁.

With these premises, Taylor takes himself to be explicating the standard notions of a necessary and a sufficient condition. Yet he seems to recognize that his application of the notions is somewhat atypical, so he provides further comments to illuminate the relations based on them. He states their relata need not exist at the same moment and that the relations are not “logical,” that is, conceptual, so they can hold between things in the world and not merely one’s ideas or representations of such things.1 Moreover, Taylor makes clear that the relations are neither physical nor nomological and, in particular, have nothing to do with causation. Thus, for example, one state of affairs being a sufficient condition for another is no indication that the former is a cause of the latter or connected to it by any law of nature. Taylor is explicit that his argument makes no recourse to physics (1962a, 42) and that his conclusions with respect to contingency are made “without any reference to causation” (48).

The other premises of Taylor’s argument are more obviously substantial, pertaining to truth, the necessary connections between things in the world, and the nature of the world in time. The first is an assumption of bivalence: every proposition whatsoever is either true or, if not true, false. The second is presented as the claim that no agent can do something if some necessary condition for that action is lacking. Although this makes clear the relevance of Taylor’s discussion to concerns about agency, the point can be generalized to draw out its deeper metaphysical consequences: no state of affairs or condition or feature of the world can occur if some necessary condition for that state of affairs does not exist.

The premise about temporal reality is the least perspicuous. As Taylor states it, it is that “time is not by itself ‘efficacious,’” (1962a, 44) and immediately explains that by this he means that “the mere passage of time” (44) does not affect the capacities of a thing. So if a thing loses some capacity, this is because it has undergone some change in its nature and not merely because the moment at which it exists has gone, for instance, from future to present or present to past. The crux of the premise is, then, that when a thing exists does not by itself affect the nature of that thing. Thus, Taylor seems to be assuming a familiar view of the nature of temporal reality, one on which it is ontologically homogeneous, that is, the view that there are many moments of time and that all these moments, and the things existing at them, are equally real.2 This interpretation of Taylor’s premise is corroborated by comments he makes later in his discussion,³ most clearly when he states that fatalism follows not from

the mere temporal relations between… states of affairs, but the very existence of those states of affairs themselves; and according to our first presupposition [that every proposition whatsoever is either true or, if not true, false] the fact of tomorrow’s containing, or lacking, [some state of affairs], as the case may be, is no less a fact than yesterday’s containing or lacking one.

(Taylor 1962A, 48)

The Argument

Given these premises, Taylor, alluding to Aristotle, presents a picturesque argument that there is no contingency. He imagines that he is a naval commander and that his order is a sufficient condition for a sea battle tomorrow; likewise, a different order is sufficient for there to be no battle. Stripped of extraneous detail, Taylor’s argument is this: a given state of affairs, s (e.g., an order for battle’s being given), at an imminent moment, m₁, is a sufficient condition for a distinct state of affairs, n (e.g., a sea battle), at some subsequent moment, m₂. The absence of s at m₁, that is, the existence then of the incompatible state of affairs, −s, is a sufficient condition for a distinct state of affairs, −n, which is incompatible with n, at m₂. The assumption that at m₁ both s is possible and −s is possible is untenable. This is because it is supposed that the proposition that n occurs at m₂ is true or, if not true, false; if it is false, then the proposition that −n occurs at m₂ is true. Consequently, it is supposed that at m₂ either n occurs or −n does. If n occurs, then −n cannot, in which case −s cannot occur (because −n is a necessary condition of −s); on the other hand, if −n occurs, then n cannot, in which case s cannot occur. Thus, either s or −s is impossible at m₁. One of the two occurs at m₁, and whichever actually does, that state of affairs must occur. This contravenes the possibility of a nonactualized state of affairs at m₁. Since there is nothing special about these moments or states of affairs or the relations in which they stand, this argument demonstrates that given Taylor’s premises there is no contingency.

Indeed from Taylor’s assumption regarding the ontological homogeneity of temporal reality, it is clear that it follows that any state of affairs is related to some other in such a way as to make the above argument applicable: The existence of some state of affairs, s, at some moment is a sufficient condition for the subsequent existence at m_w, say, one week later, of a distinct state of affairs, n, namely, the state of affairs that s occurs one week prior to m_w. The argument, therefore, has purchase on the world, and it is perplexing, for it demonstrates that there is no contingency—something that seems to be an immediate and, hence, indubitable feature of the world—from premises all of which are familiar and not without appeal.

Taylor’s Conclusion

Taylor, however, does not accept fatalism. This should be clear from his own response to the foregoing argument or, if not from this, from his body of work.4 It is odd, then, that he is often thought to be a fatalist (by, for example, Bruce Aune [1962, 69] and Wallace [1985, 143]). Rather, following Aristotle, Taylor concludes that given contingency, one must reject at least one of the premises of the fatalistic argument. So he denies that every proposition whatsoever is either true or, if not true, false. Denying this allows him to reject the claim, crucial to the argument, that the proposition that n occurs at m₂ is true or, if not true, false.

Taylor recognizes that rejecting such claims about the future requires a revision of one’s account of temporal reality. If it is not the case that the proposition that n occurs at m₂ is true or, if not true, false, then—assuming that any true proposition is grounded in some feature of reality5—it is not the case that either n or −n occurs at m₂. It presumably is the case, however, that for any state of affairs, either it or its contradictory occurs or has occurred (suspending, for the nonce, any concerns about vagueness). Therefore, with respect to the existence of states of affairs, reality as it is subsequent to this moment has a different status from the way it is (or was); the ontological homogeneity of temporal reality is false, so when a thing exists does by itself affect the nature of that thing. Taylor acknowledges that these premises about truth and the world in time are “inseparably linked, standing or falling together” (Taylor 1962a, 51).

I think the rejection of these two related assumptions about truth and temporal reality is, in the end, the appropriate response to Taylor’s Aristotelian argument. Most, however, disagree. Many believe Taylor’s argument is fallacious and maintain that there is no need to revise the familiar assumptions on which it is based. I think such criticism is mistaken, though not unreasonable. Taylor leaves some key assumptions tacit, and this has misled his critics. These assumptions need to be articulated to appreciate the force of the argument.

METAPHYSICS AND THE SOURCE OF NECESSITY AND POSSIBILITY

Many of those who have been critical of Taylor, especially his contemporaries, have simply misunderstood his project and the basis of his Aristotelian argument. The misunderstanding concerns the modality—the source of necessity and possibility—pertinent to the argument and the very conception of metaphysics associated with this modality. Confusion on these grounds has led several philosophers to make criticisms that fail even to engage with the argument.

