This Week’s Hype

Commenter CIP pointed out that today’s New York Times has one of the worst examples of string theory hype I’ve seen in a while. Based on this observation of an expected QFT anomaly effect in a condensed matter system, the NYT has an article An Experiment in Zurich Brings Us Nearer to a Black Hole’s Mysteries. Not only is the headline nonsense, but the article ends with

The experiment is also a success for string theory, a branch of esoteric mathematics that physicists have used to try to tie gravity into the Standard Model, the laws of physics that describe the other forces in the universe. But string theory has been maligned because it makes predictions that cannot be tested.

Here, Dr. Landsteiner said, string theory was used to calculate the expected anomaly. “It puts string theory onto a firm basis as a tool for doing physics, real physics,” he said. “It seems incredible even to me that all this works, falls all together and can be converted into something so down to earth as an electric current.”

There’s no connection at all to string theory here. The NYT seems to have been taken in by string theorist Landsteiner and press release hype like this, not noticing that the paper had no mention of string theory in it. The hype is timed to the paper’s publication in Nature, where the editor’s summary gets it right, referring to QFT not string theory:

Johannes Gooth et al. now provide another intriguing connection to quantum field theory. They show that a condensed-matter analogue of curved space time can add an additional, gravitational component to the chiral anomaly in Weyl semimetals. The work opens the door to further experimental exploration of previously undetected quantum field effects.

Someone really should contact the NYT and get them to issue a correction. In particular, any string theorists who care about the credibility of their field should be doing this.

Update: For a couple more stories about this, IEEE Spectrum has Black Hole Power: How String Theory Idea Could Lead to New Thermal-Energy Harvesting Tech, Nature has Big Bang gravitational effect observed in lab crystal.

Update: The author of the NYT piece did make some changes in the last two paragraphs to make things less misleading.

Update: This has finally appeared in print today, in an abbreviated version, minus among other things the string theory business.

This entry was posted in This Week's Hype. Bookmark the permalink.

54 Responses to This Week’s Hype

  1. Mitchell Porter says:

    String theory aside, is this, or is this not, an observation of an effect attributable to a gravitational anomaly?

    There’s a type of “condensed-matter hype” in which observation of monopoles, Hawking radiation, etc, is claimed, when of course the reality is just that something *analogous* to the claimed phenomenon was observed.

    What I get from skimming the theory papers is that an effective field theory for hydrodynamics in curved space-time can have a term whose coefficient is determined by the existence of the anomaly. Then I see remarks that this can still be true in flat space-time too. And finally, I see statements that in this experimental work, there’s a temperature gradient which is analogous to space-time curvature in its effects.

    So what’s going on? Is this another example of condensed-matter hype, that should be rectified by giving things their true names?

  2. Shantanu says:

    Dear readers,
    I suggest you try to write an email to the writer of this article pointing them this faux pas. On many times when Peter or others posted such hype articles I tried to find the email address of the person who wrote the article and in many cases it was hard to find. But I am guessing its easier in this case.

  3. Giulio says:

    Another piece of fake news from NYT. SO SAD!
    D.

  4. Citchell says:

    The section titled “1 Connection between the mixed axial-gravitational anomaly and thermal transport” directly references work by Landsteiner, Witten and Alvarez-Gaume, and Eguchi. Also a guy who has worked on born reprocity.

    http://www.sciencedirect.com/science/article/pii/055032138490066X
    https://en.wikipedia.org/wiki/Eguchi%E2%80%93Hanson_space
    https://en.wikipedia.org/wiki/Born_reciprocity

  5. bpv says:

    I sent a comment to the author of the article, Kenneth Chang, and he replied almost immediately with a likely pre-written email that he must be sending to everyone complaining about the article (he refers to them as “those who are outraged”).

    He basically says that he is aware of the reasons why knowladgeable people are concerned, but for most laymen the mention of string theory was just fine.

    Let me translate: He can say whatever he wants to people who have no capacity to check the truthfulness of the statements.

