East Coast versus West Coast

Back when I was a student I remember learning that there were two possible sign conventions to use for the Minkowski space metric: the “East coast” one, mostly + signs, favored by relativists and Steven Weinberg, and the “West coast” one, mostly – signs, which was the one in Bjorken and Drell. Then, as now, the big centers of influence in US particle theory were on the coasts: Princeton and Harvard on the East coast, Berkeley, Stanford and Caltech on the West (these days one might want to add the KITP at Santa Barbara). I was educated by the Eastern establishment (who sometimes used the “West coast” sign convention), where gauge theory and the new standard model were the order of the day. The West coast, with its remaining pockets of S-matrix theorists and authors of popular books like “The Tao of Physics” and “The Dancing Wu Li Masters”, was considered to be rather behind the times and a bit soft in the head, perhaps attributable to too much time spent in hot tubs at Esalen and too much use of the agricultural products of Mendocino and Humboldt county.

Remarkably, these days the two coasts remain dominant, with US Fundamental Physics Prizes going only to theorists living within a relatively short drive to an ocean beach. An East coast – West coast disjunction in interests remains, one that it remains tempting to speculate may have something to do with California’s main cash crop. For quite a few years the West coast has been the center of multiverse-mania, and I’ve often wondered what the theorists there would turn to when that lost its “new cutting-edge theory” shine.

It remains unclear what will happen in the long-term, but there’s now a new hot topic in California these past few months. It was the subject a few weeks ago of a workshop (AKA “brain-storming session”) with 50 or so in attendance at Stanford, and is being described by Raphael Bousso (across the Bay at Berkeley) as “this is probably the most exciting thing that’s happened to me since I entered physics.” A full-blown conference is rumoured for April.

LA-based science writer Jennifer Ouellette does a characteristically excellent job of covering the story, starting here by explaining the appeal of the subject (as well as the problem with explaining it):

To include every last detail, the piece would have had to be a good 6000 words long, and frankly, very few general readers would care to slog through all the gory details. So why even bother to try, if one can’t be comprehensive? Because FIREWALLS! That’s why! Seriously, how cool is this concept? There’s nothing more crowd-pleasing than death by black hole (just ask Neil de Grasse Tyson) and now there could be more than one way to die. Spaghetiffication, or incineration? Take your pick.

Another version of that post, which includes an excellent set of references is here. For her full treatment, see the version at Simons Science News.

All of this started with “AMPS” a July paper by four Santa Barbara physicists that already has 25 citations and counting (although of the three papers on the topic by Susskind, one is already “Withdrawn because the author no longer thinks it is correct”). For more on the topic, you can try Bousso’s Strings 2012 talk, blog entries by Polchinski at Cosmic Variance, Caltech’s John Preskill at Quantum Frontiers, Santa Barbara’s Aron Wall at his Physics and Theology blog, or Robert Helling here. I haven’t myself tried reading these papers, partly because I strongly suspect that I’d end up with the same reaction as Robert:

Now, of course I had to read (some of) the papers and I have to say that I am confused. I admit, I did not get the point. Even more, I cannot understand a large part of the discussion. There is a lot of prose and very little formulas and I have failed to translate the prose to formulas or hard facts for myself. Many of the statements taken at face value do not make sense to me but on the other hand, I know the authors to be extremely clever people and thus the problem is most likely on my end.

The problem may be that Robert isn’t in California, but, like me, is too far East. Someone else brought up in the East coast tradition (and now so far East he has kind of fallen off the edge…) is Lubos Motl, whose reaction to this topic is that the whole thing is Peter Woit’s fault:

AMPS isn’t as bad or as obviously wrong as “gravity as an entropic force” but it’s still wrong and what’s worse about it is that it is pushed by some of the names that are more famous than Erik Verlinde’s name. None of those bad apples would really destroy an otherwise healthy research community but the main problem I see is that the bad apples can no longer be efficiently wrestled with. Or it’s not happening. It doesn’t look like anyone cares at all. Instead, it seems to me that people are defending their subjective and increasingly non-quantitative (and often downright wrong) ideas and these people’s connectedness to the journalists and other folks outside the research community itself and the related populism – instead of the scientific evaluation by those who actually understand the things as experts – have become the key determinants of success.

