VOICEOVER:         We are setting off on a journey fraught with dangerous notions: eternity, the creation of all things from nothing, and dare I utter a word not to be thrown around with reckless abandon among those of reason, infinity. A case in point: the Library of Babel. It is a library that holds every book that has ever been written and every book that ever will be written. Now here lies the madness: each book is filled [00:00:30] with random strings of 26 letters, commas, periods, & spaces. And each book is different. Every volume in every one of these galleries has 410 pages, each page has 40 lines, each line has 80 letters, each exallius – that is, randomly through nothing but chance. So for every sensible line of straightforward statement, there are [00:01:00] leagues and leagues and leagues of senseless jumbles and incoherencies. For example, one volume is made up in its entirety of the letters M, C, V – perversely repeated from the very first line to the last. Another is mostly filled with gibberish until you reach the next the last page where you see, “Oh time thy pyramids.” [00:01:30] It is unbearable knowledge that in some gallery, on some shelf, a precious, cogent book exists yet remains inaccessible. And though one cannot know, my belief is the library stretches to – dare I say it – the infinite. Many say I’m mistaken that the possible number of books does have [00:02:00] a limit. And while that number is beyond my understanding, I am gladdened by its distant hope that however improbable I may one day uncover meaning.

HOLT:                       Hello! [00:02:30] I brought my lunch. So we just saw something about the Library of Babel which is an interesting image for the universe. But today you’re going to hear from some people who will give you a much more expansive idea of what reality is. So instead of the universe being like the Library of Babel, it will be like [00:03:00] one letter and one book from the Library of Babel. So how did we get to this this point in scientific inquiry, into reality? Let’s go back about 117 years to the year 1900. What did reality consist of? What was physical reality? Everyone thought it was the Milky Way sitting in this otherwise infinite space. It was eternal, wasn’t created, and would last forever. How naive [00:03:30] that proved to be. Because about 20 years from then, in about 1920, it was discovered that the universe couldn’t be static – it had to be expanding. And this could be empirically verified. You could see the galaxies pulling away from us. So if you extrapolate back in time, you get this origin to everything: the Big Bang. So now we have a completely different view of the universe. Instead of it being eternal and static, it had a finite [00:04:00] origin in the past and it might end in a finite future in a Big Crunch, kind of like human life. You’re born (it’s the Big Bang), you die (it’s the Big Crunch) – that’s it. So a completely different view of the universe. And this was actually exciting to some people with a religious sensibility. Pope Pius XXII said, when the Big Bang was discovered, that this is scientific proof of creation. So it really overturned our notion of reality. By the way, the Big Bang [00:04:30] Theory was confirmed in the 1960s when the echo of the Big Bang the cosmic microwave background was detected almost by accident. At first, the scientists thought that it was bird guano on the antenna that was causing this interference but it turns out it’s the echo of the Big Bang. So now the question is, ‘What does this Big Bang advance?’ And by the 1970s, some guy had come along. Two of them are here, fortunately. It’s really [00:05:00] thrilling to have them here. Then they came up with the theory of the Big Bang. Let’s see – do we have images of the two in their ’70s hair? There they are. Not bad for ’70s hair. So they set out to answer the question. What is it that banged? Why did it bang? And what was going on before it banged? And they came up with a theory which is called the Theory of Cosmic Inflation. And so now, once again [00:05:30] there is a radical shift in our idea of reality. First, it was a static, finite universe: the Milky Way. Then, it was the Big Bang universe: it blows up out of nothing in this seeming creation event and then goes on its merry history. According to the new theory of inflation, our universe is just one member of this vast, perhaps infinite ensemble of universes called the multiverse. So this is where we are right now. We’re [00:06:00] going to be talking a lot about the multiverse tonight and the notion of infinity. And we’re also going to address the question, ‘What are the fundamental constituents of nature?’ And for that we have a string theorist who is going to give us a rather radical, revisionist notion of what the ultimate constituents of reality are. These are the imponderabilia that will agitate and distend your mind for the next 90 minutes. It’ll be infinity, eternity and the idea of creation, the fundamental [00:06:30] nature of reality, and also the limits of science. What, if anything, lies beyond the scope of science? And we may even get into that into the notion of free will because one of our participants has just written a really interesting book about the relationship between physics and freewill, and things that are of great human interest. So let me tell you who the these people are. By the way, this is a record. We’ll have [00:07:00] six people on the stage in addition to me, so you’re really getting your money’s worth. This ties the record for the World Science Festival. And two of them are inflationistas – the creators of the theory of inflation. They have a very expansive view of reality. We have a string theorists who takes a very radical, revisionary view of reality. We have a person who takes a rather conservative view of reality. Then we have [00:07:30] two philosophers who are going to say, ‘Hey wait a minute, what do you mean by reality?’ So let me tell you who these people are. First of all, we have Andre Linda who is – oh I’m sorry, I did this completely wrong. I was going to say, ‘Where did our universe come from? Nobody knows, but our next our first guest will tell you.’ His name is Andre Linda. He is a professor [00:08:00] of physics at Stanford. In addition to being the father of this theory of Chaotic Inflation, He’s a sleight of hand expert and can do acrobatics. And he’s done many backflips in defending the theory of inflation, I think. Our next guest is a professor of philosophy at Columbia – he’s the Woodbridge Professor of Philosophy at Columbia. Also, he has a Ph.D. in Theoretical Physics from [00:08:30] Rockefeller University, so he’s actually a physicist by training and a philosopher by lack of training. David Albert! Great. Our next guest is a theoretical physicist and a string theorist. She’s one of the founding members of the Center for Quantum Mathematics and Physics at the University of California at Davis, and she hopes to elucidate the fundamental nature of space time, particularly [00:09:00] interested in something called the holographic principle which we’ll talk about tonight. And she works in the deeper understanding of black holes and quantum information theory. Veronika Hubeny! She’s already there. Our next participant is a professor emeritus of Applied Mathematics at the University of Cape Town – investigates cosmology, the nature of time, the emergence of complexity. The co-author with [00:09:30] Stephen Hawking of “The Large Scale Structure of Space-Time” (I think that’s the book, not the thing). He’s a past president of the International Society of General Relativity and Gravitation, he was awarded the Herschel medal of the Royal Academy of South Africa by Nelson Mandela, and he won the Templeton Prize in the year 2004. George Ellis! Our next guest is a philosopher of science. He’s the only [00:10:00] non-physicist on stage. But I think if he really had to, he could probably calculate the eigenvalues of a harmonic oscillator. Is a genius? I don’t know, but he has genius hair, clearly. Barry Loewer! And finally, this is the other father of inflation, the man who really first created it in 1979. He’s a professor of Physics at MIT. Interesting enough, he discovered the theory [00:10:30] of inflation when he was a post-doc, and so if there are any post-docs out there, there’s hope for you. All you have to do is discover a paradigm-revolutionizing theory and you’ll get an actual professorship. Which he did. Professor Guth is famous for saying, “The universe is the ultimate free lunch.” But on investigation, it turns out tax and tip are not included. And you’ll hear about that. So anyway, these are our guests and if you have questions, the hashtag is #WSF17. We may be able to take some questions at the end over the iPad, not live. So let’s get [00:11:00] this thing going. We saw this thing about infinity and we have somebody on stage with a more conservative [00:11:30] view of reality, which is Professor Ellis, and he believes that nothing real can have an infinite number of components. So first of all, I want to ask you, how do you know that? And second of all, are any of you people who are committed to a reality that’s infinite in some respects, like the multiverse, going to take issue with that? So how do you defend that? How do you know that?

ELLIS:                       I’ll take a double stand on this. First, for me it’s a philosophical principle and philosophical [00:12:00] principles underlie what we do in science. And I’m going with David Hilbert, who said the infinity never occurs in physical reality and that’s because infinity is beyond what we can ever attain. Doesn’t matter how many things you’ve done, how much stuff you’ve counted, or how far you’ve gone. You haven’t taken the first step on the road to infinity. And I think people forget that when they talk about infinity in loose ways. But more specifically, if you claim there’s an infinite number of anything in the universe – galaxies, people or something – it’s scientifically [00:12:30] unprovable because firstly, you can’t see them. But secondly, if you could see them, you could never prove they were infinite because no matter how many hundred thousand million million million million you have proved, you haven’t taken the first step on the way to proving there’s infinite. Now, I believe a scientific statement should be provable, so I think any statement about science that involves the word “infinity” is not provable. Therefore it’s not science.

