Latest news this evening from Scott Aaronson at the IAS in Princeton:
Tonight, David Nirenberg, Director of the IAS and a medieval historian, gave an after-dinner speech to our workshop, centered around how auspicious it was that the workshop was being held a mere week after the momentous announcement that a wormhole had been created on a microchip (!!)—in a feat that experts were calling the first-ever laboratory investigation of quantum gravity, and a new frontier for experimental physics itself. Nirenberg speculated that, a century from today, people might look back on the wormhole achievement as today we look back on Eddington’s 1919 eclipse observations providing the evidence for general relativity.
I confess: this was the first time I felt visceral anger, rather than mere bemusement, over this wormhole affair. Before, I had implicitly assumed: no one was actually hoodwinked by this. No one really, literally believed that this little 9-qubit simulation opened up a wormhole, or helped prove the holographic nature of the real universe, or anything like that. I was wrong.
Scott has been the one person in this field I’m aware of who has tried to do something about the out-of-control hype problem that has been going from bad to worse. I do disagree with him about one thing. He goes on to write:
I don’t blame the It from Qubit community—most of which, I can report, was grinding its teeth and turning red in the face right alongside me. I don’t even blame most of the authors of the wormhole paper, such as Daniel Jafferis, who gave a perfectly sober, reasonable, technical talk at the workshop…
I do blame all those people. Unlike Scott, they’ve been either participating in hype for years, or staying quiet and enjoying the benefits of it. Grinding their teeth and turning red in the face is not enough. They need to finally say something and take action.
Update: Still unclear to me what the ultimate fallout of the publicity stunt will be. One thing that is becoming clear is that the publicity stunt is part of a vigorous and very effective campaign to mislead funding agencies and those making funding decisions. The goal is to convince them that “quantum gravity in a lab” is a real thing and the way forward for fundamental theoretical physics. The bogus Quanta story, video, headlines are not a bug, but a feature. Among the funding agencies, DOE is on board, their grant funding the publicity stunt, and they are advertising it prominently in their presentation to HEPAP today (see page 7 here). At the IAS, it seems director Nirenberg’s claims that the “first quantum gravity experiment on a chip” was possibly the biggest breakthrough in a century were not off-the-cuff comments based on what he had read in the paper or at Quanta, but prepared remarks based on conversations with IAS senior faculty. I gather that what he has been hearing from at least some of them is that the wormhole “experiment” vindicates their past research and justifies supporting them in the direction they are choosing for the future. He has though now been hearing other viewpoints.
If researchers in this field want to know what they can do about the problem, contacting places like Quanta to get them to fix their coverage is one thing, another is contacting people with funding responsibilities (ie. program officers at funding agencies, directors of institutes) who seem to have been misled by the hype campaign.
Update: Quanta hasn’t done anything more to fix the wormhole article, but they have now updated their original “Physicists have built a wormhole and successfully sent information from one end to the other.” tweet. New one reads “Experimental physicists built the mathematical analog of a wormhole inside a quantum computer by simulating a system of entangled particles.”
Update: Today at HEPAP the Fermilab director was prominently advertising the wormhole publicity stunt as a Fermilab initiative (see slide 67 here). She describes future plans for more of the same, with these calculations being performed not on one quantum computer, but on two spatially separated quantum computers connected by a quantum network. This would somehow allow for a big increase in the “quantum gravity in the lab” hype with a new Nature cover story: “FNAL scientists connect two quantum computers by a wormhole between two labs”. A question for those in the “It from Qubit” field. Are you willing to contact those responsible for funding this, who are now prominently advertising this work as a major success and new direction of research they intend to fund (e.g. Glen Crawford at DOE, and Lia Merminga at FNAL)?
Update: 4gravitons has a blogpost about the Quanta article, concentrating on the issue of “tone” of the coverage. That’s relevant for the usual problematic sort of physics coverage, but in this case something much more unusual is going on. This was a well-organized publicity stunt designed to justify funding “quantum gravity in a lab” research. Quanta was taken in more so than many other journalism venues. But the really disturbing part of this story is who else was taken in: the IAS director, the Fermilab director, the DOE division director and others, who are sophisticated consumers of science journalism, and independently getting their information from senior scientists in the field.
The worst are the people who’ve been brainwashed by the “experts” but don’t themselves understand enough of the topic to actually have an own opinion. Like Lockyer going on about naturalness and why that means bigger colliders will find something (see here). I actually wrote to him asking what he was going on about in this interview just to make sure. In explanation he sent me a link to a blogpost by Matt Strassler, not kidding.
