Shut Up and Calculate!?

I noticed recently that Nima Arkani-Hamed was giving a talk at Cornell, with the title Three Cheers For “Shut Up And Calculate!” In Fundamental Physics. No idea whether or not video is now or will become available.

From the abstract one can more or less guess what sort of argument he likely was making, and it’s one I’m mostly in agreement with. “Shut Up and Calculate!” is pretty much my unspoken reaction to almost everything I read purporting to be about foundational issues in quantum mechanics. I have in mind in particular discussions of the measurement problem, which often consist of endless natural language text where one struggles to figure out exactly what the author is claiming. An actual calculation showing what happens in a precise mathematical model of a “measurement” would be extremely helpful and likely make much clearer exactly what the problem is (or, sometimes, whether or not there even is a problem…). Such calculations are all too few in a huge literature.

Over the last few years, while teaching and writing a book about the mathematics of quantum mechanics, the tedious exercise of trying to get all signs right in calculations has sometimes turned out to be quite illuminating, with tracking down a mysterious inconsistency of minus sign leading me to realize that I wasn’t thinking correctly about what I was doing. I’m all too aware that this kind of calculational effort is something I too often avoid through laziness, in favor trying to see my way through a problem in some way that avoids calculation.

On the other hand, I’m not quite ready to sign up for “Three Cheers”, might just stick to “Two Cheers”. For a perfect example of what’s wrong with the “Shut Up and Calculate!” philosophy, one can take a look at the forthcoming Workshop on Data Science and String Theory planned for Northeastern in a month or so. They have a Goals and Vision statement which tells us that they plan to:

treat the landscape as what it clearly is: a big data problem. In fact, the data that arise in string theory may be some of the largest in science.

About being the “largest”, I think they’re right. The traditional number of 10500 string theory vacua has now been replaced by 10272,000 (and I think this is per geometry. With 10755 geometries the number should be 10272,755). It’s also the case that “big data” is now about the trendiest topic around, and surely there are lots of new calculational techniques available.

The problem with all this is pretty obvious: what if your “data set” is huge but meaningless, with nothing in it of any significance for the problem you are interested in (explaining the Standard Model)? This is not a new project, it’s an outgrowth of the String Vacuum Project, which I wrote about here, here and here. This started with a 2005 funding proposal, ended up getting funded by the NSF during 2010-2014. From the beginning there were obvious reasons this sort of calculational activity couldn’t lead to anything interesting, and as far as I can tell, nothing of any value came out of it.

For an opposite take to mine on all this, see the paper Big Numbers in String Theory, by Bert Schellekens. It contains an odd June 2017 preface explaining that it was supposed to be part of special issue of Advances in High Energy Physics devoted to “Big Data” in particle and string phenomenology (“all the ways we use high performance computing in addressing issues in high energy physics, and (in particular) the construction of databases of string vacua”). This issue was cancelled “as requested by the Guest Editors”. I wonder what the reason for this cancellation was, in particular whether it had anything to do with part of the topic of the special issue being considered by some to be obvious nonsense.

This entry was posted in Multiverse Mania. Bookmark the permalink.

61 Responses to Shut Up and Calculate!?

  1. paddy says:

    As an adherent of the shut-up-and-calculate school of thought, I was once asked how then to think of Schroedinger’s cat. My answer: I don’t, I’m a dog person.

  2. Armin says:

    “…the tedious exercise of trying to get all signs right in calculations has sometimes turned out to be quite illuminating, with tracking down a mysterious inconsistency of minus sign leading me to realize that I wasn’t thinking correctly about what I was doing.”

    Could you please elaborate a little? What was the “mysterious inconsistency”? the nature of the “illumination”? How does it relate to “shut up and calculate”?

    As stated, the paragraph gives the impression you are implying that many interpretational problems of quantum mechanics simply evaporate if one just does the calculations properly. Is that what you meant?

