Various news about the usual topics:
- Natalie Wolchover at Quanta magazine keeps coming up with great, in-depth stories about interesting new topics in physics that are getting no attention elsewhere. Her latest is about the universality of the Tracy-Widom distribution.
- The LHC is cooling down, in preparation for a restart early next year. Nature has a good story about what is going on here. Latest status and plans are described here. The current plan is to start beam recommissioning next March, have 1 fb-1 by mid-June, in time to perhaps have some results to report at EPS-HEP2015 at the end of July. Another 10 fb-1 would be accumulated later on, before a heavy-ion run late in the year.
In the long term, by 2023 there should be 300 fb-1 and many components of the machine and the experiments will start to become unusable due to radiation damage. Planning is going ahead for “Phase-II”, or the HL-LHC, with Bertolucci’s comment that “It is inconceivable under any reasonable scenario to stop the LHC program at that point”.
- Nature has an editorial this week about What lessons can be learned from the presentation of the gravitational-waves story?, pointing to a planned discussion next week about Lessons in the communication of science from the BICEP2 story. I’ve already written extensively about this, but since the editorial refers to bloggers (and I know some people at Nature were unhappy with my blog entry about this, which was poorly worded), I’ll take another opportunity to do so.
From the purely scientific point of view, this is a pretty straightforward situation. The BICEP2 people fooled themselves into thinking that they had something much more exciting (primordial gravitational waves + evidence for inflation) than what they really had (a good measurement of B-mode polarization at one frequency). They then wrote a paper with over-optimistic claims, which later blew up in their face. This is perfectly normal science.
What’s not normal science is the behavior of a lot of theorists in response to the BICEP2 claims. The Stanford University Linde video and its 3 million downloads will live forever as an example of misguided PR for science. The comments from theorists about the significance of this for string theory that Nature quoted were an embarrassment for the field (why not just say that you could get any value of r out of string theory?), and even worse were the publicity campaigns from Linde, Guth and Carroll aiming to convince the public that this was evidence for the multiverse.
What’s the lesson for science journalists? Take a hard look at the behavior of some prominent theorists in this story, and draw the obvious conclusions for your future coverage of developments in this field of science.
- Just noticed that Sean Carroll is now trying to raise research funding online with a website devoted to attracting private funding. Will be interesting to see if that works, maybe it will become a model for how to fund this kind of research.
- One of the few things I’d change about my book written ten years ago would be the discussion of the philosophy of science “demarcation problem”, that of deciding what is science and what isn’t. Only after writing the book did I learn about the distinction between a “progressive” and “degenerating” research program due to Lakatos, which is a very good way of addressing the question of how to evaluate string theory. I also missed a paper that came out a few years ago by Johannson and Matsubara on String theory and general methodology. At one point they write that the string theory landscape business shows that:
String theory is a degenerative programme, according to Lakatos’ criterion.
There’s a lot more in the paper, it’s a good example of what I’ve seen too little of, philosophers of science engaging with the real issues here.
- For the latest on the string landscape, there was a conference last week on Fine-tuning, Anthropics and the String Landscape. See if you can find anything there like a plausible idea for how to get any testable physics, I couldn’t. Alan Guth’s introductory talk mainly explains why the measure problem means you can’t predict anything, but then ends with a claim that physicists take the multiverse seriously anyway, quoting Weinberg from 2005 about Martin Rees’s dog.
Back in 2004-5, the expectation was that the string theory landscape could be used to predict whether SUSY breaking would take place at a high or low scale (see for instance here). That idea is long dead, and no other proposal for a prediction has replaced it. So, the string landscape is itself a degenerating research program. What do philosophers of science call it when a research program degenerates into something else, and that research program in turn degenerates. A (degenerating)2 research program?