  6. Ed says:

    OT: could topological physics be the field to finally provide further insights and discoveries to break out from the stagnated situation on particle physics?
    https://www.nature.com/news/the-strange-topology-that-is-reshaping-physics-1.22316

  7. Thomas says:

    I think your outrageous hype detector is a little too sensitive. Yes, I would prefer if the author of the NYT article adds a statement along the lines of “this result does not have immediate implications for attempts to unify the SM with gravity”, but overall the article is quite accurate. In particular, I see absolutely nothing wrong with the Landsteiner quote, or with the “This work opens the door .. ” quote.

    People have now discovered a number of transport phenomena in flat space that are most easily determined by putting QFT in curved space. This is more than mere analogy as some commenters here seem to think. It is using QFT in curved space to implement the constraints that follow from Ward identities, anomlies, etc. As a result certain transport coefficients that look like ordinary themopower or magnetoresistance in flat space are indeed fixed by gravitational anomalies. That’s pretty cool, and it is remarkable that we can now engineer condensed matter systems in which these effects are seen.

    It is also the case that some of these transport coefficients that could have been discovered by Landau and Lisfschitz by analyzing the symmetry constraints for currents in hydrodynamics were in fact by analyzing holographic setups. This does mean that AdS/CFT is an efficient method to think about difficult problems in QFT.

  8. Peter Woit says:

    Thomas,

    The problem is that this has nothing at all to do with string theory. It’s purely a QFT story (as the Nature editor recognized). Going to the press with a claim “string theory was used to calculate the expected anomaly” is an intentional attempt to mislead people about the science here.

  9. Thomas says:

    but what if it is historically accurate?

  10. Peter Woit says:

    Ed,
    That’s a very good article that actually explains some of the real new connections between math and physics being investigated. I don’t think it has anything to do with HEP physics.

  11. Peter Woit says:

    Thomas,
    It’s not in any sense “historically accurate”. The anomaly calculation in the graphic associated with the article never was a “string theory calculation”. It’s a QFT calculation, reflecting the fundamental relation between the Dirac operator coupled to a gauge field and the index theorem. This was a major topic during the early eighties when I was a grad student, before string theory came into the picture.

  12. Thomas says:

    The connection to transport in flat space was discovered this way.

  13. Peter Woit says:

    Thomas,
    Even if there is some non-trivial connection to Maldacena’s AdS/CFT conjecture (which I don’t see, but I’m no expert) that doesn’t turn a straightforward QFT calculation into a “string theory calculation”.

  14. Thomas says:

    The anomalous transport calculation in QFT is not straightforward, and historically it was done using AdS/CFT.

  15. Peter Woit says:

    Thomas,
    I don’t think this is relevant to the fact that the claims about “string theory calculations” in the article are flat-out incorrect, but I’m surprised that when I tried looking through the references in the paper, I find nothing about AdS/CFT. The few that refer to “holography” seem to only be using standard QFT ideas about anomalies
    Can you provide references that would justify the NYT graphic as a consequence of AdS/CFT? I’ve spent a little time looking at the references in the paper, nothing uses AdS/CFT, the few that refer to “holography” seem to have to do not with AdS/CFT but with standard old facts about anomalies and domain walls.

  16. Chris Herzog says:

    Perhaps they were limited in the number of references they could include for a Nature paper. Landsteiner cites his own review article [5], https://arxiv.org/pdf/1610.04413.pdf. Section 6 seems to be relevant to the argument of how important AdS/CFT was in the computation of anomalous transport, as is this sentence from the conclusions: “While holography can probably not claim to have discovered anomalous transport it has certainly played a major role in gaining a better understanding.”

    As Thomas mentions, these anomalous transport calculations are not straightforward. Mistakes in the textbooks were first noticed by doing analogous calculations in AdS/CFT.

  17. Thomas says:

    I think a nice historical summary of the role of AdS/CFT in anomalous transport is the intro of Son & Surowka, https://arxiv.org/abs/0906.5044 . The NYT graphic presumably refers to something like https://arxiv.org/abs/1107.0368

  18. Peter Woit says:

    Chris Herzog and Thomas,
    Thanks for the references. I had looked at the Landsteiner review article, noticed that he invokes AdS/CFT in that one section, but I still don’t see AdS/CFT getting used to explain significant. My reading of the Son/Surowka article is that their topic is effects of anomalies on relativistic hydrodynamics, that they only use AdS/CFT as a test calculation to check consistency.