Scott is asking what went wrong and what can be done about it. I know I sound like a broken record, but here we go again. Large communities of likeminded people are affected by social reinforcement, aka, group think. This, I cannot stress often enough, is a well documented phenomenon and, no, intelligence does not protect from it. The more you hear other people talk about X, the more you think X is important, and the more you yourself will talk about it. In the end, everyone thinks X is great and important because everyone else thinks that, not because it’s actually important.
(I actually suspect the problem is worse among intelligent people because they believe they’re too intelligent to be affected, but that’s just a conjecture on my part.)
It’s easy enough to address the problem: Give every scientist a basic education on the sociology and philosophy of science, and social and cognitive biases. I believe that if they think about it, they should realize they don’t actually want to waste their time on research that will turn out to be entirely irrelevant. Though there is a case to be made that it makes money and popularity in the short run, I don’t think it’s what most of them are driven by.
I have a longer list with recommendations for what can be done by scientists, higher-ed admin, policy makers, and members of the public in the appendix of my 2018 book which is available here.
What went wrong: it is probably not a coincidence that this is happening in fundamental physics, where actual progress has been rather slow for several decades (according to some of the blogs I read).
What can be done about it: change the system so that people do not have to overhype their results in order to find a job / keep it / get funding. A tall order, perhaps.
Wow. Just wow. I really, really want to ask the journalists involved if they can hear this, and then think they communicated the result properly. This is not Jo Bloggs who gets science news fourth-hand on Facebook. This is someone who should be able to go and ask the experts down the hall to explain at length over coffee.
Next time I read a Quanta article, whatever the topic, the first thing I’ll put in my comment is “And did they also create a wormhole in the lab?”
I’ll stop when I see a meme on the net along the lines “Chuck Norris creates a wormhole…”. 🙂
After dinner speech… maybe D. Nirenberg were just a little drunk 😉
or is there something rotten in the kingdom of Princeton ?
Yesterday in my intermediate mechanics class a student pulled out a copy of the latest Nature issue and asked about the wormhole article. He had some very good questions, which indicated that he was pretty suspicious of what he was reading. I was able to confirm his suspicions and explain that most of the claims in the article were nonsense, and that Nature had been hoodwinked.
Toward the end of my time as an editor at Physical Review A (15 years was quite enough, thank you), we started getting papers where all the authors had done was run something on a quantum computer. Sometimes this was even phrased as, “We did an experiment.” Real experiments are hard work, and running something on a cloud quantum computer does not count. Claiming you created a wormhole is worse. Most of these papers were rejected out of hand. If the paper contained something new that was run on a quantum computer as an illustration, that was different. On that basis, and, for the moment, ignoring the hype, the “wormhole” paper would probably be publishable in Physical Review D (this is a guess, I know nothing about quantum gravity, and I don’t want to speak for the PRD editors). The hype, however, is appalling, and it is clearly causing a lot of damage, which will only get worse when people realize how vacuous the claims are.
Oh yeah, a lot of people were taken in by this publicity stunt.
In the comment section under the Ars article I pointed out the reaction of specialists, including Peter Woit and Scott Aaronson, to the stunt — with quotations and links. (I definitely did not try to pose as an expert myself)
I’ve got angry responses, attacking me and the experts, some saying that these reactions were “dumb” etc.
Out of curiosity I’ve checked the comment history of the people responding in this way. They seem to be intelligent techies, who made thoughtful and informed contributions to discussions of computer and technology topics, not some trolls.
I don’t think that these people are going to be convinced by blogs like this one; at this point only serious pushback by media and (ideally) the scientists involved in that research can help…
I agree, this has gotten out of control. Perhaps it might be appropriate to contact the American Physical Society. Is there any way someone with some clout in the field could draft a letter and then invite cosignatures?
Mark Hillery,
Some of the papers that you did not reject out of hand came to me to referee. I then rejected them. The most ludicrous was one claiming they had done an experiment with indefinite causal orders by implementing a causally ordered quantum circuit in a causally ordered quantum computer.
Sabine,
I think you’re right about the group-think problem, but in this particular case there’s something else even more disturbing going on. Most of the people in this field are well-aware the wormhole thing was a publicity stunt, but saying so publicly, or doing anything about it more privately, would carry personal costs they are not willing to bear. The main author of the stunt, Spiropulu, has spent four years working on it, funded by a DOE grant. Getting a “physicists create wormholes” headline in this case (unless it blows up in her face…) carries major professional and financial rewards, which may have a lot more to do with this than group-think.