HOLT:                       What if space were infinite, and it were uniform, and there were galaxies uniformally distributed throughout space? [00:13:00] Are you saying that’s metaphysically impossible? Surely someone is going to object to that.

ELLIS:                       That is a “what if” statement. So, what if? Yeah, well that might be true. But we can see up to 42 billion light years away, we can’t see anything further. So you might say it’s infinite. I might say, you know, it isn’t, it closes up on itself on a finite radius. Or I can say that it’s bounded by singularity at a finite distance. You can’t prove you’re right. I can’t prove I’m right. You can say anything you like, I can say anything I like.

HOLT:                       By the way, that’s a very controversial [00:13:30] statement because there are many cosmologists now who believe that our part of the universe is infinite, it’s flat, and it’s uniformly filled with with matter – including Brian Greene, I think, who is the father of this whole thing. But you know, we have two proponents of the multiverse here. Surely you’re going to take issue with this rather metaphysical claim that he’s making. Or maybe not.

GUTH:                      I think I’d like to make important distinction between what we view as space and what we view as time. Space, I think we [00:14:00] have no way of knowing directly whether it’s infinite or not. As you say, we can only see a certain distance and we speculate beyond that. Same thing is true about time, except for time evolution we kind of have a theory: the Schrödinger equation of quantum mechanics. And quantum mechanics does not seem to have any natural way of ending time. Therefore, I think the reasonable assumption to pursue is that time is infinite.

ELLIS:                       Let me make a correction. The correct statement is not that time is infinite. The correct statement is that time will [00:14:30] be infinite in the far future, which we will never reach. Because no matter how far time has gone you haven’t reached the far future where it is infinite. It’s always in the future. So it isn’t true that it is infinite.

GUTH:                      I don’t think distinction has any real importance to us. I mean, spacetime exists. We talk about space time. And if we know it will exist, that’s as good as – I don’t know what the tense “now” means in the context [00:15:00] of a spacetime picture of reality.

HOLT:                       We need a philosopher. Barry, want to go first?

LOEWER:                I just have one remark from philosophy. Aristotle, that guy a long time ago, made a distinction between the actual infinite and the potential infinite. I think George’s view is that there’s maybe a potential infinite, there’s no boundary. Alan’s view might be there’s an actual infinite. As far as I can see, there’s nothing contradictory about space or time [00:15:30] actually being infinite. We know there are models in mathematics which are consistent and which are actually infinitely many abstract entities at any rate (numbers). Whether any scientific theory needs to posit the existence of an actual infinite is a open question, I think.

GUTH:                      I think I would like to add one important statement there which is that given what we call in physics “Coleman-de Lucia Tunneling” – technical word – if you have an infinity of time, you’ll automatically get an infinity of space [00:16:00] coming along with it. Space and Time are more or less interchangeable in the context of general relativity.

ALBERT:                  I can pick up on that. I mean, science is full of claims. You know, current science is committed to all kinds of claims about what sorts of laws things are obeying in the centers of distant stars that we’re never going to be able to witness. But we take it to be a natural extrapolation [00:16:30] of things that we take ourselves to have good reason for believing. So the observation (which is surely true) that there’s never going to be a direct empirical proof that the world is infinite or that the world is finite. Doesn’t seem to me to be equivalent to a claim like, ‘We couldn’t imagine ourselves having what we take to be good scientific reasons,’ for supposing that the world is infinite or finite in this or that respect. I mean, Barry just mentioned one. We have the Schrödinger equation. We have all [00:17:00] sorts of good empirical reasons for having faith in the Schrödinger equation and the Schrödinger equation doesn’t present us with a natural opportunity for the time parameter to come to an end.   So it seems like a plausible thing to say that it’s infinite. You’re absolutely right in pointing out that that’s going to be one among you know uncountable numbers of [00:17:30] claims about the world that our science is committed to which we’re never going to be able to directly empirically confirm or refute. You seem to be equating that with a claim that ‘We couldn’t have anything that counts as a good scientific reason to believe one way or another about such claims’ and I guess I don’t see how that follows.

HOLT:                       And Andre, you are the father of the theory of eternal inflation. And eternity has kind of an intimate relationship [00:18:00] to infinity. Give us your take on this and make it funny.

LINDE:                     I can take the infinitely large amount of time for that, but I probably will be stopped. And that actually is similar to what actually happens in our universe. Most probably now. Right now, our universe tends to be – well, at least started at about 5 billion years ago – exponentially expanding. And if this continues, then our [00:18:30] part of the universe eventually will become empty. And we will find ourselves in something which is called “visible space.” And visible space is the largest distance from which we can ever get any information. This is called the visitor’s horizon. So the whole universe, the whole visible universe, may be infinite but only part of it will be accessible to us. So in a sense, then you’d say, “Okay, so what is, then, the multiverse if you have just a finite part of everything to explore?” On the other hand, if we study [00:19:00] it a little bit deeper and study what happens in string theory, for example. In string theory, [there are] a lot of different vacuum states. And our vacuum state may decay in the future. And when it decays our horizon may expand and expand and become infinite. Or we may all collapse and die instantly. So there are some possibilities, which right now [00:19:30] we cannot even fully foresee – all of this picture of the universe didn’t exist (in the sense which we will proceed with right now) about 35 years ago. So two comments: I would forbid this picture being discussed because it – well, against my general principle, this would probably be…well I would warn us against following the lead of those who want to forbid something. But they don’t [00:20:00] think that George wants to forbid. He wants to put the question mark here: ‘be careful because there is something we should be taking into account seriously,’ and I absolutely agree with this position.

ELLIS:                       I particularly want to put a question mark against statements coming out in the popular journals saying, “How far away is your next self?” And I just think to talk about yourself having infinite replications is kind of really stretching beyond-

HOLT:                       Right, and this notion of universe – ‘there are exact replicas 10 to the 10 to 100th angstroms or light years – it doesn’t [00:20:30] matter that scale – away at all variations, yeah.

LINDE:                     So George, do I understand you correctly that you decided to go against infinity because you want to avoid the possibility of Horov meeting yourself doubled.

HOLT:                       I think you’re right [00:21:00] .

HOLT:                       So Professor Guth, when you originally came up with the theory of inflation in 1979 it was to solve some puzzles about the observable universe: its smoothness, its uniformity, the absence of magnetic monopoles and so forth. And your original conception – it was simply a sudden inflation in a single [00:21:30] universe very early on, just ‘first second after the big bang, in 10 to the -30 seconds there was an expansion of 10 to the 30th power – something like that. But it was a single universe right?

GUTH:                      Yes.

HOLT:                       So that was 1979. And you, in 1983, came along with a variant of the theory of inflation that solve some technical problems in your theory that implied that this was an eternal [00:22:00] process and it implied the existence of a multiverse. So am I correct in that bit of history?

LINDE:                     Approximately.

HOLT:                       Well that’s that’s a lot of journalists that’s all.

HOLT:                       And so this was a really radical revision of our notion of reality. I mean first of all, inflation tells you that the observable universe is a tiny tiny fraction of our pocket universe. And then, all these [00:22:30] other pocket universes – and the whole scheme keeps regenerating into the inf – I’m sorry, I won’t say “infinite future” – eternally.

HOLT:                       So did it have a beginning?

LINDE:                     Be careful with words, because “multiverse” is used like a slogan by many. And if you try to Google “the multiverse” – who studies it, how the tendencies… tendency grows up all the time. And then you will do a more careful search, “who actually…” dialing on Google… It appears [00:23:00] that most of the people who do it are from Peru.

LINDE:                     And then the next position is from France.

LINDE:                     So then, you know, ‘What the hell is going on here?’ And then you will fine-tune the search, and it appears that people in Peru are playing the game “Dragon Multiverse.”

LINDE:                     [00:23:30] What kind of game do we play? And this is actually very, very simple. Just think about our own Earth. There are many, many countries on our Earth. Each country has its own laws, okay?

