Debates in the Metaphysics of Time 9781472594112, 9781780937489

Citation preview

For Linda

Preface T

he origin of this collection of original essays is both unusual and amusing. Colleen Coalter, the Philosophy editor at Bloomsbury, asked me to write something for the back cover of a book on metaphysics that Bloomsbury was going to publish and I was happy to do so. Evidently, Colleen liked what I had written and as a result a few months later she asked me to give my opinion on a proposal submitted to her for publication. I agreed to do it, read the proposal and filled out the questionnaire regarding it. After answering the obvious questions, I came to the end and the last question was whether there were other topics for which a book was needed. I answered that question as follows: “An anthology on the philosophy of time itself or a larger one relating time to other issues in philosophy (metaphysics, philosophy of religion, philosophy of mind) would be useful for classroom use and for professional philosophers.” Colleen was interested and I volunteered to edit the anthology. I gathered together philosophers many of whom I have known for some time, and all of whom I greatly respect, and the result is this book. The 13 original essays collected here are in one way or another connected to the philosophy of time, and indeed make explicit the importance of time in metaphysics, phenomenology, the philosophy of mind, and the philosophy of religion. As in most of the anthologies that I have edited or co-edited, my wife, Linda Galang Oaklander, has been an invaluable aid in formatting the manuscript and references, and reading my introductions. This collection is no different. I also wish to thank Linda for compiling the index. L. Nathan Oaklander Ann Arbor, Michigan USA, January 2014

Introduction W

hat is the debate in the metaphysics of time about? What is or are the central issues? The chapters in this book address both questions. McTaggart set the stage for much of the contemporary discussion by distinguishing the A-series whose terms are past, present, and future, and the B-series whose terms are earlier/later than and simultaneity. From these two series there arose, starting with Gale’s terminology, the A-theory, and the B-theory. A-theorists are those who claimed that the A-series was the essence of time from which the B-series could be derived, and the B-theorists are those who maintained that the B-series is ontologically fundamental and from which the A-series could be derived. The chapters in this volume reveal that the discussion has broadened considerably since then, so that it is not particularly helpful to approach the issue(s) by characterizing the debate as between the A-theory and the B-theory, or for that matter as between presentists and eternalists, or tensed and tenseless views of time. The chapters that follow contain several different analyses of time that cannot be easily classified into any of these dichotomies, nor should they be. In the first section of this book the question is raised as to whether there is a coherent debate in the metaphysics of time? Dolev argues that the usual way in which the debate is framed— for example, does only the present exist or does the past, present, and future exist—is ultimately incoherent because the metaphysical theories that give rise to them are unintelligible because they depend on empty notions such as “tenseless relations” and “the moving now.” He believes that by coming to this realization, “a systematic dissolution of the traditional philosophical questions concerning time is brought about” (Dolev 2007: 60). The results are “anti-metaphysical perspectives” regarding time and realism, as the subtitle of his book, Time and Realism: Metaphysical and Anti-Metaphysical Perspectives, makes clear. Oaklander demurs. It is, naturally, a tough pill to swallow: the implication that the value of traditional metaphysical questions concerning time rests on providing the necessary steps to the anti-metaphysical perspective, and that once those steps have been climbed the ontological debates that gave rise to those questions and tenseless and tensed theories is to be discarded. Rather than swallow it, Oaklander criticizes the arguments Dolev gives against tenseless relations and argues that they do not establish that they

Introduction

xi

are empty notions, but rather are objects of acquaintance that can ground the dynamic aspect of time (see also the chapters by Tegtmeier and Mozersky in this volume). Oaklander further argues that a more fundamental debate than that over whether what Dolev calls the “ontological assumption” that centers the debate on whether there are ontological differences between past, present, and future objects, is that which concerns the ontological status of temporal relations. Naturally, Dolev defends his thesis regarding the need to transcend ontology against Oaklander’s critique, and goes on the offensive by criticizing the notion of “the specious present” and offering his own account of experience of motion and change that appeals to the mind-independent property of presentness, and not to succession. Orilia takes up Dolev’s view of the present as well as his overall stance that A- and B-theories of time must be transcended before offering his own analysis of time that he calls “moderate presentism” since it adopts eternalism with respect to moments of absolute time, but presentism with regard to events or states of affairs. Orilia’s own analysis, he claims, is best suited to explain the important phenomenological fact that the past is closed and the future is open, and resolve the Augustinian paradox regarding the duration of the present. His critique of Dolev’s account of the present is intended to demonstrate that there are reasons to believe Dolev “endorses an incoherent mixture of A-theoretical and B-theoretical claims” (Orilia, this volume, p. 60), and thus does not effectively transcend either of the views he claims are ultimately unintelligible. Tegtmeier embodies his account of succession and time in the framework of his more general ontology of facts, things, and forms. He presents a purely relationist account of time that seeks to ground our experience of time and tense, including its dynamic aspects in temporal relations. In the course of his discussion he criticizes Aristotle’s presentism, Donald Williams’ four-dimensionalism, and McTaggart’s distinction between the A-series and B-series that has thrown discussions on the philosophy of time off the track by arguing that time requires change when in fact being the subject of change is not the hallmark of the temporal. Time contains change and so cannot itself change. Fiocco offers a rather novel account of time. He claims that if there is real dynamism or novelty in temporal reality, then it must come about through becoming. He distinguishes temporal becoming which is the coming to be of events or objects at a moment, absolute becoming which is the coming to be and immediately ceasing to be of moments, and atemporal becoming. Fiocco says that since a moment neither comes to be at a moment nor changes when it ceases to exist and is replaced by a new moment, it comes into being outside of time. For that reason, a moment is an atemporal entity and undergoes atemporal becoming. While Tegtmeier and Fiocco talk about change in their chapters, Mozersky defends an analysis of it that applies to temporal change or becoming—the

xii Introduction

change of an event or time from the future to the present and into the past— and to ordinary change of say, an apple, from green to red. On the face of it, both types of change involve a single event, object or time having incompatible properties, which is contradictory. To avoid this problem, Mozersky defends the view that ordinary monadic predicates, such as “is green” or “is present,” are two place relations that hold between objects and times, and the contradiction vanishes. For there is no incompatibility between “x being green at t1” and “x being red at t2” or between “e being present at t1” and “e being past at t2” where the relations in the change of tense are the temporal relations “is simultaneous with” and “is earlier than” respectively. Although Tegtmeier, Fiocco, and Mozersky do not directly address one another in their chapters, there are important areas of agreement and disagreement. For example, Mozersky and Tegtmeier argue against presentism and absolute becoming whereas Fiocco argues for those views and against Mozersky’s tenseless B-theoretic world-view that all events past, present, and future exist equally in the B-series. Mozersky and Tegtmeier agree that there is a sense in which time passes (although Tegtmeier prefers to characterize the phenomenon as succession), and has a dynamic aspect, but they disagree over the proper analysis of temporal relations. Tegtmeier implicitly criticizes Mozersky’s (by criticizing Donald Williams’ structurally analogous) view that temporal relations are ordered pairs of objects and times. In the chapters by Lee, Pelczar, and Dainton, the topic of time is approached via the phenomenology and ontology of temporal experience. What makes the chapters so stimulating and fascinating is that there are a few basic issues that are dealt with differently, but getting clear on what is the issue exactly is also approached in a unique fashion by each of the authors. Rather than summarize the different ways in which each of the authors expresses the issue, and critique Dainton’s response to it, I shall offer my own account of what I take to be common to all three and the problems they are attempting to solve. Traditionally, the problem is how can a person perceive in the present or presently perceive the duration of, say, that whistle of a train as it speeds by, or a succession of rapidly changing notes on a keyboard. The problem seems to be that the present has no scope during which something could have a duration or change. This is a problem of perceiving a change or continuation in a single unified act of temporal consciousness. There is, however, also the experience of flow from one unified temporal consciousness to the next that presents a different problem of unity or rather continuity. As we hear a series of notes that are not presently given, but are given as forming a melody over a successive series of presently given tones, our experiences are unified in a way that my experience of the alarm going off in the morning is not unified with your experience of the one o’clock siren in the afternoon. How do we

Introduction

xiii

account for the continuity of the experiences of the former and discontinuity of the experiences of the latter? As the authors of this section demonstrate, what makes this tricky to explain briefly is the radical disagreements over what forms of continuity do in fact exist in our streams of consciousness.1 Lee first briefly argues for atomism—the view, most notably, but not exclusively, that momentary (or very brief) experiences (without distinct experiences as temporal parts) can have contents that appear to possess duration and succession—and then Lee replies to objections that Dainton has raised against it. Pelczar, on the other hand, criticizes Dainton’s extensionalism—the view that the experience of flow or succession we immediately experience over short intervals, and the continuity of our experience over longer intervals, requires both that our experiences and their contents are extended and that temporally extended experiences overlap by sharing parts—without presupposing atomism. Dainton responds to Lee’s defense of atomism, gives additional arguments against it, and defends himself against Lee’s and Pelczar’s criticism of his version of extensionalism. What is also important to note is that issues in the metaphysics of time are relevant to the analysis of temporal consciousness. Thus, for example, it seems that if you are a presentist then you must be an atomist, and if one adopts the extensionalist viewpoint then one cannot be a presentist. Dainton argues, however, that an extensionalist can still maintain that “If the continuity and flow which exist in our experience are themselves physical features of this universe, then this very real form of passage is also a feature of the universe” (this volume, p. 202). In Part III, on God, Time, and Human Freedom, Diekemper and Leftow begin by debating the questions of whether there are instantaneous events or are events necessarily temporally extended entities? Do temporally extended divine events necessarily exist, and is, therefore, God temporal? In the first section of his chapter, Diekemper argues that events cannot be instantaneous, and defends that view against four objections raised in earlier articles by Leftow. In the second section, he argues that divine events exist and hence God is temporal. In his chapter, Leftow gives several original arguments for instantaneous events, one of which connects with presentism, critiques Diekemper’s arguments for divine temporality, and replies to Diekemper’s objections to instantaneous events. The chapters by Rhoda and Rogers are also concerned with God and time, and also address the issue of whether or not God is temporal or timeless, but that is not the main issue, as in the previous debate. Rather, the main concern is with the traditional problem of the possibility of reconciling divine foreknowledge with human freedom. Rhoda argues that God is temporal, but even if God were timeless it would not solve the issue and in fact would prevent reconciling the two. Moreover, he argues that the only way of avoiding

xiv Introduction

fatalism and of reconciling divine foreknowledge and human freedom is by adopting a particular version of the open future theory, according to which there are many possible futures, but they all exist. How he arrives at this view and why he rejects all versions, including Rogers’ version of eternalism, forms the content of his chapter. Rogers defends the view she calls “isotemporalism,” that shares with “eternalism” and four-dimensionalism the view that all objects in the linear temporal sequence that constitutes the history of the world exist equally, regardless of whether these objects are past, present or future. Like the typical B-theorist, Rogers holds that past, present, and future are subjective in that they depend on the perspective from which an event is viewed. If I am at t1, then t0 is past and t2 is future, and from the perspective of t2, t0, and t1 are past, t2 is present, and t3 is future. She argues that in such a world God can “see” everything that he has created “at once,” including the choice that a person has made on the basis of their own freewill. Thus, the direction of explanation of why God knows what we will do is not that God knows it or causes it, but rather God knows what free actions we engage in because we freely choose them. God’s knowledge is explanatorily dependent on our choices rather than the other way around. Rhoda argues that given God’s timelessness and the principle that truth supervenes on being, Rogers’ view of time does not avoid fatalism. This debate draws together debates in the metaphysics of time, God’s relation to time and creation, and freewill. The topics discussed in this book, and the chapters on those topics, draw our attention to the cutting-edge work being done in the philosophy of time and related topics. The give and take provides the reader with at least two sides of an issue and leaves the discussion open for readers to arrive at their own conclusion. It demonstrates the importance of time in contemporary debates in metaphysics.2

Notes 1 Here and elsewhere in this Introduction I am grateful for helpful comments by Barry Dainton. 2 I wish to thank Francesco Orilia for his comments on a earlier version of this Introduction.

Reference Dolev, Y. (2007), Time and Realism: Metaphysical and Antimetaphysical Perspectives. Cambridge, MA: MIT Press.

1 Dolev’s Anti-Metaphysical Realism: A Critique L. Nathan Oaklander

I

n Time and Realism: Metaphysical and Antimetaphysical Perspectives, Yuval Dolev (2007) argues that the metaphysical debate between tensed and tenseless theorists (presentists and eternalists; A-theorists and B-theorists) in the philosophy of time is ultimately incoherent.1 His perspective is unlike those who argue that these views are indistinguishable because the apparent debate is either based on semantic confusions centering on the equivocal uses of the word “exist,” or on the indistinguishability of tensed-transition and tenseless-transition.2 Although Dolev believes that working through the metaphysics of the tensed and tenseless views is indispensable if we are to arrive at a proper understanding of time, he argues that the debate must ultimately be transcended by recognizing that both views rest on a common thread—an ontological assumption—that cannot be sustained. The ontological assumption is that “tense concerns the ontological status of things” (p. ix); which implies that the central question in the philosophy of time is whether past, present, and future objects are ontologically on a par, being equally real, or are there real ontological differences between these tense determinations? For example, does the present have a reality that the past and future do not have? Dolev argues that these queries depend on the metaphysical theories that give rise to them and since those theories make use of terms, such as, “tenseless relations” and “the moving now,” that are themselves unintelligible and empty notions, the theories and questions they give rise to and sought to answer are also unintelligible and empty. He believes that by coming to this realization, “a systematic dissolution of the traditional philosophical questions concerning time is brought about” (p. 60). The aim of this chapter is to defend the metaphysics of time against Dolev’s attack. My response to Dolev is to demonstrate that his understanding of the

4

DEBATES IN THE METAPHYSICS OF TIME

debate, as centering on the reality of past, present, and future events and things, while typical, is not sufficiently sensitive to the more fundamental ontological differences between the disputants in the metaphysics of time. For that reason he fails to consider a third ontological alternative to his characterization of the tensed and tenseless views. With the emergence of this third metaphysical theory of time, we shall see that Dolev’s main arguments against the coherence of traditional metaphysical questions concerning time can be refuted. First, a few words about ontology. Ontology has as its subject matter everything that exists or all the entities there are, and its aim with regard to that subject matter is to determine what categories or most general principles of classification there are, and then to say something about the relations between those categories. Of course ontology does not consider each existent one by one, but is concerned primarily with the most general categories (for example, things, relations, qualities, identity), founded on the most ubiquitous phenomena. Its aim is to specify to what category or categories certain general classes of phenomena belong, and then to say something about that category.3 Thus, to answer the ontological question “What is the nature of time?” is to give an inventory of all temporal entities, or rather, of the category or categories of entities they belong to. Certainly, time is a basic and ubiquitous phenomenon and thus is within the purview of ontological explanation. What then are the temporal phenomena and on what category or categories are temporal phenomena based? Before addressing those questions I want to briefly consider aspects of C. D. Broad’s conception of philosophy and relate it to the ontology of time. This background will set the stage for a consideration of Dolev’s critique of the metaphysics of time, and my critique of Dolev. According to Broad, one aspect of the subject matter of philosophy— critical philosophy—consists of “the analysis and definition of our fundamental concepts, and the clear statement and resolute criticism of our fundamental beliefs” (Broad 1923: 18). In our everyday dealings with the world, we make use of general concepts and apply them without having a clear idea of their meaning or their relations. Common sense constantly uses concepts in terms of which it interprets experience, for example, when “it talks of things of various kinds; it says that they have places and dates, that they change, and that changes in one cause changes in others, and so on. Thus it makes constant use of such concepts or categories as thinghood, space, time, change, cause, etc” (1923: 15). For ordinary purposes it does not matter that we are not clear about the precise meaning and relations between and among the concepts we employ, but for the purposes of determining what entities fall under these concepts their meaning must be clear and unambiguous. The second task of critical philosophy is to take those uncritically accepted, deeply rooted beliefs that we employ in ordinary life and in the sciences, to



Dolev’s Anti-Metaphysical Realism: A Critique

5

state them clearly and then to subject them to criticism. Of course, in order to state clearly our deeply held beliefs such as that “events pass or flow through time from the future to the present and into the past,” or that “every change has a cause,” we must first know exactly what is meant by time’s flow or passage, the notions of past, present, and future, and the concept of change and cause. Thus, the critical examination of beliefs presupposes an analysis of the notions employed, and they too must be subject to critical examination. Only in that way can we have some degree of certainty that we have arrived at the truth. Russell once expressed the sentiments involved in Broad’s notion of “Critical Philosophy” in the following passage: The process of sound philosophizing, to my mind, consists mainly in passing from those obvious, vague, ambiguous things, that we feel quite sure of, to something precise, clear, definite, which by reflection and analysis we find is involved in the vague thing that we start from, and is, so to speak, the real truth of which that vague thing is a sort of shadow. (Russell 1918/64: 179–80) For Broad, the process of arriving at “the real truth of which that vague thing is a sort of a shadow”—that is, the proper analysis or description of a concept, and the phenomena on which that concept is based—is facilitated by making use of what Broad calls The Principle of Pickwickian Senses. According to this principle, the proper analysis of a phenomenon or a concept may not be what common sense implicitly and unknowingly takes it to be, since the implicit analysis, if in fact there is one, may be subject to dialectical difficulties and not hold up to critical examination. The Pickwickian sense of a concept, although perhaps not intuitive, has the advantage that it is quite certain that there is something that answers to it, “whereas with the other definitions of the same entities this cannot be shown to be so” (1924: 93). Broad gives as an example, Whitehead’s Pickwickian definitions of points, and moments in which it is certain: (α) that they exist; (β) that they have to each other the sort of relations which we expect points and moments to have; and (γ) that there is an intelligible and useful, though Pickwickian, sense in which we can say that volumes are “composed of” points, and durations of moments. (Broad 1924: 95) Similarly, the existence of certain phenomenological facts that need an ontological ground may be accounted for by a set of facts that might not be what common sense implicitly takes them to be, but they may be the best

6

DEBATES IN THE METAPHYSICS OF TIME

we can get and entirely suitable to account for the phenomenology of the situation. In other words, the principle allows that even if an analysis is not the one that is implicitly assumed by common sense (assuming with Broad, for the moment, that there is an implicit ontology in common sense), we can still accept that there exist entities that fall under the concept or ground the phenomenology, and justify the application of the concept. Broad uses the concepts of the “self” and “matter” to explain an error that can come about from failing to see the distinction implied by The Principle of Pickwickian Senses between our ordinary concepts, beliefs, and phenomenological data on the one hand, and the analysis or proper description of those concepts, beliefs, and data on the other. One such error occurs in the following passage where Broad notes that questions such as “Does matter exist?” or “Is the self real?” cannot be answered with a simple yes or no. He continues, Unquestionably there are facts in the world to which the names “matter” and “self” apply; and in that sense they are names of something real. But it is vitally important to distinguish between facts and the proper analysis or description of facts. The words “matter” and “self,” as commonly used, do suggest certain theories about the facts to which they are applied. These theories are never clearly recognized or explicitly stated by commonsense; and, on critical analysis, they are often found to consist of a number of propositions of very different degrees of importance and certainty. E.g., I think there is very little doubt that the word “self,” as commonly used, implies something like the Pure Ego theory of the structure of those entities which we call “selves.” Hence anyone who rejects the Pure Ego theory is, in one sense, “denying the reality of the self.” But, if he offers an alternative analysis, which does equal justice to the peculiar unity which we find in the things called “selves,” he is, in another sense, “accepting the reality of the self.” Whenever one particular way of analyzing a certain concept has been almost universally, though tacitly, assumed, a man who rejects this analysis will seem to others (and often to himself) to be rejecting the concept itself. (1924: 94–5) According to Broad, however, that would be a mistake. We must distinguish, Broad says, between facts, what I shall call “common-sense facts,” and the proper analysis or description of those facts, what I shall call “ontological facts.” There are, undoubtedly, common-sense facts to which the term “self” applies, that is, there are selves commonsensically speaking, and it may be, as Broad suggests, that a particular ontological analysis of the self is tacitly accepted (though never clearly recognized or explicitly stated) by common sense. However, if upon critical examination it is seen that there are reasons



Dolev’s Anti-Metaphysical Realism: A Critique

7

to reject that analysis and if an alternative analysis of the self can be given that accounts for all of the common sense and phenomenological facts that need to be accounted for, such as “the unity of the self,” then one can accept that the concept of the self has an application and thus that the self exists. Thus, it is a mistake to assume that if one rejects what is tacitly assumed by common sense (or a specific ontological analysis), one is thereby rejecting the concept and denying that there exist entities that fall under that concept. Broad gives an example of this mistake in the following passage: Thus, James raises the question: “Does Consciousness Exist?,” and suggests a negative answer. But really neither James nor anyone else in his senses doubts the existence of certain facts to which we apply the name “consciousness.” The whole question is: “What is the right analysis of these facts?” Do they involve an unique kind of stuff, which does not occur in non-conscious facts; or is their peculiarity only one of structure? To deny the first alternative is not really to deny the existence of consciousness; it is merely to deny an almost universally held theory about consciousness. (1924: 95) In this passage, Broad is drawing a distinction between the common-sense facts or pre-analytic data and the ontological facts or the ground of that data; a distinction that is at the heart of Broad’s doctrine of Pickwickian senses. He believes that failing to recognize it can easily lead to “unprofitable discussions” (1924: 95). In his book, On Philosophical Method (1980), Hector-Neri Castañeda articulates in detail something like this distinction.4 However, whereas Castañeda speaks of “protophilosophical data” and philosophical theories that compete in trying to elucidate these data, Broad seems to believe that common sense contains or implies an articulate, although perhaps unacceptable, ontological theory of various phenomena. Indeed, his doctrine of “Pickwickian” senses is meant to highlight that the correct philosophical analysis of a category is intended to be taken in a sense other than the literal one implicit in common sense. Of course, one may question, as I do, whether common sense or intuitive beliefs about the self, matter, consciousness, time, or whatever, actually reflect an implicit ontology at all, and even if they do, whether that has ontological significance.5 However, regardless of whether or not ordinary concepts and beliefs have an implicit, unrecognized ontology, Broad and Russell are surely right in maintaining that the difference between the phenomenological data and pre-analytic, common-sense facts, on the one hand, and the analysis of that data or the ontological facts that are their ground, on the other, is important and useful in arriving at the real truth that

8

DEBATES IN THE METAPHYSICS OF TIME

underlies the vague facts we start off with. In other words, even if common sense has an implicit theory about the meaning of a concept it employs, that does not imply that the proper ontological analysis of those facts is implicit in or given by common sense. In addition to overlooking a third metaphysic of time, one of the major pitfalls in Dolev’s critique of “tenseless relations” and more generally his anti-metaphysical stance regarding the philosophy of time is his failure to properly recognize Broad’s Principle of Pickwickian Senses and, more broadly, the distinction between common sense and ontology that follows from it. To begin to see what is involved in these claims and to defend them, I shall first turn to a brief discussion of temporal phenomenology. There are two aspects of our experience, thought and language of time that philosophers of time have taken to be of crucial importance. Broad refers to these features as the extensive and the transitory aspects of time. The extensive aspect of time consists of the fact that any two experiences of the same person stand to each other in a determinate temporal relation of earlier/later than or simultaneity.6 The transitory aspect of time consists of the fact that events or experiences that are once wholly in the future keep on becoming less and less remotely future, eventually become present, and then cease to be present, recede into the immediate past, and then keep on becoming more and more remotely past. I should add that when some philosophers talk of the transitory aspect of time they also have in mind the idea that time is dynamic, that time involves a flow or flux from one event to another one, and not a static relation between them. Broad claims, regarding the transitory aspect: There is no doubt that the sentences which I have just been quoting [e.g., “Thank God (on the theistic hypothesis) that’s over now!”] record facts and that such facts are of the very essence of Time. But it is, of course, quite possible that the grammatical form of these sentences is highly misleading. It may dispose people to take for granted a certain view of the structure and the elements of these facts and this view may be mistaken and may lead to difficulties and contradictions. (1938: 267) This passage is ambiguous because the notion of “fact” is ambiguous. At the level of common sense it is a fact—a “common-sense fact”—that events, including experiences, stand in the relations of earlier/later than and simultaneous with, other events. It is also a common-sense fact that events which were once in the future become present and then recede into the past. At the pre-analytic level of common sense, the existence of such facts is ontologically neutral in that it may suggest a certain ontological analysis, but it does not commit one to that analysis. Thus, when Broad says that the



Dolev’s Anti-Metaphysical Realism: A Critique

9

facts recorded by sentences such as “I am going to have a painful experience at the dentist’s tomorrow,” are of the essence of time, he may be using “fact” to refer to a common-sense fact. In that case, no specific view of the ontological analysis of the structure of that fact is required or implied based on the grammatical form of the natural language tensed sentence used to record it or the phenomena on which it is based. However, when Broad says that sentences about the becoming of events record facts that are of the “very essence of Time,” he may be using the notion of “fact” in a metaphysically loaded sense. In that case, tensed sentences in a metaphysically perspicuous language represent ontological facts that are the result of philosophical analysis about the essence of temporal reality. Without prejudging what Broad takes the structure and the elements of the ontological facts of time to be, we can use these two notions of “fact” to both clarify one understanding of “tenseless relations” and “tenseless time” found in Broad and Russell (hereafter called the “B/R” theory), and demonstrate that Dolev fails to consider this version of “tenseless time” in his critique of the traditional problems of the metaphysics of time. In his first writings on time, while under the influence of Russell (1915), Broad claimed that the experience of succession is the foundation of all other awareness of temporal phenomena; that all the common-sense facts about time can be grounded in ontological facts based on the extensive aspect of time. Consider this passage from Broad’s encyclopedia article on “Time,” Temporal characteristics are among the most fundamental in the objects of our experience, and therefore cannot be defined. We must start by admitting that we can in certain cases judge that one experienced event is later than another, in the same immediate way as we can judge that one seen object is to the right of another. A good example of the immediate judgment in question is when we hear a tune and judge that of two notes, both of which come in our specious present, one precedes the other. Another direct judgment about earlier and later is made in genuine memory. On these relations of before and after which we immediately recognize in certain objects of our experience all further knowledge of time is built. (1921: 143) By taking the relation of before and after (or earlier than/later than) as fundamental and indefinable, and claiming that all further knowledge of time is built from those relations, Broad is asserting that temporal relations must be taken as simple, unanalyzable entities of one’s ontology. To say that they are simple and undefined implies that they are irreducible and so cannot be analyzed or reduced to the non-relational temporal properties of pastness, presentness, and futurity of their terms and for the early Broad (and Russell)

10

DEBATES IN THE METAPHYSICS OF TIME

there are no such mind-independent or mind-dependent monadic temporal properties.7 It means, moreover, that temporal relations can exist even if their terms do not have non-relational temporal properties. In the TENSEless (B/R) theory, time consists solely of relations. Of course, there are other temporal entities. The terms of temporal relations, and the ontological facts that have temporal relations as constituents, may also be called “temporal” in virtue of their connection with temporal relations which are thereby the only intrinsically temporal entities in this relational ontology of time. On the B/R theory, there are no temporal individuals, such as moments or time points; there are no monadic temporal TENSED properties; and there is no absolute becoming understood either as the coming into and going out of existence of objects or events, as the donning and doffing of TENSED properties, or in any other way, for example, the accretion of facts.8 When Broad asserts that all further knowledge of time is built up from the relations of before and after, we should not take him to be denying the reality of passage or the transitory aspect of time, nor need he be interpreted as claiming that the passage of time is an illusion, but rather to be claiming that the common-sense fact that time passes or flows can be grounded with ontological facts involving exclusively temporal relations between events including mental events. Dolev claims that “phenomenology preempts ontological issues before they can rear their head” (p. 10), but for some, and I would count Broad among them, ontology and phenomenology go together. Phenomenology provides the subject matter of ontology; it does not preempt ontological issues since different ontological theories may be claimed to comport with the temporal phenomena. The B/R theorist can agree that the passage of time involves an experience of “the flow of successively existing events” (Paul 2010: 334), or as Donald C. Williams once put it, “we are immediately and poignantly involved in the jerk and whoosh of process, and the felt flow of one moment to the next” (Williams 1951: 465–6). What I am suggesting is that when Broad and Russell appeal to the experience of succession and characterize that relation as the simple and unanalyzable ground of “all further knowledge of time,” it is consistent with such a view and the phenomenological data to maintain that the ground of the flow of successively existing events, and the felt flow from one moment to the next, is in these B/R TENSEless relations alone. On this interpretation of Broad and Russell, TENSEless relations are dynamic and not static, and the ontological ground of the extensive and transitory aspects of time is built up from those relations. In this version of the TENSEless view, although not in some others, it is a mistake to claim that in the TENSEless view, TENSE is a mind-dependent illusion. In this regard, the B/R view is a third metaphysical alternative distinguishable from other versions of the tenseless view.



Dolev’s Anti-Metaphysical Realism: A Critique

11

Thus, for example, Laurie Paul sets herself the task of explaining how the existence of a static, four-dimensional universe of a series of changeless events standing in unchanging temporal relations can explain the “flow of successively existing events … responsible for the animated character or flow of change” (Paul 2010: 334). Paul responds by arguing that even in the static universe of the four-dimensionalist, the reductionist can provide “an account of how temporal experience could arise from the way the brains of conscious beings experience and interpret cognitive inputs from series of static events” (2010: 339). Her explanation goes something like this: When we have an experience as of passage, we can interpret this as an experience that is the result of the brain producing a neural state that represents inputs from earlier and later temporal stages and simply “fills in” the representation of motion or of changes. Thus, according to the reductionist, there is no real flow or animation in changes that occur across time. Rather, a stage of one’s brain creates the illusion of such flow, as the causal effect of prior stages on (this stage of) one’s brain. (2010: 352) Paul is claiming that our experience of passage is an illusion, and therefore while time seems to pass from one moment to the next it does not really do so: it is just a mind-dependent phenomenon with no objective reality. Paul’s B-theoretic move is to make transition something that does not exist in the world. In her version of the tenseless theory, there are durationless events that are temporally related, but there is no objective transition or becoming. Thus, the tenseless view is called the “static view” of time because the experience of dynamism does not represent any flow from one time to another since there is none. The assumption of this line of reasoning is that the only kind of objective flow is A-theoretic or TENSED. Actually, Paul’s view is that transition is a double illusion. First, it is not a feature of the static events that cause it, and furthermore, we are not really aware of transition. Temporal phenomena seem to pass, but passage as we experience it is different from how it seems, if that makes any sense. The experience of flow between events “just gives the impression of being filled in. There is no ‘figment’ as Dennett would say” (Paul 2010: 353n. 33). Thus, the second illusion consists of the fact that what appears to be the experience of transition is not really the experience of transition at all.9 Barry Dainton criticizes Paul and adopts a weaker (less illusory) form of the experience of passage, compatible with a different version of the tenseless view. He maintains that our experience of the dynamic aspects of time are fully real experientially, and they do possess dynamic qualities—the flux and flow we find in our experience is not an illusion—but what is an illusion is the belief that these features of experience represent a mind-independent reality that contains metaphysical, that is, A-theoretic passage. Dainton claims that

12

DEBATES IN THE METAPHYSICS OF TIME

… if our universe is of the Block variety then it is certainly the case that no form of M-[metaphysical] passage exists—this holds by definition. But we can be certain that E [experiential]-passage exists, as certain as we are of anything and we can conclude from this that our universe contains at least one significant form of passage—that certain regions of it have an inherently dynamic intrinsic nature. And this result holds even if our universe is entirely devoid of any form of M-passage. (2012: 132) Like Paul, Dainton’s assumption is that if mind-independent (metaphysical) passage exists, then it must be A-theoretic or TENSED. Nevertheless, he claims that There is thus a substantial difference between the position that I have been recommending, and what is being advocated by Paul. I agree that E-[experiential] passage exists in the realm of appearances, and that to the extent that these appearances misrepresent the (non-dynamic) external physical reality they can in this respect be construed as misleading or illusory. Nonetheless, the appearances in question are nonetheless fully real experientially, and the experiences in question really do possess dynamic characteristics. In this sense, there is nothing in the least illusory about the flux and flow we find in our experience. (2012: 133; emphasis added) In asserting the appearance–reality distinction with regard to temporal passage or dynamism, Dainton is clearly assuming that our experience of passage is TENSED in an A-theoretic sense, and for that reason “misrepresents the (non-dynamic) external physical reality” (2012: 133). The B/R theory rejects both the strong (there is no passage phenomenologically or ontologically) and weak versions of the tenseless view (there is experiential but no mind-independent passage) since it affirms that we do experience passage and that in so doing we are directly aware of mindindependent—albeit B/R-theoretic and not A-theoretic—passage. Thus, the B/R theory rejects the assumption that if metaphysical or experiential passage exists then the TENSED theory in some form must be true. As we have just seen, Paul and Dainton believe that assumption, but there is an alternative. Recalling our earlier discussion of Broad and Russell, we can say that the vague truth that we start off from is that time has a dynamic character. There is a flow, flux or whoosh to time and that is something that is given to us in our immediate experience. This experience is open to many different ontological interpretations, but the real truth that underlies the experience and is its ontological ground is the existence of unanalyzable temporal relations between temporal objects. The B/R theorist who is a phenomenological



Dolev’s Anti-Metaphysical Realism: A Critique

13

realist will reject the view that our experience of passage is an illusion or an appearance that “misrepresent the non-dynamic external physical reality” (Dainton 2012: 133). The experience of the dynamic aspect of time is not the experience of a mind-dependent object that misrepresents a static reality, but is the perception of a mind-independent reality that is grounded in a temporal (dynamic) simple B/R relation that is different from all other relations. Thus, it is a mistake to claim that there is a distinction between the succession as we experience it and succession as it is in itself; the former being dynamic and illusory, and the latter static and real. In our experience of the phenomenon of succession which grounds the dynamic aspect of time, we are directly acquainted with a TENSEless B/R-theoretic mind-independent feature of reality. To think otherwise is to assume that the dynamic aspect of time given in experience is founded on the subjective appearance of ontological TENSE, and it is that which a B/R theorist will deny.10 There is more that can and should be said about the B/R account of the transitory aspect of time, and the various phenomenological data that are connected with it, for example, the different psychological attitudes toward past, present, and future events. Enough has been said, however, to see how the B/R theory I have described is a third metaphysics of time distinguishable from the tensed theory and both Paul’s and Dainton’s versions of the tenseless view. With this background, we are, therefore, ready to turn to Dolev’s critique of the metaphysics of time. The overall structure of Dolev’s argument against tenseless time can be stated as follows: Dolev construes our ordinary concept of time, as well as our experience and language of time, as inescapably and indispensably tensed. He also maintains that the tenseless theory, while maintaining that tense is an indispensable and inescapable mode of thought, experience and language, denies the reality of temporal passage and past, present, and future, claiming that it is an illusion. He concludes that the “tenseless view is in a way self-refuting” (p. 99); that “the notion of purely tenseless relations is empty” (p. 95) and that the tenseless view is unknowable. We shall see, however, that Dolev’s arguments against the coherence of tenseless relations fail primarily for two reasons. First, they ignore Broad’s Principle of Pickwickian Senses, since he confuses the pre-analytic data, or common-sense facts, with the ontological analysis of them. This error manifests itself in an equivocation of the notion of “tenseless relations.” Second, Dolev does not consider the B/R analysis of TENSEless temporal relations. For those reasons his overall argument that the metaphysical theories and the queries they give rise to are empty, since the notion of “tenseless relations” is unintelligible, can be set aside. To establish his conclusion he would have to establish that the same arguments apply equally to B/R relations, and that he does not and cannot do. To see why, I shall next turn to his account of “tenseless relations.”

14

DEBATES IN THE METAPHYSICS OF TIME

Dolev initially defines “tenseless relations” as follows: “In general tenseless relations are defined to be relations of succession: we give the tenseless relation between events e1 and e2 when we say that e1 is later than, or earlier than, or simultaneous with e2” (pp. 4–5). This way of defining “tenseless relations” suggests that the copula is tenseless, that is, does not involve grammatical tense or ontological TENSE. For that reason, the meaning of “tenseless” cannot be cashed out in terms of “is now,” “was” or “will be” (as some have claimed)11 since a tenseless sentence is not omnitensed, but literally without tense. To interpret tenselessness as omnitensed would make Dolev’s claim that there are no purely tenseless sentences about events trivially true, and it would support the thesis that relations of succession are permanent. Critics of the tenseless view often argue that since “e1 is earlier than e2” is always true it follows that the fact in virtue of which it is always true is eternal. Since to be “eternal” is to be “everlasting” or to “always exist” or to have “endless duration,” it is concluded that tenseless facts are eternal facts and so are permanent. That, however, is a mistake, since temporal relational facts do not exist at any time; much less do they exist at every time.12 Nor does “tenselessness” mean the same thing as “timelessness” since the terms of the tenseless relation of earlier than are not timeless entities like numbers, but temporal objects like particulars. Furthermore, the copula involved in sentences expressing TENSEless temporal relations is tenseless, but temporal, because it is asserting that one temporal object is earlier than another temporal object.13 So, just as is now is a temporal copula (although a tensed one), is earlier than is a temporal copula as well (although a tenseless one). Leaving these distinctions aside, in so far as Dolev’s initial characterization of “tenseless relations” is simply meant to be “relations of succession” without ontological implications, his description is unproblematic. Dolev makes other remarks about tenseless relations that are more controversial. He says that “Such relations, we are told, are easily recognized: their conspicuous hallmark is that sentences describing them are true, if true, regardless of when they are tokened” (p. 94). Indeed, he claims that “the tenseless view is sustained by the straightforward distinction between tenseless and tensed sentences … tenseless sentences state tenseless relations, namely, relations of precedence, succession, and simultaneity” (p. 79; emphasis added). Thus, Dolev views tenseless relations as “sustained” and “recognized” by certain facts about the sentences describing or stating them. These claims clearly show that Dolev assumes that the bases of the notion of tenseless relations are tenseless sentences, but that is incorrect. The existence of TENSEless temporal relations as entities in the ontology of time is recognized by our experience of time, and not by the language that philosophers may use to express that experience. For that reason, Dolev is



Dolev’s Anti-Metaphysical Realism: A Critique

15

mistaken in claiming that the notion of a TENSEless relation depends upon a language from which tense has been eliminated. The language that philosophers use to express temporal relations between and among events starts from phenomenology and then attempts to ground that phenomenology in an ontology that can account for it. To consider and further clarify his remarks concerning “tenseless relations,” I want to discuss his account of the debate between tensed and tenseless views of time. Dolev explains what he takes to be at issue between the tensed and tenseless views in the following passage: We all know that the American Revolution preceded the French Revolution. We supposedly all know, or at least ready to admit once it is explained to us, that this fact constitutes a tenseless relation between the two events. Indeed, tenseless theorists take it for granted that there are such tenseless relations. Their quarrel with tensed theorists concerns the question of whether in reality the only type of temporal relations are tenseless relations, or whether there are in addition, also tensed relations. The point of contention concerns the exclusivity of tenseless relations, not their existence. (p. 79; emphasis in last sentence added) Before proceeding to criticize the contents of this passage, I shall explain Dolev’s notion of a “tensed relation.” He says that “The location of an event with respect to the present is referred to as the tensed relation of the event … Now we may ask: are tensed relations part of reality, is there a present with respect to which events really stand in a temporal relation, or is it the case that, as was claimed about color at one time … there is no present outside our apprehension and so nothing for events to have a tensed relation to?” (p. 5). A tensed relation is a temporal relation (presumably earlier/later than) between the present and past and future events, but what is the ontological status of this relation? Is it an entity in its own right over and above the terms it relates, or is it grounded in the non-relational properties of its terms? And if it is an entity, does it depend on its terms having TENSED properties? Dolev clearly believes that the tensed view is committed to (non-relational) TENSED properties, since he treats the tensed view as maintaining that “tensed properties of events must be included in our conception of reality” (p. 39), and “The difference between the tensed and the tenseless accounts is that in the former truth conditions are non-relational, whereas in the latter they are relational” (p. 21). However, as we shall see, these remarks concerning tensed and tenseless relations render his account of the difference between these two theories of time problematic. Dolev claims that “the American Revolution preceded the French Revolution” states a tenseless fact whose constituents are a tenseless

16

DEBATES IN THE METAPHYSICS OF TIME

relation and two events, but surely that the American Revolution preceded the French Revolution does not constitute a tenseless relation between the two events, but a tensed relation since the events related are past in relation to the present. Furthermore, given Dolev’s characterization of “tenseless relations” as those described by sentences with an unchanging truth value, it follows that “the American Revolution preceded the French Revolution” is a tensed sentence, since it has a changing truth value: false before the American Revolution and true after the French Revolution. Therefore, the fact constitutes a tensed relation between events, and the sentence describes a tensed fact, not a tenseless one. Dolev claims that there is a basic agreement between the tensed and tenseless views and in so doing his characterization of the debate begs the question against the tenseless view and misses the ontologically fundamental issue between them, an issue that is more basic than the ontological assumption he rejects. Clearly, no tenseless theorist, including the B/R theorist, would accept that the ontological fact that the American Revolution preceded the French Revolution constitutes a tenseless relation between two events, since the sentence expressing this fact is tensed and therefore from an ontological point of view reflects that the terms of the relation have the TENSED property of pastness. Thus, Dolev’s characterization of tenseless relations assumes that the ontological analysis of them is that they are relations with TENSED determinations, and that the tenseless facts temporal relations enter into exist in time since the sentences that express them, for example, “the American Revolution preceded the French Revolution,” change their truth value. In so doing, his characterization of tenseless temporal relations renders the tenseless view contradictory and so in the debate with tensed theorists begs the question against it. Furthermore, to assume at the outset that the tensed and the tenseless views agree on the existence of tenseless relations and only disagree about the existence of tensed relations is to misunderstand the metaphysical dispute in the philosophy of time, since in the B/R theory, “TENSEless” relations are universals whose terms are particulars that do not exemplify TENSED properties (pastness, presentness and futurity), and the facts they enter into are timeless in just this sense: though they do not exist in time since they do not exemplify temporal relations, time (temporal relations) exist in them. No tensed theorist could accept that time contains a conjunction of such ontological facts. Hence there is a fundamental dispute about the existence of, and not merely the “exclusivity” of, tenseless relations that Dolev fails to see by overlooking the B/R account. Admittedly, if by “tenseless relations” Dolev means the common-sense fact that events stand in temporal relations, or even more neutrally, that sentences such as “the American Revolution preceded the French Revolution” are true,



Dolev’s Anti-Metaphysical Realism: A Critique

17

then both the tenseless and the tensed views agree, pre-analytically, that there are tenseless relations.14 However, they would disagree about the analysis of tenseless relations or the ontological facts they involve, as well as the relation of those facts to time. Of course, if Dolev then shifts from common sense to the ontological ground of the pre-theoretical data, and claims that the tensed and tenseless views disagree with regard to the existence of tensed relations, then what he saying would be true. That is, the tensed and tenseless views agree commonsensically that there are temporal relations and disagree ontolo­ gically over whether or not there are TENSEless or TENSED temporal relations and facts. However, once common sense and ontology are kept separate, it should be clear how and why these views do not even partially agree. In Chapter 4, “Tense Beyond Ontology,” Dolev gives his main arguments against the tenseless view. He reasons that since tense is inseparable and ineliminable from language, thought and experience, “the notion of ‘tenseless relations’ remains empty” (94); “there is not one fact we can point to as truly tenseless” (94); there is not a separable concept of tenseless time, and indeed, tenseless relations are something “we-know-not-what” beyond the veil of perception. In what follows, I shall consider his arguments for these radical theses beginning with the claim that since tense is ineliminable from language and “all factual utterances are always infused with tense” (p. 92) there are no tenseless facts. My first response to Dolev’s argument against tenseless facts is that it equivocates on the notion of “factual utterance” when he says that “all factual utterances are always infused with tense” (p. 92). If a factual utterance is one that states a common-sense fact, then even if in ordinary language all common-sense facts are infused with tense, it does not follow, given Broad’s Principle of Pickwickian Senses, that all ontologically factual utterances are infused with TENSE. On the other hand, if by “factual utterances” Dolev means “utterances that state ontological facts,” then what he is saying is either false or begs the question of whether from the ineliminability of tense in ordinary language, thought, and experience it follows that all factual utterances in a metaphysically perspicuous language are infused with TENSE. Dolev argues that the ineliminability and indispensability of tense in ordinary language, thought, and experience is evidence that TENSE is ineliminable from all factual utterances and hence from the ontological facts they describe. In other words, if there are no factually tenseless sentences then there can be no TENSEless facts. How then does he argue that all purportedly tenseless sentences are in fact tensed? He considers two sentences that are or contain tenseless sentences and argues they are tensed: “John Kennedy was assassinated in 1963” and “Event e occurs in 2007.” His argument against both is that they involve dates, and dates assume a prior understanding of tense. Thus, he says:

18

DEBATES IN THE METAPHYSICS OF TIME

Consider again the sentence “Kennedy was assassinated in 1963.” To understand it—for such a sentence to transmit knowledge—one must know what “1963” refers to. It is not enough to know that these numerals indicate a counting that starts at a certain chosen point in time, that 1,963 years separate between that point of origin and the assassination. This tenseless fact is a useless fact to anyone who does not know when the point of origin is with respect to the present. (p. 91) Of course, Dolev is correct that “Kennedy was assassinated in 1963” presupposes the concept of tense since it is a tensed sentence! The question, however, is whether present tense is necessary and indispensable to understand dates and our ability to act and communicate successfully. If so, then presumably that is because reality is TENSED; that TENSE must figure in some way in the fact that event e occurs at time t, or that e1 is earlier than e2, and thus that there are no TENSEless facts. Dolev argues that, “If one never knows the present date, then one can never use information about the dates of events to act successfully” (p. 91). Thus, for example, it is not sufficient to know that the meeting starts (or tenselessly occurs) at 1 pm to know that I need to take action to go to the meeting which starts now, that is, that it is now 1pm, and that the further knowledge involves the time possessing the tensed property of presentness. Since the tensed thought is indispensable to timely action, tensed facts or relations are indispensable too. However, tensed thoughts are not indispensable. Admittedly, knowing the TENSEless fact that the meeting starts at 1 pm is not sufficient to get me to act, nor is the clock reading one o’clock sufficient, but I can know it is time to act by being conscious of perceiving the clock striking one, and judging that this perception of the clock is roughly simultaneous with the start of the meeting. In other words, I can know what event or time is present without knowing that the event or time has the TENSED property of presentness. Similarly, I can understand that the date 1945 is the date of my birth without taking that date to be in a tensed relation to the present. I can understand “today is the date of my birth” by judging reflectively that the date of my birth is 69 years earlier than this thought (of being born in 1945), and “the date of my birth is 1945” simply means that the event of my birth is simultaneous with the 1945th revolution of the earth around the sun since the birth of Christ or some other suitable event. Dolev is perfectly aware that there are tenseless theorists such as Mellor who agree that tense is ineliminable from ordinary language (and commonsense facts), but deny that therefore TENSE is a feature of reality since the ontological facts that ground tensed sentences in ordinary language can be given in terms of tenseless sentences that express TENSEless facts. Nevertheless, Dolev objects that the purported tenseless sentences are not



Dolev’s Anti-Metaphysical Realism: A Critique

19

purely tenseless, since they presuppose a tensed context. Thus, he argues that the token-reflexive account, according to which the truth condition of “x is present” is the fact that “x is simultaneous with the utterance ‘x is present’” or “x is simultaneous with this utterance” does not eliminate tense. As he puts it: The problem with the token-reflexive account is that, contrary to the supposition of those who rely on it, the sentences and relations employed in handling tensed relations are not themselves purely tenseless. It is true that “Today’s date is February 15, 2007” can be explained by saying that the date of the tokening of this sentence is February 15, 2007. But that is only because the context makes it clear that the phrase “this sentence” refers to the sentence tokened now. Before the “tenseless” formulation is “tense-ized” by the context, it cannot do anything by way of clarifying the original sentence. (pp. 92–3) Is Dolev correct that the use of the phrase “this sentence” or “this perception” refers to the sentence or perception tokened now and thus cannot be part of an analysis that eliminates tense or TENSE? I think not. In fact, Smart (1963: 194) has already replied to this line of thinking by noting that “it is simply a dogmatic rejection of the analysis in terms of token-reflexiveness. On this analysis ‘now’ is elucidated in terms of ‘this utterance,’ and not vice versa.” This elucidation, notes Smart, relies on taking a token of “this utterance” (or “this token”) as referring to itself directly, that is, without recourse to properties that identify it, and thus, in particular, without recourse to a tensed property of presentness or nowness. Dolev and other objectors (Broad 1928; Gale 1967; Ludlow 1999) appear to claim precisely the opposite, but I agree with Smart (1963: 195) that “it is not at all evident why the objector should think that an utterance like ‘this utterance’ cannot be directly self-referential. We hear a token of the form ‘this utterance’ and simply understand that this token utterance is the one referred to.” That Dolev thinks otherwise and insists on bringing TENSE into the picture may well be a byproduct of a presupposed allegiance to the tensed theory, but the B/R theorist will of course disavow this.15 Dolev also argues that there are no tenseless facts on other grounds, namely, that we have no notion of purely tenseless relations or facts since tensed and tenseless concepts are inseparable. Thus, he claims that … we need to have some grasp of temporal succession in order to understand sentences about the present; and vice versa, we need to have some mastery of tense to understand succession. To understand the tensed sentence “What you are hearing now is the sound of thunder” we need

20

DEBATES IN THE METAPHYSICS OF TIME

to know that the explanation refers to the sound that is simultaneous, or cotemporal, with it [the sentence “What you are hearing now is the sound of thunder.”] And we are made to understand the succession report “Thunder comes after lightning” on occasions in which we can associate these words with sounds and sights that are experienced now. (p. 92) In other words, Dolev’s point is that since our experience of the relation of succession is tensed, “there is no fact that we can point to as truly tenseless” (p. 94), and the notion of “tenseless relations” remains empty. Again, we must distinguish the pre-analytic data and temporal phenomena from the ontological analysis or fact that is its ground. Even if our concept of succession (or objects experienced in succession) is inseparable from objects that are experienced now, and in that sense the concept of succession is inseparable from the concept of now, it would not follow that there are no TENSEless relations in reality. For the ontological ground of the presentness of objects experienced is simply that they are presented in perception. In a specious present one experiences a temporal relation but does not experience the property of presentness. As Russell once put it, Succession is a relation which may hold between two parts of one sensation, for instance between parts of a swift movement which is the object of one sensation; it may then and perhaps also when one or both objects are objects of immediate memory, be immediately experienced, and extended by inference to cases where one or both of the terms of the relation are not present. (Russell 1915: 213) Thus, for Russell (as for Broad), temporal phenomenology is mind-independent since we are acquainted with the (TENSEless) relation earlier than when we see a rapid movement or hear a sequence of two tones, and there is no tense involved. Dolev’s argument that we cannot conceive of tenseless relations is based on the thesis that our experience is intrinsically tensed, that is, that we do not have non-tensed experiences. This view has been echoed by Jonathan Tallant when he says: Our experience of this “earlier” and “later” structure is intrinsically tensed. That is to say that when I experience the extended nature of the specious present, when I experience temporal priority, it is as a part of the now. There are B-theorists [Falk] who have explicitly acknowledged this: “I for one cannot have non-A-perceptions” (Falk 2003: 221). (Tallant 2007: 152)16 There are, however, two problems with this argument. First, I do not accept the claim that one cannot have non-A- (TENSED) perceptions. We do not



Dolev’s Anti-Metaphysical Realism: A Critique

21

experience the property of presentness when we perceive a present event or a succession of rapid events. When we have an unreflexive awareness of a present event, we are in fact experiencing nothing more than the occurrence of the event; we are not apprehending the irreducible property of presentness. Clifford Williams supports this very line of argument by means of a spatial analogue.17 Williams writes: “There are occasions, perhaps quite numerous, when we unreflexively experience, simply as existing, objects that happen to be in the proximity of our locations, without also experiencing them as being here” (1992: 370). Just as it is the case that we do not experience or apprehend the hereness of objects when we unreflexively experience objects that are here, so it is the case that we do not experience the presentness of events when we unreflexively experience present events. In other words, to experience something as now does not imply that what is experienced is TENSED or has the property of presentness, but only that it is experienced as presented. Second, and more importantly, it is simply not the case that our experience of the “earlier” and “later” structure is intrinsically tensed, that is, somehow founded on A-properties or A-facts. Since the “now” is, for Russell, what is simultaneous with this, where this is an object of perception, it is consistent to say that I experience temporal priority as part of the now without that implying that the temporal phenomena, that is, the experience of succession, is intrinsically tensed, or founded on tensed properties or facts. All that is required, as in fact occasionally is the case, is that we can unreflectively perceive a durational present within which a rapid succession or change can occur.18 Dolev continues his attack on the intelligibility of the tenseless theory by arguing that in the tenseless view tense is an inescapable global illusion and thus tenseless reality is an unknowable—“we-know-not-what” (p. 94). In the tenseless view, as Dolev characterizes it, nothing is in reality past, present or future, and thus our experience of time is an illusion; it belongs to how we experience things, but not to how they are. As Dolev also puts it, “in relation to time our experience is quite unlike what we experience” (p. 29); although time is TENSEless, our experience of time is TENSED. Dolev claims that events we perceive are tensedly located and that to be tensedly located is for them to possess tensed attributes (pp. 4, 97). One could no more perceive objects as lacking TENSE than one could perceive objects as lacking color. Indeed, he construes the tenseless view as maintaining that tense is analogous to color, that is, a secondary quality that exists in the mind, but not in reality. Consider the following passage: Tense—as ubiquitous in experience, thought, and language as, say, color—gives rise today to questions of the sort that troubled early modern

22

DEBATES IN THE METAPHYSICS OF TIME

philosophers concerning the so-called secondary qualities: does it belong to the things perceived or only to our perceptions of things? Or, more generally, is reality tensed, or does it only appear tensed to us? (p. 4) As Dolev interprets it, the tenseless view denies the existence of TENSED properties and so must claim that although we cannot help experiencing TENSE, we mistakenly believe it exists in reality. For that reason, the defender of tenselessness will argue, we perceive events as though they are tensely located, “But this is nothing but the way the inescapable global illusion of tense manifests itself ” (p. 100). Dolev argues that this account of temporal experience renders tenseless time an unknowable something “we-know-not-what” behind the veil of illusions. For in the case of tense, there are no conditions that we are aware of that would present us with a nontensed reality that could be the backdrop against which how things appear could be labeled “an illusion” (p. 102). In other words, since tense is an inescapable illusion, we cannot, under any circumstances, have access to veridical perceptions, and so are stripped of the condition that must obtain when one is subject to an illusion. Thus, there is a fundamental dissimilarity between the sense in which seeing the road as having a puddle is an illusion and seeing the presently setting sun is an illusion. In the one case we know it is an illusion because we can compare it with cases where a puddle on the road is real and cases where the reality is seen and known to be different from the appearance, but in the case of TENSE we cannot do that because TENSE is an inescapable illusion. To put the point otherwise, if all we perceive is an illusion, then the notion of an illusion becomes meaningless because we need to be able to compare the illusion with reality, and in the case of TENSE we cannot do that because we have no experience of a TENSEless reality. Dolev summarizes his argument in the following passage I shall quote at length: But, as before, this conclusion [that tense is an illusion] is self-refuting: conceiving veridical perceptions as unattainable in principle nullifies the logical condition presupposed by any talk of illusion. Non-tensed reality is the only source of terms required for describing how things are, a description that is a necessary backdrop against which how things appear can be labeled “an illusion.” But if tense is an inescapable illusion, we never access this non-tensed reality, and so are stripped of the condition that must obtain if we are to call tense “an illusion.” So if tense is an inescapable illusion, we have no means for saying that it is. Tenseless theorists help themselves to both ends of the stick: they acknowledge, and even insist, that tense is an inescapable mode of



Dolev’s Anti-Metaphysical Realism: A Critique

23

thought and experience, which veils tenseless reality from cognition and at the same time, they offer a theory that reveals the tenseless truth behind the veil. This cannot work. If tense is truly inescapable, then there is no way we can remove ourselves from our heads and take our invariably tensed minds for a stroll in the hidden tenseless fields of reality. And if, on the other hand, we can understand that reality is tenseless, then tense is no longer inescapable. Either way, tense cannot be thought of as an illusion. (p. 102; emphasis added) The conclusion of this argument is true—tense cannot be thought of as an inescapable illusion or appearance—but that does not constitute a refutation of the tenseless view understood as the B/R theory. An illusion or appearance is a mind-dependent object of perception, but the B/R theorist does not recognize TENSED properties as mind-independent or mind-dependent properties of experience since to do so is to give them ontological status— even mind-dependent entities are existents—and that is something the B/R view is not willing to do. Thus, Dolev sets up a false dilemma when he asks if tensed properties are real (that is, mind-independent) or if tensed properties belong merely to the way we perceive things—pure appearance, like secondary qualities and hence mind-dependent? Our experience of the present is not the experience of a mind-independent TENSED property, but it is not the experience of a mind-dependent non-relational TENSED property either. Nor is it correct to say that TENSEless reality is veiled behind the appearance of TENSE and so a something “we-know-not-what.” In the B/R view, the perceptual now is mind-dependent only in the sense that we would have no idea of it without our perception of objects, but it does not follow that the objects we perceive do not contain real mind-independent time, that is, parts that occur in succession. In summarizing the arguments of this chapter, I would say that Dolev’s critique of the tenseless view and his attack on the metaphysics of time is based on two errors. First, he overlooks the implications of Broad’s Principle of Pickwickian Senses by confusing the common-sense facts regarding our ordinary concept of time with the ontological facts that, in the tensed theory, are their ground. For that reason, Dolev assumes that if one rejects TENSED properties, as the B/R view does, then one must also reject the concept of time and the common-sense fact that time has a transitory aspect. In other words, Dolev blurs the pre-analytic data or common-sense fact, for example, that we can perceive an event as present, with a particular ontological analysis of it, and concludes that if one rejects the ontological analysis then one must reject the common-sense fact as well, or that since the common-sense fact is unassailable the TENSEless ontological analysis is thereby refuted. Thus he

24

DEBATES IN THE METAPHYSICS OF TIME

argues that since ordinary language, thought, and experience time is tensed, an analysis that denies TENSE is incompatible with the language and concept of time, and the facts of temporal experience. This, however, is to misunderstand the dispute by confusing one theory about the correct ontological analysis of time with the concept of time itself. Moreover, it rests on the false premise that ordinary language, thought, and experience is inescapably TENSED. Second, Dolev overlooks the B/R version of the “tenseless relations” by taking the tenseless and tensed views of time to agree about the existence of tenseless relations and in so doing he misunderstands the ontological nature of the dispute. Recall that Dolev claims that there is a basic agreement between the tensed and tenseless view regarding the nature of temporal relations. He claims that both the tensed and tenseless views have in common the belief in tenseless relations. What then does he means by “tenseless relations”? If “tenseless relations” are B/R relations then since B/R relations are temporal relations between terms without TENSED properties, his claim that the tensed and tenseless views both believe in tenseless relations implies that there are no tensed relations, and that therefore the tensed view is false. On the other hand, if “tenseless relations” are also tensed relations (as Dolev implies in his characterization of the dispute) then temporal relations obtain between terms with TENSED properties and the tenseless view is false. For these reasons it is preferable and more accurate to treat the issue between tensed and tenseless (including the B/R) views not as about the exclusivity of “tenseless” temporal relations but about the ontological status—the existence and nature—of temporal relations, and their relation to time. Once we recognize a third metaphysical view of time, the B/R theory, then we can see that the ontological status of temporal relations is a more fundamental issue than the “ontological assumption” that Dolev claims undermines the legitimacy of traditional metaphysical questions concerning time. Are temporal relations analyzable in term of the TENSES or are TENSEless temporal relations simple and unanalyzable or analyzable in terms of say (non-A-theoretic) causal relations? The question of whether past, present, and future, or present and past, or only present events and things, exist is parasitic on this more fundamental question regarding temporal relations. If the B/R view is right, then temporal relations are universals whose terms are temporal objects none of which are intrinsically past, present or future, that is, none of which exemplify TENSED properties, and in that sense, past, present, and future objects do not exist. If the ground of temporal relations is founded upon one of its terms—a strong version of internal relations—then only the present exists. If temporal relations are founded upon the coming into existence and continued existence of what did not previously exist, then the past and present exist. If there can be temporal relations only if their



Dolev’s Anti-Metaphysical Realism: A Critique

25

terms have TENSED characteristics, then past, present, and future exists and so does the moving now. So there is a debate over what he calls “the ontological assumption,” but that debate depends on the more fundamental issue of the ontological status of temporal relations. Since Dolev fails to recognize that debate, and the B/R version of TENSEless time that is at the heart of it, his arguments do not establish that both theories are untenable or that “we don’t really know how to understand either theory” (p. 60), much less that there is no genuine dispute in the metaphysics of time.

Notes   1 Henceforth, following Dolev, I will just use the tense/tenseless distinction to characterize these views.   2 For a discussion of these arguments and others surrounding the question of the genuineness of the so-called presentist/eternalist debate, see, for example, Dorato (2006), Savitt (2002, 2006), Callender (2012), Crisp (2004a, 2004b), Ludlow (2004), Oaklander (2001, 2008, “General Introduction,” vol. I, 1–11, 2012), Clifford Williams (1996, 1998a, 1998b, 2003), Lombard (1999, 2010), Meyer (2005), Sider (2001, 2006), Deng (2010), and Burley (2006).   3 See Tegtmeier (2012) and Grossmann (1992).   4 My thanks to Francesco Orilia for this reference.   5 In correspondence regarding the distinction between common sense and ontology, Erwin Tegtmeier commented that “The discussion in the analytical philosophy of time (as well as in other parts of analytical philosophy) seems to me unscientific. It does not take into account whole theories (in this case ontological theories) and it is unaware of the ontological alternatives. It is much too coarse. [Gustav] Bergmann would say that it is ontologically inarticulate and merely metaphorical. One could call it folk philosophy of time. The only technical component of it is mathematical logic (including set theory and the physics of time). Imagine folk physics thinking about mass without the context of a physical theory and starting from common conceptions or coarse-grained classifications. Imagine a discussion and a decision about the classical and the quantum theory of mass based on vague conceptions and without taking into account the whole of classical mechanics and the whole of quantum theory and their precise details.”   6 This is somewhat inaccurate for two reasons. First, Broad also includes the fact that every experience has some duration as one aspect of the extensive aspect of time. Second, he claims, given that our experiences overlap it is not always true that there is a definite temporal relation between any two of them. Having mentioned these qualifications, we can, however, safely ignore them in what follows.   7 The denial of A-properties as monadic is, of course, compatible with A-predicates being meaningful. The early Broad and Russell gave their meaning in terms of the token-reflexive of the psychological approach.

26

DEBATES IN THE METAPHYSICS OF TIME According to the former, if I utter a token n of “now” in uttering “it is now raining,” the sentence token can mean that rain is simultaneous with the utterance of n. According to the latter, to say that an “event e is present” means that e is simultaneous with this where this is an object of perception or the perception of an object. I shall have something more to say about both of these analyses when we consider Dolev’s arguments against tenseless time.

  8 For further explanation of the B/R theory, its difference from some versions of the B-theory, and a defense of the legitimacy of the dispute against some who attempt to debunk it, see Oaklander (2012).   9 For a criticism of this aspect of Paul’s view, see Barry Dainton (2011: 388–9; and 2012: 130–3). 10 For these reasons, I find Paul’s view that there is no passage a peculiar position to take since in countenancing “inputs from earlier and later temporal stages” (Paul 2010: 354; emphasis added), Paul is already acknowledging tenseless relations. Thus, unless Paul’s temporal relations are unlike B/R relations, her denial of the animated or dynamic character of change in our experience, which, on the version of the tenseless view I am suggesting, is grounded in primitive temporal relations, makes no sense. For if passage exists in reality in the form of inputs from earlier and later temporal stages, then any explanation of our experience as of passage that results from those inputs cannot demonstrate that passage does not exist phenomenologically or mind-independently since it assumes B/R-theoretic passage. 11 See, for example, Dorato (2006), Savitt (2002). 12 Dolev makes this error when he uses the permanence of tenseless facts in his support of arguments that Schlesinger and Prior give against the tenseless view (see Dolev 2007: 39–40). For a discussion of this error with regard to Prior’s “Thank Goodness” argument, see Oaklander (1993). 13 Since ersatz presentism takes temporal relations to be between times construed as abstract objects, the B/R view is not compatible with presentism, contrary to Rasmussen (2012). 14 Francesco Orilia has pointed out that this may be problematic because the presentist is an A-theorist but he may not want to agree pre-analytically that there is a relational fact “e1 before e2,” because it may seem to commit him to the existence of a past entity, namely e1. See also Crisp (2005). I would reply that since by “tenseless relations” Dolev means “relations of succession” at the pre-analytic level, even a presentist cannot deny them since they are committed to the view that events come into and go out of existence successively. 15 For a recent defense of the token-reflexive account of tense, see Orilia and Oaklander (forthcoming 2014). 16 For a critique of Tallant (2007), see Oaklander and White (2007). 17 C. Williams (1992). See also Oaklander (2004c). 18 Sean Power (2012) argues that for the B-theorist, the present is not “specious” since it is a present duration during which change can occur.



Dolev’s Anti-Metaphysical Realism: A Critique

27

References Broad, C. D. (1921), “Time,” in James Hastings (ed.), Encyclopedia of Religion and Ethics. Edinburgh and New York: T. & T. Clark and Scribners, Vol. 12, pp. 334–45. http://www.ditext.com/broad/time/timeframe.html; reprinted in Oaklander 2008, Vol. I, pp. 143–73; refs. to this repr. —(1923), Scientific Thought. London: Routledge and Kegan Paul Ltd. —(1924), “Critical and Speculative Philosophy,” in J. H. Muirhead (ed.), Contemporary British Philosophy: Personal Statements (First Series). London: G. Allen and Unwin, pp. 77–100. —(1928), “Time and change,” Proceedings of the Aristotelian Society, Supplementary Vol. 8, pp. 175–88. —(1933–8), Examination of McTaggart’s Philosophy (2). Cambridge: Cambridge University Press. Burley, M. (2006), “Beyond A- and B-time,” Philosophia, 34, 411–16. Callender, C. (2012), “Time’s Ontic Voltage,” in A. Bardon (ed.), The Future of the Philosophy of Time. New York: Routledge, pp. 73–98. Castañeda, H. N. (1980), On Philosophical Method. Detroit: NOÛS Publications, 1. Crisp, Thomas (2004a), “On Presentism and Triviality,” in D. Zimmerman (ed.), Oxford Studies in Metaphysics. Oxford: Oxford University Press; reprinted in Magalhães and Oaklander (2010), pp. 109–14. —(2004b), “Reply to Ludlow,” in D. Zimmerman (ed.). Oxford Studies in Metaphysics. Oxford: Oxford University Press, pp. 37–46. —(2005) “Presentism and cross-time relations,” American Philosophical Quarterly, 42, 5–17. Dainton, B. (2011), “Time, Passage and Immediate Experience,” in Craig Callender (ed.), The Oxford Handbook of Philosophy of Time. Oxford: Oxford University Press, pp. 382–419. —(2012), “Time and Temporal Experience,” in A. Bardon, A. (ed.), The Future of the Philosophy of Time. New York: Routledge, pp. 123–48. Deng, N. (2010), “‘Beyond A- and B-time’ reconsidered,” Philosophia, 38, 741–53. Dennett, D. (1991), Consciousness Explained. Boston, MA: Little Brown & Co. Dolev, Y. (2007), Time and Realism: Metaphysical and Antimetaphysical Perspectives, Cambridge, MA: MIT Press. Dorato, M. (2006), “The Irrelevance of the Presentist/Eternalist Debate for the Ontology of Minkowski Spacetime,” in D. Dieks (ed.), Philosophy and Foundations of Physics: The Ontology of Spacetime. New York: Elsevier, pp. 91–107. Falk, A. (2003), “Time Plus the Whoosh and Whiz,” in A. Jokic and Q. Smith (eds), Time, Tense, and Reference. Cambridge, MA: MIT Press, pp. 211–50. Gale, R. (1962), “Tensed statements,” Philosophical Quarterly, 12, 53–9; repr. in Oaklander, 2008, Vol. I, pp. 210–17; refs. to this reprint. Grossmann, R. (1992), The Existence of the World: An Introduction to Ontology. New York: Routledge. Lombard, L. B. (1999), “On the alleged incompatibility of presentism and temporal parts,” Philosophia, 27, 253–60.

28

DEBATES IN THE METAPHYSICS OF TIME

—(2010), “Time for a Change: A Polemic Against the Presentist/eternalism Debate,” in J. K. Campbell, M. O’Rourke and H. Silverstein (eds), Topics in Contemporary Philosophy, Vol. 6: Time and Identity. Cambridge, MA: MIT Press, pp. 49–77. Ludlow, P. (1999), Semantics, Tense, and Time: An Essay in the Metaphysics of Natural Language. Cambridge, MA: MIT Press. —(2004), “Presentism, Triviality and the Varieties of Tensism,” in D. Zimmerman (ed.), Oxford Studies in Metaphysics. Oxford: Oxford University Press, pp. 21–36. Magalhães, E. and Oaklander, L. N. (eds) (2010), Presentism: Essential Readings. Lanham, MD: Rowman and Littlefield. Mellor, D. H. (1998) Real Time II. London: Routledge. Meyer, U. (2005), “The presentist’s dilemma,” Philosophical Studies, 122, 213–25; reprinted in Magalhães and Oaklander (2010), pp. 99–108. Oaklander, L. N. (1993), “On the experience of tenseless time,” Journal of Philosophical Research, 18, 159–66.; repr. in Oaklander, L. Nathan (2004a), The Ontology of Time. Amherst, NY: Prometheus Books, pp. 227–34. —(2001), “Is there a difference between the metaphysics of A- and B-time?,” Journal of Philosophical Research, 26, 23–36; reprinted in Oaklander (2004a), pp. 37–49. —(2002a) “McTaggart’s paradox defended,” Metaphysica: International Journal of Ontology and Metaphysics, 3, 1, 11–25; reprinted in Oaklander (2004a), pp. 51–62. —(2004a), The Ontology of Time, Amherst, NY: Prometheus Books. —(2004b), “Absolute becoming and the myth of passage,” Philo, 7, 1, 36–46. —(2004c), “Craig on the experience of tense,” in Oaklander (2004a), pp. 332–40. —(2012), “A-, B- and R-Theories of Time: A Debate,” in Bardon, The Future of the Philosophy of Time, Adrian Bardon (ed.), New York: Routledge, pp. 1–24. —(2014), “Temporal Realism and the R-theory,” in Guido Bonino, Javier Cumpa and Greg Jesson (eds), Defending Realism: Ontological and Epistemological Investigations. Berlin: Walter de Gruyter. —(ed.) (2008), The Philosophy of Time: Critical Concepts in Philosophy. Four vols. New York: Routledge. Oaklander, L. N. and White, V. A. (2007), “B-time: a reply to Tallant,” Analysis, 67, 4, 332–40. Orilia, F. and Oaklander, L. N. (2013), “Do we really need a new B-theory of time?, Topoi, Special Issue on “Time and Time Experience,” edited by Roberto Ciuni and Giuliano Torrengo. DOI 10.1007/s11245–013–9179–6. Paul, L. (2010), “Temporal experience,” The Journal of Philosophy, CVII, 333–59. Power, S. E. (2012), “The metaphysics of the ‘specious’ present,” Erkenntnis, 77, 121–32. Rassmussen, J. (2012), “Presentists may say goodbye to A-properties,” Analysis, 72, 2, 270–6. Russell, B. (1915), “Our experience of time,” Monist, 25, 212–33. —(1918/64), “The Philosophy of Logical Atomism,” in Robert C. Marsh (ed.), Logic and Knowledge. London: George Allen & Unwin, pp. 177–281. Savitt, S. (2002), “On absolute becoming and the myth of passage,” in Craig Callender (ed.), Time, Reality, and Experience. The Royal Institute of Philosophy Supplement 50. Cambridge: Cambridge University Press, pp. 163–7.