The Notion of a Modality

A modality is a basis of related conditions from which to qualify a claim’s truth. Given this basis, a claim must, could not, or could be true; that is, relative to a modality, the claim has the status of being necessary, impossible, or possible. More generally, a modality determines certain relations in which things stand; hence, given a particular basis of related conditions, two things are necessarily connected (i.e., the one necessitates or entails the other), or incompatible (i.e., the one precludes the other), or compatible with each other.

If there is contingency, there are claims that, despite not representing what actually is the case, could be true; if there is no contingency—if fatalism is true—there are no such claims. The question of what modality is relevant here is of the utmost importance. Examples of different modalities are: the rules of a natural language, the class of concepts by which one identifies things and characterizes the world, the natural laws governing the material world, the laws of logic, the natures of things in themselves, the claims one takes oneself to know. Thus, given the rules of English, the claim that all bachelors are unmarried must be true; given (the current understanding of) the natural laws of interest to physicists, the claim that there is a vehicle that travels faster than light could not be true. Given a traditional account of concepts, the claim that there is water that is not H₂O could be true. Similarly, given the natural laws of interest to physicists, combustion necessitates oxygen; given these laws, combustion is compatible with wood.

Two Understandings of Metaphysics

When Taylor’s argument appeared in the early 1960s it was orthodox—absolutely taken for granted—that the only modalities pertinent to philosophy were linguistic: those of logic, as codified by different formal languages, and the rules of a natural language, including the concepts associated with these. Of course, the modality of the laws of nature was legitimate, but it was the domain of natural scientists.

This orthodoxy reflects a particular understanding of the world, one deriving from a number of sources, notably Hume and Kant, and the impetus for the so-called linguistic turn of mainstream analytic philosophy through most of the twentieth century. On this understanding, much of the basal structure of reality—what kinds of things exist, the natures of individual things, the relations in which they stand—is a result of the interaction with the world of the minds of conscious beings. In particular, the structure arises from how conscious beings react to a differentiated yet amorphous world, how they think about it and express these thoughts via language. Precisely how much worldly structure arises in this way is a profound and disputatious question; however, what is supposed not to be controversial is that since much of this structure comes from the activity of conscious beings, necessary connections—and, consequently, what is compatible and incompatible—are reflections of the workings of mind. They are relations of ideas or concepts or the expression of such in language: hence, the only legitimate metaphysical modalities are linguistic. From this understanding comes a particular conception of metaphysics, one on which its purpose is to make clear the linguistic-cum-conceptual rules by which reality is structured. These rules and their relations, which are supposed to be accessible to any competent thinker or speaker, are to be illuminated by a process of reflection and analysis.

In opposition to this is a quite different understanding of the world, one rooted in Aristotle and decidedly unfashionable in the twentieth century. On this understanding, the basal structure of reality is entirely independent of the minds of conscious beings: there are many kinds and individual instances of these kinds that have natures that have nothing to do with the response to them of minds; they are as they are in themselves and relate as they do because of their natures. Necessary connections and what things are compatible or incompatible are determined by these mind-independent natures. Hence, there is a modality—the one composed of the natures of things in themselves—that is in the world yet has nothing to do with language or concepts. From this understanding arises a different conception of metaphysics, one on which its purpose is to make clear how things are in themselves and how these things relate. To be sure, this understanding raises deep epistemological questions (the most pressing of which is how the mind-independent nature of a thing can be known, when that thing can only be known via the mind), but this sort of question is posterior to the metaphysics.

The Criticism of Taylor’s Contemporaries

The understanding of the world underlying Taylor’s argument is the Aristotelian one. This is clear from his acknowledgment of necessary and sufficient—and essential—connections that are neither “logical” nor physical (1962a, 42, 43). I think there are compelling reasons to embrace this understanding, but for present purposes I merely accept it as correct. Given this understanding of the world, Taylor presumes that the modality relevant to metaphysical investigation is one based on the natures of things in themselves. This modality is not linguistic; moreover, despite not being based on causation nor any laws of nature, it is nonetheless fully in the world. It is, then, from this basis that Taylor argues that his familiar premises lead to fatalism. As noted above, this understanding of the world and the conception of metaphysics that accompanies it were at odds with the views prevailing when Taylor’s argument appeared.

This argument purports to show that there is no contingency and, therefore, no agency and, a fortiori, no free action. Presumably since it is concerns about one’s personal freedom that make urgent recondite metaphysical discussions of contingency, Taylor presents the Aristotelian argument with a focus on its consequences for one’s freedom. As a result, critical discussion of it has had this focus and so has been couched in terms of agency and free will rather than in the more general ones of contingency and its conditions.

The upshot of Taylor’s argument is that at some imminent moment one of the states of affairs in every contradictory pair, s and −s, is impossible. Suppose it is s. The grounds that Taylor adduces for the claim that s is impossible is that a necessary condition for s, some state of affairs, n, does not occur. Since n is necessary for s, s is impossible because some necessary condition of it does not exist. If s is impossible, an agent lacks the power or ability to bring it about: one cannot do what is impossible.

One of the first criticisms of Taylor’s argument, leveled by John Turk Saunders (1962a), is that it does not follow from the fact that a necessary condition of some state of affairs does not exist that one lacks the power or ability to bring that state of affairs about. So just as having in one’s hands a violin is a necessary condition for playing a violin, from the fact that one does not have a violin in one’s hands it does not follow that one lacks the power or ability to play. One has this ability—and so it is not impossible that one play—it is just that at some moments circumstances prevent the ability from being exercised. Accordingly, it is incorrect to conclude, merely on the grounds that a necessary condition for it fails to exist, that s is impossible. Insofar as one has the ability to bring about s and the ability to bring about −s, both are possible. Thus, there is, pace Taylor, contingency.

Taylor (1962b, 57) concedes Saunders’s point: one can have the power or ability to bring about s despite a necessary condition for s failing to exist. Yet he disagrees with Saunders about its significance. Saunders maintains that it follows conceptually, or from the rules of English, that if an agent has the ability to bring about s (at moment m), then s is possible at m. Taylor holds that an agent might have such an ability and that it nevertheless be the case that s is impossible at m. Despite the ability, the possibility of s might be precluded by other features of the world.

