C. D. Broad’s Philosophy of Time 9780415998123

Citation preview

C. D. Broad’s Philosophy of Time

In this study, Oaklander’s primary aim is to critically examine C. D. Broad’s changing views of time and in so doing clarify the central disputes in the philosophy of time, explicate the various positions Broad took regarding them, and develop his own responses both to Broad and the issues debated. L. Nathan Oaklander is David M. French Professor Emeritus and Professor Emeritus of philosophy at the University of Michigan-Flint. Publications include The Ontology of Time (2004); The Philosophy of Time: Critical Studies in Philosophy, 4 vols. (Routledge, 2008); and Debates in the Metaphysics of Time (2014).

Routledge Studies in Twentieth-Century Philosophy

Bertrand Russell and the Nature of Propositions A History and Defence of the Multiple Relation Theory of Judgement Samuel Lebens Benjamin, Adorno, and the Experience of Literature Edited by Corey McCall and Nathan Ross Heidegger on Technology Edited by Aaron James Wendland, Christopher Mervin, and Christos Hadjioannou Revolution and History in Walter Benjamin A Conceptual Analysis Alison Ross Literary and Cultural Alternatives to Modernism Unsettling Presences Edited by Kostas Boyiopoulos, Anthony Patterson, and Mark Sandy Heidegger’s Concept of Philosophical Method Innovating Philosophy in the Age of Global Warming Vincent Blok Wittgenstein and the Limits of Language Edited by Hanne Appelqvist C. D. Broad’s Philosophy of Time L. Nathan Oaklander For more information about this series, please visit: www.routledge. com/Routledge-Studies-in-Twentieth-Century-Philosophy/book-series/ SE0438

C. D. Broad’s Philosophy of Time L. Nathan Oaklander

First published 2020 by Routledge 52 Vanderbilt Avenue, New York, NY 10017 and by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Routledge is an imprint of the Taylor & Francis Group, an informa business © 2020 Taylor & Francis The right of L. Nathan Oaklander to be identified as author of this work has been asserted by him in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging-in-Publication Data Names: Oaklander, L. Nathan, 1945– author. Title: C. D. Broad’s philosophy of time / L. Nathan Oaklander. Description: New York and London : Routledge, 2020. | Series: Routledge studies in twentieth-century philosophy ; 49 | Includes bibliographical references and index. Identifiers: LCCN 2020006339 (print) | LCCN 2020006340 (ebook) | ISBN 9780415998123 (hardback) | ISBN 9781315772806 (ebook) Subjects: LCSH: Time. | Broad, C. D. (Charlie Dunbar), 1887–1971. Classification: LCC BD638 .O15 2020 (print) | LCC BD638 (ebook) | DDC 115.092—dc23 LC record available at https://lccn.loc.gov/2020006339 LC ebook record available at https://lccn.loc.gov/2020006340 ISBN: 978-0-415-99812-3 (hbk) ISBN: 978-1-315-77280-6 (ebk) Typeset in Sabon by Apex CoVantage, LLC

To Silvano Miracchi

Contents

ContentsContents

Acknowledgementsviii Prefaceix 1 Problems in the Ontology of Time

1

2 The Early Years: The Russellian Theory of Time

25

3 The Middle Period: The Growing Block Theory of Time

54

4 The Later Years: The Full-Future Theory, Presentism and McTaggart’s Paradox

86

5 Independent Account of McTaggart’s Paradox and the R-theory of Time

119

6 The Self and Time

145

7 The Philosophical Implications of Foreknowledge: Precognition, Fatalism and Time

164

8 Conclusion: Broad, the R-theory and Time

184

Appendix: Is There a Difference Between Absolute and Relative Space?207 Index215

Acknowledgements

AcknowledgementsAcknowledgements

I wish to thank Taylor and Francis for permission to reprint excerpts from C. D. Broad, “The General Problem of Time and Change,” in Scientific Thought (Routledge & Kegan Paul, 1923), pp. 53–84; and Cambridge University Press for permission to reprint excerpts from C. D. Broad, “Ostensible Temporality,” in Examination of McTaggart’s Philosophy, vol 2, pt. 1 (Cambridge University Press), pp. 264–317. I also wish to acknowledge use of material from L. N. Oaklander, “Ingthorsson, McTaggart’s Paradox and the R-Theory of Time,” in Patrick Blackburn, Per Hasle & Peter Øhrstrøm (eds.), Logic and Philosophy of Time: Further Themes from Prior, Volume 2 (Aalborg: Aalborg University Press, 2019), pp. 73–104; and from L. N. Oaklander “Is There a Difference Between Absolute and Relative Space,” in Guido Bonino, and Rosaria Egidi (eds.), Fostering the Ontological Turn: Gustav Bergmann (1906– 1987) (Frankfurt: Ontos Verlag, 2008), pp. 242–252. I am exceedingly appreciative of the careful reading of three drafts of the entire manuscript by Silvano Miracchi, to whom this book is dedicated. His invaluable corrections to numerous stylistic and substantive matters captured infelicities that would otherwise have gone unnoticed. I have also benefited from Erwin Tegtmeier, with whom I frequently correspond on topics in the ontology of time, and whose writings on time are enlightening. Finally, I would like to acknowledge and thank my wife, Linda Galang Oaklander. Her patience and support have been a source of strength in completing this project. She has prepared the references for each chapter and the index with an unfailing skill for which I am grateful. Her background in philosophy and information science has made working with her on this project a delight.

Preface

PrefacePreface

My interest in C. D. Broad’s philosophy of time began in the fall of 1968 when I took a graduate seminar on the philosophy of time at the University of Iowa with Laird Addis. The text for the course was Richard Gale’s anthology, The Philosophy of Time (1967), and I was assigned a report on Broad’s famous chapter “Ostensible Temporality” in his Examination of McTaggart’s Philosophy (1938). The following year I took a seminar from Gustav Bergmann, who was my mentor, on Broad’s The Mind and Its Place in Nature (1925). These courses, taken over 50 years ago, began my abiding interest in Broad and the philosophy of time. This book is the culmination of those concerns. I think it is fair to say that C. D. Broad’s various writings on time, along with McTaggart’s “The Unreality of Time” (1908) and “Time,” in The Nature of Existence (1927), and Russell’s “On the Experience of Time” (1915) were all extremely influential and their authors are the most important philosophers of time in the first half of the twentieth century. However, despite Broad’s importance in the philosophy of time and the recent resurgence of interest in his work (see Thomas (2019)), there has yet to be a book devoted to his philosophy of time. The primary aim of this work is to fill that lacuna, and secondarily to defend what I have come to call “the R-theory of time,” because it has its roots in Russell’s logical atomism and his philosophy of time. There is a close connection between these two goals. Gale has claimed that “The father of the modern version of the B-theory is Bertrand Russell” (1968, p. 16). According to the B-theory, the “B-series” (so-called by McTaggart) of events ordered by the B-relations of earlier than or later than “alone is sufficient to account for time” (Gale, 1968, p. 16). In Broad’s first essay on “Time,” for the Encyclopedia of Religion and Ethics (1921), he was a staunch defender of Russell. Broad acknowledges that when he says, “At the period when I wrote the Encyclopaedia article . . . I was almost completely under the influence of Bertrand Russell in his extreme realist phase, and of Meinong as I understood him” (Broad, 1959, p. 765). Nevertheless, some of Broad’s infelicitous claims in that article have been taken over by critics and supporters

x  Preface of the B-theory, to the detriment of the R-theory. Diagnosing Broad’s mistakes is instructive in clarifying Russell’s authentic view in contrast to the B-theory, which is McTaggart’s misrepresentation of Russell, but has nevertheless been accepted generally as Russellian. As the title makes clear, the central focus of this book is Broad’s philosophy of time, but the importance of understanding McTaggart’s and Russell’s views on time cannot be underestimated in any attempt to understand Broad. The importance of McTaggart in Broad’s changing philosophies of time is explicit in the following passage: At the back of all of them is McTaggart’s paper “The Unreality of Time” published in Mind in 1908. I felt from the first and still feel, that the difficulty which he raises is (a) embarrassing enough prima facie to demand the serious attention of anyone who philosophises about time; but (b) almost certainly due to some purely linguistic source (common and perhaps peculiar, to the Indo-European verbsystem), which it ought to be possible to indicate and make harmless. (1959, p. 765) The importance of McTaggart’s positive and negative views of time will be evident throughout this book. In Chapter 1, I explain the central problems in the ontology of time as Broad understands them. Broad’s responses to McTaggart’s paradox will be a central theme in Chapters 2–4, where I critically examine his changing theories of time. Then, in Chapter 5, I provide an independent discussion of McTaggart’s argument for the unreality of time that includes a defense of the R-theory and a critique of R. D. Ingthorsson’s McTaggart’s Paradox (2016). There are several reasons why studying Broad’s varied approaches to the philosophy of time is especially relevant to the current state of the subject. First, by studying Broad’s changing views of time, we can better get a historical perspective on the contemporary issues currently being debated and various theories of time regarding them. Each of his theories of time have been discussed and subsequently developed in one form or another, and they are still prevalent today. Examining Broad’s changing philosophy of time can not only help to clarify and see the historical development of current philosophical theories of time, but can foster a critical appraisal of the strengths and weaknesses of the views put forth. Second, Broad realized, as many philosophers now do, the importance of time to other areas of philosophy. It is for that reason that the connection between Broad’s views on time and the self is discussed in Chapter 6, and in Chapter 7, time and memory, foreknowledge and fatalism are explored. Since Broad’s discussion of the unity of the mind in Chapter 6 contains an important discussion of different versions of absolute and

Preface  xi relative space and time, I explore that topic in the Appendix: “Is there a Difference Between Absolute and Relative Space?” In Chapter 8, the Conclusion, I review Broad’s early view and explain how it may have led to his later objections to the Russellian theory, most notably that “the [R-] theory leaves altogether out of account the transitory aspect of time” (1938, p. 61). The final chapter features an extended response to that objection. The previous monographs I wrote, Temporal Relations and Temporal Becoming: A Defense of a Russellian Theory of Time (1984) and The Ontology of Time (2004), have intended to be a vigorous defense of the B-theory. Readers who are familiar with my earlier work, or even those who are not, may wonder how the R-theory differs from the B-theory and why have I changed my view. The answer is that in many ways I have not radically changed my thinking on time. Indeed, I have not essentially changed my view at all, but am using the new name “R-theory” as a device to call attention to ways in which the Russellian theory of time has been misunderstood or misinterpreted by critics and defenders alike. My intention is to clarify the R-theory in contrast to McTaggart’s misinterpretation of it, with which it is often identified. Whether the R-theory is a different and better view of time than the B-theory, or a preferable or correct version of the B-theory, is not important. What is important is the adequacy of the R-theory and that, of course, depends on the soundness of the arguments put forth in defense of it. With regard to my understanding of Broad’s changing philosophy of time, and the insights his writings on the subject contain, I hope to have done them justice. With regards to my own interpretation of the Russellian theory of time, I can only echo Broad’s sentiments when he says with typical modesty, at the end of Chapter 2 in Scientific Thought: I can hardly hope that what I have been saying about Time and Change will satisfy most of my readers, or indeed, that it is more than a shadow of the truth, if that. It is admitted that this is the hardest knot in the whole of philosophy. The Dean of Carlisle judiciously remarks that “We cannot understand Time, but we shall not understand it better by talking nonsense about it.” In the hope that I have not darkened counsel by words without understanding, I leave this most difficult subject to return at a later stage. (1923, p. 82) L. Nathan Oaklander Ann Arbor, MI USA January 2020

xii  Preface

References Broad, C. D. (1921). Time. In J. Hastings et al. (Eds.), Encyclopedia of religion and ethics (Vol. 12, pp. 334–345). New York: Scribner’s. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 143–173). ———. (1923). Scientific thought (Chap. 2). London: Routledge & Kegan Paul. References are to this reprint. L. N. Oaklander (Ed.). (2008). Vol. II. Time and metaphysics (pp. 63–83). ———. (1938). Examination of McTaggart’s philosophy: ostensible temporality (Vol. II (1), Chap. XXXV, pp. 265–323). Cambridge: Cambridge University Press. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 36–68). References are to this reprint. ———. (1959). A reply to my critics. In P. A. Schilpp (Ed.), Philosophy of C. D. Broad (pp. 709–832). New York: Tudor Publishing Company. Gale, R. (1967). The philosophy of time. New York: Anchor Books. ———. (1968). The language of time. London: Routledge & Kegan Paul. Ingthorsson, R. D. (2016). McTaggart’s paradox. London: Routledge. McTaggart, J. M. E. (1908, Oct.). The unreality of time. Mind, 17 (68), 457–474. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 21–35). ———. (1927). The nature of existence (Vol. II, C. D. Broad, Ed.). Cambridge: Cambridge University Press. Oaklander, L. N. (1984). Temporal relations and temporal becoming: a defense of a Russellian theory of time. Lanham, MD: University Press of America. ———. (2004). The ontology of time. Amherst, NY: Prometheus Books. ———. (Ed.). (2008). The philosophy of time: critical concepts in philosophy (Vols. 1–4). New York and London: Routledge. Russell, B. (1915, Apr.). On the experience of time. Monist, 25 (2), 212–233. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 174–187). Thomas, E. (2019). The roots of C. D. Broad’s growing block theory of time. Mind, 128, 527–549.

1 Problems in the Ontology of Time Problems in the Ontology of TimeProblems in the Ontology of Time

I. Introduction My mentor, Gustav Bergmann (1906–1987), would occasionally remark that philosophers stand on each other’s shoulders and that great philosophers go back to the same fundamental issues over and over again. Clearly, over the past hundred years, philosophers interested in the philosophy of time have benefited from Broad’s careful analysis of the problems of time and have stood on his shoulders. Broad certainly benefited from contemporaries such as Bertrand Russell, John M. E. McTaggart, Alfred North Whitehead, G. F. Stout, Samuel Alexander, G. E. Moore, W. E. Johnson and other historical figures, of course. Like other great philosophers, Broad also returned with increasing originality and subtlety to the problems of time. What, then, were the problems in the philosophy of time to which Broad returned throughout his career, and how did he attempt to resolve them? In this introductory chapter, I shall discuss the first question, and in the remaining chapters, I shall explore Broad’s changing responses to the second. In “A Reply to My Critics” (1959), Broad tells us that “at the back of all [his different accounts of time] is McTaggart’s paper ‘The Unreality of Time,’ published in Mind in 1908” (1959, p. 765). Indeed, McTaggart’s views on time play a central role in many of Broad’s writings on time (see, for example, 1921, 1923, 1928, 1938). Arguably, one of Broad’s most well-known and famous writing on time is his discussion of McTaggart’s chapter on “Time” (1927) in Examination of McTaggart’s Philosophy (Broad, 1938). It will be useful, therefore, to approach Broad’s philosophy of time by a consideration of McTaggart. In this way we can better understand the fundamental problems of time to which Broad frequently returned. Since McTaggart and Broad are ontologists, it is appropriate to discuss the aims of ontology as I understand them. 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 the most general categories and general principles of classification, and then to say something about the relations between those

2  Problems in the Ontology of Time 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, fact), founded on the most widespread, pervasive or omnipresent phenomena. Surely, time is a ubiquitous phenomenon. Therefore, the ontologist of time asks what category or categories of intrinsically temporal entity or entities there are, or must be, to explain the temporal phenomena. To answer the ontological question “What is the nature of time?” is to give an inventory of all temporal entities, or rather, an inventory of all the kinds of such entities there are.1 Are there intrinsically temporal individuals or particulars; time points or moments of absolute time capable of existing unoccupied? Does time consist of relations, and if so, are those relations internal, grounded in the nature or properties of one or more of their terms, or are they external, grounded in an additional entity obtaining between its terms, or some combination of both? Are temporal relations reducible to other relations, such as causation or entropic increase, or are they irreducible? Are there intrinsically temporal (non-relational) properties, such as pastness, presentness and futurity (hereafter “A-properties”), introduced by McTaggart (1908, 1927) to distinguish events that are past, present or future? Are there the so-called “coordinate qualities,” to use Gustav Bergmann’s terminology (1967, p. 24), or “temporal positional qualities” as Broad calls them (1925, p. 592), such as being at t1 and being at t2?2 Is there a special category of temporal change such as temporal becoming that is, a change of A-properties or absolute becoming—the coming into and going out of existence of temporal objects—or is the passage or flow of time or events in time to be understood in terms of some other category already delineated? If temporal or absolute becoming is primitive and ontologically basic, then how is it to be understood? Certainly, time is a basic and fundamental phenomenon within the purview of ontological explanation. We must ask, then, what temporal phenomena there are, and to what category or categories do temporal phenomena belong? In the next section, I shall explore some answers to these last two questions and, in so doing, identify one central debate in the ontology of time.

II.  Temporal Phenomena and the Ontology of Time In the section from “Ostensible Temporality” titled an “Independent account of the phenomenology of time” (1938, p. 36), Broad distinguishes two phenomena that are at the forefront of debates in the philosophy of time: the transitory aspect of temporal facts and the extensive aspect of temporal facts. The phenomena in question roughly correspond to McTaggart’s A-series of events that runs from future to present to past, and the B-series of events that are ordered by the relations of earlier than/later than. As we shall see, McTaggart has already built into the A-series and B-series a specific ontological account of what entities

Problems in the Ontology of Time

3

constitute those series, if time is real. I shall eventually argue that his analysis is mistaken, but in this introduction, to help structure the debate in the ontology of time, I will continue to use the A-series and B-series terminology that has become standard. Later in this chapter and especially in the chapters that follow, I shall introduce and explain the Russellian view, or R-theory, and be more precise about McTaggart’s conception of the B-series and B-relations and how they differ from the R-series and R-relations. I shall also argue that while the transitory and the extensive aspects of time are nominally different and expressed differently in language, they are fundamentally grounded in entities belonging to the same ontological category. When McTaggart characterized time in terms of the A-series, he and subsequent philosophers meant that we experience time, or events in time, as moving or flowing—as having a transitory or dynamic quality. Time involves a flow or flux from one event to another, and not a “static” temporal relation between them. For Broad, transition in time also means that we experience events as moving from the distant future to the near future to the present, and then receding into the more and more distant past. He expresses this by means of an expression that A. N. Prior (1959) made famous more than 20 years later, after a visit to the dentist writing, “Thank God (on the theistic hypothesis) that is over now!” (Broad, 1938, p. 38). There is a sense in which time or events in time seems to flow or move through time and this puzzling, yet pervasive, aspect of time is reflected in ordinary thought, language and experience. In our thoughts and emotions, the transitory aspect of time is reflected in the different attitudes we have toward events depending on their temporal position relative to us. For example, it is natural to anticipate, perhaps with dread, an unpleasant future event, perceive or directly experience the unpleasant event when it is happening now or at present, and then to remember it with relief when we reflect on its passing into or becoming past. Thus, we think differently about events and experiences that appear to be moving toward us or away from us; a pleasant future event is thought of with joyful anticipation whereas the same event when it becomes past is remembered with nostalgia. The transitory aspect of time also has a foothold in the language of time. For we speak, write and talk of past, present and future events, and we use tensed language to reflect the unique form of change events undergo when they move from the future to the present and into the past. For example, it is now true to say that “I will have dinner in a few hours,” and in a few hours it will be true that “I am now eating dinner.” Still later tonight, it will be true to say that “I had dinner already.” It is natural to say, as Broad does, that these sentences accurately reflect facts about events in time that strike at something essential to its nature (1938, p. 38). More specifically, such sentences suggest that events pass through

4  Problems in the Ontology of Time time by changing their A-properties; first possessing futurity, then presentness and then pastness. Moreover, since each of these sentences (or the propositions they express), change their truth value with the passage of time it would appear, for that reason, they cannot be translated or eliminated in favor of sentences that solely express propositions about temporal relations between events whose truth values are unchanging. Our experience of time is also seemingly inseparable from time’s dynamism since the present which is always experienced as changing has a force and vivacity about it, to use Hume’s characterization of impressions, that the past and future do not have. What is present or now seems to have a greater reality than what is past or future. George Schlesinger once expressed this, when he said, “The NOW is, of course, not conceived as some sort of object but rather as the point in time at which any individual who is temporally extended is alive, real or Exists with a capital E” (Schlesinger, 1980, p. 23). Given that time’s dynamism (the whoosh, flow or passage of our experiences and events in time) and its transiency (or flow of events from future to present to past) are so deeply embedded in our thinking, language and experience of time, it would seem to be an act of folly to deny they exist. That may well be true, but it doesn’t necessarily follow that the ontological explanation requires an A-theoretic ontology (after McTaggart’s A-series) or that one abandon the B-theory (after McTaggart’s B-series). Indeed, Broad, in his earliest reflections on time, to be discussed in the next chapter, maintained that our thought, language and experience of time could be explained without positing temporal properties or temporal becoming as some A-theory advocates interpret those notions. Another temporal phenomenon that Broad calls to our attention is what he calls the “extensive” aspect of time. It consists of the fact that two experiences of the same person may stand to each other in a determinate temporal relation of earlier than/later than or simultaneous with, or they may partly or entirely overlap. In fact, he distinguishes several different relations that experiences can have to each other; for example, two experiences can be such that they start simultaneously, but the first lasts longer than the second (Broad, 1938, p. 37). We experience events occurring successively and simultaneously. This happens, for example, when someone rapidly knocks on the front door and you hear a series of sounds, one after the other; or when one hears a whistle sounding and hears the conductor yelling “all aboard” at the same time as the whistle, we experience the simultaneity of two events. We use temporal relational language to place events in time, as in “President Kennedy was assassinated almost 50 years before President Obama’s second inauguration,” or “I’ll see you later.” In reverie we may think to ourselves, “How did I get by before I met you?” Some have claimed that in contrast to the transitory aspect of time and experience, the extensive aspect is thought to be static, lacking any

Problems in the Ontology of Time  5 dynamism, whoosh or transiency. For that reason, it is argued that the transitory A-properties are needed to account for the passage of time and its dynamic aspect. Yet those who emphasize temporal relations over temporal properties as foundational in the ontology of time are not without an explanation of the phenomena in question. We experience succession when at noon the church bell rings 12 times and we can hear the first chime occurring earlier than the second, the second earlier than the third and so on. Arguably, the experience of succession just is the experience of time’s dynamism or transition. Or, to give other examples, upon looking at the second hand of a watch, we can perceive that it is located at one position before it is at another, and before it is at still another and so on. When we go to the doctor for a checkup and feel her rapidly tapping on our chest or our abdomen or gently squeezing the glands under our neck, we experience the taping and squeezing sensations occurring in succession. Surely, one of the ways in which we experience time passing is by experiencing the successive flow of events from earlier to later. And is this not a direct experience of temporal transition and temporal passage even though it does not appeal to A-properties? I shall return to this question in Chapters 4, 5 and 8. A further feature of time is that we experience events in time differently from how we experience objects in space; time has an intrinsic sense or a direction. Broad once expressed this basic difference between space and time by saying that whereas a series of points in space have an intrinsic order, the peculiarity of a series of events in time is that they not only have an intrinsic order, but also an intrinsic sense. As Broad puts it: Three points on a line have an intrinsic order, i.e., B is between A and C, or C is between B and A, or A is between C and B. This order is independent of any tacit reference to something traversing the line in a certain direction. By a difference in sense I mean the sort of difference which there is between say, ABC and CBA. Now points on a straight line do not have an intrinsic sense. A sense is only assigned to them by correlation with the left and right hands of an imaginary observer, or by thinking of a moving body traversing the line in such a way that its presence at A is earlier than its presence at B, and the latter is earlier than its presence at C. (1923, p. 66; italics added) Or again, in a later writing he makes the same point in the following passage: In the temporal series of experiences which constitutes a person’s mental history there is a genuine dyadic relation which is intrinsic to the series and involves no reference to any term outside the latter. This is the relation “earlier than”. It is the fundamental relation here,

6  Problems in the Ontology of Time and temporal betweenness is definable in terms of it. In the temporal series there are two intrinsically opposite directions, earlier-to-later and later-to-earlier. In the linear spatial series there is no intrinsic direction. (Broad, 1938, p. 39) To clarify Broad’s point, note that the change of an apple from green to red is a change in a given direction because the apple is first green and then red, or synonymously, it is green before it is red from any point of view. Our experience of time as having an intrinsic direction is thus connected with the difference between our experience of change in time of an apple from green to red and our experience of a change in the color of a lawn from green to brown. In this respect, time differs from space since a spatial series has a direction only in reference to something external to the series. For that reason, to account for change, that is, how one and the same thing can first have a property and then lose it, we must be able to account for the direction of time and the difference between the change of the color of an apple over time and the spatial change of the color of a lawn over space or an area. McTaggart concurs that there must be something more to temporal change or change though time than there is in an ordered series of qualitatively different terms. Genuine change requires a temporal series; a series whose generating relation is “earlier than.” He then claims, More is wanted, however, for the genesis of a B-series and time than simply the C-series and the fact of change. For change must be in a particular direction. And the C-series, while it determines the order, does not determine the direction. If the C-series runs M, N, O, P, then the B-series . . . can run either M, N, O, P (so that M is earliest and P latest) or else P, O, N, M (so that P is earliest and M latest). And there is nothing either in the C-series or in the fact of change to determine which it will be. (McTaggart, 1908, p. 25; emphasis added) According to McTaggart, the C-series is a series with order in the terms but no sense or direction. Hence it alone cannot constitute a temporal series. Thus, by the fact of change in a non-temporal C-series, McTaggart means only that the terms of the C-series are qualitatively different. Clearly then, for McTaggart, something more is essential to temporal relations and change in time than an ordered, but not directed, C-series, regardless of whether or not those terms have ordinary or A-properties. But what more? Alternatively, how are we to account for the intrinsic direction of time and the lack of direction in space? Thus, there are three prima facie phenomenological distinctions that an adequate ontology of time must be able to explain, and they raise

Problems in the Ontology of Time  7 interrelated issues that concerned Broad throughout his changing views on time. First, what is the ontological basis of the transitory and extensive aspects of time? Second, what is the basis of the order and intrinsic sense of time and the order, but lack of direction of space? Third, what is the difference between genuine change or change in time, and “change” in space. Alternatively, how are we to account of the persistence of objects through time and change? Speaking commonsensically and pre-analytically, without ontological commitment or prejudging any ontological issue, we can note that there are temporal relations and temporal relational facts, as in “e1 is earlier than e2,” and facts reflecting temporal becoming, as in “I will have my dinner,” “I am having my dinner” and “I had my dinner.” When viewed extensively or in terms of temporal relations, we conceive of time or events in time as having an unchanging quality, fixed in their location on the series of events that occur in an unchanging succession. When time is viewed as transitory, we conceive of events as changing their temporal position as they move from the future to the present and into the past. The problem then is to reconcile these two seemingly incompatible ways in which we experience, think and speak about time. How can one and the same thing, time or events in time be flowing and still, dynamic and fixed, variable and unchanging? To answer the question “what is time?” we must provide an ontological ground for those purportedly incompatible commonsense facts about the phenomenology of time. Can the ground of the transitory aspect of time—temporal becoming—be the ground of the direction of time and the difference between temporal change and spatial change, or can temporal relations alone ground all the phenomena including temporal becoming? Insofar as the B-series is construed as spacelike or merely an ordered series with certain logical properties, the problem for B-theorists is to explain how the B-series alone can ground the intrinsic sense or the direction of time, and with it the transitory and flowing aspects of time. For those who conceive of time in terms of the A-series, temporal properties, and temporal becoming, the question is how those features can ground temporal relations and the purported extensive, fixed and unchanging aspect of time. Indeed, one way of structuring the debate in the ontology of time would be to ask: Can the entities employed to account for the transitory aspect of time also account for temporal relations, or can the entities employed to account for time relations also account for the transitory aspects of time? Finally, could the ground of time be a combination of entities: those that account for time relations and those that account for transition in time? Or are there still other alternatives? For example, could some form of presentism—the view that neither the past nor the future, but only the present exists—account for all temporal phenomena?

8  Problems in the Ontology of Time Broad changed his thinking on these crucial questions concerning time. Initially, he maintained that we can account for all temporal phenomena, including the transitory aspects of time, by means of temporal relations. Broad expressed this view as follows: 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, p. 143) In other words, according to Broad, we are given in experience the temporal relations of before and after, earlier and later, and simultaneity as ontologically simple, unanalyzable in terms of monadic temporal properties of their relata, that are just different from spatial relations. From these relations all knowledge of time is derived, including those phenomena that fall under the rubric of “the transitory aspect of time.” Thus, for example, Broad says, The distinction between past, present, and future is not one which, like that between before and after, lies wholly in the experienced objects, but is one that rests on the relations between experienced objects and the states of mind in which they are experienced. (1921, p. 145; my emphasis) Other followers of Russell (1915) and contemporaries of Broad, such as Richard Braithwaite (1928) and R. M. Blake (1925), also argue that temporal relations alone, properly understood, could ground all the phenomenological facts about time. For example, with regard to the intrinsic sense of time and its difference from space, Braithwaite says: It is important to notice that the relation of succession as given in experience has an intrinsic sense. We perceive directly that there is an intrinsic difference between A succeeding B and B succeeding A. As Dr. Broad points out, in the perception of spatial relations on a straight line, we perceive an intrinsic order, e.g., that B is between A and C, but the series ABC only acquires a sense through some

Problems in the Ontology of Time  9 extrinsic determination. Temporal relations, however, have an intrinsic sense as well as an intrinsic order, but (and here Dr. Broad goes wrong) this intrinsic sense of the fundamental temporal relation of succession is as much an immediately experienced fact as the intrinsic order, and is in no way derived from the distinction between present, past and future. (Braithwaite, 1928, p. 164)3 In this passage, Braithwaite accounts for the intrinsic direction of time as we experience it by specifying a primitive temporal relation of succession (henceforth called an “R-relation”), to serve as its ground. Indeed, as I shall argue, not only the direction of time, but also the transitory aspect of time and the analysis of change can be all grounded on the fundamental temporal relation of succession. This is a crucial feature of the Russellian view or R-theory that I shall develop in this book. There is, of course, an alternative analysis, first adopted by McTaggart (1908, 1927) and then by Broad. On this account, temporal properties and becoming are more fundamental than temporal relations. McTaggart expressed this point in the following passage: The series of earlier and later is a time series. We cannot have time without change, and the only possible change is from future to present, and from present to past. Thus, until the terms are taken as passing from future to present, and from present to past, they cannot be taken as in time, or as earlier or later; and not only the conception of presentness, but those of pastness and futurity must be reached before the conceptions of earlier and later and not vice versa. (1927, p. 271 fn1) Shortly after Broad published his Russellian analysis of time (hereafter the “R-theory”) in (1921), he came to believe in the primacy of the A-series (or at least the past and present part of it) and absolute becoming; the coming into existence of a present event from a non-existent future (Broad, 1923). Subsequently (Broad, 1937), he argued that at the level of ontology, temporal relations could not exist without the properties of pastness, presentness and futurity and temporal becoming; events changing their A-properties. Ten years after McTaggart, Broad expressed this view in the following passage: I am inclined to think that the notion of earlier and later is inextricably interwoven with the notions of becoming and of pastness, presentness, and futurity; so that the suggestion that there might be a series of terms to which the latter notions did not apply, but

10  Problems in the Ontology of Time which were nevertheless ordered by the relations of earlier and later, is really unintelligible. (1937, pp. 242–243) This passage reflects not only McTaggart’s influence on Broad, but more importantly, a fundamental difference between A-theorists and R-theorists. It also suggests one way to structure the debate in the philosophy of time. For Broad (1923, 1925, 1937, 1938), McTaggart (1908, 1927) and other A-theorists, the ontology of time requires (some or all of) the monadic temporal properties of pastness, presentness and futurity as well as temporal becoming understood as the donning and doffing of such properties. These properties and temporal becoming, construed as basic and fundamental categories of intrinsically temporal entities, are required in the ontological analysis of temporal relations. For R-theorists, by contrast, it is temporal relations alone that are the fundamental temporal entities in terms of which past, present and future, as well as temporal passage, temporal becoming and change are to be understood. Of course, each of these ontological alternatives face challenges. A-theorists must be able to understand how events can acquire and shed temporal properties with the passage of time without being subject to McTaggart’s paradox to be discussed in Chapters 2–5. Clearly, if, as A-theorists maintain, temporal relations depend on temporal becoming, then temporal becoming must be rendered consistent without an appeal to earlier than/later than, but how can that be done? R-theorists would insist that temporal relations alone can ground all temporal phenomena, including temporal becoming, but how can that be done? For R-theorists, one central question becomes: can temporal relations account for the dynamic aspect of time, the phenomena of temporal passage, and our experience of time, or events in time, flowing from the future to the present and into the past? John Bigelow has argued that temporal relations can be defined in terms of A-properties and the passage of time, by an appeal to possible worlds. Apart from how he would do this, the fact that he attempts to define temporal relations or explain what makes a relation temporal in terms of temporal passage shows a difference between A-theorists and R-theorists. He states this point nicely in the following passage: What, we may ask, makes the earlier/later relation a distinctively temporal ordering? It is precisely the passage of time which is required to make the earlier/later relation a temporal one. In other words, in order for the earlier/later relation to be a temporal relation it must be defined in such a way as to ensure that when what is earlier is present then what is later is still future, and when what is later is present then what is earlier will be past—that is, the earlier/later relation is a temporal one only if it concerns something which passes . . . This entails

Problems in the Ontology of Time  11 that earlier and later must be analyzed in terms of the passage of time, not the other way around. And therefore, the passage of time, in turn must be analyzed and shown to be consistent without presupposing the ordering of events under the relation of earlier to later. (Bigelow, 1991, pp. 3–5; emphasis added) I have critically discussed Bigelow’s interesting analysis of temporal relations in terms of possible worlds and the passage of time elsewhere (Oaklander, 1994). Here I wish to illustrate what is taken to be a central divide in the ontology of time: between those who assert that temporal passage is more fundamental than temporal relations, and those who claim that temporal relations are more fundamental than temporal passage. In the literature, this has become a major point of contention between A- and B-theorists. But, as I shall argue in the next chapter and throughout the book, in an important sense this is a false dichotomy. For in the R-theory, there is no difference between temporal passage and temporal R-relations. Some philosophers of A-time have accepted the definitional approach for determining ontological commitments, as evidenced by their claim to have analyzed B-relations in terms of A-determinations and temporal becoming and by following McTaggart in holding that relations without the passage of time are non-temporal. To illustrate the definitional approach, after offering three different reductive analyses or definitions of the earlier than/later than relations in terms of tense determinations, William Lane Craig, a defender of presentism, says that, It seems to me that in light of the foregoing, it is now open to the A-theorist to turn to his B-theoretical interlocutor and say, “I have an ontological foundation in my metaphysic of time for affirming the existence of the (tenseless) temporal relations earlier than/ later than. But what entitles you, having stripped time of all its tense-determinations, to assume that what remains is really time? Why should we regard those relations obtaining between tenselessly existing events in the B-series as earlier than/later than rather than as some atemporal relations analogous to the less than/larger than relations obtaining between the members of the natural number series? Indeed, why think that any such relations exist at all, in addition to the relations which obtain between, say, the points on a spatial line? On my theory, moreover, these relations are not only entailed by the reality of tensed facts, but the existence of these relations entails the existence of tensed facts; they could not exist apart from tensed facts. So why, if there is no past, present, and future, as you claim, should we think that there are still earlier and later? (Craig, 1999, p. 363)4

12  Problems in the Ontology of Time Craig is drawing ontological implications about what are the fundamental categories (the reality of tense and temporal becoming) from logical considerations (the reducibility of B-theoretic discourse). The implication concerns the correct ontological analysis of temporal relations. For Craig, there are no temporal relations apart from tensed facts and temporal becoming. For the R-theorist there are no temporal relations with tensed properties and tensed facts as McTaggart purported to demonstrate. Clearly, then, there is a debate in the philosophy of time concerning the ontological status of temporal R-relations and temporal A-properties. I have been arguing that the ontological status of temporal relations and succession is an important ontological issue in the debate in the philosophy of time. This position is worth highlighting, since by following Broad’s changing views on temporal relations, we can understand his changing views on time. Furthermore, structuring the debate as I do will help bring out some flaws in the current debate and discussion in the ontology of time (in Chapter 3). And, finally, it will provide a fruitful model with roots in Russell (1915) and Broad (1921) for an adequate ontology of time. As I have mentioned already, and as we will see in detail in what follows, Broad’s ontology of time covered many different theories. Before turning to those theories, I want to consider one method Broad adopted in answering the question “What is the fundamental nature of time?” By “fundamental,” he means “ontologically fundamental,” and the ontologically fundamental are those simple unanalyzable entities that make up the furniture of the world. For that purpose, it will be useful to consider the relation between commonsense and ontology as Broad and others conceive of it, and relate Broad’s view to his method and ontology of time.

III. Ontology and Method In this section, I shall contrast three different approaches to commonsense and ontology, beginning with G. E. Moore, then Broad and Lynne Rudder Baker before considering Russell’s approach and offering my own that is derived from Russell. We can all agree that commonsense facts are expressed in ordinary language. Those facts must, however, be distinguished from the ontological facts that are their analysis. The everyday world contains objects, and commonsense facts about those objects, that no one wishes to dispute. Thus, for example, in his symposium paper “Is there ‘Knowledge by Acquaintance’?” Moore (1919, p. 191) argues that when Russell uses the term “acquaintance” he is generally appealing to a relation that we have to objects of sense when we perceive, hear, touch, taste or smell something. In this sense, according to Moore, it is quite clear that we are acquainted with objects. Thus, when Neutral

Problems in the Ontology of Time  13 Monists claim that in cases where two objects are experienced by the same individual there is, strictly speaking, nothing which experiences either of them, they should not be viewed as denying acquaintance with sense data, but with a specific analysis of the self as a Pure Ego. In the Neutral Monist analysis, there is no subject or self that stands in the relation of direct acquaintance with the object experienced. In Russell’s view of acquaintance, Neutral Monism is mistaken since he holds that when two objects are experienced by the same person, there is a subject S, that does the experiencing, and that it is distinct from the objects experienced. These are two different analyses of what is meant by “acquaintance.” Moore continues, I do not, when I assert that I certainly am acquainted with sensedata, in the least wish to imply that the Neutral Monists are wrong in their analysis of the facts: I only wish to assert an indisputable fact of the kind they are trying to analyse. (1919, p. 186) Generally, when Russell uses the term “acquaintance,” he is using it simply in connection with an indisputable fact; a commonsense fact that nobody would ever think of disputing. Moore proceeds, What I wish to make clear is that Neutral Monists do not for a moment deny the existence of what I am calling acquaintance with sense-data, and what I take Mr. Russell to generally to have meant by that term. All that they do is to offer a particular analysis of the kind of fact which I express by saying that I am acquainted with sensedata, without, of course, denying, any more than anybody else does, the existence of facts of the kind they are analyzing. (1919, p. 185) There are many different analyses of the indisputable facts that can, of course, be disputed, but what cannot be disputed, much less denied, according to Moore, are the commonsense, phenomenologically given facts philosophers seek to analyze or, if I may so put it, ground ontologically or seek to explain by specifying entities of various kinds. Moore reaffirms this position in the following passage: I am not at all sceptical as to the truth of such propositions as “The earth has existed for many years past.” “Many human bodies have each lived for many years upon it” . . . on the contrary, I hold that we all know, with certainty, many such propositions to be true. But I am very sceptical as to what, in certain respects, the correct analysis of such propositions is. (1925, p. 52)

14  Problems in the Ontology of Time Thus, for Moore, there is a clear and unambiguous bifurcation between commonsense and ontology. There is a second stance that presents a more ontologically loaded view of commonsense but that still agrees that there are certain indisputable facts based on everyday life. For example, that the self exists, but builds into commonsense a certain ontological analysis of the self. In Broad’s discussion of the distinction between Center and Non-Center Theories of the unity of the mind (that I discuss in Chapter 6), he claims that the indisputable facts strongly suggest a particular analysis of the self: The prima facie presumption in favour of Central theories and against Non-Central theories is the common usage of language, which strongly suggests the existence of a Centre. We say: “I am thinking of this book, and wanting my tea, and feeling tired, and remembering the tie that my friend wore yesterday.” This certainly suggests that “I” is the proper name of a certain existent which stands in a common asymmetric relation to all those contemporary mental events. We say further: “I, who am now doing and feeling these things, was yesterday doing, thinking, wanting, and feeling such and such other things.” And this certainly suggests that “I” is the proper name of something which existed and was a centre yesterday as well as to-day . . . Now, I am not suggesting that we should accept a theory because it seems to be implied by the statements of plain men. God forbid! But I do suggest that any satisfactory theory must account for the fact that plain men and philosophers in ordinary life express themselves in language which strongly favours one alternative. (Broad, 1925, pp. 584–585; emphasis added) For Broad, there is an implicit ontology in the concepts we employ in ordinary life and language that must be given weight, but it does not require that the correct ontology is to be found in commonsense or ordinary language. One can reject that analysis if the facts for which it is introduced can be explained otherwise, and the original analysis is, for example, dialectically untenable. Thus, Broad differs from Moore in thinking that commonsense facts strongly suggest or even imply a certain theory or analysis of the concept or phenomena under investigation, but agrees with Moore in maintaining that one can deny a specific analysis of certain phenomena without denying the phenomena upon which those prima facie or commonsense facts are based. A third way of understanding commonsense and its relation to ontology is to treat it as not only containing indubitable facts, but also identifying those ordinary facts with a certain ontology. Lynne Rudder Baker (2007) seems to be sympathetic with this view since she maintains that manifest objects and everyday facts reflect the metaphysics of everyday

Problems in the Ontology of Time  15 life and are irreducibly real. Baker claims that “the term ‘irreducibly real’ and its variants refer to objects that belong in ontology: objects that exist and are not reducible to anything ‘else’” (Baker, 2007, p. 4). Included in an “ontological account” of the everyday world is a “complete inventory of what exists” (2007, p. 3). “A complete ontology—comprising everything that is irreducibly real—will include manifest objects like ­ tables” (2007, p. 4). In other words, the distinction between the manifest objects we encounter in everyday life and ontology is much more intimate in Baker than in Moore, where there is virtually no connection, or in Broad, where an ontology is strongly suggested by language, but can be rejected. For Baker, commonsense is not ontologically neutral, nor does ordinary language and experience merely provide some initially sufficient reasons to support one rather than another ontology. The shared world that we encounter, that is given to us in our experience and is reflected in everyday discourse, represents a specific ontology and an irreducible reality the indubitability of which overwhelms all arguments that can be given against it. As Baker puts it: In sum we have overwhelmingly greater reason to believe in the irreducible reality of ordinary objects and properties than to believe in any theory that denies that they are irreducibly real. The evidence of our senses, of which the commonsense tradition avails itself, trumps arcane arguments leading to anti-commonsense conclusions cut off from anything we can confirm in experience. (2007, p. 9) This suggests that commonsense has an articulate ontology, that this ontology cannot be denied without thereby also denying our commonsense beliefs and experience, and, therefore, that any argument to the contrary must itself be rejected. There are, it seems to me, two questions that need to be separated. First, do our commonsense beliefs about time suggest a specific metaphysics of time? Second, if they do, is Baker’s description of that metaphysics the one suggested by everyday life? These are important questions, but I don’t think the irreducible realities that are required to answer them is implicit in commonsense. I would suggest, contrary to Baker, that commonsense is ontologically neutral. To put my point differently, one can agree with Baker that for example, it is manifest in our experience that “we cannot imagine living in a world without the passage of time” (2007, p. 157), but it does not follow that the commonsense fact that time passes or that language contains tenses reflects or implies an A-theoretic account of the ontological facts that explains our experience of time. Nor does it follow that to deny A-series properties is to deny the passage of time or commonsense. Thus, I am sympathetic with the first stance regarding

16  Problems in the Ontology of Time the relation between commonsense and ontology. This position is nicely stated by Russell: 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, pp. 179–180) Those ambiguous beliefs that we feel quite sure of, upon reflection and analysis, are often seen to be incompatible and give rise to puzzles. The task of philosophy is to show how, through reflection and analysis, the prima facie paradoxes involved in our everyday beliefs or commonsense facts can be resolved through a dialectically adequate ontological analysis of the phenomenological data in question. Russell seems to explicitly follow this methodology in search of a satisfactory analysis of the relation between the experiencing subject and the experienced object when he says, Before embarking on our analysis let us again take stock of those relevant facts which are least open to doubt. From the diversity of philosophical theories on the subject, it is evident that the true analysis, whatever it may be, cannot itself be among the facts that are evident at once, but must be reached, like a scientific hypothesis, as the theoretic residue left by the comparison of data. Here, as in philosophy generally, it is not the few logically simplest facts that form our data, but a large mass of everyday facts, of which the analysis offers fresh difficulties and doubts at every step. For this reason, if we wish to start with what is undeniable, we have to use words, at first, which, though familiar, stand in need of a dissection and definition only possible at a later stage. (Russell, 1914, p. 160) As we have seen, in Broad’s view, the connection between commonsense and ontology is closer than in Russell’s view expressed in the previous paragraph. Yet, they seem to agree that commonsense is not sacrosanct when the goal is to find the correct ontology. According to Broad, one aspect of “the subject matter of Philosophy— Critical Philosophy—consists in the analysis and definition of our fundamental concepts, and the clear statement and resolute criticism of our fundamental beliefs” (Broad, 1952, p. 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. The “meaning” of a

Problems in the Ontology of Time  17 term is given by the ordinary concept we have associated with that term, according to one sense of the term, but that is problematic since such a meaning is unsophisticated and naïve with respect to ontological questions. It is the meaning associated with commonsense. Commonsense constantly uses concepts in terms of what it interprets of the 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. (1952, p. 15; emphasis added) 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 or belong to those categories, their meaning or ontological analysis must be clear and unambiguous. According to Broad, 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, and to state them clearly and then to subject them to criticism. Of course, to state our deeply held beliefs clearly, such as “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 concepts 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. For Broad, the process of arriving at “the real truth of which that vague thing is a sort of a shadow” (Russell, 1918, 2001, pp. 179–80), 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 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 or phenomenological analysis. 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” (Broad, 1924, p. 93). Thus, a Pickwickian definition of a concept would be one that is intended or taken in a sense other than the obvious or literal one.

18  Problems in the Ontology of Time 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. (1924, p. 95)5 Similarly, the existence of certain phenomenological facts, such as the appearance of the transitory aspect of time, may require an ontological explanation that is not intuitive. They may be accounted for by a set of facts that is not what commonsense implicitly take them to be, but that may be the best we can get and entirely suitable to account for the phenomenology of the situation. In other words, the Pickwickian Principle allows that even if an analysis is not the one that is implicitly assumed by commonsense, we can still accept it if the analysis establishes that there exist entities that fall under the concept, or justify the application of the concept and explain the phenomena in question. Broad uses the concepts of “the self” and “matter” to explain an error that we may fall into by ignoring the distinction implied by The Principle of Pickwickian Senses between our ordinary concepts, beliefs and phenomenological data, and the analysis or proper description of those concepts, beliefs and data. One such error occurs in the following passage where Broad notes that questions like “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 common-sense; and, on critical analysis, they are often found to consist of a number of propositions of very different degrees of importance and certainly. 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

Problems in the Ontology of Time  19 tacitly, assumed, a man who rejects this analysis will seem to others (and often to himself) to be rejecting the concept itself. (1924, pp. 94–95) According to Broad, however, that would be a mistake. We must distinguish, Broad says, between facts, what I have called “commonsense facts,” and the proper analysis or description of those facts, or “ontological facts.” There are, undoubtedly, commonsense facts to which the term “self” applies, 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 commonsense. However, if upon critical examination it is seen that there are reasons to reject that analysis, and if an alternative analysis of the self can account for the commonsense and phenomenological facts about it, such as “the unity of the self,” then one can accept that the concept of the self has an application and therefore that the self exists. Thus, even if one rejects the commonsense view of the self and its presumed ontological analysis, one is not thereby rejecting the concept or 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 a 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, p. 95) In this passage, Broad is drawing a distinction between commonsense facts or pre-analytic data and 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 the failure to recognize it can easily lead to “unprofitable discussions” (1924, p. 95).6 Of course, one may question, as I do, whether commonsense or our “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. However, whether ordinary concepts and beliefs have any, one or more implicit, unrecognized ontologies, Broad and Russell are surely right to maintain that the difference between the phenomenological data and pre-analytic, commonsense facts and the analysis of that data or the ontological facts that are their ground, is important and useful in arriving at the real truth that underlies the

20  Problems in the Ontology of Time vague facts we start off with. In other words, even if commonsense has an implicit theory about the meaning of a concept it employs, that does not imply that the proper ontological analysis of the facts in which the concept occurs is implicit in or given by commonsense. Thus, even if we reject the commonsense concepts of the self, consciousness or time, for example, it does not follow that there is nothing that falls under those concepts or that the self, consciousness or time do not exist. At this point, I believe, it is important to see how Broad makes use of the Principle of Pickwickian Senses even before he explicitly formulated it. According to this principle, the question “Is time real?” does not have a simple “yes” or “no” answer. It may be that our commonsense belief in the passage of time and temporal becoming has an implicit ontology that suggests an A-theoretic ontology. That is, it may be that the commonsense truth that time passes and the existence of facts with temporal relations imply an A-theoretic ontology (although I would dispute this). Thus, if one denies such an ontology, as McTaggart does, then in that sense “time is unreal.” However, if there are dialectical, phenomenological or conceptual problems with the commonsense ontology, and there is another adequate ontology that can provide a ground for the phenomenological facts in question, then in that sense time would still be real. In other words, there is nothing about commonsense that is inviolable. The ground of time may be Pickwickian and not what we ordinarily believe it to be, but in matters of ontology, commonsense should not be our guide or at least not have the final word. In his earliest writing on “Time” (1921), while still under the influence of Russell, Broad discusses the question “Is time real?” He argues that what McTaggart and others who argue for the unreality of time have shown, at most, is that their analyses of time do not correspond to anything in the world. However, to demonstrate that an analysis or theory of time is mistaken is not to demonstrate that the phenomena that the analysis is concerned with does not exist or that there is nothing in reality that is its ground. The phenomena that cannot be doubted may comport with a different analysis. In the following passage, Broad emphasizes that temporal succession and duration are distinctly given to us in introspection and perception and can therefore hardly be doubted: It is a matter of direct inspection that the immediate objects of some of our states of mind have temporal characteristics. It is as certain that one note in a heard melody is after another in the same specious present and that each has some duration as that some objects in my field of view are red or square and to the right or left of each other. It is then quite certain that some objects in the world have temporal characteristics, viz. the immediate objects of some states of mind. Now it is also certain that these objects exist at least as

Problems in the Ontology of Time  21 long as I am aware of them, for in such cases I am obviously not aware of nothing. Hence there cannot be anything self-contradictory in the temporal characteristics found in these objects, for otherwise we should have to admit the existence of objects with incompatible characteristics. Hence there is no obvious reason why temporal characteristics should not also apply to what is not the immediate object of any state of mind. It follows, then, that criticism cannot reasonably be directed against temporal characteristics as such, but only against the descriptions that we give of the temporal characteristics of experienced objects, and the conclusions that we draw from them or the constructions that we base on them. . . . If we suppose that such criticisms are successful, the conclusion ought not to be either that reality has no temporal characteristics (for it is quite certain that at least some parts of it have), or that time, as an inference or construction extending the temporal characteristics of experienced objects to others, is unreal (for this goes much too far). The only justifiable conclusion would be that one particular way of describing and extending the temporal characteristics of experienced objects is unsatisfactory, and that it behooves us to look for a better one. This point has not commonly been grasped by philosophers who claimed to disprove the reality or time. (1921, p. 151; emphasis added) Broad’s point is that temporal phenomena are conveyed in ordinary language and commonsense facts are undeniable, for it is certain that “some objects in the world have temporal characteristics” (1921, p. 151). What is open to dispute is the proper description or ontological analysis of those characteristics. Consequently, those who deny the implicit and unrecognized ontology in commonsense temporal facts (if it has an implicit or articulate ontology at all), are not thereby denying the reality of time, but are proposing an alternative theory of the real truth that underlies the commonsense, phenomenological descriptions. In other words, one can reject McTaggart’s or any other analysis of the transitory or dynamic aspects of time without denying that there is temporal transition or dynamism; and without thereby concluding that time is unreal. Eventually, I shall attempt to show that some defenders and opponents of the B-theory and the A-theory fail to properly recognize Broad’s Principle of Pickwickian Senses and the distinction between commonsense and ontology that follows from it. For, even if some version of the A-theory is susceptible to McTaggart’s paradox and the B-theory cannot account for the transitory aspect of time, those arguments do not apply to the R-theory and for that reason an R-theoretic ontology may do just fine in accounting for the phenomenological facts regarding temporal passage and the like.

22  Problems in the Ontology of Time

Notes 1. See Tegtmeier (2012) and Grossmann (1992) for good introductions to the scope and methods of ontology. 2. See also, Nelson Goodman (1977). 3. In this passage, Braithwaite is criticizing Broad’s later views of time in Scientific Thought (1923), but the view Braithwaite is expressing is clearly sympathetic with Broad’s early view on time and one of Russell’s (see 1938). 4. See also, Craig (2000). I criticize Craig’s presentism in Oaklander (2002a, 2002b). 5. Broad also discusses this Principle where he says: § 12. This principle has always been familiar in Theology. When theologians say that the Second Person of the Trinity is the son of the First Person, they are using the word “son” in a highly Pickwickian sense. Anyone who will read, e.g., St. Thomas’s brilliant discussion of this subject in the Summa contra Gentiles will see how careful St. Thomas is to point out in his own language that phrases like “sonship” and “begetting” cannot be interpreted literally here, and will further see what an elaborate and metaphorical interpretation St. Thomas puts upon such phrases. Now Whitehead and Russell have explicitly carried this principle over into philosophy, where I am quite sure that it is destined to play a most important part. (1924, p. 93) 6. Francesco Orilia has pointed out to me that in his book On Philosophical Method (1980), Hector-Neri Castañeda articulates something like this distinction in detail. However, whereas Castañeda speaks of “protophilosophical data” and philosophical theories that compete in trying to elucidate these data, Broad seems to believe that commonsense 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 may be understood in a sense other than the literal one implicit in commonsense, assuming of course, that commonsense has an ontology, which is open to question.

References Baker, L. R. (2007). The metaphysics of everyday life: an essay in practical realism. Cambridge: Cambridge University Press. Bergmann, G. (1967). Realism: a critique of Brentano and Meinong. Madison, WI: The University of Wisconsin Press. Bigelow, J. (1991, Mar.). Worlds enough for time. Noûs, 25 (1), 1–20. doi:10.2307/ 2216090 Blake, R. M. (1925). On Mr. Broad’s theory of time. Mind, 34, 418–435. https:// doi.org/10.1093/mind/XXXIV.136.418 Braithwaite, R. B. (1928, Jul.). Symposium: time and change. Aristotelian Society Supplementary Volume, 8 (1), 162–174. Broad, C. D. (1921). Time. In J. Hastings et al. (Eds.), Encyclopedia of religion and ethics (Vol. 12, pp. 334–345). New York: Scribner’s. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 143–173). References are to this reprint.

Problems in the Ontology of Time  23 ———. (1923). Scientific thought. Chap. II. The general problem of time and change (pp. 53–84). London: Routledge and Kegan Paul. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. II. Time and metaphysics (pp. 63–83). References are to this reprint. ———. (1924). Critical and speculative philosophy. In J. H. Muirhead (Ed.), Contemporary British philosophy: personal statements (First Series, pp. 77–100). London: G. Allen & Unwin. ———. (1925). Mind and its place in nature. London: Routledge & Kegan Paul. ———. (1928, Jul.). Symposium: time and change. Aristotelian Society Supplementary Volume, 8 (1), 175–188. https://doi.org/10.1093/aristoteliansupp/8.1.143 ———. (1937). The philosophical implications of precognition. Aristotelian Society Supplement, 16, 229–245. ———. (1938). Examination of McTaggart’s philosophy: ostensible temporality (Vol. II (1), Chap. XXXV, pp. 265–323). Cambridge: Cambridge University Press. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 36–68). References are to this reprint. ———. (1952). Scientific thought. New York: Harcourt, Brace & Co.; London: Routledge & Kegan Paul. ———. (1959). A reply to my critics. In P. A. Schilpp (Ed.), Philosophy of C. D. Broad (pp. 711–830). New York: Tudor Publishing Company. Castañeda, H. N. (1980). On philosophical method. Bloomington, IN: Indiana University Press. Craig, W. L. (1999). Temporal becoming and the direction of time. Philosophy and Theology, 11 (2), 349–66. doi:10.5840/philtheol19991126 ———. (2000). The tenseless theory of time: a critical evaluation. Dordrecht: Kluwer Academic Publishers. Goodman, N. (1977). The structure of appearance. Dordrecht: Reidel (Originally published in 1951). Grossmann, R. (1992). The existence of the world: an introduction to ontology. London: Routledge. Marsh, R. C. (Ed.). (2001). Bertrand Russell. Logic and knowledge: essays 1901– 1950. London: Routledge (Originally published in 1956). McTaggart, J. M. E. (1908, Oct.). The unreality of time. Mind, New Series, 17 (68), 457–474. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 21–35). References to this reprint. ———. (1927). The nature of existence (Vol. II, C. D. Broad, Ed.). Cambridge: Cambridge University Press. Moore, G. E. (1919, Jul.). Symposium: is there “knowledge by acquaintance”? Aristotelian Society Supplementary Volume, 2 (1), 159–220. https://doi. org/10.1093/aristoteliansupp/2.1.159 ———. (1925). A defence of common sense. In J. H. Muirhead (Ed.), Contemporary British philosophy (2nd series). Reprinted in G. E. Moore. (1966). Philosophical papers (pp. 32–59). New York: Collier Books. References are to this reprint. Oaklander, L. N. (1994, Oct.). Bigelow, possible worlds and the passage of time. Analysis, 54 (4), 244–248. https://doi.org/10.1093/analys/54.4.244. Reprinted in L. N. Oaklander. (2004). The ontology of time (pp. 71–75). Amherst, NY: Prometheus Books, and in L. N. Oaklander (Ed.). (2008). Vol. II. Time and metaphysics (pp. 31–35). London: Routledge.