LINDE:                     China, Russia, the U.S.

LINDE:                     And then imagine that the size of the universe blows up to [an] incredible size and that is the main idea [00:24:00] of inflation.

LINDE:                     So you have China so large that anybody living in China would have no idea about the United States. And the United States would not have any trade war with China because it will be a not infinitely, but practically infinite distance. And there would be no Internet connection or whatever. So this is a very simple very practical idea.

LINDE:                     And then somebody comes and says ‘Oh I live in the place where this set of laws [00:24:30] and the set of laws is unique and other laws are impossible.’ And if I have water then it’s liquid water. And the captain of the ship just takes the ship, and then suddenly at night he crashes into solid water and the Titanic goes down.

LINDE:                     So the fact that we have seen something experienced something does not necessarily prove that something else is not possible. And a broader set of possibilities (that is, a larger set of rules) is more natural from my perspective than forbidding something beforehand.

HOLT:                       [00:25:00] I’m dying to bring Veronica into this but I don’t want to bring you in prematurely. As a string theorist – and, of course, the string theory feeds into the idea of the inflationary multiverse because it produces something called a “landscape multiverse.” There are 10 to the 500 different ways of threading flux lines through Calabi-Yau spaces. And each of these corresponds [00:25:30] to a world where there are particular properties, forces, and masses of particles and different Higgs fields and so forth and radically different physical worlds. And so you have a slightly coy and cagey relationship to string theory because you see it more as a tool rather than as a fun a real description of reality. But I’ll let you talk.

HUBENY:                That’s right. [00:26:00] I mean I view it as a framework which allows all these possibilities depending on what you put in. So it doesn’t really bother me that the framework allows all these possibilities and whether or not there’s infinitely many of them doesn’t threaten our notion of reality and so forth. But let me put in a word for infinity in a context of, ‘Can infinity be useful in physics?’ So far we have been talking about, ‘Is our universe infinite?’ or something like that. And I sympathize with all that has been said on stage here, that if you count physical objects then of course it can never get infinitely many of them. But it might [00:26:30] still be mathematically useful notion in the sense that if you allow me zero number of certain types of objects, then I can just say well let me talk about one over that. That’s a [00:27:00] bit silly example but it’s meant to illustrate that there could be different descriptions of the same physical reality, and those differing descriptions could be useful for various different things. And it could be that in one of these descriptions the thing that’s most natural is the one over what ends up being zero.

ELLIS:                       I think this is a very important point. I think just as infinity [00:27:30] doesn’t exist in physics, zero doesn’t exist in physics either. And that’s what the quantum vacuum tells us.

HUBENY:                I could take one minus that and…anyways.

ELLIS:                       I don’t believe that between my fingers there’s an uncountably infinite number of physically different points. And that’s equivalent to saying that those points have zero size. In fact, in the real universe I believe that spacetime must be quantifiable, have a little finite size, and then there won’t be an infinite number of points between my fingers [00:28:00] .

ALBERT:                  Two things I wanted to add to this conversation. One was about what Veronica just said in this discussion with George, and that is if you can count to infinity if it took half as long to count the next thing as it did the previous thing then of course that would be a series that converges to infinity. I’m sure George thinks that that’s an impossibility. The other thing that I-

HOLT:                       Does everyone get what he’s talking about? You can count to infinity. Say it takes [00:28:30] you one second to say one, a half of a second to say two, a quarter of a second say three. And so you run through all the numbers in two seconds that way.

ALBERT:                  The other thing I noticed was the different attitudes that Andrei and George have to what exists. Andrei’s idea is, ‘if it’s not forbidden, it’s there.’ George’s is, ‘you’ve got to show me it’s there,’ that it’s there only if it’s demonstrated to be there.

ELLIS:                       His is a philosophical position, mine’s a physics position.

LOEWER:                Given what [00:29:00] I know about Andrei’s background-

HOLT:                       “Philosophy” in physics is shorthand for “waste of time.”

HOLT:                       Andrei, you were accused of being philosophical. Those are fighting words.

LINDE:                     I came two years from Russia and one of my friends there expressed his attitude of the difference between how things work in different countries. And he said in Russia, everything that is not explicitly allowed is forbidden. [00:29:30] In the United States, everything that is not explicitly forbidden is allowed. And I much prefer this last possibility.

ALBERT:                  So I think that’s really very good in politics, maybe not so good in epistemology.

HOLT:                       So we’ve done infinity [00:30:00] and eternity now, so let’s move on to something – how about creation?

HOLT:                       Andre and Alan both years ago had a really clever creation ex the slow scheme and I understood it and I thought, ‘I don’t know whether it’s true or not,’ but it’s an alternative to God bringing the world into existence and all of that. And the idea was that out of sheer nothing a little nugget of false vacuum could [00:30:30] quantum tunnel into existence. And by sheer nothing, they mean something very particular: it’s a closed spacetime with zero radius. So it’s like taking a balloon and shrinking it down to a point. And that’s the closest you can get, mathematically speaking, to defining nothingness.

HOLT:                       So out of this beginning of nothing, a little nugget of false vacuum quantum tunnels into existence, and then by the magic of inflation this blows up into the multiverse. OK. And I think Lawrence Krauss has made [00:31:00] much of this notion. But now everyone has stopped believing in it and you believe that the universe the multiverse didn’t have a in origin in a finite time in the past. It’s eternal looking back into the past and eternal looking into the future. And you know, I frankly don’t understand this. Why were you wrong about saying it had a finite past in the first place? Alan, you had a mathematical proof that the universe had a finite beginning [00:31:30] in the past. And now – what was wrong with the proof? Why have you completely changed your mind on this, and why should we believe you on global warming?

GUTH:                      Well nothing was wrong with the proof. What’s wrong is your summary of the argument.

GUTH:                      The statement was never that the universe necessarily had a beginning. The statement was that if we follow the era of inflation backwards, the era of inflation had [00:32:00] to have a beginning. That we still think is true.

GUTH:                      And certainly also I think you exaggerated tremendously when you said that now all of you people don’t believe that the universe had a beginning. As far as I know, I’m really just speaking here for myself and Sean Carroll, who’ve working together on a model of the universe that would be eternal. And in this model that Sean and I are working on – it really was originally proposed by Sean, to give credit where credit is due – we’ve been living in an era where time is going forward [00:32:30] and entropy that is disorder, which we think of as almost a proxy for what defines the arrow of time this disorder is growing towards the future and getting smaller towards the past. And in these models with a definite beginning the universe would have begun at some finite time in the past with essentially zero entropy, zero disorder. And that’s still a possibility; I don’t mean to say that that’s not possible.

GUTH:                      What appeals to Sean and me about this alternative approach has to do with symmetry in physics, symmetry [00:33:00] of time reversal invariance. All of the laws of physics that we know of are time reversal invariant. That is, we here can be sitting on the stage and talking and time is going forward. And there is also a quantum state where all of that is reversed and everything is happening exactly the same, but in the opposite order. The laws of physics don’t care about distinctions between future and past. And what one puts in the beginning of this sort. You’re putting in a distinction between the future [00:33:30] and the past. We have a beginning in the past. So the goal of the picture Sean and I are pushing – which we haven’t published yet so I probably shouldn’t even really be talking about this, but here I am – the goal is to produce a picture of the universe or a multiverse where nowhere does one put in anything that violates this symmetry between future and past. So the way the model works, there’s an intermediate period which you might think of as the logical starting point [00:34:00] where the arrow of time is not well defined. Disorder might be growing, might be shrinking, doing one thing one place, one thing a different place… just chaos, basically. But then if you follow that forward in time, there are laws that we think we understand which say that disorder will start to grow. Entropy will grow. An arrow of time will develop going that way. And we also have in mind that all of this is going to be undergoing inflation, and in fact eternal inflation. If you follow the same thing backwards in time, the same thing happens but in the opposite order. The laws of physics, as I mentioned, don’t care [00:34:30] about t versus -t. So if you follow it backwards in time, entropy, disorder starts to grow in that direction. Inflation happens backwards in that direction. So you have this two-headed arrow of time picture, and the beauty of that is the whole system is completely time reversal invariant. Vroom. Same thing. Back. Vroom. So you have a nice picture of the multiverse which automatically contains an hour of time [00:35:00] which was not put in by hand but arises naturally just from the evolution of laws of physics that we think we already understand.

