It seems that a couple of the authors of the recent Cosmic Controversy letter (discussed here) are going on a campaign to embarrass the 29 physicists who were convinced to sign their letter. Andrei Linde has gone to Lubos Motl’s blog to thank him for his blog entry which lauded Linde as having eaten from the biblical tree of knowledge and which denounced his critics as imbeciles. To deal with Linde and his claims, Ijjas, Steinhardt and Loeb have added a new webpage to their website called Fact Checking. It lists the four “predictions” of inflation claimed to agree with experiment by Linde et al. and gives four references to papers published by Linde touting different “predictions” for the same quantities, predictions not agreeing with experiment.
This month’s Scientific American has a remarkable cover story, The Quantum Multiverse from one of the other four letter authors, Yasunori Nomura. I’ve seen some fairly bizarre stories about fundamental physics in Scientific American over the years, but this one sets a new standard for outrageous nonsense, and I’m wondering whether it too may cause some of the 29 co-signers of the letter co-authored by Nomura to question the wisdom of joining with him and Linde. Nomura is well known for a definite prediction based on the multiverse: in 2009 he co-authored a paper claiming that the multiverse predicted the Higgs mass would be 141 GeV +/- 2 GeV. This played a major role in the film Particle Fever. That three years later the Higgs was discovered at 125 GeV seems to have had no effect on his multiverse enthusiasm.
The new SciAm cover story is not about anything new, but is based on a 6 year old paper by Nomura discussed here. At the time I wrote about this “I’m having trouble making sense of any of these papers” and quoted Lubos’s evaluation: “They’re on crack”. Nothing I’ve seen about this over the past six years seems to me to make any sense at all, including the new SciAm cover story, which just seems even more content-free and meaningless than previous efforts to explain this “multiverse interpretation of quantum mechanics”. On the obvious question: how would you test this, Nomura just has this to say:
Evidence so far indicates that the cosmos is flat, but experiments studying how distant light bends as it travels through the cosmos are likely to improve measures of the curvature of our universe by about two orders of magnitude in the next few decades. If these experiments find any amount of negative curvature, they will support the multiverse concept because, although such curvature is technically possible in a single universe, it is implausible there. Specifically, a discovery supports the quantum multiverse picture described here because it can naturally lead to curvature large enough to be detected, whereas the traditional inflationary picture of the multiverse tends to produce negative curvature many orders of magnitude smaller than we can hope to measure.
This paragraph manages to put together three different misleading and unsupported claims:
- “If these experiments find any amount of negative curvature, they will support the multiverse concept because, although such curvature is technically possible in a single universe, it is implausible there.” This is just nonsense.
- “the traditional inflationary picture of the multiverse tends to produce negative curvature many orders of magnitude smaller than we can hope to measure”. What is the inflationary multiverse “prediction” for negative curvature? As far as I can tell it’s compatible with pretty much any level we might observe.
- “the quantum multiverse picture described here because it can naturally lead to curvature large enough to be detected.” I can’t find anywhere a calculation of the negative curvature expected by the “quantum multiverse picture”, and I don’t believe any such calculation is possible.
Given some of the outrageous hype I’ve seen in recent years in respectable publications, it’s gotten rather hard to shock me with this sort of thing, but I do find this Scientific American cover story shocking.
Update: For some reason this was not mentioned in the SciAm article, but the paper justifying Nomura’s claims about negative curvature is here.
Update: A modest proposal: Given the situation, I think someone needs to write a letter to SciAm complaining about the Nomura article and get leaders of the community to sign in support of it. They could start gathering signatures by writing to the 29 signers of the earlier letter. If these people were willing to object to the Steinhardt et al. article, they should be willing to object to the far worse Nomura article.
Pingback: A Cosmic Controversy | Not Even Wrong
“Can Quantum Mechanics Save the Cosmic Multiverse?”
Another example of Betteridge’s Law of headlines?
SciAm doing this kind of thing doesn’t surprise me at all. SciAm today is a far cry from what it was in the 70s when I was doing my Physics degree. These days if I want to read a sensible magazine about science I pick up a copy of American Scientist from our local bookstore.
Yes, SciAm has changed since the 1970s, but so has the physics community, with pseudo-scientific claims like this about the multiverse unthinkable among serious physicists back then. This does seem to me to be a new low for SciAm, not clear whether this has to do with their own efforts, or whether they’re just tracking theoretical physicists downwards.
The IS&L Fact Checking page has a link to an important lecture from Richard Feynman [https://m.youtube.com/watch?v=0KmimDq4cSU at 5:00 minute mark].