Dolev’s Anti-Metaphysical Realism: A Critique

29

—(2006), “Presentism and Eternalism in Perspective,” in D. Dieks (ed.), Philosophy and Foundations of Physics: The Ontology of Spacetime. New York: Elsevier, pp. 109–25. Schlesinger, G. (1980), Aspects of Time. Indianapolis, IN: Hackett Publishing Company. Sider, T. (2001), Four-Dimensionalism: An Ontology of Persistence and Time. Oxford: Clarendon Press. —(2006), “Quantifiers and temporal ontology,” Mind, 115, 75–97. Smart, J. J. C. (1963), Philosophy and Scientific Realism, London: Routledge & Kegan Paul; partial repr. in Oaklander, 2008, Vol. I, pp. 188–97; refs. to this repr. Tallant, J. (2007), “What is B-time?,” Analysis, 67, 147–56. Tegtmeier, E. (2012), “McTaggart’s error: temporal change,” Revue Romaine de Philosophie, 56, 89–96. Williams, C. (1992), “The phenomenology of B-time,” Southern Journal of Philosophy, 30, 123–37. —(1996), “The metaphysics of A and B-time,” The Philosophical Quarterly, 46, 379–93. —(1998a), “B-time transition,” Philosophical Inquiry, 20, 59–63. —(1998b), “A Bergsonian approach to A- and B-time,” Philosophia, 73, 379–93. —(2003), “Beyond A- and B-time,” Philosophia, 31, 75–91. Williams, D. C. (1951), “The myth of passage,” Journal of Philosophy, 48, 457–72. Zimmerman, D. (2005), “The A-theory of time, The B-theory of time, and ‘Taking tense seriously’,” Dialectica, 59, 401–57.

2 Motion and Passage: The Old B-Theory and Phenomenology1 Yuval Dolev

The old B-theory resurfaces

L

ately a new tone has been heard among B-theorists, marked mostly by being unapologetic. The founding fathers of the theory got it right, goes this recent campaign, and the onslaught they encountered, which resulted in the emergence of the softened “new B-Theory”, was ungrounded. The crux of the criticism against the B-theory was that it clashed with experience, specifically in that it turned tense and passage into illusions. Obviously, once it had been established that reality as it is portrayed by a theory is unlike what we thought it was, the theory finds itself on the defensive. Why should we believe a theory that is not corroborated by experience, or even worse, actually runs against it? And so the next generation of B-theorists went back to the drawing boards and devised an ingenious twist to the plot. Reality is indeed tenseless and, yes, passage is an illusion. But, they claimed, there is nothing mysterious about this illusion or about its existence. It can be accounted for and even shown to be an outcome, a blessed outcome, of evolutionary processes. To act successfully we need to be equipped with A-beliefs, beliefs in which events are located with respect to the present. Reality is tenseless and there are no A-facts, so such beliefs cannot be grounded in A-facts. But that does not mean they are false or groundless. There are B-facts, tenseless relations, which endow A-beliefs and A-utterances with their meaning and truth. With the manifold of A-beliefs in place and solidly anchored to the unchanging, stable ground of B-facts, we can enjoy its uses, position ourselves at the right places at the right times, time our actions so that they

32

DEBATES IN THE METAPHYSICS OF TIME

are efficacious, but without having to carry the metaphysically superfluous baggage of A-facts. The new theory offers, as Mellor puts it (1998: 23), a trade in of tensed facts for tensed beliefs, an exchange which allows us to confine tense to the only place in which it is needed—our heads. Tense is indeed an illusion, but one we live with happily, even thankfully. Note, however, that the new B-theory acknowledges the existence of a substantial gap between how things appear and how they are in reality— they appear to be tensed, but in reality, as conceived by the theory, they are tenseless. And as effective as accounting for this gap may be, it is inevitably accompanied by an apologetic undertone. That is not how things stood when the B-theory was expounded originally, claim contemporary thinkers such as Dieks (2006) and Oaklander (2012)2. The theory’s theses were perhaps surprising but were certainly not viewed as conflicting with experience. Tense and passage were never said to be illusions. Rather, they were shown to be something other than what naïve understanding makes them out to be. Instead of the notion that passage and tense consist of something that passes, some “moving Now”, the B-theory offered an explanation of how they originate with tenseless relations. But something went wrong. An attendant thesis to the original theory’s tenseless ontology was that a tenseless language, one which corresponds to how things really are, can be devised. Soon enough, however, it became clear that a tenseless language is bound to remain an unrealizable fantasy, and the original theory was abandoned for the sake of the new one, which embraces tense as part of language while denying it ontological status. Thus, a theory purporting to harmonize with experience was displaced by a new version in which the absence of such harmony is admitted but shown to be harmless. Now we are witnessing a revival of the original spirit of the B-theory. The new wave consists of renditions of the theory that pride themselves on meshing perfectly with experience. Accusations claiming the existence of gaps between how the theory conceives reality and how reality is experienced are flatly rejected. There is even a re-evaluation of the feasibility of a tenseless language3. This chapter is devoted to Oaklander’s defense of the old B-theory.4 Oaklander calls the theory he defends the B/R theory (after Broad and Russell, who pioneered the view), and I will follow suit. In the B/R theory the world is dynamic, just as it appears to be. Temporal relations are indeed tenseless, but they are not static. Oaklander reverts to capital letters to distinguish the old B-theory’s dynamic TENSEless relations from the static tenseless relations of the new B-theory. Here, too, I will follow him. In the new B-theory we are subject to a constant and inescapable illusion, not to say, error—that of taking the world to be dynamic when in fact it is a frozen block. Thus, the bulk of the effort expended by new B-theorists goes into accounting for how a static tenseless world gives rise to the dynamic



Motion and Passage: The Old B-Theory and Phenomenology 33

experience of flow. The old B-theory, and Oaklander’s B/R theory, will have none of this. The task is not to offer excuses for the theory’s deviation from how we experience temporal relations, but to show that what we actually experience are TENSEless relations. To further explain this, Oaklander invokes what he calls, again following Broad, The Principle of Pickwickian Senses: According to this principle, the proper analysis of a phenomenon or a concept may not be what commonsense implicitly and unknowingly takes it to be, since the implicit analysis, if in fact there is one, may be subject to dialectical difficulties and not hold up to critical examination. The Pickwickian sense of a concept, although perhaps not intuitive, has the advantage that it is quite certain that there is something that answers to it, “whereas with the other definitions of the same entities this cannot be shown to be so” (1924: 93). (Oaklander 2014: 5) Experience is not faulty. But uncritical appeal to it may lead to unwarranted conclusions. Thus, naïve, pre-critical common sense may tacitly acquiesce to the notion that “only the present is real” and mistakenly construe the ontology underpinning our temporal experience as consisting of A-facts, of a “moving Now” that renders ontologically superior the events it visits. It is not that experience introduces us to a kind of ontological hierarchy in which present events are ontologically distinguished from those that are not present. It is that common sense, tacitly or explicitly, gleans such distinctions from experience. But critical examination, which goes beyond common sense, reveals that no such distinctions are to be found in experience. Critical examination does not encounter the properties of presentness, pastness, or futurity in experience or in the events experienced, at any rate, not the tensed properties championed by common sense. Our experience of passage is not an illusion. But it is not what common sense is prone to make of it. One should shun the errors of pre-critical thinking but remain a realist about passage by construing temporal relations as dynamic TENSEless relations. By prompting us to be realists in this way, critical examination earns us “the real truth that underlies the vague facts we start off with” (Oaklander 2014: 7–8). Let me at this point digress for a moment to assess Oaklander’s criticism of my analysis of the tenseless/tensed debate.5 Oaklander levels two principal charges against me: one, that I fail to recognize the Principle of Pickwickian Senses; and second, that I overlook a third metaphysics of time, the old B-theory. I am more than willing to admit to the second charge. My target in Time and Realism as well as other publications was indeed current views, namely, the A- and the new B-theories. I still hold that everything I said about the new

34

DEBATES IN THE METAPHYSICS OF TIME

B-theory and about the nature of the A–new B-theory debate stands, and there is nothing in what Oaklander writes that indicates he would disagree. But it is now emerging that the old B-theory, thought falsified and forgotten, was eulogized prematurely and must be re-engaged. The bulk of what follows is devoted to the old B-theory. My conclusion will be that despite its merits, it too is untenable, and that my former claim that the A–B theory debate must be superseded by a phenomenological inquiry is revalidated. As for the first charge, I must reject it. I have not and do not defend a common-sense view of time, when this is taken to denote some naïve view that remains on the level of pre-critical, vague conceptions. I, too, think that critical examination leads to a better understanding of time, and that, as will be seen, on some issues this new understanding deviates significantly from our naïve, pre-critical thinking. But I disagree with Oaklander that the Pickwickian principle distinguishes “between commonsense and ontology.” Why ontology? Why must that be what the Pickwickian principle contrasts common sense with? The contrast should be with any analysis that promotes our understanding of the concept. And it certainly must not be with any phenomenologically unviable analysis, as the one suggested, I will argue, by the old B-theory. Thus, to reject the old B-theory is not to reject the Pickwickian principle, but to favor a different analysis of time as the source for the clarity that is absent from the common-sense apprehension of it. As I hope to show, far from indicating a failure to appreciate the significance of the Pickwickian principle, I think the alternative I propose is a good example of how to implement it. Before proceeding, let me note that Oaklander’s specific objections are all offshoots of one of the above two main criticisms. To give an example, when Oaklander says that “to assume at the outset that the tensed and the tenseless views agree on the existence of ‘tenseless relations’ and only disagree about the existence of tensed relations is to misunderstand the metaphysical dispute in the philosophy of time” (p. 16), the reason he gives is that the B/R theory’s TENSEless relations cannot be part of such an agreement. In other words, the agreement I pointed out indeed exists, but only between the tensed and the tenseless theory (the new B-theory), and not with the TENSEless theory. Oaklander’s objection, then, is that my analysis overlooked TENSEless relations, a fault, it must be said, it shares with new B-theorists, who are thus also guilty of misunderstanding the metaphysical dispute in the philosophy of time. In what follows I respond to these kinds of charges by responding to Oaklander’s two chief criticisms. Oaklander claims that on my construal of the tenseless view, the relations that supposedly obtain between events are not even temporal. Events turn out to constitute a C-series, not a B-series. Oaklander’s critical remarks about Laurie Paul’s position suggest that he indeed believes that the new B-theory



Motion and Passage: The Old B-Theory and Phenomenology 35

can hardly be called a theory of time because the static relations it takes to obtain between events are not temporal. By contrast, the axes around which the B-theory revolves are dynamic TENSEless relations. My analysis, he states, begs the question: it targets tenseless relations that are not even temporal and then proceeds to criticize the tenseless view for not succeeding to capture the essence of time. Needless to say, Oaklander continues, this maneuver cannot be effective against the old B-theory. Dynamic TENSEless relations grant us, as part of reality and not merely of how we experience things, all the temporality we are familiar with from experience. But, of course, whether or not the old B-theory delivers the goods is the key question here. We need a theory of time that is phenomenologically viable, one which does not harbor gaps between the way things appear to us and the way they really are. So the question is whether the original B-theory, in contrast with its successor (and, as I argued at the time, with the A-theory), offers an account of time which harmonizes our experience with reality. If it does, then this theory is very attractive indeed, and it would be difficult, and superfluous, to come up with a reason to reject it.

The old B-theory, the specious present and phenomenology But can a theory that denies that some events have the objective property of being present cohere with experience? I will argue that the answer is “No.” The first thing to note when we turn to assess how the old B-theory fairs phenomenologically is that, in contrast with the new theory, the old theory does not flat-out deny that some events are objectively present. Its relations are TENSEless, not tenseless, they are dynamic, not static. Specifically, some events occur simultaneously with one’s experience of them, the simultaneity in question being a TENSEless relation. And here is the crucial point: according to this theory there is nothing more, phenomenologically speaking, to experiencing the presentness of an event than experiencing its TENSEless simultaneity with the experience. Likewise, to undergo a series of experiences that TENSElessly succeed each other just is to experience passage. But, it turns out that in order to give the theory phenomenological viability, another element is required: the so-called “specious present.” The history of the specious present, from its introduction by E. R. Clay and its appropriation by James, to its current revival, for example in the work of Dainton, has been relayed often enough. Oaklander’s invocation of the specious present follows Russell and Broad. For them, the specious present is the arena for

36

DEBATES IN THE METAPHYSICS OF TIME

experiencing time’s dynamism. It is in the specious present that, for example, a chain of musical notes is apprehended as a flowing unity which constitutes a tune. But why is a specious present required for the experience of precedence, for the experience on which “all further knowledge of time is built,” to use one of Broad’s formulations which Oaklander rehearses approvingly (p. 9)? Why can’t two notes be experienced as TENSElessly following each other without the specious present? Supposedly, because to experience them separately is not to experience them related, not even temporally. To experience B temporally succeeding A is more than just experiencing A and experiencing B; it is to experience the temporal relationship between them. But to experience them related, A and B have to somehow be experienced together. An account of how this is possible consists of two stages: first it needs to be stated what unifies the elements of the succession; then it needs to be explained how whatever unifies them executes the unification. Concerning the first stage, one option is that it is by being co-present in consciousness that A and B are experienced together, for example as constituting a tune.6 Thus, by adopting the specious present, the B/R theory addresses the first part of the challenge posed by the experience of succession, motion, and change, and what’s left is to explain how a present fulfills the task of unifying temporally distant events. Moreover, it may initially appear as though the B/R theory embarks upon a phenomenologically more promising track than its new successor, for it delivers an account of change that is dynamic and in which the present and transience play a vital role. The new B-theory, in contrast, is forced to incorporate change into a tenseless world that is static, a mission that verges on the oxymoronic. But the truth is that the specious present fails precisely on the point which it is supposed to bolster, namely, phenomenology. The reason, in a nutshell, is that there is nothing specious about the present—or, at least, nothing in or about experience attaches so much as a sliver of speciousness to the present. Before elaborating this claim, let me make two notes. The first is that once the specious present becomes part of the old B-theory it can no longer be maintained that TENSEless relations account for the experience of passage—they do so only when augmented by the specious present, which otherwise would not immediately appear as a natural component of a B-theory of time. A truly tenseless theory should not be tainted by a present, “real” or “specious.” The second note concerns the aforementioned question as to how the elements of a succession are unified. It turns out that in most of its versions, the specious present is a contraption designed to perform magic, namely, conjugate what are unconjugable, distant events, for example, blending together the distinct notes of the sequence A–B–C–D into a tune. The specious present was invented so that there is something that will tolerate



Motion and Passage: The Old B-Theory and Phenomenology 37

what actual motion and change do not, namely, the coexistence of phases of the succession. The notes A–B–C–D are temporally distant and so cannot coexist in reality, but they can be co-conscious in the specious present, where they constitute a melody rather than a sequence of independent notes. How does this magic happen? How does the temporal gap keeping these notes apart in reality vanish when they arrive at the theater in which they are heard—at the specious present? There are various models that purport to explain this. Here is a brief sketch of two. In what Dainton calls the “awareness-overlap” model, to which Broad subscribed at some time, there is a distinction between phenomenal contents and the acts of awareness in which these contents are apprehended. To hear A–B–C–D is to undergo a series of acts of awareness, one including A–B, another including B–C, and so on. On the level of content, the notes are separated, but they are brought together in the act of awareness. How? Nothing is said. The model is rejected for other structural difficulties, such as that the same content is apprehended by more than one act of awareness—B figures both in A–B and in B–C. But the more troubling problem is that to simply state that the temporally distant contents A and B become co-conscious in A–B is to sweep under the rug rather than solve the difficulty for which the specious present was introduced to begin with. In the two-dimensional model, to which Broad turned later, the same experiential contents linger throughout the succession of acts of awareness, but with a decreasing degree of, what Broad called, “presentedness.” Thus, the content A is not temporally confined to a point but rather spreads in time far enough so as to overlap with B, albeit with diminished “presentedness.” Temporally distant contents such as A and B are conjoined by overlapping. There are many structural problems with this model as well (cf. Dainton 2010: 111). Here I would just like to remark that there is no phenomenological evidence that such spreading of contents takes place, and no sign of the existence of the property of “presentedness” with its varying degrees of intensity. My point is that when we look inside the specious present and inspect the mechanism which is supposed to explain how succession is perceived and how the experience of flow and passage is created, we find more questions than answers. But this state of affairs is tolerated precisely because this present is specious. Speaking of the two-dimensional theory, Dainton notes that “since the posited additional dimension is located within consciousness rather than the world itself, it is not open to the objection that we have no reason to believe that such a thing exists” (p. 110, emphasis added). In other words, because it is specious this present can allow for things that “the world itself” cannot. For example, in it unjoinable contents can be joined. The perception of change and motion seems to confront us with an

38

DEBATES IN THE METAPHYSICS OF TIME

antinomy: it can be achieved only if temporally distant contents are apprehended together. The solution is to create a specious zone that is not subject to the conceptual and logical constraints our descriptions of the world itself must respect, and let perception of change take place there. Seen in this light, as the only solution of an antinomy, we can almost say that the present must be specious. Unless, that is, there is another way to account for the perception of motion. If there is one, and shortly I will argue that there is, then the specious present had better be abandoned, at least by those guided by the conviction that when doing metaphysics, phenomenology must constantly be consulted. As was just illustrated, the route leading to the specious present is not phenomenological but conceptual, or logical. First, one determines that there is no such thing as presentness in “the world itself,” for example on account of McTaggart’s argument, or of special relativity. Or one surmises that the “real present” does not cohere with experience, for example because the real present is a volumeless point whereas the experienced present is extended. Then, one concludes that there really is no choice but to posit an “experienced present” which is specious. In other words, the speciousness of the present is the product of a derivation, and not of any reflection on the temporal aspects of experience. Indeed, it is acknowledged despite everything that can be gleaned from a phenomenological study of experience. Positing entities on the basis of conceptual analysis is a legitimate mode of reasoning, of course, which emulates something we are familiar with from scientific investigation. “Specious” entities often become central to a theory. Electromagnetic potentials7 are a famous example. The neutrino started its career as a somewhat specious particle, the existence of which was contested by leading physicists such as Bohr. The devastating difference between the specious electromagnetic potentials and the specious present is that we have no expectation to encounter the electromagnetic potentials in experience, whereas the specious present not only figures centrally in a theory that is supposed to be in as much harmony with experience as possible, but is in fact introduced into the theory precisely for the sake of rendering it harmonious with experience. Unlike its new successor, the old B-theory will not castigate flow or tense, or any component of our experience, as illusory. But then comes the branding of the present as “specious.” True, calling the present specious is not as bad as calling it an illusion, but it is in the same spirit. We don’t speak of specious successions, or volumes or shapes or sizes, though in all these cases, and in many more, arguments regarding the speciousness of these things can be, and have been, put forth.8 In general, where possible, we refuse to admit a distinction between how things are experienced and how they are “in the



Motion and Passage: The Old B-Theory and Phenomenology 39

world.” But such a distinction is precisely what is implied by tagging the present as specious. For a theory that purports to be phenomenologically attuned and attentive, this is a troubling issue. The old B-theory is seeing a renaissance because thinkers such as Oaklander are not satisfied with the gap that exists between the new B-theory’s description of reality and how reality is experienced, a gap expressed by the assertion that tense and passage are a myth, or an illusion. A theory which allows such a gap can hardly be a theory of time: “Tenseless” becomes “Timeless” in such a theory. My claim is, however, that in the end, because it relies on a phenomenologically alien, artificial construct—the specious present—the B/R theory suffers from the same weakness. Can the distinction between common sense and the Pickwickian sense of a concept be of help here? I think not. The Pickwickian understanding of a concept is not supposed to clash with experience, but to clarify it. On the contrary, it is the common-sense conception that is supposed to generate uneasiness, that is supposed to discord with experience, in a manner that will trigger a “critical examination” and inspire a more refined, more accurate and phenomenologically more satisfactory Pickwickian understanding of the concept. But in our case, a new and no less disturbing uneasiness surfaces when the common-sense conception of tense is replaced by a theory that is structured around a specious present. To summarize, Oaklander charges me with directing my criticism at a version of the B-theory, the new B-theory, which, by his lights, is indefensible. In fact, what Oaklander says about the new theory is more lethal than anything that emanates from my criticism, for according to him the new theory cannot count as a theory of time at all. But that is why, according to Oaklander, my criticism of the B-theory fails: it begs the question by targeting tenseless relations—those of the new theory—that are damned and untenable from the outset. Tenseless relations must be understood in the way they were conceived originally, in the B/R theory, as dynamic, and as consisting of everything temporal we find in experience. However, as I tried to establish, this view of the old theory cannot be sustained once the phenomenological oddity of the present’s supposed speciousness is exposed.

Natural realism about tense Still, the “TENSEless relations plus specious present” theory may be the best account of time and of experience we have. Indeed, it is, I believe, superior to any other version of either the A or B theories. I wish, however, to outline the alternative I have been advocating, and to argue that it reflects experience better than all the theories making up the A–B debate. To introduce it, let me

40

DEBATES IN THE METAPHYSICS OF TIME

briefly discuss a version of the specious present doctrine that has not been mentioned so far—Dainton’s overlap theory.9 The important advantage of this theory is its simplicity. It does not get entangled with repeated contents that are experienced many times over, or with contents that linger and form a second temporal dimension, or with other architecturally contrived complexes. It even manages to free itself from the cumbersome distinction between phenomenal contents and acts of awareness that apprehend these contents. The key to the model’s simplicity is the realization that the apprehended event and the event which is the apprehending are both extended in time and run concurrently. In this theory, acts of awareness are themselves spread in time, and they partially overlap. Thus, seeing a light fade consists of many extended experiences, each overlapping with those immediately before and after it. The problem of repeated contents that afflicts the “awarenessoverlap” model vanishes because the acts of awareness which form the seeing of a light fade overlap, with each phase of the light’s fading figuring in many of them, rather than being replicated many times over so as to be part of each of them. And there is no need for a second dimension, for example of diminishing presentedness, or of retentions, because each act of awareness is spread over time and so covers many of the phases of the light’s fading. The alternative I defend has a similar structure. In particular, it shares the idea that the event and the experience thereof run concurrently. But it differs on two basic, and related, issues: on my proposal the present is not specious, and, just as there is one event being apprehended, there is just one act of apprehension and not many. Let me explain. Dainton’s theory, like other specious present theories, is engaged with explaining how successive, that is, temporally separated, contents that cannot be brought together nevertheless figure together in consciousness. Their so-called solution for this enigma is the specious present, which sustains extended and overlapping acts of awareness, a present which, by virtue of its own unity, bestows unity upon these acts of awareness. The internal structure of the overlap model is simpler and sturdier than that of other versions of the specious present, but its function is the same—to constitute a zone which, by being specious, allows what “the world itself” doesn’t, namely, the coexistence of things that in reality do not coexist. Note that here too the argument for speciousness is logical or conceptual, rather than phenomenological; it is not that studying experience leads to the present’s speciousness in a positive way, but that the non-speciousness of the present is ruled out conceptually. But perhaps there is something wrong with the way the challenge is set up, namely, with the notion that the task facing our analysis of the experience of motion is that of uniting elements of a succession, or, to revert to this term once again, of undoing an antinomy? Because Dainton, too, becomes knotted with



Motion and Passage: The Old B-Theory and Phenomenology 41

the antonymous nature of united successions, he, too, resorts to the present’s speciousness. If, however, the antinomy is removed, the impetus to conceive of the present as specious is also eliminated. To be more exact, “remove” is not what needs to be done with the antinomy, for, as I will suggest, the antinomy was never there to begin with. Setting up the conceptual challenge as that of negotiating successions is, I wish to propose now, already phenomenologically misguided. The experience of motion does not consist of a perception of succession nor is it derived from a succession of perceptions. Succession plays no role in the perception of motion, or, for that matter, in motion itself. Seeing the cat cross the lawn does not consist of having a succession of perceptions, nor of perceiving succession. The experience simply does not break down to components that have to then be reunited. Hard as we try, when we scrutinize our experience we seek in vain for past bits that somehow coalesce with present ones to form the perception of motion. Here is what we do find. The first thing we find when focusing attention to experience are events that are experienced as present, by which I mean that we find tensed properties that belong to the events we experience, and not merely to how these events are experienced. Then, if we attend to them, we find that our experiences of these events are themselves present events.10 Of course, so as not to beg the question, I am not at this point assuming anything about this present, for example that it is real rather than specious. The second thing we note is that a part of an event is a separate event. But the part–whole relationship is not one thing but many. Rivers have parts, as do cars, sentences, vector spaces, communities, and laws. Sometimes things are dividable into their parts, sometimes they are not. Let us return for a moment to Frege’s contextuality principle. A sentence, even though it is composed of words, is the basic unit of meaning. And the meaning of a sentence is not the aggregate of the meanings of its words. The individual words in isolation do not have meanings. But this principle is easily expanded. Take the conjunction “Inflation is rising despite the lowering of interest.” Formally, this is a conjunction. But “despite” does not mean “and,” and the meaning of the sentence is not captured by, or reducible to, the meaning of the conjuncts making it up. In the proper context, this sentence has a unity that is lost when it is broken down into its components, which are also sentences. The same can be said of paragraphs and essays. The musical analogies are ready at hand—a melody is not a collection of notes, and so on. Events are made up of parts which are also events, not in the way that a car is made up of parts but rather in the way that paragraphs are made up of sentences, or symphonies are made up of parts, and parts of symphonies from phrases. Moving from an event to its part is a matter of changing contexts and not of taking apart.

42

DEBATES IN THE METAPHYSICS OF TIME

This observation carries with it a consequence that turns out to be far reaching for our inquiry, namely, that present events have parts that are not themselves present. Thus, there is no contradiction between saying that the orchestra is now playing Beethoven’s 3rd Symphony and saying that the third part of the symphony will only start in ten minutes. The first statement is about one event and the second is about a different event. That the second event is part of the first is immaterial, for the temporal attributes of these events, at least those relevant to the current discussion, do not pass down from the mother event to the part or from the part up to the mother event. This deserves further clarification. The thesis is not that there is a context in which the playing of the symphony’s third part is a present event and another context in which it is a future event. Rather, in all contexts in which the playing of the symphony’s third part figures at all it invariably appears, and is referred to, as a future event. There are no contexts in which it figures as a present event, only contexts in which it does not figure at all, as when it is proclaimed that the concert is being broadcast on the radio or that it is being conducted by so and so. That the playing of the symphony’s third part will be broadcast on the radio and conducted by so and so are separate statements that pertain to a separate event, a future event—the playing of the symphony’s third part. Of course, with time’s passage contexts change, and half an hour from now the orchestra will be playing the symphony’s third part, which will then be a present event. But at the moment there is no context in which that event, the playing of the third part, can be correctly spoken of as present. This is as simple as it is absolute: just as we cannot now hear what will be heard 30 minutes hence, so we cannot now say or think what will be said or thought 30 minutes hence (there is a familiar analogy that is useful in explaining this: I understand perfectly well what John is saying when he proclaims “I am hungry,” but I can never say what he says with this sentence). The notion that the playing of the symphony’s third part can in any way be thought of or spoken of as a present event is simply a mistake. Let us add another, related point, namely, that the individuation of events is a matter of context. Reality does not break down into events; rather, different contexts capture different events. That does not mean that “we make the world,” that cutting reality up into events is arbitrary, or a matter of choice, or anything of the like. Recognizing the inevitable, inescapable, irremovable role of contexts is not tantamount to relinquishing realism, or to rendering events “mind-dependent” in any way. Contextualism of this kind is not an extravaganza anymore than an oven is in the case of baking—it is a prerequisite. And let us recognize that different contexts pick out different present events, and that some present events are parts of other present events. “The symphony is being broadcast live on the radio, but only its second part, which is being



Motion and Passage: The Old B-Theory and Phenomenology 43

played now, is broadcast live on TV”—the first part of this sentence refers to one present event, the second part to another present event, which is a part of the first present event. Now, everything said thus far about events applies to those events that are our experiences. There is a context in which we are hearing the orchestra play Beethoven’s 3rd Symphony, and another in which we are hearing it play the symphony’s second part. In both, the hearing of the symphony’s third part does not figure at all. In both contexts it is neither past nor is it not past. It may be tempting to say of the first context that in it we are hearing the orchestra playing Beethoven’s 3rd Symphony in its entirety. But that would be dangerously misleading, for it could be taken as tacitly referring to the symphony’s parts and as implying that we are hearing the orchestra playing all the parts of the symphony, (“together,” we might add in accordance with the aforementioned “successionist” theories). We avoid this mistake by rehearsing that what we experience are present events, period, not in their entirety, and not not in their entirety. Turning attention towards a part of an event means shifting contexts and focusing on a different event, which may or may not be present. Again, that we are not yet hearing the playing of the symphony’s third part does not in any way clash with the fact that we are now listening to the symphony—not in its entirety and not to a part of it: the part–whole relationship is simply not applicable here. The upshot of all this is that phenomenology does not point to successions playing any role in the experience of change or motion. In particular, such experiences do not consist of, and are not underpinned by, successive bits of experiential inputs somehow being integrated so that they can figure together. Earlier bits do not need to linger, either by having their presentedness gradually diminish, as in the theory held by the later Broad, or in the form of retentions, as in Husserl’s theory, or of memory traces, as in the cinematic model held by several new B-theorists. What is true of those events that are our experiences is also true of what we experience. Events “in the world” as well as the acts of consciousness by which these events are apprehended are extended in time. Events indeed have parts, but they are not made up of their parts in the way that a car is made up of parts. Rather, they are distinguished from their parts by contexts. We can focus on an event, or on one of its parts, but we do not need to bring parts together so as to form the event. It is not that events enjoy a unity that ties together their parts. It is that the question of how an event’s parts combine to form it cannot be coherently posed. So my view shares with Dainton’s overlap model the contention that events, in particular those that are our acts of awareness, are extended in time, and that the events we experience and our experience of them run concurrently. But I hold the overlap model to be motivated by the phenomenologically and

44

DEBATES IN THE METAPHYSICS OF TIME

conceptually ungrounded presupposition that events consist of parts that are somehow brought together to form a unity. If the present does not have to constitute an extraterritorial safe-zone where the impossible construction of successive parts that are hopelessly separated into an indivisible unity occurs, then it no longer has to be shrouded in the shadows of speciousness. Which brings us to our final, and main, topic—the reality of tense. After concluding that there is no need to demote the present to speciousness, what can we say about its reality? Let me start by stating again in what sense it is not real. The reality of the present does not consist of any kind of ontological supremacy that present events enjoy over events that are not present. Tense does not consist of ontological hierarchies. Thus, the manner in which the various versions of the A-theory analyze tense cannot be accepted.11 But nor is it the case that all events are “equally real,” or “exist” in some tenseless sense. These locutions, too, place events in an ontological hierarchy, only one in which they all occupy the same stratum. Tense cannot be analyzed in terms of reality claims of any kind. In the approach I have been advocating, phenomenology should come before metaphysics. We have to understand tense through its phenomenological manifestations, by which I mean the manners in which it exhibits itself emotionally and cognitively. For example, an analysis and description of the anxiety before a bungee jump (an anxiety of a very peculiar flavor, which is felt, notice, only with regard to future experiences of a certain kind) can serve as a component in our understanding of what the future is. The experience of deliberating, for example where to go on vacation, provides input of a different character to our conception of the future. It is not that we have a theory telling us what the future is and then go on to adjust our emotive dispositions and cognitive practices to it; rather, we learn what the future is through our emotive dispositions and cognitive practices. The same goes for the present and the past, and also for the differences between them. A study of how our emotional and cognitive dispositions towards an event or an experience alter as the event shifts from being future to being present, and alter again when it becomes past, contributes crucial ingredients to our understanding of the nature of the past, present, and future. Let us say more about the present. To be present is to be something that can be experienced by someone properly situated—I will speak of being experientiable. This calls for many clarifications. Does being experienced by a cat, or an ameba, count? Does being experienced indirectly, for example, by means of a particle accelerator, count? What about very distant things that are happening now but cannot be experienced? And what are the kinds of things that can be experienced—objects, such as apples? Properties, such as sweetness, hardness, greenness? Events and states of affairs? Also, experientiability applies to all events, at the moment of their occurrence. So



Motion and Passage: The Old B-Theory and Phenomenology 45

how are events of one moment distinguished from those of other moments? Doesn’t our characterization reduce tense to a vacuous tautology, namely, that everything is present when it happens? It is not possible within the span of this chapter to delve into these issues. While they are important and undoubtedly difficult, more pertinent to our topic is the following question: “How can tying presentness to experienciability be the basis for a conception of tense according to which tense is an objective property of events rather than a feature of how we apprehend them?” This proposal seems to make presentness explicitly dependent upon human experience. But then, how can it presumably be a form of realism? Would it still be the case under this view that in a lifeless universe some events would be present while others would not? The answer is “yes.” Let us distinguish between the assertion that the qualitative attributes of our experiences play a role in the formation of the means with which we conceive tense, and the claim that our experience is constitutive of tense. Only in this latter claim does the explicit connection between presentness and experientiability render tense experience-dependent. My contention is, rather, that while tense is not some kind of private, subjective “quale,” presentness cannot be grasped or analyzed in complete detachment from how it is experienced. The experiencer cannot be removed from how she conceptualizes tense, not because she “makes” it, à la Goodman, but because she is an irremovable component of the interaction that generates the terms with which the conceptualization is executed. Certain elements of the fundamental structure of reality, and specifically tense, while utterly not dependent on any observer, are nevertheless given only in terms of how they would influence a properly situated observer.12 In a lifeless universe no one would be experiencing the motion or the whiteness of the snow falling on Mt. Washington, and there would be no one for whom the snow would be falling now. Still, this event would be present, just as the snow would still be white and in motion—a suitably situated observer would experience the falling of the snow as present just as he would experience the snow as white. To further explain this, allow me to invoke an analogy I first made in Time and Realism and have been reverting to since. Notice first that being present is different from being, for example, green or sweet or audible, in that the tense property is shared by the thing being experienced and by the (perhaps hypothetical) experience. When I see the cat cross the lawn, the cat’s crossing of the lawn and the act of awareness consisting of my seeing it cross the lawn are co-present, but my seeing the grass green does not involve any greenness of my experience, and there is nothing sweet about my experience of eating ice cream. The thesis I presented at the time was that the tensed properties of our experiences figure as standards for tense terms, analogously to the way that, say, the standard meter rod is a standard

46

DEBATES IN THE METAPHYSICS OF TIME

for the meter. The presentness of my present experience of seeing the cat cross the lawn serves me in much the same way that the meterhood of the standard meter serves me. The former is my handle on tense, the latter on distances. In a lifeless universe there would not be a standard meter. The making of standards is a human act, without which there is no talk or thought of distances. But the distance from the earth to the moon would still be 300,000 kilometers. Here, too, a feature of the fundamental structure of reality is given in terms that are inextricably entangled with human experience and action. But once the term is there, we can use it to discuss reality as it would be if no humans were part of it. In the same fashion, it can be asserted that in a lifeless universe the snow over Mt. Washington would be falling now. Much work has been done to explain why acknowledging the role experience has in fleshing out the objective properties of objects and events does not entail an abandonment of realism. One argument in defense of this takes bivalence13 to be the hallmark of realism. Thus, when inquiring, for example, whether the dispositional theory of colors deserves to count as a form of realism, the test is to check whether on this theory bivalence applies to sentences reporting the colors of objects.14 Likewise, in my analysis, bivalence applies to tensed descriptions of events and so it constitutes a form of realism about tense. Together with the above comments concerning the applicability of tense to reports about events in a lifeless universe, this observation should help remove any residue of “subjectiveness” that may still be clinging to the analysis of presentness in terms of experienciablity. Before concluding, let me return to Oaklander’s criticisms again. Oaklander accuses me of not being “sufficiently sensitive to the more fundamental ontological differences between disputants in the metaphysics of time.” This charge is prompted by the claim I made at the time that the A/B theories share what I called “the ontological assumption,” which, roughly, states that time and tense must be fleshed out in terms of reality claims and ontological hierarchies. As already indicated above, I argued that this assumption must be transcended, that tense cannot be explained in terms of reality claims, and that an alternative approach is required. Oaklander contends that the real dispute is not about the ontological assumption and its attendant ontological hierarchies, but about the ontological status of temporal relations, a contention derived from the reintroduction into the discussion of the old B-theory. I don’t think this shift of focus proves any kind of insensitivity on my part to the fundamental issues in the metaphysics of time. Moreover, I don’t think that reviving the old B-theory renders the questions my analysis concentrated on irrelevant. On the contrary, because it is presented as an ontological theory, in assessing it, it is crucial to ascertain whether it, like its successor the new B-theory, is committed to the thesis that all events are on an ontological par.



Motion and Passage: The Old B-Theory and Phenomenology 47

If it is, then it is subject to the criticism I have put forth against this thesis. If it is not, then I am not sure why it counts as a tenseless theory at all. In such a case it seems to me that the TENSEless theory no longer needs the crutches of a specious present and can merge with the view I have been defending. Oaklander also attributes to me the claim that “there is no genuine dispute in the metaphysics of time.” I should stress that I have always maintained that, to the contrary, the debates in the metaphysics of time are genuine not only in the sense that they are prompted by real bewilderment, but also because they constitute invaluable and unavoidable steps along the quest for clarity about time. My claim was and still is that they are not the final destination of this voyage, only stages, albeit crucial stages, along the way.

Conclusion The view I have been defending has much in common with the old B-theory (and very little with the new one, or with any of the versions of the A-theory). In both, phenomenology is of crucial importance. In both, presentness does not consist of some monadic relation, such as “being real.”15 Both reject “static” pictures of time, such as the block universe picture.16 In both, finally, the present has a vital experiential role. Oaklander would probably even agree that tense and passage are objective, provided that they are construed as features of TENSEless relations, which, in his onotolgy, are the only thing they, or, indeed, any temporal attributes, can be. And the agreement goes in the other direction as well: I could accept that tense and passage are what TENSEless theorists say they are, as long as they amount to what we know from experience. So where is the disagreement? The significant difference is that in the old B-theory the present is specious while I maintain that being present is as much a property of events as, say, their duration or special location. I think it is uncontroversial that the presentness of an event (or its pastness or futurity) strikes our pre-reflective sensibility as an objective feature of the event itself. The disagreement is about how tense should be conceived by our mature, elaborate and nuanced understanding of time. I tried to show that the conclusion that the present is specious is arrived at through conceptual arguments, and not via a phenomenological study. If these arguments were conclusive, and in the absence of an alternative, it would have been appropriate to invoke the Pickwickian principle and accept the present’s speciousness as a profound product of a meticulous study of time. But, as I have tried to show, a phenomenologically based understanding of tense, which strives to remain faithful to the fact that tense and passage experientially appear as features of reality and so is reluctant to relegate

48

DEBATES IN THE METAPHYSICS OF TIME

them to the realm of the mental or the spurious or the specious, is in fact available. If my conception of tense is viable, then there is little justification for preferring a theory which introduces into our understanding of time phenomenologically extraneous elements and whose depiction of reality does not square comfortably with experience. The alternative I propose does not signify a return to our naïve conception of tense and passage. To repeat, the notion that to be present is to be real, or to exist in some special manner, has no place in it. But our understanding that tense is objective, and belongs to events rather than merely to how we experience them, is maintained in it, as I believe it ought to be in any Pickwickian understanding of time we arrive at.

Notes   1 I wish to thank Nathan Oaklander, whose classical work on time has enlightened the writings of so many, and conversations with whom have always been inspirational, in more than one way.   2 Dieks and Oaklander, it should be noted, are engaged in very different projects. Yet they do share the insistence that the B-theory is faithful to experience.   3 Cf. “Do we really need a new B-theory of time?”, 2014.   4 I discuss Dieks’s position in “Relativity, Global Tense and Phenomenology”, in Cosmological and Psychological Time (forthcoming, Springer). Also, with the exception of a few cursory remarks, this chapter does not discuss the arguments for the B-theory, such as McTaggart’s or the argument from relativity, or the arguments for the A-theory.   5 As expounded in my Time and Realism.   6 The cinematic, retentional, and extensional theories propose a host of very different manners of fleshing this out.   7 Prior to their materialization as physical entities following the Bohm-Aharonov effect.   8 Sometimes in a skeptical vein, but also in the course of phenomenological studies, as in the work of Husserl, Merlau Ponty, and others, who, of course, were not claiming that objects don’t have real size and shapes, but were using such claims as part of their investigations.   9 Not to be confused with the “awareness-overlap” model discussed above. 10 I am ignoring the cases in which the event experienced and the experiencing are spatially, and therefore temporally, separated by significant spacio-temporal intervals. 11 Time and Realism, esp. ch. 3, presents a host of arguments in defense of this claim. 12 I, for one, tend to hold that this is true of all aspects and features of reality.



Motion and Passage: The Old B-Theory and Phenomenology 49 What it is like to see the color of objects, but also their size and shape, or to experience the duration of events and their spatial properties, figures in the inception of the terms with which we think and speak about these things.

13 While Putnam and Dummett represent the two opposing sides of the realism–antirealism debate, they agree that fleshing out the positions in terms of mind-dependence/independence will not do and that a semantic characterization of both positions is indispensible. 14 Setting aside a host of other but unrelated issues, such as vagueness. 15 Though I am not sure that framing the analysis of tense in terms of monadic versus relational properties is helpful. 16 Though, again, I think it is misleading to subject temporality to the static– dynamic dichotomy. Events, just like time itself, are neither dynamic nor static. A bus moves, and a picture on a wall may be described as static, but not events, or states of affairs, or moments of time. Future events become present and present events past: that is what time’s passage comes down to, but there is nothing dynamic (nor static) about time’s passage or about events.

References Dainton, B. (2001; 2nd edn 2010), Time and Space. Durham: Acumen. Dieks, D. (2006), “Becoming, relativity and locality”, in The Ontology of Space Time. Boston: Elsevier, 157–76. Dolev, Y. (2007), Time and Realism. Cambridge, MA: MIT Press. Mellor, D. H. (1998), Real Time II. London: Routledge. Oaklander, L. N. (2012), “A-, B- and R-Theories of Time: A Debate,” in Adrian Bardon (ed.), The Future of the Philosophy of Time. New York: Routledge, 1–24. —(2014), “Dolev’s Anti-Metaphysical Realism: A Critique,” in L. Nathan Oaklander (ed.), Debates in the Metaphysics of Time. London: Bloomsbury. Orilia, F. and Oaklander, L. N. (2013), “Do we really need a new B-theory of time?,” Topoi, Special Issue on “Time and Time Experience,” edited by Roberto Ciuni and Giuliano Torrengo. DOI 10.1007/s11245–013–9179–6.

3 Two Metaphysical Perspectiveson the Duration of the Present Francesco Orilia

Introduction

I

n his intriguing book Time and Realism (2007), Dolev criticizes some A- and B-theories discussed in current analytic metaphysics of time and proposes his own stance on temporal matters, whose basic tenets we also find in his rejoinder (2014) to Oaklander (2014) in this volume.1 This stance benefits in Dolev’s opinion from giving up the “ontological assumption” that undermines the analytic debate and thus Dolev thinks that he is engaged in phenomenology rather than ontology. To be sure, he urges us to pay more attention than has hitherto been done to data and approaches provided by so-called phenomenologists such as Husserl, Heidegger, and Levinas (p.  213), but nevertheless there are good reasons to consider Dolev’s positive view as an ontological or metaphysical theory of the kind one already finds in the analytic marketplace (for present purposes, we need not distinguish between ontology and metaphysics). In order to substantiate this claim, I shall show how ontological commitments of the eternalist variety can be extracted from Dolev’s anti-Augustinian account of the duration of the present. Dolev’s insistence on phenomenological data is commendable, but among such data we find our deep-seated feelings that the past is gone and the future is open, which are in tension with eternalism. Presentism is best fit to accommodate such feelings and Dolev’s views on the present can be resisted in favor of the Augustinian line that sees it as durationless, thereby avoiding eternalism

52

DEBATES IN THE METAPHYSICS OF TIME

and making room for presentism. Yet, a durationless present has a number of problems in store for presentism, at least as traditionally understood. I shall thus end up proposing a moderate presentism, which makes some concession to eternalism and yet preserves the gone past and the open future.

Some terminological clarifications Dolev claims (p. 3) that his work points to “resolutions for some of the central issues that traditionally make up the metaphysics of time,” however, … not by offering a metaphysical theory in which answers are given and explanations expounded, but rather, in the tradition of philosophers such as James and Wittgenstein, by working through these issues to the point at which the intelligibility of the theories generated in response to them begins to falter. The theories in question are identified by Dolev as the tensed or presentist and the tenseless or eternalist views (p. viii n.2). Dolev has in mind a big subdivision into two broad perspectives, which are often labeled as “the A-theory” and “the B-theory,” roughly because in the former the so-called A-properties, past, present, and future, have an essential and irreducible role, whereas in the latter B-relations, such as earlier and simultaneous, come to the fore. To use the adjectives “tensed” and “tenseless” to qualify the A-theory and the B-theory, respectively, is also appropriate, because, roughly, in the former, but not in the latter, tenses are accorded a crucial irreducible role in our effort to express what time is all about. The labels “presentist” and “eternalist,” as used by Dolev, may, however, be misleading, for, according to a widespread use that they have, (i) there can be an eternalist of the A-theoretical variety, and (ii) “presentism” may refer to a doctrine that not all A-theorists endorse (Zimmerman 2005). In the use of “eternalism” that I have in mind, this word refers to a position according to which the ontological inventory, all that there is or exists, includes past and future events or states of affairs2 just as it includes the present ones; it includes, for example, the present event of my writing these words, the birth of Napoleon and (let us suppose) the birth of the first child of 2020.3 The B-theorist can hardly avoid being an eternalist in this sense, for in her view the presentness of an event is a merely perspectival matter, pretty much like the being here of a place in my vicinity. An event is present in this weak sense, presentw , we may say, only in that it is simultaneous with some



Two Metaphysical Perspectives

53

other event such as the tokening by a subject of certain words (e.g. “now” or a verb in the present tense) or the subject’s having certain thoughts or sensations. And, similarly, other events are past or future in an analogously weak sense, inasmuch as they are earlier or later than present events. In this perspective, the pastnessw or futurityw of an event does not cancel the event from the ontological inventory, just as New York is not canceled from it because of the mere fact that I cannot classify it as “here,” while I am in Italy. In sum, we may say, to be a B-theorist amounts to being a B-eternalist, and vice versa. Dolev considers in detail two representatives of this view: Mellor (1981, 1998), and Parfit (on the basis on an unpublished draft).4 An A-theorist, on the other hand, may or may not be an eternalist. According to the A-theorist, the presentness of an event is an objective matter. That an event is present in this strong sense, presents , we may say, has nothing to do with its simultaneity with other (mental) events. Had there not been any thinking subjects, the events that I am observing right now, the leaves’ moving, the sun’s shining and so on and so forth would have still been present. An A-theorist is also an eternalist, if she insists that (i) when an event ceases to be present, it does not go out of the ontological inventory, but merely changes from being present to being past, and (ii) when an event happens to be present, it has not simply popped into existence, but has merely changed from being future to being present. Dolev discusses a view of this kind, a “moving now theory,” in the version defended by Schlesinger (1980, 1982, 1991).5 The A-theorist who is an eternalist, the A-eternalist, we may say, will typically add that present events are somehow privileged, that in some sense they are more important, perhaps even more real than past or future ones. For this reason, the A-theorist may be called a presentist.6 Dolev, I think, is using the label “presentism” in this sense (p. 5) and thus classifies Schlesinger as a presentist. But it is much more common nowadays to use this label in another way, according to which a presentist is a non-eternalist A-theorist who claims that only what is present exists and thus that no past or future events are to be found in the ontological inventory.7 Dolev calls this view (or at least something in its vicinity) “solipsism of the present moment” and discusses it (in § 3.2) not by focusing on its typical representatives (such as Prior and, more recently, e.g. P. Ludlow and C. Bourne), but rather on the Dummett of “The Reality of The Past” (see Dummett 1978: ch. 21). To avoid confusions, I keep in line with the dominant terminological trend and disallow the term “presentism”for moving now A-eternalist views such as Schlesinger’s. I thus reserve it for a standpoint according to which only present events exist. My usage is a bit more liberal than the usual one, according to which this term refers, we may say, to typical or traditional presentism, that is, the view claiming that everything, not just every event, is present. If this is accepted,

54

DEBATES IN THE METAPHYSICS OF TIME

my moderate presentism counts as presentism (otherwise my use of “moderate” should be taken like “fake” in “fake diamond”).8

Ontology and the “ontological assumption” As noted above, according to Dolev, the very intelligibility of the A- and B-theories of time is questionable. This is because they suffer from inner tensions that ultimately depend on an “ontological assumption,” which they both share. Here is his explanation of what this assumption amounts to (pp. 8–9): [T]here is an ontology here waiting to be fleshed out, or, what amounts to the same, the idea that reality claims—claims to the effect that events and objects are or are not “real”—are the key to the philosophical understanding of time. I will call this idea the ontological assumption … the shared assumption is that if there are real differences between the past, present, and future then they are ontological differences. Thus, the ontological assumption is the claim that events and objects, depending on whether they are past, present or future, may or may not be “real,” so that the key problem in the philosophy of time is deciding whether past, present, and future objects and events differ in terms of reality: (i) they are equally real (B-eternalism), (ii) they are all real although perhaps in different ways or degrees (A-eternalism), and (iii) the present is real and the past and future are unreal (presentism).9 According to Dolev, the ontological assumption traps the supporters of all these approaches in an “ontological debate,” which must be somehow transcended by abandoning the assumption in order to make real progress in our understanding of time. This leads to a “post-ontological” or “phenomenological” view bereft of the ontological assumption. Such a view is, according to Dolev, neither an A- nor a B-theory, and captures, as we may say in Sellars’ terminology, both our manifest and scientific (relativistic) images regarding time much more closely than A- or B-theoretical approaches can ever aspire to do. Such approaches should be viewed merely as stages or phases (Dolev 2009: 2; in Hegelian fashion, I would add) wherein no truth in the full sense of the term can emerge, for the theses of one stage are contradicted by those of the other, in such a way that one feels trapped in inconsistency. These stages are, however, necessary in order to reach a superior enlightened phase that supersedes them and in which real truth emerges. Thus, according to Dolev, we must go through the ontological debate in which the supporters of the Aand B- theories are involved in order to have access to the more satisfactory



Two Metaphysical Perspectives

55

post-ontological phase.10 I shall argue below that Dolev does not ever get to a post-ontological phase, as he proposes his own ontological theory. But before moving to this, let me briefly explain why I am not convinced by how Dolev reviews the debate in order to justify the need to turn to an allegedly post-ontological phase. Dolev criticizes both A- and B-theories in a general way (§ 4.2), because of the fact that they share the ontological assumption. But basically he objects to insisting on a word, “real,” that allows for different and contrasting meanings and is often employed in ordinary language in a way not altogether in line with typical philosophical usages. However, as we shall see below, this problem is easily circumvented, once we clearly focus on the main task of ontology. Moreover, when Dolev turns his attention to specific A- or B-theories, (i) it is not clear that his criticism of them really depends on attributing to them the ontological assumption, and (ii) the selection of theories from the debate is too incomplete and idiosyncratic to license any conclusion about a need to dismiss all sorts of A- and B-theories (Meyer 2009; Tallant 2009). In particular, Dolev’s criticism of B-eternalism focuses on the new B-theory,11 according to which, roughly, reality is tenseless, although language and thought are tensed, and tries to capitalize on the inner tension that this duality generates in a view of this sort. However, the B-theorist, despite what is typically assumed nowadays, may well dwell on the old B-theory, according to which both reality and thought and language are tenseless (Orilia and Oaklander, forthcoming 2014). Thus, even if the tension that Dolev tries to bring to the surface could not really be resolved (but see Oaklander 2014), the B-theorist can eschew it at the outset.12 Moreover, Dolev’s criticism of the A-theory focuses only on Schlesinger’s A-eternalism and Dummett’s presentism. The former assumes possible worlds (differing from each other as regards which moment in them is present) in accounting for the “moving now” in a way that most A-theorists would find questionable and unnecessary, while the latter is a highly heterodox version of presentism. Whereas the typical presentist tries to respond to the truthmaker objection to presentism by positing entities that can work as truthmakers of our intuitively true assertions about the past (Magalhães and Oaklander 2010: Part IV, Sect. 3), Dummett argues that many such assertions may well be neither true nor false. Thus, I do not think that, through criticisms of these specific theories, Dolev can show that the debate must be superseded in order to reach a post-ontological view that is neither A nor B. But is it desirable or even feasible that we reach a post-ontological view, a view about time that is not ontological? I do not think so. The key issue in ontology is, as Quine teaches us, what there is, or, equivalently, what exists or has existence.13 In addition, one may also inquire on the issue of whether what exists, or some of what exists, is mind-dependent or objective

56

DEBATES IN THE METAPHYSICS OF TIME

(mind-independent). In the ontological investigation of time, both issues have center stage. As we have seen, the presentist argues that only present events exist, that is, there are only present events. In contrast, according to the eternalist, there are also past and future events. And, within the eternalist camp, there is disagreement over whether events are (mind-dependently) pastw, presentw or futurew, as the B-eternalist has it, or (objectively) pasts, presents or futures, as the A-eternalist claims. The word “real” can cause misunderstandings because it is used in relation to both types of inquiries, since, in philosophical usage, “is real” can simply mean “is” or “exists,” but it can also mean “is mind-independent.” But, once equivocations due to this double use are set aside, the issues are sufficiently clear. Hence, in the following I shall avoid the use of “real”as equivalent to “mind-independent.” With this proviso, I shall go on to argue that Dolev has a definite ontological position in that he is committed to the existence of past, present, and future events. Whether he is an A- or a B-eternalist is, as we shall see, less clear.14 But, in any case, according to his world-view, there are events that the presentist would not acknowledge in her ontological inventory.

Dolev’s “post-ontological” perspective on the duration of the present It may be wrong to deduce Dolev’s positive view of time from his critique of the A- and B-theories, since, as we have seen, he regards them as stages that must be superseded, but I suppose it can be legitimately done on the basis of what he says from the allegedly post-ontological perspective of his Chapter 5. And therein Dolev commits himself to an ontological standpoint, namely eternalism. I shall now discuss Dolev’s account of the duration of the present in § 5.1 to support this claim (some reference to the subsequent § 5.2, dealing with “the presence of experience” will also be relevant). Dolev takes his clue from a reconstruction by Gale (1968: 4) of the famous argument by Augustine about the duration of the present (Confessions, Book XI, Sect. 15). This is usually taken to address the issue of the length of the present, understood as a moment of time, with the conclusion that it is a durationless (pointlike) instant (see, e.g. Dainton 2000: 120): a year cannot be present as a whole, because, if we are, let us suppose, in May, it is at best this month which is present, since the period January–April precedes it and is thus past rather than present, whereas the June–December period follows it and is therefore future and not past; similarly, for the month, the week, the day, and so on, until one reaches the durationless instant and can no longer apply this line of reasoning. In contrast, Dolev extracts from Gale’s



Two Metaphysical Perspectives

57

reconstruction of it a “retrenchability argument” (following Westphal 2002) that is meant to show that the length of the present, whether understood as a time (occupied by certain events) or as an event or sum of events (occupying a time), is a contextual matter. This is possible, claims Dolev, after having removed “several hidden assumptions underlying it, which belong to the ontological framework” (p. 120). In particular, Dolev has in mind the presupposition that the present is ontologically privileged in that only the present exists, since “the past now is not” and “the future is not yet.” Once this is given up, argues Dolev, the only reason why it may not be appropriate to call a year (and the events within it) present is that the context makes it more apt to focus, say, on a monthly period; similarly for the month, the week, and so on. In other words, whether a certain time or event is present or not is a merely contextual matter. Let us consider more details of the picture that Dolev proposes and then contemplate the ontology that emerges from it. A crucial notion of this picture is cotemporality, which unfortunately is never explicitly defined. One might surmise that “cotemporal” is used as synonymous with “simultaneous,”15 but it is not so, because the latter is used for events that occur at the same time and thus presumably have the same duration, whereas the former may apply to events of different durations: “Events or states of affairs can be cotemporal without being equal in duration” (pp. 141–2). From the examples offered by Dolev, we can, I think, infer that he takes two events to be cotemporal if one of them occupies a time interval included in or coinciding with the time interval occupied by the other. Thus, a dog’s barking may be cotemporal with a broadcasting of news on TV, because the time occupied by the latter includes the time occupied by the former. But the dog’s barking can also be simultaneous with a bell’s ringing, because they occupy the same time interval. As we shall now see, Dolev appeals to cotemporality, as so understood, in order to clarify what presentness, in his opinion, amounts to. Events are said to be present in relation to our “first-hand experiences” (or immediate experiences, as I’d rather say), which work as standards in assigning A-properties to time and events, pretty much as the standard meter in assigning a length to other objects.16 Thus, by focusing on a certain immediate experience, e (say, my experience of hearing a certain sound), I take a certain event, for example the bell’s ringing, as present, because it is cotemporal with e (pp. 140–1). Or I take the long and complex event consisting of the current Olympic Games as present by taking it to be cotemporal with my experience e’ of watching on TV, for instance, a certain soccer match in the Olympic program. Similarly, I can take another event, for example the Battle of Waterloo, to be past, because it ended before e or e’. Derivatively, presentness and other A-properties can be assigned to times, to the extent that they contain events with such A-properties (pp. 124–5).

58

DEBATES IN THE METAPHYSICS OF TIME

However, these A-properties do not accrue to events and their times intrinsically, but in relation to a context determined by goals and interests (pp. 122–3). Thus, given a certain context, one can take as present either an entire soccer match or the second half of the match (and, derivatively, the times of their occurrences). But no such choice is inherently privileged. Given the appropriate context, claims Dolev, a present event can be of whatever duration (and, derivatively, a present time can be of whatever length). For example, a nine-year-long event such as the current Mars probing, or even some gigantic cosmic event taking millions of years, can be taken to be present (cotemporal with some experience working as standard). One could think that, of the immediate experiences, it is possible to say that they are present tout court, independently of any context. But Dolev argues (p. 139) that even for them we need a context, just as we need an appropriate context to say that the standard meter is one meter long (and this context makes sense only because the meter in Paris has already been established as a unit of measure). From this perspective, the Augustinian argument simply shows that, by repeatedly asking appropriate questions, one can keep changing the context in such a way that, from the standpoint of the newly chosen context, the interval that counts as present is smaller than the one that counted as present in the previously chosen context, until one finally gets a context according to which the time is simply a durationless instant. But since no context is privileged or “more real” than any other, the argument at most shows that, from the point of view of a certain context, the present is pointlike, not that it is pointlike in an absolute sense. In essence, Dolev is taking very seriously the well-known fact that in ordinary life we use “present” and, similarly, “now,” in relation to events or times of different lengths. By taking ordinary life so seriously, Dolev thinks that he is “transcending ontology,” so as to offer a “post-ontological” (phenomenological) perspective on time, which is meant to be radically different from the one emerging from the A- and B-theories. But I think that it is not the case that Dolev is really presenting an alternative, or at least not a coherent one, since, in taking so seriously this use of “present,” he is reproposing eternalism, although he seems willing to place it neither in the A- nor in the B-theoretical camp (hence the suspicion that his proposal is not in the end fully coherent). Let us see first why Dolev counts as an eternalist (whether of the A- or B-variety). As we saw, the basic point in attributing to someone an ontological position such as eternalism is to look at her ontological commitments. And Dolev’s account of the duration of the present clearly shows that he is committed to the thesis that there are the events typically acknowledged by the eternalist and rejected by the presentist. For example, suppose that I am now watching a live broadcast of a certain soccer match that is part of the



Two Metaphysical Perspectives

59

Olympic Games and is taking place, let us suppose, a week after the inauguration of the Games. Then, there is, according to Dolev, an event such as the one consisting of the whole occurrence of the Olympic Games (it can even count as present, given the appropriate context). But this event is made up of course of many other events, for example, let us suppose, the event e1, consisting of athlete A’s winning the first race of the Games in their first day, and the event en, consisting of athlete Z’s winning the last race of the Games in their last day. Clearly, Dolev is committed to there being such events. Indeed, as is evident also from Dolev’s most recent reflection on a comparable example (2014: 42), in his view they may even all count as present, given the appropriate context.17 Nevertheless, since e1 occurs before my watching the soccer match, and en occurs after it, they can hardly be present, at least from the point of view of a presentist, who will classify them as past and future, respectively, and will not acknowledge that there really are such things. She will admit at most that there are descriptions such as “the victory of athlete A” or true propositions such as that athlete A won (it remains to be seen what these descriptions refer to and what makes such propositions true, according to the presentist). Not so, however, from the eternalist’s point of view and similarly not so from Dolev’s point of view. Since these examples could be multiplied ad infinitum, clearly Dolev is committed to precisely the same events as an eternalist and thus it is fair to classify him as such.18 Depending on whether, according to Dolev, the pastness or futurity of events such as the victory of athlete A and the victory of athlete Z are merely pastnessw and futurityw or (also) pastnesss and futuritys, Dolev’s eternalism is either a B- or an A-eternalism. There are reasons, as we shall now see, in favor of both options. In favor of an attribution of A-eternalism, one could note that Dolev explicitly insists (p. 128) that his contextualist approach to the present does not make the present mind-dependent and that an event happening now, such as the merging of two clouds, would be present even if there were no experiences working as standards, just as Mount Everest would still be 8,847 meters high even if nobody had selected the Paris rod as standard (p. 143). Moreover, one could remark that his criticism of Schlesinger’s A-eternalism in § 4.6 could be easily dispelled. The problem individuated by Dolev is that this position is self-contradictory, because, on the one hand, it claims that only the present is real, but, on the other hand, it also takes past and future events as real in that (i) they can work, as I would put it, as truthmakers of sentences about the past or future,19 and (ii) as the moving now proceeds, unreal future events become real (p. 109). The problem is not serious, however. We can, for example, distinguish degrees of reality or different senses of the word “real.” On the other hand, in favor of an attribution of B-eternalism, one could insist that the reason offered by Dolev for the compatibility of his contextualism and

60

DEBATES IN THE METAPHYSICS OF TIME

the objectivity of the present is not convincing. Dolev’s point is that, once a context is set, so that, for example, the whole soccer match that I am watching (rather than, say, its first half) turns out to be present, its duration, as measured by a non-mental entity such as a clock, is an objective matter (p. 128). But this is in line with what a B-theorist would say: what is objective is that the soccer match and, say, the movement of the clock’s hands from position x to position y are simultaneous, which is perfectly compatible with denying that these two events are objectively present, in opposition to the A-theory. Moreover, Dolev’s account of relativity theory at p. 201 is definitely B-theoretic: simultaneity and copresentness are frame-dependent and thus not transitive, so that there is not a single present but only presentness in a frame of reference.20 My final diagnosis is that Dolev endorses an incoherent mixture of A-theoretical and B-theoretical claims.

A presentist’s perspective on the duration of the present Thus, it is not clear whether we should attribute to Dolev an A- or a B-theory and perhaps he would say that this is as it should be, since his postontological view supersedes both (see, e.g., p. 13). But whatever we make of this, he is definitely committed to eternalism. Now, given Dolev’s just desire to save the phenomenological data about time, this is seriously problematic, since eternalism is in tension with two most important such data, namely our deep-seated feelings that (i) the past is gone, so that whatever agreeable or awful events have turned up they are no longer around with their pleasantness or dreadfulness, and (ii) the future is open, so that, depending on our free choices, agreeable or awful events can be brought about. If eternalism is true, the terrible pain that Arthur suffered from years ago somehow still is, the ontological inventory contains it. In other words, there is the unpleasant event of Arthur’s being in pain and thus the world, we may say, is affected by a certain degree of unpleasantness, in spite of the fact that Arthur exclaimed with relief, “thank goodness, it’s over.” In contrast, if presentism is true, Arthur’s relief is well justified, for the unpleasant event no longer exists and the world is, at least to that extent, a better world. Similarly, if eternalism is true, there is, let us suppose, the pleasant event of Roderick’s getting an A in his homework due next week. We may not know which of the two propositions, that Roderick gets an A and that Roderick does not get an A, is true. But in fact one of them is definitely true, because there is its truthmaker. Given our supposition, it is the proposition that Roderick gets an A that has



Two Metaphysical Perspectives

61

a truthmaker, namely the event of Roderick’s getting an A. If, in a compatibilist fashion, we see free will as merely depending on a causal connection between certain intentions and volitions and certain subsequent events, we can still say that Roderick’s success in the homework depends on his free choice of working hard for it, but certainly non-compatibilist accounts of free will, which presuppose that neither of those two propositions already has a truthmaker, are doomed at the outset. Vice versa, if presentism is true, no such truthmakers are around and the way is open for viewing these propositions as somehow undetermined, with elbow room for non-compatibilist accounts of free will and more generally for our feeling that the future is open. Presentism is thus able to save the gone past and the open future by acknowledging only present events in its ontology. This feature of presentism is in my view especially valuable and most crucial to motivate this view. We saw, however, that Dolev’s interpretation of the Augustinian argument and his consequent account of the extension of the present leads to a forestalling of presentism in favor of eternalism. I shall thus look back at the Augustinian argument to see whether one can support a different, more traditional, account of the argument, one that does not lead to eternalism. The traditional interpretation of the Augustinian argument, which takes it as aiming to establish that, strictly speaking, only a durationless instant can be present, must certainly be squared with the linguistic fact on which Dolev tries to capitalize, namely that we can use “present” and “now” to refer, depending on the context, to temporal intervals of various lengths. We can do this by distinguishing, in Chisholm’s well-known terminology, between a “strict and philosophical” and a “loose and popular” sense of these words. Just as in a loose sense we can call “identical”two twins who look alike, although in the strict sense of “identity” they are not identical, we can similarly call “present” in a loose sense, say, a minute, although in the strict sense of “present” it is not present. The Augustinian argument must then be understood as having to do with the attribution of presentness in a strict and philosophical sense. But why is the presentness of a durationless instant, as opposed to that of, say, a minute, strict and philosophical? As I see it, the reason is that only in the former case is this presentness compatible with these two intuitive principles (at play in the Augustinian argument): (P1) Whatever is past or future is not present. (P2) If something temporally precedes or follows what is present, then it is past in the former case, and future in the latter. The incompatibility in the latter case emerges because the attribution of presentness to a minute, effected, say, at its 30th second, seemingly involves

62

DEBATES IN THE METAPHYSICS OF TIME

the attribution of it to all its parts and thus, for example, to the part going up to its 29th second, which temporally precedes the 30th second and should then be, by (P2), past, and, by (P1), not present.21 In sum, it seems that the principles (P1) and (P2) cannot be retained without also buying the Augustinian argument in its traditional interpretation. On the other hand, they had better be saved, if possible, for they are part of our pre-theoretical data about time, and in general it is preferable to have, ceteris paribus, a philosophical theory that preserves the pre-theoretical data. Of course, there is nothing wrong, in the appropriate context, to use “present” in ordinary life to speak of a minute or an hour, but in the philosophical effort of constructing a theory about time, it is legitimate to propose that, strictly speaking, presentness complies with (P1) and (P2) and thus cannot be attributed to an extended interval. In sum, the principles (P1) and (P2) support the traditional interpretation of the Augustinian argument and thus defuse Dolev’s account of it, which, as we saw, leads to eternalism. However, once we combine them with presentism, some serious problems arise. First of all, (P1) and (P2), given presentism, imply that there are no extended intervals of time.22 As repeatedly noted by Augustine, this is very puzzling, for we measure such intervals (Confessions, XI, 16, 21) and, we can add, we also refer to them with dates such as “the year 2013” or “February 2013.” Moreover, (P1) and (P2), given presentism, imply that there is no instant preceding or following the present one, since an instant simply ceases to exist when it becomes past. But this is puzzling too, because, as also noted by Augustine, the present is a present time, and “not eternity,” in so far as “it passes into time past” (Confessions, XI, 14). In other words, the present instant is an instant of time, because it becomes past and this becoming past should perhaps be looked at as a sort of transformation, rather than as a failing to exist. And in fact we seem to be able to refer to past instants with dates such as “June 23, 2013 at 3 o’clock.” Further, it has been noted that, given presentism, the Augustinian argument seems to imply that experiences are durationless and can only occupy a durationless instant (Dainton 2000: 120; Le Poidevin 2007: 79). Given the Augustinian argument, only a durationless instant is present and, according to presentism, there are only present events. Thus, presumably, all events occupy a durationless instant. But experiences are themselves events and thus they must all be durationless. This, however, is perplexing, for our experiences typically appear to involve a duration, with an earlier and a later part. For example, if we look at a moving billiard ball, we seem to have an experience of movement involving, in one fell swoop, the impression of a ball first in a certain position and after in another position.23 Finally, independently of the Augustinian argument, which focuses on intervals and instants rather than events, (P1) and (P2), given presentism,



Two Metaphysical Perspectives

63

seem to force us to deny that there are dynamic events and to acknowledge only static events (Orilia 2012a). Following Casati and Varzi (2010), the former are those that, intuitively, require an extended interval of time, such as the movement of a ball from one place to another, whereas the latter are those that, intuitively, do not require an interval and, at least in principle, can occur at an instant, for example a ball’s being precisely at a certain position in space. Dynamic events are problematic, given presentism, (P1) and (P3), because they seem to involve an earlier and a later part. For example, if there is the event of a billiard ball moving from place p1 at time t1 to place pn at time tn, then presumably there is the earlier static event of the ball’s being, say, in the intermediate position p3 at time t3, followed by the later static event of the ball’s being in a successive position p4 at time t4. But the earlier static event is past and not present in view of (P2) and (P1) (suppose the ball is presently in p4) and thus non-existent, if we accept presentism. And, if we accept presentism, the dynamic event as a whole can hardly count as existent, since it would have non-existent parts. What can we make of all this? I propose to deal with these issues by advancing moderate presentism, which retains the basic intuition of typical presentism, according to which only present events exist, but is prepared to acknowledge past and future durationless instants in addition to the present one and thus, one may say, extended intervals, as somehow made up of such instants. Such moments of time (instants and intervals) should be conceived of in a substantialist, Newtonian, fashion, and not as somehow arising from classes of simultaneous events (Russell 1914; Whitehead 1929) or mereological sums thereof (Pianesi and Varzi 1996), as the opposite relationalist account has it. For otherwise the idea that there are only present events would be immediately given up. If, for example, the instant referred to by “26 October 1860 at 12.30 p.m., Italian time” were nothing over and above a class or mereological sum of events, including, inter alia, the event of Garibaldi and King Vittorio Emanuele II’s meeting in Teano, then our ontological inventory should acknowledge this past event. If we want to deny that there are such events and yet admit that there are past (and future) instants, we have to see instants as items over and above the events that occur at them. The idea then is that the present instant is the instant at which, objectively, events occur, the present events (the only ones that there are), whereas the past and future instants are those that precede and follow the present instant, in a temporal order somehow primitively given in such a way as to give a direction to time (as perhaps a substantialist eternalist may have it). Such past and future instants are, we may say, “empty,” since events no longer or not yet occur at them. The becoming present of a future instant is then not its coming to be, but, as hinted above, a sort of “transformation” from its being empty to its being such that events occur at it. Similarly, the becoming past of a present

64

DEBATES IN THE METAPHYSICS OF TIME

instant is its becoming empty, after having hosted events. In contrast, events, in keeping with the crucial intuition of presentism, are not subject to a transformation, to becoming present after having been future and past after having been present. They are rather subject to absolute becoming; they come into being when a certain property is exemplified by an object or a relation is jointly exemplified by some objects. Let us now briefly see how we can deal with dynamic events in the light of these additional resources.24 Consider again the ball’s moving from position p1 to position pn. We can now say that, at any time during this movement, the ball is not only exemplifying “present-centered” properties such as, say, being in position p3, but also “past-oriented” (“time-indexed”) properties such as having been in position p2 at time t2 and (“time-indexed”) “futureoriented” ones, whose nature I now turn to explain, by focusing for illustrative purposes on the ball’s presently being in position p3 at time t3 in the course of its movement. We can understand future-oriented properties in two ways, depending on whether or not we take it as fully determined at t3 that the ball would then be in p4 at time t4. If it is fully determined, then the ball has at time t1 the future-oriented property of going to be in p4 at time t4. On the other hand, if we do not take it to be fully determined at t3 that the ball will be in p4 at time t4, the future-oriented property that the ball has at t3 is a mere propensity, a property such as being potentially in place p4 at time t4. We can conceive of it as a property that the ball exemplifies in so far as, roughly speaking, it is storing (say, by having been pushed) some kinetic energy leading in a certain direction. Having this property does not necessarily result in its being in p4 at time t4, since, for example, there can be an intervening obstacle. According to this perspective, dynamic events, rather than being discarded, are, we may say, supervenient on static events, involving present-centered and past- and future-oriented properties, all occurring at the present instant. Thus, for example, there is the dynamic event of our moving ball in so far as there are (now, at t3) static events such as, inter alia, the ball’s having been in p2 at t2, the ball’s being in p3, and the ball’s being potentially in p4 at t4. Or, if one wishes, a dynamic event is a “conjunction” of static events of this sort and thus a conjunctive event. In this way of dealing with dynamic events, moderate presentism should perhaps be prepared to make another concession to eternalism and admit in its ontological inventory past objects; following Williamson (2002), they could be viewed as ex-concrete objects, objects that used to be in space and were thus concrete, but are no longer in space. For it seems we can observe dynamic events involving the ceasing to be of objects, their turning to be ex-concrete. For example, our billiard ball could for some reason explode in reaching position p3. Be that as it may, admitting ex-concrete objects allows



Two Metaphysical Perspectives

65

moderate presentism to deal with the notorious problem of the truthmaking objection to presentism in a way that is not open to typical presentism.25 For example, suppose it is now true that Arthur, who passed away long ago, suffered from a headache at noon on May 4, 1956. This proposition has a truthmaker, that is, the event consisting of Arthur’s having had a headache at noon on May 4, 1956, since Arthur, albeit ex-concrete, presently exemplifies the past-oriented property of having had a headache at noon on May 4, 1956: after becoming ex-concrete, an object does not lose the past-oriented properties that it has accumulated as time goes by, and, we may add, it can even gain new present-centered intentional properties, such as being observed (think of a star exploded long ago and yet in our firmament) or being admired (think of Arthur Prior). But it cannot have present-centered physical or mental properties such as having a certain mass or a headache. Let us finally deal with the issue of the apparent duration of experiences. From the perspective of a presentism that takes the present as a durationless instant, we can deal with this, by appealing to Husserl’s retentions, impressions, and protentions. This is in contrast with Dolev’s (2014: 43) misgivings about Husserl’s approach, but in line with Dolev’s recommendation to pay attention to phenomenologists, which is not at all in tension with doing ontology. The idea is that an experience, while occupying, as any other event, a durationless instant, may well look extended because it involves not only impressions, but also retentions and protentions (Dainton 2010: § 6.3). With past-oriented, present-centered and future-oriented properties in our ontology, we can understand retentions, impressions, and protentions as mental representations of three kinds of static events: objects exemplifying past-oriented properties (the ball’s having been in p2 at t2), objects exemplifying present-centered properties (the ball’s being in p3), and objects exemplifying future-oriented properties (the ball’s potentially being in p4 at t4).