    From your comments it’s looking to me like the only relation to AdS/CFT is that researchers in this area did analogous calculations in AdS/CFT and this may have been a useful part of the process of sorting out what was going on, but that’s very different than the claim that anomalous transport is based on AdS/CFT.

  19. william e emba says:

    Researchers in fluid analogues to black holes (sometimes called “dumb holes”) freely describe just about everything they do using the language of general relativity. The very title of https://www.nature.com/nphys/journal/v10/n11/full/nphys3104.html refers to “black hole” and “Hawking radiation”. They were studying a Bose-Einstein condensate in their lab.

  20. Peter Woit says:

    william e emba,
    Yes, but the question isn’t whether GR is being used in this kind of modeling, it’s whether string theory is being used, as claimed in the NYT article.

  21. Andreas Karch says:

    Hey Peter,

    my local condensed matter colleagues here are quite adamant that the effects seen in this and similar experiments should not be interpreted as being related to the mixed gravitational field theory anomaly, so I think there are some legitimate theoretical questions one can ask about this paper. But I also think you are really trying too hard to undermine the AdS/CFT connection.

    These anomalous transport coefficients have first been calculated in AdS/CFT. The relevant references are [8], [9] and [10] in the Son/Surowka paper. In the AdS/CFT calculations these particular transport coefficients only arise due to Chern-Simons terms, which are the bulk manifestation of the field theory anomalies. At that point it was obvious to many of us that there should be a purely field theory based calculation, only using anomalies, that can derive these terms. Son and Surowka knew about this. They were sitting next door to me when they started these calculations. Many of us tried to find these purely field theory based arguments and failed. Son and Surowka succeeded.

    If you ask anyone serious about applying AdS/CFT to strongly coupled field theories why they are doing this, they would (hopefully) give you an answer along the lines of “AdS/CFT provides us with toy models of strongly coupled dynamics. While the field theories that have classical AdS duals are rather special, we can still learn important qualitative insights and find new ways to think about strongly coupled field theories.” Once AdS/CFT stumbles on a new phenomenon in these solvable toy models, we want to go back to see whether we can understand it without the crutch of having to rely on AdS/CFT. Any result that only applies in theories with holographic dual is somewhat limited in its applications. In this sense, anomalous transport is a poster child for what AdS/CFT can be used for: a new phenomenon that had been missed completely by people studying field theory gets uncovered by studying these toy models. Once we knew what to look for, a purely field theoretic argument was found that made the AdS/CFT derivation obsolete.

    This is applied AdS/CFT as it should be. Solvable examples exhibit new connections which then can be proven to be correct more generally and are not limited to the toy models that were used to uncover them.

  22. william e emba says:

    So? Results from XYZ are lifted to hydrodynamics and touted as such even the work is obviously not XYZ, you say this is acceptable when XYZ=GR and obnoxious hype when XYZ=string theory? I don’t buy that.

    I see this type of idea credit in other fields. I’m perfectly happy with the terminology “simulated annealing” or “genetic algorithms” or “evolutionary programming” or “ant-colony optimization”. There’s no point in hiding the inspiration, especially if more progress is expected from applying the sources.

  23. Peter Woit says:

    Andreas,

    Many thanks for your comment, which clarifies a lot, in particular making clear to me why I was having so much trouble seeing the relation to AdS/CFT in calculations that didn’t seem to be using it. I gather that the answer is that, as Thomas suggested, the relation is a historical one, with AdS/CFT research what led to a current theoretical understanding that doesn’t use AdS/CFT.

    I think it should be clear that my problem with the NYT article has nothing to do with any complaints about this sort of research, but with its use to obfuscate the situation of string theory as a unified theory.

    Here is a question for Andreas, Thomas and Chris, about the NYT article. What do you think of the last two paragraphs of the article that I quoted? Are they scientifically justifiable or seriously misleading/incorrect? I’m making the argument to the NYT that these two paragraphs should be deleted. What say you?