When I listen to Feynman’s discussion of “vague theories” and then read about the multiverse, it makes me wonder what Feynman would say about the inflationary theory if he were alive today.
It’s pretty clear now that inflation was always flexible about every prediction. The IS&L page points out that Linde even used inflation to predict the universe is curved – of course, that was before astronomers showed that it is not.
Whatever the argument over “inflation” I think it’s clear what Feynman would have to say about the multiverse. It’s a ridiculous situation that objections to multiverse pseudoscience are not coming from leading figures in the theoretical physics community. It’s their job, and they’re not doing it. Instead they’re supporting this pseudoscience by signing on to a letter supporting Linde and Nomura and arguing against a perfectly sensible Scientific American article by Steinhardt and co-authors.
Given the situation, I think someone needs to write a letter to SciAm complaining about the Nomura article and get leaders of the community to sign in support of it. They could start gathering signatures by writing to the 29 signers of the earlier letter.
You may find it “nonsense”, but the argument that curvature is implausible in a single universe based on the canonical measure over the space of cosmological histories is pretty conventional at this point, see e.g. https://arxiv.org/abs/1406.3057. I don’t necessarily want to defend the rest of what Nomura is saying, though…
Peter Woit : ” It’s a ridiculous situation that objections to multiverse pseudoscience are not coming from leading figures in the theoretical physics community. It’s their job, and they’re not doing it. ”
Well spoken. It shows a deep lack of responsability, not only towards the public, the community who funds them, but also towards young physics students, for whom they should be an example of what a serious scientific method is all about.
Whereas now this future generation of physicists become accustomed to the notion that it is ok to make claims with an air of despotism for decades while not providing real support from observations for those claims. And that endless ‘curve fitting’ is portrayed as a professional modus operandi in this context.
One could indeed speculate about Feynman’s take on this, but what about Einstein himself ? I think he would be seriously depressed by now if he had to watch this spectacle. I believe his words were ‘It’s adding Ptolemaic epicycles all over again’.
Respect for Steinhardt&co.
cosmology grad student,
That really is complete nonsense. Carroll’s long discussion of the “natural measure on trajectories” is just the usual Hamiltonian formalism of classical mechanics. Applying it to this question is nonsense since
1. You don’t have any reason to believe classical mechanics applies.
2. You don’t know what your phase space is.
3. You don’t know what your Hamiltonian is.
That anyone considers this a “pretty conventional” argument is deeply depressing.
Peter Woit: “A modest proposal: Given the situation, I think someone needs to write a letter to SciAm complaining about the Nomura article and get leaders of the community to sign in support of it.”
As you have said earlier, Peter, some sort of weird lettter-writing petition is no way to settle a scientific argument.
Yes, but I don’t think there’s any scientific argument in the Nomura article, it’s a different sort of problem facing the physics community than a bad argument.
Scientific American sure can pick ’em… First they give a platform to Steinhardt et al’s campaign to smear inflation as unscientific; and then they make a cover story out of the worst “idea” in multiverse theory (I don’t just mean Nomura’s paper, it’s also dumb when Bousso and Susskind, and Tegmark and Aguirre, do it).
P.S. Just to be clear, what I mean by “the worst idea”, is the idea that the technical problems of eternal inflation (measure problem, holography) are solved by introducing the many-worlds interpretation of quantum mechanics into the mix. The proof that it’s a bad idea may be seen in the papers that try to make it work.
So far all the comments seem focused on the nonsense bullshit Sci. Ame. article. But, who are backing up the writers and their followers? From where have they got the money? Should funding agencies/academic departments/media be also held culpable, in addition to those big names? Would they prosper without any social/public support?
Great Article in the Guardian today:
“One of the study’s authors, Professor Tom Shanks of Durham University, told the RAS, “We can’t entirely rule out that the Spot is caused by an unlikely fluctuation explained by the standard [theory of the Big Bang]. But if that isn’t the answer, then there are more exotic explanations. Perhaps the most exciting of these is that the Cold Spot was caused by a collision between our universe and another bubble universe. If further, more detailed, analysis … proves this to be the case then the Cold Spot might be taken as the first evidence for the multiverse.”
Yes, more detailed analysis is all that is needed.
Since eternal inflation was introduced in 1983 and Feynman died in 1988, he might have actually said something about it.
The fact checking is the most pedagogical presentation of the subject of cosmology that I ever saw.
