Sometimes the journalists and editors get it right. In fact, they get it right a lot of the time, but you hear more about the complaints (sometimes from me, sometimes elsewhere) about them getting it wrong, when it comes to things like science coverage especially. What am I talking about? I’m talking about the set of questions and answers that are in a new article on MSNBC that a number of people pointed out to me yesterday and today. It starts out as an article about Brian Greene’s science outreach efforts (books, and tv and movie appearances, including a new one), with some discussion of how this is regarded by his colleagues, the value it has had in raising public awareness of physics (and fundamental science in general, I would argue), and so forth. All that is interesting, but not nearly as interesting to me right now as the later parts of the article which is simply a question and answer session. (Picture above right is from a fun joke I carried out last year that you can read here – be sure to read the comments too.)
Alan Boyle, the science editor, asks Brian a series of very thoughtful questions, and Brian gives some very thoughtful answers. The topics include research in string theory (of course), hopes and possibilities for experimental and observational results (such as from the LHC and Planck) that can inform and ultimately test the ideas coming from string theory and open up new vistas in fundamental physics, research on issues such as the landscape, the idea of multiple universes, research on better developing our understanding of string theory (to the point where we can, it is hoped, extract firm predictions from it), and many other things. (I wrote an introduction to aspects of the landscape issue here – see also the comments – and talked a bit about a Tom Siegfried article on the discussion amongst researchers here.)
It is nice to see an honest, non-inflammatory and non-hyped conversation about the issues, and read Brian’s personal take on some of these matters. The bottom line is, of course, that we (the community that is working on string theory) don’t know the answers, but we are working on some of the tools and ideas that we regard as very promising directions to follow. There’s a great deal to do, and progress continues to be made.
Other people are free to make the case for other directions of research. There’s value in multiple approaches. Nobody is wedded to string theory for its own sake. If something demonstrably better or more promising is presented as an alternative, lots of people will work on it and help it develop further, and it may well strengthen into the dominant approach to finding the answers we seek about nature. This is the way it has always been in research. We just want to know what works. With regards to those who don’t like string theory, make overly strong (and sometimes just plain wrong) public statements about and against research in that area, you’ve read my writing here and the writing of others here and elsewhere on why science simply does not and has not ever worked that way. Here is a comment from UCSB’s Mark Srednicki earlier today on just this point (follow the link for the full comment:
We see that the big issue for Brian, and for just about all scientists (though with the apparent exceptions of Lee Smolin and Peter Woit), is what is TRUE. Not what corresponds to some philosophy of what science is or is not. Lee writes that the landscape must be rejected because “it would mean the end of our field†(page 165). It should be obvious that this is not the basis that is traditionally used for accepting or rejecting a theory! Peter’s (essentially the same) argument that string theory must be rejected because (at the moment) it does not appear to be sufficiently predictive (for Peter) is also irrelevant to the question of whether or not string theory is TRUE.
Well, I’ll let you read the article, but will end with one of the questions and its answer, which echoes things I keep saying here. Before I quote it, I’d like to say one more thing. I’ve found that different people have different takes on what it means to have a “theory of everything”. There is a popular idea (perhaps the most common) that this somehow means that this theory will describe (at least in principle) all known basic physical phenomena (constituents and their interactions, if you like) once and for all. Others mean something less ambitious, a theory that consistently describes the four fundamental forces and the things that interact with them, achieving a unification of all the forces and phenomena that we currently understand. I personally think that the first idea of a theory of everything is rather naive, and my personal hunch (and bias from what I’ve learned about the history of science) is that there is simply no such thing. There’ll always be new layers, new questions we simply have not even thought of. I find it hard to believe that it is possible to write something down that can answer all questions, about everything, for all time. (I think I’ve said this at greater length in a comment elsewhere, but I can’t find it, so I will update later with a link to it, perhaps.)
What do you think?
Anyway, in reading the question below, I find myself wondering which definition the questioner is using, and which definition the answerer is using, and also which definition Hawking (who is referred to) is using. Interesting to wonder. Well here it is:
Boyle: Hawking has said that there could be a “theory of everything†produced in the next 20 years, or by 2020. Do you get that same sense? Or will there ever be a theory of everything?
Greene: Well, I always find it difficult to make predictions that are tied to a specific time frame, because as we all know, one of the exciting things about science is that you don’t know when the big break is going to happen. It could happen tomorrow, it could happen 10 years from now, it could happen a century from now. So you just keep pressing on, making progress, and hope that you reach these major milestones — ideally in your own lifetime, but who knows? So I don’t know if 2020 is the right number to say. But I would say that string theory has a chance of being that unified theory, and we are learning more and more about it. Every day, every week, every month there are fantastically interesting developments.
Will it all come together by 2020, where we can actually have experimental proof and the theory develops to the point that it really makes definitive statements that can be tested? I don’t know. I hope so. But hope is not the thing that determines what will actually happen. It’s the hard work of scientists around the world.
-cvj
(Thanks Tameem, Sean, and Mark!)
…Although such a theory of natural organisation does also resolve the problems of how mind and consciousness evolved from inanimate matter and how there can be many different immaterial minds and how consciousness is indivisible.
No theory of everything could provide solutions to all theoretical problems in physics. But I propose that a theory can be developed that can at least redolve problems 2 and 5 on Lee Smolin’s list and render problem 1 redundent.