Will firewall-fever spread from the West coast, or is it just a flash in the pan? Time will tell…

On a personal note, blogging may be lighter than usual for the next couple weeks or so, as I travel further East for a vacation in Spain, Portugal and Paris.

Update: Bee’s comment reminds me that I had planned to include a link to George Musser’s SciAm piece about this.

This entry was posted in Uncategorized. Bookmark the permalink.

35 Responses to East Coast versus West Coast

  1. Anonyrat says:

    Well, it is quite sane to believe that but for Woit and Smolin, who have to sign off on all funding and tenure decisions in all North American HEP departments, particle physicists would be safely working on string theory, instead of on useless thought experiments at black hole horizons. :)

  2. Aaron Sputzland says:

    Peter, are you suggesting that West coast physicists are all potheads?

  3. Bee says:

    It did spread to Stockholm. If you have a lot of time, watch this


    (Don’t ask, I wasn’t there and I didn’t watch it.)

  4. Bee says:

    Btw, you can avoid the whole “firewall” issue by just accepting that the BH entropy does not count the number of microstates. Unfortunately, this discussion is dominated by string theorists who can’t get themselves to even think about this possibility, which I think is the obvious one.

    Also, George Musser wrote a very readable summary here


  5. Lee Smolin says:

    I’ve read several of the firewall papers and it seems to me this is the reducto of two assumptions 1) as Bee says, that the BH entropy counts the number of possible states in the interior of the horizon rather than the entanglement entropy or channel capacity of the horizon and 2) that all the information must get back out to scri-plus so evolution neglecting the region to the future of the singularity is unitary.

    The problem is that they are focusing on the horizon rather than where the real quantum gravity physics must take place, which is the singularity. It is remarkable that the authors of this paper believe they can resolve the puzzles of black holes while ignoring the only region of strong fields, which is the singularity. It is natural to suppose-and model calculations support-the hypothesis that in quantum gravity the singularity is removed and everything that approached the singularity and the information it carries proceeds to a new region of spacetime to the future of the singularity. There is no conflict with QM-evolution including this new region can be unitary.

    Some object to this scenario with arguments about remnants but these are unconvincing as we explain in detail with Hossenfelder in arXiv:0901.3156.

    Whether the region to the future of the singularity becomes classical and develops into a baby universe which is permanently disconnected or reconnects to future infinity, as shown in detail by Ashtekar et al in the CGHS model, depends on the details of the quantum geometry dynamics. But in no case is there conflict with the principles of QM, with the equivalence principle or with usual notions of locality.


    Lee Smolin

    ps see recent papers of Bianchi for new evidence BH entropy is entanglement entropy.

  6. Peter Woit says:


    Thanks a lot for your take on the issue.


    Thanks for the links, I was planning to include George Musser’s article, will add that to the posting.

    Also worth pointing out is this from Bee:

  7. Peter Woit says:


    Certainly not. I don’t have any data on the question of why the Multiverse and black hole firewalls are mostly popular on the California coast, maybe others do. Could be lots of reasons: a friend who grew up in California, then moved to Princeton and finally England, told me that the main differences noticed while moving East were that members of the opposite sex got less attractive, and the cheddar cheese got better. So, could be one of those…

    Besides the Scandinavian connection pointed out by Bee, there are other non-Californians involved, including the Dutch here
    which would argue against this having to do with local California culture. Then again, I just remembered some of the local Amsterdam culture I once encountered…

  8. Aron Wall says:

    I think the AMPS argument uses your assumption #2, but not necessarily assumption #1. So long as the infalling observer always sees an approximate vacuum, the Hawking radiation coming out near the horizon is thermal, but if the radiation coming out is thermal, then the information doesn’t leak out.