  3. Peter Woit says:

    Armin,
    The second paragraph was the one referring to interpretational problems of quantum mechanics, and it’s there that I’m suspicious that some problems disappear if you start to look carefully at and actually calculate what is happening when you have a specific system that is “performing a measurement”. For the kind of all too rare thing I have in mind, an example would be
    http://blog.jessriedel.com/2017/10/12/models-of-decoherence-and-branching/

    In my class and in my book there’s nothing about the measurement problem or interpretational issues, the topic is the mathematical structure of the subject, and that’s what I was referring to in the third paragraph. In that case, finding a minus sign made clear to me that I really needed to be using the dual of a vector space at some point when I thought I needed to be using the vector space itself. The clarification was just one of the mathematical structure of the problem, had nothing to do with interpretational issues.

  4. Mike McCracken says:

    Hi, Peter,

    Sorry to focus on a parenthetical here, but I’m intrigued by the following:
    “(and I think this is per geometry. With $10^{755}$ geometries the number should be $10^{272,755}$.)” Can you elaborate on this or provide a link? I’m not interested in strings, but this sounds like a deep geometric idea on which I’m obviously ignorant.

    Thanks,
    Mike

  5. Peter Woit says:

    Mike McCracken,
    You can find references in
    https://arxiv.org/abs/1707.00655
    No, there is no “deep geometric idea” here, just some complicated, ugly and uninteresting constructions. I wouldn’t advise spending time on looking into the details of this, I’m certainly not going to.

  6. Bee says:

    No no, “big data” is so yesterday. Today is “deep learning” and “neural networks.” Wait for it!

  7. Luca says:

    Yep, and the most amazing trait of the Deep Learning techniques is that algorithms can give you the right answer, but it’s impossible to understand why you’ve got that particular answer. In the end one could find the right vacuum from a zillion of possibilities and still having absolutely no explanation at all.

  8. tulpoeid says:

    Yep, and tomorrow’s thing is emergent computation. It could run on the landscape and realize that all solutions point to only one viable one, namely Standard Model with QFT. It would be a triumph of strings’ predictive power.

  9. Dave Miller says:

    Peter,

    As someone with an interest going back decades in the foundations of quantum mechanics, I generally agree. Your recent post regarding the Maudlin-‘t Hooft discussion is a case in point: logically something along the lines of what ‘t Hooft proposes might be possible. But until he or someone actually gives at least a toy model… well, it is very difficult to evaluate his Big Idea.

    On the other hand, if no one ever gives any thought to “superdeterminism,” “retrocausation,” etc., we might fail to trigger the random thoughts that eventuate in a real theory. Einstein did, after all, get a lot of mileage out of his youthful curiosity about what it would be like to ride a light wave!

    I don’t think decoherence will solve the “measurement problem,” but, like you, I think it is worth pursuing to sharpen our intuition and understanding of the deeper issues. And — who knows? — maybe the big contribution of string theory will be to sharpen thoughts about temporality, causation, etc. Polchinski suggested something along those lines some years ago, I think in his Rev. Mod. Phys. article.

    Dave

  10. Atreat says:

    Here is the problem: “shut up and calculate” reduces to “shut up” if you don’t know how to calculate. And in the case of quantum interpretations that leaves everyone silent.

    A better slogan might be, “By all means keep talking, but in the absence of the ability to calculate let’s not fool ourselves that we are close to a solution.”

  11. Adrian C Keister says:

    As a “hype iconoclast”, you might be interested to know, in the unlikely case you don’t already, that there is just as much empty hype in “big data” as there is in string theory, SUSY, etc. Probably more, actually. As you mentioned, if there’s no inherent meaning in the data, you can run it through fancy algorithms all day long and produce absolutely no new insights whatsoever.

    @Atreat: Your “better slogan” is certainly better, but I’m just not sure it’s got that catchiness we expect of meaningless slogans. You should work on that.

  12. Peter Woit says:

    Bee,
    They’re on it.

    “In comparison, a renewed thrust to study the landscape (articulated e.g. at String Phenomenology 2017) has many of the same goals as the SVP but places an increased emphasis on using modern techniques in data science to study the landscape. This approach is worth pursuing since the past decade has seen rapid advances in the areas of machine learning, distributed artificial intelligence, and generative adversarial networks. ”

    “generative adversarial networks” sounds like the latest thing, right out of Blade Runner 2049….

