48. The Weizmann Experience: discussions on the future of cosmology

Together with Francoise Combes, who was recently appointed as a professor in the most prestigeous institution in France, Le College de France, and Benoit Famaey, who is an expert on Milgromian dynamics and its deeper foundations (e.g. Famaey & McGaugh 2012), we were invited by Mordehai (Moti) Milgrom to spend a whole week at the Department of Particle Physics and Astrophysics in the Weizmann Institute in Rehovot, Israel. A link to the video (dubbed in English) of the inaugural lecture given by Francoise Combes for her new chair and the introduction by Serge Haroche (Nobel Prize 2012 in physics) is available here (alternatives to the dark matter approach are explicitly mentioned by both).

I met Benoit at Frankfurt airport in the very early morning (he was heading in some random direction) since we had booked the same Lufthansa flight to Tel Aviv. We arrived on Sunday, March 6th, and met Moti at his office in the late afternoon.


In the entrance hall of the Department. From left to right: Einstein’s field equation without Lambda, Francoise Combes, Mordehai Milgrom, Pavel Kroupa and Benoit Famaey.

Coming to know the place and first discussions

I am very impressed by the size and beautiful campus of the whole Weizmann Institut, and how pleasant the entire ambiente is.


Chairs and a pond in front of the Department.

The people are very friendly and  helpful. And interested. I was staying at the spacious and luxurious San Martin Faculty Clubhouse. At night the various buildings and park areas in the Weizmann Institute are illuminated beautifully, with warm lights setting accents and emphasizing a welcoming atmosphere.

The highly-ranked  Weizmann Institute consists of many departments of various natural sciences and seems to be perfectly created for academic pursuit, including leisure areas. Its success in the pursuit of basic research in the natural and exact sciences and in acquiring funding is evident through the architecture, spaciousness, and general design.

There was no planned agenda for us, apart that Benoit was to give a talk on Wednesday, 9th of March, at 11:15, and for Francoise Combes to give a departmental colloquium on Thursday, 10th of March at 11:15. In between these talks we could do either nothing and hang about enjoying the sunshine and exquisite weather and pool, or engage in intense discussions. Perhaps due to the ambiente and of course our comparable research interests, we largely chose the latter.

On Monday, 7th of March, we had a very relaxed day, meeting with Moti at the Department in the late morning and spending our time debating. Typical discussion points (largely between Francoise, Benoit and myself) throughout the visit were the local major underdensity and its possible implications on the value of the cosmological Lambda, the underlying theory of MOND and whether it is due to a “dark” fluid which behaves like dark matter on large scales (e.g. Luc Blanchet’s dipoles and Justin Khoury’s condensate)

Given that Lambda was missing in the equation displayed in the entrance hall of the Department (see first photo above), we began to discuss it. And this is where the “local” underdensity now plays a possibly important role, see this figure from Kroupa (2015),


The underdensity is significant, according to the shown data, and may challenge any cosmological model. From Kroupa (2015).

and in contrast the very recent work by Whitbourn & Shanks where the authors explicitly state agreement with the previous survey by Kennen et al. (2014). The independent finding by Karachentsev (2012) on the local 50 Mpc scale appears to naturally continue the trend evident from the Kennan et al. data (see the figure on the left), IF one assumes the same baryonic to dark-matter ratio as at larger distances. The actually measured stellar density remains similar to the Keenan et al. value at small distance. So the baryonic density (assuming the gas to star ratio and the contribution by dwarf galaxies to remain unchanged out to distances of 800 Mpc [redshift of 0.2]) then within 300 Mpc there is at least a decrease in the baryonic density by factor of two. Conversely, taking Karachentsev’s measurement, we would see a disappearance of dark matter nearby to us since the stellar density remains similar to the Kennen measurement within 150 Mpc while the dark matter density decreases further. So the measurements appear to imply the following picture: within 400 Mpc the luminous (and thus baryonic) matter density decreases significantly by a factor of two. At the same time, the ratio of dark matter to baryonic matter decreases even more. Both findings violate the cosmological principle.

