No more ripples?

Well, that’s the Ripples in the Cosmos meeting in Durham over and done with, and I’m back in Newcastle for a few days before moving on to Edinburgh next week. I’m not sure I’ll be able to blog much over the next few days because my internet connectivity will be a bit limited.

Anyway, the meeting was very exciting, as you can tell from the picture showing me (with the beard) and Brian Schmidt (with the Nobel Prize):


Yesterday it was my job to round off the meeting with some concluding remarks leading into a panel discussion. I have to admit that although the programme for the conference was clearly designed in order to generate provoke discussion, I was a little disappointed that so few people said anything controversial. I’ve long held that there are too many cosmologists willing to believe too much, and this was further evidence that the scepticism that is a necessary part of a healthy science has been replaced by widespread conformity, especially among the young; when I was a lad the students and postdocs were a lot more vocal at meetings than they are now. Perhaps this is characteristic of a change in culture of cosmology? To get a job nowadays it’s virtually essential to climb onto one of the big bandwagon projects, and to keep your place you have to toe the party line, refrain from rocking the boat, not speak out of turn, and avoid making ripples (That’s enough metaphors. Ed).

Anyway, I think there are still a great many things in modern cosmology we don’t understand at all, and I think a few more of the older generation should show the way by questioning things in public. In fact only got asked to do the concluding remarks because Jim Peebles was unable to come to the meeting. Jim’s an immensely distinguished physicist who has probably done more than any other living person to develop the standard cosmology, but he’s also never been afraid to play devil’s advocate. We need more like him, willing to articulate the doubts that too many of us feel the need to suppress.

It’s amazing how much progress we have made in cosmology over the last few decades, but we shouldn’t use that as an excuse to get complacent. Cosmology is about the biggest questions in science. That alone makes it an exciting subject to work in. It’s an adventure. And the last thing you want on an adventure is for the journey to be too comfortable.

96 Responses to “No more ripples?”

  1. Anton Garrett Says:

    I don’t understand why you are grumbling. If you enjoy speculation and seeing the line blurred between science and science fiction then cosmology is EXACTLY the part of physical science to be in (especially quantum cosmology).

  2. Lloyd Knox Says:

    I checked out the programme — there was really good stuff there that could have been the basis for stimulating discussion. Maybe everyone was just worn out after an exhausting week?

  3. Brian Schmidt Says:

    I at least had the excuse of being in the middle of a migraine when that picture was taken!

  4. Tamara Davis Says:

    It may be that the young ones have been academically born and bred with the standard cosmological model in place. With enormous data sets backing it up and no (or little) sign of deviation during their careers. They see little reason to be skeptical — even if we don’t know what dark matter and dark energy actually are.

    I was surprised at the number of times “measuring dark energy” was equated with “solving dark energy”. Measurement is not explanation, no matter how precise we get.

    • I think there are several reasons. Malcolm Longair, in an article in QJRAS (how I miss that journal!), wrote that when he started out, there were two-and-one-half facts in cosmology. When he wrote the article, there were nine. Now there are many more. Back then, alternative hypotheses couldn’t be shot down quite so easily. These days, they have to explain all the observational data we have, in addition to whatever the motivation is for putting them forward in the first place. There is just so much more to know, and so much more to calculate. There are only a handful of people in the world with the resources to analyze, say, the Planck data. The hurdle for putting forward an alternative theory is much higher.

      Also, some people (not all, but many) 25 years ago, if they could have had a glimpse of the future, probably would have been astounded that the cosmological constant is part of the standard cosmological model now. How “out there” do you want to be? People tend to get used to things. The Rolling Stones, say, used to be provocative; now they are part of the establishment. (Interestingly, they didn’t change, but society did. I once saw an interview with Mick Jagger when he was about 21 or so. He said he did what he liked, and then it was provocative, but that wasn’t the reason to do it (in contrast to, say, punk). He even said that in the future, he’d still be doing the same thing, but people would then say he was conservative. Obvious in hindsight, but he said it at the time.)

      Also, one needs to learn a lot just to get up to speed, and at least to some extent one needs to know the standard stuff before going beyond it.

      Another point is that it is more difficult to get a job in cosmology today, and more mainstream folks might be more successful.

  5. While I wasn’t at the meeting, spies tell me that there was some
    controversial topics, particularly in the “Galaxy formation challenges and bias” (cards on table – I was involved in the work presented by Annette Ferguson).

  6. Shantanu Says:

    what was the reaction after Kroupa’s talk (which looked most controversial)? are videos of this conference online? was there any discussion of the following issues (which should be of concern to every cosmologist)

    o no sign of dark matter particle even though first phase of LHC is over?
    o no clues about energy scale of inflation (or even if inflation is the right solution to horizon,flatness problem)?

    • Did you expect the LHC to have discovered dark matter by now? If so, on what grounds?

      Inflation is difficult to prove, although there is now more evidence for it than there once was. It used to be “there isn’t any better explanation”. While there might be better explanations for some of the problems it solves (if they are indeed problems), for others it is difficult. In my view, the isotropy problem (probably what you mean by the horizon problem) is difficult to explain without inflation. Yes, people have looked at a variable speed of light and so on, but such explanations seem rather ad hoc.

      • Shantanu Says:

        Philip, before LHC started most particle physicists were pretty confident that LHC would discover supersymmetry (which is the edifice for thermal cold dark matter aka “WIMP miracle”) in its first year itself. In fact many of them made bets to that effect. J. Distler has recently lost one to T. Dorigo and F. Wilczek has one with G. Lisi that SUSY will be discovered by 2015. but LHC
        has already out large swaths of parameter space favored earlier on.
        Also, the limit on WIMP-nucleon x-section from direct detection experiments is about 10^-43 cm^2 (which is same order as weak interaction scale and this is the text book argument as to why a WIMP is a cold dark matter candidate.)
        Of course this does not mean SUSY is ruled and given the no of free parameters it can never be ruled out, even if nothing is seen with a machine 100 times more luminous than LHC.
        But surely any astrophysicist who posits the existence of Cold dark matter in their day to day work (which is almost everyone except those who work on solar system or globular clusters) should be rattled by the fact we do not even know if dark matter is a fermion or a boson and there is no laboratory evidence for any proposed CDM candidate.

        Regarding inflation there are so many inflationary models (see
        for eg., it is not even clear to me at this moment whether inflation is a theory,paradigm or what . If you have so many models, by random chance one of them has to be right. but certainly that was not the original motivation for inflation.

      • You have a very particle-physics approach. Yes, some other candidates (primordial black holes, say, at least in certain mass ranges) have been ruled out, but that does not mean that WIMPs are the only possible dark-matter candidates.

        I don’t see anything wrong with “using” dark matter (more correctly, non-baryonic matter) in astrophysics or cosmology without knowing what it is. Should we deny evidence for its existence just because we don’t know what it is? There are many examples where the observation came first, the explanation later (for example, QSO spectra, or indeed helium lines in the Sun).

        How long did it take after Pauli postulated the neutrino until it was detected directly?