    I say this because Don Marolf and I have been arguing for a viewpoint where any information that falls into the black hole from outside will always escape, but there can still be information in the interior which cannot be measured from outside at any time (if you consider all possible black hole states, not just those that formed from collapse). See http://arxiv.org/abs/1210.3590 and some previous cited work by Don. This would imply that #1 is false but #2 is true.

    I think you’re right, though, that the AMPS argument might not apply if the information doesn’t come out until after the remnant forms. Seems to me that people on the “West Coast” disbelieve in remnants primarily on the basis of AdS/CFT.

    Thanks for noticing my blog! I’ve been thinking about linking to your blog for some time now, seeing as I follow it, but I haven’t gotten around to doing it yet. I went to grad school at Maryland, partly because I wanted to study quantum gravity at a place that wasn’t completely dominated by string theorists. But since then I learned which kind of string theorists are worth talking to about fundamental conceptual issues, and I’m very happy here at UCSB.

  9. lun says:

    Judging from
    the East Coast (with a fluctuation towards the north-center) is steadily falling into a black hole of mathematically sophisticated analysis of non-existent theories.
    It remains to be seen how thick is the firewall separating this black hole from the real world.

  10. P says:


    in which sense you do mean “non-existent theories?” Please prove me wrong, but essentially any way I slice it, this seems like a pretty ignorant comment.

    Nima and company are arguing that this is a generic structure of QFT – yes, they use N=4 (which DOES exist as a theory, though doesn’t decribe particle physics) SYM as an example, but they go into detail about how to generalize to theories with less or no supersymmetry. It is far too premature to make the claims you are making. Time will tell how applicable these techniques are to non supersymmetric theories, but the techniques are deep and potentially very important.


    “The problem is that they are focusing on the horizon rather than where the real quantum gravity physics must take place, which is the singularity. It is remarkable that the authors of this paper believe they can resolve the puzzles of black holes while ignoring the only region of strong fields, which is the singularity.”

    Your slightly condescending tone is also remarkable. Isn’t their whole point that new physics was never expected to be encountered at the horizon!? Telling them that the singularity is the issue seems to be dodging their entire argument. They’re not claiming to solve every “puzzle” of black hole physics, as you imply, by focusing on the horizon – but are merely pointing out that there is an issue with our current conceptual understanding of horizon-scale physics. Of course these authors know that the singularity is important for issues of quantum gravity . . .

  11. Peter Woit says:

    P and lun,

    Enough about the “positive Grassmanian” paper here. I’m all in favor of an informed discussion of that topic at some point (maybe someday I’ll know enough about the subject to host such a thing..), but it’s an extremely technical one, and I see no value to this kind of arguing about it.

  12. Bob Jones says:

    The mathematical techniques that Arkani-Hamed and his collaborators are developing is useful for studying all sorts of thories — not just N=4 SYM. As pointed out by P, their results can be used to study scattering amplitudes in non-supersymmetric theories. The cluster algebra formalism that they’re using is also important in two-dimensional conformal field theory, wall-crossing, and other areas.

    Sorry to keep harping on this paper, Peter. But lun’s comment is just outrageous…

  13. lun says:

    Since you are allowing a whole bunch of critical comments, I hope a reply is also acceptable. What I meant by “non-existing” is that theories examined in 99% of that paper, as far as I know, do not exist. Real QFTs, as far as we know, are far away from the planar limit and do not have any SUSYs. There might be profound reasons for this, or not, we do not know. Something relevant to the real world might be derived from this or not (read my comment, I say this explicitly), we do not know. Any applications of these techniques to understanding non-planar non-SUSY theories (you know… the real world) are at the moment remote.

    That such a profound amount of work has gone into a deep mathematical understanding of such non-existing (in the above sense) theories is remarkable. Since these theories are not realized in the real world, all this might be as “scientific” as the multiverse or firewalls, so it is not irrelevant to this discussion.
    It also has a direct connection to the East-West Coast issue, since these techniques descend, in their philosophy, from the S-Matrix approach (rather directly, read the introduction about what is relevant and what isn’t in a theory!), yet the best East coast physicists are now working on this.