Reading through that Carroll paper and your objections I am reminded of your often repeated wish that physicists would learn more from mathematicians about process and rigor. The assumptions in the paper are not stated clearly – e.g. what we don’t know – and thus the impact on conclusions is similarly unclear.
As for Feynman, if he couldn’t explain it to undergrads he didn’t consider it properly understood it. That is where I think he would take issue with these current theories: he would object that even more that they have no footing in experiment, they do not help us truly *understand* anything about our physical world.
In this context, I think ISL and the rebuttal are really arguing over whether inflation and the multiverse theories have actually helped us understand something real or whether we are just as muddled about the things they supposedly help to explain as before.
Endless conjectures upon assumptions upon conjectures isn’t real understanding and the limitless flexibility of these theories just means conjectures and assumptions all the way down.
That quote from Tom Shanks is laughable BS. It is complete and utter nonsense, unsupported by any actual rigorous analysis in his or any other published paper, to say that the Cold Spot supports the multiverse.
In fact it is worse than that, since even the claim that the Cold Spot is ‘an unlikely fluctuation’ in need of any ‘explanation’ at all is highly dubious. Put in language that HEP people will find familiar, it is only at most a ~2.5 sigma anomaly with respect to the standard cosmological picture even without accounting for a massive ‘look elsewhere’ effect in the analysis; any sensible consideration of this effect and the significance completely drops away.
A couple of relevant references: https://arxiv.org/abs/0908.3988 and https://arxiv.org/abs/1408.4720 (Figure 6 in particular), among others.
I’m wondering whether there are any examples of media stories about the multiverse that are not nonsense, or whether the only accurate possible characterization of them is by degree of nonsense (with “laughable” about the median).
I’m curious, what does it say about the physics community that Nomura, who is a full professor at one of the top physics departments in the world, senior researcher at one of the most prestigious labs in the world, and principal investigator at a top institute, is writing papers which make no sense? No one seems surprised. Math profs at Berkeley don’t write papers which make no sense. Math profs at my university don’t write papers which make no sense, and I’m not at a research one school.
Well, who is supposed to be the consumer of all this sound and fury, signifying nothing?
To the lay public, this would look disturbingly like a bunch of quackery, and not even a particularly good one. Information about astral projection, crystal healing, communication with the dead, ghost manifestations and cold spots in the attic can be had in dedicated magazines and tradeshows without having a tedious boffin spoil the show.
I noticed that the Russian government news service is running the following story about this “proof of a multiverse” and how it ties in with string theory
The source of these “utter nonsense” stories that do great damage to the credibility of science has always mystified me.
Maybe it’s all a Russian disinformation plot, designed to destabilize the Western (scientific) establishment, by spreading fake physics, and getting leading scientists to start fighting amongst themselves, writing “open letters” and the like. I can’t think of a more plausible explanation for the SciAm Nomura article…
“Math profs at my university don’t write papers which make no sense”
I’m not so sure. Probably not at the same level as the multiverse/multimess, but still, there is this thing with Mochizuki’s proof of the abc conjecture that nobody seems to understand, and the computer-assisted proofs of various theorems that half of the math community doesn’t consider as “proper” proofs, etc. So math is also not completely insulated from social controversies.
“Maybe it’s all a Russian disinformation plot, designed to destabilize the Western (scientific) establishment”
If only that were true, we would have much less of a problem. But alas, I think this “disinformation plot” actually comes straight from the very Western scientific establishment itself (or one part of it), probably trying to gain celebrity status in popular media, and maybe fool others into thinking that they deserve a Nobel or something… I fear that it’s a much deeper issue than mere political propaganda.
I fear you are right about the origin of the “disinformation plot”…
But I don’t think there is any analog in the conventional math literature of things like Nomura and his “quantum multiverse”. It’s basically a sweeping claim about what fundamental physics is, backed by no sensible non-trivial idea. To get an analog in math, it would be something like papers claiming to solve the Riemann hypothesis by some incoherent invocation of Godel. Such things do exist, but they are ignored, don’t end up in respectable journals or Scientific American. And Mochizuki/abc really is something very different.
Well, ok, I agree that the abc conjecture thing is probably not a good example. But the physics discussions “what constitutes proper science” and math discussions “what constitutes proper proof” (in the context of computer-assisted debates) seem to have quite a lot in common. Of course, there are also big differences as well, but still…
Off topic, but can you comment on Bill Unruh’s latest paper (https://journals.aps.org/prd/abstract/10.1103/PhysRevD.95.103504) on Dark Energy – it’s been like a week and literally none of the blogs have picked it up given how massive this thing might end up being. I’ve gone through it, and while I don’t have domain expertise to determine whether all the assumptions and approximations are strictly valid, it seems pretty easy to follow and the result is striking. No more cosmological fine tuning problem and no more dark energy, and all done with relatively simple concepts that have been lying around for decades! Best, Steve.