Although this is a theory that no physicist is ever likely to develop. Also, in all probability, most physicists would be very reluctant to accept this theory as scientifically valid.
And such is a theory that justifies and describes details of a cause to explain how the universe has become and can remain naturally organised on both the subatomic and astronomical scale and despite the action of the forces.
If I am understanding what you write about a theory of everything correctly I agree with you. I believe that every answer brings more questions. A state where an answer will end all questions (the state of theory of everything) is not a realistic one.
The quantum theory of the standard model can’t explain how matter can be or remain organised out of its subatomic parts or disprove the theory that a distinct cause (like the quantum potential) acts universally and nonlocally in addition to the forces.
So the attempt to develop any theory of everything that unifies QFT with relativity could be all in vain.
Whereas instead one should be attempting to develop a theory that examins enough available evidence together so as to justify and describe enough details of a further unversal cause and its effects upon matter and energy.
Google> foranewageofreason
Mark,
Re: #19,
Isn’t there an argument from the holographic principle that ensures that no region containing information can have zero surface area and hence zero volume (singularity)? I realize the holographic principle is not an experimentally verified theory but after all when we are talking about black holes.
With no singularity, there is no paradox.
Elliot
Thomas: In any theory with weakly interacting particles whose number is not fixed, we must be able to define creation and annihilation operators, labeled by momentum, that create and destroy a single particle. Then we can Fourier transform those operators to get fields. Thus, QM of weakly interacting particles always has a field-theory formulation. But of course, premises must always be carefully stated for any no-go theorem. And even then, exceptions that violate the premises in surprising ways are often the most interesting result; eg, supersymmetry for Coleman-Mandula and AdS/CFT for Weinberg-Witten. But I don’t think a better modeling of observers in QFT is going to help with the “nice slice” arguments in the case of BH evaporation. (Though I would be happy to be convinced otherwise!)
Lee: Thanks. Of course I agree as well.
Gina: LOL!
gina
Excellent!!!
Dirac says, “If LQG has muscles, then it will stand in its own merit, not as a socialogical force but as a fundamental area of research.†Indeed, and it is doing so, more and more. As I have said elsewhere the most compelling results of LQG have come in the last few years. The only issue is with people who choose to remain ignorant about them because they believe they already know the answer to quantum gravity.
To Mark, of course. For one thing the existence of the string landscape has been, at least for me, a great stimulus to revising the notion of time in quantum cosmology. Beyond that the context in which the role of time in quantum cosmology has to be discussed is that of attempts to formulate background independent theories, to the extent that efforts are made to construct a manifestly background independent framework for string theory in the compact case-with no asymptotic symmetries or boundary conditions, the problem of time has to be confronted.
Dear Peter,
This amazing “admitting failure” concept of yours is really a new contribution you made to the scientific discussion. We can imagine the following dialogue between PW and a scientist S (any scientist in any field).
PW: Why are you studying what you are studying?
S: Beacuse I think it is interesting!
PW: No, you are studying it because you do not want to admit failure!
S: No, I really think it is interesting.
PW: No, No, admit failure at once!
S: If I admit failure, can I still continue studying what I am studying?
PW: Yes. but why would you?
S: because I think it is interesting!
Holmes,
I don’t think Susskind would claim this is well-understood, just that there is zero evidence that better understanding it will lead to a conventionally predictive model of particle physics. I agree with him about this. Maybe better understanding will show that you can’t get anything like the real world this way, but I think you need more than hope to show an idea is wrong or wishful thinking that it is right to motivate a widely sold research program. Right now, the only motivation I see for people pursuing this is not that they have a scientific argument about why it might work out, but that they are forced into it in order to avoid admitting failure of the string unification program.
Mark 19: I wasn’t aware that the black hole information paradox could be phrased thusly, but it seems like the kind of no-go theorem where one should state the premises very carefully. Shouldn’t it read: BH evaporation either violates causality or QFT? It is quite conceivable that QM is right but QFT is wrong. After all, QM is clean and beautiful, whereas QFT requires kludgy renormalization and is incompatible with gravity.
One modification (ok, the modification) of QFT which does not affect QM is this. Quantize not only the fields but also the observer, and in particular the observer’s clock, which is the operational definition of time. In a scattering experiment, this modification should lead to a small broadening of the peaks, because the observer recoils due to the act of observation. This does not change QM, i.e. QFT in 0+1 D, simply because there is nowhere that the observer could recoil in zero space dimensions. Explicit calculation for free theories support this conclusion.
Causality should not be abandoned lightly.
Peter Woit said:
“What is the evidence that the Coleman-DeLuccia mechanism is going to determine things like the standard model gauge groups and couplings? I think that it is this kind of thing that Susskind is claiming there is no evidence for except wishful thinking”
OK, I see what you are saying now. Well, Susskind believes, for reasons best known to himself, that [a] Whatever it is that governs bubble nucleation — and Coleman and De Luccia never pretended that their model was anything but a toy model — is well understood, and [b] that this mechanism produces simple, well-understood spacetimes that lead naturally to inflation etc etc etc. I put it to you that neither of these beliefs is supported by any evidence. In reality one would expect bubble nucleation to be a very complicated process indeed, and furthermore it is far from clear what will happen when perturbations are taken into account. It could easily be very very tough to get anything that even vaguely resembles our world.