        Is MOND a better answer? Apart from the fact that Kroupa apparently didn’t convince anyone, even many MOND supporters accept the fact that non-baryonic matter is needed. (I don’t think that this weakens the MOND case; the universe can be more complex. If one explanation fits the data, fine, but if not, then a more detailed theory is needed.)

        Astronomers have often turned out to be right. I remember the time when many people didn’t even consider cosmological models other than the Einstein-de Sitter model, since they didn’t see how inflation could produce them. Peter (and George Ellis) wrote a book claiming that Omega is less than 1, which at the time was a radical idea. Those who trusted the observations turned out to be right in the end. That’s not a guarantee, of course, but my assessment of dark/non-baryonic matter is independent of whether particle-physics people understand it. (Remember, these are the same folks who are off by 120 orders of magnitude when calculating the cosmological constant!)

      • Shantanu Says:

        I am not supporting MOND, which has lot more problems than dark matter. In fact there is no laboratory evidence for deviations from Newtonian gravity at low accelerations.

        But certainly some of the most well motivated candidates such
        as WIMPs, axions and primordial black holes are ruled out over the favored parameter space.
        Some of the other DM candidates such as Q-balls, WIMPzillas, mirror matter etc are closer to science fiction.

        I think there should be a lot more effort to determine the nature of CDM from astronomical observations alone, even in addressing rudimentary questions such as whether it is a thermal relic or not, whether it is a fermion or a boson etc.

        From the programme of this meeting, I saw only 1 talk (by Medvedev) on this (from the simulation side) and in general
        one finds very few talks on this at any cosmology meeting.

        (for one exception see this nice talk by A. Peter at STSCI
        But maybe as you said most astronomers are happy with the existence of cold dark matter and don’t care about its actual nature.

      • I’m not sure how astronomical observations could determine the nature of non-baryonic matter. Of course, at one time people thought we would never know the constitution of the stars, but spectroscopy stepped in. However, the whole point of dark matter is that it is dark, and of WIMPs that they are (only) weakly (and, of course, gravitationally) interacting.

        We knew that the Sun was hot before we knew its energy source.

        As long as astrophysicists and cosmologists limit themselves to non-baryonic and/or dark matter as implied from astronomical observations, I don’t see any problem. It’s not really their job to learn what makes it up.

      • Shantanu Says:

        Philip, for some examples of what kind of differences you can get , you can look at P. Sikivie’s talk at the KITP meeting
        The phase space densities of dark matter is different depending on whether its a fermion or boson and if its a boson you could get B-E condensation (as Sikivie has argued in that talk) and the density distribution of dark matter is completely different from NFW. Also free-streaming length scale of WIMPs is different from an axion. so you get a difference in mass of the first structure which could potentially form,
        K. Freese and colleagues at Michigan have
        worked on dark matter stars (which happens if dark matter is made up of WIMPs)
        But of course these are just exceptions and most astronomers
        don’t care about some of the above or related work.
        In fact, no one (or almost no-one) who does CDM n-body simulations puts any of this microphysics.
        Everyone is just happy with positing cold dark matter haloes
        in their simulations (or if they are observers they are satisfied with fitting NFW models to observations or measuring concentration index).
        Ten or so years ago probably that would have been fine.
        But even though we live in the era of precision cosmology
        the only progress we have made since Zwicky’s discovery of dark matter is that it is cold and non-baryonic

      • As an observer, it is very hard to be satisfied with NFW fits to data.

        Rotation curve data contain a lot more information about the mass distribution of individual galaxies than NFW or Einasto or pseudo-isothermal or name-your-halo model. Using these models to describe the data is like using a parabola to describe the CMB power spectrum. Maybe you’d fit the first peak, but you’d miss a lot of information.

        Since most people seem to have no idea what I’m talking about, see ApJ, 609, 652. This is purely empirical and complete/y general. The dark matter distribution is known, yet people still insist on inconvenient oversimplifications like NFW.

      • “Since most people seem to have no idea what I’m talking about, see ApJ, 609, 652.”

        I just read it. I recommend that everyone in this thread read it. Anyone with a background in astrophysics should be able to understand the main arguments, though of course there is a huge amount of astronomy and astrophysics in the details (measuring rotation curves, photometry, stellar population synthesis etc).

        The references include a huge number of Dutch astronomers, many from Groningen. I’m surprised we didn’t meet when I was in Groningen.

        You make the point that even if MOND as a theory is wrong, the empirical data are still there. I think a similar point can be made about lambda-CDM: even if things like galaxy formation in numerical simulations look difficult, the values for the cosmological parameters, including the amount of dark matter, look rather firm. So the latter could be correct even if the former are not, just as the observations supporting MOND are correct even if MOND itself is wrong. (Martin Rees, in an article in the sorely missed QJRAS, made a similar point about “the big bang”: that the universe is expanding from a hot and dense state is well established, while things like the exact mechanism of inflation (and indeed whether it even occurred) are less certain. Just as discovering a new species of animal doesn’t mean that “biology books have to be rewritten”, it is possible in cosmology to have basic results which are robust and details which are less so.

      • Exactly, Phillip. No one here is suggesting the Big Bang didn’t happen, nor even that we don’t have to understand why large scale probes persistently return LCDM as the right set of parameters. Of course we do. The point has been made that this is a tall order for alternative ideas. Indeed. But the same goes for galaxy dynamics: we are still obliged to understand why the data do what they do. That is a much taller order than seems to be widely appreciated.

      • Are there any galaxy simulations detailed enough to be able to demonstrate Renzo’s rule? If so, do they demonstrate it? (I’m guessing “no” for the second question, but don’t know the answer to the first.)

      • I am not aware of any simulations that reproduce Renzo’s rule. Mostly, as Tom complains, people who work on such things seem eager to dismiss it rather than explain it. That isn’t to say the simulations haven’t improved. Some now seem to do a decent job of both the normalization of Tully-Fisher and the scale length of disks, albeit over a limited dynamic range in mass. They still tend put too much mass near the center, but that in itself isn’t too terrible. I am concerned that the successes they have are obtained with unrealistic feedback models. (Mass loadings of ~10 instead of observationally motivated < 0.5.) What's an order of magnitude or two among astronomers?

      • C. Saxton Says:

        It’s nice if some galaxy simulations reproduce Tully-Fisher relations, but is that decisive? I wouldn’t like to invest too heavily in feedbackology. It’s just about the worst kind of phenomenon for numerical investigation.

        Analytically, consider a 1D shock propagating through a smooth gaseous medium with optically thin radiative cooling (according to some prescribed function that depends on metallicity). Calculate the first three to ten complex eigenfrequencies of oscillation and thermal instability.

        Now, run a long hydro simulation of this test problem, and analyse the wrinkles and oscillations in Fourier power spectra. Do you match any of the true eigenvalues? Double the resolution. Double it again. Is there any sign of convergence? Some entire classes of numerical algorithms may be incapable of converging towards realism. The artefacts just change from one type to another. That’s if you’re perceptive enough to recognise them as artefacts. …. We might well guess that this is one of the reasons why superpowers occasionally need to test their deadly hardware in real life.