24  Problems in the Ontology of Time ———. (2002a). Presentism: a critique. In H. Lillehammer & G. R. Pereyra (Eds.),  Real metaphysics: essays in honour of D. H. Mellor, with his replies (pp. 196–211). London: Routledge. Reprinted in L. N. Oaklander. (2004). The ontology of time (pp. 101–115). Amherst, NY: Prometheus Books. ———. (2002b). Presentism, ontology and temporal experience. In C. Callender (Ed.), The royal institute of philosophy supplement 50: time, reality, and experience (pp. 73–90). Cambridge: Cambridge University Press. Reprinted in L. N. Oaklander. (2004). The ontology of time (pp. 83–99). Amherst, NY: Prometheus Books. ———. (Ed.). (2008). The philosophy of time: critical concepts in philosophy (Vols. 1–IV). London: Routledge. Prior, A. N. (1959). Thank goodness that’s over. Philosophy, 34 (128), 12–17. https://doi.org/10.1017/S0031819100029685. Reprinted in L. N. Oaklander & Q. Smith. (1994). The new theory of time. New Haven, CT: Yale University Press, and Oaklander, L. N. (2008). Vol. 3. Time, experience and freedom (pp. 7–12). Russell, B. (1914, Jul.). On the nature of acquaintance. Monist, 24 (3), 435–453. https://doi.org/10.5840/monist191424311. Reprinted in R. C. Marsh (Ed.). (2001). (pp. 125–174). References are to this reprint. ———. (1915, Apr.). On the experience of time. Monist, 25 (2), 212–233. https:// doi.org/10.5840/monist191525217. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 174–187). ———. (1918, Oct; 1919, Jan, Apr, Jul.). Lectures on the philosophy of logical atomism. Monist, 28 (4), 495–527; 29, 32–63, 190–222, 345–380. https:// doi.org/10.5840/monist19182843. Reprinted in Marsh, R. C (Ed.). (2001). (pp. 177–281). London: Routledge. References are to this reprint. Lectures also appear in J. Slater (Ed.). (1988). Vol. 8. Collected papers of Bertrand ­Russell. London: Routledge. ______. (1938). The principles of mathematics. New York: Norton. (originally published in 1903). Schlesinger, G. (1980). Aspects of time. Indianapolis: Hackett Publishing Co. Tegtmeier, E. (2012). McTaggart’s error: temporal change. Revue Roumaine de Philosophie, 56, 89–96.

2 The Early Years

The Early YearsThe Early Years

The Russellian Theory of Time

I. Introduction The origin of the debate between A- and B-theories is between the followers of McTaggart (1908, 1927) and followers of Russell (1915). However, Russell’s view, as it was developed and adopted by followers such as A. J. Ayer (1956, 1972), Broad (1921), R. M. Blake (1925) and Richard Braithwaite (1928), is different in important ways from the B-theory as expounded by McTaggart and other critics, including the later Broad. In Broad’s first writing on time, to which this chapter is devoted, Broad states that he “was almost completely under the influence of Russell in his extreme realist phase and Meinong as I understood him” (Broad, 1959, p. 765). It will therefore be useful to clarify Broad’s early view by also discussing the Russellian theory of time. By doing so, we can understand not only the origins of the R-theory, but also see how Broad’s errors in his exposition of Russell’s view may have influenced his turn to the A-theory and invited objections to the B-theory by others. In this chapter I want to show how Broad’s early reflections on time exhibit original insights that have played a significant role in later discussions of the B-theory. Perhaps nowhere is this more evident than in his discussion of the metaphysical implications of temporal language. To see how and why this is, I will first turn to the section on the connection of time with logic found in Broad’s encyclopedia entry on “Time” (1921).

II. On the Relation of Time to Logic: Broad’s Analysis of Tensed Language Broad distinguishes three kinds of propositions as they are related to time. First, there are (seemingly) tensed propositions with truth values that are at the mercy of time. For example, “Queen Anne is dead” expresses a proposition that is false before she died and true after. We shall see shortly that Broad ultimately believes that there are no propositions of the first kind, since all tensed propositions can be reduced to tenseless propositions whose truth value is timeless, hence unchanging. On the

26  The Early Years other hand, there are other propositions that are totally independent of time, e.g., 2 x 2 = 4. “These are sometimes called eternal truths; they always state relations between universals, and all our a priori knowledge is of such propositions” (1921, p. 147). Finally, there are propositions that are true at any time they are uttered and make an essential reference to time. He gives as an example, “whenever it rains, and I am out without my umbrella, I get wet” (1921, p. 147). There is another set of propositions that make an essential reference to time, that Broad also calls “eternal” or “timeless,” that state temporal relations between particulars or events, for example, “a is earlier than b.” These propositions are called “B-propositions” since they are about the temporal relations between the terms in McTaggart’s B-series. The truth value of B-propositions are unchanging, and thus, if they are true at all, then they are always true, eternally true or timelessly true. Broad uses these terms more or less interchangeably, but clarifying the different possible meanings of them will be useful in what follows. Broad calls the copula in B-propositions and the propositions themselves “timeless,” but that is potentially misleading and problematic since it may suggest that B-propositions also state relations between universals. While B-propositions and mathematical propositions are both in some sense eternal truths and timeless, the notions of “eternal” and “timeless” have a different ontological meaning in each case, as I shall explain by distinguishing between three different kinds of “eternally true” statements or propositions. Although Broad would agree that “the death of Plato is earlier than the death of Aristotle” is an “eternally true” statement, he is not as clear as he should be regarding the ontological ground of “eternal truths” since the notion of “eternal” is open to different interpretations that he blurs. A proposition is eternally1 true or is about an eternal1 fact, if it is always or permanently true in virtue of a fact that always exists or exists at every time. Thus, an eternal1 proposition or fact has an endless duration. For the Russellian, B-statements are always true, or true at every time they are expressed, but they are not true in virtue of something that always exists. For neither the death of Plato, the death of Aristotle, the relation between them, nor the fact that Plato’s death precedes Aristotle’s always exist. In a second sense, eternally2 propositions are true independent of time since they state relations among universals, and the facts they are about are outside of time and none of their constituents are intrinsically temporal. They are outside of time since to be “in time” is to stand in temporal relations, and eternal2 propositions or facts do not stand in temporal relations to anything. Eternal2 facts are independent of time since none of their constituents is in time and they do not contain any intrinsically temporal constituents. For that reason, an eternal2 entity would exist even if time did not exist.

The Early Years  27 Although the fact that e1 is earlier than e2 is outside of time since it is not a term of a temporal relation, it is not entirely independent of time, as eternal2 facts are, since not all of its constituents are universals, and more importantly, it contains a temporal relation as a constituent. Thus, temporal relational facts are outside of time; but they are not eternal in precisely the same sense as eternal2 facts are. Nor are temporal relational facts eternal1 since they do not exist at every time. Rather they are “eternal” in a third sense. Eternal3 facts exist outside of time (as all facts do), yet they are temporal facts since they contain time, that is, intrinsically temporal or time relations whose terms, particulars, are thereby in time. Consequently, although the terms of temporal relations are in time, the fact that Plato’s death is earlier than Aristotle’s death does not exist in time, nor a fortiori does it exist at every time. Indeed, R-facts do not exist in time at all. This point is important enough to be worth repeating. To be eternal3 means to exist outside of time, although not, as eternal2 entities are, entirely independent of it. An eternal3 entity is related to time in the following way: it is a whole that contains successive (or simultaneous) parts. We could say that although an eternal3 entity is not contained in time, time is a constituent it. Thus, the fact that WWII is later than WWI is eternal3 since it is not itself a term of a temporal relation. For that reason, though time contains temporal relations, time does not exemplify them, and so it is outside of time. Eternal3 facts contain temporal relations and, in that sense, are temporal facts and are neither timeless2 as, say, “3 is a prime number” is, nor eternal1 as, say, a being with an endless duration would be. This is an important point since failure to recognize it can give rise to a mistaken understanding of the Russellian theory of time. Unfortunately Broad’s exposition and later criticism of the R-theory (1938) is guilty of that misunderstanding. With this background we can return to Broad’s early view of time. The early B-theory of time, as originated with Frege (1956) and Russell (1906) and adopted by Broad, had a semantic component and an ontological component. In its semantic dimension it maintained that for a statement to be meaningful, it must express a proposition with an unchanging truth value. For example, on the date-analysis of tensed statements, “It is now raining” appears to express a proposition that prima facie changes its truth value but does not really do so. Indeed, for Russell and Broad, a tensed statement without a temporal reference does not express a complete proposition at all. However, once the statement is given a reference by means of the time at which it is uttered, written or thought, its meaning is fixed and the tenses are reduced to or eliminated by tenseless copulas plus dates. For example, a token of “It is now raining,” uttered at time t, expresses the unchanging proposition that “It is (tenselessly) raining at t.” The eliminability of the tenses in ordinary language for the purposes of having a logic of propositions with unchanging truth values is believed

28  The Early Years to have ontological significance. It implies a view of time that is devoid of what the tenses seem to refer, namely, the past, present and future, or, to speak ontologically, the monadic, non-relational temporal properties of pastness, presentness and futurity. This is the old tenseless or B-theory of time, and it relates to how Broad and many after him understand the task of philosophy. If we could translate all the information in a tensed statement without any loss of meaning by a tenseless one, then what is represented by the tenses would be eliminated from reality, or equivalently analyzed in terms of what is represented by the tenseless translation. In other words, to translate a tensed statement in ordinary language by means of a tenseless one without loss of meaning would, according to Broad, demonstrate that reality is tenseless, that is, without A-properties. To analyze a concept is to give an account of its meaning; a sentence that expresses a proposition that perspicuously (ontologically speaking) represents the reality that the concept is about. For Broad, the meaning of a concept is not simply about words; it is not a purely verbal matter, but it is about what those words represent; and so, he is concerned with referential meaning. Broad expresses this view in the following passage: At this point a criticism may be made which had better be met at once. It may be said: “By your own admission the task of Philosophy is purely verbal; it consists entirely of discussions about the meanings of words.” This criticism is of course absolutely wide of the mark. When we say that Philosophy tries to clear up the meanings of concepts, we do not mean that it is simply concerned to substitute some long phrase for some familiar word. Any analysis, when once it has been made, is naturally expressed in words; but so too is any other discovery. When Cantor gave his definition of Continuity, the final result of his work was expressed by saying that you can substitute for the word “continuous” such and such a verbal phrase. But the essential part of the work was to find out exactly what properties are present in objects when we predicate continuity of them, and what properties are absent when we refuse to predicate continuity. This was evidently not a question of words but of things and their properties. (Broad, 1952, p. 17) Similarly, the meaning of temporal predicates—“is past,” “is present” and “is future”—is not a question of words, but what those predicates represent in reality. Do those predicates represent non-relational temporal properties, as A-theorists maintain, or can they be replaced without loss of meaning by tenseless sentences that represent temporal relations among objects and events without ontological tense, as early B-theorists maintained? If the latter, then the correlates of tenseless relational sentences reflect a more accurate ontology of time.

The Early Years  29 According to Broad, “Critical Philosophy” involves the analysis of concepts and the justification of beliefs. We have seen that for Broad, ordinary language suggests a particular ontology that may be mistaken. Thus, Broad distinguishes between grammatical or verbal form and logical form. The verbal form of a statement in ordinary language suggests a particular ontology, but that ontology may be open to criticism, and so not reflect the logical or ontological form of reality. The logical form of a sentence in an ideal or ontologically perspicuous language reflects the true reality that underlies the vague truth that ordinary language may suggest. To give an example of the distinction between grammatical form and logical form, consider the following sentences: “My eating breakfast is past” and “My eating dinner is future.” These sentences suggest that an event has the properties of being past and being future, and yet we should not be bound by the grammatical form to depict reality if it leads to insurmountable dialectical difficulties, such as McTaggart’s paradox. Ordinary language may not be able to do without temporal predicates and grammatical tenses for communication, but a logically adequate language, not used for communication, constructed to represent the world, may and perhaps must do so if we are to arrive at an adequate ontology of time expressible in an ideal language. For Broad, the logical form that results from the meaning of a sentence token is a representation of the ontological structure of reality. What then is the meaning (or analysis) of the concepts derived from experience that pertain to time? Since meaning is expressed in language, we may put the question differently by asking: What is the meaning of the language of time? Alternatively, what is the proper analysis of propositions expressed by tensed sentences and tenseless temporal sentences? Are tensed sentences ineliminable and irreducible? If they are, then from the perspective of the A-theorist, that provides a strong reason to believe in the reality of A-properties and the special reality of the present. Or, are tensed sentences reducible to tenseless sentences that express eternal3 propositions about temporal relations between objects or events, thereby providing evidence that tensed properties can be eliminated from the ontology of time, as old B-theorists including Broad maintain? This way of structuring the debate—in terms of the reducibility of tensed to tenseless sentences or vice versa—was influential both for Broad and for others undertaking the philosophy of time for most of the twentieth century until the development of the new theory of time. It is typically thought that different sentence tokens of the same tensed sentence type express propositions that change their truth value with the passage of time. This is alleged to be evidence for and representative of the passage of time and the reality of A-properties. In other words, since the sentence “JFK is dead” expresses a proposition that is true now but was false 100 years ago, the changing truth value of that proposition

30  The Early Years has ontological significance for the philosophy of time. It implies some version of the A-theory of time, that time passes, construed either as temporal becoming (changing A-properties) or absolute becoming (changing existential status), or some other A-theoretic analysis of time. In Broad’s view of time, there are no A-propositions—whether construed as propositions that either don’t have a truth value or propositions that change their truth value with the passage of time. Since they can be eliminated or reduced to B-propositions, the reduction demonstrates that there are no A-properties, and there is no absolute becoming. Broad’s view of what kinds of temporal propositions or statements there are reflects both a view about language and a view about reality. The view about language is that the meaning of a tensed sentence can be given in terms of a sentence that does not contain grammatical tense or represent via the proposition it expresses an ontological tense. The view about reality is that tenseless sentences express propositions and facts ontologically tenseless since they do not contain or refer to the non-relational transitory temporal A-properties of pastness, presentness or futurity, but only earlier/later than and simultaneous with relations. Broad explains the token-reflexive account of tensed language in the following passage: If I say, “It is raining,” this verbal expression, since it clearly intends to refer to a particular event, is incomplete and stands for no definite proposition; for it says nothing about the time at which it rains. It therefore seems to be sometimes true and sometimes false. But, as actually asserted, the words would be taken to express my judgment of the proposition, “It is raining at the time at which I say, ‘It is raining.’” And this proposition is timelessly true or false, subject to a further correction which we shall add in a moment. In fact, whenever we are told that a proposition is sometimes true and sometimes false, we know that we are dealing with an incomplete statement about an event, and that the real state of affairs is that a propositional function of the form ‘e happens at t’ gives true propositions for some values of t and false propositions for other values. But the propositions themselves are timelessly true or false. (1921, pp. 147–148) At the beginning of the passage, Broad gives a token-reflexive analysis of tense. According to it, the tensed statement “It is raining” means that “It is raining occurs at the same time as or is simultaneous with my saying that ‘It is raining.’” Apart from the temporal context, the present tense statement “It is raining” does not express a proposition, and so it is neither true nor false. It neither corresponds nor fails to correspond to a fact and is, in that sense, an incomplete symbol. Thus, the verbal form of “It is raining” is misleading since it mistakenly suggests that tense is

The Early Years  31 an indispensable feature of the proposition expressed and if true, the fact recorded. Whereas the logical form requires reference to the time of utterance, and when it is, tense is eliminated from the proposition expressed and the reality (or fact) to which it corresponds. Broad raises an objection to the token-reflexive analysis. He argues that the meaning of “It is raining at the time at which I say ‘it is raining’” is ambiguous and the truth value is indeterminate until it is specified which token of “It is raining” is being referred to. Thus, it appears that we need to reintroduce the notion of now by specifying that we are talking about the present utterance of “It is raining.” In that case, we are once again dealing with a statement that expresses a proposition that is sometimes true and sometimes false with the ontological implications of tensed time. Broad’s response to this criticism of the token-reflexive analysis occurs in the following passage: It is important to notice that in practice there is always the possibility of any verbal statement about events, no matter how carefully put, being sometimes true and sometimes false . . . Take, e.g., the amended expression offered above: “It is raining at the time at which I say, ‘It is raining.’” To any reader of this article the expression remains ambiguous, because he knows the event that is used for dating only by the very ambiguous description, “The writer’s statement of the words ‘It is raining’”—a description which applies to dozens of different events. In practice the difficulty is solved in conversation by the fact that all the manifold circumstances under which the particular conversation takes place go into the description and make it practically unambiguous. In writing, the difficulty is solved practically by using as the origin of dates some event, such as the birth of Christ, whose full description is so complicated that it is almost certain that only one event answers to it. But the theoretical difficulty remains, and so we are tempted to say that any proposition about events is sometimes true and sometimes false. But the proper thing to say is that any verbal expression referring to events, no matter how carefully put, always runs a theoretical risk of ambiguity—i.e., it might with equal propriety make one reader think of one proposition which is true, and another of another proposition which is false. (1921, p. 148) Broad’s interesting reply is that there may be a theoretical ambiguity in the question of which token is fixing the time of the event in question, but it does not follow that the ambiguity reintroduces tense, or that the statement expresses a proposition with a changing truth value. Rather, Broad is claiming that the verbal (or written) expression runs the risk of having different readers apprehend different propositions with different

32  The Early Years truth values (since different people may think of the token that fixes the reference as occurring at different times). Nevertheless, in each case the proposition expressed or apprehended has an unchanging truth value; it is (in Broad’s terminology) timelessly true or false. To put the point otherwise, if I wish to communicate the information that “It is raining now,” then I can do that by saying that “It is raining simultaneously with (or at the same time as) the utterance of the sentence ‘It is now raining,’ or with this sentence,” where “this sentence” refers directly to the sentence being uttered. If it is uttered, then the circumstance will make it clear which token of “It is now raining” is being referenced. If it is written then it may not be entirely clear which sentence is being referred to, but even so, it does not follow that tense (presentness) must be introduced to capture the meaning since what the sentence expresses for whoever reads it is still a timeless proposition.1 For Broad, the significance of the reducibility of the meaning of A-­statements in terms of B-statements is ontological; reducibility demonstrates that McTaggart’s positive account of time is wrong—A-properties are not objective features of reality. In 1921, Broad clarifies this point by explaining what is meant by the A-theorist’s thesis that events are sometimes future, then present and finally past. The statement that “e is present” is incomplete and expresses the proposition that “e is present at time t.” According to Broad, this means (assuming that there are moments) That e is at the moment t in an analogous way to that in which an object is at a position in space. The statement “e is present at t” may be compared with the statement “Mr. Asquith is present at the meeting,” which means that his body is in the place where the meeting is held. In all complete statements of the form “e is at t” we must understand the word “is” as standing for a timeless copula and distinguish it from the ‘is’ of the present tense, which is contrasted with “was” and “will be.” Let us denote the “is” of the present tense by “is now.” Then the statement “e is now present” is an incomplete statement which is interpreted in use to mean “e is at (or occupies) the same moment as my assertion that it is now present”; “e is now past” = “e was present” = “e is at a moment earlier than my assertion that e is now past.” Similarly, “e will be present” = “e is now future” = “e is at a moment subsequent to my statement that e is now future.” The laws of logic are of course concerned with the timeless copula, and they presuppose that statements containing tenses are reduced in the way suggested above. (1921, pp. 148–149; emphasis added) The previous passage is potentially open to misunderstanding, and in Broad’s later writings about time, he does offer criticism of his earlier view that based on that misunderstanding. First, the analogy between

The Early Years  33 space and time—meaning that events are at times in a way that is analogous to how objects are in places or in space—is or may be problematic. Since all the positions in space occur at the same time, it might be concluded that positions (and objects) in time occur at the same time (as a simultaneous whole or totum simul) or at different times in a second time dimension. From this point of view, the world is an unchanging block, or a world of particulars that persist in a second time dimension. Second, the notion of a “timeless” copula is ambiguous as between the copula in eternally2 statements (“2 + 2 = 4”) and eternal3 statements (“a is earlier than b”). Failure to keep these two kinds of “timeless” statements separate can lead to the fallacious conclusion that the terms of temporal relations are abstract objects or universals. The possibility of blurring the differences between space and time, and a temporal series and a timeless or everlasting one, emerges with unfortunate consequences in Broad’s discussion of the similarities and dissimilarities between time and space, which I shall turn to next.

III. The Similarities and Dissimilarities Between Time and Space In the preceding section, I have shown how Broad appeals to the logic of temporal language to draw implications for the ontology of time. In this section, I shall show how Broad uses metaphysical arguments regarding the similarities between time and space to support the R-theory. Paradoxically, what he says in this section may have also planted the seeds for his rejection of the R-theory in subsequent writings. For in his attempt to argue that time is analogous to space, he foreshadows mistaken claims that the R-theory is committed to a static, block universe. I shall argue that though some interpretations of the B-theory are susceptible to these criticisms, they leave the R-theory unscathed. Broad summarizes the senses in which time is like space in the following passages that I shall quote at length:

(a) Likenesses 1. Most objects of immediate experience possess a kind of magnitude called extensity, and such objects stand in certain immediately recognizable relations to other objects of the same sense experienced along with them. Also, the parts of any one such object have relations of this kind to each other. Similarly, the objects of our experience have another kind of magnitude called propensity or duration. Such objects have to others of the same kind the relation of partial (or, in special cases, total) precedence and this relation can be recognized immediately. Likewise the parts of a single specious present can be seen to have this relation to each other. (1921, p. 145; my emphasis)

34  The Early Years 2. The relations in each case have magnitude. Just as one object in the field of view can be more to the right of another than a third, so one event in the field of memory or in the specious present can precede another event by a longer interval than some third one. 3. In each region there is the same close and peculiar connexion between the kind of magnitude possessed by the terms and the kind possessed by the relations. It is possible to say that the interval between two events A and B is as long as the duration of some event C, just as it is possible to say that the distance between two sticks laid in the same straight line is the same as the length of some third stick. 4. It is commonly believed that, when the analysis is made into moments and momentary events, all the events in the history of the world fall into their places in a single series of moments. So too it is supposed that, when the analysis is made into material and geometrical points, all the points in the world take their places in a single three-dimensional series of geometrical points. We shall consider later what the Theory of Relativity has to say as to the impossibility of separating time and space as to the notion of one single time series. (1921, p. 145, 173 fn) Regarding the first similarity between space and time, I think it is important to point out, as we already have, that the temporal relation of precedence, whether partial or total, can be recognized immediately, meaning that it can be the object of an act of direct acquaintance. This is important and it contrasts quite clearly with the distinction between past, present and future that, in Broad’s view, does not lie “wholly in the experienced objects, but one that rests on the relations between experienced objects and the states of mind in which they are experienced” (1921, p. 145). In other words, while we are acquainted with earlier/ later and simultaneity relations, we are not acquainted with the properties of pastness, presentness and futurity. Here Broad is propounding a psychological reduction of tense in terms of different temporal relations between certain psychological states of mind and the object of those states. If an object is earlier than a memory of it, then it is past; simultaneous with a perception of it, then it is present; and later than an anticipation of it, then it is future. The alleged differences between time and space Broad reduces to three: (1) It is commonly held that all events have temporal relations to each other, but that psychical events have no spatial relations. This is denied by a small number of philosophers, notably by Samuel Alexander. Without questioning the possibility of correlating psychical events with positions in space, we must hold that this alleged difference is a genuine one. (1921, p. 145)

The Early Years  35 Broad later expresses this by saying that the difference between space and time consists in the fact that judgments about time are in time, whereas judgments about space are not in space (1921, p. 149). In that sense, time is more fundamental than space since there are things that exist in time but not in space, but nothing exists in space that is not also in time. Broad claims that there are two other differences between time and space that are natural to make, but rest on confusions. He states them as follows: (2) A much more important point is that time is said essentially to involve the distinction between past, present, and future as well as that between before and after. Now nothing in space obviously corresponds to these distinctions in time. (3) Closely connected with this alleged difference are a number of rather vague statements often made—e.g., that parts of space co-exist, but that only the present moment exists. (1921, p. 145) Regarding difference (2), Broad argues that time is similar to space since the notions of past, present and future can be understood relationally as not requiring non-relational A-properties. Positions in A-time are analogous to relations between where we are located or near and elsewhere in space. He puts it: the distinction between present and not-present at any rate may be usefully compared with that between here and elsewhere in space. Here means near my body; elsewhere means distant from my body. If we want an analogy to the distinction between past and future, we can find one in the distinction between things before and things behind our body. It is true, however, that this analogy is incomplete, and that for an important reason, though one extraneous to the nature of time. The reason is that our practical and cognitive relations towards the future are different from those towards the past. We know a part of the past at any rate directly by memory, but we know the future only indirectly by probable inference. There is no analogy to this in space; our knowledge of what is behind our body is of the same kind and of the same degree of certainty as our knowledge of what is in front of it. But we may imagine that a distinction like that between past and future would have arisen for space also, if we had been able to see straight in front of us but had never been able to turn our heads or our bodies around. (1921, p. 335; emphasis added) Broad is here foreshadowing the account of “here” and “now” as indexicals that refer to the place and time at which they are tokened.

36  The Early Years When we say, “I am here,” we do not mean to assert that this place has the property of hereness, but rather that “here” refers to the place where I am located. Similarly, when we say an event is now, we do not mean to assert that the event has the property of nowness, but that it is simultaneous with my utterance or happens near or at the time I am located when I say that “It is raining now.” Indeed, just as we can experience objects to be here by unreflectively experiencing them as existing in close proximity to our bodies (spatial locations) without apprehending them as exemplifying hereness, we can experience an object as present by experiencing it as existing in close temporal proximity—roughly simultaneous—with an unreflective perception of it, without experiencing the object exemplifying presentness. In other words, since time itself does not essentially involve the notions of past, present and future, we can maintain a similarity between time and space by analyzing past, present and future as analogous to the relations of behind, at this location and in front of. I should add that the analogy between here and now, though useful to a point, leads Broad and B-theorists into difficulties for it may give rise to a pernicious spatialization of time. Since all the parts of a spatial whole co-exist regardless of what distance they are from me, it might appear to follow that all the events of a temporal whole co-exist regardless of their temporal distance from my location. Given this consequence, and the further claim “that past present and future events are always equally real” (Broad, 1921, p. 155), it is easy to move to the claim that the history of the world “always co-exists.” Thus, if the world’s history co-exists in a spatial sense, and for that reason, events co-exist at the same time, we are led to the view that the world is sempiternal, that is, a totality that always exists. But that is absurd: the world is not a totum simul or a simultaneous whole or totality that persists eternally1. In Broad’s view, A-properties are analyzed into facts regarding the temporal relations between mental acts, sentence tokens, utterances or dates, and the temporal objects—particulars or events—they are about, depending on which reduction he adopts. Thus, if facts that are the analysis of A-properties always exist, then the terms or relata of these relational facts always exist, with disastrous consequences. Broad warns against this confusion, but ultimately succumbs to it, as we shall see. The alleged third dissimilarity between space and time is that all the parts of space co-exist, but in time only the present exists. Broad attributes the alleged dissimilarity to a confusion between two different senses in which the parts of a whole “co-exist.” In one sense the parts of a spatial whole, say, a table, co-exist because they all exist at the same time. In another sense, the whole history of the world is a series of terms that exist in succession, but do not co-exist in succession. If one confuses these two senses in which the parts of a whole exist or co-exist, one may argue that since all the parts of time co-exist analogous to the way in which all the parts of space exist, then all of the (past, present and future) parts of

The Early Years  37 time co-exist simultaneously. Since that is absurd, one may conclude that all the parts of time—the past, present and future—do not exist, but that only the present exists. Admittedly, in ordinary language we say such things as the past no longer exists and the future does not yet exist, but that only the present exists, and this may seem to imply the presentist view that, as Broad puts it, “the present is a mere transition from one infinite non-existent and another” (1921, p. 150), but that would be a mistake. In the following passage, Broad argues that one can explain the sense in which the past no longer exists and the future does not yet exist without being committed to presentism. He also maintains that one can give a sense in which the parts of time exist is similar to the parts of space. Unfortunately, the way in which he explains these notions may succumb to the objection he is attempting to avoid, of spatializing time: It is perfectly true, of course, that the whole history of the world is not a complex of co-existing parts (in the sense of parts existing at the same time), as a table is. But this does not mean that it is not a whole, or that one part of it exists any less than any other part. To say that x no longer exists, or does not yet exist, simply means that it occupies a moment before or after my statement about it. At another moment I may make another statement of the same verbal form about x, and, since this no longer stands for the same proposition, it may no longer be true (i.e. no longer stand for a true proposition). But this involves no change in x itself. That x exists at a certain moment simply means that x occupies that moment, and this is timelessly true. Similarly, the fact that this moment has a certain temporal relation to any definite assertion that I make about x is timelessly a fact. That it has different and incompatible temporal relations to various assertions of the same verbal form made by me is also timelessly true and is not merely compatible with but also a necessary consequence of x’s existence at its own moment. An event must continue to be, if it is to continue to stand in relations; the battle of Hastings continues to precede the battle of Waterloo, and therefore both these events must eternally be at their own respective moments. (1921, p. 150; emphasis added) In this passage, Broad is clearly rejecting presentism, since he is rejecting the notion of absolute becoming—that events come into and go out of existence as they “move” from the non-existent future to the present and into the non-existent past. For Broad, different statements of the same verbal form as “x no longer exists,” uttered at different times, express different propositions that have different truth values, but that is all that is involved in the “change” from the non-existent future to the existing present and the non-existing past. If “x no longer exists” is uttered earlier

38  The Early Years than x, then the judgment is false; and if “x no longer exists” is uttered later than x, then the judgment is true. There is nothing more to it. Broad’s difficulties begin, however, with his claim that events have and “continue to have temporal relations . . . and therefore must eternally be at their own respective moments” (1921, p. 150). Broad is suggesting that the tenseless or timeless truth of propositions about temporal relations between events implies that events continue to have those relations. The notion of “continue” is, however, a temporal notion that means “remain in existence” or “persist through time.” Thus, if an event continues to stand in a temporal relation, then not only the terms of the relation, but the fact of which they are constituents must continue to exist, and therefore the totality of all such “facts” that constitutes the whole history of the world must also continue to exist—but that is the likeness of time to space he sought to avoid. The whole history of the world constitutes all of time so that it cannot exist in time, and much less at every time as it must if the totality of temporal facts that constitutes the history of the world continues to exist. To put the same point differently, if the whole course of history forms a related whole whose parts continue to have being, then each part would endure throughout all of time and so each term would exist forever. The parts of time would literally resemble the parts of space; they would constitute a whole whose parts all exist at the same time. Of course, if events are at their respective moments eternally1 or with unending duration, then they are simultaneous and not successive. Thus, the problem Broad rails against—of confusing the sense in which the parts of space and time “co-exist”—arises in his account of how to avoid it. It is always true that “the battle of Hastings precedes the Battle of Waterloo,” and therefore, whenever that sentence is asserted it expresses a true proposition. It does not follow, however, that the fact itself exists at every time (eternal1) or that it is timeless2 or eternal2 containing only timeless constituents. For if the terms “continue” eternally2 then they cannot exist as terms of temporal relations, since only temporal objects are terms of temporal relations, not timeless ones. Thus, to characterize events as “continuing to be” or “eternally” being at their own respective moments is open to misunderstanding. It suggests that the terms of temporal relations are either persistent particulars (eternal1) or timeless objects like numbers (eternal2), and both those interpretations are mistaken. An event or particular exists as a term of a temporal relation, as a constituent of an eternal3 fact that is outside of time, but that does not require or imply either the term or the temporal relational fact is eternal1 or are eternal2. Thus, to assert, as Broad does, that for events to “continue to have temporal relations . . . [they] therefore must eternally be at their own respective moments” (1921, p. 150) is fraught with difficulties. Broad warns against the mistake of treating the history of the world, or the entire world, as a spatial whole whose terms co-exist at the same

The Early Years  39 time, and yet, it seems that he falls prey to that conclusion. Consider, for example, the following passage: The fallacy which we have to avoid is that of confusing two different senses of co-existence. In one sense the parts of any related whole coexist; in another only those events that occupy the same moment of time co-exist. It is clear that the whole course of history does not coexist in the second sense, and it is thought that this prevents it from co-existing in the first. Yet this is necessarily false, since it is admitted that events do have and continue to have temporal relations, and therefore they must form a related whole all of whose parts have being. (See Meinong) The confusion is increased by the belief that past, present, and future are essential characteristics of objects in time in the same way as before and after are, instead of being analysable into the temporal relations of states of mind and their objects. (1921, p. 150; emphasis added) Broad has claimed that simple and unanalyzable temporal R-­relations are the ontological and epistemological ground of time, and in this, I believe, he is correct. Where he goes wrong is by insisting that to account for the unchanging and “timeless” or “eternal” character of R-relations and R-facts, he must treat the terms of R-relations, R-­relations ­themselves and the R-facts of which they are constituents as continuing to exist. Although Broad obviously would find this consequence repugnant since it implies the denial of R-relations, his remarks in the preceding passage and elsewhere in “Time” (1921) unwittingly embrace it, as for example, when he says, “past, present, and future are all always equally real” (1921, p. 155; emphasis added) and that “the whole course of events is in a certain sense a totum simul” (1921, p. 150). Both these remarks imply that R-facts and their constituents exist at the same time, which is absurd, given that R-facts involve the succession of temporal objects, and so are eternal3. Broad argues that the whole course of history does not co-exist in the sense that the simultaneous parts of a table co-exist, and yet in a footnote he clarifies the analogy between space and time by claiming that “The point can perhaps be made clearer by reflecting that a tune has a pattern in time in exactly the same sense as a wallpaper has a pattern in space” (1921, p. 173, n 3). This can easily be read as spatializing time—as treating time as space-like—since the wallpaper representation treats times as spread out in space, as a static representation of a dynamic process; and so as static and not dynamic. For that reason, it is not difficult to understand how interpreters can construe Broad as guilty of the error of spatializing time. The tendency of B-theorists to spatialize time exacerbates that confusion. For instance, Donald C. Williams, in his classic paper “The Myth

40  The Early Years of Passage” says, “Time ‘flows’ only in the sense in which a line flows or a landscape recedes to the west. That is, it is an ordered extension” (Williams, 1954, p. 463). In the R-theory, however, time is fundamentally different from space since primitive temporal relations are dynamic, and thus intrinsically different from spatial relations. One way to avoid confusing time and space (or spatializing time) is to distinguish between a dynamic, temporal series, such as a tune played on a piano, and a static, spatial representation of a temporal series, such as a tune on a sheet of music. In other words, to distinguish a representation of a temporal series from what it is a representation of.2 To clarify, Frank Ramsey, in a handwritten manuscript subsequently published as “Ramsey’s ‘Note on Time’,” (Ramsey, 2006), distinguishes two ways of understanding the way events are spread out in time, and warns against a confusion—easy to make—between dynamic time itself and a static representation of it. He explains how the different ways of representing a temporal series may mistakenly lead us to think that we need to add something more to a sequence to make it temporal. Ramsey notes that when we try to imagine a series of events in time, we can do this either by going “through the events one after another in the order in which they happened, as when one rehearses a tune in one’s mind” (2006, p. 157). Or we may want to have all the events in our minds at once in order to better see their relations, we then imagine them spread out before us along a line like the notes in a score. (2006, p. 157) The resulting series is, of course, a spatial representation of a temporal series, and then qua spatial series the “sense” of time from earlier to later is lost. If, however, one fails to distinguish these two understandings of time—a representation of temporal phenomena and the phenomena itself—or confuses the dynamic R-relation of “earlier than” with its static spatial representation of points on a line or time qua notes in a score, one may conclude that we must introduce something, for example, the NOW, over and above the series of notes spread out in space as something that “moves” and gives the series its sense and direction, as some A-theorists do. To counteract these moves which lead to unpalatable consequences Ramsey continues, But clearly the whole difficulty is a mistake; the events are really in temporal order one before the other; each is present to or simultaneous with itself, future to the preceding one’s past to the subsequent. The moving present is really the series of events themselves; only when the temporal series is replaced in imagination by a spatial

The Early Years  41 series, do we try to restore its temporal quality by introducing presentness from the outside. This is not to say we cannot legitimately represent a temporal series by a spatial one, provided we are prepared to keep to it to all (say) “to the left of” to stand by convention for “before” and not attempt simultaneously an imaginative realization of the temporal relationship. (2006, pp. 157–158; emphasis added) While I certainly agree with Ramsey in maintaining that the representation should not be confused with the represented, the task of providing an ontological ground for temporal relations still remains. I have already suggested, as has Broad, that for a Russellian ontologist, simple and unanalyzable relational universals between particulars or events (not absolute moments) are sufficient to do that. C. W. K. Mundle, in his critical essay on Broad’s changing philosophy of time (1959), accuses Broad and Russell of spatializing time. Mundle argues that Broad spatializes time when he says, “past present and future are analyzable in terms of temporal relations between states of mind and their objects” (Broad, 1921, p. 150). For, Mundle claims that what Broad says here is very reminiscent of the view expressed earlier by Russell: Past, present and future arise from time-relations of subject and object, while earlier and later arise from time-relations of object and object. In a world in which there was no experience, there would be no past, present or future, but there might well be earlier and later. (Russell, 1915, p. 212) Broad seems, like Russell, to have been thinking spatially of time, conceiving “objects” (events) as constituting a linear series along which “subjects” travel, and assuming that “past” “present” and “future” refer to external relations between subjects and object. The model implicit in such thinking seems to be that which Broad describes in his later accounts as the policeman’s bull’s-eye theory. (Mundle, 1959, p. 356) The idea that past, present and future are analyzed in terms of a self or consciousness moving along “a linear series” grossly misunderstands Russell’s and Broad’s early view. First and foremost, the subject and object are parts of the same time-series—the self or consciousness does not exist outside of it. Admittedly, the subject has different psychological attitudes and different temporal relations to the objects it cognizes, thereby determining whether the objects are past, present or future. However, that is not because consciousness moves like a flashlight along a non-temporal series of objects. In the policeman’s bulls-eye theory, the row of houses

42  The Early Years along which the flashlight moves is a non-temporal spatial series, and the earlier than relation and the direction of time is analyzed or grounded in terms of the movement of presentness along a non-temporal series. On the R-theory, R-relations are not analyzable in terms of the movement of presentness or anything else. The temporal series and the R-relations that generate it are primitive time relations. For that reason, the moving spotlight theory cannot be used to characterize Broad’s (1921) or Russell’s (1915) analysis of time. Mundle’s way of understanding the Russellian view, and Broad’s sometimes misleading statement of Russell, has led to the R-theory being characterized as a block universe, a static view, and anti-realist because it denies temporal passage. All of these characterizations imply the criticism that something is missing from time, namely, succession. Thus, it is not surprising that in characterizing the B-theory Richard Gale says, “This is the static or tenseless way of conceiving time, in which the history of the world is viewed in a God-like manner, all events being given at once in a nunc stans” (Gale, 1968, p. 67). In this understanding, time involves neither change nor succession, an interpretation Broad lends credence to when he likens the history of the world to a totum simul (1921, p. 150). To view time as a block, or as space-like, is to deny the transitory aspect of time and implies that temporal passage is an illusion, or a minddependent characteristic of experienced objects with no objective validity. It is argued that the block view eliminates from time what is essential to it. As events neither have the intrinsic properties of pastness, presentness or futurity, nor undergo either temporal becoming or absolute becoming, the transitory aspect of time is eliminated from the nature of time. Since the transitory aspect is arguably essential to time, it is thought by the later Broad (1938) that the Russellian view leaves out something essential to the nature of time. Indeed, without the transitory aspect what is left is an ordered series whose terms no more compose a temporal series than do the series of points on a line or numbers in order of magnitude. In this interpretation, time is an unchanging block that is space-like or timeless. In the next section I shall offer an R-theoretic response to this criticism.

IV.  The R-Theory of Time For the Russellian, temporal relational states of affairs or R-facts—while they are not themselves in time—are indeed temporal since they contain temporal relations, and it is the relation of succession between temporal objects that is the ground of temporal passage and the transitory aspect of time. R-facts are entities in their own right over and above their constituents, and as such, they are not in time in that they do not exemplify non-relational temporal properties (since there are none), occupy moments (since there are none) or stand in temporal relations. In that sense, time, understood as a Russellian series composed of a conjunction

The Early Years  43 of R-facts, is a timeless3, eternal3 conjunctive fact. This view gives some meaning to an aphorism I favor, namely, time is timeless, or eternal3 in just this sense: though time contains temporal relations, time does not exemplify them. Whether Broad thinks of temporal relational facts as I do, namely, as entities that do not exist in time but contain time as a constituent, is not entirely clear. He makes claims such as that time is similar to space, that temporal relational facts always exist, and that “the whole course of events is in a certain sense a totum simul” (1921, p. 150), and they are certainly liable to be interpreted as spatializing time. However, in emphasizing the primitive and irreducible nature of temporal relations with which we are directly acquainted, together with his denial of A-­properties, absolute becoming and presentism, and affirming the eternal3 nature of the facts about temporal relations between and among objects, Broad is propounding a Russellian theory of time. In addition to the accuracy of an exegesis of Broad’s text, what is important is the notion that facts, as I understand them, are capable of grounding both the transitory and eternal3 nature or “timelessness” of time. These facts are not eternal2 in the sense of literally having constituents that are all timeless, as say, Platonic universals are. Nor are these facts eternal1 in the sense of having everlasting existence; they do not exist at every time or forever. Rather, temporal relational facts are eternal3 in the sense that while they do not exist in time, since they do not stand in temporal relations to anything, temporal relations are constituents in R-facts, and for that reason are temporal facts. Given a realist ontology of R-relations and R-facts, the Russellian theory can account for both the dynamic and the eternal aspects of time. That both aspects are part of the reality of time is stated by early R-­theorist R. M. Blake in his critique of Broad’s growing block theory, to be considered in the next chapter. In Scientific Thought (1923) and elsewhere, Broad maintains that the Russellian view leaves out something that is fundamental to the nature of time. Blake raises and responds to Broad’s objection in the following passage that I shall quote at length. His [Broad’s] concept of an unanalyzable “becoming” is very similar to M. Bergson’s equally ultimate “duration” and to Mr. Whitehead’s “passage of nature,” or “moving on” (Concept of Nature, p. 54). As Mr. Broad says (p. 59), “We are naturally tempted to regard the history of the world as existing eternally in a certain order of events.” The trouble with this is that it seems to take the temporal character of succession out of time and to make it “static,” or, as M. Bergson puts it, to “spatialize” time. Now there seems to me to be a strange mixture of truth and illusion in all this. There is certainly a unique character about time which cannot be reduced to anything else. Time is filled with “events, and events are happenings,” things that “come

44  The Early Years to pass,” that succeed one another in a fixed direction of earlier and later. This feature of time is revealed to us in our immediate experience of duration and the passage of events. But we may be equally certain that, however much of succession there may be in events, every event has in the order of succession just the place that it has and none other. The order as a whole, however much it may be an order of change and of succession, must in a sense be “static,” for it must be true that it is what it is. Let fluidity be never so fluid, the fact that it is so remains unaltered. These are simply the necessities of logic. (Blake, 1925, pp. 434–435; my emphasis) Blake is attempting to reconcile the fundamental features of our experience of time and its true nature. On the one hand, time has a unique and irreducible character that distinguishes it from space that is revealed to us through our experience of the succession of events. On the other hand, time forms an ordered series of terms in which every item has just the place it has and no other, and that the whole conjunction of facts is unchanging. How are we to reconcile the “eternal” character of time with the dynamic character of our immediate experience? I am suggesting that to understand these two aspects of time, one must recognize that time contains timeless yet dynamic R-relations and temporal yet eternal3 R-facts. To countenance R-facts in addition to R-relations is crucial to providing a ground for the unchanging character of time as a whole, and the dynamic nature of time within it. Although Broad is not always as clear as he should be, I believe that like Russell’s (1915), Broad’s metaphysics of time in 1921 differs from McTaggart’s since Broad rejects A-characteristics as the ground of the transitory aspect of time, and that he uses our experience of temporal facts (the phenomenology of time), the logic of temporal language and the ontology of time to establish that point. For McTaggart, that event e is present, was future and will be past, expresses the propositions that e exemplifies presentness at a present moment of time, futurity at a past moment of time and pastness at a future moment of time—whereas for Broad, there is no such implication. For Broad, sentences that express the passage of time are analyzed and reduced to sentences which contain no tenses and require no A-determinations, and therefore do not express propositions that change their truth value or represent or correspond to tensed events or tensed facts that exemplify A-properties. Tensed sentences are incomplete symbols that, once completed, represent unchanging temporal relations between events and the psychological or sentence tokens about them. An event e is past if and only if the time at which the sentence that asserts “e is past” is later than the event, present if the time at which the sentence “e is present” is simultaneous with the event and future if the sentence that asserts that “e is future” is earlier

The Early Years  45 than the event it is about. Thus the “change” an event undergoes when it “moves” from the future to the present to the past is no more and no less than the transition from the times at which a future tense utterance is earlier than the event, to a present tense utterance is simultaneous with the event, and to a past tense utterance is later than the event. It is the succession of different sentence tokens or cognitive acts that have different and unchanging temporal relations to one and the same event that constitutes the reality that underlies temporal becoming, and our experience of it. Note, however, that the passage of time stands for two different kinds of phenomena. The first is simply the phenomena of transition of succession. The second is an R-theoretic account of temporal becoming in terms of a succession of different sentence tokens or psychological attitudes toward the same event. While the utterances or cognitive states occur in succession, the facts that reflect the change of an event from future to present to past do not change, since in reality the event itself does not change when it undergoes temporal becoming (even if becoming is a kind of change). It is a single mind having successively different psychological attitudes toward events that occur in a succession. Richard Braithwaite expresses this idea nicely: Thus, change in persistent objects is to be analyzed into a succession of events which are the states of the object, change of the time of events into a succession of events which are experienced by one conscious mind. The specifically temporal element in each series of events is the relation of succession, which is directly experienced and has an intrinsic sense given in experience. I cannot see that there are any facts about change that are unaccounted for on this theory . . . This theory makes the change of the time of events consist of a change in the experiences of an observer and can in a sense be called a subjective change. (Braithwaite, 1928, p. 171) Clearly, then, if there were no experienced objects there would be no distinction between past, present and future events. In that sense, A-­determinations are not objective characteristics of events, as McTaggart would have it, if time were real, but it does not follow that past, present and future are illusions or non-relational properties of the experience of an observer. In fact, Broad denies this. For Russell and Broad, nonrelational A-properties do not exist ontologically or epistemologically— as subjective or objective. The passage of events or objects from the future to the present to the past is not a change in the temporal characteristics of the event in question, as the A-property view maintains, but a change in the psychological attitudes of a single conscious mind toward the same event, each from a different temporal perspective. Thus, in the R-theory—with or without

46  The Early Years consciousness—nothing is really or intrinsically past, present or future. As Bertrand Russell put it in Mysticism and Logic: Every future will someday be past: if we see the past truly now, it must, when it was still future, have been just what we now see it to be, and what is now future must be just what we shall see it to be when it has become past. The felt difference of quality between past and future, therefore, is not an intrinsic difference, but only a difference in relation to us: to impartial contemplation, it ceases to exist. (Russell, 1957, pp. 20–21) Nor does temporal passage consist in absolute becoming as the presentist conceives of it, namely, as a transition from non-existence to existence and vice versa. In the R-theory, temporal transition or passage is understood in terms of succession, and if one thing succeeds another, they must both exist. For that reason, becoming, understood in terms of something emerging and passing away, is not a matter of successively changing with respect to existence. Rather, to paraphrase Erwin Tegtmeier, to emerge is to have a beginning and to pass away is to have an end. The beginning of a thing is its earliest temporal stage and the end is its latest temporal stage. This non-existential (non-presentist) view of becoming is compatible with the relational theory that nothing can acquire and shed existence. In the R-theory, temporal relations and the facts those relations enter, for example, a is earlier than b, constitute transition or passage. The Russellian view maintains that the ground of the passage of time is succession; a relation between objects that does not itself exist in time. To be in time is to be a term of a temporal relation, but in the R-theory, temporal relations are timeless since it is impossible that the earlier than relation is earlier than or later than itself, since only particulars can stand in temporal relations. Moreover, even though the R-fact that, say, a is earlier than b does not change, it is not the case that the fact is permanent in the sense that it always exists, or exists at every time. For facts, too, are timeless, since they are never a term of a temporal relation. Thus, since the essence of time is passage, the foundations of which are timeless, temporal relations and temporal eternal3 R-facts, we may conclude that although time contains temporal passage, the passage of time and indeed time itself is timeless in the sense of being outside of time. To countenance temporal relational facts and temporal relations is crucial to providing a ground for the unchanging (timeless) character of time as a whole and the dynamic flow or passage of time within it. J. S. Mackenzie, in an encyclopedia entry on “Eternity,” stated the connection between the timeless and the transitory aspects of time well when he wrote, There is no time outside the process. Hence the process as a whole might be said to be eternal though every particular part in it has a

The Early Years  47 place in time. . . . The process, when we thus conceive of it, is not in time, rather time is in the process. Time is simply the aspect of successiveness which the eternal process contains. (Mackenzie, 1955, p. 404)

V. The Specious Present, Our Awareness of the Past, Present and Future, and the Phenomenology of Temporal Passage Broad mentions the “specious present,” which is intended to refer to a brief span of time that is the duration of a single awareness. The length of this awareness can vary from one person to another. According to Broad, we can be aware of the duration of an object in a specious present and can immediately recognize the relation of partial (or, in special cases, total) precedence between the parts of a single specious present. The awareness of temporal relations also applies to our conscious states. As Broad puts it, If in introspection we do contemplate our states of mind in the same sense as in perception we contemplate other objects, it seems clear that our states of mind show no trace of being extended or standing in spatial relations but do have duration and stand in temporal ones. (1921, p. 143) Therefore, based on experience, time has two kinds of intrinsically temporal entities, namely temporal relations and durations, such as lasts as long as and lasts longer than. Hence temporal relations are simple entities not reducible to A-properties. Temporal relations are the objects of direct awareness and therefore exist. These relations, as given in experience, “connect . . . events that have some duration, and not momentary events or moments. We are not directly aware of events without duration and therefore are not directly aware of the relations between such objects” (1921, p. 143). Broad maintains that we have immediate knowledge of the past and the present, and introduces the notion of the specious present to explain how this is so. The specious present is the length of time we are aware of in a cognitive act. It is called “specious” because what is cognized is given to us as present, but since the object of cognition is not instantaneous or punctiform, but has a duration—it is not really present. Broad says that: If an object be known directly by a state of mind which succeeds it by more than a certain short time t, which seems to be fairly constant for a given individual, the state counts introspectively as a memory, and the object is judged to be past. If the period between the object and

48  The Early Years the direct awareness of it be not greater than t, the awareness does not count for introspection as a memory, and the object is judged to be present. To say, then, that an object has been present and is now past means that (a) it is (timelessly) the object of an immediate awareness which succeeds it by less than t, and (b) that my statement “It is now past’” succeeds it by more than t. (1921, pp. 149–150) For Broad, we can have direct or immediate knowledge of an object or event that lies outside of our specious present, i.e., an event that is earlier than the whole of the specious present, and this occurs in memory. We can also have an immediate awareness or consciousness that is a short time t later than the object, and this is the experience of the present. These experiences reflect our experience of the passage of the object from the present to the past. Thus, judgments that claim that an event or object is past do not require anything more than temporal relations between the object and a mental act to serve as their phenomenological basis and ontological ground. At different times we have different cognitive attitudes and temporal relations to one and the same object or event. While this supports Broad’s view that the distinctions between past, present and future do not confirm a difference in the ontological status of such events—either their existence or their monadic properties—it does not support the view, implied by some critiques, that Broad spatializes time or treats time as space-like (in which all events in the history of the world co-exist at the same time). Although Broad analyzes the presentness of an object or event in terms of a temporal relation between a state of mind and its object, he gives a different account of the presentness of a state of mind. He says: We have still, however, to consider what is meant by the presentness of a state of mind. This seems to mean that, if a state of mind be the object of an act of introspection which succeeds it by less than a certain short period, the state presents a certain peculiar characteristic which it does not present to any later act of introspection. (1921, p. 404) This is a strange claim for Broad to make and foreshadows his later view of the specious present, since it seems to suggest that states of mind are present in virtue of having “a certain peculiar characteristic,” but what can that characteristic be? It is certainly not presentness, since he rejects such a property. This may be the first indication of “presentedness,” a characteristic he introduces later in his account of the specious present in Examination of McTaggart’s Philosophy. If it is the characteristic of “being present” or presentness, then while objects of all states of mind except introspection are not intrinsically past or future, the presentness

The Early Years  49 of a state of mind is a further quality that the state of mind exemplifies. While this is compatible with the mind-dependence of temporal characteristics, it is not compatible with the Russellian view, for it suggests that states of mind (or mental acts themselves) have temporal characteristics that they don and doff with the passage of time. Of course, this will lead to precisely the problems that McTaggart sought to draw our attention to in his argument against the reality of time. If states of mind have tensed properties—or only the property of presentness—that they lose and thus are not apprehended at a later or earlier time, then his analysis of the meaning of past, present and future in terms of token reflexives will not work. Since the state of mind itself changes its temporal qualities and not merely its temporal relations to a later or earlier introspective state, there is a type of change not recognized by the Russellian view, and for that reason Broad’s account of the presentness of a state of mind is incompatible with the generally expressed Russellian theory (Broad, 1921).

VI. The Reality of Time and Broad’s Response to McTaggart’s Argument for the Unreality of Time Broad emphasizes that “we are not directly aware of events without duration, still less of moments of empty time” (1921, p. 143). Broad does not infer that there are in nature no momentary events or moments of time, but only that they cannot be the building blocks of the ontology and epistemology of time. Thus, he maintains that we must start with the experienced facts of time and then make inferences and develop the scientific notion of time from them. These characteristics are given in experience and cannot be denied. This is an important point about the reality of time and hence the impossibility of the unreality of time. For this reason, all arguments against the reality of time must be mistaken. In considering arguments against the reality of time, Broad makes an important point about the limits of what they can hope to prove. To be successful, any account of time cannot be inconsistent with our experience, direct awareness or acquaintance. In other words, temporal characteristics cannot be entirely removed from reality since they are given to us in experience. As Broad argued in the passage quoted in Chapter 1, at most we can say is that some or all temporal characteristics do not characterize reality exactly as they would appear to do in a particular philosophical analysis. The immediate objects of some states of mind have temporal characteristics. What may not be real is the description or analysis we give to these direct objects or the inferences we draw from them about objects that are not the direct objects of states of mind, or are constructions based upon them, since in all these matters, we may be mistaken. To question whether our perceptions are veridical (or erroneous) in that sense is one thing. To question whether our perceptions are veridical in the sense of picking

50  The Early Years out something that is contained in reality in the broadest sense in quite another. Broad’s point is that temporal characteristics are features of reality, and that cannot be denied. With that background, he turns to arguments for the unreality of time. Broad first formulates an argument for the unreality of time that he attributes to Leibniz, and then dismisses it in the following passage: Arguments against the reality of time which turn on the distinction of past, present, and future may be dealt with shortly. One argument asserts that the past and the future do not exist, and that the present is a mere point without duration. It is then supposed that what occupies no finite duration cannot be real, and this disposes of the present. An argument of this kind is used by Leibniz against absolute time, though it would presumably apply to events just as well. It is met, of course, by the consideration that past, present, and future are all always equally real, and that these characteristics do not belong to events as such, but in virtue of the temporal relations between them and certain psychical events. (1921, p. 155; emphasis added) Broad’s refutation of this argument is based on two claims: that all times or events (past, present and future) are always equally real, “and that these characteristics do not belong to events as such, but in virtue of the temporal relations between them and certain psychical events” (1921, p. 155). Of course, and unfortunately, Broad talking about past, present and future as always being equally real is problematic since (a) in his view nothing really is past, present or future; and (b) the notion of “always” is a temporal notion and that suggests that either events are continuants, or that timeless facts are temporal, and in either case we are liable to transform Broad’s Russellian view into dialectically unacceptable ontology. In spite of Broad’s sometimes misleading way of expressing his R-view, I think there should be no doubt where he stands regarding the ontological explanation of our changing thoughts about time. The debate may be stated as follows: Are different R-series relations between our different psychological attitudes toward the same event the foundation and origin of our experience of the A-series change from future to present to past, or are changing A-series properties of pastness, presentness and futurity the foundation of changing psychological states? In other words, do we dread an unpleasant event because it has futurity and is moving toward us, and feel relief that the event is over because it is changing from the having presentness to having pastness, or do we experience certain unpleasant events with dread and relief because we are conscious of those unpleasant events as later than and earlier than our different psychological attitudes toward them? In short, is the R-series or the A-series more basic in our ontological analysis of the emotions and feelings toward events? In

The Early Years  51 rejecting A-characteristics in terms of R-relations, I think Broad would say that it is the different cognitive relations to earlier and later events that is the ground of the attitudes rather than the attitudes being evidence for the reality of A-properties. Broad next considers McTaggart’s paradox. He first briefly characterizes McTaggart’s argument as follows: Every event is past, present and future, which is a contradiction since these are incompatible properties. McTaggart’s “way out” of the contradiction is to claim that no term has those properties at the same time, but rather, say, e is now past, has been present and future. It would seem there is no contradiction in those statements, but since upon analysis those statements make use of the tenses, McTaggart argues that the result is another series with incompatible temporal properties, and so on ad infinitum. Broad dispenses with McTaggart’s paradox by showing that there is no contradiction to begin with once a correct analysis (or meaning) of the tenses is provided. For example, Take a definite statement by McTaggart that Queen Anne’s death is now past and has been present and future. Suppose we interpret this to mean that Queen Anne’s death is not the direct object of any awareness (even a memory) which is contemporary with McTaggart’s statement, but that it is contemporary with some states of mind (e.g., Lord Bolingbroke’s) which precede McTaggart’s statement; and that it is later than some thoughts about it (e.g., William III’s), which also precede the statement. Then those three propositions seem to be timelessly true, perfectly compatible, and to contain all that is meant in the assertion by McTaggart that Queen Anne’s death is past and has been present and future. (1921, p. 156) In other words, for McTaggart, but not for Broad, there are intrinsically temporal A-characteristics. For Broad, but not for McTaggart, A-­properties are analyzable, reducible or definable in terms of R-relations between psychic events (or states of mind) and the events that are claimed to be past, present or future. For example, McTaggart’s statement “Event e (Queen Anne’s death) is now past” means that Queen Anne’s death is not contemporary (simultaneous) with (a direct object of) a state of mind that is simultaneous with McTaggart’s statement that “Queen Anne’s death is now past.” But that in itself is not enough to make Queen Anne’s death now past (earlier than McTaggart’s statement) and opposed to now future (later than McTaggart’s statement). However, if we add that an awareness of Queen Anne’s death is simultaneous with a state of mind that is earlier than McTaggart’s statement, and her death is later than a still earlier thought about it, then we have expressed what is meant by McTaggart’s statement that “Queen Anne’s death is now past and

52  The Early Years has been present and future,” and have provided an ontological ground for it. Although Broad is one of the first B- or R-theorists to respond to McTaggart’s paradox in this way, he came to reject both his account of the meaning of tensed statements and the ontology of time propounded in his article on “Time.” In what follows, I shall examine the various alternatives Broad puts in their place, and the reasons he gives for them.