Conclusion Any theory has its own ontological commitments and in theorizing about time there is no exception to this. It is thus impossible to be engaged in “post-ontology” or mere phenomenology, nor should this worry us, as long as we have the appropriate canons of rigor and avoid equivocations. Thus, Dolev himself has his own ontological, or metaphysical, theory of time, which commits him to eternalism. Eternalism, however, cannot easily accommodate our feelings that the past is gone and the future open. Presentism is best suited to this task, and moderate presentism can accomplish it without the difficulties of typical presentism.26

66

DEBATES IN THE METAPHYSICS OF TIME

Notes   1 Unless otherwise indicated, page and paragraph references in the following are to Dolev’s book and my attributions of views to Dolev are mainly based on it. However, occasional references to Dolev (2014), and to another work by Dolev (2009) will also be appropriate.   2 I treat “event” and “state of affairs” as synonymous (I think this is basically in line with Dolev’s usage, at least as regards the issues discussed here).   3 To call a view of this kind “eternalist” is not of course meant to imply that such events are eternal in the sense of being everlasting, i.e. existing at all times (see Oaklander 2014).   4 The version of the old B-theory discussed by Dolev (2014) in this volume is another representative.   5 One should not be misled into thinking that A-eternalism is simply B-eternalism with a moving now superadded, so to speak, to a temporal series of times or events resulting from B-relations, in such a way that we could get B-eternalism back, by taking the moving now away. For A-eternalism is in the first place an A-theory and thus considers A-properties essential for there to be time at all. Thus, from the perspective of A-eternalism, since the exemplification of A-properties by times or events is dependent on the moving now, without such a moving now there would be no temporal series. To put it otherwise, the B-series would not be a temporal series.   6 And in fact there are A-theorists, such as W. L. Craig and Q. Smith, who classify themselves as presentists, even though they are best viewed, given my preferred terminology, as A-eternalists (Zimmerman 2005).   7 There may be qualms over this way of classifying approaches. For example, Oaklander, in private correspondence, has commented thus on this matter (with reference to the B/R theory, discussed in Oaklander 2014 (basically, a B-theory, as I see it); “B” and “R” remind us of Broad and Russell, respectively):   [This classification] assumes that both “eternalists” accept the existence of temporal relations and just debate over the ontological status of the present, but that is potentially damning for the B/R theory (since the A-eternalist would deny that without the privileged now we have temporal relations) or damning for the A-theorist eternalist, since temporal relations (of the R-theoretic stripe [B-relations, or at least akin to them]) cannot obtain with terms that have A-properties. Thus, if the temporal relations that R-theorists accept is common to both forms of eternalism, then A-theoretic eternalism must deny temporal relations and so morphs into true presentism.

I think that the B-theorist would insist that a privileged now is not required for the instantiation of temporal B-relations, and the eternalist A-theorist would argue that she can understand temporal relations in way that does not imply presentism. For present purposes, however, I think that we can put these issues aside, since they do not hinge on what I want to focus on here.



Two Metaphysical Perspectives

67

  8 It should be noted that the presentist, whether traditional or moderate, does not deny that, e.g. the birth of Napoleon or (perhaps) the birth of the first child of 2021 existed (was in the ontological inventory) or will exist (will be in the ontological inventory). The point of the presentist is that they do not exist now. However, to the extent that she is committed to the truth now of the past tense proposition that Napoleon’s birth existed and the future tense proposition that the birth of the first child of 2021 will exist, she owes us an account of what makes these propositions true compatible with her view of what the ontological inventory contains now. More on this later.   9 There is a fourth option seriously considered in the current literature, namely that the past and present are real whereas the future is unreal or branching. But since Dolev does not discuss it, I have neglected it here. 10 In an ideal reconstruction, I myself would rather distinguish (following Castañeda 1980) a phase dedicated to the collection of pre-theoretical data, which typically involves puzzles and thus conflicting intuitions (“protophilosophy”), a subsequent phase of theory construction (“symphilosophy”) where any apparent protophilosophical contradiction is removed, and finally a phase of intertheoretic comparisons (“diaphilosophy”). 11 Mellor, on whom Dolev focuses, is a typical supporter of the new B-theory. The fact that Dolev also discusses Parfit’s unpublished version of the B-theory does not alter the basic fact that Dolev’s misgivings about the B-theory ultimately derive from his focusing on the new B-theory. 12 In his rejoinder to Oaklander (2014), Dolev (2014) has now taken up the old B-theory as well, or at least Oaklander’s attempt to revive it, and claimed that the ontological assumption afflicts it as well, to the extent that, in Dolev’s words (2014), “it is presented as an ontological theory.” I cannot engage in a full discussion of this here, but it should be evident that my points against Dolev’s misgivings regarding the ontological assumptions also apply to his criticism of the old B-theory. 13 I basically side with Meyer (2009: 97) when he urges that in looking at the A vs. B dispute we should not worry about how “real” is used and should rather concentrate on what exists, according to the parties. 14 Thus, I do not have Meyer’s (2009: 9, n. 7) impression that Dolev’s sympathies simply lie with the B-theory. Meyer’s impression may seem to be confirmed by this explicit claim by Dolev (2014: 47): “The view I have been defending has much in common with the old B-theory (and very little with the new one, or with any versions of the A-theory).” Yet, as we shall see, there are significant A-theoretical elements in Dolev’s standpoint. 15 The occurrence of these terms at p. 92 may suggest this. Dolev also uses another related term, “copresent,” in discussing relativity theory, but seems to take it as equivalent to “simultaneous” (see p. 201). 16 The standard meter is one and public, whereas experiences are many and private; thus, Dolev prefers to say that the latter work as “quasi-standards” (p. 140). 17 It may be objected that in making such assertions I am attributing to Dolev a view that is not really his, for he has explicitly disavowed “the

68

DEBATES IN THE METAPHYSICS OF TIME phenomenologically and conceptually ungrounded presupposition that events [such as the Olympic Games] consist of parts [such as the victories of athlete A and athlete Z] that are somehow brought together to form a unity” (Dolev 2014: 43–4). Yet, Dolev himself speaks of events in part/whole terms: “some present events [the second part of a symphony, broadcast live on TV] are parts of other present events [the whole symphony, broadcast live on the radio]” (Dolev 2014: 42).

18 This is confirmed by what Dolev says of past and future objects and events, e.g., in relation to memory at p. 153, to future contingents at pp. 187–204, and to relativity theory at p. 202. And also by what he says of time passage, which is understood as the becoming present and then past of future objects and events, with the proviso that this does not imply their turning from being “not real” to being “real” and then to “not real” again, since the ontological assumption has been given up (pp. 164–5). The fact that there are, in Dolev’s opinion, the objects and events subject to this becoming is sufficient to nail him to the eternalist’s ontological commitments, in spite of his refusal to use the words “real” or “not real” for them. 19 In Dolev’s terminology (p. 109), there are “conditions obtaining at any time, past, present, or future, as truth conditions, as that the obtaining of which establishes the truth or falsity of sentences.” 20 This is B-theoretic up to a point, because, given a space-like separation between two events, even B-relations linking such events are frame-relative. But this is a problem (underestimated, in my view) for any approach that wants to reconcile relativity theory with the B-theory and I shall set this issue aside here. 21 It is not clear to me whether Dolev would agree or disagree with the idea that the attribution of presentness to the minute implies an attribution of presentness to all its parts. In favor of the second hypothesis there is his claim that “present events have parts that are not themselves present” (Dolev 2014: 42). In favor of the first hypothesis, there is his claim that “some present events are parts of other present events” (Dolev 2014: 42). Be this as it may, I understand that one could insist that the being present of the minute does not imply that all its parts are also present, but only that a certain instant within it is such. But this is another way of saying that only a durantionless instant is, strictly speaking, present, which is what I am pressing. 22 Since, as we saw, they cannot be present and thus do not exist, according to presentism. 23 Dolev (2014: 42–3) tries to exploit the durational character of experiences in order to support his conception of time, but he does so by also claiming that experiences, and more generally events, do not have successive parts that somehow form a unity. In contrast, it seems to me that our experiences, e.g., as of a fast moving object, do appear to have successive (earlier and later) parts. There is no room to try a diagnosis of this disagreement here. 24 See Orilia (2012a) and Orilia (2014) for some additional details. 25 By appealing to ex-concrete objects, moderate presentism can also deal with intertemporal relations such as causation, in a way that is not open to



Two Metaphysical Perspectives

69

traditional presentism (Orilia 2012b: § 6.7). Reasons of space prevent me from going into this here. 26 I wish to thank L. Nathan Oaklander for his valuable advice and his comments on a previous version of this chapter.

References Casati, R. and Varzi, A. (2010), “Events,” in E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy. (Spring 2010 edn), http://plato.stanford.edu/ archives/spr2010/entries/events/. Castañeda, H.-N.(1980), On Philosophical Method. Bloomington, IN: Noûs Publications. Dainton, B. (2000), Stream of Consciousness. London: Routledge. —(2010), “Temporal Consciousness,” in E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Spring 2010 edn), http://plato.stanford.edu/ archives/fall2010/entries/consciousness-temporal/. Dolev, Y. (2007), Time and Realism. Cambridge, MA: MIT Press. —(2009), “Time and ontology: a reply to Meyer,” Iyyun. The Jerusalem Philosophical Quarterly, 58, 292–300. —(2014), “Motion and Passage: The Old B-Theory and Phenomenology,” in L. N. Oaklander (ed.), Debates in the Metaphysics of Time. London: Bloomsbury. Dummett, M. (1978), Truth and Other Enigmas. Cambridge, MA: Harvard University Press. Gale, R. (ed.) (1968), The Philosophy of Time. New Jersey: Humanities Press. Le Poidevin, R. (2007), The Images of Time. Oxford: Oxford University Press. Magalhães E. and Oaklander, N. (eds) (2010), Presentism. Essential Readings, Lanham, MD: Lexington Books. Mellor, D. H. (1981), Real Time. Cambridge: Cambridge University Press. —(1998), Real Time II. London: Routledge. Meyer, U. (2009), “Yuval Dolev, Time and Realism (The MIT Press, 2007),” Iyyun. The Jerusalem Philosophical Quarterly, 58, 92–101. Oaklander, L. N. (2014), “Dolev’s Anti-Metaphysical Realism: A Critique,” in L. N. Oaklander (ed.), Debates in the Metaphysics of Time. London: Bloomsbury. Orilia, F. (2012a), “Dynamic events and presentism,” Philosophical Studies, 160, 407–14. —(2012b), Filosofia del Tempo. Roma: Carocci. —(2014).”This Moment and the Next Moment,” in V. Fano, F. Orilia and G. Macchia (eds), Space and Time. A Priori and A Posteriori Studies. Berlin: De Gruyter, 171–94. Orilia, F. and Oaklander, L. N. (forthcoming), “Do we really need a new B-theory of time?,” Topoi, Special Issue on “Time and Time Experience”, edited by Roberto Ciuni and Giuliano Torrengo. DOI 10.1007/s11245–013–9179–6. Pianesi, F. and Varzi, A. C. (1996), “Events, topology, and temporal relations,” The Monist, 78, 89–116. Russell, B. (1914), Our Knowledge of the External World. London: Allen and Unwin. Schlesinger, G. N. (1980), Aspects of Time. Indianapolis: Hackett.

70

DEBATES IN THE METAPHYSICS OF TIME

—(1982), “How time flies,” Mind, 91, 501–23. —(1991), ”E pur si muove,” Philosophical Quarterly, 41, 427–41. Tallant, J. (2009), “Review of Time and Realism: Metaphysical and Antimetaphysical Perspectives, by Yuval Dolev,” Analysis, 69, 2, 372–4. Westphal, J. (2002), “The retrenchability of ‘the present’,” Analysis, 62, 4–10. Whitehead, A. N. (1929), Process and Reality. An Essay in Cosmology. New York: Macmillan. Williamson, T. (2002), “Necessary Existents,” in A. O’Hear (ed.), Logic, Thought and Language. Cambridge: Cambridge University Press, pp. 233–51. Zimmerman, D. (2005), “The A-theory of time, the B-theory of time and ‘taking tense seriously’,” dialectica, 59, 401–57.

4 Temporal Succession and Tense Erwin Tegtmeier

The phenomenon of temporal succession

T

here are phenomena and there are analyses of the phenomena applying scientific theories. That holds also for ontology. The phenomena of interest for ontology are elementary and ubiquitous. The ontologist applies his theories of categories to analyze the phenomena. The succession of two tones and the succession of two light signals are elementary examples. There are, of course, longer successions but they can be derived from pair successions. However, if one considers longer temporal successions one feels led forward step by step. Looking backward by remembering after having followed through the longer succession, I may express the impression by saying that time passes. That may apply even more where the longer succession is a successive change of some material object. There are two alternative ontological analyses of temporal succession based on different views of time. According to the absolutist view of time, it consists of time points; according to the relationist view, time is nothing but a group of relations. Consider a temporal succession from e to e’. The absolutist analysis would be that e is located at time point t1 and e’ at time point t2, and that in the order of time points t1 comes before t2.The relationist assumes an earlier relation between e and e’. That is his entire analysis.

The phenomenon of tense Traditionally, the phenomenon of temporal succession has been closely connected with that of tense. Tense consists primarily in distinguishing and

74

DEBATES IN THE METAPHYSICS OF TIME

singling out some objects as present and thus having a division between present objects and non-present objects. It is roughly a division between objects which present themselves in perception and those which do not. The non-present objects have to be apprehended by other kinds of mental states. Obviously, the division between present and non-present does not involve a temporal succession. That division has as such nothing to do with temporal succession. But it succumbs, so to speak, to temporal succession. It is drawn into the phenomenon of succession. There arise more presents which succeed each other. Only by taking into account the temporal succession of presents are the other tenses of past and future discovered. They become accessible by the mental states of memory and expectation. As to the ontological analysis, there is again the opposition between the absolutist and the relationist. According to the absolutist, what is distinguished as present is a time point or a time interval. The relationist takes objects or events to be present. Both face the question: “What makes those entities present?” The simplest answer which suggests itself is a property of being temporally present.

Is there a property of being temporally present? The best answer would be to refer to a non-relational property of being present. There is reason to conclude that there is no such (non-relational) property. The reason is epistemological. We easily and with certainty recognize that the objects we perceive in our vicinity are temporally present.1 Nevertheless, we are unable to discern a property which all those objects share and which could be called presentness. No property of presentness is given to us. Hence, what allows us to realize so easily that those objects we see near us are present must be something else than a property and something else other than such a property. As has frequently been pointed out, it is a semantic mistake to assume that “now” stands for a single property like all ordinary adjectives. “Now” is an indexical, and indexicals do not always stand for the same. What linguists emphasize is that the reference of indexicals depends on the utterance of the respective sentence and its circumstances. The semantic considerations suggest the assumption of relational property. For the absolutist it is the property of being localized at the time point at which a certain utterance of “now” is located, and for the relationist it is the property of being simultaneous with that utterance. Thus tense has to do with simultaneity to an utterance. That is a relevant phenomenon because perception and effectuation are normally restricted to roughly simultaneous objects.



Temporal Succession and Tense

75

Temporal succession as existential passage Aristotle is a temporal absolutist but his temporal continuum is an undifferentiated whole. Therefore, he needs to introduce time points to explain the phenomenon of succession. He draws on nows (presents) to furnish such points. Thus he relates succession to tense. Aristotle grounds the two phenomena together. Whence their fusion in much of traditional and contemporary philosophy of time. Aristotle holds that a present has zero duration and extension and is like a point. There are many present or now points and the temporal continuum Aristotle assumes allows for infinitely many points. It has the potential for infinitely many points. Aristotle analyzes tense in terms of present points (“now points”) which cut the continuum of time into two halves which are the past and the future. He conceives of the present points and thus of his time points as two-sided boundaries which are designed to avoid difficulties about zero extension. Aristotle distinguishes between the now points which all differ from each other and a seemingly persisting now which is always the same.2 However, he holds that a general now is a mere abstraction of that which makes all now points now points and he does not countenance a persistent now which moves across time or some series of events. Aristotle is a presentist who holds that no two different now points exist together. The existence of a now point excludes the existence of all other now points. How does he ground succession? Not only by relating to now points and thus to time points. Aristotle adds a changing existence of the now points. In the way of ordinary presentism, he assumes that no two time points exist together and that existence passes between times. The latter is presumably designed to generate the order of time points and the direction of time, and that role would justify Aristotle’s assumption of a passage of existence. One could therefore say that Aristotle analyzes temporal succession as a passage of existence between now points which are points on the temporal continuum. But what does this passage mean in ontological detail? There are two time points and their existence, and wherein consists the passage according to Aristotle’s ontology? It cannot consist of anything else than the relating of existence to time points. More precisely in a symbolized way: t1 (E,t1) and -(t2 (E, t1)) (“E” symbolizes existence); in words: t1 exists at t1 but not at t2. Two points can be made about this kind of ontological analysis. First, the order of time points is presupposed and not grounded. Therefore, the passage of existence between now and time points does not add anything to the ontological analysis of advancements such as that between tones and flashes of light. It leads back to the same time points in terms of which they

76

DEBATES IN THE METAPHYSICS OF TIME

can be analyzed without detour via tense. Second, temporal succession is understood as a kind of change, namely existential change. Obviously, Aristotle would not agree with the absolutist analysis of tense suggested by the semantics of temporal indicator words. In Aristotle, tense has not only a selective but a constitutive role. The structure of the temporal is clearly taken to be furnished by tense, although Aristotle would insist that the structure is inherent as potentiality to the temporal continuum. Tense is, according to Aristotle, an objective matter and a matter of existence and non-existence. Aristotle is aware of the grave difficulties of presentism and of an ontological analysis of time derived mainly from tense.3 He is to a certain extent at a loss, having to realize that time is composed of two parts which do not exist (the past and the future) while the part which exists (the present) is merely a boundary. The latter makes the present more dubious since the question arises as to how there can be a boundary for what does not exist.

Temporal succession as passage of the present While Aristotle relies on a tense-oriented analysis of temporal succession in spite of its grave difficulties, because of his view of the temporal continuum McTaggart embraces certain of those difficulties as a means to show that time is unreal. There are also two other contrasts between Aristotle and McTaggart. First, McTaggart starts not from an absolutist but from a relationist view of time, which is Russell’s view. Second, McTaggart assumes a persistent and moving now. He seems to take for granted that there is such a now and not to care about the skepticism of the tradition with respect to it. The moving now plays a crucial role in his argument for the unreality of time. That argument continues to exert a strong influence on the contemporary philosophy of time. Although no one accepts the conclusion of that argument, everyone adopts McTaggart’s pattern of the problem and the alternative solutions, particularly his distinction between A-series and B-series, which leads to the allegation that a theory of time is either an A-theory of a B-theory and to the ill-conceived characterization of Russell’s view of time as “eternalism” just because it is in terms of facts and because facts do not change. Naturally, facts do not change. Facts must not be assumed to be changing since change is based, according to fact ontologies such as Russell’s, on facts. Moreover, facts do not have any duration, hence they cannot last eternally (have an infinite duration). The influence of McTaggart illustrates the point that even the philosophers of time who hold in contrast to McTaggart that tense is not essential to time agree that time is static without a moving present.



Temporal Succession and Tense

77

The concept of “eternalism” originates not only from McTaggart’s misguided argument that a Russellian temporal series cannot possibly change, but also from the traditional tense-oriented view that time as a whole is an eternity divided by a present which serves as a boundary into the past and the future. It is correct, of course, to claim that Russell’s analysis is not presentist and that it rejects the assumption that only what is present exists, but it would be a misinterpretation to say that according to that analysis what exists does so eternally. Rather, the Russellian analysis require a separation of time and existence.4 The phenomenon of temporal succession is mostly designated as “the passage of time,” particularly if longer and many parallel successions are considered. One could also say “the passage of the present.” Thus, I would argue that the term “passage of time” suggests McTaggart’s analysis of the phenomenon. I used the term “temporal succession” because it seems to me more neutral with respect to the alternative ontological analyses. McTaggart prefers to view the passage of time as the moving of the present and therefore he is reliant on a persisting now. He needs a non-relative now. He needs a now which is not relative to utterances or mental acts. If the nows were relative then he would not have one and the same now but several different nows. Hence, one wonders whether he really succeeds in determining such a now. He is disappointingly elusive at this crucial point. Moreover, he is not consistent. On the one hand, he explains that being present is “simple and undefinable.”5 On the other hand, he holds it to be a relation to “an entity X outside” the A-series of events.6 The latter implies that being present can be defined in terms of the respective relation and X. McTaggart does not even give examples for objects which are second relata X for the relation of being present. I guess he has in mind objects X which overlap the first relatum temporally on both sides, which thus begin earlier and end later than the first relatum. Take as an example a climbing of a mountain as a first relatum, and the mountain as a second relatum of McTaggart’s presence relation. McTaggart would argue that before climbing, the climbing was not present, during the climbing it was present, and after the climbing it was not present for the mountain. Since the climbing thus stands in McTaggart’s relation of being present for the mountain and does not stand in it (before and after the climbing) there is clearly a contradiction. And that is just what McTaggart is getting at. After having argued that the present and the two other tenses are essential to time, he arrives at the conclusion that time is contradictory and therefore cannot be real. Initially (The Nature of Existence II: 9), McTaggart announced that he would put forward a new reason for holding, like Spinoza, Kant, and Hegel, that time is unreal. Now, from the contradictoriness of an analysis or a theory follows that it has no model, that is, nothing in the world corresponds to it. But it does not

78

DEBATES IN THE METAPHYSICS OF TIME

follow that the phenomenon with which the analysis or theory is concerned does not exist. Rather, the analysis or the theory might be inadequate or wrong. This conclusion is more reasonable in the case of time since temporal phenomena are crucial and ubiquitous in our world. Moreover, as C. D. Broad and others have emphasized, temporal succession and duration are distinctly given to us in introspection and perception and can therefore hardly be doubted.7 Of course, there are ontological analyses and theories of time which are free from contradictions. However, McTaggart rejects non-presentist ontologies of time. Are there then presentist ontologies of time which are consistent? Indeed there are, for a long time: Aristotle’s, for example. Aristotle does not run into the contradictions McTaggart uses to argue against the reality of time because he holds that only the respective present exists. That excludes the movement of a persistent present. McTaggart did not really take into account the central tenet of presentism (that only the present exists) although he is a presentist in so far as he insists that tense is essential for time.

An implicit moving present McTaggart works towards a moving present and needs a non-relative present for an argument against time. But there is also an ordinary and a scientific view of temporal succession on a larger scale which implies a moving and non-relative present. I mean the concept of development, including the development of the whole universe. Compare two statements: 1. “The development of the universe reached stage s,” and 2. “The development of the universe reached stage s at time t.” The second statement relates the stage to time. We have an aversion against the time relativity of developmental stages and we prefer the non-relative statement 1. We have a general aversion to the temporal relativization of attributions. For example, we would not want to say that a certain leaf on a tree is yellow at the present date and we would not even say that it is yellow now. Rather, we would insist that the leaf is simply yellow. We insist although we know very well that the leaf has not been yellow and will no longer be yellow and that it is only temporarily yellow. From an ontological point of view, I would agree that attributions are not time relative. Relativity of attributions to time would turn properties into relations and would not even be possible in an ontology without time points. However, if attributions are taken to be of persistent and changing particulars, as is customary and common sense, then past and future attributions of



Temporal Succession and Tense

79

particulars have to be invalidated by McTaggart’s moving present or by a structurally equivalent entity. Otherwise, Parmenides’ contradiction of change threatens. What the persistent moving present would do (if it existed) is to set off exactly one member of a temporal series and suppress all the other members. That is how a competition and contradiction with the other members is avoided and, it is purported that the ultimately contradictory attribution of incompatible tenses to the persistent and changing particular is all right. The ontologically tenable solution would be to attribute not to the changing particular but to its temporal parts. However, neither commonsensical thinking about attributions nor the scientific concept of development is meant to involve a moving present and a crucial role of the tenses. It is just an unnoticed implication. Moreover, scientists applying the concept of development clearly think of the temporal phenomenon indicated not in terms of a passage of time but in terms of a temporal succession. They should realize that there is no outstanding (present) stage of development and that all stages are equally related to time. For the relationalist, with respect to time, that means they are simultaneous to some event of time measurement. The insight to be acquired with respect to the concept of development is that there is no absolute stage reached but only a stage related to a certain date, a stage at a certain date. A development is, of course, also a temporal succession.

McTaggart’s mistakes and his fundamental error The customary view of the development of the world involves the same fundamental mistake as was made by McTaggart in his attempt to improve Russell. He wanted to mobilize Russell’s temporal sequence by running a present along it. If one analyzes what McTaggart proposes completely, one arrives merely at another temporal sequence. I mean the temporal sequence of being present of the members of the Russellian sequence a1, a2, a3, where a1 is earlier than a2 and a3, and a2 is earlier than a3. McTaggart pretends to mobilize that sequence by another Russellian sequence, namely by p(a1), p(a2), p(a3), where “p“ stands for “being present.“ The mobilizing Russellian sequence would then be the fact that p(a1) is earlier than p(a2) and p(a3), and that p(a2) is earlier than p(a3). McTaggart does not realize that he offers nothing but another sequence to mobilize a sequence because he compares what does not match, namely the Russell ontological analysis with his own indications concerning the movement of the present which he avoids transforming into an explicit ontological analysis. He argues that Russell’s sequence is not temporal because it does not change. However, as

80

DEBATES IN THE METAPHYSICS OF TIME

soon as his indications are turned into ontological analysis his sequence does not change either. McTaggart also misunderstands Russell’s characterization of the relation of temporal sequence as an order relation. He relies on the customary opposition of order and dynamism, although Russell defines an order relation as a relation which satisfies the conditions of asymmetry and transitivity. That is in no way incompatible with being dynamic. The view I called the customary non-presentist view is, of course, not due to a misunderstanding of Russell’s temporal sequence. It may be due to an attempt to re-enact in the mind the development of the world by moving mentally along the sequence of its stages. Nevertheless, this attempt commits the same mistake as McTaggart. There are two reasons why I consider McTaggart’s view of temporal succession mistaken. First, the impression of dynamism of the A-series which McTaggart emphasizes arises from incomplete analysis. The sequence of A-determinations is not analyzed. If it were analyzed, it would turn out to be a B-series. Second, McTaggart commits what I would call the discursive mistake. Let me explain. When we want to grasp a series we mentally run through it, maybe also physically. One could call it the method of discursion. We apply it not only to temporal series but to series of all kinds. Naturally, the running through is also a series. It is a temporal series and it is clearly different from the series to be grasped. It would be a mistake to identify the two series and it would be just the mistake I called “the discursive mistake.” McTaggart’s mistakes are ultimately rooted in his error concerning the essence of time. It is not only a misunderstanding of Russell’s temporal series but more deeply the misunderstanding of temporal succession as a change. In his criticism, McTaggart uses the premise that the mark of time is change. This premise is fundamentally false. Being subject to change is not a hallmark of the temporal as McTaggart presupposes. On the contrary, temporal attributes are the exception. Temporal attributes are the basis of change and therefore must not change themselves. An ontological analysis of temporal change reduces it to an absurd relating of time to itself and reveals that temporal change does not exist. It can also be shown that McTaggart’s moving of the present, if analyzed ontologically, turns out not to be a change.8 One can understand Aristotle’s as well as MacTaggart’s as well as the customary implicit mobilization of succession by means of tense as a misguided attempt to view temporal succession as a change. Tellingly, the contradictions arising from the transitions of presentness which McTaggart uses to discredit time are after all merely a case of Parmenides’ contradictions of change (which contemporary analytic philosophers attribute to David Lewis).



Temporal Succession and Tense

81

A Russellian temporal series According to his principle of acquaintance (which demands all entities assumed in an ontology must be empirically given), Russell introduces his temporal relations ostensibly by examples. These examples show that those are dynamic relations if anything is dynamic. Imagine someone is practicing the musical scale. First a c-tone, then a d-tone and then an e-tone sounds. What we hear, according to Russell, when we hear the c-tone preceding the d-tone is the relational universals of “occurring earlier than” together with its relata. We hear nothing else. Let us assume that we don’t recognize the first tone as a c and the second as a d. Thus we hear only a temporal fact which as such is a dynamic fact which analyzes a case of the phenomenon of succession. Since it consists in addition to the tones of a relational universal, this relational universal must be the dynamic constituent. If the fact is dynamic, which one can take for granted, the relational universal in it must be dynamic, too. Now, Russell introduces the relational universal as the one which holds between the two tones in the fact of our example. One can conclude that the relation “occurring earlier than” is clearly designed to analyze the phenomenon of temporal succession. Let the three tones succeed each other so quickly that one hears that the c-tone occurs earlier than the d-tone and earlier than the e-tone, and (without having to resort to memory) that the d-tone occurs earlier than the e-tone. Thus one hears the conjunctive fact E(c, d) & E(c, e) & E(d, e). Universals, relational universals, facts, and conjunctive facts are categories of Russell’s ontology. It is an application of these categories to categorize what is heard in our example as a conjunctive fact. Since there is a law for E according to which it fulfills the conditions of asymmetry and transitivity, the conjunctive fact is a series—a Russellian temporal series. Russell’s ontological analysis of the successions of the whole universe is that it is such a conjunctive fact with a very great number of conjuncts. Now, the succession is wholly based on the E-relation. Looking again at the example by which E was introduced supports this ontological view. There is no point in running through a Russellian E-series because it results in exactly the same series. Thus the discursive mistake is transparent. It shows itself to be a mistake immediately. To see that the E-relation is sufficient to ground the succession, one has to take into account that the Russellian view of time is relational. According to this view, time is nothing but a group of relations. It is directed against the absolutist view, mentioned already, that time is a continuum of time points. If time is viewed as a continuum of time points, a mobilization, a running through that continuum may seem necessary to do justice to the

82

DEBATES IN THE METAPHYSICS OF TIME

phenomenon of temporal advancement. Strictly speaking it is not. One has merely to introduce relations between the time points.

The 4-D view Mostly, the Russellian view is equated with the 4-D view. But there are fundamental differences. The former offers an explicit ontological analysis and is based on a complete ontology. The latter is hardly connected with an ontology. The former is based on a relational analysis of time; the latter implies the opposite, namely an absolutist view of time. An influential advocate of a kind of 4-D view was D. C. Williams with his paper “The myth of passage.” In his analyses of temporal successions, Williams aligns himself with the mathematical theory of manifold which takes space and time to be a set of points. Time is understood as the fourth dimension of such a point space.9 Concrete entities get their temporal determination by being localized in the fourth dimension. Our example of the three tones, c, d, and e, would be analyzed by Williams’ 4-D view by localizing c at a time point t1, d at a time point t2, and e at a time point t3. No temporal relation between the tones is called into play. A relation between the time points which orders them is, however, involved. It is not a universal, of course, but a set of ordered pairs. Hence the 4-D analysis is markedly different from the Russellian. Moreover, in contrast to the latter, the 4-D analysis does not attach great importance to the accordance with what is given to us in perception. It assumes time points but, obviously, time points are not given to us in perception. All perceptions of temporal locations are relative to some anchor object such as a time signal. Hence, a relation which would order time points according to the 4-D theorist could not be given to us in perception either. McTaggart diagnosed a Russellian E-series as non-temporal, as merely extensive and not transitory (to use C. D. Broad’s terms). That was a misunderstanding. But his diagnosis would be true of the 4-D view as advocated by Williams who writes that “Time ‘flows’ only in the sense in which a line flows or a landscape ‘recedes into the west’. That is, it is an ordered extension. And each of us proceeds through time only as a fence proceeds across a farm: that is, parts of its being, and the fence’s, occupy successive instants and points, respectively.”10 By contrast, the Russellian relation E is clearly a relation of transition which is temporal and definitely not spatial. It does definitely not hold between the intervals of a fence. E is transitory in a sense in which only temporal relations can be transitory.



Temporal Succession and Tense

83

Mathematics and ontology There is also a methodological difference between the 4-D and the Russellian view: the former applies a mathematical theory (especially set theory and general topology), the latter an ontological one (Russell’s logically atomistic ontology). From an ontological point of view, several of Williams’ positions and solutions are unsatisfactory. He adopts the customary conception of becoming (that is emergence or passing away) as a transition between non-existence and existence. He writes (to quote more extensively what was referred to in part above): Time “flows” only in the sense in which a line flows or a landscape “recedes into the west”. That is, as an ordered extension. And each of us proceeds through time as a fence proceeds across a farm: that is, parts of our being, and the fences occupy successive instants and points, respectively. There is passage, but it is nothing extra. It is the mere happening of things, their existence strung along in the manifold. (Williams 1951: 105) By “theory of the manifold” Williams means “set theory,” I suppose. What he then states is that the reality behind the myth of passage is nothing but what the set theorists call “serial order.” It entails that all serial orders exhibit a passage. His phrase “their existence strung along in the manifold” implies that in the case of a temporal passage where the manifold is an ordered temporal extension the date of a thing (i.e. the respective time interval in which it is located) is its span of existence and that it does not exist outside of the interval. It implies also that in Williams’ sense of “passage” there is a passage into existence and passage out of existence. It seems that Williams could not escape completely the deeply rooted customary presentist thinking, though he explicitly rejects presentism. The presentist leftover still leads to the contradictions of becoming as transition between non-existence and existence which cannot be resolved by the introduction of temporal parts since before the emergence and after the passing away of an object there is nothing of which temporal parts can be temporal parts.11 Another case in point is Williams’ analysis of the sense of a series.12 He wants to defend the 4-D view against the objection from Broad that it cannot ground the direction of time. He argues that it can easily be grounded by the ordered pairs and that to see the directions of the ordered pairs a;z and z;a one has merely to take into account the numerical identity and diversity of terms. However, while the spatial order of the symbols on paper is clear, that of the particulars they represent is a mystery. Ordered pairs are not a

84

DEBATES IN THE METAPHYSICS OF TIME

philosophical paradigm, as Quine suggests, but rather a gap in the explanatory structure of set theory. They are no ordinary sets. That is why some set theorists (Kuratowski and Wiener) wanted to reduce them to ordinary sets. At any rate, ordered pairs do not satisfy the standards of articulateness required in ontology.13 Russell assumes in his ontological analysis of order in relational facts order relations such as “being the first relatum” and “being the second relatum.” These relations hold between relata and facts. As was mentioned already, series are analyzed ontologically as conjunctive facts in a Russellian ontology. Now, the direction of a whole series is aggregated from the order relations for all the relational facts of which the respective conjunctive fact consists. From Russell’s analysis of the direction of a series follows that Williams is right to claim that all series have directions. However, that implies that the direction of a temporal series cannot explain the transitory character. That is done in Russell’s ontological analysis by the specific temporal relation E. One cannot argue, of course, that ontology is always preferable to mathematics. It depends on the problem to be solved. With respect to such a simple and fundamental problem as the explanation of the phenomenon of temporal succession, ontology seems to me the appropriate science.

Notes   1 The temporal sense of “present” has to be distinguished from the spatial sense. It is common knowledge today that we can be sure about the temporal presence of the spatially present (i.e., near) only.   2 Aristotle (1987): 218a.   3 Ibid.   4 As I argued in 1997 and in 1999.   5 McTaggart (1927: 19f).   6 Loc. cit.   7 See Broad (1921).   8 See also Tegtmeier (2012).   9 The 4-D view goes back to Minkowski’s representation of Einstein’s Special Theory of Relativity in terms of a spacetime. The Minkowski space has in contrast to the three-dimensional Euclidean space a fourth time-like dimension. 10 Williams (1951: 105). 11 See Tegtmeier (1999). 12 Williams (1951: 108). 13 See Tegtmeier (1995).



Temporal Succession and Tense

85

References Aristotle (1987), “Physics,” in J. L. Ackrill (ed.), A New Aristotle Reader. Princeton, NJ: Princeton University Press. Broad, C. D. (1921), “Time,” in J. Hastings et al. (eds), Encyclopedia of Religion and Ethics. Vol. 12. Edinburgh and New York: T. & T. Clark and Scribners, pp. 334–45. http://www.ditext.com/broad/time/timeframe.html McTaggart, J. M. E. (1927), The Nature of Existence II. Cambridge: The University of Cambridge Press. Tegtmeier, E. (1995), “Ein vernachlässigtes ontologisches Problem der Relationslogik,” in J. Brandl (ed.), Metaphysik. Neue Zugänge zu alten Fragen. Sankt Augustin: Academia, pp. 83–95. —(1997), Zeit und Existenz. Tübingen: Mohr. —(1999), “Parmenides’ problem of becoming and its solution,” Logical Analysis and Philosophy of History, 2, 51–66. —(2012), “McTaggart’s error: temporal change,” Revue Roumaine de Philosophie, 56, 89–96. Williams, D. C. (1951), “The myth of passage,” Journal of Philosophy 48 (15), 457–72, reprinted in R. Gale (ed.) (1967), The Philosophy of Time. New York: Anchor Books, pp. 98–116; ref. to this repr.

5 Becoming: Temporal, Absolute, and Atemporal M. Oreste Fiocco

Introduction

T

here are two conspicuous and inescapable features of this world in which time is real. One experiences a world in flux, a transient world in which things constantly come into existence, change and cease to be. One also experiences a stable world, one in which how things are at any given moment is permanent, unchangeable. Thus, one can contemplate in silence—then be startled by a flash and accompanying boom—then return to silence, and although the flash and boom are gone, it seems indubitable that something remains unchanged, at least in so far as it must be true that a flash and boom precede this silent moment. There is transience and permanence. Yet these two features of the world seem incompatible. However, focusing on one can yield only an objectionable metaphysics of time to the extent that the other, itself a compelling feature of the world, is neglected. The primary purpose of this chapter is to sketch a metaphysics of time that embraces both features. Given a certain view of the nature of reality and of the structure it contains, from a basis of uncontroversial claims about time and change, I show that utter stasis and continuous dynamism can both be genuine and objective features of reality. Crucial to this undertaking is the notion of becoming, that is, coming into existence. I distinguish three distinct phenomena of becoming: temporal, absolute, and atemporal. The last is the least familiar of these; it is the phenomenon of coming into existence outside of time. Although the idea that there are things that do not exist in time is not unfamiliar, it is largely taken for

88

DEBATES IN THE METAPHYSICS OF TIME

granted that if anything comes into existence, it must do so in time. Indeed, I suspect many think that the idea of a thing coming to be outside of time is simply incoherent. It would be if becoming were a process—but it is not. In this chapter, I articulate and defend the notion of atemporal becoming. It is by means of this notion that one can develop a fully satisfactory metaphysics of time, one that honors both transience and permanence and finds for each its proper domain within the world. A discussion of the fundamentals of the metaphysics of time requires an explicit account of what time itself is. So I begin with such. Making clear the nature of time per se enables me to distinguish time from temporal reality and to present the two generic positions regarding temporal reality. I then propound an account of atemporal reality, the world outside of time. Having laid this foundation, I consider the undeniable phenomenon of temporal becoming, which underlies the flux in the world, and discuss attempts to situate this phenomenon in different accounts of temporal reality. I maintain that the only feasible way of doing so is by recognizing absolute becoming. Absolute becoming, however, seems to compromise the patent stability of the world. This leads to the key notion of atemporal becoming. After illuminating this phenomenon, I defend its place in a fully satisfactory metaphysics of time.

Time and temporal reality As a prologue, I make a brief statement of the view of the world in itself— reality as it is prior to any conceptualization by thinking beings or their linguistic or social interactions—that underlies the present discussion. Although I believe there are strong reasons for accepting this view, I leave its defense to another occasion; it is feasible enough for present purposes. I believe that the world in itself contains a great variety of individual substances, instances of genuine kinds of thing, and ways (both universal and particular) that these things are. Each entity has a nature determined by what it is and stands in relations to entities of the same and other categories. These relations and the constraints thereby placed on their relata is the structure in reality. All features of the world have an ontological basis in this structure and, in so far as they admit of explanation, are ultimately explicable in terms of it. This picture is robustly realist, then, about distinct categories (including substances, kinds, properties, and modes), the structure in the world and explanation. A metaphysics of time is a theory of the peculiarly temporal features of the world, those that arise given that time is real. I maintain that a distinction between the world in time and the world outside it is crucial to a fully



Becoming: Temporal, Absolute, and Atemporal

89

satisfactory metaphysics of time. In order to get purchase on this distinction, one needs an account of the nature of time itself. With such an account, one can articulate the relations between time and other entities—and see past the mere spatial connotations of “in” and “outside”—to acquire a better sense of the worldly structure associated with temporality.

What time is Discussions of time are contentious. In so far as there is here a subject, though, there is some common ground from which discussion arises, certain phenomena that motivate inquiry and are thought to go together. The most conspicuous phenomenon associated with time is change. Change is an incontestable feature of the world. Moreover, what change is is uncontroversial: an entity changes if and only if it in itself is one way at one moment and an incompatible way at a distinct moment. Everyone who recognizes change can accept this account; it is neutral on any substantive issue. Change is thought to require time. Although there has been debate regarding whether there could be time without change—and some accept there could1—no one maintains that there could be change without time. Thus, that change requires time seems a truism. I suspect some accept it because, in light of the foregoing account of change, they accept that there could be no moments without time; or perhaps they just conflate time, itself, with times, that is, moments. The former is more insightful. Regardless, however, of why the truism is accepted, it leaves open what time itself is. Despite interest in time throughout the history of Western philosophy, there is very little discussion of time per se. This claim is perhaps surprising, but when one recognizes that investigations pertaining to time tend to focus on issues attendant on time—like change, becoming, tense, persistence, temporal experience—rather than time itself, one can see its truth. Some ecumenical account of time is needed, then; some account that illuminates these issues and unites the different factions all of whom take themselves to be investigating the metaphysics of time. In light of these considerations, I provide an account of time per se: I submit that time is a thing, namely the thing in virtue of which any entity changes.2 An entity might have, by its very nature, the capacity to change—but without time itself it could not change. Time is, therefore, the thing that makes change possible. It does so by yielding the moments required for change. Time itself is distinct from any moment or collection of moments. This brief account of time itself has at least two benefits. It makes explicit the connection whereby change requires time and it also provides the ontological basis of a phenomenon, to wit, change, the complexity of which

90

DEBATES IN THE METAPHYSICS OF TIME

seems to be ill-suited to be taken as primitive. There is, in the structure of the world, some explanation of how change occurs and this explanation is rooted in time. Time itself is simple and essentially existent; by its very nature it must exist.3 Some might object to this account of time itself because of the hypostasis it requires. One should recognize, however, that this sort of objection is based on ontological scruples, pertaining to parsimony or perhaps an aversion to abstracta, rather than on any consideration arising from the metaphysics of time itself. Indeed, there is nothing objectionable about this account from the perspective of the metaphysics of time; it is wholly neutral in regards to any controversial issue. Therefore, in an attempt to propound a fully satisfactory metaphysics of time, I consider the proposed account of time per se to be not only acceptable but also illuminating.

What temporal reality is: The world in time If time is the thing that makes change possible, there could be many entities that exist in virtue of time, for example, any moment that exists (or has any being whatsoever); any properties that are borne primarily by moments (e.g. being present, being past, being future) and any relations that are borne primarily by moments (e.g. earlier than, later than, simultaneous with). Moreover, there are things that do not exist in virtue of time yet are intimately related to it in that they require a moment at which to exist, for example, any entity whose very nature includes the capacity to change. Such a thing is a temporal entity and exists at a moment if it exists at all. Temporal reality is all that exists in virtue of time and every temporal entity—it is the world in time. Note that despite being included in temporal reality, a moment itself is not a temporal entity, for no moment exists at a moment and so is not a thing that could change. Time itself is neither a temporal entity, for it exists at no moment, nor included in temporal reality, for nothing, including time, exists in virtue of itself. I think the existence of time, as characterized above, is highly plausible (if not beyond dispute) and that it is this notion of time that those who accept that time is real accept. Similarly, I think the existence of temporal reality is highly plausible (if not beyond dispute). Anyone who bothers with a critical examination of time, then, recognizes both time and temporal reality. However, one needs to recognize the distinction between the two: time is a particular entity; temporal reality is a collection of entities (including concrete particulars, moments, temporal properties, temporal relations) essentially related to time. So with these ecumenical notions of time, temporal reality— and change—there is a good deal of common ground in the metaphysics of time.



Becoming: Temporal, Absolute, and Atemporal

91

Yet, as noted above, discussions of time are contentious. One might accept the existence of both time and temporal reality as characterized and yet disagree about the extent or contents of temporal reality. For example, the metaphysics of time that I ultimately propose herein is one in which no more than a single moment of time ever exists (in any sense) and only the temporal property being present and the temporal relation simultaneous with are instantiated. Such a view is determined by a number of controversial considerations. I do not think it an exaggeration, however, to maintain that the source of all the contention in the metaphysics of time is a single ontological issue: whether things come into or go out of being simpliciter. Disagreement on this issue leads to two generic positions regarding the nature of temporal reality. Those who deny that a thing comes into or goes out of being simpliciter believe that at no moment does anything that exists in time absolutely fail to be a constituent of reality. This is not to say that in this view each thing always exists, for there can indeed be moments at which a thing does not exist. However, even at these moments, that thing EXISTS (tenselessly) relative to some other moment. Each thing has a permanent, tenseless existence at any moment at which it ever exists. This leads to a position regarding temporal reality in which all the many moments of time—and everything that exists at them—are equally real; it is a view in which temporal reality is ontologically homogeneous. Proponents of this position are called B-theorists, eternalists, block theorists and, somewhat misleadingly, tenseless theorists.4 Those who believe that a thing comes into or goes out of being simpliciter accept that something, which in no way exists, can come into being at a moment or that a thing can cease to be in every way and, hence, bear no properties or stand in any relation to anything (including any moment). This leads to a view of temporal reality in which there are ontological differences in the world in time. Given that this position is based on difference, it admits of a variety of more specific views, the unifying feature of which is the supposition that temporal reality is ontologically heterogeneous. For example, the view that there is but one moment and nothing before or after it; the view that there are many moments of time with different ontological statuses; the view that this moment and what is prior to it have the same ontological status though what is subsequent to this moment does not exist at all, all posit ontological heterogeneity in temporal reality. Presentists, A-theorists, growing block theorists and, perhaps, tensed theorists5 represent this generic position on temporal reality. In both positions, a thing exists at the moments at which it exists. The key distinction between the two is the ontological status of a thing at some moment, m, at which it does not exist. In the former position, although it does not exist at m, it does EXIST (tenselessly) at some other moment;

92

DEBATES IN THE METAPHYSICS OF TIME

in the latter position, at m, it does not exist at all, in any way—it is not a constituent of reality. In light of this pivotal difference, it is perhaps clear what would motivate one to adopt the position that temporal reality is ontologically homogeneous: one’s experience of a stable world, in which how things are at any given moment is permanent and unchangeable. Likewise, the motivation for the position that temporal reality is ontologically heterogeneous is one’s experience of a world continuously in flux, in which how things are at any given moment is transient. These two positions regarding the nature of temporal reality are incompatible. The position one adopts presumably turns on whether one regards permanence or transience as the dominant feature of a world in which time is real. But since both are irrefragable features of one’s experience, to neglect either can lead only to an objectionable metaphysics of time. Fortunately, the appearance of a dilemma here is based on a false assumption. This is the assumption that how the world is, given that time is real, must be accounted for entirely by an account of temporal reality. This is false. There is more to the world than the world in time—what more there is provides the means of presenting a fully satisfactory account of a world in which time is real.

Atemporality I have discussed the notions of time and of temporal reality in some detail in order to make clear the notion of atemporality, of existing outside of time, that is, without temporal reality. As mentioned above, this notion is, I believe, of the utmost importance to a satisfactory metaphysics of time.

The world outside time A straightforward account of atemporality emerges from the foregoing. If a thing is in time, then it is related to time in a suitably intimate way. One such way is to be related as to partake of the very nature of time. Since time is just the thing that makes change possible, doing so by yielding moments, it is plausible that what it is to partake of the very nature of time is to exist at a moment. Hence, existence in time is existence at some moment. Every temporal entity, that is, every entity whose very nature includes the capacity to change, exists in time. Since everything is related to everything else in some way (if in no other, at least in being similar with respect to being real), a thing outside of time does not fail to be related to time. Rather, such a thing is outside of time in that it fails to be related to time in a suitably intimate way; it fails to partake of the



Becoming: Temporal, Absolute, and Atemporal

93

very nature of time. Since, again, time is just the thing that makes change possible—by yielding moments—a thing that fails to exist at a moment clearly fails to partake of time’s nature. What it is, then, to not exist in time, to exist outside of time, is to exist but at no moment. As such, an entity that exists outside of time—an atemporal entity—fails to meet a necessary condition of change and so cannot change. An atemporal entity is just as real as a temporal one, but given their respective relations to time, the former is immutable whereas the latter is mutable.

An attempt to reject atemporality altogether One might reject this notion of timelessness, or any other, because one might maintain that all entities are capable of changing in some way and, hence, exist at some moment. If this is correct, there simply are no atemporal entities and, a fortiori, no atemporality (and no atemporal becoming). The claim that every entity is capable of changing is plausible only if one counts as change the gain (and corresponding loss) of any property contrary to one a thing has at some moment. If one were to do this, one would hold that when Xantippe goes from being wife to widow upon the death of Socrates that she has undergone a genuine change. This, however, does not accord with the uncontroversial account of change introduced above, for the properties being a wife and being a widow do not characterize Xantippe in herself. The putative change that Xantippe undergoes upon the death of Socrates needs to be distinguished from the sort of change Xantippe undergoes when, say, she stands up after sitting. In the latter case, the change arises from the gain of a property that is incompatible with one that Xantippe has in virtue of how she herself is. In the former case, the change does not arise in virtue of Xantippe herself; rather, in a clear sense, the “change”—so-called “Cambridge change” of a “Cambridge property”6—has crucially to do with something entirely distinct. In this context, one in which the bone of contention is the mutability of all things—whether it is compatible with the nature of each thing to change—the latter notion of change, the uncontroversial one presented above, is clearly the operative one.7 Bearing this notion in mind, there are some things—for example, numbers, properties, propositions, moments themselves—that, in so far as their natures are understood at all, seem to be incapable of change. It certainly seems that things of these kinds are not one way at one moment, then an incompatible way at another in virtue of how they themselves are. Indeed, I am aware of no consideration that even suggests that such things can undergo genuine change.8 In what follows, then, I take for granted that there are atemporal entities, things that exist without temporal reality, as real as anything else, but not at any moment.

94

DEBATES IN THE METAPHYSICS OF TIME

Two views of timelessness (neither of which is adequate) So there is no good reason to think that everything must or can change.9 Nevertheless, one might still resist the proposed account of atemporality. There are traditionally two notions of timelessness. In one, an atemporal thing is eternal: it is supposed to exist outside of time, that is, at no moment. In the other, a timeless thing is sempiternal: it exists unchanging at each moment. Thus, one might acknowledge that there are immutable things, but maintain that such things still exist in time in the sense provided above: they exist unchanging at every moment. The capacity to change, however, is such a basic feature of an entity that it is plausibly regarded as partially definitive of a thing’s nature and so, in this way, essential to anything that has it. Thus, it is of the very nature of a mutable thing to be capable of changing, in that it is not possible for that very thing to exist as the very thing it is and yet lack the capacity to change. A mutable thing must be a temporal thing in that it must exist at a moment if it exists at all. Similarly, for those things that are incapable of changing, it is of their very nature to be incapable of changing. Nothing mutable is possibly immutable or vice versa; a fortiori, nothing mutable becomes immutable (or vice versa). Temporal and atemporal entities are necessarily mutually exclusive types of entity. This is important for it provides grounds for rejecting the claim that atemporal entities are sempiternal, “timeless” yet existing at moments. If one assumes that an entity that exists at a moment is ipso facto susceptible to change, then it follows immediately that no atemporal entity can exist at a moment. If one forgoes this assumption, alternative considerations support the same conclusion. If an atemporal entity is essentially immutable, it is not immutable because it exists outside of time. Rather, it fails to exist at a moment because, by its very nature, it is not susceptible to change. If, however, one assumes that an atemporal entity exists unchanging at every moment, one is left to account for why it exists at a moment if it cannot change. The assumption that it exists in time is entirely gratuitous. The simpler view, in that it does not require one to account for what would be a perplexing feature of atemporal entities, is the one in which atemporal entities just do not, by their very nature, exist at moments. Perhaps those who accept sempiternal entities maintain that such things exist among the temporal entities—that is, at moments—because they see no other way to understand timeless structure in the world. But the preceding account of time itself provides the means to do so: timeless things exist, yet at no moment(s), for they are not related to time in a suitably intimate way. There is, then, little motivation for maintaining that an immutable, atemporal entity exists at all moments. This does not mean, however, that I



Becoming: Temporal, Absolute, and Atemporal

95

think the traditional notion of eternal existence provides an accurate account of atemporality. This notion brings with it the idea of essential existence, in that an eternal entity is supposed to be one that must exist by its very nature. It would be inappropriate to say that such a thing has always existed, for this is to characterize an atemporal entity with a temporal notion. But it is apt to say that such a thing has no origin; it simply has to be. I want to clearly distinguish essential existence (and, thus, eternal existence) from timeless existence, for I believe there are things that are timeless that nevertheless have origins. Such things do not exist essentially, yet they exist, so they do, in a literal sense, become. However, they come into existence outside of time. Each is an example of atemporal becoming.

Becoming: Temporal and absolute I fully admit that the notion of atemporal becoming likely sounds incoherent. The purpose of this chapter, though, is to show that not only is it coherent, but that atemporal becoming plays an important role in the metaphysics of time. Before expounding this phenomenon, it is helpful to discuss a vexed, but more familiar, one, namely temporal becoming. As observed at the outset, one of the conspicuous and inescapable features of a world in which time is real is a certain dynamism: one experiences a world in flux in which things constantly come into existence, change, and cease to be. Temporal becoming is standardly supposed to underlie this dynamism. An adequate account of this phenomenon makes clear the need for atemporal becoming.

Dynamism: Change, temporal becoming, and novelty Change is indubitable. Since there can be no change without dynamism, both positions regarding the nature of temporal reality—that it is ontologically homogeneous and that it is ontologically heterogeneous—must provide some account of temporal becoming. In so far as temporal becoming is a real phenomenon, it has an ontological basis among the things that exist and the relations in which they stand. However, no thing is in itself dynamic, and no entity in isolation of every other thing changes. Certainly no property itself changes, nor any moment; even a paradigmatic mutable entity, a concrete individual substance, which obviously has the capacity to change, is not in itself dynamic. The dynamism in the world comes not from any thing, but arises from the relations in which entities stand. (Recall that in the operative account of change, change occurs when a substance bears incompatible

96

DEBATES IN THE METAPHYSICS OF TIME

properties at distinct moments.) Therefore, dynamism and the phenomenon of temporal becoming that underlies it are structural features of the world. This is apparent when one recognizes that temporal becoming is the coming into being of some entity at a moment. Since change requires a temporal entity, assuming (as seems plausible) that no mutable, that is, temporal, entity is essentially existent, it follows that change requires temporal becoming. If change is the mark of dynamism, this shows why temporal becoming is supposed to underlie the dynamism in the world. As intimated by the discussion above, incompatible accounts of temporal becoming can be regarded as definitive of the two opposing positions regarding the nature of temporal reality. In one position, at any moment before a thing comes into existence at moment m, it EXISTS10 (tenselessly) at m; in the other, before a thing comes into existence at a moment, it is in no sense a constituent of reality. Both positions, then, provide some account of temporal becoming; moreover, partisans of both positions can accept the very same notion of change. However, in considering the dynamism in the world, one must consider more than merely temporal becoming and change. It is universally acknowledged—embraced by partisans of both positions on the nature of temporal reality—that there is an especial impression of novelty arising from being in a world in which time is real. This impression of novelty has traditionally been characterized as a sense of flow or passage. Thus, D. C. Williams asserts that we are “immediately and poignantly involved in the jerk and whoosh of process, the felt flow of one moment into the next,”11 and J. J. C. Smart maintains that “certainly we feel that time flows.”12 Tim Maudlin states that there is a “manifest fact that the world is given to us as changing, and time as passing.”13 Bradford Skow characterizes the impression by observing that “Of all the experiences I will ever have, some of them are special. Those are the ones that I am having NOW. All those others are ghostly and insubstantial,”14 and Laurie Paul observes that “I … feel the cool breeze on my face. I feel the freshness of the cool breeze now, and, as the breeze dies down, I notice that time is passing.”15 This impression of novelty is crucially associated with the dynamism in time. A full account of this dynamism and, hence, a fully satisfactory metaphysics of time, needs to include not only accounts of temporal becoming and change, but also some account of this special sort of novelty. Although partisans of both positions regarding the nature of temporal reality agree that there appears to be a further dynamic feature of reality, they disagree on what is needed to account adequately for it. It is one of the most disputatious points in the metaphysics of time whether the position that temporal reality is ontologically homogeneous has the means to do so. A pivotal question, then, is how one who thinks that the world in time is ontologically homogeneous



Becoming: Temporal, Absolute, and Atemporal

97

accounts for the impression of novelty, the phenomenon that, combined with temporal becoming and change, yields the experience of a world in flux.

Novelty and ontological homogeneity of temporal reality If there are many moments of time and everything in temporal reality—all the many moments and anything that exists at any one of them—has the same ontological status, nothing in time is genuinely novel. Every mutable thing EXISTS (tenselessly) and is always a constituent of reality, in the sense that at every moment each temporal entity EXISTS at some moment. Since there is nothing novel in reality itself, one must account for the impression of novelty by citing some interaction between a thinking being and a permanent feature of the world. Any novelty must be projected onto the world by one’s subjective experience of it, presumably as one encounters a permanent thing or moment for the first time. Therefore, one must account for novelty in terms of the changes in thinking beings who EXIST (tenselessly) at moments. As is clear from their proposals to account for the impression of novelty, this point is recognized by proponents of the ontological homogeneity of temporal reality. J. J. C. Smart suggests that it is one’s confusion regarding how certain predicates in natural language (like “is past”, “is present,” and “is future”) work that is the basis of this novelty.16 D. H. Mellor suggests that it is one’s different beliefs at different moments that is the subjective “truth in the metaphysical falsehood that time flows.”17 In the same vein, Laurie Paul suggests that the impression of novelty arises from “the way brains of conscious beings experience and interpret cognitive inputs from series of static events.”18 But this approach does not seem feasible. Grant that it is some mental (or neuro-physiological) state of a thinking being that is the basis of the impression of novelty. This impression is distinctive in that it marks a particular moment as special (as novel ). It can only do this, though, if no other moment is marked in the same way. However, assuming temporal reality is ontologically homogeneous, nothing in time is any more or any less real than anything else. This includes not only every thinking being, but also every mental state of every thinking being. If this is so, and the impression of novelty is inexorable, then at every moment at which one exists, one has the impression that that moment is novel. In which case, for any thinking being there are many moments marked as novel; indeed every moment at which one exists presents itself as novel. Thus, the mental state underlying the impression of novelty cannot be the source of one’s sense that there is something special about this moment, now; for one has the same state (or a relevantly similar one) at every moment at which one EXISTS, and one EXISTS

98

DEBATES IN THE METAPHYSICS OF TIME

at many moments. An explanation of one’s sense of the novelty of a particular moment—this one, now—is still needed. The problem here indicates a much deeper problem for the position that temporal reality is ontologically homogeneous. The position does not present an accurate account of how one experiences a world in which time is real. Consider again Paul’s account of the sense that time passes. At moment m1, one feels the freshness of a cool breeze; at m2, the breeze has died down to the extent that one no longer feels it. If both moments of time are equally real, one both FEELS a cool breeze (at m1) and FEELS no breeze (at m2). One does not feel the breeze and feel no breeze simultaneously; nevertheless, both experiences are as real and so should be equally arresting. But, of course, one HAS no such odd and conflicting experience of a breeze and no breeze. Nota bene (and this point cannot be stressed enough): It makes no difference that one experiences the cool breeze at m1 and feels no breeze at m2, for both moments are equally real and one is just as much at one moment, experiencing the cool breeze, as one is at the other, experiencing no breeze.19 (Nor does it make any difference if it is merely a temporal part of one at each moment, if a temporal part of a thinking being is sufficient for that thinking being to be at and experiencing a moment.) Yet at most a single moment is ever salient to one (in the familiar way). Even if one posits some sort of mental (or neurophysiological) mechanism that at each moment dampens one’s experience of every other moment, one must account for the fact that one experiences each equally real moment, in order, as if new. So the need for an account of the basis of the impression of novelty is urgent, but given that in this position, such an account can only be in terms of the states of thinking beings in time, one does not seem forthcoming. Therefore, it seems that if one accepts that temporal reality is ontologically homogeneous, one must posit something extraneous to any thinking being— some objective feature of temporal reality—as the basis of the undisputed impression of novelty and the means of a plausible account of the phenomenology of being in time.

Novelty and passage What the preceding section reveals is that there can be temporal becoming and much change in the world—and even a sort of ceasing to be in time, viz., EXISTENCE at some moments, but not at others—and yet there fail to be any real dynamism. The temporal becoming and change accepted by those who think temporal reality is ontologically homogeneous is consistent with an utter lack of novelty, both in the world itself and in one’s experience



Becoming: Temporal, Absolute, and Atemporal

99

of it. Many have conceded this point and in light of it have proposed that temporal reality includes a phenomenon whereby the things in time, primarily moments, gain and lose intrinsic properties in an orderly way, independently of any thinking being. This sort of objective change, traditionally regarded as true passage in time, is supposed by some to be the basis of the impression of novelty. The inclusion of objective intrinsic temporal properties among the things in temporal reality has long been thought to be the definitive feature of the position in opposition to the one in which the world in time is ontologically homogeneous. Although this is indeed the difference between some (problematic) versions of the so-called A-theory and the B-theory of time, it is a mistake to think that this difference is the basic one in the metaphysics of time. If one accepts that an entity must exist in order to be any way whatsoever—including past, present, and future—then a view in which every moment and anything that exists at that moment has some intrinsic temporal property bears a greater affinity to the position that temporal reality is ontologically homogeneous (it entails it) than the one that posits ontological differences in the world in time.20 This point is noticed by some,21 but overlooked by many.22 In this view, the objective basis of novelty in temporal reality is the continuous gain and subsequent loss, by moments, of the intrinsic (temporal) properties, pastness, presentness, futurity. It is crucial to recognize that nothing bears any one of these properties permanently. Thus, a moment that is future momentarily takes on the property of being present and then bears the property of being past (and subsequently, perhaps, the properties associated with being further and further past). This is the traditional notion of the passage of time; it is explicated in terms of literal change. This account of the dynamism in temporal reality is illustrated by several familiar metaphors: the moving spotlight—just as a spotlight illumines a particular (unchanging) area as it courses over a building, presentness “illumines” a single moment; a projected film—just as one (unchanging) frame of a film is shown as it passes the projector bulb, one moment is “shown” as it momentarily bears the property of being present; a flip-book— just as one views as animated a series of static drawings as they quickly pass by, one “views” a sequence of unchanging moments as each takes on the property of presentness before becoming past. The images are well worn, as is the underlying idea: there can be dynamism in a sequence of unchanging things, if each, in succession, temporarily bears some special feature. But one need only state the well-worn idea to raise concerns about its coherence. In any coherent account, what it is for something to have a feature temporarily is for that one thing to have the feature at one moment, m1, and for that very thing to fail to have it at a distinct moment, m2. But

100

DEBATES IN THE METAPHYSICS OF TIME

moments themselves cannot undergo this sort of change. Even granting that a moment exists at itself, no moment can exist at a distinct moment, a fortiori, no moment can have any property—temporal or otherwise—temporarily. The incoherence of this view can be drawn out even more starkly. If the objective basis of novelty in temporal reality were the continuous gain and subsequent loss, by moments, of intrinsic temporal properties, then, given that no moment can exist at a distinct moment, a moment would have to have a given feature (say, presentness) temporarily in the sense that it has that feature at m and subsequently fails to have that feature at m. But this is contradictory in two ways: nothing can both have and fail to have a feature at m; moreover, something at m cannot be subsequent to something at m. Any view in which equally real moments are supposed to change leads to contradiction. This is the key insight of McTaggart’s discussion of the metaphysics of time and the significance of the paradox named for him.23 McTaggart thought, incorrectly, that contradiction here shows that time is unreal. Rather, it shows that an account of the nature of temporal reality, to wit, one in which it is ontologically homogeneous, is incompatible with this account of the dynamism and the undisputed sense of novelty associated with being in the temporal world.24 One might try to introduce a higherorder temporal dimension to resolve the incoherence, but this leads to an infinite regress of temporal dimensions. Even if such a regress is in itself not problematic, it cannot solve the problem here, for such a structure could include no novelty. The present task is to provide a plausible account of the ontological basis of the novelty of being in time. Hence, the too familiar images of the “passage” of time are entirely misleading. In each case—the spotlight, the film, the flip-book—there are things that persist at different moments with different properties—a particular portion of a building, a frame on film, a drawing. Each thing has temporarily, in the standard sense, some property, though perhaps merely a relational one. One cannot make literal sense of these metaphors in the case of a series of moments; moments do not persist and change requires persistence. This important point can be made in another way. Each of the familiar metaphors employs some process. Any sequence of changes that is appropriately considered a process requires the persistence of some entity through some stages of the process. (There can be different persistent entities at different stages of the same process, but persistence is required nonetheless). Therefore, a process is something that occurs in time, over time, that is, at distinct moments. But the objective basis of novelty in temporal reality cannot be something that occurs at distinct moments. One has the full impression of novelty at each moment—rather than partial impressions at distinct moments—and so the ontological basis of the dynamism in the world cannot be any process. Nor can the objective basis of novelty be something



Becoming: Temporal, Absolute, and Atemporal

101

that occurs at a single moment, for if temporal reality is ontologically homogeneous, everything that occurs at a single moment is static and permanent. Even if temporal reality is ontologically heterogeneous, what occurs at a single moment is static. But if the objective basis of the experience of novelty cannot occur at distinct moments, nor at a single moment, then it cannot occur in time at all. In so far as there is an objective basis of the experience of novelty, then, it must be something that is not temporal yet happens to moments.

Absolute becoming If there is real dynamism—novelty—in temporal reality, it cannot come through process, the mere change of equally real things that exist at different moments, nor can it come through the moments themselves taking on and shedding intrinsic temporal properties (or properties of any sort). Change itself cannot provide the ontological basis for the sort of dynamism that partisans of both accounts of the world in time acknowledge. If one insists that novelty come through change, one demands more from change than can possibly be given. There is nothing more to change than the account provided above—a sequence of changeless moments with the same persistent entity at both—and such change can be fully accommodated by both accounts of temporal reality. One who takes the novelty in temporal reality seriously need not think that there is anything more to change.25 However, one must accept that there is, in a sense, more to temporal reality than a sequence of moments all of which EXIST (tenselessly). If there is an ontological basis to the certain appearance of novelty, it arises through moments in time coming into and going out of existence completely. A moment comes into being, lasts but an instant, and then ceases to be entirely. It does not cease to be relative to some other moment even as it REMAINS a constituent of reality. It ceases in every sense to be a part of reality. This phenomenon of coming to be is absolute becoming. The name and original articulation of the notion comes from C. D. Broad.26 As observed at the end of the preceding section, the objective basis of novelty needs to be something that is not temporal yet happens to moments, despite their incapacity to change. Absolute becoming provides the means of accounting for what happens to them: they come into being despite previously existing in no sense. Given this phenomenon, there is literal novelty in temporal reality and not merely new acquaintance with something that EXISTS (tenselessly) at many moments. Absolute becoming is not the coming to be at a moment, which is temporal becoming, but rather the coming to be of a moment. This is no process and yet the source of dynamism and novelty.

102

DEBATES IN THE METAPHYSICS OF TIME

Therefore, the dynamism in the world is a structural feature: it is not found in anything in itself, rather it is found among the things that exist, in the relations they stand to one another—including failing to relate in any sense whatsoever (which occurs when, for example, a moment ceases to be a constituent of reality). Given absolute becoming, a moment comes into being and ceases to be, immediately replaced by a wholly novel moment. In so far as one must recognize absolute becoming to account for the impression of novelty in the world—and its objective ontological basis—one must reject the position that temporal reality is ontologically homogeneous. Thus, the world in time is ontologically heterogeneous.

Atemporal becoming Although each moment lasts but an instant, before ceasing to be a constituent of reality, it does not follow that what exists at each moment goes out of existence with each moment. Familiar concrete objects persist through time, so literally the same one can exist at different moments. What it is for a mutable entity to come into being is for it not to exist (or EXIST) at any moment, then to exist at one. Regardless of one’s account of temporal becoming, what it is for something to come into existence in time is for it to come to be at a moment. Existence at a moment is the mark of a temporal entity. Yet each temporal entity, which has the capacity to change, presumably can be destroyed and, hence, cease to be. Consequently, everything that exists in time eventually ceases to be. If this is so, there appears to be no lasting stability, no true permanence in the world. This, however, is problematic. After all, there are two conspicuous and inescapable features of this world in which time is real. Just as much as one experiences a world in flux, one experiences a stable world. The phenomenon of absolute becoming might be the ontological basis of genuine novelty in the world, but, if this is so, the indubitable permanence in the world seems to be lost. One who acknowledges absolute becoming must account for the stability of a world in which time is real. It remains to be seen, though, whether a view that accepts that temporal reality is ontologically heterogeneous has the means to do so. I believe the phenomenon of atemporal becoming reconciles radical, continuous novelty in a world in which time is real with abiding permanence. If temporal becoming, coming into being in time, is to come to be at a moment, then atemporal becoming, coming into being outside of time, is to come into being, but not at a moment. A thing that comes into being outside of time is an atemporal entity. Above, it was noted that the mark of a temporal



Becoming: Temporal, Absolute, and Atemporal

103

being is the capacity to change; the mark, then, of an atemporal entity is immutability. An atemporal being is one that does not and could not possibly change. It comes into being outside of time because, by its very nature, it is the sort of thing that cannot change. What must be noted immediately is that coming into existence is not any sort of change. An entity that comes into being is not first one way— non-existent—and then a different way—existent—having persisted through some change. Becoming is no sort of process; a fortiori, atemporal becoming is not either. Of course, once an atemporal entity comes into existence it cannot change. But the crucial point to recognize is that there is no contradiction in saying that a thing that comes into existence does so outside of time, that it undergoes no change in coming to be and, therefore, is an atemporal entity. Note that it is a mistake to try to time, as it were, the coming into existence of an atemporal entity. It might be tempting to say that the existence of some new kind of thing comes into being simultaneously with the coming to be of the first instance, some individual substance, of that kind. Temptation here should be resisted, for this way of regarding the situation is misleading and confusing. The coming to be of the two entities, the new kind and its first instance, is not simultaneous, because what it is to be simultaneous is to occur at the same moment. In this case, however, one entity—the individual substance—comes to be at a moment, the other—its kind—comes to be, but at no moment at all. It is literally false that the two origins are simultaneous. I fear that some will suspect chicanery. One might insist that coming to be seems like a process—something must change when a novel entity exists though it did not before—so the notion of atemporal becoming is just double-talk. I maintain, however, that any suspicions here must be based on unexamined and too coarse notions of becoming and process (and perhaps non-existence). The phenomenon of atemporal becoming articulated above is based on explicit and plausible accounts of change, time, and temporal becoming. Therefore, it seems to be a genuine phenomenon.