Anyway, the general point is this: the mere existence of 10^whatever solutions does not in itself invalidate predictability. You need to understand the means by which possible worlds are *actually realized*. Our understanding of that is so poor that we just don’t know whether string theory is predictive or not. It’s like saying, hey GR has white hole solutions but nobody has ever seen anything even remotely like that! This would be a problem only if somebody could show that there is an actual physical mechanism that should be producing white holes all over the place. There is no such mechanism, hence no problem with predictability in GR. Your problem is with Susskind’s [probably] vastly over-optimistic views on what bubble nucleation can actually accomplish, not [necessarily] with string theory.
It is fairly evident as an outsider that the current string debate has reached a level of madness
that is simply not able to produce anything constructive. I think that there
is virtue in people like Srednicki, Johnson and Smolin debating on aspects of quantum gravity in public sphere. However,
I feel that its tragic that Smolin has to join in with someone who doesnt have any credential to fight
the war against string theory. If LQG has muscles, then it will stand in its own merit, not as a socialogical force
but as a fundamental area of research.
I normally dont comment on these threads; I found it
rather surprising that people who dont have any original publication in years are leading the
scientific debate in fundamental physics. If you want to say something publish original academic work.
I am still an undergraduate but I have the feeling that the whole debate about string theory
in public has misled people and perhaps become an agenda to sell popular books.I suggest people who were not able to make it in academia(although I am sympathetic) to stop disgruntling
about what they could not be and do some work original work as opposed to repetitive ranting.
Well said Clifford!
Gina; but also…others:
Math and logic (among many other disciplines) are based on axiomatics…which are agreed to ‘on faith’, “without proof,” and are used to develop self-consistent systems. If one follows the rules, derived corollaries are absolutely TRUE (with possible Goedelian qualifications). Theoretical science can build self-consistent models with the tools of math and logic, or with wishy-washy heuristic tools.
But the scientific ‘truth’ of models depends on how closely their predictions match observable evidence of experimental science, NOT on how logically consistent are their mathematical structures.
The symmetries of the equivalence principles led to self-consistent, Conservation Laws which were all ‘obviously’ TRUE. But observation ‘proved’ that Parity and CP are not always conserved…and so led ultimately to the Standard Model…despite that mathematical TRUTH.
Empirical ‘truth’ is quite different. It can’t be proved because observation is always incomplete, no matter how often it has been repeated. A sample of a ‘whole’ can never be PROVED to be a fair sample. So, as established by David Hume, our ‘belief’ in inductive reasoning must ultimately depend upon an implicit or explicit “faith” in the ‘truth’ of what ‘seems’ to be implied by ‘facts’ of observation. But empirical truth is our only connection with reality. And that’s why only models which, at least in principle, can be confirmed by observation, are scientific models.
And until they’re confirmed, they sit in purgatory, as a kind of science fiction.
String Theory, SETI, Higgs bosons, axions, etc. are all in limbo. Some will be resurrected by evidence…others are doomed. Obviously, the harder it is to confirm them, the longer they’ll sit in limbo, and the less patience the scientific enterprise will have for giving them continued attention (and resources).
I believe this is closer to Lee’s and Peter’s view of science, than to that of Gina, Clifford, Mark and Lubos.
Lee, I like that answer. However, it seems to me to be a very different answer than the one in your book, where you say that it is “most rational” to deny possibility 1, and that, “as scientists”, we should not even “consider” it. (Full quote in post 27 above.)
A follow-up question: is there any reason that your suspicion that “by the time we understand quantum cosmology the notion of time will have been so altered as to make your question either nonsense or uninteresting” could not arise within string theory? It seems to me (and I feel sure you will agree) that string theory has not helped much (if at all) with this sort of question, and that therefore any future discoveries related to, say, semiclassical emergent time, might turn out to be applicable within the framework of string theory. Would you agree or disagree with that?
Dear all, Peter Woit wrote “You seem to believe that string theory would be a legitimate, viable physical theory if it was just shown to be logically consistent, while predicting nothing about the physical world. I don’t know what to call such a theory, but whatever it is, it’s not science.”
It looks to me that showing logical consistency (even for older successful physics theories) will be a big step forward.
As for what is science, I tend to think now that mathematics itself is, in fact, science. At least science of some sort. Isn’t it?
We are all agreed that your theory is crazy. The question which divides us is whether it is crazy enough to have a chance of being correct. My own feeling is that it is not crazy enough.
— Niels Bohr
I consider this an interesting support for ST.
The idea that it should be called the String Hypothesis is interesting. In addition to the definition of the word theory where GR is a theory, and most consider it fact (truth), there is this idea of a theory being a body of ideas. ST qualifies on that count, even if it turns out to be wrong.
Linde:During inflation, quantum fluctuations can produce not only galaxies, but also new parts of the universe.
For the lay person I thought it was important that this be understood. As well, to get this sense of how such a mechanism may work?
For more information see here
Is there no comparison satisfactory that allows the move from the false vacuum to the true, to have said that we see such a result.