The work by David Wiltshire (his lecture notes) and Thomas Buchert already indicates that inhomogeneities could possibly make the Universe appear to an observer situated within such an underdensity as if it’s expansion is accelerating, although in truth it is not. That is, the inhomogeneities appear to be of the correct magnitude to eliminate the need for Lambda, Lambda (dark energy) merely being an apparent effect mis-interpreted by the supernova type 1a data. The reason lies in that a distant object’s observed redshift depends in reality on the exact paths the photons travel in a universe which consists of time-changing voids and over-densities, and this is a different redshift computed assuming a homogeneous and isotropic expanding Universe.

But we need more detailed calculations taking into account the constraints from the observed under-density shown in the figure to be assured that Lamba=0. It is certainly true that Lambda=0 may be more in line with theoretical ideas than the very small value deduced to explain an apparently accelerating Universe, because it is actually predicted, from quantum field theoretical calculations of the vacuum (for details see e.g. Padilla 2015), to have a value some 60 to 120 orders of magnitude larger. It should be emphasized, though, that “MOND likes Lambda“, in the words of Moti. The reason is that the Lambda derived from astronomical observations (e.g. from supernovae of type 1a observations) and Milgrom’s constant a_0 appear to be naturally related, and MOND may be derivable from vacuum processes (Milgrom 1999).

Within about 300 Mpc, where we can say that we have the best measurements, the Universe is nicely consistent with MOND. The mass-to-light ratios of galaxy groups are less than 10 (Milgrom 1998 and Milgrom 2002), i.e. there is only baryonic matter. The observationally inferred increased density of baryonic matter at distances larger than 300 Mpc would then perhaps be due to cosmological models being inappropriate, i.e. that the currently used red-shift–distance relation may be wrong.

We also debated galaxy evolution, the fraction of elliptical galaxies and the redshift dependence of this fraction. Notably, fig.7 in Conselice (2012)  shows that the observed fraction of massive galaxies does not evolve although the LCDM model predicts a strong evolution due to merging. This is consistent with the independent finding by Sachdeva & Saha (2016) that mergers are not a driving mechanism for galaxy evolution, and this is in turn consistent with the independent findings reached by Lena et al. (2014)  on the same issue.

We further talked about how LCDM is faring on large, intermediate and small  scales, how stellar populations change with physical conditions, the variation of the IMF, as well as political topics. The discussions were far from reaching consensus, we had different views and data sets we could quote on various problems, and time flew by such that we barely noticed.

However, Moti managed to drag us away from his Department, and showed us around the Weizmann institute. An particular station was the famous landmark tower which once housed the Koffler Accelerator and which now houses, in its “bubble”,


The tower which housed the Koffler Accelerator and which now houses a conference room (in its “bubble”) and the Martin S. Kraar Observatory.

a conference room and also the Martin S. Kraar     observatory which is also used in international top-level    research projects. The director of the observatory, Ilan   Manulis, kindly explained to us in much detail its   functionality and design for full remote-observations   without human interference.


Viewing the lands from the top of the Koffler Accelerator Building. From left to right: Benoit Famaey, Francoise Combes and Mordehai Milgrom.


Part of the Weizmann Institute as viewed from the top of the Koffler Accelerator Building.


The Group at the Koffler Accelerator. From right to left: Benoit Famaey, Francoise Combes, Mordehai Milgrom and Pavel Kroupa

On this Monday Moti took us to lunch at the Lebanese restaurant Petra located in Nes-Ziona, a town 5 minutes drive from the Weizmann Institute. The Lebanese cuisine was fabulous, and I ate far too much.

A diversion to history

And, on Tuesday, 8th of March, Moti and his wife Ivon took us on a drive-around nearby Israel. This trip, involved about 4 hours of driving by Moti, and while driving we discussed, amongst other topics, the new study by Papastergis et al. (2016) in which they use 97 gas-dominated galaxies from the ALFALFA 21cm survey to construct their estimate of the baryonic Tully-Fisher relation showing excellent agreement with the expectations from Milgromian dynamics.