Sorry, this is off-topic and you need someone more expert than me in these issues about the CC. There’s a long history of claims of ideas of how to get rid of the CC (my roommate Nathan Myrhvold was working on one more than 30 years ago), none of which ever seem to have gotten significant acceptance. You could try Sabine Hossenfelder at Backreaction…
There is an issue in math over computer assisted proofs, though it’s still pretty small scale. I was actually involved with one, there’s a well known proof, due to Viswanath, that random Fibonacci sequences grow exponentially which was computer assisted, published, accepted. The basic idea was Viswanath controlled round off error. A colleague and I came up with a very elementary proof of the same fact (you could explain it to a high school student), though we only got rough bounds on the growth rate, the computer proof gave a very good estimate. We, however, could generalize to more elaborate random sequences where you not only randomize plus/minus, but you allow for multiples of previous terms. In that case we got a sharp bound for when growth can occur (which again confirmed a computer estimate for the number, by Embree and Trefalyn).
I think it’s different than Nomura though, there are serious, long term discussions of what is valid, and why or why not.
Does there really still exist significant controversy about computer-assisted proofs? In my experience some mathematicians could say that a proof requiring extensive computer calculations is perhaps not very beautiful or satisfying, but I don’t think that they’d call it invalid.
This is getting far off-topic, but no, I don’t think that computer-assisted proofs are “controversial”.
The controversy here over inflation is I think very different. The question at issue is whether scientists pursuing these models have made them so flexible that they now don’t solve a problem they are advertised as solving (the fine-tuning problem) and can be fit to any conceivable new data, so “inflation” has become an idea immune to falsification.
Pingback: Podcast CB S&R 111: Popurrí de noticias | Cambio climático | La Ciencia de la Mula Francis
My reading of the letter to Scientific American by the 33 was that the target audience was purely the editors – not for public attention, with a general feeling of trying to pressurise the editors: “Don’t try anything like that again if you know what’s good for you”. I’ll say no more, but I didn’t like that side of it.
I’ve just had another look at Nomura’s 2011 paper and I want to moderate my earlier remarks somewhat. I do think that these attempts to apply the many-worlds interpretation of quantum mechanics to cosmology are seriously flawed. But there’s another side to Nomura 2011, and to Bousso & Susskind 2011, which is the concept that de Sitter universes are metastable and eventually decay into a flat supersymmetric universe. It should be possible to get rid of the Everett-like many-worlds part of the argument, and replace it with a more conventional understanding of quantum mechanics. What would be left is just a physical hypothesis that could then be discussed sensibly.
The problem is that most proponents of the “many-worlds interpretation of quantum mechanics” claim that they just mean conventional QM (state space + Schrodinger equation) applies to the whole system including the measurement apparatus. In the case of early universe cosmology, this is pretty uncontroversial, the default assumption is that it is governed by conventional QM.
With many-worlds, if you try in the non-cosmological case to start claiming this implies branching into multiple universes, you immediately run into the problem that you don’t have a useful description of the branching process or how to characterize possibilities and how to count them. This is on top of the usual problems that you don’t even know what space you are trying to put a measure on.
Sure, Nomura can claim that many-worlds applies in the cosmological case, but as far as I can tell it solves no problems, just introduces more issues you don’t understand. If he were able to derive anything about the real world from this, that would make it worth paying attention and seeing what he is doing. Instead this seems to be a completely ill-defined proposal that leads nowhere. Again, why is such a thing on the cover of Scientific American???
When you talk about early universe cosmology, do you just mean that in the early universe essentially the whole universe has a wave function? In essence, there is no measurement apparatus? If that’s it, then how do you account for the first wave function collapse? I mean why wouldn’t the universe just evolve as a wave function satisfying the Schrodinger equation?
Can somebody (or somebody’s friends) comment on whether Scientific American papers in biology, chemistry, or some other non-physics field have undergone a substantial drop in quality over the last few decades. Then we’d know whether this phenomenon stems from the high-energy physics community, or from the editors of Scientific American.