Atemporal entities An atemporal entity is one that exists immutably outside of time. Such an entity might lack an origin or, as argued above, an atemporal entity might have an origin: it might come into being outside of time and EXIST (tenselessly) immutably. Examples of the former might be the number 2, God and the property of being self-identical or being real. Examples of the latter might be things like the property of being good or being salty or the kind water or platypus or microwave-oven or fan of the New York Yankees.

104

DEBATES IN THE METAPHYSICS OF TIME

In the present context, a particularly relevant kind of atemporal entity that has an origin is a simple fact. I have discussed simple facts in detail elsewhere, so I provide here only a very brief account of them.27 Simple facts are mereologically simple and ontologically independent entities that are both the truthmakers for every true representation and also provide, in many cases, the ontological basis of the structure in the world. Examples of simple facts are the simple fact that there is no tiger in my office (at m), the simple fact that I am not eight feet tall (at m), the simple fact that the branches on the tree outside my window are moving in the breeze (at m) and, importantly, the simple fact that Aristotle is a philosopher (at m), where m is a moment that no longer exists. Such entities are of crucial importance because they are the ontological basis of the permanence in a world that also includes absolute becoming and radical novelty. In such a world, every moment and every mutable entity at some point ceases to be. However, there are simple facts that come into existence to ground every true account of every moment and everything in time. Given atemporal becoming—and the existence of simple facts—one who accepts that temporal reality is ontologically heterogeneous, and so can adequately account for the experience of novelty, can also account for the indubitable permanence in the world. Things, the temporal ones, come and go, but some things, the atemporal ones, abide.

Conclusion Reality is structured: it is, in itself, a world of natured entities standing in relations. The proper understanding of these entities and their natures and relations enables one to see that there is time and change and mutable things; there are also immutable things, entities not intimately related to time. The proper understanding of process and its ontological basis, absolute becoming, enables one to distinguish temporal becoming from atemporal becoming. Atemporal becoming is the phenomenon that reconciles the two seemingly irreconcilable features of being in a world in which time is real: there is flux and transience and there is stability and permanence. Every truth about some changing feature of the world is grounded by an unchanging and immutable entity in the atemporal world (viz., some simple fact). So transience is in the temporal world, permanence in the atemporal world. Reality includes both the temporal and atemporal. I believe the pressing sense that reality is both transient and permanent, and the inability of the two general positions regarding the nature of temporal



Becoming: Temporal, Absolute, and Atemporal

105

reality to account for this sense in themselves has been the source of much, if not all, of the contention in contemporary discussions about the metaphysics of time. If one takes the beginning of the modern development of the metaphysics of time to be McTaggart’s seminal argument for the unreality of time, one sees from the outset the struggle to provide a satisfactory account of the dynamism in a world of static moments. What has been neglected in the metaphysics of time is the notion that there is more to the world than the world in time, and the initially perplexing phenomenon of atemporal becoming. The preceding discussion is my attempt to redress this neglect and thereby sketch a fully satisfactory metaphysics of time, one that embraces both of the conspicuous and inescapable features of this world in which time is real. Thus, one can contemplate in silence—then be startled by a flash and accompanying boom—then return to silence, and although the flash and boom are gone in every sense, there is something that remains unchanged, a simple fact regarding the flash and boom that no longer exist. This fact, having come into existence outside of time, is as permanent and stable as a thing could be.28

Notes   1 See, for example, Shoemaker (1969); for an opposing view, see Lowe (2002: 247–9). Time without change is consistent with the account of time per se I am about to present.   2 A thing is certain ways because of what it is and via this nature contributes to the structure in reality. I make no distinction between thing, entity, existent, or being; any difference in usage is merely stylistic.   3 In brief, the argument for this claim is that since change is actual, it must be possible; the thing that makes it possible therefore must exist.   4 The tenseless theory is a misleading name for this position because it is based on a semantic thesis that is actually compatible with the other position on temporal reality about to be presented in the text. For a discussion of the tenseless theory, see Mellor (1981, 1998).   5 The tensed theory is a misleading representative of this class of views because it is based on a semantic thesis that is actually orthogonal to the key ontological issues. See the previous note.   6 These terms come from Peter Geach (see Geach 1969: 71–2) in (perhaps backhanded) acknowledgment of the Cambridge philosophers, like McTaggart and Russell, who employed the notions.   7 Note that there can be real changes that are relational. Changes incurred when a thing takes on a part that it did not have or when it simply moves are, in both cases, in virtue of how that thing itself is (how it is composed or where it is located).

106

DEBATES IN THE METAPHYSICS OF TIME

  8 Quentin Smith argues that all entities, including abstract ones such as propositions, exist in time (see Smith 1998: 157–61). However, his argument for this is based on the claim that an object undergoes genuine change when it changes with respect to its “Cambridge properties.” He is explicit about this (Smith 1998: 148). So he maintains that when I cease to believe a certain proposition, p, this is a change in p. But clearly what grounds this change is some difference in my mental states, not in how p is in itself.   9 Given this real distinction between the world within time and the world without, I must disagree with Chisholm and Zimmerman, who maintain that there is no reason to take “tenselessness” seriously (Chisholm and Zimmerman 1997). 10 In the context of any discussion of the homogeneity of temporal reality, all verbs must be read as tenseless. I only put select verbs in ALL CAPS in order to emphasize their tenseless reading, but this does not mean the other verbs in that context are not tenseless. This point should be borne in mind throughout this chapter. 11 Williams (1951: 465–6). 12 Smart (1980: 3). 13 Maudlin (2007: 135). 14 Skow (2009: Section IV). 15 Paul (2010: 333). For further recognition of this impression of novelty, see, for example, Schlesinger (1982: 501, 515) and Mellor (1998: 66–7). 16 Smart (1967). 17 Mellor (1998: 66). Mellor (1981: 116) characterizes change in one’s beliefs as the “psychological reality behind the myth of passage.” 18 Paul (2010: 339). 19 For related discussion of this point, see Fiocco (2010). 20 Hence, I disagree with Laurie Paul when she maintains that “the nexus of a philosophical debate over the ontology of time” is whether the “temporal properties of now and passage exist” (Paul 2010: 338). 21 See, for instance, Skow (2012: 223), where Skow asserts that his “A-theory” of time is a “version of eternalism.” 22 For a prominent example, see the work of D. H. Mellor (1981, 1998). 23 See McTaggart (1908). 24 Thus, Skow’s view, which explicitly combines the ontological homogeneity of temporal reality with changing moments, seems problematic. See Note 21. 25 Pace Laurie Paul’s claim at Paul (2010: 334). 26 See Broad (1938: ch. 35, vol. II). For discussion of the notion, see Savitt (2002: 159ff.) and Fiocco (2007). Savitt interprets Broad’s account of absolute becoming very differently than I do. 27 See Fiocco (2014). 28 I would like to thank Michael Brent and Nathan Oaklander for their interesting and insightful comments on a draft of this chapter.



Becoming: Temporal, Absolute, and Atemporal

107

References Broad, C. D. (1938), An Examination of McTaggart’s Philosophy. Cambridge: Cambridge University Press. Chisholm, R. and Zimmerman, D. (1997), “Theology and tense,” Noûs, 31, 262–5. Edward, P. (ed.) (1967), The Encyclopedia of Philosophy. New York: Macmillan. Fiocco, M. O. (2007), “Passage, becoming and the nature of temporal reality,” Philosophia, 35, 1–21. —(2010), “Temporary intrinsics and relativization,” Pacific Philosophical Quarterly, 91, 64–77. —(2014), “On Simple Facts,” Res Philosophica, 91(3). Geach, P. (1969), God and the Soul. London: Routledge. van Inwagen, P. (ed.) (1980), Time and Cause: Essays Presented to Richard Taylor. Dordrecht: Reidel. Le Poidevin, R. (ed.) (1998), Questions of Time and Tense. Oxford: Oxford University Press. Lowe, E. J. (2002), A Survey of Metaphysics. Oxford: Oxford University Press. Maudlin, T. (2007), The Metaphysics within Physics. New York: Oxford University Press. McTaggart, J. M. E. (1908), “The unreality of time,” Mind, 18, 457–84. Mellor, D. H. (1981), Real Time. Cambridge: Cambridge University Press. —(1998), Real Time II. London and New York: Routledge. Paul, L. (2010), “Temporal experience,” Journal of Philosophy, 107, 333–59. Savitt, S. (2002), “On Absolute Becoming and the Myth of Passage,” in Craig Callender (ed.), Time, Reality and Experience. Cambridge: Cambridge University Press, pp. 153–67. Schlesinger, G. (1982), “How time flies,” Mind, 91, 501–23. Shoemaker, S. (1969), “Time without change,” Journal of Philosophy, 66, 363–81. Skow, B. (2009). “Relativity and the moving spotlight,” Journal of Philosophy, 106, 666–78. —(2012), “Why does time pass?,” Noûs, 46, 223–42. Smart, J. J. C. (1967), “Time,” in P. Edward (ed.), The Encyclopedia of Philosophy. New York: Macmillan. —(1980), “Time and Becoming,” in P. van Inwagen, Time and Cause: Essays Presented to Richard Taylor. Dordrecht: Reidel, p. 3–15. Smith, Q. (1998), “Absolute Simultaneity and the Infinity of Time,” in R. Le Poidevin (ed.), Questions of Time and Tense. Oxford: Oxford University Press, pp. 135–83. Williams, D. C. (1951), “The myth of passage,” Journal of Philosophy, 48, 457–72.

6 Temporal Predicates and the Passage of Time M. Joshua Mozersky

Introduction

T

raditionally, advocates of the tenseless, B-theory of time deny the reality of temporal passage (e.g. Smart 1963, 1980; Mellor 1981, 1998; Horwich 1987; Price 1996). I argue that this is a mistake. B-theorists should accept that time passes not only because there is overwhelming evidence that it does, but also because the B-theory provides the best resources for making sense of the passage of time. I defend this view via a consideration of the puzzle of change: how is it possible to make sense of a single object having incompatible properties at different times? I present and defend a solution to this puzzle that needn’t worry B-theorists and allows for a coherent and satisfying theory of change, persistence, and temporal passage.

The logical form of temporal predicates Consider the temporal predicates “x is past,” “x is present,” and “x is future,” which I shall refer to as “A-predicates.” Assume that the variables take individuated entities, such as events or times, as values. Let us begin with a simple question about these predicates: are they one-place (monadic) or are they many-place (relational)? If the former, then their logical form is as follows: (A) F(x), G(x), and H(x);

110

DEBATES IN THE METAPHYSICS OF TIME

if the latter, then their logical form is: (B) F(x, y), G(x, y), and H(x, y),1 which I shall refer to as “B-relations.”2 How might we decide between these two options? A useful place to start is with a consideration of the nature of change. The concept of change can certainly appear puzzling. On the one hand, change requires difference: if x changes from F to ~F, then x is no longer exactly the same before and after the change. On the other hand, change requires identity: if x changes from F to ~F, then it must be x that is both F and ~F; if distinct entities are F and ~F respectively, then nothing has changed. Yet identity and difference appear to conflict with each other. If x = y, then x and y cannot differ in any way (Leibniz’s Law); but, if x changes from F to ~F, then it seems that whatever it is that is F cannot be identical to whatever it is that is ~F. Hence, change appears to be impossible (see Hinchliff 1996). We can generate a similar puzzle with A-predicates. The reason for this is that if these predicates are to model time, then if anything satisfies a given A-predicate, it must also satisfy the others (let us ignore the complication of a first or last moment of time, as this will not impact the arguments here). If some event is, say, present, it does not remain present for eternity; it is eventually past and something else is present, and so on. But then one and the same entity must be both F (present) and ~F (past), and the conflict between identity and difference arises for anything that satisfies A-predicates. I propose that the solution to the puzzle of change and the decision between (A) and (B) are the same. So, let us start with change. Suppose some object, x, goes through a change, say from being (entirely) blue to being (entirely) green. If we let “F(x)” = “x is blue” and “G(x)” = “x is green,” and assume that F(x) entails ~G(x) (and vice versa), then the formal description of this change appears to be the logically impossible: (1) F(x) &G(x). In other words, when ordinary predicates are understood monadically, attributions of change are outright contradictions. One can, however, reconcile identity and difference while avoiding contradictions in one fell swoop. The trick is to assume that ordinary predicates, such as “x is blue” and “x is green,” are two-place relations that hold between objects and times. In that case, the logical form of the predicate “x is blue” is: (2) F(x, t)



Temporal Predicates and the Passage of Time

111

and the logical from of the predicate “x is green” is: (3) G(x, t). Given this, the proposition that some object has changed from blue to green is: (4) ( x)(F(x, t1) &G(x, t2)). Notice that in (4) it is one entity, x, that satisfies both relational predicates, so identity through change is preserved in this description. Furthermore, in (4) x is blue at one time, green at another, so difference is represented. Finally, note that (4) is consistent since two otherwise incompatible relations can be combined without contradiction if they relate an entity to two different times. So, the problem of how to coherently describe change has been resolved. Returning to A-predicates, we can engage in a similar line of reasoning. Suppose that some event, e, is present, then past. How are we to represent this? If A-predicates are monadic, our choice would seem to be (1) again, which is a contradiction because F(x) entails ~G(x) and G(x) entails ~F(x). Let us, then, try an analogous solution: “e is past” will be understood as a relation between an entity and a time: (5) F(e, t). Similarly, “e is present” will be: (6) G(e, t). Now the change an event undergoes from being present to being past can be expressed as follows: (7) ( e)F(e, t1) &G(e, t2). This proposition is logically consistent and reconciles identity and difference. Accordingly, (B) is to be preferred over (A) and we may conclude that A-predicates are relations. It appears that the simple question with which we began has been answered. This argument may look familiar. It is, with one important difference, very much like McTaggart’s argument against the coherence of the A-series (McTaggart 1908, 1927). The important difference is that McTaggart thought that propositions such as (7) preserve identity and logical consistency at the cost of difference; in other words, he believed that neither (4) nor (7)

112

DEBATES IN THE METAPHYSICS OF TIME

represent change. Hence he took the argument against (A) to be an argument against the existence of time on the grounds that time entails change. But why follow McTaggart here? I turn next to three lines of reasoning intended to indict (4) and (7) as unsatisfactory accounts of change.

Objections to the relational theory McTaggart The first objection, which is noted above, is due to McTaggart who argued that propositions such as (7) and (4) are not truly representations of change because their truth-values are unchanging: The fact that [a poker] is hot at one point in a series and cold at other points cannot give change, if neither of these facts change … It follows from what we have said that there can be no change unless some propositions are sometimes true and sometimes false. (McTaggart 1927: 15) B-relations have temporally stable extensions. If x is blue at t, then x is blue at t in a temporally insensitive way: it will not one day become false that x is blue at t because events in time are not variable in that way. So McTaggart is right that a proposition of the form of (7) is, if true, a tenseless truth. Must we, however, follow McTaggart in concluding that (7) cannot represent change? I don’t see why we should. Even if a representation does not itself change, it does not follow that it is not a representation of change. Heather Dyke has very importantly drawn attention to what she calls the “representational fallacy” (Dyke 2008), which is, roughly, the attempt to read off features of non-linguistic reality from features of our linguistic representations.3 I think she is right to be suspicious of such reasoning. There are many instances in which features of our representations—for example, the presence of words—fail to be features of what we represent. In general, therefore, a property of a representation isn’t necessarily a property of that which is represented. McTaggart’s suspicion of (7) appears to rest, however, on the even more dubious inverse of the representational fallacy. One would only follow McTaggart’s argument that (7) does not accurately represent change if one were to suppose that a property of what is represented must be a property of the representation. In this particular case, McTaggart assumes that the world can exhibit change in properties only if propositions about that world change their properties, in particular their semantic properties. This is, however,



Temporal Predicates and the Passage of Time

113

simply not credible for, if generalized, it would entail that any representation of a red item must itself be red, or any story about evil must itself be evil, or that a picture of a living person must itself be alive, and so on. There is no reason to suppose that there is anything like a one-to-one correspondence here. What we may conclude is that McTaggart had good reason to insist that the only coherent conception of A-predicates is relational, but wrong to assume that this meant the unreality of change.

Lewis It has been argued that the relational account is incompatible with the idea of intrinsic change. David Lewis gives forceful expression to this concern: Persisting things change their intrinsic properties. For instance shape: when I sit, I have a bent shape; when I stand, I have a straightened shape. Both shapes are temporary intrinsic properties; I have them only some of the time. How is such change possible? I know of only three solutions … First solution: contrary to what we might think, shapes are not genuine intrinsic properties. They are disguised relations, which an enduring thing may bear to times. One and the same enduring thing may bear the bentshape relation to some times, and the straight-shape relation to others. In itself, considered apart from its relations to other things, it [i.e., a changing object] has no shape at all. And likewise for all other seeming temporary intrinsics; all of them must be reinterpreted as relations that something with an absolutely unchanging intrinsic nature bears to different times. The solution to the problem of temporary intrinsics is that there aren’t any … This is simply incredible … If we know what shape is, we know that it is a property not a relation. (Lewis 1986: 203–4) One can identify three complaints in this passage: first, that the relational solution reduces objects to loci of unchanging intrinsic natures; second, and relatedly, that the relational view entails that nothing is temporarily intrinsic to an object; and third, that it is simply unbelievable that, for example, an object’s shape could be a relation between it and a time. I address each of these in turn. Concerning the first objection, it is hard to see why expressing ordinary change by way of propositions of the form of (4) above reduces x to something that has an unchanging intrinsic nature. Perhaps the thought is that if, for example, shape is a relation between x and t, then x in fact lacks a shape. But just because it is only meaningful to predicate some shape of x if

114

DEBATES IN THE METAPHYSICS OF TIME

there is some time or other at which x has that shape, it doesn’t follow that x itself is somehow shapeless. One has to keep in mind that we are concerned here with relations between material objects and times, and if there is no time at which x is shaped, well, then x has no shape. For a material object to have a shape is for it to be shaped at some time or other, and this does not entail that the object lacks shape. It simply entails that shape predications only make sense in reference to a time at which the shape is exemplified.4 So, in other words, it is truly x that has a shape in the relational view; it is just that the having of its shape is not something that can be rendered sensible without reference to some time or another. Hence, I think that the relational view does not render objects featureless. If this is right, then the reply to Lewis’s second objection is as follows. It is true that ascriptions of shape, size, color, mass, and so on, are necessarily relative to times. This, however, is compatible with shape, size, color, mass, and so on, being intrinsic to objects because according to (4) above it is precisely the object, x, and not something else that has shape, size, color, mass, and so on, at a time. In particular, it is not the object–time pair that is, say, round, or large, or blue, and so on. Suppose, for comparison, that John is inside a house. This is a relation between John and a house, and the relation requires the existence of both, so in that sense one might want to say that the pair instantiates the “is inside of” relation. Nevertheless, in this pair it is John who is inside; the John–house pair isn’t inside of anything. Now, of course, this isn’t an intrinsic property of John, so the comparison is misleading in that respect. However, the point I am trying to emphasize is this: the fact that x stands in relation R to y doesn’t entail that there is no asymmetry, with respect to R, between x and the pair . So, even though x’s being blue entails that there is a t at which x is blue, and in this sense can be said to entail that the pair instantiates the “is blue at” relation, this is all compatible with it being x that is blue rather than the pair (how could such a thing be blue anyway?).5 I think we can see that this is sufficiently “intrinsic” if we examine this concept in a bit more detail. Lewis writes that “A thing has its intrinsic properties in virtue of the way that thing itself, and nothing else, is” (Lewis 1983: 197). It is important to note that, according to the relational theory defended above, although x is red at t in virtue of standing in the “is red at” relation to t, x’s being red doesn’t depend on the way that t is. It depends, rather, on there being a t at which x is red. Lewis in fact defends an analogous position, for in his view (Lewis 1986) for x to be F is for x to be F in some world or other. Lewis does not, nor should he, conclude on that basis that for any F, x’s being F depends in part on the “way x’s world is” and so isn’t intrinsic. To be F is to be F in some world, for Lewis, even if F is intrinsic. Analogously, in the relational view, for any material object to be F is for it to be F at some time, even if F is intrinsic.



Temporal Predicates and the Passage of Time

115

Here is more from Lewis: The intrinsic properties of something depend only on that thing; whereas the extrinsic properties of something may depend, wholly or partly, on something else. If something has an intrinsic property, then so does any perfect duplicate of that thing; whereas duplicates situated in different surroundings will differ in their extrinsic properties. (Lewis 1983: 197) I think, however, that the first sentence is wrong. Since to be a material object is to occupy space and time, all properties of material bodies depend on space and time for their instantiation; material bodies would not exist without space and time. Nonetheless, there is still a difference between those properties that are properties of the object (its being round) and those that are not, or not wholly, of the object (e.g., its being to the left of something). It should be added, moreover, that in the relational view of predication, a perfect duplicate of a material object would not differ intrinsically from the original object. What follows from the relational view is that for an object and its duplicate, any specification that the object or duplicate is F must make reference to time, that is, to when the object is F. This is in fact a strength of the relational view, for we don’t want to commit to an account of material objects that does not tie their descriptions to spatial and temporal locations. An object is material only if it instantiates its predicates at some time (place) or another and this will be true of both an object and its duplicate. So, the relational view appears to be compatible with the distinction between those predications that describe an object itself and those that do not. The distinction depends on whether one needs to specify a time/place in describing the object or whether one needs also to specify other objects. If one insists that even in the former case, the concept of something being intrinsic to an object has been destroyed, then my reply is “Let us lay this concept to rest in peace,” for it has become impossible to suppose that any predication ever captures anything intrinsic about a material object without denying that material objects are necessarily related to at least one time (and place); that is, without denying something that is essential to materiality. The relational account has an alternative understanding of “intrinsic” close at hand. As for Lewis’s third objection, I think that the sting is taken out of it by the foregoing considerations: the relational theory of temporal predication seems less bizarre once we realize that it doesn’t render objects featureless or lacking intrinsic characterizations. What’s more, it captures something we want any theory of objects to capture, namely that it makes no sense to predicate anything of a material object without entailing that there is some time at which that predicate is instantiated. I conclude that Lewis’s concerns are not telling against the relational theory.

116

DEBATES IN THE METAPHYSICS OF TIME

Mellor D. H. Mellor argues that the relational account of predication fails because: … relations generally do not require the entities they link to share locations in space and time. My being taller than Napoleon, for example, is quite consistent with his dying before I was conceived; while my conception’s being later than his death positively requires it. (Mellor 1998: 93–4) So, argues Mellor, “x is blue at t” cannot be a relation between x and t because if x is blue at t then x must exist at t but R(x, t) does not entail that x coexists with t. As a result, Mellor argues that predicates such as “x is blue” are monadic but that temporal predications must include a temporal operator, for example: (8) At t: Fx. If x changes from F to ~F, then: (9) At t’: ~Fx. For Mellor, the operator indicates the location of a fact (that x is F and ~F respectively). In this way, consistent descriptions of change are possible. As Mellor himself observes (Mellor 1998: 94), some relations do entail sameness of temporal location; his example is “x is simultaneous with y.” But, we should note, there are very many relations that entail temporal (and, in many cases, spatial) coincidence or overlap. Here are just a few examples: “x is in (physical) contact with y,” “x is above y,” “x is beside y,” “x is talking to y,” “x is at y,” and so on. So in some cases the fact that x stands in R to y entails that x and y overlap temporally; in other cases it does not. It is, therefore, hard to see why Mellor’s words in the quotation above count, in any decisive way, against the relational view of predication. What appears to be the case is that some relations are spatial- or temporal-overlap entailing and some are not, but that this difference is not the result of logical form. Mellor presents a response to this sort of challenge: But it is no answer to say that changeable properties too are relations which entail this [i.e., sameness of temporal location]. For what makes [“a is F at t”] entail that a is located at t, if not the fact that, as I argued in §3, F’s being a non-relational property requires a to be located wherever and whenever a is F. (Mellor 1998: 94)



Temporal Predicates and the Passage of Time

117

This is an interesting suggestion, but ultimately fails as an argument against the relational view. The reason for this is that the argument in §3 to which Mellor alludes is that the gain or loss of any of x’s real properties must have its first effects on, in or at least near x: The birth of another child to my parents is not a change in me, because its first effects are neither on nor near me … In short, real changes of properties need effects, and for them to be changes in the things to which we ascribe those properties, that is where their first effects must be. (Mellor 1998: 87–8) Notice, however, that “e’s first effects are at/on x” (where e indicates an event that is a change in x) is a relational predicate, and if it can entail that e and x are co-located, then so can “x is blue at t,” or “x is green at t” entail that x and t are co-located. Perhaps Mellor has in mind that there is something about causation that does the essential work here, that is, that it is only because e is the effect of the gain or loss of a property of x that e and x are co-located. This does not strike me as plausible because causally disconnected events or objects can coincide or overlap. However, even if we accept this line of reasoning, nothing prevents the advocate of the relational account of temporal predication from helping herself to a similar story. That is to say, the relationalist can argue that it is because becoming blue has its first effects, say at t, on x that “x is blue at t” entails that x and t are co-located; this is fine and doesn’t change the fact that “x is blue at t” is a relational predicate.6 Mellor also makes the following, further point: The causal test for properties is related to another one, namely that a thing’s properties should be detectable just by inspecting that thing. (Mellor 1998: 88) However, as my argument above entails, there is no way to inspect something without inspecting it at some time or other. Hence, there is no way to detect anything about some object, x, independent of time. The fact that everything manifested by some object is manifested relative to a time in no way excludes the possibility of that object’s properties being detectable by inspection of that object alone. It is simply a mistake to treat relations to times as akin to relations to other objects in this respect. It makes perfect sense to suppose that a material object, x, is red independent of its relations to any other objects in the universe, that is, that it would still be red even if they did not exist; this may turn out to be physically impossible,7 but it is a coherent supposition.

118

DEBATES IN THE METAPHYSICS OF TIME

What is not coherent, however, is to suppose that x is red independent of its relations to time, for that would entail that x could be red even if there were never a time at which it is red (or that we could inspect it for color but not at a particular time), which is absurd.8 We can extend these observations to temporal operator accounts of predication in general. In indicating a temporal location, the operator “At t” in “At t: Fx” sets up a relation, in particular the relation of temporal co-location, between t and x, or between t and that fact that x is F.9 “At t,” in other words, indicates the time at which x is F. Since there will, in general, be some times at which x is F and some times at which it is not, the formula: (8) At t: Fx. will be satisfied by ordered pairs, , of objects and times, which is another way of saying that (8) expresses a relation.10 There is simply no getting around it: temporary predicates, including A-predicates, are relational in form.11

The passage of time I have argued that the relational account of temporal predication leads to a coherent and satisfactory account of change. If I am right, then change is compatible with a tenseless, or B-theoretic, world-view. It seems to me to follow from this that the passage of time is itself compatible with a tenseless, B-theoretic world-view. I propose the following: time passes if and only if first one thing (or a set of things) happens, then another. Temporal passage is, in other words, the ordering of events by the B-relation, “x is earlier than y.” People age and die, species arise and fall, fruit ripens and rots, and planets orbit stars. All of this is describable in the tenseless, relational view of temporal predication. Most authors who defend a tenseless, B-theoretic account of time, however, deny that time passes; for example: It is clear, then, that we cannot talk about time as a river, about the flow of time, of our advance through time, or of the irreversibility of time without being in great danger of falling into absurdity. (Smart 1949: 485) There is no real passage of time. What we refer to by “the passage of time” is an illusory feature of conscious experience. (Prosser 2007: 81)



Temporal Predicates and the Passage of Time

119

My present thesis would resolve the antinomy by rejecting the extra idea of passage as spurious altogether. (Williams 1951: 462) There are many other examples one could mention (e.g. Smart 1963, 1980; Mellor 1981, 1998; Price 1996). I think that this is a mistake. We all have overwhelming evidence that time passes: I am now typing on a computer but earlier today I was not; I used to carry my daughter on my forearm but she is now too big for that; 25 years ago nobody had a smart phone while today they are ubiquitous; and so on. Accordingly, the denial of temporal passage lends unneeded credibility to the opponents of the tenseless, relational view of time. So the question I want to address next is: “Why do B-theorists reject the passage of time?”12 I think there are two reasons. The first is that many B-theorists accept McTaggart’s view that only the A-series can explain passage. Mellor, for example, writes: One author, however, I will acknowledge: J. E. McTaggart, who proved the unreality of tense and of the flow of time. (Mellor 1981: 3) What is wrong with McTaggart is not his attack on time’s flow but his view that change requires it. (Mellor 1998: 72) Here we see a move from the tenseless view of the world (i.e., the denial of the tensed view) to the unreality of the flow of time. But such a transition can now be seen to be unnecessary and unmotivated. If the passage of time required monadic A-predicates, then there would indeed be no passage of time. But what authors such as Mellor overlook is the possibility of a tenseless, B-theoretic temporal passage to be equated, I argue, with the temporal ordering of events. The tenseless, B-theoretic world-view is one in which the sum total of events, objects, and processes is unchanging (which is why the extensions of temporal predicates are not temporally variable). This, however, is compatible with change: … change is always variation in one thing with respect to another, the totality of absolute facts about those functional relations remaining forever constant. (Horwich 1987: 25) Horwich is right: the existence of genuine change does not require the totality of facts to change with respect to time. Therefore, the passage of time does not require such change either. Michael Dummett disagrees:

120

DEBATES IN THE METAPHYSICS OF TIME

Now if time were real, then … there would be no such thing as the complete description of reality. There would be one, as it were, maximal description of reality in which the statement “The event M is happening” figured, others which contained the statement “The event M happened”, and yet others which contain “The event M is going to happen”. (Dummett 1978: 356) The problem is that this argument presupposes that real change or passage requires absolute (i.e., non-relational) A-predicates to apply to events. Without this assumption, then what “M is happening” or “M is now” expresses is a temporally invariant relation between M and a particular time, and this relation is expressible at other times. As I have argued above, the non-relational account of A-predicates is untenable and I see no reason to cling to its ghostly apparition, that is, to the idea that without it nothing really changes. The second reason that B-theorists deny temporal passage is that they think of it as a kind of motion, a motion whose rate is indefinable. Here, for example, is Smart: Contrast the pseudo-question “how fast am I advancing through time?” or “How fast did time flow yesterday?.” We do not know how we ought to set about answering it. What sort of measurement ought we to make? We do not even know the sort of units in which our answer should be expressed. “I am advancing through time at how many seconds per—?” we might begin, and then we should have to stop. What could possibly fill the blank? Not “seconds” surely. In that case the most we could hope for would be the not very illuminating remark that there is just one second in every second. (Smart 1949: 485) A more recent variation is due to Huw Price: Indeed, perhaps the strongest reason for denying the objectivity of the present is that it is so difficult to make sense of the notion of an objective flow or passage of time. Why? Well, the stock objection is that if it made sense to say that time flows then it would make sense to ask how fast it flows … Some people reply that time flows at one second per second, but even if we could live with the lack of other possibilities, this answer misses the more basic aspect of the objection. A rate of seconds per second is not a rate at all in physical terms. It is a dimensionless quantity, rather than a rate of any sort. (Price 1996: 13) There is, however, nothing in the relational account of change that entails that temporal passage must be thought of as a kind of motion. Our



Temporal Predicates and the Passage of Time

121

language is perhaps a bit misleading here. We speak of the passage of time as though some single, individuated entity, Time Itself, is changing or moving, when in fact what is really going on is that changes occur within or with respect to time. So, “the passage of time” strikes me as a summary concept that refers to the ongoing processes of change that constantly occur, in order, with respect to time.13 The only thing all these processes have in common is that they involve events related by “is earlier than”; it is this relation and not the concept of motion that captures the essence of temporal passage. Suppose that a person begins the day with nervous energy, becomes stressed as the day progresses, and then feels a late afternoon bout of sadness before feeling happy and relieved in the evening. What could the rate of this change possibly be: so many moods per hour? Even if there is no answer to this question (perhaps moods are not divisible in the requisite way), it is obvious that this person has undergone various emotional changes over the course of the day. So, it is not essential that there be a well-defined rate of change in order for change to occur (and we certainly don’t need to know that there is such a rate in order to correctly conclude that change has occurred).14 In sum, passage needn’t be a kind of motion nor explained in terms of monadic A-predicates in order to be real. There is, therefore, no conflict between the tenseless, B-theoretic world-view and the reality of temporal passage.

On two rival views: Temporal parts theory and presentism The relational view of temporal predication, change, and passage that I defend above differs in important respects from two popular philosophical theses: presentism and perdurantism. First, the relational view ontologically commits to more than one time (e.g. (4) above) which conflicts with the presentist proposition that, necessarily, all and only that which is present exists. Second, in the relational view, when an object changes it is numerically the same object before and after the change—for example x in (4) above takes the same object as its value in both places—and this conflicts with the perdurance view that objects persist in virtue of having distinct temporal parts that are located at different times. Let me make a few brief remarks on presentism and perdurantism in turn. Arthur Prior sets up one of his many defenses of presentism by noting the need to make room for a concept of temporal passage:

122

DEBATES IN THE METAPHYSICS OF TIME

I believe that what we see as a progress of events is a progress of events, a coming to pass of one thing after another, and not just a timeless tapestry with everything stuck there for good and all. (Prior 1996: 47) I have argued, however, that the relational view is perfectly consistent with genuine and objective change as well as temporal passage. So there is no need to appeal to the coming into (and passing out of) being of events to make sense of the passage of time. This is a good thing, for I consider such a process to be deeply mysterious. I also prefer a view that allows past and future times, objects and events to serve as the referents and truthmakers of propositions about the past and future, since this strikes me as the most straightforward way of making sense of thought and talk about the past and future (I discuss this issue in Mozersky 2011). The relational theory allows for all this. Others defend presentism on other grounds. For example, Craig (1998) argues that presentism is the only way to solve the apparent contradiction entailed by A-predication.15 The arguments above suggest otherwise. Others suggest that presentism is the common-sense view (Markosian 2004; Zimmerman 2008). Note, however, that the relational account defended above is based on the proposition that objects persist through change, then uses rather straightforward logical considerations to draw conclusions, including that the passage of time is real; I submit that there is little to offend common sense here. In short, by committing to the reality of change and passage, avoiding McTaggart’s contradiction and cohering with common sense, the relational view allows one to resist many of the arguments used to push one toward the presentist ontology. As I mention above, I take this to be a consideration in favor of the relational view. According to perdurance theory, on the other hand: … change over time is the possession of different properties by different temporal parts of an object. (Hawley 2001: 12) It is worth noting that in this kind of view the primary bearers of predicates are temporal parts of objects, not objects; x is F at t in virtue of having a temporal part, p, located at, and only at, t such that p is F. I find perdurance theories to be less extravagant than presentism—no mysterious coming into existence from nothing—but the former still incur the disadvantage of committing to entities, namely temporal parts, that are unfamiliar. Is the commitment to such entities necessary? Sider argues that one reason for believing in temporal parts is that it solves the puzzle of change:



Temporal Predicates and the Passage of Time

123

The temporal parts account of change is that incompatible properties are had by different objects, different temporal parts of the whole. Change is therefore no more remarkable than the variation of a road with some bumpy stretches and some smooth stretches. (Sider 2001: 93) This is by no means the only or even the central argument Sider provides for a temporal parts ontology, but the relational view does remove a motivation for the view by providing an alternate account that has what I believe to be the distinct advantage that objects retain their strict, numerical identity as they change. Sider adopts “stage theory” according to which later temporal parts of an object, though numerically distinct from earlier ones, are the object at later times (they are temporal counterparts). So, he does not take perdurantism to be the denial that objects persist. However, as the quotation above makes clear, his view is still the denial that there is a single material object involved when, say, an apple changes color. Hence, his view does, I think, retain a curious consequence that the relational theory avoids. Again, I see this as an advantage of the relational view, which does not move us to reinterpret our ordinary way of thinking about persistence in terms of numerical identity; a way of thinking that is, I think, reinforced regularly by our interactions with the world. I take it that the foregoing considerations suggest that the relational view retains a comparative advantage over presentism and perdurantism as a result of its relatively minimal metaphysical and revisionary commitments. It is, in short, an economical, plausible, and satisfactory account of change, persistence, and passage that lacks the drawbacks of its competitors and ought, therefore, to be preferred to them.

Conclusion A seemingly simple question about the nature of temporal predicates, answered through a consideration of the puzzle of change, has led to a number of interesting conclusions. First, that any predicate of a material object that indicates a changeable feature of that object expresses a relation between the object and a time. Second, that the logical form of predications of time, tense, and change is tenseless. Third, that the passage of time is real and also tenseless, describable by relations whose extensions are temporally invariant. In sum, in so far as one believes in genuine change and temporal passage, one should accept the tenseless, B-theoretic world-view.16

124

DEBATES IN THE METAPHYSICS OF TIME

Notes   1 For simplicity I shall let two-place predicates stand for relational predicates in general; the difference between two-place and higher-order relational predicates is of no relevance here.   2 Following McTaggart, I will assume here that A-predicates order a set of events as an A-series; B-relations order a set of events as a B-series.   3 Savitt (2002) also points out the need to distinguish properties of a representation from properties of that which is represented.   4 What about abstract objects, such as geometric squares? Wouldn’t they have shape but not at any time? I think that the right thing to say is that geometric squares don’t have shape but, rather, are shapes. If, however, one were to insist that they have shapes, such shapes are essential to their objects and so cannot be temporary intrinsics and hence cannot be implicated in the puzzle of change.   5 Thanks to Donald Baxter for raising this issue.   6 Temporal parts theories (more on these below) explain how an object, x, is, say, red at t, by positing the existence of a red temporal part, p, that is located at t. Such accounts are, therefore, committed to the idea that some relations are existence entailing and are not, accordingly, in a position of relative advantage over the relational view with respect to the issue of explaining what it is for an object to exist/be located at a time.   7 It is possible, for example, that physics will discover that being red involves molecules on the surface of an object entering into quantum entanglement with particles on the far side of the galaxy.   8 It might be argued that the phenomenal form of temporal predicates such as “x is red” is monadic, expressing a non-relational property of the apple. Perhaps, but I have doubts. First, I am convinced by the foregoing that for an apple to be red is for it to stand in relation to time; so, when we notice an apple, we notice something that is related to time. Second, for an observer to notice the color of an apple is for her to stand in relation to time, so the experience itself is best understood as a relation between an observer and a time. I doubt, therefore, that there is a monadic phenomenal predicate or property available. Suppose, however, that there is. Then what follows, I think, is that this predicate constitutes an incorrect representation of the apple (rather than, say, of the experience of the apple). If our experience of the apple convinces us that “x is red” is non-relational, then our experience is, I believe, misleading. I am willing to accept that experience is misleading in this way, but I don’t think it gives us any reason to doubt the relational account of temporal predication anymore than the fact that experience convinces us that the sun rises gives us reason to doubt the heliocentric model of the solar system. Thanks to Nathan Oaklander for suggesting I clarify this point.   9 Indeed, it is hard to think of a more paradigmatically relational term than the preposition “at.” 10 If times are substantive entities whose existence is independent of that of any events, then B-relations relate substantive entities in general. If times



Temporal Predicates and the Passage of Time

125

are simultaneity classes of events, then B-relations will typically relate classes, though in some cases they will relate substantive entities such as objects to such classes. Either way, the logical analysis above goes through since the logical form of the relations is permissive with respect to the kinds of entities that satisfy them. Thanks to Nathan Oaklander for suggesting I clarify this. 11 Could temporal operators be functions from propositions to truth-values? Perhaps, but in that case propositions lack truth-values when considered independently of time. In other words, it would be impossible to express something determinately true or false about a material object’s properties without reference to time, in which case the essential features of the relational account will transfer. 12 Notable exceptions include Beer (1988) and Oaklander (1984, 2004). What follows is in broad agreement with these works; indeed, it is very much indebted to them. 13 In this respect, it is like the term “natural selection,” which is a general phrase that refers to the processes by which inherited traits result in differential rates of survival and reproduction in a given environment; there needn’t be any single process that occurs in all cases. 14 My principal goal in the foregoing is to defend the notion of temporal passage without appeal to monadic A-predicates or the concept of motion. Accordingly, I identify the passage of time with the existence of B-relations between times and events. Since B-relations share a logical form with other tenseless relations, such as “x is red at t,” it might be thought that my account is incomplete for it lacks an explanation as to what differentiates B-relations from other logically similar relations such that the former but not the latter determine that time has passed. At least part of what is required here is an account of the direction of time. One of the ways in which B-relations differ from, say, color relations is that the former are directed. Now, this is not the place to outline an account of the direction of time, but I think the view defended above has the strength of being compatible with various accounts of direction: however one fills in the theory of temporal asymmetry, the logical form of temporal relations will remain as outlined above (the pairs related by B-relations are ordered pairs, after all). Indeed, I think the account here is compatible with all manner of ontological views on the nature of B-relations, from nominalism—that is, the view that relations are identified with sets of ordered pairs—to more robust views in which the B-relation has ontological standing in addition to the relata. If, say, nominalism is correct, then it is an ontologically basic fact that some entities are ordered by the asymmetric “x is earlier than y” relation and others are not; nothing will be added by positing the existence of a relation in addition to the ordered pairs. Indeed, perhaps the asymmetry of the relation is primitive as well. If, on the other hand, an ontology that recognizes relations in addition to the relata turns out to be the best account of the temporal asymmetry, then the logical structure of the account defended above will not be overturned. So, I think that whichever way these debates turn out, the essence of the relational account can be preserved. Thanks to Nathan Oaklander for bringing these issues to my attention.

126

DEBATES IN THE METAPHYSICS OF TIME

15 To be precise, Craig defends the claim that only presentism can solve McTaggart’s paradox within the context of a tensed theory of time. He rejects a relational solution in part because it leads to a tenseless theory. I think, however, that his approach puts the cart before the horse: let us first give the best account of change and persistence and only thereafter worry about whether the world is tensed or tenseless, rather than accepting or rejecting solutions to the puzzle of change on the basis of a position on the tensed–tenseless debate. I recommend this in part because metaphysical theses such as presentism are substantial: they ask us to believe that events come into existence from nothing and then pass back into nothingness; they ask us to believe that propositions about the past can be true even if nothing past is real; and so on. Better, I propose, to solve philosophical puzzles prior to taking on such commitments, if at all possible. 16 Some of the material from this chapter was presented at Queen’s University, the University of Connecticut (Storrs), and a meeting of the Pacific Division of the American Philosophical Association. I would like to thank audience members, my APA commentator, Michael Tooley, and L. Nathan Oaklander for their suggestions and comments.

References Beer, M. (1988), “Temporal indexicals and the passage of time,” Philosophical Quarterly, 38, 158–64. Craig, W. L. (1998), “McTaggart’s paradox and the problem of temporary intrinsics,” Analysis, 58, 122–7. Dummett, M. (1978), “A Defence of McTaggart’s Argument Against the Reality of Time,” in Truth and Other Enigmas. Cambridge, MA: Harvard University Press, pp. 351–7. Dyke, H. (2008), Metaphysics and the Representational Fallacy. New York: Routledge. Hawley, K. (2001), How Things Persist. Oxford: Oxford University Press. Hinchliff, M. (1996), “The Puzzle of Change,” in J. E. Tomberlin (ed.), Philosophical Perspectives 10: Metaphysics. Oxford: Blackwell, pp. 119–36. Horwich, P. (1987), Asymmetries in Time. Cambridge, MA: MIT Press. Lewis, D. (1983), “Extrinsic properties,” Philosophical Studies, 44, 197–200. —(1986), On the Plurality of Worlds. Oxford: Blackwell. Markosian, N. (2004), “A Defense of Presentism,” in D. W. Zimmerman (ed.), Oxford Studies in Metaphysics, Vol. I. Oxford: Oxford University Press, pp. 47–82. McTaggart, J. M. E. (1908), “The unreality of time,” Mind, 17, 457–74. —(1927), The Nature of Existence, Vol. II. Cambridge: Cambridge University Press. Mellor, D. H. (1981), Real Time. Cambridge: Cambridge University Press. —(1998), Real Time II. London: Routledge. Mozersky, M. J. (2011), “Presentism,” in C. Callender (ed.), The Oxford Handbook of Philosophy of Time. Oxford: Oxford University Press, pp. 122–44. Oaklander, L. N. (1984), Temporal Relations and Temporal Becoming. Lanham, MD: University Press of America.



Temporal Predicates and the Passage of Time

127

—(2004), The Ontology of Time. Amherst, NY: Prometheus Books. Price, H., (1996), Time’s Arrow and Archimedes’ Point: New Directions for the Physics of Time. New York: Oxford University Press. Prior, A. N. (1996), “Some Free Thinking About Time,” in J. Copeland (ed.), Logic and Reality: Essays on the Legacy of Arthur Prior. Oxford: Oxford University Press, pp. 47–51. Prosser, Simon (2007), “Could we experience the passage of time?,” Ratio, 20, 1, 75–90. Savitt, S. (2002), “On Absolute Becoming and the Myth of Passage,” in C. Callender (ed.), Time, Reality and Experience. Cambridge: Cambridge University Press, pp. 153–67. Sider, T. (2001), Four-Dimensionalism: An Ontology of Persistence and Time. Oxford: Oxford University Press. Smart, J. J. C. (1949), “The river of time,” Mind, 58, 483–94. —(1963), Philosophy and Scientific Realism. London: Routledge & Kegan Paul. —(1980), “Time and Becoming,” in P. van Inwagen (ed.), Time and Cause: Essays Presented to Richard Taylor. Dordrecht: D. Reidel Publishing Company, pp. 3–15. Williams, D. C. (1951), “The myth of passage,” The Journal of Philosophy, 48, 457–72. Zimmerman, D. (2008), “The Privileged Present: Defending an ‘A-Theory’ of Time,” in T. Sider, J. Hawthorne and D. W. Zimmerman (eds), Contemporary Debates in Metaphysics. Malden, MA: Blackwell Publishing, pp. 211–25.

7 Physical Time, Phenomenal Time, and the Symmetry of Nature Michael Pelczar

Introduction

A

stream of consciousness is a series of experiences whose adjacent members relate to one another as my present state of mind relates to the state of mind I was just in, rather than as my first mental state this morning related to my last mental state last night, or as my present state of mind relates to the state of mind that you were just in. The exact nature of this relationship is debatable; here, I am not offering an analysis of the concept of a stream of consciousness, but merely drawing attention to the phenomenon that I wish to discuss. A typical human stream of consciousness extends over a period of hours, from one episode of dreamless sleep to the next. I wake up, eat breakfast, brush my teeth, take a shower, get dressed, walk to campus, check my e-mail, walk to the lecture theater, give a lecture, drink another cup of coffee, and check my e-mail again. I am conscious the whole time, and the experiences I have at each stage of my morning belong to a single stream of conscious experience. But there is, for me, no such thing as an experience as of waking up, eating breakfast, brushing my teeth, taking a shower, getting dressed, walking to campus, checking my e-mail, walking to the lecture theater, giving a lecture, drinking a cup of coffee, and then checking my e-mail again. I have no experience characterized by both shower-taking and lecture-giving

132

DEBATES IN THE METAPHYSICS OF TIME

phenomenology. The shower-taking qualia and the lecture-giving qualia all contribute to a single stream of consciousness, but not by contributing to a single experience—a single conscious state of mind.1 Innocent as it seems, the fact that the experiences that make up a stream of consciousness do not all belong to a single, extended experience gives rise to a serious paradox. But to see how, we first have to inquire more closely into the way in which successive experiences in a stream of consciousness relate to one another. Suppose you are standing before the mirror in your bathrobe, brushing your teeth, having the sort of experiences you normally do during this part of your morning routine, when suddenly you find yourself—or so it seems to you— chewing on a piece of beef jerky while riding horseback through a cactus forest. (Maybe your toothpaste is laced with LSD.) Overcoming your initial shock, you begin to suspect that you are dreaming, and decide to appreciate the desert landscape while it lasts. Now let’s modify the example. As before, you are brushing your teeth, and your conscious state of mind is what it usually is when doing so. This time, however, your tooth-brushing state of mind gets interrupted by a mental state phenomenally identical to the state of mind that Wyatt Earp was in on a certain October afternoon in 1881: same visual images of cactuses, same auditory sensations of jangling spurs, same cognitive and proprioceptive qualia, same “sense of self,” and so on. The mental state is not characterized by any feeling of puzzlement or surprise, since no such feeling characterized Earp’s state of mind on the afternoon in question. The sequence of experiences just described is not a stream of consciousness. Rather, it is as if you had been suddenly annihilated and someone having a totally different kind of experience created in your place. The horseback-riding experiences occur immediately after the tooth-brushing experiences, and are even generated by the same brain as the tooth-brushing experiences, but they are not, for all that, co-streamal with the tooth-brushing experiences. Why not? The reason is that there is no phenomenal overlap between successive experiences in the second scenario, whereas there is such overlap in the first. Two experiences phenomenally overlap just in case one is completely characterizable (in terms of its phenomenality) as an experience as of a sequence of states or events X … Y, and the other completely characterizable as an experience as of a sequence Y… Z (for some X, Y, and Z). When there is a phenomenal overlap between two experiences, the phenomenal content of one of the experiences is partially offset towards the future relative to the phenomenal content of the other experience. Let us call any series of experiences all of whose adjacent members phenomenally overlap a phenomenally integrated series. The experiences



Physical Time, Phenomenal Time

133

that you have in the first scenario, but not the second, form a phenomenally integrated series. In both scenarios, you begin with an experience of moving your toothbrush right then left. But in the second scenario, this experience is followed—not by an experience as of moving your brush left then chomping on jerky, but—by an experience that exhibits no toothbrushrelated phenomenology at all. In the second scenario, there is no Z such that you have (1) an experience as of brush-right then brush-left, and (2) an experience as of brush-left then Z. There is no experience in this scenario that possesses both tooth-brushing and horseback-riding qualia: none of the tooth-brushing states of mind contains any hint of horseback riding, and none of the horseback-riding mental states contains any trace of toothbrushing. That is why it is, phenomenologically, as if someone brushing his teeth had been annihilated and someone riding a horse created in his place.2

A paradox The fact that phenomenal integration is necessary for co-streamality poses a challenge to attempts to account for the stream of consciousness in a phenomenologically realistic way. Suppose that I am listening to someone play scales on the piano, and suppose (somewhat unrealistically) that the only conscious experiences I have while listening are auditory experiences of the various notes being struck. Naively, we might try to represent my stream of consciousness in this situation as follows: (1) (Do, Re, Mi, Fa, So, La, Ti) —where “Do” designates an experience as of the note C being struck, “Re” an experience as of D being struck, and so on. This representation is naive, because it does not capture the phenomenal integration of the series of experiences it attempts to represent. (1) might equally well represent the series of experiences I would have if I was first in a mental state indistinguishable from that of someone who hears only a solitary C, and then in a mental state indistinguishable from that of someone who hears only a solitary D, and then in a mental state indistinguishable from someone who hears only a solitary E, and so on. A seemingly natural fix is to represent my auditory stream of consciousness like this instead: (2) (Do…Re, Mi…Fa, So…La, Ti)

134

DEBATES IN THE METAPHYSICS OF TIME

—where “Do … Re” represents an experience of C followed by D, “Mi … Fa” an experience of E followed by F, and so on. But this will not do either. For while it is true that my stream included an experience of C followed by D as well as an experience of E followed by F, it also included an experience of D followed by E; but this latter experience goes unrepresented here. Very well then: let’s just insert a new element into the series, to represent the D-followed-by-E experience (as well as the other, similar experiences that (2) omits): (3) (Do…Re, Re…Mi, Mi…Fa, Fa…So, So…La, La…Ti) Problem solved? No. What we have now represented is not the stream of conscious experience that I had while listening to the scales, but the stream that I would have had if I had heard someone play the following sequence of notes: CDDEEFFGGAAB. In fact, (3) doesn’t even succeed in representing that (counterfactual) stream! This is for the same reason as (1) fails to represent my actual stream: (3) fails to capture the phenomenal integration of the series of experiences I would have if I heard someone playing the “stutter” scale. To get an adequate representation of the stutter-scale stream, we could try: (4) (Do…Re, Re…Re, Re…Mi, Mi…Mi, Mi…Fa, Fa…Fa, Fa…So, So…So, So…La, La…La…, La…Ti) It’s clear where this is going. In order for the experiences I have while hearing the scales to belong to a single a stream of consciousness, they must constitute a phenomenally integrated series. But in order to constitute a phenomenally integrated series, it seems my experiences must include many that I simply do not have—and even if we suppose that I did have such experiences, we still wouldn’t have a stream of consciousness, without positing yet further experiences, ad infinitum. Of course it would be different if, while listening to the scales, I had just a single, lengthy experience: an experience as of C followed by D followed by E followed by F followed by G followed by A followed by B. But I had no such experience. (Or, if I did, we need only consider a longer stretch of time over which I listen to the player play the scales repeatedly, perhaps for an hour.) And so we have a paradox. In order for our experiences to occur as parts of (lengthy) streams of consciousness, our experiences must be phenomenally integrated, without collectively constituting a single (lengthy) experience. But when we try to describe a stream of consciousness in a way that respects these constraints, we end up badly misdescribing it as including many experiences that it does not, in fact, include.



Physical Time, Phenomenal Time

135

This is what Barry Dainton calls the problem of repeated contents. The problem is as important in its own way as the famous Eleatic paradoxes of motion and change. There is even a structural similarity between the paradoxes. What generates the Arrow Paradox, for example, is the seeming need for the arrow to traverse infinitely many intervals of time and space in order to make any progress along its path. What generates the paradox of the stream of consciousness is the seeming need for the mind to traverse infinitely many experiences in order to make any progress through its stream of consciousness.3 Grappling with Zeno’s paradoxes has led to valuable insights into the nature of time, space, and the infinite. As we shall see, grappling with the paradox of the stream of consciousness stands to yield valuable clues to the relationship between spacetime and experience.

Dainton’s theory Dainton solves the problem of repeated contents by supposing that the pairwise phenomenally overlapping experiences that make up a stream of consciousness also overlap in the literal sense that they have experiential parts in common. For example, according to Dainton, my Do … Re experience has two briefer experiences as parts: (a) an auditory experience as of C, and (b) a slightly later auditory experience as of D. Similarly, my Re … Mi experience comprises (c) an experience as of note D being played, and (d) a slightly later experience as of E being played. But—and this is the key idea behind Dainton’s theory—experience (b) is one and the same experience as experience (c). So the Do … Re experience and the Re … Mi experience have between them only three basic constituents: an auditory “Do” experience, an auditory “Re” experience, and an auditory “Mi” experience. Since I do not, by this account, have two experiences as of “Re,” the account escapes the problem of repeated contents.4 Dainton’s solution to the problem of repeated contents relies on the idea that some of our experiences are extended in time, consisting of successive, briefer, and (ultimately) temporally basic experiences none of which have successive experiences as parts. In Dainton’s terms, a stream of consciousness consists of overlapping sequences of “diachronically co-conscious” experiences, diachronic co-consciousness being the relation by virtue of standing in which a number of successive experiences constitute a longer, complex experience.5 Dainton’s proposal conflicts with what Dainton (following Izchak Miller) calls “the Principle of Simultaneous Awareness” or PSA. This is the view that

136

DEBATES IN THE METAPHYSICS OF TIME

each of our conscious experiences is temporally basic, and in fact strictly momentary, having no objective duration whatsoever.6 While the PSA has a certain initial plausibility, there are, Dainton argues, no obviously compelling reasons to accept it. A less extreme view is that some, but not necessarily all, of our conscious experiences, including presumably our simplest or most basic experiences, are as the PSA says (perhaps erroneously) all experiences are. It may be that consciousness is wholly present at some moments of my waking life, but that I also have extended experiences comprising the contents of successive waking moments. This position is consistent both with the PSA and with the existence of diachronically co-conscious experiences.7 I believe that this more moderate view—that our temporally basic experiences have no objective temporal extent—is true, and that its truth has unexpected implications for Dainton’s theory of the stream of consciousness.8 From here I proceed as follows. In the next secion, I give an argument (based on some remarks from Bertrand Russell) in support of the more moderate view described above, which I shall call Russell’s Thesis. In the section after that I anticipate and address the main objection to this argument. I will follow this by arguing that Russell’s Thesis, when combined with Dainton’s theory of the stream of consciousness, entails a violation of an otherwise universally observed symmetry of nature connected with the so-called reversibility of fundamental natural laws.

Russell’s Thesis In a paper from 1914, Russell writes: Two events which are simultaneous in my experience may be spatially separate in psychical space, e.g. when I see two stars at once. But in physical space these two events are not separated, and indeed they occur in the same place in space-time. Thus in this respect relativity theory has complicated the relation between perception and physics.9 Here Russell claims that the phenomenal events that characterize any moment of one’s conscious mental life are confined to a single point of spacetime; let us call this Russell’s Thesis. While Russell’s argument for the thesis is not entirely explicit, it clearly has something to do with the relativistic structure of spacetime. In a relativistic context, if it is possible correctly to describe a given pair of events as occurring simultaneously at some distance from one another (in terms of a given inertial



Physical Time, Phenomenal Time

137

coordinate system or “frame of reference”), it is also possible to describe those events as occurring in temporal succession (in terms of a different but equally good frame of reference). This is in contrast to classical Newtonian physics, in which two spatially separated events are simultaneous in terms of one frame of reference only if they are simultaneous in terms of every frame of reference. We can put this by saying that simultaneity is “absolute” in a Newtonian context, but not, except in special cases, in a relativistic context. In relativistic spacetime, two events occur absolutely simultaneously (i.e. are assigned identical time coordinates by every inertial coordinate system) only if they occur not only at the same point in time, but at the same point in space. I reconstruct Russell’s argument for his thesis as follows: R1. If consciousness is in spacetime, each of my temporally basic experiences instantiates its qualia absolutely simultaneously. R2. Properties that get instantiated absolutely simultaneously get instantiated at the same point of spacetime. R3. So, if consciousness is in spacetime, each of my temporally basic experiences instantiates its qualia at a single point of spacetime. (R1, R2) R4.  Any experience that instantiates its qualia at a single point of spacetime is a point-event (an event confined to a single point of spacetime). R5. So, if consciousness is in spacetime, each of my temporally basic experiences is a point-event. (R3, R4) Let us call this the Relativistic Argument. Russell seems to regard its first premise as evident from introspection. There is certainly something to this. When I look at a banana, I have a conscious experience that is simultaneously characterized by a certain phenomenal shape, a certain phenomenal size, and a certain phenomenal color. My experience’s simultaneous possession of these phenomenal properties is moreover absolute, in the sense that the experience instantiates the properties simultaneously according to every complete and accurate description of the world. If you were to describe me as having first an experience as of the bananas’s shape, followed by an experience as of its size, and only then an experience as of its color, you would badly misdescribe my experience of the banana. R2 says that absolutely simultaneous property instantiations are confined to a single point of spacetime. As Russell points out, this is a direct consequence of the fact that in relativistic spacetime, absolute simultaneity entails spatial and temporal co-location. There are two ways you might try to dispute this fact (short of denying the relativistic structure of spacetime). First, you might argue that conscious

138

DEBATES IN THE METAPHYSICS OF TIME

experience occurs in physical time, but not in physical space, and therefore not in spacetime (and not, a fortiori, in relativistic spacetime). (By “physical time” I mean the time in which physical events take place; it may be that there are non-physical events that occur in physical time.) But an event (whether an experience or anything else) cannot occur in physical time except by occurring in physical spacetime. A description of events purely in terms of physical time is an underdescription of events, not only because it does not tell us how the events relate to one another in space, but also (and partly because of this) because it does not tell us which events occur absolutely before, after, or simultaneously with which others. The statement that event E1 occurs at the same time as event E2 does not convey whether the simultaneity of E1 and E2 is absolute or merely relative; for that, we need to know whether E1 and E2 occurred at the same place (in addition to the same time). A description of physical events in purely temporal terms therefore fails to specify any absolute temporal structure for those events. To say that experiences occur in physical time but not in physical space would therefore be to claim, implicitly, that there was no fact of the matter concerning the temporal relationships between our conscious experiences and ordinary physical events. But this would be as much as to say that conscious experience does not occur in physical time at all. A different objection to R2 is that conscious experiences might occur in physical spacetime, but in a non-relativistic way. Granted that our experiences occur in spacetime along with physical events, why should we think that the relativistic conception of spacetime applies to our experiences? But if phenomenal events occur in the same spacetime as physical events, it must be possible for a phenomenal event to coincide with a physical event. This is particularly so in view of the fact that physical events occur throughout spacetime (e.g. as components of the cosmic microwave background). Consider two spatially separated, and supposedly absolutely simultaneous, phenomenal events, A1 and A2, and a pair of physical events, B1 and B2, with which the phenomenal events respectively coincide. Then A1 is absolutely simultaneous with B1, and A2 with B2. But then it must also be the case that B1 is absolutely simultaneous with B2, since A1 is absolutely simultaneous with A2, and absolute simultaneity is transitive. The upshot is that if spatially separated phenomenal events can be absolutely simultaneous, spatially separated physical events can be absolutely simultaneous too. Since separate physical events cannot be absolutely simultaneous, neither can separate phenomenal events. The final premise of the Relativistic Argument (R4) states that any co-instantiation of qualia at a point in spacetime is a conscious experience. This is just an a priori truth about consciousness. It can be expressed as a



Physical Time, Phenomenal Time

139

conditional: if some qualia are co-instantiated at a point in spacetime (or by the contents of a point in spacetime), then that point contains conscious experience. One should accept this conditional, even if one rejects its antecedent.

The now of experience If the Relativistic Argument has a vulnerability, it is R1: the claim that a temporally basic experience instantiates its qualia absolutely simultaneously (assuming that temporally basic experiences occur in time). One cannot very well object to R1 that a temporally basic experience might instantiate some (or all) of its qualia in temporal succession. Any experience that instantiates qualia successively is, de facto, not temporally basic, since it has non-simultaneous experiences as parts or phases (corresponding to the successive qualia instantiations). The more serious objection to R1 is that it relies on a dichotomy, allegedly false, between (1) some qualia getting instantiated in temporal succession, and (2) those qualia getting instantiated absolutely simultaneously (and so at a single point of spacetime). There is, one might claim, a third option: (3) the qualia get instantiated by the same objective temporal sequence of point-events, not by successive point-events, and not by a single, solitary point-event. Call option (3) serialism. A serialist disputes R1 on the grounds that every experience is a temporal sequence of intrinsically non-experiential events. He claims that the instantiators of qualia are not point-events or the contents of individual moments of time, but whole temporal series of point-events. Serialism is antithetical to what I would like to call the Presence Principle. This is the principle (tacitly assumed by Russell in his discussion) that in order for consciousness to exist in time (or spacetime), there must be moments (at least one moment) whose contents are sufficient for conscious experience—that is, whose contents are such that there is something that it’s like for someone, or something, for those contents to exist. According to the Presence Principle, consciousness is sometimes (indeed, often) wholly present; according to serialism, consciousness is never wholly present. Which is right? A gladiator lies in the dust. He has, let us suppose, a conscious visual experience as of the Colosseum: the flying banners, the jeering throng, the Emperor’s wavering thumb. Call this experience V. According to the serialist, V is an instantiation of qualia by some objective temporal sequence of

140

DEBATES IN THE METAPHYSICS OF TIME

non-experiential events, say e1, e2, and e3. But what, according to the serialist, is the gladiator’s state of mind shortly before e3 occurs—say, when e2 takes place? Suppose that the gladiator had died at e2, or between between e2 and e3. Would he have had V? Not according to the serialist: according to him, the e1-e2-e3 event-sequence is what instantiates the qualia that characterize V (if anything does). But this event-sequence does not exist in a hypothetical scenario in which the gladiator does not live long enough for e3 to occur. So, the serialist has to say that the gladiator does not, in that hypothetical scenario, have V, and therefore does not have the same quantity and quality of conscious experience as the actual gladiator (who makes it past e3 and who does have V ). But now suppose that at or around the time that e2 takes place, but before e3, the gladiator poses himself the question: Am I having V?—or, as he might more naturally put it: Am I having this experience? How should he answer himself? Assuming that e3 does eventually occur, he does have V. But when he asks himself whether he is having V, he might have reason to believe that e3 will not occur (he knows the Emperor has always given the thumbs-down in the past). On this basis, he will judge that he is not having V—he will judge that he is not having the experience, even though he is having the experience. This is odd. But the serialist may argue that in reality, there is not enough time between e2 and e3—or for that matter within the whole [e1, e3] interval— for the gladiator to pose himself any question, or perform any other mental act. Our thought processes unfold at a slower pace than our conscious visual processes, and this prevents us from ever getting into the dubious cognitive condition attributed to the gladiator in the preceding paragraph. But even if evolution has not equipped us with cognitive machinery fast enough to form thoughts between the successive events that constitute an experience (on the serialist view), it might have done so; or, if not, we might bring our cognitive machinery up to speed by artificial means. Suppose that the gladiator’s cognitive capacities are greatly enhanced. Suppose that his cognitive centers gain high-bandwidth access to real-time information about the low-level states of his perceptual centers. And suppose that he benefits from cognitive enhancements that allow him to process this information as it streams in. His experience-producing perceptual centers themselves are left alone. This fast-thinking gladiator can make deliberate judgments between e2 and e3. So what does he judge then? Presumably this depends on what information he has. But if he has information implying that it is unlikely that e3 will occur, then, if he is rational, he must (assuming the truth of serialism) judge that he is not having V, even though he is. One might insist that all this cognitive enhancement (high-bandwidth access to microphysical brain-activity, high-speed information processing



Physical Time, Phenomenal Time

141

capabilities, and so on) would have phenomenological repercussions. That is probably right; at least, one would expect the cognitive enhancement to come with some novel cognitive phenomenology, and possibly novel perceptual or proprioceptive phenomenology too. But this additional phenomenology need not prevent the gladiator from having the conscious experiences he would have had without the enhancement, or experiences much like them. For example, it need not prevent the gladiator from having more or less ordinary visual experience. Serialism therefore appears to entail that it is possible to become rationally less confident that you are having conscious experience, or a given kind of conscious experience, just by acquiring additional information about a conscious experience that you are, in fact, having. But this is unacceptable. Gaining more information about a conscious experience that you are having cannot make you rationally less confident that you are having a conscious experience of that kind. The fundamental argument for the Presence Principle (and against serialism), and the linchpin of the case for Russell’s Thesis, is that if you reject the Principle, you must say, wrongly, that one can become rationally less confident that one is having (say) visual experience, just by gaining more information about some visual experience that one is, in fact, having.

The law of experience Our conscious mental lives are teeming with a huge variety of experiences. But within this teeming variety we can discern a considerable amount of order. By and large, it is possible to interpret our experiences as perceptions of a universe of objects behaving, and events unfolding, in accordance with certain laws—the laws of physics. The possibility of interpreting our experiences this way is more obvious in the case of some laws than in others. We can hardly help interpreting our experiences as including perceptions of spatiotemporal continuants. If we think of the existence of such continuants as a law of nature, we can say that our experiences are obviously interpretable as perceptions of a world that conforms to this law. Our experiences are also interpretable as perceptions of a world that conforms to the Einstein Field Equations. The possibility of interpreting them thus is far less obvious—otherwise, it would not have taken people so long to interpret them that way. Between these two extremes, we have the possibility of interpreting our experiences as perceptions of a world in which an object’s resistance to

142

DEBATES IN THE METAPHYSICS OF TIME

change in its state of motion upon application of some force is independent of where, when, or in what direction the force is applied. And there are many other intermediate cases as well, some closer to the law of continuants, some closer to the Field Equations. Not all experiences are interpretable as conforming to all, or perhaps even to any, of these laws. Pathological experiences are possible, which, considered in isolation, resist interpretation as perceptions of any lawlike state of affairs. Still, taken all together as a collective whole, our experiences do admit of interpretation as perceptions of a world that conforms to such laws as those mentioned above. We can think of this property of human experiences as corresponding to a law in its own right, a “law of experience.” Our experiences conform to this law by tending to occur in such a way as to admit of interpretation as perceptions of a universe that conforms to those other laws. The law of experience is undoubtedly a law. But is it a fundamental law? It is hard to see how it could not be, unless phenomenological states of affairs reduce to purely physical states of affairs. If dualism is true, the law of experience is bound to be fundamental (even if it correlates with some purely physical law or laws), and if phenomenalism is true, the law of experience is likely to be not just fundamental but architectonically central. But if physicalism is true, then presumably the order we find in our conscious experience is not fundamental, but reduces to some purely physical feature of the world. Like Dainton, I am deeply skeptical of physicalism. I will not rehearse the well-known arguments against physicalism here, but only point out that Russell’s Thesis poses an additional problem for the physicalist position.10 In a traditional conceptual setting, where we think of time and its determinations as absolute, the physicalist would naturally identify temporally basic experiences with complexes of simultaneous physical events—say, neuronscale events occurring simultaneously within a human brain. From here, he might build up to more complex experiences, but at the most basic level, one would normally expect a physicalistic experience to consist of a collection of simultaneous physical events distributed throughout some spatially extended network. But if Russell’s Thesis is correct, then many of our most basic experiences are mere point-events. The traditional physicalist picture is therefore untenable. (I take it that this is why Russell says that “relativity theory complicates the relation between perception and physics.”) A physicalist might argue for an identification of temporally basic experiences with individual physical point-events. The connection of these individual point-events to larger-scale physiological activity would have to be spelled out, but as long as there is some exceptionless correlation between a given kind of physical point-event and a given kind of phenomenally-individuated



Physical Time, Phenomenal Time

143

temporally basic experience, there is a basis (a physicalist might argue) for equating the two. However, this proposal requires us to accept that the world contains far more experience than we ordinarily suppose. Any physical point-event that occurs in the body of a sentient being can also occur outside of any such body. What is distinctive about, for example, a human body, is not that it is made up of distinctive sub-atomic events, but that the sub-atomic events that constitute it relate to one another so as to form distinctive physical states and processes extended in time and space. So if we are to identify various experiences with physical point-events, we will have to identify them with events that occur, or could occur, in rocks, plants, sausages, and other prima facie unconscious things. Furthermore, it is likely that for every sub-atomic physical event that occurs in the body of a bat (or any other creature), there occurs an exactly similar sub-atomic event in my body. Here, the implication of the physicalist’s proposal is that there occur in my body the same kinds of temporally basic experiences as occur in the bodies of bats and other conscious animals.11 The appeal of physicalism was always its promise to make sense of consciousness as something that explains various aspects of our behavior, by construing conscious mental states as mediators of environmental input and behavioral output. A theory that attributes consciousness to sausages cannot possibly deliver on this promise. Setting aside physicalism, it seems appropriate to regard the law of experience as a fundamental law, like the fundamental laws of physics. Now, the fundamental laws of physics all exhibit a certain symmetry. In classical Newtonian physics, the fundamental laws of motion are timereversal invariant, in the sense that they would also hold good in a time-reversed counterpart of our world, in which events occurred in the opposite temporal order from that in which they actually occur, and with opposite time-dependent properties from those which they actually possess. The situation is different, but similar, in contemporary physics. There are, it turns out, physical processes governed by laws that are not time-reversal invariant. A time-reversed counterpart of our universe would not conform to all the fundamental laws that govern our (actual) universe, since it would violate the fundamental laws that determine the probabilities for various forms of elementary particle decay.12 Nevertheless, the fundamental laws are invariant under a more inclusive reversal of time, spatial position, and charge. In other words, even though a merely time-reversed counterpart of our universe would violate certain fundamental laws, a counterpart of our universe that was reversed with respect to time (as described earlier), space (as in a mirror reflection), and charge (with positive charges replacing negative charges and vice versa) would obey the same fundamental laws as the actual world obeys.

