The Future is Unreal — Part 1

0:00 Thank you very much. Can everyone hear me okay? Okay. So, my abstract promises more than I will deliver. It was, of course, not really an abstract in the sense of something abstracted from a finished piece, but a promise. I'm going to mostly talk about McTaggart's argument and won't say much of anything about Alexander or I had one quote from Broad that, but it turns out I think that's appropriate. It's been almost 100 years since McTaggart's argument about the unreality of time, and I suspect it's been refuted many times successfully, but at least there's no agreed upon. reputation of it. And so I'm going to put my hand to it. I suppose if I say anything true, it's pretty surely not original. If I say anything original, it's pretty surely not true. But there's some good jokes. I begin not with MacTaggart, however, but with fatalism, an ancient argument for fatalism. By fatalism, I do not, of course, mean that we are fated like to do something terrible at some point in the future, no matter what choices we make now. I mean the philosophical doctrine that we can do nothing at all to affect the future in any way. For most of us, this will mean we are not only fated in the first sense to do terrible things, although hopefully not as terrible as Oedipus was fated to do, but that even the root to these terrible deeds is not in our power to alter. I adopt the principle I'll call fatalistic entitlement, namely that we are entitled to all the distinctions we need to avoid fatalism. I will see what these are and then maintain that they show us the way to avoid McTaggart's argument and that in doing that we will see although there is no reason to agree with McTaggart that time is unreal, the future is in a fairly clear sense, not real. The fatalistic argument goes like this. The proposition that Hillary Clinton will be inaugurated president in 2009 is either true or false, but not both.

2:30 Second step, if Hillary Clinton will be inaugurated president in 2009, then that proposition is true. It's got to be either true or false. If she's going to be inaugurated, obviously it's true. Propositions do not change their truth values. So, fourth step, if the proposition that Hillary Clinton will be inaugurated president in 2009 is true today, then it was true a year ago. Next step, you cannot change the past. So, if something was true a year ago, no one can do anything now or at any time later than now that will affect its truth value. So, if Hillary Clinton will be inaugurated president in 2009, there is nothing she can do or Bill Clinton can do or Jeb Bush can do today, or anything that anyone else can do today, or at any day in the future, to prevent it. Here's the dubious step. As important as Hillary Clinton is, and as important as the issue of her inauguration is, there is nothing in principle that makes this event special in terms of this argument. So, if something is going to there is nothing anyone can do now or at any time to prevent it. So that's a version of an argument many of you I'm sure have taught and criticized in your classes and so forth. The main problem with this argument, it seems to me, is that it does not recognize that many propositions, certainly including propositions about who wins elections and who is then inaugurated as president of the US the following January, are made true or false by events. In this case, crucially, the elections that are held in November preceding the inauguration and the events between then and the next January 20th, which may include dubious Supreme Court decisions and the like. If it turns out that Hillary Clinton wins the 2008 U.S. presidential election and is then inaugurated in 2009, this will be on account of events, many of which will not happen until Election Day, the first Tuesday of November 2008. Suppose that Hillary and Bill Clinton have a big in late September 2008. It would be pleasant to provide details concerning what this fight might be about, but I will leave that to your imagination. Feeling hurt and unappreciated, it occurs to Bill that he could make a speech in which he details every nasty thing Hillary has ever done or said and asserts that he wouldn't vote for her for dog catcher, much less for president,

5:00 and that by doing so he could prevent her election and hence her inauguration. He can prevent her from being inaugurated. But remembering all the good times he had in the White House, he doesn't do it, and she is elected and inaugurated. Still, if he had done it, he would have been affecting the future, not the past. He wouldn't have changed the past at all. have changed the past, for he will not do anything that makes false something that has already been made true, nor does he do anything that makes true something that has already been made false. We need to think about the truth of propositions in some way that at least does not rule out this common sense response to the fatalist argument. Even if we are hard determinists, that is, even if we believe that determinism is true and it rules out freedom, we should still not fatalism, for if fatalism is true, determinism is really quite beside the point. It seems to me that to allow this common sense response, we need to recognize two concepts of truth in relation to propositions. The first is a property of propositions, a proposition be true or be false, or if you prefer, be true or be false of a world. It's best to avoid using tense with this concept, hence the untensed B. Second, events up to a certain time make a proposition true or make it false or leave its truth or falsity open. So events up through 2009, say March, will make it true or make it false that Hillary Clinton is inaugurated president for that term. They presumably will leave it open whether or not, say, Jenna Bush is inaugurated president in 2032 or 33. So there is an important connection between these concepts. If events up to a certain time make a proposition true or false, then it be true or false. The converse principle does not hold. There are, or at least seem to be, propositions that are not made true by events but are nevertheless be true, such as the propositions of logic and mathematics and other necessary truths. If we believe that the laws of nature are contingent

