Talking Back to Your Elders

When I was young and as yet unformed as a theoretical physicist, cosmology was in a transitional period. We had certainly moved beyond the relatively barren landscape of the 60’s and 70’s, when pretty much the only things one had to hang one’s hat on were very basic features like expansion, rough homogeneity, and the (existence of) the cosmic microwave background. By the late 80’s we were beginning to see the first surveys of large-scale structure, there was good evidence for dark matter, and the inflationary paradigm was somewhat developed. In the 90’s things changed quite rapidly, unbelievably so in retrospect. We detected primordial anisotropies in the CMB and began to study them in detail, large-scale-structure surveys really took off, we discovered the acceleration of the universe, and techniques like gravitational lensing matured into usefulness.

My students and postdocs will readily testify that I am fond of complaining how much harder it is to come up with interesting new ideas that aren’t already ruled out by the data.

In an interesting and provocative post, Peter Coles bemoans a generational shift among cosmologists: “When I was a lad the students and postdocs were a lot more vocal at meetings than they are now.” In particular, Peter is worried that people in the field (young and old) are “willing to believe too much,” and correspondingly unwilling to propose dramatic new ideas that might run counter to received opinion. Or even, presumably, just to express doubt that received opinion is on the right track. After all, even with all we’ve learned, there’s certainly much we don’t yet know.

I’m not sure whether there really has been a shift or not; there’s a big observational bias from the fact that I used to be one of those young folks, and now I am a wise old head. (Old, anyway.) But it’s completely plausible. Is it a bad thing?

There’s an argument to be made that widespread agreement with a basic paradigm is actually a good thing. People agree on what the important questions are and how to go about answering them. Ideas are held to a higher standard. Furthermore, it would be very hard to blame a young scientist who wanted to play by the rules rather than rocking the boat. It’s easy to say “challenge conventional wisdom!”, but the thing about conventional wisdom in a mature field is that it’s usually right. The exceptions are important and memorable (remember when everyone thought the cosmological constant was zero?), but most controversial new ideas are just wrong. Being wrong is an important part of the progress of science, but it’s hard to tell other people that they should be wrong more often.

At the end of the day, though, I agree with the spirit of Peter’s lament. I do think that the discourse within cosmology has become tamer and less willing to try out new ideas. Dark matter is well-established empirically, but we certainly don’t know that it’s WIMPs (or even axions). Inflation has had some successes, but we are very far indeed from knowing that it happened (and the problems with eternal inflation and predictability are extremely real). I have my own prejudices about what’s settled and what are interesting open questions, but the field would be healthier if youngsters would challenge people like me and make up their own minds.

Then again, you gotta eat. People need jobs and all that. I can’t possibly blame anyone who loves science and chooses to research ideas that are established and have a high probability of yielding productive results. The real responsibility shouldn’t be on young people to be bomb-throwers; it should be on the older generation, who need to be willing to occasionally take a bomb to the face, and even thank the bomb-thrower for making the effort. Who knows when an explosion might unearth some unexpected treasure?

This entry was posted in Academia, Science. Bookmark the permalink.

59 Responses to Talking Back to Your Elders

  1. When faced with dark matter, perhaps there is a way to cloak regular matter? For higher dimensional theories is there a way to make a higher dimensional phase space without making more spatial dimensions?

  2. I made a similar comment over on Peter’s blog: With so much data today, any new hypothesis which doesn’t fit the data is easily ruled out. Back when there wasn’t so much data, it was easier to come up with alternative explanations.

    Related to this is the fact that, until quite recently, one could actually know all there was to know about cosmology, and that at a young age. That is no longer possible. It is thus harder to get past those people who know more, who can easily find a flaw in your pet theory by mentioning something you don’t even know about.

    In summary: rebellion for the sake of rebellion is wrong. It was wrong then, and it was wrong now. However, in the old days, “positive rebellion” was certainly easier than it is now.

  3. Sean Carroll says:

    That sounds right to me. The question is, how to encourage a healthy amount of rebellion (whatever that might be) in times when a successful paradigm is firmly entrenched?

