Critical views of science in the news

Physicist George Ellis Knocks Physicists for Knocking Philosophy, Falsification, Free Will


Biologist Rupert Sheldrake, whom I interviewed in my last post, wasn't the only fascinating scientist I hung out with recently at Howthelightgetsin, a festival hosted by the Institute of Arts & Ideas. I also befriended George F. R. Ellis, the physicist-mathematician-cosmologist, an authority on the Big Bang and other cosmic mysteries. Ellis and I hit it off initially because we share some—how shall I put it?—concerns about the direction of physics, but I soon discovered that his interests range far beyond physics. He has published papers and books not only on physics and cosmology (including the 1973 classic The Large-Scale Structure of Space-Time, co-authored with Stephen Hawking) but also on philosophy, complexity theory, neuroscience, education and even low-income housing. (See his website, and his terrific 2011 critique of multiverse theories in Scientific American.) A native of South Africa, Ellis is professor emeritus at the University of Cape Town, where he taught for decades, and has also held positions at Cambridge, the University of Texas, the Fermi Institute and other institutions around the globe. I admire Ellis's social activism as well as his scientific work. He was an early critic of apartheid, and in 1999 Nelson Mandela awarded him the Order of the Star of South Africa. Ellis has a big brain and a big heart.

"You cannot do physics or cosmology without an assumed philosophical basis," says George Ellis. Photo: David Monniaux, Wikimedia Commons.

Horgan: At the conference where we met, Howthelightsgetsin, you were in a session called "The end of experiment." What was that about?

Ellis: Well this was just echoing what you have already said: many of the possible high-energy physics experiments and astronomy observations relevant to cosmology are now in essence nearly complete. Physics experiments are approaching the highest energies it will ever be possible to test by any collider experiment, both for financial and technical reasons. We can’t build a collider bigger than the surface of the Earth. Thus our ability to test high energy physics – and hence structures on the smallest physical scales – is approaching its limits. Astronomical observations at all wavelengths are now probing the most distant cosmological events that will ever be “seeable” by any kinds of radiation whatever, because of visual horizons for each form of radiation.

It’s rather like the situation as regards exploring the Earth: once upon a time we had only fragmentary knowledge of what is there. Then we obtained a global picture of the Earth’s surface, including detailed satellite images of the entire land mass. Once you have seen it all, you have seen it all; apart from finer and finer details, there is nothing more to find. You might respond, But we can’t see to the bottom of the oceans. However, we do indeed now have quite good maps of the ocean floor too, through various sounding techniques. This is similar to the way we have seen right back to the last scattering surface in the early universe at a redshift of 1200 (the analogue of seeing the entire surface of the Earth from space) with satellites such as COBE, WMAP, and Planck, and also (indirectly) to the time of emission of gravitational waves by Bicep2 (the analogue of seeing to the bottom of the ocean). We’ll sort out that controversy in the next couple of years.

So what we can see at the largest and smallest scales is approaching what will ever be possible, except for refining the details.

But I emphasize that this comment does not apply to complex systems. Complexity is almost unbounded – microbiology, biology, the brain will give us work to do for many centuries more, what we may find may be very unexpected. That might also apply to the foundations of quantum physics, and its relation to complexity. But – barring something very unforeseen - the possible tests of the very large and the very small are coming towards the limits of whatever will be possible.

Yes I know this kind of thing has been said before. That was before we had explored the entire visible universe at all possible wavelengths. I concede that observations relevant to structure formation in the universe – galaxies, stars, planets – have a good while to go, they are in essence verging to the side of studying complexity, and still have life in them yet, they are very interesting studies.

Horgan: Lawrence Krauss, in A Universe from Nothing, claims that physics has basically solved the mystery of why there is something rather than nothing. Do you agree?

Ellis: Certainly not. He is presenting untested speculative theories of how things came into existence out of a pre-existing complex of entities, including variational principles, quantum field theory, specific symmetry groups, a bubbling vacuum, all the components of the standard model of particle physics, and so on. He does not explain in what way these entities could have pre-existed the coming into being of the universe, why they should have existed at all, or why they should have had the form they did. And he gives no experimental or observational process whereby we could test these vivid speculations of the supposed universe-generation mechanism. How indeed can you test what existed before the universe existed? You can’t.