The drive was incredible, as we saw places with many thousands of years of history dating back to the Caananite peoples. It is this land which took the central role in the evolution of the Mediteranean-Sea-engulfing Roman Empire to a Christian empire. It contains the scars of the episodes of the invasion by a newer religion of christian lands, christian reconquest, and reconquest by the newer religion, till the foundation of Israel, issues which remain current to this day.

We visited Caesarea:


The author amongst the ruins of Caesarea. “What was the fate of Caesarea’s inhabitants when it fell to the Mamluks?”


Caesarea, once a thriving port for many centuries, from where Paulus was imprissioned and sent to Rome for his hearing at the emperor’s court, was wiped out in the 13th century.


The Group in front of the Roman ampitheater in windy Caesarea, nearly but not quite ready. From right to left: Mordehai Milgrom, Francoise Combes, Benoit Famaey, Pavel Kroupa.

The thriving thousand-year old medieval city of Caesarea, named by King Herod after Octavian (i.e. Augustus Caesar) and which was once the main port in his kingdom, was finally obliterated from existence after a siege by a Mamluk army in the thirteenth century.

Acre: the chief port in Palestine  during the crusader epoch still boasting major remains of the huge crusader’s fortress:

Acre: the remains of the Crusader port.

Acre: the remains of the Crusader port.


Acre, once a blossoming port and a gate-way to the holy lands for christian pilgrims.

After a wonderful dinner at the seashore between Tel Aviv and old Jaffa at the restaurant Manta Ray, where some action happened just before we arrived judging from the large number of police and other forces around, we visited very beautiful Old Jaffa:

Old Jaffa, which dates back to a history of 4000 years and where alrady the Egyptian empire stationed a garrison.

Old Jaffa, which dates back to a history of 4000 years and where alrady the Egyptian empire stationed a garrison.


Old Jaffa.

The restoration of the archeological sites of   Caesarea, Acre and of Old Jaffa brings to mind   how  incredibly rich and beautiful the thousand   year old places are along the Mediterranean coast   throughout the middle East and northern Africa, if   upheld with the corresponding desire to show   this history.

Back to science

On Wednesday, 9th of March, we spend the whole day in discussions with staff of the Institute. It began with Benoit Famaey’s presentation on the latest numerical results of modelling the Sagittarius satellite galaxy and its stream in Milgromian dynamics by Strasbourg-PhD student Guillaume Thomas. Natural solutions appear to emerge and this will, once published, clearly add spice to the discussions, given that the only solutions available in LCDM by Law & Majewski (2010) are unnatural in that the dark matter halo of the Milky Way needs to be oblate at right angle to the Milky Way, a solution which poses severe dynamical instabilities for the Milky Way disk. Notably, this polar oblate dark matter halo of the Milky Way alignes with the vast-polar structure (the VPOS) of all satellite galaxies, young halo globular clusters and stellar and gas streams.

In these discussions with the staff members during the aftenoon, we dealt with supernova rates and explosions and types in different galaxies, the relevance to the variation of the IMF in various environments (e.g. metal-poor dwarf galaxies vs metal-rich massive galaxies and the dependency of the IMF on density and metallicity), and cosmological problems such as the local massive under-density mentioned above.

An important point I tried to emphasize repeatedly is that if Milgromian dynamics is the correct description of galactic dynamics, then we must keep an open mind concerning the possibility that all of cosmological theory may have to be rewritten and the large-redshift data may need to be reinterpreted in terms of different redshift–distance and redshift–age relations.

In the evening of Wednesday I tried out the swimming pool on campus, and their sauna as well. I had access to this swimming pool by staying in The San Martin Faculty Clubhouse and the Hermann Mayer Campus Guesthouse – Maison de France. I must admit, that the day was near to being perfect with the sunshine and a closing dinner with Francoise and Benoit again in our meanwhile standard kosher restaurant (Cafe Mada) nearby the San Martin guest house.

On Thursday, 10th of March, Francoise Combes gave her interdepartmental presentation on “The Molecular Universe” which was well visited, and afterwards we went together with some staff of the Weizmann Institute for lunch at Cafe Mada, where a lively and very entertaining discussion ensued on religeos questions. In the late afternoon we joined the Whisky lounge, in which anyone traveling back to Rehovot from abroad can bring a duty-free bottle of Whisky to and donate it to this lounge.