144

DEBATES IN THE METAPHYSICS OF TIME

In so far as our universe is governed by the fundamental laws of physics, it is symmetrical in this way; the fundamental laws of physics are CPT-reversal invariant. What about the law of experience? If, as Dainton contends, our streams of consciousness consist of diachronically complex experiences (comprising simpler diachronically co-conscious experiences), the law of experience is not CPT-reversal invariant. The reasons for this are as follows. Let us call the CPT-reversed counterpart of the actual world “Unworld.” Unworld contains all the same temporally basic experiences as the actual world, since each of these experiences is a mere point-event. If there actually occurs a temporally basic experience as of a bird gliding from left to right, then the same experience occurs in Unworld. The situation would be different if temporally basic experiences were, as the serialist maintains, temporally extended sequences of non-experiential events; in that case, one could argue that a reversal in the temporal order of those non-experiential events would involve a corresponding reversal of the phenomenology of the experience that those events constituted. But if, as Russell’s Thesis states, temporally basic experiences are confined to individual points of spacetime, they have no temporal structure to reverse. Now, in Unworld, our temporally basic experiences occur in the opposite temporal order from that in which they actually occur in us. Suppose I am watching a pendulum. And suppose that while watching, I have, among others, a diachronically complex experience as of the pendulum moving from A to B to C to B to A. This experience, we may assume, consists of four temporally basic appearances: (1) an appearance of the pendulum swinging from A to B, (2) an appearance of the pendulum swinging from B to C, (3) an appearance of the pendulum swinging from C to B, and (4) an appearance of the pendulum swinging from B to A (see Figure 7.1). In Unworld, the temporal order of these four experiences is reversed. Therefore, if I have a diachronically complex experience consisting of them, it is an experience as of the pendulum moving first from B to A, then from C to B, then from B to C, then from A to B. Such an experience, considered by itself, might not be an experience as of a physically impossible sequence of events. One could imagine that the particles making up the pendulum decohere and recohere in astronomically unlikely, but physically possible, ways. But in Unworld, all diachronically complex experiences—which means very many of the Unworlders’ experiences—are like the bizarre pendular experience just described. If, in the actual world, I have a diachronically complex experience as of a ball rolling down an inclined plane from point G through point H to point I—an experience consisting of two temporally basic experiences (as of the ball rolling from G to H, and as of the ball rolling from H to I), my counterpart in Unworld has



Physical Time, Phenomenal Time

145

FIGURE 7.1 an experience as of the ball rolling from H to I and then—without any intermission—from G to H. Considered as a whole, the Unworld experiences do not admit of interpretation as perceptions of a world of physical things conforming to the laws of physics. As the foregoing examples illustrate, this includes even very simple laws, such as those dictating that physical objects enjoy spatiotemporally continuous existences. Of course, Unworld does obey the laws of physics. In Unworld, pendula, balls, and all other things obey fundamental physical laws. But if, as Dainton’s theory requires, our streams of consciousness consist of sequences of overlapping temporally extended experiences having successive temporally basic experiences as constituents, Unworld does not conform to the law of experience. Dainton must therefore regard the law of experience as a fundamental law of nature that, unlike other fundamental laws, violates CPT-reversal invariance.

Conclusion I conclude that while Dainton’s theory of the stream of consciousness is consistent with Russell’s Thesis (that our temporally basic experiences are wholly present at individual points of spacetime), it combines with Russell’s Thesis to violate an otherwise universally observed symmetry of nature. One response would be to conclude that CPT-reversal invariance is not, after all, sine qua non for fundamental lawhood. It’s not as if this sort of thing hasn’t happened before. People used to think that all fundamental laws were time-reversal invariant. Dainton might reason that since experiences must occur in time, and since the only way they can do so without impossible phenomenal repetition is by occurring as his theory describes, and since,

146

DEBATES IN THE METAPHYSICS OF TIME

finally, his theory implies a violation of CPT-reversal invariance, we must conclude that CPT-reversal invariance is violated. On the other hand, if we want to say that the stream of consciousness unfolds in objective, physical time without violating CPT-reversal invariance, we have no alternative but to affirm the Principle of Simultaneous Awareness. If all of our experiences are temporally basic, then there need be no violation of time- or CPT-reversal invariance. We need only suppose that the objective order of experiences in a stream of consciousness is phenomenologically meaningless. And this is not an obviously unreasonable thing to say. For surely the really remarkable thing about conscious experience, apart from its very existence, is that it is possible to interpret the totality of it as comprising, by and large, perceptions of a common, intelligible world of events in time and space. The idea that the objective temporal order of the experiences in a stream of consciousness has no phenomenal implications may seem strange, or even absurd. But here we should remember that we never introspect more than one experience in a single act of introspection, unless it is by introspecting multiple experiences that occur as phenomenal constituents of a single, complex experience. The basic point is due to Susan Hurley. If I can introspect multiple experiences in a single act of introspection, those experiences must belong to a single experience. What would it be like, for introspection to reveal to me that I had two experiences that did not belong to the same state of consciousness? The answer is that there can be nothing that this is like. For, if there were something it was like, then the introspected experiences would belong to a single experience: an experience by virtue of having which there was something it was like for me to introspect myself as having the two experiences.13 Unfortunately, the suggestion that all experiences are temporally basic takes us back to square one with the stream of consciousness, since you can’t construct a Daintonian stream with temporally basic experiences alone, and Daintonian streams are the only ones described so far that avoid the problem of repeated contents. A more radical way to preserve CPT-reversal invariance that does not lead back to the problem of repeated contents would be to suppose that conscious experiences do not occur in time, or spacetime, at all, but function more like an atemporal subvenient base for physical reality (the reality of things in physical time and space). After all, if the temporal order of our experiences is phenomenologically meaningless, why suppose that they have any temporal order at all? And if experiences do not even exist in time, then repeated contents are impossible ab initio. Of course, it is far from obvious that a timeless conception of experience ultimately makes sense; this is a question too large to pursue here.



Physical Time, Phenomenal Time

147

Dainton’s theory of the stream of consciousness has the great virtue of solving the problem of repeated contents, being apparently the only theory, among those that construe experiences as occurring in physical time, to do so. The upshot of the present discussion is that the solution does not come for free. Diachronic co-consciousness is not a minor technical innovation, but a weighty metaphysical posit that requires us to see consciousness as extending through time differently from all other temporal phenomena. How this figures into the ultimate cost-benefit analysis of Dainton’s theory remains, I think, uncertain, depending on large questions about the more general relationship between time and consciousness. But whichever way the balance of considerations finally tips, it is Dainton who gets the credit for showing us how to work the scales.

Notes   1 The word “experience” has many meanings, and it might be that in one of them, the statement that all my day’s sensations belong to a single experience comes out true; presumably Michael Tye is using the word in some such sense when he claims that all of the experiences in a stream of consciousness belong to a single, temporally extended experience (Tye 2003: 108). In any event, it is, I take it, uncontroversial that the qualia that characterize the various phases of an afternoon of conscious mental life do not characterize a single “experience” in the same sense as that in which the various qualia that characterize each phase individually do so, and it is this latter sense of “experience” that is relevant to our purposes here (see Dainton 2008: 71–3).   2 Phenomenal integration is necessary, but not sufficient, for co-streamality. If I have a stream of consciousness consisting of successive experiences I1, I2, I3, … In, and my counterpart on Twin Earth has a phenomenologically indistinguishable stream consisting of experiences T1, T2, T3, … Tn, then the series I1, T2, I3, T4, … In–1, Tn is phenomenally integrated, but it isn’t a stream of consciousness.   3 See Dainton (2006: 141–2, 156–9).   4 See Dainton (2006: 162–77).   5 Might all conscious experiences consist of briefer successive experiences? That is, might consciousness consist of temporally extended phenomenal “gunk”? Elsewhere I argue not; here, I simply assume that human experience has a logically atomic structure.   6 See Dainton (2006: 131–5). Past proponents of the PSA include, apparently, St. Augustine, William James, C. D. Broad, Edmund Husserl, and Michael Lockwood.   7 For critical discussion of the PSA, see Dainton (2006: 162–6, 179–82).   8 By a “temporally basic experience,” I mean an experience that doesn’t have non-simultaneous experiences as parts.

148

DEBATES IN THE METAPHYSICS OF TIME

  9 Russell (1914: 130). 10 For classic arguments against physicalism, see Chapter 2 of Broad (1925), Campbell (1970: 100–1), Jackson (1982), and Chalmers (1996: 93–140). For Dainton’s skepticism, see Dainton (2006: 5–10). 11 It is also not obvious that there is enough variety among physical pointevents to reflect the phenomenal variety of temporally basic experiences. This is connected with the so-called “grain problem”: see Lockwood (1993: 275–6). 12 The experiments that originally established the violation of time-reversal invariance are described in Lee and Yang (1956) and Christenson et al. (1964). 13 See Hurley (1998: 164–5), and for relevant discussion, Bayne and Chalmers (2010: 510–11, 527–9).

References Bayne, T. and Chalmers, D. (2010), “What is the Unity of Consciousness?,” in The Character of Consciousness. Oxford: Oxford University Press, pp. 497–539. Broad, C. D. (1925), Mind and Its Place in Nature. New York: Harcourt, Brace & Co. Campbell, K. (1970), Body and Mind. London and Toronto: Macmillan. Chalmers, D. (1996), The Conscious Mind: In Search of a Fundamental Theory. Oxford: Oxford University Press. Christenson, J. H., Cronin, J. W., Fitch, V. L. and Turlay, R. (1964), “Evidence for the 2π decay of the K2 0 meson,” Physical Review Letters, 13, 4. Dainton, Barry (2006), Stream of Consciousness: Unity and Continuity in Conscious Experience. New York: Routledge. —(2008). The Phenomenal Self. Oxford: Oxford University Press. Hurley, S. L. (1998), Consciousness in Action. Cambridge, MA: Harvard University Press. Jackson, F. (1982), “Epiphenomenal qualia,” The Philosophical Quarterly, 32, 127, 23–36. Lee, T. D. and Yang, C. N. (1956), “Question of parity conservation in weak interactions,” Physical Review, 104, 1, 254–8. Lockwood, M. (1993), “The Grain Problem,” in Howard Robinson (ed.), Objections to Physicalism. Oxford: Clarendon Press, pp. 271–91. Russell, B. (1914), “The World of Physics and the World of Sense,” in Our Knowledge of the External World. London: George Allen & Unwin, pp. 106–34. Tye, M. (2003), Consciousness and Persons. Cambridge, MA: MIT Press.

8 Extensionalism, Atomism, and Continuity Geoffrey Lee

Introduction: Extensionalism and atomism

A

temporal experience is an experience that presents to its subject states of affairs that manifestly involve duration and change over time, such as the temporal order of sounds, the velocity of moving objects, or the duration of a brief flash of light in the visual field. There is a disagreement about temporal experiences between extensionalists, atomists, and snapshot theorists, which will be my subject here. According to extensionalists like Barry Dainton, temporal experiences are themselves temporally extended processes that play out over time, having experiential stages that mirror the stages of the presented events (see Figure 8.1). For example, an extensionalist will typically think that experiencing the temporal order of two sounds S1 and S2 involves first experiencing S1, then experiencing S2, and a suitable relation holding between these experiences. By contrast, atomists think that temporal experiences do not themselves have temporal structure that mirrors the temporal structure represented in their content. On this view, an experience of a temporallly structured state of affairs like the temporal order of two sounds is itself temporally unstructured, in that it does not have distinct experiences as proper temporal parts; for example, a temporal order experience does not involve first experiencing one event and then experiencing a second event (see Figure 8.2); it involves experiencing both events at the same time, even though they are experienced as happening at different times. For this reason, atomic temporal experiences might even be instantaneous, although atomists needn’t think this, as I will explain below.

150

DEBATES IN THE METAPHYSICS OF TIME

FIGURE 8.1   The Extensionalist View

FIGURE 8.2   The Atomist View Extensionalists and atomists both disagree with snapshot theorists, who deny that we have temporal experiences. On their view, the stream of consciousness is a series of photo-like presentations of properties of objects like their shape, spatial arrangement, surface texture, color or illumination, presentations that do not concern how the world is changing over time. In so far as we are able to make accurate judgments about temporal features, this involves a post-experiential process of comparing the snapshots, a process that doesn’t require actually experiencing the temporal features. I will set aside the snapshot view here (for a defense of it, see Chuard 2011), as it is incompatible with obvious phenomenological data. Take auditory experience, for example— most of the information in it pertains to how events are organized over time; it is very hard to make sense of auditory experience without temporal content. My aim in this chapter is to defend atomism, by responding to a number of arguments against it that appear in the literature, particularly those given by Dainton (2000, 2010) and Phillips (2010). Phillips (2010) argues that



Extensionalism, Atomism, and Continuity

151

atomism is incompatible with the kind of introspective knowledge we have of how experience is changing over time. Dainton argues that atomists can’t adequately capture the sense in which the stream of consciousness is continuous and connected over time. I will respond to these arguments, among others. Elsewhere Lee (2014a) I give a detailed positive argument for atomism and against extensionalism, which I will also briefly recap here. Lee (2014a) also contains a detailed discussion of what exactly is at issue between extensionalists and atomists. Here I will state more briefly what I see as the main issues. It is helpful to distinguish three closely related but distinct ideas that extensionalists may or may not subscribe to. In increasing order of strength, these are: The process view: Temporal experiences are experiential processes that unfold over time by having shorter experiences as proper temporal parts.1 The mirroring view: The process view, with the additional claim that the experiential temporal parts of a temporal experience are arranged in time in a way that mirrors or matches the temporal structure represented in their content. The representation by resemblance view: The mirroring view, with the additional claim that the correspondence between the objective temporal structure of experience and the temporal structure in the experience’s content is explained by the fact that temporal properties are represented by resemblance—time is used to represent time. I treat experiences as instantiations of experiential properties by subjects (or parts of subjects, like brain areas). So an extensional temporal experience is a process involving distinct property instantiations at different times. In many cases, these will be qualitatively distinct, involving different experiential properties, for example when you experience an object moving or changing illumination, or hear the temporal relations between different sounds. However, we can also have distinct instantiations of the same experiential property as different temporal stages; for example, this is presumably what the extensionalist will say is happening when you have a qualitatively constant experience as of an unchanging scene. Not any old series of experiences can make an extensional temporal experience. The process theorist thinks that the stages of a temporal experience have to be unified, or “co-conscious,” in order to form an experience. So, for example, if I experience one event, then undergo a complete brain reset, and then experience another event, I will not have a unified experience as of

152

DEBATES IN THE METAPHYSICS OF TIME

the temporal order of the events. What the relevant kind of “unity” is here is a difficult question. Dainton assumes that this diachronic unity relation is (at least) the same unity relation that holds synchronically between different parts of a subject’s field of awareness: they are “experienced together” in a way that two random experiences enjoyed by different subjects are not. It is controversial even in the synchronic case what the unity relation is (or even whether there is a single relation here to focus on). I will grant Dainton that there is such a relation. The fact that in his view it holds diachronically is supposed to be a significant advantage over other views—I will discuss this in detail below. The idea that temporal experiences are experiential processes is usually held in conjunction with a commitment to the mirroring view (at least a weak version of it)—indeed I would interpret extensionalists as thinking that the process view is true because mirroring obtains. Mirroring comes in various strengths. Extensionalists typically at least believe in topological mirroring—the view that experiencing A as happening before B requires first experiencing A and then B. One could also subscribe to the stronger Metrical mirroring thesis, that experiences of duration and rates of change are mirrored by the duration relations between the parts of the experience itself, so that, for example, an experience as of an event lasting 1 second itself lasts 1 second. Dainton himself rejects metrical mirroring; a process theorist who accepts mirroring for all experienced temporal features is Phillips (2010), although his position on metrical content is qualified in his more recent work (Phillips 2013). The mirroring thesis is itself neutral on why mirroring obtains. If mirroring obtains, one possible explanation is that the mirrored temporal features are represented through resemblance—time in experience is represented by time itself. Recent defenders of extensionalism—including Dainton—do not in fact offer this as the explanation, so we probably shouldn’t assume that representation by resemblance is part of the view, even though it is a very natural extension of it. This won’t matter here, as I will be concerned only with the weaker claim that temporal experiences are process-like. “Atomism” is not a term used by Dainton in his influential taxonomies of views in this area. I define atomism as the view that temporal experiences are never process-like, a view which implies that the process view is false, and therefore by implication, so are the mirroring and resemblance views. Atomism needs to be understood carefully. For one thing, atomic experiences needn’t be instantaneous events. They might be realized by extended physical processes in the brain—indeed I would argue that all experiences are extended in this sense, because experiences require extended processes like neural firings in order to exist. Atomic experiences are atomic in the sense that they do not contain shorter experiences as temporal parts. If an atomic experience is realized by an



Extensionalism, Atomism, and Continuity

153

extended process (like 40hz neural firing), then this process may have shorter physical events as proper temporal parts, even if it doesn’t have experiences as proper temporal parts; for example, these shorter physical events may not be sufficient for any experiences to exist. To be clear, if an experiential property instantiation occurs fundamentally over a short interval in this way, we can say that it has shorter experiential temporal parts in a derivative sense. For example, if I am feeling a certain painful sensation over interval I, there is a derivative sense in which I am feeling it at each time during I. The atomist can admit that experiences have such derivative temporal parts. The process theorist thinks that temporal experiences have temporal parts in a stronger sense than this. This is obvious when the temporal parts are qualitatively varied, but even if they are not, the idea of the process view is that experiential stages of temporal experiences are distinct property instantiations—for example, they are realized by different physical events happening at different times. Atomism encompasses a number of different views that Dainton contrasts with extensionalism. Dainton himself focuses attention on “retensionalists,” who think that atomic temporal experiences have a complex structure involving memory-like retentional experiences, direct perceptual experience of the present, and possibly also a “protentional” anticipatory awareness of what is about to occur (Husserl is the most famous proponent). Atomists needn’t think that temporal experiences have this tripartite structure, however. Dainton rightly complains that we have awareness of temporal properties that is just as immediate as our awareness of “static” features like shape and color, and that the retentional view can’t account for this fact. A better version of atomism would hold that there is a single kind of perceptual experience that presents both non-temporal features and facts about how these features are changing over time. Let us call this “non-retentional atomism.” Dainton does acknowledge the possibility of such a view (versions of which have been defended by Broad 1925 and Grush 2005), but it doesn’t fall very neatly in his taxonomy. To my mind, it is the most promising competitor to extensionalism.

Defending atomism A full defense of atomism would involve a positive case against the process view and a defense of atomism against objections. Here I will focus more on objections against the view, although it will be worthwhile to first briefly describe the positive case for it (see Lee 2014 for more detail). The reason why atomism is true is that temporal experiences are realized by physical events in the brain that do not code temporal information in a way that could realize an extensional experience; in particular, the neural

154

DEBATES IN THE METAPHYSICS OF TIME

realization of a temporal experience does not have distinct temporal stages corresponding to the allegedly temporally separated stages of the experience itself. Rather, even if an experience involves being presented with events as happening at different times and standing in certain temporal relations, the experiences of these different events are each realized by neural events that happen at the same time (or over the same short interval). We can infer from this that the different parts of the temporal experience themselves happen at the same time, contrary to the process view. Why think that the different parts of a temporal experience are realized at the same time? Information at the periphery of the perceptual system—for example, on the retina—is typically represented in a form that is process-like. For example, if A happens before B, then typically light from A will stimulate the retina, and then light from B will stimulate the retina (although due to different latencies in transmission from different sources, there is not always such a neat correspondence between arrival time and transmission time, something that the system is capable of correcting for, at least to a limited extent). Theories of how temporal information is extracted from such inputs typically assume that the point is to compare or integrate the different stages of the input to get an explicit representation of temporal information that is available “all at once.” For example, the retinal stimulation from event A might leave a trace which is simultaneously compared with a trace from the retinal stimulation from event B, to get a simultaneous representation of “A before B.” This might in turn enable the subject to act on the basis of this temporal information, for example by correctly pressing the “A before B” button, rather than the “B before A” button. The reason such simultaneous integration seems necessary is that without it temporal information is not in a form that could cause appropriate effects downstream of experience, such as verbally reporting the experience, putting the information in memory, or pressing the appropriate button. The fact that the content of experience is typically accessible in this way is the main reason for thinking that temporal experiences are realized by the output of such a process of simultaneous “trace integration,” rather than by earlier events in the processing stream, such as those on retina, which do in some sense “code time by time.” If this is right, then we get the result that these experiences are not structured in time as the process theorist alleges, but rather represent temporal information all at once, as the atomist thinks. A fuller explanation and defense of this argument is given in Lee (2014a). I will not pursue it here, instead focusing on the negative arguments that might be given against atomism. Whereas the Trace Integration argument is an empirical argument, these negative arguments tend to be based on phenomenological observation and philosophical considerations. Although such considerations are relevant to the debate at certain points, ultimately



Extensionalism, Atomism, and Continuity

155

it seems to me that a debate about the timing of experience has to take into account empirical evidence pertaining to the timing of the neural events that underpin experience, and that these considerations have far stronger weight than any phenomenological or philosophical considerations that point in a different direction. That said, even if we have independent reason for skepticism about the weight of such considerations, a satisfactory defense of atomism should have something to say about them, and saying it is my aim here.

The simultaneity argument and the multiple presentations argument One argument against atomism is that if temporally separated events appear in your experience all at the same time, then they will appear to be happening at the same time—for example, tones occurring in experience at a single time will form a chord, or the different positions of a moving object would appear smeared through space. This is contrary to the atomist’s claim that one can perceive, at a single time, consecutive events as happening at different times. This is obviously a question-begging objection, however—the whole point of atomism is that simultaneously presented events need not appear as simultaneous. This only seems problematic if one assumes a mirroring constraint, but the atomist rejects such constraints. Another easily deflected objection is the “multiple presentations” objection. Since the content of atomic experiences covers a temporally extended portion of what is happening in the world, a single event may appear in the content of a series of consecutive experiences in the stream, not just a single experience. This is supposed to be counterintuitive because we do not seem to experience the event over and over again. However, all that “multiple presentations” really amounts to is the claim that you experience the event for an extended period of time, longer than the duration of a “single experience” (which presumably is the duration of the minimal amount of neural activity sufficient for an experience to exist). There is nothing counterintuitive about this. It is true that the onset of an event may be presented for more than a single moment, but this does not mean that the event will seem to you to be starting over and over again; all the experiences you have present it as starting only once (Tye (2003) makes the same point in response to this objection.) Phillips (2010) thinks that this response to the multiple presentations objection—and atomism in general—requires we lack introspective access to how our experience is changing over time—in this case, because we are not aware of enjoying more than one experience of the onset of an event.

156

DEBATES IN THE METAPHYSICS OF TIME

He thinks that we have introspective awareness of the temporal layout of experience in a stronger sense than this allows for, and that introspection reveals that temporal experience does not in fact have an atomic structure, but rather is process-like. Let’s look at this argument.

Phillips’ transparency argument Phillips thinks that we have introspective access to the temporal relations between our experiences (I’ll call this the “introspectibility premise”), but also that experience is “transparent”: introspection of perceptual experience only reveals the properties that external events are experienced as having, not any psychological properties (the “transparency premise”). He thinks that we can reconcile these two data by holding that we introspect the temporal features of experience by introspecting the apparent temporal layout of external events (the “reflection premise”). Thus, an experience as of A happening before B also is an appearance that the experience of A happened before the experience of B. Furthermore, he takes it as a premise that it is part of the nature of experience that if experience appears a certain way to me, then it must be that way (the “infallibility premise”). So if it seems introspectively as if my experience of A happens before my experience of B, then these experiences really must be related in this way. Clearly, if the apparent temporal layout of the world is an infallible guide to the temporal layout of experience itself, then a strong form of mirroring (and hence a process view) must be correct. Phillips’ argument is valid; atomists must reject one of the premises. The atomist could deny that by presenting a certain temporal layout, an experience automatically seems to have a corresponding temporal layout (the reflection premise); for example, they could deny that an experience of temporal order automatically seems to have a corresponding temporal order. An analogy with the spatial features of experience is relevant. If physicalism is true, then experiences are physical events happening in spacetime and therefore have spatio-temporal features like all other physical events—for example, at a given time, an experience will be occurring in a region of space with a certain size (for example a region of a brain). These spatial features of experiences are obviously not made available to you by introspecting the apparent spatial layout of the world; for example, an experience of a teacup does not in any sense appear to occupy a teacup-shaped region. The atomist could say something similar about temporal experience. The spatial features of experience are not available to introspection at all (this is presumably why people are sometimes tempted to say that experiences do not happen in physical space). It seems too strong to make the



Extensionalism, Atomism, and Continuity

157

analogous claim that we have no introspective sense of our experiences as events happening in time. We talk about the “stream of consciousness,” a description that seems apt given the way experience seems from the inside. Kant described time as the “form of inner sense,” again giving voice to the intuition that we are introspectively aware of experiences as happening in time. Admittedly, if the atomist denies the reflection premise, they must give an alternative account of how we are aware of experiences as playing out in time, or at least explain away the strong appearance that we have such awareness. Consider the option of denying any introspective awareness of experience changing in time. It might be that at any given moment you are simply enjoying the atomic experience you are having, and have no introspective access at all to the atomic experiences that were happening at other times. If you try to introspect the temporal relations between experiences, you will only be able to attend to the experienced temporal relations between events presented by your current atomic experience. Perhaps there is a tendency to confuse these temporal relations with those that hold between experiences themselves, giving us the mistaken impression that we are aware of experience itself changing, when really we are only aware of the external world changing. Although I think Phillips has not done enough to disarm this view, and that the view is not obviously false, I also think it may be too strong. For one thing, instead of denying introspectibility, the atomist could have a view that rejects transparency, or a view that accepts reflection but denies infallibility; if either of these options can be made to work, they will allow for introspective awareness of experience changing in time. Consider the latter option. The atomist could say that, after all, we can become aware of how experience is changing by attending to worldly changes (the reflection premise), but that the argument fails because apparent external changes are not an infallible guide to experiential changes. For example, it may very often be true that if you experience A as before B, then you had an experience of A before you had any experience of B, because A was detected before B, and the information that A occurred was available before the information that B occurred was. This is something that is perfectly consistent with atomism (although if atomism is true, your experience of B may be accompanied by a further experience of A). There may be exceptions to this rule of thumb,2 but if it is usually true, then we can use temporal appearances as a rough guide to the temporal structure of experience, gaining knowledge of experiential structure that is consistent with atomism. Phillips will reply that this alleged status as a rough rule of thumb is incompatible with the principle that there is no appearance–reality distinction for

158

DEBATES IN THE METAPHYSICS OF TIME

experience. But in so far as there is a correct principle of this kind, it cannot apply to all the properties of experience. Take spatial properties again: it might be that experiences as of different regions of space are typically located in distinct neural regions. If this were right, we could, after all, use the apparent spatial properties of things as a reliable, but not infallible, guide to certain spatial properties of experiences themselves. This obviously wouldn’t conflict with whatever the truth is behind a “no appearance–reality distinction” principle. The other way the atomist could block the argument without rejecting introspectibility would be to reject transparency. If transparency fails, then even if we aren’t aware of the temporal features of experience via reflection, the implausible conclusion that we have no awareness of experience changing over time does not follow. In fact, a pretty strong case can be made that transparency fails in a way that is relevant here: we can be aware of our experience changing in a way that does not go via awareness of external changes. Consider shifting your attention from one object to another. These could be covert attention shifts that do not require moving your eyes or other body parts. And the scene you are looking at might not be changing at all. Still, you experience a change as happening—a psychological change, not an external change. Note that it is not merely that your experience changes when you shift your attention: you have an experience of change happening. You can even be aware of certain features of this change. For example, if you shift attention back and forth between two objects at a certain rate, then you can experience the rate that this is happening, just as if you perceive a light turning on and off at a certain rate, you can be aware of this rate. The proponent of transparency makes the mistake of thinking that perception is a kind of bare confrontation with the world. However, perception involves actively shifting your attention around the environment, and an awareness of this shifting contributes to what it feels like to consciously perceive. We might compare such awareness of psychological changes with the awareness of bodily changes that can occur during touch. We can conceive of an analogue of touch that is more “transparent” than our actual sense of touch: a creature might get information about the surfaces of objects by touching them, but the resulting experiences might present the properties of the surfaces without ever making the subject consciously aware of the process of exploring the surface of the object with a part of the body; such bodily interactions with an object might be controlled entirely sub-personally. Similarly, visual awareness could have been designed to make you aware of the properties of external things without any experienced sense of visually exploring the world through controlled changes in attention, but this is not the experience we actually have. Even if we are aware of attentional changes in this transparency-violating way (the idea deserves further discussion, but I will not pursue it here), it is



Extensionalism, Atomism, and Continuity

159

not clear that they exhaust our awareness of changes in experience. Below I will consider another kind of experience we have—the experience of a constant flowing change in our temporal perspective—that is arguably also inconsistent with transparency, and that might constitute another way in which we are aware of experience changing over time (these experiences are also part of my response to Dainton, which is why I will delay discussing until later). Furthermore, as I already mentioned, it may be that there is something right about the reflection premise: perhaps we can be aware of experience changing, albeit in a fallible way, just by attending to external changes. I think it is plausible that postulating these various forms of awareness of experiential changes is enough to explain our introspective sense of a “stream of consciousness” and the fact that we seem to be aware of the temporal features of experience in a way that we are not aware of its spatial features; no appeal to an infallible awareness of the kind Phillips postulates is necessary. To sum up, Phillips has not made a compelling case that external temporal appearances are an infallible guide to the temporal structure of experiences itself: there are a number of plausible moves an atomist can make in response. They can deny introspectibility, holding that we mistake awareness of external changes for awareness of internal changes. Or they can deny infallibility, holding that awareness of external changes is a fallible, not infallible, guide to internal changes. Finally, there is a good case to be made that transparency is false: we can be aware of experience changing in a way that does not depend on awareness of external changes (as I said, I will discuss another possible counter-example below). Even though I think that Phillips’ argument can be resisted, I think it has the great merit of emphasizing that there is a real puzzle accounting for the sense in which we are aware of experience changing over time: it is really not obvious how, if at all, this happens (for example, I am suggesting in this chapter that we should believe in a kind of temporal introspection that violates transparency, but I will not be able to give a full account of how exactly it is that such inner awareness exists). Having looked at Phillips’ case against atomism, let us now move on to consider the reasons Dainton gives for rejecting it.

Dainton’s continuity argument Dainton argues that atomism cannot capture the sense in which experience is continuous through time, whereas extensionalism is tailor-made to account for this. The idea that experience is “continuous” is extremely ambiguous, and to understand his argument properly, we need to first think about the different meanings that we could attach to this.

160

DEBATES IN THE METAPHYSICS OF TIME

Experience could be continuous in the sense that it does not have any gaps in time. If time itself is continuous, this means that a section of the stream fills a continuum of moments from its beginning to end. This is weaker than saying that experience is continuous in the sense that it changes continuously (i.e., it can be represented as a continuous function from real numbers representing times to sets of real numbers representing experience states). For example, experience might change discretely, even if it is non-gappy. It is also weaker than saying that experiences are instantaneous (not just that they exist derivatively at instants), and that these instantaneous experiences form a continuum in time. The stream could fill a continuum of time, without itself forming a precise continuum in this way. For example, imagine a pain whose intensity is realized by the firing rate of a single neuron (obviously in reality many neurons would be involved). Each intensity is realized over a period of time (because a firing rate is an average calculated over a period of time); nonetheless, the firing rate, and hence the intensity of the pain, can increase gradually over time. In this way, the temporal parts of an experience might be all temporally extended, but overlap in time in such a way as to form a non-gappy covering of all the instants in a certain interval of continuous time, without precisely occupying any of these instants. It is an interesting question as to whether experience is continuous in any of these senses (for some relevant empirical literature, see Van Rullen and Koch, C. 2003; Van Rullen et al. 2011; Kline et al. 2004; Mathewson et al. 2009, 2011). Atomism is consistent with continuity in any of these senses, but also with strong discontinuity—for example, it is consistent with experiences coming in discrete bursts that are separated in time from each other. Some may think that it is obvious from introspection that consciousness is not gappy in this way. However, it is not clear that a gappy structure would be apparent introspectively; awareness of it would seem to require a higherorder quasi-perceptual monitoring of first-order experiences that is set up to be sensitive to such gaps, and we may lack any such capacity. Furthermore, the continuity intuition can be explained away as confusing the fact that we lack an awareness of gaps, with our having an awareness of a lack of gaps. The moral is that to discover whether experience is discretely gappy, we need to figure out empirically what the temporal structure of neural states underwriting experience is, not search for gaps introspectively.3 Dainton’s objection to atomism is not that it can’t accommodate continuity in any of the above senses. The kind of continuity that he is interested in (he calls it “strong continuity”) involves a feature of experience that we allegedly experience from the inside. It is a little hard to pin down what it is exactly, but it involves a sense of the interconnectedness of experience over time, of one experience flowing into the next—he expresses this by saying that consecutive experiences seem to be experientially connected. He quotes William James as one source of the idea:



Extensionalism, Atomism, and Continuity

161

My experiences and your experiences are “with” each other in various external ways, but mine pass into mine, and yours pass into yours in a way in which yours and mine never pass into one another. (James 1912: 47) … the parts of experience hold together from next to next by relations that are themselves parts of experience. The directly apprehended universe needs, in short, no extraneous trans-empirical connective support, but possesses in its own right a concatenated or continuous structure. (James 1912: xii) One example Dainton gives to try to make this vivid is the case of hearing a long continuous tone; one’s current experience seems to flow directly from previous experiences of earlier stages of the tone. Or when you apprehend a melody, your experience of one sound or a portion of melody seems to flow seamlessly into your apprehension of the next part of the melody. Allegedly this strong continuity can easily be accounted for in Dainton’s version of extensionalism. He postulates diachronic unity relations linking together one moment of experience with the next. He thinks the unity relation holds between all experiential stages that are sufficiently close together in time, so that if we consider maximal groups of mutually co-conscious experience-stages (i.e., “total experiences”), we will find that these form overlapping blocks in time. Hence he calls his version of extensionalism “the overlap model” (see Figure 8.3).

FIGURE 8.3   Extensionalism with Overlap Supposedly the fact that total experiences overlap by sharing experiential parts, and the fact that the whole stream is interconnected by diachronic unity, gives us an explanation of the felt sense of continuity, connection, and flow within conscious experience. On the other hand, on the atomic view, consecutive experiences within the stream of consciousness are not connected together by unity—they do not form larger experiential wholes

162

DEBATES IN THE METAPHYSICS OF TIME

(although see the discussion below). In this sense, there is a kind of independence between your experience at one moment and the very next moment for the atomist. Dainton thinks that this is implausible. One objection is phenomenological—it doesn’t capture the felt flowing connectedness of experience. Another is more metaphysical: if successive experiences are independent in this way, then they are no more connected than experiences in completely different streams of consciousness: In the absence of experienced transitions between pulses, successive pulses might as well belong in entirely different streams of consciousness. (Dainton 2004: 17) … individual acts are totally isolated from one another. From a purely experiential perspective, the successive phases of our streams of consciousness might as well exist in different universes. (Dainton 2004: 21) One way that the atomist could respond here is by saying that it is obscure what Jamesian experiential flow is supposed to be, and so it is unclear what exactly the atomist is allegedly failing to account for; pending clarification, there is no real objection here. I think they can do better than this, however. There is something intuitive about the idea of experiential flow, and there are fairly well-defined ways to elaborate an atomist view in order to accommodate it. Moreover, the atomist can challenge whether overlap extensionalism really does itself accommodate the relevant phenomena here. Finally, the atomist can also explain why their atoms are not “totally isolated” from one another in an implausible way. Let us discuss each of these points, beginning by addressing the metaphysical worry. According to Dainton, consecutive atomic experiences “might as well be in completely different universes.” I find it hard to see a compelling worry here. The atomist should reply by detailing the ways in which experiential stages belonging to a single stream may well be intimately connected with each other in the atomic view, connections that won’t hold between experiences in different subjects, or in completely different universes. As we will see, the atomist can even acknowledge a limited kind of experiential unity across times. First, what one experiences at a particular moment in time may well be causally relevant to what one experiences a moment later, so the stages of an atomist’s stream are at least bound by causation. To be clear, I do not think that there is any a priori reason to think that there are direct causal relations between experience stages. For all we know a priori, there is no top-down feedback within perceptual processing, so that consecutive experiences can be appropriately compared with consecutive images projected on a screen



Extensionalism, Atomism, and Continuity

163

by a movie projector, which are not directly causally related to each other, despite having correlated properties. In the experience case, if the “images” are close enough in time, they may have overlapping causes at the input end (because, as I mentioned above, perceptual processing integrates information that arrives over an extended interval), but unless the output of one chain of perceptual processing feeds back into the next chain, consecutive experiential “outputs” will not be directly causally related. As it happens though, there is empirical evidence that there are such feedback connections in perceptual processing. One reason for thinking this is that there are neural connections feeding back within many different stages of perceptual processing (see, for example, Lamme et al. 1998). Another is that computational models that assume such feedback give good predictions. Grush (2005) describes one such model in giving his “emulation theory,” in which perceptual and motor control systems estimate the next state of the world by comparing current input with a prediction of how the world will evolve (based on previous estimations of the state of the world and reference copies of current motor instructions), and creates a new model by computing a weighted average between the prediction and the input-based estimate. It is as if visual experience has a natural path that it will follow on its own (like a ball rolling down a hill), and the role of external input is as an external force correcting the direction it travels. If a model like this is correct, then it is plausible that there are feedback relations causally linking adjacent stages of an atomic stream. Another kind of intimate connection that might exist between temporally adjacent atomic experiences is overlapping realization. If an experience is neurally realized by temporally extended neural activity, then another experience close enough in time might be partly realized by some of the same neural activity. Consider again the example of the intensity of a pain being realized by the firing rate of a neuron. Because temporally adjacent firing rates might be realized by some of the same neural firings, temporally adjacent experiences of different intensity might be realized by some of the same neural activity. Thus two experiences with different intensities may “overlap” in the sense that they have overlapping realizers, even if they don’t overlap by sharing experiential parts (the overlapping neural activity may be too brief to realize any experiences). In this way, stages of a stream of atomic experiences that are close enough in time may not be completely metaphysically independent events. Dainton’s response is likely to be that the existence of both causal and realization connections between stages is not the same as the existence of the kind of experiential connections that he thinks exist, and which the atomist allegedly can’t account for. In particular, he thinks we experience the transition from one experience into another, in virtue of diachronic unity

164

DEBATES IN THE METAPHYSICS OF TIME

relations holding between them. I think the points I have made so far at least show that different atoms are intimately connected in ways that make the “different universes” complaint at least somewhat misleading. I now want to argue that (1) there is a limited kind of diachronic unity between experience that the atomist can acknowledge, and (2) that the atomist can explain the sense of continuity that Dainton is gesturing at. With regard to (1), recall that above I said that atomists probably ought not to hold that their atomic experiences are instantaneous—they are best off saying that they are extended in time, because they are realized by extended processes like neural firings. If all experiences are extended in this way, this makes it tricky to draw a sharp line between synchronic unity and diachronic unity, suggesting a form of atomism in which synchronic unity inevitably gives us a form of diachronic unity also. For example, suppose we believe, as theorists like Bayne (2010) do, that synchronic experiences belonging to a single subject at a single time are typically “unified” into a single field of awareness (a weak version of what Bayne calls the “unity thesis”). If experiences are extended in time, the “at a single time” constraint is rather fuzzy. How are we to interpret it? As meaning overlapping in time, or something else? Here is one way in which the unity thesis might work out in an atomic view, with synchrony interpreted as temporal overlap (see Figure 8.4). Consider modality-specific sub-streams, such as visual experience and auditory experience (I will assume that experience can be divided up this way). Suppose it is usually the case that (1) visual experiences which overlap in time with auditory experiences are unified with them, but (2) that consecutive visual experiences (which may overlap in time, in virtue of being realized over overlapping temporal intervals, as in the pain intensity/firing rate case discussed above) are not unified. (1) is a kind of unity thesis, and (2) is plausible, because the continuous updating of visual experience is likely to mean that adjacent atomic visual experiences have incompatible contents, even if they overlap in time. Given this set-up, it is possible that a particular auditory experience, such as A2 in Figure 8.4, will be unified with consecutive stages of visual experience (V1,V2, and V3) that are not themselves unified and which do not overlap by sharing experiential parts (even though they overlap in time). If a “simultaneous total experience” at time t is a group of mutually unified experiences that all happen during a period that includes t, then the situation depicted would involve A2 being contained in three different total experiences that occur over different intervals, total experiences that contain different experiential elements (A2+V1, A2+V2, A2+V3). This shows how, even in an atomic view, total experiences at slightly different times can overlap by sharing experiential parts (the overlap is quite different from



Extensionalism, Atomism, and Continuity

165

FIGURE 8.4 the kind postulated in Dainton’s overlap model, however). By implication, the momentary stages of an atomic stream of consciousness may not be experientially isolated islands, but may be connected to each other by a crisscrossing web of unity relations. That said, I should stress that the atomic view is also compatible with a lack of any such diachronic connections, and further empirical evidence may reveal that they do not exist. For example, if your total experience is a series of discrete pulses, there may be no overlapping realization or unity connections between stages. I don’t want to suggest that I think that these diachronic unity relations, if they exist, would explain our sense of one experience flowing into the next. I don’t think they would. But then again I don’t think Dainton’s diachronic unity relations would explain the sense of flow either. I bring them up simply to point out that even on an atomic view, there can be experiential connections between stages that are similar to those that appear in the extensionalist view. So, let us now turn to the question as to whether the atomist can explain the sense of “flow” in experience. Before getting to that, one point that an atomist should make here is that there are reasons in advance to doubt whether the extensionalist can do any better than them in capturing this extra sense of “flow.” Consider a total experience that involves apprehending, for example, a section of music, including the temporal relations between various sounds. The atomist and the extensionalist can agree about what the overall content of this total experience is, and agree that it has as parts experiences

166

DEBATES IN THE METAPHYSICS OF TIME

of individual sounds and their relations, and agree that these parts are unified as part of the complex whole. The main difference between the theories involves the temporal arrangement of these parts (this is especially clear in Dainton’s version, because he thinks that the unity relation that holds diachronically is the same as the one that holds synchronically). Since, aside from this temporal difference, atomic total experiences can be isomorphic to extensional ones, it is not at all clear that there is any phenomenal fact that can be explained by extensionalists but not by atomists. So if a “feeling of flow” is not accounted for in the atomic view, this suggests that it is also left out in the overlap model4. Dainton’s view must be that the mere fact that, in the overlap model, the relevant parts of temporal experience happen sequentially in time rather than at a single time, gives a feeling of flow from one part of the experience to the next, which would otherwise be absent. But this is problematic. First, it is not at all clear why having a sequential rather than synchronic arrangement in time would make any phenomenal difference at all, let alone why it would produce a sense of “flow” (when we consider the kinds of features that would explain “flow” we will see that this is especially clear). Second, whatever phenomenal role is played by these temporal relations in Dainton’s model has an analogue in the atomic view in the temporal content of an atomic experience. For example, the temporal order of the parts of an extensional experience has an analogue in the presentation of temporal order in an atomic experience. So if the mere temporal arrangement of an extensional experience can explain a sense of “flow,” it is unclear why the corresponding temporal content of an atomic experience cannot do the same work.5 So, it is unclear that extensionalists do any better than atomists in explaining the sense of “flow.” What would explain it? It may be that it can be fully explained in terms of our awareness of changes in external events, in ways that I will describe immediately below. But if not, I would suggest that this is because there is a kind of awareness of our experiential perspective changing over time that is inconsistent with temporal transparency: I will try to describe what I mean by this in more detail below. As just mentioned, I think our experience of the continuity of external events and processes is probably at least one source of the sense of “flow.” External events may be experienced as continuing on from before we apprehended them, or at least we may lack an experience of them as starting when we start to experience them. Take the simple example of experiencing a long continuous tone. Your auditory system—indeed your perceptual system in general—is designed to detect discontinuities in stimuli, and make them perceptually salient to you. There is therefore a big perceptual difference between experiencing a sound as starting, and experiencing a sound without experiencing it as starting. There may also be such a thing as positively



Extensionalism, Atomism, and Continuity

167

experiencing the sound as continuing on from before. All three of these can easily be accommodated by an atomist, because they involve different conditions under which experience is veridical, and there is no reason why an atomic experience cannot have the relevant veridicality condition. The tone example is not an isolated curiosity—we constantly perceive processes that are not bounded in time at the point of apprehension. Much of ordinary experience is of activities and processes that are already in motion; more generally, even the unchanging state of a boring material object like a teacup or muffin is an example of something continuing on from before that we either do not perceive as bounded, or positively sense as continuing on. Interestingly, it is not actually clear how an extensionalist will capture the distinctions we are talking about here. Suppose, for example, that I have a “total experience”—a maximally unified experience—that presents a sound as continuing on from before. What is the difference between this and a case where the sound is heard as bounded in time? Dainton’s idea seems to be that it is the fact that the beginning of the total experience is unified with a prior experience of sound, rather than an experience of silence, that explains the appearance of continuity. But this connection with earlier parts of the stream is actually an extrinsic property of the total experience, whereas the appearance of continuity is surely an intrinsic property of the experience, having to do with how it presents a current sound as related to the past. This suggests that the atomist may actually do better than the extensionalist in explaining our experience of continuity. Whether or not they have a response to this last point, the extensionalist might object that Jamesian continuity is a quite different phenomenon from an experience of temporal continuity in the world. Jamesian continuity is a matter of feeling the flow from one experience to another, not from one external state of the world to another. It is this that the atomist cannot capture. I suspect that theorists like Michael Tye, who often press the idea that experience is transparent to introspection (for example, Tye 2002), will be inclined to deny the phenomenological intuition here. They will say that introspection only reveals how the world appears to be arranged, and so any sensed flow or continuity must be an apparent feature of external events. We have already discussed transparency, and noted that it may be subject to counter-examples. I want to end by considering an additional kind of awareness we have that would fit Dainton’s job description of an experience of flow very nicely, and is an additional counter-example to temporal transparency. As we will see, “flow” in this sense is perfectly compatible with atomism. The relevant “flow” is best described as a sense of a constant change in your temporal perspective, that is, of which events are presented by

168

DEBATES IN THE METAPHYSICS OF TIME

experience as being in the present. It is related to the idea that we are aware of time itself passing—not just in the sense that we are aware of temporal relations between external events, such as temporal order and distance relations, but in the sense that we are aware of a constant flowing change in what is in the present. If there is an experience of this kind, it means that there is a sense in which we “experience time” that does not just involve perceiving relations like temporal order and duration between external events. Now, I prefer a metaphysical view of time in which it doesn’t literally involve a “moving present”—a fortiori we don’t perceive such a thing. Nonetheless, it seems to me that we do genuinely have a sense of a constant change in what is present, a sense which arguably helps ground the intuitions about time that believers in a moving present have6 (I will not say any more about the fascinating topic of the metaphysics of the moving now here). Supposing this is right, what would account for it? Let us begin by noting that our perceptual system does, in some sense, present events as happening in the present. We automatically assume that what we see is happening in the present—for example, we make present tense judgments based on experience, and act on the assumption that perceived events are current. (This can be true even if we are also capable of accommodating cases where we know that there has been a significant delay between information transmission and reception at our sensory periphery.) Slightly more contentiously, we have a conscious sense of events having happened in the immediate past. For example, a briefly perceived event like a bright flash of light in some way lingers in consciousness longer than the brief moment when it is perceived as present. Or the presently experienced passage in a piece of music is somehow experienced as having a particular musical context (the phenomenology I am talking about is also at its most obvious in cases where we are anticipating that an event will happen, and deliberately attend to it when it does in fact happen). These phenomenological claims are partially substantiated by evidence for the existence of short-term memory mechanisms linking perception with other processes downstream. For example, Pöppel (2004) cites various pieces of evidence in favor of the hypothesis that we have working memory representations in vision and audition that represent temporal information from the last 2–3 seconds of perceptual experience: for example, subjects are able to accurately reproduce visual or auditory information perceived within the last 2–3 seconds, but their performance rapidly drops off beyond this range. Perhaps these working memory representations form a “long specious present,” which contrasts with a shorter temporal window of events presented in vivid phenomenal awareness, and which gives us a sense of the immediate past of events that are experienced as present. If this is right, then we can think of perception as a conveyor belt of information, being first processed unconsciously, then appearing in phenomenal



Extensionalism, Atomism, and Continuity

169

consciousness as a presentation of a short temporal array of current events, and then briefly lingering in short-term memory, still contributing to what it is like for the subject, as a conscious sense of the immediate past. In order to keep up with present events, notice that there must be a constant updating of the information that is represented at each stage—the conveyor belt has to keep moving, otherwise we won’t keep up with the world. The existence of these constantly updated conscious representations of what is present and immediately past is perfectly consistent with the basic version of atomism, and probably with extensionalism too (although maybe it is easier for an atomist to accommodate a sense of what is immediately past as being immediately past). Now take note of the following important point: the mere existence of these constant changes is not enough to give us a sense of the flowing present—for that, it seems that we need to be in some sense aware of the process of updating occurring, not merely be such that it is in fact occurring. Consider an experience of a brief event happening and then seeming to quickly fade into the past. To get such a sense of fading, it will not be enough that the representation of the event is in fact moved along the conveyor belt from perception-as-present into conscious short-term memory, and then out of consciousness entirely. This process will have to be one that we have a higher-order awareness of. Simply having a series of first-order experiences of external world events (whether atomic or extensional), even if they are accompanied by short-term memory experiences, will not give us this awareness. That would only amount to a “frozen” sense of the layout of events from one temporal perspective, not a sense of a constant flowing change of temporal perspective. Arguably, to explain the sense of experience flowing from one moment to the next, we need to postulate a higher-order awareness that can be directed to the changing array of first-order conscious experiences. Notice that, since first-order experiences, even if tensed, could not on their own give us a sense of changing temporal perspective, it must be that this higher-order awareness is inconsistent with “transparency” as articulated by Phillips. If such higher-order awareness exists, then we are not stuck inside our present perspective, aware only of how it presents the world, but have a sense of this perspective itself evolving through time. (Note that there is a difference between this awareness of temporal flow, and a mere awareness that our temporal perspective has changed, of a kind that we might get from episodic memory.) I am not the first person to suggest that such a thing exists—Husserl held a similar position; furthermore, he may have intended it to explain the sense of temporal flow (see Miller (1985) for an interpretation of Husserl along these lines). I also do not say that the idea is without problems, or that it is completely obvious that there is any need for such a thing. We might

170

DEBATES IN THE METAPHYSICS OF TIME

FIGURE 8.5

wonder if these higher-order states are quasi-perceptual experiences that can misrepresent first-order states, or stand in some other kind of relation to their targets. We might also wonder if there is any independent reason for thinking that they exist. These are important questions, but I will not address them here. Perhaps pursuing them will reveal that I have misdescribed experiential flow, or that the idea that it exists is an illusion. What I am saying here can perhaps best be put as a dilemma. If there is a flow in experience that needs explaining, it either has to do with a sense of external events continuing on from before, or a sense of a constant change in temporal perspective of the kind discussed here. The former can easily be accounted for by the atomist in terms of the contents of first-order experiences, as explained above. One might doubt that there really is an experience of the latter kind, but in so far as there is, a story of the kind that I have just told, involving a higherorder awareness of changes in your temporal perspective, might be true. Furthermore, this is a story that is perfectly compatible with atomism, and extensionalists do not have an alternative story to offer that requires their view to be true. This shifts the burden to the extensionalist to explain how there is a kind of experienced continuity or flow between experiences that is incompatible with atomism.



Extensionalism, Atomism, and Continuity

171

Conclusion The main aim of this discussion has been to defend atomism against the criticisms of it that appear in the current literature, especially those in Phillips (2010) and Dainton (2000, 2010). I looked at Phillips’ (2010) argument, suggesting that one or more of the main premises can plausibly be rejected by the atomists. I then considered Dainton’s argument from Jamesian continuity, suggesting that, if anything, atomists are better placed to explain our sense of flow and continuity in experience than extensionalists are.

Notes 1 These temporal parts need not be completely independent events capable of existing on their own. What matters is that they are distinct events capable of happening at different times; this may (or may not) be compatible with various forms of mutual interdependence (for example, they might have independent core realizations, but identical or overlapping total realizations). For more discussion of various forms of interdependence between experiences, see Lee (2014b). 2 Perceptual systems are able to recalibrate information in time to account for discrepancies in transmission time from different sources. For example, simultaneous taps on the nose and feet feel simultaneous even though it takes longer for the signal from the feet to be transmitted to the brain. Consider a case where your nose is tapped slightly in advance of your feet. You might correctly experience this temporal order, even though the signal from the feet does not arrive before the signal from the nose. Given atomism, this makes it at least doubtful whether you experience the foot tapping before you experience the nose tapping. 3 Interestingly, it is not clear whether extensionalism is compatible with such a discrete gappy structure: the extensionalist’s commitment to mirroring might suggest that if experience had this structure, events in the external world would seem to also have a discrete gappy structure, which they do not. I suspect that there are versions of extensionalism that avoid this problem by adopting a sufficiently weak version of mirroring, so I won’t pursue the objection here. 4 One way to see this is, is to note that Overlap Extensionalism is (apparently) completely consistent with strong transparency. So even if Overlap Extensionalism is true, we may only ever introspectively apprehend the worldly appearances provided by total experiences, not diachronic connections between experiences themselves. In other words, it is not clear why the fact that, on this view, total experiences overlap with each other, or the fact they are spread out in time, is something that would be revealed by introspection, or contribute to a sense of continuity or flow that is absent on the Atomic picture.

172

DEBATES IN THE METAPHYSICS OF TIME

5 It is true that the Atomist will deny that having an experience with a certain temporal content is the same thing as having awareness of how experience is changing over time. But as we saw above in discussing Phillips’ argument, it is contentious to assume that ordinary temporal phenomenology, involving experiences of events in the external world playing out in time, also involves direct awareness of experience itself changing. If the idea of Jamesian flow depends on such an assumption, then that atomist can reasonably say that it is suspect. 6 We can distinguish two aspects to these intuitions : (1) an intuition that the present moment is metaphysically special; and (2) an intuition that there is a constant change in which events are highlighted as present. The psychological phenomenon I am interested in is more relevant to (2). For attempts to explain (1), see Butterfield (1984) and Callendar (2008). A recent attempt to explain (2) (or something close to it), in terms different from those I discuss here, is Paul (2012).

References Bayne, T. (2010). The Unity of Consciousness. Oxford: Oxford University Press. Broad, C. D. (1925), The Mind and its Place in Nature. London: Routledge & Kegan Paul. Butterfield, J. (1984), “Seeing the present,” Mind, 93, 161–76. Callendar, C. (2008), “The common now,” Philosophical Issues, 18, 339–61. Chuard, P. (2011), “Temporal experiences and their parts,” Philosopher’s Imprint, 11. Dainton, B. (2000), Stream of Consciousness: Unity and Continuity in Conscious Experience. London: Routledge. —(2004), “Precis of ‘stream of consciousness’,” Psyche, 10, 1. —(2010), “Temporal Consciousness,” in Edward N. Zalta (ed.), The Stanford Online Encyclopedia of Philosophy. http://plato.stanford.edu/archives/fall2010/ entries/consciousness-temporal/ Grush, R. (2005), “Internal models and the construction of time: generalizing from state estimation to trajectory estimation to address temporal features of perception, including temporal illusions,” Journal of Neural Engineering, 2, 3, S209–18. James, W. (1912), Essays in Radical Empiricism. New York: Longmans. Kline, K., Holcombe, A. O., and Eagleman, D. M. (2004), “Illusory motion reversal is caused by rivalry, not by perceptual snapshots of the visual field,” Vision Research, 44, 23, 2653–8. Lamme, V. A., Super, H. and Spekreijse, H. (1998), “Feedforward, horizontal, and feedback processing in the visual cortex,” Current Opinion in Neurobiology, 8, 4, 529–35. Lee, G. (2014a), “Temporal experience and the temporal structure of experience,” Philosopher’s Imprint, 14, 3, 1–21. —(2014b), “Experiences and Their Parts,” in D. Bennet and C. Hill (eds), Sensory Integration and the Unity of Consciousness. Cambridge, MA: MIT Press. Mathewson, K. E., Gratton, G., Fabiani, M., Beck, D. M., and Ro, T. (2009), “To



Extensionalism, Atomism, and Continuity

173

see or not to see: prestimulus α phase predicts visual awareness,” Journal of Neuroscience, 29, 9, 2725–32. Mathewson, K. E., Lleras, A., Beck, D. M., Fabiani, M., Ro, T., and Gratton, G. (2011), “Pulsed out of awareness: EEG alpha oscillations represent a pulsedinhibition of ongoing cortical processing,” Frontiers in Psychology, 2. Miller, I. (1985), Husserl, Perception and Temporal Awareness. Cambridge, MA: MIT Press. Paul, L. (2012), “Temporal experience,” Journal of Philosophy, CVII, 7, 333–59. Phillips, I. (2010), “Perceiving temporal properties,” European Journal of Philosophy, 18, 2, 176–202. —(2013), “Perceiving the passage of time,” Proceedings of the Aristotelian Society, CXIII. Pöppel, E. (2004), “Lost in time: a historical frame, elementary processing units and the 3-second window,” Acta Neurobiologiae Experimentalis (Wars), 64, 295–301. Tye, M. (2002), “Representationalism and the transparency of experience,” Noûs 36, 1, 137–51. —(2003), Consciousness and Persons: Unity and Identity. Cambridge, MA: MIT Press. Van Rullen, R. and Koch, C. (2003), “Is perception discrete or continuous?,” Trends in Cognitive Sciences, 7, 5, 207–13.

9 Flows, Repetitions, and Symmetries: Replies to Lee and Pelczar Barry Dainton

O

f the many commonplace observations that could be made regarding ordinary human lives, there are two that are relevant to the issues with which I will be dealing here. First, most of us spend most of our lives awake, rather than asleep or otherwise unconscious. Second, during our waking hours we are continuously conscious—we are experiencing continuously. For the most part we are always aware of something or other, and more often than not one of these somethings will be changing, in one way or another. Given these two hard-to-dispute facts, it would be natural to suppose that it would have by now proved possible to arrive at an account of the general structure of our streams of consciousness that would command broad assent. Indeed, one might think that nothing would be less obvious or controversial. After all, these streams contain the experiences which constitute our conscious lives, and their basic structural features are comparatively constant: what it feels like to be an ordinary awake-and-experiencing human subject does not change enormously from one moment, or hour (or day or month) to the next. But this is far from being the case. There are very different accounts of the structure of our streams of consciousness on offer, and as yet nothing approaching agreement as to which of these is closest to the truth— as the contributions of Lee and Pelczar to this volume attest. If the nature of time itself is among the more intractable issues in metaphysics, the nature of our temporal experience must count as among the most intractable issues in phenomenology and the philosophy of mind.

176

DEBATES IN THE METAPHYSICS OF TIME

Timeless experience One basic division of opinion concerns the manner and extent to which our experience really is temporal in character. A useful way of bringing the (apparent) temporality of our ordinary experience into clear view is by trying to imagine what it would be like to be a wholly timeless being. In attempting just this, Walker (1978: 40–2) envisages a subject whose sensory experience is restricted to the visual, and who sees a spatial array of colored solids at differing distances. Needless to say, this atemporal subject’s visual experience is utterly static and devoid of change—there can be no change without time. So too is the subject’s “internal” mental life: they do not enjoy an ongoing (and hard to stop) inner soliloquy, they do not feel their stomach’s rumbling or their limbs changing position. Nor do they experience anything as persisting or enduring. For someone to see an object remaining the same color takes some time, and time is something Walker’s timeless subject does not have. Consequently, this subject’s experience will be both static and stroboscopic: it will consist of a single “flash” without discernible duration.

Variety is the spice of life

viewpoint FIGURE 9.1   Experience in a Timeless World



Flows, Repetitions, and Symmetries

177

This extreme brevity may not rule out the apprehension of meaning. Walker suggests a timeless subject may be able to understand the meanings of inscriptions on the blocks it sees (as shown in Figure 9.1). But since these meanings must also be grasped all at once in a flash, the corresponding messages must be short and simple (assuming the being’s cognitive powers are similar to our own). Also, it is not for nothing that Walker’s subject is confined to a wholly visual world. For it seems plausible to suppose that auditory experience is essentially temporal; a sound that has no discernible duration at all would not be a sound. Now, our own experience is manifestly very different from that of this timeless subject.1 We do hear sounds, and not just individual sounds but whole seamless successions of them. On listening to a sequence of brief notes of equal duration, C–D–E–F–G–A–B, by the time we hear the last note B we will no longer be able to hear the initial note C, but we will hear each note flowing into its immediate successor. We too can see static ensembles of objects, but unlike the timeless subject, our experience is not confined to frozen snapshot-like instants, even on those rare occasions when we are gazing at an array of motionless objects. If we turn our heads quickly, we see our surroundings zip by in a blur; if we turn our heads more slowly, we see our surroundings slide by in a clear non-blurry fashion; if we don’t turn our heads (or move our eyes) we see the objects we are looking at continuing on, in a way that is not available to the timeless subject. We too can take in meanings in a glance (when looking at the inscription in Figure 9.1, for example). But we can also apprehend meanings over time. We do so when we listen to someone talking, or read through a lengthy paragraph of text, or when our inner soliloquy is conducted in our acoustic imaginations—as it often is.

Atomism v. extensionalism When it comes to making sense of how our experience can be as it is, we are confronted with the problem of explaining how our experience can have those dynamic, time-consuming characteristics which differentiate it from the consciousness of a timeless subject. In responding to this question, some have claimed—for example, Chuard (2010)—that our streams of experience are in fact composed of successions of momentary phases whose contents are just as static (or frozen) as those apprehended by Walker’s timeless subject.2 This is a bold, even heroic, proposal, but it is not a very promising one. Given the dramatic phenomenological differences between our own streams of consciousness and those of a timeless subject—a subject whose experience is not stream-like at all—the claim

178

DEBATES IN THE METAPHYSICS OF TIME

that fundamentally our experiences are of precisely the same sort is not a plausible one, to put it mildly. If we reject the claim that our experience is fundamentally timeless—as I propose to do here—we are left with the task of comprehending how our experience can be both highly dynamic (over short intervals) and continuous (over longer intervals) in the way it seems to be. Here too there are various options and possibilities, but I want to focus here on just two: the “extensionalist” view that I have explored and defended on several previous occasions, and the view to which Lee and Pelczar are both sympathetic.3 Following the lead of Lee, I will call the latter “atomism.” Both views fully acknowledge that over short intervals we are directly aware of change and succession. So in the case of the sequence of brief tones mentioned above, even though we don’t apprehend the entirety of C–D–E–F–G–A–B in a single unified experiential episode—by the time B occurs we are no longer experiencing C—we do hear each tone giving way to its immediate successor. We hear the C-tone flowing into (or being succeeded by) the D-tone, the latter flowing into the E-tone, and so on. If this is the case, then our stream of consciousness must contain auditory experience with the following contents: (C–D), (D–E), (E–F), (F–G) … Each of these is a unified episode of experience—a “specious present” to use the usual jargon—with a dynamic content which appears to extend a short way through time. I am assuming here that any two of these tones fully occupy a single specious present, hence the brackets mark the outermost boundaries of these unified experiential wholes. But do these appearances correspond with reality? According to the extensionalists they do: each of these episodes—each of these specious presents—extends a short distance through time, in much the way it seems to. Hence (C–D) is a temporally extended experience, with briefer experiences as parts, and these include (at the very least) an experiencing of C and an experiencing of D, with the latter occurring later than the former. C may be experienced before D, but it is nonetheless experienced with D, as parts of a single unified experience of C-being-followed-by-D. According to the atomist, in sharp contrast, episodes such as (C–D) do not extend through time. They consist of momentary (or very brief) experiences with contents which appear to extend over a short period (e.g., a second or so), even though they do not in fact do so. Consequently, unlike their extensionalist counterparts, atomist specious presents do not consist of a succession of briefer experiences. Experienced duration and succession exist in experiences which do not themselves possess duration and succession. In effect, in this view the temporality of experience lies orthogonal to objective temporality, as illustrated in Figure 9.2. Extensionalism and atomism differ in how they situate experienced



Flows, Repetitions, and Symmetries

179

FIGURE 9.2   Directly experienced succession as seen by the Extensionalist (on the left) and the Atomist (on the right). Each rectangle represents a single specious present.

temporal extension into objective time. But there is another important divergence: the accounts they offer of the structure of streams of consciousness are also very different. According to the atomist, a stream consists of a succession of discrete momentary (or near-momentary) experiences, each of which possesses apparent temporal depth. For the extensionalist, a stream consists of a succession of temporal extended specious presents which overlap by sharing parts, in the manner indicated in Figure 9.3.4 There is more to be said about each of these theories, but this brief outline is sufficient to reveal that we are being offered very different accounts of temporal experience.5 How should we go about choosing between them?

FIGURE 9.3   Extensionalist specious presents overlap by sharing parts. Hence in the case of the stream-phases (C–D) and (D–E) shown on the left, the D-tone in the former is numerically the same as the D-tone in the latter. Atomist specious presents, as shown on the right, are entirely distinct and discrete episodes of experience.

180

DEBATES IN THE METAPHYSICS OF TIME

Repetitions On various occasions I have argued that the extensional model has greater phenomenological plausibility than rivals such as atomism. Lee disagrees, arguing that phenomenological considerations do not provide us with a basis for preferring one model over the other. One issue is related to the “problem of repeated contents.” As Pelczar makes clear in his excellent—and in some respects novel—presentation of the problem, the atomist is confronted with an awkward dilemma. In the case of the auditory succession C–D–E–F–G–A, the atomist could say that this experience consists of just three specious presents (C–D), (E–F), and (G–A). But this uncomplicated proposal faces a problem. In reality, tone D is experienced as flowing into E, and F is experienced as flowing into G, but neither of these transitions would be experienced under the current proposal. There is an obvious solution to this problem. The atomist can fill the gaps by holding that there are five specious presents not three: (C–D), (D–E), (E–F), (F–G), (G–A). Unfortunately, this move also proves problematic. Every transition between notes is now experienced, but at the cost of too many notes being experienced. The subject of this stream would experience C–D–D–E–E–F–F– G–G–A. A normal stream of auditory experience has been transformed into a stream with a stutter, as Pelczar aptly describes it. These phenomenologically unrealistic repetitions are an inevitable consequence of the atomist’s attempt to capture all the experienced transitions which actually exist in streams of consciousness. In fact, as Pelczar points out, the situation rapidly goes from bad to worse. In our actual streams, it is typically the case that every brief phase of experience is experienced as flowing into the next. Given this, the new specious presents just introduced, namely (D–E) and (F–G), should themselves be experientially continuous with their neighbors. The only way of achieving this within the atomist’s framework is by introducing yet more specious presents, for example we will need a further specious present (D–D) to bridge (C–D) and (D–E). And since these new atomic phases will themselves need linking to their neighbors, we are embarked on a disastrous infinite regress. The extensionalist model captures all the experienced transitions without introducing any unrealistic stuttering. In the simple case envisaged, we again have just five specious presents: (C–D), (D–E), (E–F), (F–G), (G–A). Moreover, instead of D (for example) being experienced twice over at different times, it is experienced just once. This is because the D-tone in (C–D) is numerically identical with the D-tone in (D–E), and similarly for the Fs and Gs. For Pelczar, the one great advantage of the extensional model



Flows, Repetitions, and Symmetries

181

lies in the simple and effective solution it offers to the repeated contents problem. For Lee, in contrast, the repeated contents issue is an “easily deflected” pseudo-problem. Since Lee believes atomists should accept that our streams of consciousness do not possess the same depth of unity as in the extensional model—a claim to which I shall return—he has little to fear from the threatened infinite regress. In the case of our example above, he could hold that the experiencing of C–D–E–F–G–A involves just five specious presents, (C–D), (D–E), (E–F), (F–G), (G–A), and argue that the additional “linking” specious presents—such as (D–D) and (E–E)—are surplus to requirements. Even so, don’t these five specious presents create an unrealistic stutter? Isn’t it implausible to suppose that in hearing this sequence of notes we hear C–D–D–E–E–F–F–G–G–A? In response, Lee has this to say: … all that “multiple presentations” really amounts to is the claim that you experience the event for an extended period of time, longer than the duration of a “single experience” … There is nothing counterintuitive about this. It is true that the onset of an event may be presented for more than a single moment, but this does not mean that the event will seem to you to be starting over and over again; all the experiences you have present it as starting only once. (Lee 2014: 155) One’s musical abilities do not need to be particularly acute to appreciate that the sequence C–D–E sounds very different from C–D–D–E. (Don’t forget that here and throughout we are assuming that the tones are of equal duration.) Indeed, our ability to appreciate music depends in large part on our being very sensitive to this sort of difference. A D-tone which is twice as long as a particular C-tone sounds very different from a D-tone of the same duration as that same C-tone. So on the face of it at least, and contrary to what Lee claims, it seems that it does matter if the repeated contents problem leads to events being experienced for extended periods of time. In the second part of the passage cited above, Lee makes a further claim. He suggests that the repeated contents are not presented as new, and so will not be noticed by us. In the case of C–D–D–E, the second D-tone exists, but since the experience is “presented” to us as being the same experience as the first D-tone, we experience D only as a single tone, not two. We can, I think, agree that if the repeated contents are undetectable—if they are entirely invisible to introspection—there is no longer any problem, at least at the phenomenological level. But it is difficult to see why the additional tones would be undetectable. Suppose someone plays the sequence of notes C–D–D–E on a piano standing a few feet away from you; the pianist is

182

DEBATES IN THE METAPHYSICS OF TIME

sufficiently accomplished that the tones are all of precisely the same duration, and volume. So far as its auditory properties go, the second D-note is intrinsically just like the first. Evidently, in these circumstances there are two easily detectable D-tones present in your stream of consciousness. According to Lee, when you listen to the three-note sequence C–D–E, you first experience (C–D) and a moment or two later (D–E). So far as their auditory phenomenal properties go, both D-tones are precisely the same. Given this, why would the second D-tone be an entirely undetectable presence in your stream of consciousness? Why wouldn’t it be just as detectable as the second D in the four-note sequence played on the piano? There are in fact easily conceivable circumstances in which a repeated content would be interpreted as being nothing more a re-presentation of the original. Suppose, for example, that I have a digital sound recorder which has—in effect—developed a regular stutter: after every second or so, the acoustic content recorded over the previous half-second is duplicated by the machine, and this addition content gets added to the sounds actually recorded when the latter are played back. Since I am well aware of this machine’s fault, if I heard the note sequence C–D–D–E emerging from this device, and durations of these notes is such that C–D together last for a second, I would naturally assume that the second D-tone is a product of the recorder’s malfunctioning systems, and does not correspond to any actual sound. So I would then (rightly) conclude that the sequence of sounds which actually occurred consisted of the sequence C–D–E. Consequently, I would hear the second tone as a re-presentation (or reproduction) of the initial D-tone. In a similar fashion, if my hearing were similarly afflicted—by virtue of, say, a malfunctioning cochlear implant—I would make similar judgments about what I was hearing: I would appreciate that only some of my auditory experiences correspond with actual sounds in the environment. In scenarios of this type there is a divergence between the character of one’s experiences and what one takes these experiences to represent; but in virtue of the fact that the divergence is systematic and predictable, it is nonetheless possible to arrive at an accurate picture of what is going on about one on the basis of one’s experience. But even though in these circumstances some experiences are not taken at face value, they are perfectly ordinary components of streams of consciousness, and just as easily discernible introspectively as any other experience. The second D-tone is experienced as a re-presentation of the earlier D-tone, but it is a perfectly ordinary auditory experience in its own right. Lee claims that such experiences would be undetectable, invisible to our introspective scrutiny. But as far as I can see, there is no reason to believe he is correct. He has certainly provided us with no such reason.6



Flows, Repetitions, and Symmetries

183

In reply, Lee could argue that it is at least possible that the repetitions exist, but go undetected by us. If Descartes’ all-powerful Demon could get us to believe that 2 + 2 makes 5, couldn’t the Demon also ensure that we don’t notice the repeated contents the atomist’s model of temporal experience brings with it? And if an all-powerful Demon could manage this, couldn’t our own brains, or cognitive systems, manage to do the same? If our experience is structured in the way the atomist suggests, perhaps our brains have developed compensatory mechanisms, and ensure that the duplicate experiences they themselves are creating are introspectively invisible to us, and are not cognized or remembered. There is, after all, a good reason why our brains would develop such mechanisms: in their absence, our sensory experience would be a very unreliable guide to what is going on around us. This may all be possible, but it is not a very economical way for our brains to function, to put it mildly. We would be justified in taking this scenario seriously if the atomist’s model of temporal experience were the only available option, but it isn’t. The extensional alternative is simpler and more economical: it brings with it neither the duplicated experiential contents, nor the compensatory mechanisms which ensure that we are systematically deluded as to the experiences we are actually having, from moment to moment. In the absence of countervailing considerations which favor atomism, it looks as though the repeated contents problem does provide us with a reason for favoring the extensionalist’s position.7 However, there is a further twist in the tale. The issue of whether it really is possible for experiences in our streams of consciousness to “disappear” without leaving any phenomenological trace is linked to another issue: the sorts of continuity which exists in these same streams.

Sensible continuity The fact that the extensionalist’s account of the structure of our streams of consciousness does not give rise to the problem of repeated contents is one reason for preferring it over atomism. A second reason is its ability to accommodate any distinctive continuity that is characteristic of our ordinary streams of consciousness, or so I have argued previously. Once again, Lee demurs, arguing that when it comes to experiential continuity, the explanatory and phenomenological resources of atomism are precisely equivalent to those of extensionalism. Once again, I am not convinced by his arguments, though the issues here are admittedly rather more nuanced phenomenologically speaking than in the repeated contents case.