Scott describes the quantum gravity research community as “grinding its teeth and turning red in the face”, but not as doing anything. I gather no one spoke up at the time to publicly tell Nirenberg he had just embarrassed himself and the IAS by being the victim of a publicity stunt. Did anyone explain this to him in private later? If not, could it be not groupthink, but not wanting to take the professional cost of being the one to explain to someone in a powerful position that you and your community have a serious hype problem, highly damaging to his institution, which it is unwilling to do anything about?
Allow me to seriously doubt about the sincerity of these “teeth grinding” and “red faces” that Scott reports. I mean, this has been happening in the field for years, many of them probably contributed with both nonsensical research and hype themselves, but now suddenly they feel ashamed? I think they are actually feeling shame because of how grotesque and caricaturesque it all resulted at the end. I bet that, in their minds, they were hoping for something much more serious, credible and professionally looking. But, alas, it’s just the mirror: sometimes it gives you back an image which ain’t pretty… if you have nothing to offer to it, the mirror cannot do magic and put substance where there’s none.
The entire reason this happened is because the general public – which is illiterate on the subject matter at hand – passionately wants to believe that their science fiction dreams are becoming real. The editors/writers at Quanta, the NYTimes, Reuters, and ABC perfectly understand this is the hook for their audience: report that their science fiction dreams are slowly becoming real and make money by feeding this audience.
The purveyors of this hype – whether it was the experimentalists or the theorists or both – who actively drove Quanta and these other media outlets to write about it – knew perfectly well the incentives at play. Had they soberly explained the research and not hyped it, then it would not have gotten the worldwide media fame that it has gotten them. Whether there is self-deception here on the part of the scientists I can’t say. But the whole field does take a hit when that eager audience for their science dreams come true realizes they were bamboozled.
Adam Treat,
I think you’re partially blaming the victim here. If physicists actually created wormholes in a lab in Santa Barbara, most people would sensibly be interested in hearing more about this. The problem isn’t an illiterate and unsophisticated general public, since the people believing this include the IAS director, a Pulitzer prize-winning science journalist and editors at Nature. Best to put the blame where it belongs: those directly responsible for the stunt, and the larger community of researchers which enabled them.
Scott is someone who enjoys science communication, is good at it, and is willing to devote significant time to it. As a result, he has a blog where he publicly posts his opinions on things like this. Most physicists do not have a blog, do not regularly engage in science communication and so did not publicly comment on the article or hype. If everyone working in the field other than Scott should be blamed for this whole mess, is your claim that any practising physicist has a moral obligation to make a blog? That takes a lot of time, effort and skill as I assume you know.
As someone in the field, I have been on multiple emails threads in the last week about how embarrassing a failure of science communication this was, and how we need to get better at communicating to journalists to avoid such events in the future. I was also at the dinner with Scott last night and there was loud heckling throughout the talk (think: “I don’t know whether this will be as impactful as the eclipse observations in 1919” “I really think we do”) to an extent that I have never seen or even imagined was possible in such a setting.
I certainly don’t blame David at all, and clearly as a field we have some collective responsibility to not let stuff like this happen. But all the actual It from Qubit researchers quoted as independent experts (Scott, Dan Harlow etc.) made skeptical comments, and even the authors of the paper who actually work in the field were not the ones saying completely ridiculous stuff. Daniel Jafferis gave a nice and entirely reasonable talk on the paper this Monday. In particular, I’m not really sure what the vast majority of people in the field, who all first learned about the paper when they saw the quanta/NYT articles, were supposed to do to prevent this whole thing.
“If everyone working in the field other than Scott should be blamed for this whole mess, is your claim that any practising physicist has a moral obligation to make a blog?”
I believe there exists a serious (very serious) fallacy in this “argument.”
Geoff Penington,
I’ve also been having multiple email threads the past few days about this, with people’s position either “nothing wrong here” (IAS and Quanta point of view) or “yes, this was bad, I don’t see anything though I can do about it”. Scott’s taking a “this is bad and I’m trying to do something about it” position is an exception.
My comments blaming all involved in this except Scott should be taken in that context, and in the context of unsuccessfully complaining about disturbing levels of hype in this field (string theory – > AdS/CFT -> “It from Qubit”) for decades, watching it getting worse and worse in recent years. This past week it has gone off the charts on the crazy meter.