  13. Tim Maudlin says:

    This is a very telling sentence: “I have in mind in particular discussions of the measurement problem, which often consist of endless natural language text where one struggles to figure out exactly what the author is claiming.”

    Now I understand why you don’t my style. It all natural language and no math. Indeed, I have often said that the way many physicists read papers is the exact complement to how I read them: they read the equations and skip the prose and I read the prose and skip the equations. I figure they have not made a mathematical mistake, and the whole issue is *how the mathematics is being used as a way to represent physical reality*. A piece of mathematics is just that: a piece of mathematics. It isn’t, by itself, a physical theory. And it doesn’t magically become a physical theory by just explaining how to use it to make some predictions. Then it is part of a predictive apparatus. To become a physical theory, or rather be part of a physical theory, a mathematical formalism has to be interpreted, or, as I say, accompanied by a commentary. The commentary explains exactly what is being posited to exist by the theory (the ontology) and how that is being represented by the mathematics. To take is simple example, the usual mathematical formalism of classical electromagnetism can be used to express completely different physical theories depending on whether one takes the (mathematical) electric and magnetic fields to represent something physically real or the scalar and vector potentials to represent something physically real. And no amount of staring at the math will tell you. That is in the prose commentary.

    There is an extremely effective predictive apparatus called “quantum theory” that is not, in the sense I just articulated, a physical theory at all. So much of the discussion in foundations of quantum theory is about the commentary, not the math. If you are struggling through the prose (which in many cases can be obscure and poorly expressed because the ideas being defended are simply unclear) you are struggling though the essential part. Looking for equations will not be much help.

    John Bell, in “Against ‘Measurement'”, clearly explained the problems with standard presentations of the quantum theory. The problems are not in the math, they are in the prose. And on the other side, a theory like Bohmian mechanics, which completely solves the measurement problem, is not appreciated at all because there is no change to the math of Schrödinger’s equation. There is, of course, one extra equation, the guidance equation, but to appreciate its significance you have to read the prose. And then there is a lot of detailed mathematical analysis of those equations that has been worked out over 40 years that almost no one is even aware of or reads.

    This is a mess. But the way out of the mess is not to give even one cheer for “Shut Up and Calculate”. Pablo Echenique-Robba had it much closer to the mark in the title of his paper: “Shut up and let me think. Or why you should work on the foundations of quantum mechanics as much as you please.”

  14. Frederica says:

    Peter,

    there are quite a number of calculations of decoherence times, and measurements of the values. Are they not convincing for you?

  15. Peter Woit says:

    Tim Maudlin,
    The earlier objection to your style was about something different, things like : “And for years and years ‘t Hooft gives replies that only verify that he has not understood Bell’s work and keeps at it.”

    About arguments over interpretations, I understand this has nothing to do with the mathematical formalism. My problem when I start reading such arguments often has to do with imprecision of language: when you and ‘t Hooft are using the term “superdeterminism”, what exactly are you talking about? Of course part of my problem is a lack of expertise, insufficient time spent immersed in these debates and thus insufficiently aware of conventions for exactly how terms are used and ambiguities resolved. My sympathy here with “shut up and calculate!” is that often when I can’t tell exactly what argument is being made, my high school English teacher’s demand “Be specific, give examples!” comes to mind. Extremely helpful often would be not necessarily equations, but an explanation of what the issue is in the context of a specific physical measurement process.

    Frederica,
    I have no problem with calculations relevant to interpretational issues, I’m just saying that many interpretational debates I read could be clarified a lot by reference to specific calculations (see above).

  16. Tim Maudlin says:

    Peter,

    You will be glad to know that as a consequence of this discussion I have recommended dropping the word “superdeterminism” in favor of “hyperfine tuning”, which is fine tuning of the initial conditions of a theory not just to some macrostate (e.g. the Past Hypothesis) but to a microscopically specified set of states. I think it is a more descriptive term. In fact, I offered a whole set of exact definitions.