Linde:To use the ball analogy, if it experiences small perturbations as it rolls, it might at some point roll over into the next valley, initiating a new inflationary process, he said.
gerry atric,
What is the evidence that the Coleman-DeLuccia mechanism is going to determine things like the standard model gauge groups and couplings? I think that it is this kind of thing that Susskind is claiming there is no evidence for except wishful thinking. He does make the claim that Coleman-DeLuccia constrains the spatial curvature of the universe (others seem to disagree), but this is very different.
Pingback: Not Even Wrong » Blog Archive » Censored Comments From Asymptotia
Dear Mark,
Sorry, I missed your question.. Do I believe it is impossible in principle that possibility 1, that ours is one of a vast collection of universes with random laws, is true? If that’s not what I mean, what should we do if possibility 1 is, in fact, true?â€
I can only tell you my suspicion, which I certainly don’t have a rigorous argument for. It is that possibility 1 is not in fact a logical possibility in the correct theory of quantum cosmology. Why? Because random refers to a probability measure and that in turn refers to a notion of time. There are two kinds of random, random with respect to possible states at a given time or random with respect to the ensemble at all times. If we mean the latter there are tricky questions because trying to define an ensemble including instances of X at all present and future times can lead to paradoxes such as the doomsday argument that starts, “I am a typical member of the ensemble of all people who will ever live.†There may be good reasons why such ensembles cannot be reasoned about in the same way one can reason with the ensemble of all X that exist at the present time or at a given past time.
The alternative is to consider a random ensemble of all universes at a given time, but that requires a global multi-universal time. It also requires a mechanism to have acted to randomize the ensemble of all universes in the multiverse by a given time. My hunch is that the difficulty people have had defining static measures on models of multiverses are indicative of this being a question that, when we understand the notion of time more deeply, will turn out to have been ill posed logically.
Note that in biology we never have to make reference to a timeless ensemble such as all possible species or collections of species. We do quite well with local (in the space of genotypes) statements such as that typical viable creatures are as fit or more fit than almost all of their single mutation variants. This is all we need to make heaps of falsifiable predictions. I suspect the same is true in cosmology.
Someone might have said to Einstien, Isn’t it logically possible that there is a stationary aether and its dynamics are arranged so that motion through it is meaningful but undetectable? Einstein didn’t spend time to answer this, he just developed a radically different notion of spacetime. When he was done, such a question was no longer interesting. Now, none of us living are remotely in the class of Einstein, but perhaps we can learn something from his methodology. So I suspect that by the time we understand quantum cosmology the notion of time will have been so altered as to make your question either nonsense or uninteresting. There certainly is lots of evidence that the idea common in quantum cosmology, that time is not fundamental but “emergent in the semiclassical limit,†cannot be made sense of for real theories, beyond simple models. This is tied up with the very challenging question of what is the right physical inner product in quantum gravity theories. This is why my own work in this area is focusing on the nature of time in cosmological theories.
Thanks,
Lee
PW said: ”
all the evidence points to quite the opposite.”
Well, no it doesn’t actually. The CdL mechanism [apparently ]*always* produces universes of a very specific kind —- it doesn’t produce just any old thing. So certainly there are vast swathes of the landscape that are mathematically compatible with string theory, but which cannot actually be realized by the CdL mechanism. It is perfectly conceivable that , as the CdL mechanism is better understood [in terms of realistic cosmology] it will turn out that only a tiny fraction of the landscape can actually be realized. So one will have to say that there are vast numbers of string models that exist as mathematical abstractions but not as real universes. Just as white holes exist as solutions of GR but our world doesn’t know how they can actually be manufactured.
I think we can all agree that efforts to understand just what CdL allows is real science. Maybe I am wrong, and you really can get just about anything out of it. Peter can come back then and tell us that all is lost. But the work hasn’t been done yet. Why not wait until it is?
The answer I would give to MS’s question to LS is: we should always proceed on the assumption that a testable theory is out there. It is possible that option 1 is true, but we should always proceed as if it were not. This may sound rather gross 🙂 but I think that that is what we should do.
Everyone:- As may or may not be well known, I routinely delete comments from Lubos Motl on this blog without reading them. I also delete comments that reply to him or try to provoke him. So I’m sorry, but I have deleted two or three comments of his, and also a reply to him from Lee Smolin that started by directly addressing him. I don’t know what was in those comments. Since I have a great deal of respect for his ability as a physicist, however, if he was making a physics point in his comments, perhaps he might make it on his blog and link to this discussion via trackback. I thank him for his physics contributions and widening the discussion.
Best,
-cvj
Lee, you didn’t answer a question I asked you on the earlier thread, so I’ll repeat it here. In The Trouble with Physics, you wrote
“…when it comes to the biofriendliness of our universe, we have at least three possibilities:1. Ours is one of a vast collection of universes with random laws. 2. There was an intelligent designer. 3. There is a so-far unknown mechanism that will both explain the biofriendliness of our universe and make testable predictions by which it can be confirmed or falsified.
“Given that the first two possibilities are untestable in principle, it is most rational to hold out for the third possibility. Indeed, that is the only possibility we should consider as scientists, because accepting either of the first two would mean the end of our field.â€
It is not clear to me what you mean when you say that we “should not consider” possibility 1. Do you believe it is impossible in principle that possibility 1 is true? If that’s not what you mean, what should we do if possibility 1 is, in fact, true?
gerry atric,
What you’re postulating is some unknown vacuum selection principle that will pick something like our universe out of the landscape. I’m with Susskind on this issue: no one has any evidence for such a thing other than wishful thinking, all the evidence points to quite the opposite.
Actuallly, all of you are right. Peter W would have a point IF we knew that string theory can produce model universes with arbitrarily prescribed properties. While L Susskind may believe that, there is no reason to do so. The most plausible mechanism for producing *actually existent* universes [as opposed to mathematical possibilities] is the Coleman-DeLuccia mechanism. But nobody really knows whether you can get something that resembles our universe that way. 10^500 may sound like a lot, but if all of them have to be produced in this way, it may be that few [or none] of them are realistic. So Clifford and MS are also right: the thing to do is to continue research until this question is settled.
Mark,
Personally I’ve never been a big fan of the kind of pure quantum gravity research that you describe, precisely because you can’t hope to resolve issues that come up by appeal to experiment in the forseeable future. But it’s science, I’m glad other people who like to think about it do so, and it’s not impossible they’ll come up with something interesting. The reasons these issues aren’t amenable to experiment is that the relevant scales are hard to access, not that the models are made extremely complicated in an attempt to evade contradicton with experiment and testability.
This is very different than the landscape, which purports to explain physics at experimentally accessible scales, but actually doesn’t do so, instead just providing an elaborate excuse for failure.
In your hypothetical scenario, “The Trouble With Physics” and “Not Even Wrong” would not have been written, so I don’t see their relevance. Yes, the standard model is science and is testable, whether or not governments decide to fund the accelerators to perform the tests.
If you want to help the cause of defunding of particle physics, go right ahead promoting landscape research and publicly promoting the idea of inherently untestable science. Do you ever talk to non-string theorist scientists and do you have any idea what impression things like Susskind’s “The Cosmic Landscape” is making on them?
Clifford,
As time goes on and the failure of string theory becomes more apparent, you are starting to rant […snip … – personal reference deleted -cvj]. You should get a grip.
Mark asks, “What does LQG have to say about the information paradox?†Let me first say something I hope no one can disagree with: either there is still a singularity to the future of BH horizons even after quantum gravity effects have been taken into account or there is not. If there is no singularity then spacetime survives to the immediate future of where GR predicts the singularity. Let us call this new region F. This means that any matter, and any quantum information associated with their state continues to survive in F. There then can be no violation of local quantum evolution rules, ie the whole evolution on a spatial slice is governed by a Hamiltonian which means outside the horizon and in F the evolution of density matrixes is given separately by CP maps. There is then no puzzle, the quantum information the photons in the hawking radiation are entangled with continues to exist forever in F.
Does this information ever travel again outside the horizon? It depends on the dynamics. If F causally reconnects with the spacetime far from the black hole it does, if however F has no future causal connection with ScrI+ then the quantum information remains in F.
My understanding is that the “information paradox†arose because Hawking assumed without proof that there was still a singularity even when quantum effects were taken into account.
What does LQG say about whether there is a singularity or whether the information propagates into the new region F? A number of models of the region inside the horizon have been studied using methods from LQG. The results, by Modesto, Husain and Winkler, Ashtekar, Bojowald et al and others all show that generically the singularity is eliminated and there is a bounce creating the region F.
These however are all models where the degrees of freedom are reduced before quantization. While the results are suggestive, we have no result yet from the full QFT as to whether the singularity avoidance mechanisms of the models work there,
Thanks,
Lee
Ps let me anticipate the usual objection based on an incorrect application of the Bekenstein entropy to rule out ‘remnants’. The error in this reasoning is to take the Bekenstein entropy to measure the “information content inside the horizon†when all the arguments we have for it show it measures the information measurable on the horizon by an observer just outside it. For more on this see hep-th/0003056
.
Peter, I was refering to the debate as to whether evaoporating black holes violate macroscopic causality or quantum mechanics (one or the other being necessary). There is no practical prospect for checking which one is right experimentally in anything like the forseeable future, even with the most advanced technology that is realistically imaginable. So you must see this debate as “pretty uninteresting”. Is that right?
Here’s another hypothetical: suppose no one had ever come up with the quark model or QCD, and we were trying to do hadronic physics with effective field theories of hadrons with hundreds or thousands of parameters. Then, one day, some genius announces that the whole shebang can be explained by “quarks” and “gluons”, involving the group SU(3), and just 12 parameters: the “strong coupling constant” (evaluated at some arbitrarily chosen renormalization point), six quark masses, four “mixing angles”, and a “vacuum angle” that must be set to zero (for unknown reasons).
The next day, the world’s governments, having read “The Trouble with Physics” and “Not Even Wrong”, decide to stop wasting money on particle physics, and shut down all accelerators. No further tests of this new model are possible for the forseeable future.
Question: in this scenario, should the new model be considered science, or not?
Peter,
What’s really amazing is your continued ability to unashamedly repeat yourself in condemning a body of ongoing research without (a) actually understanding it very well at a technical level, and (b) being able to back up your assertions about how it is wrong. Why don’t you just let us get on with our research, and you get on with yours?
We still await your promised demonstration of your assertions’ validity. Where is it, Peter? Before you refer to your slide show again, (or a new place where it might be hidden this time, in the Easter egg hunt for the proof) I’ll refer you and readers to my previous commentary on that, to which you had no contentful answer.
As to your “Physics is not math, and physicists don’t actually have theorems”. What!?! Do you actually teach any young people physics? Or any science? I dearly hope not. No-go theorems (and things in that spirit) are often the starting point for wonderful discoveries in science. It is very useful and instructive to try and examine what you can and cannot do within a framework based on your existing assumptions. It sometimes then turns out that re-examining the assumptions leads you to new places… new physics. It does not mean that the theorem was wrong. The theorem was right, and told you to go and look again at your assumptions. Physics works this way a lot. I can’t believe you said the above. You must surely know this, and of several examples, being an educator. I can only think that you’re once again leading with your anti-string prejudice in any argument, rather than using the knowledge that you have to take part in an open-minded discussion.
As to Mark’s discussion of no-go theorems, as I understand it, he was not claiming that this is the way things will turn out… he was merely (generously assuming that he was talking among sensible, open-minded people) and offering a speculative scenario for how things could turn out in the process of trying to see what framework works in describing Nature.
I don’t see why this always has to turn into a predictable round of “anti-string vs string” back and forth. Nobody knows the answers (despite your unverified claims, Peter), and all we want to do is find out what works. Why don’t you do something constructive and try to find out what works too, rather than spreading deception to the general public?
Now please try to take part in a discussion of the physics with an open mind. Without turning every thread into a repetitive string vs anti-string discussion. Just try. Once. Please. You can do it.
-cvj
Mark,
“Is this debate part of science?” I can’t tell what debate you are talking about. Black holes?
Whatever your proposed theory of quantum gravity/black holes is, it ultimately has to provide some sort of convincing experimental prediction that in principle can be checked against observation. If the only predictions of your theory are impractical to check, you’re doing a pretty uninteresting sort of science, because no one will ever know if you are right. If you have no predictions at all, you’re not doing science.
You seem to believe that string theory would be a legitimate, viable physical theory if it was just shown to be logically consistent, while predicting nothing about the physical world. I don’t know what to call such a theory, but whatever it is, it’s not science. As for a hypothetical no-go theorem that only string theory can explain gravity, there are several problems with that. You know what the general fate of no-go theorems is: people find ways to get around them (e.g. Coleman-Mandula). Physics is not math, and physicists don’t actually have theorems, what they have are non-rigorous arguments of varying degrees of reliability that are based on lots of assumptions about the real world. You can’t use this kind of argument to claim to know what is true about the world, you have to be able to go and check it.
I find this whole discussion really remarkable. What’s going on with the string theory landscape is nothing unusual. A common fate of speculative ideas, once one understands their implications, is to find that they turn out to be wrong because they are vacuous and you can’t predict anything using them. That’s all that has happened here, but instead of admitting failure people are trying to change the rules of the game, announcing that conventional ideas about how one tests speculative scientific ideas no longer hold. It’s a pretty amazing phenomenon to watch.
Eliot: look up the “black hole information paradox”. Very briefly, black-hole evaporation necessarily violates either quantum mechanics or causality (as conventionally realized in field theory). There has been a 30+ year debate on which it will turn out to be, with Stephen Hawking the most visible proponent of modifying quantum mechanics. But in 2004, he changed his mind, famously conceding a bet on the question to John Preskill (with the payoff being a baseball encyclopedia; this is part of a joke, since one of the thought experients on information loss involved throwing an encyclopedia into a black hole). Hawking seems to have been persuaded largely by arguments from the AdS/CFT correspondence of string theory, though he has his own way of presenting his new opinion.
Question for Peter: is this debate part of science, or not? Note that there is really no chance that we will ever be able to tell experimentally whether or not black holes evolve from pure to mixed states. (It would involve having a very large number of identically prepared black holes, and nearly perfect measurement of all aspects of every escaping particle for every one of them.)
Question for Lee: What does LQG have to say about the information paradox? I’ve suddenly realized that I don’t know the answer to this.
Matteoeo: the sort of theorem I have in mind is the Coleman-Mandula theorem or the Weinberg-Witten theorem. For example, it might be possible to show that any consistent theory that obeys quantum mechanics and has general coordinate invariance must have an infinite tower of particle states of ever increasing mass when expanded around a flat background, and that the interactions of these states is always consistent with a string interpretation.
Gina: you wrote, “if string theorists will convincingly prove (and this looks unlikely) that no conventional scientific prediction is possible from string theory, then this proof will be a first-rate scientific achievment and then it will be for string theorists and other physicists to think how to proceed from there.” Beautifully put! I agree completely.
Hopefully this link is suitable enough.
What if, what if, what if… That’s so silly to do science with such wisdom. But if you care so much, anyone could take it seriously and compete with you for the most radical and imaginative “What if” ever. For example: WHAT IF it turned out not to be supersimmetry in LHC? WHAT IF no go theorems should be demonstrated for String Theory? WHAT IF gravitons were shown not to exist? WHAT IF black holes were shown not to exist, or the cosmological costant were shown to be zero? WHAT IF a new theory should come up and unify gravity and the quantum and be testable?
Does this “what if” battle have any sense? No it hasn’t. Your argument is a kind of tautology: if my theory were shown to be good (or equivalently, to be the onyl un-false theory, because un-falsifiable), than it would be good, so my theory is good.