The “many worlds interpretation” is basically just the claim that there is no wavefunction collapse, that you can make sense of quantum mechanics without invoking that as something separate. As far as I can tell, saying “I’m going to use the many worlds interpretation to do cosmology” is an empty statement, since that’s what people already were doing, since there is no separate experimental apparatus. It’s not surprising then that his supposed “prediction” is the same one already in Susskind’s book about the string landscape over a decade ago.
Here is a 2007 article decrying SciAm publishing an article by Peter Duesberg (full context in the article):
Here is a 2010 article concerned about a decline in the quality of SciAm coverage of climate change:
So now I’m confused. From what I remember, many worlds says that when the wavefunction would collapse, it doesn’t collapse, it splits, possibly into many separate wave functions. All of which go on happily evolving according to Schrodinger until they split again. So no collapse, but something takes it’s place, and has an effect on the universe (which is just bigger than the normal one, since it contains many copies every time there’s a split). So it eliminates collapse, that’s true, but in cosmology, where people don’t think about collapse I guess since there’s no observer, there is a single strand in the universe, and under many worlds there are many many strands, all separate and unable to communicate with each other. And in each strand it LOOKS like the wave function collapses, right? That’s the whole point, you can’t possibly tell, many worlds is pure interpretation, and can’t ever be shown to be false.
I really don’t want to try and get into the details of quantum mechanical interpretation issues, and exactly what different people mean when they say “many worlds”, there are lots of places you can read about that, it’s a complicated story. Maybe some day I’ll write a post about that, especially if I run across something that explains the issues well. As I wrote earlier, I don’t see how taking those complexities into account helps at all with the conventional cosmological multiverse problems, and that’s what Nomura seems to be claiming.
The reason SciAm publishes these articles is because they are losing money. Controversy sells. If academics want impartial publications they should tell their universities board of directors to form an association with other universities and buy SciAm as well as other important magazines, (they will never be cheaper then they are now). Universities are rolling in cash (look at football coaches salaries) they can afford to support these magazines. If you have an association of ten or more schools you can avoid the problems of arxiv.
These days universities (at least in the US) are looking for ways to make more money, not money-losing operations to take over. Foundations like the Simons Foundation are the organizations not looking for profit-making opportunities. They already have Quanta magazine, and maybe the way the world is going, they’ll start playing the role that SciAm used to play.
I was just about to mention Quanta. Between that and their support for the arxiv, it’s impressive how much the Simons Foundation has done for physics and science in general.
Nautilus is another magazine that’s been successfully operating in the same space as Quanta, although I’ve heard that they are feeling some financial pressure lately. It’s a bit depressing to think that having a billionaire patron is the only way to run a non-hype driven popular science magazine.
Peter is very generous to let the readership off the hook, but I think we bear some responsibility. The wisest thing to do with clickbait is ignore it, hence putting a disincentive on producing it. But does the typical reader resist that urge? Of course not. I’m just as guilty as anyone. Even when we know it’s complete B.S. we find it hard to resist.
I resolve to try to do better.
This story has reached a mainstream outlet:
I’m not the least bit interested that it’s a “literally cool idea”: I seriously considered about the level of intellectual hygiene reaching the public. What are us outsiders missing? Should we keep our mouths shut and let this run its course?
Verlinde has now made it to ArsTechnica:
It even mentions the false idea that glass flows, albeit slowly. The quality of popular science reporting is going down in general, I think.
Re analogous controversies in math:
In math, there is usually no real controversy about truth. That is, computer-assisted proofs are accepted as correct (maybe not enlightening), and the general notion of correctness in mathematics has not really changed since the 20s.
Instead, we mathematicians fight over “elegance”, “relevance”, whether “this is how you should view this”. And yes, in my opinion there are entire subfields in mathematics that have lost touch with reality and produce mostly bullshit. For example the community that does “mathematical neuro-stuff”, i.e. coupled oscillator networks, chimera states, etc appears to be pretty bullshit (to me). This does not mean that the claimed theorems are wrong, but it means that the “claimed motivation” is wrong: The theorem might be correct, but it is irrelevant and only pursued in order to follow the latest funding fads; it makes no progress at all for the actual questions that are claimed as motivations.
There is no shortage of true theorems in mathematics; the difficulty lies in deciding what questions to ask, and (progressing towards) answering these relevant questions.
Also, see thurston “on proof and progress in mathematics”. It is quite clear that many computer-assisted proofs provide no real insight. People can rightfully argue that generating insight is a large part of what it means to be a mathematician; hence, it makes sense to be dissatisfied with some computer-assisted proofs. On the other hand, some computer-assisted proofs contain a lot of insight!