        Next, let’s consider a single supernova (analogous to a bomb) detonating in a fractal ISM, comprising multiple phases that span orders of magnitude in density and temperature. Never mind magnetic fields. Next, let’s consider hundreds of SN in a parsec-sized box (or blurry SPH blob) where all those phases have been averaged together. What’s going to happen? How about we make a whole galaxy out of these blurred-out abstractions? …. The leeway for fine-tuning and wishful thinking is greater than the scope of the original physical proposition.

        If you ask a defence physicist, he probably wouldn’t touch this category of problem with a barge-pole. If you ask an astrophysicist, he’ll probably say: “it must be OK, because all the other astrophysicists are doing it already.” Hmmmm….. 😦

      • “What’s an order of magnitude or two among astronomers?”

        A factor of about two-and-a-half or of about six-and-a-quarter? 🙂 (I’ll get me coat.)

    • telescoper Says:

      To answer the original question, I’m afraid most people just ignored Pavel Kroupa’s talk. It was poorly argued and full of grossly exaggerated claims. I think most people thought as I did that it wasn’t worth the effort arguing with him. In any case I wouldn’t have known where to start.

      • Did anyone here attend the “great debate” between Pavel Kroupa and Simon White a while back?

      • Well, I for one am certainly rattled that the particle physics evidence for SUSY looks so grim.

        We have spent a generation quite sure that the WIMP miracle would lead to the detection of said particles in experiments like CDMS. For the past 25 years, I have every 5 years heard the statement “In 5 years we’ll know what the dark matter is.” Hmm. All that has happened is that the direct detection experiments have excluded the parameter space that looked like a sure bet just a few years ago (just as they did with the first round of predictions decades ago).

        WIMPs need supersymmetry to exist. Anybody else remember when the Bs decay was the “golden test” for supersymmetry? But – D’oh! – it turns out to be perfectly standard model. SUSY only remains viable because, as Shantanu notes, we can make it arbitrarily complicated.

        We astronomers are happy as long as the dark matter is non-baryonic and dynamically cold (or at least not too hot). Well, at least as long as we don’t look too hard at the small scale data. So sure, the dark matter can always be something other than what we have been sure it was for a quarter century.

        As an astronomer, I am more rattled by the galaxy scale evidence. Why do the baryons alone suffice to predict the dynamics dominated by non-baryonic dark matter? I have been unable to come up with a viable explanation for this observed phenomenon. Most of the explanations I have heard offered boil down to “we’re sure CDM exists, so surely it will work out!”

        I think as scientists we should hold ourselves to rather higher standards.

      • A word in defence of Pavel. He did make science arguments against LCDM. The first was based on simulation evidence that tidal dwarf galaxies are baryon dominated and he argues that at least some of the Milky Way’s dwarfs are tidally produced. The fact that they fit on the Tully-Fisher relation with galaxies that are Dark Matter dominated he sees as evidence against the idea that any galaxy is exotic particle dominated and as positive evidence for baryon domination and MOND. I agree with the first conclusion but not the second. He also argued against the hierarchical LCDM picture on the basis the the MW and Andromeda dwarfs seem to occupy planes which he thinks is unnatural in a bottom-up galaxy formation model. Annette Ferguson also gave a very nice talk on the Andromeda dwarf plane although she was careful to say that it was too early to draw too strong conclusions on this.

        The main issue against the TGD part of Pavel’s talk was raised by Richard Bower’s question as to how his TGD simulations dealt with feedback. Now this might actually strengthen Pavel’s TGD argument since most feedback mechanisms are usually ad hoc ways to fix up LCDM or in this case to kill Pavel’s argument. But I am not an expert on any of this…

    • David Merritt Says:

      I didn’t attend the Durham meeting but I can relate the reaction to Kroupa’s (apparently very similar) talk at the “Probes of Dark Matter” meeting in Monterey. Kroupa had highlighted the many failures of the standard model to reproduce observed properties of the Local Group. During the panel discussion, a young observer told us that we could “ignore all Local Group data”. A young theorist said that it would be “5, 10 or 15 years” before the standard model would be able to make useful predictions and so we should stop trying. In other words: “Close your eyes and have faith”.

      As someone who has zero stake in either point of view, let me make an observation. Since the 1980s, people like Milgrom have been pointing out some striking regularities in the data — regularities that have (apparently) no explanations in the standard model. In the past, the discovery of such regularities has often (not always) pointed toward a breakdown of the standard model. (I’m thinking of things like the periodic table of elements; Mendel’s laws of inheritance; the constancy of the speed of light etc.) Surely there is a finite chance that we are on the brink of such a paradigm shift in cosmology; lord knows we are overdue. I think we owe people like Kroupa a respectful hearing, at the very least.

      • telescoper Says:

        He got such a hearing.

        I should make it clear that – as I said in my concluding remarks – I do think there are problems with the standard model on small scales. The problem with Pavel Kroupa’s approach is that he asserts the existence of “theorems” that are in fact no such thing, just hunches. I think the correct approach is to keep an open mind on difficult issues, not form dogmatic camps.

      • Beth Willman Says:

        I think I’m ” the young observer” on the panel you refer to, David. I didn’t claim that we could ignore all Local Group data. I even clarified my point at the end of the panel because I was frustrated that Pavel misquoted me as saying such. I did say that responsible comparison between observations and models must be very careful to consider issues of halo-to-halo scatter when drawing sweeping cosmological conclusions from single systems, and must be mindful of how uncertain we are about what baryons do on Fornax and smaller scales. I therefore think we will learn more about galaxy formation than we will about dark matter from Local Group observations in the next 5 years.

        Also, I don’t think it is fair to quote Manoj as saying “we should stop trying.”

      • Fortunately, I don’t know what Manoj said, so I can’t misquote him. But the spirit of “wait ten years and all problems will be solved” I first heard 15 years ago. So I share David’s frustration to hear that again fresh – and I do hear it repeated a lot, but many people. So it does seem like a fair characterization: “close your eyes and have faith.”

        It seems every generation has to figure things out for itself. Or not.

  7. “I was a little disappointed that so few people said anything controversial.”

    Wasn’t Tom there? 🙂

    • I was there. Now to be honest I didnt have much hope for the Panel discussion at the end of Ripples despite it being led by Telescoper himself! But it proved absolutely rivetting. Did I really hear panellist Nick Kaiser say that both Lambda and CDM are an “embarrassment”. Then shortly afterwards did Carlos Frenk really say that he was “getting sceptical about the existence of a CDM particle given the null results from the LHC and the intrinsic unnaturalness of the SUSY models that are left to occupy the Planck parameter space. Celine Boehm had given a nice talk on this unnaturalness on the first day of the conference. This has clearly moved at least CSF away from CDM and more towards WDM. Carlos going against CDM is something that I never thought I’d see!

  8. […] an interesting and provocative post, Peter Coles bemoans a generational shift among cosmologists: “When I was a lad the students […]

  9. Unfortunately I could not attend the conference in Durham, but I just downloaded the slides from Kroupas talk. It is available on the following webside:

    From those slides, it seems that Kroupa may like provocative ideas and strong statements, but he DOES have a point. Dwarf galaxies are indeed extremely strange (to say the least) if one tries to understand them within the LCDM-model, while their properties seem to be a more or less natural consequence of their nature if one considers them to be tidal dwarf galaxies in a universe without cold dark matter.