    As to the particular sentence of mine you cite, I am puzzled. It is blunt, but accurate. The title of your blog is also blunt, but accurate. But maybe it’s not worth the effort to figure out your problem here.

  17. vmarko says:

    Regarding the interpretations of quantum mechanics, they are always in the prose (as Tim says), but this is basically by definition, since interpretations should all agree on the math, and hence the math is the same in each interpretation. If two models do not agree on the math, then they are different theories rather than different interpretations.

    But the measurement problem is not an interpretational issue, it is rather a bit more serious. QM as a theory, both the math and the prose included, fails to tell us which processes constitute measurements (and, say, lead to the collapse of the wavefunction) and which don’t. The notion of measurement is always introduced intuitively, on a case-by-case basis, and plugged into the QM formalism by fiat, for every particular situation. Either by saying that the wavefunction collapses at some random moment of evolution, or by specifying the pointer basis for the decoherence and picking one particular vector from it, or by saying that this particular beable was detected in the apparatus, or otherwise. I agree with Peter that just speaking in prose about the measurement problem is not enough, and that one should instead write some equations that produce an actual model for the measurement process. And it should be clear that these equations should supplement QM, rather than be derived from it. Simply talking prose about it will never fill in for such equations.

    HTH, 🙂
    Marko

  18. Robert A.Wilson says:

    As an amateur in the field, I find the maxim “shut up and listen to the experts squabble” very useful. It does not take all that much listening to distinguish the experts who have thought deeply, but may ultimately be misguided, from those who shout loudly but have nothing to say.

  19. a1 says:

    Reading “How the Hippies saved Physics’ could remind you of some things the Calculators missed (not just the fun). And, by the way, the post/review/ from 2011 seems more nuanced than today’s.

  20. Robert A.Wilson says:

    On a more serious, but less important, note, several comments above make the distinctions between mathematical errors, physical errors, and interpretational errors. All three need to be addressed. As a mathematician, I am best qualified to comment on the mathematical errors. In the way the standard model is routinely explained, there are many serious mathematical errors. Many of these have been pointed out in this blog, such as the particularly egregious “so(3,1)=su(2)+su(2)”. Physicists rightly question whether this really makes any difference. And if you just “shut up and calculate” then most of the time the answer is no, it makes no difference. Clearly, the standard model works, and the calculations give the right answer. Most of the time. But this doesn’t address the issue. Much of the physical argument is based on an inconsistent, not to say incoherent, interpretation of the mathematics. What should a mathematician who is concerned about this do? If he says “the mathematics is wrong”, he is told “it doesn’t matter, it works”; if he says “the physical interpretation is wrong”, he is told “shut up, what do you know about it?”. And heaven help him if he tries to say “the physics is wrong”! But if we are ever to solve this problem, we have to start from the assumption that probably all three are wrong, and start listening to constructive suggestions for what to do about it. In particular I whole-heartedly agree with Peter that getting the signs right is an educative experience, that is not about getting the right answer, but about understanding the nature of the underlying model, and, dare I say it, reality. I don’t think anyone could say it doesn’t matter whether the Lorentz group is SO(3,1) or SO(4).

  21. tulpoeid says:

    (Tim Maudlin, he refers to your tone verging on the sarcastic and even conceivably impolite.)

  22. a says:

    I think most agree that t’Hooft was one of the greatest physicists of the latter half of the 20th century. And apparently he is one of the most straightforward and honest people you’ll see on the web, judging by his comments. But it looked to me like he was thinking a little fuzzily on the quantum interpretation business. Yes, the commentary was, um, rude, and I doubt t’Hooft would have taken that tone even in his heyday (which included some decidely modern topics) but the comments made some sense and were helpful (to me, at least, a nonexpert) in sorting through the issues. It also motivated me to reaquaint myself with Bell’s papers (ok…basically just speakable and unspeakable); his style is very appealing (for example, he was a great Einstein fan but maybe not so much Bohr, as was made clear in the very amusing first appendix to Bertlmann’s socks).