- The standard defence of string theory these days acknowledges that it can’t explain particle physics, but claims it has had great success in quantum gravity. Next spring the KITP will have a program on quantum gravity foundations. The description of the program has a lot to say about “deep connections between quantum information theory and gravity”, no mention of string theory. There seems to be a move away from string theory and a convergence between the KITP and the sort of alternative research favored by the Perimeter Institute.
- Speaking next month on Quantum Mechanics and Spacetime in the 21st Century at Perimeter will be Nima Arkani-Hamed, one of the organizers of the KITP program. Not clear what he’ll be arguing for then, but he did just give a talk at an Oxford workshop on New geometric structures in scattering amplitudes, with the title “The Amplituhedron, Scattering Amplitudes, and the Wavefunction of the Universe”. I’m curious to see how he gets the Wavefunction of the Universe, although I suppose one should keep in mind his comments here.
- Not announced yet what the price of tickets to Arkani-Hamed will be. For a real rock star of physics though, I think you want Brian Cox, who is on tour in Australia. Premium tickets there are about $175 US.
Update: The latest on the Journal of K-theory situation, from algtop-l
The time has come to advise your librarians to cancel the subscription to the Journal of K-Theory. The precious money could be better spent elsewhere.
As you know the journal is going through a crisis. The most recent development is that the Bak family has written to Cambridge University Press informing them that they are under a contractual obligation to keep publishing the journal through the end of 2017, whether they like it or not. I haven’t seen the contract in question, not have I seen the letter from the Bak family to Cambridge University Press, hence I cannot comment on the legal merits of the case. The Baks evidently feel confident, Tony Bak has accepted at least one paper for the 2015 edition of the journal without clearing it with any of the other editors.
The Baks might be right, Cambridge University Press might have no choice but to continue publishing the journal. But the vast majority of the editors will be walking out and the scientific standards of the journal are bound to plummet. It would be a waste of money to continue subscribing.
Just curious if you’ve seen an advance copy of Lee Smolin’s new book, being published next month by Cambridge. It sounds interesting and I’d love to hear your thoughts.
I haven’t seen the new book. From the synopsis, it sounds like it’s largely along similar lines to Smolin’s last book, which I disagreed a lot with, see here
He and I may agree about a lot of the problems with string theory, but we have completely different visions about how mathematics and physics relate to each other.
“…Sean Carroll is now trying to raise research funding online…”
And I notice that his vehicle of choice, benefunder.org, says they’ll take a flat 10% fee off all donations (or perhaps weaseled to >15% since they also claim researchers receive “nearly 85%”.) Leaving aside the value of the actual research, doesn’t a slapped-together website with 10% overheads seem like a bit of a scam?
Peter, I think PI is also gravitating more towards conventional physics in the last few years. You can get an idea by looking at their list of seminars and colloquia in the last few years compared to when they initially started.
Will be funny if places like the KITP become the hosts for exotic speculation (eg. gravity is information), and PI becomes the place for staid, conventional research…
My impression is that a 10-15% overhead expense for non-profit fundraising is not bad at all. And I thought the website (at least Carroll’s part of it that I saw) was a well-put together professional effort.
One danger of this kind of thing is not benefunder scamming the public, but the people advertising for funding doing it. When the NSF gives a scientist a grant, there are elaborate structures in place to make sure it is spent as intended, I wonder whether that’s true in this set up. Is there really anything there to stop me from putting up a glitzy web-site advertising for funding for my investigations into the nature of reality and mathematics, but then using whatever comes in to finance my next vacation in Europe?
This kind of funding raises a lot of interesting questions. Carroll’s site makes all sorts of claims (just as a random example, that his “work is unique because he takes a big picture point of view that answers questions about the universe in the broadest possible terms”) that would quickly get an NSF grant application put in the discard pile as not serious science. This is advertising aimed at the general public, something very different than making a scientific case to one’s peers. The NSF is cutting back on support for HEP theory research, and a lot of new funding is coming in from private sources. So far, most of the private money has come from wealthy foundations, which often have a sophisticated staff to judge proposals and make sure the money is spent appropriately (as well as sometimes people like Jim Simons who can make scientifically informed judgements). That situation raises one set of issues, which I’ve discussed here often. Fundraising via direct advertising to the public, without oversight by foundation officers, is something I haven’t seen before, I wonder how it will play out. What would theoretical physics look like 100 years from now if this becomes the model for how to fund it?