    That is apparently the main theme in Kroupas talk, and he has some arguments that cannot be dismissed easily. Consider for instance the fact that the local population of dwarf galaxies are apparently very prone to form rotating disks around major galaxies (i.e. the Milky Way and Andromeda). This property would be natural to a population of dwarf galaxies that formed within tidal arms after a close encounter between already existing major galaxies, while groups of DM-dominated dwarf galaxies that are accreted by a larger galaxy (the commonly suggested explanation for the rotating disks within the LCDM-model) are to extended to form disks of the observed thinness.

    Why wouldn’t that be interesting to people who (as I think rightfully) complain about to much conformism in cosmology?

    • telescoper Says:

      I see where you’re coming from…despite the pseudonym.

      Your email and IP address indicate that you are one of Pavel Kroupa’s collaborators on the galaxies work. I wonder why you chose to conceal that fact when you posted that comment. This kind of thing has happened before on this blog. I think it’s a very weird way to carry on.

      • I am sorry – I did not know that using a pseudonym would be a problem, since I have seen other people using pseudonyms before me on this blog (telescoper is one, for instance). It is true that I did NOT hide my IP address and even used an e-mail address that clearly spells out my name for submitting my comment. Thus, finding out who I am was more than just simple and I was well aware of that. You may see that as proof that I did not actually try to hide my identity. However, for dispelling your suspicions about me wanting to hide my identity, I would like to reveal my official name to everyone. My official name is Joerg Dabringhausen and I have no problems with making my above statements under that name as well.

        It is also true that I am collaborating with Pavel Kroupa on dwarf galaxies and I can tell from his slides that a part of his talk apparently summarized my own work on the topic. I was therefore quite delighted by your blog entry, which essentially states that cosmology would profit every now and then from ideas a bit off the trodden paths. However, the following statement by you (telescoper) was then a surprise to me:

        ” […] I’m afraid most people just ignored Pavel Kroupa’s talk. It was poorly argued and full of grossly exaggerated claims. I think most people thought as I did that it wasn’t worth the effort arguing with him. In any case I wouldn’t have known where to start.”

        I was also somewhat annoyed by that statement, since also some of my work is thereby dismissed as not even worth criticizing – by someone who said that he would like to see more controversy. My work is published in a refereed journal, which means that I have proof that more people than just Pavel Kroupa and me think that this is a worthy contribution to the field. You can find it here:

        In fact, most of Pavel Kroupas talk was apparently based on refereed works. Even if you do not like Pavel Kroupas way of giving talks, I would like to invite you to have a look at the articles Pavel Kroupa mentioned in his talk, if you are interested in finding controversial ideas that are at the same time reasonable enough to survive the refereeing process. There you will find plenty of those. At the same time, I would like to ask you to point out some papers on cosmology that contain controversial ideas which you would think worthwhile.

      • telescoper Says:

        I think that there should be more sensible, rational, scientific debate about things that are often taken for granted. That’s not the same as dogmatically adopting a contrary position.

        I think there are issues to do with LCDM on small scales which merit a great deal of further study, but that’s difficult because you’re dealing with highly non-linear dynamical systems which are so complicated that clean results are hard to come by. In my view the more fruitful approach is to concentrate on things where there’s more chance of a clear shot at a potential Achilles’ Heel. Among these I’d suggest: large-scale inhomogeneity and/or anisotropy (including, e.g., parity violation) and evidence for large-scale departures from GR.

        I also think the Dark Matter sector may well be much more complicated than we imagined. There was only one talk at Durham on “exotic” dark matter, for example, which I thought was a shame..

      • I have for a long time advocated that there should be more sensible, rational, scientific debate about things that are often taken for granted. In my experience, that is hard to obtain in practice. Attaining this goal is not helped by asserting that others are “dogmatically adopting a contrary position.” If you want to hear debate, you had better be prepared to hear things you don’t like.

        I have heard many times the assertion that galaxies are “highly non-linear dynamical systems which are so complicated that clean results are hard to come by.” Aye, that is what they *should* be in LCDM. Problem is, they aren’t. They are simple dynamical systems follow scaling relations so uniform they can be summarized by a single effective force law.

        If that does not confuse you, you are not thinking clearly.

      • telescoper Says:

        Even the three body problem in Newtonian gravity is insoluble. I’m therefore not clear what you mean by stating that a collection of millions of stars, possibly with an unknown component of dark matter, evolving from an unknown set of initial conditions, is “simple”…

        They may be simple at the level of phenomenological scaling laws, but that’s not the same thing as being understood in physical terms.

        And I can think of many examples of observational “facts” that have turned out to be nothing of the sort, just effects of observational selection or bias. Observers are just as liable as theorists to believe too much….

      • So the 3 body problem can’t be solved analytically. Are you then implying that Newton’s inverse square law provides no physical understanding of the solar system?

        If I know Newton’s Law, I can solve for the motions of objects in the solar system. I can apply it to binary stars, etc. I cannot apply it to galaxies with the same success. But I do know of a formula that does work. Does that teach us nothing?

      • telescoper Says:

        The Solar System is dominated by one central mass, so that’s a very poor example to compare with a system of stellar masses.

        And it seems quite reasonable to me to suggest that the reason you can’t apply Newton’s law is that you don’t know where all the matter is. Just like how new planets are discovered…

      • Yes, galactic dynamics is harder than the solar system. But you can measure the luminous mass distribution. You can numerically solve the Poisson equation for that observed distribution. And you can use that information to predict the dynamics.

        You do not need to know anything about the mass you can’t see to do this. How does this make sense?

      • “I was also somewhat annoyed by that statement, since also some of my work is thereby dismissed as not even worth criticizing”

        Actually, Peter criticized the presentation, not the content. If the presentation is not good, the content is difficult to judge. You’ve provided links to your own work, and I put links to a huge review article by two of the other commentators on this post, so interested readers know where to look. I’ve never met Stacy, but my impression from the literature is that his MOND work is indeed clearly argued and not dogmatic. Even if MOND is not the answer, one really needs to be familiar with the observations MOND people talk about in order to have a healthy debate, and I dare say that most conventional cosmologists are not. Note that these observations are usually at a scale which is too small for most cosmologists, to whom galaxy clusters are specks of dust. 🙂

        I wasn’t there so I can’t judge Pavel’s talk. Of course, one can do good research just be bad speaker (for a variety of reasons), or just have a bad day. (Research quality doesn’t necessarily correlate with other qualities; one can be good at research and be bad at collaborating (no-one was awarded a doctorate for working with Einstein, for instance).)

  10. “I have to admit that although the programme for the conference was clearly designed in order to generate provoke discussion, I was a little disappointed that so few people said anything controversial.”

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

  11. I also attended the great debate between P Kroupa and SDM White in Bonn a few years ago. At the time I thought Simon won, mostly because of the strength of the CMB argument. Pavel argued around the CMB on the basis of a model that included WDM and MOND. Now apart from the MOND bit, I thought the arguments that Pavel gave on the MW satellites issue and WDM werent that far from what Carlos Frenk was saying 6 months later. Although both will of course deny it.