184

DEBATES IN THE METAPHYSICS OF TIME

The phenomenological feature in question is the sort of continuity which makes it appropriate to describe our consciousness as stream-like in the first place, the sort of continuity which James calls “sensible continuity.” This form of continuity is at its most conspicuous when we are perceiving ongoing processes which are themselves continuous: when watching river water flowing by, or listening to a sustained note played on a church organ, or watching an ice skater gliding around the rink, or enjoying the feel of a hot shower against one’s skin—the examples can be multiplied endlessly and easily. In such cases each phase of one’s experience has a dynamic content, but each phase also seamlessly flows into the next. Sensible continuity is readily accommodated by the extensional model, provided at least that the latter includes the form of overlap via part-sharing mentioned earlier. We see the ice skater glide from X to Y and then from Y to Z. The seamlessly flowing character of the whole experience is explained by the fact that the second half of the first episode is numerically identical with the first half of the second episode. Overlap by part-sharing supplies a simple but effective underpinning for sensible continuity. If this form of overlap does exist, there is no mystery as to why our streams of consciousness can be as continuous as they appear to be. Is sensible continuity so easily accommodated by atomism? On the face of it, this seems unlikely. After all, the defining feature of atomism lies precisely in its atomistic conception of the stream of consciousness. Streams are composed of entirely discrete momentary (or near-momentary) experiences. True, the contents of these momentary experiences are dynamic, and seem to extend through a brief-ish interval of time. But it remains the case that the atomic experiences themselves are self-contained, and so are not experientially connected to any earlier or later experiences in the same stream. Lee thinks none of this matters. Why? Because he also thinks that “Atomic total experiences can be isomorphic to extensional ones” in all respects save their orientation with regard to ordinary objective time. A “total experience” is a maximal unified episode of experience, possessing dynamic contents which themselves possess apparent temporal extension—experiences otherwise known as “specious presents.” Lee invites us to take as an example an auditory total experience which consists of the hearing of a short stretch of music, consisting of a succession of notes experienced as a succession. So far as the contents of this total experience are concerned, atomists and extensionalists are in total agreement, or so Lee claims: “Atomist and extensionalist can agree about what the overall content of this total experience is, and agree that it has as parts experiences of individual sounds and their relations, and agree that these parts are unified as parts of the complex whole” (2014: 165–6) Since they agree on so much, it is difficult to see how there can be any phenomenal fact or feature—such as phenomenal flow or



Flows, Repetitions, and Symmetries

185

sensible continuity—which can be explained by the extensionalist but not by the atomist. The only difference between the two camps is whether the contents of the total experience occur in parallel with ordinary time (as the extensionalist maintains) or lie orthogonal to it (as the atomist claims). But Lee cannot see how this difference in objective temporal orientation can have any phenomenological significance at all, given that—as he nicely puts it—”the temporal order of the parts of an extensional experience has an analogue in the presentation of temporal order in an atomic experience” (2014: 166). I think Lee is right about one thing. Take two streams of consciousness, S1 and S2. These two streams are, let us suppose, as similar as can be, save for the fact that S1 is a stream structured along extensional lines, and S2 is a stream structured along atomist lines. If S1 and S2 were totally indistinguishable phenomenally, then both streams would be sensibly continuous in the same ways and to the same extent. But can the two streams be phenomenally indistinguishable, given the dramatic differences in the way they are structured? It is here that Lee and I start to diverge.

Structures and relations To keep matters as simple as possible, let us return to Lee’s auditory total experience which consists of a succession of several brief sound-sensations. We will call it “T-Ex1,” and stipulate that it consists of three successive auditory experiences which we will label “S1,” “S2,” and “S3.” We will also suppose that these experiences are the initial phase of a stream of consciousness which consists of six successive auditory sensations, with S4, S5, and S6 following on from S3. Just as S1, S2, and S3 are sufficiently brief to fit into a single total experience (or specious present), so too are S4, S5, and S6. As this experienced succession unfolds, the atomist and extensionalist will both claim that the following sequences of total experiences will occur. The initial T-Ex1 consisting of [S1-S2-S3] will be followed by a second total experience (which we can call) T-Ex2 consisting of the experienced succession [S2-S3-S4], which is followed by a third total experience, T-Ex3, consisting of [S3-S4-S5], and then finally a fourth, T-Ex4, which comprises [S4-S5-S6]. When viewed in this way – see Figure 9.4 overleaf – it may seem just obvious that there is no phenomenological difference between the two ways of construing stream-structure. In fact, this is far from being the case. Consider first the atomist’s matrix on left of Figure 9.4, and the two initial total experiences, T-Ex1 and T-Ex2. There is a good deal of apparent overlap in the contents of these total experiences, but here appearances are misleading. For the atomist—as Lee fully accepts—the reoccurrences of

186

DEBATES IN THE METAPHYSICS OF TIME

FIGURE 9.4 the S2- and S3-type experiences in T-Ex2 may derive from the same external stimuli, but they are very definitely not the same token experiences as their counterparts in T-Ex1. Since the S2- and S3-type tokens in T-Ex1 and T-Ex2 belong to total experiences which exist at different times, and which do not overlap by part sharing, there is no option but to accept that they are numerically distinct.8 And the same applies to the S3-type experience in T-Ex3: since the latter occurs in a different momentary (or near-momentary) episode of experience to its counterpart in T-Ex2, it too must consist of a numerically distinct experience (or part of an experience). Furthermore, in the case of the S4 type, it appears in three distinct total experiences, existing as it does in the final phase of T-Ex2, in the middle of T-Ex3, and the start of T-Ex4. Again, each of these occurrences is a numerically distinct token experience in its own right. In Figure 9.5, below I have used “*” and “**” to register the numerical distinctness of these (and similar) experiences. The picture drawn by the extensionalist is very different. Here, the S2and S3-type sounds which occur in T-Ex2 are numerically identical with their counterparts in T-Ex1, and similarly for S3- and S4-types in T-Ex3. In Figure 9.5 below, the experiences on the right that are shown in bold are not new experiences, but the originals located in a succession of partially overlapping temporal wholes.

FIGURE 9.5



Flows, Repetitions, and Symmetries

187

When the situation is clarified in this way, no doubt the most obvious difference between the two analyses of this short and simple stream of consciousness lies with the number of times individual sounds get experienced. According to the extensionalist, there are six sound-experiences in total, whereas for the atomist there are twelve. This difference is a further illustration, needless to say, of the repeated contents problem. There is no need to dwell further on this here, so for now let us set it aside and concentrate instead on another difference, one that is less obvious, but no less important. Aside from its orientation in (ordinary) time, the initial total experience T-Ex1 is exactly the same on both accounts. It consists of a unified experiencing of a succession of three tones, S1–S2–S3. In this three-phase succession, S2 is experienced in two sub-phases, (S1–S2), and (S2–S3); in the first, S1 is experienced as flowing into S2, and in the second, S2 is experienced as flowing into S3. As is evident, the fact that S2 exists in these two distinct sub-phases does not mean that S2 exists twice over in T-Ex1. It doesn’t; it occurs just once. This is because the two sub-phases overlap by sharing a common part, namely (the one and only) S2. Or to put it another way, the second part of the first sub-phase is numerically the same experience as the first part of the second sub-phase. We can see from this that there is overlap via part-sharing on the atomist model too, or at least, there is within the confines of single total experiences (or specious presents). Unlike the extensionalist, the atomist does not permit partsharing between successive total experiences. Does this confinement have phenomenological implications? It does, but before turning to these I want to bring to the fore one particular aspect of the part-sharing as it exists in T-Ex1.

Identity preservation Consider again the two sub-phases (S1–S2) and (S2–S3). These occur in succession, so we hear S1 being followed by S2, and S2 being followed by S3. Now, in these two sub-phases it is a basic phenomenological datum that the S2 which occurs in (S1–S2) is numerically identical with the S2 in (S2–S3). This is a corollary of the fact that (S1–S2) and (S2–S3) are parts of a single unified episode of experience—a fact that atomists and extensionalists both accept. These identity-preserving successions (as we can call them) have their counterparts in the spatial realm. Imagine looking at a picture on a wall, and divide your visual experience at a given time into three spatially adjacent phases, which we can label A, B, and C. The experiences consisting of the spatial expanses (A–B) and (B–C) overlap by possessing a (spatial) part in common, and—in this case at least—there is no doubt at all that the B in (A–B) and the B in (B–C) is numerically the same experience. So much is

188

DEBATES IN THE METAPHYSICS OF TIME

obvious; what we can now appreciate—explicitly—is that a closely analogous form of overlap exists in the temporal case also. However, what we can also appreciate is that there is a very significant divergence between the identity-preserving (IP-) successions which exist in the atomist’s way of construing our streams of consciousness and those which exist according to the extensionalist. Figure 9.6 below depicts the two sequences of total experiences we have been discussing, with the experiencephases which form parts of IP-successions now indicated by underlining.

FIGURE 9.6 Looking first to the atomist’s matrix on the left, only S2, S3*, S4*, and S5 are underlined, because these are the only experiences (or experience-phases) which occur in the midst of IP-successions. Recalling the typographic terminology, S1, S2*, S3**, and S4** are all orphans (i.e., they occur at the start of the line, preceded by nothing), and S3, S4, S5, and S6 are all widows (i.e., they are at the end of the line, followed by nothing). In the case of the extensionalist’s view of things, as seen on the right of Figure 9.6, experiences S2, S3, S4, and S5 each occur in the midst of IP-successions. S2 occurs in the center of the IP-succession T-Ex1. Why is S3 also underlined in the T-Ex1? Because it occupies center-stage in the IP-succession T-Ex2, and (as is now familiar), for the extensionalist these token S3 experiences are numerically identical. Since the same applies to S4 and S5, the only experiences which do not occupy center-stage in IP-successions are the first and last. Hence there is just a single widow and a single orphan: S1 and S6 respectively. If we now ask “Which of these ways of construing this stream of consciousness best reflects the phenomenology?” I think the answer is plain. When we listen to a sequence of brief tones such as S1–S6, we hear an initial tone that is preceded by silence (or nothing at all), we then hear each tone-phase smoothly sliding into the next, before hearing a final tone which is followed by nothing. With the exception of the first and last tones in the sequence, we experience each tone-phase (a) occurring just once, and (b) occurring in the midst of an IP-succession.



Flows, Repetitions, and Symmetries

189

This corresponds exactly with the extensionalist’s account, as depicted on the right. It does not remotely correspond with the account the atomist is offering. If our experience were as shown on the left, only one-third of the tonephases would be experienced as occurring in the midst of IP-successions. The remaining two-thirds are either widows or orphans, that is, they are experienced as being preceded by silence or succeeded by nothing. The atomist could respond thus: “If the matrix on the left were an accurate representation of the course of our experience then it would be unrealistic phenomenologically. But it isn’t. In reality, the repeated contents are not experienced as such. Take the repeated contents out of the picture and the problem is solved.” For the reasons outlined earlier, I do not think it is plausible to suppose the repeated contents would be undetectable. However, even if we were to grant Lee this, the atomist’s predicament is scarcely any better. In the matrix below, the repeated contents (those marked with *) are—we can suppose—not experienced; the contents which do get experienced are in bold. T-Ex1 T-Ex2 T-Ex3 T-Ex4

= = = =

[S1–S2–S3] [S2*–S3*–S4] [S3**–S4*–S5] [S4**–S5*–S6]

So in this case, the stream in question consists of S1–S2–S3–S4–S5–S6. As is apparent, if anything, omitting the repeated contents only makes the atomist’s predicament worse. For although each of S1, S2 … S6 is experienced only once, only S2 in T-Ex1 is experienced in the midst of an IP-succession (hence the underlining), and S4, S5, and S6 are each orphans and widows: each of them is experienced in complete isolation from the other tone-phases. This clearly fails to correspond with the actual character of our streams of consciousness. The situation is much the same if we take a different selection of contents to represent our stream: T-Ex1 T-Ex2 T-Ex3 T-Ex4

= = = =

[S1–S2–S3] [S2*–S3*-S4] [S3**–S4*–S5] [S4**–S5*–S6]

A further possibility is depicted below: T-Ex1 = [S1–S2–S3] T-Ex2 = [S2*–S3*–S4]

190

DEBATES IN THE METAPHYSICS OF TIME

T-Ex3 = [S3**–S4*–S5] T-Ex4 = [S4**–S5*–S6] The latter scenario avoids repeated contents by taking the stream to be composed of just two non-overlapping total experiences, one with content S1–S2–S3, the other with content S4**–S5*–S6. There is still a problem with a surplus of widows and orphans. But more seriously, there is now a transition between tones which does not get experienced at all. The first total experience ends with S3, and the second starts with S4, but there is no experience of S3 flowing into S4. So once again we have a highly problematic result: in this case, an unrealistic fragmentation of the stream in question. I suggested above that in hearing a sequence such as S1–S6, we experience each tone-phase as (a) occurring just once, and (b) occurring in the midst of an IP-succession—with (of course) the exception of the first and last tones in the sequence. As is by now very clear, if this is the case, then it looks to be impossible for the atomist to accommodate this basic phenomenological datum. In contrast, the extensionalist can accommodate it fully and easily. If this is not already obvious, Figure 9.7 below makes it so. Here each of the double-headed horizontal arrows indicates a different total experience (or specious present), and the vertical arrows indicate the tone-phases which are centrally located in IP-successions. Since these total experiences overlap by part-sharing, there is no unrealistic repetition of contents.

Experiencing Discontinuity By virtue of dividing our apparently seamless streams of consciousness into experientially isolated fragments, the atomist both reduces the quantity

FIGURE 9.7



Flows, Repetitions, and Symmetries

191

of experienced continuity these streams possess, and—as a direct and inevitable consequence—increases the number of phenomenal widows and orphans, that is, experiences which are either followed by nothing, or preceded by nothing. Or so I have been arguing. A natural query at this point might be: “Yes, the atomist is committed to the existence of a great number of additional widows and orphans, but is this a problem? Would their presence make a phenomenal difference? Or to put it another way, are discontinuities in experience really discernible in the way you are suggesting?” I think the answer to this is “Yes,” and so does Lee: Take the simple example of experiencing a long continuous tone. Your auditory system—indeed your perceptual system in general—is designed to detect discontinuities in stimuli, and make them perceptually salient to you. There is therefore a big perceptual difference between experiencing a sound as starting, and experiencing a sound without experiencing it as starting. There may also be such a thing as positively experiencing the sound as continuing on from before. (2014: 166–7) There is nothing here with which I would disagree. However, Lee goes on to suggest that the atomist is better placed to explain how discontinuities can be experienced than the extensionalist. I do not think this is correct, but once again exploring where Lee’s argument goes astray will prove useful. To keep things as simple as possible, we will stay with Lee’s example of a continuous sound, which we’ll take to be a prolonged C-tone. From the extensional perspective, a single total experience (or specious present) filled with this tone, let’s call it “T,” can be represented thus: T = [C–C–C] It seems to the subject of T—or so we can stipulate—that this C-tone is part of an ongoing sound. It is here that Lee sees a potential problem. To reveal the problem we need to introduce another total experience into consideration: T* = [C–C–C] Both T and T* are experiences of an exactly similar C-tone that appears to extend over a similar short interval of time. However, the two experiences are nonetheless different in character. Whereas T is (seemingly) an experience of an ongoing open-ended sound, T* is not: this sound (or so we can stipulate) doesn’t seem to be continuing on from a previous tone, and its subject has no sense that it will continue on either. Hence Lee’s query: how can the extensionalist explain the difference between T and T*, since they are intrinsically

192

DEBATES IN THE METAPHYSICS OF TIME

just the same? For the extensionalist there is a difference between T and T*, in that the former is unified with a prior experience of a C-tone, whereas the latter is unified with a prior experience of silence, but Lee claims that “this connection with the earlier parts of the stream is actually an extrinsic property of the total experience, whereas the appearance of continuity is surely an intrinsic property of the experience, having to do with how it presents a current sound as related to the past” (2014: 167) In reply, there are a number of points extensionalists can make. First and most obviously, they will maintain that it is misleading to consider total experiences such as T and T* in isolation. If these experiences form parts of more extensive streams of consciousness, then the extensionalist will hold that they are connected by experienced relations to earlier and later stream-phases, and these relations play a key role in the experiencing of both continuity and discontinuity. For example, T might find itself being experienced amid the sequence of sounds depicted in Figure 9.8 below.

FIGURE 9.8 In this diagram “S” represents an absence of auditory experience, so a period of silence, and “D” the experiencing of a D-tone.9 Since the total experiences overlap by part-sharing, only three C-tones are experienced during the period depicted. As can be seen, the earlier phases of T are directly experienced with—and following on from—the preceding period of silence, whereas the later phases of T are experienced together with an emerging series of D-tones. The continuous auditory experience is thus interwoven with the preceding and succeeding stream-phases in a way that is not possible for the atomist. A second point is no less important. Lee is right that as our experience unfolds from moment to moment we often have a sense of where it is going, and where it has come from. This general sense can vary in strength and specificity, and have different sources. It can involve memories of what has just occurred, or anticipations (involving our sensory imaginations) of what is to come, or even conscious thoughts (e.g., “This has been going on for a while”). But more often it involves no more than a feeling which—if we were asked to



Flows, Repetitions, and Symmetries

193

put it into words—we might describe as “I had the impression that it had been going on for a while/or had just started.” In the absence of a widely used alternative, I will use the expression temporal intimations to refer to these varied and often hard-to-pin-down (but nonetheless real) aspects of our experience.10 Lee is right that temporal intimations of this sort exist, but completely wrong to think that extensionalists cannot incorporate them into their account of the overall content and structure of our streams of consciousness. I for one have previously argued (Dainton 2003) that we should recognize that temporal intimations are aspects of our consciousness which exist alongside (or pervade) our basic sensory experience.11 So in this respect the explanatory resources available to the extensionalist and the atomist are precisely the same. Consequently, even if we do consider total experiences such as T and T* in complete isolation, and ignore their experienced relationships with earlier and later experiences, the extensionalist can also appeal to temporal intimations in explaining the residual phenomenological differences between them.

Phenomenal fulcrums There is a further respect in which the atomist’s account is deficient with respect to the extensionalist’s. To bring it out requires a small modification to the scenarios we have been considering thus far. We have been focusing on simple illustrative examples featuring total auditory experiences divided into three sub-phases. Let us vary things slightly, by supposing that the auditory experiences have a slightly longer duration, so each occupies a half (rather than a third) of the total experiences to which they belong. With this small adjustment made, now consider a brief stream of consciousness consisting of a sequence of four auditory experiences, S1–S2–S3–S4. Given the lengths of the tones, this will divide into three overlapping sub-phases: (S1–S2), (S2–S3), and (S3–S4). Now, if you were to hear this sound-sequence, it is plausible to suppose that it would take the form shown in Figure 9.9 below. That is, you would first hear S1 flowing into S2; you would then hear S2 flowing into S3; and finally hear S3 flowing into S4. I think it is also plausible to suppose that you would hear both S2 and S3 occurring in IP-successions. That is, when you hear S1 flowing into S2, and S2 flowing into S3, it is manifestly the case that the S2-phase which follows S1 is numerically the same tone-experience as the S2-phase which flows into S3 (and similarly, mutatis mutandis, for S2–S3 and S4). At the very least, if people were asked to characterize their experience of this succession of tones, this description would strike most people as completely apt.

194

DEBATES IN THE METAPHYSICS OF TIME

FIGURE 9.9 That said, it is important to recognize that we are now dealing with a fundamentally different form of IP succession. In the three-phase total experience (S1–S2–S3) we encountered earlier, the S2 phase occurs between S1 and S3, and all three phases belong to a single unified experience, and so are experienced together as parts of a whole. In the sort of case we are now considering—for example (S1–S2), (S2–S3)—the central phase S2 forms the second half of an earlier total experience and the first half of a later (and partially overlapping) total experience. But although S1 is experienced as flowing into S2, and the latter is experienced as flowing into S3, it is not the case that S1, S2, and S3 all form parts of a single unified experience. However, despite this difference, it remains the case that S1, S2, and S3 do form parts of an IP-succession. For as just noted, if we were asked to describe our experience it would seem very natural to say “I experienced S1 flowing into S2, and S2—the very same S2—flowing into S3.” We can call experiences such as S2 phenomenal fulcrums. These fulcrums bind experiences belonging to distinct total experiences (such as S1 and S3) into IP-successions. If phenomenal fulcrums didn’t exist, our streams of consciousness would not be continuous in all the ways they actually are. The relevance of phenomenal fulcrums to our present concerns is straightforward. Whereas the extensionalist can easily acknowledge their existence and importance, the atomist cannot accommodate them at all. In the case of (S1–S2) and (S2–S3), the extensionalist will hold that the S2 in both total experiences is numerically the same experience, and hence does in fact occur in the midst of an IP-succession. There is no such possibility open to the atomist. For, once again, since the total experiences in question are completely separate and non-overlapping, the S2-type experience in the first



Flows, Repetitions, and Symmetries

195

must be numerically distinct from the S2-type experience in the second. A succession of this kind—that is, (S1–S2), (S2*–S3)—lacks a phenomenal fulcrum, and so does not constitute an IP-succession. This is not to say that the atomist cannot accommodate some IP-successions. If the contents are sufficiently brief, as was the case with the three brief sub-phases of T-Ex1 above, then successions of the identity-preserving variety can exist within the atomist’s framework. But this possibility evaporates for contents with slightly longer durations. Is this very plausible? Again, I suggest not. Identity-preserving succession is not just confined to the contents of single total experiences in the way the atomist alleges. In order to explain the appearances here, what the atomist can do—and what Lee does—is appeal to temporal intimations, as we called them in the previous section. We experience S1 flowing into S2 in the first total experience, and S2 is accompanied by the intimation that “this isn’t the end/ another sound will follow this one.” As I have acknowledged, intimations of this kind do exist, and our overall experience would be significantly different in their absence. But the atomist is asking a very great deal from these subtle insinuations. In reality, if the atomist is correct, our consciousness consists of brief discrete pulses, and so the flow of our experience is being continually brought to an abrupt and complete halt, every second or so. We don’t notice these total interruptions because we experience subtle (and often non-sensory) intimations that this is not the case. In contrast, the extensionalist can accept that our sensory experience is just as it seems: we experience S1 flowing into S2, and S2 flowing into S3 with no interruption. As we do so, we may well have temporal intimations as well, and these contribute to our sense that our experience has been flowing and will continue to flow. But these exist in addition to the sensory continuities which exist in our streams of consciousness; they do not replace or wholly constitute them.

Fulcrums and repetitions These considerations are also very much relevant to the problem of repeated contents that we looked at earlier. As will be recalled, in the case of the auditory succession C–D–E we encountered at the start, the atomist holds that this stream of auditory consciousness consists of two atomic total experiences, with contents (C–D) and (D-E). This gives rise to a problem: how is it that we hear just three tones [C–D–E], and not four [C–D–D–E]? To avoid this unrealistic stuttering, one option for the atomist is to claim that the second (repeated) occurrence

196

DEBATES IN THE METAPHYSICS OF TIME

of the D-tone is introspectively invisible. A question that was raised earlier, but postponed, was whether duplicates such as this D-tone could be “made to disappear” from our streams of consciousness, that is, be such that they exist, but are introspectively invisible. We are now in a better position to address this question, and see why such disappearances are more problematic than they might initially appear. So far as appearances go, both atomists and extensionalists agree that the experiencing of the succession C–D–E seems to unfold over two total experiences: in the first we hear C flowing into D, and in the second we hear D flowing into E. Given this, if my claims in the previous section are correct, the experience of middle D-tone is a phenomenal fulcrum, that is, it is an experience occurring in the midst of an IP-succession. As we have also seen, IP-successions of this sort simply cannot be accommodated within the atomist’s framework. For the atomist, this sequence of experiences unfolds thus: (C–D), (D*–E). Even if we suppose the repetition of the D-tone goes unnoticed, we still have a phenomenologically unrealistic result. We hear C followed by D, but we do not hear that same D-tone being followed by E. The experiencing of E is not phenomenally connected—not diachronically co-conscious—with the experiencing of the preceding D-tone. But this is not how things actually seem: we hear C flowing into D, and D (the same D!) flowing into E. There is a further aspect of the problem, one which is independent of IP-successions. By rendering the D* inaudible, the atomist makes it impossible for us to experience a D-tone flowing into the E-tone. Since in the envisaged case we do hear a D directly flowing into the E, this too is unrealistic. The only obvious way to remedy this deficit—within the confines of atomism, at least— is by reintroducing D*, and allowing us to hear it. But this is no solution at all: for we are now back with the problem of repeated contents. *** Returning to the bigger picture, Lee claims that atomists and extensionalists can agree on what the overall content of a given total experience is, and also “agree that it has as parts experiences of individual sounds and their relations, and agree that these parts are unified as parts of the complex whole.” We can now see that this is only partially correct. So far as individual total experiences are concerned, both camps agree that it consists of a succession of parts that are experienced as unified. But when it comes to the relations which exist among the parts of different (but successive) total experiences, the accounts we have been considering have very different tales to tell. As we have also seen, the tale told by the extensionalist is both simpler and more plausible.



Flows, Repetitions, and Symmetries

197

Symmetries Pelczar may admire the way in which the extensional approach solves (or sidesteps) the repeated contents problem, but he nonetheless finds the doctrine to be problematical, albeit in other respects. His main complaint centers on symmetry. The fundamental laws of our most successful physical theories—most versions of quantum mechanics and general relativity—are often said to be time-reversible. To a first approximation, it is not difficult to grasp what this means. If a sequence of physical events E1–E2–E3 … En are permitted by the laws in question, then the laws will also permit the time-reversed sequence consisting of En … E3–E2–E1. Imagine a video film depicting (say) the way the balls on a pool table move over the course of a game—so as to focus on the action as far as possible, the director of the film has ensured that only the on-table action is visible onscreen. If this video were played backwards, the movements of the balls you see onscreen is also physically possible. This is a consequence of the time-reversibility of the laws of motion governing the balls, along with all other physical objects in our universe. However, as Pelczar points out, the symmetries at the level of fundamental physical laws which actually obtain in our universe are rather more complicated than this simple illustrative example. The actual symmetry is known as “charge, parity, and time (CPT-) invariance.” The CPT-invariance theorem states, roughly, that our actual physical laws also apply to a universe in which (i) time is reversed, (ii) charge is reversed, along with other quantum quantities, and (iii) all spatial relations between material bodies are the mirrorimage of those which exist in our universe. The upshot of (ii) is that material objects in our universe are replaced by their anti-matter counterparts in the CPT-reversed world. Hence Pelczar’s complaint. It is generally accepted that our basic laws are CPT-invariant. However, if the extensionalist account is correct, the basic laws cannot all be CPT-invariant. This is because the extensional theory is incompatible with temporal reversibility: in a time-reversed world, very peculiar things would happen at the experiential level, things which would not happen if the relevant laws were time-reversal invariant. So the extensionalist account of temporal experience brings with it deep and revisionary commitments. Pelczar concedes that this does not necessarily mean that the theory is false. Sometimes commitments need to be revised, usually in order to accommodate new empirical data; it may be that the character of our temporal experience is such that it requires revisions to current physical theories. But if so, Pelczar is right to point out that this is a non-trivial step, one that isn’t to be taken lightly.

198

DEBATES IN THE METAPHYSICS OF TIME

The unworld case But why think that the extensionalist is committed to a revisionary position with regard to time-reversibility? It is here that Pelczar’s “Unworld” scenario comes into play. The Unworld is a temporally reversed duplicate of our world. Physical events in the Unworld are just like their counterparts in this world, only their temporal orientation is different. But if the extensionalist’s account is correct, the experience of Unworld subjects would be radically unlike our own, or so Pelczar claims. In setting up his scenario, Pelczar makes a number of assumptions. The first cluster of assumptions concern the relationship between the experiential and the physical. Pelczar finds physicalism implausible, so here works on the assumption that some form of dualism is true.12 He also assumes, plausibly enough, that even if our experiences are themselves non-physical, they are related to physical events in our brains in systematic, law-governed ways. We still lack a detailed understanding of the ways in which neural activity give rise to our different forms of consciousness, but at a higher level of generality we are nonetheless in a position to say something about these psycho-physical laws. The laws are such as to ensure that, in general, the character of human experience is such as to be interpretable as being of a world which conforms to the laws of physics. So bodies move continuously when forces are applied to them, conform roughly to Newton’s laws of motion, and more precisely to Einstein’s, and so forth. Pelczar calls this high-level nomological constraint on the psycho-physical relationship which exists between our own brains and our own experiences “the law of experience.”13 He also claims that it is plausible to think that this is a fundamental law, and as such that it will conform to CPT-invariance, in the manner of the other fundamental laws governing the universe. The second cluster of assumptions concerns the nature of our temporal experience. Pelczar assumes that some of our experiences are temporally basic, that is, they do not possess briefer experiences as parts. He takes the extensionalist to be proposing that (a) our streams of consciousness are composed of these basic experiences, and (b) over short intervals temporally basic experiences can be unified into wholes—by the diachronic co-consciousness relationship. Pelczar also works on the assumption that if temporally basic experiences exist in time at all, they must instantiate all the phenomenal properties (or qualia) they possess absolutely simultaneously. More controversially, drawing on relativistic considerations advanced earlier in his chapter, Pelczar maintains that the physical correlates of these temporally basic experiences must be point-events in spacetime. There are questionable elements in Pelczar’s defence of this last claim—as we shall see in due course—but let us grant them for the time being.



Flows, Repetitions, and Symmetries

199

We are now ready for the Unworld case proper. Pelczar envisages himself observing a pendulum swing back and forth. He sees it move from A to B, from B to C, then (as it reverses direction) from C to B, and from B to A. Let us suppose, following Pelczar, that the experiencing of (A–B), (B–C), (C–B), and (B–A) are temporally basic experiences possessing dynamic contents—in this instance, of a brief movement. Next we are invited to consider the experiences Pelczar’s counterpart in the timereversed Unworld will undergo. Pelczar claims that his counterpart would experience (B–A), then (C–B), then (B–C), and finally (A–B). Instead of seeing a pendulum move smoothly back and forth, Pelczar’s unfortunate time-reversed counterpart sees it moving from B to A, then leaping forward to C before sliding back to B, then moving from B back to C, at which point it vanishes, reappears at A, and moves smoothly to B. Evidently, if this were the case, the envisaged temporal reversal has failed to reverse Pelczar’s stream of consciousness. If his consciousness had been reversed, his counterpart would have experienced the smooth succession (A–B), (B–C), (C–B), (B–A), and so would not have experienced the pendulum moving unpredictably and discontinuously. Moreover, as Pelczar points out, this result generalizes to all experiences of continuous motion: subjects in the time-reversed world will never experience objects moving smoothly from one location to another. As a consequence, their experience will not be such as to make it plausible that they live in a world governed by Newton’s laws of motion, let alone Einstein’s. In which case, the law of experience fails to hold in the time-reversed universe. Since the law of experience is a fundamental law, it should be as time-reversible as any other fundamental law, but if we adopt the extensionalist’s view of temporal experience, it isn’t. The extensionalist’s view thus comes with a heavy cost: it requires us to qualify or reject what is otherwise a universal symmetry of nature.

Analyses and alternatives Pelczar’s line of argument is not only novel, but brings into play considerations relating to our basic physical theories that are not often considered in the context of arguments relating to temporal experience.14 Even if it does not succeed—and for reasons which will emerge, I have my doubts on this score—it introduces welcome new challenges. One obvious line of response open to the extensionalist is to question an assumption that Pelczar relies on but does not defend. The fundamental laws of physics may be time-reversible, but why assume the same applies to the laws of psycho-physics, that is, the laws relating physical events

200

DEBATES IN THE METAPHYSICS OF TIME

to experiential events? Perhaps the psycho-physical laws are not timereversible; perhaps they only function properly—that is, so as to generate perceptual experiences that are an accurate and reliable guide to our physical surroundings—in worlds with the same temporal orientation as our own. However, on reflection, Pelczar’s assumption does not seem unreasonable. After all, if the basic physical laws are fully time-reversible, the actual world and its time-reversed counterpart are completely indistinguishable in all purely physical respects. Given this, it is not easy to see how the psycho-physical laws could be different in the two worlds. If your brain states are physically indistinguishable from those of your time-reversed twin’s, if at a given time you are enjoying experiences with a certain character, so too will your twin. Of course, the situation might be different if the physical laws privileged a particular temporal direction—as we shall see shortly, this may well be the case in our world. In any event, there are other more pressing problems with Pelczar’s scenario. Let us return to the pendulum’s perceived motion. In the actual world, Pelczar sees it moving from A to C via B and back again, and so has the following four experiences: (i) [A to B], (ii) [B to C], (iii) [C to B], (iv) [B to A] If we follow Pelczar, then these experiences are non-physical occurrences, but they also have distinctive neural correlates, that is, the physical states which occur simultaneously with these experiences, and which are related to them in a law-like way. We can indicate these thus: NC1: [A to B]; NC2: [B to C]; NC3: [C to B]; NC4: [B to A] If we assume the psycho-physical laws are deterministic, then all occurrences of NC1 type physical states generates an experience that is qualitatively just like Pelczar’s experience of seeing the pendulum moving from A to B, and similarly for NC2, NC3, and NC4. To simplify, let us make this assumption. Now, Pelczar suggests that his time-reversed counterpart will have experiences of precisely the same intrinsic character as those he enjoyed in the actual world, but in reverse order: (iv) [B to A], (iii) [C to B], (ii) [B to C], (i) [A to B] Let us grant that a sequence of experiences with this disorderly character is at least logically possible. It might also seem plausible to think that the time-reversed counterpart must have experiences of this sort, given that their brains go through this sequence of neural states:



Flows, Repetitions, and Symmetries

201

NC4—NC3—NC2—NC1 Since we know in our world that momentary NC4-type neural states give rise to a [B to A]-type experience, won’t it do so in the time-reversed world as well? And won’t NC3-type states also give rise to a [B to C]-type experience? As far as I can see, while this nomological arrangement may well be possible, there is nothing in the least necessary about it. We are, after all, currently working within a dualistic framework, and the relevant laws hold between physical states and conscious states. Since conscious states are not themselves physical, there is no reason to think there is any necessary constraints on the kinds of experiences any given physical state can produce. In which case, a nomological arrangement of the following sort is perfectly possible: NC4: [A to B]; NC3: [B to C]; NC2: [C to B]; NC1: [B to A] So instead of brain state NC4 giving rise to a visual experience of the pendulum moving back from B to A, it gives rise to a visual experience of the pendulum moving from A to B—and similarly for NC3, NC2, and NC1. Is there anything which renders this psycho-physical correlation any more puzzling or problematic than Pelczar’s alternative? I cannot see that there is; within a dualistic framework, both options seem equally viable. If time-reversed Pelczar’s brain operates under this nomological arrangement, the experiences he enjoys when watching the pendulum swing back and forth are not only perfectly orderly in their own right, but correspond closely with the actual movements of the pendulum. Evidently, this nomological arrangement satisfies the criteria for being a “law of experience” as Pelczar uses the expression. Moreover, we have every reason for supposing that it is this nomological arrangement, and not the arrangement Pelczar introduces, which would be in place in a fully and genuinely timereversed world. After all, this arrangement effectively reverses the direction of subjects’ experiences, whereas Pelczar’s scrambles them hopelessly. We have been working thus far on the assumption that some form of dualism is true. There is at least one other view of the matter–consciousness relationship which (in my view at least) is a live option: the position that has become known in the recent literature as Russellian monism. This is a kind of physicalism, but one which does not expel phenomenal properties from the physical realm, nor attempt to reduce them to anything that is non-experiential in nature. Instead, it accepts that phenomenal properties are both fully real and wholly irreducible ingredients of the physical world itself. This is achieved by holding that some parts of the physical world possess intrinsic qualities, of an experiential kind, over and above all those properties that are recognized

202

DEBATES IN THE METAPHYSICS OF TIME

by current sciences such as physics, chemistry, biology, and neuroscience. The Russellian monist solves the problem of relating the experiential and the physical by dramatically expanding the range of properties physical things— such as brains—can possess. If we take this step, then our temporal experience, along with its intrinsic dynamic character, also enters the physical realm. Our typical experiences of continuous processes, such as a pendulum swinging back and forth, have an immanent directedness: the experience seems to flow or unfold in a forward- or future-directed way: we see the pendulum smoothly moving from A to B, then from B to C, from C to B, and from B back to A. For the Russellian monist, this intrinsically directional experiential flow is itself a part of the physical world. There may be no intrinsic directionality in the physical world as conceived by current (unaugmented) physics, but according to the Russellian, this conception of the physical world is incomplete. Phenomenal properties exist, and are fully a part of physical reality. And not only that: most phenomenal properties (perhaps all) are to be found in experiences with a dynamic flowing character, and this dynamism and directedness themselves belong fully to physical reality. One implication of incorporating the flow of experience into the physical world is for the reality of temporal passage. Even if it turns out that we live in an eternally real “Block universe,” the claim that in such a universe temporal passage is wholly unreal looks more than a little implausible if Russellian monism is true. If the continuity and flow which exist in our experience are themselves physical features of this universe, then this very real form of passage is also a feature of the universe.15 Less momentously, perhaps, but also importantly, there are also implications for Pelczar’s argument. If intrinsically directed experiential passage is a physical feature of the universe, then in a fully time-reversed universe the direction of this flow will also be reversed. The reason for this is straightforward: a universe in which all physical properties apart from experiential continuity are time-reversed would only be partially time-reversed, given that experience and its properties are themselves physical. In which case, Pelczar’s experience of the moving pendulum in the time-reversed universe would not be disjointed or chaotic. He would experience it smoothly moving from A to B, from B to C, from C to B, and from B back to A. His experience would be just as it is in the actual world, only unfolding in the opposite temporal direction.



Flows, Repetitions, and Symmetries

203

Complications and conundrums So Pelczar’s claim that adopting the extensionalist model leads to serious problems with temporal reversibility looks to be unfounded; at the very least the argument is inconclusive. He has overlooked two things. First, the possibility of an alternative form for the dualist’s psycho-physical laws to take. Not only do the alternative psycho-physical laws not yield chaotic streams of consciousness under temporal reversal, but they are the laws one would expect to find in a fully time-reversed world. Second, Pelczar ignores the possibility of the phenomenal being physical, in the way the Russellian monist proposes. If the latter’s doctrine is true, then since experiential flow is itself a physical process it too would be reversed (smoothly and completely) in a time-reversed duplicate of our universe. Once again, the result is an orderly rather than chaotic experience of motion. I have been proceeding thus far as if Pelczar’s scenario is itself unproblematic. In fact, this is not entirely the case. The trouble lies not with time-reversal itself, but firstly with the diachronic experiential structures Pelczar assumes must exist, and secondly with this assumption that timereversal poses a problem only for the extensionalist. In the perceived-pendulum case, as we have seen, Pelczar suggests that in the actual (non-time-reversed world) he has four total experiences, with overlapping contents—(A–B), (B–C), (C–B), (B–A)—as he watches the pendulum swing first away from, and then back to its original position. He also holds that the physical correlates of these four experiences are themselves momentary, and hence the problem. If the neural correlates of these experiences lack temporal duration, it is not obvious that the experiences themselves could overlap by part-sharing in the way they must, according to the extensionalist (who, after all, is Pelczar’s target). Certainly, if the Russellian monist is correct, and experiences are themselves physical phenomena, then (A–B) and (B–C) would each be identical with a distinct momentary physical state, and since momentary states cannot overlap by part-sharing, it looks as though the experiences with which these states are identical could not overlap by part-sharing either. Admittedly, the situation is less clear-cut for a dualist. Perhaps the laws relating the neural correlates of experiences with experiences themselves permit the experiences to overlap even when the physical states which produce them do not (and cannot). Given that Pelczar is himself (temporarily) working within a dualistic framework, he may well have assumed precisely this. In any event, it is difficult to see that the problem his scenario reveals— assuming it is a genuine problem—is one that confronts only the extensionalist. If we construe (A–B), (B–C), (C–B), (B–A) as the atomist does, that is, as

204

DEBATES IN THE METAPHYSICS OF TIME

discrete and non-overlapping, then if Pelczar’s time-reversed counterpart stream of consciousness takes the form Pelczar predicts, namely (B–A), (C–B), (B–C), (A–B), then the atomist is in trouble too! That said, since the same remedies to this problem are available to both camps, the atomist may not be unduly troubled by this fact.16 Another part of Pelczar’s discussion raises an issue that is both larger and more difficult to resolve. Drawing on some brief remarks of Russell, Pelczar maintains that the physical correlates of our experiences must be point-events in space-time. Russell rightly observed that it is very natural to suppose that the contents of our consciousness at any one time are simultaneous in an absolute fashion. If when looking at a banana at a given time, I feel a pinprick, my visual experience of the fruit and the jolt of pain are experienced together, simultaneously, as parts of a total state of consciousness. It is difficult to take seriously the idea that this simultaneity could be frame-relative, so that these experiences are not in fact simultaneous from the point of view of someone in motion with respect to me. Or as Pelczar puts it, “My [total] experience’s possession of these phenomenal properties is … absolute, is the sense that the experience instantiates the properties simultaneously according to every complete and accurate description of the world” (2014: 137). But how are we to reconcile this with Einstein’s special relativity, according to which the spatial and temporal distances between events vary from one (equally valid) inertial frame of reference to another, and where the absolute simultaneity is confined to point-events? If spacetime has a relativistic structure, then Pelczar can see no easy options. If the Russellian monist is correct, and our consciousness is itself physical, then our consciousness itself at any given moment must be condensed into a single spacetime point. If instead we assume some form of dualism is true, the physical correlate of our consciousness—the region of the brain where mind–body causal transactions take place—must also be a single spacetime point. As Pelczar is well aware, the notion that our consciousness at a given time occupies (or interacts with) no more than a single point-sized region of the physical world is highly problematic. Certainly, consciousness being thus concentrated is difficult to reconcile with the evidence that our minds depend on neural structures spanning large tracts of the brain. Neither is it obvious how all the complexity we find in our conscious states could exist in a dimensionless point. That said, the proposal may not be entirely absurd. Seth Lloyd has argued that it is possible in principle for black holes to carry out complex computations in their interiors, and being singularities, black holes are themselves point-like.17 But since we have no reason for thinking that these very distinctive singularities exist within average human brains—which is just as well—this does not provide us with very much assistance. Are



Flows, Repetitions, and Symmetries

205

there no alternatives to the condensing of consciousness that Pelczar would have us embrace? Given our current ignorance concerning the key issue of how activity in brains gives rise to consciousness in the first place, there is little here that can be said with any confidence. Even so, there are alternative possibilities which are worth exploring. The first point to note is that the relativity of simultaneity only extends to space-like separated events, that is, pairs of events that are sufficiently far apart in space, or close to one another in time, that a signal would have to be travelling faster than the speed of light to connect them. Light travels fast: 186,000 miles per second, or 983,571,056 feet per second, or about one foot per nanosecond. If a typical human brain is six inches in diameter, light will traverse this sort of distance in around half a billionth of a second. From this it follows that the temporal ordering of non-simultaneous neural events in different parts of the brain that are separated by more than about a nanosecond will be invariant, that is, the same in all reference frames. The relativity of simultaneity can only have an impact on our experience if the character of the latter can be influenced by neural processes occurring on the sub-nanometre scale. It is far from obvious that this has to be the case. If we suppose that the briefest discernible experiential events are of the order of a thousandth of a second, these are separated by six orders of magnitude from neural events whose temporal ordering is frame-dependent. There is thus plenty of room, temporally speaking, for consciousness to depend only on intermediate-level processes which exist within these six orders of magnitude. A second point to bear in mind is that causality is frame-invariant. If an event E1 is a cause of E2, then E1 is earlier than E2 from the vantage points of all inertial frames of reference. This immediately opens up one way of reconciling the absoluteness of phenomenal temporal relations with the relativity of physical temporal relations. If the existence and character of consciousness is dependent upon causal processes in the brain, then consciousness would be as immune to relativistic effects as the underlying causal processes. The frame-relativity of simultaneity does not render our computers incapable of following their programs because any physically realized computation is itself a causal process, the successive stages of which are successive for all observers in all frames of reference. This fact may not be irrelevant to conscious mentality. According to Chalmers’ familiar proposal (1996)—subsequently refined by Tononi (2008)—consciousness is a non-physical phenomenon that is nomologically correlated with computational processes in our brains. If this hypothesis is along the right lines, then the relationship between our brains and our consciousness would be largely immune to relativistic effects.18 Chalmers’ computational hypothesis is a form of dualism. What if consciousness is itself physical, and in no way separate from the

206

DEBATES IN THE METAPHYSICS OF TIME

physical activity going on in our brains? Here too there are alternatives to confining consciousness to point-sized regions. Being co-located at a single spacetime point is one way of instantiating absolute simultaneity within a relativistic universe, but there is another. The events located on the same light cone can also be taken to be absolutely simultaneous. After all, events on a light cone are connected by light rays—or some other form of electromagnetic radiation—and from the frame of reference of a light ray any spatial distance is traversed instantaneously. It is for this reason that some presentists have argued that the surfaces of past light cones are good candidates for the present moment in relativistic universes.19 Bearing this in mind, the Russellian monist might well be tempted to suggest that the phenomenal qualities which exist in momentary conscious states are absolutely simultaneous because they are instantiated on the surfaces of light cones. For this to be possible, at least some electromagnetic fields— some of those found in our brains, for example—must possess intrinsic natures which are experiential, and it is within these fields that our own consciousness exists. This is all highly speculative, it goes without saying. But then such speculation is unavoidable, given that we know so little about precisely how brains give rise to consciousness. In a similarly speculative vein, it would be wrong to close off entirely a further possibility: that spacetime itself does not possess a relativistic structure. Since physicists have yet to solve the problem of reconciling quantum mechanics with general relativity—since we lack a viable quantum gravity theory—at present we cannot be sure what story physics will eventually tell about the nature of space and time. But one thing at least is clear: physicists are no longer completely hostile to the notion that there is in fact a privileged universe-wide plane of simultaneity, Einstein and his relativity theories notwithstanding. There has long been tension between the relativity theories and some aspects of quantum mechanics. The well-known phenomenon of quantum entanglement, where a measurement made on one particle can instantly affect the properties of another particle even when the particles are far apart, does not sit easily with the non-absoluteness of simultaneity the relativity theories bring with them. But it has not proved easy to devise a theory which (a) has a preferred plane of simultaneity, and (b) matches the empirical predictions of general relativity (in particular). This has recently changed. The theory known as shape dynamics satisfies both (a) and (b). Shape dynamics is, in effect, a reformulation of general relativity which has the same empirical content—it makes the same predictions for observable effects—but also possesses a privileged temporal frame of reference. There is one other significant divergence. In general relativity, time is relative, but size is not; objects retain their size as they move through space, and it makes



Flows, Repetitions, and Symmetries

207

sense to compare the sizes of two objects even when they are separated by a vast spatio-temporal interval. In shape dynamics, time is absolute, but size is not: it makes no sense to compare the sizes of spatially distant objects; only their shapes are invariant.20 One thing shape dynamics does not do is provide us with a way of determining the preferred temporal reference frame locally, that is, by any observable effects in a small region; the global simultaneity plane is determined by the distribution of mass energy across the entire universe. There is thus room for the Russellian monist to venture another tempting speculation. If consciousness is physical, and the simultaneity relation in consciousness is absolute, might our consciousness not provide us with what we would otherwise lack: evidence pertaining to the preferred temporal reference frame that is both empirical and local?

Symmetries revisited The notion that experience itself might provide us with a guide to the preferred physical frame of temporal reference is in some ways an appealing one, but it might also prove fanciful. Shape dynamics may not prove the promising way forward in reconciling relativity and quantum mechanics that it currently seems. Time will tell. In the meantime we can conclude only this: there is more than one way in which physics and temporal phenomenology can interact. With this in mind, and by way of a conclusion, let us return to the symmetries in basic physical theories to which Pelczar has drawn our attention. As I noted earlier, the symmetry which physicists believe applies to our world is not simple temporal reversibility, but the more complex CPT-invariance property. Intriguingly, it turns out that this more complex symmetry in fact makes it possible for some physical processes to be non-symmetrical temporally. Since the 1960s, violations of CP-symmetries have been experimentally detected—it turns out that neutral K-mesons don’t decay into their anti-particles at the same rates in both temporal directions. In 2001, the BaBar experiment carried out at the Stanford Linear Accelerator Center also discovered CP-violations in B meson decays. These results can be accommodated by the CPT-theorem, but there is a cost. If CP-violations exist in nature, so too must “T-violations”: there will be some physical phenomena which are not time-symmetrical. To put it another way, for some physical processes there will be a preferred direction of time. This purely theoretical predication has only recently received empirical confirmation: in 2012, the BaBar team found evidence that B meson decay does not occur in precisely the same ways in both temporal directions.21

208

DEBATES IN THE METAPHYSICS OF TIME

The ultimate significance of these T-violations is at present unclear. It is possible to argue, certainly, that since the temporal asymmetries that have been discovered occur among exotic elementary particles, under very special conditions, they have very little relevance to ordinary macroscopic events and processes, many of which are wholly constituted of fundamental processes that are fully time-symmetric. But it is also possible to argue that they are of immense importance, at least with regard to the overall physical character of the universe. Not the least of the as-yet unexplained mysteries confronting cosmologists is why we live in a universe where there is vastly more matter than anti-matter. If CP-invariance was never violated, under any conditions, then there is no route recognized by current physics by which physical processes could lead to more matter than anti-matter being produced in the early universe. However, if some nuclear processes are not CP-invariant, if some processes end up producing more matter than anti-matter (even by a small amount), then matter-dominance is no longer a mystery, as Sakharov pointed out in a ground-breaking paper in 1967. That said, the CP-violations involving the weak force that have been discovered so far are insufficient to explain the extent to which matter predominates anti-matter in our universe. But since it is by no means out of the question that there are other sources of CP-violations, this is by no means the end of the story. It may well turn out to be the case that this type of symmetry violation is the origin of the universe as we know it. Although the precise relevance of this for our own experience is as yet even more unclear, and likely to remain so for the foreseeable future, there are nonetheless a couple of points worth noting. I noted earlier that Pelczar’s claim, in a dualistic context, that psychophysical laws (relating non-physical experience to physical states) will be time-reversible was a reasonable one, if it is the case that there is no preferred direction of time in physical reality. But if, as seems to be the case, there is a preferred temporal direction in the physical realm, then it is not out of the question that psycho-physical laws might not be time-reversible. At the very least, we no longer have a compelling reason for supposing they must be. From the vantage point of the Russellian monist, the immediate implications are perhaps less obvious. Even so, we do know that our experience possesses an intrinsic directional flow, and hence is a temporally asymmetrical phenomenon. If our fundamental physical laws themselves permit temporally asymmetric phenomena, the Russellian will find it tempting to suggest that this may not be a coincidence.22



Flows, Repetitions, and Symmetries

209

Notes   1 I will not attempt here to adjudicate on the issue of whether or not Walker has succeeded in describing a world (and subject) that is truly timeless. The subject knows no change or persistence, and that is enough for present purposes.   2 Hoerl (forthcoming) doesn’t go quite this far but does suggest that our visual experience of motion has this character.   3 See Dainton (2010) for an overview of a wider range of options.   4 These two competing views of the structure of a stream of consciousness may well be a more fundamental difference between atomism and extensionalism than their differing relationships to ordinary objective time. For example, it may be possible for a stream of consciousness consisting of overlapping specious presents to exist in the absence of any objective time (e.g., some forms of idealism have this consequence). It may also be possible, logically at least, for such a stream to exist in a single moment of objective time. In the latter case, the stream would be subjectively extended (seemingly for hours, days, or even longer), but objectively momentary.   5 To keep things as simple as possible, I will be focusing here on the most straightforward form of atomism, the version where neighboring atomic experiences in the same stream of consciousness are completely distinct and discrete from one another. In his contribution to this volume, Lee introduces interesting variants in which the (so-called) atomic experiences are brief, but partially overlap by part-sharing—the overlap is partial because it is restricted to experiences in different sensory modalities, rather than entire stream-phases. I will leave for another occasion a discussion of whether this model of temporal experience is viable, and whether it should really be classified as a variant of extensionalism.   6 The situation would be very different if one were to hold that the content and character of a subject’s stream of consciousness are determined entirely by the representational content of the experiences it contains, where the latter content is (roughly) how one takes the external world to be on the basis of the perceptual experiences one is having at the time. For anyone who subscribes to this “strong representationalism,” the repeated contents we have been considering here would not feature in one’s stream of consciousness, since their representational content is entirely discounted. In this view, if a given sequence of sounds leads one to conclude that the tones C–D–E have just been played on a piano, the only experiences one has are of a C-tone, a D-tone, and an E-tone. This is why, I take it, Michael Tye does not think the repeated content problem is really a problem. But Tye’s brand of representationalism is itself a contentious doctrine and I for one do not find it plausible. When Lee suggests that he and Tye are making essentially the same point when both claim that the repeated contents (such as the duplicated D-tone) would not be invisible to introspection, it is not clear that this is the case, because it is not clear that Lee is himself committed to Tye’s representationalism. Indeed, in his comments on Phillips’ “transparency argument,” he suggests that there are reasons for rejecting

210

DEBATES IN THE METAPHYSICS OF TIME the claim that “experience … only reveals the properties that external events are experienced as having, not any psychological properties.”

  7 Lee mentions empirical considerations which favor atomism, in the form of the “trace integration argument” outlined in his chapter. Since I cannot discuss this properly here I will restrict myself to one brief comment. It may well be that our brains process some temporal information in the way Lee describes (i.e. representing temporally extended phenomena in neural representations that are not themselves temporally extended). But it does not follow from this that the neural processes responsible for temporal experience take this form.   8 If we assume that experiences are individuated by subjective character, time of occurrence, and subject, then the experiences in question are distinct by virtue of existing at different times. The same applies a fortiori if we individuate by reference to subjective character, time of occurrence, and neural realization (rather than by subject). The experiences in question not only exist at different times, but their neural realizations also differ. Lee is neutral between these options; I prefer the second. I noted earlier that in his chapter, Lee goes on to consider whether there might be a version of atomism in which the neural realizations of some co-streamal total experiences do overlap. The result is an interesting halfway house between the standard form of atomism and extensionalism. For present purposes what matters is that for Lee, the partial overlap version of atomism is not the standard form: in the latter, successive total experiences (along with their neural realizations) do not overlap.   9 If the subject of these experiences is aware of these periods of silence, then they are, presumably, having experiences of a non-auditory kind during the period in question—if they weren’t, these silences would not be featuring in streams of consciousness at all. With a view to keeping things as simple as possible, these additional experiences are not depicted. 10 In his 1909 A Textbook of Psychology, Titchener calls them “attitudes of consciousness” and writes that “What precisely these attitudes are … is still a matter of dispute. They are reported as vague and elusive processes, which carry as if in a nutshell the entire meaning of a situation … We may have, under the guise of an attitude, the consciousness that something is real, that it is lasting a long time, that it is over more quickly than we had expected, that it is the same as what came before, that it is incompatible with some other thing, that it makes sense, that it is novel, that it is on the tip of the tongue, that it will be difficult … There seems literally to be no end, till we have exhausted the resources of the language, to the catalogue of possible attitudes” (505–6). Also see Findlay (1955). James also drew attention to these aspects of consciousness in the Principles—and with the recent resurgence of interest in cognitive phenomenology, they are once again being given their proper due. 11 There I call them “fringe feelings” rather than temporal intimations, but the intent is the same. 12 Pelczar falls short of endorsing dualism: he is also open to the possibility that a form of phenomenalism is true, and hence that fundamentally only



Flows, Repetitions, and Symmetries

211

experiences (and perhaps their subjects) exist. Phenomenalism may well be more defensible than is often assumed, but to simplify the discussion I will focus solely on the dualist option. 13 If phenomenalism is true, then the law of experience exists solely as a constraint on human experience. 14 One notable exception is Lee’s own “Consciousness in a Space-time World” (2007). 15 Or in other words, passage—or one significant form of it—exists, even if the universe-wide creation and/or annihilation which are features of the presentist and growing block metaphysical pictures do not exist. For more on this theme, see Dainton (2012). 16 If the atomist holds—as Pelczar appears to—that the objective temporal locations of a specious present are irrelevant to the subjective character of the streams of consciousness to which they belong, then the option exists of claiming that the (B–A), (C–B), (B–C), (A–B) sequence in the time-reversed world would be experienced as the orderly and lawful succession (A–B), (B–C), (C–B), (B–A). Even if this were the case, the dislocation between the subjective appearances and the objective temporal locations of the external causes of these subjective appearances would make timely action problematic—and hence there are other problems in store for the atomist who is tempted by this line. 17 See Lloyd’s “Ultimate physical limits to computation” (1999). 18 In this connection it is also worth bearing in mind that while the digital computers commonly employed in everyday life depend on global internal clocks as they run through their programs, step by step, this is by no means essential to computation per se: so called asynchronous computers do not rely on central clocks. In “clockless chips,” the basic processing modules only operate when there is some work for them to do, and not whenever a central clock ticks; entirely local “handshaking rules” govern the flow of data between modules. In an asynchronous computer, what is simultaneous functionally need not be simultaneous temporally, and it is the former which is directly relevant to experience—assuming, of course, that consciousness and computation are linked in the way Chalmers and Tononi propose. See Sutherland and Ebergen (2002) for an overview of asynchronous computing. 19 For more on cone presentism, see Hinchliff (1998). 20 For more on shape dyanamics, see Smolin (2013: 164–71), and Barbour, Koslowski and Mercati (2013). For a different approach—featuring the “emergent block universe”—see Ellis (2006). 21 For full details, see Lees et al. (2012). 22 For helpful comments on an earlier draft of this chapter, my thanks to Nathan Oaklander, Geoff Lee, and Michael Pelczar.

212

DEBATES IN THE METAPHYSICS OF TIME

References Barbour, J., Koslowski, T. and Mercati, F. (2013), “The Solution to the Problem of Time in Shape Dynamics.” http://arxiv.org/abs/1302.6264 Chalmers, D. (1996), The Conscious Mind. Oxford: Oxford University Press. Chuard, P. (2011), “Temporal experiences and their parts,” Philosopher’s Imprint, 11. Dainton, B. (2003), “Time in experience: reply to Gallagher,” Psyche, 9, 12. —(2010), “Temporal Consciousness,” in Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy. http://plato.stanford.edu/archives/fall2010/entries/ consciousness-temporal/ —(2012), “Time and Temporal Experience,” in A. Bardon (ed.), The Future of the Philosophy of Time. New York: Routledge, pp. 123–48. Ellis, G. (2006), “Physics in the real universe: time and spacetime,” General Relativity and Gravitation, 38, 1797–824. Findlay, J. N. (1955), “The logic of Bewusstseinslagen,” Philosophical Quarterly, 5, 18, 57–8. Hinchliff, M. (1998), “A defense of presentism in a relativistic setting,” Proceedings of the 1998 Biennial Meeting of the Philosophy of Science Association, 67, 3, part II, supplement. Hoerl, C. (2014), “Do we (seem to) perceive passage?,” Philosophical Explorations, 10.1080/13869795.2013.852615 Lee, G. (2007), “Consciousness in a space-time world,” in J. Hawthorne (ed.), Philosophical Perspectives, 21, 1, 341–74. —(2014), “Extensionalism, Atomism, and Continuity,” in L. N. Oaklander (ed.), Debates in the Metaphysics of Time. London: Bloomsbury. Lees, J. P. et al. Stanford Linear Accelerator Center. (2012), “Observation of time-reversal violation in the B0 Meson system,” Phys.Rev.Lett., 109, 21. Lloyd, S. (1999), “Ultimate physical limits to computation,” Nature, 406, 1047–54. Pelczar, M. (2014), “Physical Time, Phenomenal Time, and the Symmetry of Nature,” in L. N. Oaklander (ed.), Debates in the Metaphysics of Time. London: Bloomsbury. Sakharov, A. D. (1967), “Violation of CP invariance, C asymmetry, and baryon asymmetry of the universe,” Pisma Zh.Eksp.Teor.Fiz., 5, 32–5. Smolin, L. (2013), Time Reborn. London: Allen Lane. Sutherland, I. E. and Ebergen, J. (2002), “Computers without clocks,” Scientific American, August. Titchener, E. (1909), A Textbook of Psychology. New York: Macmillan. Tononi, T. (2008), “Consciousness as integrated information: a provisional manifesto,” Biological Bulletin, 215, 216–42. Tye, Michael (2003), Consciousness and Persons: Unity and Identity. Cambridge, MA: MIT Press. Walker, R. (1978), Kant. London: Routledge & Kegan Paul.

10 Divine Events* Joseph Diekemper

Introduction The purpose of this chapter is to approach the question of God’s temporality via a study of the nature of events. If events are necessarily extended entities involving change, then there is no such thing as a non-temporal event. And if God is necessarily the subject of such events, then he is necessarily temporal. In the first section I will defend the essential temporality of events, and in the subsequent section I will consider various ways of conceiving of divine events, concluding that these necessarily exist.

The ontology of events In this section I will argue that we do not have good reason to allow instantaneous events into our ontology, and therefore that events should be conceived as essentially temporal and extended entities which involve change.1 Why think that events cannot be instantaneous? For the simple and common-sense reason that events are, intuitively, processes of change. Lombard (1986) has argued that events are processes of change; and it certainly is integral to the common-sense concept of an event—or “happening”—that it involve change.2 But while common sense may be a good starting point in metaphysics, it can only take us so far. Once we start applying a little metaphysical pressure to a common-sense concept, we often find that the concept cannot hold up under the strain, and that it requires some revision. So even if we normally conceive of events as processes of change, is this necessarily the case? Leftow (2002) has argued that events

216

DEBATES IN THE METAPHYSICS OF TIME

need not be extended, and therefore that they need not involve change. Clearly, change cannot occur at an instant, so if an event can occur at an instant, then events are not necessarily processes of change. So Leftow applies some metaphysical pressure to the notion that events cannot be instantaneous, and concludes that this notion is mistaken. This paves the way for him to argue that God can be the subject of an instantaneous, timeless event; and this, in turn, helps to provide a coherent conception of Boethius’ notion of the divine “eternal present.” Leftow brings four main points to bear against the thesis that events cannot be instantaneous. I will consider these in turn, and will conclude that the common-sense conception of events as essentially temporally extended entities can hold up against the pressure Leftow applies to it.

Instantaneous velocity First, Leftow considers the case of motion (Leftow 2002: 26). We can take any instantaneous slice of a given motion (such as my walking from A to B) and claim that that slice is a temporal part of the event that is my walking from A to B. There is, for example, an instant at which I am midway between A and B, and my being in that position at that instant is a part of my walking from A to B. But, according to Leftow, a temporal part of an event can only be an event itself. Events only have events as temporal parts, and motions only have movings (which are events) as temporal parts. Why think that my instantaneously being midway between A and B is a moving? Because, according to Leftow, I have an instantaneous velocity at that point, and obviously only objects that move have velocity. So the instantaneous slice of my motion is an instantaneous moving and is therefore an event. But is there any reason to think that instantaneous velocity, which is introduced in a scientific context by special definition in terms of limits, is an actual velocity possessed by an object at an instant? Suppose I reach the midway point between A and B at time t. And suppose we measure my velocity over increasingly smaller and smaller intervals of time ending at t, and these measurements get closer and closer to 1 m/s. Further, suppose that we measure my velocity over increasingly smaller and smaller intervals of time beginning at t; and suppose that these measurements also get closer and closer to 1 m/s. This, according to definition, yields an instantaneous velocity of 1 m/s at time t. But note that the definition is given in terms of intervals of time, and that it is over intervals of time that the velocity is actually measured. Furthermore, notice that the definition treats the instant t not as containing an event, but as the bounding point of intervals over which velocity is measured. Although the definition is neutral with respect to the ontological



Divine Events

217

status of events, it nonetheless treats instantaneous velocity as a purely theoretical concept derived from velocities measured over temporal intervals; thus the definition does not support the inference from an instantaneous velocity to an instantaneous moving. Given how instantaneous velocities are defined, Swinburne (1994) argues that we should not think that they are somehow discovered with the use of limits; rather, according to Swinburne, an instantaneous velocity is just a limit of velocities measured over series of intervals—intervals which are bounded by the instant to which the instantaneous velocity is, by definition, attributed (Swinburne 1994: 73). If this is correct, then there seems little reason to suppose that instants are temporal parts of events. In fact, it seems wholly implausible that some contiguous set of instants of zero duration could somehow add up to an interval or event of non-zero duration. If, however, instants are merely the bounding points of temporal intervals, then there is no pressure to admit instantaneous events into our ontology (other than, perhaps, as theoretical entities).3

Succeeding events So Swinburne endorses (what I am calling) the common-sense view, according to which instants are merely the bounding points of temporal intervals, and things happen over temporal intervals, rather than at instants. Leftow, however, puts further pressure on the common-sense view by claiming that the instantaneous moving, which he takes to be part of a motion, can be thought of as the successful culmination of a process which leads up to being at that location. And this, according to Leftow, “sounds like an event” (Leftow 2002). The thought is that, in ordinary language, we speak of reaching a particular place (such as my reaching the midway point between A and B), where the reaching implies succeeding; and, according to Leftow, a succeeding is an event. Thus, a perfectly legitimate answer to the question, “What happened at t?” is, “I reached the midway point between A and B at t.” So reaching there must be an instantaneous event which happened at t. Leftow acknowledges that there is another way of describing the same situation: perhaps reaching there only refers to the event involving changes leading up to being there. But being there is not an event, it is a state “which terminates the events leading up to my being there” (Leftow 2002). So here we have two different ways of describing the same situation, one of which characterizes my location at t as an instantaneous event, and the other which characterizes it merely as the terminus of an event. How shall we decide between these two characterizations? It is not clear that ordinary language is going to help us adjudicate in these kinds of ontological

218

DEBATES IN THE METAPHYSICS OF TIME

disputes. If we took the ontological commitments of ordinary language at face value, we would have a far more bloated ontology than most of us would be prepared to allow. Yes, we can say things like, “it happened that I reached that point at that instant,” without offending the ear. But just as most of us would not want to allow the existence of a property for every predicate of language, we also should not affirm the existence of an event for every linguistically proper use of the verb “to happen.” Events are concrete particulars which serve as the relata of causal relations, and are therefore more coarse-grained than mere facts, which have a one-to-one correspondence with every true, linguistically correct description of reality (i.e. a different description corresponds to a different fact). So just as one and the same event can admit of several different descriptions, without implying that more than one event is being referred to, so, too, not every true, linguistically correct description which sounds like an event is an event.4 What we should ask ourselves is what our fundamental theory of the world commits us to; and while we should not avoid an abundant ontology as a matter of principle, we should endeavor to reduce our ontological commitments wherever doing so does not reduce the explanatory power of our theory. Given this desideratum, does the explanation of my reaching the midway point between A and B really require both the event of my travelling from A to the midpoint of AB, and the event of my arriving at the midpoint of AB (at t)? Or does is it only require that the former event terminate at time t? I submit that the ontological parsimony of the latter explanation is, for that reason, more virtuous. There is a potential worry, however, with this line of argumentation. I am claiming that my location at t is the terminus of an event involving my walking from A to the specified location, and as such does not constitute a distinct event. Can we not, however, conceive of a possible world which consists only in my being at that location at t, and nothing else? And if such a world is possible, then what should we say about the ontological status of that instant? If it is just the terminus of an event which, ex hypothesi, does not exist, then it appears that the instant also cannot exist.5 But would I really wish to claim that the described world is not possible? My answer is that there might be possible worlds in which objects admit of properties and relations which never change, but that there is no sense in which anything occurs in such worlds, and therefore no sense in which time exists in such worlds.6 So if there is a world in which I am located at the midpoint between A and B, but in which nothing ever happens, then there are neither temporal instants nor events in that world (see “Changeless instants,” below).



Divine Events

219

Coming to be Leftow’s next point focuses on the notion of coming to be. He argues that coming to be can only be a change if it is a change in the entity that comes to be; but, since the entity that comes to be does not exist until the instant at which the process of coming to be is complete, coming to be cannot be a change in the entity that comes to be. Thus, coming to be is an instantaneous event. Suppose, for example, that x comes to be at t. Prior to t, x does not exist; subsequent to t, x already exists. So x’s coming to be is not a change in the individual x that comes to be. Thus t cannot be conceived as the terminal instant of an extended event (i.e. x’s coming to be), since there is no subject of change prior to t. Therefore coming to be is an instantaneous event, since t is the first instant of x’s existence. My response to this argument is to claim that it conflates qualitative change with substantial change. Obviously x’s coming to be is not a process of qualitative change, since x does not exist to undergo change during the process of its coming to be. But clearly something is changing in the coming to be of a new individual. For Aristotle, coming to be is a case of substantial change, since it involves the coming into existence of a substance that did not previously exist. So what is the subject of change in a case of substantial change? It is the matter of which x is formed. If x is a member of the substantial kind K, then x’s coming to be is a change in the matter which eventually constitutes x—that change being the instantiation of the substantial kind K. So x’s coming to be is not something which happens to x, but it is something which happens to x’s matter, and thus it is a temporally extended event involving change. Of course, we could characterize x’s coming to be as a process involving substantial change, and at the same time characterize x’s coming to be (at t) as the successful completion of that process, and therefore as an instantaneous event. Again, however, the latter is merely a linguistic characterization, and it is one that does no ontological work (given the argument in the previous section). Leftow considers three examples in this context: one is the first moment of time, one is the coming to be of the universe according to Big Bang cosmology, and the third is the coming to be of Michelangelo’s David (Leftow 2002: 27–8). Leftow takes all three examples to be of a par, and the notion of coming to be to be a perfectly general one. However, given my response to his argument, it is not clear that I can treat all three examples as involving the same kind of event. It is one thing to analyze David’s coming to be as a case of substantial change in David’s matter; it is another to analyze the first moment of time or the coming to be of the universe as a case of substantial change—substantial change in what? Setting aside, for the moment, the

220

DEBATES IN THE METAPHYSICS OF TIME

example of the first moment of time, I will focus first on the example from Big Bang cosmology. According to that cosmology (and according to the theological doctrine of creation ex nihilo), there is no matter prior to the existence of the universe, and therefore no potential subject of change. This entails, according to Leftow, that the universe’s coming to be is not a change but an instantaneous event. There are two ways to consider the metaphysics of this situation: one is atheologically, and the other theologically. Atheologically, either the initial appearance of the singularity—from which the universe expanded—was a contingent, yet uncaused, cause, or it was an effect of some prior event (for example, the big crunch of an earlier universe). Only in the former construal would it be true to say that the first moment of the universe was an instantaneous event; since in the latter construal, the appearance of the “original” singularity is the terminus of an event whose origin is the big crunch of an earlier universe. But how plausible is the former construal? If the scientific explanation for any event or state is made by reference to its causal antecedents (a plausible assumption), then the existence of an original, contingent, yet uncaused, cause would entail the impossibility of explaining all that is causally downstream from the event. This strikes me as a particularly unpalatable option for those engaged in atheological explanation, since it blocks a certain response to the cosmological argument. Proponents of that argument will claim that atheists are unable to ultimately explain existence. In response, atheists will say it suffices for explanation to be able to explain any event in terms of earlier events—that is as “ultimate” an explanation as is required. Clearly, this response loses its force if there is a contingent, first cause. Furthermore, although Leftow speaks in terms of the “appearance” of the Big Bang singularity, and I have retained that language above for the sake of argument, it may misrepresent the actual Big Bang cosmological models. Halvorson and Kragh (2011) claim that in the singular spacetime models which most closely model our universe, if t0 is the absolute lower bound of the time parameter t, then as t decreases towards t0, “t0 is an ideal point that is never reached: the universe exists at all times after t0, but not before or at time t0.” Thus, in these models, there is no initial state of the singularity or first moment of time, even though the universe is finitely old; there is only a first interval of time. This would entail that, from the atheological perspective, neither the Big Bang example nor the first moment of time example can do the work that Leftow intends them to. In any event, given the context of Leftow’s argument, as one in which conceptual room is being made for the notion of a divine “eternal present,” atheological considerations are not, perhaps, terribly relevant. Considering the metaphysics of the situation from the theological perspective, the appearance of the singularity is, once again, either an original



Divine Events

221

singularity, or it is an effect of some earlier cosmic event (big crunch, etc.). In either case, however, God is the ultimate, necessary cause of whatever is the genuine original singularity (we are assuming, for the sake of the argument for instantaneous events, that there was such an entity). One question here is how we should characterize the change—if, as I maintain, it is a change— involved in the first act of creation. It cannot be substantial change in the Aristotelian sense, since there exists nothing material prior to the appearance of the original singularity. Leftow considers and rejects the thought that the first act of creation would be a change in the “way things are,” and I think he is correct that this is too diffuse a notion to serve as the subject of change. But what about God? Before his first act of creation he has never created (trivially), so can’t his first act of creation be a (relational) change in him? Well, this description of the first act of creation entails that God is temporal, and that is ultimately what is at stake here. Similar worries about circularity attend the example of the first moment of time, since a first moment of time is only plausible on the assumption that God is atemporal: if God is necessarily eternal and temporal, and time is necessarily unified, then there could be no first moment of time. The lesson here is, I think, that the examples of creation ex nihilo and the first moment of time (from the theological perspective) cannot be used to demonstrate either the possibility or impossibility of instantaneous events. One must assume a conception of divine eternity in order to employ the examples in such a way, and the conclusions reached bear directly on the nature of divine eternity.