Taylor’s position leads Saunders (1962b) to charge that he is merely abusing language.6 However, the real bone of contention is not linguistic; rather the issue here is fundamental disagreement about the modality operative in Taylor’s argument. Saunders is presuming the relevant modality is linguistic-cum-conceptual; given the English language, the attribution of an ability to bring about s necessitates the possibility of s. Yet Taylor is basing his claims about what is possible on nonlinguistic features of the world; given these features, viz., the absence of a necessary condition for s, s is impossible—despite anyone’s abilities. With such disagreement, Saunders fails even to engage Taylor’s argument and so provides no reason for thinking it faulty.

Similar failure to engage with Taylor’s Aristotelian argument—arising from the prevailing view that the only modalities pertinent to philosophy per se are linguistic—can be found in all the best-known critiques of Taylor by his contemporaries. Thus, Raziel Abelson (1963) maintains that Taylor “systematically equivocates” between two senses of modal terms like “necessary,” “sufficient,” “can,” etc. Abelson assumes that these notions are either “logical,” i.e., pertaining to language and concepts, or “causal,” i.e., pertaining to the causes of things in the extralinguistic world.

According to Abelson, then, Taylor begins with the linguistic notion of necessity when he claims that some state of affairs, s, is a sufficient condition of another state of affairs, n. Given this notion, it is clearly true, if n, a necessary condition of s, does not occur, it is impossible that s does. If being unmarried is a necessary condition for being a bachelor, and one is married, then it is impossible that one is a bachelor. This is just a linguistic-cum-conceptual point and clearly holds. Yet Taylor goes on to make claims about states of affairs that are “logically unrelated”—states not connected by linguistic rules or concepts—and so it seems the modal terms in his argument cannot be construed in the “logical” sense.

If, however, the terms are not construed in this sense, Abelson maintains that Taylor cannot legitimately derive the strong conclusion that s is impossible if n does not exist. Abelson assumes that in the causal sense, if a necessary condition of s does not exist, all that follows is that s does not occur, not that s is impossible. If one does not have a violin in one’s hands at a given moment, one does not play; it is not impossible that one play at that moment, for one might have picked up a violin just prior to it.

So Abelson accuses Taylor of equivocating, of deriving a strong modal conclusion from a modal notion too weak to support it. What Abelson fails to see is that Taylor is considering a modality that is neither logical nor causal. Taylor maintains that with this modality, one based on features of the world that are not merely physical or causal, the lack of a necessary condition for a state of affairs indeed renders that state impossible. Abelson simply fails to recognize the modality pertinent to Taylor’s argument and, hence, just like Saunders, fails even to engage it.

Considerations very similar to Abelson’s underlie the critique of Taylor by Charles D. Brown (1965). Relying on his ear, on what sounds colloquial, Brown draws a distinction between necessary conditions of, which are purely logical relations, and necessary conditions for, which are causal or physical and so relate features of the world. Taylor denies that the relations pertinent to his argument are logical or conceptual, so Brown maintains he must be using the causal notions. (Of course, Taylor explicitly denies his discussion has anything to do with causation—see above.) Yet if this is the case, Brown alleges, Taylor’s conclusion that the absence of a necessary condition for a particular state of affairs renders that state impossible must be incorrect.

Brown allows that there can be states of affairs that are necessary conditions of prior states of affairs, but such relations, Brown assumes, must be purely logical. What Brown denies is that a subsequent state of affairs can be a necessary condition for a prior state of affairs. A subsequent state of affairs, n, cannot be a necessary condition for a prior one, s, since n occurs after s: n can exist and still not causally necessitate some preceding state of affairs, for an effect cannot precede its cause. Thus, s could occur or it could fail to occur regardless of n. It is, however, crucial to Taylor’s argument that a state of affairs, s, at an imminent moment is impossible because a necessary condition for it, namely the existence of state of affairs, n, is lacking at a later moment. Thus, Brown concludes, Taylor’s argument is fallacious.

What determines whether a certain state of affairs, s, occurs are the causal conditions prior to it—what happens subsequently to s, according to Brown, has nothing to do with whether s occurs and, thus, with whether s is possible. Taylor, however, is arguing that features of the world subsequent to s can render s impossible. But the relevant modality is not causal (nor conceptual), and Brown simply neglects to consider any other modality—indeed, Brown states: “I fail to see how ‘necessary conditions for’ can be legitimately interpreted other than causally” (1965, 129). So, yet again, there is a failure to engage with Taylor’s argument.

In sum, Taylor’s Aristotelian influences led him to recognize and take seriously a modality based on features of the world independent of language or concepts. Hence, one of the key assumptions of Taylor’s argument that he leaves tacit is that the modality relevant to it is based on the natures of things in themselves. Such a basis seems to have been alien to his contemporaries; so strange indeed that the critiques appearing soon after Taylor’s argument failed even to engage—let alone debunk—it.7

NECESSITY AND POSSIBILITY IN A TEMPORAL WORLD

After much attention in the first few years following its appearance, Taylor’s fatalistic argument was less discussed. However, such a straightforward argument with a conclusion inimical to so familiar a conception of the world and one’s place in it was hardly forgotten. Two decades after the spate of publications responding to Taylor, David Foster Wallace, dissatisfied with previous attempts to debunk the argument, made his own. Wallace’s discussion is sophisticated and precocious yet still fails really to engage Taylor’s Aristotelian argument. Nevertheless, it is important because Wallace’s critique brings to the fore the crucial notion for illuminating not only Taylor’s argument but contingency itself.

Wallace on Taylor

By the 1980s, when Wallace was considering Taylor, metaphysical inquiry had been sufficiently revived—through the efforts of Saul Kripke, Hilary Putnam, Alvin Plantinga, Roderick Chisholm, David Armstrong, and David Lewis, to name a few—that it was no longer simply taken for granted that the relevant modality in a philosophical context was linguistic. Consequently, Wallace recognizes the need at the outset of his critique to get clear on the modality relevant to Taylor’s discussion. He determines that since Taylor obviously takes himself to be talking about the world itself—the realm of agency and free action—rather than how one speaks or thinks of it, the modality is not “logical,” that is, linguistic-cum-conceptual. He states, then, that the modality and the “relations treated of here by Taylor must be regarded as physical and causal, not logical” (Wallace 1985, 147). As pointed out above, though, Taylor is explicit that his argument does not rely on any claims about the physical world and has nothing to do with causation. Wallace is not unaware of this. Yet he insists that the relevant modality is physical-cum-causal “even though Taylor maintains, confusingly, that ‘Our problem has been formulated without any reference whatever to causation’” (147).