    Another bit of sociology – why is Modified Gravity that includes DM deemed to be “respectable” whereas MOND that tries to explain flat rotation curves without DM deemed less so!

    Note again that I do not advocate LCDM, MOND or Modified Gravity – I am still trying to make GR plus the what you see is what you get approach work – despite everything!

    • “Another bit of sociology – why is Modified Gravity that includes DM deemed to be “respectable” whereas MOND that tries to explain flat rotation curves without DM deemed less so!”

      Because the latter is a bigger change. New species of animals and plants are still being discovered, but when that happens, no-one says “that means we need a new biology”. Like all analogies, it’s not precise, but in some sense dark matter is just some matter we didn’t know about before, so it’s a bit like discovering a new species. Interesting, especially if one has indirect evidence for its existence before the discovery, but not game-changing. MOND would be like discovering some form of life (terrestrial or extraterrestrial) not based on DNA.

      Whether one is respectable and the other not depends on one’s point of view.

      To me, I don’t see a problem in dark matter per se; why should everything be visible to us? I also don’t see a problem with non-baryonic matter per se; why should everything be made out of the same stuff we are? Some people see dark and/or non-baryonic matter as an “epicycle”, but I don’t follow this argument. Sure, one starts with a simple theory and refines it if necessary; that’s science. This isn’t an attempt to “save the appearances”. If enough evidence accumulates, change is accepted (witness the fact that a non-zero cosmological constant is standard; it wasn’t that long ago that it was sometimes difficult to get something published which even examined the consequences of a non-zero cosmological constant, even if it didn’t claim a detection).

      “Note again that I do not advocate LCDM, MOND or Modified Gravity – I am still trying to make GR plus the what you see is what you get approach work – despite everything!”

      But it’s become more difficult, right? It wasn’t that long ago that a Hubble constant of 30 would have solved many problems. I think this is an example of what Peter referred to. When there wasn’t a great amount of data, a simple new idea couldn’t be ruled out. Many observations are proportional to the product of Omega and H so a low H would allow a higher Omega. Today, however, H and Omega are constrained individually, as well as their product, and there are allowed combinations of Omega and lambda and so on, so any alternative idea has to be more complex so that it fits the observations as well as solving the problem it is intended to solve.

      • Let me also echo the sentiment that there needs to be more productive discussion, rather than rival camps. Having spent a couple of years of coffee breaks with Bob Sanders in Groningen, I have some idea of MOND, the motivation for it, the people who work on it etc. Most of them are quite knowledgeable about traditional astronomy and astrophysics, both theory and observation whereas many CDM folks are those who do simulations (or particle physics types) who are a bit further removed from traditional astronomy and astrophysics. My impression is that the problems the MOND folks point to are real and need an explanation, but that doesn’t necessarily mean that MOND is the best answer.

        For those who want to know more: (also at (free to read and not financed by the authors!)).

    • “Note again that I do not advocate LCDM, MOND or Modified Gravity – I am still trying to make GR plus the what you see is what you get approach work – despite everything!”

      What do you think about which claims: We find H=37.7±0.4?

  12. C. Saxton Says:

    I agree with the commentary, but it’s not just a problem in cosmology. This generational malaise afflicts many areas of physics and astrophysics. “Concordance” and consensus are now valued above diversity and novelty. There’s too much groupthink; too many yes-men. Often the wrong kinds of people are given the most secure jobs, and too early in life. The better kinds of youngsters are afraid to express (or develop) their scepticism and creativity. Escaping from self-perpetuating serfdom depends on patronage.

    A few powerful professors (with initially good intentions) win the enjoyable luck of devising top-down “strategies” and “prioritisations” for everyone else. Junior researchers become collectible and expendable pawns. The concentration of funding towards national and international mega-projects makes matters worse. Gigantic “consortia” are dedicated to churning out massively multi-authored papers; over and over and over again. Bibliometric inflation goes on forever. Personal paper-counts and citations depend on membership of those clubs. Nobody, however talented, can compete with that kind of organised efflux. The game is rigged. Everyone plays along. Timidity is rewarded.

    May I suggest a remedy? Give tenure to every person who publishes a single-author refereed paper. Automatically; regardless of cost. Without prejudice. If that blows the budget, then make your cuts according to rankings of single-author or two-author publications (allowing for student-mentor pairs).

    • “Give tenure to every person who publishes a single-author refereed paper.”

      Where do I apply? 😐

      • C. Saxton Says:

        I propose that it should be automatic at whichever institution you’ve listed as “affiliation” on your single-author paper. If there are several affiliations, take your pick. 🙂

      • My last single-author paper (in MNRAS) listed no affiliation. 😦 Or maybe in that case it defaults to all the places where I thought about some aspect of the stuff I wrote up, so I should be looking for a joint appointment in Hamburg, Jodrell Bank and Groningen! 🙂

  13. Interestingly, I don’t see Stacy’s comment in the RSS feed, but do see it here in the thread. Does anyone else have this problem? (Let’s call it the dark-feed problem.)

  14. “I also attended the great debate between P Kroupa and SDM White in Bonn a few years ago.”

    There was Thomas Crapper, who had a role in the development of the flush toilet; there is the astronomer Alan Heavens. I’m sure there are some fishermen named Fisher. But if Simon had been named, say, Charles or Carl, there would be an even better relationship between content and presenter.

    (There is actually another astronomer named Simon D. White, so he needs the “M” to be unique (at least in bibliographies).)

  15. John Peacock Says:

    Many years ago, Geoff Burbidge used to attend pretty well all cosmology conferences. On several occasions, he chewed me out for being (as he saw it) too unquestioningly conformist for a young person (as I then was). I tried to argue to him that his point of view was conditioned by his experience: he was fortunate enough to have participated in the revolution that brought nucleosynthesis into astronomy, but having once tasted the feeling of ripping up the rule-book, he mistakenly assumed that this is how it always went. So he spent the rest of his life as a big-bang denier, which was sad. Now I wouldn’t say that CDM is yet on as sound grounds as the hot big bang, but it seems that many people have a strangely Burbidgean view of it – they can’t accept even the possibility that something invented only 3 decades ago might actually be right. But I think it was Weinberg who said, “the trouble with cosmologists is not that they’re always having crazy ideas, but that they frequently fail to take the simplest predictions of their theories seriously enough”.

    • That might be the reason for Geoff’s behaviour; changing the paradigm worked once, so why not change it again? My experience with him was always friendly enough, though. I never met Hoyle, but many people say that, in his later years, he became very bitter. Then there are folks like Philip Morrison, who were once Steady Statesmen but then converted when the evidence for the big bang arrived. (The steady-state cosmology was a very good theory: it got a lot of predictions out of very basic principles. As a scientific theory, they were falsifiable, and indeed they were falsified. Many people today say that the CMB ruled out the steady-state theory. Actually, it doesn’t, though it has no natural explanation. (In the steady-state model, things are as they are because they were as they were, so there could have always been a CMB. Unlikely, perhaps, but not something which really rules out the steady-state theory.) What really ruled it out were radio-source counts, which showed that the population changes with time. Another thing which doesn’t rule it out but which seems strange (but, as far as I know, no-one at the time mentioned) is that in the steady-state model it is just coincidence that the Hubble time is of the same order of magnitude as the age of a galaxy, while in the big-bang scenario this is natural and fine-tuning is required to get a universe with an age substantially different from the Hubble time.)