Notes 1. For a defense of the token-reflexive analysis of the tenses, see Orilia and Oaklander (2015). For a variety of papers that discuss the debate between the old and new B-theory of time, see Oaklander and Smith (1994) and Oaklander (2008, Vol. I). 2. This distinction is nicely made in Savitt (2002). I critically discuss Savitt’s paper in Oaklander (2004).

References Ayer, A. J. (1956). The problem of knowledge. London: Macmillan. ———. (1972). Statements about the past. In Philosophical essays (pp. 167–190). London: Palgrave Macmillan. Blake, R. M. (1925, Oct.). On Mr. Broad’s theory of time. Mind, 34 (136), 418– 435. https://doi.org/10.1093/mind/XXXIV.136.418 Braithwaite, R. B. (1928). Symposium: time and change. Aristotelian Society Supplementary Volume, 8 (1), 162–174. Broad, C. D. (1921). Time. In J. Hastings et al. (Eds.), Encyclopedia of religion and ethics (Vol. 12, pp. 334–345). New York: Scribner’s. Reprinted in L. N. Oaklander. (Ed.), (2008). Vol. I. The reality and language of time (pp. 143– 173). References are to this reprint. ———. (1923). Scientific thought. Chap. II. The general problem of time and change (pp. 53–84). London: Routledge and Kegan Paul. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. II. Time and metaphysics (pp. 63–83). References are to this reprint. ———. (1938). Examination of McTaggart’s philosophy: ostensible temporality (Vol. II (1), Chap. XXXV, pp. 265–323). Cambridge: Cambridge University Press. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 36–68). References are to this reprint. ———. (1952). Scientific thought. London: Routledge & Kegan Paul (Originally published in 1923). ———. (1959). A reply to my critics. In P. A. Schilpp (Ed.), Philosophy of C. D. Broad (pp. 709–832). New York: Tudor Publishing Company. Frege, G. (1956). The thought: a logical inquiry. Mind, New Series, 65 (259), 289–311. https://doi.org/10.1093/mind/65.1.289. Reprinted in M. Beaney. (1997). The Frege reader (pp. 323–345). Cambridge, MA: Blackwell Publisher. Gale, R. (1968). The language of time. London: Routledge. Mackenzie, J. S. (1955). Eternity. In J. Hastings et al. (Eds.), Encyclopedia of religion and ethics (Vol. 5, pp. 401–405). New York: Scribner’s (Originally published in 1913).

The Early Years  53 Marsh, R. C. (Ed.). (2001). Bertrand Russell. Logic and knowledge: essays 1901– 1950. London: Routledge (Originally published in 1956). McTaggart, J. M. E. (1908, Oct.). The unreality of time. Mind, 17 (68), 457–474. https://doi.org/10.1093/mind/XVII.4.457. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 21–35). References are to this reprint. ———. (1927). The nature of existence book II (C. D. Broad, Ed.). Cambridge: Cambridge University Press. Mundle, C. W. K. (1959). Broad’s views about time. In P. A. Schilpp (Ed.), The philosophy of C.D. Broad (pp. 353–374). New York: Tudor Publishing Co. Oaklander, L. N. (2004). Absolute becoming and the myth of passage. Philo, 7 (1), 36–46. doi:10.5840/philo2004713 ———. (Ed.). (2008). The philosophy of time: critical concepts in philosophy (Vols. I–IV). London: Routledge. Oaklander, L. N. & Smith, Q. (1994). The new theory of time. New Haven, CT: Yale University Press. Orilia, F. & Oaklander, L. N. (2015). Do we really need a new B-theory of time? Topoi, 34 (1), 157–170. https://doi.org/10.1007/s11245-013-9179-6 Ramsey, F. P. (2006). Ramsey’s ‘note on time’. In M. C. Galavotti (Ed.), Cambridge and Vienna: Frank P. Ramsey and the Vienna circle (pp. 155–166). Dordrecht, The Netherlands: Springer. Russell, B. (1906, Apr.). MacColl, H. Symbolic logic and its applications. Mind, New Series, 15 (58), 255–260. ———. (1914, Jul.). Mysticism and logic. Hibbert Journal. Reprinted in (1957). Mysticism and logic (pp. 1–31). New York: Anchor Books. References are to this reprint. ———. (1915, Apr.). On the experience of time. Monist, 25 (2), 212–233. https:// doi.org/10.5840/monist191525217. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. 1. The philosophy of time (pp. 174–187). Savitt, S. (2001). On absolute becoming and the myth of passage. Royal Institute of Philosophy Supplement, 50, 153–167. doi:10.1017/S1358246100010559. In C. Callender (Ed.). Time, reality and experience. Cambridge: Cambridge University Press. Williams, D. (1954). The myth of passage. Journal of Philosophy, 48 (15), 457– 472. doi:10.2307/2021694

3 The Middle Period

The Middle PeriodThe Middle Period

The Growing Block Theory of Time

I. Introduction There is clearly a difference between Broad’s views on time as reflected in his article on “Time” (1921), and in Chapter II “The General Problem of Time and Change,” in Scientific Thought (1923). The difference concerns his analysis of temporal relations and temporal becoming. The reason for his shift is somewhat unclear because in his later work, he does not directly address the question of why he changed his views on time so fundamentally. In a recent paper, Emily Thomas (2019) gives a convincing historical explanation of Broad’s shift, namely, the influence of Samuel Alexander on him. However, I think the influence of McTaggart— with whom Broad had a close relationship, both as a student and a ­colleague—is also clear. Broad’s analysis of temporal relations in terms of tensed notions is basically the same as McTaggart’s in The Nature of Existence (1927, p. 271). In addition, structurally, the seeds of Broad’s later views of time can be found in his ambiguous account of temporal relations in “Time” (1921). In “Time,” Broad maintains that primitive temporal or R-relations are given in experience and are the building blocks of all other aspects of time, both ontological and epistemological. R-relations can account for both the “eternal” and the “dynamic” aspects of time, and in the Russellian view, the direction of time is primitive; not founded upon any property of a term or terms of the relation, or some other notion such as primitive, absolute becoming. In Scientific Thought, rather than treat temporal relations as the building blocks of time, Broad maintains that A-determinations provide the ground of temporal relations and that becoming, understood in terms of the coming into existence of what previously did not exist, is the ground of A-determinations. Thus, in his discussion of time in 1923 (1923) he rejects unanalyzable R-relations countenanced earlier in 1921 (1921). Recall that in “Time,” Broad argued that we can understand the notions of past, present and future by analogy with the spatial relations of behind, in the immediate vicinity and in front of. Of course, this is not

The Middle Period  55 to say that R-relations that explain the tenses can be reduced to spatial relations, but rather that non-relational temporal properties are no more needed to account for the phenomenological meaning of “past,” “present” and “future” than non-relational spatial properties are needed to account for the phenomenological meaning of “here,” “there” and “elsewhere.” There is no presentness or pastness in the ontology of time any more than there is hereness or thereness in the ontology of space. In Scientific Thought, Broad is again concerned with how far the analogy of space and time can go. Thus, he wonders if the notions of past, present and future have a spatial analog, for example, can “now” be understood in terms of “here”? He also considers whether the changing tenses can be understood as analogous to a spotlight (presentness) moving along a fixed row of houses (events), or whether past, present and future can be reduced to cognitive relations between events and the minds that experience them. Broad argues that none of these alternatives provide an adequate account of the tenses. I contend that these arguments for the irreducibility and ineliminability of the tenses rest on the assumption that temporal relations are analyzable in terms of tensed notions, and for that reason they beg the question against Broad’s earlier R-theory. I shall also argue that the same assumption infuses Broad’s account of the growing block theory of time with problematic consequences. To begin to see what is involved, I will first consider and then critique Broad’s initial account of temporal relations.

II. A Critique of Broad’s Preliminary Analysis of Temporal Relations in Scientific Thought In Scientific Thought, Broad maintains that the intrinsic sense or direction of a temporal series, that is, the difference in a series of events a, b and c going from (c is earlier than b is earlier than a) rather than from (a is earlier than b is earlier to c) depends on the (changing) ontological tense of their terms. As he puts it: Now the intrinsic sense of a series of events in Time is essentially bound up with the distinction between past, present, and future. A precedes B because A is past when B is present. (1923, p. 66; emphasis added) We shall see that this assumption renders any analogy between space and time impossible and represents the first of his writings where the shift from his Russellian to his McTaggartian view takes place, since the A-series becomes primary and the B-series derived. Before exploring that shift, I want to point out a problem with Broad’s analysis that mirrors a problem with McTaggart’s analogous account; namely, they both assume without argument that the A-series is fundamental and that B-relations

56  The Middle Period are analyzable in terms of A-properties, and so begs the question against the earlier R-theory of time. Furthermore, there is reason to believe that this assumption is false or circular for the same reason that McTaggart’s analogous definition is circular. The analysis of temporal relations in terms of the conceptions of tense assumes that time has an intrinsic direction, and so cannot explain it. McTaggart’s reduction of B-relations to A-determinations proceeds as follows: The term P is earlier than the term Q, if it is ever past while Q is present, or present while Q is future. (McTaggart, 1927, p. 271; emphasis added) The problem with this account is that “while” is a temporal notion, and thus McTaggart’s use of it implicitly reintroduces the notion of time he sought to eliminate. For, to take the first disjunct, “P is past while Q is present” implicitly asserts that “P is past at t1” and “Q is present at t1,” and that certainly does imply that “P is earlier than Q” because “is past at t1” and “is present at t1” means the same as “is earlier than t1” and “is simultaneous with t1,” respectively. What then is the referent of “t1”? Whether we are talking of absolute time in the form of temporal individuals (moments) or relational time as sets of simultaneous events, McTaggart’s reduction is untenable. If t1 is a moment, then it must stand in a temporal relation to other moments and that must itself be analyzed. If t1 is a set of simultaneous events, then we still need an account of the sense or the temporal relation that generates the direction of the series of sets of simultaneous events. Hence, McTaggart’s analysis is implicitly circular since it assumes a ground for earlier than it but does not analyze it. Similarly, the problem with Broad’s account of “precedes” is that it assumes the reality of temporal relations but does not provide an analysis of them. In asserting that “A precedes B because A is past when B is present,” (1923, p. 66; emphasis added), Broad is neither accounting for the direction of time, nor reducing temporal relations to tensed properties or facts. For Broad’s analysis implies that “when” is a temporal notion that means, “at the same time as.” Thus, he is assuming time in addition to the tensed facts that A is past and B is present to account for the temporality and sense or direction of the series and for that reason his analysis is circular. The problem with Broad and McTaggart’s analyses of the sense or direction of a temporal series can be stated slightly differently. To determine which set of simultaneous events (or which moments) comes first, or is earlier than the other, and thus to determine their direction, there must be a ground for when A is present and B is future, and when A is past and B is present. The times or when A and B have different and incompatible tensed properties must be grounded to avoid the contradiction of A and

The Middle Period  57 B having incompatible temporal properties timelessly, and that requires an analysis of “precedes” and the intrinsic direction of time. By incorporating time (when or while) into their analyses of “precedes,” Broad and McTaggart fail to answer the question of what accounts for temporal relations and the intrinsic sense of the temporal series. In what follows, I shall argue that Broad assumes his definition of “earlier than” in his critique of the Russellian view that he maintained just a few years earlier, and also in his defense of the growing block view of time. If I am correct, then his arguments in 1923 neither undermines the R-theory, nor adequately defends the growing block theory.

III.  Can We Give a Spatial Analysis of the Tenses? Since temporal relations and the intrinsic sense of time are founded on the characteristics of past, present and future, Broad asks: how are we to understand them? He suggests first that “The obvious analogy to Now in Time is Here in Space” (1923, p. 66), but claims that the analogy is mistaken since the meaning of “here” “involves a reference to these very temporal characteristics on which it is supposed to throw light” (1923, p. 66), namely, now in time. We shall see, however, that although Broad has a point, it is not the point he thinks it is. Broad claims that “By Here I always mean that region which is near me at the time of speaking” (1923, p. 67). In this sense, “here” has a different meaning when used by two people at the same time, and it may have a different meaning when used by the same person at different times. In this so-called “referential” or “indexical” meaning of “here,” the content of a proposition expressed by the sentence in which it occurs refers to a particular spatial region near the person using it at the time of utterance. There is, however, also a common meaning (what David Kaplan calls the “character” (1989, pp. 126–127) to all the utterances in which “here” occurs. As Broad explains, We can, of course, extract a general meaning of “hereness”; it means “nearness to an observer who uses the word Here, at the time when he uses it.” But obviously Here is a descriptive phrase with a double ambiguity, since it refers both to a certain person and to a certain date in his history and does not become definite till these two blanks have been filled in by the context. (1923, p. 67) On the basis of this common meaning (or character) and the referential meaning (or content), Broad concludes that: It is evident then that Here is not going to help us to understand Now, since it contains an essential reference to Now. We must therefore

58  The Middle Period treat past, present and future on their own account, without expecting any help from spatial analogies. (1923, p. 67) The fallacy in Broad’s argument is that it rests on treating dates or times as tensed. While the character of “here” does presuppose the notion of a date or time, it does not follow that the date or time involved in the meaning of “here” refers to a tensed date or time, that is, a term with the property of nowness or presentness. To believe it involves tense in its meaning is to beg the question against the R-theory of time. In the R-theory, the analogy between “here” and “now” is that just as “here” can refer to a place without presupposing that there is a property of hereness, so “now” can refer to a time or date or, as on the token-reflexive account, a simultaneity relation between a token in which “now” occurs and what that token is about, without ascribing the property of nowness or presentness to that item. Thus, the question of whether we can understand the tenses by means of a spatial analogy cannot be simply dismissed by asserting that spatial attribution requires temporal attribution and so presupposes the notion of tense. The word “here” does make an essential reference to time, but it does not follow that it makes an essential reference to a tensed time. In the R-theory, to assert that “I am here now” means that the person who utters that sentence is in close proximity or near the place where she utters it, and temporally close or simultaneous with the uttering of it. To believe otherwise is to assume that dates are tensed rather than tenseless, an assumption that rests on the meaning of “precedes” in terms of tense.

IV. The Moving Now: A Relational Analysis of the Tenses and the Direction of Time How can we explain the passage of time, the sense of time flowing from earlier to later? Broad assumes that “earlier than” or “precedes,” and hence the flow of time, can be analyzed in terms of the notions of past, present and future. Can these notions be understood by an analogy of the changing relations of a spotlight moving along a fixed series of houses? I shall take up this question next. Broad gives a second understanding to the direction of time from earlier to later events as grounded in presentness moving along a fixed series of events. What is present is now related to the Now, what is past has been related to the Now and what is future will be related to the Now. Clearly this account of the past, present and future is circular since it employs the tenses in an attempt to understand them. This view has come to be called the “moving spotlight” or “moving now” view of temporal passage, and in his statement of it we perhaps find the origin of the term

The Middle Period  59 “eternalism,” used nowadays to characterize the B-theory. This is what Broad says: We are naturally tempted to regard the history of the world as existing eternally in a certain order of events. Along this, and in a fixed direction, we imagine the characteristic of presentness as moving, somewhat like the spot of light from a policeman’s bulls-eye traversing the fronts of houses in a street . . . On this view the series of events has an intrinsic order, but no intrinsic sense. It gains a sense, and we become able to talk of one event as earlier than another, and not merely of one event as between two others, because the attribute of presentness moves along the series in a fixed direction. (1923, p. 67) To characterize the history of the world as “existing eternally” is typical of B-theorists and their critics, and is a central tenet of the B-theory known as “eternalism.” Such an understanding begins by assuming that the events in the history of the world do not in and of themselves stand in temporal relations to each other. The notion of the world as a series of events existing eternally with an intrinsic order, but no intrinsic sense, is not the notion of a temporal series, but a non-temporal series that requires a moving now or transitory temporal properties to ground time relations to each other. It takes for granted that the temporal relations and the direction of time is understood in terms of the members of the series changing A-properties or changing their relations to the Now. I am not suggesting that the B-theory is committed to this view, but only how easy it is to construe it in that manner by characterizing the B-theory as a form of eternalism, and for that reason, R-theorists feel compelled to disavow that label. Indeed, such an account of the B-series with all A-properties, or only presentness moving along eternally existing events, has nothing in common (except the logical properties that generate a series) with the Russellian account of the R-series and the time relations that generate it. This understanding of B-relations found in the moving spotlight view has its roots in McTaggart’s (1908) conception of the B-series. McTaggart claims that the B-series is in part a non-temporal C-series (whose generating relation is more/less inclusive) whose terms have A-properties or stand in A-relations to the Now. Of course, more is required to generate a temporal (B-) series, namely “temporal becoming,” understood as the Now moving along a static, non-temporal C-series, with the result that the terms of the “eternally existing” series change their A-properties or A-relations to a term outside the series (McTaggart, 1908, pp. 25–26). If time is introduced in the form of a moving now, then time presupposes time since motion implies being at different positions at different times. Thus, a second time dimension is required, and other problems

60  The Middle Period follow as well. The light of presentness, now on one event and now on another event, are themselves events—each event becoming present—and ought therefore, to be part of a second series of events. Whether an event becoming present is in the same series as the original series of events or in a second order series of events, that new series of events itself will need to be given a sense and direction (see also, (Broad, 1938, pp. 45–46). When we understand temporal passage or temporal becoming as involving the Now moving along “the history of the world as existing eternally in a certain order of events” (Broad, 1923, p. 59), we are assuming that events in the history of the world do not in and of themselves stand in dynamic temporal relations that ground the passage of time. That assumption implies that the history of the world is static and not dynamic, and for that reason does not contain transition, and without transition there is no change (although there can be transition without change). In the moving spotlight view, for genuine dynamism to be ­introduced into a fixed series, a moving NOW that entails changing A-properties must be introduced. However, if what generates temporal relations and gives time a direction is a temporal dynamism, assumed A-theoretically, then the moving spotlight view assumes that the present, or what is illuminated, is at one point in the series earlier than it is at another. For if that is not the case, and the Now is at many points timelessly or simultaneously, then paradox ensues. For then each point is either timelessly or simultaneously present and that is absurd. If the Now is at different points successively, then paradox is avoided at the cost of assuming that time has a direction it does not account for. In characterizing the “history of the world as existing eternally in a certain order of events” (1923, p. 59) along which presentness moves, Broad is confusing a temporal series with a simultaneous block or a totum simul, such as a spatial series, all of whose terms are sempiternal (eternal1). Or, as a timeless series in which all terms do not exist in time at all (eternal2). What, then, we must ask, is the ground of A having been illuminated already, B being illuminated now and C not yet being illuminated rather than the other way around? To ground the direction of time, A, B and C must be present at different (successive) times, or at successive nows, and for that, in Broad’s analysis of “precedes,” we need A-characteristics. So, either the analysis is contradictory, circular or it presupposes a primitive relation of succession that renders the moving NOW superfluous. The moving NOW does not explain what is meant by “past,” “present” and “future” since it leaves unexplained what is the ground of an event having been illuminated (past) rather than will be illuminated (future), and so it also leaves unexplained the intrinsic sense or the direction of time. Broad makes this point by saying that if events have no intrinsic sense but only an intrinsic order, what meaning can we give to the assertion that the characteristic of

The Middle Period  61 presentness traverses the series of events in a fixed direction? All that we can mean is that the characteristic is present at B when it is past at A. Thus, all the problems which the policeman’s bull’s eye analogy was invented to solve are simply taken out of other events to be heaped on that particular series of events which is the movement of the bull’s eye. (1923, pp. 67–68) The appeal to “when” A is past and B is present assumes that A is earlier than B—it does not account for it. We cannot therefore understand the notions of past and present in terms of the moving spotlight, since the moving spotlight assumes that the light is at one point in the series earlier than another, and thus that the series has an intrinsic direction prior to the introduction of the moving NOW. Since we do not need a moving NOW to give the series a sense, we can dispense with it, and conclude that the introduction of A-properties is neither necessary nor sufficient to explain the direction or sense of the series. In contrast to Broad, the key feature of the R-theory I wish to emphasize, that distinguishes it from the previous characterizations of time relations, is that R-relations are primitive and unanalyzable external relations.1 Russell explains what he means by an external relation in the following passage: “Every relation is grounded in the natures of the related terms.” Let us call this the axiom of internal relations. If this axiom holds, the fact that two objects have a certain relation implies complexity in each of the two objects, i.e., it implies something in the natures of the two objects, in virtue of which they have the relation in question. According to the opposite view, which is the one I advocate, there are such facts as that one object has a certain relation to another, and that such facts cannot be reduced or inferred from, a fact about the one object only together with a fact about the other object only; they do not imply that the two objects have any complexity, or any intrinsic property distinguishing them from two objects which do not have the relation. (Russell, 1966, pp. 139–140) Clearly then, whether we accept McTaggart’s strong version of internal relations, according to which B-relations are entirely grounded on A-properties; the weak version of internal relations, according to which B-relations are grounded on a non-temporal C-relation whose terms have A-properties, plus A-theoretic becoming; or Broad’s later view (1938) that B-relations are dependent upon A-determinations, the resulting ontologies of time are decidedly un-Russellian. In the R-theory, there are no primitive non-relational temporal A-properties of pastness, presentness

62  The Middle Period and futurity, and there is no A-theoretic temporal becoming; particulars or events changing with respect to those properties. For the R-theorist, the foundation of real time is the R-relation and the R-facts that it enters. R-relations and R-facts are unlike the temporal relations and temporal facts as conceived by A-theorists or (reductionist) B-theorists, as I argue in Chapters 5 and 8.2 Unfortunately, McTaggart’s account of the B-series and his attack on the Russellian view tends to confuse these disparate views; an error that has been taken over by A- and B-theorists alike. To see how and why this is so, I shall turn next to an unhelpful characterizations of the A-theory/B-theory debate. There is one prevalent interpretation of the A-theory/B-theory debate that has its origin in McTaggart’s mistaken understanding of B-relations and was exacerbated by Broad’s discussion of the moving spotlight view we have been discussing. According to this interpretation, A- and B-­theories agree that there are B-relations, but they disagree over whether there are non-relational temporal properties in addition. So, for example, Steven Savitt interprets McTaggart’s positive view of time, before he argues that time is unreal, as claiming that there are A-properties “in addition to the B-series and its unchanging relations” (Savitt, 2001, p. 261). More recently, Kristie Miller makes the same claim when she says, The B-theory contrasts with the A-theory, according to which in addition to relations of earlier than, later than, and simultaneous with, there exists properties of pastness, presentness and futurity and these properties are had by different sets of events at different times. (Miller, 2013, p. 97) Bradford Skow offers an analogous statement in the following characterization of the A/B debate: In one corner we have the B-theory. The B-theory says there are times; the times are structured by the relation x is r seconds earlier than y; this relation gives time the same order and metric structure as the real numbers. And that is all. In the other corner we have the moving spotlight theory. The moving spotlight theory says that the B-theory leaves something out. In addition to the characteristics the B-theory says time has, there is also this: exactly one time the intrinsic property presentness. (2011, p. 359) As a final example, Yuval Dolev puts the dispute in similar terms when he characterizes the tenseless theorists’ quarrel with tensed theorists as over the question of whether in reality the only type of temporal relations are tenseless relations, or whether there are in addition, also tensed

The Middle Period  63 relations. The point of contention concerns the exclusivity of tenseless relations, not their existence. (Dolev, 2007, p. 97; emphasis added in last sentence) From the R-theorist’s perspective, there are problems with all of these analogous characterizations of the A/B debate. For the claim that A- and B-theorists agree over the ontological status of B-relations either ignores the R-theory entirely, or if not, then it assumes at the outset that either the R-theory or the A-theory is false. In the R-theory, temporal relations are external relations, meaning that they are simple, unanalyzable and irreducible temporal relations in their own right, not grounded entirely or in part on the nature of the related terms, and thus, R-relations exist independently of A-properties. For A-theorists who countenance A-properties, the “B-relations”, whose terms exemplify them, are internal relations either wholly or partially dependent on the non-relational temporal A- properties of their terms. Thus, if the “commonly agreed upon” relations were B-relations, as A-theorists conceive of them, then R-relations, too, could not exist without A-properties and for that reason, there could be no R-relations, and the R-theory would be false. On the other hand, if the commonly agreed upon time relations were R-relations, then the A-theory would be false. R-relations are external time relations that necessarily hold between terms without A-properties. But for the A-theorist, objects without A-properties are not terms of time relations. Thus, whether the commonly agreed upon relations are R-­theoretic or A-theoretic, it follows that to frame the debate in terms of the exclusivity of tenseless relations or B-relations, and not their nature and existence, as Dolev, Miller, Savitt and Skow do, begs the question against either the A-theory or the R-theory. One reason why characterizing the debate partly in terms of commonly agreed upon relations is so prevalent is because it confuses a non-­temporal series with a temporal series or process. The linguistic representation of a non-temporal series takes time to express, and the pictorial representation of a temporal series may be momentary and static. For example, when we count numbers to ourselves it takes time and we may represent a temporal series, a real progression or process such as the playing of a musical score as a static series, for example, as a written score. Nevertheless, R-relations are unique in that they involve a process of transition or a succession from earlier to later temporal objects. Representing time in a non-temporal way, for example as a spatial series, e.g., a row of houses, as Broad does in the moving spotlight analogy is fraught with difficulties, since succession in time is phenomenologically and ontologically different from “succession” in space, as well as from the successor relation in the number series, any isomorphism notwithstanding. The earlier series is a special series, a real progression from earlier to later events, and not the other way around, and this, in

64  The Middle Period one of Russell’s views of the direction of a relation, is grounded in the relational fact that contains the relation of succession from earlier to later temporal objects. The Russellian will reject claims that without tense and becoming the temporality and direction of time is without foundation. It should be clear that Broad’s characterization of the moving spotlight analysis of passage does not have anything in common with the R-theory, unless one confuses the R-series with a non-temporal series. For if one removes the moving NOW from the spotlight view then one is left with a series without A-determinations and so, by hypothesis, the resulting series is non-temporal. Thus, there is no common feature in the ontology of the moving NOW and the defenders of the R-theory. What follows from this is that the ontological issue concerning time goes beyond whether there is something more to time than B-relations as they are typically understood; all A-theorists and most B-theorists do not countenance temporal relations in the same way as R-theorists. Thus, to construe the debate between A- and B-theorists as over whether there is something more to time than the B-series runs the risk of turning the B-series into a static non-temporal series, and also of failing to recognize R-relations as dynamic and the R-series as an intrinsically temporal series.

V. A Cognitive Analysis of the Tenses Broad rejects another analysis of past, present and future that is reminiscent of, but markedly different from, the psychological analysis he adopted in his 1921 paper. He says that: It is extremely tempting to try to resolve the difference between past, present, and future into differences in the cognitive relations of our minds to different events in a series which has intrinsic order but no intrinsic sense. (1923, p. 68; emphasis added) Recall, the difference between a series that has an order and one that also has a sense is the difference between a spatial series and a temporal series. Given the points in space A B C, B is between A and C, but there are two different directions A B C and C B A. The series does not have an intrinsic direction. Regarding events in a temporal series, say, experiences in the history of a person, they all flow or pass in a definite direction from earlier to later. For Broad, the direction or sense of a series is connected—actually dependent on—the difference between past, present and future. Broad’s second account of the tenses shares an analogous assumption since it begins by assuming that apart from minds, events do not form a temporal series. For that reason, his argument against the cognitive account of the senses does not constitute an argument against the R-theory. Let me explain.

The Middle Period  65 In the cognitive analysis of the tenses, as Broad understands it, if we consider events that fall within the knowledge of a certain observer O, we can say that an event is present for O, meaning that it is perceived or simultaneous with something that O can perceive; for an event to be past for O means it can be remembered or simultaneous with something known by memory; and for an event to be future for O means that it can be anticipated or simultaneous with something that is anticipated. According to Broad, such a view assumes temporal relations, and therefore A-characteristics, and so neither grounds the intrinsic direction of time, nor explains the distinction between past, present and future. He argues that since every event is both perceived and remembered by O, it follows in this analysis that each event is both perceived and remembered (and so is present and past), and that is a contradiction unless events are perceived before they are remembered. Since, however, the notion of before or earlier than is analyzed in terms of past and present, the cognitive analysis is circular; grounding the direction of time in terms of a succession of mental acts that requires them to have A-properties in succession. Although this analysis of the tenses has a superficial similarity to Russell’s and Broad’s earlier psychological reduction, there is a crucial difference between the analysis Broad criticizes here and the one he maintained in “Time” (1921). In Scientific Thought, but not in his earlier article, Broad assumes that the series of events that we cognitively apprehend is a non-temporal series. It has an order, but to generate a sense of direction in the series, the terms must change with respect to A-properties from the future to the present to past, and therein lies the problem. In the later cognitive analysis, the tensed notions are grounded in the different cognitive relations between minds and events in an ordered series. The problem, according to Broad, is that if events stand in different cognitive relations to a certain observer O then each event has incompatible cognitive relations to the same mind, which is a contradiction unless O has those cognitive relations at different times. Alternatively, the event e is perceived (present) before it is remembered (past), and that reintroduces the tenses since “earlier” is defined in terms of past and present, and a vicious infinite regress results. Broad states this argument in the following passage: Clearly, we cannot simply define an event as present for O if O can perceive it or if it is contemporary with something that O can perceive. For we shall then have to define an event as past for O if O cannot perceive it but can either remember it or something contemporary with it. Now, of course, every event that falls within O’s knowledge has these two incompatible relations to O; though, as we put it, it has them at different times. He can first perceive, but not remember the event, and can then remember but not perceive

66  The Middle Period it. Hence these cognitive characteristics do not suffice to distinguish a past from a present event, since every event that O knows of has both these relations to him. If you add that an event always has the perceptual relation to O before it has the memory relation, you only mean that the event of remembering something is present when the event of perceiving it is past, and you have simply defined present and past for O’s objects in terms of present and past for his cognitive acts. If you then try to define the latter in terms of different relations to O’s acts of introspection, you simply start on an infinite regress, in which past, and present remain obstinately undefined at any place where you choose to stop. (1923, p. 68; some emphasis added) Note that Broad’s charge of circularity presupposes that precedes is analyzable in terms of past and present, and thus that there are no simple and unanalyzable temporal relations. This important point, as we have seen, is an assumption that he makes throughout; but without having argued for it anywhere. Broad concludes that past and present cannot be analyzed away in this manner out of “Reality as a whole,” which of course includes observing minds as well as what they observe. In other words, A-properties are subjective: there are real A-properties, but they exist in the mind. Or, more accurately, cognitive acts are characterized by A-properties. Broad adopts this point of view in the following passage: It does not of course follow that past and present in external Nature may not be reducible to certain relations between objective events and minds which observe them; but it does follow that these characteristics cannot be analysed away in this manner out of Reality as a whole, which of course includes observing minds as well as what they observe. (1923, p. 68) Thus, according to Broad, the cognitive analysis of the tenses leads to a mind-dependent thesis regarding A-properties: A-properties are properties of states of mind and not of the objects of those states. But again, Broad’s argument is based on an account of temporal relations that the R-theorist will not accept, namely, that the series of objects without A-properties is non-temporal. If one accepts R-relations, then Broad’s analysis (1921) of past, present and future as R-relations between states of mind and their objects does not involve a contradiction because the perception occurs before the memory and that is an unanalyzable fact. However, in Scientific Thought (1923), Broad rejects simple R-relations and therefore denies that A-determinations can be analyzed in terms of R-relations.

The Middle Period  67

VI. Ordinary Change, Temporal Change and (Absolute) Becoming Having considered and rejected three different analyses of the tenses, Broad turns his attention to what he considers to be two prima facie paradoxes concerning past, present and future: (1) Every event has all three of these incompatible properties and (2) events change with respect to these characteristics. The problem, according to Broad, is that we get into the second paradox when we take the obvious move to avoid the first (1923, pp. 68–69). I shall summarize the problems surrounding these paradoxes, and deal with them more extensively in the next two chapters when we consider McTaggart’s paradox. For now, it is sufficient to note that if events have all three incompatible A-characteristics we have a contradiction unless they have them at different times. But if events have them at different times, then it would appear that there is a fundamental feature of time that is “deeper” or “more fundamental” than past, present and future. How, then, are we to understand “time?” There are, as we know, different ways in which one can construe those times. Unfortunately, all attempts to define “time” so as to avoid the incompatible properties paradox, assuming of course that there is a genuine paradox to begin with, are circular and lead to a vicious infinite regress. If one appeals to times qua time points or moments of absolute time, then the analysis is circular. According to McTaggart, moments are temporally related and have a direction only because they have A-properties and change with respect to them. Thus, the appeal to moments cannot avoid the incompatible A-properties problem because moments depend on possessing incompatible A-determinations, leading to an infinite regress. Similarly, since temporal relations are analyzable in terms of A-determinations, if one avoids the incompatibility problem by claiming that a is present before a is past, one is clearly presupposing A-change, but is not accounting for it. Finally, if one appeals to the tenses, so that A is now past, was present and was (earlier) future; or A is now present, will be past and was future; or A is now future, and will be present and (still later) will be past, then one is obviously using A-time to explain A-change. Is there a way then, for events to have A-characteristics and change with respect to them without contradiction, circularity or infinite regress? Broad begins by rejecting the view that we can understand the way events change their A-properties by analogy with the way things change their ordinary properties, that is, in terms of ordinary change. We cannot analyze temporal change or A-change as analogous to an apple changing from green to red in terms of a suitably related succession of events that are qualitatively different. For to say that an ordinary thing changes—for example, a traffic light changing from red to green—“simply means that

68  The Middle Period its history can be cut up into a series of adjacent short slices, and that two adjacent slices may have qualitative differences” (1923, p. 69), and that the red section occurs before the green section. Since “before” or “precedes” is analyzed in terms of changing A-characteristics, ordinary change presupposes A-properties and A-change and therefore cannot explain it. Broad explains this criticism in the following passage: We have assumed that the history of our signal lamp can be analyzed into a series of shorter adjacent events, and that it was true of a certain pair of these that the earlier was red and the later green. But to say that this series of events passes from earlier to later (which is necessary if we are to distinguish between a change from red to green and a change from green to red) simply means that the red sections are past when the green ones are present, and the red ones are present when the green ones are future. Thus, the notion of the history of the lamp as divisible into a series of sections, following each other in a certain direction, depends on the fact that each of these sections itself changes from future, through present, to past. It would therefore be circular to attempt to analyse changes in events in the way in which we have analysed changes in things, since the latter imply the former. (1923, p. 70; emphasis added) To summarize, for Broad, things changing their ordinary, non-­temporal properties depends on succession and succession depends on events changing their non-relational temporal properties. For that reason, to explain temporal becoming—events changing their A-properties—in terms of things changing their ordinary properties would either be a circular explanation or assume that we already understand a non-A-­ theoretic account of succession and temporal becoming. The second problem with the attempt to understand temporal becoming in terms of ordinary change is that it does not fit the phenomenological facts. When we say that an event passes through time, from present to past, we mean that the entire event passes through time—that it undergoes temporal becoming as a whole, and not that it is split into parts, each with a different A-characteristic. To divide the event to say, the red section into past, present and future parts that occur successively, would not account for the temporal change of the red section as a whole; nor is it how temporal becoming is presented to us. For the fact is that the whole event was future, became present, and is now past. Clearly, no analysis which splits up the event into successive sections with different characteristics is going to account for the change in the temporal attributes of the event as a whole. (1923, p. 70)

The Middle Period  69 What is to be done then? What is the ground of a red section or temporal slice of, say, a traffic light preceding the green section rather than the other way around? Broad says it involves the red section changing from being present to being past, and the green section changing from future to present. How, then, does Broad explain the change that an event undergoes when it passes from future through present to past? What is his account of temporal becoming, or what he later came to call the “transitory aspect of time”? By turning to that question, we will arrive at Broad’s version of the growing block theory of time.

VI. Broad’s Growing Block Theory of Time For Broad, the problem of time is to explain how an event can have A-characteristics and change with respect to them in such a way that the direction of time, for example, a occurring before b and not the other way around, is grounded. Since ordinary change requires succession and succession depends on events changing their A-characteristics, he rejects ordinary change as a model for A-change. Instead, he turns to a new and unique type of relational change to explain both qualitative change and temporal change. There are, he says, two ways an entity can change its relational properties. The first is when a thing changes its relation to something that already exists, for example, when the child of Tom Smith (T) becomes taller than her father. The other is when something acquires a relation to an entity that previously did not exist, for example, when the child of Tom Smith changes from being the youngest child to becoming the last child but one. In the second case: a certain entity has changed its relational properties because a second entity, which did not formerly exist (and therefore could stand in no relation whatever to T), has begun to exist, and consequently to stand in certain relations to T, who is a member of the same universe as it. (1923, pp. 65–66) Broad characterizes this last change as becoming—what he eventually comes to call “absolute becoming”—and claims that it is the basis of temporal change and qualitative change. Thus, Broad distinguishes three kinds of change: 1. The change of the ordinary properties of a thing (qualitative becoming). When an apple changes from green to red, the green temporal slice of the apple is green before the red slice. Since before is analyzed in terms of changing A-characteristics, to account for a thing changing from green to red, one must account for the green section becoming past and the red section becoming present.

70  The Middle Period 2. The change of the temporal properties of events (temporal becoming). For Broad, the qualitative change of things implies events changing their A-characteristics of presentness for pastness. A change in an event occurs when, for example, the green section ceases to be present and becomes past. What, then, is the ground of an event changing from present to past? The answer lies in the third kind of change. 3. Broad claims that both kinds of change depend on the increase in the sum total of existence beyond the limit that it had when our given event came into existence (absolute becoming). This “change” occurs when the future (which does not exist) becomes present. The green section of the apple is part of the sum of existence that does not include the red section, and then there is another moment when the sum total of existence included all that was included in the first moment and also the red (section) event. Thus, the qualitative change of things is found to involve temporal becoming and that, in turn, involves absolute becoming or the coming into existence of events. In an important sense, absolute becoming is not really a change at all since, When we say that a thing changes in quality, or that an event changes in pastness, we are talking of entities that exist both before and after the moment at which the change takes place, and so it continues to exist. But, when an event becomes, it comes into existence; and it was not anything at all until it had become. (1923, p. 72) In Broad’s analysis of change, the first kind of change depends on the second and both depend on the third—absolute becoming or what Broad calls, simply, “becoming.” Thus, Broad is claiming that the change an event undergoes when it changes from present to past is like the change an event undergoes when it changes from being the youngest child to being the next to the youngest child. This claim, if taken literally, as Broad seems to do, is manifestly false. For the change that a child undergoes requires that the youngest child slice is different from the slice that is no longer the youngest child. If it is the same slice, then that slice has the incompatible properties of both being and not being the youngest child, and the original problem of incompatible temporal properties or of incompatible ordinary properties is left unresolved. To avoid a contradiction there must be two child slices or events (given Broad’s analysis of an object) of the same child, one without a relation to a younger sibling, the other with a relation to a younger sibling with the former slice occurring before the latter, and circularity arises once again.

The Middle Period  71 In other words, since the person who ceases to be the youngest child does so in virtue of a temporal slice now exemplifying the relational property of being the next to the youngest child, that slice is present. In that case, the past event does not lose the property of being the youngest child and acquire the new relation of being the next to youngest child, because the event of being the youngest child is past, and so is earlier than the new event of being the next to the youngest child, which is present. Thus, either Broad’s analysis of an event changing from present to past fails because it implies that the same event has and does not have incompatible relational properties at present, which is a contradiction. Or, if there are two events, one that had a relational property of being the youngest child (that is a past event) and another that now has the relational property of no longer being the youngest child (a present event), and one event is before the other, then his analysis is circular since the introduction of absolute becoming is supposed to provide an analysis of temporal becoming and ordinary qualitative change that explains temporal relations, and not presuppose them. To put my point otherwise, the analogy Broad employs is problematic because when one and the same event both have the relation of being the youngest child, and acquires the relation of being the next to the youngest child, it must exist both before and after the change. However, if an event acquires a relation at a certain present time it must exist at that time, but in Broad’s analysis, the event acquires the new relation after it has come into existence and thus acquires this relation in the past, and that is impossible. For a past event that is the youngest child cannot now acquire the relation of being the next to the youngest child, unless that event is now present; and thus the same event is present and past. Recall that Broad maintains that an event does not lose any relations it previously had, it just acquires new ones. He says: When an event, which was present, becomes past, it does not change or lose any of the relations which it had before; it simply acquires in addition new relations which it could not have before, because the terms to which it now has these relations were then simply non-entities. (1923, p. 66) Hence, the temporal part or event that has the relational property of being the youngest child of T does not lose it when it acquires the new relational property of no longer being the youngest child. The change of an event involves an entity that exists before and after the change takes place. The event that is present and now has the relational property of no longer being the youngest child cannot be identical with the past event that was the youngest child, for then the same event would either have incompatible relational properties in the present, or the

72  The Middle Period same event would be past and present. In other words, if the event of no longer being the youngest child does exist now, then it has just come into existence, and it cannot be identical with the event that was the youngest child, since that event does not exist now. Indeed, if it did exist now, then the same event would now have incompatible properties, and that is a contradiction, unless the same event has those relational properties successively, and that would require another round of absolute becoming. The objection can be reinforced by considering the following quote: “Thus the becoming of a long event is just the successive becoming of its shorter sections” (1923, p. 69; my emphasis). This passage reveals that Broad’s analysis of becoming, upon which temporal becoming and qualitative change is based, presupposes a primitive and unanalyzable notion of succession, which is incompatible with absolute becoming (since becoming is supposed to ground or explain the direction of time), or assumes an A-theoretic account of succession, and that renders his account viciously circular. It either case, becoming is not an ontological primitive by means of which temporal becoming, qualitative change, and the direction or sense of time can be explained. Broad’s growing block view is also expressed in The Mind and Its Place in Nature (1925), in the chapter on memory, where we read: Queen Anne’s death now precedes Queen Victoria’s by so many years and will do so forever; but there was a time when Queen Anne’s death preceded nothing. And, until Queen Victoria had died, Queen Anne’s death stood in no relation whatever to the event which we now call “Queen Victoria’s death”. For there was then no such event; and an event cannot stand in any relation to a mere nonentity.3 (1925, p. 256) Thus, the change from the future to the present requires that events undergo becoming. When becoming present, an event comes into existence; after coming into existence, it “continues” to exist. It appears to me that, once an event has happened, it exists eternally; all that happens henceforth to it is that, as more and more events occur and take their permanent place in the ever lengthening temporal order of the universe, it retreats into the more and more distant past. (1925, p. 252; emphasis added) Yet, taken literally, the notion of an event that “exists eternally” is problematic. For to be eternal typically means to be timeless, whereas for Broad, to exist means to be in time. What he must mean is that an event, once it comes into being, exists at every moment thereafter. This is

The Middle Period  73 reinforced by his mention of the “permanent place” of events that have come into the universe. Now, if one assumes that future events do not exist until they have become, and that once an event has become it continues to exist, then, Broad holds, one can also account for the direction of change: [W]hen we say that the red section precedes the green section, we mean that there was a moment when the sum total of existence included the red event and did not include the green one, and that there was another moment at which the sum total of existence included all that was included at the first moment and also the green event. (1923, p. 72) Or again, The sum total of existence is always increasing, and it is this which gives the time-series a sense as well as an order. A moment t is later than a moment t’ when the sum total of existence at t includes the sum total of existence at t’ together with something more.7 (1923, p. 72) That leaves no doubt. In Broad’s world, one essential feature of time and change is becoming, accounted for by the continual increase in the sum total of existents. There are, it seems to me, three problems with Broad’s analysis. First, it turns momentary events into continuants. Second, it gives an inadequate account of temporal relations, that is, it cannot account for the direction of time and change. And third, it is subject to McTaggart’s paradox. Let me briefly consider each of these points in turn. If, as Broad says, There is no such thing as ceasing to exist; what has become exists henceforth forever. When we say that something has ceased to exist, we only mean that it has ceased to be present; and this only means that the sum total of existence has increased since any part of the history of the thing became, and that the later addition contain no events sufficiently alike to and sufficiently continuous with the history of the thing in question to count as a continuation of it. (1923, p. 73) It would appear to follow that events, such as the death of Queen Anne, are continuants and not the momentary events (or at least short-lived entities) they are normally taken to be, since once they come into existence they continue to exist at every time thereafter. (p. 80).

74  The Middle Period Furthermore, if, as Broad says, a moment t is later than a moment t’ because t includes the sum total of existents at t’ plus something more, his account of time and change collapses. Since the sum total of existents at t includes the sum total at t’, there is no longer a sum total at t’ for the sum total at t to be later than. In other words, since the sum total at t’ is included in the sum total at t, when t comes into existence, all that exists at t is present and thus none of the different existents at t can be past or earlier than the others. To put the same point otherwise, in Broad’s view, when an event becomes past it now acquires relations to what previously did not exist and is now included in the sum total of existents, that is, it exists at the time that just came into existence, and for that reason must be present. On the other hand, by now acquiring relations that it did not have previously, it now becomes past. Thus, at the moment when an event becomes past it is also present and McTaggart’s paradox rears its ugly head. It is, of course, no use appealing to the different times at which the event is past and present since that would require another time dimension (since on the first level it is past and present at the same time), and the problem would then be to account for the direction of time and change in the second time dimension, and so on ad infinitum.

VII. Tooley’s Growing Block Theory We have seen that Broad’s view fails because the sum total of existents cannot be different at different times if everything that exists at the earlier time also exists at the later time, for then the two times overlap and the earlier is simultaneous with the latter, which is absurd. The problem is that, in Broad’s view, once an event comes into existence “it continues to exist forever after.” It doesn’t cease to exist, but it exists “eternally.” One might object by asking: Why does the later time not have some additional content not present in the earlier time? The response is that in Broad’s analysis there is no ground for there being two times, one earlier than the other. For when the sum total of existence increases, there no longer exists an earlier sum total since the “earlier” sum total is now simultaneous with or at the same time as the events that have just come into existence. However, if it is simultaneous with what has just come into existence, then it is present. Hence, if what is past is present, then the past does not exist, and the present sum total of existence does not contain as a part the earlier sum total of existence since there is no earlier sum total of existence. To put the same point differently, if there is only one time dimension, and what comes into existence are momentary events, then if momentary events “remain” in existence as new events come into existence, then it exists at the moment the new events come into existence. Hence there are no earlier sum totals, hence there is no

The Middle Period  75 increase in the sum total of existence since there is only the present sum total of existence. Is there, then, a consistent model of a growing universe that can avoid these difficulties? Michael Tooley (1997) Tooley thinks there is, but it requires a novel combination of tenets from A- and B-theories. Thus, time is dynamic—what previously did not exist comes into existence— and it is also B-theoretic in that the basic facts that come into existence are tenseless. His fundamental gambit is to introduce two notions of existence: “actuality simpliciter” and “actuality at a time” and two notions of truth: “truth simpliciter” and “truth at a time.” What is actual simpliciter is the sum total of what exists tenselessly, that is, the sum total of all those tenseless facts that are actual as of one time or another. What is actual as of a time t are those tenseless facts that have t or some time earlier than t as a coordinate. With the passage or flow of time, the universe grows as new tenseless facts come into existence so that reality or what is actual as of a time differs at different times (but the sum total of tenseless facts that exist simpliciter does not change). Since what comes into existence are only tenseless facts (facts that do not have tensed ties or exemplify tensed properties), and “since one event’s being earlier than another is . . . a tenseless state of affairs” (1997, p. 187), Tooley is not susceptible to the criticism that events that come into existence as of different times are nonetheless simultaneous, since no term is now earlier than another (although it may be tenselessly earlier as of the present moment). Moreover, to exist as of a time is not the same as to exist at that time. For that reason, the growing block does not turn momentary events into continuants. A momentary event that comes into existence at t1 need not exist at t2 to be part of the sum total that exists as of time t2. Nor does Tooley believe his theory is susceptible to McTaggart’s paradox. Since he rejects McTaggart’s trio of monadic tensed properties and tensed exemplification ties there is no incompatible properties problem to be avoided. Although I have tried to offer a sympathetic reconstruction of Tooley’s open future theory, I do not think it is tenable. I cannot understand how he can countenance, on the one hand, a R-theoretic ontology where all the basic facts are tenseless facts that exist eternally3, and so, I take it, without regard to what time it is, and on the other hand, a dynamic or A-theoretic ontology according to which tenseless facts come into existence at certain times and add themselves onto those tenseless facts that already exist “as of” that time—a concept I confess I do not fully comprehend. If the succession of events (and the totality of all states of affairs) exist eternally3 (regardless of whether they are past, present or future), then how can what tenseless states of affairs are actual as of one time differ from those that are actual as of another time? If Tooley means simply that different states of affairs, or more accurately,

76  The Middle Period different events, are tenselessly located at different times, construed relationally, then that is a view compatible with the R-theory of time. To put my objection otherwise, on the one hand, Tooley wants to claim that as of today, there are no tenseless states of affairs such as Oaklander ties (tenselessly) his shoelaces tomorrow, but there is simpliciter the tenseless state of affairs (call it “O”) that Oaklander ties his shoes on November 12, 2019. If Tooley means that neither the state of affairs O nor the event of tying my shoes contained in it exist today, then I would agree. And if he means that today, the eternal3 truth of the sentence that “Oaklander ties (tenselessly) his shoes tomorrow” depends on what I do tomorrow, then I would also concur. But if he means that the tenseless state of affairs “O” exists eternally3, and so I would say, does not have a temporal location, i.e., is not a term of a temporal relation, nevertheless comes into existence at a certain time, then there is an inconsistency. And an infinite regress arises as well, based on an argument I owe to Robin Le Poidevin. For suppose the fact (call it “E”) e occurs at t comes into existence at time t. Then, there is another fact (call it “F”), namely, the fact that E occurs at t and so on ad infinitum. Thus, I think we should say that states of affairs do not exist in time at all, although of course some of their constituents may. Rather than pursue Tooley’s valiant attempt to combine what he considers the insights of the tensed or A-theory with the B-theory, I shall return in section VIII to Broad’s account of propositions about the future. In section IX, I will discuss Broad’s response to McTaggart’s paradox and his attempt to avoid it. Finally, in subsection A. of section IX, I shall briefly discussion a problem with Broad’s views on propositions, facts and time.3 This last discussion, though important in its own right, has significance and relevance to Broad’s discussion of the unity of the mind in Chapter 6, and the philosophical implications of foreknowledge in Chapter 7, with which it is incompatible.

VIII.  Broad’s Analysis of Propositions about Future There are three final questions Broad discusses in the chapter on “Time and Change”: (1) what are judgments about the future about and (2) what are the facts, if any, they refer to and (3) how do we avoid McTaggart’s paradox of the same event having all three incompatible A-­properties? To address these questions, we need to revisit the crucial distinctions between grammatical and logical form, and correspondingly between commonsense and ontological facts. Consider the sentence “It will rain” (or “It’s raining is future”). Their grammatical forms correspond respectively to the commonsense facts It will rain and It’s raining is future that obtain if the statements are true. In “Time,” Broad maintains that the logical form of the sentence “It will

The Middle Period  77 rain” is that “It rains later than my anticipation of it raining,” or “It is raining later than the utterance of the sentence that ‘It will rain.’” The logical form each of those sentences are about their corresponding ontological facts. The point is that grammatical form is not a clear indication of logical form, and since the logical form reflects the ontological structure of the fact depicted, by attending to the grammatical form we could easily be misled into an erroneous metaphysics. Broad makes the same point in Scientific Thought when he says the grammatical form of “Puck exists” or “Puck churned the milk” suggests that these sentences are about Puck, but that is a mistake. There is the distinction between what a judgment is ostensibly about and what it is really about (1923, p. 76). . . . The judgment which is grammatically about ‘Puck’ proves to be logically about the set of characteristics by which the assertor describes Puck to himself. (1923, p. 75) The distinction is nicely put in the following passage: Roughly speaking, we may say that what a judgment professes to be about can be determined by a grammatical analysis of the sentence in which the judgment is expressed. Although there is always a connexion between the grammatical structure of a sentence and the logical structure of a judgment, it is highly dangerous to suppose that what the sentence is grammatically about is the name of what the judgment is logically about. (1923, p. 77) Thus, the sentence “Puck churned the milk” is grammatically about Puck, but since there is no such person, it is logically about the set of characteristics used to describe Puck. Similarly, “S is future” is verbally of the same form as “X is red,” and for that reason it seems to attribute the property of futurity to S, but this must be a mistake; since to have a characteristic implies to exist (at any rate in the case of particulars like events), and the future does not exist so long as it is future. (1923, p. 73) Thus, logically speaking, the proper analysis of future-tense judgment reveals that it is not a characterizing judgment, nor an existential or relational judgment, but a unique kind of judgment about a set of characteristics that will characterize some part of what will become. Broad calls this a genetic judgment.

78  The Middle Period What, then, are judgments of the future about? We are thinking of something, not of nothing, when we judge that “It will rain,” so what is the content of judgments about the future? He distinguishes two different meanings of this question: What is the grammatical subject of the judgment and what is the logical subject? What are we really talking about when we make judgments about the future versus what are we ostensibly talking about? Judgments about the future are like judgments about non-existent objects in that their grammatical subject is not their logical subject. In both cases, they are about some characteristics (universals) by which the subject is described by the person making the judgment. Thus, contrary to what Broad claims, the future is not a non-entity, since he maintains that propositions about the future have ontological status. What judgments about the future are really about—their logical ­subject—is a set of characteristics, namely, those by which we describe the subject of the judgment to ourselves. It is something real and independent of the judging mind; having the kind of reality and independence which is characteristics of universals, and not, of course, that which is characteristic of particular existents. (1923, p. 75) This implies that judgments or propositions are independent of time, or are timeless (eternal2), and that events are particular existents that come into existence and so exist in time and are dependent upon it. Thus, the judgment Puck exists, is about a certain set of characteristics (universals that exist independently of the mind), and what it asserts of those characteristics is that there is an existent (particular) that has those characteristics. The judgment is thereby false since it is not a particular that is characterized by the properties in question. Broad wants to account for the fact that even though there are no facts about the future, there is something that such judgments are about: our thoughts about the future intend propositions that have as constituents universals that characterize or define the subject of the judgment. Moreover, propositions about the future make assertions about those constituents, namely, that they will be instantiated, but contrary to existential judgments, they do not refer to anything. Thus, judgments which profess to be about the future do not refer to any fact, whether positive or negative, at the time when they are made. They are therefore at that time neither true nor false. They will become true or false when there is a fact for them to refer to;

The Middle Period  79 and after this they will remain true or false, as the case may be for ever and ever. (1923, pp. 75–76) Importantly, it follows from this that facts come into existence and remain in existence. Genetic judgments, such as “Biden will become the next President of the United States” cannot be reduced to characterizing judgments, for example, “Biden’s becoming President of the US is future,” that assert a particular existent has or is characterized by certain properties, e.g., futurity or existence. Of course, one can make a purely verbal reduction, since verbally a genetic judgment may be characterizing one, but logically or ontologically that would be a mistake since there is no property of futurity. For if there was such a property that was exemplified by S, then S would have to exist. In Broad’s view, the future does not exist, and for that reason, futurity is not a characteristic possessed by anything. According to Broad, the tendency to construe a genetic judgment about what will become as a characterizing judgment leads to the view that the future exists. This erroneous belief is due to the verbal ambiguity in the words “is” and “becomes.” That ambiguity might encourage misidentifying qualitative change (an apple become red) with absolute becoming (the future become present), but that would be a confusion. The word “is” can be used in an existential judgment and in a characterizing judgment; the word “becomes” can also be used in the sense of coming into existence or to express qualitative change. “X is red,” and “X is future” have the same grammatical form, but “X is future” is not characterizing judgment, attributing futurity to X. Rather its correct logical form makes a peculiar and unique assertion.

IX. The Eliminability of the A-properties of Pastness, Presentness and Futurity Broad attempts to avoid the problem of incompatibly tensed properties by claiming first that the characteristics of pastness and presentness do not exist since they can be analyzed away. Second, he maintains that if we take as primitive or unanalyzable temporal notions that of will and becoming, then we can give an analysis of what judgments about the future are about and what they assert that avoids McTaggart’s conundrum. To see what is involved in the first claim, let’s turn to Broad’s analysis of the past and present. He says: Let us now take the series of judgments: “It has rained,” “It is raining,” and “It will rain,” which are about events, and contain an essential reference to time. The first may be analysed as follows:

80  The Middle Period “There is an event which is characterized by raininess, and the sum total of existence when the judgment is made includes all and more than all which it includes when this event becomes.” The second may be analysed as follows: “There is an event which is characterized by raininess, and the sum total of existence is the same when this event becomes and when the judgment is made.” (1923, p. 77; emphasis added) There are several problems with Broad’s analysis of the tenses in this passage. In making use of “when” in the analysis of “It has rained,” Broad is using the term to refer to a different time at each occasion of its use; the (present) time2 when the judgment “It has rained” is made and the (past) time1 when the event becomes. Since according to this analysis, the sum total of existence at t2 is greater than the sum total of existence at t1, it follows that t1 is earlier than t2. This presents Broad with a dilemma. On the one hand, if “earlier than” is analyzed in terms of past and present (as he initially claimed (1923, p. 66), the analysis of judgments about the past and the present is circular, and thus leaves the analysis of has been or was unanalyzed. On the other hand, if he analyzes the past in terms of becoming, since reality is greater at t2 than it is at t1, when e is at t1 present, that can only be because It is raining at t1 still exists at t2. Of course, if it is raining at t1 exists at t2, then at t2, both “It is raining” and “It has rained” are presently true of the same event, and that is contradictory. For if “It has rained” is now true at t2, then there exists at t2 the event characterized by raininess at t1 because whatever comes into existence persists forever afterwards. Since, however, the event characterized by raininess at t1 is simultaneous with the same event characterized by raininess at t2, it follows that one event is simultaneously past and present at t2, which is impossible. To put the point otherwise, when (at t2) the past-tense judgment is made, the sum total of existence includes all and more than what it includes when (at t1) the event becomes. So, all “past” events exist at t2, since all events at t1 also exist at t2. Thus, when the judgment “It has rained” is made, the sum total of existence includes at t2 the event characterized by raininess at t1. Consequently, whether a true judgment about the past or present is made, the event it is about always exists in the present. However, if the analysis of the truth of a past tense judgment is a present tense fact, then the sum total of existence is not a growing block, but a simultaneous block or a totum simul. Furthermore, Broad says that “there is an event that is characterized by raininess,” but “is” is either “is now” or tenseless, for if it were past tense, then the analysis is obviously circular analyzing “It has rained” in terms of “There was an event characterized by raininess.” If “is” means “There is now an event,” then there is a contradiction since then the

The Middle Period  81 analysis of “It is raining” applies to judgments about the past, as the past still exists in the present. If it is analyzed as a tenseless copula, then we have a view such as that put forth by Michael Tooley (1997) that I have just criticized (see also Oaklander, 1999, 2001). Broad claims that we cannot analyze the judgment “It will rain” similar to the way he analyzed past and present tense judgments, but that “will” is taken to be a primitive kind of assertion. That is fine, but in asserting that these notions are unanalyzable, Broad is committed to giving them ontological status. It is not, however, clear what their status is: What does “will” refer to? Broad is clear that it does not refer to a characteristic, but then what does it refer to? If it does not refer to anything, then how can a future-tense judgment be about it, and how can we make judgments that assert that a certain characteristic “will” characterize what will become? On the other hand, if it is about the characteristic “futurity,” the question arises whether or not there could be such a characteristic if there was nothing that ever possessed it, and the existence of the future is obviously incompatible with the growing block. Broad claims that the judgment “It will rain tomorrow” grammatically purports to be about the future and asserts that a certain event will occur, but logically, We can only restate the judgment in the form: ‘The sum total of existence will increase beyond what it is when the judgment is made, and some part of what will become will be characterized by raininess.” We cannot then analyse will away, as we can has been and is now. (1923, p. 78) The grammatical form suggests that there is such a thing as the future populated by events such as “It will rain tomorrow” or “It is raining,” but the logical form of the corresponding judgment rejects that commonsense analysis or fact and claims instead that there is no such fact, but just the proposition that the judgment is really about. Note, however, that in appealing to “will” and “becoming” as basic and unanalyzable, Broad is committed in his ontology of time to whatever they stand for. Thus, he is not quite accurate when he says that, The only ‘constituents’ of the judgment [It will rain tomorrow], when it is made, are the characteristic—which has the kind of reality which universals possess—and the concept of becoming . . . the judgment makes a unique and not further analyzable kind of assertion about these terms. (1923, p. 78) The assertion must be that the universal will come to characterize the rainy event. My point is that even if “will” is unanalyzable it must still

82  The Middle Period have a referent—indeed, because it is unanalyzable it must have a referent, and thus there will be a further entity in his ontology that needs to be categorized.