Wallace’s procrustean interpretation seems to come simply from his failure to recognize a modality that is fully in and about the world yet not based on conditions imposed by causality or physical relations in the material world (147–148). It is such a modality, based on the natures of things in themselves, that is operative in Taylor’s discussion. Wallace, therefore, like Taylor’s contemporary critics considered above, fails to appreciate from the start the nature of Taylor’s project. Nonetheless, he examines, with more determination than his predecessors, modal connections that are genuinely in the world, and this enables one to discern the crux of Taylor’s Aristotelian argument.

Wallace sees that the modality relevant to Taylor’s discussion is supposed to be in the world, yet he is unable to recognize the Aristotelian understanding of reality and conception of metaphysics motivating Taylor. This leads Wallace to question how exactly the states of affairs in Taylor’s argument are necessarily connected, and this, in turn, leads him to take up Brown’s criticism of Taylor. Like Brown, Wallace thinks Taylor’s argument is fallacious ultimately because Taylor conflates different notions of a necessary condition.

As discussed above, Brown holds there is a distinction between being a necessary condition for and being a necessary condition of; the former is a straightforward physical relation, the latter merely logical or conceptual. Wallace maintains that things in the extralinguistic world may be necessary conditions of other such things but concurs with Brown that the two conditions impose different temporal relations on their relata. Whereas if n is a necessary condition for s, s and n must either occur simultaneously or n must occur prior to s; if n is a necessary condition of s, s must occur prior to n. To illustrate: if n (the presence of fuel) is a necessary condition for s (combustion occurring), n must occur prior to s—or the two states of affairs must be simultaneous, but if n (a sea battle’s occurring) is a necessary condition of s (an order for a sea battle’s being given), s must occur prior to n.

As Wallace interprets Taylor’s argument, it requires one to consider whether a state of affairs, s, is possible in the absence of a necessary condition that occurs subsequent to it. Wallace, in agreement with Brown, thinks it obvious that s is impossible in the absence of a necessary condition (that is, precondition) for it but finds the case of an absent necessary condition of s less clear. The difficulty arises from considering a relation between states of affairs from distinct temporal perspectives, in particular, considering the modal status of s at moment m from a perspective subsequent to m at which the necessary condition of s is supposed to occur.

The Taylor Inequivalence

This line of discussion indicates that Wallace (like Brown before him) fails to acknowledge fully the premises of Taylor’s argument. In the argument, Taylor assumes that the world in time is ontologically homogeneous; hence, whether a necessary condition of a state of affairs, s, occurs prior or subsequent to s simply makes no difference. It is, nonetheless, via this line that the crucial notion for illuminating Taylor’s Aristotelian argument emerges. For through his effort to get clear on what to say about cases in which a necessary condition of a state of affairs at a particular moment subsequently fails to exist, Wallace notes an ambiguity in claims about the modal status of a feature of the world at a moment when evaluated from the perspective of a distinct moment. This ambiguity is of the utmost importance not only to Wallace’s critique and dismissal of Taylor’s argument but also to appreciating the argument.

The crux of Wallace’s criticism of Taylor is that he fails to recognize distinct claims about the modal status of features of the world in time. According to Wallace, Taylor conflates:

(MT1) if at m₂(−n), then at m₁(−Pos: s)

and

(MT2) if at m₂(−n), then −Pos(at m₁: s)8

Wallace calls the distinction here the Taylor Inequivalence. The first claim states that if a certain state of affairs, n, does not occur at m₂, then at m₁, a prior moment, another state of affairs, s, is in itself not possible. The second claim states something different: if a certain state of affairs, n, does not occur at m₂, then it is not possible that at m₁ s occurs.

The differences here are subtle. (MT1) is the claim that the lack of n at m₂ has as a consequence that at a prior moment, m₁—from the perspective of that very moment itself—s is impossible; in other words, considering only how the world is at m₁, s could not be. But (MT2) is the claim that the lack of n at m₂ has as a consequence that it is impossible that s occur at m₁. This leaves open what exactly renders s impossible at m₁. There need be no reason to think that it is how things are at m₁ that makes s impossible—it is perhaps what goes on at other moments, e.g., the lack of n at m₂.

Wallace contends that Taylor becomes confused by the language in which he expresses his argument. He fails to distinguish these elusive differences between claims about the modal status of features of the world in time. Wallace asserts: “An analysis that can show that MT1 and MT2 are not equivalent, why they are not equivalent, that MT2 not MT1 follows from Taylor’s argument, and that only MT1 would actually force fatalism on us, should represent a significant step toward solving the Taylor problem” (165).

Wallace draws out the differences between (MT1) and (MT2) with a colorful (if infelicitous) example. He supposes that a group of terrorists brings a nuclear weapon to Amherst College. The commander of this group has all the normal capacities of any healthy adult. He sits all day with his finger on the button that would detonate the weapon; the apparatus is in perfect working order. Yet he does not depress the button. A nuclear explosion on the Amherst campus yesterday would be sufficient for radiation in excess of 20 rads today, that is, radiation in excess of 20 rads today is a necessary condition of a nuclear explosion yesterday.

Now suppose that there is not radiation at Amherst today in excess of 20 rads. Two things might follow from this: in light of the connection between a nuclear explosion and the local level of radiation, from the lack of radiation today (i.e., at m₂) it might follow that there could not have been an explosion yesterday (i.e., at m₁). This is indeed a very natural and plausible conclusion. Or, what might also follow, given the lack of radiation today (at m₂), then yesterday (at m₁) a nuclear explosion was in itself impossible, that is, the lack of radiation today makes it the case that yesterday, considering it from the perspective of that very day itself, a nuclear explosion was impossible.

Wallace’s Dismissal of Taylor’s Fatalistic Argument

Wallace maintains that the second conclusion, unlike the first, is bizarre. It does not seem that the absence of some state of affairs, n (e.g., the presence of a certain level of radiation), at a later moment can constrain how things are at an earlier moment by making some other state of affairs, s (e.g., a nuclear explosion), in itself impossible at that moment. But if s is not in itself impossible, it could occur (or could have occurred) at that moment—despite −s actually occurring—and so there is contingency in the world.