      Of course, Landau remarked that cosmologists are often wrong but never in doubt. 😐

    • Tom Shanks Says:

      So is Carlos Frenk a Burbidgean for having doubts caused by the lack of LHC detection of a CDM particle?

      • telescoper Says:

        I think people are overstating the problem of non-detection of dark matter by the LHC. It’s true that this puts pressure on the simplest supersymmetric theories, but there’s still a huge parameter space.And I’ve always been sceptical of the general assumption that dark matter has to be a SUSY particle anyway…

      • Indeed, the non-minimal SUSY parameter space is huge. Vast. Perhaps endless. We can only exclude one DM candidate at a time, and are always free to make up another. In my experience, most people are unwilling to consider something as dire as MOND until we’ve exhausted all possible dark matter candidates. But is that even possible? If it isn’t, then is the concept of non-baryonic dark matter falsifiable? It is confirmable – no one would be more relieved than me if a credible detection were made. But if not? How long and hard do we look before having serious doubts that it is there at all?

      • What’s your money on?

  16. Regarding Kroupa’s talk, I find it strange to attack the form rather than the substance. He has for instance been advocating the existence of rotating disks of satellite galaxies for years and was not taken very seriously for it. Now that such configurations of satellites are indeed confirmed features of Local Group galaxies, it would be, for instance, fair to acknowledge that he was right after all, even if one disagrees with his interpretation of this observational fact. I am currently at a conference where we just had a MOND vs DM debate, and the atmosphere was one of respectful disagreement, or even sometimes agreement. Agreement that there is evidence for something behaving as a dissipationless dust fluid on the largest scales (see CMB 3rd peak, colliding galaxy clusters) but also that small-scales have a lot of problems and show regularities that have been predicted 30 years ago by Milgrom. Again, even if one disagrees with the interpretation of Milgrom’s formula, it is only fair to recognize that he has nailed a rather succesful phenomenological law for galaxies. Regarding interpretation, the MOND point of view is that the latter might be part of the fundamental lagrangian of nature, and that this does not necessarily preclude the dissipationless dust fluid behaviour on the largest scales, as Luc Blanchet for instance showed in his presentation. The point was then made that these “non-economical” MOND-inspired lagrangians have all sorts of instabilities etc., and are hence presumed guilty until they prove their innocence. And they very much lack elegance and first principles. Fair enough, but as they are motivated by observed phenomenology on galaxy scales, it is entirely acceptable to continue to search for one such formulation, like Bekenstein, Milgrom, Blanchet and others are doing, and maybe a consistent and elegant formulation could finally be found. We will only know by searching for it. It had been argued for 20 years that a relativistic formulation of MOND could never be formulated, but this was proven wrong. The proposed models are not very elegant (to say the least), but it is certainly not by not searching for one than a more elegant formulation could be found. The other point of view is that, of course, “galaxies are complicated”. But the point is that they are actually simple, and that’s precisely the problem. Of course it is not impossible in physics to have some regularity emerging from very complex physical processes, but until one can show how that would actually happen in the particle dark matter context, it is all natural to also be presumed guilty. At least until plain simple particle dark matter is indeed detected in the lab, which Milgrom said would of course make him admit defeat immediately. So, all in all, everyone is presumed guilty indeed, but that does not preclude a sane and cordial scientific debate. The slides of the conference I’m referring to are available at

  17. Time to bring this out again:

    The link is to one of Stacy’s MOND pages. I recommend that everyone, MOND sceptic, adherent or agnostic, spend a few days there and at the stuff linked to. A huge amount of information.

  18. After due reflection in the Ripples recommended Cellar Door restaurant (!), another theme from the conference springs to mind – Ryan Keenan gave an excellent talk on the Local Void, detected in 2MASS, 6dFGRS and SDSS z surveys out to 300h-1Mpc at the ~ 50 percent level via local luminosity function estimates. My student Joe Whitbourn more or less confirmed this result in his talk with a 40 percent under density detected out rto 150 h-1Mpc in the Southern Galactic Cap and a 15 percent under density in the North. These levels are getting on for explaining away the accelerating expansion as possibly being due to faster expansion rates locally due to these local underdensities.

    • If one takes the values allowed by the supernova date for Omega and lambda (from which one infers acceleration; no-one has ever “measured cosmic acceleration directly” in any meaningful sense), then this relatively local measurement indicates acceleration. However, one can fit the CMB power spectrum and rule out all models which are not accelerating. This is on a much larger scale, so even if one could explain the local acceleration by voids or whatever, one still has to fit the CMB data. Finally, if one believes in the inflationary prediction of a spatially flat universe (which had nothing to do with acceleration) and puts in the observed density according to Coles and Ellis, Shanks, Carlberg etc, i.e. an Omega of 0.3 or so, then one ends up with the same lambda as implied by the supernova stuff and, independently, by the CMB.

      • Problem is that I am unfortunately old enough to recall the CMB problems of the early 90s where ~10 ground based, independent, CMB experiments all agreed that dT/T at 1degree scale was going down below 10^-5, as expected in standard Omega_m=1 CDM model. Then along came COBE at a larger scales and reported dT/T 2x bigger than previously expected. Suddenly all the previous upper limits were turned into detections by discovering that previously discarded CMB “contaminants” were actually signal. This unfortunately has left me with the adage that 2 CMB experiments agreeing with each other are a necessary but not sufficient condition that the result is actually correct. This led me to spend a lot of time analysing WMAP and Planck data even at the raw time ordered data level to check these results. A quick summary of potential issues is given in my Ripples talk at

      • Following up the comment below(!), I also recall in the 1980s I was told that it was pointless to argue against the standard isothermal/ isocurvature model because of the strength of the gold standard evidence of the day which was the gamma=-1.8 power-law of the galaxy correlation function which clearly implied n=0 non-adiabatic initial conditions! Now hardly anyone thinks this observation is significant as the fashions have changed. Then I was told that I couldnt argue against standard Omega_m=1 CDM because of the aforementioned 1deg gold standard CMB experiments which turned out to be systematically flawed. Now…… anyway you can see where I am coming from! As you yourself quoted – frequently wrong but never in doubt – particularly about the hierarchy!

      • Actuallly the followed up comment ended above!

      • I can see where you are coming from but where are you going? 🙂

        Certainly goofs like in the past should make one wary, and indeed this happens too seldom. On the other hand, just because people goofed up in the past (or were intimidated by Rocky Kolb or David Schramm) doesn’t mean that we will never make any real progress. (See Dennis Overbye’s Lonely Hearts in the Cosmos (what a title!) for a story of Schramm intimidation, including the immortal criticism “you are thinking like an astronomer and not like a physicist”!) In other words, one needs to be aware of the “Burbidge effect” which John mentioned.