A. Propositions, Facts and Time For Broad, facts have constituents and some of these constituents of facts are in time, and so they come into existence and they persist. “As new events become Queen Anne’s death becomes a constituent of additional facts” (1923, p. 82). Indeed, he also says that “the universe of actual fact is continually increasing through the becoming of fresh events; and changes in truth, which are mere increase in the number of truths [propositions] through this cause, are logically unobjectionable” (1923, p. 82). Thus, propositions do not change their truth value, but they are not timeless as he claims them to be. If true propositions come into existence, with the becoming of particular existents that are constituents in them, then they must exist in time, and therefore must be particular existents too, and not the abstract timeless entities they are supposed to be. One of the main points that comes out of Broad’s discussion is that facts are in time because they come into existence. When events come into existence, so do facts. The universe of actual fact is continually increasing. This is a fundamental difference between R-theorists and most A-theorists. Consider the statement that “It is not true that Queen Anne’s death, when it happened, had caused Lord Bolingbroke to swear.” This does not mean that “at the time of Anne’s death there was a negative fact, containing Anne’s death and Bolingbroke’s oath as constituents, and discording with the judgment that the death causes the oath. For when Anne’s death became, there was no such entity as Lord Bolingbroke’s oath, and therefore no fact of which this is a constituent. . . . What the laws of identity, contradiction, and excluded middle, between them assert is that a proposition is either true or false, cannot be both, and cannot alter in this respect. They do not assert . . . that the number of propositions, is eternally fixed; they only assert that it cannot be diminished. But it may be increased, and it is continually increased by the process of becoming which continually augments the sum total of existence and thereby the sum total of positive and negative facts. Or, to put it in another way, the laws of logic apply to a fixed universe of discourse, and we can at any moment get a fixed universe of discourse by taking the sum total of reality up to that moment. But the universe of actual fact is continually increasing through the becoming of fresh

The Middle Period  83 events; and changes in truth, which are mere increases in the number of truths through this cause, are logically unobjectionable. (Broad, 1923, p. 82) According to Broad, the law of excluded middle does not assert that the number of propositions is eternally fixed, but if the number of propositions can be increased, then they must come into existence through becoming too. If propositions and facts can come into existence, then they cannot be the timeless entities they are supposed to be but must be particular existents. We shall see that this analysis of propositions and facts, and their relation to time, presents problems for his discussion of the unity of the mind in Chapter 6, and his analyses of memory and precognition in Chapter 7.

Notes 1. Philosophers who endorse Russell’s view that temporal relations are external relations include Braithwaite (1928), Blake (1925), Erwin Tegtmeier (2007, 2009), Oaklander (2004, 2012), Gustav Bergmann (1992) and, of course, Russell (1915, 1966). 2. See also, Oaklander (2012, 2014a, 2014b) 3. For further criticisms of Tooley, see Oaklander (1999, 2002)

References Braithwaite, R. B. (1928). Symposium: time and change. Aristotelian Society Supplementary Volume, 8 (1), 162–174. Bergmann, G. (1992). New foundations of ontology (W. Heald, Ed., with a foreword by E. B. Allaire). Madison, WI: University of Wisconsin Press. Blake, R. M. (1925, Oct.). On Mr. Broad’s theory of time. Mind, 34 (136), 418– 435. https://doi.org/10.1093/mind/XXXIV.136.418 Broad, C. D. (1921). Time. In J. Hastings et al. (Eds.), Encyclopedia of religion and ethics (Vol. 12, pp. 334–345). New York: Scribner’s. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 143– 173). References are to this reprint. ———. (1923). Scientific thought. Chap. II. The general problem of time and change (pp. 53–84). London: Routledge & Kegan Paul. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. II. Time and metaphysics (pp. 63–83). References are to this reprint. ———. (1925). Mind and its place in nature. London: Routledge & Kegan Paul. ———. (1938). Examination of McTaggart’s philosophy: ostensible temporality (Vol. II (1), Chap. XXXV, pp. 265–323). Cambridge: Cambridge University Press. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 36–68). References are to this reprint. Dolev, Y. (2007). Time and realism: metaphysical and antimetaphysical perspectives. Cambridge, MA: MIT Press.

84  The Middle Period Grossmann, R. (1992). The existence of the world: an introduction to ontology. New York: Routledge. Kaplan, D. (1989). Demonstratives: an essay on the semantics, logic, metaphysics, and epistemology of demonstratives and other indexicals. In J. Almog, J. Perry, & H. Wettstein (Eds.), Themes from Kaplan (pp. 481–563). Oxford: Oxford University Press. Reprinted in L. N. Oaklander & Q. Smith (Eds.). (1994). The new theory of time (pp. 115–135). New Haven, CT: Yale University Press. McTaggart, J. M. E. (1908, Oct.). The unreality of time. Mind, 17 (68), 457–474. https://doi.org/10.1093/mind/XVII.4.457. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 21–35). References are to this reprint. McTaggart. J. M. E. (1927). The nature of existence (Vol. II, C. D. Broad, Ed.). Cambridge: Cambridge University Press. Miller, K. (2013). Presentism, eternalism and the growing block. In H. Dyke & A. Bardon (Eds.), A companion to the philosophy of time (pp. 345–364). Oxford: Wiley-Blackwell. Oaklander, L. N. (1999). Time, tense and causation, Michael Tooley. Mind, 105, 407– 413. Reprinted in L. N. Oaklander. (2004). The ontology of time (pp. 135–144). Amherst, NY: Prometheus Books. ———. (2001). Tooley on time and tense. In L. N. Oaklander (Ed.), The importance of time: selected papers of the philosophy of time society proceedings, 1995–2000 (pp. 3–12). Reprinted in L. N. Oaklander (Ed.). (2008). Vol. II. Time and metaphysics (pp. 118–127). ———. (2002). McTaggart’s paradox defended. Metaphysica, 3 (1), 11–25. Reprinted in Oaklander (2004). The ontology of time (pp. 51–62). ———. (2004). The ontology of time. Amherst, NY: Prometheus Books. ———. (Ed.). (2008). The philosophy of time: critical concepts in philosophy (Vols. 1–4). London: Routledge. ———. (2012). A-, B- and R-theories of time: a debate. In A. Bardon (Ed.), The future of the philosophy of time (pp. 1–24). London: Routledge. ———. (2014a). Dolev’s metaphysical anti-realism: a critique. In L. N. Oaklander (Ed.), Debates in the philosophy of time and related topics (pp. 1–29). New York: Bloomsbury Publishing. ———. (2014b). Temporal realism and the R-theory. In G. Bonino, J. Cumpa, & G. Jeschild (Eds.), Defending realism: ontological and epistemological investigations (pp. 122–139). Boston, MA: De Gruyter. ———. (2015). Temporal phenomena, ontology and the R-theory. Metaphysica, 16 (2), 253–269. https://doi.org/10.1515/mp-2015-0018 Russell, B. (1915, Apr.). On the experience of time. Monist, 25 (2), 212–233. https://doi.org/10.5840/monist191525217. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. 1. The reality and language of time (pp. 174–187). ———. (1966). The monistic theory of truth. In Philosophical essays (pp. 131– 146). London: George Allen & Unwin Ltd. Savitt, S. (2001, Jul.). A limited defense of passage. American Philosophical Quarterly, 38 (3), 261–270. Skow, B. (2011). Experience and the passage of time. Philosophical Perspectives: Metaphysics, 25, 359–387. doi:10.1111/j.1520-8583.2011.00220.x

The Middle Period  85 Tegtmeier, E. (2007). Three flawed distinctions in the philosophy of time. Metaphysica, 8 (1), 53–59. doi:10.1007/s12133-007-0005-8 ———. (2009). Ontology of time and hyperdynamism. Metaphysica, 10 (2), 185–198. Thomas, E. (2019, Apr.). The roots of C. D. Broad’s growing block theory of time. Mind, 128 (510), 527–549. https://doi.org/10.1093/mind/fzx020 Tooley, M. (1997). Time, tense, and causation. Oxford: Oxford University Press.

4 The Later Years

The Later YearsThe Later Years

The Full-Future Theory, Presentism and McTaggart’s Paradox

I. Introduction In Chapter XXXV “Ostensible Temporality” of Examination of McTaggart’s Philosophy (1938), Broad adopts a different account of the nature of time once again. In fact, there is textual evidence, as I shall show, that he there adopts two different A-theories of time: the full-future theory and presentism. According to the full-future theory, past, present and future events exist in virtue of exemplifying A-properties, and events only have B-relations if their terms have A-determinations. Presentism is the view that only the present exists, the past no longer exists, the future does not yet exist and there are no A-determinations with the possible exception of presentness. Broad’s commitment to both these views in the same chapter is problematic since they are incompatible. To develop and defend these claims I shall argue that in his initial account of the transitory aspect of time—the sense in which events flow from the future to the present and into the past—Broad rejects the fullfuture theory and adopts a presentist analysis of transition construing it as absolute becoming. I shall then examine Broad’s discussion of McTaggart’s critique of Russell’s view of time and change, and McTaggart’s positive view of time. Next, I consider Broad’s critique of Russell’s view, and argue that a proper understanding of the R-theory can avoid his objections. It will become clear that Broad sees himself as defending the full-future theory and McTaggart’s view that “the notions of Time and Qualitative Change involve A-characteristics as well as B-characteristics” (1938, p. 62). Finally, I shall discuss Broad’s analysis of McTaggart’s paradox.

II. Independent Account of the Phenomenology of Time Broad begins the section “Independent Account of the Phenomenology of Time” (1938, p. 36) by distinguishing sentences that record temporal facts from sentences that record non-temporal facts. Of the former kind he gives the following examples,

The Later Years  87 “My grandfather died before I was born”, “I am now writing”, ‘I had my breakfast before I began writing this sentence, and I shall have my lunch after I have stopped writing.” (Broad, 1938, p. 36) Examples of non-temporal facts are “Twice two is four,” “Red is a color” and “Purple is darker than yellow.” These examples are telling since they suggest, as Broad evidently believes, that temporal facts are recorded by sentences that have temporal or tensed copulas, such as “is now,” “was” and “will be,” whereas non-temporal facts have non-temporal, timeless or tenseless copulas. Thus, linguistically, the distinction between sentences that record temporal and non-temporal facts concern the kinds of copulas in each. Broad interprets this linguistic point about the two different kinds of copulas as implying a certain analysis of the ontological difference between temporal and non-temporal facts. All temporal facts, including those that contain temporal relations between events, are tensed, (the italics indicating either the possession of A-properties by the terms of the relation, or the special ontological status of the present). All non-temporal facts are tenseless since their constituents do not have A-properties, nor undergo absolute becoming but instead involve either timeless universals or persistent (everlasting) particulars. Thus, the distinction between temporal and non-temporal facts gives rise to potential trouble at the outset. For as we shall see, Broad believes that since sentences in ordinary language that record temporal relations between events are tensed, the facts they record must be tensed. Since, however, for the Russellian, sentences in an ideal or ontologically perspicuous language that record temporal relations between particulars are tenseless, the R-facts they record must be tenseless. It is then a natural, but mistaken, step to construe R-facts as not temporal but eternal2 in the sense in which “Red is darker than yellow” or “2 + 2 = 4” are, or eternal1 in the sense in which points on a line or places in absolute space, if there are such, would be. We shall see in the following that this error does in fact arise in Broad’s critique of the R-theory. Broad is concerned with temporal facts whose constituents all fall within the experience of a single individual that he calls “intra-subjective experiences.” An example would be, “I saw a bright flash, and almost immediately afterward I heard a loud bang” (1938, p. 265). There are three features to be considered when discussing intra-subjective experiences: 1. Experiences have some duration. 2. Any two experiences of the same person stand next to each other in a certain determinate form of a temporal relation (of which there are 13 alternatives including partial and total overlap).

88  The Later Years 3. Our experiences flow from the future, become present and then recede into the past (1938, p. 37). He calls the first two features the “extensive aspects of time” due to their similarity to the extensive aspects of space. Just as a person’s mental history as a whole has a duration whose parts stand in temporal relations, a cord made up of shorter strands has a spatial extension whose parts are spatially related. Although time and space are in some ways similar, in the following passage Broad asserts a fundamental difference between them: In a linear spatial series there is no asymmetric dyadic relation intrinsic to the series, but in the temporal series of experiences which constitutes a person’s mental history there is a genuine dyadic relation which is intrinsic to the series and involves no reference to any term outside the latter. This is the relation “earlier than”. It is the fundamental relation here, and temporal betweenness is definable in terms of it. In the temporal series there are two intrinsically opposite directions, earlier-to-later and later-to-earlier. In the linear spatial series there is no intrinsic direction. If direction is to be introduced, this must be done extrinsically, either by reference to motion along the line (and therefore to time), or by reference to the right and left hands of an external observer, or in some other way. (1938, p. 39) We have seen in Chapter 3 that points on a line have an order but not an intrinsic sense or direction. Experiences in a person’s mental life history, however, not only have an order but also have an intrinsic direction from earlier to later. In Scientific Thought, the ground of the intrinsic direction of time is the continual increase in the sum total of existence. Event e2 is later than event e1 since the sum total of existence at e2 includes everything that exists at e1 together with something more. In Examination of McTaggart’s Philosophy, or at least in “Ostensible Temporality,” he rejects the growing block theory, but what he puts in its place is difficult to ascertain.1 Broad next turns to the transitory aspect of temporal facts where there are two features that need to be considered: (i). the characteristics of pastness, presentness, and futurity; and (ii) the fact that every event is continually changing in respect of these characteristics. It is continually becoming less and less remotely future, then it becomes present, and then it continually becomes more and more remotely past. (1938, pp. 40–41)

The Later Years  89 In discussing the sentences that express the transitory aspect of time, Broad makes two points: one linguistic, the other non-linguistic. The linguistic point is that in ordinary language we can record transitory temporal facts in two ways. The first way is by means of differences of tense or temporal copulas with certain temporal adverbs. “Thus, I should most naturally say, ‘I had my breakfast lately’, ‘I am writing now’, ‘I shall be eating my lunch soon’, and so on” (Broad, 1938, p. 41). The second way is by means of a single uniform copula with different temporal adjectives or predicates. “Thus, I might have said, ‘Eating my breakfast is just past’, ‘My writing is present’, ‘Eating my lunch is slightly future’.” (1938, p. 41) The second point is not purely linguistic; it is that we cannot eliminate temporal copulas entirely. Broad expresses this point in the following passage: By using various forms of temporal adjective we may be able to reduce the number of forms of temporal copula needed in recording temporal facts to a single copula. We can, e.g., replace the sentence “I was eating my breakfast, I am writing, and I shall be eating my lunch” by the sentence “Eating my breakfast is past, this spell of writing is present, and eating my lunch is future.” But the “is” is the temporal copula “is now,” which a person would use if he said of me “He is now writing”; it is not the non-temporal copula which would be used if one said “37 is a prime number” or “Scarlet is a determinate form of red.” (1938, p. 41) This non-linguistic point concerning the ineliminability of the temporal copula “is now” is important because it has several ontological implications. First, Broad claims that “This point is highly relevant in connexion with McTaggart’s argument against the reality of Time” (1938, p. 42). McTaggart’s error in his argument for the unreality of time, Broad will argue, is to think that temporal or tensed copulas can be eliminated by means of non-temporal or tenseless copulas and temporal predicates. For example, McTaggart thought that “It was raining,” means “It is (tenselessly) raining at t1 and t1 is past,” and similarly for present- and future-tense sentences. So McTaggart believed that tensed sentences can be translated or eliminated without loss of meaning in terms of tenseless ones. This cannot be done, according to Broad, but the belief that it can is what gives rise to McTaggart’s paradox (see pp. 113–116). Second, the ineliminability of tense is also relevant to Broad’s dialectical argument against the Russellian theory of time. For Broad argues that in Russell’s (and his own earlier) view, the tenseless way of expressing temporal facts is tantamount to reducing or eliminating temporal facts in

90  The Later Years terms of non-temporal facts about abstract timeless objects such as integers or persistent particulars such as points on a line (1938, p. 61). This consequence, if correct, would be devastating for the R-theory because it implies that there is no temporal sequence or R-series. A third (non-linguistic) point Broad draws from the ineliminability of the present-tense copula regards the ontological nature of temporal facts. It demonstrates that sentences that record temporal facts imply the existence of tense or A-characteristics as constituents upon which temporal facts depend. Thus, Broad construes a fact about ordinary language as both supporting an A-theoretic ontology of time and undermining the Russellian view. I shall discuss each of these alleged implications later in this chapter, but of more immediate concern is Broad’s argument for the ineliminability of the present tense or temporal copula “is now.” Broad’s argument for the irreducibility of the present tense copula is found in the following passage: Suppose that, on a certain occasion, I utter the sentence “The event e is present.” And suppose that this utterance records a fact. If the word “is” in it were a non-temporal copula, every utterance by me of the same sentence would record the same fact, no matter whether it were earlier than, contemporary with, or later than this utterance of mine. But actually the only utterances of this sentence which would record the same fact as this utterance of mine would be those which are contemporary with my utterance. . . . It is clear then that there can be no question of getting rid altogether of temporal copulas and replacing them by a single non-temporal copula and temporal predicates. (1938, pp. 41–42) The conclusion Broad draws, that tense is ineliminable from ordinary language, does not follow. For the premise upon which it is based, namely, that if the copula is non-temporal (or tenseless) then all sentence tokens of “Event e is present” would record the same fact, is either false or question begging. If, for example, one accepts the token-reflexive analysis of tensed sentences, then it is false that every utterance of “Event e is present” would record the same fact. Different tokens of the sentence type “Event e is present” uttered at a different time, and even contemporary utterances of “Event e is present,” would record different facts because the different (simultaneous or successive) tokens would be constituents of the different tenseless facts recorded. Of course, if one assumes that the fact recorded by “Event e is present” is that “E exemplifies presentness,” then all utterances of that sentence type would record the same fact. However, in the token-reflexive analysis the logical form of the fact recorded by “Event e is present” is “Event e is simultaneous with this token,” where “this token” refers directly to a different sentence

The Later Years  91 token, and therefore to a different fact each time a sentence of that type is uttered, written down or thought. Thus, to assume Event e exemplifies presentness is the fact recorded by “Event e is present” begs the question against the R-theorist who denies the existence of all A-properties, including presentness. Broad’s assumption of some version of the tensed theory of time is also strongly suggested, if not implied, by his explication of time’s transitory aspect. According to Broad, the transitory aspect of time consists in the fact that experiences undergo temporal becoming; an experience is at one time wholly in the future, it becomes less and less remotely future, and eventually “this particular experience [a visit to the dentist] becomes present and then slips into the immediate past. There is the fact which one records by saying “Thank God (on the theistic hypothesis) that’s over now!” (1938, p. 38). As he puts it: 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, p. 38) This passage is ambiguous because the notion of “fact” is ambiguous. At the level of commonsense it is a fact—or as I shall call it, a “commonsense fact”—that events, including experiences, stand in the earlier than/ later than and simultaneity relations to other events. It is also a commonsense fact that events that were once in the future become present and then recede into the past. At the pre-analytic level of commonsense, the existence of such phenomenologically given “facts” is ontologically neutral. Thus, when Broad says that the 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 commonsense fact. In that case, no specific view of the ontological analysis of the structure of that fact is required or implied. 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. He could be suggesting that what we commonsensically express in ordinary language by means of tensed sentences are “ontological facts” about the essence of temporal reality, although the grammatical form may be misleading regarding what specific tensed ontology is implied. These two notions of “fact”—commonsense fact and ontological fact—get blurred along with the distinction between grammatical and logical form. Before turning to

92  The Later Years this last point, we shall first consider Broad’s remarks about the temporal characteristics of pastness, presentness and futurity.

III.  Pastness, Presentness and Futurity Broad’s view of the ontological status of A-determinations is unclear as he both affirms and denies their existence. On the one hand, his ontology makes use of instantaneous event particles, “which have an absolutely determinate temporal position, but no temporal boundary” (1938, p. 44), and claims regarding instantaneous events that “the properties of pastness, presentness and futurity belong only to event particles and not to processes” (1938, p. 44).2 On the other hand, he argues that it is a mistake to treat temporal becoming in the model of qualitative change. It must instead be treated as primitive—absolute becoming—coming into and going out of existence. Thus, from the A-theoretic perspective, he seems to adopt both the full-future theory, according to which, past, present and future exist, and presentism, the view that only the present exists—a combination that is, prima facie, inconsistent. Broad’s reason for rejecting A-properties is that we cannot analyze ­temporal becoming in terms of either motion (the moving present), or qualitative change (events changing their A-properties), since such analyses would be exposed to objections like those raised in Scientific Thought. For example, in the moving spotlight model of temporal becoming, the present moves along a row of eternally existing events, and so it makes sense to ask: How fast does time (the moving present) move? But there is no meaningful way of answering this question. Broad puts this objection as follows: To ask, “How fast does it move?” . . . is equivalent to asking “How great a distance will it have traversed in a unit time-lapse?” But here the series along which presentness is supposed to move is temporal and not spatial. In it “distance” is time-lapse. So the question becomes “How great a time-lapse will presentness have traversed in a unit time-lapse?” And this questions seems to be meaningless. (1938, p. 45) Furthermore, if temporal becoming is understood in terms of event particles acquiring and shedding the property of presentness, then any instantaneous event in the first time-series will acquire and lose presentness, and all other A-properties, in a second time-series. Thus, a momentary event in the first time-series will have an indefinite duration in a second time-series. Since events such as e1 acquiring presentness are themselves events in the second-time dimension, they too must change their A-­characteristics as they undergo temporal becoming. In that case, however, we will need another time dimension in which they endure through a change of A-properties, and so on ad infinitum (Broad, 1938, pp. 46–47).

The Later Years  93 For these reasons, we cannot understand temporal becoming in terms of motion or in terms of qualitative change. According to Broad, the error that underlies these analyses of temporal becoming is that they confuse the grammatical form of the phrase “to become present” with the logical form of “to become hot,” and so it is thought that “X becomes present” attributes the property of presentness to persistent object x. As he puts it, The phrase “to become present” is grammatically of the same form as the phrase ‘to become hot” or “to become louder”. We are tempted to think that sentences like “This event became present” record facts of the same kind as those which are recorded by sentences like “This water became hot” or “This noise became louder.” Now a little reflection is enough to show that this is a mistake. (1938, p. 47) Although “X becomes present” is of the same grammatical form as “X becomes louder,” it is not of the same logical form. For something to become louder it must exist before and after the change. However, for something to become present, it cannot exist before and after the change, since only instantaneous event particles can become present. Also, to become hot involves X taking on a monadic quality and that suggests the same thing is the case regarding “becoming present,” but Broad rejects that similarity. Thus, Broad maintains that the fact recorded by “X becomes present” is not one of qualitative change—although it is grammatically similar, it is logically quite different. As he says: Any subject of which we can significantly say that it “became hot” must be a more or less persistent substance, which before and after the date at which it became hot. . . . But a literally instantaneous event-particle can significantly be said to “become present”; and indeed, in the strict sense of “present” only instantaneous event-­ particles can be said to “become present”. (1938, p. 47) How, then, are we to understand the change that an event undergoes when it becomes present? In response Broad says, To “become present” is, in fact, just to “become”, in an absolute sense; i.e., to “come to pass” in the Biblical phraseology, or more simply, to “happen”. Sentences like “This water became hot” or “This noise became louder” record facts of qualitative change. Sentences like “This event became present” record facts of absolute becoming. . . . I do not suppose that so simple and fundamental a notion as that of absolute becoming can be analysed, and I am quite

94  The Later Years certain that it cannot be analysed in terms of a non-temporal copula and some kind of temporal predicate. (1938, p. 47) I do not think that Broad’s appeal to “Biblical phraseology” is helpful; his earlier analysis in “Time” and the R-theorists’ analysis in terms of emerging and passing away is preferable. Absolute becoming, as Broad understands it, involves a presentist analysis of generation and annihilation that I shall criticize in the next chapter. Although Broad does give a different analysis of absolute becoming that I shall consider later in this chapter, I shall argue his analysis is either lacking ontological content or acknowledges A-properties and so is susceptible to the same problems he sought to avoid by apparently adopting a presentist metaphysics. By appealing to absolute becoming, Broad is shifting from the property acquisition and loss model—or the full-future theory—of temporal becoming, to the existential model or presentism without acknowledging that he does so. According to the existential model, temporal becoming or an event’s changing from the future to the present and into the past, is grounded in absolute becoming: the coming into existence of what did not previously exist and the ceasing to exist of what previously did exist. Unfortunately, Broad’s presentist analysis of absolute becoming in terms of existential change is inconsistent with his account of temporal becoming that requires event particles possessing determinate degrees of pastness, presentness and futurity. This difficulty reemerges in his discussion of McTaggart that I shall turn to next.

IV.  McTaggart’s Account of the Phenomenology of Time Broad’s discussion of McTaggart’s philosophy of time in 1938 is one of the writings for which he is most famous. His painstaking analysis of McTaggart’s account of the phenomenology of time, his critique of McTaggart’s positive argument for the thesis that time and change require A-properties and negative view that time is unreal, are all thorough, clear and insightful. In the following sections of Part IV, I shall discuss (A) Broad on McTaggart’s treatment of the A- and B-series and his positive account of time; (B) Broad’s analysis of generation and annihilation; (C) Broad’s critique of McTaggart’s critique of Russell’s account of time and change and his objection to what he takes to be Russell’s token-reflexive analysis of tensed sentences; and (D) Broad’s criticism of the R-theory. A. The B-series and the A-series and McTaggart’s Positive Account of Time According to McTaggart, prima facie positions in time form a B-series and an A-series. In ordinary language, thought and experience we

The Later Years  95 distinguish between the B-series, whose generating relations are the B-relations of earlier than/later than and simultaneity, and the A-series determinations of past, present and future. As Broad interprets McTaggart, the B-series is composed of event particles that are either simultaneous or successive. Simultaneity is a symmetrical transitive relation; succession is an asymmetrical transitive relation. Event particles can be classified into a series of successive sets of mutually simultaneous event particles whose generating relation is “earlier than” (1938, p. 48). The A-series is “formed by the various possible degrees of pastness in decreasing order of magnitude, the characteristic of strict presentness, and the various possible degrees of futurity in increasing order of magnitude” (1938, p. 49). The terms of the A-series and the terms of the B-series constitute the same series of terms under different descriptions. They do so in virtue of the fact that at any moment the terms of the B-series are characterized by one and only one characteristic of the terms in the A-series, and the terms of the A-series are in one-to-one correspondence with the terms of the B-series.3 Thus, if e1 is earlier than e2, then at the same time, e1 is past and e2 is present or e1 is present and e2 is future. Commonsensically, the distinction between these two series, with their B-relations and A-determinations, is unproblematic. We speak of events as being earlier or later than another and as being past, present and future. What is open to dispute is the proper analysis of the constituents and the structure of the phenomenological facts recorded by statements involving B-relations and A-determinations. To get at McTaggart’s analysis of the B-series and the A-series, Broad raises the question of how the two series are related and claims that for McTaggart, the B-series and B-relations are essential to the reality of time, but that A-determinations are necessarily involved in the notion of B-relations. For that reason, if A-properties turn out to be delusive or unreal, then we should have to reject B-relations as unreal, and therefore, time as well. The rest of the chapter will explain McTaggart’s arguments for these claims, beginning with his positive account of time. Broad summarizes McTaggart’s argument for A-characteristics as follows. (i) The A series cannot count as a process of perpetual qualitative change unless the asymmetrical transitive relation generating it is the relation “earlier than” and the series is thus a B-series. (ii) If the generating relation is “earlier than,” then there is a B-series and the series can be counted as in perpetual qualitative change. (iii) In order for there to be a process of perpetual change and hence a B-series, each term must change with respect to a certain characteristic. (iv) The only characteristic each term can perpetually change is with respect to its pastness, presentness or futurity. (v) Therefore, if there are no A-characteristics, there is no perpetual qualitative change and there is no relation of “earlier than.” Without such a relation there can be no B-series, and hence there can be no time (1938, p. 54).

96  The Later Years B. Broad’s Critique of McTaggart’s Critique of Russell What, then, is the argument for McTaggart’s claim that change requires A-characteristics? Briefly, it is this: time requires B-relations, but B-relations alone cannot account for change. For in a B-series, nothing changes. The position of an event in the series never changes its relations to another event in the series. If a is earlier than b, then it is always the case that a is earlier than b and it is never the case that a is not earlier than b. Nor do events in the B-series change by coming into existence or ceasing to exist. In other words, there is no generation and annihilation in the B-series. Since if it is ever the case that a is earlier than b, then it is always the case that a is earlier than b, and so a and b always exist and can neither be annihilated nor generated. McTaggart concludes that in order for the B-series to change, or contain change, events in the B-series must have A-properties and change with respect to them. McTaggart also argues that events do not undergo qualitative change in the B-series. For if an event e has a quality before another event e’ has a different quality, then it is always the case that e has that quality at that time and it will always be the case that e has that property at that time. For example, the event of Queen Anne’s death is always before the event of Lord Bolingbroke’s death, and so it does not change its position in the B-series or any other property it possesses unless its position in the A-series changes. Moreover, the fact that say, a poker is hot at t1 and cold at t2 does not constitute a change either, since it is always a fact that the poker is hot at t1 and cold at t2. Those facts would constitute a change if t1 and t2 are moments of time in a B-series. However, in order for the B-series to be a temporal series, there must be change, and in order for there to be change, the terms of the B-series, whether moments, events or facts, must have A-properties and change with respect to them. Thus, without A-properties there are no B-relations and, for that reason, Russell is mistaken in believing that the B-series alone is sufficient to constitute time (Broad, 1938, pp. 50–51). Broad first replies to McTaggart’s argument against qualitative change. He claims that McTaggart’s arguments against the possibility of qualitative change in the B-series without an A-series fail since they ignore Russell’s account of change. Russell’s account rests on the premise that it is things that change and not events or facts. In Russell’s view, a thing changes when there are qualitative differences between successive events or particulars that are united by causal laws and spatio-temporal continuity. Broad states Russell’s view of qualitative change as follows. In Russell’s view, Events are neither generated nor annihilated, nor do they change in respect of any of their characteristics. There are certain series of successive events, such that the members of any one such series are intimately interconnected by certain spatial, causal, and other relations

The Later Years  97 which do not interconnect members of any two such series. Each such series is counted as the history of a different thing. Now successive members of one such series may differ in respect of a certain quality; e.g., one term may have a determinable quality Q in the determinate for q1 and a later term may have Q in the form q2. The statement “The thing T changes from q1 to q2” is completely analyzable into a statement of the following kind. “There is a certain series of successive events so interrelated that it counts as the history of a certain thing T; e1 and e2 are two successive adjoined phases in the series; and e1 has Q in the form q1 whilst e2 has Q in the form q2. (1938, p. 54)4 McTaggart objects to this analysis of change by claiming, in Broad’s words, it is always a fact about this series that it contains a term which has q1, and a term which has q2, and that the former immediately precedes the latter. Hence this fact cannot be what is referred to when we say that T has changed in respect of Q from q1 to q2 (1938, p. 54) In other words, McTaggart argues that Russell’s analysis fails because in it nothing—that is, no event or fact about an event—changes unless we introduce A-properties. According to Broad, that is an irrelevant objection since in Russell’s view, change is not to be found in events or facts changing, but in things changing; and things change in accordance with the analysis he summarized in the previous passage. Broad states his reply to McTaggart in the following passage: Now this [argument against Russell] seems to me to be irrelevant. Certainly, on this [Russell’s] view of qualitative change, no fact and no event changes. It is alleged, instead, by the supporters of this view that the fact of change consists in a conjunction of facts which neither change nor are about change. To this McTaggart merely makes the counter-assertion that there can be no change unless certain facts about events change, i.e., unless events of the first order are subjects of events of the second-order. And the only ground which he has given for this is the argument in paragraphs 310 and 311 [in McTaggart, 1927, where he argued that that there is no change unless the terms of the B-series have A-properties and events change with respect to them] where he ignored the present alternative and assumed that he had exhausted all the possible alternative views about qualitative change. (1938, pp. 54–55)

98  The Later Years Admittedly, an event does not change, but it does not follow that there is no change in a thing as a whole. In Broad’s interpretation of Russell, change is grounded in causal and spatio-temporal continuity between qualitatively different momentary parts of a single thing. To deny that such an account gives us change, since nothing in it changes, mistakenly assumes that the analysis of change must itself change; that the analysis must imitate its object. Interestingly, Broad makes the same criticism of Bergson, as he does of McTaggart, when he says, “Bergson’s arguments seems to rest partly . . . on the erroneous view that a whole of related states cannot be a change unless each of its terms be a change” (1921, p. 172). Broad’s important point has also been made by Erwin Tegtmeier: As to McTaggart’s argument that the B-series cannot be temporal because it does not change, it is misleading and wrong mainly for two reasons. Firstly, the task is to analyse ontologically the general structure of temporal phenomena, the task is to analyse the dynamics, not to dynamise the analysis. The task of science, including philosophy, is to find out what the entities involved in its research object are and what their laws are; it is to describe and explain, not to imitate the object. (Tegtmeier, 2007, p. 154) Thus, the argument by McTaggart that Russell’s analysis is inadequate insofar as it is unchanging arises from a methodological muddle. The ontological analysis of time and change need not, and indeed must not, attempt to duplicate its object or be changing itself. Broad returns to McTaggart’s argument in sections 310 and 311 for the claim that there is no existential change in the B-series; that generation and annihilation are incompatible with events standing in B-relations. McTaggart’s argument rests on the premise, or principle (hereafter “the temporal principle” or “TP” for short), that if a term is ever in a temporal relation to another term, then it is always in that temporal relation to that other term, and if a term ever has a property at a certain time, then it always has that property at that time. The temporal principle underlies McTaggart’s argument that no term can change by either coming into existence or going out of existence, i.e., generation and annihilation is impossible in the B-series. For if X ever precedes Y by a certain amount, then X always precedes Y by that amount, and therefore X and Y always exists. Hence neither X nor Y can change by coming into and going out of existence. Clearly, the basis of McTaggart’s rejection of change in the B-series is the temporal principle which implies that if a temporal relational fact holds at any time (if it ever holds) then it holds at every time (it always holds), and thus that B-facts exist in time, if they exist at all. As we have

The Later Years  99 seen in Chapter 2, this is true of Broad’s earlier view of temporal relations where he maintains that temporal relations must continue to exist, and it is also true of McTaggart view of B-relations and B-facts, but it is not true on the Russellian view of R-relations and R-facts. McTaggart gives an interpretation of TP that implies that B-facts exist in time and continue to exist throughout all of time. The R-theorist will maintain, however, that McTaggart’s and, as we shall see, Broad’s reasoning confuses the language of time (“always true”) with the ontology of time (“always exists”). As a claim about temporal language, TP is true. “Plato’s birth is earlier than Aristotle’s birth” is always true. As a claim about temporal ontology it is not true, since R-facts do not always exist. Indeed, they do not exist in time at all, much less at every time. Moreover, an R-theoretic interpretation of TP is compatible with generation and annihilation, but not in the sense that Broad intends it. There is emerging and passing away in the R-theory, but there is no generation and annihilation as understood in either the presentist theory (absolute becoming) or the full-future theory (changing A-properties), as I shall explain later. First, I want to explain Broad’s response to McTaggart. Broad argues that once the temporal principle is correctly understood, then it is compatible with the existence of temporal relations and with generation and annihilation. To see why, consider the following passage where Broad explains the sense in which TP is true: Let us take as examples the Battle of Hastings and the Battle of Waterloo. Before either battle had happened, it would have been true to say “There will be a battle at Hastings and there will be a battle at Waterloo 749 years later.” . . . During the Battle of Hastings, it would have been true to say, “There is a battle going on at Hastings and there will be a battle at Waterloo 749 years later.” . . . At any moment after the Battle of Waterloo it is true to say, “There was a battle at Hastings and there was a battle at Waterloo 749 years later.” . . . These expressions, all of which involve temporal copulas, are the natural and the accurate ways of recording facts about relations of precedence. (1938, p. 55; emphasis added) We can agree with Broad that these expressions are, grammatically, the “natural” way of recording commonsense facts about relations of precedence, but whether they are logically the “accurate” ways of recording ontological facts is the question at issue in an argument against the R-theory. Are what ordinary temporal sentences profess to be about what they really are about? Does the grammatical form represent the logical form of sentences about temporal relations? Broad explicates TP by claiming that sentences that involve temporal copulas are accurate ways of recording facts about relations of

100  The Later Years precedence. He is thus making the point that the tenses have ontological status. There are two problems with Broad’s interpretation of TP. First, in an argument purportedly against McTaggart’s use of TP to show that temporal relations require terms with changing A-properties, Broad is assuming what McTaggart is attempted to prove, he is not refuting it. Broad’s interpretation of TP builds tensed or A-determinations into relations of precedence, and in doing so he assumes that grammatically accurate statements in ordinary language, such as “Socrates was born before Plato was born,” are logically accurate statements in an ideal ontological language, and so implies that the terms of temporal relations are tensed.5 From the R-theoretic standpoint, Broad is confusing grammatical form with logical form. Moreover, Broad’s analysis of TP is markedly different from the R-theoretic analysis since he provides a ground for the statement that “a is earlier than b” is always true by providing at each moment a different tensed fact that grounds its truth. Although, in the R-theory “a is earlier than b” is also always true, the ground of its being always true is not something that exists at every time, but something that exists outside of time. For the Russellian, a is earlier than b is an eternal3 fact and facts do not exist in time, although sometimes, as in this case, a temporal relation (earlier than) that is a universal is a constituent of an temporal fact. C.  Broad’s Analysis of Generation and Annihilation Broad maintains that there is no incompatibility between the “prima facie appearance that events become and pass away and that they stand to each other in relations of temporal sequence, and simultaneity” (1938, p. 67). True enough, but the whole issue depends on whether or not we can provide an ontological analysis of these prima facie appearances in a way that is both phenomenologically and dialectically adequate. Statements in ordinary language do not wear their ontology on their sleeve, and one could only think they do by confusing commonsense with ontology or grammatical form with logical form. To emphasize the grammatical tenses in an account of the language of time, including claims about things coming to be and passing away, is one thing. To provide an ontological analysis of the phenomena that ground the commonsense facts that are expressed in ordinary language is quite another. Broad claims that Russell denies that events are generated or annihilated, and if Broad means that Russell denies absolute becoming, then he is correct. If Broad means that Russell—or more generally, the R-theory— denies the phenomena in question, then he is mistaken. Broad, who rejects the R-theory, does give a sense in which there is generation and annihilation that he claims is compatible with any two events “always” standing to each other in any temporal relation in which they “ever stand.” I shall argue that, for Broad, they are not since his account of generation

The Later Years  101 (coming to be) and annihilation (passing away) is presentist, whereas his account of temporal relations and TP entails the full-future theory. Thus, his ontological analysis of the phenomena in question are incompatible, and hence inadequate. Consider what he says about generation and annihilation: The only sense in which an event e is “annihilated” is that there was and no longer is an event answering to the description of e. The only sense in which an event e is “generated” is that there was not and now is an event answering to the description of e. In this sense events are “generated” and “annihilated”, and this is compatible with any two of them “always” standing to each other in any temporal relation in which they “ever” stand. (1938, p. 56) If a proposition describing an event now has something falling under it, then that something now exists or is generated. If a proposition describing an event does not now have something falling under it, then either the event has been but no longer is now existing, and so is annihilated, or the event has not yet but will come into existence and then will be generated. In any case, we are still left with the question of the ground of “there has been (or was) and no longer is an event” and “there is not yet but will be an event.” In other words, what is the ground of statements about the past and the future. What is the ontological status of the past and the future? Depending on Broad’s answer, his account of generation and annihilation either assumes some tensed version of the A-theory (which seems likely) or it is ontologically neutral and just expressing what we ordinarily believe. If the non-existence of the future and past is grounded on their not existing at the time of utterance of future-tense statements (but later) or the time of a past-tense statement (but earlier), then Broad’s view is compatible with an R-theoretic analysis similar to the one he maintained in 1921 (Broad, 1921). If the analysis is in terms of A-properties, then an event is generated if it has the property of presentness and annihilated if it loses presentness and acquires has the property of pastness. But Broad rejected the analysis of temporal becoming as involving the full-future theory or the donning and doffing of A-properties. To be consistent he would have to reject that analysis, and interpret generation and annihilation as a presentist would, in terms of absolute becoming. Is presentism compatible with events always standing to each other in the same temporal relation in which they “ever” stand? Indeed, is generation and annihilation compatible with their ever standing in a temporal relation? If generation and annihilation is interpreted as the presentist does, as seemingly Broad would have it, namely, as absolute becoming, then arguably there are no temporal relations, or at a minimum there are

102  The Later Years no R-relations. For if only the present exists, and the present is instantaneous, then there are no temporal objects earlier or later than the present; only objects simultaneous with each other exist. The problem I am raising regarding Broad’s analysis of TP, annihilation and generation can be stated this way. Suppose we assume that Broad is adopting presentism. Unfortunately, that is inconsistent with his analysis of TP since in it, B-relations depend on A-characteristics. However, if there are A-characteristics of pastness, presentness and futurity, then there must be past, present and future entities that exemplify them and therefore, presentism is false. If Broad rejects the full-future theory and adopts presentism, then it is questionable that Broad can defend TP, for in presentist analysis there are no temporal relations. Conversely, given Broad’s interpretation of TP, the reality of temporal relations is incompatible with presentism. To sum up the argument of this section, McTaggart claims that a B-series without an A-series or B-relations without A-determinations would not constitute a genuine temporal series since it could not contain existential or qualitative change. Broad argues that McTaggart’s argument is fallacious since TP and existential and qualitative change are compatible. The problem for Broad is that TP implies the existence of A-properties, and therefore his interpretation of TP defends, rather than refutes, McTaggart’s argument for A-change. Furthermore, Broad’s analysis of B-relations entails the full range of A-properties and is therefore incompatible with the presentist ontology he apparently employs in his analysis of generation and annihilation. If Broad’s analysis of generation and annihilation implies a presentist ontology, then not only is it incompatible with his full-future account of TP, but since presentism denies the existence of temporal relations, it is arguably also incompatible with TP, as he understands it. D. Broad’s Critique of the Token-Reflexive Analysis of Tensed Language Although Broad rejects McTaggart’s radical analysis (the doctrine of strong internal relations) of “X is earlier than Y,” according to which “ ‘earlier than’ can be defined in terms of A-characteristics” (1938, p. 52), he agrees with McTaggart that there cannot be B-relations without A-characteristics (the doctrine of weak internal relations). This agreement is peculiar since the analysis of B-relations as either dependent on or reducible to A-determinations treats temporal becoming or the transitory aspect of time as a species of qualitative change, and so entails the infinite regress of time dimensions Broad previously rejected. Nevertheless, Broad argues that Even if we reject the view that “X is earlier than Y” means that there is a difference in the A-characteristics of X and Y and that

The Later Years  103 this difference is positive, there remains another alternative which would suffice for McTaggart’s purpose. It might be suggested that the relation “earlier than” can hold only between terms which have A-characteristics; just as harmonic relations can hold only between terms which have pitch. And it might be suggested that the degree of the B-relation between two terms depends on the difference between the determinate values of their A-characteristics; just as the harmonic relations between two notes depend on the difference between the absolute pitches of the two. In fact, to use an expression of Meinong’s, we might be able to see that B-relations are “founded upon” differences in the A-characteristics of the related terms. (1938, p. 58; emphasis added) This is a weak version of internal relations, but it would suffice for McTaggart’s purpose. For if B-relations depend on the A-characteristics of their terms, then change requires A-characteristics and McTaggart’s positive view of time is justified, and his negative argument would at least have a foothold. Broad continues by stating his tentative agreement with the suggested analysis of temporal relations: This view holds the field unless it can be shown that sentences that contain the words “past”, “present”, and “future”, or their equivalents, can be translated without loss of meaning into sentences which do not contain these words or equivalents of them, but do contain the phrase “earlier than” or some equivalent of it. Now Russell and certain other philosophers have claimed that this can be done. (1938, p. 58) This passage makes it clear that the translatability of temporal predicates and tensed copulas or lack of such has ontological implications. Broad argues that the translation of A-sentences by B-sentences cannot be done since the grammatical form of sentences in ordinary language are irreducibly tensed. He believes that the use of tensed copulas is the natural and commonsense way of expressing temporal facts, and that they cannot be analyzed away. We shall see, however, that Broad does not rest his argument against Russell on linguistic considerations alone or how we ordinarily speak. Rather, Broad appeals to ontological considerations to undermine the R-theory. In the next section I shall consider Broad’s argument against the R-theory, but first, I shall consider his critique of the token-reflexive translation of tensed language. Broad summarizes the token-reflexive account, that McTaggart and Broad associate with Russell, as follows: Any utterance of a type-sentence, which is of a certain grammatical form and contains the type-word “now” or “present” or some

104  The Later Years equivalent is understood by the speaker and hearers to mean that an event of a certain kind is simultaneous with this utterance. Any utterance of a type-sentence, which is of a certain grammatical form and contains the type-word “past” or some equivalent, is understood by speaker and hearers to mean that an event of a certain kind is earlier than this utterance. And the same holds, mutatis mutandis, for any utterance of a type-sentence which is of a certain grammatical form and contains the type-word “future” or some equivalent. . . . If this be so, A-characteristics have been completely analysed in terms of B-relations. (1938, p. 59) It is important to note that Broad claims that in this view, “Unless there were people who uttered type-sentences of these kinds nothing would be past, present, or future; though events would still be simultaneous or successive” (1938, p. 304). Thus, the adequacy, or lack of such, of the linguistic analysis—the criterion of translatability—has ontological significance. If the linguistic reduction is successful, then there are no A-properties. If it is not successful, then “B-relations are ‘founded upon’ differences in the A-characteristics of the related terms” (1938, p. 304). Broad criticizes the token-reflexive analysis by claiming that it cannot possibly be what the speaker means by “It is raining now” since the speaker is not “using the utterance to express a judgment which he is making about the utterance itself” (1938, p. 57). An R-theorist will reply that one must distinguish between grammatical meaning and logical meaning. What a sentence means grammatically may be what the speaker or writer intends to convey or express by the language. Alternatively, it is what the speaker is using the sentence for, say, to communicate to the listener to take an umbrella before going outside. The logical form of the sentence is what the sentence refers to, or more accurately, depicts or stands for in reality—the ontological fact that is the phenomena being analyzed. Admittedly, ordinarily, the speaker does not use or mean by a tensed sentence the tenseless sentence that perspicuously represents its ontological analysis. In other words, the language used for the communication of commonsense facts is not the same as the language used to provide an accurate statement of the constituents and structure of the ontological facts recorded. Thus, it is compatible with the world being tenseless that sentences in ordinary language are irreducibly tensed, or so defenders of the new tenseless theory of time believe. Broad’s distinction between the grammatical form and logical form of a sentence (and his Principle of Pickwickian Senses in Chapter 1) foreshadows the distinction between the ordinary and philosophical use of language that R-theorists employ in their defense against Broad’s linguistic argument. Broad does not initially consider this R-theoretic response to his criticism of the token-reflexive view, but he does offer an alternative reduction

The Later Years  105 or account of the meaning of tensed statements consistent with it. He distinguishing between what the speaker means and what the hearer understands by a tensed sentence token. He says that when a speaker says “It is now raining” what she means is that “It is raining simultaneously with ‘this’” where “this” is a logically proper name for an object of perception or introspection of hers. What a listener understands when she hears “It is now raining” is that “It is raining simultaneously with that utterance” of the speaker. In neither case is grammatical tense or ontological tense (in the form of A-properties) an indispensable element of the judgment or what it is about. In one case, “this object” refers to the object perceived when the speaker makes the utterance, and in the other case, “that utterance” refers directly to certain auditory sensa the listener hears when the speaker utters the sentence in question. Thus, neither the speaker nor the listener is attributing the property of being present to the object, even though they both understand the information intended to be conveyed by the sentence token. As Broad puts it, We might say that what the speaker means by his utterance is that an occurrence of rain is simultaneous with this, where this is some particular, other than the utterance itself, which he prehends simultaneously with making the utterance. As regards the hearers, we may suppose that each interprets the utterance to mean that an occurrence of rain in the speaker’s neighborhood is roughly simultaneous with certain auditory sensa which that hearer is sensing, viz., those which are manifestations to him of this utterance of the speaker. (1938, p. 60) In both first- and third-person cases, the tensed thought or sentence token refers to the fact that Such and such an event is simultaneous with, or is earlier than, or is later than, this; where “this” is used as a logically proper name for some particular which the speaker or the hearer is prehending when he hears or makes the utterance. (1938, p. 60) Thus, for the speaker, the account is not token-reflexive since “this” does not refer to the utterance itself, but to some particular other than the utterance that the speaker perceives simultaneously with making the utterance. Is this analysis successful? Since it is intended as an ontological analysis, the question is, according to Broad, are the temporal facts expressed by sentences containing temporal copulas more accurately expressed by tenseless sentences? Does the analysis capture what we mean by tensed

106  The Later Years sentences? To answer this question, if the copula in the allegedly tenseless translations are really timeless. As Broad puts it, Now what kind of copula is the “is” in these substituted sentences? Is it a timeless copula, like the “is” in “3 is the immediate successor of 2” or in “13 is a prime number?” Or is it the temporal copula “is now”? Or is it some third kind of copula which logicians and metaphysicians have not clearly recognised and distinguished? If it is the timeless copula, the theory has prima facie been successful. If it is the temporal copula “is now” then the analysis has certainly failed. If it is supposed to be some third kind of copula, we must await further information from the supporters of the theory. (1938, p. 60) Broad argues that it is not the timeless copula since “no one except a philosopher doing philosophy” (1938, p. 61) would say, “My breakfast is earlier than (or precedes) my lunch.” That may well be true, but what relevance does that have to the issue of what is the best description of the fact to be analyzed? Broad appears to rule out the logical form of A-sentences in terms of tenseless sentences by appealing to grammatical form or what we ordinarily say. As Broad puts it, “No one ever does talk in this way in real life” (1938, p. 61). Since Broad maintains that all temporal facts are expressed by a linguistically temporal or tensed copula, and therefore an ontologically tensed copula, we have not analyzed A-characteristics away. Furthermore, since no third kind of copula has been introduced, it follows that “prima facie, the temporal copula has not been analysed away” (1938, p. 61). There is, however, a third alternative that Broad does not consider. A copula that is neither tensed nor timeless, but tenseless and temporal. Broad recognizes that there may be “some third kind of copula which logicians and metaphysicians have not yet clearly recognized and distinguished” (1938, p. 60), but awaits further information about it from supporters of the R-theory. Indeed, an R-theorist will maintain that not all temporal copulas are tensed. Briefly, the idea is that in ordinary language, there are grammatically two types of copulas: tensed temporal, and tenseless and non-temporal. However, in an ontologically adequate language there is a third kind of copula, one that is tenseless and temporal. Just like “is now” is a temporal copula, “is earlier than” is also a temporal copula because it connects temporal objects with temporal relations; but, it is tenseless because the terms of temporal relations do not have A-properties or undergo A-theoretic absolute becoming. These three kinds of temporal copulas—tensed and temporal (“It is now raining”), tenseless and timeless (“Red is a color”), and tenseless and temporal (“e1 is earlier than e2”)—reflect

The Later Years  107 different ontological analyses of the facts represented by them in an ideal language. The R-theory maintains that the ontological form of sentences used to record temporal facts is tenseless and the copula they contain is temporal. Thus, “a is earlier than b,” records a temporal fact because the intrinsically temporal R-relation, earlier than, is a constituent of the R-fact that a is earlier than b. It is a tenseless fact because it does not contain A-properties or any other feature that determines whether it (or its terms) are past, present or future. Moreover, an R-fact is tenseless because it is timeless3, and so does not exist in time, as a term of an R-relation. The ordinary language of precedence is tensed and thus, according to Broad, makes use of temporal copulas that are tensed. In an ontologically perspicuous ideal language sentences record temporal yet tenseless facts because they contain temporal relations and the R-facts they stand for are without tense. Broad acknowledges that there may be a logical language that does accurately represent the correct ontology of “temporal facts” even if it is not the language we use, or how we ordinarily speak. Thus, he recognizes the distinction between grammatical form and logical form when he replies to the linguistic argument against the R-theory, by saying, “Of course, it may be answered that this objection depends simply on defects in the language that we speak” (1938, p. 61). Yet, he finds that reply unavailing since the ontology of Russell’s view unsatisfactory for reasons I shall consider in the next section. E. Broad’s Critique of the R-theory According to Broad, a Russellian tenseless language implies or strongly suggests a picture of the world that is ontologically tenseless and hence non-temporal. He accuses R-theorists of using a mistaken analogy when thinking of R-relations and the R-series, that he characterizes as follows: The [R-] theory seems to presuppose that all events, past, present, and future, in some sense ‘co-exist,’ and stand to each other timelessly or sempiternally in determinate relations of temporal precedence. But how are we to think of this ‘co-existence’ of events? It seems to me that the events and their temporal relations are thought of either by analogy with timeless abstract objects such as the integers in their order of magnitude, or by analogy with simultaneous persistent particulars, like points of a line in spatial order from left to right. Neither of these analogies will bear thinking out. (1938, p. 61; my emphasis) I agree that neither of these analogies will bear thinking out, but I disagree that Russell is committed to either of them. In the R-theory,

108  The Later Years temporal relations between particulars are in some sense analogous to the relations that are obtained between universals, but we need not claim that the terms of temporal relations are to be thought of as timeless abstract objects or universals. Like relations among universals, temporal relations are universals, and like the facts that they enter, they are not located at any time or in any place. Yet, it does not follow that the terms of temporal relations co-exist timelessly in the way that universals do. Nor does the permanent truth of “e1 is earlier than e2” require that either e1 and e2, or the fact of which they are constituents, be located at all times. Why, then, does Broad think that events are persistent particulars or timeless objects? I think at least part of the reason stems from his belief that temporal facts are recorded by means of sentences with tensed or temporal copulas, and that sentences with tenseless or non-temporal copulas record non-temporal facts, and that non-temporal facts state relations between and among universals. I have claimed, however, that the logical form of an R-statement such as “e1 is earlier than e2” contains a tenseless copula that records a temporal fact that is eternal3. The problem with his critique of the Russellian view is that he reads an ontological form that is alien to the R-theoretic analysis of time into the logical form of a Russellian ideal language. Furthermore, by confusing the grammatical form of temporal relation statements, which are tensed with the logical (ontological) form of the fact they record, Broad could think that an analysis of R-relations without A-properties is a non-temporal relation. Although this is true in McTaggart’s positive view of time, it is not in Russell’s. Broad misunderstands the ontological form of the R-theory by confusing the R-theoretic account of temporal relations with B-relations as McTaggart understands them. Broad’s criticism of the R-theory is confusing McTaggart’s notion of the B-series without A-properties as a C-series (a non-temporal series without A-properties, with an order but not a direction) with the R-series. For in that case, the R-series could be confused with, say, “greater than” between integers or “to the left of” between points on a line. If, however, the R-relations and McTaggart and Broad’s understanding of the B-series as an A-series superimposed on a C-series plus temporal becoming are kept distinct, then Broad’s criticisms do not follow. R-relations are intrinsically temporal and even though they have the same logical properties as say, a number series or a spatial series, they are just different from all other such relations. Further evidence for my diagnosis of Broad’s misinterpretation of Russell stems from the next objection he makes to the Russellian analysis of time. Based on his misinterpretation of Russell, Broad claims that “the theory leaves out altogether the transitory aspect of Time” (1938, p. 61).