I agree that not everyone should be out there making public statements about this. But if people don’t want to get blamed for craziness in their field, they should do something. Here’s one example: personally I believe Quanta should pull the article and video from their site and re-report this story properly. Thomas Lin (Quanta chief editor) made the argument to me that he would not change anything about their coverage because he was not hearing from quantum gravity experts that there was a problem with it. Why not? Whenever a publication covers your field of expertise and gets it very wrong, why not write to the editor and argue that they should fix it? My impression is that everyone (except Scott…) is saying “someone else should do that, not me, not my responsibility, I don’t want the cost in time/energy/making enemies that I’d incur”.
Another comment: I think your problem is not so much being bad at communicating with journalists, but rather not being willing to take on those in your midst who are very good at communicating with journalists, but do so to self-promote and push bad science.
Peter, no. I do not believe Jafferis and members of his group are to blame, and I do not think you do either. While they had some control over the contents of the Nature article they cannot prevent their co-authors from beclowning themselves on the popular media forums.
I brought this up with some eminent Rutgers theorists and their consensus is that the questions Jafferis et al are asking about a small group of interacting fermions that mimic “…(some sort of) (TWO DIMENSIONAL) quantum gravity with a wormhole…” is not crackpot stuff. (All caps theirs!)
Unsaid but implied: calling the implementation of this on a computing device (of any sort) a “crummy wormhole” is crackpot.
I agree with Scott Aaronson: let’s not tar everyone on the author list with the same brush.
For those who say that the authors are not to blame: Has anyone watched the video clip released by Quanta (https://www.youtube.com/watch?v=uOJCS1W1uzg) in which the authors kept crying “wormhole”, “wormhole”, “This is a wormhole. I think I just saw one”, “I think we have a wormhole, guys”, “I never really believed that we’d get all the way to something that was real”…? Should we blame the popular media people for what the authors said in the video clip? Am I missing something?
Amitabh, you should ask your theory friends if they can derive Newtonian gravity by studying these small group of interacting fermions.
The fact that theorists feel duty bound to defend this garbage even in private discussions is disturbing.
That is the actual sad part of this story. Maria is a publicity seeker and thus easy to condemn. But, she was encouraged by all of these theory people to do this experiment since they claimed they were going to learn something deep about quantum gravity from it. Sadly, she is not the only experimentalist who has been misguided. There are quite a few groups around the world doing this sort of work (using cold atoms etc.) and there is nothing to be learned from these experiments – the experiments will work as predicted by well understood quantum mechanics and it is then “reinterpreted” in fancy holographic language.
First, let me say that I hardly understand this stuff at all, so I may easily be missing things.
But I am puzzled by something here. Looking at the Nature paper, Eq. 1 is essentially the original SYK model, where you construct a Hamiltonian on Majorana fermions that contains all four-element terms, and you give each one a random Gaussian coefficient. When I first heard about the SYK model, if my memory is correct, this was described as either a zero-dimensional or an ∞-dimensional theory.
With the AdS-CFT correspondence, the dimensionality of the dual should differ from this number (0 or ∞) by 1. But the article clearly says that the dual is a 2-dimensional theory (as do some of the commenters here). What am I missing? Could somebody who knows something about the SYK model enlighten me?
And another question: It takes quite a bit of work to make things behave like Majorana fermions on a quantum computer. How did the authors manage to do this? Or did they just use ordinary distinguishable qubits instead? Would doing this change the dynamics of the system so that it’s not actually equivalent to teleporting through a wormhole? (The teleportation protocol itself should work fine whether you’re using Majorana fermions or plain distinguishable qubits.)
Amitabh Lath,
Sorry, but I do think Jafferis in particular has a lot to answer for here. Yes, his part of the paper may be the part that contains some non-trivial science, but he has played a significant role in the publicity stunt aspect of this. He appears in the ludicrous Quanta video, the promotional podcast, and is quoted in the Quanta article as claiming that the stunt shows “a filament of real space-time”. If he has any problem with the way he has been used in the publicity stunt, I haven’t seen any evidence of it or heard privately anyone mention this.
Perhaps more significantly, the stunt was organized by Spiropulu, who recruited him as co-PI on a DOE grant that has funded them and the stunt. The stunt is now being advertised by DOE as a great accomplishment (this is happening today, at a HEPAP meeting, audience other funding agencies and top people in HEP, see page 7 of this presentation: https://science.osti.gov/-/media/hep/hepap/pdf/202212/Crawford_DOE_HEP_Research_Program_Status_HEPAP_202212.pdf)
So, from what I can tell, Jafferis has been deeply involved in the misleading publicity stunt, has been getting paid to do so, and has shown no signs of having any problem with it. I think it’s perfectly fair to blame him.