7:30 but also believe that they are not merely the empirical generalizations that remain true at the end of time but somehow structural principles that shape what happens then they too will not be made true by events. Events will conform to them but not make them true. So, in general, if events up to a certain time make a proposition true, then it be true. From the connection that propositions made true be true, and what we know about being true, we know that if events up to a certain time make a proposition true or false, events up to some other time, earlier or later, don't make it false or true. From this, I believe, we arrive with the correct understanding of the claim that you cannot change the past. Namely, you can't do anything to make a proposition false that has already been true. Excuse me. You can't do anything to make a proposition false that has already been made true or to make a proposition true that has already been made false. However, the following principle is not correct. If a proposition P be true, no one can do anything now that will affect its truth value. This is not correct because the events that will make P true may lie in the future and someone powerful, like Bill Clinton, may well be able to do something to prevent them. With this understanding, if we go back to the fatalistic argument, the first steps become the proposition that Hillary Clinton is inaugurated president in 2009 be true or be false, but not both. If Hillary Clinton will be inaugurated president in 2009, then that proposition be true. When the argument starts that way, nothing of significance follows. From the supposition that the proposition be true, it does not follow that it has been made true by today or yesterday or will have been made true by, say, September 2008, when Bill and Hillary have their fight. So it doesn't follow that. If Bill Clinton were to prevent it from being made true by a nasty speech in late September 2008, he would have in any way changed the past. Slight digression. For discussions of determinism, a third concept is important. If a proposition is entailed by propositions that have been made true by what has happened up to a certain time,

10:00 together with the laws of nature, and if the laws of nature are propositions that be true without having to be made true by events then I will say that the proposition is settled by the time in question although it has not yet been made true so if determinism in some form or other is true it may be settled by late September 2008 and indeed may have been settled by the time that Adam bit into the apple that Hillary Clinton would be inaugurated as president in 2009 The question of the compatibility of human freedom and determinism is whether we can, at a given time, do things to prevent events whose occurrence is already settled. But that issue I leave aside. I mention only to emphasize that being settled in this sense is one thing. Being made true by events is another. Now, let's consider a manageable series of events, say the presidential inaugurations in the United States of the 20th century. from McKinley's in 1901 to George W. Bush's in 2001. Consider the domain of all the people who appeared on the ballot for the U.S. president in the 20th century. Consider the property being inaugurated as president in D, where D is a year. So this domain and property give us a set of atomic propositions. Some of these, like that Carter be inaugurated president in 1976, true, or as here we could say, be true of this little sequence of events we're considering, and others like that Carter B inaugurated president in 1980 be false. Each atomic proposition will be made true or made false by the events that occurred in the 20th century up to March 4th of the year that is a constituent of the proposition. So as of, say, April 1975, a number of propositions that be true of the sequence were not yet made true or made false. As of April 1975, it had not been made true that Bush was inaugurated in 2001, nor it had been made false that Gore was inaugurated in 2001. So now I move to a section called chronological possibility. Now I'd like to introduce a second class of propositions of the form X in year D can prevent P, where X is a