The Later Years  109 Broad’s static interpretation of the R-theory gives rise the following objections that, due to their importance, I shall quote at length: Consider, e.g., that series of successive experiences which constitutes my mental history from the cradle to the grave. On the theory which we are discussing, there is no question of events “becoming” or “passing away”. In some sense of “is”, there “is” timelessly or sempiternally all that there ever has been or will be of the series. The qualitative changes that take place in the course of my experience are supposed to be completely analyzable into the fact that different terms of this series different in quality, as different segments of a variously coloured string different in colour. But this leaves out the fact that at any moment a certain short segment of the series is marked out from all the rest by the quality of presentedness; that at any two different moments the short segments thus marked out are different segments, though they may partially overlap . . . and that the relatum at any moment is, or is contained in, that short segment which has presentedness at that moment. Thus change has to be postulated in a sense not contemplated by the theory, viz., the steady movement of the quality of presentedness along the series in the direction of earlier to later. If we try to deal with this kind of change in the way in which the theory deals with the qualitative changes that take place in the course of my experience, we shall be committed to making each term in the original series a term in a second series in a second time-dimension. We shall have events of the second order, viz., the becoming presented of events of the first order. In fact we shall be landed in an endless series of time-dimensions and orders of events which I mentioned in Sub-section 1.22 of this chapter. And this seems to me to be a most serious difficulty. (Broad, 1938, pp. 61–62) What lies behind these objections, once again, is Broad’s view that since the language that we use to record temporal facts—prima facie facts about the transitory and extensive aspects of time—are tenseless, the facts so recorded are tenseless, and hence non-temporal. In Broad’s fullfuture view, a reality without A-properties is not a temporal reality since there are can be no B-relations without A-determinations, and without B-relations there can be no time. But that is to assume that R-relations without A-properties are C-relations, and thus question begging against a Russellian theory of time. It is not surprising, therefore, that according to Broad, both McTaggart’s view and Russell’s lead to the unreality of time. There are several claims in the long passage just quoted that are subject to criticism and deserve a response. Admittedly, there are senses in which events don’t become or pass away on the R-theory, but the facts to

110  The Later Years which those experiences refer do have an ontological ground. Becoming does not involve events losing the property of futurity and taking on the property of presentness. Nor does it involve absolute becoming as the presentist conceives of it, as an entity that comes into existence when it becomes present. Rather, in the R-theory, passage is understood in terms of succession, and if one thing succeeds another, they must both exist. For that reason, becoming and ceasing to exist are understood in terms of something emerging and passing away. To emerge is to have a beginning and to pass away is to have an end. The beginning of a thing is its earliest temporal stage and the end is its latest temporal stage. In denying the transitory aspect of time in the R-theory, Broad might mean that it leaves out the sense in which time is dynamic; the sense in which we experience terms in the R-series as flowing from the future to the present and into the past; or how our experience of what is present is constantly changing. In Chapter 8, the Conclusion, I shall explain how these problems can be explained. Broad is treating time as space-like, or rather he is arguing that in the R-theory there is no intrinsic difference between a spatial series and a temporal series. That in order to account for the R-series as being a temporal series, consciousness is moving across it. The quality of presentedness lights up one event after another. But, if the R-series is thought of in this way, then there are no temporal relations between the terms of the series apart from consciousness and the property of phenomenal presentedness. Broad introduces the phenomenal property of presentedness to account for the phenomenological ground of an object being present to us in experience. He claims that for the theory under discussion to account for the transitory aspect of time, presentedness must move along the series of experiences, from which it allegedly follows that the R-theory must give rise to a second time dimension and ultimately an endless series of time dimensions. Thus, Broad’s interpretation of the R-theory is that the transitory aspect of time consists in the movement of presentedness (or consciousness) along a series of experiences, thus leading to an infinite regress of time dimensions. This is a version of the moving spotlight view. But that objection is based on a misunderstanding of the R-series as a nontemporal series or, as McTaggart understands, the B-series as without A-properties. Since Broad, like McTaggart, misconstrues the B-series (qua R-series) without A-properties as a non-temporal series of terms, he concludes that, A-characteristics cannot be analysed completely in terms of B-­relations, and that the notions of Time and Qualitative Change involve A-characteristics as well as B-characteristics. (1938, p. 102)

The Later Years  111 By essentially agreeing with McTaggart’s positive conception of time, Broad thereby becomes susceptible to his own critique of the full-future theory, “that it is hopeless to treat temporal becoming as a particular case of qualitative change; and this, in effect, is what the present theory tries to do” (1938, p. 57). Thus, unless Broad can give another way of understanding temporal becoming, given that B-relations are dependent upon A-characteristics, he cannot avoid the difficulties with temporal becoming as a species of qualitative change. And, as I shall argue further in Chapter 5, nor will it work for Broad to adopt a presentist explanation of passage and change.

V. Broad’s Critique of McTaggart’s Argument for the Unreality of Time According to Broad, McTaggart has two objectives in the chapter on “Time” in The Nature of Existence (1927). First, he seeks to establish that for there to be change there must be a B-series whose terms change with respect to their A-characteristics. Second, “McTaggart tries to prove that there is a contradiction involved in this condition, and therefore that nothing could be a temporal series” (1938, p. 63). I have discussed the first point in the previous section. In this section, I shall attend to the second. McTaggart argues that every event has all three (or an indefinite number of) incompatible A-characteristics, and that to avoid contradiction, each event has to have them at different times, or successively. However, this prima facie refutation of the alleged contradiction does not work. For, according to McTaggart, to say that an event e is past, present and future successively means either (e is now present, was future and will be past), or (e was present, is now past and was future) or (e is future, and will be present and past). Each of these disjuncts of conjuncts may seem to avoid the contradiction, but together they do not since their analyses re-creates the original contradiction, not at the level of events, but at the level of moments. Given McTaggart’s analysis of the tenses, e was future means that there is a moment t, such that e has futurity at t, and t is past; e is now present means that there is a moment t, e has presentness at t, and t is present; and e will be past means that there is a moment t, e has pastness at t, and t is future. Since it is also the case that any term that is past is also future and present, the contradiction is not removed, but is just shifted or transferred to from events to moments. Of course, this too can seemingly be avoided by maintaining that moments have incompatible A-properties successively, but since succession is analyzed in terms of has been, will be and is now, the contradiction of incompatible A-properties of moments is transferred to a second level of moments, and so on ad infinitum.

112  The Later Years Broad offers two main criticisms of McTaggart’s argument. His first criticism is that there is no contradiction to begin with since events have incompatible A-properties successively and not simultaneously or timelessly: I cannot myself see that there is any contradiction to be avoided. When it is said that pastness, presentness, and futurity are incompatible predicates, this is true only in the sense that no one term could have two of them simultaneously or timelessly. Now no term ever appears to have any of them timelessly, and no term ever appears to have any two of them simultaneously. What appears to be the case is that certain terms have them successively. Thus there is nothing in the temporal appearances to suggest that there is a contradiction to be avoided. (1938, p. 66) There are two points to make about Broad’s reply. First, McTaggart himself points out that “no event can have more than one A-determination since they are incompatible, but that every event has all of them successively” (1938, p. 66). Thus, it is not a failure to recognize succession, but rather it is McTaggart’s A-theoretic analysis of succession that is the problem. My second critical point starts with a question: What, then, is the proper analysis of succession? For McTaggart, succession is a B-­relation, but his view is that there cannot be B-relations unless there are A-­determinations, or more radically that B-relations are analyzable in terms of A-determinations. If, however, succession is necessary for A-change and A-change is necessary for succession, then the natural way to avoid any problem with the attribution of past, present and future to events by an appeal to succession is viciously circular. McTaggart summarizes this in the following passage: The attribution of the characteristics past, present, and future to the terms of any series leads to a contradiction, unless it is specified that they have them successively. This means, as we have seen, that they have them in relation to terms specified as past, present, and future. These again to avoid a like contradiction must in turn be specified as past, present and future. And, since this continues infinitely the first set of terms never escapes from contradiction at all. (1927, sect. 332, p. 22; emphasis added) This, it seems to me, is the heart of McTaggart’s reasoning, and cannot be refuted by claiming, as Broad does, that events have incompatible A-characteristics successively. I shall explore this mistake in greater detail in Chapter 5.

The Later Years  113 Broad raises a second criticism of McTaggart’s argument for the unreality of time. McTaggart argues that a vicious infinite regress arises when one takes the obvious step to avoid the initial contradiction. For if we say that events are past, present and future at different times, then with time and becoming connected to each other, these times must themselves be (tenselessly) past, present and future. Thus, for example, if event e was present is analyzed as e is present at time t and t is past, one is eliminating the temporal copula. In that case, however, the contradiction in events is shifted to moments and an infinite regress ensues. Broad views this aspect of McTaggart’s paradox as resting on his analysis of the tenses since it is based on a linguistic mistake. McTaggart reasons that the correct analysis of a tensed sentence must be in terms of a tenseless copula and a temporal predicate. Thus, McTaggart assumes that what is meant by a sentence with a temporal copula must be completely (and more accurately) expressible by a sentence or combination of sentences in which there is no temporal copula, but only temporal predicates and non-temporal copulas. And the regress arises because there remains at every stage a copula which, if taken as non-temporal, involves the non-temporal possession by a term of certain temporal predicates which could belong to it only successively. (1938, p. 66) Broad claims that the proper interpretation of the regress is that the assumption that temporal copulas are eliminable is false, and “the regress raises no objection to the prima facie appearance that events become and pass away and that they stand to each other in relations of temporal sequence and simultaneity” (1938, p. 67). That is, if temporal copulas are ineliminable, we can accept that that the ordinary language of time accurately records both the transitory aspect of time and temporal relations. In other words, Broad claims that TP is compatible with generation and annihilation so long as we do not eliminate temporal copulas. However, earlier in this chapter I argued that his account is unsuccessful. Broad discusses why he rejects McTaggart’s assumption that the temporal copula must be eliminated. He considers two different analyses of the sentences: “It will rain,” “It is now raining” and “It has rained.” He argues that neither of the analyses gets rid of the temporal copula “is now” and at least one A-property, namely, presentness, and for that reason he sees no reason to adopt either of them. He continues: Quite apart from the fact that such “analyses” serve no useful purpose, it seems to me that they fail to express what we have in mind when we use such sentences as “It has rained” or “It will rain.” When I utter the sentence “It has rained”, I do not mean that, in

114  The Later Years some mysterious non-temporal sense of “is”, there is a rainy event, which momentarily possessed the quality of presentness and has now lost it and acquired instead some determinate form of the quality of pastness. What I mean is that raininess has been and no longer is being, manifested in my neighborhood. When I utter the sentence “It will rain”, I do not mean that, in some mysterious nontemporal sense of “is”, there is a rainy event, which now possess some determinate form of the quality of futurity and will in course of time lose futurity and acquire instead the quality of presentness. What I mean is that raininess will be but is not now being manifested in my neighbourhood. (1938, pp. 67–68) Broad claims that his analysis of the tenses and generation and corruption “is compatible with any two events ‘always’ standing to each other in any temporal relation in which they ‘ever’ stand” (1938, p. 56), but it is not clear how the account he gives grounds the two facts in question, as I have argued. Nor does Broad otherwise elucidate this claim. Admittedly, Broad provides phraseology that is compatible with both the natural ways of expressing facts about relations of precedence and generation and corruption, but he fails to explain the ontological significance of these expressions. After all, the Russellian could also give an interpretation of the sentences to which Broad appeals, so why should we suppose that there is no becoming and passing away in the R-theory? The answer must be that Broad construes his account of annihilation and generation as ontologically loaded and therefore incompatible with the R-theory. He believes in annihilation and generation as the presentist would conceive of it, but clearly the R-theorist rejects that account, and the claim that events always stand to each other in relations of temporal sequence is certainly questionable. The prima facie appearance of temporal relations and annihilation and generation as reflected in ordinary language and experience is one thing; the proper logical form of the language that records a dialectically adequate ontological account of these facts is quite another. The analysis McTaggart puts forth may fail to express what we have in mind when we use sentences like “It has rained” or “It will rain,” whereas the original sentences “express the facts in the most natural and simple way without introducing temporal predicates in addition to temporal copulas” (1938, p. 67). So, we are still left to wonder what we mean ontologically speaking when we utter such expressions. Surely, when we ordinarily and commonsensically talk about the weather, we are not giving an ontological analysis of the phenomena that give rise to tensed sentences. Broad gives us an explanation, but it does not, I suggest, take us far in avoiding the difficulties McTaggart was getting at. Broad claims that “When I utter the sentence ‘It has rained’ . . . I mean that raininess has

The Later Years  115 been, and no longer is being manifested in my neighbourhood” (1938, pp. 67–68; emphasis added), but what does that amount to ontologically? “No longer is being manifested” means at least that it is not now manifested. He continues, “When I say ‘It will rain’ . . . I mean that raininess will be but is not now being manifested in my neighbourhood” (1938, p. 68). In both cases, there is something in common to “has been” and “will be,” namely, that raininess is not now being manifested, but that does not yet ground the difference between the past (has been) and the future (will be); between the no longer and not yet. But what is the ground of the direction of change, that a will be f occurred before a is no longer F, or that a will be generated before a is annihilated? The “not yet” and “no longer” or “will be” and “has been” either do or do not have ontological significance. If they do have ontological significance, then they either express A-properties or R-relations between events; and if they do not, they express absolute becoming. If absolute becoming is understood in terms of Raininess is now being manifested and “Fineness” (of weather) will be manifested, or Raininess was manifested and Fineness is now manifested, then it is not clear how those facts can ground the existence of succession. Since if the first conjunct obtains then a rainy event exists, and if the second, then a fineness event exists—but what is the ground in either of those facts that one succeeds or precedes the other? There is simply no ground for the fact that the rainy event exists before the nice weather event or the other way around, since there is no ground for “was” or “will be” because they are literally nothing on the presentist analysis. Broad fails to ground anything in the sense that he refuses to commit himself to any explicit ontological analysis. Broad summarizes McTaggart’s mistakes as analogous to those made by the ontological argument, namely, to take presentness to be a predicate, i.e., a property of events (like taking existence to be a predicate) rather than a claim about certain characteristics being manifested or instantiated. Similarly, those who countenance A-properties mistake the relation between presentness, pastness and futurity, and what has those properties (the characterizing tie), with the description of a thing and what is instantiated by that description. The result is to think of absolute becoming as a species of temporal becoming, i.e., changing A-properties. McTaggart is guilty of this; he thinks that temporal copulas can be replaced by non-temporal copulas and temporal properties. Both claims are the result of linguistic confusions. Broad summarizes these objections in the following passage: McTaggart’s main argument against the reality of Time is a philosophical “howler” of the same kind as the Ontological Argument for the existence of God. The fallacy of the Ontological Argument consists in treating being or existence as if it were a predicate like

116  The Later Years goodness, and in treating instantial propositions as if they were characterizing propositions. The fallacy in McTaggart’s argument consists in treating absolute becoming as if it were a species of qualitative change, and in trying to replace temporal copulas by non-temporal copulas and temporal adjectives. Both of these “howlers”, like the Fall of Adam, have been over-ruled to good ends. In each case one can see that there is something radically wrong with the argument’ and one’s desire to put one’s finger on the precise point of weakness stimulates one to clear up linguistic confusions which would otherwise have remained unnoticed and unresolved. (1938, p. 68; emphasis added) To sum up, McTaggart maintains that B-relations without Adeterminations would not constitute a genuine temporal series since it would not contain existential or qualitative change. McTaggart’s argument is that the B-series implies the temporal principle and TP is incompatible with qualitative and existential change, unless we countenance A-properties. Broad argues that McTaggart’s argument is fallacious since the two notions, namely, the temporal principle and existential change are compatible, and his arguments against qualitative ignore Russell’s view and so are irrelevant. The problem for Broad is that in his interpretation of TP, it implies that there are A-properties, and therefore it defends rather than refutes McTaggart’s argument against the R-theory. Furthermore, Broad’s interpretation of existential change (generation and annihilation) suggests a presentist ontology that is arguably incompatible with the temporal principle. For in presentism, there are no temporal relations and presentism is also incompatible with his full-future account that is required for Broad’s analysis of TP. Thus, the main problem of Broad’s chapter on “Ostensible Temporality” is that he adopts both presentism and the full-future theory, but each is problematic for him. For, on the one hand, if presentism is true, then it is not clear that he could ground his commonsense claim that TP is true. On the other hand, if he maintains the full-future theory and events do become and pass away in virtue of changing A-properties, then there is the problem Broad himself raises, namely, that the full-future theory gives rise to an infinite regress of time dimensions. In the next chapter, I shall argue that the best way to avoid McTaggart’s paradox is by endorsing the R-theory and the timelessness (eternal3) of time.

Notes 1. This is clearly correct regarding what he says in “Ostensible Temporality” (Broad, 1938, p. 183), but it is only there where he seems to support the growing block view. 2. Given Broad’s empiricist leanings, he attempts to provide “an account of the experiential basis of the notion of events-particles, and of the assumption that any process of finite duration can be regarded as consisting of a compact series

The Later Years  117 of successive event-particles” (1938, p. 44). The experience of spatial and temporal boundaries is the basis of lines without breadth or color, and events without duration or sound. He provides a phenomenological analysis of our experience of instantaneous events or event particles and points that have no position or magnitude by an appeal to boundaries. For a criticism of Broad’s argument for the experiential basis of the notion of even-particles see, Oliver Rashbrook, 2012. 3. Broad (1938, pp. 52, 57–58) mentions McTaggart’s (1908) view of temporal relations that I discussed in Chapter 3, what he calls the more radical view that B-relations can be defined in terms of A-determinations, and A-determinations cannot be defined in terms for B-relations. Broad’s discussion assumes that it makes sense for the B-series and the A-series to exist at a moment; and that at every moment, the B-series constitutes an A-series in virtue of its one-to-one correspondence with the terms of the series of A-characteristics and vice versa. As I have already argued, it does not make sense to speak of the B-series qua R-series at a moment (whether construed substantively or relationally) because then all the events in the world would exist at the same time, as a totum simul (or a second time dimension would be required). Moreover, it would imply that temporal relational facts and, indeed, temporal relations themselves, exist in time. To put the point otherwise, if the entire A-series exists at a moment, and the terms of the A-series and the B-series are in a one-to-one correspondence, then all the terms of the B-series must also exist at that same moment. In that case, the B-series qua R-series, could not be a temporal series, that is, a series generated by succession.    This problem arises again when Broad claims that the difference in magnitude between two events in the B-series is correlated with a difference in the degree of the A-characteristics possessed by the related events. Broad explains this in the following passage: Every event-particle is continually changing in respect of its A-characteristics; it is continually getting less and less remotely future, then present, and then more and more remotely past. But any two event-particles keep the same algebraic difference between the A-characteristics. They are always simultaneous or always successive. And, if they are successive, the earlier always precedes the later by the same amount. (1938, p. 128) Broad claims that two successive event’ particles will always be successive: “the earlier always precedes the later by the same amount.” However, if B-relations are construed as R-relations, then neither the relation nor the relational fact can always (that is, at every time) be the case, since R-relations are timeless and R-facts are eternal3. That is, e is earlier than e’ is not in time and so cannot “always” be the case, even though e and e’ exist in time. 4. In claiming that in Russell’s view “Events are neither generated nor annihilated,” Broad must mean that there is no absolute becoming. Or, as he puts it, “the present is a mere transition from one infinite non-existent to another” (1921, p. 150). It doesn’t follow from that, however, as Broad later claims that, in some sense of “is,” there “is” timelessly or sempiternally all that there ever has been or will be of the series (1938, pp. 307–308). But in the R-theory, there is a sense in which the past no longer exists, i.e., it has passed away and the future does not yet exist, i.e., it is yet to emerge. Nor does the rejection of absolute becoming require the reality of tense. 5. I am italicizing “tense” to mean that the terms—objects or facts—have ­temporal properties, or indicate some other, for example, presentist, A-theoretic (non-R-theoretic understanding) is involved.

118  The Later Years

References Broad, C. D. (1921). Time. In J. Hastings et al. (Eds.), Encyclopedia of religion and ethics (Vol. 12, pp. 334–345). New York: Scribner’s. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 143– 173). References are to this reprint. ———. (1938). Examination of McTaggart’s philosophy: ostensible temporality (Vol. II (1), Chap. XXXV, pp. 265–323). Cambridge: Cambridge University Press. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 36–68). References are to this reprint. McTaggart, J. M. E. (1908, Oct.). The unreality of time. Mind, 17 (68), 457–474. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 21–35). References are to this reprint. ———. (1921). The nature of existence (C. D. Broad, Ed.). Cambridge: Cambridge University Press. ———. (1927). The nature of existence (Vol. II, C. D. Broad, Ed.). Cambridge: Cambridge University Press. Oaklander, L. N. (Ed.). (2008). The philosophy of time: critical concepts in philosophy (Vols. I–IV). London: Routledge. Tegtmeier, E. (2007). Three flawed distinctions in the philosophy of time. Metaphysica, 8 (1), 53–59. https://doi.org/10.1007/s12133-007-0005-8

5 Independent Account of McTaggart’s Paradox and the R-theory of Time McTaggart’s Paradox and R-theory of TimeMcTaggart’s Paradox and R-theory of Time

I. Introduction In his valuable book on McTaggart’s Paradox, R. D. Ingthorsson (2016) makes some provocative claims regarding the scholarship on McTaggart’s argument for the unreality of time. Most notable is his claim that the key assumption that McTaggart employs in his argument, namely, the principle of temporal parity—the view that all times (whether A-times or B-times) exist equally or co-exist—is a basic tenet of the B-theory, and therefore McTaggart’s paradox cannot be used to support the B-theory. In other words, McTaggart’s argument for the unreality of the A-series rests on a premise that also undermines the B-series and the B-theory of time. Ingthorsson expresses this point as follows: But, in the end, the central question still is, do all times exist in parity or not? If they do [as the B view maintains], then how should we understand such a reality as temporal? That is, wherein lies the temporality of earlier than and later than? (2016, p. 141) A further provocative claim Ingthorsson makes is that virtually all commentators on McTaggart’s philosophy of time have failed to recognize the connection between his a priori metaphysics, that is, his view of Absolute Reality and his view of Present Experience. Ingthorsson maintains McTaggart’s argument for time’s unreality is intended to demonstrate that time as it is given to us in experience, as stated in the chapter “Time” in Volume II of The Nature of Existence (1927), is incompatible with his view of Absolute Reality found in Volume I (1921). For that reason, Ingthorsson is highly critical of Broad (1933, p. 9), who claims that McTaggart’s paradox is a stand-alone argument that is not dependent on the results in the first part of his enquiry. Another controversial thesis of Ingthorsson’s book is that McTaggart is not concerned with the language of time, but the ontology of time. McTaggart’s argument is not based on the proper semantic analysis of sentences which record the facts of temporal becoming, such as “It is raining” and

120  McTaggart’s Paradox and R-theory of Time “It’s raining has been future and will be past,” but with their ontological analysis. Ingthorsson argues that given the principles that underlie McTaggart’s substance ontology, the ontological ground of time as we experience it and express it in ordinary language cannot exist, and since McTaggart believes that only what exists is real, he concludes that time is unreal. Thus, it is a mistake to claim as Broad (1938), Lowe (1987) and others have done that McTaggart’s argument is based on linguistic confusions. At the outset, I would like to say that I applaud Ingthorsson’s emphasis on the ontology that underlies McTaggart’s argument, and given the ontology he attributes to McTaggart, would not contest that he demonstrates that time is unreal. So, I wholeheartedly agree with his emphasis on the ontology and not the language of time. However, I shall argue that a textual case can also be made for questioning the claim that McTaggart’s a priori metaphysics developed in 1921 is what he employs to demonstrate the unreality of time in 1927. There is an important difference between the ontology Ingthorsson attributes to McTaggart, and the ontology McTaggart employs to generate his paradox. Indeed, the ontology presupposed by McTaggart’s argument (the doctrine of external relations) is inconsistent with his prior commitments. Therefore, not only is Ingthorsson mistaken about McTaggart basing his argument on his prior ontology, I will show that McTaggart is inconsistent in that he bases the argument on an incompatible (Russellian) ontology. Furthermore, I would also question whether McTaggart’s argument for the unreality of time shares a common assumption with Russell’s view that McTaggart claims to be critiquing in the chapter on “Time”—that the B-series alone constitutes time. Ingthorsson claims that just as McTaggart’s ontology implies that all temporal positions whether past, present or future co-exist equally even if they are not present, in the B-theory all times co-exist equally even if they are not simultaneous, but successive. Indeed, it is the co-existence of past, present and future that is common to McTaggart and the B-theorists. As Ingthorsson puts it, To illustrate the fundamental difference between A theorists, on the one side, and McTaggart and the B theorists on the other [we can say:] The former cannot understand why we should think of the future and past as an existing reality—at least not one separate from what exists in the present—and thus fail to see why we should interpret claims about the future and past as claims about something non-present and yet existing, whereas the latter cannot understand how we can fail to think of the future and past as co-existing with the present, even though they are not co-located in time. (2016, p. 89) Ironically, Ingthorsson blames Broad for setting the discussion of McTaggart’s paradox on the wrong path, but it is precisely the same ontological

McTaggart’s Paradox and R-theory of Time  121 assumption—the principle of temporal parity; the view that all times or events co-exist—that Broad, like Ingthorsson, attributes to McTaggart and B-theorists, and both argue is mistaken. Ingthorsson’s thesis is that the only adequate A-theory is presentism, since it rejects temporal parity, whereas all other theories follow McTaggart and B-theorists in accepting temporal parity, and thereby fail. Thus, Ingthorsson maintains that the fundamental debate is not between A- and B-theorists, but between those who adopt the temporal parity thesis (McTaggart, B-theorists and others)1 and those who deny it (presentists). I shall argue, however, that there is a third alternative that involves a Russellian (or R-theoretic) ontology of time that differs from presentism, McTaggart and the B-theory, as Ingthorsson understands them. The debate between R-theorists and Ingthorsson’s version of presentism is between those who view the spatio-temporal universe as the sole purview of reality and those who view the world as greater than the universe, including non-spatial and non-temporal objects. The fundamental debate, in other words, is between naturalists and ontologists—to use terms that Reinhardt Grossmann (1992) did to d ­ istinguish a fundamental philosophical divide. To explain what I consider to be wrong with how Ingthorsson characterizes the fundamental agreement between McTaggart and the B-theorists, it will be necessary to distinguish McTaggart’s understanding of B-relations and the B-series from a Russellian understanding of R-relations and the R-series.2 Given the incommensurability between McTaggart’s and Russell’s analyses of the “B-series,” it is a mistake to believe that the temporal parity thesis in McTaggart’s argument against the A- and B-series can also be employed against the R-series to demonstrate that “the conception of temporal reality as a series of events . . . related as earlier/later than each other . . . is a conception of a changeless reality and consequently not a conception of a temporal reality” (2016, p. 92). I shall proceed by briefly summarizing McTaggart’s ontology as Ingthorsson understands it, and how it differs from Russell’s ontology, in section II. Then, in section III, I will explain key elements of the Russellian or R-theory of time and the principle of temporal parity, demonstrating how they differ. In the fourth section, I discuss McTaggart’s paradox and show how it not only undermines the A-series, but that it is also applicable to presentism as Ingthorsson conceives of it. Since his paradox can be interpreted to show that succession, construed R-theoretically, is the ground of the passage of time, and since presentism rejects cross-temporal relations, McTaggart’s argument undermines Ingthorsson’s presentism as well.

II. McTaggart’s Substance Ontology and Russell’s Fact Ontology According to Ingthorsson, McTaggart’s metaphysics “is the ultimate substance ontology and alluring in its simplicity. Reality is grounded in

122  McTaggart’s Paradox and R-theory of Time the existence of substances that bear properties and stand in relations” (2016, p. 15). Regarding substance, McTaggart claims There is a substance which contains all existent content, and of which every other substance is a part. This substance is called the Universe. A universe contains all existent content. Or we can define the Universe as a substance of which all other substances are parts. (1921, sect. 135, p. 148) As Ingthorsson interprets him, McTaggart presupposes that time, if real, is a compound whole whose every part is equally existent and real. . . . Since existence and reality coincide, and every existing thing is in time, then by time he includes the whole of reality. There is no distinction to be drawn between existence, reality and time, in terms of their constituents. . . . In other words, time cannot be external to existent reality but must be an existing part of it. (2016, pp. 23–24, 29; emphasis added) McTaggart says that if any reality is in time, then it must exist. That may be true, but it does not follow that time itself is in time or that time is an existent somehow in the universe (the realm of concrete objects). Clearly, there are things in the universe that change, but time, which is necessary for change, need itself not be in the universe and it need not change. There may be a content greater than the universe; there may be a world that contains constituents that exist but are not in time. I shall argue that in such a world, temporal relations and temporal facts (and all other relations and facts) exist. These relations and facts do not change but are what make change possible. Thus, it is a mistake to argue, as McTaggart does, that since the relations between events in a time-series do not change and the fact that say, an apple is green before it is red does not change, there is no time or change, unless A-change or becoming is introduced. To see what is involved in these points, it will be useful to consider a distinction that Ingthorsson blurs between the notion of “fact” in McTaggart’s substance ontology and in Russell’s fact ontology. On a substance ontology, the only ontological categories are substances, qualities and relations. Facts do not belong to a category of their own, and so must be understood as being either a substance, quality or relation. This is McTaggart’s view, according to Ingthorsson: We need to become familiar with McTaggart’s understanding not only of the general nature of the fundamental building blocks (substances, qualities, relations), but also of whatever has to do with

McTaggart’s Paradox and R-theory of Time  123 the distinction and connection between thought and reality, such as beliefs, assumptions, assertions, facts, truth and falsity. With respect to the latter, one must understand that they do not form a separate category of entities distinct from substances, qualities, and relations. (2016, p. 16; some emphasis added) On the other hand, Ingthorsson also claims that McTaggart’s account of facts is in fact very similar to the account given by the logical atomists of an atomic fact, notably that it consisted “either in the possession by a particular of a characteristic, or in a relation holding between two or more particulars” (Urmson, 1960, p. 17, 2016, p. 28; emphasis added) Surely, these two notions of “fact,” although verbally similar—both are called “states of affairs”—are ontologically quite different. In a fact ontology, such as logical atomism, there are particulars, nonrelational and relational universals, and facts. If a particular exemplifies a non-relational quality, those two different kinds combine to form a fact which is an entity over and above its constituents. It is a unity with complexity. If some particular stands in a relation to another, for example, if a is earlier than b, there is a temporal relational fact. Facts do not exist in time (or space), and the relational universals that are constituents in them do not exist in time (or space) either, although the terms (particulars) of temporal relations do exist in time. The philosophy of logical atomism and the notion of an atomic fact is associated with Russell. He treats facts as a category of their own over and above their constituents when he says, Facts are . . . plainly something you have to take account of if you are going to give a complete account of the world. You cannot do that by merely enumerating the particular things in it: you must also mention the relations of these things, and their properties, and so forth, all of which are facts, so that facts certainly belong to an account of the objective world . . . and the things and their qualities or relations are quite clearly in some sense or other components of the facts that have those qualities or relations. (Russell, 1918, pp. 191–192) A Russellian fact is a single (unity), yet complex entity that cannot be reduced to its constituents, and for that reason forms a separate category of entities distinct from substances (particulars), qualities and relations. Despite Ingthorsson’s attributing a substance ontology to McTaggart and his subsequent denial of the separate category of facts, McTaggart

124  McTaggart’s Paradox and R-theory of Time seems to recognize the category of fact and, like Russell, construes (some) facts as timeless even before his denial of the reality of time, since he says: When the subjective belief is eliminated, it seems to me that the truth goes with it and that we find ourselves left, not with a timeless, nonexistent, and true proposition, but with nothing but the fact, which is not true (though it determines the truth of beliefs), which may or may not be timeless, and which, as we have seen above, is always in one way or another, existent. (1921, sect. 38, pp. 35–36) In the context of discussing the question of whether there are entities that are real but non-existent, McTaggart distinguishes substance, quality and relations categorically: Qualities and relations are very different from substances, and the fact that a substance cannot be both existent and non-existent does not prove that qualities and relations—which are universal, and not particular as substances are—could not be existent in one aspect, and non-existent in another. (1921, sect. 5, p. 6) Although McTaggart ultimately rejects the reality of qualities and relations as real but non-existent (unexemplified), he accepts Russell’s distinction between universals and particulars. McTaggart claims “I am aware of an object or am acquainted with an object—the phrases are used synonymously—when ‘I have a direct cognitive relation to that object’” (1921, sect. 44, p. 40). He then approvingly quotes Russell: “When we ask what are the kinds of objects with which we are acquainted, the first and most obvious example is sense-data. When I see a colour or hear a noise, I have direct acquaintance with the colour or the noise. . . . But in addition to the awareness of the above kind of objects, which may be called awareness of particulars, we have also . . . what may be called awareness of universals. And universal relations, too, are objects of awareness; up and down, before and after, resemblance, and so on, would seem to be all of them objects of which we can be aware.” (From Russell, 1912, pp. 209–212; and quoted in McTaggart, 1921, sect. 44, pp. 40–41; emphasis added) This is striking because Russell used direct acquaintance with before and after as grounds for maintaining that they are primitive, simple temporal relations.

McTaggart’s Paradox and R-theory of Time  125 Further evidence that McTaggart takes qualities and relations to be ultimate, and indefinable, ineliminable universals and distinct from particulars is his appeal to Russell’s argument against reducing the common quality of two particulars to the simple relation of exact likeness between them. McTaggart (1921, sect. 83, n1, p. 83) cites Russell’s argument in “On the Relations of Universals and Particulars” to justify the irreducibility of qualities as universals because the argument against qualities requires relations as (timeless) universals: Likeness at least, therefore, must be admitted as a universal, and, having admitted one universal we have no longer any reason to reject others. Thus, the whole complicated theory, which had no motive except to avoid universals, falls to the ground. . . . There must be relations which are universals in the sense that (a) they are concepts, not percepts; (b) they do not exist in time; (c) they are verbs, not substantives. (Russell, 1911–12, p. 9) So, we can say that in certain respects, McTaggart’s ontology has similarities with Russell’s. There exist substances (particulars) that stand in relations and have non-relational qualities. Qualities and relations are universals (they do not exist in space or time) and substances are particulars in that, if time exists, are the terms of temporal relations and so exist in time. Surprisingly, then, in 1921 (McTaggart, 1921), Russell’s influence on McTaggart is apparent in his treatment of relations, although there is also a fundamental difference. McTaggart claims that the conception of relations is indefinable, since it is impossible to substitute any other concepts for it which can be taken as equivalent (1921, sect. 80, p. 80). He notes that theories have been put forth to demonstrate “that relation, though valid of existence, was not ultimate, but definable in terms of quality, so that statements about relations could be translated into statements about qualities” (1921, sect. 80, p. 81; emphasis added). McTaggart maintains, however, that these theories are mistaken. “No fact which can be stated in terms of relations between substances can ever be stated in terms which omit the conception of relation” (1921, sect. 82, pp. 82–83). In other words, McTaggart initially rejects the view that relations can be reduced to or eliminated by the qualities of one or both of the terms of the relation. Although McTaggart argues that relations are indefinable, irreducible and ineliminable in terms of qualities, he does say that if there are relations then there are qualities generated in each of the terms of the relation. McTaggart summarizes these points in the following passage: The conception of relation, then, must be accepted as valid of the existent. But it might be admitted to be valid, and yet denied to be

126  McTaggart’s Paradox and R-theory of Time ultimate and indefinable. It might be said that it really was true that substances were in relations, but that the fact expressed in this way could be expressed in terms of qualities only, without bringing in relations. But this also is false . . . A relation may no doubt be based on a quality in each of its terms. But this does not mean that it can be reduced to those qualities. If A is larger than B, this relation may depend on the fact that A covers a square mile, and B covers an acre. . . . But a statement of the size of A and a statement of the size of B are not equivalent to a statement that A is larger than B, though the latter may be a certain and immediate conclusion from them. (1921, sect. 82, pp. 82–83) It is quite clear, therefore, that McTaggart (1921) accepts the existence of relations, including temporal relations, and in his discussion of cognition (1921), he quotes Russell who maintains that we are acquainted with them. Thus, there are three views of relations that McTaggart considers. First, that relations are definable, reducible or eliminable in terms of qualities; a view McTaggart consistently rejects in 1921. Second, that relations are not definable or reducible to the qualities of their terms but depend on them; McTaggart’s view in the passage just quoted. Third, Russell’s view that relations are indefinable and irreducible to qualities and are not dependent on the qualities (temporal or otherwise) of their terms. McTaggart seems to have some sympathy with this Russellian view in his discussion of cognition quoted previously. In the next section, I shall argue that McTaggart’s argument against the B-series, the A-series, and thus against the reality of time does not depend on the principle of temporal parity as Ingthorsson understands it, but on his changing analysis of temporal relations. For that reason, McTaggart’s argument in 1927 should or at least could, as Broad maintained, be understood as a standalone argument that does not depend on his a priori metaphysics in 1921.

III.  The R-theory and the Principle of Temporal Parity Richard Gale refers to Russell as “the father of the B theory,” and indeed, there is a good reason for him to do so, since McTaggart claims that “Mr. Russell’s time-series . . . is identical with our B series . . . and the relation which unites the terms of the B-series is the relation of earlier and later” (1927, sect. 351, p. 31). However, the ontology of the B-series as McTaggart understands it is not the same as the Russellian time-series as it should be understood. It will be useful, therefore, to clarify Russell’s authentic view in contrast to the B-theory which is McTaggart’s misrepresentation of Russell but has nevertheless been accepted generally as Russellian.3 A brief discussion of some of the differences between the

McTaggart’s Paradox and R-theory of Time  127 Russellian theory of time (“R-theory” for short), and standard B-theoretic accounts as understood by defenders and critics alike is in order. In typical interpretations, the ontology of B-time is construed as antirealist because it denies that temporal passage is an objective, mind-­ independent feature of reality. For that reason, B-relations and the B-facts they enter, that alone constitute the foundation of the B-theory of time, are “nontransient” and static in that what appears to be the flow and flux of events in time—time’s dynamism—is an illusion that would not exist without consciousness.4 On the other hand, R-relations as given in experience are not static, but dynamic, and are the basis of our experience of transition and the passage of time. Since the R-theory rejects the common view that B-time is a static, block universe, this last point deserves attention. I have the knowledge that time has passed when, for example, I reflect on successive changes along my life’s way (when I got married, when my first child was born, when my first grandchild was born and so on), or when I see that the position of the minute hand on my watch has changed its position. There is, however, a more immediate way in which we know that time is passing. This occurs when we are directly aware of passage, when we experience change in a single act of awareness. For example, if we look at a second hand of a watch or a flickering flame, we see the moving second hand at one place on the watch before the other, or one flicker of the flame occur before another in a single act of awareness. These are cases where we experience two stages of the second hand or flame occur in succession, one earlier than the other, and in so doing we are directly aware of a temporal transition or passage from one stage to the other. Similarly, when we hear the successive notes of a tune or feel the successive taps of a physician on our abdomen during a physical, we directly experience that time is passing. In the R-theory, an appeal to the direct experience of succession in a single act of awareness is the basis for grounding our most basic experience of the flow or passage of time on mind-independent temporal earlier/later than relations alone; a view I will call the temporal relational theory or Russellian theory. In the R-theory, the commonsense belief that time passes is to be understood in terms of the relation of succession between earlier and later temporal objects. Thus, the passage of time consists in the succession of events throughout the history of the universe; one event—the earlier—being followed by another—the later. Russell gives a colorful example of our experience of the earlier than relation in the following passage: Immediate experience provides us with two time-relations among events: they may be simultaneous, or one may be earlier and the other later. These two are both part of the crude data; it is not the case that only the events are given, and their time-order, within certain limits,

128  McTaggart’s Paradox and R-theory of Time is as much given as the events. In any story of adventure you will find such passages as the following: “With a cynical smile he pointed the revolver at the breast of the dauntless youth. ‘At the word three I shall fire,’ he said. The words one and two had already been spoken with a cool and deliberate distinctness. The word three forming on his lips. At this moment a blinding flash of lightning rent the air.” Here we have simultaneity—not due, as Kant would have us believe, to the subjective mental apparatus of the dauntless youth, but given as objectively as the revolver and the lightening. And it is equally given in immediate experience that the words one and two come earlier than the flash. (Russell, 1914, pp. 116–117) Returning to the differences between B- and R-theorists, in standard (reductionist) B-theories, B-relations are analyzable in terms of causal relations whereas the R-theory takes R-relations as primitive and unanalyzable, relational universals that can be directly experienced. Russellian temporal relations are external relations, since there are such facts as that one object has a certain relation to another, and that such facts cannot be reduced or inferred from, a fact about the one object only together with a fact about the other object only: they do not imply that the two objects have any complexity, or any intrinsic property distinguishing them from two objects which do not have the relation in question. (Russell, 1966, pp. 139–140) For that reason, R-relations are neither analyzable in terms of A-­properties of their terms nor do they depend on A-properties. Indeed, in the R-theory there are no such properties. In the Russellian view, temporal relations are not to be identified with causal, spatial, entropic or any other kind of relations. Thus, the R-theory differs from most B-theorists who analyze the direction of time from earlier to later in terms of the direction of causality (Leininger, 2014; Mellor, 2009; Le Poidevin, 1991, 2007). For in the standard B-theory, the direction of time is founded upon the direction of causation or entropy, but in a Russellian relational ontology there are reasons to reject such grounding. First, the phenomenon of temporal succession is fundamental, whereas causation and entropy are rather derived and complicated relations. A Russellian ontological analysis complies with the principle that a fundamental phenomenon should be grounded on a simple entity if at all possible. Second, Russellians will reject causal accounts of the direction of time since they adopt an empirical principle of acquaintance according to which we must be acquainted with the simple entities of one’s ontology. What excludes causal theories of time is that we perceive

McTaggart’s Paradox and R-theory of Time  129 many temporal successions while we don’t perceive the relation “causes” or “has greater entropy than” in those cases. Furthermore, Russellians will argue that causation and entropy are circular as earlier relations since they presuppose temporal succession and the direction of time. A last difference is particularly important for the discussion to follow. The B-theory is often identified with McTaggart’s (1927) misinterpretation of Russell, according to which B-relations are unchanging and B-facts are permanent in that if a is ever (at any time) earlier than b, then a is always earlier than b. In contrast, R-theorists do not believe that either R-relations or R-facts exist in time, much less at every time, as McTaggart’s interpretation implies. Earlier than is a timeless yet dynamic temporal relation. It is timeless because it does not exist in time as a term of a temporal relation. It is dynamic because it is the ground of our experience of the passage of time—of successively existing temporal objects that exist tenselessly, that is, without tensed or A-properties. Similarly, time, understood as a Russellian series composed of a conjunction of R-facts, is timeless or atemporal. This view gives some meaning to an aphorism I favor, namely, time is timeless, or eternal3 in just this sense: though time contains temporal relations, time does not exemplify them. Ingthorsson argues that McTaggart’s paradox rests on a premise that can also be used to undermine the B-theory of time. The only view left standing is presentism. The premise is the “temporal parity thesis,” the view that all times exist equally or co-exist, in a sense that is compatible with their being successive and not simultaneous or timeless. I disagree and shall argue that McTaggart’s argument rests on a premise other than temporal parity, whose implications refute the B-series (as McTaggart understands it), the A-series (whose terms are past, present and future) and the A-theory including presentism, but when the dust settles, leaves the R-theory unscathed. This is a surprising conclusion, but true nonetheless, or so I shall argue. The premise that plays a crucial role in McTaggart’s argument against the B-series, the A-series and the unreality of time is not the principle of temporal parity, as Ingthorsson understands it, but McTaggart’s analysis of “earlier than.” He states that analysis in the following passage: The series of past, present and future is what we have called the A series, on which the B series of earlier and later is dependent. The term P is earlier than the term Q, if it is ever past while Q is present, or present while Q is future. (1927, sect. 610, p. 271; emphasis added) This premise is nowhere argued for but is assumed in his argument against the B-series and the A-series. Moreover, it is incompatible with his account of relations in 1921 (McTaggart, 1921), and so it is incompatible with the a priori metaphysics found there. Recall, that in 1921

130  McTaggart’s Paradox and R-theory of Time (1921) he argued that relations are indefinable, and that while the terms of relations have qualities on which they may be based, they cannot be reduced to those qualities. However, in a footnote accompanying the previous passage, McTaggart claims that “earlier than” can be defined in terms of temporal qualities (or relations to a term X outside the A-series). I shall argue that this analysis is crucial to his argument for the unreality of time, but is incompatible with the temporal parity thesis and his earlier view of relations. For that reason, McTaggart’s (1927) argument can be construed as a stand-alone argument not dependent upon his a priori metaphysics in 1921 (1921). We shall see that, even more importantly, McTaggart defends his analysis of earlier than by appealing to his rejection of the B-series, but his rejection of the B-series implicitly appeals to his analysis, and so massively begs the question against an R-theoretic interpretation of the B-series. It also implies an understanding of the temporal parity thesis that underscores a difference between McTaggart and R-theorists, even if McTaggart and B-theorists can be grouped together. To see what is involved in these points, let us return to McTaggart’s text. One problem with McTaggart’s account of “earlier than” is that the word “while” implies that each of the disjuncts exist in time and thus the analysis is circular. For if P is past at t1, and Q is present at t1, then that is reducing “is past at” and “is present at” to the relations “is earlier than” and “is simultaneous with.” If he denies absolute time, then “while” would imply simultaneity. Then “Socrates is past” is just as real as or is simultaneous with “Oaklander is present.” In that case, temporal parity implies co-existence in the sense of simultaneity, and that is incompatible with co-existing terms being successive. Finally, if “while” means co-present or existing at the same NOW, then we still haven’t grounded for example, Socrates existing earlier than Oaklander because Socrates being past and Oaklander being present are both facts that exist now.5 Leaving that problem aside, McTaggart says that there seems to be a counterexample to his account of “earlier than” since in a durational present we are acquainted with the earlier than relation where both terms are present and thus neither are past or future. His reply is instructive: Two terms may both be present together, although one is earlier than the other. This is due to the fact that the present is a duration, and not an indivisible point. But the statement in the text remains an adequate definition of “earlier than,” for although P and Q may at one time be in the same present, yet, before that, P is present while Q is future and after that, P is past while Q is present. (1927, sect. 610, n 1, p. 271; emphasis added) McTaggart’s response to the alleged counterexample is that if it is ever the case that P is past while Q is present, or P is present while Q is

McTaggart’s Paradox and R-theory of Time  131 future, then P is earlier than Q and the antecedent is satisfied because before P and Q are both present, P is present while Q is future, and after they are both present, P is past while Q is present. Thus, McTaggart says that the statement in the text remains an adequate definition of “earlier than.” However, to avoid an objection to his definition of “earlier than” by appealing to “before” and “after” is obviously circular. We saw that his appeal to “while” in his statement of the analysis of “earlier than” is also problematic. This is important because it shows the need for primitive R-relations to account for the phenomenology and ontology of the earlier than relation and temporal passage. Indeed, the McTaggart paradox is the result if we attempt to ground time without them, as we shall see. McTaggart raises another objection to his analysis of “earlier than” that mirrors Russell’s analysis of the tenses in “Our Experience of Time” (1915) that McTaggart was obviously familiar with. McTaggart says that since we can perceive the earlier than relation without perceiving the A-properties of pastness and futurity, perhaps we can take earlier than as primitive (as the R-theorist would have it) and define the future as what is later than the present, and the past as what is earlier than the present. He says, Since the present comprises different terms, of which any one will be earlier or later than any other, it might be thought that the fact that P was earlier than Q would be perceived when they were both present, and that “earlier than” need not be defined in terms of the A series. After this, it might be thought, the future may be defined as what is later than the present, and the past as what is earlier than the present. Thus, the A series would be defined in terms of the B series, instead of the B series in terms of the A series. (1927, sect. 610, n 1, p. 271; emphasis added) In other words, McTaggart considers the view that the B-series alone constitutes time and that the A-series is not needed in a complete ontology of time, given that we are directly acquainted with earlier than without being acquainted with A-properties. McTaggart’s response is familiar since it harkens back to his argument against the B-series in his chapter on “Time” (1927). He replies to this argument by saying that to suppose that the B-series alone constitutes time would be a mistake. For the series of earlier and later is a time series. We cannot have time without change, and the only possible change is from future to present, and from present to past. Thus, until the terms are taken as passing from future to present, and from present to past, they cannot be taken as in time or as earlier and later; and

132  McTaggart’s Paradox and R-theory of Time not only the conception of presentness, but those of pastness and futurity, must be reached before the conceptions of earlier and later, and not vice versa. (1927, sect. 610, fn 1, p. 271; emphasis added) McTaggart’s argument against a B- or R-theoretic reduction of A-determinations is that earlier and later are temporal relations that generate a time-series, only if their terms form an A-series and change A-properties with the passage of time. His justification of that is the argument quoted previously that says nothing changes in the B-series alone since there is no temporal passage, and without events changing their A-determinations there are no temporal relations. I shall show, however, that this does not constitute an argument against the R-theory, since it mischaracterizes the R-series as a B-series and construes the B-series in a way that assumes McTaggart’s definition of “earlier than.” McTaggart’s argument against the B-series also shows that temporal parity takes the terms of the B-series to be simultaneous, sempiternal or timeless (eternal1) and not successive, and so cannot be employed against the R-theory. To see what is involved, consider McTaggart’s argument against the view that there can be a temporal series without the A-series. McTaggart argues that since the relations between the terms of the B-series are permanent, nothing changes on the B-series by coming into and going out of existence: If N is ever earlier than O and later than M, it will always be, and has always been, earlier than O and later than M since the relations of earlier and later are permanent. N will always be in the B series. And as, by our own hypothesis, a B series by itself constitutes time, N will always have a position in a time-series, and always has had one. That is, it always has been an event, and always will be one, and cannot begin or cease to be an event. (1927, sect. 310, p. 12) Generally, McTaggart says what he means, and means what he says. He does not say that “if it is ever true that N is earlier than O and later than M, it always will be true and always has been true that.” He is not talking about beliefs, but about facts—about time itself. So, the principle, if it is ever that case that P, then it is always the case that P, is stating a truth about the universe (the existent) that requires an ontological ground that always exists.6 McTaggart’s argument against the B-series and for the A-series (whether construed as terms having A-properties or standing in A-relations to a term outside the series) misinterprets the R-theory and assumes the existence of the A-series. McTaggart’s argument misinterprets Russell since in the R-theory, temporal relational facts are timeless in the sense that they do

McTaggart’s Paradox and R-theory of Time  133 not exist in time, i.e., they do not occupy moments (since there are none), they do not exemplify non-relational temporal properties (since there are none), and they do not stand in temporal relations (since relations of the first order can only have particulars as terms). In McTaggart’s analysis of the B-series, however, B-relations between terms are permanent—they always have been and always will be. In other words, B-facts and the terms of those facts always exist; they exist at every time. “If N is ever earlier than O and later than M, it will always be, and always has been, earlier than O and later than M” (1927, sect. 310, p. 12; emphasis added). This contradicts the notion of R-relations and R-facts, since if R-relations and R-facts do not exist in time then they cannot be permanent, and therefore cannot exist at every time. Why, then, in an argument against Russell would McTaggart assume that the B-series is permanent or always exists? It is at this point that McTaggart’s assumption of his definition of “earlier than” comes into play. If B-relations are analyzed in terms of different substances/events having incompatible A-properties at the same time (P is present while Q is future or P is past while Q is present, and so on), then the B-series would exist at every time. The B-series would be a permanent fact whose terms exists at every time with different temporal A-properties. Thus, without A-properties, the series of the terms of the B-series would not be successive but either a simultaneous, sempiternal block or timeless. Admittedly, in such an interpretation of the B-series, nothing would change by coming into and going out of existence or by acquiring and losing a property. However, in an argument against R-relations, that are primitive, unanalyzable and indefinable relations, it obviously misunderstands R-relations (or the B-series as a Russellian would understand it) and begs the question. McTaggart’s argument that the characteristics of pastness, presentness and futurity must be relations to a term outside the time-series and not qualities supports my reading of temporal parity as excluding co-existing yet successive times: Let us first examine the supposition that they are relations. In that case only one term of each relation can be an event or a moment. The other term must be something outside the time-series. For the relations of the A series are changing relations, and the relation of terms of the time-series to one another do not change. Two events are exactly in the same places in the time-series, relatively to one another, a million years before they take place, while each of them is taking place, and when they are a million years in the past. The same is true of the relation of moments to each other. Again, if the moments of time are to be distinguished as separate realities from the events which happen in them, the relation between an event and a moment is unvarying. Each event is in the same moment in the future, in the present, and in the past. (McTaggart, 1908, p. 29; emphasis added)

134  McTaggart’s Paradox and R-theory of Time Thus, for McTaggart, to say that a is earlier than b does not change means that they are in the same position in the time-series before they take place, after they take place, and when a and b are happening. However, if B-series facts exist at every time, then their constituents, a and b, always exist, and so are simultaneous or sempiternal, and not successive. Thus, NOW a is earlier than b, and it always has been the case and always will be the case that a is earlier than b. This is the ground of B-series facts always existing, but obviously, that cannot be an argument against the existence of the R-series without the A-series, unless it assumes an analysis of the B-series in terms of the A-series and confuses the R-series with the B-series. Interestingly, some of the textual evidence that Ingthorsson gives to support attributing the temporal parity principle to McTaggart implies, it seems to me, that past, present and future events do all co-exist, but simultaneously, and not successively. McTaggart says, “Now tomorrow’s weather is existent, for existence is as much a predicate of the future and past as of the present” (1921, sect. 6, p. 7, n 1). I think it is important to note that this quote implies not only that “Now tomorrows weather is existent,” but also that “Now yesterday’s weather is existent” and “Now, today’s weather is existent.” In other words, McTaggart’s assertion of temporal parity—that past, present and future exist equally or co-exist—does imply that all the tensed facts in a single A-series are now, hence simultaneous, and does not support that they exist successively. McTaggart is not claiming that a single event is past, present and future simultaneously, but that, say, Socrates is past, Oaklander is present and the 100th president of the U.S. is future—are all contents in a single A-series, and thus all exist NOW, at the same time or simultaneously. Thus, there is no ground for the different terms of a single A-series being successive even though they are equally real or co-existent. For McTaggart (1927), temporal parity does not allow for all times to be co-existent and still be successive. McTaggart’s notion of co-existence as applied to the terms of the A-series is simultaneity or timelessness, and therefore the temporal parity thesis is not something that McTaggart and R-theorists have in common. Thus, it is a mistake to claim that for the R-theorist “Socrates is in reality just as existent and real as we are now,” (Ingthorsson, 2016, p. 83) since for the R-theorist, Socrates is past and Oaklander is present are not facts that exist now because they do not exist in time at all. By treating McTaggart and B-theorists on a par in accepting the principle of temporal parity, Ingthorsson fails to see that there is a fundamental difference between McTaggart and R-theorists. For McTaggart, two objects/facts in an A-series cannot co-exist unless they are both NOW, and so exist at the same time. The R-theorist need not accept temporal parity in that sense. The R-theorist does not accept that Socrates is just as real and existent as we are now, and that there is a sense in which Socrates “still exists.”