Moreover, with what kind of math argument could you ever imagine to construct a valid no-go theorem for any other theory but yours, which doesn’t rely on aruments taken from your own theory or any other theory but the one being dismantled, and that relies only on physical principles, if you yourself dismiss physical principles such as causality with such ease, and basing upon some very speculative idea?
No-go theorem can only be generated out of that theory itself. I personally would only trust a no-go theorem relying on some very deep and profound logic truth such as Goedel’s, and on good math; anything else depending on some interpretation of physics, based on some other unreliable theory which might eventually turn out to be false or approximate, is not a theorem to me.
Peter asked: “Suppose for the sake of argument that the landscape is an inherent part of string theory, and its nature is that no conventional scientific prediction can be generated by it. The only evidence for string theory will always and forever be circumstantial things like “it explains gravityâ€. Then, is string theory science, yes or no? ”
Lets start with a simpler questions: “Suppose for the sake of argument that string theory will fail. either because a better alternative will be found or because some serious intrinsic problems will be found or because it will be falsified by experiments. Then is string theory science, yes or no?”
And the answer is “Yes, of course it is science.” Not just because (quoting you, Peter) it already constributed important insights to physics and to mathematics, but mainly because failure is a possibility in any worth while endeavor and, in fact, this is an important distinction between scientific truth-searching and relegious truth-searching where failure is not an option.
Your specific question, with the wording “will always and forever be” has a theological ring to it. But if string theorists will convincingly prove (and this looks unlikely) that no conventional scientific prediction is possible from string theory, then this proof will be a first-rate scientific achievment and then it will be for string theorists and other physicists to think how to proceed from there.
Mark,
If string theory predicts violations of causality, I am not sure this is type of assertion that would lead many “string agnostics” (of which I count myself as a member) to readily adopt the faith. Of all the scientific principles, causality is in my mind one of the fundamental cornerstones.
Elliot
Clifford wrote: ” Phase I of the modern discussion of a landscape of solutions of string theory, and the phenomenological issues it raised, began around then… in the 1980s. ”
So, then, what is the best reference from the 80s where landscape is discussed.
And: “But there were several serious discussions of the other possibility, I believe, although I do not know if much of it made it to print.”
Giving correct credits in science is probably a more complicated issue than discovering the secrets of nature. (So its is unclear why to get to this at all in this already complicated debate.) But an important part of getting credit is realizing that something is important enough to get it in print. What was not made itself to print does not count so much. (Maybe blogs will change matters; we will see.)
Trying hard with non-string approaches and no go theorems would be appropriate for a scientific discussion. But if you abandon the scientific method, “string wars” becomes a religion war. Religion wars have their own methods: a good start might be proclaiming that Smolin & co are crackpots in a Harvard blog with censorship.
There may well turn out to be ways to test string theory that are not yet forseen. For example, the string picture of black hole evaporation requires huge violations of macroscopic causality; maybe there are ways to trigger this in less extreme situations that could be observed.
Maybe not. But what if string theory turns out to be the only consistent framework for quantum gravity that ever gets found? (I consider the various flavors of LQC that are under investigation to fall far short of reliably being theories of quantum gravity that have flar-space solutions with gravition excitations.) What if “no go” theorems are proved for all other approaches? What if, after decades or even centuries of work, we never find any loopholes in those theorems? Wouldn’t we be pretty confident that string theory is right, even if we can’t ever do further experiments to test it?
Part of comment I placed on the other thread:
Lee,
I also have to say that I find this claim of yours to have invented, or introduced, or discovered the landscape in string theory to be far from convincing. I don’t see (in broad brushstrokes at least) how the idea of a landscape of solutions was not evident right after the first (for example) Calabi-Yau compactification, combined with the knowledge that there is a vast number of CY manifolds. Phase I of the modern discussion of a landscape of solutions of string theory, and the phenomenological issues it raised, began around then… in the 1980s. I don’t recall you being on any of those papers. Most people did, I’ll agree, prefer to learn more about the theory first before concluding from perturbation theory alone that that was all there was. But there were several serious discussions of the other possibility, I believe, although I do not know if much of it made it to print. Phase II (where some researchers are now) came about with constructing landscapes of long-lived vacua with finite cosmological constant, with the aid of some improved knowledge of control of some non-perturbative issues (but far from anything like a satisfactory non-perturbative understanding of the theory).
So I would say that your discussions several years later are notable, and probably interesting, but it is really quite a distortion (by almost any definition) to say that you discovered/invented the landscape in string theory and was just “waiting for the string community to catch upâ€. I will grant you that you might have been one of the first to try to see what would happen if the landscape of solutions was all there was, and what that would mean. But that is a bit different from what you are claiming.
Cheers,
-cvj
Pingback: Saturday evening link fest « Entertaining Research
Hi Lee. Thanks. I will place a link in your comment above to help guide people directly to what you said.
-cvj
Dear Clifford and friends,
The quote from Mark above hugely distorts my views on the landscape. This is disheartening as he ignores the fact that not only do I take the possibility that the landscape is real seriously, I was the first to point out the issue, to give it the name “landscape”, and to consider and reject the AP as a solution to it. Moreover, I went on, back in 1992 to propose a solution to the landscape problem that implies falsifiable predictions for doable experiments.