The Local Group of galaxies is highly symmetrical, with all non-satellite dwarf galaxies lying in two planes symmetrically and equidistantly situated around the axis joining the Milky Way and Andromeda. From Pawlowski et al. (2013).

The Local Group of galaxies is highly symmetrical, with all non-satellite dwarf galaxies lying in two planes symmetrically and equidistantly situated around the axis joining the Milky Way and Andromeda. From Pawlowski et al. (2013).

Young researchers meet every Thursday (remember, this is in Israel the end of the week) to sip Whisky and thereby to elaborate on various problems, such as in our case on the local underdensity, or how the two critical constraints we have from the highly organized structure of the Local Group of galaxies and the CMB together constrain the cosmological model.

An interesting statement made was that while one needs about ten LCDM Universes to get one Bullet cluster (Kraljic & Sarkar 2015), an infinite number of LCDM Universes will not give a single Local Group with its symmetries.

At least these are some of the questions we discussed while there on this Thursday. We were also impressed by all the connections of this Department with Princeton, Caltech and Harvard.

Friday and Saturday

Shops begin to close down and it becomes a challenge to find food and Francoise left for France. In the morning I went for a swim and sauna, and for luch Benoit and myself had to go out of the Weizmann Institute (exit Main Gate and turn left) to find a sandwich place.


The Basha Bar in Tel Aviv.

After some work and then in the evening and at about 18:00 we decided to take a taxi to Tel Aviv. We arrived at the Basha Bar by about 18:30 and stayed for three hours (see photo).


The Basha Bar, enjoying a three-hour shisha smoke and many Tuborg beers.

On Saturday, the kosher breakfast in the guest    house was as excellent as ever, but it was    interesting for me to note that neither the   toaster nor the coffee machine were to be  used,  while the water boiler was   on so we  could still have hot Turkish coffee (which we  also drink in Bohemia, by the way, so not   much      new for me here). Nearly everything is closed. Benoit   and myself met for lunch and walked outside the Main   Gate turning right, over the bridge to reach the   Science Park finding bistro Cezar for lunch.

In the evening Moti picked us up for a dinner at his home with Ivon, where we had a long discussion also on the dynamic situation in Germany, Europe and the future.

At the home of Moti in Rehovot.

At the home of Moti in Rehovot. From right to left: Moti, Benoit and the author.

Final comments

Benoit and myself stayed on until Monday, joining the astrophysics journal club which serves lunch at the Department on Sunday. I spent most of the afternoon discussing with Boaz Katz how star clusters may be relevant for type 1a supernovae. In the evening of Monday Benoit and I went again to Cafe Mada for a final dinner and drinks. On Monday, 14.03., we flew out around 16:00, taking a taxi to the Tel Aviv airport at 13:00 from the Department. We shared the same flight back. Again the 4+ hour long Lufthansa stretch without personal-screen-based entertainment system! But, this gave Benoit and myself a chance to further discuss at length the above mentioned Khoury condensate and the Blanchet dipoles as models for galaxy-scale MOND and cosmology-scale dark-matter-like behaviour. But I note that these are not dark matter models. During pauses my thinking was that as the coastal line of Tel Aviv receded in the setting Sun we left a small fraction of the Levant and northernmost Africa, all once pat of the Roman Empire, at a level of civilisation mirrored by the clear, brllliantly lit vast and dynamic power- and resource-hungry central-European night with full autobahns, radiant towns and illuminated football fields in nearly every village. In Frankfurt our ways parted after a last small dinner in the train station, Benoit taking a bus to Strasbourg at about 21:30, and me starting my odessey to Bonn at the same time using the available train connections (German trains all too often run late, these days).

The visit was most memorable for all of us, and Benoit and myself agree that we would like to return. We did not reach any conclusions but we came to know many new people and perhaps helped to underscore the very seriousness of alternative concepts to dark matter and the many failures of the LCDM model.

In closing it is probably fair to say that Milgrom contributed the greatest advance on gravitational physics since Newton and Einstein.