McTaggart’s Paradox and R-theory of Time  135 It might be objected that if R-relations are atemporal, then that is one notion of “permanence,” and in this context to be permanent means to exist independently of any change. Thus, McTaggart could be making the point that the B-series does not involve change since the relations that generate it do not change. But it is a mistake to suppose that the basis of change must itself change; that the ground of temporality in the world must itself be temporal in the sense of existence in time.7

IV.  McTaggart’s Paradox To begin our discussion of McTaggart’s paradox,8 let me call your attention to an ambiguity, already alluded to, in the notion of the A-series. The A-series may be a single A1-series, or the A-series can be an A2-series of A1-series. Consider the following quote where McTaggart characterizes an A1-series: We must begin with the A series, rather than with past, present, and future, as separate terms. And we must say that a series is an A series when each of its terms has, to an entity X outside the series, one, and only one, of three indefinable relations, pastness, presentness, and futurity, which are such that all the terms which have the relation of presentness to X fall between all the terms which have the relation of pastness to X, on the one hand, and all the terms which have the relation of futurity to X, on the other hand. (1927, sect. 328, p. 20; emphasis added) A single A-series has terms that each co-exist, and Ingthorsson claims that co-existence is compatible with their existing in succession, but that is incorrect since each term in a single A-series is now either past, present or future. It is correct that a, b and c exist equally, but not that they exist in succession. There are two reasons why a single A1-series is not a genuine temporal series whose terms exist in succession. First, because there is no change in a single A-series because there is no term that has a property and then loses it. Second, although his definition of “earlier than” would suggest that a single A-series whose terms have different A-properties is a temporal series, that is not in fact the case; for if P is past while Q is present, then P and Q exist at the same time either simultaneously, if in B-time, presently, in A-time, or at t1, if time is absolute. Thus, there is no ground for P is earlier than Q in a single A-series, for to generate a temporal series that contains “real change,” there must be temporal passage and that requires a series of A-series—an A2-series of A1-series in which each different A1-series has terms with incompatible A-properties. For example, in one A1a-series, e is present and e’ is future, and in another A1b-series, e is past and e’ is present. Of course, an A2-series of such a

136  McTaggart’s Paradox and R-theory of Time conjunction of A1-series is not yet a temporal series for unless we introduce time in some way then the A2-series is contradictory. What, then, is to be done? McTaggart’s answer is startling: When we say that the B series is a series of changes, we do not, of course, mean that the terms change their places in the series. If one term is ever earlier than another, it is always earlier than that other. But the B series is a time-series, and time involves change. And the change in the terms of the B series is that they are successively present (passing from futurity to presentness, and from presentness to pastness). It is first an earlier term which is present, and then a later one. (1927, sect. 698, p. 347; emphasis added) What is key here is that the ground of the passage of time from the future to the present is the relation of succession. The A2 series of A1series is a temporal series because the generating relation is earlier than since “it is first an earlier term [in an A1a -series] which is present, and then a later term (in an A1b-series) that is present.” In other words, A-theoretic change or temporal passage presupposes succession.9 It is not sufficient for change that one and the same event/object/time have incompatible temporal properties, they must have them successively. However, the appeal to succession to ground the passage of time gives rise to either the unreality of time due to a vicious infinite regress if succession is analyzed A-theoretically in terms of McTaggart’s definition of “earlier than,” or the R-theory of time if succession is analyzed R-theoretically, since then R-relations are more fundamental than A-properties, and temporal passage and temporal becoming are grounded in R-relations. To put this point slightly differently, let’s go back to his definition: “The term P is earlier than the term Q, if it is ever past while Q is present, or present while Q is future” (1927, sect. 610, p. 271). Since neither disjunct alone could ground P is earlier than Q (even if each could entail it), the disjunction should be a conjunction. In that case, P is past while Q is present, and P is present while Q is future. If “while” is non-temporal, then a contradiction exists since P and Q would timelessly have incompatible properties. On the other hand, if “while” is temporal, then it is assuming time and so, we must ask, what more is needed to give us time and change? McTaggart answers that question, and his answer is quite revealing. To repeat: And the change in their terms of the B series is that they are successively present (passing from futurity to presentness, and from presentness to pastness). It is first an earlier term which is present, and then a later one. (1927, sect. 698, p. 347) In other words, the ground of change in the B-series is the transition from one present to another; the transition from one A1a-series to

McTaggart’s Paradox and R-theory of Time  137 another A1b -series is that the terms of the A2-series are successively present! However, if succession is a primitive R-relation and the basis of temporal passage or temporal transition, then McTaggart’s definition of “earlier than” in terms of A-properties is undermined. On the other hand, if succession from one present to another is analyzed in terms of his definition of “earlier than,” that would give rise to another third level A3-series whose terms are a series of A2-series, with incompatible A-properties, that is contradictory and static without introducing succession and thus leading to a vicious infinite regress. Thus, without R-relations and the R-series as the basis of passage, the result is a contradiction or a vicious infinite regress, but with R-relations, the A-series is not needed for time to pass. For that reason, the significance of McTaggart’s argument is not that time is unreal, but that time requires temporal passage which can only be grounded by appealing to a primitive R-relation of succession. McTaggart’s argument can also be applied to presentism. Suppose we consider his definition of “earlier than” and say that only one of the disjuncts, that is, only one A-series exists, the one that is now. Since it is the case that, say, P is past while Q is present, we can then maintain that “while” designates the present time. The seeming advantage of presentism in this case is that all tensed judgments, including those about the future and the past, for example “It will rain” and “It did snow” and “It is now sunny,” are all grounded in the present—the single A-series that is present. Moreover, given his definition of “earlier than,” a single A-series presumably also grounds the existence of that relation. Importantly, there is no contradiction since no single term of an A-series has incompatible A-properties. These are all advantages of a McTaggartian conception of presentism. Nevertheless, these apparent advantages for presentism come to naught since a single (present) A-series is not sufficient to constitute time. To have time there must be passage and, for that, presentism must account for continual becoming or absolute becoming. Passage involves a transition from the existence of one single A-series to the existence of another single A-series. In other words, one present (a single A-series) must go out of existence and another come into existence. However, transition is from existence to non-existence/existence (existence of an A1a -series to ceasing to exist of A1a and coming to exist of A1b), and that involves succession. In other words, a single A-series presupposes a series of A-series since, as Richard Gale has noted: It can easily be shown that if there is one A-series there must be a series of A-series. Assume that the A-series consist of events M, N and O, which are respectively past, present and future. A past (future) event by definition is one which was (will be) present. Thus, if there is one A-series there is becoming—a series of A-series; and if the A-series is objective then so too is becoming. (Gale, 1968, p. 190)

138  McTaggart’s Paradox and R-theory of Time Note, however, that McTaggart’s point is that a sequence of A-series does not constitute a temporal sequence or series unless the generating relation is a temporal relation. Thus, if P(Q) is ever past (present), then at an earlier time, P(Q) had to be present (future). In other words, a single A-series is not itself a temporal series since it does not involve change. What is needed to generate time is passage, but passage requires a succession of present A1-series or at least a succession of present times for different A1 series to become present. Thus, without a primitive notion of earlier than, a single A-series whose terms are past, present and future could not exist. Of course, with a primitive earlier than relation there would not be an A-series either since a primitive earlier than relation is nothing other than the R-relation which therefore must exist, and an R-relation being an external relation does not have terms with A-properties. Clearly, Ingthorsson would reject this McTaggartian understanding of presentism because he rejects A-properties (past and future events), and “denies that tensed passage occurs at all” (2016, p. 128). Nevertheless, I think the line of argument I have developed in interpreting McTaggart applies to his form of presentism as well. For although Ingthorsson rejects tensed passage, he undoubtedly accepts temporal passage since he believes that the ordinary conception of change is dependent on it. He says, Change is a difference in the properties of an object that remains numerically identical through the change, i.e., “genuine change”. This conception of change requires that things persist by enduring, i.e., that they come to exist at many times by passing as numerically identical three-dimensional bodies through time. . . . It is the view that ordinary material objects are three-dimensional things that move as numerical wholes through a succession of times. (2016, p. 94; emphasis added) Ingthorsson highlights the need for temporal passage in endurantism when he says, It is indeed some form of temporal passage that is supposed to allow three-dimensional objects to come to exist at many times and yet exist completely and only at each of those times., i.e. without having parts “lying around” at other times. (2016, p. 95; emphasis added) Ingthorsson makes it clear that temporal passage is “what allows threedimensional particulars to be at many times in succession” (2016, p. 138) and to be wholly present at those times. His form of presentism involves “permanent” material substances that do not come into being or go out

McTaggart’s Paradox and R-theory of Time  139 of being, but “ ‘remains’ continuously in the present” (2016, p. 138) as they pass through the flow of time and change. There are several problems with this analysis that leads us once again, it seems to me, to the R-theory. Ingthorsson maintains that just as an object can move from one place to another only if it passes as a whole from one place to another, an object can pass from one moment to another only if it is wholly contained in each moment through which it passes (2006, p. 99). The first point I want to make is that the analogy assumes the existence of times that exist before and after a substance arrives at them, just as spatial places must exist before and after an object leaves and arrives at them. The idea of motion assumes that the place we left from still exists and the place we are moving to already exists. So, if an object x leaves one time, arrives at another time and moves toward another time, this implies that more than the present time exists. Thus, the notion of temporal passage, as Ingthorsson unwittingly conceives of it, involves the co-existing of times that are not successive (if only the present exists), but are simultaneous; or, they are successive, but then it is not the case that only the present exists and is incompatible with his denial of crosstemporal relations. Thus, Ingthorsson is faced with a dilemma. If the passage of objects through time is grounded in the succession of times through which objects move, then succession is a cross-temporal relation, and everything cannot be grounded in the present. Moreover, it treats the time-series as a series through which objects move, leading to all the problems of the moving spotlight view and further undermining presentism. Similarly, if a substance moves from one time to another time, then those times must exist (or permanently remain in existence) for substances to move toward and away from. If moments permanently remain in existence, however, then they are simultaneous and sempiternal and so are not successive, unless one accepts McTaggart’s definition of “earlier than” with the subsequent vicious infinite regress to follow. On the other hand, if past times no longer exist and future times do not yet exist, then the notion of passing from one time to another makes no sense, since there are no times or a temporal series through which an object moves. If a substance comes to exist at a certain time, it is moving toward something that already exists “waiting” for the substance to arrive. A substance cannot move to a time and thus come to exist at that time unless there is something it is moving toward. The best it can do is come into existence, through absolutely becoming, at a time that also comes into existence. That would preserve presentism, but it would do away with temporal passage because the notion of moving successively from one time to another has no ground if there does not exist a temporal series, that is, if only the present time exists. Ingthorsson claims that if there is no passage, as in the B-theory, then everything remains permanently at their own temporal positions. But

140  McTaggart’s Paradox and R-theory of Time that implies that without passage the world is a totum simul. The result is that there are no temporal relations. However, McTaggart’s point is that temporal passage requires succession. Both the presentist and the R-theorist reject temporal passage as tensed property change. The Russellian rejects it because there are primitive temporal relations whose terms do not exemplify A-properties. Ingthorsson rejects tensed properties and R-relations (cross-temporal relations) because the past and the future do not exist, but puts nothing in their place. Therefore, succession has no ontological ground in his universe, and therefore, the positions his substances move through are not temporal positions or times, but spatial positions that are sempiternal or timeless positions. Certainly, they are not successive. Change for Ingthorsson is the objective loss and acquisition of properties by an enduring portion of matter. When something changes a qualitative state ceases to exist as another begins to exist, and never is there a cross-time relation between two qualitative states of the same substance existing at different times. Change is when “one state goes out of being while another begins to be” (pp. 135–36). However, if the time at which a qualitative state of a substance wholly present comes to exist is simultaneous with the substance’s arrival, and a qualitative state of that “same” substance ceases to exist at the same time as the substance’s arrival, then it is not the same substance that loses and acquires a qualitative state. If a time comes into existence absolutely, then everything at that time must also come into existence absolutely. There is no temporal passage from one time to another, just creation ex nihilo at that time. For if there is no passage, or no succession of times though which substances move, then not only the qualitative state, but the substance having that qualitative state, goes out of existence and the substance that comes into existence with a qualitative state is not the same as the substance that ceased to exist with a different qualitative state. Thus, there is not a single substance that changes from one present to another. Ingthorsson claims that, The alterations that take place are a question of qualitative states coming into and going out of being, and they are provoked by the influence exerted between different portions of matter. Here we find presentism thoroughly embedded in a metaphysics of material nature. And it is the most sparse ontology I know of. (2016, p. 138) Indeed, it is, or seems to be, a naturalist ontology where everything exists in space and time. This is reinforced by his claim that “The core idea of

McTaggart’s Paradox and R-theory of Time  141 presentism, as standardly formulated, is that the concretely existing present is meant to ground everything” (p. 135). How, then, can there be a succession of times that substances pass through? Surely, the succession of times cannot be grounded in a concretely existing present unless he accepts McTaggart’s definition of succession that we have seen, leads to a vicious infinite regress, that is, McTaggart’s paradox. For the R-theorist, on the other hand, in order to “ground everything” we must recognize not only the concrete particular existent, but the abstract non-spatial and non-temporal realm. Thus, the dispute between presentism and the R-theory is a debate between the naturalist and the ontologist.

V. Conclusion My understanding of McTaggart’s refutation of the A-theory, including presentism, also enables us to understand the overarching point of McTaggart’s main argument for the unreality of time. After arguing for the existence of A-change, he claims that the obvious reply to the attribution of incompatible A-properties/relations to events is that they have those properties successively. What, then, we must ask “does it mean to say that events or moments of time have A-properties successively?” His reply appeals to moments of time and A-properties (has been, will be and is now) as he later claims in his definition of “earlier than” in section 610. Obviously, that does not work, as he will subsequently show, since the passage of time involves different events/times successively becoming present. Thus, if succession is analyzed in terms of a single A1-series, or an A2-series of A1-series, or an A3-series of A2-series, the analysis either does not yield change or yields a contradiction that cannot be removed without appealing to succession and the subsequent infinite regress. McTaggart summarizes his argument in the following passage: The attribution of the characteristics past, present, and future to the terms of any series leads to a contradiction, unless it is specified that they have them successively. This means, as we have seen, that they have them in relation to terms specified as past, present, and future. These again, to avoid a like contradiction, must in turn be specified as past, present, and future. And, since this continues infinitely, the first set of terms never escapes from contradiction at all. (1927, sect. 332, p. 22; emphasis added) We see then that his argument in 1927 does not depend on the temporal parity as Ingthorsson understands it, or the a priori metaphysics of 1921, but in his analysis of succession that is not argued for, but assumed. In a footnote to the passage just quoted, McTaggart makes it clear that the vicious infinite regress arises from the attempt to avoid the contradiction in the attribution of incompatible A-characteristics to the terms in each A1-series by appealing to succession and then analyzing succession

142  McTaggart’s Paradox and R-theory of Time in terms of incompatible A-characteristics. This passage is important enough to quote at length: It may be worthwhile to point out that the vicious infinite does not arise from the impossibility of defining past, present, and future, without using the terms in their own definitions. On the contrary, we have admitted these terms to be indefinable. It arises from the fact that the nature of the terms involves a contradiction, and that the attempt to remove the contradiction involves the employment of the terms and the generation of a similar contradiction. (1927, sect. 332, p. 22, n 1) To conclude, I believe I have shown three things. First, that McTaggart’s argument against Russell begins by assuming his analysis of the “earlier than” relation as stated in 1927 section 610, and he uses that analysis in his main argument against the B-series. This has given rise to an A-theoretic misunderstanding that distorts the R-theory. Second, McTaggart’s analysis of “earlier than” undermines all versions of the property and the presentist versions of the A-theory, as his argument for the unreality of the A-series sought to demonstrate. Third, since McTaggart is clear that the passage of time and the possibility of change requires succession, I conclude that McTaggart has shown that given his analysis of “earlier than,” time and change are unreal. Therefore, to preserve the reality of time and change, we must reject McTaggart’s analysis of “earlier than,” recognize that it presupposes a primitive notion of succession and assert that the relation that generates change is the simple, unanalyzable R-relation of earlier than.10

Notes 1. Ingthorsson tends to include A-B hybrids, moving spotlight theorists and even growing blockers among those who accept temporal parity. 2. See also, Oaklander (2012, 2014a, 2014b, 2015, 2016, 2017, 2019). 3. For a good discussion of McTaggart’s misinterpretation of Russell’s view, see Tegtmeier (2012). 4. Not all B-theorists think of B-relations as static. See, for example, Savitt (2002), Deng (2013a), (2013b), and Mozersky (2015). For a critique of those who claim that the A-series, temporal passage and the dynamic aspect of time are illusions, see Boccardi and Perelda (2018). 5. Hope Sample has suggested to me that “while” could be understood in an atemporal sense of co-exist, but then co-existence would imply that each of the terms in each disjunct exists “eternally” and not successively, raising the specter of McTaggart’s paradox. 6. Although Ingthorsson would disagree with my literal interpretation of this passage (see 2016, p. 39), he does say “McTaggart consistently stays in object-language mode, assuming that he is talking about the world, and only ever takes a step back to talk about our talk of the world when he thinks

McTaggart’s Paradox and R-theory of Time  143 that his particular use of words invites the risk of misunderstanding” (2016, pp. 89–90), and that would support my interpretation. 7. For a fuller discussion of this point including a reply to the “No change,” see Chapter 8. For an account of emerging and passing away consistent with the R-theory, see Tegtmeier (1999). 8. For my earlier account of McTaggart’s paradox, see Oaklander (2002). 9. This point is also argued for in Boccardi (2018). 10. I wish to thank Emiliano Boccardi, Silvano Miracchi, Hope Sample and Erwin Tegtmeier for their helpful comments on earlier drafts of this chapter.

References Boccardi, E. (2018). Turning the tables on McTaggart. Philosophy, 3, 1–16. https://doi.org/10.1017/S0031819118000141 Boccardi, E. & Perelda, F. (2018). The delusive illusion of passage. Analysis, 78 (3), 387–396. http://doi.org/10.1093/analys/anx128 Broad, C. D. (1933). An examination of McTaggart’s philosophy (Vol. I). Cambridge: Cambridge University Press. ———. (1938). Examination of McTaggart’s philosophy: ostensible temporality (Vol. II (1), Chap. XXXV, pp. 265–323). Cambridge: Cambridge University Press. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 36–68). References are to this reprint. Deng, N. (2013a). Fine’s McTaggart, temporal passage, and the A versus B-debate. Ratio, 26 (1), 19–34. https://doi.org/10.1111/j.1467-9329.2012.00526.x ———. (2013b). Our experience of passage on the B-theory. Erkenntnis, 78 (4), 713–726. Gale, R. M. (1968). The language of time. London: Routledge & Kegan Paul. Grossmann, R. (1992). The existence of the world: an introduction to ontology. London: Routledge. Ingthorsson, R. D. (2016). McTaggart’s paradox. London: Routledge. Leininger, L. (2014). On Mellor and the future direction of time. Analysis Reviews, 74 (1), 148–157. doi:10.1093/analys/ant088 Le Poidevin, R. (1991). Change cause and contradiction: a defence of the tenseless theory of time. New York: St. Martin’s Press. ———. (2007). The images of time: an essay on temporal representation. Oxford: Oxford University Press. Lowe, E. J. (1987, Jan.). The indexical fallacy in McTaggart’s proof of the unreality of time. Mind, 96 (301), 62–70. https://doi.org/10.1093/mind/XCVI.381.62 Marsh, R. C. (2001). Bertrand Russell. Logic and knowledge: essays 1901-1950. Robert Marsh (Ed.), London: Allen & Unwin. McTaggart, J. M. E. (1908, Oct.). The unreality of time. Mind, 17 (68), 457–474. https://doi.org/10.1093/mind/XVII.4.457. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 21–35). References are to this reprint. ———. (1921). The nature of existence (Vol. I). Cambridge: Cambridge University Press. ———. (1927). The nature of existence (Vol. II, C. D. Broad, Ed.). Cambridge: Cambridge University Press.

144  McTaggart’s Paradox and R-theory of Time Mellor, D. H. (2009). Real time II. London: Routledge (Originally published in 1998). Mozersky, M. J. (2015). Time, language, and ontology: the world from the B-theoretic perspective. Oxford: Oxford University Press. Oaklander, L. N. (2002). McTaggart’s paradox defended. Metaphysica, 3 (1), 11–25. Reprinted in (2004). The ontology of time (pp. 51–62). Amherst, NY: Prometheus Books, and (2008). Vol. 1. The reality and language of time (pp. 108–120). ———. (2012). A-, B- and R-theories of time: a debate. In A. Bardon (Ed.), The future of the philosophy of time (pp. 1–24). London: Routledge. ———. (2014a). Dolev’s metaphysical anti-realism: a critique. In L. N. Oaklander (Ed.), Debates in the philosophy of time (pp. 1–29). New York: Bloomsbury. ———. (2014b). Temporal realism and the R-theory. In G. Bonino, J. Cumpa, & G. Jesson (Eds.), Defending realism: ontological and epistemological investigations (pp. 122–139). Boston: De Gruyter. ———. (2015). Temporal phenomena, ontology and the R-theory. Metaphysica, 16 (2), 253–269. doi:10.1515/mp-2015-0018 ———. (2016, Dec.). Common sense, ontology and time: a critique of Lynne Rudder Baker’s view of temporal reality. Manuscrito, 39 (4), 117–156. http:// dx.doi.org/10.1590/0100-6045.2016.v39n4.no ———. (2017). The best of times. The Philosophers’ Magazine, 79 (4th Quarter), 44–49. doi:10.5840/tpm201779106 ———. (2019). R. D. Ingthorsson: McTaggart’s paradox (Book Review). Metaphysica, 20 (2), 255–267. https://doi.org/10.1515/mp-2019-2015 Russell, B. (1911–1912). On the relation of universals and particulars. Proceedings of the Aristotelian Society, New Series. Reprinted in R. C. Marsh. (2001). (pp. 115–124). References are to this reprint. ———. (1914). Our knowledge of the external world. London: Allen & Unwin. ———. (1915, Apr.). On the experience of time. Monist, 25 (2), 212–233. https:// doi.org/10.5840/monist191525217. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 174–187). References are to this reprint. ———. (1918). Lectures on the philosophy of logical atomism. Monist, 29 (1), 32–63. https://doi.org/10.5840/monist191929120. Reprinted in R. C. Marsh. (2001). (pp. 176–281). References are to this reprint. ———. (1966). The monistic theory of truth. In Philosophical essays (pp. 131– 146). London: Allen & Unwin. Savitt, S. (2002). On absolute becoming and the myth of passage. Royal Institute of Philosophy Supplement, 50, 153–167. doi:10.1017/S1358246100010559. Reprinted in C. Callender (Ed.). Time, reality and experience. Cambridge: Cambridge University Press. Tegtmeier, E. (1999, Nov.). Parmenides’ problem of becoming and its solution. Logical Analysis and History of Philosophy, 2 (1), 51–65. https://doi. org/10.30965/26664275-00201006 ———. (2012). McTaggart’s error: temporal change. Revue Romaine de Philosophie, 56, 89–96. Urmson, J. O. (1960). Philosophical analysis. Oxford: Oxford University Press.

6 The Self and Time

The Self and TimeThe Self and Time

I. Introduction Broad’s philosophy of time is not only interesting and insightful in its own right, but it is also important for the connections he makes between issues in the philosophy of time and other important areas of philosophy. In this chapter and the next, I shall explore topics in Broad’s philosophy in which time plays a crucial role. First, I shall discuss Broad’s views in The Mind and Its Place in Nature (MPN), Chapter XIII “The Unity of the Mind” (1925), and relate them to his growing block theory of becoming, and his views on absolute space and time. In the next chapter, I shall connect his views on time to his views on precognition, memory and fatalism. In the course of his analysis of different views of the unity of the mind (or the “self” as he also refers to it), Broad makes use of another fundamental ontological distinction, “a real and irreducible one” (1925, p. 18), between existents and abstracta. He characterizes the difference as follows: Existents and Abstracta. The first great division within Reality as a whole which strikes one is the distinction between the part which exists and the part which is real but not existent. The contents of the latter I call “Abstracta”. (1925, p. 18) Entities that are real, but do not exist, are abstracta. The basis of the distinction is time: existents are in time, whereas abstracta do not exist, they subsist, and are timeless. Included in the realm of abstracta are numbers, propositions, qualities, relations and facts.1 Since all existents are in time and only particulars are in time, it follows that only particulars exist (1925, p. 19). What, then, is a particular existent? And what is a proposition or fact of which they are sometimes constituents? These are important questions, for facts play a crucial role in his account of the unity of the mind or self in this chapter, and propositions play a vital role in his account of memory and precognition in the next.

146  The Self and Time I shall argue, given Broad’s growing block view of time, propositions and facts, as entities that reside in the realm of abstracta, collapse into particular existents. This consequence, if true, supports the view, argued for in the previous chapter, that a main area of debate between A-theorists (including presentists and growing block theorists) on the one side, and R-theorists on the other, concerns whether temporal facts are eternal3 entities over and above their constituents that are timeless in the sense that they are outside of time. This debate is also relevant to the main topic of this chapter, namely, the unity of the mind. For unless facts are timeless abstracta, Broad’s preferred account of the self is doomed to failure. To begin to see why this is so, I shall turn to Broad’s account of alternative theories of the unity of the mind by first indicating the data that, he maintains, need to be accounted for in any adequate account of the mind. According to Broad, there are four facts about the mind that need to be provided an ontological ground. (1) A single state of mind at a moment may be a complex, distinguished qualitatively and objectively, or in both ways. Feeling tired and wanting tea would be an example of a state of mind that is at one time qualitatively distinguished. Thinking of my tea, of my book and of the multiplication table would be an example of a state of mind that has different objective constituents (1925, p. 556). What accounts for the unity of various simultaneous experiences of both kinds belonging to a single state of mind? (2) There are two different kinds of mental events: referential (or intentional), such as “I am seeing a chair” and non-referential (or non-intentional), such as “I am feeling tired” (1925, p. 556). In the one case, the mental event intends or is about an object other than itself, whereas in the other case, there is no distinction between the feeling and tiredness. (3) Two contemporary mental events can have the same determinate qualities and same object, but they cannot belong to the same mind/person. For example, two simultaneous mental events can both involve thinking of X and feeling Y, but they are still two and not one since they each belong to different minds. Thus, an adequate account of the self must be able to individuate them. What, then, accounts for their being two mental events and not one (1925, p. 557)? (4) Finally, “Certain series of successive total mental states are said to ‘belong to a single mind’” (1925, p. 557). So what distinguishes a series of successive total mental states that belong to one mind or self from those that belong to different minds? These are questions that an adequate account of the unity of the mind should be able to answer.

II.  Alternative Theories About the Unity of the Mind A.  Center Theories Although Broad discusses a variety of alternative theories about the unity of the mind, they all fall into two general categories: There are Center

The Self and Time  147 and Non-Center theories. In the most common view, the Center theory, the unity of the mind is grounded in a Center—the Pure Ego—that differs in kind from an event. The Pure Ego stands in a common asymmetrical relation to all mental events that belong to the same mind. In Non-Center theories, the unity of mind eliminates an existent Center and instead grounds a particular mind at a moment and over time, in terms of mental events being interrelated in characteristic ways. It is important to note, as Broad later does, that even Non-Center theories do not entirely do away with the assumption of a Center, but only with an existent Center. In Non-Center theories, there is a subsistent center that is neither an existent Pure Ego nor an existent central event, but a subsistent fact that certain successive events are suitably related to constitute oneself over time. I shall explain this point further after considering two versions of the Pure Ego Center theory. According to Broad, a Pure Ego is a particular existent that he characterizes as a “timeless particular” that differs solo numero from every other Pure Ego (see 1925, p. 279). In saying that one Pure Ego differs solo numero from another, Broad means that they are just numerically different; their difference is primitive and not based on a difference in qualities or relations to other particulars. Furthermore, since in Broad’s ontology “timeless” means outside of time, and all particulars exist in time, a “timeless particular” is, if taken literally, a contradiction in terms, unless he means “timeless1” or a particular of unending duration. Since he maintains that the mind comes into existence at a certain level of physical development, and he highlights the dependence of the mind on the body, Broad’s notion of “timeless” Pure Ego is not a particular with an unending duration, much less one that is timeless as abstracta are. Therefore, in calling a Pure Ego a “timeless particular” Broad must mean a particular or substance that exists as the same substance throughout all of its existence; it doesn’t change internally since it has no parts. In this view, the unity of a single mind is grounded in the simultaneous and successive mental events being “owned” at every moment of its existence by the same Pure Ego. A crucial distinction is the difference between a Pure Ego and an event. A Pure Ego “owns” various events, but it is not an event. Broad distinguishes between two versions of the Pure Ego Center theory depending on how they each account for ownership of a mental event. He explains two views of an event and two ways in which a Pure Ego “owns” events depending on the relation between a Pure Ego and an event. In the first version of the Center theory, mental events are facts about Pure Egos. An event is the fact that a Pure Ego is characterized by some determinate form of some determinable non-causal quality at a certain moment or has a determinate relation to some object or other at that time. For example, the event of my feeling tired and my thinking of going to bed is grounded in the fact that the Pure Ego I am has the quality

148  The Self and Time of tiredness and stands in the relation of thinking about going to bed. In this view, mental events are subsistent (timeless) facts about the qualities and relations of a Pure Ego, whereas the Pure Ego is a unique kind of particular existent common to each event belonging to the same mind. In this first version of the Center Theory, the unity of the mind thus founded on facts that have the same Pure Ego as a constituent and the “ownership” of events by the same mind is grounded in a “formal” relation between the Pure Ego and those events, construed as subsistent facts, of which it is a constituent. The unity of simultaneous mental events belonging to the same mind is based on a single Pure Ego that is a common constituent of all those events. The unity of the mind over time is founded on the same Pure Ego being a constituent of successive mental events that are ordinarily said to belong to the same self. One question this analysis raises is how can an event that is a subsistent fact occur before or after another event given that both are abstracta and do not exist in time? Clearly, the first version of the Pure Ego view depends on the notion of fact as an entity that subsists. An event is a substantive, that is, it can only be the logical subject of a proposition or a fact, but there are two different kinds of substantives, namely, those that exist in time and those that subsist timelessly. In this version of the Pure Ego theory, events are subsistent substantives whereas the Pure Ego is an existent substantive. Thus, a Pure Ego and an event are of different ontological categories: a Pure Ego is a particular existent and therefore in time, whereas an event is a subsistent abstractum; a timeless fact that has a Pure Ego as a constituent. The second version of the Pure Ego theory construes mental events as existent substantives in their own right, and not merely subsistent substantives or facts about the qualities and relations of Pure Egos (1925, p. 562). In this view, an event is a particular existent characterized by qualities and stands in relations so it is the logical subject of the facts in which it occurs. “Ownership” is not a “formal” relation of a Pure Ego to the fact of which it is a constituent, as in the first version, but a peculiar asymmetrical “material” relation between a Pure Ego and an event. Thus, in the second version of the Center theory, both the Pure Ego and the event it owns are existent substantives, but they are different kinds of existent substantives. In the first Center theory, an event is a subsistent fact. Whereas in the second, an event is a particular existent that has qualities and relations, and an event having qualities and standing in relations to other events is a fact. Broad summarizes these two versions of the Pure Ego theory in the following passage: On the first theory, the Pure Ego is characterised directly by tiredness; on the second theory, the Pure Ego has to the quality of tiredness a compound relation which is the logical product of the two

The Self and Time  149 relations of “owning” and “being characterised by”. For, on the second theory, the Pure Ego owns something which is characterised by tiredness. The difference is that, on the first theory, the relation between the Pure Ego and the quality is direct, like that of father to son; whilst, on the second theory, it is indirect, like that of uncle to nephew. Again, on the first theory, the relation of Pure Ego to mental event is the formal relation of a subject to a fact about that subject; whilst, on the second theory, it is the non-formal relation of “ownership” between one existent substantive of a certain kind and another existent substantive of a different kind. (1925, p. 563; emphasis added) In the second theory, events and Pure Ego are “different kinds” of existent substantives, but how are they different? What is the ground of an event being a particular existent of a different kind from a Pure Ego? An event is not a substance since then it would not be the instantaneous particular that Broad claims it to be (1925, p. 563). Nor is an event a bare particular—a particular that differs solo numero from all other bare particulars—for then it would not be a substantive existent of a different kind from a Pure Ego. My suggestion for the ontological status of particular existents that are mental events is that events are clusters of tropes (quality instances). I shall argue for my interpretation after considering the second great distinction regarding the theories of the unity of the mind, namely, the Non-Center theory. In another version of the Center theory, there is no assumption of a Pure Ego. Broad considers an event such as a mass of bodily feelings to be the center of which other events are related. It may be related to other mental events in the same two ways in which events are related to a Pure Ego. There are facts that a single central event is a bodily feeling that has qualities and stands in relations to various objects. Or, the single central event is a bodily feeling that owns various other events. In Oaklander (2013), I argue that one problem with identifying bodily feelings as the Central Event and thus the basis of the unity of the mind is that they cannot account for the difference between two different kinds of mental events: referential (or intentional) events, such as “I am seeing a chair” and non-referential (or non-intentional), such as “I am feeling tired.” The same problem arises if one construes mental events as bodily feelings being characterized by qualities and standing in relations as one may in a Non-Central theory of the self. B.  Non-Central Theories Non-Center theories of what constitutes a single mind dispense with an existent center, whether a pure ego or a mass of bodily feelings, and attempt to account for the unity of the mind at a moment and over time,

150  The Self and Time by means of different characteristic relations between the events that constitute the mind at a moment and over time. Broad considers and rejects one objection to the non-central theory based on a logical point. The objection is that our use of personal pronouns like “I,” “You” etc. presupposes that we recognize the existence of centers; and that Non-Central Theories are necessarily incapable of accounting for this fact. (1925, p. 570) Broad rejects that argument; he explains that if there is a relation between the members of a series then there is a further entity, a subsistent Center, that is the referent of the “I” to which all events have an asymmetrical relation. Thus, even if ordinary language suggests a certain ontological analysis, it does not follow that the analysis is implied by ordinary language. There may be a different analysis that is compatible with the linguistic fact in question. As Broad explains it, If a number of terms be interrelated directly in a characteristic way it follows analytically that there is something to which they all stand in a common asymmetrical relation, even though there be no Existent Centre in the system. For each of them is a constituent in the fact that they are all related to each other in this particular way; and so, this fact stands in a common asymmetrical relation to all these terms. Thus, even if a number of interrelated terms have no Existent Centre, there is always a certain substantive, which subsists though it does not exist, which stands in a common asymmetrical relation to all of them and might be called their “Subsistent Centre’. (1925, p. 570; emphasis added) In other words, even non-Center theories can account for our use of personal pronouns and what we commonsensically say or think about the self without there being a particular existent that is the Center. For our purposes, the crucial point to highlight is that for Broad, the difference between the Center and Non-Center theories concerns the ontological status of the “center.” In the unity of Center theories, the center is an existent substantive, that is, it exists in time, whereas in the Non-Center theories, there is a “center,” but it is a timeless subsistent fact, and thus not a temporal particular existent. Broad summarizes his response to the objection based on our ordinary language use of personal pronouns as follows: We see that this objection is baseless. Even on Non-Central Theories there is necessarily something which can be called “I” or “You”. This

The Self and Time  151 something is a substantive, and it stands in a common asymmetrical relation to “my” state or to “your” states respectively. The only difference between Central and Non-Central Theories is about the logical nature of this substantive. On Central Theories it is a particular existent, either a Pure Ego or a Central Event. On Non-Central Theories this substantive is a Fact about certain mental events and their interrelations, and so its mode of being is subsistence and not existence. (1925, p. 570)2 Thus, in both the Center and the Non-Center theories, Broad’s ontology is committed to facts as real but non-existent (subsistent) entities or abstracta. In the first version of the Pure Ego theory, an event is a fact about the qualities or relations of a Pure Ego. In the second version, a fact is an event; a particular of a different kind from a Pure Ego that is characterized by qualities and relations, and is owned by a Pure Ego. In the Non-Center theory, a single mind is a fact; the interrelation of simultaneous and successive mental events. Thus, in all versions of the Center and Non-Center theories of the unity of the mind, the self depends on the bifurcation of reality into temporal particular existents and subsistent, timeless abstracta. C. A Critique of Broad’s Analysis of Center and Non-Center Theories In MPN, Broad’s commitment to facts as timeless subsistents that have particulars existents as constituents is incompatible with the growing block theory of time also maintained in the work. For given his view of time, Broad cannot provide a ground for facts as real, but subsistent abstracta, since propositions and facts collapse into particular existents. That is, he cannot ground the difference between those entities that are dependent on time and those whose reality is independent of it. If, however, the distinction between existents and abstracta collapses, then so does his view of the unity of the mind, regardless of which view he adopts. For if there are no facts, we cannot make any sense of either the Center or the Non-Center theories of the unity of mind since they both, in whatever form they take, presuppose the subsistence of facts as abstracta. Recall that for Broad, propositions and facts are abstracta that are not in time, and for that reason, neither come into nor cease to exist, nor can they change their constituents. He maintains that propositions and facts are single, complex entities in that they have constituents and yet they are one.3 There is textual evidence to support that Broad believes that propositions and facts contain existents. All existents, he says, are logical subjects, which, as he speaks, means that existents can occur only in

152  The Self and Time the subject place of a proposition. In this regard, the following passage is relevant: Whatever exists can occur in a proposition only as a logical subject. Of course the name of an existent may appear in a sentence as a grammatical object and in other positions too. E.g., in the sentence “Smith dislikes Jones” the only grammatical subject is the word “Smith”, and the word “Jones” counts as a grammatical object. Nevertheless, the men Smith and Jones are both logical subjects of the propositions for which this sentence stands. (1925, pp. 18–19) Thus, not only is Broad’s ontology committed to the category of facts, he sometimes also speaks of facts and propositions as containing existents. The presence of existents in propositions and facts will prove troublesome for Broad’s various accounts of the unity of the mind. Another especially revealing passage is also pertinent to my claim that Broadian propositions and facts contain existents is the following: But there are two different kinds of substantives, viz., those which exist and those which only subsist. A Pure Ego, if there be such a thing, is an existent substantive. A fact or a proposition is a substantive, in the sense defined above. We can say that “The execution of Charles I was a political mistake” or that “It is probable that Edwin will marry Angelina”. Here we have facts or propositions functioning as subjects of other propositions. And they cannot play any other part in a proposition. They are therefore substantives. But they do not exist (though they may contain existents as constituents); they merely subsist. (1925, pp. 558–559) Recall that for Broad, propositions and facts are abstracta that are not in time, and for that reason neither come into existence nor cease to exist nor change their constituents. However, if facts may contain existents, then, given his views on time, they become existents. To see why this is so, consider the fact that “This is green” where the word “this” stands for a particular existent. Since in Broad’s view, the future does not exist, existents are at some times but not at others. How, then, can an existent be a constituent of a fact that is supposed to be a timeless and changeless abstractum? How can a fact contain what “this” stands for when what it stands for is at sometimes literally nothing? In other words, the fact that “this is green” will first lack an existent and then forever contain one. But this is against the very notion of a proposition or fact. For, if a fact must change in this way then it is, in an obvious sense, in time, and therefore, not an abstractum but itself an existent.

The Self and Time  153 The same point may be made in a slightly different way. Broad assumes that existence involves time since in the growing block theory, the sum total of existence increases with the coming into existence of new slices of (particular) existents. Now, if a fact may literally contain existents, then a fact cannot be without reference to the existents it contains. That is, a fact cannot be until the time when the existent it contains comes into existence. But then, such a fact is no longer an abstractum, that is, the timeless entity that Broad intends it to be, but rather, itself is an existent. Furthermore, since in Broad’s view, all existents are particulars, it follow that such “facts” themselves are, or, as I put it earlier, they collapse into particular existents. This consequence has dire implications for his analyses of the self, so the situation is worse than anticipated. The subsistence of facts as a different kind of entity from particular existents is necessary to distinguish different versions of the Pure Ego Center theories from each other, and the Center theory from the Non-Center theory. Both Center and Non-Center theories, according to Broad, have a “center,” but in the Non-Center theory, the center is a subsistent fact, whereas in the Center theory, the center is a particular existent. Since, however, subsistent facts contain particular existents, they cannot become until the existents they contain come into existence. For that reason, they cannot be timeless, subsistent facts as they must be if any of Broad’s accounts of the unity of the mind is to be possible. We may conclude that Broad must either reject his growing block theory of time or deny that a fact may contain particular existents. Clearly, Broad faces an unacceptable dilemma. Broad cannot consistently treat a fact as a cluster of abstracta, since subsistent facts that contain particular existents are employed in his distinction between Center and Non-Center theories of the unity of the mind. Therefore, the only way his ontology could retain facts as timeless entities would be to either reject his growing block theory of time or treat propositions and facts as clusters of abstracta. Unfortunately, as we shall see shortly, Broad’s views on time during the middle period are also incompatible with the reality of propositions as clusters of abstracta. I shall return to this problem in the next section. Putting these problems aside, I shall next turn to Broad’s discussion of the pros and cons of the Center and Non-Center theories of the self.

III. Discussion of Alternative Theories of the Unity of the Mind Broad claims that the most important issue to consider in making a tentative decision about the unity of the mind concerns the nature of mental events, which he states as follows: Can we take the notion of ‘mental event’ as fundamental, and define the notion of ‘mental substance’ in terms of mental events and certain

154  The Self and Time relations between them? Or must we conceive a ‘mental event’ as consisting in the fact that a certain Centre has at a certain time such and such a determinate quality, or such and such a determinate relation to other things? (1925, p. 587) He approaches this question by considering the analogous question: Can we define the notion of “material substance” in terms of “material events”? Are they suitably related or must we define a mental event as consisting in the fact that a substance has a quality or stands in a relation at a certain time? If the former, then perhaps the same can be done for mental substance in which case a Non-Center theory is possible. If the latter, then the analogous view of mental events would be tantamount to the Pure Ego Center theory of the Self. With that in mind, Broad then considers if the reduction is possible. In the first purported reduction of material substance to material events, Broad argues that the apparent success of the reduction is “due to the fact that most of us tacitly assume something like the Newtonian theory of Absolute Space” (1925, p. 588). For that reason, we have not really gotten rid of substance as a fundamental notion. The analogous view of mental substance in terms of mental events would be the first version of the Pure Ego Center theory, according to which a mental event is a Pure Ego characterized by distinct qualities and standing in relations. Thus, the proposed reduction of mental substance to mental events would fail. To explain how the proposed reduction would work and why it fails, suppose we begin by stating the two most fundamental features of the Newtonian view as Broad understands it. The first is that space and time are logically prior to matter or things and events. That is, space could exist, even if there were nothing anywhere in it, and time could be, even if there had never been any events. Or, as it has been put, after Newton, space and time are the containers or receptacles which things occupy and in which events happen. Now there are two possible relations between space and time and a “thing.” A thing may be a substance or a particular event that occupies the distinct entities which are space-time regions. Or a “thing” may be a space-time region which is pervaded or characterized by qualities. Upon the latter notion, but not the former, the essence of the “container view” is that if all the nonspatial and nontemporal qualities pervading a space-time region where taken away, there would still be the space-time region. Broad uses this notion of a thing to define a mental event. For, If we think of Space as a kind of pre-existing substance, we can of course think of a material event as the fact that a certain region of Space is characterised throughout at a certain moment by a certain determinate form determinate form of a certain terminable quality (e.g., by a certain shade of a certain colour). (1925, p. 588)

The Self and Time  155 The second fundamental feature of the Newtonian view is that there is more to space and time than spatial and temporal relations. This feature follows from the first: If spatio-temporal relations were the only category of entities in space and time, then they would obtain not between places and moments, but rather between and among things and events. Therefore, if we took away things and events there would not be space or time. However, according to the container view, there may be space and time without any things and events. Thus, if we assume the container view then spatio-temporal relations cannot, except derivatively, obtain between and among things and events, but must directly obtain between and among places and moments. Hence, in the Newtonian view, places and moments have ontological status. To say that places and moments have ontological status is one thing, but to say what kind of entities they are is quite another. Newton did not need to, and in fact did not, specify what ontological category places and moments belonged to. As a scientist, he did not need to concern himself with whether they are substances or qualities. Scientists may legitimately worry about whether there are places and moments, and, if so, whether they could exist without things and events, but they do not worry about the ontological category to which something belongs. Broad, however, interprets the Newtonian view to commit scientists to holding that space and time are substances, and for that reason maintains that material substances cannot be analyzed or derived from material events. More specifically, Broad objects to the first analysis of material substance in terms of material events because the notion of a material event is a derivative of the notion of substantival absolute space. He states this criticism in the following passage: For what is a material event, on this theory, but the fact that such and such a region of Space is characterised throughout by such and such determinate forms of such and such determinable qualities? And what is a region of Space, on this theory, but a timeless particular in which sometimes one quality, sometimes several qualities, and sometime perhaps no qualities inhere? And what is a plurality of different regions of Space, in terms of which the plurality of coexisting material substances is defined on this theory, but a plurality of timeless particulars which differ solo numero? It is plain that no form of Non-Central theory about mental events and mental substances could be at all closely analogous to the above theory about material events and substances. For the theory just described is essentially a peculiar form of Central Theory. (1925, p. 590) Broad explains why this is so and he is correct. The analogous view of mental events would be a version of the first form of the Pure Ego theory.4 Since material events are analyzed in terms of facts about a substance—a

156  The Self and Time region of absolute space—having qualities and standing in relations at a moment of absolute time, the analogous view of mental events would be analyzed in terms of facts about a substance—a Pure Ego—having qualities and standing in relations at a moment. For that reason, the analysis of mental events would not do away with mental substance. Broad argues, however, that it is possible to take the notion of material event as fundamental and construct the notion of material substance out of it without assuming absolute space in Newton’s sense. In this view, places and times are not temporal or spatial individuals, but determinate forms of some “determinable Positional Qualities, viz., Temporal and Spatial Position” like “ ‘being in such and such a place’ or ‘being at such and such a date’” (1925, p. 592). Broad employs these positional qualities in his analysis of events in terms with which he seeks to eliminate the notion of material substance. Here is the connection between Broad’s notion of material substance, including his criterion of identity through time and his version of “absolute” space and time. The sizes, shapes and so on are of and between particulars. Yet spatio-temporal qualities of particulars depend on or are founded upon the spatial and temporal positional qualities that they possess. Specifically, there are two second-order or determinable positional qualities—“being a time” and “being a place”—and falling under these determinables, a continuous one-dimensional order of determinate temporal positions, (. . . T1, T2, Ti . . .) on the one hand, as well as a continuous three-dimensional order of determinate spatial positions ( . . . P1, P2, Pj . . .) on the other5 (1925, p. 594). A set of particular existents whose temporal qualities vary continuously from one moment to another have a certain duration that depends on the relations between the temporal qualities that characterize the first and the last point-instance of the set. Similarly, the shape of a whole composed of particular existents with the same temporal position, whose spatial qualities vary continuously, depends on the relation between the spatial qualities that characterize the existents that form the boundary of the set. And finally, the various temporal, spatial and spatio-temporal relations that characterize particulars are grounded on the determinate spatial and temporal positional qualities that characterize them (1925, pp. 595–596). An event is an instantaneous particular that has positional and nonpositional qualities. Broad summarizes this analysis in a passage I shall quote at length: We have assumed that there are spatial and temporal positional qualities, and that spatial and temporal relations depend on them. Thus, our theory of Space and Time is absolute, in the sense that it is not purely relational. It is not absolute, in the sense that it makes the points of Space and the moments of Time to be existent substantives of a peculiar kind, as Newton’s theory does. The only existent

The Self and Time  157 substantives which we assume are instantaneous punctiform particulars, which have determinate forms of Spatial and Temporal Position and determinate forms of determinable Non-positional Qualities. Certain sets of these form wholes which have qualities of shape, size, and duration in terms of the relations between their positional qualities. (1925, pp. 597–598) In this view, to say that a thing changes means that its history can be cut up into a series of adjacent thin slices or events (mental states), and that two adjacent slices have characteristic qualitative agreements and disagreements and characteristic interrelations, including causal connections. Broad states the need for causality in this account of a persistent material substance in the following passage: No doubt everyone would admit that something more than this is needed to complete the notion of persistent material substances. But it might be suggested that the “something more” is merely a causal unity between those successive events which are counted as successive total states of the same material substance. This causal unity would consist in the fact that the variations in the determinate forms of these determinable qualities which characterise successive total states of a single material substance follow certain laws. (1923, p. 589) And, of course, the spatio-temporal position of each slice is grounded in the spatio-temporal positional qualities that characterize it.6 Thus, material substances can be analyzed in terms of material events or particular existents that are of a different kind from material substances. Similarly, Broad suggests that a mental substance or the Pure Ego can be analyzed in terms of suitably related mental events. Broad raises a problem with this view that clarifies what he takes an event to be and how it differs from the Pure Ego view of the mind he wishes to avoid. Since mental events are not extended, they do not have a spatial position, and for that reason, the first hurdle this Non-Center theory faces is the problem of individuating two mental events occurring at the same time with the same qualities or relations to the same object. He states the problem in the following passage. It is logically possible for there to be two contemporary mental events which have the same determinate form of the same Mental Quality (e.g., it is logically possible that there might be two precisely similar contemporary thoughts of the same object, even if there is reason to think that this is causally improbable or impossible). Now there seem to be only two alternative ways of explaining this fact. The

158  The Self and Time two precisely similar thoughts must either belong to different Pure Egos, or there must be some non-spatio-temporal Positional Quality of which they possess different determinate forms. (1925, p. 599) Since Broad is attempting to provide an analysis of the mind that answers the problem of individuation without introducing a Pure Ego, his response to this problem is to introduce a new and shadowy positional quality, mental position. He continues: We must suppose that every mental event is an instantaneous particular which has a certain determinate Temporal Position; and two mental events may agree in every other respect, provided that they differ in Mental Position, but they must have different determinate forms of one or other of these Positional Qualities. (1925, p. 599). With this assumption it would, I think, be possible to take the notion of “a mind” as definable. A total state of mind would be an instantaneous particular existent, which (a) has a determinate quality of Temporal Position, (b) has a determinate quality of Mental Position, and (c) is an instance of several different Mental Qualities. (1925, p. 597; emphasis added) A mental event is an instantaneous particular existent, but what is that? An event is not a universal or a bundle of universals. Moreover, it is not a bare particular, for if it were, Broad would not need to introduce mental positional qualities to individuate two mental events with the same qualities and relations. To put this point differently, if the particular existent “in” a mental event is a bare particular, then the question of individuating two contemporary thoughts about the same object can be answered by saying that they each belong to a numerically diverse particular, or different Pure Egos. Yet Broad does not take this way out, for he is attempting to see if a Non-Center view of the Self is possible. Thus he maintains that regarding mental events, in addition to temporal positions there are mental positions, and the latter individuate two minds thinking about the same thing at the same time. To put the point otherwise, if a peculiar, particular existent were a bare particular, a mere individuator and support of qualities, then mental positional qualities would be wholly unnecessary for individuation. Since Broad stresses the need for mental positions to solve the individuation problem it follows that his particulars are not bare, but quality instances or tropes such as “this pain,” “this mental position,” “this time” and so on. I conclude that in Broad’s ontology, particular existents or events are clusters of quality instances or tropes; that is, simple particulars which are supposed to be numerically different from, but qualitatively similar to, all other particulars of the same kind.

The Self and Time  159 Is there further evidence that Broad construes a particular (event) as a cluster of tropes? I submit that there is. In the first place, it allows one to reconcile certain otherwise incompatible statements about the status of qualities. On the one hand, he speaks of qualities as universals that can be, even though they are neither thought of by anyone nor characterize an existent. Of this line the following passage is typical: The quality redness is timeless; nevertheless it sometimes characterise one thing, sometimes another, and sometimes perhaps nothing at all. Again, it is something thought of by me, sometimes by you, and sometimes perhaps by no one at all. (1925, p. 257) On the other hand, he insists that the appearance or occurrence of sensible qualities is dependent on particular existents. That is, sensible qualities are dependent on the arrangements of particles moving in certain ways; if certain existent particles are not there, certain sensible qualities will not be there either. Therefore, in order to avoid the contradiction that qualities are both dependent and independent of existents, Broad must be equivocating on the term “quality.” The only way to make him consistent is to attribute to him the view that, while “dependent” sense qualities are particulars, “independent” qualities are abstract or Platonic universals. And, of course, as for non-positional qualities, so for positional determinates. Assuming that Broad construes sensible qualities to be particulars, one might expect him to hold that we always perceive particulars and never universals. In an article written five years before MPN, he in fact articulates this position. Consider the following passages: It is clear that every immediate object of our senses both exists and is real in the primary meaning of these terms. . . . For it seems to be a synthetic a priori proposition that anything of which we can be directly aware by our senses is both real and particular; and what is real and particular exists in the primary meaning of that word. . . . If we are directly aware of a universal, it is the object of thought, and it is clearly something real in the same sense in which a particular flash of light is real when it is the object of our senses. It does not, however, exist in the primary sense, because it is not a particular. (1919, p. 558) In MPN, too, Broad says that when sensing a sensum, e.g., a red patch, one is in a unique way related to a particular existent. Again, in the same book, in a different context: When I try to introspect the sensing of a noise or the feeling of a pang of toothache the only particular existents which are intuitively

160  The Self and Time presented to me are the noise of the toothache and certain bodily feelings. (1925, p. 309) Since a noise, a toothache or a red patch obviously have some qualitative aspects, and since these entities are particulars, it follows that particulars are quality instances. I believe, therefore, that there is ample evidence for my interpretation of events qua particular existents as clusters of quality instances or tropes, and that propositions that are about them contain the abstracta that characterize them. Thus, consider the proposition that “This is green.” What is the referent of “this”? When a proposition is about a material existent, the word “this” refers to the dyadic cluster of a spatial and a temporal positional quality, e.g., P1 and T1, at a minimum. Thus, the proposition that “This is green” contains, at a minimum, the abstracta (P1,T1,G). The following passage supports this interpretation: We take the fundamental constituents of the material world to be instantaneous punctiform particulars, each of which has a determinate quality of Temporal Position, a determinate quality of Spatial Position, and determinate forms of one or more Non-Positional Qualities. (1925, pp. 593–594) Since every material particular existent is characterized by some Pj, Ti and at least one non-positional quality, Fk, one cannot but conclude that in this world, a proposition that is about a material particular existent, without of course containing it, does contain the cluster of the abstracta, positional and otherwise, that characterize it. For, if the proposition did not contain those abstracta, how could it possibly be about that existent? Thus, we shall assume that a particular existent is a cluster of tropes, and that the temporal determination in two simultaneous events (particulars) is numerically different although qualitatively the same. When representing the abstract temporal positional qualities as . . . T1, T2, Ti, . . . one might represent instances of, say, Ti as ti1, ti2, . . . tij . . . Similarly, if two events “occupy” the “same” place at different times, then each will contain numerically diverse—though qualitatively the same—spatial determinations, say pj1, pj2, . . . pji. Thus, we would represent two simultaneous events, say (p11, t11, f11), (p21, t12, f21); and two events occurring a different times at the same place, say (p11, t11, f11), (p12, t21, f21) with F1 and F2 being two non-positional abstracta under which f11 and f21 fall, or synonymously, in which they participate. Broad argues that the view of mental unity that takes events as fundamental and avoids the existence of a Pure Ego is slightly preferable because it enables one to explain abnormal phenomena. He rejects the

The Self and Time  161 arguments from experience against the existence of the Pure Ego theory or the central event theory for reasons that we need not pursue here. Interestingly, while he considers and rejects arguments from experience against the existence of a Pure Ego, and also from those who claim we are not acquainted with the relation between a Pure Ego and its object, he is not concerned with arguments from acquaintance with positional qualities. In the Appendix I shall provide an independent discussion of the ontology of absolute and relative space and time, and their connection with the existence of bare particulars and the topic of acquaintance. What is worth noting, however, is that given his commitment in 1925 to the non-relational qualities of pastness and presentness; and his view of propositions as clusters of abstracta, and events as clusters of quality instances or tropes; propositions collapse into particular existents, and events give rise to an infinite regress of time dimensions and McTaggart’s paradox. The argument is as follows: Pastness and presentness are either universals or particulars. Assume that they are universals and consider the following two propositions about one and the same mental event. “This is a painful feeling” and “This was a painful feeling.” The assays will yield the following abstracta: clusters (Mj, Ti, Pk, Pr) and (Mj, Ti, Pk, Pt). Unfortunately, though, the two propositions are incompatible; the only way to make them compatible is to specify the times at which the one and the other are true. Before the single event which allegedly they are both about has become, they are both false. At the moment the event they are about comes into existence, the one containing Pr becomes true; the other containing Pt, false; and at the next moment and from there on, the latter becomes true while the former from then on remains false. But this apparent way out implies that either of these propositions, true at one time and false at another, does not have a “timeless” true value. Or, to say the same thing differently, if pastness and presentness are universals, then propositions are in time, and therefore, upon Broad’s criterion of existence, are not abstracta but existents. Furthermore, if there are two universals Pt and Pr, then there are in this world also the corresponding tropes (and conversely). From this an infinite regress of time-series ensues as follows. The event, a painful feeling, contains a particular presentness and a particular pastness. This, though, is intelligible only if it first contains the presentness and then the pastness. That is the event which is present must exist earlier than the event which is past. But, given that for Broad, temporal relations depend on temporal properties, since he says, Point instants are thus ordered in various ways, and stand in various temporal spatial and spatiotemporal relations to each other in virtue of the determinate qualities of Temporal and Spatial Position which characterise each point-event. (1925, p. 594)

162  The Self and Time there must be a second level or series of moments (construed as temporal positional qualities) at which the event in question becomes present and then becomes past. For in the first level of time, the same event is present and past at the same time, as the analysis shows. But with becoming and time being essential to each other, members of the series of temporal positions at which events become first present and then past requires a second series, or, as I put it, a second time dimension of temporal positional qualities, and so on. How then do propositions and facts fare in an ontology of time such a Broad’s? In this chapter, I have argued that they are incompatible with his growing block view of time. For Broad, propositions and facts are abstracta—they do not exist in time; they are subsistent, but given his view of time, they collapse into particular existents that are subject to McTaggart’s argument. This is important for it shows that neither presentism, as Ingthorsson conceives of it, nor the growing block view, as Broad understands it, can countenance facts that contain particulars (or temporal relations between particulars) as a category in their ontology. In the R-theory, facts contain particulars that are temporally related and are thereby temporal, timeless and eternal3. Such R-facts are, I believe, necessary for the reality of time, but are incompatible with Broad’s ontology during the middle period.

Notes 1. Broad says of abstracta, “Some minds, and especially Dr Whitehead’s, seem to spend a good deal of their time in contemplating, reasoning about, and feeling approval or disapproval towards objects which are, on the face of them, neither material nor mental, e.g., numbers, propositions, and the formal relations of such objects among themselves. And it is certainly arguable that a mind could go little if any distance in cognizing objects which are physical or mental if it did not have the power of cognizing objects which are neither” (1925, p. 5). 2. Here, the notion of the “logical nature of this substantive” means the ontological nature. 3. See (1923, p. 252) where Broad says, “An act [my perceiving] is something which cannot exist by itself but can only exist as a constituent in a complex, whose other constituent is its object. Broad calls this whole complex situation “(my perceiving)—of—(table)” a fact. Clearly, therefore, facts are essential to his ontology. 4. But it has some peculiar consequences, e.g., the same center (region of Absolute Space) at different times can be a center of states that belong to different minds. 5. I ignore, as one safely may, the three dimensionality by using only one superscript. 6. He says that duration is an internal relation (1925, p. 598) and also that in this adjectival view of Absolute Space and Time (so-called after W. E. Johnson’s (1922) terminology), spatial and temporal relations depend on positional qualities (1925, p. 599), that is, they are internal relations.

The Self and Time  163

References Broad, C. D. (1919). Reality. In J. Hastings et al. (Eds.), Encyclopedia of religion and ethics (Vol. 10, pp. 587–592). New York: Scribner’s. ———. (1925). Mind and its place in nature. London: Routledge & Kegan Paul. ———. (1952). Scientific thought. London: Routledge & Kegan Paul (Originally published in 1923). Johnson, W. E. (1922). Logic (Vol. II). Cambridge: Cambridge University Press. Oaklander, L. N. (2013). C. D. Broad’s ontology of mind. Boston, MA: De Gruyter (Originally published in 2006).