Taylor’s fatalistic argument is supposed to demonstrate that the absence of some state of affairs, n, that is a necessary condition of a state of affairs, s, at a previous moment, m, shows s is impossible at m. If this conclusion is understood as (MT1), then s is in itself impossible at m. Consequently, Taylor’s argument shows that there is no contingency. But if the conclusion of the argument is understood as (MT2), s is not in itself impossible at m, and fatalism does not follow. Therefore, Wallace thinks if he can only show that (MT1) and (MT2) are inequivalent, then he has undermined Taylor’s argument, for this provides a way of interpreting its conclusion that is perfectly compatible with contingency. One can accept all the premises of the argument without accepting fatalism.

Most of Wallace’s subsequent discussion is taken up with a presentation of a formal system in which to codify interpretations of the two claims; this is supposed to “provide formal reasons for thinking that propositions such as [MT1 and MT2] are not equivalent” (177). But the formalism is really superfluous. (MT1) and (MT2) are both clearly legitimate and different claims that one can make about the modal features of the world in time. Wallace’s informal example suffices to illustrate these differences. The formalism he presents merely provides a way of regimenting the two claims; if one does not grasp their differences—if one did not already know how to interpret them—the formal system would be of no help.9 So one can grant that (MT1) and (MT2) are not equivalent.

Even granting this inequivalence, however, Wallace has not undermined Taylor’s argument. He has indeed shown that there are two ways of understanding its conclusion, one consistent with contingency, one not. Were it the case that he and Taylor were making all the same assumptions, then Wallace’s reading of the conclusion would show that fatalism does not validly follow from Taylor’s argument’s premises. But Wallace and Taylor are actually making incompatible assumptions about the nature of contingency; each is presupposing a totally different view of the modal features of the world in time. Given Taylor’s view, the conclusion of his argument must be interpreted in line with (MT1); Wallace concedes that this leads to fatalism. Given Wallace’s view, the conclusion of the argument may be interpreted in line with (MT2), and this is compatible with contingency. Therefore, it is on the truth of this fundamental and unstated assumption—regarding synchronic possibility—that the appropriate conclusion of Taylor’s argument turns.

TWO METAPHYSICS OF CONTINGENCY

Wallace takes the modality relevant to Taylor’s discussion to be physical-cum-causal, one based on those connections among things determined by physics or the laws of nature. He maintains that what Taylor’s argument demonstrates is that the absence of some necessary consequence of a certain state of affairs, s, shows only that s did not occur at an earlier moment, m—it could not have occurred given the laws of nature and the absence of its necessary consequence. According to Wallace, the argument does not show that s is in itself impossible at m. Indeed, Wallace takes for granted, considering s in itself, insofar as s itself is compatible with the laws of nature, that although s did not occur at m, it could have. This is precisely what Taylor rejects.

Synchronic Possibility

Taylor assumes that if s does not (or did not) occur at m, then s could not occur (could not have occurred) at m. This is not, however, merely to presume fatalism. Rather, it is to accept a different view of contingency than the one presupposed by Wallace and all Taylor’s other critics. The different views turn on the modal status of features of the world at a moment considered at that moment. The key issue is whether at a given moment, m, considering some state of affairs, s, at m, whether s—at m—could be otherwise than it in fact is, that is, whether at m there are nonactualized possibilities. The issue here, in other words, is whether there is synchronic possibility in the world.

Among the ancients and medievals, it appears to have been universally accepted that there was no synchronic possibility: at each existent moment everything must be just as it is at that moment.10 The notion that there were unactualized possibilities at a moment was revolutionary. It is thought to have been introduced by John Duns Scotus in the fourteenth century.¹¹ The notion naturally accompanied discussions of modality in the early- to mid-twentieth century, where the only modality thought to be relevant to philosophy per se was linguistic-cum-conceptual. It is certainly consistent with how we speak or conceive of the world that things be otherwise than how they are at a given moment: although I wear a white shirt at m, it is not a contradiction in terms or conceptually incoherent to hold that I wear a blue shirt at m.¹²

Moreover, the notion of synchronic possibility also naturally accompanies talk of possible worlds, which since the end of the 1960s has been widespread. Regardless of how they are conceived—as maximally consistent sets of sentences or propositions, as properties the actual world might have, as complete concrete universes—it seems plausible that there are possible worlds in which things are otherwise than they are at this very moment. Thus, there is a maximally consistent set of sentences that represents me as now wearing a blue shirt, though I wear a white one, or perhaps there is someone quite like me in some other concrete world now wearing a blue shirt.

Possibilities at a Moment Versus Possibility from a Moment

Given the notion of synchronic possibility, two very different pictures of how possibility is located in the world arise; depending on this notion, then, there are two distinct metaphysics of contingency.

If one accepts synchronic possibility, one gets an account of contingency on which at every moment that exists there is an array of distinct possibilities. Possibility is located at each actual moment: at each moment reality could be otherwise than it in fact is. On this picture, speaking figuratively, modal reality overlays temporal reality; possibilities burgeon throughout time.

If one denies synchronic possibility, at every moment that exists things must be just as they are. At no moment does nonactualized possibility exist. Thus, insofar as there is unactualized possibility, it begins where the world in time ends, and, consequently, there is not a never-ending succession of moments each of which is equally real. Presumably, there is now no subsequent moment. Everything must exist at any moment it does and just as it is; hence, all true claims about the temporal world are necessarily true. Nevertheless, all this is compatible with contingency, for the world could be ever so many different ways subsequent to this moment: possibility arises not at this moment but from it. There is more to the actual world because it could become more in time. On this picture, speaking figuratively, modal reality abuts temporal reality; the future is the realm of possibility.

Taylor does not explicitly deny synchronic possibility—indeed, he does not mention the notion at all—yet it is clear that he rejects unactualized possibilities at a moment. Thus, his rejection of synchronic possibility is a second key assumption of Taylor’s fatalistic argument that he leaves tacit, accompanying the assumption that the modality relevant to the argument is based on the natures of things in themselves.