HOLT:                       David, you’re the natural person to critique this because you – this is the man who coined the term “the past hypothesis,” which is a very simple coinage, but it’s really caught on. The past hypothesis – explain what the past hypothesis is.

ALBERT:                  Although I don’t think it’s in such a dramatic conflict with this, usually before we were [00:35:30] considering theories of the kind that were just described, we used to think that we needed to make a special posit to the effect that there was an initial state of the universe where the entropy was very low. What’s going on in Sean’s proposal is that the ideas being floated, that maybe the cosmos doesn’t have anything like [00:36:00] an equilibrium condition that is the cosmos, that there isn’t any upper bound to the entropy that the cosmos can have. So wherever you start out entropy is going to go up from both sides of that, it doesn’t matter what kind of state that is. That’s still going to be a situation in which on either side of this minimum, you have the equivalent of a past hypothesis. You’re just not claiming that things [00:36:30] can’t be traced back even behind that. So this seems to be not something that contradicts the past hypothesis, it’s an attempt to put something like a past hypothesis in a larger framework.

HOLT:                       By the way, Alan, I had a conversation with Sean Carroll a couple of years ago in Los Angeles, and I said, “Do you believe [00:37:00] that the universe has a finite age?” And he said, “No.” And I said, “Well, you know Alan Guth proved that it does.” And he said, “No, in fact I’m working on a paper with Alan Guth to show just the opposite.” And I thought, you know, physics is like Hollywood. Nobody really knows anything.

HOLT:                       You know at the moment there’s this huge controversy over inflationary [00:37:30] theories in the multiverse. We have two of the creators of the theory of cosmic inflation here on stage.

HOLT:                       Arguably the third father of the theory of inflation is Paul Steinhart, who’s the Albert Einstein Professor of Physics at Princeton, and he’s now become an adversary of the theory of inflation. And in the Scientific American I believe, there was an article published by Steinhart and some collaborators arguing [00:38:00] that the theory of inflation was a bit of a failed research paradigm essentially. Maybe I’m putting it too dramatically. And there was a response by a couple of dozen physicists including the two of you, and including Ed Witten, the father of strength th- a major string theory figure, and Steven Weinberg, the Nobel Laureate, saying “No, no, no, inflation isn’t a failed paradigm.” Can someone say something about- I mean you two are both inflationists so you’ll have to shut [00:38:30] up for a second. I want someone to say something skeptical. Tell us why there is so much suspicion about the theory of inflation among perfectly reputable physicists. Barry’s a good very good friend of Steinhart’s so he can give the position.

LOEWER:                I know Steinhart and I know the authors, but I’m not a partied to that because I’m not a cosmologist. But it does remind me of a famous remark that Lev Landau, the famous Russian physicist, made about cosmologists. I think that remark goes something like this, “Cosmologists: often in error, never in doubt.” I think all [00:39:00] cosmologists must be taught this at the very beginning, so-

HOLT:                       Now Barry, because you got to make that joke, and it wasn’t even your joke, explain to us why they’re unhappy. I mean, there are two reasons as far as I understand it: there are intrinsic problems with the theory of inflation. In a way it’s too elastic. There are too many versions – there are hundreds of proposals for a theory of inflation, and they [00:39:30] all hypothesize a field that does just what you want the field to do and there’s no internal evidence that this field exists. And the other problem is that it implies the existence of a multiverse in which anything that can happen will happen. So how do you test the damn thing. That’s a very crude version.

LOEWER:                I think three reasons – I think there’s reason to think that the kinds of fields responsible for inflation could exist. And so I don’t think that that’s their objection so much. But I do think [00:40:00] that there are objections that are interesting to philosophers that you mentioned. One is that as I think you mentioned earlier, inflation was introduced in order to explain certain things about what were thought to be the initial conditions in the early universe, like that the universe has a very flat geometry, that the universe is very homogeneous. And these need to be put it in by hand in the Big Bang account without inflation. Inflation looked as [00:40:30] though it was giving an explanation of what looked like fine tuning. But their worry – and also Roger Penrose’s worry – is that more fine tuning is really needed to get inflation going in the first place.

LOEWER:                That’s one worry. Another is that inflation is really a whole model a big framework of theories with a lot of free parameters in them and by changing the values the parameters the potential energy of the field and the curve of the field and so on, you can get [00:41:00] it to adjust various empirical consequences and that that objection seems somewhat at odds with their other main and very interesting to a philosopher objection, and that is if inflation generically leads to internal inflation, the production of all of these pocket universes, and if in these pocket universes can be very different, maybe even fill up the whole landscape of string theory, as I think some people think then it looks like [00:41:30] the theory itself is predicting that anything that can happen will happen as Alan has been quoted as saying. And it looks as though a theory like that just can’t have empirical evidence for it.

LOEWER:                And that’s a really interesting issue to hash through, maybe a little bit too in house for here, I don’t know, but it’s-

HOLT:                       I want to give you an idea of the of the different proposals for theories of inflation. Yours is the first – you use the term “inflation” first to [00:42:00] refer to this cosmological theory.

HOLT:                       Then you came up with “chaotic inflation” which became “eternal inflation”, Steinhart was “extended inflation”… and then we have also “double inflation”, “triple inflation”, “hybrid inflation”, “mutated hybrid inflation”, “tilted hybrid inflation”, “hyperextended inflation”, and, invariably, “supernatural inflation”. And there are hundreds of others, not all of which have names. I’m sorry-

LINDE:                     You’re listing them too fast because [00:42:30] there are much more.

LINDE:                     But what is important about them is that all of them are just versions of one general principle. And this principle is that in the very beginning of the universe, for whatever reasons – and we can speculate whether it was tunnelling from nothing, as Alex Vilenkin suggested, or some features of wave functions of the universe, whatever – there is a region of space, a tiny, tiny region [00:43:00] of space filled with some special kind of matter which is not very specific. It’s like Higgs field, which was just discovered…nothing special about that. And then this space, if you follow the equations of motion which anybody can solve, it’s really, really simple. Start expanding exponentially. And then after that, it produces a big chunks space which is much bigger than people expected before, and after that decaying field for all matter in the universe [00:43:30] . And it looks like science fiction, you solve the equation, you see that’s how it is. So that’s the basic principle.

LINDE:                     After that, you’ll start fine tuning. What if I need to adjust some parameters or the theory to describe what our data, the experiment is going to bring us. And what if I tried to implement that in string theory, super gravity a grand unified theory? That’s why many people suggest many different versions. They are based on the same principle. Now, let’s compare [this] with the [00:44:00] history of standard model of elementary particle physics, describing weak and strong electromagnetic interactions. Just in describing weak and electromagnetic interactions requires 20 parameters to adjust the theory to describe experimental data. Most of the Higgs boson are very large, nobody could predict it. Most of the quark were very large, but people knew that quark work is necessary [00:44:30] . So what is necessary to achieve is some framework, and then feed the data. At the moment there are some inflationary models which require just a single parameter to describe all inflation-related data which Planck circulates and other experiments give. One parameter is enough. Is it complicated? I don’t think so.

ALBERT:                  Let me just press a point that Barry already made though that might [00:45:00] lead to to an illuminating discussion. Look, there are people – and maybe I should issue a disclaimer here – in past occasions when I’ve appeared on panels at the science festival, they’ve been panels set to discuss a topic on which in some sense fell within my professional areas of expertise. It’s not at all true here and I take it I’ve been invited [00:45:30] here in a different spirit to sort of ask questions or to express kinds of puzzlement that might be on the minds of people outside the field who are doing their best to understand what’s going on and who are intrigued and curious about the talk they hear. So in that spirit: look, here’s something that people may be puzzled about. One hears two claims about inflation as it’s practiced now. One, that there [00:46:00] is spectacular, enormously impressive agreement between its predictions and what we see. Second, that it’s a theory that predicts that everything that can happen, will, OK? And will for certain. And there is prime aphasia, an obvious puzzle about how to put these two remarks together with one another. If [00:46:30] you go back to your analogy of the Earth expanding, there are all these different countries, they have different laws, there are an infinite number of them. In fact they have very very different laws. Things are very different going from one country to another. Somebody is about to open his eyes and would like information from science about what he should expect to see. It seems like if you have if you have a situation [00:47:00] that the picture of the world is there’s this enormous earth on which there are an infinite number of places and everything describable is happening in one or another of those places, this is going to give the person exactly zero information about what he should expect to see. And in this case it seems puzzling. Good. So presumably you understand the puzzlement. Good.