7 The Philosophical Implications of Foreknowledge Implications of ForeknowledgeImplications of Foreknowledge

Precognition, Fatalism and Time I.  The Epistemological Objection to Precognition Broad is concerned with the philosophical implications of precognition, or foreknowledge, in both his early work and later work (1921, 1923, 1937a, 1937b). Indeed, it is a topic he considers throughout his career. Broad defines “foreknowledge” (or “precognition”), as “a non-inferential veridical cognition of events in the future” (1937a, p. 180). In this chapter, I shall explore his attempt to defend the possibility of precognition against two a priori objections, one epistemological and the other metaphysical, and then I shall briefly consider an R-theoretic response to the challenge of fatalism. For our purposes, what is particularly important about his account of precognition (and memory) and fatalism is their connection with time. The epistemological objection to precognition can be succinctly stated as follows: To know a future event, to have direct acquaintance with it or to prehend it (terms Broad uses more or less synonymously), the future must exist. Precognition, like any veridical cognition, is a relation (prehension) between a present act of cognition and a cognized object. Since, however, a cognitive relation cannot hold “until and unless what it is about exists” (1937a, p. 180; emphasis added), and since the future does not yet exist, it follows that precognition is impossible. Broad states the objection as follows: To say that a person P had a non-inferential veridical cognition of an object O at a moment t is to say that the object O stood at the moment t in a certain relation to the Person P, viz., in the relation of being cognized by P. Now an object cannot stand in any relations to anything unless and until it exists. But to say that P had a noninferential veridical precognition of O at the moment t implies that O did not exist at t, but only began to exist at some later moment t1. So the phrase “non-inferential veridical precognition by P of O at t” involves a plain contradiction. It implies that O stood in a certain relation to P at a time when O did not exist, and therefore could not stand in any relation to anything. (1937a, p. 180; emphasis added)

Implications of Foreknowledge  165 This argument assumes that if at a time t we have a veridical precognition, then the object of that cognition must exist at t, but that is clearly incompatible with foreknowledge; a veridical cognition of an object that exists after the time it is cognized. Broad argues for the implausibility of the epistemological argument by first noting that an analogous argument against memory knowledge would rest entirely on the tacit assumption that to remember an event is to have a present prehension of a past event. Since prehension or direct acquaintance requires a relation between existents, and memory knowledge and foreknowledge are about what no longer exists or does not yet exist, they are both equally impossible. Broad draws the analogy between memory knowledge and foreknowledge in the following passage: If it is obvious that a term which does not yet exist cannot yet stand in any relation to anything, it is equally obvious that a term which no longer exists can no longer stand in any relation to anything. But to say that I remember at t2 an event which happened at t1 is to say that at t2 this event has the relational property of being cognized by me. On the other hand, since the event no longer exists at t2, it can have no relations to anything at that time. (1937a, p. 180; emphasis added)1 Thus, if the epistemological argument demonstrates that foreknowledge is impossible, then an analogous epistemological knowledge argument would establish that memory knowledge is impossible. In a talk on “Philosophical Implications of Precognition” published in The Listener (1947), Broad states his reply to the epistemological argument against precognition: I am sure that there is nothing in this objection. In the first place we may note that, if it were valid, it could equally be used to disprove the possibility of ordinary memory of events. For a remembered event has ceased to be actual by the time when it is remembered. (1947, p. 709) Thus, Broad’s initial response to the epistemological argument against foreknowledge is that if the argument against precognition were sound, then it would also apply to memory knowledge since memory knowledge involves a cognitive relation to what does not exist or no longer exists. Since memory knowledge is not impossible, he considers this argument to be a reductio of the claim that foreknowledge is impossible. Broad is right, “there is nothing in this objection,” but I suggest that the error in the argument is the view of time that it assumes, namely, either presentism or the growing block view. The issue is, can there be

166  Implications of Foreknowledge a relation between a present veridical cognition and a future (or later) and a past (or earlier) object/event/fact? If presentism is the case, then the answer is “no” since only the present exists. In the growing block theory, memory knowledge would be possible since the past exists, but foreknowledge would be impossible since the future does not exist. Clearly, the epistemological argument against foreknowledge assumes that if a person P stands in a veridical precognitive relation to some object O, then the act of prehending O must be contemporary with O (or O must be or grounded in something that exists in the present); a view that is implied by presentism, but not by the R-theory. Indeed, why can’t an act of cognition have as an object an event that exists later (or earlier) than t, the time of cognition? Admittedly, a relation requires relata, but if a temporal relation of succession can obtain between entities that exist but are not simultaneous, then why not a cognitive relation as well? In the R-theory, a veridical cognitive relation between a mental act and its object involves a simple, unanalyzable intentional relation that obtains between existents. To say that the past no longer exists, or that the future does not yet exist, implies that past and future events do not exist simultaneously with my thinking of them, but that is compatible with having a cognitive act such as memory that intends an event that is earlier, or an anticipation of an event that is later than the time of cognition.2 To further develop this R-theory response to the epistemological argument against foreknowledge and memory note first that there is an ambiguity in the phrase “non-inferential veridical precognition (or postcognition) P of O at t1.” We can get at the different interpretations by abbreviating the ambiguous phrase by: (a) at-t1, P is related to O, (b) P-at-t1 is related to O-at-t1, (c) P-at-t0 is related to O-at-t1, (d) P-at-t2 is related to O-at-t1. Admittedly, if (a) at-t1, P is cognitively related to O, so that the fact that P is related to O exists at time t1, then one may naturally infer that P and O exist at-t1, in which case the cognitive relation between them is neither foreseeing nor remembering. Similarly, if (b) then once again, P and O are contemporary, and therefore, P cannot have precognition (foreknowledge) or postcognition (memory) of O. However, to say that O does not yet exist at t1 does not rule out the possibility that O exists later than t1. Thus, (c) is compatible with then O existing later than t0–the time of P’s intending O-at-t1. Since O does not exist at the time of P’s precognition of O, P is aware of O as not yet existing by being conscious of the precognition (or anticipation or expectation) of O as being earlier than O-at t1. Finally, if (d), then O-at-t1 is earlier than P-at-t2, and P is conscious that the memory of O is not simultaneous with O-at-t1, but later. For that

Implications of Foreknowledge  167 reason, P is aware of O as not existing or no longer existing at the time of memory (or postcognition), but earlier. Thus, (a) and (b) are not compatible with the possibility of “non-inferential veridical precognition or postcognition by P of O-at-t1,” but (c) and (d) are. In other words, if one rejects the view that facts exist in time and the presentist view that only the present exists, as R-theorists do, then I see no a priori reason why a non-inferential veridical cognitive (intentional) relation could not bridge a temporal gap and connect an earlier act of foreseeing with a later event foreseen, or an earlier event remembered with a later memory of it. Broad, however, does not take this way out since he rejects the R-theory during the period when he was working out his A-theoretic ideas while examining McTaggart’s and Russell’s views on time. Rather, he argues that since we have ostensible non-inferential veridical memory, and since we are not prepared to reject memory knowledge, there must be something wrong with the analogous arguments against both memory and precognition. Broad claims that the flaw in the epistemological objection is to suppose that foreknowledge (and memory) requires direct acquaintance (prehension) with a future (or past) object that is not present. He demonstrates that by providing analogous analyses of veridical memory and precognition that do not require direct acquaintance or a direct cognitive relation between an act of memory or anticipation of an object that is present or simultaneously with the act. Broad begins by explaining how memory knowledge is to be analyzed. In the case of memory, Broad first defines “ostensible memory” as a purely psychological term regarding how memory appears to us with no implication about truth or falsity of what the memory belief is about. He then claims that in an “ostensible memory” there is an acquaintance with a particular, that is, an existent image (not the past event, but a “retro-presentative” image) and a non-inferential judgment/belief about something (a timeless proposition) based upon this prehended image that a certain event has happened. If the judgment or proposition is true, then there was such an event that corresponds to or accords with the proposition about it. Broad gives his preferred alternative analysis of memory in the following passage: On the right analysis something is prehended, viz., an image. But this is contemporary, and it is not the remembered event. Again, something is judged or believed on the basis of this prehended image. This something is a proposition, to the effect that there was an event of such and such a kind in the experient’s past life and that the prehended image is its present representative. This proposition, like all propositions, has no date; it is not an event or a thing or a person, though it is about a person and about a past event. There is, therefore, no difficulty in the fact that it can be the object of a present act of believing. Lastly, if, and only if, the ostensible remembering is

168  Implications of Foreknowledge veridical, there actually was such an event in the experient’s past life as he believes there to have been on the basis of the present image which he is now prehending. In that case, and only in that case, there is a relation, though a very indirect one, between this past event and the present experience of ostensible remembering. It is this. The past event then corresponds to or accords with the present belief about his own past which the experient automatically and uncritically bases on his present image. (1937a, p. 185) Thus, in Broad’s analysis of memory we are not directly acquainted with or cognitively related to a past event, but we are able to have knowledge about it by knowing that the timeless proposition is true. Therefore, although there is no direct cognitive relation that exists between P-at-t1 and something that does not exist at the time of cognition, namely, O-at-t0, there is a relation between the proposition believed and the event (or fact) that there was such a past event. If the memory judgment is veridical, there is a past event that corresponds to or accords with the proposition in question. To summarize, in Broad’s analysis of memory the past is not actual (does not exist now), but it need not exist now as an object of prehension for us to have non-inferential cognition of it. If it was actual, or if it did exist, then it can be indirectly cognized. Thus, what is mistaken in the epistemological argument, according to Broad (1937a), is not that it assumes a mistaken presentist analysis of time, but that it has a mistaken analysis (direct realist view) of memory. Broad claims that a precisely analogous analysis can be given of veridical precognition. There is a present image that gives rise to a belief in a proposition that there will be an event in the person’s future, and if the proposition is true, then it corresponds to a future event. Broad maintains that the epistemological objection can be avoided by analyzing ostensible foreknowledge in the same way in which he analyzed ostensible memory. There is an act of imaging, or prehending of an image—a particular existent—that occurs simultaneously with the act. And then, that ­situation triggers the person to Automatically, uncritically and non-inferentially base upon this prehension of this image a belief that there will be an event of a certain kind, of which this image is a present representative. If his ostensible precognition is veridical, this present belief will eventually be verified by the occurrence of such an event as he believes to be going to happen. (1937a, p. 187) Therefore, in the correct analysis of memory and foreknowledge, he believes the a priori epistemological argument vanishes.

Implications of Foreknowledge  169 One question that arises in Broad’s analysis of memory knowledge (that also applies to foreknowledge) is its connection with time. More specifically, according to Broad, memory knowledge involves making a true judgment about a past event; a judgment to the effect that the past event corresponds to the proposition about it. How, then, is the correspondence relation between the proposition and the past event related to time? We know that the proposition is timeless, and the event it is about is in time. What, then, of the relation of correspondence between them? Is it tenseless (non-temporal) or tensed (temporal)? Given Broad’s A-theoretic ontology of time, both alternatives are problematic. For if the correspondence relation is tenseless, and hence, in Broad’s view nontemporal, then the fact that a proposition tenselessly accords with or corresponds to a past event is timeless. However, if a proposition corresponding to a past event is a timeless fact, then it is an unchanging fact. But it is not an unchanging fact. Before the event became past it was present, and at that time it is not true that there was such an event in the experient’s past life. In other words, propositions about the past cannot tenselessly correspond to past events, since when an event is not past, but present, the past-tense proposition about that event is false. However, if the fact that a past-tense proposition corresponds to a past event cannot be timeless, the proposition cannot tenselessly correspond to it. Moreover, if true propositions about the past do tenselessly corresponds to past events, then the true propositions about the present and future also tenselessly correspond to present and future events. It follows that it is timelessly a fact that true past-, present- and future-tense propositions about the same event correspond to the same event that is past, present and future, and that is impossible. Nor would it suffice to say that the event has incompatible A-properties successively or one earlier than another because for Broad, temporal relations are founded upon the notions of past, present, future and becoming (1937b, p. 243), and his analysis would be viciously circular. Hence, the correspondence relation must be tensed, and for that reason, if there is a veridical precognition (or memory) or postcognition (or memory’), then the proposition triggered by the prehended image that there was (or will be) such an event now corresponds to that past (or future) event. The problem with this analysis is that propositions must change their truth value with the passage of the event through time. When the proposition that an event will occur is now true, the propositions that the event is now occurring and did occur are now both false. Then, when the event becomes present, the present-tensed proposition becomes true, and the future-tense proposition becomes false and remains false; and when the past-tense proposition becomes true, then the present-tense proposition becomes false and remains false. However, if the propositions required for Broad’s analysis of veridical intentional acts such as memory, foreseeing and perception change their truth value then they cannot be the

170  Implications of Foreknowledge timeless and hence unchanging propositions Broad claims them to be, but collapse into particular existents. We see, therefore, that whether the correspondence between timeless propositions and intrinsically past, present and future events are tenseless or tensed, Broad’s analysis of memory knowledge and foreknowledge are incompatible with his tensed theory of time. Timeless, yet tensed, propositions and tensed events are of fundamental importance not only to Broad’s view of memory and foreknowledge, but to his entire philosophy of mind. For consistency he would have to hold, and I believe very probably does hold, that all cognitions that go beyond direct acquaintance essentially involve propositions. For that reason, one key constituent of his philosophy of mind is incompatible with his views on tensed properties. Broad claims that if there is veridical foreknowledge, then the proposition believed will eventually be verified. This suggests that in foreknowledge the correspondence between a future event and the proposition about it is tensed, but not present tensed (1937a, p. 185). In precognition, there is a future event that will correspond to the future-tense proposition about it, and in memory the belief in a proposition that there was an event in the past, if true, did correspond to that past event. The problem with this analysis is that it is subject to a version of the epistemological problem he sought to avoid. Recall the problem with precognition was that it seemed to require the present prehending an object in the future that did not yet exist. Similarly, if foreknowledge involves believing a proposition about the future that will correspond to a future event, then the correspondence between the proposition and the event does not yet exist and so the proposition cannot be true at the time it is believed. And if the proposition is not true at the time it is believed, then the belief in the future event cannot be foreknowledge. Similarly, if in memory, there was a correspondence relation between a proposition about the past and the past event then there no longer is or no longer exists that relation, and therefore the belief in the existence of the past event cannot now be true. Thus, in Broad’s analysis of precognition, in order to avoid the epistemological objection the proposition about the future must now correspond to the future event. However, if the present belief that there will be an event now corresponds to the fact that such an event is future, then fatalistic worries emerge, as we shall see in the next section.

II. Broad’s Critique of the Fatalistic Objection to Precognition Broad considers what he calls “the fatalistic objection against precognition.” To explain what the fatalistic objection is, he first defines the phrases, “dependent on a voluntary action” and “predetermined” as they apply to events. An event e is dependent on a voluntary action or decision d when e would not have taken place if we had chosen differently, given

Implications of Foreknowledge  171 all the same circumstances. In other words, some events are brought about by us, and wouldn’t have happened if we had chosen differently (or if we did not bring them about), even if all the antecedent conditions (circumstances) of e, together with the laws of nature, had been the same. According to Broad, we believe that many events are such that they are dependent upon voluntary actions in that sense. To claim that an action is “already completely predetermined” at a certain earlier time t means that: There is a set of facts about the dispositions, the mutual relations, and the internal states at or before the moment t of the various substances then existing, which, together with the laws of matter and of mind, logically entails that an event exactly like e will happen after an interval t1—t in the place or the mind in which e did happen. (1937a, p. 204) Broad goes on to argue that if an event e is dependent upon d, and d is not completely predetermined, then e is not completely determined at any moment t before d. According to the fatalistic argument, as Broad understands it, only if both of those conditions are met, that is, e is dependent on d and d is not completely determined, can e be freely chosen and not fated. Hence, if e is predetermined then it is not dependent on d, and in that sense, fated. The next premise in the argument against foreknowledge is something Broad claims that people find self-evident, namely, that If an ostensible precognition occurs and is subsequently fulfilled, then, unless this is a mere chance coincidence, the event which subsequently fulfilled it must have been completely predetermined at the time when the ostensible cognition took place. (1937a, p. 206) In other words, if we can know that e will occur, and e occurs after our knowledge of it, then e is predetermined at the time of cognition. Given that allegedly self-evident premise, the argument against precognition may then be stated as follows: 1. Suppose that there is a veridical non-inferential precognition of an event before the time when the (ostensibly) voluntary action upon which the event depends occurs. For example, suppose e at t3 is known at t1, and e depends on a decision made at t2. 2. Suppose as we have assumed, that if an action is voluntary then it would have been different if the person had chosen differently, all other circumstances being the same. 3. Suppose further that if the precognition of an event occurs (before the decision to bring it about), then the event is predetermined.

172  Implications of Foreknowledge 4. It follows either that: a. The event which subsequently fulfilled the precognition did not really depend on the “voluntary decision” on which it seemed to depend, since the event could not have been different given the prior circumstances and laws of nature, even if I chose differently. b. If the event did depend on the voluntary action, then the “voluntary decision” must have been already completely predetermined at the time when the precognition took place. 5. Both consequences (4a) and (4b) are unacceptable since they each entail the unacceptable claim that there are no voluntary actions. 6. Therefore, step (1) and, hence, foreknowledge, is impossible. As Broad puts it, we are intellectually and emotionally committed to the notion that some of our actions do depend on us in a sense that precludes their being already completely determined. For if they were completely determined then nothing depends on us, and our decisions are unfree. Broad calls this the “fatalistic objection” to foreknowledge. Broad considers this to be a valid argument, but unsound since he maintains that premise (3) is false. He demonstrates that it is false by arguing that it rests on a confusion between “predetermined” and “predeterminate.” If an event is “predetermined” then it must be “predeterminate,” but the converse does not hold. The determinateness of future events at t3 that are known at t1 does not imply that future events are already predetermined at t1. Predetermined involves causation, but predeterminate involves no such notion. Broad explains the sense in which a future event is determinate as follows: Now consider an event e which actually happened at a certain moment t1 in a certain place or in the mind of a certain person. What would be meant by saying that e “was already completely predeterminate” at a certain earlier moment t? It would have the following meaning: If c be any characteristic which e manifests, then a judgment made at t to the effect that there will be a manifestation of c at t1 in this place or in this mind would already have been true at t. (1937a, p. 206) An event e is predeterminate when a judgment at t about a future event e, such as “event e will manifest the characteristic c at t1,” is either “already true” or “already false” at time t. It does not follow, however, that at time t e is already predetermined. Broad’s rejection of premise (3) regarding “determinate” may avoid the fatalistic objection to foreknowledge as Broad understands it, but it does not avoid a more serious fatalistic argument, as I shall argue.

Implications of Foreknowledge  173 A true proposition about the future already accords with or corresponds to a fact, then at the earlier time there is something that grounds or corresponds to the future event. If it is already true at t that there will be a manifestation of C at t1, then the future-tense proposition now corresponds to the fact that e will be c at t3, or e manifests c at t3 and t3 is future, and for that reason that fact already exists, or exists now, at t. Of course, one might dispute, as an R-theorist certainly would, and as I shall argue in the next section, that judgments about the future are grounded in facts that already exist. One may agree that a judgment that “a will occur,” if true, is already true in the sense that it is true when it is uttered or thought but deny that the truth maker already exists or exists at the time it is true. In other words, one may certainly ask, if a proposition about the future is true today, what does that imply about the temporal location, if any, of the fact to which it corresponds? That is one issue, but it is not the issue that concerns Broad. Rather, Broad is concerned with the connection or lack of such between predetermined and predeterminate events. Does the fact that it is already true (predeterminate) imply that sufficient causal conditions exist that the event will occur and so is determined? Broad argues that it does not. According to Broad, it is possible that at t2 a person’s voluntary action was the cause upon which the event at t3 depends, even if the proposition that e will have c at t3 is already predeterminate at t1. To put the point differently, there can now be a true proposition about the future, and thus a ground for the existence of an event that will occur, without implying that either the event or the decision that brought it about is determined. Since it is only predetermined decisions or events that are incompatible with an action being dependent upon a voluntary decision, the fact that foreknowledge implies that something is predeterminate at an earlier time does not entail that the event foreseen is predetermined.3 With the distinction between predeterminate and predetermined in hand, Broad continues: Now that this confusion has been removed, we can easily settle the question for ourselves. Suppose that the occurrence of e at t3 was foreseen by A at t1. Suppose, further, that the occurrence of e at t3 was in fact dependent on the occurrence of a certain voluntary decision in B at an intermediate date t2. Does this entail that the occurrence of this decision in this person at t2 was already predetermined at t1? (1937a, p. 207) No, it only entails that the decision at t2 was predeterminate at t1. The possibility of precognition or pre-prehending of a future event rests on the event being predeterminate or already (is now) true at t1, that the

174  Implications of Foreknowledge event will occur and that requires a tensed view of time, but it does not require that the future event or the decision to bring it about is causally determined. Even if it is known at the earlier time, and thus even if the fact that e will occur at t3 exists when it is known (at t1), it does not follow that the decision at t2 to perform that action at t3 was predetermined; that e’s having c at a future time t3 was not dependent on a voluntary action at t2. Broad argues that if we preprehend at t1, or have a direct acquaintance at t1 with an event e manifesting the characteristic c at t3, then we are having foreknowledge of the future event. We may not verify that the judgment is now true until e manifests c at t3, but we can have foreknowledge of it before we can verify it. Broad expresses this point in the following passage: If an event can be pre-prehended, it must in some sense co-exist with the pre-prehension of it; and the precognition must consist in knowing by acquaintance that it has such and such characteristics. This would be impossible unless it is in some sense already true that it has these characteristics, i.e., unless it is in some sense already predeterminate. Supposing that a meaning can be given to the notion pre-prehension, it is quite clear that an event need not be completely predetermined. All that is necessary is that it should then be predeterminate. (1937a, p. 207) If we can pre-prehend (know) that e will occur, then we must know by acquaintance that an event e is (now) manifesting such characteristics at t3. Since Broad rejects the existence of a non-temporal copula when dealing with temporal facts, then what is pre-prehended must be a futuretensed event or fact that exists now. Hence the problem: since Broad says that “precognition must consist in knowing by acquaintance that it [an event] has such and such characteristics,” (1937a, p. 207), he must be talking about a present-tense fact (that event e is now future), something that “is in some sense already true that it has these characteristics, i.e., unless it is in some sense already predeterminate” (1937a, p. 207), and for that reason, the problem of fatalism emerges. For that means that it is true before the time at which the event happens, more specifically, that it is true now; i.e., e will manifest characteristic c in the future. However, that implies that there now exists a ground for its truth or it is presently existing; it is already a fact. Whether it is a future-tense fact that e will manifest c at t3, or a present-tense fact that e manifests c at t3, and t3 is now future, makes no difference. If a fact (e manifests C at t3) about the future already exists, then there is nothing I can do now to prevent it or bring it about. If the veridical non-inferential precognition of the future depends on a future event existing in the present, then the present is big with the future, or the

Implications of Foreknowledge  175 future pre-exists in the present. Thus, it is the determinateness of the future together with a tensed view of time and not its predetermination that gives rise to the a priori metaphysical problem with foreknowledge. The criticism I am making against Broad can be stated differently, as follows. Broad claims that propositions about the future are determinate, i.e., they are already (now) true or already (now) false. Broad recognizes this, for if judgments about the future are determinate, i.e., already true, then there already exists, it now corresponds, to a future event that can be directly prehended. Recall, Broad explains what he means by “predeterminate” in the following passage: Suppose that a judgment is made at any moment t to the effect that an event manifesting the characteristic c will happen in a certain place or in a certain mind at a certain future moment t1. Then this judgment is already true, or it is already false at the time t when it is made. The actual course of future history will show that it was true or will show that it was false, as the case may be; but the judgment will not become true or become false, from being neither the one nor the other, when the moment t1 is reached. (1937a, p. 206) However, if a judgment made at t to the effect that there will be a manifestation of c at t1 in this place or in this mind is already true at t, then it already (now) corresponds to or accords with an event that “co-exists” with the pre-prehension, at t. In saying it is now already true, he thus implies that the future-tensed fact already exists, or exists now, and therefore the future already exists, that is, it exists in the present. If the fact will correspond but does not yet, then it would not already be true, but would become true. Yet Broad explicitly denies that the judgment will become true and asserts that it is already true, and therefore the judgment or belief that x will manifest c at t1 now corresponds to the event or fact. If there is such a fact now, then no later decision can bring it about or prevent it. This has several consequences. The reason for introducing the notion of predeterminate was to remove the specter of causal fatalism from the possibility of precognition, but (logical) fatalism remains. For if I will choose to do e at t3 already exists at t1, then what I will do tomorrow is not dependent upon me. That is, all things could not be the same and yet I choose to do something different. For if the ground of what happens tomorrow already exists today, not causally but logically, then the decision to choose to bring about e is c at t3 is such that I could not have decided to do something different, since the fact that I will choose at t2 that e is c at t3 already exists now, at t1. I cannot choose or decide at t2 to bring about something different tomorrow from event e’s occurring at t3 if my decision (and the event occurring at t3) to bring it about tomorrow already exist today.

176  Implications of Foreknowledge Thus, Broad’s attempt to preserve foreknowledge against the fatalistic argument fails. Broad fails to see this because (1) he confuses the ontological ground of the truth of a proposition about the future with the epistemological ground (cannot be verified). Presumably, since we do not know it was already true until the event happens, it does not already exist. Nevertheless, its being already true before the event it is about happens does take away our freedom—not its being already verified. (2) He does not to realize that fatalism is not solely about the determinateness of the future (and the law of excluded middle) and but also about the ontology of time. I will clarify this point in the next section. (3) Broad does not clearly distinguish causal and logical fatalism. The determinateness of the future together with his tensed theory of time creates fatalistic worries, even if the determinateness of the future does not imply that it is predetermined. In short, the issue of fatalism is not epistemological nor is it causal; the issue of fatalism is logical. Thus, although Broad’s view implies neither epistemological fatalism nor causal fatalism, it does imply logical fatalism.

III.  R-Theoretic Response to Fatalism Suppose, then, that at t1 it is true that a sea fight will take place tomorrow. What, we must ask, is the state of affairs in virtue of which it is true? In the argument for fatalism it seems clear that in the case of propositions about the future that are already true they correspond to events or facts that already exist, or co-exist with the prehensions or precognitions of them. What is also clear is that if a proposition is now true (at t1) then the state of affairs in virtue of which it is true must now exist (at t1). In other words, truth goes with existence, and present truth with present existence. Therefore, if a proposition about the future is now true, then it corresponds to an existing state of affairs; and if a state of affairs exists then it is present. Once we build into the fatalist argument the claim that (i) truth involves correspondence between propositions and existing states of affairs and that (ii) existing states of affairs are present, that is, co-exist with the precognition of them, the fatalist argument becomes very forceful. For if it is now true that “there will (or will not) be a sea fight tomorrow,” then there now exists the fact that there will (or will not) be a sea fight tomorrow, and consequently I can neither bring about nor prevent a sea fight tomorrow. More generally, I cannot bring about a state of affairs in the future because if it is already true today that I will bring about or prevent a certain state of affairs tomorrow, then it already exists today, and I cannot bring about or prevent tomorrow what already exists today. Unpacking the fatalist argument in the way I have done makes clear that it is not simply the logical law of excluded middle that implies that no person has free will, but rather it is that law coupled with a theory of temporal truth and a decidedly anti-Russellian theory of time. For the

Implications of Foreknowledge  177 argument assumes that true propositions correspond to facts (or truths) that exist in time, and that what exists in time exists in the present. If, however, at time t, a sentence about the future is now true, or already true, then the future is fixed and there is nothing we can do to prevent it. And if, at time t, a sentence about the future is now false, or already false, then there is nothing we can do to bring it about. In either case, as Aristotle puts it, “nothing is or takes place fortuitously, either in the present or in the future, and there are no real alternatives; everything takes place of necessity and is fixed” (1941, p. 49). The crucial assumption in this argument for fatalism is that if a sentence is now true at a time t, then there now exists (or “already exists”) at time t a fact in virtue of which it is true. In that case, the future exists in the present and fatalism is the result. Fortunately, the R-theorist need not accept the assumption since it depends on the reality of tense. Consider the future-tense sentence, “There will be a sea fight.” According to the R-theory as I understand it, the fact referred to by that sentence will vary depending on when it is uttered or inscribed. For example, if the sentence “There will be a sea fight,” call it S1, occurs before the sea fight, then the fact that exists if S1 is true is A sea fight occurs later than S1. However, that fact is not located at the time at which the future-tense sentence is uttered, and it is not located at any later (or earlier) time. Of course something is located when S1 is uttered, namely, the inscription or utterance that “There will be a sea fight,” and indeed, assuming the utterance is true, something is located at a time later than S1, namely, the sea fight, but the whole state of affairs that a sea fight occurs later than S1 is not located when S1 is, and it is not located at any later or earlier time, much less at every time. Indeed, it is not located in time at all.4 Facts that contain temporal relations between particulars or events are eternal3 in the sense of existing outside the network of temporal relations, but not in the sense of existing or persisting throughout all of time (eternally1), nor in the sense of being a relation among universals, existing eternally2. Any “eternal3” facts are timeless in the sense that they are outside of time because they do have temporal relations to any entity. Nevertheless, an eternal3 entity is related to time: it is a whole which contains successive parts. We could say that although an eternal3 entity is not contained in time, time is contained in it. Thus, the fact that World War II is later than World War I is eternal3, because although not itself located in time (or a term of a temporal relation), it contains time (a temporal relation) as a constituent. It is thus both a timeless and a temporal fact. Consequently, the truth of a future-tense sentence does not imply that the future “pre-exists” in the present, or that the future “already exists.” Of course, if one confuses the tenseless sentence “A sea fight occurs later than S1”—that represents the reality that exists if “There will be a sea fight” or “S1” is true—with the (tenseless temporal relational) fact represented, namely, a sea fight occurs later than S1, then fatalistic

178  Implications of Foreknowledge worries will emerge. For the tenseless sentence describing the fact that exists if the sea fight will occur, is tenselessly true. Its truth value is unchanging; if it is ever true then it is always true. If one then confuses the tenseless sentence “A sea-fight occurs later than S1” with the tenseless fact that it represents, one might conclude tenseless facts are sempiternal—are always existing—and that implies fatalism. For if a sea fight occurs later than S1 or a sea fight occurs simultaneous with S2 (where S2 is a later true token of “The sea fight is now occurring”) always exists and thus “already” exists when S1 is uttered, that is, at a time before the sea fight occurs, then my choice is an illusion, for I can neither bring about nor prevent at a time later than S1, or what already exists simultaneously with S1. In the R-theory, since the proper transcription of a future-tense sentence S is a tenseless sentence that represents a tenseless fact about the event occurring later than the time of the sentence, and not in virtue of anything that is located at the time of utterance, the difficulty disappears. One might still object that the existence of events that are tenselessly later than the sentences that are about them takes away our freedom. For if there is the timeless (eternal3) fact that my voting for Joe Biden on November 5, 2020 is later than my utterance that “I will vote for Joe Biden on November 5, 2020,” then, arguably, that fact necessitates the occurrence of my voting for Joe Biden in the year 2020, or synonymously, that it is not within my power, in 2020, to prevent my voting for Joe Biden. It is not at all clear, however, that the existence of an eternal3 fact necessitates anything. The necessity does not follow from the principle of bivalence. The law necessitates that one of a pair of contradictory statements is true, but it does not imply that either “P” is necessarily true or “not-P” is necessarily true. If I vote for Joe Biden on November 5, 2020, then the sentence on January 20, 2020 “I will vote for Joe Biden on November 5, 2020” (call it “W1”) is true, and if I do not vote for her, then the sentence on January 20, 2020 “I will not vote for Joe Biden in 2020” (call it “W2”) is true. Thus, which sentence is true depends on whether or not the fact I vote for Joe Biden is later than W1 exists or the fact that I don’t vote for Joe Biden is later than W2 exists, but which of those two facts exists depends upon what I will choose to do on election day, November 5, 2020. In other words, it is my later decision that determines which of two contradictory statements about the future is true, since it is my later decision that determines what eternal3 fact exists. Thus, the existence of eternal3 facts is not incompatible with our having it within our power to bring about or prevent certain events. Indeed, it is because we do have it within our power (because we do choose or do not choose) to bring about or prevent certain events that certain eternal3 facts exist.5 One might still wonder how the existence of facts that are not located in time can depend on something that can be located in time. How can

Implications of Foreknowledge  179 the “eternal” fact that I vote for Joe Biden later than W1 depend for its existence on something that occurs in time, that is, on what I do on November 5, 2020? Although eternal3 facts do not exist in time (i.e., as terms of temporal relations), they do contain temporal relations and their relata as constituents. Thus, the fact that I am going to vote for Elizabeth Warren, if I do, includes my decision in November 2020 to do so. That decision is an event that exists in time, and if it does not occur on November 5, 2020, then there is no such fact as my voting for Joe Biden on that date. For that reason, the truth of a future-tense sentence does not entail the existence of a fact which in turn determines my choice. Rather, my choice in November 2020 determines what (tenseless) fact exists, and hence what future-tense sentence is true. Interestingly, Storrs McCall, an A-theorist, who once argued that propositions about the future must have an indeterminate truth value, has recently changed his mind. And what he says on this topic closely reflects the R-theorist’s view: We shall say that the truth of an empirical proposition supervenes upon events in the sense of being wholly dependent upon them, while at the same time events in no way supervene upon truth. Thus the truth of the proposition that X is in Warsaw town square at noon next Friday depends upon what happens next Friday, and in this way the sting of “logical determinism” ’ is drawn. . . . What is true today depends upon what happens tomorrow, not the other way round. The set of true propositions in no way determines what the future is like. Instead, what the future is like determines the set of true propositions. (McCall, 1984, p. 14) Nevertheless, there is an intuition that we all share according to which the past is unalterable, fixed, and already settled and thus no longer within our power to control, whereas we are free to make choices that will determine how the future will be. What account of the phenomenological asymmetry of openness with respect to the past and future can the R-theorist give? In this concluding section, I will mention three possibilities congenial to the R-theory that are worthy of further exploration. The asymmetry of openness is analyzed by David Lewis as an asymmetry of counterfactual dependence. The future depends counterfactually on the present in a way in which the past does not so depend. As Lewis puts it, What we can do by way of “changing the future” . . . is to bring it about that the future is the way it actually will be, rather than any of the other ways it would have been if we acted differently in the

180  Implications of Foreknowledge present. . . . Likewise, something we ordinarily cannot do by way of “changing the past” is to bring it about that the past is the way it actually was, rather than some other way it would have been if we acted differently in the present. (1991, p. 53) This suggestion by Lewis is quite plausible and it is compatible with both the R-theory and with determinism. My major hesitation concerns Lewis’ analysis of counterfactual dependence and the subsequent commitment to possible worlds that it entails. Thus, in order to avoid the thorny problem of giving an adequate account of the truth conditions of counterfactuals, I therefore prefer, although I will not here explore in any detail, two suggestions put forth by Broad during his (early) R-theory stage. Broad says that there are two senses in which the past is fixed and unalterable, while the future depends, in part, on our volitions. He says: (i.) However much I may know about the laws of nature, I cannot make probable inferences from the future to the past, because I am not directly acquainted with the future, but I can make probable inferences from the past to the future; i.e., although every possible proposition about the future is even now determinately true or false, I may be able to judge now, from my knowledge of the past and present and of the laws of nature, that some propositions about future events are much more likely to be true than others. . . . (ii.) I know with regard to certain classes of events that such events never occur unless preceded by a desire for their occurrence, and that such desires are generally followed by the occurrence of the corresponding events. But the existence of a desire for x does not increase the probability that x has happened. If it did, we might be said to affect the past in exactly the same sense in which we can affect the future. Thus the assertion that we can affect the future but not the past seems to come down to this; (a) that propositions about the future can be inferred to be highly probable from a knowledge of the past and present, but not conversely, because of our lack of direct acquaintance with the future; and (b) that the general laws connecting a desire for x with the occurrence of x always contain x as a consequent and never as an antecedent. (1921, p. 146) According to Broad’s first point, the asymmetry of openness is epistemological. The second point seems to me to be more convincing. Our experience of the openness of the future is based upon the awareness that the desire for the occurrence of a certain event is something that can be causally efficacious or lawfully connected with the desired event only

Implications of Foreknowledge  181 when the event is later than the desire. And our experience of fixity stems from our knowledge that a desire or wish that something had happened is not lawfully connected with that event having happened. For these reasons, we do not experience the past as something we have any control over, whereas we do experience the future as something over which we do have some control. I think the R-theorist’s account of coming into existence and ceasing to exist is relevant here. For the openness of the future is, in part, based on our experience of an event that does not yet exist (i.e., is not located at the time of my decision to bring it about, but later), emerging or being located at a time after I decide to bring it about, and experiencing it as simultaneous with my expression of, or thoughts about it. On the other hand, the experience of the closed past, or the fixity of the past, stems from the experience or realization that my desire (at t2) for a certain event to have happened in the past (at t1) does not bring that event into existence. Similarly, my decision (at t2) to end an unpleasant event (at t3) may be lawfully connected with that event passing away (or not being located at t3), but my wish (at t3) for an unpleasant past event (at t2) not to have happened (or not to be located at t2) does not make a difference. These experiences, which form the basis of the asymmetry of openness between the future and the past, can be explained in the R-theory. Although much more could and should be said concerning the phenomenology of freedom, these brief remarks suggest the direction a R-theorist of time could consistently take.

Notes 1. Notice that this argument does not apply to his earlier view of time (Broad, 1921) since he there maintains that “past, present and future are equally real.” 2. Why would Broad assume that if P cognizes O, then P and O exist simultaneously? That P cognizes O is a fact. If that fact exists in time, then the terms of the relation must also exist at that time. Presentist denies the reality of facts as timeless and holds that all “facts” are present, and the present is always changing. Still, those facts that exist are now, or exist at the present time. Hence, the terms of cognitive relations must be co-present, and for that reason, the ground of veridical memory and foreknowledge must be found in the present. This line of reasoning misunderstands the nature of facts, that, in the R-theory, are eternal3. For, if one rejects presentism, and countenances facts as eternal3, then the a priori epistemological objection is unsound, not because it gives an incorrect analysis of cognition as a direct awareness of the future, but because it assumes a mistaken conception of time. 3. The distinction between determinate and determined, and the argument for why the former does not imply the latter, was made by A. Grünbaum (1976). It is also a point that Broad makes during his Russellian period. As he puts it: When it is once recognized that the whole course of events is in a certain sense a totum simul, it becomes easy to see the answer to the famous theological problem: How can God’s foreknowledge of men’s actions be compatible with the freedom of men’s wills? The answer is as follows. Whether

182  Implications of Foreknowledge men’s wills be free or not, every man’s future actions are as completely determinate as his past ones; this is a mere consequence of the laws of logic. If indeterminism be true, then no amount of knowledge about events previous to a moment t, and about the general laws of nature or the particular habits of a man, will enable us or even God to infer with certainty what the man’s volition at t actually is, although it it eternally perfectly determinate. These two statements are clearly quite compatible. Finally, in spite of the fact that God cannot infer the man’s volition at t, He may at any and every moment be directly aware of it in precisely the same way as we are aware directly (and not merely inferentially) of certain events through memories which are themselves later events. The facts that at a certain moment t1 God can have a state of mind whose immediate object is the volition of a man at some later moment t2, and that no amount of knowledge of events before t2 would enable Him to infer the volition at t2, are perfectly compatible; and they cease to be even paradoxical when we compare the case of memory and note that there is no essential difference between past, present, and future. (1921, pp. 150–151) 4. Robin Le Poidevin has suggested to me another argument for the claim that facts do not exist in time. Suppose that the fact (call it “F”) that e occurs at t exists at t. Then there is another fact, namely the fact that F exists at t, and so on ad infinitum. 5. Perhaps one will ask how, say, Jones is able to not eat dinner at t1 if it is true (or a fact) that Jones does eat dinner at t1. Very briefly, I would say, that the notion of “ability” or what we “can” do is ambiguous. Relative to one set of facts someone may be able to do something which relative to another set of facts one is unable to do. Thus, given the fact Jones eats dinner at t1 there cannot also be the fact that Jones does not eat dinner at t1, but given certain other facts which we ordinarily take to be relevant to what we can or cannot do, it can be the case that Jones does not eat dinner at time1. Given facts about Jones’s physical capacity to drive a car and to stop by the local pub, his propensity to have a drink and the typical rush-hour traffic around dinner time, and so on, he can avoid eating dinner at time1, but he will not. See David Lewis (1993).

References Aristotle. (1941). De Interpretatione. In R. McKeon (Ed.), The basic works of Aristotle. New York: Random House. Broad, C. D. (1921). Time. In J. Hastings et al. (Eds.), Encyclopedia of religion and ethics (Vol. 12, pp. 334–345). New York: Scribner’s. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 143– 173). References are to this reprint. ———. (1923). Scientific thought (Chap. 2). London: Routledge & Kegan Paul. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. II. Time and metaphysics (pp. 63–83). ———. (1937a). The philosophical implications of foreknowledge. Proceedings of the Aristotelian Society Supplementary Volumes, 16, 177–209. ———. (1937b). The philosophical implications of precognition. Aristotelian Society Supplement, 16, 229–245.

Implications of Foreknowledge  183 ———. (1947, May). The philosophical implications of precognition. The Listener, 37 (8), 709–710. Grünbaum, A. (1976). The exclusion of becoming from the physical world. In M. Capek (Ed.), The concepts of space and time (pp. 471–500). Boston: Reidel. Lewis, D. (1991). Counterfactual dependence and time’s arrow. In F. Jackson (Ed.), Conditionals (pp. 76–101). New York: Oxford University Press. ———. (1993). The paradoxes of time travel. In R. Le Poidevin & M. MacBeath (Eds.), The philosophy of time (pp. 135–146). Oxford: Oxford University Press. McCall, S. (1994). A model of the universe: space-time, probability, and decision. Oxford: Clarendon Press.

8 Conclusion

ConclusionConclusion

Broad, the R-theory and Time

In the conclusion, I want to bring together some of the main themes that I have been stressing throughout this book by recounting Broad’s views on time and my version of the R-theory. In his first writings, Broad was under the influence of Russell (1915, 1938), and adopts an essentially Russellian theory of time according to which temporal relations are the foundation of our knowledge of time and the ontological ground of all temporal facts. Initially, Broad explicitly rejects presentism and absolute becoming, the full-future theory and the growing block theory of time. The textual evidence for these claims abounds. Unfortunately, in his exposition of the R-theory, Broad says things about temporal relations and the events or particulars that are their terms that are misleading and ambiguous, and render his early view of time susceptible to objections, including those made by Broad himself in his later writings. In what follows I shall review Broad’s early view, explain how it may have led to his later objections to the Russellian theory in Broad (1938), and then summarize my R-theoretic responses to them.

I.  Language, Ontology and Time The “B-theory” is the name coined by Richard Gale after McTaggart’s B-series, whose generating relations are earlier than/later than. In the B-theory, “the B-series alone is sufficient to account for time” (Gale, 1967, p. 69). It appears, however, that what McTaggart meant by the B-series is not what Russell meant by it. This has led critics of the B-theory to overlook the R-theory, to which those criticisms do not apply. A good statement of the R-theory is found in the opening paragraph of Broad’s Encyclopedia entry on “Time” where he says, 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. . . . On these

Conclusion  185 relations of before and after which we immediately recognize in certain objects of our experience all further knowledge of time is built. (1921, p. 143) For the Russellian, B-relations are simple unanalyzable external relations, meaning that they cannot be defined in terms of causation or any other relation, nor can they be analyzed in terms of the A-properties of pastness, presentness and futurity of their terms. These relations are more aptly called “R-relations” to distinguish them from McTaggart’s view of B-relations. R-relations are known by acquaintance and are the basis of “all further knowledge of time” including our knowledge of the past, present and future. By adopting the R-theory, Broad rejects the full-future view that construes temporal becoming as the acquiring and shedding of A-properties as events pass from the future to the present and into the past. One reason Broad gives for rejecting A-properties is that while we are acquainted with time relations, we do not directly experience non-relational A-properties. As Broad puts it: The distinction between past, present, and future is not one which, like that between before and after, lies wholly in the experienced objects, but is one that rests on the relations between experienced objects and the states of mind in which they are experienced. (1921, p. 145; emphasis added) For that reason, Broad notes in several places that past, present and future are not “essential characteristics of objects in time in the same way as before and after are, instead of being analysable into the temporal relations of states of mind and their objects” (1921, p. 150). For example, an event is future if it is later than an anticipation or expectation of it, present if it is simultaneous with a perception of it and past if it is earlier than a memory of it. Unfortunately, Broad misleads the reader when, in an endnote, he says, “The point can perhaps be made clearer by reflecting that a tune has a pattern in time in exactly the same sense as a wallpaper has a pattern in space” (1921, p. 173, n 3). This quote tends to reinforce the mistaken view that the R-theory spatializes time, and that time is a “block universe” or a totum simul; a term Broad uses to characterize his view. Broad’s view on the relation between logic and language support his rejection of A-properties, and therefore also support his rejection of the full-future theory. He maintains that the laws of logic dictate that all propositions are timelessly true or false and therefore tensed sentences such as “It is now raining” and “It will rain” are incomplete until the time at which they are uttered, written or thought is included in the propositions they express. Thus, the proposition expressed by the sentence

186  Conclusion token “It is now raining” would be “It is raining simultaneously with this token of ‘It is now raining,’” and, in that way, grammatical present tense is reduced to a tenseless sentence that expresses a tenseless proposition or refers to a tenseless fact. Analogous reductions would apply to sentences about the past and future. Thus, Broad concludes, “The laws of logic are of course concerned with the timeless copula, and they presuppose that statements containing tenses are reduced in the way suggested above” (1921, p. 149; emphasis added). Broad further claims that “Whether men’s wills be free or not, every man’s future actions are as completely determinate as his past ones; this is a mere consequence of the laws of logic” (1921, pp. 150–151). For that reason, true judgments about the future correspond to facts that contain events or particulars that exist later than the time the judgment about them is made. Thus, Broad’s early view of time rejects the growing block theory according to which the future does not exist, although the past and present do. Finally, Broad rejects presentism, the view that “only the present exists, or that the present is a mere transition from one infinite non-existent to another. These phrases are mere rhetoric rooted in confusions” (1921, p. 150). Presentism confuses two senses in which the parts of a whole co-exist by implicitly arguing that because the parts of the whole course of history cannot co-exist at the same time, as the parts of a spatial whole do, they cannot co-exist as parts of a whole that occur successively as they allegedly do on Broad’s view. Yet this is necessarily false, since it is admitted that events do have and continue to have temporal relations, and therefore they must form a related whole all of whose parts have being. (Broad, 1921, p. 150) In other words, An event must continue to be, if it is to continue to stand in relations; the battle of Hastings continues to precede the battle of Waterloo, and therefore both these events must eternally be at their own respective moments. (1921, p. 150; emphasis added) These passages demonstrate that Broad is rejecting presentism and absolute becoming or the existential change objects undergo when they come into and go out of existence. For in his early musings on time, “To say that x no longer exists, or does not yet exist, simply means that it occupies a moment before or after my statement about it” (1921, p. 150). Nor do events undergo temporal becoming understood as the acquiring and shedding of non-relational temporal A-properties since

Conclusion  187 “these characteristics do not belong to events as such, but in virtue of the temporal relations between them and certain psychical events” (1921, p. 155). It is clear, therefore, that in “Time,” Broad rejects presentism, the growing block and the full-future theories and adopts some version of the R-theory of time. Unfortunately, many of the things he says in his early paper can easily give rise to objections. For in claiming that events must “eternally” be at their own respective moments; that events do have and “continue” to have temporal relations, and that “the whole course of events is in a certain sense a totum simul” (1921, p. 150), Broad may be accused of being committed to a static, block universe in which nothing changes. For if events continue to have temporal relations, then they must persist, and if they are eternally at a time, then they must be timeless. The conception of time suggested by such language, as Broad later alleges (1938, p. 61), is a series of “persistent particulars” or “abstract objects” in which the reality of temporal relations and the transitory aspect of time must be denied. I have argued that the R-theory has adequate responses to these objections. Before I summarize them, I want to first review Broad’s understanding of the relation between ordinary language and commonsense facts, on the one hand, and the methodology and ontology of time, on the other. For, as we shall see, their connection leads to Broad’s later views on time and his critique of the R-theory. According to Broad, “the most fundamental task of Philosophy is to take the concepts that we daily use in common life and science, to analyse them, and thus to determine their precise meanings and their mutual relations” (1952, p. 16). And, to subject those concepts, and the beliefs we have regarding them, to justification and criticism. The concept of time involves both an extensive and a transitory aspect. These two features of time are reflected in ordinarily life and language by means of tensed expressions. Some examples Broad gives are: “ ‘My grandfather died before I was born’, ‘I am now writing’, and ‘I had my breakfast before I began writing this sentence, and I shall have my lunch after I stopped writing it’”(1938, p. 36). Language also suggests that events move from the future to the present and fade into the past when we reflect upon an unpleasant event and assert “Thank goodness that’s over.” For Broad, all of these temporal facts are expressed in tensed language and for that reason, they prima facie presume a tensed or A-theory of time. In other words, since temporal facts are recorded by sentences that contain temporal or tensed copulas, ordinary language strongly suggests that some version of the A-theory is true. Of course, Broad is not asserting that we should accept a theory because it seems to be implied by the statements of plain men. God forbid! But I do suggest that any satisfactory theory must account for the fact that plain men and

188  Conclusion philosophers in ordinary life express themselves in language which strongly favours one alternative. (1925, p. 585) Although these comments occur in connection with the debate between the Center and Non-Center views of the self where Broad clearly believes that ordinary language and commonsense facts favor some version of the Pure Ego Center theory; he also believes that ordinary language favors some version of the A-theory as well. Indeed, he employs this attitude in his argument first, for the full-future theory of time, and then for what appears to be a defense of presentism (see Chapter 4). However, one might argue that the appeal to commonsense facts and ordinary language should not be given the importance Broad gives it in determining ontological commitments. To illustrate this point, I shall return to R. D. Ingthorsson’s defense of presentism (see Chapter 5). One overarching theme in Ingthorsson’s McTaggart’s Paradox (2016) is his appeal to commonsense or what we intuitively believe about time and change. He uses commonsense in his critique of McTaggart’s argument for the unreality of time, in his critique of the B-theory view of change and in his defense of presentism. For example, he says The problem with McTaggart’s argument is not that it is obscure or invalid. The problem is that the conception of an A series with which he finds fault is not how time appears to be, it is not how the man on the street in general thinks that time is . . . nor is it how the majority of A view sympathisers depict time as being. The problem is simply that it is a refutation of a conception that hardly anybody believes in. (Ingthorsson, 2016, p. 76) Similarly, Ingthorsson’s claims that “endurantism represents what has been the commonsense view of persistence” and identity through change, and that, furthermore, “Most people will take for granted that only those things exist that exist now” (2016, p. 89); that presentism is the case. There are, however, several problems with Ingthorsson’s appeal to commonsense to defend these positions. Ingthorsson mentions Augustine, who famously recognizes that what “we learned as boys, and as we have also taught boys” (Confessions: Ch. XVII, sect. 22; quoted on 2016, p. 115) is that the passage of time involves the movement of events from the future through the present and into the past, and therefore must exist. Moreover, he recognizes that there is an incompatibility or at least an “unresolved tension” (2016, p. 115) between different commonsense intuitions or what most people believe about time, since he says, “Although it may be true that most people believed that only the present exists, most people also believed that time really passed from future to

Conclusion  189 present, and from present to past” (2016, p. 114). For that reason, it is a mistake for Ingthorsson to appeal to what most people believe to justify the presentist view of tense and reject McTaggart’s positive view that past, present and future exist, rather than to accept McTaggart’s positive view and reject presentism. What we need to do, it seems to me, is to distinguish between commonsense and ontology. It is thus a mistake to pack into commonsense a specific ontological analysis, and then to criticize an alternative analysis by claiming it is incompatible with commonsense. What needs to be grounded are the phenomenological data. Oftentimes, the data are reflected in our commonsense beliefs, for example, that time passes and objects change, but our commonsense beliefs should not be identified with a specific ontological analysis while simultaneously assuming that analysis to be strongly suggestive or intuitively true. In fairness to Broad, he does not consider ordinary language or commonsense as conclusive regarding matters of ontology. He recognizes that the grammatical structure of the tensed language that records temporal facts may be misleading and suggest an ontological analysis of those facts that is mistaken (1938, p. 38). Nevertheless, in his later writings, he never really doubts some version of the A-theory and his reasoning moves, in part, from the ineliminability of the temporal copula “is now” in sentences that record temporal facts, to the ineliminability of tense from the ontology of time. For Broad maintains that all temporal facts are recorded in ordinary language by tensed or temporal copulas. Thus, if sentences that recorded facts about the R-relations between events employed tenseless or non-temporal copulas, then the facts recorded would be non-temporal or timeless. Broad concludes that if we construe the language of time as tenseless, as the R-view does, then the R-series and R-facts would neither have tensed properties among their constituents nor undergo A-theoretic becoming, and thus they would be timeless. Here, for a fact to be “timeless” means that it has no intrinsically temporal entities among its constituents. Hence the tenseless R-series composed of a conjunction of many R-facts would be an “eternally” existing series of terms that are either timelessly eternal2 or temporally eternal1 (sempiternal). In other words, the R-series would be a series analogous to a timeless series of integers, or a spatial series of points on a line, and so would not be a temporal series generated by a time relation. Broad is mistakenly reading into a grammatically tenseless sentence such as “e1 precedes e2,” which stands for an R-theoretic ontologically tenseless fact, an analysis that an R-theorist need not and would not accept. Why does he assume that a sentence without a temporal or tensed copula implies that the fact it records is a non-temporal or timeless fact? Perhaps he reasons that if the copula is tenseless then the fact it records does not contain temporal properties, and without temporal properties the fact cannot have any intrinsically temporal entity, for example the

190  Conclusion earlier than relation, as a constituent. At this point, we can see why Broad thinks the tenseless copula or tenseless “precedes” would suggest that the terms of temporal relations somehow co-exist either timelessly or sempiternally; it is because he either implicitly assumes that Russell’s account of R-relations is indistinguishable from McTaggart’s account of B-relations as internal relations, or because Broad fails to recognize the R-theory is an analysis of temporal facts, that is, the temporal phenomena given to us in experience, and it is not an analysis of temporal language. To clarify this point, recall that in one notion, a fact is non-temporal or timeless if it does not contain or have an intrinsically temporal entity among its constituents. In a second notion, a non-temporal fact or timeless fact is outside of time, which, since in the R-theory time is purely relational, means that no fact is a term of a temporal relation. In the second sense all facts are timeless, but as I use the term, only temporal facts are eternal3, meaning that they have intrinsically temporal relations as constituents, and are also timeless since they are not themselves in time. Thus, in the R-theory “e1 is earlier than e2’” records a timeless fact because it is not temporally related to anything; that is, it does not stand in a temporal relation to anything. It is, however, also a temporal fact because it contains time in the form of a time relation, earlier than, as a constituent. In ordinary language, only sentences with tensed temporal copulas record commonsense temporal facts, but in an ideal language constructed to be “stand ins” for, or depict, the correct ontology, there are no tensed copulas, but there are tenseless yet temporal copulas, that perspicuously represent the ontological nature of temporal facts and time. These facts are outside of time, yet temporal, because they contain temporal relations and so ground the temporal phenomena. Thus, the R-theoretic reply is that even though the copula is tenseless linguistically, and the ontological form is tenseless, it does not follow that the fact recorded is non-temporal in the sense of being sempiternal or everlasting (eternal1), or in the sense of being composed entirely of universals (eternal2). A temporal fact is timeless or eternal3 since although it is outside of time, time is contained in it. To recognize R-relations and R-facts as distinct categories of entities in one’s ontology is to recognize the R-theory as distinct from Broad’s (and McTaggart’s) interpretation of Russell, and it is, in the sense I have just explained to understand what is meant by the timelessness of time. Broad says, regarding the tenseless (hence tenseless) ideal language of the R-theoretic description of reality that “no one except a philosopher doing philosophy ever does talk in this way” (1938, p. 61). He rejects the claim that the A-theoretic ontology that he believes is suggested or implied by ordinary language may be a defect in the language that we speak (1938, p. 61). Indeed, he claims that the R-theoretical ontology

Conclusion  191 associated with a (tenseless) language, and thus a timeless ontology, does away with the transitory aspect of time, and so is static, and cannot account for change without succumbing to an infinite regress of time dimensions. Ironically, it seems to me that without the R-series, it is the A-theory that is static; and that without the A-series, it is the B-series—properly understood as the R-series, whose terms are ordered by Russellian temporal relations—that is dynamic. The claim that the A-theory without temporal relations cannot account for the dynamic aspect of time, and is thus static, is not as controversial as it might appear. A-theorists beginning with McTaggart, as I have shown in Chapter 5, and more recently Kit Fine, explicitly acknowledge that succession is essential to passage whether construed as the donning and doffing of A-properties, the successive actualization of possible worlds, the transition from one time to the another, or the coming into and going out of existence of events. Therefore, without succession (which is nothing other than an earlier temporal object being followed by a later one), the A-theory in its various embodiments is static and leaves out something essential to time. Fine argues that succession is essential to account for temporal passage since in order for there to be time and change, it is not enough that events or times have A-determinations or in particular the property of presentness, but they must successively have that property. Fine puts it as follows: The passage of time can be taken to consist in the successive possession of the absolute property of PRESENT or NOW. This property passes as it were from one moment to the next and it is in its passage . . . that the passage of time can be taken to consist. (Fine, 2006, p. 404) Fine understands passage in terms of the successive possession by moments of the property of presentness, and for that reason, the tensed fact that a particular time t0 has the property of presentness is alone insufficient to provide an adequate ground for the passage of time. He explicitly makes this point in the following passage: But although the realist possesses the right concept of the present in terms of which an explanation of the proposed sort might be given, he does not possess the right metaphysics by reference to which it might actually be sustained. For all he can properly say is that a particular time t0, which happens to be present, possesses the absolute property of being present. But what we wanted was the successive possession of the property of being present, not merely its current possession. (2006, p. 405; emphasis added)

192  Conclusion He continues: His [the tense realist] conception of temporal reality, for all that he has said, may be as static or block-like as the antirealist’s, the only difference lying in the fact that his block has a privileged “center”. (2006, pp. 405–406) Of course, the R-theorist would argue that temporal reality is not static or block-like at all since R-relations generate a dynamic temporal series. Of that, more again, later. In the rest of the paper, Fine (2006) develops two non-standard accounts of realism to account for the successive possession of presentness, but the point I want to emphasize here is that Fine clearly believes that tense realism alone is insufficient for temporal realism since tensed facts are insufficient to give us passage or distinguish space from time. To ground the dynamic aspect of time we must have an adequate account of succession, or the transition from earlier to later temporal objects. This point can also be supported by William Lane Craig’s account of temporal becoming. For Craig, temporal becoming is modeled on the different members of the A-series coming into existence successively, as successive times become present. He says, the doctrine of objective becoming . . . could be graphically displayed as the successive actualization of the history of the actual world. It is this model of a successively instantiated, rather than tenselessly existing, actual world that precludes the existence of a “totality of facts.” (Craig, 2000, p. 207; emphasis added) The appeal to succession implies the existence of temporal relations, and thus without temporal relations, Craig’s presentist world is static and not dynamic. Similarly, Yoval Dolev claims that “On the anti-reductive account, transience has to do with events’ becoming present after having been future and before becoming past” (Dolev, 2012, p. 70), and thus without the before/after relation(s) transience would not exist. What, then, is the proper metaphysical analysis of succession (or temporal relations) and the transitory or dynamic aspect of time? Broad argues that since transition is the essence of time and the foundation of our emotions, thoughts, feelings, experience and language of time, the B-series alone and the tenseless language philosophers use to characterize time cannot record the facts that ground the reality of time. For what is left when transition is omitted from time is a changeless, timeless static series of timeless abstract objects or persistent particulars. However, Broad’s Principle of Pickwickian Senses (see Chapter 1) provides a response to his objection. According to this principle, there are certain circumstances in which we can abandon the ordinary way of thinking and talking about things and adopt another that comports with the commonsense facts and

Conclusion  193 phenomena, but is otherwise preferable. That is, even if Broad is correct in maintaining that the natural and accurate way of expressing temporal relations between events involves using the tenses, it does not follow that the correct ontological analysis of those facts requires some version of the tensed or A-theory of time. There may be dialectical and/or phenomenological reasons to reject the commonsense account and maintain that the commonsense judgments about time are true, nevertheless. What is required is that the temporal phenomena can be adequately explained with an alternative ontological analysis. Thus, if we can provide another analysis that provides an adequate account of the temporal phenomena in question, then we can still preserve commonsense (assuming that is desirable), by grounding: (1) temporal passage or the transitory aspect of time, (2) identity through time and change, and (3) the direction of time and the difference between space and time. For Russellian temporal relations are given in experience as phenomenologically simple relations, and thus are taken as unanalyzable and irreducible, mind-independent entities in the ontology of time. In the following sections, I shall indicate the directions an R-theorist could take in accounting for these phenomena.