The best evidence that Taylor rejects synchronic possibility is that Taylor’s argument is obviously Aristotelian, and Aristotle is explicit in his discussion of time and modality that he rejects synchronic possibility. Consider: “What is, necessarily is, when it is; and what is not, necessarily is not, when it is not.”13 Aristotle’s discussion here is the inspiration for and basis of Taylor’s ruminations on fatalism. Moreover, Taylor (1963, 85) takes himself to be presenting the same sort of argument that disturbed St. Augustine and Boethius. Such an argument would bring with it the ancient presuppositions about the nature of reality, and, as noted above, one of these presuppositions is that there is no synchronic possibility. Last, given the assumption that there is no synchronic possibility—and a modality based on the natures of things in themselves—Taylor’s Aristotelian argument is undeniably valid. (I defend this claim below.)

It is as clear that Taylor’s critics are assuming there is synchronic possibility as it is that Taylor assumes there is not. When considering the Taylor Inequivalence in the context of his colorful example, Wallace claims that Taylor takes it to be an upshot of his argument that at some moment, m, when the commander of the terrorists sat with his finger on the button that would detonate the nuclear weapon at Amherst and did not depress the button, that it was impossible for him to do so. To this, Wallace responds: “this is clearly just plain wrong: I have constructed the case in just such a way that under any halfway reasonable definition of situational physical possibility [i.e., the modality relevant to the argument] it is physically possible at [m] for the explosion to occur at [m]” (171).

If, however, there is no synchronic possibility and there is no explosion at m, then it is indeed impossible for there to be an explosion at m. Thus, Wallace asserts—without any consideration—that a key assumption of Taylor’s argument, one crucial not only to understanding that argument but also Taylor’s position and motivation more generally, is “clearly wrong,” not even “halfway reasonable.” It is no wonder, then, that Wallace dismisses Taylor’s argument as fallacious!

The present point—that Wallace takes for granted synchronic possibility—might be obscured by several comments he makes in the course of distinguishing the physical-cum-causal modality he thinks is relevant to Taylor’s argument from the modality he thinks typically underlies metaphysical discussions. The notion of possibility accompanying the latter is, according to Wallace, understood “in terms of a synchronic relation between alternative, simultaneous possible ‘worlds’… while I will be arguing that physical possibility is best understood as a diachronic relation of compatibility under causal laws between sets of conditions as the condition-sets stand in appropriate relations through time” (177, see also 180–181). Despite appearances, this quotation actually corroborates the present point. Wallace is proposing a restriction on “logical” or “alethic” modality, the notion that he thinks underlies contemporary metaphysical discussions, because he thinks this notion is not sensitive to changing features of the world. I believe this is mistaken, a vestige—which Wallace accepts uncritically—of the prevailing view for several decades that the modality relevant to philosophy is linguistic-cum-conceptual. But the “logical” modality that Wallace proposes to restrict, by his own acknowledgment, includes synchronic possibility, so it is part of the framework he employs in attempting to undermine Taylor’s argument. The importance of synchronic possibility to Wallace’s thinking about how possibility is in the world is apparent from the “visual apparatus” he presents to illustrate his discussion: although the focus is on the relations among worlds at moments, an essential feature of these moments is that there are many possibilities at any given one (Wallace 1985, 184–186).

Therefore, Wallace summarily rejects, as one that could not be right, a key assumption of Taylor’s argument. He was not alone in doing this. Abelson, Taylor’s contemporary critic, notes that it follows from Taylor’s assumptions that “since every event is a necessary and sufficient condition for itself, the nonoccurrence of an event would render that event impossible while the occurrence would render it necessary” (Abelson 1963, 82). Abelson concludes that this result arises from fallacious reasoning and leads trivially to fatalism. But he is incorrect on both counts. The result is not a consequence of Taylor’s argument; it is, rather, a presupposition of it—it is the assumption that there is no synchronic possibility. This assumption, however, leads to fatalism only if one rejects the alternative metaphysics of contingency (the one that Taylor, in the end, accepts). Others, like Brown (1965, 127), who charged that Taylor begs the question of fatalism at the outset of his discussion were in error for similar reasons.

The Proper Interpretation of Taylor’s Argument Turns on the Notion of Synchronic Possibility

Taylor’s denial of synchronic possibility, inspired and motivated by Aristotle, is the key both to appreciating his fatalistic argument and also to understanding his attitude toward it. He was quite confident in the argument and certain of its validity (see Taylor 1962b, 1963, 1964). As presented, the focus of the argument is on the necessary connections between states of affairs. Underlying the argument, though, is the notion of synchronic possibility.

Bearing these points in mind, the crux of the argument can be presented in a form that is transparent and simple: Suppose that there is no synchronic possibility. The way things are at a certain moment, m, must be as they are, and so a given state of affairs, s₁, at m is necessary. If this is so, and there is a necessary connection (based, one may assume, on the very natures of the things in themselves) between s₁ and another state of affairs, s₂, at a different moment, then s₂ must exist at that moment. The reasoning here: Necessarily p and necessarily (if p, then q), therefore necessarily q is clearly valid—indeed it is taken as an axiom of any standard system of modal logic.

This view of the argument not only makes patent its validity but also provides the means for simplifying the long and complicated dialectic surrounding it. Taylor’s contemporaries focused their criticism on the necessary connections between states of affairs that are crucial to Taylor’s argument. Beginning with the linguistic-cum-conceptual modality, which they presumed was the only legitimate one, they argued in different ways that these necessary connections were problematic. Thus, Saunders maintains that it does not follow from the lack of a necessary condition for some state of affairs, s, that one cannot bring s about; Abelson and Brown maintain that the necessary connections Taylor relies on are illegitimate because of equivocation or some other sort of confusion. Because of their failure to recognize the modality operative in Taylor’s argument, such criticisms failed even to engage it. However, the more important point for present purposes is that with this focus on the necessary connections employed in the argument, the issue of synchronic possibility does not become salient.

With the revival of metaphysical inquiry and the recognition of the philosophical legitimacy of different modalities, some of which are fully in the world, i.e., not linguistic-cum-conceptual, consideration of the argument shifted from its crucial necessary connections between states of affairs, which now seemed less alien, to what is the case—or must be the case—at a particular moment. Wallace takes it for granted that at any given moment there are myriad unactualized possible states of affairs. Presuming this, then, it is not the case that at a certain moment, m, a given state of affairs, s₁, is necessary. If this is so, one can concede a necessary connection between s₁ and another state of affairs, s₂, at a different moment, and still it does not follow that s₂ is necessary. Indeed, it clearly does not follow that s₂ is necessary. This line of reasoning explains Wallace’s attitude toward Taylor’s argument: given Wallace’s assumptions, it is fallacious and obviously so.