  4. James Salsman says:

    Much of the problem is that scientists will often say that some theory is “ruled out by the data” without talking at all about the certainty level or the details. Every conversation about which theory is more parsimonious with observations should be a comparison of probability values, not Boolean truth values. You would not believe the number of practicing astrophysics Ph.Ds who tell me that dark matter can not be black holes because microlensing, but can only cite support for tiny black hole mass ranges which nobody actually is suggesting comprise any substantial fraction of them.

    We need a comprehensive database of theories, the evidence supporting those theories, the strength of the evidence, the strength of the resulting theories, tied together all up and down the different observations. Back in the 1980s, artificial intelligence researchers used to call such databases and the programs which showed their consistency or lack thereof (and sometimes could suggest ways to improve consistency) “truth maintenance systems.” Has there ever been a project to build such projects for physics?

  5. John Duffield says:

    You don’t really have any problems in cosmology. Not compared to HEP.

    By the way, I noticed this:

    “Inflation has had some successes, but we are very far indeed from knowing that it happened”

    You must know it happened. You know the universe has always been expanding. You know about gravitational time dilation and how the universe can be likened to a black hole. So you must know that to gedanken observers at the time and to real observers at a later epoch, the initial expansion must have looked really really fast.

  6. Eric Habegger says:

    “So you must know that to gedanken observers at the time and to real observers at a later epoch, the initial expansion must have looked really really fast.”

    There is no physical evidence that time flow was the same in the early universe as it is now. What you are doing is assuming inflation is true and then inventing a fact to support it. If the energy density of the quantum vacuum is not constant during the cosmic expansion then there is every reason to believe that time flow sped up in the compressed phase. Like some other commenter said here a while back (and I’m paraphrasing) “Perhaps time contracts just like space does”.

    Now that’s a beautiful idea. And it even makes Gedanken sense.

  7. Stephen says:

    Scientific fields come and go. If I may offer my own two cents: High-energy particle physics, as most people realize, will enter a major decline due to the particle desert and non-discovery of supersymmetry. The really hot and interesting topics of the coming decades, I think, will be quantum computing, quantum biology, exoplanetary astronomy, and (gasp!) a revival of parapsychology due to the discovery of neuronal correlates of presentiment and telepathy in the vast majority of EEG/fMRI investigations.

    I don’t think cosmology will go into decline any time soon. You guys have your work cut out for you in terms of dark matter/energy and inflation. But poor particle physicists … 🙁

  8. Meh says:

    fight the power. right on.

    The problem is that there are so many snark laden jerks out there who have no idea how to present their disagreement with a successful idea without completely insulting the person(s) they disagree with…you bunch of ignorant asses! Am I the only one who sees these things?! 😀

    Every STEM major should be required to take a class called “principles of disagreement” or “intro to admitting fault”.

    Look at the interaction between Einstein and Bohr; they heavily disagreed on some things but never became complete assholes about it. Or the fact that Feynman and Carl Sagan probably smoked a joint together on more than 1 occasion. The problem we face in this new era is an inability to respectfully disagree…you dumb bastards.

  9. LadyAtheist says:

    I work at Ball State, so you can guess where I stand!

  10. John Duffield says:

    Eric: time doesn’t literally “flow”. But clocks do run slower where gravitational potential is lower, which is where energy density is higher. If you had a light-clock fixed to the surface of the Earth and you somehow made the Earth smaller whilst keeping its mass-energy the same, you’d find the clock runs slower. The “coordinate speed of light” reduces, so the clock rate reduces. Shrink the Earth down to a black hole, and at the event horizon, the clock rate is zero. It’s the same kind of thing when you wind back the universe. Only the expansion of the universe right now isn’t limited to the speed of light, so it’s only reasonable to assume that the expansion of the early universe wasn’t either. So if it was expanding whilst the clock rate was zero, the expansion looks instantaneous. That’s plain-vanilla general relativity for you.

    Sounds a lot like inflation to me.

    Stephen: I think high-energy particle physics is going to undergo a renaissance myself. Exciting times and all that.

  11. Iain says:

    Hi Sean, speaking as one of those young cosmologists – well, youngish – I’ve seen these causes for alarm for my entire career. The acceptance amongst young cosmologists of the standard paradigm, and the willingness to immediately ascribe direct physical significance to a phenomenological observable, has concerned me since at least 2004 when I first became aware of it.

    The thing is that I can see two reasons for this.