  19. Then there was Planck – superb results that generally support the LCDM model. Lensing results look particularly interesting. But there is the odd anomaly – for example the CMB multipoles measured by Planck below l<50 currently are lower than the standard model prediction at the few sigma level – to the point where they are not included in parameter fitting. This is also reflected in the lack of expected large scale anti-correlation in the CMB correlation function. The constraints on non-Gaussianity also appear to depend on the sky cut – probably foreground but if 2/3 of the sky is included rather than 1/3 then the histogram tails start to look more non-Gaussian. Then there is the continued existence of cold spots that are too cold etc and the continued sensitivity of the C_l to the Planck beam estimates. So if no smoking gun, then at least some room for new physics over LCDM!

    • Hi Tom. Keep up the good work.

      When you say “lack of expected large scale anti-correlation in the CMB correlation function,” are you referring to the fact that there is essentially zero power on angular scales > 60 degrees? I recall that from WMAP; I have not seen the same from Planck. (I have read some of those papers, but certainly not all.) This leaps out at you when plotting against theta rather than ell. Can someone comment on this in the Planck data? Is the same feature still there? What does it mean? Seems very strange for any flavor of scale-free theory.

      • According to Dominik Schwarz’s review of CMB Anomalies, Planck sees the same lack of anticorrelation as WMAP. Again you have to worry about post-hoc analysis of cosmic variance limited measurements but there still seems to be an issue here.

      • Indeed. Essentially, the WMAP anomalies (axis of evil, cold spot, lack of large-scale power etc) are more significant in the Planck data.

  20. Then the supernova Hubble diagram – a tale of two talks by Brian Schmidt and Mark Sullivan. Brian’s review was excellent. His take was that dust absorption could still produce systematics at a low level but in his view other significant effects such as SNIa evolution with redshift were ruled out. But in Mark Sullivan’s talk there seemed to be potential issues everywhere from different SNIa species inhabiting early and late-type galaxies to superluminous supernovae that although rare would frighten the amateur like me in terms of SNIA systematics. Now in interest of balance, Mark’s talk didnt give quite the crisis spin that I am implying but he certainly raised some issues!

    • Of course, if the “standard concordance model” is correct, then there is the firm prediction that at higher redshift objects should be brighter, not fainter, than in a non-accelerating model. This can’t be done with dust. Superluminous supernova would have the opposite effect, as the main result is that the objects are fainter than “expected”.

      • Think you have got fainter and brighter wrong way round in 1st sentence but not in 2nd? Anyway the main issue is whether SNIa are standard candles because if they’re not then all sorts of systematics can arise in the Hubble diagram. Mark Sullivan’s
        talk raised plenty of doubts on this point even over and above the usual maximum luminosity – decay time correlation. On its own the latter correlation means SNIa are not absolute standard candles and in my opinion at least this leaves them more open to evolutionary corrections than they would be otherwise. SNIa evolution plus the Local Void/Hole look to be the possible escape routes out of the 1 part in 10^100 fine tuning implied by the SNIa Hubble diagram result.

      • Sorry, more detail: At low redshifts, the Hubble diagram is linear, and the parameters other than the Hubble constant don’t matter. For the current concordance model, at moderate redshifts objects are fainter “than expected”. However, at even higher redshifts, they are brighter than in the fiducial comparison model. This is a) a firm prediction and b) obviously something like dust would have to increase the dimming effect with increasing redshift. Of course, evolutionary effects could mimic an arbitrary Hubble diagram, but based on the current model we have an explicit prediction of the behaviour at higher redshifts. Unless someone comes up with a plausible model using evolution or whatever to get the same behaviour, this brightening, if observed, would increase the confidence in the interpretation of the almost-flat accelerating universe. Of course, the behaviour of the Hubble diagram at large redshifts has been known for decades, so I would be sceptical of some other explanation which was put forward only after this behaviour was observed. The grey dust, IIRC, was a direct reaction to the fainter than expected SNIa. This doesn’t mean it is wrong, but it does seem rather ad-hoc. But it’s testable, and that’s what counts, and it’s a simple model. Any evolutionary or other model which has a dimming then a brightening seems even more ad-hoc, especially if no-one thought of it before 1998.

  21. Then there was the “tension” between Planck estimate of H0 (67) and distance scale estimates (72). Personally if I got H0 estimates this close I would be claiming that this was an observational and theoretical triumph. But my old sparring partner, Prof GP Efstathiou had a different view and went so far as to suggest that parallax distances to Galactic stars and Cepheids must be significantly in error since they disagreed with Planck’s estimate. Now Brian Schmidt had advocated “blind” analysis of future cosmological tests to prevent confirmation bias but this discussion seemed not to meet these high standards as H0 estimates changed by the second. But then who am I to complain about bias about H0?

    • Thanks, Tom. I totally agree with you about H0. Indeed, I did read enough of the Planck papers to see that the real constraint was closely approximated by Omega*h^3 = constant. If you plot that line, all the WMAP best fits fall along it. So I have a very hard time accepting that Planck’s H0=67 is somehow better than the all the direct measurements giving low 70s. What is so terrible if we insist on H0=73 (say) as a strong prior? Omega goes back down a bit; anything objectionable besides rabid allegiance to MCMC?

      Amazing we can even worry about this level of difference in H0, if indeed we should. I also worry about confirmation bias, with future measurements likely to find the “right” value. This has already happened in BBN, where D and Y track the CMB value with short time delays, while it seems to be widely presumed that Lithium must just be wrong. Or maybe that work is done by stellar astronomers who don’t pay such close attention to the CMB.

    • Back when the debate was between 40 and 80, this was mirrored in the debate about the time delay in the first-discovered gravitational lens system, 0957+561, with a long delay (meaning longer distance scale and hence lower Hubble constant) favoured by Press et al. and a short delay (meaning higher Hubble constant) favoured by Jaan Pelt and the Hamburg group. Paul Schechter cried out from the audience during a debate on this topic at the 1993 gravitational-lens conference in Liege “But they agree at 3 sigma!”

  22. There were many other interesting issues raised in the Ripples conference but personally I am going to have a lie down now!
    These conference reports come courtesy of hospitality at‎
    Note that other Durham restaurant brands are also available!

  23. […] shamelessly to rip off an interesting comment on a blog post by Sean Caroll which picked up on the theme I posted about a few days ago, namely my perception that the current generation of cosmologists seems rather reluctant to […]

  24. Peter Coles, alias telescoper,

    after the Durham meeting I flew directly to Prague for the FQMT13 conference, where I was until now with a full schedule: . The conference had 400+ physicists with a convincing and very successful synthesis between science and music at historical venues. There I gave two presentations, on Tuesday the same one as in Durham, and on Thursday a one-hour long public talk which was recorded. At this occasion I was handed the Silver Commemorative Medal of the Senate of the Czech Republic
    (for more info see ).