Therefore, assuming there can be necessary connections between states of affairs in the world,14 if one accepts that there are synchronic possibilities, as Wallace—and many other contemporary philosophers do—Taylor’s Aristotelian argument is clearly invalid; if one denies synchronic possibility, as Taylor and the ancients and most medievals did, the argument is clearly valid.

Regardless of whether there is synchronic possibility, one can accept contingency, nonactualized possibility in the world. There are, however, two very different metaphysics of contingency depending on whether there is synchronic possibility. If there is, things now could be otherwise than they in fact are; if there is not, things must be just as they are (and were) though the world could be different—at a subsequent moment. Rejecting synchronic possibility, Taylor presupposed an unfamiliar way of understanding how possibility is in the world and then shows, with his argument, that this way is incompatible with contingency given certain popular assumptions. Taylor is not guilty of any bad reasoning. He might, however, be charged with fostering confusion by failing to articulate the Aristotelian presuppositions regarding modality and metaphysics and synchronic possibility underlying his discussion.

CONCLUSION

What emerges from examining Taylor’s argument is that there are two different ways of locating possibility in the world, two different metaphysics of contingency. On one, there is possibility at a moment; on the other, there is possibility from a moment. If the former view is correct, Taylor’s fatalistic argument is invalid—if the latter is correct, the argument is valid, and, hence, in order to accept contingency, one must reject some of the popular assumptions on which the argument is based. Yet either way there is contingency: one need have no fears regarding fatalism.

One should not presume that just because there is an assumption one could make, viz., accepting synchronic possibility, from which it follows that Taylor’s argument is invalid, that this assumption is true. The assumption entails a particular metaphysics of contingency. There are, however, two incompatible metaphysics of contingency—the question of which is correct remains. Given its connection to modality and the world in time and truth, this is a question of profound and far-reaching metaphysical importance.

Taylor (like Aristotle) merely takes for granted that there is no synchronic possibility; his critics, including Wallace, merely presume there is. Therefore, Taylor’s argument itself and all the critiques of it really provide no insight into this issue. To determine the correct account of contingency, one must go beyond this argument. Above, I stated that I thought the conclusions about the world in time and truth about the world that Taylor, following Aristotle, draws from his argument are correct. This is because I believe there are compelling reasons to reject synchronic possibility. This is not the place for a full discussion of these issues,15 but in closing, I present some of these reasons and then ameliorate the conclusion that there are not many moments of time, all of which are equally real, and that it is not the case that every proposition whatsoever is either true or, if not true, false.

Reasons for Thinking There Is No Synchronic Possibility

The first thing to note is how odd the assumption is that right now things could be other than they actually are. As things are, I now wear a white shirt. It is certainly natural to think that with respect to what I am wearing at this moment reality is complete—is full—and, therefore, must be as it is. Yet if there is synchronic possibility, it is possible that I now wear a blue shirt. This might seem plausible, for one might suppose it possible that earlier today I put on a blue shirt (rather than a white one). However, this sort of consideration provides no grounds for thinking that right now things could be other than they are, for it merely presupposes synchronic possibility: one who advances it presumes that at the very moment at which I actually put on a white shirt I might have put on a blue one.

It is difficult to see how one could provide grounds for the claim that there is synchronic possibility. Indeed, to my knowledge, not a single contemporary philosopher has provided any reason whatsoever for this claim—yet that there is synchronic possibility is universally accepted (to my knowledge). The considerations Duns Scotus adduces for synchronic possibility when introducing the notion are based on doctrinal assumptions, ones that many today would find implausible or otherwise objectionable (see again Normore 2002). Of course, if one believes that the source of necessity and possibility is ultimately how the minds of conscious beings interact with the world (and, hence, that the only legitimate philosophical modality is linguistic-cum-conceptual), then synchronic possibility is corollary. But such a position is much less accepted in contemporary metaphysics than it was throughout the twentieth century. I believe that philosophers have simply failed to recognize that when this position is dismissed, so is any reason for accepting synchronic possibility. Moreover, the reliance on and familiarity of talk of possible worlds, which abet the acceptance of synchronic possibility, has obscured the fact that this acceptance is entirely unmotivated.

These negative considerations are not meant to be a strong argument against the claim that there is synchronic possibility; rather, they are meant to expose the tenuousness of this claim. I present now a positive argument against synchronic possibility. The combination of this argument with the foregoing considerations seems to me to provide a compelling reason to reject synchronic possibility and, hence, to accept the metaphysics of contingency that eschews this notion.

Suppose at moment m there is a cat before one. At m, the cat is white. It is plausible that the cat, being an instance of a certain kind of thing, has essential features, features it must have in virtue of being a cat (perhaps, being a living thing, being a mammal, being extended). It is implausible that its color is one of these essential features. There are cats of many different colors, and this cat could surely survive and be grey. Considering only the cat, then, it might seem that although at m the cat is white, it is possible that at m the cat is grey (or at m the cat is possibly grey)—there is, after all, nothing in its nature that precludes it from being grey at m. However, there is more in the world to consider.

At moment m, the world is a unique and particular way, e.g., with respect to the color of the cat. If the cat and its nature cannot ground and explain this particularity—and they cannot if the cat could be white and could be grey—then, presumably, there is something else in the world that can. Perhaps it is the state of affairs of the cat being white at m or the cat’s particular whiteness at m (that is, a trope or mode) or the simple fact that the cat is white at m. Whatever it is, whatever is the ontological grounds of this particular feature of the world at m, it is not feasible to think that it is compatible with the nature of this thing that the cat be grey at m. Thus, for example, it does not seem that the state of affairs of the cat being white at m could exist yet it be possible that the cat be grey at m—the very nature of this entity, unlike the cat itself, seems to preclude the possibility.

The example generalizes: insofar as the world has a unique nature at each moment, and insofar as there are grounds for and explanation of the particular features that underlie this uniqueness, there are entities to whose nature it is incompatible that the world be otherwise at that moment. Therefore, there are in the world things that exclude synchronic possibility.

Why the Resulting Picture of the World Is Not So Bad

As noted above, the upshot of this conclusion is that if there is contingency, then temporal reality is not ontologically homogeneous, and every proposition is not either true or false. I find neither consequence unfeasible.