12:30 candidate from our domain, one of the nominees for president. D is a year in the 20th century, and P is one of our original atomic propositions about a guy or a girl or a woman is inaugurated in a year. So examples are Carter in 1980 can prevent that Reagan be inaugurated in 1981. Carter and Reagan, you may remember, ran against each other in 1980, and Reagan won and was thus in 1981. So the proposition I have in mind now is Carter, in 1980, can prevent than one of the original atomic propositions that Reagan be inaugurated in 1981. This would seem to be true if in 1980 there was some set of basic bodily movements that Carter could have made or refrained from making, which had he made them or refrained from making them, other circumstances being what they were, this would have led to his winning the election of November 1980 rather than Reagan. Many analysts think that had Carter been less obsessed about the hostage situation in Iran and not tried to micromanage the issue, he would have defeated Reagan. If they are correct, this proposition is true. The following proposition is almost certainly true, that Reagan in 1980 can prevent that Reagan be inaugurated in 1981. Reagan could have withdrawn from the race. Or just to put some more intriguing possibilities before us, he could have taken off all of his clothes in the middle of a speech, proclaimed that he was a nudist, and would never wear another stitch, even if elected president. Or he could have divorced his wife Nancy and married Charo, a venerable San Francisco institution, who at about that time was the hoochie-coochie girl on the Merv Griffin television show. I could go on, but maybe I've made the point. At each time, there are propositions that have been made true and propositions that have been made false, and in addition, many propositions that have not been made true or false. There is a certain kind of impossibility involved with respect to propositions that have already been made true or false by a certain time. There is nothing anyone can do, and for that matter, nothing that can happen, whether done by a person or simply the result of non-human causes that or true. The kind of possibility involved is not the same as logical possibility. That is, the falsity

15:00 or truth of these propositions does not involve a contradiction. Nor is it a matter of pure metaphysical impossibility. The propositions involved are not guaranteed to be true by whatever deep structural facts there are about properties, relations, and objects. I'll call this sort of impossibility and possibility. of possibility and impossibility appealed to but misused in the fatalist argument, chronological possibility. Propositions that at a given time might still be made true by events are chronologically possible at that time. Those that can no longer be made true are chronologically impossible at that time. Section three, McTaggart's B series and C series. That said, I turn to John McTaggart Ellis McTaggart. I love his name, John McTaggart Ellis McTaggart. If you ever put it in an article, the editor always queries whether you accidentally repeated his name, but that's his name, John McTaggart Ellis McTaggart. Seems like real class to an American, you know. But I digress. That said, I turn to John McTaggart Ellis McTaggart in his famous argument of almost 100 years ago that time is unreal. He describes three series of events, which he calls the C-series, the B-series, and the A-series. I suppose a fact that is painfully well-known to this audience by now. We'll put the A-series aside for the moment and discuss the C-series and the B-series. The C-series and the B-series both comprise all the events in history in order. that the B series also includes the temporal direction of the events. Suppose that in the course of archival investigations you came across a list of leaders of some small country which you'd never heard of, the land of woe, say. It looks like this. Year 100, the reign of Elwood the Unready begins. Year 110, the reign of Gretchen the Inept begins. Year 120, the reign of Ephraim the ignorant begins. So you know the order, in MacTaggart's sense, of the rulers, but you don't know the direction because you don't know whether the years listed are B.C. or A.D. Thus you don't know whether Elwood came before or after Gretchen. You know that Elwood's

17:30 reign was next to Gretchen's and not next to Ephraim's, but you don't know which came first. You have order but no direction in MacTaggart's sense. Basically, if you can came between which events, you know the order. If you could say which came first, you know the direction. McTaggart believed that both order and direction are part of our concept of time and that the direction of time is the direction of change. So the C series plus the direction of change should give us the B series. McTaggart can come up with no coherent account of the direction of temporal change, so he concluded that time is unreal. Just to keep things manageable, let's take a miniature list of the history of the world, let us limit ourselves again to events that are inaugurations of United States presence. This is, I realize, in many ways an extremely unpleasant object to continue to think about. But I will plunge ahead nevertheless. I considered using the inaugurations of governors of California instead, since that would provide a pleasant bond between Austria and California, since California's present governor is our beloved Austrian Arnold Schwarzenegger. However, there is a great deal of jealousy among Americans from other states, because California has a European governor. And since people like John Herman from the state of Pennsylvania were here, and I didn't want his jealous rages breaking out, I to stick to the example of presidents. Now, you are required to be 35 years old to be inaugurated as president in the United States, so we can be sure that all of the presidents inaugurated up until 2037 are already alive. So I will limit myself to the succession of inaugurations that begin with Franklin Delano Roosevelt's second inauguration in 1937, continue through George W. January, and then continue with Hillary Clinton's two inaugurations in 2009, 2013, Laura Bush's two inaugurations in 2017 and 2021, Chelsea Clinton's two inaugurations in 2025 and 2029, Jenna Bush's

20:00 inauguration in 2033, and her twin sister Barbara's in 2037. I'm not entirely certain of these results, But I've done a fair amount of research, and I'm fairly confident. So that gives us 100 years of inaugural events and 100 propositions, all of the form, that X be inaugurated succeeding Y. Bush be inaugurated succeeding Clinton, Bush be inaugurated succeeding himself, Hillary Clinton be inaugurated succeeding Bush, so forth. Starting with that FDR be inaugurated succeeding himself, which happened in 1937, and continuing, too, that Barbara Bush, not the grandmother but the twin, that Barbara Bush be inaugurated, succeeding Jenna Bush. Let us give this series of propositions a name, say, dismal. If we think of dismal as an ordered sequence of propositions, then, of course, it cannot be changed. It is a set theoretical object defined by its members. Now, if all of the propositions in dismal are true, then dismal is a B series, that is. I'm taking any subset of the B-series, the set of everything that ever happens in order and direction, and calling any subset of it a A-B-series. I didn't realize that that would be so confusing to say A-B-series. I'd say and-B-series. Now, even if dismal is a B-series, and even if we are all powerless to change the membership of dismal, since it's just a set of some sort, can do something that will make it not a B-series by doing something that will prevent one of the propositions in it from being made true. For example, Jenna and Barbara Bush might become nuns, in which case they'll be prevented by papal decree from taking part in American politics. That is not likely, but it could happen. Even if dismal be a B-series, which requires that all the propositions in it be true, some of the events necessary to make it a B-Series have not yet happened. We, or at least powerful politicians, can do things that affect which propositions have any to do with which future elections, having to do with future elections. All we cannot do is affect truth values of the propositions that have already been made true or false. So I will assume, nevertheless, that dismal is a B-Series and plunge on. The next section is called the A-Series and

22:30 the D series. The first thing we need to do now is to pull out of dismal all the information about the direction of the time. Dismal is the B series. We pull out all the information about the direction of time and we will have what MacTaggart called a C series. I'll call it dismal sub C. So this is a series of propositions of the form X be inaugurated next to Y, rather than X be inaugurated succeeding one. The series contains information about which inaugurations were between which other inaugurations, but not about which came earlier and later. Now how do we put back in what we have just taken out? To get back to a B series, to get from dismal sub C back to dismal, we have to add information that imposes a direction on dismal sub C. McTaggart tells us three things about what we need to add. First, it is related to temporal change, which he takes to be something we experience, or at least seem to experience, for according to him, it turns out to be an illusion. Second, temporal change is fundamentally different from the sort of change we talk about when we say, for example, that a stretch of Highway 4 in the Sierra Nevada, of a stretch of Highway 4 in the Sierra Nevada mountains. For each town from Angel's Camp to Ebbets Pass, the elevation above sea level increases as the distance from the Pacific Ocean increases. So he doesn't think temporal change is just a matter of one, of some parameter changing along a dimension. I am inclined to agree with McTaggart about both of these things today, maybe not in the future. Although I do not think that our experience of change is an illusion, I must say, and I'm happy to be here in Austria, where, as far as I know, there is a long tradition of philosophers that deny the reality of time, unlike Germany and England. I must say, I really think our experience of time must be one of the fundamental ways which we're locked into very basic facts about the universe. I mean, I can believe that. You know, political parties and nations and maybe even other people or maybe even ourselves are in some way mostly construction and illusion. But time and space seem to me, boy, if we're not right about those, I'm pretty depressed.

25:00 And I just don't know how McTaggart and Kant and all those guys live with themselves. But I digress. I really believe in space and time. I'm a big space-time enthusiast. I've spent my whole life in space and time, and I intend to spend the rest of it there, too. I'm going to demonstrate my free will now, just to remind myself of, see, I mean, notice, see the possibilities being eliminated as I? Anyway, the third thing McTaggart tells us is that what we need to add to the C-Series to add direction is what he calls the A-Series, which is a series of events ordered by whether they are in the present, past, or future. And, if in the past or future, how distant. The fact that Truman's inauguration came after Roosevelt's fourth inauguration, rather than being merely next to it, And before Eisenhower's first inauguration, rather than merely being next to it, consists in the fact that Roosevelt's fourth inauguration is more distantly past than Truman's, and Truman's more distantly past than Eisenhower's. This idea of McTaggart seems to me to have been rather unfortunate, a wrong turn in the philosophy of time. He was quite confident of this. he was confident that the only thing you could add to the C series to get the B series was the A series. And that, therefore, if he could show that the A series couldn't be added to the C series in a coherent way, he had shown that our concept of time really had no application. He had not merely shown that his own analysis was wrong, as a more modest person might have thought, but had shown that time was unreal. So he was quite certain. But this really seems to me to be quite weird because it doesn't seem to me that the first thing I would pick about time to get at change would be the movement of events from the future into the present and then into the past, which is what he thought the A-Series. I would pick something like what you see when you do something, namely change. But I'll get back to this.

27:30 I believe that what we need to add to dismal sub C in order to return to dismal is facts about chronological possibilities, which I will call the D series, or in this case, dismal sub D. That is, we need to add propositions about what the chronological possibilities were at the time of Roosevelt's third and fourth inauguration, Truman's inauguration, and so forth, right down through the chronological possibilities at the time of Bush's first inauguration, Bush's second inauguration, Hillary Clinton's first inauguration, and beyond. I am allowing myself the concept of simultaneity, for as I understand McTaggart, this is allowed in the C-Series. Events X and Y occur at the same time if there is no event between them, and we can have this information even if we do not have the information about the direction of So we can add dismal sub D, this sequence of changing possibilities, to dismal sub C, giving us a sequence of propositions about the occurrence of inaugurations and the chronological possibility and possibilities at the times those inaugurations occurred. During his third term, Roosevelt's vice president was Henry Wallace, a Democrat from the left wing of the party, perhaps correctly called, as he was by many Republicans, a socialist. not a term of praise in the United States even then. Had Roosevelt chosen Wallace for his running mate for the 1944 election, as he certainly could have, then it seems that either one of two things would have happened. One is that Thomas Dewey, the Republican candidate in 1944, would have defeated Roosevelt. The other is that Roosevelt would have won, and Wallace, rather than Truman, would have succeeded to the presidency in 1945 when Roosevelt died. If either of these courses' events had occurred, never would have been present, for he was an amiable Missouri senator with no presidential ambitions. So it seems that at the same time as Roosevelt was inaugurated for his third term in 1941, it was chronologically possible that Truman not be inaugurated in 1949, as he in fact was, having served out the rest of Roosevelt's fourth term and having defeated Dewey in 1948. That was, by the way, my earliest political memory. My parents, who were devout Republicans, had a big party

30:00 to celebrate Dewey's election on election night. And I remember, I think because it was so emotionally, the House that evening, as my father and mother and uncles and friends just listened in disbelief to Dewey's going down to flaming defeat at Harry Truman's hands. What that has to do with McTaggart, I don't know. But I just thought I'd share that with you. At any rate, by the time of Eisenhower's first inauguration in 1953, it was no longer chronologically possible that Truman not be inaugurated in 1949. about it. Not Dewey, not my father, not my uncle. This was when I was introduced to the powerlessness of human beings. When we add true propositions about chronological possibilities for other inaugurations to dismal sub C, our little C series, a clear pattern emerges that gives us two clear directions, one of which is the direction of temporal change. As we move one direction through augmented dismal sub-C, possibilities will increase. And as we move the other direction, possibilities will decrease. The possibility that Gore be inaugurated in 2001 is present at Roosevelt's second inauguration in 1937, and at every subsequent inauguration through Clinton's second inauguration in 1997. Then it disappears, sad to say. The direction of time and of change is the direction of decreasing possibilities. If we confine ourselves to any finite subset of the contingent propositions that be true, more and more will be made true and more and more rendered impossible as we consider events that are later in time. And fewer and fewer will have been made true and fewer and fewer rendered impossible as we consider events that are earlier in time. The next section is called Change and the D Series. I now turn to the question of whether the D-series meets McTaggart's requirements for what needs to be added to a C-series to get a B-series, namely, whether it is something that characterizes our experience of temporal change and not merely a matter of change along a dimension.

32:30 Angel's Camp, Murphy's and Arnold, as I mentioned, are three towns one drives through as one travels east along California's Highway 4, which runs from the Bay Area through the Sierra Nevada Mountains to the state of Nevada. Consider the facts about the elevations above sea level of these three towns, roughly 1,500 feet, 2,500 feet, and 3,500 feet. One experiences this change in elevation of towns along Highway 4 relative to the distance from the Pacific Ocean as one travels along the highway. Each of these facts by itself does not necessitate the others. Angels could be 1,500 feet even if Murphy's were not 2,500 feet and vice versa. for example. On the other hand, if we travel along Highway 4 from west to east, and by the time we get to Murphy's, we have eliminated the possibility that Angel's Camp has any elevation other than 1,500 feet, and that Murphy's has any elevation other than 2,500 feet, but we have not eliminated the possibility that Arnold has some elevation other than 3,500 feet. So there is a progression, a change in what is possible as we travel from west to east. This is, however, much different possibility. For one thing, the change in what is possible in this case has to do with what the subject knows. It is a matter of epistemic possibility. Related to this, the situation is reversible. If you travel from east to west, and when you get to Murphy's, it will not be possible that Arnold has any elevation other than 3,500 feet, but it will be possible that Angel's camp has some elevation other than 1,500 feet. Chronological possibility is not simply a matter of epistemic possibility, and it is not reversible in this way. Let us return to our friend Bill Clinton, late September 2008, contemplating whether to make a speech that will destroy Hillary Clinton's chance of being inaugurated the following January. He might think that it is possible for him to destroy her chances. It is something he can do. It is a possibility for him. It is up to him. On the other hand, at that point in time, it is not possible for him to destroy her chances for being nominated, which will have happened in summer 2008. That was once a possibility, say in the early days of the Democratic Convention and before, but it is no longer a possibility by September 2008. In other words, Clinton and all of us are aware of chronological possibilities and chronological impossibilities. As we experience change, we experience the decrease in chronological possibilities and the increase in chronological

35:00 opportunities. Opportunities once missed are gone. The direction in which chronological impossibilities increase and the chronological possibilities decrease is the direction of change as we experience. As we experience change, it involves events occurring which we perceive and participate in, which make it the case that various propositions are true or false, and leave no possibility of our doing anything or anything happening that will make them otherwise. Past events and future events play different roles in our lives, cognitive, emotional, and practical, all connected to our sense of chronological possibility. We regret the past and try to make up for it, or cover it up, or change its consequences, but we do not try to change the past itself. We try to shape the future to make some events more likely to occur and others less likely to occur. We remember the past, we anticipate the future. Everyone seems to agree about this. People move from west to east and east to west, and as they do, they remember where they have been and anticipate the experience of what they have yet to encounter. But all are anticipating the future and remembering the past. No one moves from future to past trying to affect the past and remembering the future. No one, of course, except our beloved Austrian governor, Arnold Schwarzenegger, who in Terminator 1 and Terminator 2 but I won't go there. So I conclude chronological change meets McTaggart's requirements and that by augmenting the dismal sub C with dismal sub D, we have returned to our starting point dismal, a B series, and have done what McTaggart thought could not be done. But you won't be convinced yet, not until we've discussed the A series and some other issues about the B series. So the next section is called McTaggart and the A series. McTaggart thought that you had to add the A series, the one that divides events into past, present, and future, to the B series to get the C series. He did not think there was a coherent way to do this and so concluded that time was unreal. That is, we have a clear concept of what time would have to be, a combination of the C series and the A series into a B series, And his argument gives us a clear idea that we can't do this,

37:30 so we see that our concept applies to nothing. So according to McTaggart, at any rate, the fact that Truman's inauguration came after Roosevelt's fourth inauguration rather than before it or merely being next to it consists in the fact that Roosevelt's fourth inauguration is further in the past than Truman's. I must admit, I cannot see any merit, or at least almost no merit, in McTaggart's idea. Suppose I say, truthfully if sadly, George W. Bush's second inauguration is past. That makes it sound like I'm sad because I was anticipating it so much. That's not what I had in mind. Anyway, what fact makes my statement, George W. Bush's second inauguration is past, true? The reasonable theory, like mine, says that my utterance, let's call it you, is true simply because George Bush's second inauguration precedes you, the utterance itself. So if we want to put the fact that I seem to be getting at when I say George Bush's second inauguration is passed in the dismal sub C, we ought to add the proposition that George Bush's second inauguration precedes my utterance U. But that proposition is already a B-series proposition. If we're going to add a B-series proposition to dismal sub C in order to get dismal, we may as well simply jump back to dismal and be done with it. Now, I should point out that one can agree with this point, even if one thinks, as I do, that David Kaplan's theory is also correct in saying that the proposition that George Bush's inauguration precedes you is not the proposition expressed by my utterance of George Bush's inauguration is past. According to history, the proposition expressed by my utterance of George Bush's inauguration is past is not a proposition about my utterance itself, but roughly a proposition about the time T, the time at which my utterance occurs, to the effect that George Bush's inauguration precedes T. This proposition would be true even if I had not made the utterance, although proposition that George Bush's inauguration precedes my utterance would not be true if I hadn't made the utterance. So they're not the same proposition. However, if you trace Kaplan's analysis through, you'll see that my utterance will be true if and only if George Bush's

40:00 election precedes it, for only under those conditions will it express a true proposition about the time at which it occurs. So the express proposition that George Bush's inauguration proposition and not a C proposition. So it's just a little unclear how you can add these or why it would do any good. I do not mean to imply that McTaggart missed the fact that adding these propositions to the C series would be begging the question and perhaps not very helpful. But he did not think the words past, present, and future were basically getting at relations to utterance of sentences containing them, but they get at some property of events. Indeed, he thought the passage of events from future to present and present to past was the essence of temporal change. He thought that to find real temporal change, we needed to find some way in which the, some temporal way in which the B series changes that corresponds to our subjective sense of events passing from the future into the present and then into the past. McTaggart thought there can be no change in the B series. The B series contains all events and to never change. As he puts it, if M is ever earlier than N, it is always earlier. So our concept of time demands that the B series is the A series added to the C series, but the C series will have to involve events changing from future to present to past if we add the A series as he conceived it to the C series, and this makes no sense. It seems to me that the idea of an event moving from the future into the present to the past is not any very basic thing about our experience of time. It's rather a metaphor and a misleading one at that. What's basic to our theory of time is a perception of change and the asymmetry between what is made possible and what is made impossible as change occurs and the conditions for more and more propositions completed. And this idea of an event moving from the future into the present and into the past really connects with none of that except through the words past, present, and future. So in spite of the incredible amount of ink that has been spilled, including some by me, on this, it seems

42:30 me, it just wasn't a very good idea. However, I need to see if I have evaded the conundrum that bothers McTaggart, that you have all the changes incorporated into the B-series, and yet somehow there's something about temporal change that requires the B-series to change. How can that be? So the next section of my paper is called The Future is Unreal. But I only mean the part that hasn't happened yet. As I said, early in his discussion, McTigert says, where M and N are events, if M is ever earlier than N, it is always earlier. So these basic propositions in the B series, he's saying, they're true, they're always true. Well, this seems to me to be a mistake. Consider Bush's second inauguration. It has happened, while Hillary Clinton's first inauguration has not happened yet. Is it correct to say that Bush's second inauguration is earlier than Hillary Clinton's first inauguration? Certainly after Hillary Clinton's first inauguration, we can say that Bush's inauguration was earlier than it. But was it earlier before Hillary's first inauguration happened? Is it now earlier than her inauguration? It seems to me that Hillary Clinton's inauguration is not now later than Bush's second inauguration because Hillary Clinton's inauguration does not yet exist. By Hillary Clinton's inauguration, I mean the concrete event, not a description of it, or an abstract object that characterizes it. The concrete event, it seems to me, has no reality at all until it happens, even if the propositions that say it will happen be true, or for that matter, even if it is settled, in the sense I adumbrated in the brief discussion of determinism early on, before it happens. The concrete has no existence, no reality, until it happens. That's what we perceive. We perceive things happening, events coming into changing. My argument for this is that the status of events that will occur before they occur is the same as the status of the events that might occur in their stead but will not. All are possibilities,

45:00 not realities, before one of them occurs. It becomes real. The others do not. According to the picture I have put forward, at each time there is a future which is characterized by a number of contingent possibilities or propositions, none of which have been made true at that time, and all of which might still be made true, but only one of which will be made true or factual, a garden of forking paths, as Borges put it. Then, at a later time, some of these possibilities will be eliminated or made false. Now, the simplest explanation for the legitimacy of this picture, as far as I can see, is that at each time all the possibilities for the future have the same status. They are mere possibilities. That is, until an event happens, and so makes the proposition according to which it happens true, the event is a merely possible event and not a real event. And by saying it is merely a possible event, I mean to say basically that it is not an event at all. There There are descriptions of it.