    Its amazing how many feathers one can ruffle by saying non-controversial truths.

  14. Peter Woit says:

    OK, enough about this, really.

    I do think though that the historical comment is correct: this latest East coast effort is a descendent of the West coast S-matrix theory program of the 60s. Whether the arguments about the desirability of abandoning local field theory will work out better this time around is still to be seen.

  15. Yatima says:

    Okay, so the gravastar idea is back?

    Note that a “firewall” used to designate a fire-resistant partition in a building. Now it designates a membrane when/where you are instantly de-rezzed.

    Oh Tempora, Oh Language Perversions.

  16. Michael Welford says:


    Your list of West Coast books leaves out “Quantum Reality”. It’s the best popular account of quantum theory that I’ve ever read. And it would still be the best book even if it didn’t include the best description ever of nonlocality.

    I noticed that one of the preconditions for the firewall is the principle of Conservation of Information, a string theorist fantasy that they’ve somehow bamboozled real physicists into believing is an actual law of nature. In fact, Rolf Landauer established the relation between irreversable information changes and irreversable thermodynamic changes back in 1961. And this didn’t come from out of the blue. Chemists had figured out decades before that increasing entropy was associated with information loss and physicists had built much of the mathematic needed. Since Shannon gave the expression for information entropy in 1948, this discovery was arguably a decade overdue. In the years since no has built a computer or discovered a biological process that defies Landauers Principle.

    By an odd coincidence, recently I was watching one of the principals on the Firewall Debate, John Preskill, lecturing on quantum information. It was boring, so I skipped through it in case he said something interesting. Nope! But then I saw this post, and decided to give the lecture another chance. And I fell asleep. But third time is the charm and the lecture turned out to be more or less adequate. Funny thing though, the lecture was explicitly about quantum information, but there wasn’t a word about information conservation. Could it be that Preskill accurately gauged the reaction from theorists at having one of the fundamental principles of their discipline contradicted?

    And speaking of quantum computing/information theorists, it’s past time that the Big Names got off their asses and raised a howl about this Conservation of Information business. Or better yet, they should get their asses back in their swivel chairs and type up some refutations.

  17. Bee says:

    Aron: “People on the West Cost” don’t believe in remnants primarily because of the “pair production problem”, which however has never been shown to exist under reasonable conditions. The only cases I know where it occurs assume the validity of effective field theory and treat the inside of the remnant as having a small volume. Now it is not clear exactly why effective field theory should fail at the horizon far away from the Planckian curvature regime, but nobody seriously expects effective field theory to be valid in the strong curvature regime, which is what you have to deal with if you have Planck scale remnants. So why does anybody care about an argument that creates an alleged problem based on a method that one expects to fail? It’s a mystery to me.

  18. Ian Durham says:

    @Michael Welford: Just out of curiosity, why would anyone in quantum information even attempt to refute information conservation? It would invalidate large swaths of our work. Personally, I’d be open to the idea, but it would require some major reconsiderations of our not just our results but our methods as well.

    Personally, I think one of the major problems in this debate is that quantum theory and relativity treat space and time in fundamentally different ways. Thus, it’s really a debate between what Chris Granade has called the (in an intentional pun) “psi-ontologists” and the psi-epistemics. The former believe that the spacetime manifold, not to mention fields in general, possess some ontic status. This sort of world view is built right into relativity itself. As it relates to the firewall problem, this means that the event horizon represents something about space and time itself.

    On the other hand a psi-epistemicist might be persuaded that the event horizon represents, rather, something about an interaction and thus can be interpreted as some kind of information-theoretic boundary. This is attractive if spacetime is an emergent phenomenon and there is, indeed, a growing feeling that it very well may be.

    My own feeling (though I am not the first or only to consider this possibility) is that the interior and exterior of the horizon represent different superselection sectors and thus there is a superselection rule in place that manifests itself classically as this apparent loss of information. On the other hand, quantum resource theory includes techniques that can be used to overcome such superselection rules and thus the information can be recovered in a quantum information theoretic kind of way.