  23. Tim Maudlin says:

    Tulopoid: Yes, it is. So is “Not Even Wrong”: it means that a theory is so confused that you can’t even say that it is false because it doesn’t say anything coherent. Indeed, the overall tone of my original post—which was not addressed directly to Prof. ‘t Hooft but at a wide swatch of the field—should come across as extremely exasperated because I was, and am, extremely exasperated. Literally years of theoretical effort has been wasted because people do not understand Bell’s Theorem, which is nearly a mathematical triviality. The sociology of how this could happen is long and complicated, but the effect is there. And the systematic confusion of the notion of a “free variable” in a physical theory with the notion of “free will” as it applies to humans is so obvious as to be inexcusable. Just as Prof. Woit got rightly exasperated by the dominance of string theory in theoretical physics, people who work in the foundations of physics are exasperated that the most important and often simplest results in that field are systematically misreported and misunderstood.

    It was not my intention, but if some naked exasperation (including occasional sarcasm) manages to bring some attention and light to these issues, that is a good thing. Certainly the very measured and elegant and immaculately careful and concise writing of Bell himself didn’t do the trick. Sometimes people have to hit by a 2×4 to pay attention. That would have been beneath Bell, but I admit it is not beneath me.

  24. Dave Miller says:

    Re Bee’s and Peter’s comments on “deep learning” and physics: An interesting paper was posted on the arXiv earlier this year entitled Deep-Learning the Landscape! Despite the provocative title, this was basically an exercise at using deep neural nets (AKA multi-layer perceptrons) to “learn” various calculated quantities in algebraic geometry (specifically concerning Calabi-Yau manifolds).

    The paper looks legit, and the results are interesting, though I think it is fair to say that the paper shows that neural nets are not yet ready to replace human mathematicians!

  25. tulpoeid says:

    (Maudlin: I see, I think this point was worth clarifying.)

  26. Chris W. says:

    Speaking of deep learning, see yesterday’s news from Google’s DeepMind team on AlphaGo Zero.

    Echoing Luca’s comment, the recurring question with such accomplishments is how and when they are useful. With AlphaGo, Go players can study and learn from the strategies that it discovers. That the strategies are useful and effective is readily verifiable, and this is obviously essential to AlphaGo’s learning process. The fact that its learning—its experimentation—takes place in a completely automated way within a self-contained simulation environment on a high performance computing platform makes the process extremely rapid in human terms.

    Leaving aside the problematic motivations for even attempting to apply such techniques to understanding the string theory landscape, how much of the above can be said to apply to physics? Obviously, if part of the learning involves doing actual experiments with real physical systems then it can’t help be much slower, and subject to all the constraints that human experimentalists face every day. Learning non-trivial things about the physical world is much harder than learning something as artificial and well-structured as a board game.

    By the way, I completely agree about the current level of hype in “big data”. When commercial applications and marketing get involved, the amount of BS in circulation always rises precipitously.

  27. JoKing says:

    Hey Tim Maudlin,

    You see, the reason that physicists when pressed won’t posit any ontology for their equations is that this is not only not the most important part of what you might conceive of as a physical theory, but that it is in fact totally redundant and misplaced.

    To give any credence to that statement that “such and such entity exists and this is how it is represented in the equations”, you have to say what effect that entity will have in the real world. (Or else what do you mean?) Without an operational prescription, that statement is physically totally empty and can at best be viewed as a purely mathematical existential quantification.

    What’s more, the physical theory known as quantum mechanics in fact forbids you of talking about the physical reality of many of the objects you talk about when explaining what you think Bell did. (And I think Mermin now understands that meaning of “shut up” as well.)

    My point is indeed that your grief with the physics community arises very early on from your epistemological misconception of a “physical theory” rather than from their misunderstanding of Bell.

  28. Bill Bailey says:

    Tim Maudlin,
    There are many things natural language can’t get at that math can. The math needs no prose commentary because the math is telling you what the abstract relationships are itself. Physicists read the equations to understand what is going on without needing a running prose commentary. Some things like “free will” are metaphysical speculations that may actually be incoherent under precise examination based on the structure of the mathematical relationships that physicists discover. These problems like “the measurement problem” may only arise because one wants to force a translation between our prose language and the mathematics. One may be introducing problems that are not there. But that’s in your area. The dissolution of paradoxes and confusions.