HOLT:                       Alan, you want [00:47:30] to answer this but feel free to jump in any moment. Go ahead.

GUTH:                      Well, since I’m the one who keeps getting quoted saying that “in the multiverse, anything that can happen, will happen,” an infinite number of times – let me just clarify that sentence. If something has a probability of 10 to the -6, it means if you do something 5 times, it’s very unlikely to happen. If you do it 10 to the 6 times, it’s likely to happen once. If you do it an [00:48:00] infinite number of times, it has to happen, in fact, an infinite number of times. But the probability is still 10 to -1/6. If you only make a finite number of observations, the probability of you seeing that is very small.

GUTH:                      As I like to say even if there is a multiverse, unfortunately the probability of my winning the lottery tomorrow is not high. It remains small. But-

ALBERT:                  But Alan, let me just jump in there for a second. What you’re saying implies that the picture comes along [00:48:30] with some set of probabilities of this being the world that’s realized or that being the world that’s realized. And my understanding of the current status of the theory in which people are looking for a so-called “measure” over these worlds, which would provide us with with the probability you’re talking about, is an element of the theory which is so far not in place. Which we’re hoping is going to be in place and is going to give us [00:49:00] the right probabilities, is going to identify what we see as a high probability kind of an event. Would somebody be wrong to say, “Oh, I see so the hard part hasn’t even begun yet. All of the physics of this is going to rest in the choice of a measure. That’s an ocean of uncertainty onto which we haven’t even dipped our toes yet.” Would that [00:49:30] be a fair thing to say?

GUTH:                      I think it would be a slight exaggeration. I mean a lot of what Dave just said is true. There is a measure problem, and I was just going to get there honestly, I really was-

HOLT:                       Whenever physicists talk about the measure problem they’re talking about a fiendishly difficult problem having to do with defining probabilities when you have an infinite number of possible events, and all physicists tend to skirt it. Because it [00:50:00] does seem to be insoluble in every context in which it arises. But so the measure problem means a problem with dividing probabilities. That’s all. Go ahead.

GUTH:                      I agree with that clarification. So people who work on inflationary cosmology are aware of this measure problem. When we say it’s an unsolved problem, I don’t know. It’s either an unsolved problem or a problem for which we have multiple solutions. Those are kind of equivalent statements, more or less. We don’t yet know what the right solution is. [00:50:30] We certainly do have proposals which define probabilities in ways which seem to be perfectly consistent with everything that we know of. So we think of this decision of how to define a measure as an empirical decision. We do have ways which work empirically. I should also emphasize that there’s no doubt that it’s possible to define probabilities on infinite spaces. Mathematicians have been doing that for ages – for a century, anyway. There are many choices usually. And that’s the issue here. So we do have to find the right choice, but there doesn’t seem [00:51:00] to be any problem in finding choices that are very sensible. What we still need, and hope to find, is a way of definitively determining what the right measure is.

HOLT:                       We want to get to Veronika’s revisionist view of reality soon. So let’s wrap this up. You have something to say and you have something to say and you’ve been preparing an annihilating remark, I can tell. You’ve got this mischievous look in your eyes, you always do. So you, and then George. [00:51:30] So let’s do that quickly and then we’ll get to Veronica because… feel free to interrupt.

LOEWER:                So Andrei is going to make an annihilating remark? He wants to bring us back to nothing.

LINDE:                     I’m talking? OK. So I know when I was born. I know where I was born. I don’t know what was the probability [00:52:00] that I was born in Russia as compared to being born in China, or in New York, or whatever. Well I know that after I’d been born, I start asking questions. Once I look around, if I see everybody around speaking Chinese, that would be a problem, that would be an invasion. So there is something, some stage where I start asking questions on the basis of something previous. If I am looking on our universe, I know something [00:52:30] about our universe. At this moment, starting from this moment. The measure problem doesn’t apply already. The measure problem is mostly, ‘what is the part of the universe where I first opened my eyes?’ After I open my eyes, my wave function, in some interpretations, is reduced to the part of the multiverse where I’m going to live in some other interpretations. I just happen to be in [00:53:00] the universe – whatever interpretation it is. Starting from this moment, this problem disappears. And then after that, what appears? Well you still have a problem to predicting, on the basis which you assumed about your part of the universe. What are going to be the consequences of that? And these consequences are more or less easily predictable and so happened that for whatever mysterious reasons, it could be just luck. But for whatever mysterious reasons, cosmologists made lot of predictions [00:53:30] based on inflationary theory. And so far, so good. So either we just dreamed right answers or we some somehow know how to get it.

HOLT:                       That seemed very profound to me, but I am still mystified by it.

HOLT:                       Would anyone like to add a clarificatory codicil to that?

ALBERT:                  So, a couple of follow [00:54:00] up questions. First of all, it doesn’t seem like – so the talk about choosing a measure is a little bit mysterious and I’d like to know more about it.

ALBERT:                  I take it if we really had a full and satisfactory theory in front of us, then, as in the ordinary quantum mechanical case, the right measure is going to follow from the dynamics from whatever dynamics generates the new pocket universes. They’ll be a definite – presumably a “Born rule” – kind of probability [00:54:30] given the evolution of the universal wave function, of what the probabilities are of producing this kind of pocket universe, or that kind of pocket universe. So I’m a little puzzled by talking about finding a measure or choosing a measure. If we really had the theory in front of us, the theory would there wouldn’t be any choices of the measure. The dynamics of the theory would give us the measure. Am I right in thinking that?

GUTH:                      I don’t think so.

ALBERT:                  I see.

GUTH:                      Here we do [00:55:00] think we do have the full theory. I mean, we don’t know the details of the full theory, of course. But we can invent a full model theory of the multiverse that we’ve made up a lot of the things to fill in the blanks, and it still has this measure problem.

ALBERT:                  But I don’t get it. So we’ve got a complete theory that says, how the, say, wave function of the entire universe evolves?

GUTH:                      Correct.

ALBERT:                  It gives us every detail? In principle, there’s no question about what’s physically there? But there remains [00:55:30] a question about what the probability is of our seeing this or that.

ALBERT:                  So I’m puzzled by this. Once we’ve said the totality of the history of the world, how could there be something that still remains unspecified?

LINDE:                     George, can we see the totality of the world?

ALBERT:                  His claim was that we have a theory that gives us the complete dynamics.

LINDE:                     Whose claim?

GUTH:                      My claim. Not that we have it, but that we could model the multiverse.

LINDE:                     Alan? No, he never said that.

ALBERT:                  That’s what he just said!

GUTH:                      Well no, I didn’t say we had it, he was close. I certainly did not [00:56:00] say that we had such a theory. I just said we could model the multiverse, filling in the blanks with guesses. And we can make definite guesses and we still have the measure problem.

ALBERT:                  Right, that is – let’s be more clear. The question was, if we had such a theory in front of us-

LINDE:                     Which includes measure? Measure is not a part of the theory?

ALBERT:                  No. No. It gives us the full dynamics of the theory.

LINDE:                     And quantum mechanics, were the rules of quantum-

ALBERT:                  With the rules of quantum…

LINDE:                     So rules are [00:56:30] a part of the theory?

ALBERT:                  Yeah, the Born rule is a part of the theory.

LINDE:                     And measure is a rule.

ALBERT:                  No no no, fine.

LINDE:                     Measure is the part of the theory, right?

ALBERT:                  So maybe there’s no disagreement here. If we had a full quantum mechanical theory of the evolution of the wave function of the world. Alan, I thought you were just saying there would still be a question that remains yet answered. And it sounds like you’re denying it.

GUTH:                      What I would say is that this full [00:57:00] wave wavefunction would tell you what the probability might be of finding something here or finding something there.