    That’s a very rough and brief summary of the idea and there’s much more to it, but it at least may give an idea of why some of us have no desire to jump on the refutation bandwagon.

  19. Ian Durham says:

    I should note that the superselection explanation only works if one takes a psi-epistemic view. The difference between a timelike and a spacelike geodesic couldn’t be ontic.

  20. Aron Wall says:

    Of course there is more than one person on the West Coast, and even when they believe the same thing they don’t necessarily believe it for the same reason. From my limited perspective here at UCSB, most people here (except Steve Giddings) seem to regard AdS/CFT as a more definitive refutation than the pair production argument, which as you say makes some assumptions from field theory which might well not be valid.

    As for the information question, I am a reluctant convert to the idea that information isn’t lost. Most of the arguments against info loss are rather silly (it’s not any more a “violation of quantum mechanics” than the other view) but Don Marolf’s argument for holography here: arxiv:0808.2842 is hard to get around since it follows almost trivially from the facts that (i) the energy can be measured at infinity in general relativity, and that (ii) the energy generates time translations in quantum mechanics.

    Now that firewalls have come along, I’m reconsidering whether I should convert back, but I still don’t see any good way around Don’s argument.

  21. Michael Welford says:


    The Landauer Principle has real world consequances. Todays computer chips are only a couple of orders of magnitude from theoretical limits on energy consumption. And we probably won’t get all that efficient, because a low energy chip is a reversible chip is a slow chip. Once we approach the Landauer wall we’ll need all sorts of optimization decisions: getting by with smaller registers?, strategic use of qbits?, how about analog?, new fabrication technologies? Now imagine some funding functionary telling you that your proposal is irrelevant because information is conserved so we’ve already achieved perfect efficiency.

    With regard to your deeper remarks, I would like to point that I am well acquainted with the epic-ontic controversy. I just can’t remember which is which.

    Ending this comment before Peter slaps us both down for drifting off topic.

  22. Peter Woit says:

    Yes, I really can’t moderate a general discussion of quantum theory. If it’s not about firewalls or East/West coast physics, find someplace else to discuss it…

  23. Daisuke says:

    A couple odd thoughts–

    Is it actually certain that EP=no horizon wall? There are some rather dramatic phenomena in classical black hole dynamics, the inner horizons of charged and rotating holes for instance–you’d certainly expect matter falling into a near-extremal black hole to encounter a “wall” of that sort under the event horizon. IIRC the long-term evolution of a BKL singularity, while lacking a “creep” outward to the event horizon, tends toward some sort of abrupt “wall” rather than a drawn-out “spaghettification” near the final approach. Given the generally poor state of understanding of the evolution of BH interiors, is it possible that the classical evolution of the interior could produce a horizon wall of some sort by the Page time, thereby preserving the EP?

    Bousso’s preferred solution to firewalls seems to be a limited relaxation of normal causality requirements, which brought to mind the Feynman-Wheeler absorber model of radiation–and now I’m wondering how that would behave in highly curved spacetime. The behaviour of advanced waves “approaching” an event horizon would have to be peculiar indeed…

  24. Ian Durham says:

    @Michael (and Peter): My comments were about the firewall issue specifically (I really have no opinion on the East/West Coast matter). Discussing the Landauer Principle would be getting off topic and I have no intention of debating that topic here.

    But the superselection question is very germane to the firewall question as is the ontic v. epistemic debate since which view one holds determines the route one takes to attempt to solve the firewall problem (as I clearly pointed out above).

  25. Peter Shor says:

    If you were a firm believer in black hole complementarity (and I believe that the West Coast was the epicenter for this belief), then the quantum information argument showing that something is really wrong with the traditional view of black hole complementarity needs to be addressed. Firewalls seem to me to be a desperate attempt to preserve some of the ideas of complementarity, although my intuition is that they really can’t work.

  26. Fry and Laurie says:

    How do you know so much about Mendocino and Humboldt counties?