Changeless instants In the final argument I wish to consider, Leftow takes the occurrence of an instant to be an instantaneous event which does not involve change. He argues that the only way to deny this is to endorse the reduction of time to actual events and their relations. According to Leftow, however, this reduction has the consequence that “nothing could have occurred at any time save what actually did occur then,” and this is not plausible (Leftow 2002: 32).7 I will argue that the reduction of time to actual events and their relations can avoid this alleged ramification, and thus that this should not be cited as a reason to reject the reduction. And as Leftow acknowledges, if one were to reduce time in this manner, then one can reject the claim that changeless instants occur.8 I will start my argument by acknowledging agreement with Leftow on his main point in this context. He argues that the reduction of time to actual events and their relations (henceforth “TR” (Temporal Reductionism)) entails

222

DEBATES IN THE METAPHYSICS OF TIME

essentialism about times. If times are nothing more than simultaneous sets of events, then the events simultaneous with any given time t are essential to t. Thus t does not exist in worlds in which the set of events which actually comprises t does not exist. So, suppose in the actual world @ that I drive to Donaghadee at t. Now suppose that in some other, very close possible world w, I instead drive to Bangor, where my driving to Bangor is simultaneous with all the events in w that my driving to Donaghadee is simultaneous with in @. Since my driving to Donaghadee at t is an essential constituent of t, t does not exist in w, and so it is not true in w that I drive to Bangor at t. So I agree with Leftow that this essentialism about times cannot allow for alternative possibilities to obtain at times that are transworld identical. But shouldn’t our metaphysics allow for this? Well, certainly our metaphysics should allow for claims such as, “I might have driven to Bangor at t,” and the question is whether essentialism about times can allow for that claim without transworld identity for times. I think it can, by employing the concept of a counterpart.9 A modal counterpart to an actual individual is an individual existing in some other possible world which bears a relation of similarity to the actual individual. The closer the world, the closer the relation of similarity. The modal counterpart in w of actual time t is the time at which I drive to Bangor instead of driving to Donaghadee (call it tw). So the claim, “I might have driven to Bangor at t” is true in virtue of my driving to Bangor in w at tw; and I think that this is all that is required to allow that things might have gone differently. So although alternative possibilities cannot obtain at (strictly) identical times, they can obtain at times that are otherwise identical to actual times (otherwise but for the non-occurence of the event actually occurring at that time). Finally, it is important to stress that this account of counterpart times need not buy into the counterpart theory of modal realism. If one endorses transworld identity for other kinds of individuals, and rejects modal realism and the indexicality of actuality, then one may continue to do so while acknowledging that there are counterparts to actual times. What we have seen in this section is that the common-sense view that events are processes of change can withstand the sustained metaphysical pressure that Leftow brings to bear on it. And if events are processes of change, then they are temporally extended entities; that is, there are no non-temporal events. Assuming that one accepts the common-sense status I have accorded the thesis that events are processes of change, then this section provides considerable evidence in favor of the thesis.



Divine Events

223

Divine events Given the conclusion of the previous section, if God is a subject of events, then God is temporal.10 In other words, if divine events exist, then God is in time. There are several concerns associated with the concept of a divine event, but most of these stem from the worry that God can only be the subject of an event in virtue of his interaction with creation, since it would be inappropriate and overly anthropomorphic to characterize God’s mental life as involving discrete, ordered events. Furthermore, the objector will note that even if we assume God’s mental life to involve events, and take this to imply that God is temporal, then we must say that God waited an infinite amount of time before creation (since a temporal God’s life extends an infinite amount of time into the past). Lacking some principled reason for why God would wait an infinite amount of time to create, the objector concludes that it is absurd, and that we should therefore reject the initial assumption. I will consider three different responses to this set of concerns: the possibility of co-eternal creation; the possibility of changeless time prior to creation; and the possibility that reflection on God’s nature as a person can answer the objections.

Co-eternal creation According to the standard theistic view, God is the creator, but suppose that his being creative is not an accidental property; that is, suppose that he is essentially creative. Perhaps, then, he would not await an infinite amount of time to create. The objector argues from the absurdity of God waiting an infinite amount of time to create to the conclusion that he must be timeless. But if God is temporal, and he does not have a principled reason for delaying creation, then surely it is possible that he did not delay; that is, he must always have been creating. Clearly, if we take the doctrine of creation ex nihilo and Big Bang cosmology seriously, with their implications of a finitely old universe, then our universe cannot be co-eternal with God.11 But this hardly rules out co-eternal creation, since God’s creative activities could go well beyond our own universe. If God is temporal, presumably he created the heavenly places and its denizens (e.g. angels) before he created our universe. But if his creative activities are to fill an infinite past, then there must surely have been more than this. For this reason, the doctrine of co-eternal creation seems to imply the existence of a multiverse: God has been creating other universes for an infinite amount of time. If co-eternal creation can be made sense of, then one could claim both that God is the subject of temporal events (in virtue of his interaction with his creation), and that there is no puzzle about

224

DEBATES IN THE METAPHYSICS OF TIME

why he waited to create. There are, however, some potential pitfalls to this approach. For one thing, there is a concern about whether a temporal God’s ontological priority can be confirmed unless he also has temporal priority over his creation. Second, if God has always been creating, then perhaps there are an infinite number of universes among the multiverse, and this seems problematic. I will consider these two problems in turn. First, I suppose the defender of co-eternal creation would urge us to get away from thinking, “first God existed, then he created,” in order to affirm his ontological priority over creation. It is not as though God came into existence, then decided to create; he has always been existing, so (perhaps) he has always been creating. Augustine attributes the doctrine of co-eternal creation to the Platonists, though the context there is different, since they are defending the view that the human soul is co-eternal with God (City of God, bk. 10, ch. 31).12 But Augustine provides them with a helpful analogy: a foot eternally planted in the dust. There is no temporal priority of the foot and the footprint if they are co-eternal, but there seems to be no question that the print was formed by the pressure of the foot: without the foot, there could not have been a print. In this way, the creation might ontologically depend upon God, even though some of the creation is co-eternal with God (again, even if not our universe, then some universes, and presumably angels). Furthermore, the Thomistic doctrine of creatio continuans could be employed here to help preserve God’s ontological priority over a co-eternal creation. According to this doctrine, God’s causing things to exist is a matter of him continually sustaining them wholly by his power. If that power were removed by God, then the creation would cease to exist. This seems to imply that the causal relation between God and his creation is primarily metaphysical and not temporally ordered from earlier than to later than, in which case, God does not require temporal priority over his creation in order to have ontological priority.13 The second worry, however, seems more problematic. If God has been creating for an infinite amount of time, then this implies that his creation is infinitely old. Since we have good reason to believe that our universe is not infinitely old, and since, as I argued above, co-eternal creation implies the existence of a multiverse (again, assuming that our universe is not infinitely old), it also implies that there are an infinite number of universes in the multiverse. The existence of a multiverse is difficult enough for many theists to embrace, but even those who do, argue that the justification for such ontological extravagance is that it can help explain how the creation as a whole is the unique best possible world (i.e., the universe of goodmaking properties in all the universes, taken together, maximally outweigh the bad-making properties of our universe).14 But if the multiverse is the unique, best possible world, then it appears that it cannot be infinite, since I



Divine Events

225

take it that an infinite number of entities cannot be unique. So this gives us little independent reason for embracing such an extravagant ontology as an infinitely large multiverse.

Changeless time prior to creation The second possibility is articulated by Swinburne (1994). He argues for the possibility that God, prior to creation, was the subject of a single, undifferentiated event—perhaps a mental act of self-awareness. According to Swinburne, during this pre-creation temporal interval there would be no metric of time. This is because a metric of time is contingent and dependent upon natural laws, inasmuch as measurement of temporal intervals requires the regularity of law-governed physical processes. For Swinburne, this means that the occurrence of events is sufficient for the existence, and ordering relations, of time; but for there to be a fact of the matter about the measurement of those relations, there must also exist laws of nature. Thus Swinburne argues that as long as God’s pre-creation mental act of selfawareness did not involve any change throughout its duration, there would be no fact of the matter about whether the event lasted an instant or an infinite amount of time. So, we can have time before creation, without any absurdity associated with how long God waited to create. How can there be time with no succession of events, and no change throughout the event in question? For Swinburne, the interval during which the act of self-awareness takes place is still a temporal one, because it is possible throughout that interval that some change take place. That is to say, Swinburne endorses the modal reduction of time, according to which time is reduced to relations between actual and possible events (as opposed to relations only between actual events). I think this solution is problematic for two reasons. In the first place, I reject the modal reduction, so in my ontology there can be no time “during” the act of self-awareness.15 This means that if I wish to avail myself of Swinburne’s answer, I must affirm that time did not start “until” the first moment of creation. This view, which is as difficult to articulate as it is to believe, is the one adopted by Craig (2001). It is an odd view, and considered problematic by many. Consider, in this view, that there is no “before” God has created, since before is a temporal relation and, in this view, God is timeless without creation. The view also renders God’s temporality an accidental feature of his existence (if we assume that he need not have created). But even if I were to endorse the modal reduction of time to which Swinburne subscribes, and so avoid Craig’s “timeless sans creation” conception of divine eternity, there would still be a problem: I do not think

226

DEBATES IN THE METAPHYSICS OF TIME

that the act of self-awareness by God which Swinburne considers is possible. To see why, we need to look at a third possible response to the concerns attending divine events before creation. This is the response that I favor.

The personhood of God This response claims that God’s essential personhood requires that his thoughts are dynamic, ordered events. According to this response, the best way to interpret the concept of the Imago Dei (“image of God”), in which the Bible claims human beings are made, is in terms of personhood.16 It is therefore the category of personhood that we have in common with God. Furthermore, there is good reason to believe, in this conception, that the category of personhood is essentially relational. Emil Brunner (1952) argues that the Imago Dei is not some intrinsic property which we have independently of our relation to God; rather, it entails a responsive relationship to God. According to Brunner, inasmuch as God reveals himself to us through Christ, Christ is the center from which we must consider the concept of the image of God. What is this revelation? It is the revelation of “the One who imparts Himself to me in freedom, since as Holy Love He claims me wholly for Himself” (Brunner 1952: 55). Furthermore, since Christ reveals himself to us in this way, his revelation of himself also provides us with a revelation of ourselves: “He is the One who wills to have from me a free response to His love, a response which gives back love for love” (Brunner 1952: 55). So as Christ reveals himself to me, I learn something about myself: I am designed to freely be in a loving relationship with him. According to Brunner, once we realize the intimate connection between knowing God through Christ and being known by him, we cannot but conclude that to be in relation to him is part of our nature. Thus, in creating human beings in his image, God has made us to reflect his freely given love by freely responding to him in love. In being made in God’s image or likeness, of course, our freedom is not just like God’s; his freedom is unlimited, whereas ours is limited. Indeed, given that we are made to respond in love to God, we have a responsibility to do so, and as Brunner points out, responsibility involves a restricted freedom: it entails that we must be free to choose to respond (since we do not hold creatures who lack freedom responsible), but it also entails that we must respond appropriately in order to fulfill our responsibility. I say respond “appropriately,” since Brunner takes it that every human being makes some response to the relational nature of her existence, even if she does not recognize it as such—the unbeliever responds to the call by turning away from God. So the responsibility to freely respond in love to God is part of the nature and



Divine Events

227

existence of every human being. Brunner says of this responsibility that it “is part of the unchangeable structure of man’s being … he has been made to respond—to God” (Brunner 1952: 57). When this Christ-centered conception of the image of God is characterized in terms of personhood, it entails that the personhood of both God and human beings is fundamentally relational and loving, and of course this is supported by the apostle John’s assertion that “God is love” (1 John 4.8; emphasis added). In view of this conception of God’s personhood, we have much more in common with God than the classical conception allows. The classical conception claims that God is simple, impassible, and strongly immutable; the relational conception of the image of God entails that he changes in his relations with created persons and is, to some extent, affected by those relations; and, of course, all of this entails that he is not simple. So in the relational conception, concerns about excessive anthropomorphism are often exaggerated and misplaced. We unacceptably anthropomorphize God only when we project our finite attributes onto him. It is the infinity of his attributes which distinguishes God from human beings, not the generic attributes themselves. So, plausibly, God and human beings share not only love, but also such attributes as creativity, intentionality, imagination, rationality, knowledge, power, goodness, and so on, but where God’s capacities in realizing these attributes are unlimited, ours are limited. Once we are freed from worries about anthropomorphism, then we are able to take at face value the depiction of God in the Bible, where such mental states as love, anger, pleasure, sorrow, regret, and jealousy are attributed to God. So we should consider what God’s mental life would need to be like in order to experience these mental states.17 If God’s mental life did not consist of discrete, ordered events, then all of his experiences would be accessible at every time. So, at any given time t, he might be experiencing love for a humble servant, sadness for a lost soul, anger at someone who had led believers astray, and pleasure in a particular creative plan of his (and of course, an infinite amount of other experiences). What sense can be made of the claim that any person, whether they are infinite or not, can have all of these experiences simultaneously? One begins to feel that what we must be dealing with here is not a person, but merely an infinite knower. We can make sense of an infinite being knowing everything at every time, but I would reject that any sense can be made of any being experiencing all of their intentional, emotional states at every time. And for those of us who believe in a personal God—the God of the Bible—it seems essential that we attribute such experiences to God. If this is correct, it implies that God experiences emotions in a way that is structurally similar to ours; and this entails that God’s mental life does not just consist of static, propositional attitude states, but necessarily involves

228

DEBATES IN THE METAPHYSICS OF TIME

the dynamic experience of events. Furthermore, as implied by the arguments of Leftow (1991), in which he posits the pleasure of anticipation experienced by God prior to creation (as an explanation of why God did not create sooner), these emotional experiences need not depend upon God’s already having created. We have every reason to believe that God’s inner mental life would be wonderfully rich even prior to interacting with his creation. Think of God’s experiences prior to creation by analogy with an author who savors the working out of his story. God’s being omniscient does not make this analogy incoherent, it just means that, unlike a human author, he does not need to work it out. This, however, does not entail that he does not enjoy working it out. And working it out requires thought processes that consist of ordered mental events. This analogy with an author suggests another way of interpreting the doctrine of co-eternal creation. If we agree with the advocate of co-eternal creation that God is essentially creative and has always been exercising his creative power, and if we think that God’s inner mental life is far richer than Swinburne’s example of an undifferentiated act of self-awareness would suggest, then there is scope to claim that, prior to the creation of the universe, God’s creative power was exercised in his inner mental life. Consider how much pleasure we, as persons, derive from the exercise of our creative imagination; and then consider how much more the pleasure for God must be in such an exercise. It is true, for human beings the creative process can also often be both excruciating and cathartic, and these aspects of the process would clearly not apply in God’s case; but it is plausible that the pleasurable aspect of the creative process would apply in God’s case. In this rendering of co-eternal creation, according to which it is the exercise of his creative power in his inner mental life that is co-eternal with God, there are no worries about ontological priority or infinite universes; but there are also no worries about why God waited to create, because he did not wait.

Conclusion In the preceding section I have offered some ways of thinking about God’s mental life which, if correct, entail that he is necessarily the subject of events both prior and subsequent to creation. It is God’s shared personhood with ourselves that allows us to conceive of God’s mental life in this way. If the essential temporality of events defended in the first section is also correct, then God is necessarily temporal.



Divine Events

229

Notes  

* Earlier versions of this chapter were presented in conferences at Queen’s University Belfast and Cambridge University, and I am grateful to the audiences at those events for their stimulating and helpful comments. I am particularly grateful to Brian Leftow for written comments on an early draft. I would also like to acknowledge the John Templeton Foundation, whose support made possible the research for this chapter.

  1 Here I will be treating the question of whether events are necessarily processes of change as equivalent to the question of whether events are necessarily extended. This treatment assumes that there could be no temporally extended unchanges. I argue for this assumption in Diekemper (unpublished manuscript).   2 Cleland (1991) also argues that events are processes of change, but her account, unlike Lombard’s, allows for the possibility of non-spatial events, since she does not identify material objects as the subjects of change; rather, she identifies determinable tropes, which change in respect of the determinate properties they exemplify, as the subjects of change.   3 On instantaneous events as theoretical entities, see Simons (2003: 377). In written comments, Leftow has cited, as grounds for endorsing instantaneous events, the causal work done by such events in a presentist conception of time. I grant that for presentism to be viable it probably requires instantaneous events with causal powers, but I do not think presentism is viable. See Diekemper (2005, 2013).   4 There are a lot of issues lurking in the background here, which I do not have the space to address. Bennett (1988) and Lowe (1998) argue that events should not be assimilated to facts, but Bennett thinks facts can still play a causal role, as does Mellor (1995). Although Kim (1976) rejects fact causation and does not assimilate events to facts, the result of his version of the property exemplification view is that events are nearly as fine-grained as facts. I assume, however, in view of how Leftow intends to employ events, that he would agree to the categorical distinction between events and facts, and that he would reject fact causation.   5 Thanks to Brian Leftow for raising this worry in written comments.   6 So if God is necessarily temporal, such worlds are not possible.   7 Le Poidevin (1993) makes the same point.   8 I argue in Diekemper (unpublished manuscript) that there are good, positive reasons to reduce time to actual events and their relations.   9 I thank Dean Zimmerman for this suggestion. 10 It is another question as to what sense is God temporal. I address this issue in Diekemper (unpublished manuscript). 11 Interestingly, however, Halvorson and Kragh (2011) report the observation of Misner that the finite/infinite time distinction “might lack intrinsic physical or theological significance.” According to Misner, even in spacetime models that begin with singularities, “the Universe is meaningfully infinitely old because infinitely many things have happened since the beginning.”

230

DEBATES IN THE METAPHYSICS OF TIME

12 Cf. Zimmerman (2002: 78n. 17). 13 Halvorson and Kragh (2011) make this point in the context of reconciling theism with a steady-state cosmology according to which the universe is infinitely old, but the point also has application in the present context. 14 See Kraay (2010). 15 See Diekemper (unpublished manuscript). 16 Henceforth I will be assuming Christian theism. 17 I make no claim that, for example, God’s jealousy is just like ours, since his jealousy is necessarily a righteous jealousy, and of course ours is inevitably not. So, too, for his sorrow and regret, since these will be tempered by his knowledge of his providential plan.

References Bennett, J. (1988), Events and Their Names. Indianapolis and Cambridge: Hackett Publishing. Brunner, E. (1952), The Christian Doctrine of Creation and Redemption, Dogmatics, 2, trans. O. Wyon. London: Lutterworth Press. Cleland, Carol E. (1991), “On the individuation of events,” Synthese, 86, 229–54. Craig, W. L. (2001), “Timelessness and Omnitemporality,” in G. Ganssle (ed.), God and Time: Four Views. Downers Grove: Inter-varsity Press, ch. 4. Diekemper, J. (2005), “Presentism and ontological symmetry,” Australasian Journal of Philosophy, 83, 2, 223–40. —(2014), “The existence of the past,” Synthese, 191(6), 1085–104. —(unpublished manuscript), Events: Temporal and Eternal. Ganssle, G. and Woodruff, D. (eds) (2002), God and Time: Essays on the Divine Nature. Oxford: Oxford University Press. Halvorson, H. and Kragh, H. (2011), “Cosmology and theology,” in Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Winter). http://plato. stanford.edu/archives/win2011/entries/cosmology-theology/ Kim, J. (1976), “Events as Property Exemplifications,” in M. Brand and D. Walton (eds), Action Theory. Dordrecht: Reidel, pp. 159–77. Kraay, K. (2010), “Theism, possible worlds, and the multiverse,” Philosophical Studies, 147, 355–68. Leftow, B. (1991), “Why didn’t God create the world sooner?,” Religious Studies, 27, 157–72. —(2002), “The Eternal Present,” in G. Ganssle and D. Woodruff (eds), God and Time: Essays on the Divine Nature. Oxford: Oxford University Press, pp. 21–48. Le Poidevin, R. (1993). “Relationism and Temporal Topology: Physics or Metaphysics?,” and “Postscript,” in R. Le Poidevin and M. MacBeath (eds), The Philosophy of Time. Oxford: Oxford University Press. pp. 149–67. Lombard, L. (1986), Events in Metaphysical Study. London: Routledge & Kegan Paul. Lowe, J. (1998), The Possibility of Metaphysics. Substance, Identity and Time. Oxford: Clarendon Press.



Divine Events

231

Mellor, H. (1995), The Facts of Causation. London: Routledge. Simons, P. (2003), “Events,” in M. Loux and D. Zimmerman (eds), The Oxford Handbook of Metaphysics. Oxford: Oxford University Press, pp. 357–85. Swinburne, R. (1994), The Christian God. Oxford: Oxford University Press. Zimmerman, D. (2002), “God Inside Time and Before Creation,” in G. Ganssle and D. Woodruff (eds), God and Time: Essays on the Divine Nature. Oxford: Oxford University Press, pp. 75–94.

11 Instants, Events, and God Brian Leftow

Diekemper offers an argument for divine temporality with three premises: (1) God is necessarily the subject of (mental) events before and after creation.1 (2) Events are necessarily temporal.2 (3) Only a temporal being can be subject to temporal events. He rests (2) on the premise that (4) Events are necessarily temporally extended changes. I now contend that this argument fails. I do not advance a positive case for divine atemporality: space constraints preclude this, and I do so elsewhere.3 I begin with Diekemper’s treatment of (1).

God Let us take (1) as a conjunction of: 1a. God is necessarily the subject of mental events before creation. 1b. God is necessarily the subject of mental events after creation. In favor of (1a), Diekemper suggests that before creation, God is like an author savoring the working out of his story, a story he says God does not need to work out.4 But if He does not need to work it out, He possibly has the story without working it out.5 If so, He is not necessarily the subject of story-working-out events prior to creation. And of course God doesn’t need

234

DEBATES IN THE METAPHYSICS OF TIME

to work it out. Even we can frame a (simple) story for ourselves in a single flash. No matter what time you might suppose a perfect mind would need to work a story out, a truly perfect mind would need less than that. So no extended time, however small, is a time that would be needed: a perfect mind could do this instantaneously. So this way to back (1a) fails. Diekemper also mentions my old argument that a temporal God might spend time pre-creation enjoying the pleasure of anticipating creating,6 but this won’t help him here. For there is no necessity that God act so; certainly I argued none. As Diekemper mentions no other sort of pre-creation event, he simply hasn’t backed (1a). Further, (1a) is quite plausibly false. (1a) entails that there necessarily is a “before” creation—that it is not so much as possible that past time and the universe’s past are co-extensive. But Big Bang scenarios in which time and the universe are finite and co-extensive pastward seem eminently possible; so do views with an infinite past in which a universe has always existed. If (1a) is false, so is (1). So we could just stop here. But let us consider Diekemper’s case for (1b), which hinges on God’s emotions. As he sees it, God has many different ones, and if He did not shift from one to another, … at any given time (God) might be experiencing love for a humble servant, sadness for a lost soul, anger at someone (evil, etc.) … What sense can be made of the claim that any person … infinite or not, can have all of these experiences simultaneously? One begins to feel that what we [have] here is not a person, but merely an infinite knower. We can make sense of an infinite being knowing everything at every time, but I … reject that any sense can be made of any being experiencing all of their … emotional states at every time.7 But if the personhood seems to Diekemper to go in one view, the infinity seems to go in Diekemper’s. If we can’t have all these emotions at once, that is a function of our cognitive and affective limits. But anyone who thinks one cannot feel love and hate for the same person at once has had a remarkably tranquil life: for most people, “mixed emotions” are common. Now suppose that I see before me my true love and the vile Dastardly Dick. I always react emotionally to what I see. It makes perfect sense that I would simultaneously feel love for the one and hate for the other. I think I can, though I will focus more on one, and the other emotion will be in the background; I might for instance really focus on hating DD, then shift suddenly to concentrating on my love, so that the hate for DD is a sort of afterglow in which the new emotion is framed. If I can’t do this, it is simply because I can’t have both my love and DD in the forefront of my mind simultaneously: if I so focus on my love that I feel love for her, I don’t focus on DD. But that is just a matter of



Instants, Events, and God

235

how human brains happen to work, as far as I can see. It is equally a matter of our brains that I cannot think of all propositions at once, yet Diekemper has no qualms about God’s infinity letting him do this, though in this case, too, we can’t imagine what it would be like to do it. If inability to imagine is insufficient to convince us that God is cognitively limited, I do not see why it should convince us that a God with the infinity Diekemper insists on would have to suffer emotional limitation. I do not see why we should think that all possible minds can foreground only some humans at once in the way that evokes emotion. Further, this limitation might entail that God cannot respond to events perfectly. If God’s foreground attention is of limited capacity, it may be possible that so many events happen so quickly that He fails to react to some with appropriate emotion. Some events, after all, deserve not just sorrow, but sorrow for a while. If your spouse of 30 happy years dies, you should not get over it in five minutes. Suppose that event e at t deserves a minute’s sorrow and event e* at t plus one second lasts for 30 seconds and deserves happiness: if God shifts to e* in time He short-changes e, and if He does not, He does not give e* its due. Further, some events deserve emotion-based intentional responses. But if God can feel only so many emotions at once or over a period, He might not be able to form enough responsive intentions in time: if I cannot at once love my love and hate DD, neither can I at once form a love-based intention to help my love and a hate-based one to harm DD. I thus suspect that even many temporalists would be uncomfortable with Diekemper’s argument.8 In any case, appeal to successive emotional states doesn’t give us reason to accept (1b)’s claim that God necessarily undergoes successive mental events after creation. God could have made an emotionally simple world. He could have made one whose only creatures were immortal happy clams, each everlastingly just humming one note contentedly in its shell, all hums (and clams) qualitatively identical. God could only sensibly react to such a world with one kind of emotion: nothing emotionally relevant changes.9 We might gradually become bored. Surely that is a defect to which God is not prey.10 Diekemper’s case is for a succession of kinds of emotion. But in clam-world, God would not have different kinds of emotion, and so not have them successively. Diekemper might instead appeal to change in pleasure of anticipation of various particular events, but if all that ever happens is qualitatively identical clam-humming, and He has this now, it is hard to see why or how He could thrill with anticipation of the next round. We might also take as an argument for (1b) Diekemper’s claim that we are in the image of God, that this “image” involves us essentially in relation to God, and that “the relational conception of the image of God entails that He changes in his relations with created persons and is, to some extent, affected

236

DEBATES IN THE METAPHYSICS OF TIME

by those relations,”11, adding that the effect is in affect or cognitive state. But if this is an inference from image to original,12 it is too quick. Two things are relevant in thinking about images: the nature of the original and that of the reflecting medium. Some media do not reflect accurately, and produce systematically distorted images of the original. Why so sure that change and affectedness in the image comes from the original, not the medium? We might be funhouse mirrors, reflecting God in systematically warped ways. That seems true in the moral sphere. God’s personhood is fundamentally loving.13 Contra Diekemper, it is hard to believe this of ours as now constituted.14 And again, it is not necessary that God create persons, and so even if this were a good argument that God now changes mentally, it would not show that He necessarily changes mentally after creating. Thus Diekemper fails to back (1b). This does not, of course, entail that (1b) is false. If there necessarily is time and time is necessarily the sort that “passes,” then arguably if God is necessarily omniscient, his cognitive state necessarily changes before and after creation, as He notes the ever-new time it is. But the claims about time would require some metaphysical heavy lifting, and all it would get us is a modalized version of the old argument from omniscience and immutability, to which I reply elsewhere.15 (1) is not just plausibly false (via (1a)) but question-begging. It assumes that there necessarily is time before creation, though atemporalists will rightly insist that in their view time is a divine creation, and so as there is no such thing as time before time, there is no such thing as time before creation. Again, if an atemporal God need not have created, then if time is a divine creation, there need not have been any time. Again, atemporalists are happy to hold that in some possible world an atemporal God atemporally creates only atemporal entities, for example a space without time; thus even if God necessarily creates, an atemporal God need not have created time. The best solution, I think, would be to substitute for (1) a simple claim that, 1c. God is necessarily the subject of mental events. But if (2) is in the offing, I wonder whether (1c) is quite neutral on divine temporality. God is certainly necessarily the subject of token mental states. An atemporalist can say this without fear, as states need not be temporal: if there is a timeless Platonic number three, it is in the state of being prime. If we take (2) seriously, then the difference between state and event is precisely that between something neutral on temporality and something essentially involving temporality. If so, one can legitimately demand rather better support for (1c) than Diekemper provides for (1).16 I now turn to (2) and (4).



Instants, Events, and God

237

Intuition and events Diekemper thinks that (4) is the “intuitive” view of events.17 Not so: intuitively, some events are extended and some aren’t. A runner’s winning a race is an event. It is one we intuitively recognize: most of The Folk would call it an event. It occurs as soon as the right part of the winner’s body reaches the finish line. It does not take any time: as soon as the winner makes contact with the plane extending upward from the line, it is over; the winner has won. Winning occurs. It happens that someone wins. But it seems to us intuitively that winning takes no time. “How long did it take to win?” makes sense if it asks for the duration of a race, or of the crucial part of it when it was determined who would win, but we all think that this question makes no sense asked of the winning itself. Winning is what ends the long event which is the race. Any number of other things The Folk would call events are like this—the ceasing or beginning of many events and processes. The Folk would express their intuitions in calling these things events. Intuition has a more complex view of events, and that full complexity deserves philosophical respect. Now the folk understanding of winning, just explicated, involves a tacit idealization. The finish line is not a geometric straight line and does not have the sort of sharp point-thick boundary from which a plane would extend upward. It is thick, irregular, and vaguely bounded. So is the runner’s body. So there is no such event as the leading surface of the runner’s body contacting the plane of the finish line’s closer edge. But my argument concerned what folk intuition—the sort we all bring to philosophy—favors. That folk intuition favors the picture I sketched is confirmed by the fact that The Folk find it intuitive that one can assign a single correct time to a race—for example, that someone ran the 100 meters in 10.0 seconds. This supposes the Folk picture. In the second picture, all one can say is that the runner ran it in some time between (say) 9.99 and 10.01 seconds. In the second picture, the race is won when some part of the runner’s body coincides with some part of a thick zone extending upward from the thick finish line inscribed on the dirt. But in the second picture, for any time at which some part of the runner’s body so coincides, there is a prior time at which some smaller body-part or something which is not definitely part or not part of the body did so. So it seems that any assignment of a time to the run will have to be arbitrary. Tell this to The Folk and they will be disturbed. The second picture may not really get rid of instantaneous events anyway. For perceivers judge winning: in the Olympics, say, the winner wins when a judge sees or a camera “sees” the smallest visible part of the runner’s body coincide with what the judge sees or the camera reveals to be the smallest visible part of the thick zone extending upward from the finish line.18 This

238

DEBATES IN THE METAPHYSICS OF TIME

coincidence takes place instantaneously. First there is not a sufficient intersection to be visible, then all at once, there is: this follows from the fact that the smallest visible parts are what is relevant, because the race is won as soon as the intersection is visible.

Instants Intuition, then, favors instantaneous events, and even the picture that replaces what is initially more intuitive generates them. So, are there instantaneous events? A prior question is whether there are instants. I give a simple argument for them, which Diekemper does not address.19 Consider an object moving continuously, and a volume of space along its path shaped so that the object exactly fills it.20 The object passes through the volume. It does not stop there. There is a fact about how long it was there. If it was moving continuously and the volume exactly fits it, it can only have been there for an instant. So in whatever way there are times, there are zerolength times: instants. What Diekemper does say about instants is puzzling: … there seems little reason to suppose that instants are temporal parts … it seems wholly implausible that some contiguous set of instants of zero duration could somehow add up to an interval … of non-zero duration. If, however, instants are merely the bounding points of temporal intervals, then there is no pressure to admit instantaneous events into our ontology (other than, perhaps, as theoretical entities).21 Both extended parts and bounding points could be real. To call an entity theoretical is to say what sort of reason we have to believe in it. It says nothing at all about whether it is real. It may be that we do not literally observe sub-atomic particles through an electron microscope, instead seeing only patterns of their effects. If that is correct, electrons and quarks are theoretical rather than observational entities, yet we all believe they exist. So to admit a theoretical entity could well be to admit something fundamentally real, as quarks may be. At page 218, Diekemper seems to allow that unless God is necessarily temporal, my location at t could exist apart from the journey it terminates, though (he claims) it wouldn’t be an instant then. Whether God has this character is a matter extrinsic to a bounding point. So it sounds like Diekemper thinks that nothing intrinsic to bounding points rules it out that they are real, independent entities in good standing, which have temporality as an extrinsic property. Yet given the reduction Diekemper endorses,22 if there are no instantaneous events or event-parts, there are no instants: denial



Instants, Events, and God

239

of instants is denial of instantaneous events. So Diekemper should not think that bounding points are really there, even as entities somehow dependent on the periods they bound. If denial of instants is to help Diekemper argue against instantaneous events and for divine temporality, he needs an argument against instants independent of the reduction noted above or of divine temporality. What he gives us is an allusion to a Zeno paradox, that if we put any number of instants (per impossibile)23 “next to” one another, the result will still have zero extent. Even continuum-many instants could not in this way yield an extension. But these days no one worth his salt thinks instants “add up to” periods this way. If there are instants, periods are instants with distance-relations between them. The relations, not their relata, account for periods’ extension: that is why (in the paradox) without putting distance-relations between the points, we don’t get an extension. So it is no argument against instants that they cannot account for extension apart from the relations. Further, even if the argument was good, it would not rule out the existence of instants. It would merely rule it out that they “add up to” periods. They might still be there, bounding the items that “add up.” They might even count as parts of the periods: they would after all stand in the mereological relation of overlap to them. Even if instants are “merely bounding points,” if they are real bounding points, that is enough (contra Deikemper) to create pressure to admit instantaneous events in most theories of events. If events are havings of properties at times, or subjects’ having properties at times, or fillers of 4D regions, then if there are instants, instants are as good a value of “t” in such accounts as any other. Let me put the point more strongly: any theory that involves a time in the constitution of an event generates instantaneous events if there are instants to plug in for “t.” It would complicate the theories to restrict the permissible values of “t” to extended periods, and such complication of ideology would at least partly offset whatever parsimony one might gain by rejecting instants. But one would not really gain parsimony this way. One way to measure parsimony is at the level of fundamental kinds—that is the way nominalism is more parsimonious than Platonism about numbers, though if there are numbers at all, there are the same number of numbers whether nominalism or Platonism is the true story about their nature. If we believe in times at all, we could achieve equal parsimony of fundamental kinds by treating intervals as constructions out of real instants rather than vice versa: either way, times would come in one real fundamental kind. Further, if instants are fundamentally real, there are continuum-many of them, and if intervals are fundamentally real, then equally there are continuum-many of them. So we have the same number of individuals either way. Actually, then, there is no parsimony argument in favor of rejecting instants. What should tip the scale—in favor of them—are two things, the

240

DEBATES IN THE METAPHYSICS OF TIME

simpler ideology of event-theories that do not include arbitrary restrictions against instants, and that admitting instants leaves us less conflict with intuition: rejecting instants requires us to deny events we intuitively think are there, while the other move allows us genuine fusions of point-events, which are what intervals are, and so preserves every event we believe in. I now turn to some reasons to believe in instantaneous events.

Instantaneous events Even if events were necessarily changes, as Diekemper thinks, they would not necessarily be temporally extended. Diekemper has it that “change cannot occur at an instant.”24 True, a process of change cannot. But the change toward which it is a process, its terminus ad quem, the very last bit of the process by which it arrives where it was going, can and does. It is in and only in the last instant that the change-process gets to where it was going. This last bit is as good a change as any other bit of the process, since in it something comes to be different than it was at any point earlier in the process. But if a change, then an event.25 Diekemper suggests that its character as an event may be extrinsic—may depend on there having been a process leading up to it, so that in a world one instant thick, the same thing could be there but not be an event.26 That does not matter for whether it is actually an event. It is also implausible: could a metaphysical category really be settled extrinsically? Of course, if there are no instants, there is no instantaneous terminus ad quem event: but I await a better argument for that. Again, the most natural account of occupation pairs parts with occupied regions 1:1. So if there are instants, as I have argued, the most natural account of occupation would have it that events have instantaneous temporal parts paired 1:1 with instants they occupy. Again, consider a possible world containing objects that can pass through one another. Consider two such objects, concentric and of precisely the same shape. The inner one is changeless. The larger, outer one begins to shrink. It shrinks continuously and evenly, retaining its shape at all times. After a bit, it precisely coincides with the smaller, then as it continues to shrink it is within the smaller. The larger object’s precisely coinciding with the smaller was an event. It happened. But as the larger was shrinking continuously, this event can only have occurred for an instant.27 So it seems that instantaneous events are at least possible, which is all I need to defeat (4). Here is a structurally similar actual-world case.28 Consider a car decelerating continuously from 10 mph to rest. Ask whether it ever travels at precisely 5 mph. “No” would be hard to understand: how get from 10 to 0 while never passing through 5? Any reason to say “no” here would equally



Instants, Events, and God

241

be reason to say “no” throughout the deceleration, with the result that the car would have no particular definite velocity at any time during the process. Further, if “no” is the right answer, then most moving items most of the time have no definite velocity, since things are rarely if ever neither accelerating nor decelerating. Instead, most objects mostly have only ranges of velocities over periods of time: between t and t*, say, an object has a velocity between n and n*, and that is the only fact about its velocity. Or perhaps objects just have average velocities though never definite velocities of which they are averages, and that is all that is true. Either would be a hard doctrine indeed. Yet suppose that the car does at some time travel at precisely 5 mph. If its deceleration is continuous, it can only do so for an instant. So, for an instant, it travels at 5 mph. Traveling at 5 mph is a way of moving. Movings are events. So here we have an instantaneous event. Further, it is an event in which a car has a genuine instantaneous velocity. Again, events have as temporal parts only smaller events. If for part of the period of a supposed event nothing is going on, the event is not going on during that supposed part, and no event has any part at a time at which it does not occur. (Football games have half-time breaks, but these are not part of the game; the game is not going on during the break.) If only events are temporal parts of events, then if events have instantaneous temporal parts, there are instantaneous events. Intuitively, events do have such parts. A race is an event. The second-place finisher’s reaching the finish line takes place during the race. It is part of the race. The event in which some part of that runner’s body reaches the finish line (or reaches a position at which it is judged to do so, in my alternate account) is instantaneous. So the race has an instantaneous temporal part, and so there are instantaneous events. My arguments so far do not depend on any particular view of time. But theories of time in which time “passes” generate their own reasons to believe in instantaneous events. Let us first consider presentism with an instant-thick present. Consider an extended process. In this sort of presentism, only the instantaneous bit of the process going on now is real. It is all there is to the process. The rest has no place in reality at all. That surely seems enough to distinguish it from the rest of the process as a part in its own right.29 But processes are events, and events have only events as temporal parts. Instantaneous events do causal work in presentism. For in presentism, only what is going on in the present instant is available to be a causal relatum or transmit causation. So if events are causal relata, instantaneous events are causal relata. Extended past events can act only in a derivative sense, through having had influence on what is going on now. Consider an instantaneous slice of my walking, what I’m doing at an instant during it. In presentism, there is only this to cause later phases of the walking. On suitably broad principles of recombination, an indiscernible universe could have begun at that instant,

242

DEBATES IN THE METAPHYSICS OF TIME

with me in that state of motion, and continued just as the actual universe did. In this case it would not have been possible to explain my succeeding motion by my motion over a preceding period. Only my motion at that instant could do the explaining. But if this was an indiscernible universe, actually my motion at that instant is also a cause, though not one which competes with the prior period’s motion for causal honors. The prior motion acts through it, if there was any. But causes are either events or substances—facts or states of affairs are relata of explanation but not causation—and my motion is not a substance. If what is going on at that instant is a cause, that is because it is a moving. Plausibly its being fit to be a cause does not depend on there actually being time following it. It is the other way around: because it is intrinsically fit to cause motion, it does cause motion if there is time following it. But it is fit intrinsically to cause motion only because it is, intrinsically, a moving— whether or not there is time following it. If all this is plausible, then plausibly it would be a moving even in a world in which there was not only nothing preceding it but also nothing following it. Similar things to these are true in a “growing block” theory of time that allows a present instant. In such a view, past and present equally exist, but as causal processes are continuous, the past still has causal efficacy only through the present, and the causal efficacy in another world begins only at that present. These are mostly new arguments. Diekemper discusses some older ones.

Temporal Parts In the paper Diekemper discusses, my first argument for instantaneous events is an actual-world case similar to the coinciding-object case above30: let one object be me and the other a precisely me-shaped volume of space into which I fit at one instant during a continuous walk. Then I come to coincide with the volume, but again, can only do so instantaneously. So it seems that instantaneous events actually occur. Here the premises are: 1. My walking has a part in which I just fill a particular region of space

for an instant. 2. Events have only events as temporal parts.

Diekemper denies neither premise, nor does he deny validity. Instead he claims that my reason to think that what fills that region is a moving is that it has an instantaneous velocity. That is wrong. My reason is just that it is part of a continuous walking. Diekemper misreads the role of instantaneous



Instants, Events, and God

243

velocity in my discussion. Having argued from (5) and (6), I brought up a complication: on an at-at theory of motion, the distinction between being in motion and being at rest derives from inclusion in a larger whole which is a motion; intrinsically, an object at a place is neither moving nor at rest.31 My response was that this does not deny that there are instantaneous movings. It entails that being a moving is an extrinsic property of an object’s being at a place at a time. But extrinsic properties can be real. It would take an argument to show that this particular one is not. Instantaneous velocity is something Wesley Salmon brought into his exposition of the at-at theory; my only use for the concept was to point out that Salmon’s use of it supports the claim that in the at-at theory, there are instantaneous movings. Thus Diekemper fails to address my argument. Nor does his discussion achieve much in its own right. He reminds us that we measure instantaneous velocity by measuring velocity over periods.32 But this tells us nothing about whether instantaneous velocity is a real property of instantaneous events. We often learn about real properties at scales beyond our perceptual discrimination by learning about related properties: we cannot see atoms, but we can see electron microscope images. And if a property is extrinsic, to learn whether an item has it, we have to look not just at the item, but at the thing(s) to which the item bears (or not) the relevant relations—in this case, surrounding periods. When Diekemper writes that “the definition treats the instant t not as containing an event, but as the bounding point of intervals over which velocity is measured,”33 he is about half-right. Yes, it measures what is at the instant by treating its contents in terms of a limit. But that does not imply that t’s contents are somehow a non-event. The definition is neutral about this. It defines instantaneous velocity in terms of what we can measure, to let us compute it. Only a verificationist would infer from a definition designed for computation that to say that the car has an instantaneous velocity of 5 is simply to say that there are certain measurable properties of intervals, such that their limit is 5. Diekemper writes that the definition of instantaneous velocity “does not support the inference from an instantaneous velocity to an instantaneous moving.”34 That is fine; I never claimed that it did.

Success My paper’s second argument appeals to success-events like winning. Diekemper claims that parsimony rules against these.35 But my parsimony argument above for instants repeats at the level of instant-fillers. Beyond that, consider two possible worlds: in one, my journey toward a midpoint is open toward that midpoint, and t. I never actually arrive. I wink out of existence

244

DEBATES IN THE METAPHYSICS OF TIME

by t. In the other, I arrive. It seems natural to say that in the second world, something happens which did not happen in the first. The difference between them is my getting there, and what is my getting there if not an event? It can be a relatum of a causal relation: you can see me arrive, though you cannot discriminate the case in which you do from the other-world case in which you don’t, and as argued above, my arriving can cause my further motion. It can be a causal relatum even on the least favorable assumptions, 4D eternalism plus an at-at theory of motion. Even supposing these, spacetime has timewise slices a point thick. (Why not? Diekemper has yet to give us an argument against real bounding points.) There could be an eternalist world beginning with me at the arrival point and continuing with me moving, and there will still be causal relations between my being at that point and what follows.

Coming to Be Diekemper next addresses my discussion of coming to exist. He says that “clearly something is changing in the coming to be of a new individual.” New individuals might include discarnate souls or angels. All are immaterial; there can be nothing changing for them. If they come to be, they simply appear: first they are not here, and then, all at once, they are. A process of change goes through mid-stages between origin and completion. There are no mid-stages between non-being and being. There are no intermediate states to be in. This is part of why Aristotelians distinguish substantial from qualitative change. The change from these things’ not existing to their existing cannot be a process. Nor is there any process on which it supervenes or with which it is correlated. Again, consider classes. If a material thing A comes to exist, so does {A}. But the process that moves toward A’s existence is not gradually forming {A} as it gradually forms A, and there is no correlated process of gradually forming {A}. {A} simply appears when A does. Something similar is true if there are tropes. Again, as I have noted,36 if there can be instants, there can be a first instant of time. Anything, even a material thing, might begin to exist then without a prior process of change, and of course without a process during that instant. Even if a material thing’s matter gradually changes till a new individual comes to be, the new individual’s finally coming to be cannot be a process of change in anything. It has to be instantaneous. Again, first the individual is not there, then all at once it is, though that “all at once” had a gradual process of material change leading up to it. Diekemper asserts that talk of instantaneous appearances is just talk—that such events do no “ontological work.”37 On the contrary, first, note that there is a real difference between closed and open



Instants, Events, and God

245

intervals. The same interval which is open in one world can be closed by something else in another. Since we can have the interval without the closing thing, the closing thing has to be real and really distinct from the interval: it is separable from it. More importantly, in the case of a process of coming to be, unless this last distinct event happens, we do not get a new existent. Let us do the two-world trick again. In one world, the process of change in x’s matter never terminates. The world lasts only as long as the process leading toward x. We get no x. In the other, it lasts longer and we do get x. Again, we have a new causal relatum. We can see x begin to exist in the one world, for we can begin to see x. We cannot in the other. By virtue of this extra, instantaneous event we have x in one world and not the other. That is ontological work with a vengeance. Moving to Diekemper’s atheistic treatment of the Bang singularity, I do not think atheists balk at inability to give an ultimate explanation. On the contrary, they are happy to do without one to precisely the extent that they are happy to be atheists, as the desire to have one would push them toward a cosmological argument for theism. Atheists who made the move Diekemper suggests would not save themselves from theism. Even if there is always an earlier event to explain a later, none of them, nor all together, answer the question of why there are ever any events at all. That is a legitimate question if it is an ultimate explanation you want. You don’t have one if you cannot answer it, and the quest for an answer takes us beyond the universe. As to models without a real t0, in a way, I don’t need to worry here. It is surely possible that there be a physically real singularity. It is not a necessary truth that physical universes are so well behaved as never to yield infinite values in equations describing them. A possible universe with a first instantaneous event does as well for my purposes as an actual one, since Diekemper claims that (4) provides us with a necessary truth about the nature of events. Now to the theistic treatment of the Bang. Diekemper suggests that the change involved in the singularity’s appearance is in God. I do not see how this would help. Given his approach to physical change, Diekemper needs to show that the appearance of a singularity is, supervenes upon or terminates a process of change in an underlying material subject. God is not material and does not underlie the appearance of a singularity. Change in God might accompany this appearance, but he needs a claim about the appearance of the singularity, not about a change in God accompanying it. Further, it is not clear that even an accompanying relational change in God would be a process. If at first creation is not there and then all at once it is, the relational change in him by which He comes to count as its creator happens all at once—instantaneously. If there is no first instant at which creation exists, the picture is this: at every time, either it hasn’t happened yet or it has already happened; again, there is no process here. This problem is independent of

246

DEBATES IN THE METAPHYSICS OF TIME

any issue about whether God is temporal: even if He is, no change in him can play the role needed. There is also independence on the other side. One need not think God atemporal to offer a first instant of time as an example, for one needn’t think God atemporal to posit a first instant of time. A deity in hypertime could create a time with a first instant. Diekemper neither explains nor supports the claim that time is necessarily unified; if he means it to rule hypertime out, he needs an argument.

Changeless instants Finally, we come to changeless instants. Diekemper wants to invoke counterpart theory for times to avoid a move I make. That comes at a price. What counts as a counterpart to what depends on which aspects of similarity are most heavily weighted. If what happens during t* is otherwise like what happens during t, things are straightforward. Both overall intrinsic similarity and t*’s relations to bits of actual history make t* the counterpart of t. But overall similarity and relations to actual history can come apart: suppose that in an otherwise close world, the events of t (including the Bangor trip) are just as they actually were intrinsically, but are located a year earlier, while events at t* are wildly unlike those at t. We can continue to hold that t* is t’s counterpart and the Bangor trip was a year earlier, because we can weight similarity as context dictates, and say that in this context, similarity with respect to distance from actual events before t weighs most heavily. However, things might have wildly differed from actuality all the way back through history. Let us consider what to say about times in such worlds. One option is to say that if the world was too dissimilar to ours, t has no counterpart—that is, things did not happen at the times actual events did. But that is very counter-intuitive. Intuitively, even if everything had been very different and dinosaurs still ruled the earth, it would still now be the time we call 2014. Another option is to say that all that really matters for preserving a counterpart relation to t is distance from some actual event: but this will not deal with worlds with no events in common with ours. We might suggest that all that matters is distance from a beginning event: but then no time in our world is a counterpart of any time in a world in which time/events never began, and that is again counter-intuitive. Intuitively, it would still be 2014 years after the date traditionally assigned to the birth of Christ even if there had been an infinite number of years before that. That is, intuitively, this very year could have occurred in a world with an infinite past and no events at any time which actually happened. It is not evident how the counterpart move can



Instants, Events, and God

247

handle this, and I think it would be a fairly heavy price to have to deny that this is possible.

Atemporal events In sum, I do not see anything in Diekemper’s discussion to shake one’s confidence in my earlier arguments, and I have given new arguments besides. So (4) appears false. So Diekemper gives us no reason to accept (2). And (2) is not obvious. As I argue elsewhere, it is possible that there be a second temporal series. If it is, events are not necessarily located in our time. Further, the possibility of a second series only an instant thick cannot be ruled out, on general combinatorial grounds: if things can happen instantaneously, they can happen instantaneously and not be part of an extended temporal series. Diekemper’s denial of this is just table-pounding: he simply insists that if nothing changes, nothing counts as happening, even if the contents of what he insists is not then a time are exactly what they are actually.38 Most theories of events disagree. And surely whether something is an event is not determined extrinsically. It is a matter of ontological category. Categories seem to be determined intrinsically. It is just what I am intrinsically that makes me a substance. It is just what yellowness itself is intrinsically that makes yellowness a property. Why would it not be what my buttering toast is, just itself, intrinsically, that makes it an event? Diekemper might insist that there is a necessary connection between temporal bounding points and temporal intervals bounded, but even if there is, it doesn’t follow that there is a necessary connection between the contents of the first and the contents of the second (unless Diekemper can force his reduction of times to events on us), and in any case, this claimed necessary connection is (so far at least) brute, unexplained, mysterious, and so deserving of suspicion. I submit, then, that there could be a temporal series without temporal parts. So nothing about an atemporal event’s part-structure or lack of location in our time is reason to call one impossible. And if events need not have temporal extension, it is hard to see why they would need temporal location. Events in an instant-thick time series would differ from the contents of what tradition calls an atemporal present only modally: they could have had a successor or a predecessor, but an atemporal being’s life could not have. I say that there are two kinds of events, those that can have successors or predecessors, and those that cannot. To defeat this claim, someone would have to show without begging the question that being an event a priori entails being able to have a successor or a predecessor. I do not see how. Events of the second sort do not pass away, for if they did, they would be succeeded

248

DEBATES IN THE METAPHYSICS OF TIME

by their not occurring. So they just occur—period. Events of the first sort are temporal. Events of the second sort are not. You may want to protest: how could something happen but not be temporal? I do not think that the concept of happening is thick enough to yield anything relevant by analysis. Here we are in the realm of pictures. So here is mine. Take it as given that lives are the sort of thing that happen. One main source of medieval atemporalism, Boethius, introduces his concept of eternality as a way of being alive.39 One sort of life, ours, is essentially realized a bit at a time. It is a life dribbled out. This is biological life, which consists of various sorts of processes going on. Theists think there is non-biological life, for they accept that there is an immaterial intentional agent, and anything that intentionally brings about its own actions has to count as alive. Boethius suggests that one non-biological sort of life is not spread out, but lived all at once—intensely, rather thinned by stretching out. Someone alive this way lives all at once everything he ever lives. This would be the briefest possible life if it passed away, but Boethius adds that none of God’s life ever passes away. The events of God’s life can have no successors. They just occur—period. Because of this, God is outside the order of time, the order of what passes. Really, now, what is so impossible about that?

Notes   1 Diekemper (2014: 227–8).   2 Ibid., p. 215–16.   3 Leftow (1991a).   4 Diekemper (2014: 228).   5 I am reading “does not need to” as implying “does not necessarily.” Diekemper could reply, I suppose, that he means “need” purely in a psychological sense, as referring to what does or doesn’t drive God to act so, and say that even if He need not, He necessarily wants to, and wants this more than He wants anything incompatible. Well, he could say it. Arguing it would be a tall order.   6 Leftow (1991b).   7 Diekemper (2014: 227).   8 And that is even apart from the list of mental states Diekemper is willing to take as literally present in God, which is worrisome. Jealousy (ibid., p. 227) isn’t a good state. It doesn’t matter who you’re jealous of or why. It is plausibly a vice. God has no vices. The case isn’t quite so clear for anger (ibid., p. 227). It is often useful (e.g. when fighting for one’s life), and we think there are things we can justifiably be angry about (e.g., someone’s beating up our child). But whenever we respond with anger, it would be



Instants, Events, and God

249

morally better (and often more effective as well) to respond without it. God’s responses can’t be morally sub-optimal.   9 If Diekemper were to insist that God could focus on only a given number of clams at once and so would have to change the focus of his attention, I would stipulate that God creates no more than that number. Were he to point to the ever-varied whirl of the clams’ atoms, I would stipulate that this is an Aristotelian world, not an atomist one. 10 Were Diekemper to carry his anthropomorphism so far as to say that this world would bore God, I would reply that God can create short-duration worlds, and this world lasts less time than it would take for it to bore him. 11 Ibid., p. 227. 12 Diekemper could just mean to raise the sort of argument I bring up in the text’s next paragraph. But if that is his intent, he does not make it clear, nor discuss the varied responses this argument has gotten since Kretzmann brought it into the contemporary discussion. See Kretzmann (1966). 13 Ibid., p. 227. 14 Ibid., p. 227. 15 Leftow (1991a: 315–48). 16 I have kept to the main thread of Diekemper’s argument in the text, but let me note some other dissatisfactions with Diekemper on God. Diekemper rejects the claim that God has eternally created. So do I, but some of what he says against it puzzles me. He says that an infinity of entities cannot be unique—but there is just one hierarchy Ø, {Ø}, {{Ø}}…, and that hierarchy has infinitely many members. He says that co-eternal creation would imply an infinity of universes, but surely there could be just ours and another one infinitely old. He says that if   co-eternal creation … is the exercise of his creative power in his inner mental life that is co-eternal with God, there are no worries about … why God waited to create, because he did not. Diekemper (2014: 228).

God did not wait to think creatively, true. But there still remains the question of why God waited to create a concrete actual universe. That He thinks creatively does nothing to address this. Again, it would not help Diekemper make Swinburne’s move to hold that time did not start till creation did (ibid., p. 225). That would instead preclude Swinburne’s move: no pre-creation temporality, no temporally extended pre-creation divine awareness. I do not see what is odd about time starting with creation (16): why would there have to be a “before” creation? And I note that multiverse defenders might well be happy with the claim that this is a rather than the best possible world; that would do fine for theodicy, for instance.

17 Ibid., p. 217. 18 Then just what time the race took is relative to the level of visual resolution at which the judge judges. And it does take some time to win, for it takes some time for that much of the runner’s body to first enter and then fully occupy that region. 19 Leftow (2002).

250

DEBATES IN THE METAPHYSICS OF TIME

20 My talk of volumes is neutral between substantival and relationalist construals. The latter would work fine. I also do not assume that the object or the volume have 0-thickness surfaces or absolutely determinate mereologies. Objects with vague or extended boundaries can coincide with regions with vague or extended boundaries. This requires that every definite part of the object coincide with an appropriate definite part of the region and that every thing not definitely part or not part of the object occupies something not definitely part or not part of the region. 21 Diekemper (2014: 217). 22 Ibid., p. 229n. 8. 23 As time-intervals are dense if there are instants, instants could not be contiguous. Nor need they be to be parts of intervals. 24 Ibid., p. 216. 25 This is not an entirely new consideration: winning is one such event, terminating a process of racing. 26 Ibid., p. 218. 27 This illustration might be indebted to Dean Zimmerman. 28 Leftow (1997). 29 One might think to avoid this by applying a non–1:1 account of occupation, in which events do not have parts paired 1:1 with sub-regions of the region they occupy, but instead just bear a whole-to-partial-region occupation relation to each sub-region (see Hudson 2008). But this would require saying that the whole event is in some way “there” to bear the relation, and does so, even though only part of it can fit into the region, but has no part to fit into the region and nothing outside the region is in any way real (given presentism). I cannot see how this could be. 30 Leftow (2002: 26). 31 There is also another sort of theory, in which being a motion is intrinsic to what is going on at an instant. I did not mention that one because it is obviously friendlier to my case. 32 Diekemper (2014: 216). 33 Ibid., p. 216. 34 Ibid., p. 217. 35 Ibid., p. 218. 36 Leftow (2002: 27). 37 Diekemper (2014: 219). 38 Ibid., p. 218n. 6. It should be clear that his footnote’s invocation of divine necessary temporality can do no argumentative work here, since that is precisely the conclusion he is trying to argue. 39 The Consolation of Philosophy V, prose 6.



Instants, Events, and God

251

References Diekemper, J. (2014) “Divine Events,” in L. N. Oaklander (ed.) Debates in the Metaphysics of Time. London: Bloomsbury. Hudson, H. (2008), “Omnipresence,” in T. Flint and M. Rea (eds), The Oxford Handbook of Philosophical Theology. Oxford: Oxford University Press. Kretzmann, N. (1966), “Omniscience and immutability,” Journal of Philosophy, 63, 409–21. Leftow, B. (1991a), Time and Eternity. Ithaca, NY: Cornell University Press. —(1991b), “Why didn’t God create the world sooner?,” Religious Studies, 27, 159–72. —(1997), “Eternity,” in P. Quinn and C. Taliaferro (eds), The Cambridge Companion to Philosophy of Religion. New York: Cambridge University Press, pp. 257–63. —(2002), “The Eternal Present,” in G. Ganssle and D. Woodruff (eds), God and Time. Oxford: Oxford University Press, p. 26.

12 Foreknowledge and Fatalism: Why Divine Timelessness Doesn’t Help Alan R. Rhoda

Introduction

T

he problem of divine foreknowledge and creaturely freedom or, more generally, the problem of divine knowledge of future contingents, has long been a matter of controversy. If someone—say, God—knows that some event—say, a sea battle—occurs tomorrow, can it be undetermined today whether that event occurs tomorrow, and if so, how? Conversely, if some possible future event is not now determined either to occur or not to occur—in other words, if it is a future contingent—then how can it be either known to occur or known not to occur in the future? It seems that, until it actually occurs, a future contingent lacks the definiteness required to be a proper object of knowledge. At any rate, the problem is especially pressing for theists, most of whom believe both that there are future contingents, especially human libertarian free choices,1 and that God has always known which future contingents are going to happen. Despite two millennia of active debate, there is still no consensus about whether the problem can be solved, and, if so, what a philosophically and theologically acceptable solution might look like.2 In this chapter I analyze the problem as a specific instance of the more general problem of fatalism, and I argue that, as with any (valid) argument for fatalism, there are only two possible solutions. One solution is to say that God’s foreknowledge—for purposes of argument I shall assume throughout

254

DEBATES IN THE METAPHYSICS OF TIME

that there is God—does not single out any possible future, any unique and complete sequence of post-present events, as the actual future. This is the “open future” solution. While currently championed by some theists,3 many believe this solution unacceptable, in large part because it categorically denies the traditional view that God has advance knowledge of everything that ever comes to pass.4 The other possible solution is to say that God’s foreknowledge is explanatorily dependent upon the actual occurrences of future contingent events.5 This is the “preventable future” solution, upon which I will be focusing most of my attention. I consider the bearing of this solution on both the doctrine of divine timelessness and matters of temporal ontology. Since Boethius,6 divine timelessness has often been thought essential to any acceptable solution to the foreknowledge/future contingency problem. I argue to the contrary. In the next section, I examine the more general problem of fatalism and show that fatalism (the denial of future contingency) follows if and only if there is a fixed or now-unpreventable “future specifier.” Since God’s foreknowledge, as traditionally understood, is a future specifier, traditional theistic anti-fatalists must hold that God’s foreknowledge (in so far as it concerns future contingents) is not fixed but, rather, is “explanatorily dependent” upon the actual occurrences of future contingent events. After elaborating on this, the preventable future response, and clarifying the key notion of explanatory dependence, I then consider its implications for temporal ontology. I argue that, given that (contingent) truth supervenes on being, the traditional conception of divine foreknowledge requires an “ontically settled” or linear block future according to which there (tenselessly) exists a unique and complete sequence of future events. Next, I examine the implications of preventable futurism for divine timelessness by engaging with Katherin Rogers’ recent (2008) Anselmian response to the foreknowledge/future contingency problem. Rogers’ proposal combines (1) divine timelessness, (2) an ontically settled future, and (3) the preventable future response. Pace Rogers, I argue that (1) and (2) are each incompatible with (3). Hence, divine timelessness doesn’t help the anti-fatalist. While it does not itself entail fatalism, it blocks preventable futurism, which is the anti-fatalist’s only hope for reconciling future contingency with a traditional conception of divine foreknowledge.

The challenge of fatalism To set up the problem of foreknowledge/future contingency with maximum generality, we must step back and consider fatalistic arguments in general. Doing so will give us a clear sense of what our basic theoretical options



Foreknowledge and Fatalism

255

are for rebutting so-called “theological fatalism,” the contention that divine foreknowledge is incompatible with future contingency. Roughly, fatalism is the doctrine that there is a precise way the future is going to go and that there is now nothing that can be done about it. In short, fatalism says that the future that is going to be is—to use Prior’s apt expression—“now-unpreventably” going to be.7 By a “future” I mean an abstract representation of a unique, complete, linear extension of the actual past. For a future to be “now-unpreventable” is for it to be causally necessary, or such that it obtains in all logically possible worlds8 having the same causal laws and the same causal/explanatory history9 as the actual world as of the present. Fatalism entails that there is only one causally possible future or, in other words, that the future is “causally settled.” Fatalism thus entails that there are no “future contingents,” no events that occur in some, but not all, causally possible futures. Of course, if humans or other creatures possess libertarian freedom, or if there is genuine causal indeterminism in nature (at, say, the quantum level), then there are future contingents, fatalism is false, and the future is not causally settled, but rather “causally open.”10 Only two substantive assumptions are needed to construct valid arguments for fatalism. In terms of my opening characterization, they are simply (1) that there is a precise way the future is going to go, and (2) that there is now nothing that can be done about it. I call these, or rather their precisifications, the “specified future” (SF) and “unpreventability” (NP) theses, respectively. Concerning SF, since the fatalist’s conclusion is that there is only one causally possible future, which future is therefore inevitably going to be the actual future, the premises of any valid fatalistic argument must posit something that singles out a unique possible future as the actual one. Let us call that something a “future specifier.” For example, alethic arguments for fatalism (or for what is often misleadingly called “logical fatalism”)11 begin by assuming or attempting to establish that there is a collection of truths about the future—a complete, true story of the future, if you will— that specifies how the future is going to go. The existence of such a story amounts to the future’s being “alethically settled.” Alethic arguments for fatalism then attempt to show that if the future is alethically settled then it must also be causally settled. Likewise, epistemic arguments for fatalism (or for what is often misleadingly called “theological fatalism”) posit a complete, known story of the future. Usually this story is held to exist in the mind of God. The existence of such a story amounts to the future’s being “epistemically settled.” Epistemic arguments for fatalism then attempt to show that if the future is epistemically settled then it must also be causally settled. But clearly the mere existence of a future specifier is not enough to warrant fatalism. A future specifier ensures that the specified future will

256

DEBATES IN THE METAPHYSICS OF TIME

happen, but fatalism makes the stronger claim that it must happen in the sense of being now-unpreventable. To see how the fatalist must try to bridge this gap, consider that some facts are unquestionably fixed or now-unpreventable such that we no longer (and perhaps never did) have any say about them. Plausible candidates include the laws of logic, mathematical truths, the laws of nature, the basic principles of moral law, and the actual past. What the fatalist proposes is that the fixed facts, whatever they are, collectively constitute a future specifier. If so, then there is not merely a specified future (SF) but an unpreventably (NP) specified future. From that, fatalism follows. To see that fatalism follows from SF and NP, let us sketch out the reasoning. Given SF, there is a future specifier, S, the existence of which entails a specific future, F, that is, (1) (S ⊃ F). Given NP, S is now-unpreventable, or such that it will obtain no matter which causally possible future eventuates. Using N(X) to stand for , we can write (2) N(S). Since the entailment in (1) is also unpreventable—if it is logically necessary that S ⊃ F, then there cannot be a future in which S obtains and F doesn’t—we can rewrite (1) using the N operator: (3) N(S ⊃ F). Finally, we can represent the fatalistic conclusion: (4) N(F). (4) says that all causally possible futures are F futures, or equivalently, that F is the only causally possible future. All that remains is to show that (4) follows from (2) and (3) in virtue of the following transfer of necessity principle: (5) [N(p ⊃ q) ∧ Np] ⊃ Nq. The validity of this principle can easily be established by comparison with the transfer of logical necessity, that is, [(p ⊃ q) ∧ p] ⊃ q. The latter is an axiom in every standard system of modal logic, and for good reason. If all possible worlds are ones in which p ⊃ q is true, and if all are ones in which p



Foreknowledge and Fatalism

257

is true, then there are no worlds in which q is false. Exactly parallel reasoning underwrites (5) by substituting “causally possible futures” for “possible worlds.” Hence, the inference from (2) and (3) to (4) via (5) is demonstrably valid. What we have here is a minimal valid recipe for fatalism: simply establish that there is a future specifier among the fixed facts. Different fatalistic arguments posit different future specifiers and use different strategies to establish their fixity, but, in so far as they are valid, they all follow this basic recipe. Since the inference from (2) and (3) to (4) is logically impeccable, anti-fatalists have but two options for rebutting any given instance of this fatalistic argument schema. The first is to deny SF, that is, to deny that any future specifier of the posited type exists. This was Aristotle’s response to the alethic argument for fatalism,12 in which a complete, true story of the future plays the role of the future specifier. To deny SF in this context is simply to deny that there is any such story. The future, in this view, is not alethically settled, but alethically open. Likewise, in response to the epistemic argument for fatalism based on God’s foreknowledge, one might deny SF either by denying God’s existence, denying or restrictively qualifying God’s omniscience, or by arguing, as some theists do, that the content of an omniscient God’s knowledge does not constitute a future specifier.13 The future, in any of these views, is not epistemically settled, but epistemically open. Analogous “open future” or SF-denying responses can be given to any valid argument for fatalism. The anti-fatalist’s second option is to deny NP. This was Ockham’s response to both alethic and epistemic arguments for fatalism.14 Ockham conceded to fatalism the existence of at least two future specifiers: (i) a complete, true story of the future, and (ii) God’s having knowledge of such a story. Contra fatalism, however, Ockham maintained that because there are future contingents, the truth of that story and God’s knowledge of it are still preventable in virtue of there being causally possible futures in which some things actually true about the future are not true and in which some things actually foreknown by God are not foreknown. Analogous “preventable future” or NP-denying responses can be given to any valid argument for fatalism. In this chapter I will not have much more to say about open future responses. While that is the type of response that I prefer,15 my primary goal here is to explore the tenability of a preventable future response to the epistemic argument for fatalism, so to that task I now turn.

258

DEBATES IN THE METAPHYSICS OF TIME

Preventable future specifiers and explanatory dependence Since the existence of an unpreventable future specifier entails fatalism, antifatalists must either deny that there are any future specifiers or maintain that those which exist are still preventable. But what is it for a future specifier to be preventable? More specifically, given that divine foreknowledge is a future specifier, what is it for such knowledge to be preventable? Let us begin by clearing away one possible misconception: unpreventability does not entail temporality. The fatalistic argument schema outlined above only requires that an unpreventable future specifier exist. Whether it is temporally situated or not is a further, and tangential, question. Thus, the epistemic argument for fatalism does not depend on God’s literally having foreknowledge (i.e., temporally prior knowledge of events), but rather on God’s having unpreventable knowledge of the future, that is, of what we temporally situated beings would regard as the future. While I will continue to speak of God’s “foreknowledge,” as is customary in the literature, this should be understood in the latter, knowledge-of-the-future sense, which is neutral concerning God’s relation to time. Consequently, the foreknowledge/future contingency problem cannot be solved simply by appealing to divine timelessness or to creaturely power of some sort—counterfactual, causal, or otherwise—over the past.16 One major virtue of approaching the epistemic argument for fatalism by first considering fatalistic arguments in general is that it allows us to sidestep complications like the notoriously vexed “hard fact”/“soft fact” distinction,17 which is relevant only to fatalistic arguments that rely on the fixity of the past. What matters for fatalism is not the temporal relation between future specifiers and future events, but the explanatory relation between them. The central issue is whether the posited future specifier is fixed independently of the actual occurrences of future contingent events or whether it is fixed (in part) by their occurrences. Thus, the preventable futurist must say that if tomorrow I make a libertarian free choice between, say, vanilla and chocolate ice cream, and choose vanilla,18 then God will have always (or eternally)19 known that I was going to choose vanilla, and he will have known that in virtue of my so choosing. And if I should choose chocolate instead, as by hypothesis I have the power to do, then God will have always known that I was going to choose chocolate, and he will have known that in virtue of my so choosing. Hence, my free choice to do this (rather than that) brings about God’s having always known that I will do this (rather than that) in the future. As Figure 12.1 shows, if it is a future contingent whether I choose chocolate or vanilla ice cream tomorrow, and if it is causally necessary that I do exactly one of those two things, then the set of causally possible futures



Foreknowledge and Fatalism

259

FIGURE 12.1 can be partitioned into “vanilla futures,” {fv1, fv2, …}, and “chocolate futures,” {fc1, fc2, …}. Since a future specifier entails the coming to pass of its corresponding future, if a future specifier exists then it must specify either a vanilla future or a chocolate future, and so it must either be a “vanilla specifier,” {sv1, sv2, …}, or a “chocolate specifier,” {sc1, sc2, …}. But since which type of future comes to pass—vanilla or chocolate—is up to me and is brought about in part by my free choice, which type of specifier exists—vanilla or chocolate—is also up to me and is brought about in part by my free choice. And, clearly, if something is brought about in part by my free choice, then it is explained in part by my free choice. The core of the preventable future response to fatalism is, therefore, simply this: for any given future specifier, its existence is explanatorily dependent on, and brought about by, the actual occurrences of the future contingent events that it specifies.20 The whole point of NP, the fatalist’s unpreventability assumption, is to block this response by ensuring that the existence of the posited future specifier is explanatorily independent of, and thus not even partly brought about by, the actual occurrences of future contingent events. Consider, for example, the openly fatalistic position of theistic determinism, the view that God is the ultimate sufficient cause of all creaturely events. In this view, if God knows that I will choose vanilla tomorrow, God does so not in virtue of anything I do tomorrow, but in virtue of God’s having sovereignly decreed that I choose vanilla and God’s having set in place causes sufficient to bring that about. In this model, God’s knowledge is borne out by creaturely events, but never brought about by them. The explanatory arrow runs from God to creaturely events, and there is no explanatory arrow running in the other direction. But despite what the example of theistic determinism may suggest, “explanatory independence” is simply a denial of explanatorily dependence.

260

DEBATES IN THE METAPHYSICS OF TIME

What is essential for fatalism is that the future specifiers not depend explanatorily on future contingents. It is not necessary that future events depend explanatorily on the future specifiers for, as Jonathan Edwards famously pointed out, “Infallible Foreknowledge may prove [i.e., establish] the Necessity of the event foreknown, and yet not be the thing which causes the Necessity.”21 In other words, if a future specifier is explanatorily independent of the actual occurrences of any future contingent events, then it entails the unpreventability of the future that it specifies even if it doesn’t itself render that future unpreventable. Finally, while the explanatory dependence of future specifier S on future event E licenses a counterfactual, namely, , explanatory dependence is not reducible to counterfactual dependence. In the first place, explanatory dependence is transitive and (at least) anti-symmetric and non-reflexive,22 whereas counterfactual dependence is non-transitive,23 non-symmetric, and reflexive. In the second place, if explanatory dependence were reducible to counterfactual dependence, then preventable futurism would fail as a counter to fatalism. After all, fatalists themselves would insist that future specifiers are counterfactually dependent on the events they specify. It follows from theistic determinism, for example, that if (counterfactually) I were to choose chocolate over the divinely predestined vanilla, then God would have foreknown (because he would have predestined) that I was going to choose chocolate. The presence of a counterfactual arrow running from future events to a future specifier is thus compatible with the fatalist’s insistence that no relevant explanatory arrows run in that direction. Having clarified both the core structure of fatalistic arguments and the preventable future response, we are now in a position to consider what sort of temporal ontology could underwrite the epistemically settled future that God’s foreknowledge has traditionally been thought to entail.