  24. Peter Woit says:

    william e. emba,

    The point is that the status of GR as a tested theory is uncontroversial, but here the invocation of string theory is explicitly being made so as to obfuscate the failure of the heavily promoted string theory unification program. Instead of acknowledging this failure (which took the form that the idea turned out to be inherently untestable), claims are made (as in the next to last paragraph of the article) that string theory is testable, and has been successfully tested. The paragraphs I quoted are the result of an intentional not very honest campaign to mislead people.

  25. Chris Herzog says:

    Hi Peter.

    I hesitate to make a judgment about the news article and to extrapolate from what is said there to what is actually believed by the scientists involved. So often, things are misquoted or stretched in misleading ways. I know Karl, and I suspect Andreas, Karl, and I are all pretty much on the same page when it comes to AdS/CFT and the role of string theory therein. I do not have as strong a reaction to those two paragraphs as you did, and I can see my vision, perhaps in a distorted form, behind what is written there.

    I see AdS/CFT as a tool for doing certain types of field theory calculations, a tool like the epsilon expansion or like lattice gauge theory or like integrable systems. AdS/CFT is at its most rigorous when embedded into a string theory. The fact that one can calculate conductivities and viscosities in maximally supersymmetric Yang-Mills for me already puts AdS/CFT and by extension string theory on a solid footing, or at least on a footing worthy of a substantial research effort. That we have learned something about anomalous transport from AdS/CFT and that anomalous transport could be seen — not in holographic systems, but nevertheless seen — in the lab is icing on the cake.

  26. william e emba says:

    Fair point, although from my point of view, it’s success in mathematics (including that of three Fields medalists, Witten, Borcherds, Kontsevich) means it’s as legitimate as GR is when making analogies. String theory has proved its chops, just not in the way the hypesters intended.

  27. Peter Woit says:

    william e emba,
    I think it actually matters to distinguish between what has been successful about “string theory” and what has been a failure. My problem with this article is that it misleads people about this.

    By the way, Witten’s Fields medal was for work (Chern-Simons) on QFT, not string theory.

  28. Andreas Karch says:

    Hey Peter,

    I am not nearly as disturbed as you are by the last two paragraphs. AdS/CFT certainly has been used “as a tool for doing physics, real physics”. I think there is no doubt about it. This seems to be in the end the only thing that is really claimed. It’s a tool that has its limitations (you study the wrong theory), but it works in situations where we have basically no other tools available (time dependent processes in strongly coupled theories ). In detail, of course things are complicated. The stringy toy models were only part of the story. The way the second to last paragraph (talking about the standard model and unification) leads into the last paragraph (describing how string theory was useful in this situation) is open to the misinterpretation that they tested quantum gravity in the lab (which they didn’t). I have different problems with the article that concern me more, but they are of a more technical nature.

  29. Peter Woit says:

    Chris and Andreas,

    To make the question more specific, recall that the article states
    “string theory has been maligned because it makes predictions that cannot be tested”
    and then goes on to imply that the maligners have now been shown to be wrong, quoting Landsteiner about string theory now being on “a firm basis as a tool for doing physics”. Is this or is it not highly misleading (in particular the move without any acknowledgement from string theory as a 10/11 d unified theory to gauge-gravity duality as a tool in condensed matter physics)?

    I can see many good arguments for pursuing AdS/CFT as a tool for understanding strong coupling QFT, but why not sell this for what it is (and acknowledge that maybe the maligners of string theory unification were right?).

  30. william e emba says:

    Atiyah’s laudatio of Witten mentioned three things: TQFT (which paper referred explicitly to using string theory arguments), new proof of positive energy theorem (using SUGRA), and new formulation of Morse theory (using SUSY). That’s string theory enough for me.

  31. Peter Woit says:

    william e emba,
    I’ve found from bitter experience that it’s hopeless to argue with people about this. That Witten won the Fields medal because of his work on string theory is now a deeply entrenched alternative fact. If you are interested in real facts, recall that TQFT/SUSY/SUGRA are not string theory and take a look at
    http://www.math.columbia.edu/~woit/wordpress/?p=99

  32. Matt Grayson says:

    I’ve used this analogy. If we make no distinction between Newton’s theory of gravitation and the mathematical tools he developed for it, then I could say that I’ve put Newton’s gravitation theory on a firm footing by doing a calculus problem.