ALBERT:                  Indeed, what it WOULD be.

GUTH:                      What would what it would be, OK agreed. But still, if you ask – suppose all I know is my situation: I’m seeing this room, I was born at such-and-such time, and so on. There’d be an infinite number of copies of such a situation in the multiverse. And therefore there’d be an infinite number of copies where people will hear the word “yes” following whatever time we start, an infinite number will be “no”, and you need a rule to [00:57:30] compare that. That’s the measure.

ALBERT:                  But do you think such a rule is going to be part of the list of physical facts about how the world is?

GUTH:                      Well, it might be an independent law of physics, yes.

ALBERT:                  Huh. But it wouldn’t be a law that says anything about how things physically stand in the world, it would be a new and very mysterious kind of law.

LINDE:                     What do you mean by “world”?

ALBERT:                  I thought philosophers were supposed to say stuff like that.

LINDE:                     No, I’m not kidding. Either it’s the observable part of the universe-

ALBERT:                  No. Not the observable part. The world in general.

LINDE:                     About the world in general, very hard [00:58:00] to say anything.

ALBERT:                  I thought that’s what we were talking about.

LINDE:                     But about the world which was once observed, you can say quite a lot.

HOLT:                       Barry, make a final comment on this, and then we have to get on to the nature of reality.

LOEWER:                Well I want to say a couple of things. One is that I think Andrei asked a great question. Sometimes I think to say what the difference between philosophers of physics and physicists is, is that it’s [00:58:30] your job as physicists to come up with accounts of what the laws are; the job of philosophers of physics is to say something about what laws are. These are two different things. I do think there’s a way of thinking about laws and probabilities, which gives an answer to David’s question more in line with what Alan might be thinking. In fact, we’ve been there before and even been talking about it just 25 minutes ago, talking about the increase in entropy. So the second law of thermodynamics [00:59:00] is something that’s added to the universe and people were very puzzled, because how could it be added to the universe given that the fundamental dynamics with the dynamical equations of classical mechanics? And particularly since there seem to be a mismatch since, as Alan was saying, that the fundamental laws of time reversal invariant, in classical mechanics as well as in quantum mechanics. While the [00:59:30] second law of thermodynamics has a temporal direction in it. And Boltzmann solved that problem by adding a probability distribution to the possible trajectories that classical mechanical particles could travel on in the universe.

LOEWER:                One that that needed to be supplemented by – you mentioned, David – adding the past hypothesis to it. So it sort of brings things together. The question, though, that David was pressing is ‘I just don’t [01:00:00] see what there could be in the inflationary picture that could play the role that is played in classical mechanics that gives rise to that probability distribution.’

LOEWER:                So the question that a philosopher would really be interested in, is ‘What kind of fact about the universe could it be’ – this is the question that David is pressing – ‘that makes it the case that that measure is the right measure?’ I don’t mean that that’s the right one to think about, or that that’s the one we believe in. That’s what [01:00:30] Andre was really saying. Andre was saying, ‘Well we know what the right measure is because it’ll give us a universe like ours.’ But what fact about the universe makes that the right probability? That’s really a question about the meaning of probability.

HOLT:                       OK, let’s get away from the measure problem. It’s just too hard. And let’s get on to something simple, like anti-de Sitter space, conformal field theory, duality, and the holographic principle. Which you have to talk about because we haven’t heard enough from you.

HOLT:                       These guys have been gassing on forever.

HOLT:                       [01:01:00] You have two interesting takes on the fundamental nature of reality: one take is, you’re not sure that space and time – the arena of reality – are themselves fundamental constituents. They might emerge from something at a more fundamental level. And then the other-

HUBENY:                Oh, I wouldn’t say “might”, I think they will-

HOLT:                       The other thing is this notion of duality, that we have two ways of making a theory about the world: one is the theory in the bulk, like inside the room, and then the theory of what’s going [01:01:30] on on the walls of the room. And both of these theories – the theory of the interior of the room and the theory of the walls of the room – seem to be equivalent. I’m giving a very garbled caricature of it. Why don’t you tell us? You’re the expert.

HUBENY:                Well that was pretty fair. So in string theory, we have seen these remarkable statements which we refer to as “dualities”, which is an interesting situation, that the same physics is described by different theories, different [01:02:00] mathematical languages if you will. And so there isn’t a sensible way to ask the question ‘Which one is correct?’ They are they’re both correct. They’re just different ways to describe th-

Speaker 52:         Tell us about the two theories. One is the theory with gravity that has more dimensions and that’s the theory of the inside of the room, and the theory of the walls of the room doesn’t have gravity and has fewer dimensions. And so something inside the room like a black hole a black hole – you know we’re told black holes are real, we can see them forming and at the center of the universe and so forth. But the theory on the walls of the room – there are [01:02:30] no black holes there’s just a bath of strongly interacting particles, and that’s all a black hole really is. And so your revisionist picture says there’s really no answer as to whether black holes genuinely exist. I’m putting this in a provocative way just to – so just tell us a little bit about the two theories. That both give [01:03:00] you the same information about reality. But one of them might be more convenient to use for certain calculations than the other.

HUBENY:                So in this particular example of duality, which is referred to as “AdS/CFT duality” or “gauge gravity duality”, the two sides that describe the same physics are a theory of gravity (in particular string theory) living in [01:03:30] higher dimensions, which is precisely equivalent to a theory without gravity – a so called “gauge theory” or “conformal field theory” – living in lower dimensions. In fact, you can think of it living on the boundary of the spacetime that the gravity lives in. And, well, it is believed more generally – so studying black holes – that any theory of gravity has this holographic nature [01:04:00] to it that you can encode all the information that’s happening in such a theory with a lower dimensional theory-

HOLT:                       If you could only see the shadows on the wall, you couldn’t reconstruct everything that was going on at the cocktail party in the room, right?

HUBENY:                That’s right. So we don’t have just shadows-

HOLT:                       But what’s so striking about AdS/CFT duality, is that there’s enough information-

WOMAN:               Let her speak please!

HOLT:                       Oh, I’m sorry. I always get heckled. That always happens to me. Do I talk too much? OK, I’ll shut up.

HUBENY:                Well, so, with shadows, indeed you couldn’t tell enough. Neither can [01:04:30] we tell with movies, which seem a little bit better than shadows and yet they’re supposed to be two dimensional representation of our three dimensional world, because you don’t see what’s behind the foreground [01:05:00] things or stuff like that. So here in this holographic duality, we would have clever shadows, or things that actually know the entire series. So indeed that’s very counter-intuitive. You fit all this information on a lower dimensional theory. But maybe the point to make is that a theory of gravity already does not have as much room in it as you might [01:05:30] have thought. And the reason for that is gravity. Once you have gravity, you can collapse black holes, for example. Which is a very generic type of behavior. And if you try to, say, pack in as much information as you can in a given region, say, in this room, now we can think, you bring your book and we can [01:06:00] read it and there’s some amount of information, I bring another book, Andrei brings one more book, and so forth. And we can pack this room with books and there’s no problem with that. But once we try packing so many books that we would sort of push at the limit of how much information can be packed in, gravity takes over and the whole thing will collapse into a black hole. And now we’ll see that, in fact, the amount [01:06:30] of information scales only the surface area of this room rather than with volume. Gravity has somehow built in the structure of being describable by a lower dimensional theory. So that’s called the “holographic principle”. Now this duality that you are prompting me to tell us about is a concrete realization of that – the so-called “AdS/CFT duality”. The AdS side stands for “anti-de Sitter”. It’s a particular [01:07:00] spacetime that has a negative cosmological constant – so very different from the expanding universe that we’re living in. But nevertheless, because it’s a theory of gravity, it contains black holes and in fact we believe that black holes there behave just like the black holes that we see out in our universe.

HUBENY:                And now, the miracle is that yes, there is this other description – this lower dimensional description. That, now, does not [01:07:30] have gravity. It does not have black holes. But everything that is happening around us, all these black holes and stuff like that, are described in this completely different language. It’s very scrambled but the same physics is happening in this.