II.  The Passage of Time According to Broad, a consequence of Russell’s view is that “the theory leaves altogether out of account the transitory aspect of time” (1938, p. 61). It is a frequent criticism of the B-theory that it is a static unchanging totality. Hence, it leaves out what is essential to time. This is a common interpretation of the B-theory, fostered by proponents and critics alike, that the ontology of B-time is anti-realist because it denies that temporal becoming or passage is an objective, mind-independent feature of reality. The mind-dependence thesis is thought to be a natural consequence of the B-theory, which is characterized as a block, static universe. The implication of these appellations is that all events, past, present and future exist eternally in unchanging and fixed temporal relations. Such a world view seems to imply that the flow and flux of events in time—time’s dynamism—is an illusion that would not exist without consciousness, much like secondary qualities would not exist without consciousness. For these reasons, the mind-dependence of passage is thought to be the death knell of the B-theory. It is argued that in denying objective becoming or transition, the B-theory removes what is essential to time and what is left is really a block or series of blocks that is indistinguishable from a spatial object; a house of blocks, as it were. Furthermore, if passage was not an objective mind-independent feature of reality, then we could not explain our experience of the dynamic aspect of time. Barry Dainton puts this in terms of a challenge to the B-theory: How can our experience have the flowing, stream-like character that it does in a passage-free universe? How can the calm, eternal

194  Conclusion character of the Block universe be reconciled with the turbulent, dynamic character of our immediate experience? . . . [The failure to account for this is what gives rise] to the most deep-seated resistance to the Block conception, the conviction—shared by many—that it omits what is most distinctively timelike about time. (Dainton, 2011, p. 385, n 3) Dainton’s criticism of the B-theory dovetails with Broad’s criticism of Russell; that he leaves out the transitory aspect of time. I have argued, however, that their criticisms do not apply to the R-theory. For the Russellian, R-facts while they are not themselves in time, are indeed temporal since they contain temporal relations. R-facts are entities in their own right over and above their constituents, and as such, they are not in time in that they do not exemplify temporal relations. In that sense, time, understood as a Russellian series composed of a conjunction of R-facts, is timeless. It might be thought, however, that if the whole process (or conjunction of R-facts) is not in time, then there is something missing, namely, the transition from one time to another; or the sense that R-facts are dynamic and not static, and that time involves passage. But this strikes me as a mistake. Transition and passage do exist in the Russellian theory, and they are compatible with an R-theoretic ontology. That the “is earlier than” relation is dynamic is argued for by Tegtmeier in the following passage: 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 recognise 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. . . . 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 a dynamic relation. (Tegtmeier, 2010, p. 42)1 Thus, the notion that Russellian time is static and A-time is dynamic is rooted in confusion. McTaggart thinks that an order relation, such as temporal sequence, which satisfies the conditions of asymmetry and transitivity, is thereby not a dynamic relation—but that is a further consequence of confusing a Russellian temporal series, which is dynamic, with McTaggart’s C-series, which is static. In the R-theory, an appeal to the direct experience of succession in a single act of awareness is consistent with founding or grounding our most basic experience of the flow or passage of time on mind-independent temporal relations alone.

Conclusion  195 Although the R-theory understands and grounds the passage of time— the flow and transition from one event to another—by appealing to the mind-independent temporal relation of succession, there is a sense in which the passage of time is mind-dependent. For, the passage of time also refers to the phenomenon of temporal becoming or the “movement” of events from the future to the present and into the past, and though there is a close connection between these two temporal phenomena, there is an important difference between them. In the R-theory, temporal becoming is dependent on the perspective from which events are viewed and the successively changing psychological attitudes we have toward the same event, whereas the passage from earlier to later events is not so dependent. When I am consciously anticipating a later unpleasant event, say, an upcoming root canal, the event is future and I may think of it with dread. Later, when I am consciously perceiving or otherwise experience the (roughly simultaneous) unpleasant event, the event is present and I am experiencing it (with, perhaps, some distress). Finally, when I am consciously remembering the earlier event, I feel relieved that it is over or past. Thus, temporal becoming—events moving through time from the future to the present and into the past—is a change, but it is a change brought about by the succession of different psychological states in me. Since it is a change of mental states toward the same object, temporal becoming is mind-dependent, but it is founded on the mind-independent relation of succession. The passage of events or objects from the future to the present to the past is not a change in the temporal characteristics of the event in question, as the property view maintains, but a change in the psychological attitudes of single conscious mind toward the same event, each from a different temporal perspective. Richard Braithwaite nicely expresses this view as follows: This theory makes the change of the time of events consist of a change in the experiences of an observer and can in a sense be called a subjective change. What from one point of view is past may from another point of view be future. Change of the time of events can be analyzed into a succession of events which are experienced by one conscious mind. The specifically temporal element in each series of events is the relation of succession, which is directly experienced and has an intrinsic sense given in experience. (Braithwaite, 1928, p. 171) At this stage, it is reasonable to ask, what then is the “I” that has the successive experiences and constitutes “one conscious mind?” Is it a persisting substance or self, or a succession of temporal stages that are suitably related, or is there some third alternative? To answer that question requires an account of change consistent with the R-theory.

196  Conclusion I have pursued some of Broad’s reflections on this topic in Chapter 6, and I shall in the next section briefly summarize two answers consistent with the R-theory. First, it will be instructive to consider another standard objection to the R-theoretic account of the phenomenon of temporal becoming. In his reply to Braithwaite’s version of the R-theory, Broad claims that there is something missing from the view of time that analyzes the changing tense (that is, A-properties) of events in terms of unchanging relations between mental and physical events, namely, it fails to account for the fact that there is one experience that is special—the present experience that is alive and real. Broad states the criticism as follows: Granted that there is a certain series of mental events which can be called “my experiences,” then it seems to me that at any moment there is one of these which has a certain characteristic which some of them “have had but no longer have,” and which the rest of them “will have but have not yet had.” And I do not see how this could be accounted for, as I understand Mr. Braithwaite claims to do, by the unchanging relations between two series of unchanging events, one subjective and the other objective. (Broad, 1928, p. 188) Broad’s argument is that since in Braithwaite’s R-theoretic view there is nothing that picks out the present experience from those that are past or future, the R-theory cannot account for temporal becoming, or our experience of events moving from the future to the present and into the past. It seems to me that the R-theory can account for the presence of experience, and pick out the present experience, but before I explain how, I want to explore more carefully why Broad and others think it cannot. I suggest that their reasoning is as follows. In the R-theory, relations between events are unchanging because the copula is tenseless, from which it follows that neither the temporal relational fact nor the terms of the relation exemplify tensed properties. Since there is no nonrelational property or properties, like pastness, presentness and futurity that pick out the present experience from those that have been or are not yet present, it follows that the unchanging relation between events in such a series is non-temporal. Indeed, the notion that there could be a temporal series without an objective non-relational ground for past, present and future is not only mistaken, it is unintelligible. As Broad once put it, I am inclined to think that the notion of earlier and later is inextricably interwoven with the notions of becoming and of pastness, presentness, and futurity; so that the suggestion that there might be a series of terms to which the latter notions did not apply, but

Conclusion  197 which were nevertheless ordered by the relations of earlier and later, is really unintelligible. (Broad, 1937, pp. 242–243) Broad is arguing that since, in the R-theory, temporal becoming requires an unchanging temporal series of cognitive attitudes toward a single event, and since the terms of the R-series do not have A-properties or undergo A-theoretic becoming, the notion of a temporal series of cognitive attitudes is impossible and even unintelligible. Hence, neither the experience of the present nor temporal becoming have an ontological ground. Of course we see once again that this argument assumes a McTaggartian A-theoretic analysis of succession and so begs the question against the R-theory. Both Broad and McTaggart assume that without A-properties there cannot be temporal R-relations. Furthermore, the argument is invalid, for as I shall explain next, the R-theory can give an account of the presence of experience and temporal becoming even if the R-relations between events are unchanging, do not have A-properties and do not undergo A-theoretic becoming. Out of all the experiences that constitute my life, there is one I am experiencing as present, or is picked out as present, namely my (unreflective, non-positional) consciousness of this perception of familiar faces. No other earlier or later experience is my consciousness of perceiving these friends. There may be a later consciousness of remembering my perception of familiar faces and an earlier consciousness of anticipating my perceiving familiar faces, but there is no good reason to say that those earlier and later experiences, which are no longer (they do not exist simultaneously with this utterance or this perception, but earlier) or not yet (they do not exist simultaneously with this perception, but later), cannot be distinguished from the present experience consisting of my consciousness of this perception of old friends. I just did. I can be conscious of my memory of perceiving them, but when I am, I cannot be conscious of perceiving them since that perception is earlier than this memory, and for that reason, that experience of perceiving is past, and if I judge that the perception is earlier than this memory, I can experience it as over and think of the past event with nostalgia. To put my point slightly differently, I am conscious of this perception of familiar faces that succeeds my consciousness of anticipating perceiving familiar faces and is succeeded by my consciousness of remembering perceiving familiar faces. Since this perception is the only one I am conscious of, my consciousness of it (which is not distinct from it) is the ground of the presence of the perceptual experience of seeing familiar faces. In the R-theory, different cognitive acts with the same object are successively present in experience, for I am conscious of them in succession, and therefore, the

198  Conclusion presence of experience and the phenomena of temporal becoming can be accounted for.2 It might be objected that this perception is not the only one I am conscious of since I am also tenselessly conscious of my memory and tenselessly conscious of my anticipation of familiar faces. While that is true, why should that be a problem? To argue that if I am tenselessly conscious of my anticipation, perception and then memory of an object there is nothing that accounts for the series being temporal and therefore that there is nothing that picks out the present experience begs the question against the R-theorist. It is to assume that if events stand in tenseless relations (that is, relations whose terms do not have primitive tensed properties) then those events are like timeless abstract objects like numbers or sempiternal persistent particulars like regions of absolute space or points on a line, and so cannot occur in succession. That assumption, while amenable to A-theorists who treat temporal relations as internal relations grounded on A-properties and becoming, will be rejected by the R-theorist who treats temporal relations as simple and unanalyzable external relations whose terms occur in succession. I think, therefore, that McTaggart’s notion of the B-series should not be taken to be identical with Russell’s R-series. My proposal is that we treat the B-series and the A-series as ontologically neutral. For McTaggart, the B-series is founded upon the earlier/later than relation, but what ontologically speaking those relations consist in is another question entirely.

III.  Time and Change If one thinks of the debate between A- and B-theorists or presentists and eternalists, as Ingthorsson does, as whether or not one accepts the principle of temporal parity, then the question becomes do all times exist equally or co-exist? If they do, then “wherein lies the temporality of earlier than and later than?” (Ingthorsson, 2016, p. 141). Thus, Ingthorsson and McTaggart believe that “the conception of temporal reality as a series of events that are permanently related as earlier/later than each other—the B series—is a conception of a changeless reality and consequently not a conception of a temporal reality” (2016, p. 92; emphasis added). This is also Broad’s conception of how Russell must conceive of temporal reality. It is no surprise then, in his critique, Broad compares change in Russell’s view to mere spatial qualitative variation, “as different segments of a variously coloured string differ in colour” (1938, p. 308). Given this conception of the B-series, it follows that the events, particulars or experiences in the series do not exist in time, for we have a changeless non-temporal reality. Typically, B-theorists construe temporal relations as analyzable in terms of causal relations. There is, however, an argument that Broad gives, indirectly, for not understanding temporal relations in terms of

Conclusion  199 causal relations. His point in the following passage in the context of identity is that there can be succession without causal relations. We can think of the world as a chaotic succession of events. As Broad puts it in the following passage: In ordinary life, we distinguish between an object and its history, and we are inclined to think that the former is logically prior to the latter. . . . I believe this to be a profound mistake, which arise from taking “history” in too narrow a sense. An object, separated from its history, is clearly not the kind of thing that could possibly exist. . . . Now it is conceivable that there might have been succession but no history. If so, there might have been neither an object nor a plurality of objects. (Broad, 1952, p. 407) Broad’s point here is that you can have succession without causality. When causality is introduced together with resemblance you then have an object, but before that there is just “a chaotic hail of events” (Broad, 1952, p. 407). Hence, there can be succession without causation. Therefore, causal relations are conceptually distinct from temporal relations and can exist in nature without them. Broad goes on to consider other possibilities including where we have distinct strands of history that constitute the history of a single object, But its history is more than the sum total of a number of distinct strings lying side by side. If there be causal and other regularities which hold throughout the whole period under discussion, there will be characteristic relations between the strands, and the history of the world as a whole would have more unity and complexity than is implied by the simple statement that it is composed of such and such parallel strands. (Broad, 1952, pp. 407–408) For Broad, a single object is a four-dimensional strand of history. . . . The question whether it is a changing or an unchanging object is simply the question whether successive slices of the strand, normal to the time-dimension, are exactly alike or progressively different in quality. (1952, p. 409) Tegtmeier refers to this as a “weak serial view” since the object is nothing over and above the successive strands appropriately related. The standard objection to this view is that if change is a difference in the properties of an object that remains the same through change, then

200  Conclusion the weak serial view is inadequate, for there is no object that has a property and then loses it. If Socrates is sitting at one time and standing at another time, then if literally the same individual Socrates exists equally at both times and everything exists in parity, then Socrates must be construed as a series of temporal parts, each with different properties, and in that case, Socrates doesn’t change. As Ingthorsson puts it: Ergo, if David Lewis sitting down is qualitatively discernible from David Lewis standing up, then sitting Lewis is numerically discernible from standing Lewis. The only way we can construe David Lewis as an entity that exists at many times having at those times different and incompatible properties is to think of him as a temporally extended compound object whose different parts exist at different times and have incompatible properties, ergo, David Lewis does not change but only varies qualitatively between his parts. (2016, p. 97) The objection, then, is that if a thing consists of temporal parts, then such a thing cannot change, since temporal parts do not change. But there is a view of time and change that accepts a persisting, changing object that consists of temporal parts. It is the strong serial view, where the series of parts is a further entity. In this view, a numerically identical persisting thing has temporal parts, but it is an individual thing—a further entity over and above or in addition to its temporal parts, and it can account for change. To see how this is so, note that the question: “What does it mean to say that an individual object O1 at tl that is standing, is the same individual object as O2 at t2 that is sitting?” is ambiguous. For the phrase “the individual O1 at t1’’ may mean “the temporal part O-at-tl” (more simply, “o1”). Or, it may also mean “the entity O, of which O-at-t1 (o1) is a part.” Clearly, if we mean the former then it is misleading, indeed false, to assert that O-at- t1 (o1) is the same individual as O-at-t2 (o2), but this does not imply that the individual of which these two different stages are temporal parts is not the same at these different times. For, in the second interpretation we can say that David Lewis, who at an earlier stage in his life (o1) is standing, is the very same David Lewis that at a later stage (o2) is sitting. Thus, the existence of temporal parts is compatible with treating individuals (including persons) as persisting entities which in the strictest sense persist through change. See, Reinhardt Grossmann (1984, pp. 93–101, 1980, pp. 464–465, 1974, pp. 92–93). Erwin Tegtmeier (2018, 2008, 2007) distinguishes the weak versus strong versions of the serial view. Broad does this, too, when he considers the difference between McTaggart’s view of the self from Hume’s view of the self (Broad, 1938, pp. 242–243). Tegtmeier’s view of change differs from both the strong and weak serial view. In his view, a substance has

Conclusion  201 temporal parts, but they are not parts of the substance that has them. They are particulars that stand in the “being a temporal part of” relation to that substance, but that is not the part/whole relation to the simple (has no constituents) substance that has them. Temporal parts have properties, but they do not change. In Tegtmeier’s view, the change of an apple from green to red is analyzed in terms of one temporal part being green, and another temporal part being red; and the first temporal part is earlier than the second, and both parts stand in the being a temporal part relation to the same simple substance. These facts are the ground of change, but they do not themselves change, nor do these facts exist in time, although time (i.e., the earlier than and being a temporal part of) relations exist in them. A substance that has passed away is grounded its having a last temporal part, its having emerged is based on having a first temporal part.

V. The Direction of Time and the Difference Between Space and Time Relations What, then, is the R-series? There are difficulties with the terms “B-series” and “B-theory” because they originate with McTaggart. Unfortunately, his notion of the B-series is ambiguous in a way that has led commentators to either misunderstand McTaggart on time, or prejudge the issue against the Russellian ontology of time. In the standard interpretation of McTaggart, there is a B-series, whose terms are related by B-relations; there is an A-series (or series of such), whose terms exemplify the monadic temporal properties of pastness, presentness and futurity; and there is temporal becoming, the acquiring and shedding of A-properties by the terms in the B-series. Thus, it is typical for interpreters of McTaggart to attribute the view that there are A-properties “in addition to the B-series and its unchanging relations” (Savitt, 2001, p. 261) to him. This view is called the “A/B theory of time.” It is the standard interpretation of McTaggart’s positive conception of time, but if we assume the “unchanging relation” is an unanalyzable temporal relation, it is not McTaggart’s. There is another interpretation of the B-series and therefore of the B-theory, which is McTaggart’s, but is not Russellian. According to this view, the B-series is analyzable (reducible) in terms of the application of the A-series to the non-temporal, but ordered C-series, whose generating relation is (for McTaggart) included in.3 In this view, the direction of time and its transitory or dynamic aspect is grounded in A-properties moving along a non-temporal C-series. According to McTaggart, the terms of the C-series have an intrinsic order, but do not have an intrinsic direction, or as Broad call it, an “intrinsic sense.” By an intrinsic direction or sense is meant the difference between A B C D and D C B A. In our experience of time and change, events or things in time occur in a certain direction since A is earlier than B is earlier than C is earlier than D rather than the

202  Conclusion other way around. For McTaggart, it is the application of the A-series to the C-series that generates a B-series with a direction or sense. For that reason, he says, The B series, on the other hand, is not ultimate. For given a C-series of permanent relations of terms, which is not in itself temporal and therefore is not a B-series, and given the further fact that the terms of this C-series also form an A-series, . . . it results that the terms of the C-series become a B-series, those which are placed first, in the direction from past to future, being earlier than those whose places are farther in the direction of the future. (1908, p. 26) Clearly, the resulting B-series is not Russellian. In the Russellian view, temporal R-relations are not to be identified with causal, spatial, entropic or any other kind of relations. Thus the R-theory differs from B-theorists who analyze the direction of time from earlier to later in terms of the direction of causality (Mellor, 2009; Le Poidevin, 1991, 2007). Moreover, in calling temporal relations unanalyzable, a Russellian means that they cannot be reduced to the properties of their terms and, indeed, the terms of temporal relations have no intrinsically temporal properties such as pastness, presentness and futurity, since there are none. R-relations are not analyzable, as in McTaggart, in the terms of a non-temporal series having temporal properties. R-theorists reject the moving now and the monadic property of presentness, but nevertheless maintain that genuine succession exists and that a conjunction R-facts that constitutes the R-series and time itself, contains change. R-theorists maintain that temporal relations are primitive and unanalyzable relations, and the difference between temporal and all other relations is an irreducible qualitative difference. From this perspective, it is a mistake to suppose that if time is the mere succession of events, then the change involved is exactly like the spatial “change” in the color of the lawn one observes as one walks from the front to the back. In this view, the relation that distinguishes temporal order is just different from any spatial relation in the same sense that red and green are just different. In the R-theory, the only category of intrinsically temporal entities are relations—there are no temporal individuals, such as moments or time points; there are no monadic temporal A-properties; and there is no absolute becoming understood either as the coming into and going out of existence of objects or events, or as the donning and doffing of A-properties. In other words, in the R-theory, time is relational, that is, all ontological facts about time are understood as grounded in relations and that include durations such as lasts as long as or lasts longer than. Russell’s account of the direction of the R-series also differs from McTaggart’s. According to Russell’s view in the Principles of Mathematics,

Conclusion  203 when a relation relates two individuals there is an order in that connection, in that it does so with a sense: the relation goes from one relatum to the other. As Russell puts it: it is characteristic of a relation of two terms that it proceeds, so to speak, from one to the other. This is what may be called the sense of the relation and is . . . the source of order and series. It must be held as an axiom that aRb implies and is implied by a relational proposition bR’a, In which the relation R’ proceeds from b to a and may or may not be the same relation as R. . . . The sense of a relation is a fundamental notion, which is not capable of definition. (1938, pp. 95–96) For Russell, order must be taken as primitive. What we need is an order between a and b, as related by R. Russell’s account secures this order by building it into the relation itself; relations, in Russell’s account in PM, invariably have a sense. An asymmetrical relation generates a series because it provides a structure to each fact, since each fact has an order—an intrinsic order. Russell argued that there is a difference between the two facts aRb and bRa, and if the relation is asymmetrical, then only one of those facts obtains. It is a synthetic a priori truth that if a relation is asymmetrical the converse of the relation cannot obtain. That is, for Russell, an asymmetrical two-place relation always holds in a definite direction between its terms from either a to b or from b to a.4 In Theory of Knowledge, the 1913 Manuscript (Russell, 1984), he gives a different analysis of order. According to it, Russell assumes in his ontological analysis of order in relational facts (e.g., a is earlier than b) order relations such as “being the first relatum” and “being the second relatum.” These relations hold between relata and facts. Since all series have a direction, Russell differs from McTaggart in both his account of the transitory aspect of time and of its direction. For McTaggart, what gives time a direction and its transitory character are changing A-characteristics. A Russellian will ground the transitory or dynamic aspect of time in the relation “is earlier than,” and the direction of a whole series is aggregated from the order relations for all the relational facts contained in it. I shall make no attempt to defend either of Russell’s accounts of temporal order against criticisms that have been lodged against them (although I am sympathetic with the first).5 My intention, rather, is to indicate that these accounts (and others like them) differ not only from all A-theoretic accounts, but also from the standard B-theoretic causal accounts of the direction of time as well. To sum up, the B-series can be understood, as Russell understood it, as involving unanalyzable temporal relations between its terms which are thereby temporal, or, as McTaggart understood it, as involving a

204  Conclusion non-temporal relation between timeless terms in a C-series plus something more—temporal properties and temporal becoming. These two notions of the B-series are not clearly recognized or distinguished in the literature, and a sliding from one to the other is prevalent in discussions of McTaggart and the “B-theory” with often unfortunate results both for the interpretation of McTaggart’s paradox and for the proper interpretation of the Russellian theory of time. Broad began the Introduction to Scientific Thought by stating what he considered to be the strongest objections to philosophy. One of those he considered was that there is no progress in philosophy, to which he replied, “First, as to the alleged unprogressive character of Philosophy. This is, I think, an illusion; but it is a very natural one” (1952, p. 13). Of course I agree with Broad and naturally hope that my discussions of Broad’s changing philosophy of time, and the R-theory of time, have contributed to progress in the scholarship of both areas.

Notes . In this connection, see Oaklander and White (2007). 1 2. See Oaklander (2004) for further discussion of the presence of experience. In Orilia and Oaklander (2015), we reply to the familiar objection, raised by Broad and others, that the appeal to “this utterance” or “this perception” just means the utterance tokened now, or the perception, which is present, and so does not eliminate A-properties from the analysis of tense. 3. This is not quite right since without temporal becoming the terms of the C-series cannot form a B-series. Of course, with temporal becoming, i.e., the acquiring and shedding of A-properties by the terms of the C-series, there cannot be a B-series either, which is McTaggart’s paradox. 4. For a defense of this analysis of order, see Fred Wilson (2013). 5. For criticisms of Russell’s accounts and alternative accounts of the direction of relations see, for example, Orilia (2008, 2011), Tegtmeier (2016), Bergmann (1992) and Hochberg (1987, 2001).

References Bergmann, G. (1992). New foundations of ontology (W. Heald, Ed., with a foreword by E. B. Allaire). Madison, WI: University of Wisconsin Press. Braithwaite, R. B. (1928). Symposium: time and change. Aristotelian Society Supplementary Volume, 8 (1), 162–174. Broad, C. D. (1921). Time. In J. Hastings et al. (Eds.), Encyclopedia of religion and ethics (Vol. 12, pp. 334–345). New York: Scribner’s. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 143– 173). References are to this reprint. ———. (1925). Mind and its place in nature. London: Routledge & Kegan Paul. ———. (1928). Symposium: time and change. Aristotelian Society Supplementary Volume, 8 (1), 175–188. ———. (1937). The philosophical implications of precognition. Aristotelian Society Supplement, 16, 229–245. ———. (1952). Scientific thought. London: Routledge & Kegan Paul (Originally published in 1923).

Conclusion  205 ———. (1937). The philosophical implications of precognition. Aristotelian Society Supplement, 16, 229–245. ———. (1938). Examination of McTaggart’s philosophy: ostensible temporality (Vol. II (1), Chap. XXXV, pp. 265–323). Cambridge: Cambridge University Press. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 36–68). References are to this reprint. Craig, W. L. (2000). The tenseless theory of time: a critical evaluation. Dordrecht: Kluwer Academic Publishers. Dainton, B. (2011). Time, passage and immediate experience. In C. Callender (Ed.), The Oxford handbook of philosophy of time (pp. 382–419). Oxford: Oxford University Press. Dolev, Y. (2012). Perceiving transience. In A. Bardon (Ed.), The future of the philosophy of time (pp. 56–72). New York: Routledge. Fine, K. (2006). The reality of time. Synthese, 150, 399–414. Gale, R. (1967). The philosophy of time. New York: Anchor Books. Grossmann, R. (1974). Bergmann’s ontology and the principle of acquaintance. In M. S. Gram & E. D. Klemke (Eds.), The ontological turn, studies in the philosophy of Gustav Bergmann (pp. 89–113). Iowa City: University of Iowa Press. ———. (1980). Reviewed work: Person and object: a metaphysical study. By R. M. Chisholm. Noûs, 14 (3), 457–467. doi:10.2307/2214969 ———. (1984). The categorical structure of the world. Bloomington: Indiana University Press. Hochberg, H. (1987). Russell’s analysis of relational predication and the asymmetry of the predication relation. Philosophia, 17, 439–459. ———. (2001). The positivist and the ontologist: Bergmann, Carnap and realism. Amsterdam: Rodopi. Ingthorsson, R. D. (2016). McTaggart’s paradox. London: Routledge. Le Poidevin, R. (1991). Change cause and contradiction: a defence of the tenseless theory of time. New York: St. Martin’s Press. ———. (2007). The images of time: an essay on temporal representation. Oxford: Oxford University Press. McTaggart, J. M. E. (1908, Oct.). The unreality of time. Mind, 17 (68), 457–474. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 21–35). References to this reprint. Mellor, D. H. (2009). Real time II. London: Routledge (Originally published in 1998). ———. (2004). Craig on the experience of tense. In L. N. Oaklander (Ed.), The ontology of time (pp. 235–242). Amherst, NY: Prometheus Books. ———. (Ed.). (2008). The philosophy of time: critical concepts in philosophy (Vols. 1–4). New York and London: Routledge. ———. (2015). Temporal phenomena, ontology and the R-theory. Metaphysica, 16 (2), 253–269. Oaklander, L. N. & White, V. A. (2007, Oct.). B-time: A reply to Tallant. Analysis, 67 (4), 332–340. https://doi.org/10.1093/analys/67.4.332 Orilia, F. (2008). The problem of order in relational states of affairs: a Leibnizian view. In G. Bonino & R. Egidi (Eds.), Fostering the ontological turn: essays on Gustav Bergmann (pp. 161–186). Frankfurt: Ontos Verlag. ———. (2011, May). Relational order and onto-thematic roles. Metaphysica, 12 (1), 1–18. https://doi.org/10.1007/s12133-010-0072-0

206  Conclusion Orilia, F. & Oaklander, L. N. (2015). Do we really need a new B-theory of time? Topoi, 34 (1), 157–170. https://doi.org/10.1007/s11245-013-9179-6 Russell, B. (1915, Apr.). On the experience of time. Monist, 25 (2), 212–233. https://doi.org/10.5840/monist191525217. Reprinted in L. N. Oaklander (Ed.). (2008). Vol. I. The reality and language of time (pp. 174–187). ———. (1938). The principles of mathematics. New York: Norton (Originally published in 1903). ———. (1984). Theory of knowledge: the 1913 Manuscript in E. R. Eames & K. Blackwell. (Eds.). The collected works of Bertrand Russell, vol. 7. Boston: George Allen & Unwin. Savitt, S. (2002). On absolute becoming and the myth of passage. Royal Institute of Philosophy Supplement, 50, 153–167. doi:10.1017/S1358246100010559. Reprinted in C. Callender (Ed.). Time, reality and experience. Cambridge: Cambridge University Press. ———. (2007, Apr.). Three flawed distinctions in the philosophy of time. Metaphysica, 8 (1), 53–59. https://doi.org/10.1007/s12133-007-0005-8 ———. (2008). Persistence. In C. Kanzian (Ed.), Persistence (pp. 185–195). Frankfurt: Ontos Verlag. ———. (2010). The advancement of the world. Chronos, XII, 39–50. ———. (2016). Time and order. Manuscrito, 39 (4), 157–168. ———. (2018). Identity, change, and criteria of identity. In V. Buonomo (Ed.), The persistence of persons (pp. 79–90). Neunkirchen-Seelscheid: Editiones Scholasticae. Wilson, F. (2013). Burgersdijck, Coleridge, Bradley, Russell, Bergmann, Hochberg: six philosophers on the ontology of relations. In Acquaintance, ontology, and knowledge: collected essays in ontology (pp. 275–328). Berlin: De Gruyter (Originally published in 2007). The quoted passages occur on pp. 320–321.

Appendix Is There a Difference Between Absolute and Relative Space? Between Absolute and Relative SpaceBetween Absolute and Relative Space

L. Nathan Oaklander

One issue that Gustav Bergmann discusses in his article “Synthetic A Priori” (1964) is the ontology of space. He presents his answer to the question “What kinds of spatial entities are there?” by distinguishing three answers to the question that could plausibly be called “absolutist,” and argues that his view is nonabsolute (or relative) with respect to each. For Bergmann, there is a close connection between the ontology of space and the phenomenology of space. What we know to be true about space, what needs an ontological ground is based on how space is presented to us. Conversely, according to the Principle of Acquaintance, the simple entities of one’s ontology must be objects of acquaintance that are presented to us. For that reason, Bergmann worries about the questioner and critic who asks him to direct his attention to the entity “in” say, a red, round spot, which is the bare particular. To answer that supposedly “unanswerable” question in a way that allows Bergmann to preserve his relativism is one task he sets for himself in “Synthetic A Priori.” I shall argue, however, that Bergmann is not successful in accomplishing that task since his phenomenology of particulars renders his ontology of space “absolute” in at least one sense of that ambiguous term.1 To see why, I shall begin by considering Bergmann’s three views of absolute and relative space. The first view of absolute space that Bergmann discusses is the socalled container view. He characterizes it as follows: If all “things” now ceased to exist, would space be left? The view of those who answer negatively is relative1. Those answering affirmatively hold the absolute1, or, as it is also called, the container view . . . The absolutist is committed to holding that at least some spatial entities are neither things “in” the spots nor relations among such things. (p. 285) The absolutist1 holds that regions of space—what Bergmann calls “spacethings” and others call “places”—the spatial relations between them and the shapes they exemplify are spatial entities. The cornerstone of this

208  Between Absolute and Relative Space view, as Bergmann characterizes it, is that at least one kind of spatial entity is not a constituent “in” the spots, or a relation among such things. Thus, one would be an absolutist1 if one held that places directly stand in spatial relations and that ordinary things only indirectly stand in spatial relations in virtue of occupying a place. In Bergmann’s view, the only spatial things are relations and properties. The relata of spatial relations are bare particulars and they are non-spatial. Since relations obtain directly among things and the various spatial properties are “in” the spots (that is, are constituents of them), Bergmann’s view is relativistic1. Although Bergmann “dismiss[es] all kinds of absolutism1 out of hand” (p. 286), he does, in fact, have various reasons for doing so. For one, we are not acquainted with space-things or places. For another, they are dialectically dispensable since the problem of individuation can be solved without them (with bare particulars). Furthermore, they are ontological lavish, violating Occam’s razor by unnecessarily positing a relation of occupancy as the ontological ground of a spot’s being at a place. For Bergmann, the spatial relations between the particular “in” the spot and other particulars are sufficient to ground a spot being at a place. For these reasons, he dismisses absolutism1 and the alternatives to which he turns are all relativistic1. Since the second version of the absolute/relative space distinction is the main focus of my presentation, let me briefly state the third version before I turn to a detailed discussion of the second. According to the third version of absolute space, places construed as individuals existing “in” or “outside” of each spot, are rejected and in their stead are peculiar spatial properties that Bergmann calls “coordinate qualities”—hereness, thereness and so on are introduced. If you maintain that there are such properties, then you are an absolutist3; if you deny their existence, then you are a relativist3. Bergmann is a relativist3 since he says, “There are no coordinate qualities. We are neither presented with them nor dialectically forced to ‘postulate’ them” (p. 287). In a Scotist world, all simple things are characters, including non-relational spatial and temporal characters, or coordinate qualities, that are introduced to individuate two spots with the same ordinary non-relational properties. Bergmann’s rejection of Scotism (that is, a gamma ontology), in favor of an ontology that recognizes a categorical difference between universals and particulars is part and parcel of his rejection of coordinate qualities. Bergmann characterizes the second view of absolute space as follows: In a nonscotist world there is an individual “in” each spot. Is this individual a spatial thing? Depending on whether your answer is affirmative or negative, you are an absolutist2 or a relativist2. . . . Relativism2 is the view that all simple spatial things are either properties or relations. (pp. 286, 287)

Between Absolute and Relative Space  209 Bergmann says that he is a relativist2. That can only mean that the individual in each spot, i.e., the bare particular, is a non-spatial thing. But what feature or features do spatial individuals (hereafter called “places”) possess that particulars lack? That places have some feature or features that set them apart from particulars is obvious, for otherwise we could not make sense out of the distinction between absolute2 and relative2 space. But does the distinction make sense, and is the debate between absolute2 and relative2 space really that important? One might argue that the absolute2/relative2 controversy is spurious because particulars and places are both spatial things. After all, particulars and places are both in space, since they both exemplify spatial relations, such as being at a certain distance from other particulars or places, and they both have spatial characters, such as having a certain shape. However, that would be a bad argument for rejecting the distinction. The issue separating absolutism2 and relativism2 concerns not what properties and relations the individual “in” a spot has, but concerns what the individual is in itself. To answer the ontological question “What is space?” is to give an inventory of all spatial entities, or rather, of all the kinds of such entities there are. Thus, the issue separating absolutism2 and relativism2 concerns whether or not there is a kind of individual that is spatial, which is not the question of whether the individuals in the spots are also in space, since they are, or whether they exemplify spatial properties, since they do, but whether or not the individuals “in” the spots are intrinsically spatial, or by their very nature spatial. Presumably, then, Bergmann is a relativist2 because in his ontology, the individuals in the spots are bare; they have no nature and a fortiori are not intrinsically spatial. The controversy between absolute2 and relative2 space is very important since one’s views on the ontology of space have implications for the ontological status of relations (are they internal or external?), fundamental ties or nexus (are they homogenous or inhomogeneous?) and the choice between substance and fact, and gamma and epsilon ontologies. If the individuals “in” the spots are spatial things, then they are natured (perfect) particulars (in contemporary parlance “tropes”) or substances of classic ontologies and that would be a disaster, undermining Bergmann’s ontological system completely. Furthermore, if the individual “in” the spot is intrinsically spatial then it can only be exemplifying nonmental properties and therefore, Bergmann’s view that particulars, being bare, can exemplify either non-mental properties such as being round and being red, or mental properties such as the species characters being a remembering and being a perceiving, and propositional characters such as “that the cat is on the mat” cannot be sustained. Thus, upholding the distinction between absolutism2 and relativism2, and the justifying claim that his view is relative2 (because bare particulars are nonspatial), is crucial to Bergmann’s ontology and is therefore, important indeed.

210  Between Absolute and Relative Space If, however, we are not presented with bare particulars then his relativistic2 ontology of space does not have a phenomenological ground. Since an adequate ontology must have a phenomenological ground, the critic’s requirement to “show me the particular” is one that Bergmann takes seriously. He makes the connection between the ontology of space and the phenomenology of particulars clear when he says: In my world there are neither space-things nor coordinate qualities. That makes my view relative1 as well as relative3. Whether or not I am also a relativist2 depends on whether or not the individuals which exemplify shapes are themselves spatial. That takes us back to the bare particular and the question which supposedly is unanswerable. (p. 288) Bergmann formulates the supposedly unanswerable question in the following passage that I will quote at length: Suppose someone asks me what c is? I strike the right key, strike some others, strike the first again, tell him that c is what has been presented to him on the first and last occasion but on none of the others. . . . In my ontology, what is presented on each of these two occasions is a fact, namely, a particular exemplifying a pitch, loudness, and so on. The pitch is one; the particulars are two. Suppose now that the questioner asks me to direct his attention to the particular in the way I just directed it to a pitch. Particulars, or, at least, this sort of particular being momentary, they cannot be presented twice. The questioner appreciates the point but insists that what he was in fact presented with on each of the two occasions is a pitch, a loudness, perhaps some other qualities, and nothing else. (That shows the appeal of Scotism!) Thus he keeps asking me what a bare particular is, demanding that his attention be directed to one. This is the question the critics of D2 [the doctrine that all relations are external] hold to be unanswerable. So far, the defenders have not known how to answer it. Eventually I shall propose an answer. (p. 278) If Bergmann cannot answer the critic’s demand to direct her attention to the bare particular “in” the spot, then either there is no individual “in” the spot or, assuming a nonscotist world, the individual in the spot would be spatial and space would be absolute2 with either alternative having disastrous ontological consequences. On the other hand, if we are acquainted with bare particulars, then the “unanswerable” question can be answered, the individual in the spot is non-spatial, and his relativism2 is preserved. What, then, is Bergmann’s response to the unanswerable

Between Absolute and Relative Space  211 question and is it phenomenologically adequate to satisfy the critic and ontologically adequate to insure his relativism2? Bergmann answers the allegedly unanswerable question by saying: Remember the questioner who, when presented with middle c, insisted that all the entities presented to him were properties. Suppose he gives me another chance, asks me to direct his attention to the bare particular “in” the spot. I first acquaint him with my use of ‘shape’, then tell him that the bare particular is the spot’s area. (p. 288; emphasis added) Although we are acquainted with the area of the spot, the critic could reply that since a certain area is a character “in” the spot, and for that reason the entity to which Bergmann directed our attention is this character, and not a bare particular. If this is true, as phenomenologically it may appear to be, then either there is no bare particular “in” the spot—the spot is simply a collection of characters—or there is a bare particular “in” the spot, but it is not presented to us, only its properties are. Bergmann is sensitive to this objection and has the following reply: Assume that you are presented with two spots. If they agree in all non-relational characters, including shape, they will also agree in the character you claim the entity is. How then would you know that they are two and not one? The questioner has no answer. (p. 289; emphasis added) Bergmann’s point is that the area a thing has, or its being spatially extended, is not a character of it, since if it were, then we could not know upon being presented with two spots that they were two and not one. However, this appeal to epistemological considerations (for example, how would you know there are two spots and not one) is surprising. For the issue is not how we know they are two (since the perceived distance between the two spots is sufficient for that), but rather how would the critic provide an ontological ground for their being two (since they have all their non-relational characters in common)? Bergmann’s answer is that to individuate the two spots the area must be a particular—a mere individuator—and not a character. And he clearly does identify the particular with its area when he says, perhaps infelicitously, that “The spot’s area is not only round, it also is red. I take it, then, that the bare particular ‘in’ the spot is its area” (p. 290; emphasis added). By treating the bare particular as being identical with the area of the spot, Bergmann can claim that when he sees the two spots, he is directly acquainted with different bare particulars when he is acquainted with different areas, but with that response he goes, I submit, from an unanswerable question to an unacceptable answer.

212  Between Absolute and Relative Space For Bergmann is now faced with the following dilemma. If the individual in the spot is (identical to) an area or, as he says later in the article, its (spatial) extension, then the bare particular can be presented to us, but it is no longer a mere “this,” a mere individuator since there is something about it, some feature in itself, in virtue of which it is identifiable or recognizable. It is, in other words a natured particular; not a bare this, but a “this particular area” or this particular (spatial) extension. But then, there is no basis for distinguishing the particular in the spot from an absolute2 place. Since an area or a spatial extension seems to be a simple spatial thing, if it is identical with the “bare” particular in the spot, then the resulting particular is also a spatial thing. It has (or is) a nature, identifiable and recognizable as different from all other entities of the same ontological kind. On the other hand, if Bergmann maintains that the area or spatial extension of a particular is a character external to it, and so is not grounded in the particular itself (since only then the particular is truly bare), then the critic’s unanswerable question is indeed unanswerable. In other words, if being presented with the spot’s area or spatial extension does not acquaint us with the particular itself, or the thing that has an area or is (spatially) extended, then bare particulars lack a phenomenological ground. Thus, if a bare particular is an area or spatial extension, then his ontology of space is absolute2; and if the bare particular is not identical with its area but exemplifies it, then his phenomenology of particulars is inadequate. The dilemma Bergmann is facing arises again when he tries to refute a reason why particulars may seem to be spatial things. He says: Call a word a space word if and only if, when used phenomenologically, it represents a simple spatial thing. . . . “Area” obviously is a predicate. Hence, if it were a space word, area would be a character. . . . If one holds that it is [a character] of the first type then I turn the tables on him, ask him to direct my attention to the individual that exemplifies the spot’s area. This question is unanswerable. Your only way out is to become a Scotist. Then you will need coordinate qualities. (pp. 290–291) I do not see how this argument supports Bergmann’s case, since it begs the question of whether or not we are acquainted with particulars, and thus whether or not space is relative2. The question at issue is this: how can we be aware of the individual that exemplifies the spot’s area, if the spot’s area is included among its characters and so we are only aware of characters? But isn’t that precisely the question that Bergmann needs to answer? To answer it by saying that “this question is unanswerable” is unavailing. Alternatively, Bergmann seems to be arguing that the word “area” does not represent a simple spatial character of the first type

Between Absolute and Relative Space  213 because if it did, we would not be acquainted with the individual that exemplifies it, that is, the bare particular. But we are aware of the bare particular (since a Scotist ontology is false), and therefore “area” is not a space word. Unfortunately, I doubt that response will impress the critic of the phenomenology of bare particulars, for a dialectical argument against Scotism—a gamma ontology—is not tantamount to a phenomenological argument for particulars. In any case, Bergmann must still face the problem I raised a moment ago. If the words “area” and “spatial extension” do not name a character, what they do name is a single simple individual thing “in” the spot, but then I can see no basis for distinguishing it from a single, simple spatial thing, and thus the individual “in” the spot it is not a bare (nonnatured) particular, but an absolute2 place. A similar problem arises with regard to acquaintance with awarenesses, or bare particulars that exemplify species and propositional characters. To have a shape is to be a (spatial) extension. To have duration is to be a (temporal) extension. Thus, in directing one’s attention to the bare particular in a mental act, our attention is directed to its (temporal) extension. But again, the particular extendedness “in” an act of awareness— it’s being temporally extended—is a character exemplified by the particular “in” the act or it is not. If it is a character exemplified by a particular, then being presented with it does not answer the question “Where is the particular?” It does not provide a phenomenological ground for the bare particular which is “in” a mental act. On the other hand, if the temporal extendedness of a particular is grounded in what the particular is in itself; if the particular is identical to its being temporally extended or its temporal extension, then the particular would appear to be a simple temporal thing or a moment of absolute2 time, which, of course, is anathema to Bergmann. What, then, is to be done? Bergmann does give himself a way out, although in “Synthetic A Priori” he didn’t realize he would need to use it. Immediately after raising the critic’s allegedly unanswerable question, he says parenthetically: (Let it be said once and for all that even if the question were unanswerable, it would not follow that there are no bare particulars. Should they turn out to be dialectically indispensable, an argument could be made for ‘postulating’ their existence. The proper place for such postulation, though, is in science and in science only. Thus it is much, much better not to have to make that argument.) (p. 278) Thus, the way out of the dilemma I have posed is to abandon the principle of acquaintance and his preferred phenomenology of bare particulars, and “postulate” their existence.

214  Between Absolute and Relative Space It would appear, therefore, that Bergmann’s phenomenology of particulars and ontology of space are an unhappy fit. On the one hand, to preserve relativism2 he must deny that bare particulars have or require a phenomenological ground. On the other hand, to require a phenomenological ground for bare particulars he must accept absolutism2. Thus, in response to the question of this paper, “Is there a difference between absolute and relative space?” I would say that given Bergmann’s phenomenology of particulars and his ontology of space in the second sense of the absolute/relative space controversy, the answer is “no.”

Note 1. It is a platitude to claim that Bergmann’s views have steadily and in some ways radically changed over the years but so far as I can tell, his views on space and time have remained basically unchanged throughout. In New Foundations of Ontology, Bergmann mentions that his current views on time can be found in Appendix C, but since it is not there, I assume he never wrote it, and my point holds. See Gustav Bergmann (1992, pp. 209, 221). And since he says nothing new about space in New Foundations we can assume that his views there remain the same as those expressed in earlier works.

References Bergmann, G. (1964). Synthetic a priori. In G. Bergmann (Ed.), Logic and reality (pp. 272–301). Madison, WI: University of Wisconsin Press. Unless otherwise noted, all page references in the text will be to this paper. Reprinted in E. Tegtmeier (Ed.). (2003). Vol. II. Gustav Bergmann: collected works, Selected Papers II (pp. 225–254). Frankfurt: Ontos Verlag. ———. (1992). New foundations of ontology. Madison, WI: University of Wisconsin Press.

Index

absolute becoming 2, 9, 30, 37, 42 – 43, 46, 54, 69 – 72, 79, 86 – 87, 92 – 94, 100, 106, 110, 115 – 117, 137, 184, 186, 202; see also A-theory; temporal becoming absolute space 87, 145, 154 – 156, 162, 198, 207 – 208; container view 154 – 155, 207 abstracta 145 – 148, 151 – 153, 160 – 162; see also existents A-change 67 – 69, 102, 122, 141 A-characteristic 44, 51, 60, 65, 67 – 70, 86, 90, 95, 96, 102 – 104, 110 – 112, 117, 141 – 142, 203 A-determinations 11, 44, 54, 61, 64, 66, 67, 86, 92, 95, 100, 102, 109, 117, 132, 191 A-properties 2, 4 – 6, 10, 12, 28 – 30, 32, 35 – 36, 44 – 45, 47, 51, 56, 59 – 63, 66 – 68, 79, 86 – 87, 91 – 92, 94 – 97, 99 – 102, 104 – 112, 115 – 116, 129, 131 – 133, 135 – 138, 140 – 141, 169, 185 – 186, 191, 196 – 198, 200 – 202; pastness, presentness, futurity 2, 4, 9, 10, 28, 30, 34, 42, 44, 50, 62, 77, 79, 88, 92, 94 – 95, 102, 111 – 112, 114 – 115, 131 – 133, 135 – 136, 185, 196, 201 A-relations 59, 132 A-series 2 – 4, 9, 15, 50, 55, 94 – 95, 117, 119, 121, 126, 129, 131 – 132, 134 – 138, 142, 191 – 192, 198, 201 – 202 A-theory 4, 21, 25, 30, 62 – 63, 76, 101, 121, 129, 141 – 142, 187 – 189, 191, 193 A-time 11, 35, 67, 119, 135, 194 Ayer, A. J. 25

Baker, L. R. 12, 14, 15 bare particulars 149, 158, 161, 207 – 214; see also particulars becoming see absolute becoming; A-theory; temporal becoming Bergson, H. 43, 98 B-facts 98 – 99, 127, 129, 133 Bigelow, J. 10, 11 Blake, R. M. 8, 25, 43 – 44 block universe 33, 42, 127, 185, 187, 194 B-propositions 26, 30 Braithwaite, R. B. 8, 9, 25, 45, 195 – 196 B-relations 3, 9, 11, 55 – 56, 59, 61 – 64, 86, 95 – 96, 98 – 99, 102 – 104, 108 – 109, 111 – 112, 116 – 117, 121, 127 – 129, 133, 185, 190, 201 B-series 2 – 4, 6 – 7, 9, 11, 26, 55, 59, 62, 64, 94 – 98, 102, 108, 110 – 111, 116 – 117, 119 – 121, 126, 129 – 130, 132 – 136, 142, 184, 191 – 192, 198, 201 – 204 B-statements 26, 32 B-theory 4, 9, 10, 21, 25, 27 – 28, 33, 42, 59, 62, 76, 110, 119 – 121, 126 – 127, 129, 139, 184, 188, 193 – 194, 201, 204 B-time 119, 127, 135, 193 Cantor, G. 28 causal relations 128, 198 – 199 causation 2, 128 – 129, 172, 185, 199 center and non-center theory (of the unity of the mind) 14, 150 – 151, 153, 188 co-existence 39, 107, 120, 130, 134 – 135

216 Index cognition 47, 126, 164 – 166, 168, 170 – 171; see also precognition commonsense facts see facts continuants 50, 73, 75 copula: temporal 89 – 90, 94, 99, 105 – 108, 113 – 116, 117, 189 – 190; tensed 87, 89, 103, 106, 187, 189 – 190; tenseless 27, 80, 87, 89, 108, 113, 190; timeless 32, 106, 186 correspondence relation 169 – 170 counterfactual dependence 179 – 180 Craig, W. L. 11 – 12, 192 C-relations 61, 109 C-series 6, 59, 108, 194, 201 – 202, 204 Dainton, B. 193 – 194 Dolev, Y. 62 – 63, 192 dyadic relation 5, 88 earlier than/later than 2, 4, 9 – 11, 95 endurantism 138, 188 eternalism see B-theory eternal truths: definition of 25, 26 events: material 154 – 157; mental 14, 146 – 149, 151, 153 – 158, 196; past, present and future 3, 45, 86, 134, 170; see also moments existents 74, 78, 82 – 83, 145 – 146, 149, 151 – 153, 156 – 162, 166, 170 facts: B-facts 98 – 99, 127, 129, 133; commonsense 7, 12 – 16, 19, 21, 76, 91, 99 – 100, 104, 187 – 188, 192; ontological 15, 19, 76 – 77, 91, 99, 104, 202; phenomenological 8, 18 – 21, 68, 95; R-facts 39, 42 – 44, 46, 62, 87, 99, 107, 129, 133, 162, 189 – 190, 194, 202; temporal/nontemporal 2, 21, 27, 38, 43 – 44, 62, 86 – 90, 103, 105 – 109, 122, 146, 174, 184, 187, 189, 190; temporal relational 27, 38, 43, 46, 98, 117, 123, 132, 196; tensed 11 – 12, 44, 56, 134, 192 fatalism 10, 145, 164, 174 – 178 foreknowledge 10, 76, 164 – 176; see also precognition Frege, G. 27 full-future theory 86, 92, 94, 99, 102, 111, 116, 184 – 185, 187 – 188; see also A-theory; growing block theory; presentism futurity see A-properties

Gale, R. 9, 42, 126, 137, 184 grammatical/logical form of language 29, 31, 76 – 77, 79, 81, 90 – 91, 93, 99 – 100, 103 – 104, 106 – 108, 114 grammatical subject/logical subject 78, 148, 151 – 152 Grossman, R. 121, 200 growing block theory 43, 55, 57, 69, 88, 145, 151, 153, 166, 184, 186; see also A theory; full-future theory; presentism Hume, D. 4, 200 identity 2, 82, 156, 188, 193, 199 indexicals 35, 57 Ingthorsson, R. D. 8, 10, 119 – 123, 126, 129, 134 – 135, 138 – 141, 162, 188 – 189, 198, 200 Johnson, W. E. 1, 162 judgments 35, 48, 76 – 81, 137, 173, 175, 193; genetic 77, 79 Kaplan, D. 57 language: ideal 29, 107, 108, 190; ordinary 12, 14 – 15, 2, 27 – 29, 37, 87, 89 – 91, 94, 100, 103 – 104, 106 – 107, 113 – 114, 120, 150, 187 – 190; temporal 25, 33, 44, 99, 190; tensed 25, 30, 102 – 103, 187, 189; tenseless 107, 192 law of excluded middle 83, 176 Le Poidevin, R. 76, 128, 202 Lewis, D. 179 – 180, 200 logical subject 78, 148, 151 – 152 McCall, S. 179 McTaggart’s paradox 10, 21, 29, 51 – 52, 67, 73 – 76, 86, 89, 113, 116, 119 – 121, 129, 135 – 136, 141, 161, 188, 204 meaning 16, 28 – 30 Meinong, A. 25, 39, 103 memory 8, 10, 34 – 35, 47 – 48, 51, 65 – 66, 72, 83, 145, 164 – 170, 185, 197 – 198; knowledge 165 – 167, 169 – 170; ostensible 167 – 168; veridical 167 – 168, 181 mental events see events Miller, K. 62 – 63

Index  217 moments 18, 32, 34, 37 – 38, 41 – 42, 47, 49, 56, 67, 96, 109, 111, 113, 128, 133, 139, 141, 155 – 156, 162, 186 – 187, 191, 202; see also events Moore, G. E. 1, 12 – 15 moving spotlight theory 42, 58 – 64, 92, 110, 139 neutral monists 13 Newtonian view/theory 154 – 156 non-relational temporal properties see properties “now”: here/now 35 – 36, 55, 57 – 58, 103; moving “now” 58 – 61, 64, 202 Oaklander, L. N. 11, 76, 81, 130, 134, 149 objective becoming see temporal transition ontological facts see facts ontology: substance 120 – 123 particulars 2, 26 – 27, 33, 36, 38, 41, 46, 62, 77, 87, 90, 96, 107 – 108, 123 – 125, 133, 138, 145, 147, 151, 153, 155 – 162, 177, 184, 186 – 187, 192, 198, 201; persistent 38, 90, 107 – 108, 187, 192, 198; timeless 147, 155; see also bare particulars passage 10 – 11, 21, 42, 46 – 47, 58, 60, 127, 131 – 132, 135 – 140, 191, 193 pastness, presentness, futurity see A-properties precognition 164 – 176 prehension 164 – 165, 167 – 168, 174 – 175 presentism 7, 11, 37, 43, 86, 92, 94, 101 – 102, 116 – 121, 129, 137 – 141, 162, 165 – 166, 186 – 189; see also A-theory; full-future theory; growing block theory Principle of Pickwickian Senses 17 – 18, 20 – 21, 104, 192 Prior, A. N. 3 properties: incompatible 51, 67, 70, 72, 75, 136, 200; monadic temporal 8, 10, 201 – 202; non-relational temporal 28, 30, 42, 55, 62 – 63, 68, 133, 186; ordinary 67 – 70; see also A-properties; temporal properties propositions: tensed 25, 169; timeless 32, 167 – 168, 170

Pure Ego 13, 18, 147 – 149, 151 – 158, 160 – 161, 188 relations: causal 198 – 199; dyadic 5, 88; earlier than/later than 4, 10 – 11, 9, 95; external 41, 61, 63, 120, 128, 138, 185, 198; internal 61, 63, 102 – 103, 190, 198; simultaneity 4, 8, 34, 58, 91, 95, 134; spatial 8, 34, 40, 47, 54 – 55, 202, 207 – 209; spatio-temporal 155 – 156; see also A-relations; B-relations; C-relations; R-relations; succession; temporal relations R-facts see facts R-relations 3, 11 – 12, 39, 42 – 44, 51, 54 – 55, 61 – 64, 66, 99, 102, 107 – 109, 115, 121, 127 – 129, 131, 133, 135 – 137, 140, 185, 189 – 190, 92, 197, 202 R-theory 3, 9, 11, 21, 25, 27, 33, 40, 42, 45 – 46, 55 – 58, 61, 63 – 64, 86 – 87, 90, 94, 99 – 100, 103, 106 – 110, 114, 116, 121, 126 – 129, 132, 136, 139, 141 – 142, 162, 166 – 167, 177 – 181, 184 – 185, 187, 190, 194 – 197, 202, 204 Russell, B. 1, 8, 12 – 13, 16 – 17, 19 – 20, 25, 27, 41 – 46, 61, 64 – 65, 94, 96 – 100, 103, 107 – 108, 121 – 129, 131 – 133, 142, 184, 190, 194, 198, 201 – 203 Russellian view/theory of time 3, 9, 25 – 27, 41 – 43, 46, 49 – 50, 54 – 55, 57, 59, 61 – 62, 64, 87, 89 – 90, 99, 100, 107 – 109, 114, 120 – 121, 123, 126 – 129, 133, 140, 176, 184 – 185, 194, 201 – 204; see also R-theory Savitt, S. 62 – 63, 201 Schlesinger, G. 4 Skow, B. 62 – 63 specious present 8, 20, 33, 47 – 48 substance 93, 122, 124, 138 – 140, 147, 149, 154 – 157, 195, 200 – 201, 209 succession 5, 8 – 9, 12, 20, 36, 39, 42 – 45, 60, 63 – 65, 67 – 69, 95, 111 – 112, 127 – 129, 136 – 142, 166, 191 – 192, 194 – 195, 197, 199, 202 Tegtmeier, E. 46, 98, 194, 199 – 201 temporal becoming 2, 4, 7, 9 – 12, 20, 30, 42, 45, 54, 59 – 60, 62, 68 – 72,

218 Index 91 – 94, 101 – 102, 108, 111, 115, 119, 136, 185 – 186, 192 – 193, 195 – 198, 201, 204 temporal change 2, 6, 7, 67 – 69 temporal characteristics 8, 20 – 21, 45, 49 – 50, 57, 92, 184, 195 temporal entities 10, 47, 189, 202 temporal individuals 2, 56, 202 temporal parity 119, 121, 126, 129 – 130, 132 – 134, 141, 198 temporal phenomena 2, 7 – 8, 10, 21, 40, 98, 190, 195 temporal predicates 28 – 29, 89 – 90, 103, 113 – 114 temporal properties 4 – 5, 7 – 10, 28, 42, 51, 55, 57, 59, 62, 68, 70, 115, 117, 133, 136, 161, 189, 201 – 202, 204 temporal relations 7 – 12, 20, 26 – 29, 33 – 34, 36 – 50, 54 – 57, 59 – 60, 62 – 67, 71, 73, 87 – 88, 99 – 103, 106 – 108, 110, 113 – 114, 116, 121 – 126, 128 – 129, 132 – 133, 139 – 140, 155 – 156, 161 – 162, 169, 177, 179, 184 – 187, 190 – 194, 198 – 199, 202 – 203 temporal sequence 90, 100, 113 – 114, 138, 194; series 5 – 6, 33, 40 – 42, 55 – 57, 59 – 60, 63 – 65, 88, 96, 102, 108, 110 – 111, 116 – 117, 132, 135 – 136, 138 – 139, 189, 19 temporal succession 20, 128 – 129 temporal transition 46, 21, 46, 127, 137

tensed sentences/statements 27, 29, 44, 52, 89 – 91, 94, 105, 114, 185 Thomas, E. 9, 54 time: absolute 2, 50, 56, 79, 130, 156; and change 2, 7, 73 – 74, 76, 86, 94, 96, 136, 139, 142, 188, 191, 193, 200 – 201; intrinsic direction 6, 9, 56 – 57, 61, 64 – 65, 88, 201; language of 3, 29, 99 – 100, 113, 119 – 120, 189, 192; ontology 2 – 3, 5 – 7, 10 – 12, 28 – 29, 33, 44, 52, 55, 81, 90, 99, 119, 121, 131, 162, 169, 176, 187, 189, 193, 20; passage 4 – 5, 10 – 11, 15, 20, 29 – 30, 44 – 46, 49, 58, 60, 121, 127, 129, 132, 136, 141 – 142, 188, 191, 193 – 195; tensed theory/ view 91, 170, 174 – 176; transitory aspect 3 – 4, 7 – 9, 18, 21, 42, 44, 69, 86, 89, 91, 102, 108, 113, 187, 191, 193 – 194; see also A-time; B-time; Russellian view/theory of time timeless particular 147, 155 Tooley, M. 74 – 76, 81 totum simul 33, 36, 39, 42 – 43, 60, 80, 117, 140, 181, 185, 187 TP (temporal principle) 98 – 102, 113, 116 universals 26 – 27, 33, 41, 43, 78, 81, 87, 108, 123 – 125, 128, 158 – 159, 161, 177, 190, 194, 208