    Concerning your statement above “To answer the original question, I’m afraid most people just ignored Pavel Kroupa’s talk. It was poorly argued and full of grossly exaggerated claims. I think most people thought as I did that it wasn’t worth the effort arguing with him. In any case I wouldn’t have known where to start.”:

    I do not understand this comment nor its purpose. You knew that I was invited to give the presentation at the Durham “Ripples in the Cosmos” meeting months in advance. The arguments I am using in my presentations have all been published in an invited peer-reviewed review (2012,…29..395K) as well as in many peer-reviewed research papers, and so you could have prepared detailed counter arguments. At the meeting you never approached me, nor did you ask a question after my presentation. Instead, you now use your own blog to comment negatively on my presentation to the public without my presentation having been recorded such that a reader cannot check your allegations. On your blog you are trying to have a discussion which you could have had in Durham face to face.

    Phillip Helbig commented above “I wasn’t there so I can’t judge Pavel’s talk. Of course, one can do good research just be bad speaker (for a variety of reasons), or just have a bad day. (Research quality doesn’t necessarily correlate with other qualities; one can be good at research and be bad at collaborating (no-one was awarded a doctorate for working with Einstein, for instance).)”

    So, the blog entry by Peter Coles may, apparently, be taken to suggest I am not a good speaker. Congratulations Peter! My reply to this statement is that readers of this blog may rest assured that I was in top form giving very very similar presentations in Cologne, Cambridge, at the AAS topical meeting on “Probes of Dark Matter” meeting in Monterey, in Durham at the Ripples meeting, and in Prague last week. At all occasions (except at the Durham meeting) vivid discussions ensued. At the exception, Durham, I received one question. Barely anyone at first talked to me. None of the senior cosmologists present there (apart from Tom Shanks), discussed with me my observational-data-based criticisms of the cold or warm dark matter based cosmological models. At the conference dinner I sat next to a professor form the University of California, Davis, who most of the time turned her back towards me. However, during the social events and somewhat hidden from view an appreciable number of younger people discussed with me. I did not mind this state of affairs, but on the remark by a very senior cosmologist in front of a little crowd of other people “Pavel, I do not understand your logics”, of course I would tend to reply equally brusquely. All in all, I enjoyed and I liked the Durham meeting. It was cry well organised and went smoothly. But it did appear to me to be somewhat unusually subdued, according to my experience of the many times I have given this presentation. The discussions with Tom Shanks and quite a few other people were, however, interesting and thought provoking. My personal impression was that few wanted to be seen discussing with me, perhaps given the substantial number of cosmological luminaries present and career issues. The comment(s) by Peter Coles on this blog do give a hint.

    Concerning your statement: “The problem with Pavel Kroupa’s approach is that he asserts the existence of “theorems” that are in fact no such thing, just hunches”:

    I suggest looking up the definition of what “theorem” is in mathematics. I followed this precisely in proving the dual dwarf galaxy theorem. Try disproving it in a peer-review paper (I will not engage in an endless discussion on this blog).

    Concerning the notion that the LC/WDM model cannot be tested on scales <few Mpc: this is not correct because the non-linear structures that form are well understood as long as phase-space is sampled sufficiently well (e.g. Diemand, 2008a, b:;…680L..25D; see also the seminal papers by Klypin et al. and Moore et al. on the issue of the missing satellite problem). The substructures and the spatial distribution thereof within a dark matter host halo is rather well understood.

    Creating thin vast polar structures from such sub-structures is basically ruled out (Pawlowski, 2102:…80P ), while them being created from the accretion of ultra-thin dark matter filaments has been hypothesized in Sect.14 in…29..395K , where some of the standard cosmology arguments against the LCDM skeptic are presented in a dialogue.

    Of course our team is making sure that the tests we perform are robust and largely independent of baryonic processes. I would most definitely not stand up to the many audiences I have been addressing to make claims which are not substantiated by highly significant evidence. I could otherwise loose my reputation as a scientist. But I do not blog about the style of other people's presentations, as I consider this to be bad practice.

    To make this clear please: I am not an anti-dark matter freak. Most of my research time I worked assuming cold or warm dark matter defined our reality. Only, my work led me to ever deeper tests of this (valid) hypothesis, and the many tests we have been performing, also in collaboration with international teams, has lead to the firm conclusion that dissipationless cold or warm dark matter is not present. In reaching this conclusion I follow the exact scientific method.

    Pavel Kroupa

    • telescoper Says:

      Thank you for your comments. Just to correct something: I did not mention your talk at all in the blog post. I was asked in a subsequent comment about it, so I gave my opinion. Sorry if you don’t like it, but I stand by what I said.

      Also, I was asked at rather short notice to stand in to do the concluding remarks and chair the final panel discussion. Had I been asked to prepare a detailed response to your talk, I would have done so. But I wasn’t.

  25. Anton Garrett Says:

    Peter, I’ve followed this thread with great interest but have not commented because I consider myself unqualified to.

    Re MOND, could you give ready references for the modifications that have been proposed to Newtonian dynamics, that do not go too much into the cosmology? Is the form of the modification similar to that proposed for the “fifth force” about 20 years ago?

    • telescoper Says:


      The wikipedia page on MOND is a reasonable starting point which leads to more detailed references. In its simplest form it basically involves a modification of Newton’s second law (at very low acceleration), so is not really a fifth force but is related to some of the ideas about that.


      • Just to give credit where due, Milgrom was not the first (in the modern era) to propose a modification to GR/newtonian gravity
        to explain flat galactic rotation curves. That credit goes to joel Tohline, but it was published in an unrefereed conference proceedings in 1982, which probably is hard to get hold off (and in fact I couldn’t even find it on ADS).
        (This is not mentioned in the wikipedia article)

      • Tohline did indeed seriously point out that one might also consider modifying gravity rather than invoking dark matter. I’m sure others considered it around that time as well (Bob Sanders, for example). The important difference is that many of Milgrom’s predictions came true.

        As for timing, check the date on this fascinating read:

      • telescoper Says:

        I’ve never read that paper. Can you suggest where it might be found?

      • I remember seeing Tohline’s contribution a long time ago, but am not entirely sure which one Shantanu is referring to. My best guess (that has at least an ADS abstract) is from the classic IAU 100. If memory serves, there is also a fascinating discussion after the talk of Kalnajs in that volume (back when the questions and responses were published along with the talks).

      • telescoper Says:

        We probably have that in our archive. If I find it I’ll scan it and post it on here…

      • “back when the questions and responses were published along with the talks”.

        Yes. Made proceedings much more valuable, especially as historical documents.

  26. Shantanu Says:

    Yes, I think its the one Stacy is referring to.(for some reason it is listed as 1983 in ADS which is why I missed it, although looks like the proceedings was published in 1982). There of course maybe another article by him around that time.

  27. […] Galaxy Scales" and in Durham at the "Ripples in the Cosmos" conference. The latter talk resulted in quite a discussion on Peter Coles' (aka Telescoper) blog "In the Dark", following his criticism of Pavel's talk as being "poorly argued and full of grossly […]

  28. Hi, not quite on topic, but your blog is “about the universe, and all that surrounds it.” I wonder if you have written about these ‘surrounds’ or if there is a particular blog category for them.

  29. […] and in Durham at the “Ripples in the Cosmos” conference. The latter talk resulted in quite a discussion on Peter Coles’ (aka Telescoper) blog “In the Dark”, following his criticism of Pavel’s talk as being “poorly […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: