A conversation with the philosopher of science at Columbia University.

David Albert: I'm David Albert. I'm a professor of philosophy at Columbia University. And my research is mostly concerned with issues of the foundations of physics.

Question: What is the role of a philosopher of science?

David Albert: Well, I think that philosophy of science is at its best and at its most exciting at historical moments when it's not so easy to distinguish between the activities of certain kinds of theoretical physicists and the activities of certain kinds of philosophers. Philosophy of science, I think -- or at least -- well, let me back up a bit. There's -- philosophy of science can be divided roughly into two different kinds of activities. One is an activity of raising and investigating general philosophical questions about what science is, about whether the claims of science have some kind of privileged epistemic access to the world, can be justified, attempts to systematize how science reasons, attempts to raise questions about whether we should trust the conclusions of science, so on and so forth. These are very broad, very traditionally philosophical kinds of issues.

There's another branch of philosophy of science that takes up questions that arise within particular scientific theories -- the theory of evolution, the theory of relativity, quantum mechanics, so on and so forth, and actually gets its hands dirty in the details of the structure of these scientific theories in order to try to help with problems that are often essentially scientific problems, but whose solution calls for an unusual degree of sensitivity to philosophical questions. It's the second kind of work that my own work has mostly been, and it's the second kind of work that one refers to when one refers to the foundations of physics. There are problems about the logical structure of physics, about the foundational assumptions that physics makes. Whether these problems properly belong to physics or they properly belong to philosophy when the field is healthy isn't much of an issue.

In my own case, my Ph.D. was in theoretical physics. I was a professor in physics departments before being a professor in philosophy departments. When I write a paper now, my rule is that if at the end it has more than two equations in it, I send it to a physics journal, and if it has less than two, I send it to a philosophy journal, and there's not much more of a distinction than that. When I attend conferences where people are discussing the kinds of questions that I'm interested in, about half of the people speaking at these conferences are employed in physics departments, and half are employed in philosophy departments, and it's not particularly easy to tell by listening to their talks what sort of department they're employed in. So philosophy of science, like I say, when it's healthy, is a matter of focusing with a certain level of philosophical sensitivity and sophistication on questions at the foundations of physics.

Question: What are some of the great questions in physics today?

David Albert: Sure. There's a glass of water on the table beside me. Someone asks, how do I know there's a glass of water on the table beside me? And the answer, the kind of answer that the whole structure of Western scientific knowledge is very deeply committed to, is something like this: there's light in the room, some of the light bounces off the glass, some of the light that bounces off the glass enters my retina; that causes certain electrical excitations in my retina; that in turn causes certain electrical excitations in my optic nerve; that causes various chemical and electrical changes deeper in my brain, and after some finite number of some steps, my brain is in the state that corresponds to having the impression that there's a glass of water sitting on the table. And needless to say, it's crucial to this story that every one of the steps I just described occurs in full accord with whatever the fundamental laws of physics happen to be. Good.

It was noticed about 80 years ago that if one supposes that the fundamental laws of the world are quantum mechanics, if one supposes that the fundamental physical laws of the world are the ones that we get in quantum mechanics textbooks, this story that I just told about how I know there's a glass of water on the table radically falls apart. And it can't fall apart; we can't imagine how else to begin to tell the story of how I know there's a glass of water on the table. So there's a problem; there's some kind of problem at the foundations of quantum mechanics. This problem has come to be called, for reasons that are obvious, given what I've just said, the measurement problem in quantum mechanics, or the observation problem. And there has been a project under way ever since then to imagine how to fiddle around with the fundamental laws of quantum mechanics in such a way as to hang on to all of the very good predictions that the theory makes, but so as to make it possible to tell the story of how I know that there's this glass on the table at the same time.

Now, this is, at the end of the day, a scientific project. The fact that this story collapses indicates that something must be amiss with the fundamental laws of the theory, and what needs to be done in order to correct this is an essentially scientific job of writing down new laws that are going to be compatible with all of the good predictions that quantum mechanics makes and with the possibility of telling the story that I just described. On the other hand, the problems that we run into are peculiarly philosophical sorts of problems. What is going to count as a solution to this problem? What’s the minimum that we need in order to be able to tell a story like this? These are questions, questions of epistemology, questions of the relationship of our beliefs to the external world and so on and so forth that philosophers have been dealing with for a long time.

So philosophers can be, or people with philosophical training or philosophical sensitivity can be, helpful here in trying to frame very precisely what the problem is, what would count as an adequate solution to the problem, so on and so forth. That's a good example, and an example which has been very, very active over the past 25 years or so, of a problem that comes up at what used to be thought of as the boundary between physical investigation and philosophical investigation.

Question: How does quantum mechanics contradict common sense?

David Albert: Here's the deal: quantum mechanics allows physical systems -- and the easiest systems in which to observe phenomena like this are very tiny systems like subatomic particles, electrons or neutrons or protons -- quantum mechanics apparently allows for the existence of physical conditions of material objects like electrons in which questions about where the electron is located in space seem to fail to make sense. Let me back up a little bit and explain this a little more slowly. There are experiments we can do where an electron passes through a certain apparatus, is fed into one end of an apparatus, comes out the other side of the apparatus. And the apparatus has several routes inside of it which the electron could potentially have taken from the input to the output. And there are experiments we can do with pieces of apparatus like this which, taken together, make a compelling case that although the electron went from here to there, it didn't go by route A, it also didn't go by route B, and it didn't go in any intelligible sense by both routes; that is, it didn't split in half, with one half taking one route and one half taking the other route; and it also didn't take neither route, okay?

And what's puzzling about that is that that would seem to be all the logical possibilities that there are. These experiments, you know, are now very routine experiments to do in physics laboratories. We've been good at doing experiments like this for something on the order of 70 years now. We're very good at doing them now. The results are very, very compelling. And after enormous soul-searching and puzzlement and confusion and so on and so forth, the sort of standard consensus understanding that evolved in physics of situations like this is that electrons could apparently be in situations where asking a question of the form "which route did the electron take?" was something like asking a question of the form "what is the marital status of the number five?" Or "what are the political affiliations of this tuna sandwich?" or something like that. These are questions that philosophers often refer to as category mistakes, okay? The very raising of a question about the political affiliations of a tuna sandwich or the marital status of the number five indicates that there's something basic that you're misunderstanding of what it is that you're asking a question about.

And the strikingly strange thing about quantum mechanics -- and indeed it seems to me a case could be made that this is the strangest and most unsettling result to come out of the natural sciences since the scientific revolution of the Renaissance -- is that even things like particles can be in conditions where it simply radically fails to make sense even to ask where the thing is located in space. What's particularly strange about this is that of course there are other circumstances where it does make sense to ask those questions about where it's located in space. The electron determinately goes into the box over here and comes out over there, okay? But we can give good arguments from these experiments that while it's inside, it's not merely that we don't know where it is, it's something much more radical and much more unsettling than that: that the very act of raising a question about where it is represents some kind of misunderstanding of what mode of being that electron is participating in while it's going through this device.

David Albert: Anyway, here's a further fact about electrons: if we go -- if we do one of these experiments that I just described and stop it in the middle, rip the box open, okay, and go look for the electron, as a matter of fact we always find it in some determinate position, okay? We have equations, we have basic laws of motion: the Schrödinger equation in the case of nonrelativistic quantum mechanics; in the case of relativistic quantum mechanics the fundamental equations are the Dirac equation or the Klein-Gordon equation. Anyway, we have these fundamental laws of motion for things like electrons; indeed, for all material things. And these laws are very successful at predicting when these strange -- let me back up and say these conditions in which it fails to make sense to ask whether the electron is here and here -- in a long, distinguished tradition of facing a mystery that one doesn't understand by at least making up a name for it, a name has been made up for this condition. People speak of electrons in such circumstances as being in a superposition of going along route A and going along route B. And although it's very difficult for us to get our heads around what this word means, we are very adept at treating these situations mathematically. We have very reliable equations that tell us when and under what circumstances these superpositions are going to arise and when they're going to go away, and blah blah blah blah blah. Good.

Further empirical fact: when we rip open these boxes and look for these electrons, we always find them in one position or another, okay? So that somehow the act of looking at them makes these superpositions go away, okay? Good. On the other hand, we could perform the following exercise: take these fundamental equations that we have discovered and which we have very good reason to believe are reliable at predicting when superpositions are going to arise and when they're going to go away and so on and so forth; use those equations to predict what ought to occur when we rip the box open and look inside. That may sound like a very difficult calculation to do. It involves this macroscopic human being and his brain and so on and so forth. Actually, there's a mathematical trick for getting this calculation done, as miraculous as that sounds.

And it's very easy to show that what these equations predict ought to occur when we rip this box open is that we ourselves go into a superposition of seeing the electron on route A and seeing the electron on route B, okay? That is, that we ourselves go into some condition in which not only does it fail to make sense to ask where the electron is; it fails to make sense even to ask about our beliefs about where the electron is, okay? Or it fails to make sense to ask whether we're in the brain state corresponding to believing that the electron is on route A or in the brain state corresponding to believing that the electron is on route B. Good. God knows what the hell that would feel like, okay? But the usual way of setting up this measurement problem is merely to observe that whatever it is that would feel like, that's not what happens to us when we rip open these boxes. When we rip open these boxes, there is always a perfectly determinate matter of fact about where we take the electron to be. Sometimes we see it on route A; sometimes we see it on route B; it's never the case that anything else is going on. It's never the case that it looks fuzzy, or we get nauseous, or we become disoriented, or in any sense that one can put one's finger on there fails to be a fact about where we see the electron. Good.

So we have a flat-out contradiction. And this is the more explicit version of the way this story about the glass breaks down. We have a flat-out contradiction between, on the one hand, the predictions of the fundamental quantum mechanical equations of motion about what ought to happen when we rip open these boxes, and our everyday introspective experience of what's going on when we rip open these boxes, which is that in each of those occasions we either see an electron there, or we see an electron there. These two claims flatly contradict one another. Of course, you know, the empirical claim is the one that's true; that's the one that we see from our observations. There's something wrong with these equations.

On the other hand, we also know that there's an enormous amount that's right about these equations. These equations are where, you know, indescribably vast swathes of 20th century science and technology come from. These equations get an enormous amount right. On the other hand, it couldn't be more obvious from our everyday experience of the world and of ourselves that something is wrong with them. There's a problem about how to put these two facts together. There's a problem more specifically about how to modify the theory in such a way that this contradiction goes away without ruining the rest of the good predictions of the theory. This problem, once again, is called the measurement problem.

Question: Can science give us a precise image of the universe?

David Albert: Oh, I see what you're saying. Look, that is -- throughout most of the 20th century, what was widely considered to be the lesson of quantum mechanics, what was widely considered to be the upshot, the deep upshot of our scientific investigations of subatomic particles, was precisely that: the thought that science was going to ultimately give us a picture of the world all the way to the bottom, which we were going to be able to carry around in our heads -- okay? -- which we were going to be able to understand in the way we understand billiard balls colliding with one another or something like that. It's widely been thought to be the upshot of quantum mechanics that those expectations of science have now been exposed as quaint and naïve and old-fashioned, and moreover as presumptuous, okay? Who were we to think, with these brains evolved for very different purposes of hunting and gathering and so forth, that there was going to be this kind of intelligible, mechanical model of the world that we were going to be able to get our heads around, okay?

And indeed, the response to this measurement problem throughout most of the 20th century was precisely that: look, this is where our scientific imagination gives out in its attempt to penetrate the world. We have encountered for the first time ever the ultimate limits of the capacity of the scientific project to penetrate into the foundations of the world. We're not going to get farther than this; we should be thankful enough that we have a good mechanism for predicting the behaviors of these particles, and so on and so forth. That was very much the consensus throughout most of the 20th century. And if students were to raise their hands and say, gee, how can you be so sure of this? I mean, have people tried to make these modifications, blah blah blah? -- those students would be referred to a number of famous so-called no-go theorems or no hidden variable theorems, the most famous of which for most of the century was one due to the mathematician John von Neumann.

And it wasn't until rather late in the century that attention began to be focused on the fact that these theorems and these arguments that we couldn't do better than this were hasty, were premature; that the theorems, especially the von Neumann theorem, was just a very flawed theorem. The mathematics was completely correct, but the presumptions that he started with were much too restrictive. There was no reason to be persuaded that those assumptions were true. And since then, mostly over the past 25 years, enormous progress has been made, okay, in actually proposing ways to tinker with these fundamental equations in such a way as to solve this problem, in such a way as indeed to provide us with precisely the sort of thoroughly intelligible mechanical picture of what was going on that was said for most of the 20th century to be quaint and outmoded and immature and so on and so forth.

So yeah, you're right: throughout most of the 20th century the reaction was, the lesson to take from this is that there are limits to the capacity of the human scientific imagination to penetrate the mysteries of nature; and the mature thing to do, the grownup thing to do, is to accept these limits. It isn't until recently that it's become clear that these pronouncements were enormously premature. Who knows if we're going to finish the scientific project or not? Who knows if we're every going to get to the bottom of it? But there is this really interesting episode in the 20th century where it was thought that we had. And it's now becoming clearer and clearer that the announcements of the death of this project, after Mark Twain, were greatly exaggerated.

Question: Can you give a brief overview of quantum mechanics?

David Albert: Quantum mechanics is supposed to be a completely general account of the behavior of the physical world. The way quantum mechanics emerged historically was that around the end of the 19th century, the beginning of the 20th century, there were more and more reasons to be worried that the prevailing classical physics -- that is the physics of Newton and later of Maxwell, who incorporated electromagnetic phenomena into Newton's theory -- that the Newtonian/Maxwellian model of the world that we had was apparently going to be unable to account for atomic structure, was going to be unable to account for things as simple as the stability of matter.

A famous problem that very much worried people just in the period immediately before quantum mechanics emerged was that people had experimentally seen that the atom apparently consisted of a small positively charge core with electrons rotating around it, and it's easy to show from Maxwell's equations that electrons going in a circular orbit around a core like that are going to produce huge amounts of electromagnetic radiation; they're going to lose all of the energy of their orbits to this radiation; they should very quickly crash into the nuclei, and matter should cease to exist. So there was a very acute problem about how matter could be stable at all. The more this and other related problems were looked into at the beginning of the century, the more -- the less it looked as if there was any hope at all of an explanation of these phenomena along the lines of Newtonian and Maxwellian classical physics.

It looked like this was going to call for nothing less than a global revolution in our fundamental theories of physics. And over a period of 10 years or so -- centered on the period between 1920 and 1930, say -- a new fundamental theory of the world was developed, called quantum mechanics. Quantum mechanics, then, aspires to be a complete replacement for Newtonian mechanics. It now proposes to be understood as the fundamental laws of the evolutions of the physical states of every kind of physical system. But, as it surely should have, quantum mechanics was engineered in such a way as to essentially reproduce the predictions of classical mechanics for those systems, human-size systems, macroscopic and larger systems, for which Newtonian mechanics was known to do a good job, okay? And the way it worked out was that the predictions of quantum mechanics differed significantly from those of Newtonian mechanics only for very small kinds of physical systems, subatomic systems, so on and so forth, which is exactly where we needed it to differ.

So it's important to separate how the theory was discovered and what it was originally developed for from what it proposes to be once it's there. It was originally developed to account for new phenomena that we had discovered at the subatomic level, okay? But it would be a serious misunderstanding to interpret it therefore as merely a specialized theory of subatomic objects. It's supposed to be the theory of the entire physical world, but its discovery was prompted by failures of the previously existing theory in the subatomic realm.

Question: What is string theory?

David Albert: String theory is a version of quantum mechanics. That is, quantum mechanics is less a completely specific theory than a class of theories, theories that involve this principle of superposition that we were just talking about, theories whose fundamental equations of motion have a certain particular kind of mathematical structure, so on and so forth. So one can enumerate five or six basic principles of quantum mechanics, okay, and these principles allow for a rather wide range of more specific claims of exactly what elementary systems the world is made of and so on and so forth. So string theory is one version of quantum mechanics, one version of a quantum theory. It's a theory whose fundamental ontological entities aren't particles, but these one-dimensional objects, these strings. And that fundamental ontology looks promising for all sorts of reasons, especially in regard to attempts to make a coherent quantum theory of gravitation and so on and so forth.

But in the context of our discussion here, it's one version of a quantum theory. It shares all of the weird properties that we were just talking about, the measurement problem, the principle of superposition, so on and so forth, with every other quantum theory. And these foundational problems, especially the measurement problem, come up in string theory in exactly the same way as they come up in older versions of quantum mechanics.

Question: How might we establish the truth of string theory?

David Albert: We need to smack particles together -- you know, this is of course a -- one doesn't want to anticipate what's going to happen, and maybe tomorrow somebody's going to figure out some much more clever and much cheaper experimental method of distinguishing between string theories and other quantum theories that we have -- but insofar as we know at the moment, the only way of getting a handle on whether or not string theory is true is going to be the very brute-force, very expensive project of building these huge accelerators that are going to smack particles together with such intensity, that is with energies reminiscent of the energies that the particles had in the very early milliseconds of the life of the universe, where the predictions of string theory are going to differ from the predictions of other interesting quantum theories that we have on the table. So it's going to be a matter of doing those experiments.

Question: Is the Large Hadron Collider capable of doing this?

David Albert: It's -- no, even that -- I mean, there may be -- the evidence that we would get even from that, as far as I understand the matter, would still be rather indirect, although there are things that could emerge from those experiments which would be a more comfortable fit for certain string theories than for other theories that we have on the table. But even that evidence is going to be rather indirect. You know, what most people say nowadays is, look, the best evidence we have for string theory is that there is a phenomenon of gravitation, and we don't know any way besides string theory of making gravitation compatible with quantum mechanics.

Question: Does quantum mechanics speak at all to consciousness?

David Albert: Well, it's been thought to, and presumably it does in one way or another. There have certainly been episodes in the history of struggling with the measurement problem over the past 50 years or so when distinguished physicists -- for example, Eugene Wigner, Nobel prize winner, enormously distinguished theoretical physicist of the first half and middle of the 20th century -- became convinced around the middle of the century that consciousness was going to be an absolutely essential and ineliminable ingredient of any possible solution to the measurement problem that we were talking about before. You remember that the problem was that when we rip this box open we see an electron either here or there, but the fundamental quantum mechanical equations of motion, if you apply them as well to our brains, would seem to predict the opposite, okay, that we don't distinctly see an electron here or there; rather, our brains end up in a superposition of the state associated with seeing it here and the state associated with seeing it there. That is, our brains end up in some condition where it fails even to make sense where we believe the electron to be. Okay.

Wigner took a look at this situation and said, well, so apparently what's going on here is that our brain, or at the very least our mind, seems to be evolving in a way that directly violates these fundamental equations of motion. And WIgner's approach to this was, instead of seeing this as bad news, okay, to see it as the news we've been waiting for, you know, since the beginning of science. Here is finally a proof that the mind of the observer is not a physical object and is not tied to physical objects in the way that rocks are or tables are or chairs are, so on and so forth. That is, the reason that the fundamental equations applied to our brains end up making the wrong predictions -- so said Wigner -- was because have this special additional feature of being associated with consciousness, okay?

And this had the sort of cute effect of turning the traditional mind/body problem on its head. Traditionally the worry has been that the picture of the world that's emerging from physics is hostile to mind, that there's no place for mind in it, that we can analyze everything in terms of the physics of our brains -- why I'm saying this, why I do everything I do, so on and so forth -- it's hostile to mind, it's hostile to all of these things that we associate with mind, like freedom of will, so on and so forth. Wigner thought he had an argument that as a matter of fact in quantum mechanics, precisely the opposite turns out to be true: not only is physics not hostile to the idea that there is a distinct nonphysical, mental thing intervening in the physical world; not only is it not hostile to that; it absolutely needs that in order to make the right predictions. It absolutely needs this mind to come in and violate the equations of motion in order to make this electron end up in one determine place or another, which is what we observe it doing. So Wigner thought first of all he had for the first time a clean mathematical definition of the difference between a physical entity and a mental entity. A physical entity is by definition an entity that obeys these equations of motion. A mental entity is the kind of entity that is capable of causing violations of those equations of motion. Good.

This sounds cute for about 10 minutes, but it quickly became embarrassing. I remember myself as a young graduate student being at conferences where Wigner would stand up and speculate that although dogs could likely cause violations of the equations of motion, mice probably couldn't. And it just became silly and embarrassing, and one didn't know where he was coming up with this, and one was going to be forced, in order to write down the fundamental physical laws in a clean way, to make these distinctions between conscious and not conscious; whereas what seems much more plausible to everybody is that there's some continuum going from conscious to not conscious, rather than some clean cutoff point. And it was just a mess. So this was a view that was entertained seriously for about a 15-year period from the early '50s, maybe, to the late '60s and hasn't been taken particularly seriously by physicists since then. On the other hand, the existence of this view in this earlier historical period has been a goldmine for New Age enthusiasms about quantum mechanics ever since then.

Question: What are some of the most prevalent misinterpretations of science today?

David Albert: Well, I mean, very broadly speaking, look, there are -- you know, one can contrast two very different ways of coming at the world. One can come at the world determined in advance that what one is going to find at the bottom of it is some reassuring, flattering, more or less comfortable image of oneself. Or one can come at the world in a genuinely more open way, to see what the world has to teach you, and to determine as much as one can to keep one's eyes open whether one finds what one expected to find in advance or not. There are human limitations, no doubt, to the capacity to do that, but there are some people who are more resolute about doing that than others.

And it seems to me what unifies all the New Age approaches to various kinds of scientific discoveries is that the curiosity turns out very quickly not to be authentic, okay? What one was after, it quickly emerges from the beginning, was reassurance that all sorts of things one already thought about the world and about oneself before the investigation even began are going to be confirmed. And if they're not confirmed, you're going to lie about what the science says in order to make it appear that they were confirmed. This is very much the opposite of the scientific spirit that one admires and hope to emulate as much as one can, which involves a profoundly more authentic and profoundly more courageous kind of curiosity than that.

The picture that science presents us with of ourselves, it seems to me, for all that anybody might want to say about it -- I don't know how to put this. It seems to me that the picture that we're presented with of ourselves by science is a picture -- if you really stay in it, if you really take it in -- connected with some kind of bottomless terror, okay, where the -- science presents us with a picture of ourselves as machines, okay, as billiard balls knocking into one another, at the very bottom of things, and that's why we do what we do, and that's why the world is the way it is. It seems to me that if one really opens oneself to what science has to tell you about yourself that the category of sort of esthetic reaction that it falls into is something like the uncanny, okay, especially in the way that terms is used in psychoanalysis, where it's associated with apparently animate objects that turn out to be inanimate, or the other way around. There's something really indescribably strange about the picture that we're presented with of ourselves by -- especially by fundamental physics. And it's a picture that we just don't know how to fully take in. And I think it's very, very, very disturbing. And I think that's as exactly the opposite as it could be of what typically is said to come out of science from New Age understandings of it.

Question: How is our everyday notion of time flawed?

David Albert: Well, in a number of ways. First of all, you might say that the sort of gesture with which physics gets under way in the first place is a gesture that lots of philosophers in the Continental tradition have come to call the spatialization of time. That is, the way time appears in physics is as another component, a fourth component, of the address of an event, okay? You want to say -- what there is to say about events is where they were located and how they were distributed, blah blah blah, in the X dimension, in the Y dimension, in the Z dimension and in the T dimension. And what physics aspires to tell us -- and all physics aspires to tell us, when you get right down to it -- is how matter and energy gets distributed over this four-dimensional block of X, Y, Z and T, okay? You tell me where all the particles are in X, Y, Z and T; I'll tell you the history of the world, okay? That's supposed to be, according to the model we've been dealing with in physics from the word go, all there is to say about the world.

From the word go, then, something is being done with time that does enormous violence to our prescientific intuitions about time. Prescientifically, it seems like time is as different from space as you could possibly imagine. And the way discussions of this often go is that people will immediately say, well, time moves or flows; space doesn't move or flow. I can move around at will in space; I can't move around at will in time. All these kinds of things. Time seems just as different as it could be from space, and indeed there's a tradition in Continental philosophy, in European philosophy -- you find this in people like Heidegger or Deleuze or people like that -- that it is the fundamental mistake of physics, this spatialization of time. It's from that moment on that physics had already doomed itself to not being able to say anything deep or interesting about what time was.

Physics, on the other hand, has this interesting retort that it can apparently explain why you say all these things you say about how time is different from space on a model in which time is just another parameter in the address of an event. So there is this very interesting conversation, there is this very interesting dialog, where on the one hand it looks like physics has made from the word go a profound error about what time is. Physics, on the other hand, has been up to now always in a position to come back and say, you want me to explain to you, based on physics, why you say time passes even though that doesn't actually mean anything, and why you're talking the way you do about our capacity to move around in space but not in time, so on and so forth? I can do all that. It's like -- it's very much like when Newtonian mechanics was first proposed, and it was widely objected that if the earth was indeed moving through space at the enormous velocities that it's predicted to be by Newtonian mechanics, everybody would fall off, okay? And the beauty of Newtonian mechanics was to be able to say not merely that I have an explanation of why we don't fall off, but that the very same theory that predicts that the earth is going to be moving so fast is also going to predict that we wouldn’t fall off. Indeed, it's also going to predict that we would think it isn't moving, even though it is. One wants to play this game with the scientific conception of time as well.

Maybe I should go a bit further. Specifically, there is within physics the following problem that's been sitting around for about a hundred years now, at the foundations of physics, which goes like this: imagine watching a film of two billiard balls collide. So in the first frame you just see a billiard ball sitting at the center of the frame. Then another billiard ball comes in, hits the one that's at rest. The one that was moving is now at rest, and the one that was previously at rest goes off and exits the frame on the other side. Imagine that you were shown this film in reverse, so what you're going to see is the second billiard ball coming in here, hitting the first one, which has now been at rest up until this point in the reversed movie, and now the first one goes out and leaves the frame on the other side.

Note that the set of events depicted by the movie being shown in reverse is just as much in accord with everything we believe about the laws governing collisions between billiard balls as is the movie being shown in the correct direction. That is, if you were shown a movie like this and asked to guess -- just based on your familiarity with the laws of physics, just based on your familiarity with how billiard balls behave when they collide -- if you were shown a film like this and asked to guess whether it was being shown forward or in reverse, you wouldn't be able to tell. Physicists express this by saying that the laws governing collisions between billiard balls are symmetric under time reversal, okay? And what that means more concretely is -- a law is said to be symmetric under time reversal if it's the case that for any process which is in accord with that law, the same process going in reverse -- that is, the same process as it would appear in a film going backwards -- is also in accord with that law. So we say that the laws governing collisions between pairs of billiard balls are time-reversal symmetric. Good.

It's an astonishing thing that all of the serious candidates that anybody has entertained for fundamental laws of physics, from Newton onward -- so by this I mean Newton's laws, I mean Maxwell's equations, I mean the Schrödinger equation, I mean general relativity, I mean string theory, everything -- it's an astonishing fact about all of these theories, even though they differ wildly from one another in all kinds of other respects, and even though they straddle different sides of enormous scientific revolutions, it's an astonishing fact about every single one of these theories that all of them are perfectly symmetric under time reversal. Every single one of them has the feature that for any process which is in accord with those laws, the process going in reverse is in accord with those laws as well. But this poses a problem. And once again, just as with the measurement problem, the structure of this problem is a clash between the laws of physics that we have developed by carefully observing the behaviors of microscopic systems and our everyday macroscopic experience of the world.

Okay, what's the problem? The problem is that if we get more sophisticated than collisions between billiard balls, okay -- if we imagine a film of someone walking down the street, or of a piece of paper being consumed by fire, or a time-lapse film of somebody growing old or something like that, we can damn well tell very easily whether we are being shown the film forward or in reverse, okay? Now, that is very directly at odds with what we take ourselves to have very good reasons for believing about the structure of the fundamental physical laws, which are supposed to govern all of these process; namely, that those laws are perfectly symmetric under time reversal, okay? To put it slightly differently, on the level of fundamental physical law there doesn't seem to be any distinction whatsoever between past and future, okay? Moving from the present into the future ought to look, statistically speaking, just like moving from the present into the past if those laws are true.

On the other hand, it's probably not an exaggeration to say that the most basic, primordial, unavoidable feature of our everyday experience of being in the world is that there is all the difference in the world between the past and the future. In our everyday experience of the world there is an extremely vivid and pronounced temporal direction, okay? On the level of the fundamental laws, that temporal direction apparently completely vanishes, okay? Once again, there's a question, and like I say, this question has a similar structure to the question in the measurement problem in that it's a clash between our fundamental equations of motion and our everyday experience, macroscopic experience, of being in the world. There's a question with time about how to put these two things together.

Once again, it appears as if although the theory does an extremely good job of predicting the motions of elementary particles and so on and so forth, there's got to be something wrong with it, okay, because we have -- although we have very good, clear quantitative experience in the laboratory which bears out these fully time-reversal symmetric laws, at some point there's got to be something wrong with them, because the world that we live in manifestly not even close to being time-reversal symmetric. And once again there are proposals on the table for how to fiddle around with the theory, adding a new law governing initial conditions, for example. There are all kinds of proposals about how to deal with this, but this has been -- this is a very fundamental challenge. Although we've got laws that are doing a fantastic job on the micro level, there's some way in which these laws manifestly get things wrong on the macro level, and we need to figure out what to do about it.

Recorded on December 16, 2009