Peter Shor, the SYK model has time in it (space is a point) so it is a 0+1 dimensional theory being the boundary theory of a 1+1 dimensional gravitational theory in principle….
Geoff Penington, I appreciate you commenting on this blog which is much more than what most of your colleagues do. Consider a “hypness index” = [Influence x Exposure]/ [Achievements] , then the hypness index of the “it from quibit” community is 2 orders of magnitud above any other field in physics, specially after this stunt.
Also think of the following: If any government agency gives a lot more money after this stunt to your community, will anybody within it come out and ask for moderation? Your colleagues may grind their teeth as Scott says but they will gladly take the extra money that comes from this and forget why they got it. You may say that every community has this hypocrisy problem but by my made up hypness index the most insulting and in urgent need to correct case is the “it from qubit” one.
I figured out the answer to my first question: the wormhole is 2D because it has one dimension of time and one dimension of space, since the SYK model has zero dimensions of space.
I still don’t know the consequences of replacing Majorana fermions with ordinary distinguishable qubits in the SYK model.
Well… it should be well-known that there is no m=0,s=2 particle in the representation theory of Lorentz group SO(1,d) for d=1,2, hence no propagating DoF’s in d=2,3, thus it’s questionable if it makes sense to speak of gravity theory in low-dimensions; unless one is considering near-extreme solutions of the d=3 theory. But, then, all work on JT gravity relies on heavy Wick rotation, and reading Witten+Stanford paper, https://arxiv.org/abs/1907.03363, you see computations of Z in the JT theory with genus g!=0 in Euclidean signature, so, there’s not well-defined Lorentzian structure on their computations. Hence, no graviton DoF’s propagating in their calculations, and it’s not even clear if one can recover the Lorentzian manifold structure in the end.
But if this seems already bearing on the absurd, the JT/SKY pseudo correspondence is based on a NEARLY AdS2/CFT1 “duality.” So, these It from Qubit people are studying an approximation of a toy model of an approximation of a theory which is not even clear whether it corresponds to any gravitational theory at all.
Attendee: I believe you may be misconstruing what my Rutgers theory colleagues said about this work. They remain skeptical about the main claim, but do not claim any expertise on SYK connections to QI, so from what I gather they are driving past this car wreck after giving it a short glance, without slowing down to consider it in too much detail (this is a skill you develop driving in New Jersey).
Peter: I do not know Jafferis and cannot speak to his inner thoughts and opinions. I did not watch the Quanta video but surely we can make a distinction between appearing in a video and saying something (maybe) impolitic and imprecise about “filaments of spacetime”, and what Lykken is doing going on about having created actual wormholes. A lot of my colleagues think very highly of Jafferis (he was a PD at the Rutgers NHETC) and as a community we would be better off not excommunicating him and members of his group.
Amitabh Lath,
I’m not suggesting “excommunicating” Jafferis. He’s a tenured full professor at Harvard and will do fine. He has though clearly signed up as co-organizer of the publicity stunt. If it succeeds he’ll be rewarded, if it blows up in his face he may suffer some reputational damage, and I’ll have trouble feeling too sorry for him about this.
As for the junior people in this collaboration, they’re not responsible for the hype/publicity stunt aspects. Some of them will though have tough decisions to make about how to pursue a career and try and find a permanent position. If the “quantum gravity in the lab” hype works and that’s where positions are, their best bet would be to go along with that. If it blows up and gets discredited (I’m doing what I can…) they should be looking for different sorts of projects to work on and stay away from this kind of over-hyped activity.
Since there seem to be relevant experts here, what is the (local) topology of the 1+1d spacetime with the SYK wormhole? It’s a Lorentzian surface, with boundary I guess, so it should be possible to get a concrete handle on it, no? Pun not intended…
Just a lurker, x-experimentalist; I would have heard nothing of this if I didn’t follow SA’s blog. So the general public(GP) knows nothing (IMHO), however the general trust I see in the GP is at an all time low, and this just increases the mistrust. Citation rankings reward fame, so to succeed in science you need fame, which rewards hype, and you all forgive that… because they are just trying to succeed in the fame game. I often think that so many problems are related to having the wrong feedback variable.
I have to say that the spectacle of tenured academics calling the general public illiterate and unsophisticated in this instance I find a little ironic.