    Firstly, that people are so heavily specialised that it is difficult to understand the theoretical underpinnings of cosmology while also being an expert on parameter estimations or hands-on observation, which given the current (laudable) drive towards ever-greater surveys has unfortunately had the knock-on effect of reducing possibilities for those of us who, while at a couple of removes from the data, are perhaps more aware of those underpinnings.

    Secondly, and this may have more direct relevance to Peter’s post, it simply isn’t worth the risk to your career to speak up too loudly or to challenge established professors. Certainly this can be done, and if it’s done well it can do your career absolutely no end of good, but the risk is very high and unless speaking from a position of absolute strength the impression I’ve had is that most people will not attempt it. I’m not trying to imply that within the community dissenting voices will be immediately chopped down – that would clearly be an inaccurate statement – but more that an attempt to research in non-standard fields makes it increasingly unlikely that one can secure any additional funding, whether that is direct from funding agencies, or from a professor who is also nervous about his future funding.

    The unfortunate upshot of this is that a large number of people either specialise themselves into the mainstream and are either not pursuing, or are not *capable* of pursuing, alternative approaches; or that the vast majority of the people who take a gamble are forced out of the field.

    Fortunately for the future of cosmology I can think of a number of established researchers — with tenure or at the least tenure-track positions — who are well-positioned to challenge the generally-accepted paradigms. Unfortunately, however, I do feel that their numbers are overwhelmed by researchers who are not so well-positioned. This isn’t meant to sound like an attack on the latter by any means, merely that I feel we have become somewhat unbalanced, and I am concerned that the imbalance is only going to increase over time.

  12. Did you intentionally change feed settings so only the first few sentences are available in the feed reader?

  13. Sean Carroll says:

    Not intentionally, no … I’ll check it out.

  14. “The question is, how to encourage a healthy amount of rebellion (whatever that might be) in times when a successful paradigm is firmly entrenched?”

    First, one must answer the question if rebellion should be encouraged at all. If the paradigm is successful because it works, then encouraging rebellion is not much different than creationists saying “teach the controversy”. Rebellion for its own sake is wrong.

    If one thinks that the paradigm is unjustly entrenched, then the usual stuff works: gather more data, be careful in your analysis, have a theory which makes quantitative, falsifiable predictions. Proof: Despite Rocky Kolb, most cosmologists now believe that Omega_m is substantially less than 1.

  15. “speaking as one of those young cosmologists – well, youngish”

    Iain, did we meet up north a bit more than two years ago?

  16. “You would not believe the number of practicing astrophysics Ph.Ds who tell me that dark matter can not be black holes because microlensing, but can only cite support for tiny black hole mass ranges which nobody actually is suggesting comprise any substantial fraction of them.”

    The most vocal advocate for black holes as dark matter, at least recently, has been Mike Hawkins. (Primordial black holes, so they are non-baryonic and hence there is no conflict with limits on baryon density from big-bang nucleosynthesis.) The mass range he suggested has definitely been ruled out by microlensing.

  17. “You must know it happened. You know the universe has always been expanding. You know about gravitational time dilation and how the universe can be likened to a black hole. So you must know that to gedanken observers at the time and to real observers at a later epoch, the initial expansion must have looked really really fast.”

    Some of this makes sense, some doesn’t, but nothing here has anything to do with inflation.

  18. Eric Habegger says:

    I think John has some basic conceptual confusion when he says the universe can be likened to a black hole. A black hole has extreme energy density at the center. Counterweighing that is extremely low energy density radiating outward from it per the usual with the lowest energy density in the space next to the center of mass. That low energy density surrounding the mass maintains conservation and symmetry laws for energy and time. At the very center where the singularity exists time should be moving extremely quickly. Outward from that time does that reverse and slows down to compensate. Density slowly increases outward from that point, which coincides with decreasing gravity, per the usual 1/r^2. That’s basic conceptual physics that any undergraduate should understand before even attempting the mathematical substructure.

    There is no space outside of the universe so it has nothing to with black holes. At the singularity of the big bang everything is moving fast, including time. There is nothing outside of it to compensate since there is no space, unlike gravity in the space surrounding the mass at the center of the black hole. I suggest John educates himself before he tries to tell everyone what they are doing wrong. The motto for phsycians applies here: “first do no harm”.

  19. Eric Habegger says:

    I guess I should add that nothing I’ve said has anything to do with the “event horizon”.