  29. Peter Shor says:

    Peter,

    My take on the “shut up and calculate” philosophy is that it’s one of the reasons that quantum computation and information was not discovered by mainstream physicists, but by the motley crew of non-mainstream physicists, computer scientists, mathematicians, and semi-crackpots who discovered it. (And for the curious, I’m not going to say who I think falls into which category.)

    And maybe you should consider whether “shut up and calculate” might not be a reason that people haven’t spent more time studying quantum field theory deeply, the way that you are advocating.

    On the other hand, I am also convinced that studying the foundations of quantum mechanics in the manner that many people have been doing it is a waste of time. But I believe the “shut up and calculate” philosophy’s influence has been much broader and more negative than that.

  30. F4 says:

    I attended Arkani-Hamed’s talk (though when prompted by this post I realized that I recall shockingly few details about it now (something for which my memory is solely to blame); I also had to leave just as he was finishing skipping through the roughly sixty (?) percent of the slides that he didn’t discuss at all).

    Hopefully any recording made will be decent: he was forced to suffer some inexplicable (and inexcusable, I would say) technical difficulties before commencing, had to resort to standing directly in front of the lectern microphone, and eventually seemed to give up or forget and so spent most of the talk perambulating around the front of the hall sans amplification.

    Anyway: I thought that he made it fairly clear at the beginning that both a rigorous shut-up-and-calculate pragmatism and a broader philosophical vision are necessary for real scientific progress. But indeed speculation must be grounded: a segment of the talk concerned Einstein’s perserverance in improving his mathematical knowledge and his spurious justification of failed pre-GR proposals with a variety of philosophical principles. Arkani-Hamed also said that getting the mathematics of a concept right can be a (more or less) combinatoric exercise once you have delivered yourself into the right theoretical attractor space.

    I don’t remember much if anything said about applications to string theory in particular but then again he seemed to be ready to talk for at least two more hours than the time allotted.

  31. Tim Maudlin says:

    JoKing,

    The question is not how a physical theory with clear ontological postulates (i.e. saying clearly what physically exists according to the theory) has any trouble at all with making empirical predictions, it is just the opposite: how can you derive empirical predictions from a theory that make no clear posits about what exists?

    Quantum theory does not “forbid you from talking about” anything. How could it? I mean what is quantum mechanics going to do to me if I insist on talking about things like what really exists at microscope? Bohmian mechanics, for example, says that point particles do. And that tables and chairs and cats and people are composed out of point particles. And how the point particles move (what exists and how it behaves). So it automatically makes predictions about tables and chairs and cats and “measuring apparatuses” without having to mention “measurement” or any other problematic notion.

    My conception of a proper physical theory as postulating what exists and how it behaves has zero epistemology built into it. It just isn’t an epistemological thesis of any kind. So you are confused there.

  32. Tim Maudlin says:

    Bill Bailey,

    A piece of mathematics without some indication of how it is being used as a mathematical representation of something physical just isn’t, and can’t be, a physical theory of anything. I mean, how could you tell even what it is supposed to be a theory of, what phenomena it is supposed to account for? How could you get experimental evidence for or against it? That extra information comes in the commentary. It is given in natural language, not mathematics.

    As a simple example: take the mathematical formalism of classical E & M. Accompanied by the commentary: “The mathematical E and B vector fields represent the physically real electric and magnetic fields; the mathematical A vector and phi scalar potentials are mere mathematical conveniences that represent nothing physical” that math has one sort of physical significance. Accompanied by this commentary: The A vector field and phi scalar field represent physically real things, and the E and B fields are just mathematical conveniences” the very same math underlies a completely different theory.

    I hope this helps make the point.