  33. Jeff M says:

    Just to 1/3 support Peter, Witten’s work on Morse theory was not string theory. That paper was the basis of my orals in grad school, so I know it well. Very nice math, nothing to do with strings. I’ve seen some math talks which are string theory, a few of them very interesting (Kevin Costello, on moduli spaces of Riemann surfaces, must be 20 years ago). Witten’s work on Morse theory at least wasn’t related.

  34. Deane Yang says:

    As a biased observer, I wanted to make some comments:
    1) This appears to me to be a debate on semantics. Is “string theory” a (not necessarily rigorous) mathematical tool that was developed as an effort to develop a TOE? Or is “string theory” the TOE itself?
    2) It appears to me that the main objection to the last two paragraphs is the concern that people who view string theory as the TOE will believe that Landsteiner’s work represents progress in string theory as such.
    3) On one hand, this is a valid concern that I agree with. On the other, this is not national politics. Having readers being misled on this does little harm. To me what’s more important is to have more articles about science that people want to read. If poetic license is taken, I think we should try to live with it. Mentioning string theory here, even if it’s a stretch, is good publicity. I know a big concern is the overfunding of string theory, but I think a bigger problem is the underfunding of scientific research.
    4) Please be cognizant of the limits on what a journalist can do, including the time to research the article and the length of the article. It is hard to be precise without being dry in a 500-1000 word article. It is hard to be precise, if you are not an expert in the field and have to talk to experts who are unable to express themselves clearly to a non-expert. It is hard, period, to assess who is right and who is wrong. Even in a hard science like physics, there is a lot of subjectivity.
    5) So what can you do when you don’t like what an article says? Luckily, articles can still be revised even after they’ve been posted on the web. So you can contact the journalist and ask for a correction. However, you’ll increase your chances of getting the correction made, if you make the journalist’s job easier. This means trying to isolate what part of the article you’d like revised (here, it is the last two paragraphs) and suggesting new wording that’s roughly the same number of words as the original wording *and* interesting to read. As someone who has tried to help Ken write about math, I know that this is really, really hard. Everything I write for him turns out to be far too long, literal, and dry. This is a huge handicap when trying to help a journalist write about science and math.
    6) It’s not too late to email Ken with suggested rewording. If you believe you can write the last two paragraphs in a less misleading way, please try. But it seems to me that you have to live with the quote by Landsteiner, which is not literally wrong. It’s just that you want to find a way to express that Landsteiner’s work was, if anything, a validation of one particular technique used in string theory, rather than string theory as a TOE.
    7) By the way, if you want to see a masterful job by Ken writing about math, please read his obituary of Maryam Mirzakhani.

  35. vmarko says:

    Sooo, finally string theory stops being a theory about high energy physics, and moves ever more firmly into the solid state domain. It started as an effective theory of strong interactions, had one long stint as a theory of everything, and now moves on to calculating conductivities and viscosities in solid state physics. After that, what’s next, maybe plasma physics and fusion reactor modeling using AdS/CFT techniques?

    I’ll probably have mixed feelings when the everlasting flood of papers published by string theorists moves from hep-th to cond-mat section of the arXiv. 😉 Just imagine, titles like “Electron transport characteristics in an 11D supersymmetric hexagonal crystal lattice compactified to a 3-torus with a negative cosmological constant”, and similar, appearing on cond-mat… 😀

    🙂
    Marko

  36. Mitchell Porter says:

    If any of the experts are still reading… I am still trying to understand the *mechanism* of MAGA-induced (?) anomalous transport. Surely it doesn’t involve virtual gravitons?! But then, how does it work?

  37. william e emba says:

    I personally think of SUSY/SUGRA as part of “string theory”, but I’m perfectly happy if people think that is just ignorant twaddle on my part. I was not confused about the history, and our main disagreement is simply what does “string theory” mean. I completely agree that Witten was not given the Fields medal for his work in string theory.

    So far as I know, neither Borcherds or Kontsevich did work in string theory, but they took what was there and ran with it. All I meant was that Witten did the same thing. For example, in his TQFT paper he explicitly mentions that his central arguments “will be quite recognizable to string theorists”.