In The Dark Matter Crisis by Pavel Kroupa and Marcel Pawlowski. A listing of contents of all contributions is available here.

5 thoughts on “48. The Weizmann Experience: discussions on the future of cosmology

  1. Pavel: I enjoyed your travelogue. Since you mention both my work and Moti Milgrom’s 1999 paper, I would like to make some comments on the cosmological constant problem.

    Firstly, note that my background is that of a theoretical physicist in gravity and cosmology. As such, I am keenly aware of different levels of explanation in theoretical physics. I note that the “quantum field theoretical calculations of the vacuum” which you refer to are crude estimates which do not take gravity into account. Quantum vacuum effects are understood and verified in flat space quantum field theory, giving rise to the Lamb shift and the Casimir effect, which involve the specific boundary conditions of atoms and conducting plates respectively in QED. From the point of view of general relativity, flat space field theory (classical or quantum) applies locally in an observer’s tangent space – as far as the quantum vacuum problem for the whole Universe is concerned, we are not going to understand the cosmological constant problem until we have a quantum theory of gravity. The 60-120 orders of magnitude by which crude flat space quantum field theoretical arguments are off just shows the gap between our established understanding and what quantum gravity must ultimately deliver.

    The big problem in quantum gravity is the separation of propagating from nonpropagating degrees of freedom, or alternatively phrased the isolation of the “true gravitational degrees of freedom” to quantize. The cosmological constant problem is intimately related to this, as it relates to the nonpropagating degrees of freedom. I believe that we are slowly making some progress in uncovering crucial pieces of the puzzle in understanding the gravitational degrees of freedom in quantum gravity; but it is a problem with a decades long history that is too long to summarize inside this box.

    MOND and the timescape cosmology are both phenomenological models (to varying extents) which deal with two different problems – galactic scale “dark matter” and “dark energy” respectively. While the timescape model is advanced to the extent that it makes a prediction which can be distinguished from the LCDM cosmology with the Euclid satellite – and many other predictions which are discussed in my lecture notes that you cite – it still requires a more rigorous mathematical formalism for relating local to statistical geometry. It does have a means for relating the local geometry of observers in bound structures to a statistical geometrical average of the inhomogeneous structure – this is phenomenological based on matching null cones. Although the resulting phenomenology (like MOND) is able to make a number of accurate predictions, a theoretical physicist wants a deeper level of rigour before accepting a theory – which is why neither MOND nor timescape are mainstream as yet. Phenomenological models are important – the Bohr atom being one such model – but steps analogous to the steps from the Bohr atom to quantum mechanics have to be taken at some point.

    I do believe that physicists who think about the fundamental issues, like Milgrom does, show the courage that real physical problems demand. However, we have to be realistic in weighing up the relative merits of ideas. In this vein I would point out that the suggestion that “MOND likes Lambda” is based on the crudest of analogies made in Milgrom’s paper, Phys. Lett. A 253 (1999) 273. In particular, “an observer with constant acceleration, a, in a de Sitter universe sees Unruh radiation with a temperature, T, proportional to sqrt(a^2 + Lambda/3) and T(a) – T(0) depends on a in the same way that MOND inertia does”. I along with many a theoretical physicist would question what relevance a crude analogy to an ideal uniformly accelerated quantum particle detector in a hypothetical universe containing a cosmological constant and no matter has to do with a phenomenological acceleration scale in a model dealing rotation curves of real baryonic matter in the actual universe.

    Curiously, I myself was led to a completely different “derivation” of the MOND scale, when thinking from first principles about the nature of inertia and Mach’s principle in relation to cosmological averages in my article Phys. Rev. D78 (2008) 084032. Mach’s principle – as broadly stated by Bondi in his text “Cosmology” (Cambridge Univ. Press, 1961) – is: “Local inertial frames are determined through the distributions of energy and momentum in the universe by some weighted average of the apparent motions”. It is thus intimately related to the averaging problem in cosmology, which is what my work and that of Thomas Buchert is all about – and any attempt to understand Mach’s principle without reference to the averaging problem misses its essence, I would argue. Anyway, to cut a long story short, phenomenologically in the timescape model one is naturally led to a “relative volume deceleration scale”. This volume deceleration scale varies with redshift as average structures in the universe evolve, but putting in the numbers required to match supernova distances and the like I found that for redshifts z < 0.25 the magnitude of the resulting relative deceleration scale gives a range that matches the MOND scale more closely than any rough order of magnitude estimates based on the Hubble constant that I have seen in various MOND papers.