For the full quote from that section of my book, which shows that I was referring in that quote to the AP and not the landscape issue in general, see my last post under the thread “More Scenes From the Storm in a Teacup, VII”
Lee
TBB:- Thanks! I can’t answer your later questions, but I do know that Brian knows about the prank from last year. We talked about it.
Cheers,
-cvj
disregarding the mushy philosophical mashup above (some linguistic philosophy studies might help, fellas), my take on the question: what do you think/ as regards the potential of a TOE or GUT is similar to yours Clifford. The more we increase the capacity of our epistemological bases of describing the universe, the more we tend to find more we do not know. Thus any set of answers inspires more and more questions. A potentially verifiable TOE would, in most likelihood, lead to more and more questions about the cosmos.
Could we have a show of hands from the string theorists in the audience? Suppose for the sake of argument that the landscape is an inherent part of string theory, and its nature is that no conventional scientific prediction can be generated by it. The only evidence for string theory will always and forever be circumstantial things like “it explains gravity”. Then, is string theory science, yes or no?
Clifford and Mark,
It’s funny, but I’ve been criticized by various string theorists, publicly and privately, for claiming that the failure of string theory based unification has led some string theorists to give up on science. So, I’ve tried to be careful and have been hedging this claim. Now I see that I don’t have to. Both of you seem to be quite willing to accept the idea of a scientific theory that doesn’t make predictions, basing your belief in it on “circumstantial evidence”.
No, science is not about what is “TRUE”. Religious believers also would claim that they are concerned with what is TRUE and they have circumstantial evidence for it. The difference between religion and science is in how claims to know what is TRUE are evaluated. Science has made progress in the past because it has required that theories make testable predictions that can be convincingly verified. It may be TRUE that we are a simulation in a vat overseen by beings of higher intelligence, but it’s not science because we don’t have any way to test that hypothesis and see if it’s wrong.
The Intelligent Designers are going to have a field-day with this posting. They know what is TRUE, can point to circumstantial evidence for it, and will love to find out that mainstream physicists don’t see a problem with theories that can’t be tested, that this is just a minority view held by some marginal figures named Smolin and Woit.
I do have Greene’s The Elegant Universe DVD, but as a layman I can’t comment on string theory (it’s like putting the cart before the horse, physics-wise). These “stormy” discussions over ST just further confuses some of us folks, and being that Greene’s programs have been on PBS, the confusion should be well-noted. I am waiting to see the outcome of the String Theory in two minutes or less contest that Greene will be judging. Did any scientists enter this? I wonder…
Anyway, I must say, Clifford, that your prank was clever. It sort of confused me at first – “Since when do they give Hollywood stars for physicists??” And the Little Joe comment about Bonanza was clever as well (I wouldn’t have gotten that.) Did Brian ever see the photo? If I were him that would be a keeper on my wall.
An aside: You know, if there was a “Hollywood” Walk of Fame for people bringing science to the public in a big way, the Brian Greenes, Neil deGrasse-Tysons are making a run for it. BTW, I was pleased as punch to see that Mutual of Omaha’s Wild Kingdom making a comeback on Animal Planet. Marlin Perkins, Jaques Cousteau were some early heroes of mine. Later Carl Sagan (Attenborough was not accessible). Much of their information on the shows is still valuable and correct. I don’t recall any physicists on TV from my youth – why is that?
The issue of demonstrating the merits of a theory to your peers when there’s no new experimental guidance has all been discussed here very many times, Nigel. See this comment, and the surrounding discussion. I don’t really think there’s any point in going around in circles again. If you’ve nothing new to add, let’s just agree to disagree and move on, shall we?
-cvj
“If something demonstrably better or more promising is presented as an alternative, lots of people will work on it and help it develop further, and it may well strengthen into the dominant approach to finding the answers we seek about nature. This is the way it has always been in research. We just want to know what works.” – Clifford.
What happens if it the situation is like developing general relativity, which needed to be completed in 1915and then tested in 1919, before it was widely accepted and physicists invested time in studying it? How many people joined Einstein in his quest for general relativity? The mathematician David Hilbert worked on it, but he was more in competition with Einstein. It certainly wasn’t a mainstream project with massive support.
By the time the theory became “demonstrably better or more promising” it was too late for others to contribute.
In any case, what does “demonstrably better or more promising” mean here? You may have noticed that the most motivated string critics (Lee Smolin and Peter Woit) both have alternative programmes of their own, preon unification of particles and loop quantum gravity, and representation theory to find new symmetry principles underlying the standard model. Ockham’s razor might suggest that such models are more economical as they don’t require so many speculative assumptions.
Does “demonstrably better or more promising” just mean popularity contest indicators like citation counts, which are really expressions of prejudice and the politics of science (people cite papers related to what they are funded to research, which is determined by consensus and other political-type decisions)?
[much more repetitive material snipped for brevity… -cvj]
I reckon a lot of the heat that string theory suffers is aimed at theoroticians in general, not ST in particular. It is just one of the morehigh-profile, less empirical types of theory, and this an easy target for all of us data-collecting conscripts laboring in the trenches.
For example, here’s Inkling’s take:
http://www.inklingmagazine.com/articles/believing-the-unseen/