If there are not many moments of time, all with the same ontological status, then, assuming that this moment, now, is real, there are significant ontological differences between the present and the future and past. But it is not at all farfetched to think that one differs ontologically from dinosaurs and Napoleon or from one’s great-great-great-great-grandchildren and Martian outposts; one exists, whereas the others do not in any sense. There are many developed positions in the metaphysics of time that accept such ontological heterogeneity in temporal reality, so this thesis can hardly be regarded as extreme. I myself defend a presentist view on which, though the past is real, there are no past moments—no past entities whatsoever—and on which there are no future entities and, hence, no facts about what will be (Fiocco 2007).

Even less problematic is the consequence that not all propositions are either true or false. If this is so, then some propositions are indeterminate; they lack a truth value. Indeterminacy is a phenomenon recognized in connection to many issues in the philosophy of language, in physics, and in metaphysics. Here, then, is another. To say that indeterminacy is unproblematic as a consequence of contingency is certainly not to suggest that the phenomenon, in its multiplicity, is amenable to any simple or definitive account but only to acknowledge that it is a genuine phenomenon. To deny it outright seems to me farfetched; the assumption of bivalence for all propositions seems somewhat naïve or myopic. Thus, if contingency requires it, one has merely another appearance of indeterminacy to consider.

In the end, the value of considering Taylor’s Aristotelian argument is that it brings to light two different metaphysics of contingency. Examining these, one ends up with a view on which there is contingency yet no contingent features of the world (or claims about it). Reality is such that it contains the capacity for change and evolution without any feature being able to be otherwise than it in fact is. This leaves one with a view of temporal reality on which there is nothing but possibility after this moment, now. Far from being objectionable, this view seems to be both true and just the way any self who aspires to free expression and growth would hope the world to be.

NOTES

1. See Taylor (1962a, 43). “If any state of affairs is sufficient for, though logically unrelated to, the occurrence of some further condition at the same time or any other time, then the former cannot occur without the latter occurring.”

2. This view of temporal reality would have been familiar to Taylor’s contemporaries from, e.g., Russell (1915), Williams (1951), Smart (1955), and Quine (1960). It is familiar to contemporary readers from more recent discussions, such as Sider (2001), Mellor (1998, 1981), Lewis (1986), and Oaklander (1984).

3. In particular, the section “Considerations of Time.” See Taylor (1962a, 47–48).

4. See, in particular, Taylor (1957).

5. This idea, that a true representation is made true by some feature of reality, is widely accepted in contemporary metaphysics. For defense of the claim that every true proposition must, on pain of contradiction, be made true, see Fiocco (2013). It is an idea that seems to have been accepted by both Taylor (1963, 87) and Wallace (1985, 175).

6. For related criticism, see Aune (1962, 71) and Makepeace (1962).

7. From his responses (Taylor 1962b, 1963, 1964), it seems Taylor himself was not aware of how out of step he was with his colleagues and, thus, that the source of their disagreement was radical: very different understandings of the nature of reality and ancillary views of metaphysics. In his contribution to the present volume, William Hasker also suggests that confusion regarding modality, endemic to that period of analytic philosophy, is relevant to evaluating the objections of Taylor’s contemporaries to his Aristotelian argument. See chapter 1.

8. See Wallace (1985, 164). I have changed Wallace’s notation to bring it in line with my presentation of Taylor’s argument above.

9. Hasker makes a similar point in his contribution to this volume when he suggests that Wallace may be overlooking the fact that “formal logical systems rest on intuitive foundations” (italics are Hasker’s).

10. For discussion, see Knuuttila (1993, chaps. 1–2).

11. See Knuuttila (1993, chap. 4) and Normore (2002).

12. Though, of course, it would be a contradiction in terms to hold that I wear a white shirt and a blue, i.e., nonwhite, shirt at m (assuming I wear but one shirt).

13. Note that Aristotle goes on to say: “But not everything that is, necessarily is; and not everything that is not, necessarily is not. For to say that everything that is, is of necessity, when it is, is not the same as saying unconditionally that it is of necessity” (De Interpretatione, chap. 9, 19a). Thus, denying synchronic possibility is not in itself to reject contingency.

14. One should note that none of Taylor’s critics give any reason for denying this assumption. Rejecting it is merely part of the view that the only modality relevant to philosophy per se is linguistic-cum-conceptual. This view, so dominant throughout the twentieth century, is no longer orthodox.

15. I consider them in much more detail in Fiocco (manuscript).

REFERENCES

Abelson, R. 1963. “Taylor’s Fatal Fallacy.” Philosophical Review 72, no. 1. Reprinted in Cahn and Eckert 2011, 79–83. Citations refer to the reprint.

Aristotle. 1987. De Interpretatione. In A New Aristotle Reader, ed. J. Ackrill. Princeton, N.J.: Princeton University Press.

Aune, B. 1962. “Fatalism and Professor Taylor.” Philosophical Review 71, no. 4. Reprinted in Cahn and Eckert 2011, 69–78. Citations refer to the reprint.

Brown, C. D. 1965. “Fallacies in Taylor’s ‘Fatalism.’” Journal of Philosophy 62, no. 13. Reprinted in Cahn and Eckert 2011, 127–132. Citations refer to the reprint.

Cahn, S., and M. Eckert, eds. 2011. Fate, Time, and Language: An Essay on Free Will, by David Foster Wallace. New York: Columbia University Press.

Fiocco, M. O. 2007. “A Defense of Transient Presentism.” American Philosophical Quarterly 44: 191-212

——. 2013. “An Absolute Principle of Truthmaking.” Grazer Philosophische Studien 88: 1-31.

——. Manuscript. “Truth, Modality, and Time.”

Hasker, W. 2014. “David Foster Wallace and the Fallacies of Fatalism.” Chapter 1 of the present volume.

Knuuttila, S. 1993. Modalities in Medieval Philosophy. London: Routledge.

Lewis, D. 1986. On the Plurality of Worlds. Oxford: Basil Blackwell.

Makepeace, P. 1962. “Fatalism and Ability II.” Analysis 23. Reprinted in Cahn and Eckert 2011, 61–64. Citations refer to the reprint.

Mellor, D. H. 1981. Real Time. Cambridge: Cambridge University Press.

——. 1998. Real Time II. London: Routledge.

Normore, C. 2002. “Duns Scotus’s Modal Theory.” In The Cambridge Companion to Duns Scotus, ed. T. Williams. Cambridge: Cambridge University Press.