  33. Manfred Requardt says:

    I think Peter Shor is right. I know many mathematicians who think they do understand quantum mechanics if they understand the Schroedinger equation. Mathematics is only an important tool in natural science. Most of our important ideas arise in form of images and words. As far as I can see, many of our greatest colleagues were not of the shut up and calculate type (e.g. Einstein, Bohr, Heisenberg). I believe the really deep problems can never be found by shut up and calculate.

  34. André says:

    JoKing:
    I think, “operational prescription” is just a euphemistic way of saying “shut up and calculate”. It’s of course nice to have a mathematical model that tells you what reading (with which probability) you will get from a photodetector under certain initial conditions, but at the end of the day most physicists will probably agree that the photodetector itself is built of atoms that need to be described quantum mechanically. So, from my point of view, it seems clear that the “operational prescription” can’t be more than an effective theory.
    Effective theories can be very useful, thermodynamics, for example, has quite a lot of applications for which it is completely irrelevant to understand the microscopic definition of temperature, entropy, etc., but if you want to gain some deeper understanding of the world, you will eventually end up asking what exactly these quantities are.
    The way I see it, the measurement problem is pretty much at the same level. Quantum theory is a perfectly predictive theory for the outcome of specific experiments, but it doesn’t teach us anything about what is “really” going on at a deeper level.
    Also, I disagree that quantum mechanics forbids us from talking about the physical reality of objects. It is just that the real objects in (Copenhagen) quantum mechanics are very odd ones: the measurement devices/outcomes.

    What I don’t understand – and this is a purely sociological and not a scientific observation – is how many physicists are totally obsessive about the assumption that quantum theory must somehow be the end of the story, and not just some limit of some more elegant underlying theory. This is against everything we should have learned from the history of science, and somehow against the very idea of science.

  35. André says:

    Tim:
    Although I very much agree with your point of view and enjoyed reading your conversation with ’t Hooft, a question regarding your example for classical EM:
    I could just consider a concrete set-up, say a Hall probe in an EM field, and have a perfectly well defined mathematical formalism that tells me: given this experimental set-up, my B-probe will show the value 0.5 Tesla (e.g.). For this, I don’t have to give any reality to either E and B or A and phi.
    Of course, you would still need a commentary, namely stating that the output of your mathematical algorithm gives you the reading of your B-probe, but I would still be interested about your view on such an approach, since I think it is pretty much what many physicists have in mind when they argue like JoKing and Bill Bailey.

  36. Bee says:

    @Dave Miller,

    Thanks for the reference, looks interesting!

  37. Bee says:

    @Tim Maudlin,

    Replacing one word with another word is not a definition.

  38. To become a physical theory, or rather be part of a physical theory, a mathematical formalism has to be interpreted, or, as I say, accompanied by a commentary.

    An established term for this in the philosophy of physics is coordination.

  39. Peter Woit says:

    All,
    Sorry, but I have to cut off further attempts to use this as an interpretation of quantum mechanics discussion board, in effect telling people to shut up. I’ve been trying to delete those (many) comments that don’t appear to make an argument worth reading, but have now had enough. Whatever my sins, I don’t deserve having to moderate a discussion of exactly the sort I was criticizing as a waste of time.

  40. onlygamintwn says:

    You should take a look at https://arxiv.org/abs/1709.03554
    since you are taking about string vacua or numbers like 10^{whatever}.

  41. Prof. Woit,

    I have to peacefully disagree with your assertion that the measurement problem is somehow a ill-posed natural language problem. It is mathematical.

    Consider a quantum system in the state \ket{\psi_0}. Then a measurement happens, taking it to the state \ket{\psi_1}. What is the evolution from \ket{\psi_0} to \ket{\psi_1}? Is it a projection? Is it a unitary transformation? It’s easy to prove that these alternatives are mathematically incompatible, so it can’t be both (I guess you already know this, but I’d be happy to provide a proof if you’re interested).

    Of course it might be even something else, but I guess you are even less interested in this possibility than you are in the foundations of quantum mechanics.