  27. Spencer Tracy Jr. says:

    I myself don’t know anything about those counties – but I’m thinking seriously about heading out there to see if I can find a good looking woman and some bad cheddar.

  28. OldStuff says:

    There are some matters of detail in which the firewall guys are different, but the basic idea that classical gravity can have corrections already at the horizon scale is something that Samir Mathur has been saying for a long time, and the string theory “establishment” has been consistently hostile to.

    The real reason why most string theorists are suddenly taking the firewalls seriously is because Polchinski is one of the authors.

    Of course Polchinski himself is a very fair person and he gives adequate credit to Samir in all the talks I have heard. But some of his collaborators on the other hand, try to revise history and present it as though all Samir did was to come up with one of the qubit toy models for black holes.

    Pretty dishonest, if you ask me.

  29. MathPhys says:

    Of course Polchinski himself is a very fair person and he gives adequate credit to Samir in all the talks I have heard. But some of his collaborators on the other hand, try to revise history and present it as though all Samir did was to come up with one of the qubit toy models for black holes.

    Don’t worry about it. It’s been always like that.

  30. Jean says:

    Don’t know whether it’s worth posting but Peter’s blog has been mentioned in the list of mathematics blogs here-> http://www.talkora.com/science/List-of-mathematics-blogs_112 (look for entry #2 in the list.)

  31. Miguel Carrion says:

    I’m going to have to read the papers in some detail because for the life of me I can’t imagine what would prompt Polchinski to write that “the infalling observer encounters nothing unusual at the horizon” is the most easily dropped assumption in the firewalls paper, and that “perhaps the most conservative resolution is that the infalling observer burns up at the horizon”.

    Maybe this is just philosophical prejudice on my part, but it seems to me that the regularity of the Eddington–Finkelstein coordinates around the horizon uncontroversially implied the horizon could be crossed, at least by a physical system small enough not to be destroyed by tidal forces there.

  32. DaveH says:

    Peter Woit,

    I thought your East/West and signature intro was interestingly correlated with your following discussion of AMPS for a couple reasons. First, the t/r signature actually flips at the event horizon (so East becomes West there I guess ;) (Tongue in cheek, and to avoid “helpful” replies, I know about better coordinate choices; but also see MTW section 31.3 which gives the flip more of a physical interpretation nonetheless.) Secondly, I always thought the convention difference was more GR guys vs QFT guys, not East vs West, to me the split in interpretation also seems to track this better than East vs West.


  33. Low Math, Meekly Interacting says:

    I don’t know if Dr. Woit is a connoisseur, but horticulturalists, being who they are, always strive to breed better crops. And how. SoCal has its rivals, but even locavores of today are getting a product that is astonishingly potent compared to what was available in the 70′s. If these advancements have continued to impinge on physics in any way, their effects are likely to be amplified.

  34. Sebastian Thaler says:

    Off-topic, but I can’t resist: http://news.yahoo.com/atom-smasher-hiatus-sets-stage-more-discovery-134311023.html
    I can’t wait for early 2015, when the LHC “atom smasher” will be “unlocking more mysteries,” just as last July’s discovery of the Higgs “promises a new realm of understanding of the universe,” and when it’s gained the “capacity to simulate the moments after the Big Bang nearly 14 billion years ago.”

  35. Small Bang says:

    Indeed why not? It’s a reasonable article. Read the last paragraph:

    “It will bring you more collisions. Which means that the more collisions you have, the more likely you are to see rare events,” he said. “The Higgs particle was just one of many on the wish list that we’d like to find, so higher energy increases your discovery potential.”

    That is a very fair statement. (BTW all of these machines are known as atom smashers. Get used to it. And don’t get picky that `atom bombs’ are really ‘nucleus bombs’.) The SSC was openly advertised as creating conditions not seen since the Big Bang. All of the quark-gluon plasma business is similar `not seen since the Big Bang’ stuff. Perhaps you would prefer to say “We have no clue what any of the output from the LHC means”? The obvious riposte to which is “Then why are you doing HEP?”