    I believe we agree far more than we disagree, but in my attempts at brevity I have been unclear about my meaning. I apologize.

    @Matt Grayson: considering how inconsistent the foundations of calculus were in Newton’s day, there really was a time when getting calculus to work was a point in Newton’s gravitational favor.

  38. Peter Woit says:

    Deane,

    I have written to Ken with my suggestion about how to fix the article, which is pretty simple, just delete the stuff about “string theory”. It seems we disagree about this. That’s too bad, in general I think he and others at the NYT do a very good job of carefully covering math and physics, not always an easy task.

    The fundamental problem with the NYT article is the use of the term “string theory” to mean “gauge-gravity duality calculational method”, something with zero relationship to the “string theory” that is supposed to unify physics that readers have heard about, so they’re going to be misled.

    If you no longer know what idea “X” refers to, it becomes impossible to have a sensible conversation about “X” and things descend into the realm of ideology (are you for or against the meaningless term “X”?). My take on what’s going on is that X=”string theory” is in the process of becoming not a scientific term, but a clan name. This may be good for publicity purposes, but it’s not good for science.

  39. If one replaces occurrences of “string theory” with “the mathematics of string theory” then many problems go away.

  40. Peter Woit says:

    David Roberts,

    I don’t really think so. “the mathematics of string theory” is a really ill-defined term, and under many interpretations has little or nothing to do with the condensed matter story of this article.

  41. I think it is useful to recall the relevant historical timeline.

    The connection between axial-gravitational anomaly and transport (in flat space) at finite temperature, zero chemical potential was first suggested in a paper by Landsteiner, Megías, and Pena-Benitez, https://arxiv.org/abs/1103.5006. The effect was first pointed out to be possible by Neiman and Oz (arXiv:1011.5107). Landsteiner et al. suggested that the coefficient of this effect is equal to the coefficient of axial-gravitational anomaly, multiplied by temperature squared. How did they know? They constructed a simple theory in AdS5 so that the corresponding CFT4 has an axial-gravitational anomaly. Then they considered a black-brane solution, and by using AdS/CFT, found a “chiral vortical” coefficient proportional to the coefficient of the axial-gravitational anomaly.

    I remember being very perplexed by the claimed connection between transport in flat space and anomaly in curved space appeared to me very strange. Not only that Surówka and I missed the coefficient in our paper (mentioned by Andreas), it was also that if we tried to correct our calculation (as Neiman and Oz did), all we could do was to restrict the coefficient to be an constant times T^2, with no indication that the constant is related to any anomaly. In contrast, naive power counting would imply that contribution of the axial-gravitational anomaly should appear at the third order in the derivative expansion, rather than at the first order, as Landsteiner et al. claimed.

    For a more than a year, the issue was not understood. There was strong indication in favor of the connection from AdS/CFT, but it was not clear how general it is. If the coupling in field theory is not infinite, but only finite, would the connection still hold? The first paper explaining the connection without using AdS/CFT seems to be the one by Jensen, Loganayagam, and Yarom (arXiv:1207.5824).

    What follows is my subjective opinion.

    The AdS/CFT correspondence played a crucial role in this story. Without AdS5 as a laboratory, it would be highly nontrivial to see any connection between the two effects. For one, finite temperature or temperature gradient does not lead to violation of chiral charge (Spivak’s argument). Then there is a mismatch between the orders of the derivatives mentioned above.

    Figuratively, AdS/CFT has provided a cheap laboratory for theorists to play and to “measure” different transport coefficients of an exotic medium (like the one where particles are fermions with definite chiralities), years before it became possible to do it in a real lab.

    Landsteiner’s statement quoted at the end of the NYT article, in my view, is an accurate description of the role that string theory played in his and many other theorists’ work. As often in condensed matter physics, whether the suggested theory is directly relevant to the experiment is a question that will take some time to completely settle.

  42. Peter Woit says:

    Dam Thanh Son,

    Many thanks for your comment, which completely clarifies the role of AdS/CFT here.