  20. Not exactly relevant, and not my opinion, but funny:

  21. Eric Habegger says:

    Yep, that’s funny and mostly true. And on the other extreme we have the individuals who, when being challenged, yell “Badges? Badges? I don’t need no stinkin’ Badges!”

    Well no, you shouldn’t have to have a stinkin’ badge. But one should do the do diligence required to educate oneself before interjecting oneself in the conversation. There are equivalents that a smart person can do by way of the equivalent of credentials. It sounds like this Alexander whats-his-name is going the “I don’t need no stinkin’ badges” route with absolutely no self education to make up for it. Yikes!

  22. Re: conventional wisdom in a mature field is that it’s usually right.

    I want to comment on the statement above and present the challenges I encountered myself. I am an independent researcher in QM and I had the fortune to come in contact with the wonderful ideas of Emile Grgin (a retired physicists who worked in the 70s at Yeshiva University in Peter Bergmann’s group ) who a few years ago discovered a fourth number system for QM besides the reals, complex, and quaternionic numbers. About last year, Grgin decided to upload his work on the archive and I enthusiastically endorsed his work for the QM section. The swift result was that his work was reclassified in the general section (which is known as “the land of layman’s fantasies”) and I lost my endorsing rights.

    I am currently heavily working on expanding on Grgin’s results which generalizes QM from the C* algebra formalism into the C* module formalism and has natural links with Connes’ non-commutative geometry and the Standard Model. In fact, Grgin discovered (independently of Connes) how to naturally incorporate the gauge degrees of freedom directly into QM formalism. Inspired by his work, I recently obtained a natural axiomatization of QM which this spring was uploaded in the archive only 11 days before a similar paper by Anton Kapustin who also got inspired by an old paper by Grgin from the 70s.

    So no matter the intrinsic value of your contribution, challenging the conventional wisdom is initially (very) bad for you. It takes real effort and energy for people to start taking you seriously.

    And to some extent this applies also to challenging wrong ideas supported by people more powerful than you. Case in point, Bell theorem “disprover” Joy Christian was until recently selling his “theory”. His claims were obviously wrong but I could not put my finger on it and I got interested to get to the bottom of it. So I did and I got a big black eye for it.

  23. Iain says:

    Iain, did we meet up north a bit more than two years ago?

    If by “north” you mean in Britain probably not, I’ve not been to a meeting in Britain for years. There is another Iain Brown in astronomy though – radio astronomy, I think. (There may also be more who spell their names a bit differently.) On the other hand, I have been to a couple of small meetings in Canada and I’ve been based in Oslo for the last three years, so if it was Scandinavia or Canada, it’s certainly possible.

  24. John Duffield says:

    Eric: take a look at the comments on this topic re a “frozen star” early universe with no space beyond it. And no point-singularity within it, or gravity either. Mull it over. Kick it around. If you prefer not to, well, I guess I’m just talking back to my elders.

  25. Thanks for this very interesting post. Some of the thoughts you and Peter Coles express are similar to ones that my CfA colleague Avi Loeb has written about in a series of essays, all available on the arXiv. The first one was “Taking “The Road Not Taken”: On the Benefits of Diversifying Your Academic Portfolio”:

    and there have been 3 others that at least touch on this topic:

    The final one starts with this paragraph, so you can see the relevance:

    “Too few theoretical astrophysicists are engaged in tasks that go beyond the refinement of details in a commonly accepted paradigm. It is far more straightforward today to work on these details than to review whether the paradigm itself is valid. While there is much work to be done in the analysis and interpretation of experimental data, the unfortunate by-product of the current state of affairs is that popular, mainstream paradigms within which data is interpreted are rarely challenged. Most cosmologists, for example, lay one brick of phenomenology at a time in support of the standard (inflation+Λ+Cold-Dark-Matter) cosmological model, resembling engineers that follow the blueprint of a global construction project, without pausing to question whether the architecture of the project makes sense when discrepancies between expectations and data are revealed.”

    The apparent trend for conservatism in cosmology isn’t the only thing that’s interesting to me here. Avi told me that the number of emails he received for one of these essays – I think the first – far exceeded the number he normally receives about papers. I also saw how well attended his talk was on this topic. I think part of the reason is a real lack of discussion of career planning and strategic thinking in astrophysics.