    To my mind, this "derivation" is closer to the spirit of a real derivation because it is based on a relative deceleration scale which is an actual consequence of a cosmological model, and cosmological averages which deal with actual matter – not a crude analogy to an artificial situation such as that of Unruh radiation in an empty de Sitter Universe. However, personally I would not yet claim it as a derivation, since the relative deceleration scale I calculate is between the rulers and clocks of an ideal observer at a statistically average location and those of ideal observers on the boundary of the largest bound structures – galaxy clusters. I have not yet thought about the internal dynamics of clusters, which is an altogether different problem. So I see it as another clue in the puzzle, but until one understands things at the level of explanation of quantum mechanics as compared to the Bohr atom, then one must treat it with a grain of salt. But equally on the actual evidence MOND has no preference for a cosmological constant or not, as long as it is phenomenology.

    What is clear to anyone familiar with general relativity, is that (i) there is much about it that is not Newtonian; (ii) GR far from understood in the statistical many body regime of cosmological averages. Combining these facts we need to keep an open mind, be aware of phenomenological facts, but also think self-critically in seeking explanations.


  2. “… the underlying theory of MOND and whether it is due to a “dark fluid” which behaves like dark matter on large scales …”
    The following publication should not be overlooked:
    http://arxiv.org/abs/0804.1588 “Dark Fluid: Towards a unification of theories of galaxy rotation, Inflation and Dark Energy” by HongSheng Zhao & Baojiu Li, 2008
    Deviations from Newtonian-Einsteinian gravitational theory might be caused by wave interactions among alternate universe or some “dark fluid sloshing” that goes beyond conventional theories of physics.


  3. Why should the cosmological constant be nonzero? What is the meaning of negative gravitational energy? Does dark energy obey the equivalence principle? Does dark matter obey the equivalence principle? I conjecture that dark matter is the gravitational analogue of the Čerenkov effect.
    Photons and gluons travel (on average) at the speed of light but never escape from the boundary of the multiverse into the interior of the multiverse. Gravitations travel (on average) at the speed of light but sometimes escape from the boundary of the multiverse into the interior of the multiverse, This process of escape causes dark energy, dark matter, and inflation.
    The preceding scenario might be wrong, but 2 things are clear at this stage: (1) MOND is empirically valid (according to all tests so far) and (2) the successes of MOND require a new paradigm for the foundations of physics.


  4. Wird nun doch alles ganz anders?
    Meldung von Andreas Müller 22/4/2016:
    Der zentrale Bereich der Milchstraße beheimatet eine sehr alte Generation von Sternen, die völlig andere Bewegungsmuster aufweisen als die übrigen, jüngeren Sterne.
    Müssen wir nun nochmals von vorn anfangen.
    Mit freundlichen Grüßen
    Joachim Blechle


    • Die inneren Reginen von vielen normalen Galaxien bestehen aus alten komponenten, welche elliptischen Galaxien sehr aehneln. Diese Komponenten (die sogenannten “bulges”) beinhalten demnach alte Sterne, welche auf chaotischen Bahnen laufen, waehrend die juengeren Sterne aus der Gasscheibe der Galaxie entstehen und demnach auf beinahe Kreisbahnen laufen. Diese verschiedenen Komponenten einer normalen Galaxie sind recht gut verstanden. Ein bulge kann z.B. entstehen, wenn sich Galaxien sehr nahe kommen. Dieses war vor ca 10Gyr haeufiger der Fall. Oder er kann sich direkt waehrend der initiellen Entstehung der Galaxie bilden, oder aus eine dynamischen Instabilitaet der Scheibe. Dieser Befund, den Sie ansprechen, impliziert also keinen Neuanfang in unserem Verstaendnis von Galaxien.


Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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