    I still however think that using the term “string theory” to describe these calculations is misleading (given that strings play no role). Why not just use a term like AdS/CFT or gauge/gravity duality? One argument for using “string theory” would be its historical role in AdS/CFT. If so though, in the usage of “string theory” in this article, we’re now at two different historical removes from an actual theory of strings, and the average reader is unlikely to understand at all the relation between the “string theory” being discussed here and the “string theory” they’ve heard about elsewhere.

  43. Mitchell Porter says:

    Let me distinguish between the various complaints that have been made so far.

    Peter Woit objects that a successful application of AdS/CFT to condensed matter could be mistaken for a successful application of string theory to particle physics.

    William B. Emba and I object that the actual effect at work here is not literally the mixed axial-gravitational anomaly, it’s some kind of thermal analogue of that.

    And Boris Spivak (quoted in Nature) doesn’t even think it’s the thermal anomaly producing the reported results, it’s some other process entirely.

    Obviously Spivak’s criticism is the only one of direct scientific consequence, the others pertain to how concepts are named and communicated.

    But I would like to expand a little on *my* complaint. Ironically, I find the science journalism clearer on the point that this is a physical analogy, than the actual arxiv paper! (I’m referring to 1703.10682.)

    I see nothing in that paper which would lead a person to think that the effects in question are not being attributed to literal gravitational anomalies. Several times, the authors write that the reader may wonder how a gravitational anomaly could be relevant in the absence of gravitation; but each time, we are told that the connection has nonetheless been established.

    There must be a way to understand this which is not “consider this field theory in curved space, then consider the zero-curvature limit, and find that the gravitational anomaly coefficient is still nonzero”; but which is more like, “consider this field theory in flat space, and note this algebraic isomorphism with the zero-curvature limit of the curved-space analogue”.

    But it seems that if I want to understand what’s really going on at the level of theory, I’ll have to just keep butting my head against the original papers until enlightenment dawns.

  44. DrDave says:

    I don’t think it matters what the NYT *thinks,* they should simply publish a response from Peter. That’s good journalism. Better journalism would be to get the facts right, of course. If a paper doesn’t print a well thought out, factually accurate letter, it isn’t a newspaper anymore–something the Times has been struggling with.

  45. Peter Woit says:

    DrDave,
    You’re assuming I want to write a letter to the NYT for them to publish, which is not the case (I don’t think the issues involved lend themselves to such a letter, and I’ve already wasted enough time on this). I have been in communication with the author of the article, and, given that my complaints are not about factual correctness but about misleading use of language, he’s not interested in significantly changing the article.

    Unlike most cases where people disagree with something in the NYT, at least in this one my point of view has gotten aired in great detail here (and opposing views also, thanks to commenters).

  46. Peter Woit says:

    DrDave,
    Perhaps due to our email exchange, the author of the NYT piece did make some changes in the language to make things less misleading.

  47. Chris W. says:

    “… but for most laymen the mention of string theory was just fine.”

    As much as Chang, that’s an editor talking.

    So much of the media is worse than this where science and many other subjects are concerned. I suppose one can give the New York Times some credit for resisting the all-out competition in the race to the bottom.

  48. Peter Woit says:

    Chris W.,
    I don’t think the problem is really with the journalists, it’s with the physics community. In this case the publicity for the (quite complex) scientific result seems to have largely been provided by string theorist Landsteiner, including presumably accurate quotes claiming this as a “string theory” calculation and success. And, from the comments, his colleagues in his field say they’re fine with this, even if it’s misleading. Other physicists need to step up and say that this is misleading and not fine. The author heard from me making this case, but is then put in a position of not knowing who to believe (a non-specialist has no hope understanding the role of string theory in AdS/CFT and the role of AdS/CFT in the calculation of this effect).

  49. Art says:

    This post and the comments have been educational and productive: a fair amount of history and technical detail was discussed, culminating in a revision to the original article. I hope it’s at least somewhat satisfying. The only thing missing, I think, was an explanation of the change to the article, which I’m used to seeing after an edit.

  50. G Euphrates says:

    mea culpa.
    As an ex-IRE member (Institute of Radio Engineers) I witnessed the change to IEEE (Institute of Electrical and Electronics Engineers); thus, your use of IIEE confused me.

    But excellent clarification by argumentation in the comments!

Comments are closed.