  42. Peter Woit says:

    Mateus Araujo,
    I never said the measurement problem is an ill-posed natural language problem. My comment was that many natural language discussions of it don’t seem to me to touch the actual problem. When I try and read many such discussions I soon get lost, unsure exactly what the author is saying, and thus whether a real problem is getting addressed. There are also plenty of examples where the author writes unambiguously, perhaps using a mathematical formalism to achieve precision. Typically in such cases I can at least understand what is being said, and when the author moves to discussing something I don’t think is interesting or relevant, I know when to get off the bus.

    My positive take on “shut up and calculate” here is that it would often be useful if someone making a claim to say something new about the measurement problem would discuss an actual measurement instead of making imprecise claims about all “measurements” (including being imprecise about what a “measurement” is). Pick an actual physical system and work out what happens (i.e. “calculate”), showing where the “problem” occurs and explaining what you have to say that’s new and enlightening about this problem. When I see this kind of activity, I can follow it, and often learn something interesting, often that the “problem” has some different aspect that wasn’t obvious to me before.

  43. Prof. Woit,

    Indeed, that is what you had written. Sorry for misunderstanding you.

    I think saying something *new* about the measurement problem is close to being a literal impossibility, due to the sheer amount of literature written about it. As a corollary, the solution to the measurement problem has already been written, but finding it in the literature is also nigh impossible, due to this “Library of Babel” situation.

    Nevertheless, I disagree with you that making some precise calculation about some specific physical system would be of any help. I’ve actually met a guy in a conference that claimed that he had solved the measurement problem by doing exactly this. This is his paper. It does make a nice mathematical model of a measurement, up to where the outcome should be produced, where he writes some stuff that I read as “then a miracle occurs”.

    The problem is that this sort of details hardly matters, the points about which there is disagreement are much more general features of the measurement process. This paper by Maudlin, for example, discusses in a rigorous way what the features are and where there is disagreement. I think that if your solution doesn’t address the problems he is bringing up then you are completely missing the point (and yes, it makes me feel dirty to agree with Maudlin on something).

  44. Robinson says:

    A comment section with a Tim Maudlin observable… A pleasure to read.

  45. Doug McDonald says:

    An earlier comment had a pointer to this blog:
    http://blog.jessriedel.com/2017/10/12/models-of-decoherence-and-branching/

    An even better discussion, really definitive as blogs go, is
    http://blog.jessriedel.com/2016/11/12/how-to-think-about-quantum-mechanics-part-0-measurements-are-about-bases/

    especially in part 4.

    I’m completely shut up by that blog series.

  46. Curious Mayhem says:

    Isn’t the whole “shut up and calculate” ethos just an extension of what Lee Smolin was complaining about when he criticized the whole field has having degenerated into virtuoso but mindless calculation mixed with a lot of hand-waving? You know, as in, no one is thinking about anything physically?

    We’re a long way from Einstein thinking up the invariance of light rays or the Equivalence Principle, or the pioneers of quantum mechanics struggling with classical variables in the face of the quantum of action, or Feynman’s insight that yielded the path integral. All of those required formidable mathematics to implement. But the original insights were not mathematical.

  47. Jeff M says:

    Curious

    Lee was trained by the same physicist I was, before I went over to the dark side (math). Herb was brilliant, but no computationalist. He was always very physically motivated, and interested in why things were the way they were. He went on to be very involved in quantum teleportation stuff, despite teaching at a small liberal arts college. My perspective now is somewhat different, but the hand waving drives me nuts, mathematician that I am.

  48. Peter Woit says:

    Curious Mayhem,
    To a large extent, I disagree with Lee about this, and agree with Arkani-Hamed. I think the problem with string theory is not too much virtuoso math, but that it’s a wrong physical idea about unification. The physical idea impressed a lot of people, but when you try and write down a fully consistent theory and calculate anything meaningful you find that you can’t . It’s the physical idea that is leading you down the wrong path, the actual calculations implementing it that show that this is the wrong path. Of course, once your calculations have shown you you’re on the wrong path, one possible reaction is to ignore this and do more calculations, which is what the string vacuum people are doing, showing that calculations are not always the answer.

  49. Robinson,

    I think Maudlin is actually a beable.

Comments are closed.