This kind of funding raises a lot of interesting questions. Carroll’s site makes all sorts of claims …
Hi, is something wrong about putting my own money directly to a research proposal, whatever vague it is? Does it mean that when a Joe gives a few bucks to Carroll, CERN looses some funding? According to my experience, research foundations tend to support current views and approaches with near and specific targets, thus effectively putting barriers on unbeknown ways. Any new source, especially when allowing more fundamental work (that usually needs a lot of relaxing), should be welcome, I would say.
I’m not making any simplistic claims about what is bad or good. I just think that people should think about the significance of changes in how research is funded, not just assume that all new money is a good thing . At least in the US, it is a fact that government funding for theory research is being seriously cut back, and private funding is moving in to replace it.
If the private model is a success, you’ll likely see arguments made in Congress that this is a subject where government funding should be removed. You’ll also see universities moving from hiring people who can get NSF grants to hiring people who can bring in money this way.
The NSF peer-reviewed model for funding is not ideal, suffers from problems of favoring unambitious or conventional research. But there are also problems with funding research that makes grandiose claims about ambitiously revolutionizing the subject (I get a dozen of such claims in my email each week….). Is it better for people to have to convince their peers to fund them, or have to convince the general public/rich people (who don’t actually understand anything about what they’re funding). Both have problems…
Doesn’t the core of the anthropic argument, an “unnatural” lambda, basically render the entire landscape programme infeasible?
If lambda is unnatural then finding our vacuum state would involve working out a bunch of different quantities associated with some particular 10-dimensional CY manifold and then seeing if they all cancel each other to 120 decimal places. No-one has even been able to use plain old lattice QCD in three dimensions to calculate the proton mass to one decimal place!
Yes. And even if you could do that computation, there’s this problem
The whole string landscape business is very odd. There are all sorts of strong arguments that you can’t get anywhere, have been for ten years. Everything people have done just makes clear these are insuperable problems, and yet, it somehow remains a subject people work on, and make claims to the media that this is a great insight into how physics works.
@PW: I understand that both of the approaches have issues, and that any change has two sides. Still, regarding peers: they are at conflict of interest, grouping into mafia structures, they (according to my experience) generally do not understand stuff slightly off their own work, and they tend to dislike anything that goes a different way than they do, and more it is relevant to their own work, more irrational they are. Thus I personally feel sympathies to this new public approach.
PS Is not your fear alike Anderson’s fear about SSC?
Anderson was worrying about the allocation of money between fields of physics, I think this is kind of different.
Maybe one source of my worry about this has to do with seeing how US universities have changed during my lifetime. They’re now a lot more market-driven and PR-driven, and this is not necessarily a good thing for scholarship. And, for whatever reason, I seem to have an aversion to PR as opposed to serious discussion.
I seem to have an aversion to PR as opposed to serious discussion.
That’s the point: one day you can discuss with the public on a blog, the other day you can try to convince your peers within academia. Or it can actually go like: today a nonsensical PR, tomorrow a power abuse within the institution. Thus it seems to be about the general approach, not about a currently followed path.
Whatever way it will go I wish you good luck. And I guess that this blog puts you into a good position regarding a reasonable communication with the public.
Re. BICEP2 Lawrence Krauss made an interesting case today – in a talk called “Cosmology for Philosophers” at Bonn University in Germany – that there could well be a cosmological signal in the data after all (with a lesser r, I suppose, but good enough to test inflation). And that we should all wait until the end of this year when the *joint* BICEP/Planck paper will come out. So I will.