HOLT:                       So as we were saying earlier, in the bulk theory, in the higher dimensional theory, you’ve got a black hole. In the boundary theory – which is a complete description as well – what looks [01:08:00] like a black hole is actually a path of glue on some quarks. So David, being a philosopher, is concerned with reality in itself. He wants to know, which is real? Are there black holes? Or maybe that’s a naive question.

ALBERT:                  Yeah, I’m being forced to say something that I’m not sure I believe. I mean, I guess there is – maybe here’s [01:08:30] a way to put the question. Physics has tons of mathematical equivalences in it. So we can talk about, say, the condition of a classical system of particles portrayed in three dimensional space, or we can talk about it in configuration space, or we can talk about it in phase space. All of these are in a very straightforward sense, mathematically equivalent, mappable [01:09:00] into each other, isomorphic to each other. There is a temptation to say, in the classical case, yes, but not withstanding these isomorphisms, the three dimensional space is the real one. The three dimensional space is the physical one. Is there an analogous question in the kinds of equivalences that you’re talking? Well, there may [01:09:30] be several things one could say about this. Maybe you think even in the classical case the question doesn’t make sense. Although most people would think it does.

HUBENY:                I think there’s a hidden assumption in this which is, who is asking? OK, so you are asking, and you are a local object, and to you, talking in the position space is the most natural. So then, that somehow singles out a preferred [01:10:00] description, because in some sense, you’re providing the basis on which to ask this question.

ALBERT:                  But I don’t understand – I’m part of this description too.

HUBENY:                But you’re a very specific one. You’re not, for example, spread out over this room.

ALBERT:                  I’m a very specific one, but – you know, there’s an old question attributed to Wittgenstein. People said to him – he said, “I can’t understand why the the Copernican theory was counter-intuitive,” [01:10:30] and people said, “Because it looks like the Sun goes around the Earth.”

HUBENY:                So that answers your question, no?

ALBERT:                  No, but Wittgenstein reportedly said, “What would it look like if the Earth goes around the Sun?” And so I take it the same question applies here.

HUBENY:                So is the claim – I’m sure if I prefer to describe things in three dimensional space, of course [01:11:00] I’ll be describing them in three dimensional space. If there’s some psychological feature of me which prefers three dimensional space, of course I’ll describe them in three dimensional space. And it sounds like you think the you think the right thing to say about this is just that it’s a mistake. Most people would say, in classical mechanics, “No, there’s some metaphysically significant sense in which the physical space is three dimensional-”

HUBENY:                I’m a physicist, I’m not a metaphysicist.

ALBERT:                  “-and in which the others are just sort of abstract [01:11:30] mathematical representations.”

HUBENY:                I do not subscribe to that.

ALBERT:                  Good.

LINDE:                     Maybe what he wants to ask is, ‘If your brain is splashed on the wall, are you still thinking?’

ALBERT:                  Right. Good. Exactly right.

HUBENY:                Right so then you would be naturally describing things on the wall.

ALBERT:                  But your brain is splashed on the wall! According to you.

HUBENY:                Right, and that brain is describing things –

ALBERT:                  No, but that brain is the same one as in the three-dimensional space.

HUBENY:                Right. So there you go.

ALBERT:                  What do you mean? Where do I go [01:12:00]?

HUBENY:                It’s doing both at the same time. I think that-

ALBERT:                  Those are two things? Aren’t they the same thing?

HUBENY:                Good. So if they’re the same thing, then it’s a linguistics question, right? We have different descriptions of the same thing and that is one thing, but we’re describing them in different languages. And now, you’re sort of trying to force me to say ‘Well, isn’t one preferred to the other because we somehow have this gut feeling that yes we’re living in three dimensions-’

ALBERT:                  [01:12:30] I mean, I’m not trying to force you to say anything. But it’s just a question. In the simple classical case, most people would say – indeed, this is how people are taught physics – ‘Oh I see what’s going on. The real deal is three-dimensional space and there are particles moving around. And three-dimensional space…’ There are various other abstract mathematical representations which are useful in certain contexts: configuration space, phase space, so on and so forth. These [01:13:00] are useful, there are all kinds of problems and I’m going to turn to those. But there is some sense in which I’m going to think what’s really going on is this three-dimensional space with the particles moving around in it. And it’s not that I’m claiming ‘That’s obviously right.’ It’s interesting that that’s what most people, even people with a physics education, would tend to say.

HUBENY:                It’s convenient, and it’s convenient psychologically. But why is it crucial [01:13:30]?

ALBERT:                  Because somebody wants to know, ‘What’s the dimensionality of space?’ And it sounds to them-

HOLT:                       David, what’s your answer to that, by the way? You have a very counterintuitive answer.

ALBERT:                  Yeah, that’s another story.

HOLT:                       Just tell me how many dimensions, space-wise.

ALBERT:                  n times the number of – well, the number of elementary degrees of freedom.

HOLT:                       What’s the order of magnitude? Is it about a google?

ALBERT:                  It’s really big [01:14:00] .

HOLT:                       So we have 12 minutes left. And the two topics I thought were interesting – we’re getting onto the soft ones – beauty is a guide to truth in physics… that may be to – but George is interested in the limits of physics and the limits of science, and if we knew the universal wave function, could we predict that we were going to get together and have this particular conversation and say these particular things? Does it leave any room for free [01:14:30] will? You have a notion of top-down causation, which I’m very skeptical of, but I want to hear about it.

ELLIS:                       I want to be – free will is a difficult one, but let me make the following statement: according to the standard big bang picture, let’s have a period of inflation (I’m happy with that), we’ve got the last scattering surface where matter and radiation decouple, we’ve got particles on the last scattering surface. The question is the following: if we knew everything about the last scattering surface to the finest detail, could we predict what is happening [01:15:00] in this room at this moment? Could we predict, physically, what you have just said? My claim is absolutely not, there isn’t the faintest chance that could be true. For many, many reasons. I’ll just take two. One, there’s quantum uncertainty between here and there and I could amplify on that. The second one is that what cosmologists do is characterize something called the transfer function which goes from here to here. We have difficulty even explaining the black hole, [01:15:30] dark halos, and so on. We have difficulty characterizing the distribution of galaxies. In practical terms, there’s not the faintest chance. But supposing there was a chance of doing that, then you would be telling me that somehow the words you’ve just said were written into the last scattering surface by some Demiurge and I don’t happen to believe that that could possibly have happened. So the real statement is the last scattering surface set up the conditions whereby higher level beings like you and me could come into be. Where we had [01:16:00] psychological powers and which we can argue with a psychological level and that psychological level has real causal powers which are allowed by the physics, but they are not predicated by the physics. The physics cannot tell you what you are going to say in psychological terms, because you are doing a logical argument and the physics has no concept what that logical argument is.

HOLT:                       And what you said is a highly heterodox point of view. Does anyone here agree with that? Because it seems to violate the unitarity of quantum mechanics.

ELLIS:                       I don’t happen to believe quantum mechanics is unitary, because I happen to believe that measurements take place. Every time a measurement takes place, it’s not unitary. So I don’t know what understand why quantum physicist says quantum mechanics is unitary.

HOLT:                       So do you think this is important because it does lend an opening to a robust notion of free will?

ELLIS:                       Yeah.

HOLT:                       OK. Surely this must have raised some hackles here. These are excessively polite people but they’re raging [01:17:00] inside.

LOEWER:                I would like to get one clarification from George about what a measurement is, in your view. So were there measurements before they were people and apparatus?

ELLIS:                       Yeah, absolutely. The-

HOLT:                       This is a quantum measurement problem we’re getting into, which is really deep waters here.

ELLIS:                       The word “measurement” has come because people first discovered this in laboratories. But a measurement takes place when a superposition goes to an eigenfunction. That’s a nonunitary process because you’re losing information about the superposition and you end up with unspecific – that happens all the time, everywhere. It happened [01:17:30] in the early universe, when nuclear synthesis took place, long before there was a was any-

LOEWER:                What is the law that determines when that happens?

ELLIS:                       We do not know. That’s one of the big things. We don’t know how it happens, and we have no specific characterization of when it happens. It’s one of the big questions at the foundations of quantum mechanics.

GUTH:                      And do you have any evidence that it happens?