From epistemically to ontically settled I have already introduced three senses—causal, alethic, and epistemic—in which the future may be thought of as either “open” or “settled.” By way of review, the future is causally settled just in case only one future is compatible with the causal laws plus the causal/explanatory history of the actual world as of the present, and it is causally open just in case multiple futures are compatible with those constraints. Likewise, the future is alethically settled just in case only one future is compatible with the complete collection of truths about the future, that is, just in case there is a complete, true story of the future. And



Foreknowledge and Fatalism

261

the future is alethically open just in case multiple futures are compatible with the collection of truths. Finally, the future is epistemically settled just in case only one future is compatible with all that is known, and it is epistemically open just in case multiple futures are compatible with the sum of all knowledge. I now want to introduce a fourth sense in which the openness/settledness of the future may be understood. Let us say that the future is “ontically settled” just in case only one future is compatible with the concrete totality of future events. In other words, the future is ontically settled just in case a unique, linear, and complete sequence of future events exists. Conversely, the future is “ontically open” just in case multiple futures are compatible with the concrete totality of future events. Thus, if there are no future events—as presentists and growing-blockers would have it—or if there exists a branching array of future events—as McCall (1994) would have it—then the future is ontically open. Contrastingly, if some non-branching version of eternalism is correct, such as the “moving spotlight” version of the A-theory or a linear block version of the B-theory, then the future is ontically settled. I introduce this distinction in order to ask whether an ontically settled future is needed to underwrite an epistemically settled future. As is well known, God is standardly conceived to be essentially omniscient. While there is some debate about precisely how to analyze omniscience,24 I shall take it to be the view that God essentially believes all and only truths, believes them infallibly, and is immediately and fully acquainted with all of reality. It follows that if there is a complete, true story of the future, then God knows it. An alethically settled future, therefore, entails an epistemically settled future. Conversely, since knowledge entails truth, if the future is epistemically settled it is also alethically settled. Given an essentially omniscient God, then, alethic and epistemic settledness/openness necessarily go hand in hand. Hence, we can replace the question about whether an ontically settled future is needed to underwrite an epistemically settled future, with the question of whether it is needed to underwrite an alethically settled future. If it is necessary for the latter, then it is necessary for the former. And if it is sufficient for the latter, then it is also sufficient for the former. The supposition that the future is alethically settled raises a question: “What makes this story of the future the true one?” Truth, it is plausible to suppose, supervenes on being.25 What is true is true in virtue of what is real. This is especially plausible for logically contingent truths,26 of which truths about future contingents are obviously a subset. If is true, it seems proper to ask why that is true when ex hypothesi has (we may assume) just as good a chance of being true instead. Since contingent propositions cannot be true in virtue of themselves, something else must be different about reality in virtue of which the first is true and not

262

DEBATES IN THE METAPHYSICS OF TIME

the second. Hence, if there is a complete, true story of the future, then we need an ontology robust enough to explain why this story is true as opposed to any other that might otherwise have been true. One way to ground an alethically settled future is to suppose that the future is causally settled. If it is, then a God who knows the causal laws plus the causal/explanatory history of the actual world as of the present will be able to predict with certainty exactly how the future will go. So a causally settled future, given an omniscient God, entails an epistemically settled future. And since knowledge entails truth, it also entails an alethically settled future. But it does so at the cost of giving up future contingency. If we want future contingency, we need another way to ground an alethically settled future. So let us suppose that the future is causally open. In this case, the causal laws plus the causal/explanatory history of the world as of the present leave underdetermined which future is to be the actual one. Hence, if we want to ground an alethically settled future, we will need something more in our ontology. An obvious thought is to suppose that the future is ontically settled, that is, to suppose that a unique, linear, and complete sequence of future events exists. If that is so, then, since an omniscient God is fully acquainted with all of reality, God would be fully acquainted with all actual future events, and so the future would be epistemically and, therefore, alethically settled. So unless there’s some deeper incompatibility between a causally open and an ontically settled future, this looks like an effective way to ground divine foreknowledge of future contingents. If, however, the future is neither causally nor ontically settled, then it is unclear how an alethically settled future could be grounded. Assume that the future is ontically open. In presentist and growing-block models, future events and entities do not exist and so are not available to do any grounding,27 whereas in a branching-future model like McCall’s, too many future events and entities exist to single out any one future as actual. So given ontic openness, future events and entities don’t suffice for grounding an alethically settled future. Let us now factor in non-future events and entities while assuming that the future is causally open, that is, that there are future contingents. As just noted, the causal laws plus the causal/explanatory history of the world as of the present do not suffice for grounding because they underdetermine which future is to be the actual one. But that underdetermination remains even if we add in past, present, and even timeless events and entities that are not part of that causal/explanatory history. Because such events and entities have no explanatory bearing upon which future events occur, they don’t substantively contribute toward this causally possible future’s coming to pass rather than another, and so they don’t suffice to explain its being true that this causally possible future comes to pass rather than another. In sum,



Foreknowledge and Fatalism

263

then, given both causal and ontic openness, neither future events and entities nor explanatorily relevant non-future events and entities nor explanatorily irrelevant non-future events and entities suffice, either individually or collectively, to ground an alethically settled future. But then what else is there that could provide such grounding?28 I conclude that, unless we are prepared to jettison the highly plausible principle that (contingent) truth supervenes on being, our best hope—indeed, our only hope—for reconciling an alethically and epistemically settled future with a causally open one is via an ontically settled future. I now examine a recent proposal along these lines.

An Anselmian solution? Katherin Rogers (2008) has recently endorsed and defended what she cogently argues to be Anselm’s response to the problem of divine foreknowledge and future contingency. Her objective is to describe a model of reality and of God’s relation to it that makes clear how the future can be both causally open and epistemically settled (for God). She summarizes as follows: Anselm’s solution rests on three premises: (1) the sort of “necessity” which follows upon divine foreknowledge need not conflict in any way at all with the most robust libertarian freedom because (2) God is eternal and (3) time is essentially tenseless. (Rogers 2008: 146) Each of these points requires some unpacking. I will take them in reverse order. By (3), the idea that time is essentially tenseless, Rogers means to endorse a B-theoretical version of “eternalism,” which she prefers to call “four-dimensionalism” (Rogers 2008: 158) so as to reserve “eternal” and its derivatives for God’s timeless mode of being. More precisely, Rogers must intend to endorse by (3) a “linear block” version of eternalism, such that the future is ontically settled. She must intend this because eternalism alone, in either an A- or B-theoretical interpretation, is not sufficient for her purposes. This is because eternalism—the idea that all past, present, and future events (tenselessly) exist—is compatible with a non-linear or branching future. In a “branching block” version of eternalism, there would be no unique future for God to know as the actual future. Hence, the future would be neither alethically nor epistemically settled. By (2), God’s eternality, Rogers means that God is essentially timeless (cf. Rogers 2008: 146–7). From this it follows that God’s existence is essentially beginningless and endless, that God essentially lacks any temporal

264

DEBATES IN THE METAPHYSICS OF TIME

properties, and that God is essentially immune to intrinsic change. Rogers is far from alone in thinking that divine timelessness is essential for rebutting the epistemic argument for fatalism. One common motivation behind this thought is that divine timelessness allows one to sidestep fatalistic worries about the “fixity” or “accidental necessity” of the past since a timeless God’s knowledge isn’t in the past. In addition, divine timelessness is often thought to afford a model of how God can know future contingents in that God, from a vantage point “outside” of time, is able to survey all at once the entire sweep of history. Nevertheless, as I argued above, divine timelessness is not sufficient for rebutting epistemic arguments for fatalism.29 And while one might take it to be necessary on the grounds that a temporally situated God couldn’t possibly survey the actual occurrences of all future contingent events, it is not immediately clear why a temporally situated God couldn’t have the requisite access to future events. In an A-theoretical version of eternalism, for example, such as the “moving spotlight” theory,30 all future events (tenselessly) exist and so are available for a transcendentally temporal God to be acquainted with.31 In sum, pace Rogers, (2) is at most an optional commitment of an Anselmian solution to the foreknowledge/future contingency problem. Finally, in (1) the sort of necessity which “follows upon divine foreknowledge” is, says Rogers, merely a conditional or “consequent necessity” (Rogers 2008: 158). It is the kind of necessity by which God foreknows if and only if X happens in the future (i.e., in what we temporally situated beings think of as the future). This necessity “need not conflict” with future contingency because it is compatible with the explanatory dependence of God’s knowledge upon the actual occurrences of future contingent events. As Rogers repeatedly stresses, the arrow of dependency runs from future contingent events to God: It is the fact that the agent actually chooses what he chooses that produces God’s knowledge of the choice, and so the consequent necessity involved in divine knowledge is ultimately produced by the agent making the choice … Anselm’s position entails that God “learns” from us. He knows what we choose, because we choose it. (Rogers 2008: 175–6) In summary, Rogers (following Anselm) proposes to solve the divine foreknowledge/future contingency problem by supposing a linear block model of time and a God “outside” time whose knowledge of future contingent events “in” time is quasi-perceptual and thus explanatorily dependent upon the actual occurrences of those events. One merit of this proposal is that it directly challenges the epistemic argument for fatalism based on divine foreknowledge by offering, in effect, an NP-denying or preventable future



Foreknowledge and Fatalism

265

response to fatalism. Another merit is that it provides us with a reasonably clear model for how the future could be epistemically settled for God, specifically, through God’s being acquainted with an ontically settled future. Unfortunately, as I will now argue, Rogers’ solution does not work, for divine timelessness and an ontically settled future are each incompatible with a preventable future response to fatalism. I develop my argument in three stages. First, I raise an objection against divine timelessness based on the idea that divine choices entail intrinsic change, and thus temporal sequence, in God. While the objection is not conclusive, it is instructive, for—and this is the second stage of the argument—parallel reasoning shows that preventable future specifiers cannot be atemporal. Hence, divine timelessness, when coupled with the traditional idea that God’s foreknowledge constitutes a future specifier, not only does not help solve the foreknowledge/future contingency problem, but actually entails fatalism. Finally, in the third stage, I show that if an ontically settled future must be temporally invariant, then it too is incompatible with a preventable future response, and thus entails fatalism.

First stage: A problem for divine timelessness Most theists have believed that God makes choices, including choices about whether to create, about which (type of) world to create, and how to respond to creaturely actions. But perhaps, as some have argued, the inherent diffusiveness of divine love ensures that God creates some world or other. And perhaps, as others have argued, God must create the best type of world or at least one from among the class of best strongly actualizable world types.32 Still, even if we grant all that, most theists would be inclined to think that God must have had some “open options”—for example, the option to create one more or one fewer hydrogen atoms in some far-flung corner of the universe, or the option to make it such that humans perceive an inverted color spectrum. For present purposes it does not matter what the options are. As long as God has at least one open option requiring at least one choice on God’s part, essential divine timelessness is ruled out. This is because choices are inherently temporal events essentially involving both a “before” state of contemplating the options without as yet having settled upon any of them, and an “after” state of having decided upon one of the options over the others. The relation between the two states cannot be understood as one of merely logical priority, for the states are mutually incompatible—one cannot be concurrently both undecided and decided with respect to the same option—whereas relations of merely logical priority can obtain only between things that are mutually compatible, such as the

266

DEBATES IN THE METAPHYSICS OF TIME

premises and conclusion of a valid argument. Moreover, the transition from not yet having decided to having decided constitutes an intrinsic change in the chooser. Hence, the relation must be one of temporal priority, of before and after. Because intrinsic change is impossible for a timeless being, it is impossible for a timeless God literally to choose and to remain timeless.33 An essentially timeless God therefore must not face open options. Rogers seems to realize this. Affirming essential divine timelessness, she follows Anselm in denying that God ever faces open options because “God, being the best, does the best” (p. 185). This assumes, however, that there is such a thing as a unique best that God can do—a best strongly actualizable world type (cf. Rogers 2008: 195). Rogers defends this seemingly implausible claim with vigor, though in the end she claims only that it is not “wildly implausible” or “obvious madness” to suppose that ours is the best type of world God could have actualized (Rogers 2008: 205). Be that as it may, the incompatibility of divine timelessness and open options vitiates Rogers’ preventable future response to fatalism.

Second stage: Open options and future contingency As with choices, future contingency entails open options (of a sort), namely, the different types of causally possible futures that hinge upon future contingent events. The actual occurrences of future contingent events are in fact analogous to—and in some cases are—decisions between open options. My freely choosing vanilla ice cream tomorrow over chocolate is a case in point. Prior to my choice, both chocolate-futures (in which I choose chocolate) and vanilla-futures (in which I choose vanilla) are causally possible. Both are open options for me. But even when there is no agent per se, as in the case of quantum-level indeterminism, we can still think of future contingents as providing open options for a physical system, or even for reality in general. Somehow—we need not know how—reality “chooses” to go one way rather than the other. Now recall our discussion of fatalism. If we admit future specifiers, then we avoid fatalism only by saying that which token future specifiers exist is explanatorily dependent upon the actual occurrences of future contingent events. This is the “preventable future” response to fatalism, which Rogers endorses. But now we run into another problem for divine timelessness. For if the existence of some token future specifier is explanatorily dependent upon, say, my choosing vanilla ice cream tomorrow, and if my so choosing is now a future contingent, then there are causally possible futures in which that future specifier exists and causally possible futures in which that future specifier does not exist. In other words, future specifiers that are



Foreknowledge and Fatalism

267

explanatorily dependent upon the actual occurrences of future contingent events are themselves future contingents—they obtain in some, but not all, causally possible futures. This is a crucial point. It implies that nothing can be both timeless and explanatorily dependent upon future contingents since the temporality of the latter is inherited by anything explanatorily dependent upon them. Hence, far from helping to solve the foreknowledge/future contingency problem, divine timelessness precludes a preventable future response to fatalism. This consequence is one that Rogers sometimes seems to be on the verge of grasping. Thus, as she puts it in one passage, “Anselm’s position entails that God ‘learns’ from us. He knows what we choose, because we choose it” (Rogers 2008: 176). Later, in a footnote, she writes that “God must ‘wait and see’ what created agents actually choose” (Rogers 2008: 195n. 29). In the preventable future response, this is exactly right, but if God “learns” from us and must “wait and see” what we choose, then this introduces temporal sequence into God. Just as divine choices entail a before–after sequence in God, consisting of a not-having-yet-decided state followed by a having-decided state, so also the explanatory dependence of God’s knowledge upon future contingents entails a before–after sequence in God consisting of a not-havingyet-learned state followed by a having-learned state. Rogers comes closest to realizing this when she writes that “God cannot know what the created agent chooses ‘until’ (logically, not temporally) he chooses it” (Rogers 2008: 150), but she wrongly supposes that the sequence can be merely “logical” and not “temporal.” As explained above, merely logical sequences, such as obtain between the premises and conclusion of an argument, require that the termini be mutually compatible. In the case of learning, however, as in the case of choosing, they are mutually incompatible. One cannot both have learned something and not have learned it, either at the same time, or even at the same timeless “moment.”

Third stage: Implications for temporal ontology The future is ontically settled just in case a unique, complete, and linear sequence of future events exists. If such a sequence does exist, it is a future specifier. Hence, to avoid fatalism, the ontically settled future must be preventable, such that whether this sequence of future events exists or not is explanatorily dependent upon the actual occurrences of future contingent events. But if a unique, complete, and linear sequence of future events exists, it cannot exist now (because the events are future), and so presumably it must exist tenselessly or sub specie aeternitatis. But then for the same reason that divine timelessness (plus omniscience) precludes a

268

DEBATES IN THE METAPHYSICS OF TIME

preventable future, it seems as if an ontically settled future does so too. As we have seen, preventable future specifiers are themselves future contingents. It is doubtful, however, whether a tenselessly existing sequence of events can be a future contingent. Future contingency seems to be an inherently tensed status because there invariably comes a time when the event in question is either no longer future, or no longer contingent. The proverbial sea battle tomorrow may now be a future contingent, but it won’t be one after tomorrow. At any rate, if a future specifier is explanatorily dependent upon the actual occurrences of future contingent events, then whether a token future specifier exists is something that reality must, so to speak, “wait and see” to find out. Just as with the case of God’s foreknowledge, there is a before–after sequence: the “before” state of reality’s not being yet determinate with respect to whether a sea battle occurs tomorrow, and an “after” state of reality’s being determinate in that respect. And since these states are mutually incompatible, the sequence must be temporal, not merely logical. This, however, seems to be incompatible with the supposition that the sequence of events constituting the ontically settled future exists tenselessly—at least it does if “existing tenselessly” entails either atemporality or temporal invariance.34 If this is right, then an ontically settled future cannot be a preventable future, and thus entails fatalism.

Conclusion Reflection upon fatalism has significant implications for both theology and temporal ontology. I have argued that fatalism is entailed by the existence of a fixed or unpreventable future specifier and that there are, therefore, only two ways of resisting the fatalist’s conclusion. One can adopt an “open future” strategy and deny that any future specifiers posited by the fatalist exist, or one can adopt a “preventable future” strategy and hold that which token future specifiers exist is explanatorily dependent upon the actual occurrences of future contingent events. Traditionally, most theists have thought of God’s foreknowledge as a future specifier, and so most theistic anti-fatalists have been preventable futurists. They have thought of the future as being epistemically and alethically settled but causally open. Because (contingent) truth supervenes on being, however, such theists are also implicitly committed to an ontically settled future because only thus would there be adequate grounds for the “complete, true story of the future” that God has traditionally been thought to know. Rogers’ Anselmian response to the foreknowledge/future contingency problem embraces an ontically settled future and tries to avoid fatalism by combining divine timelessness with a preventable future response. Unfortunately, neither divine timelessness nor



Foreknowledge and Fatalism

269

(arguably) an ontically settled future is compatible with the preventable future response, and thus, far from helping the anti-fatalist it actually lends support to fatalism. If my argument thus far has been successful, then to avoid fatalism we must affirm a causally and ontically open future (or find a way to affirm an ontically settled future without countenancing any temporally invariant future specifiers). Given that (contingent) truth supervenes on being, such a future must also be alethically and epistemically open. Hence, the traditional conception of divine foreknowledge, which entails an epistemically settled future, is untenable. My advice to anti-fatalistic eternalists and theists is to be thoroughgoing open-futurists, to hold that the future is causally, ontically, alethically, and epistemically open. I close by noting that it is in fact possible to affirm eternalism and divine timelessness while being a thoroughgoing open futurist. Start with McCall’s branching model of time, according to which all of the many causally possible futures exist in a branching array, with nodes representing decision points for future contingents. McCall’s model is dynamic in that, as future contingents are resolved, unchosen branches drop off or cease to exist. We can convert it into an eternalist model by setting the dynamic component aside and holding that all events that are ever causally possible (tenselessly) exist. The resulting static “branching block” model of time would be causally, ontically, alethically, and epistemically open.35 A timeless and omniscient God could be fully acquainted with the whole branching array of events. Of course, this model requires giving up the traditional idea that God’s foreknowledge constitutes a future specifier, but if my argument is sound, that idea will have to go anyway on pain of losing future contingency altogether.36

Notes   1 Most theists are deeply concerned to protect God against the charge of being ultimately responsible for human wrongdoing. But this arguably requires that humans occasionally have the ability to exercise libertarian freedom, which in turn requires indeterminism and thus future contingency. Theists who deny human libertarian freedom have a comparatively harder time with the problem of evil, for if human moral responsibility is compatible with all human behavior being (ultimately) determined by God, then it is hard to see why an all-good, all-powerful God couldn’t have and wouldn’t have created a sinless world—or at least a much less sinful one. See Rhoda (2010a).   2 Important recent studies of the problem include Craig (1991), Fischer (1989), Hasker (1989), and Zagzebski (1991).   3 For example, Rhoda (2010b) and Tuggy (2007).   4 For example, Ware (2000).

270

DEBATES IN THE METAPHYSICS OF TIME

  5 This is the gist of Ockham’s (1983) response to the foreknowledge/future contingency problem.   6 Boethius (1973: bk. V, prose 6).   7 Prior (2003).   8 It would be more exact to speak of “complete histories” rather than “worlds,” for the common assumption that each possible world essentially includes a complete history (a past, present, and future) is problematic, for reasons given in Rhoda (2010a: 284). Nevertheless, I will stick with the more familiar term “worlds.”   9 The “causal/explanatory” qualifier is important for two reasons. First, to say simply “same history” would beg the question against anti-fatalists like Ockham (1983) who want to say both (1) that there are multiple causally possible futures, and (2) that only one causally possible future is compatible with the entire actual past. Second, the causal/explanatory order is, at least arguably, not necessarily restricted to the temporal past. If backward causation is possible, for example, then the causal/explanatory “history” of an event may include future events. Alternatively, if there is a timeless God who causally sustains a temporal creation or who provides enabling “concurrences” for creaturely actions, then God’s activity is part of the causal/explanatory “history” of creaturely events even though it isn’t part of temporal history. 10 As I have characterized it, “fatalism” entails that there are not now any future contingents. It does not entail either that there never have been any future contingents or that it is metaphysically impossible that there be future contingents. While most historical fatalists would endorse either or both of those stronger claims, my justification for the weaker characterization is that what makes fatalism disturbing to most is its implication that we have no independent “say” in what course our own future will take. To learn, after discovering that one’s future is fated, that it hasn’t always been fated or that it is only contingently fated would provide no existential comfort. 11 The terms “logical fatalism” and “theological fatalism,” while common in the literature, are misleading because they suggest that these are different types of fatalism, when they are really just different ways of arguing for fatalism. 12 There is some debate about the exact nature of Aristotle’s response to the fatalistic argument. See Craig (1988) and Gaskin (1995) for discussion. 13 Open theists (e.g. Rhoda et al. 2006, and Tuggy 2007) and process theists (e.g., Viney and Shields 2003) often take this line. 14 Ockham (1983). 15 I discuss both “open future” and “preventable future” responses in greater detail in Rhoda (n.d.). 16 Of course, some formulations of the epistemic argument for fatalism (e.g. Pike 1965) do presuppose that God has temporally prior knowledge of the future. Appeal to divine timelessness clearly undercuts those formulations. Divine timelessness may also have some utility against epistemic arguments for fatalism in general, just in case it is less plausible that a timeless God’s knowledge would be among the fixed facts than that the



Foreknowledge and Fatalism

271

corresponding knowledge of a temporally everlasting God would be. On this point, see Zagzebski (1991, ch. 2). But this potential benefit of divine timelessness will be nullified below, where I argue that preventability entails temporality and thus that a God whose foreknowledge is a preventable future specifier cannot be atemporal. 17 For a good discussion of the hard fact/soft fact distinction, see the introduction to Fischer (1989). As Fischer (1994: 115) notes, “[I]t is very important to distinguish two sets of issues: first, temporal nonrelationality and relationality (i.e. hardness and softness), and second, fixity and non-fixity (i.e. being out of one’s control and being in one’s control).” Despite the amount of ink that has been spilled on the former distinction, it is the latter that is the crucial one. 18 Suppose these are my only two options, and that they are mutually exclusive. 19 In what follows, past tense expressions related to God’s knowing should be understood in a manner that is neutral on the question of whether God is timeless or not. 20 A similar point is made by Fischer et al. (2009: 255ff.) and by Finch and Rea (2008: 11ff.). “Explanatory dependence,” in my usage, is a species of what Lowe (2010) calls “ontological dependence” since it is a matter of what accounts for the existence of a token future specifier. 21 Edwards (2009 [1754]: II.12). 22 Unlike asymmetry, which precludes the joint possibility of aRb and bRa, anti-symmetry allows for their joint possibility, but only if a=b. To hold that explanatory dependence is asymmetric and irreflexive is to rule out the possibility of self-explanation. Perhaps we should rule that out, but I am unsure about this, and so regard it as safer to view explanatory dependence as anti-symmetrical and non-reflexive. At any rate, nothing in my argument turns on this point. 23 Lewis (1973: 32–5). 24 For discussion of some of the issues, see Hoffman and Rosenkrantz (2002: 111–26). 25 The truth supervenes on being (TSB) principle is weaker than its close cousin, the truthmaker (TM) principle. TM says, minimally, that for every truth there exists something that makes it true, whereas TSB only requires that reality as a whole be appropriately different from what it would have been had what is (contingently) true been false instead. The difference between TM and TSB becomes clear in the case of negative existentials. TM requires that for to be true there must exist something—a universal unicorn-excluding state of affairs—that makes it true, whereas TSB is satisfied by the non-existence of anything (e.g. a unicorn) that would make it false. 26 One might suppose that logically necessary truths, and especially analytic truths, need nothing to ground their truth. Or perhaps we should suppose that such truths are their own truthmakers. Cf. David (2009: 153). 27 Some presentists say that what makes it true now that there will be a

272

DEBATES IN THE METAPHYSICS OF TIME sea battle tomorrow is simply tomorrow’s occurrence of a sea battle. Cf. Craig (2000: 213–14). But this is a transparent dodge of the grounding requirement. If the future isn’t real (as per presentism), then tomorrow’s occurrence of a sea battle doesn’t (yet) have any ontological status, and so isn’t (yet) there to make it true now that there will be a sea battle tomorrow.

28 It may be suggested that Molinism can come to the rescue. According to Molinism, God’s knowledge of future contingents is grounded in God’s free decision of which world to create, which decision in turn is grounded in God’s “middle knowledge” of what every possible free creature would do in any possible causally specified indeterministic scenario. But this merely trades one grounding problem (what makes it true that this future is the actual one?) for another (what makes these middle knowledge conditionals true?). 29 More precisely, divine timelessness may be sufficient for rebutting some versions of the epistemic argument for fatalism, specifically, those that assume a temporally situated knower, but it is not sufficient for rebutting epistemic arguments for fatalism in their most general form. 30 The locus classicus for the moving spotlight theory is Broad (1923: 59ff.). 31 Since all A-theoretical models of time admit some tensed facts (e.g., what time it is now) as ontologically basic, an omniscient God couldn’t know such facts without intrinsic change, and therefore temporal sequence, in God. Thus, when T1 is present, God would know and not . Later, when T2 is present, God would know and not . Hence, the moving spotlight theory entails divine temporality. But it also requires that the temporal sequence of God’s life be distinct from, and transcendent over, the linear block time of creation. Hence, it requires that God be “transcendentally temporal.” 32 Something is “strongly actualizable” if God can unilaterally cause it to be. If there are future contingents, such as future human libertarian free choices, then God cannot strongly actualize those events, for that would be contrary to their status as future contingents. He can, however, strongly actualize a world type by unilaterally fixing everything in it that does not depend on future contingents, such as causal laws, initial conditions, and unilateral divine interventions. The notion of strong actualization comes from Plantinga (1974: 172–3). 33 A proponent of divine timelessness can say that God eternally wills thusly, but not that God chooses to will thusly. 34 I say “seems to be incompatible” because I am not convinced that my argument here is correct. Nevertheless, I think the reasoning is plausible enough and the conclusion significant enough that the argument deserves a wider hearing. 35 I believe this model is more plausible than McCall’s. Not only does it fit well with the Everett “many worlds” interpretation of quantum mechanics (cf. Vaidman 2008), but it also avoids an exceedingly odd consequence of McCall’s model, namely, that every time an indeterministic event occurs, a huge swath of reality—everything in the branches of all of the “nonchosen” causal possibilities—is thereby consigned to oblivion. (On the oddness of



Foreknowledge and Fatalism

273

this, see Miller 2006). To my knowledge, however, no one has yet endorsed a branching block model. 36 Portions of this chapter were presented to Notre Dame’s philosophy of religion discussion group and at the 2010 Central Division meeting of the Philosophy of Time Society. I benefitted greatly from comments received at those venues.

References Boethius (1973), The Consolation of Philosophy, Loeb Classical Library, 74. Cambridge, MA: Harvard University Press. Broad, C. D. (1923), Scientific Thought. London: Kegan Paul. Craig, W. L. (1988), The Problem of Divine Foreknowledge and Future Contingents from Aristotle to Suarez. Leiden: Brill. —(1991), Divine Foreknowledge and Human Freedom. Leiden: Brill. —(2000), The Tensed Theory of Time: A Critical Appraisal. Dordrecht: Kluwer Academic. David, M. (2009), “Truth-Making and Correspondence,” in E. J. Lowe and A. Rami (eds), Truth and Truth-Making. Montreal: McGill-Queen’s University Press, pp. 137–57. Edwards, J. (2009 [1754]), Freedom of the Will. Vancouver: Eremitical Press. Finch, A. and Rea, M. (2008), “Presentism and Ockham’s way out,” Oxford Studies in Philosophy of Religion, 1, 1–17. Fischer, J. M. (ed.) (1989), God, Foreknowledge, and Freedom. Stanford, CA: Stanford University Press. —(1994), The Metaphysics of Free Will: An Essay on Control. Oxford: Blackwell. Fischer, J. M., Todd, P. and Tognazzini, N. (2009), “Engaging with Pike: God, freedom, and time,” Philosophical Papers, 38, 2, 247–70. Gaskin, R. (1995), The Sea Battle and the Master Argument: Aristotle and Diodorus Cronus on the Metaphysics of the Future. Berlin: Walter de Gruyter. Hasker, W. (1989), God, Time, and Knowledge. Ithaca, NY: Cornell University Press. —(2001), “The foreknowledge conundrum,” International Journal for the Philosophy of Religion, 50, 97–114. Hoffman, J. and Rosenkrantz, G. S. (2002), The Divine Attributes. Oxford: Blackwell. Lewis, D. (1973), Counterfactuals. Oxford: Blackwell. Lowe, E. J. (2010), “Ontological Dependence,” in E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Spring). http://plato.stanford.edu/archives/ spr2010/entries/dependence-ontological/ McCall, S. (1994), A Model of the Universe: Space-Time, Probability, and Decision. Oxford: Oxford University Press. Miller, K. (2006), “Morality in a branching universe,” Disputatio, 20, 1, 305–25. Ockham, W. (1983), Predestination, Foreknowledge, and Future Contingents, 2nd edn, trans. M. M. Adams and N. Kretzmann. Indianapolis: Hackett. Pike, N. (1965), “Divine omniscience and voluntary action,” Philosophical Review, 74, 27–46. Plantinga, A. (1974), The Nature of Necessity. Oxford: Clarendon Press.

274

DEBATES IN THE METAPHYSICS OF TIME

Prior, A. N. (2003), “The Formation of Omniscience,” in P. Hasle, P. Øhrstrøm, T. Braüner and J. Copeland (eds), Papers on Time and Tense, new edn. Oxford: Oxford University Press, pp. 39–58. Rhoda, A. R. (2010a), “Gratuitous evil and divine providence,” Religious Studies, 46, 281–302. —(2010b), “The Fivefold Openness of the Future,” in W. Hasker, D. Zimmerman and T. J. Oord (eds), God in an Open Universe: Science, Metaphysics, and Open Theism. Eugene, OR: Pickwick, pp. 69–93. —(n.d.), “Five roads to fatalism and the openness of the future,” Unpublished manuscript. Rogers, K. A. (2008), Anselm on Freedom. Oxford: Oxford University Press. Tuggy, D. (2007), “Three roads to open theism,” Faith and Philosophy, 24, 28–51. Vaidman, L. (2008), “Many-worlds Interpretation of Quantum Mechanics,” in E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Spring). http:// plato.stanford.edu/archives/fall2008/entries/qm-manyworlds/ Viney, D. W. and Shields, G. (2003), “The Logic of Future Contingents,” in G. Shields (ed.), Process and Analysis: Essays on Whitehead, Hartshorne and the Analytic Tradition. Albany: State University of New York Press, pp. 209–46. Ware, B. A. (2000), God’s Lesser Glory: The Diminished God of Open Theism. Wheaton, IL: Crossway. Zagzebski, L. T. (1991), The Dilemma of Freedom and Foreknowledge. New York: Oxford University Press.

13 Defending the Isotemporalist Solutionto the Freedom/ Foreknowledge Dilemma: Response to Rhoda Katherin A. Rogers

A

lan Rhoda presents a thoughtful analysis of the dilemma of freedom and divine foreknowledge critiquing the isotemporalist solution offered by Anselm of Canterbury, a solution I have defended. Here I attempt to respond to Rhoda’s criticisms. First, though, I believe it will be helpful to spell out the Anselmian solution in more detail than Rhoda includes. I will move back and forth between Anselm himself, and “the Anselmian,” that is, me, as a follower of Anselm, engaged in developing and defending his solution in the contemporary idiom. Anselm is, to my knowledge, the first philosopher to develop a wellworked out, analytic, and systematic theory of libertarian free will.1 For Anselm, a core requisite of the sort of free will worth wanting, the sort which is needed for moral responsibility, is that one’s acts of will be a se, that is absolutely from oneself. In order to have aseity an act of will cannot be causally or metaphysically necessitated by anything outside of oneself. Anselm allows that an act of will which is causally necessitated by one’s character may be free in the way required for responsibility if one caused one’s character oneself through earlier, non-necessitated, a se acts of will. The character-necessitated choice can be from oneself, if one’s character is from oneself. Indeed, the important thing in Anselm’s view is not isolated choices and actions on the part of an agent, but the sort of person the

276

DEBATES IN THE METAPHYSICS OF TIME

agent becomes due to the choices he makes and the subsequent actions he engages in. By including “or metaphysically necessitated” above I mean to express the thought that an act of will that is not causally necessitated may yet be necessitated in a way that conflicts with aseity, if it is rendered necessary by something other than the agent himself actually engaging in the act. For example, contemporary Molinism holds that there are “counterfactuals of freedom,” eternally true propositions about what any possible free agent would choose in any possible situation. In the Molinist universe, it is eternally necessary that the actual agent choose what the counterfactual of freedom has him choosing, and the truth of the counterfactual of freedom is not dependent on what the actual agent actually chooses. The Anselmian argues that there are no robustly free libertarian choices in the Molinist universe.2 Or conversely, if agents make a se choices, we do not live in a Molinist universe. The point can be put another way by noting that Anselmian libertarianism entails what I will call the “grounding principle.” Let us take as an example of an Anselmian, a se choice, some agent, S, who chooses B at t2. The grounding principle holds that the indispensable, originating event in the causal process which results in anyone’s knowing that S chooses B at t2 is the actual event of S choosing B at t2. And the truth of the proposition “S chooses B at t2” absolutely depends upon the actual event of S choosing B at t2. All truth concerning, and knowledge of, an a se choice is “grounded in” the actual choice. (This is similar to Rhoda’s point about “explanatory dependence” (258–60). I will cite Rhoda’s chapter to which I am responding by page numbers in parentheses.) Libertarians usually insist not just upon a choice being “from oneself” somehow, but also that the agent confront open options. And Anselm is no different. He is motivated by the puzzle of trying to allow created agents to be able to choose a se in the universe of classical theism in which everything that is is immediately kept in being by God. Where is there room for anything to be up to the created agent? Anselm’s solution is to propose that, while all that exists is immediately caused by God, all that happens is not. In a free human choice, God causes the agent, the agent’s faculty of will, and the motivations which move the agent to will. But sometimes the agent is motivated by conflicting (divinely caused) motivations, both of which cannot be chosen. For example, say that at t1 S is in what we can call the “torn condition,” motivated to choose A and motivated to choose B, where the choice of one over the other has moral significance, and the choice of one precludes the choice of the other. In that sort of situation, Anselm holds, the choice itself is absolutely up to the agent. (Hence the conflict between causal or metaphysical necessitation on the one hand and a se choice on the other.) In our example, it is S himself who makes it the case that S chooses



Defending the Isotemporalist Solution

277

B at t2. But had he not been confronted by the alternative possibilities, that is, had he simply been motivated by one, God-given motivation, or by one, overriding, God-given motivation more strongly than any other, he would not have chosen freely. Or if, following the torn condition, he had been caused to opt for one over the other by something outside himself—God, a brain tumor, a malignant neuroscientist—such that the options were not really open, then he could not make an a se choice. And if some truth about the universe independent of what S actually chooses at t2—like the Molinist counterfactuals of freedom—had rendered it necessary that S choose one way rather than the other, then the choice could not be a se. Thus alternative possibilities are a key aspect of Anselmian libertarianism. But note that alternative possibilities are necessary as a basis for aseity only for the created agent. For an uncreated agent such as God, who exists absolutely a se, his acts of will are from himself alone, so alternatives play no part in divine freedom.3 Anselm’s insistence on aseity, which entails the grounding principle, gives rise to his version of the dilemma of freedom and divine foreknowledge. If knowledge of a free choice is absolutely grounded only in the actual making of the choice (as is the truth of propositions about the choice), how can God possibly know at t1 that S chooses B at t2? Anselm addresses the question in a later work, De Concordia, but he already had the pieces needed to solve the puzzle at hand from his earlier discussions of the nature of God, especially in the Proslogion.4 God is “that than which no greater can be conceived.” He is simple and immutable. He has all possible knowledge in the most perfect way possible and all possible power in the most perfect way possible. But what does all this mean? Especially, what does it all mean concerning the dilemma of freedom and divine foreknowledge? Divine simplicity entails that God is not spatially extended, some here, some there. And it entails that he is not temporally extended, some now, some past and gone, some not yet. Yet he knows all that happens in all of time and space. God’s knowledge is the best sort of knowledge, that is, direct knowledge. He knows everything that is by immediately thinking it. It is his thinking that makes whatever exists to exist. God’s thinking is his willing, and he thinks and wills all that exists in one, simple act of thinking/willing. So, of course, there is no change in God, since that would imply that his life consists of temporal parts. But what about foreknowledge? Well, if foreknowledge is logically possible, then an omniscient being must have it. Just as that than which a greater cannot be conceived must know all of space, and all that space contains, directly as the immediate cause of the existence of all spatial things, he must know and will directly all of time and all that all times contain. God is the immediate cause of the existence of all temporal things at whatever times they exist. (How in the world, then, is he not the cause of all human choices? Well, that

278

DEBATES IN THE METAPHYSICS OF TIME

is an interesting story which I have been trying to tell in various venues, but here is not the time or place. Suffice to say that God causes all things with ontological status, but the human choice is not a “thing.”)5 A certain theory of time follows from this unpacking of the concept of a perfect being. Were it the case that all that exists is the present moment and all its “contents”—call this presentism—such that anything that exists, including God, exists only at the present moment, then God’s knowledge and power would be radically limited. He could know and act directly only in the present moment, since the present moment is all there is. He might remember the past and anticipate the future, and he might do things now which would impact the future, but his direct knowledge and action would be limited to the present moment. And what a limited God that would be! Better to make the move that Anselm makes and adopt the theory that all times, what to the temporal perceiver at a given time seems to be present and past and future, are all equally real.6 As Rhoda notes, I had used the term “four-dimensionalism” in my 2008 book, but I have since adopted the term “isotemporalism.”7 It is preferable because it captures the equality of ontological status for all times. God directly knows and wills every moment of time and whatever existing things that moment contains.8 Assuming this is logically possible, this is a picture that ascribes to God much more knowledge and power than if we adopt presentism. It is true that isotemporalism is phenomenologically bizarre.9 It feels to us like all that exists is the present moment, for any given moment which is present to a temporal perceiver. But that feeling is misleading. It is as if the spatial perceiver were to feel as if only what he could perceive was actually existent, such that what was “here” to him was all there was. Space seems to be less of a problem for us. All of space is “there,” even if we have access only to what is “here” for us. Well, isotemporalism makes the same claim for time. Even though, at each moment, “now” for a temporal perceiver feels like all there is, in fact what is “then,” the past, or “yet to come,” the future, is equally real. Isotemporalism entails the absolute indexicality of terms like “past,” “present,” and “future.” What is past or present or future is indexed to some particular moment in time. There is no objective past, present, or future, since what is past, and so on, depends on what moment in time you pick as the vantage point from which to consider the time line. Rhoda defines a “future” as “an abstract representation of a unique, complete, linear extension of the actual past” (p. 255). I would subscribe to this definition regarding causally and metaphysically possible futures, that is, futures which are causally or metaphysically possible extensions of the actual past. But I consider the future to be not an abstract representation but rather just actual reality occurring after some time understood as present or past. This is what isotemporalism entails. Note that it does not entail that “before,” “simultaneously with,” or “after” are subjective. It does not do away with the “arrow” of time. It just says that all the sequentially ordered moments exist equally.10



Defending the Isotemporalist Solution

279

Rhoda labels my view of the universe a “linear block.” This is accurate, so long as the whole block is understood to exist equally. It is not the case that there is a past and present which are actual and a future which is merely a representation. There is just the ordered series of equally real times. It is linear because it is not branching. At the end of his chapter, Rhoda mentions a branching hypothesis as a way of maintaining divine foreknowledge and eternity, while rejecting fatalism. The thesis is that “all of the many causally possible futures exist in a branching array … [and] all events that are ever causally possible (tenselessly) exist … a timeless and omniscient God could be fully acquainted with the whole branching array of events” (p. 269). The Anselmian need not entertain this view, since, as I will try to argue, the isotemporal solution does the job. But an alternative branching hypothesis is worth mentioning here, since some metaphysicians seem to take it seriously. Rhoda’s mention of a branching universe might bring it to mind, and the Anselmian must reject it utterly. The thought is that, at points in the universe when alternatives can be realized, such as is the case with libertarian free choice, all of the alternatives are actually realized. In isotemporalism the picture of this universe would be an actually existing “block” of branches, where all of the branches actually and equally exist and contain all of the actually and equally existing moments of time. So when our exemplary agent S is in t1, it is causally and metaphysically possible for him to choose A and causally and metaphysically possible for him to choose B. At t2 the universe branches into a universe in which S chooses A and a universe in which S chooses B. But then it is metaphysically necessary, due to the branching nature of the universe, that a branch at which S chooses A exist and a branch at which S chooses B exist. This renders S’s “choice” at each branch metaphysically necessary, so it is not free in the Anselmian theory. And remember that A and B, in order to provide the sort of options in which the Anselmian is interested, must have moral significance. One option must be morally better and one must be morally worse. In this branching hypothesis, S must choose both the better and the worse option. But then the universe is morally absurd and the whole point of insisting that created agents have free choice is lost. The Anselmian universe is definitely a linear block. I do not care for the term “block” since I think it conjures up a picture of something static and unchanging. Of course the Anselmian believes in change. Change happens when some x is one way at t1 and another at t2. Of course the whole “block” does not change. That is as incoherent a thought as suggesting that the entirety of space should be able to move from where it is. Another difficulty with the term “block” is that it seems to suggest finitude. Although nothing is really riding on it, the Anselmian, relying both on revelation and physics, is happy to suppose that our spacetime continuum has a temporal beginning; a t1 in the most absolute sense. But the Anselmian

280

DEBATES IN THE METAPHYSICS OF TIME

is committed to an old-fashioned, bodily afterlife. And bodies seem to exist across space and time. So if an everlasting afterlife is embodied, then time is everlasting. I admit to having no clue as to how time and space will work “on the other side,” but I suppose (and this is the merest supposition) that time will be not wholly unlike, or discontinuous with, time as we know it. Still, it will go on forever, which doesn’t sound very “blockish” to me. But I can live with the term “block” so long as it refers to an isotemporal universe within which all the ordinary change occurs and which is everlasting. Divine eternity, then, consists in God’s immediately sustaining in being the whole, unending, linear block, while he himself is not located at, or circumscribed by, any point or points in time and space. To put it another way, he is wholly present to every point in all of time and space, and every point in all of time and space is immediately present to him. And now the Anselmian solution to the dilemma of freedom and divine foreknowledge should be obvious. God knows that S chooses B at t2 in the linear block because t2 is immediately present to him. He “sees” S choose B at t2 eternally, that is, as wholly present to all of time. The scare quotes around “sees” should be taken seriously. “Sees” seems the best way to express the view, but it should not be taken to suggest that God’s knowledge is “quasi-perspectival.” God is not a perceiver in anything like the way we are. God, in his direct and causal knowledge, is causing everything that has any ontological status in S’s choice for B. It is just that he is not causing S’s choice for B. It is S alone that is causing S’s choice for B. So the causal “arrow” representing the relationship between God’s knowledge that S chooses B at t2 and S’s choosing B at t2 runs from the latter to the former. If some temporal believer, existing at t1, should believe that God knows that S will choose B at t2, then what the believer believes at t1 is true. And so it is acceptable to say that, “At t1 God knows that S will choose B at t2,” so long as this locution is not taken to imply the objectivity of past, present, and future, or to mean that God is a temporal being who exists at t1 and not wholly and equally at all the other times there are. Anselm grants that divine foreknowledge does entail a sort of necessity regarding S’s choice for B at t2. God knows at all times, including all times before t2, that S chooses B at t2. That does entail that S cannot fail to choose B at t2. It is, to adapt some of Anselm’s terminology, consequently necessary that S choose B at t2. It is a “consequent” necessity because it “comes after” or “follows upon”—not temporally, but in the sense of depending upon—the positing of S choosing B at t2. But this consequent necessity is entirely innocuous as regards free will. It is neither causal necessity nor metaphysical necessity. It is the necessity that accrues to any actually occurring event. If in fact x happens, then x cannot fail to happen. This is a logical point, not a causal or a metaphysical point. In an isotemporal universe, propositions about



Defending the Isotemporalist Solution

281

what a free created agent chooses at any given time are true at every time, even bracketing divine knowledge. And they have a consequent necessity. But the truth of these propositions, as per the grounding principle, depends upon that created free agent making that choice at the time he makes it. So the existence of these true propositions does not conflict with the aseity of the choices. Similarly, God’s “fore”knowledge does not causally or metaphysically necessitate S’s choice for B at t2, since it is S’s choice for B at t2 which originates God’s “fore”knowledge. (“Fore” is in scare quotes because God knows all times, and the events they contain, in one act of knowing. There is no true before, during, or after for God. Or perhaps we could say it is all during for God.) True, if S chooses B at t2, then S cannot fail to choose B at t2, but since whether or not S chooses B at t2 is entirely up to S, the consequent necessity involved in divine “fore”knowledge does not conflict with the aseity of the choice. Rhoda takes this solution to be a species of the “preventable future” approach; the future is not causally closed and it is the actual event in the future on which the foreknowledge depends. And in this he is correct. I believe the term “preventable future” is misleading, though. In the Anselmian theory, the future is no more “preventable” than is the past or the present. What is future from the vantage point of some moment in time is “ontically settled” at that moment, to use Rhoda’s terminology, in exactly the same way as what is, from the vantage point of that moment, past and present. The term “preventable” suggests that some possible future exists in some way, and then is supplanted or superseded. In isotemporalism, while there are, from any given vantage point in time, many causally or metaphysically possible futures, there is only one actual future and it exists. There are many causally or metaphysically possible pasts and presents as well, but only one actual past and one actual present. Indeed, the year 1500, which is future to Anselm in 1100, is just as “fixed” in 1100 as it is in 2014. Actors before 1500 might be able to bring about what happens in 1500 in a way which actors after 1500 cannot (barring time travel), but no one can “change the course of the future.” This is true even if presentism is the case, since in presentism there is no “course of the future” to be changed. And it is true in isotemporalism since what happens, happens. If S chooses B at t2, then S chooses B at t2. The course of the future can be brought about, as S brings it about that S chooses B at t2, but it cannot be “prevented” as we would normally think of prevention.11 Thus I prefer to stick with the label “Anselmian” for the solution I propose. And so to Rhoda’s criticisms of the Anselmian solution. Rhoda first attacks the thought that God is timeless. He argues that “most theists” believe that God makes choices. But a choice must be a temporal event. There must be a time at which God is undecided about what to do, and then a subsequent

282

DEBATES IN THE METAPHYSICS OF TIME

time at which God has made his decision. The Anselmian, in accord with most of the great medieval philosophers, responds that God does not make choices under this description. God’s act of will is a single, immutable, and eternal act. Divine timelessness is such a crucial great-making property that there would have to be very, very strong reasons to insist that God makes temporal choices, and hence is a temporal being. It is not clear, though, what theological advantage is gained by supposing that God deliberates and then chooses. The image of God not quite knowing what to do is surely not biblical, nor does it accord with the tradition of classical theism. Within that tradition there are roughly two basic theses regarding whether or not God confronts open options concerning what to do, including and especially what world to create. Anselm holds that God inevitably does the best.12 In that case there is no need to deliberate. God sees the best and does it in a single, immutable, and eternal act. If we ask, “Why did God create this world?” the answer is that he knew that it was the best. But being omniscient, he does not need to deliberate about it. The alternative, championed by Thomas Aquinas, is that God has open options. Thomas makes the case in discussing the creation of the world, but I take it that a similar point applies to other divine actions—all within his single, immutable, and eternal act. Regarding creation, there is a set of possible, well-ordered, worlds that God might create, or he might not create at all. As it happens, he “opts” for one of the set—our actual world. (I include the scare quotes to indicate that this “opting” is not the temporal choosing Rhoda has in mind.) If we ask, “Why did God create this world, rather than a different world, or no world at all?” the answer is that there is no reason. Whatever reason we might propose for God having made our world—love, for example—would have equally been the reason for creating some other world, or for just sticking with the internal, dynamic relationship of the Trinity.13 So, again, there is no reason to deliberate, since God does not “opt” for one world over another on the basis of some reason. He does not need to figure out what he ought to do. Though there are options, in Thomas’s account, in the sense that there was no reason why God “must” create our world rather than do something else, there is no condition of God’s being undecided. God, in a single, immutable, and eternal act creates our world and does all he does. So, whether he does what he does because it is the best, or he does what he does, just because, he does not need to weigh his options. Having to go in for deliberation is a sign of radical imperfection.14 Only a very diminutive god would engage in making choices under Rhoda’s description, and it begs the question against Anselm to insist that the God under discussion is such a being. Rhoda offers a second criticism that I fear I may not be understanding. Rhoda uses the term “future specifier” to label anything that “singles out a



Defending the Isotemporalist Solution

283

unique possible future as the actual one” (p. 255). So, in the puzzle we are concerned about, again using our chosen example, God’s knowledge at t1 that S chooses B at t2 would constitute a future specifier regarding which causally and metaphysically possible future—the one where S chooses A or the one where S chooses B—is the actual future. At t1, and indeed at all times, God knows that S chooses B at t2, and so it cannot fail to be the case that S chooses B at t2. I have argued that the existence of this future specifier is not in conflict with S making an a se choice at t2, since it is S’s a se choice at t2 upon which God’s knowledge depends. Rhoda argues that this won’t work. He writes that: … future specifiers that are explanatorily dependent upon the actual occurrences of future contingent events are themselves future contingents—they obtain in some, but not all, causally possible futures. This is a crucial point. It implies that nothing can be both timeless and explanatorily dependent upon future contingents since the temporality of the latter is inherited by anything explanatorily dependent upon them. Hence, far from helping to solve the foreknowledge/future contingency problem, divine timelessness precludes a preventable future response to fatalism. I am not sure what the argument is here. Using our example, the future specifier is God’s knowledge that S chooses B at t2. And it is dependent upon S actually choosing B at t2. So far, so good. And this entails that God’s knowledge that S chooses B at t2, which obtains in some, but not all, causally possible worlds. That seems correct. If S at t1 really faces causally (and I would say, metaphysically) open options, then it is causally possible that at t2 he chooses A rather than B. Had he chosen A rather than B, then God’s knowledge at t1, and always, would have included that S chooses A at t2. But as it happens, S actually chooses B at t2. So it is not actually possible at t1 that S choose A at t2. Rhoda holds that the temporality of S choosing B at t2 is “inherited” by God’s knowledge at t1. I don’t know what this means in the context. All times exist equally, all times are immediately present to God, and God is wholly present to each and every time. We might say that God’s knowledge is temporal in that it is true, at any time, that God has the knowledge he has. But this does not mean that God’s knowledge is circumscribed by being in a specific moment in time, nor does it occur in some sort of sequence, extended across moments of time. My suspicion is that Rhoda has not appreciated the implications of isotemporalism (or four-dimensionalism as I called it in my 2008 book). He writes as if “the future” were objective, not subjective to a given, temporally limited, perceiver at a given moment in time. I take a contingent event to be an event, such as S’s choice for B at t2, that is

284

DEBATES IN THE METAPHYSICS OF TIME

not causally or metaphysically determined (necessitated). In isotemporalism, what might be future contingents when viewed from a certain moment in time have the same status as present and past contingents when considered objectively. If we take S’s a se choice for B at t2 to be a contingent event, by my definition, then it is a contingent event whether t2 is being considered from t1, and so is future, or from t2, and so is present, or from t3, and so is past. If by a contingent event we mean one that actually might or might not happen, consistently with all that actually happens at all the moments in the isotemporal universe, then there are no contingent events in Anselm’s universe. Actuality is all there, so to speak, in the isotemporal universe, and actuality cannot be other than it actually is. The thought that only future events can be contingent, and that they can be objectively contingent in a way that past and present events cannot be, since the past and the present are “fixed” in a way that the future is not, assumes a universe which is not isotemporal. Rhoda seems to grant the objectivity of the future. He writes that “Future contingency seems to be an inherently tensed status because there invariably comes a time when the event in question is either no longer future, or no longer contingent. The proverbial sea battle tomorrow may now be a future contingent, but it won’t be one after tomorrow” (p. 268). Here Rhoda seems to be assuming that past, present, and future are objective, rather than subjective, and so he begs the question against the isotemporalist. From the perspective of the limited, temporal, human knower, the distinction between past, present, and future is surely relevant. Quoad nos, our past and present are “fixed” in a way that our future is not. But this is an epistemic point, not a metaphysical point about what there actually is. I fear that my own explanation of the Anselmian solution may be, in part, to blame for Rhoda’s misconception. He notes that I write that, regarding God’s knowledge of what created agents freely choose, God “learns,” and must “wait and see,” and cannot know “until” the agent chooses. But, says Rhoda, if God learns and waits, then there must be a time when God does not know what the created agent chooses, and then a subsequent time when he does know. So God cannot be timeless. And this would support Rhoda’s assumption (if he is indeed assuming it) that the future is different from the present and the past in some way that is relevant to the dilemma of freedom and foreknowledge. And if God is not timeless and has access only to the present moment, then isotemporalism is false. It is only the present that exists. My language, then, was perhaps infelicitous. I had intended only to insist upon the claim (shocking to those who embrace an Augustinian or a Thomistic approach to the issue) that the causal arrow regarding God’s knowledge of the a se choice of the created agent must run from the choice to God and not



Defending the Isotemporalist Solution

285

vice versa. In the Anselmian account there is no time at which God does not know that S chooses B at t2. T2 and all times are immediately present to God. At all times at which the temporal universe exists, it is true to say that God knows that S chooses B at t2, and there is no time at which this statement is false. Rhoda’s third criticism underscores the thought that he has not fully grasped the isotemporal aspect of the Anselmian solution. (Or perhaps he just finds it incredible. But in that case he needs to mount an argument against it. As it is, he seems to assume that isotemporalism and its implications are false.) Rhoda writes that “The future is ontically settled just in case a unique, complete, and linear sequence of future events exists” (p. 267). Yes, the Anselmian says that. For any time you might pick—t1, let’s say—there is a subjective future, t2, t3, and so on. And in an isotemporal universe, t2, t3 and all that they contain exist in exactly the same way that t1 exists. If you pick t2 as your vantage point, then t1 is past and t3 is future, but, again, in fact, all exist in the linear block.15 Rhoda goes on: “But if a unique, complete, and linear sequence of future events exists, it cannot exist now (because the events are future), and so presumably it must exist tenselessly or sub specie aeternitatis” (p. 267). If what Rhoda means by this is that time is isotemporal, past, present, and future are subjective, and objectively—from the God’s-eye point of view—each moment has the same ontological status as each other, that is indeed the Anselmian understanding. Rhoda seems not to mean this since he goes on to argue that “reality must, so to speak, ‘wait and see’ to find out” (p. 268) which future contingent events actually occur. It is in this discussion that he notes how the future sea battle which was a contingent event becomes no longer contingent once it has happened. The implication is that he takes the future to be objectively open and the past and present objectively fixed, begging the question against the isotemporalist. What might it mean to say that the future exists tenselessly or sub specie aeternitatis? Many philosophers, writing against solutions to the dilemma of freedom and divine foreknowledge that appeal to divine eternity, seem to suggest that those of us who hold to God’s timelessness understand future events to exist in two ways: one at the time they occur and the other “in eternity.” And then these critics go on to argue, as does Rhoda, that positing a future event existing as fixed in eternity, on which divine knowledge can be based, undermines the contingency of the actual event in the actual future. And some proposed solutions do invite this analysis. As I interpret Boethius, he holds that there is a sense in which future events could be said to exist twice. Boethius assumes presentism and holds that God knows the future as if it were present, and so he knows what will happen when the future comes to be the present. How can God know a presently non-existent future? He

286

DEBATES IN THE METAPHYSICS OF TIME

can know because he is the absolute cause of all that has happened, is happening, and will happen. This includes human free choices. Boethius is what can be termed a theist compatibilist: you choose freely even if God causes your choice. And God knows in one, simple, immutable act all that he has caused, is causing, and will cause.16 So S’s choice for B at t2 (it is not, in this Boethian understanding, an a se choice) could be said to exist in two ways. When t2 comes to be present, it actually exists in the present moment. But it has always existed “in eternity” as part of the divine plan, as the object of God’s knowledge. Of course, it is not S’s actual choice at t2 that exists “in eternity” in this view. But in a perhaps too loose and analogical way one might say that S’s (Boethian) choice exists in eternity. But if it is not S’s actual choice, but a facet of the divine plan, which exists in eternity and grounds God’s knowledge, then S’s choice is necessitated by something outside the choice itself. If Rhoda’s understanding of future events existing tenselessly and sub specie aeternitatis is rather like what I attribute to Boethius, then Rhoda is right that this appeal to divine timelessness does not solve the dilemma. But, of course, the Anselmian solution does not suggest that “a unique, complete, and linear sequence of future events” exists in a way different from the unique, complete, and linear sequence of past and present events. The universe is a unique, complete, linear sequence of all events at all times. Perhaps there is a beginning moment, and perhaps the sequence stretches infinitely into what is to us the future. (A “complete” infinity seems an odd concept, but it’s a weird old world.) Each point in time, whether past or present or future to some perceiver at some point in time, exists and exists equally with all the other points. So does the future exist now? Well, that depends on what you mean. Say we are having our discussion across an extended stretch of time which we can label t1. And suppose, from our vantage where t1 is our present, t2 is some future time. Does t2 exist at t1? Yes and no. Yes, if we mean that it is true to say at t1, “T2 exists.” It is true at t1, and at every point in time, that the whole linear block universe exists and t2 is in it. No, if we mean to say that t2 occurs at the same time as t1, that t1 and t2 overlap temporally. They don’t. T1 is at t1, and t2 is at t2. If this is puzzling, take the spatial analogy. Every point in space exists equally. We live—as no one seems to doubt—in an isospatial universe.17 In the Anselmian account, God is wholly present to every point in space, and every point in space is immediately present to God. It does not follow that each point in space exists twice, once at its own point in space, and once, non-spatially, in divine ubiquity. No, each point exists once, at its own spatial location. Suppose I am in Newark, Delaware, and I ask—analogously to our question about “the future”—does Delhi exist here? Yes and no. No, if I mean that Delhi occupies the same spatial location as



Defending the Isotemporalist Solution

287

Newark. It is not that Newark and Delhi spatially overlap. But yes, if I mean to say that Delhi does indeed exist somewhere in the spatial universe. One way of justifying the claim that Delhi exists, even as uttered in Newark, Delaware, is that you can get to Delhi from Newark. Drive up to the (very different!) Newark, New Jersey, take the daily 8.30 p.m. flight, fly for roughly 14 hours, and get off in Delhi. And once you’re there, Delhi becomes here to you. And so with isotemporalism. As all the time travel stories posit, if you had the right means of conveyance, you could travel from what you perceive as present to what is (subjectively) past or future. And when you get there, what used to be (in your personal time line) past or future becomes your present. That is what it means to say that all the moments of time exist equally.18 But if the whole, unique, complete, linear block exists, and all events are “fixed” such that it is the case, considered from any point of time, that the events that happen at each and every point of time cannot fail to happen, then could it be that that alone is sufficient to undermine free choice? No. Suppose S makes an a se choice for B over A at t2. An a se choice is one absolutely caused by the agent. It is neither causally nor metaphysically necessitated. In the Anselmian account it is true that S’s making the a se choice for B over A at t2 is “fixed.” There is no time at which it is not the case that S makes the a se choice for B over A at t2, and so, by consequent necessity, S cannot fail to make the choice. But it is S, the agent himself, who fixed it by choosing! If this consequent necessity is among those species of necessity that conflict with freedom, then no one is ever free. Consequent necessity holds for every posited event. If A happens, then A happens. Whenever anyone at any time chooses x, they render it the case that they cannot fail to choose x. If, to choose freely, you must be able to choose in the absence of consequent necessity, then you must be able actually to choose other than you actually choose. But that is logically impossible. Surely you do not need to be able to do the logically impossible to be free! The fixity entailed by the linear block universe is just consequent necessity. It cannot and does not conflict with the sort of freedom necessary to ground moral responsibility. Contra Rhoda, the Anselmian solution to the dilemma of freedom and divine foreknowledge stands, and we do not need to take the drastic step of abandoning either robust human freedom or divine foreknowledge.

Notes   1 Rogers (2008).   2 In Rogers (2008: 148–52), I noted a number of difficulties with Molinism, but I did not mention the claim made here that Molinism entails the

288

DEBATES IN THE METAPHYSICS OF TIME metaphysical necessity of a choice, and hence conflicts with a se choice. This is a thesis that I am currently attempting to develop.

  3 To speak more correctly, God engages in one, immutable, simple, and perfect act of will. But quoad nos there look to be multiple acts, and it is legitimate, and sometimes helpful, to speak as if there are, so long as one understands that in fact there are not.   4 In the Proslogion, see especially chapters 13 and 18–21.   5 Rogers (2008: 117–21), Rogers (2012).   6 Anselm’s clearest statement of this thought is in De Concordia, bk. 1, ch. 5. See Rogers (2008: 176–84).   7 Thanks to Catherine and Michael Tkacz for the term.   8 I do not insist that time is divided up into extensionless moments. Indeed time seems to me more like a continuum. But it is easier to talk about “moments.” Just allow that these “moments” may simply be locations on the continuum.   9 I attempt to mitigate the bizarreness a bit in Rogers (2007: 13–15). 10 I do not have any well-developed thoughts on how to analyze or explain the arrow of time, but that is a separate issue. 11 This opens an interesting can of worms about what we normally mean by “prevention,” but that need not detain us here. Likely we mean something about what sort of effects were likely to result from certain causes had we not stepped in, etc. 12 Rogers (2008: 185–205). 13 Summa Theologiae I, q. 25, a.5. 14 Even Alghazali, who places God in time and has him making choices of a sort, does not suppose that God must deliberate about what to do. 15 The temporal perceiver (barring time travel) doesn’t get to pick his vantage point in that there seems to him to be an absolute now, but we are doing metaphysics and need not be bound by temporal seemings. 16 Rogers (2008: 111, 156). This seems to me to be ultimately incoherent. How, if God is absolutely immutable, can he be causing x today and then y tomorrow, as presentism would have it? Charity would, I think, suggest that Boethius is an isotemporalist, in that that is the best way to address the question at hand: “How can an immutable God know and act in a temporally extended universe, and foreknow free choices?” But the texts, I believe, support the interpretation that Boethius was not proposing isotemporalism. 17 This may be a little fast. Space is a peculiar phenomenon. But I think the point can be made without worrying about the nature of space and spatial extension. 18 An alternative analogy is the (old fashioned) film. The frames all exist equally, even though one frame is earlier in the movie and one is later. At the earlier frame one might say that the later one exists, when considering the whole film, but that would not mean that the later frame exists twice, and it would not mean that the later frame overlaps with the earlier frame. I suppose the information on disks works the same way—there is a sequence recorded on



Defending the Isotemporalist Solution

289

the disk corresponding to the sequence in the movie as viewed—but film is easier to visualize.

References Anselm of Canterbury (1998), The Major Works. Oxford: Oxford University Press. Aquinas, T. (2008), Summa Theologiae. http://www.newadvent.org/summa/ Rogers, K. A. (2007), “Anselmian eternalism,” Faith and Philosophy, 24, 3–27. —(2012), “Anselm on the ontological status of choice,” International Philosophical Quarterly, 52, 183–97.

Index 4-D view 11, 82, 83, 263, 278, 283 absolute becoming 10, 64, 88, 101, 102 absolutist view of time 73, 74, 75, 76, 81, 82 A-eternalist 53–6, 59 Anselm of Canterbury 275–82, 284 A-predicates 109–11, 113, 118–21 A-properties 21, 52, 57, 58 Aristotle 75, 76, 78, 80, 219, 257 atemporal becoming 11, 87, 88, 93, 95, 102, 103, 104, 105, 236 see also temporal becoming atemporal entity 93, 94, 95, 102, 103, 104 atemporality 92, 93, 94, 95, 233, 268 A-theory 35, 44, 47, 52, 53, 55, 60, 76, 99, 261 atomism 150–5, 157, 159, 160, 164, 167, 169–71, 178–9, 180, 183, 184, 196 Augustine 56, 62, 224 Augustinian argument 51, 58, 61, 62, 264 B-eternalist 53–6, 59 see also eternalism Big Bang cosmology 219, 220, 223, 234 block universe, 12, 47, 91, 202, 254, 261, 262, 264, 279, 280, 286, 287 Boethius 216, 248, 254, 285, 286 B/R theory 25, 27, 35, 36, 48 see also B–theory, tenseless theory of time branching hypothesis 261, 263, 269, 279 B-relations 52, 110, 112

Broad, C. D. 4–10, 13, 20, 23, 33, 36, 37, 43, 78, 83, 101 Broad/Russell theory see B/R theory Brunner, Emil 226, 227 B-theory 31, 32, 33–6, 34, 39, 43, 46, 47, 52, 54, 55, 60, 76, 91, 99, 109, 119, 120, 261 see also B/R theory, tenseless theory of time Castañeda, Hector-Neri 7 Chuard, P. 150, 177 counterfactual 134, 258, 260, 276 Craig, W. L., 122 critical philosophy 4, 5 Dainton, Barry 11–13, 35, 37, 40, 41, 43, 135, 136, 144, 145, 146, 147, 149, 151–3, 159–67 divine eternity 221, 225, 280, 285 divine foreknowledge, 253, 254, 258, 263, 264, 269, 275, 277, 279, 280, 285, 287 Dolev, Yuval 3–25, 51–65 Dummett, Michael 53, 55, 119, 120 durationless events 11, 51, 52 instant 56, 58, 61, 62, 63, 65 present 52 Dyke, Heather 112 Einstein, Albert 141, 198, 199, 204, 206 epistemically settled future 255, 260–4, 269 eternalism 51, 52, 54–6, 62, 64–5, 76–7, 244, 261, 263, 264, 269 see also B-eternalist experiential flow 162, 202, 203 explanatory dependence 254, 260, 264, 267

292 Index

fatalism 253, 254–60, 263–8, 279 four-dimensionalism see 4-D view future contingents, 253–5, 258–64, 260, 266–9, 283–5 future specifier 254–60, 265–9, 282, 283 Hurley, Susan 148 instantaneous events 152, 217, 219–21, 237–42 velocity 216, 217, 241, 243 introspectibility premise 156–9 isotemporalism 278, 279, 281, 283–5, 287 James, William 160 law of experience 142, 144, 145, 198, 201 Lee, Geoffrey 178, 180–5, 189, 191–3, 195, 196 Leftow, Brian 215–17, 219–22, 228 Lloyd, Seth 204 Lombard, L. 215 McCall, S. 80, 113, 114, 115 McTaggart, J. M. E. 6, 76–82, 100, 105, 111–13, 119, 122 Mellor, D. H. 18, 32, 53, 97, 116, 117, 119 mirroring view 151, 152 multiple presentations argument 155, 181 Newtonian physics 63, 137, 143, 198, 199 now 20, 21, 75, 77 moving now 3, 25, 32, 33, 53, 55, 59, 76, 168 Oaklander, L. Nathan 32–6, 39, 46, 47, 55 Ockham, W. 257 ontological facts 7, 9, 10, 16, 17, 18, 23 ontology 4, 6, 7, 8, 10, 14, 15, 17, 32, 33, 34, 39, 47, 54, 55, 57, 58, 61, 65, 73, 78, 81, 82, 84, 122, 123, 125, 218, 225

temporal 254, 260, 268 open future 61, 254, 257, 268 overlap model 40, 43, 161, 165, 166 Parmenides 79, 80 passage 5, 10–12, 31–3, 35, 37, 39, 42, 47, 48, 75, 77, 79, 83, 96, 99, 100, 109, 118–23, 168, 202 A-theoretic 12, 13 experiential 12, 202 temporal 12, 13, 83, 109, 118–22, 202 Paul, Laurie 11–13, 35, 96–8 Pelczar, Michael 178, 180, 197–205, 207–8 perdurantism 11–13, 35, 96–8 Phillips, L. 150, 152, 155–7, 159, 169, 171 physicalism 142, 143, 156, 198, 201 Presence Principle 139–41 presentism 51–4, 56, 60–5, 75, 76, 78, 83, 121–3, 241, 278, 281, 285 moderate 54, 63–5 presentness 10, 18–21, 33, 35, 38, 45–7, 52, 53, 57, 60–2, 74, 80, 99, 100 preventable future 254, 257–60, 264–8, 281, 283 Price, Huw 120 Principle of Pickwickian Senses 5, 6, 7, 8, 13, 17, 23, 33, 34, 39, 47, 48 Principle of Simultaneous Awareness (PSA) 135, 136, 146 Prior, Arthur 121, 122 process view 151–4, 156 reductionist 11, 139, 140, 141, 222 relational view (of predication) 113–19, 121–3 relativistic argument 137, 138, 139, 198, 204–6 repeated contents 40, 135, 146, 147, 180–3, 187, 189, 190, 197 representational fallacy 112 resemblance view 151–2 Rhoda, Alan 275, 278, 279, 281–6 Rogers, Katherin 254, 263–8

Index Russell, Bertrand 5, 7, 10, 12, 20, 21, 36, 75, 77, 79–84, 136, 137, 139, 141, 142, 144, 145, 204 Russellian monism 201–4, 206, 207, 208 serial reductionism (serialism) 139, 141 settled future 254, 260–5, 267–8 Sider, T. 122, 123 simple fact 104, 105 simultaneity 8, 14, 35, 53, 60, 74, 137, 138, 155, 204–5, 207 Smart, J. J. C, 19, 96, 97, 118, 120 snapshot theorists 149, 150, 177 specious present 2, 20, 35–40, 47, 168, 178–81, 184, 185, 187, 190, 191 stream of consciousness 131–6, 145–7, 150–1, 157, 159, 161, 165, 178, 182, 184, 193, 199, 204 succession 9, 10, 13, 14, 19–21, 23, 36–8, 40–2, 73–84, 99, 137, 139, 177–9, 184–90, 193–6, 199, 225, 235 see also passage temporal 19, 73–84, 137, 139 Swinburne, R. 217, 225, 226, 228 temporal becoming 88, 95, 96, 97, 98, 101, 102, 104 see also atemporal becoming experience 11, 22, 24, 33, 89, 149–56, 166, 173, 179, 183, 197, 199, 202

293

extension 83, 179, 184, 247 intimations 193, 195 ontology 123, 254, 260, 268 parts 79, 83, 121, 122, 123, 149, 151, 153, 160, 216, 217, 238, 241, 242, 277 passage see passage, temporal phenomenology 8, 20 predication 115, 116, 117, 118, 121, 123 reality 9, 88, 90–3, 95–102, 104 relations 10–17, 24, 25, 32, 33, 46, 81, 82, 90, 151, 154, 156, 157, 165, 166, 168 tensed properties 10, 16, 21, 23, 24, 32, 41, 45 relations 15, 16, 17, 18, 19, 24, 34 sentences 9, 14, 16, 18 theory of time 12, 13 see also A-theory; presentists theistic determinism 259, 260 transitory aspect of time 8, 10, 13 transparency argument 156–9, 166, 167, 169 truthmaker 55, 59, 60, 61, 65, 104, 122 Tye, Michael 156, 167 universals 16, 24, 81 Unworld 144, 145, 198, 199 Walker, R. 176–7 Williams, Clifford 21 Williams, Donald C. 10, 82, 83, 84, 96