Sure, there could be an observable non-dust signal, we don’t know yet, but back in March all BICEP2 had was an observation of B-modes at a level consistent with that expected from dust. Claiming a “five sigma” observation of primordial gravitational waves was just completely wrong and unjustified.
Any real evidence of such a signal will be based on joint work of the two experiments, or maybe on Planck’s all-sky results on their own. Supposedly we’ll know end of next month. Whatever happens, doesn’t change the fact that the March headlines and PR were an embarrassment and a fiasco, a good example to various people of how not to behave.
In the Linde video, the experimentalists showed up at his door unannounced, and claimed a discovery that was not supported by the evidence. How do you transmute this into a statement about “the behavior of theorists” ?
Dear Sebastian Thaler and Peter,
My new book, The Singular Universe and the Reality of Time is written with Roberto Mangabeira Unger, a Brazilian philosopher who is Professor of Law at Harvard, and is thus a very different kind of book than my previous books. I hope you will find more to agree with in it, or at least take it as a challenge to argue with. It supports the conclusions of Time Reborn, but it is more radical and goes deeper.
In the end, you and I may disagree about the nature of mathematics and its proper role in physics (about which there is much more in the new book) but I suspect we agree that what really matters is to develop theories that make testable predictions for doable experiments. The new book lays out a program for doing that and presents some early steps.
Thanks, I look forward to seeing it.
Yes, you’re right, the experimentalists have most of the blame for that video. Linde’s claims about “a smoking gun” were in the Stanford press release, not that video, see
Linde could have pointed out to the guy who showed up at his door that while r=.2 was all well, and good, he had versions of inflation for any value of r.
Not announced yet what the price of tickets to Arkani-Hamed will be. For a real rock star of physics though, I think you want Brian Cox, who is on tour in Australia. Premium tickets there are about $175 US.
The Tickets for the Arkani-Hamed talk are free.
The reason why tickets to Brian Cox show cost $175 and to Arkani-Hamed talk nothing is simple to understand.
Having attended talks by both I would say that Cox knows much less physics but is much better as entertainer, for people who don’t know the subject.
I didn’t know who Brian Cox is, but he’s apparently also writing books.
From Alan Guth’s presentation:
I’m no physicist or expert of any kind, but doesn’t point 1 completely undercut the conclusion reached in point 3 here? At any given point in time, Alan Guth’s observation that he has continued to observe a coherent world over the previous five seconds, or two weeks, or 62+ years, can only be based on what he can recall of that stretch of past time. But if what he remembers, or thinks he remembers, is a function only of what his present brain-state is … and if (by point 1) that brain-state is (barring some extra constraint) more likely a product of fluke than the kind of environment and life history which his brain state tells him that he has … then what reason does he have to trust his recollection of experiencing a normal world over the past however-long? His recollection doesn’t let him conclude a suitable extra constraint exists, because his recollection itself requires the existence of such an extra constraint to be a reliable witness to that effect (or to almost any effect).
And it obviously doesn’t work for him to say “aha, I’ll settle this through experiment, by waiting five seconds to see if I still exist and life continues as normal during that time”. At the beginning of the five seconds the results of the experiment are of course not in yet. At the end of the five seconds, the results may be in but now he’s facing the equally uncertain question of whether his last several seconds of remembered past, including the decision to test his Boltzmann-brainness by waiting five seconds, were fictitious or not!
The “Boltzmann brain” business was a hot topic in certain circles six years ago, see for instance
I don’t think anything worthwhile ever came out of that debate, and it seems to have died out for very good reason. I have no idea why Guth is reviving it.
All: please don’t try and revive this debate here. The probability of anyone having anything interesting to say about it at this point seems to me to be about the same as that of a Boltzmann brain popping into existence.
Thank you PW.
I have enjoyed very much over the last several days (1) delving into where the Natalie Wolchove Tracey-Widom article have lead me and (2) reading of Lakatos’ ideas in philososphy of science (was pretty much ignorant of them before this).