ELLIS:                       Yeah. Those pictures of the screen, with the particles coming onto the screen, and you cannot predict where the next particle come on, those wonderful pictures, the particles come on one [01:18:00] by one, nobody can predict where the next one is coming. But eventually, it builds up to those interaction patterns in which you can predict, with great fidelity, what the final statistical thing will be. But you cannot predict what the next one will be, and that is the evidence for what I’ve dispensed.

GUTH:                      Well, we’ll agree with that. But isn’t that consistent with the Everett interpretation, and many other interpretations in quantum mechanics that do not invoke this mysterious collapsing wave function?

ELLIS:                       Give me some way of experimentally showing the Everett interpretation is true. Just some hint.

GUTH:                      The ball is in your court.

ALBERT:                  [01:18:30] As Jim said, we’re getting here into altogether different territory of the measurement problem in quantum mechanics. But we now have lots of ways of thinking about the measurement problem. It’s too early to know – just as with choosing between inflationary theories, it’s too early to know which, if any of the ones we have now [01:19:00], is going to turn out to be correct. But but we have lots of ways of thinking about them which contradict crucial assumptions that you were making.

ELLIS:                       Now you are supporting the Everett interpretation which he is putting forth?

ALBERT:                  No, actually I’m a very trenchant critic of the Everett interpretation.

HOLT:                       Just for the sake of the audience, no one uses the term “Everettian” except for-

ALBERT:                  There’s a many worlds interpretation.

HOLT:                       Just explain. So there are three of three or four interpretations of quantum theory…

ALBERT:                  Good. So there is this [01:19:30] puzzle-

LINDE:                     It’s like multiverse, many worlds interpretation.

HOLT:                       One of them is a multiverse interpretation, it’s very confusing.

ELLIS                         But if we come back to my question, I think Alan says there are zillions of other worlds out there in which we each say different things, rather than what we actually said here. I think it just simpler, makes more sense to say that real emergence takes place in which we do have real psychological powers, and in which we can make logical arguments, come to logical conclusions which lead to what we say. And I think that is what has happened [01:20:00]. Not that there’s just zillions of other possible things happening out there which we cannot prove are happening.

HOLT:                       OK and I would say nobody on the stage agrees with that. But yeah, go ahead.

LINDE:                     Actually the whole discussion brings us a little bit back to what we, with Alan, discussed, and that is about the measure problem. Because right now we’re returning to the measurement problem, which is a different problem which is, however, very much related. Because in [01:20:30] both cases we’re talking about whether there was something in the early universe which was definitely defined before anybody has seen it. Because the measure problem is ‘What is the problem of the first event,’ which nobody has seen it, then the wave function was not reduced, and stuff like that. So what is interesting about that, is the measurement problem is discussed by everyone for almost a hundred years. And you can really find [01:21:00] people who completely, in all details, agree about the solution of this measurement problem. And nevertheless, quantum mechanics works pretty well, everybody agrees about that. So I think that this is very similar to what happens with inflationary cosmology which we do. There are not so many people who absolutely agree about which probability measure, in a world amenable to a many universes interpretation of the inflationary theory, is better. But more or less everybody agrees among [01:21:30] experimentalists with who we’re discussing it, who are providing us with the data. So more or less everybody agrees that if you have a given inflationary theory it gives them predictions which they check and the checking so far was pretty successful.

ELLIS:                       Let me take a simpler example. We’ve come into existence through Darwinian evolution. My claim is the following: Darwinian evolution is consistent with physics, but it is not implied by physics. There is [01:22:00] no physics textbook on which you will read a chapter on Darwin because it isn’t a physics result, it’s a biology result.

LOEWER:                Well, a couple of things. One thing that’s striking is that people could come away from here learning that there are two big issues in the philosophical foundations of physics that are on the table. As Andrei was the saying, the measurement problem – an old one in the foundations of quantum mechanics, and the measure problem, a newer one [01:22:30]. There’s an important difference in the two cases. In the case of the measurement problem, there is quantum mechanics. Everyone who learns or takes physics learns how to use it to make predictions in the experiments that they do. What they disagree with is what’s really going on when measurements are being made. What the real underlying reality is. George has one view. David may have another view. I may have another view. In the case of the measure problem, it’s a problem that arises [01:23:00] for a particular theory in cosmology: the theory of inflation. And it’s really a question of whether or not the theory itself makes any predictions at all until the measure problem is actually solved. So these are two rather different, though both very important and interesting, issues.

HOLT:                       We have three minutes left. On the notion of beauty – beauty, a very fuzzy notion but people, great physicists like Steven Weinberg, the father of the standard model says that we’re now in an era where we don’t [01:23:30] have a lot of experimental data and observational data that we’ve reached the limits – you know, there won’t be a bigger particle accelerator than the Large Hadron Collider built in the foreseeable future, we’re sort of up against an empirical wall. And when that happens, physicists rely on their sense of beauty. And beauty has been a very reliable guide to truth. Weinberg said when he was a graduate student at Harvard, Paul Dirac came and said to the graduate students, “Don’t pay attention to what your equations mean, just pay attention to how beautiful they are.” [01:24:00] And and this is kind of a weirdly mystical, lotus-eating notion, but beauty has proved to be a reliable guide to truth in the past. And now, we’re in a situation where theories like string theory, which were thought to be beautiful initially, are now looking rather ugly, and people would make the same claim about inflation. But I won’t name them, because I don’t want to give offense to you, that inflation is an ugly theory.

LINDE:                     Who said to you about string theory that it’s ugly? [01:24:30]

HOLT:                       Sheldon Glashow. He said it’s like medieval-

LINDE:                     Sheldon Glashow in ‘71 suggested he is an alternative to the standard model of electromagnetic interactions. Standard model was based on the theory – sorry, for math. SU(2) x U(1), two different groups, two different coupling constants, many particles, anomalies, doesn’t work. And he suggests that his own beautiful symmetric [01:25:00] model, vector model. No anomalies. Everything is great. And everybody says ‘Oh yeah yeah yeah.’ Electromagnetic theory, Weinberg-Salam model – no, it’s because it is not beautiful. And [the] Sheldon Glashow model is so beautiful. And then they discovered electric currents. And then, where is [the] Sheldon Glashow model? OK, he got his Nobel Prize. But the Weinberg-Salam model survived. So that’s how it was.

HOLT:                       And he got a character named after him on “The Big Bang Theory”, the TV show, right? [01:25:30] Do you have one? No, there’s no Andrei. There’s a Sheldon. Anyone else have anything to say, any epigrammatic observations to make on beauty as a guide to truth? It’s just too fatuous.

HUBENY:                Well first of all, I would dispute that string theory is not beautiful, but that’s me. But I think – OK, the word “beauty” has all these other meanings, I would I would think “simplicity” and “internal consistency” are much [01:26:00] more descriptive. And-

HOLT:                       Why not accept logical rigidity, and the example of that was general relativity: if you alter it a little bit, it crashes. Whereas string theory is very elastic.

HUBENY:                No, no. I mean internal consistency is a very powerful tool, in fact it’s very restrictive. I mean it’s clear to all of us that whatever we propose as a theory can’t be mutually self-inconsistent. But you might think [01:26:30] that that’s a useless thing, because just about anything you propose, you can fiddle around with to make it self-consistent. But that doesn’t happen. And so this takes us much further along than one might have thought. And it turns out – I mean the physics packages itself in some nice way.

LINDE:                     The theory predicts multiverse, so it must be beautiful.

HOLT:                       OK [01:27:00] . So we’ve heard we’ve come to the end – we’ve heard the we’ve heard the scientific account of the universe, and we’ve heard the philosophers commenting on it. So I think we should just give the last word to religion. So the religious story is, in the beginning there was nothing, and God said, “Let there be light,” and there was still nothing but now you could see it. And then I have to say thank you to everyone for attending this afternoon and to all of you online. Please remember the World Science Festival is on through June 4th [01:27:30], and for all the programming and scheduling information please visit us at worldsciencefestival.com. Special programs not to be missed include: “Science in a Polarized World: A Global Town Hall Meeting” tonight at 8 pm at the NYU Skirball Theater, and our final day in Times Square today with our sustainable dance floor, LED screen, and holocenes. Thank you again everyone.