Thus what he is presenting is not tested science. It’s a philosophical speculation, which he apparently believes is so compelling he does not have to give any specification of evidence that would confirm it is true. Well, you can’t get any evidence about what existed before space and time came into being. Above all he believes that these mathematically based speculations solve thousand year old philosophical conundrums, without seriously engaging those philosophical issues. The belief that all of reality can be fully comprehended in terms of physics and the equations of physics is a fantasy. As pointed out so well by Eddington in his Gifford lectures, they are partial and incomplete representations of physical, biological, psychological, and social reality.

And above all Krauss does not address why the laws of physics exist, why they have the form they have, or in what kind of manifestation they existed before the universe existed (which he must believe if he believes they brought the universe into existence). Who or what dreamt up symmetry principles, Lagrangians, specific symmetry groups, gauge theories, and so on? He does not begin to answer these questions.

It’s very ironic when he says philosophy is bunk and then himself engages in this kind of attempt at philosophy. It seems that science education should include some basic modules on Plato, Aristotle, Kant, Hume, and the other great philosophers, as well as writings of more recent philosophers such as Tim Maudlin and David Albert.

Horgan: Are you, or were you ever, a believer in a final theory of physics?

Ellis: I certainly have believed in it as a possibility. However the price of having to use higher-dimensional theories is in my view a considerable drawback and to be avoided if possible; it is certainly unproven to be the way Nature is. Rather it may be that fundamental physics in the end involves two different intermeshed theories: a Grand Unified Theory (GUT) and unimodular Loop Quantum Gravity, with no hidden dimensions and no string theory landscape. The unification will be in terms of broad approaches underlying physical theories (Lagrangians, variational principles, physical symmetries, etc.) but not necessarily one overriding theory that encompasses all fundamental physics in a unified form.

Indeed it is in my view unlikely there is a unified theory of all fundamental forces including gravity, because Einstein taught us that gravity is *not*, at a foundational level, in fact a force – rather it is an effective result of spacetime curvature. Yes of course there are various effective theories of gravity where it is seen as a force, such as Newton's theory and representing it as a spin 2 graviton, but their existence does not determine what the underlying deep theory of gravitation is. My feeling is that it will be a theory based on a discrete view of space-time with a conformal structure for the gravity sector, probably based in principles of holonomy, interacting with a unified GUT theory for the matter sector. That is not a unified final theory of physics as envisaged by string theorists. Such a theory may not exist.

Horgan: Are you a fan of multiverse theories? String theory? The anthropic principle?

No (may be true but unproveable, much too much untestable speculation about existence of infinities of entities, ill defined and untestable probability measures), no (too much speculative introduction of very complex unseeable entities, treats gravity just like any other force), yes (however one responds to it, it’s a real question that deserves consideration). Fine tuning of fundamental physics parameters is required in order that we can exist. Examining this issue has led to many very interesting studies.

Horgan: Physicist Sean Carroll has argued that falsifiability is overrated as a criterion for judging whether theories should be taken seriously. Do you agree?

Ellis: This is a major step backwards to before the evidence-based scientific revolution initiated by Galileo and Newton. The basic idea is that our speculative theories, extrapolating into the unknown and into untestable areas from well-tested areas of physics, are so good they have to be true. History proves that is the path to delusion: just because you have a good theory does not prove it is true. The other defence is that there is no other game in town. But there may not be any such game.

Scientists should strongly resist such an attack on the very foundations of its own success. Luckily it is a very small subset of scientists who are making this proposal.

Horgan: Krauss, Stephen Hawking and Neil deGrasse Tyson have been bashing philosophy as a waste of time. Do you agree?

Ellis: If they really believe this they should stop indulging in low-grade philosophy in their own writings. You cannot do physics or cosmology without an assumed philosophical basis. You can choose not to think about that basis: it will still be there as an unexamined foundation of what you do. The fact you are unwilling to examine the philosophical foundations of what you do does not mean those foundations are not there; it just means they are unexamined.

Actually philosophical speculations have led to a great deal of good science. Einstein’s musings on Mach’s principle played a key role in developing general relativity. Einstein’s debate with Bohr and the EPR paper have led to a great of deal of good physics testing the foundations of quantum physics. My own examination of the Copernican principle in cosmology has led to exploration of some great observational tests of spatial homogeneity that have turned an untested philosophical assumption into a testable – and indeed tested - scientific hypothesis. That’ s good science.

Ellis: You are a Christian, more specifically a Quaker. Does your faith have any effect on your scientific views, or vice versa?

It may affect to some degree the topics I choose to tackle, but it cannot affect the science itself, which has its own logic that must be followed wherever it leads without fear or favour, within the domain of application of the relevant theories.

My philosophical and religious views must of course take present-day science seriously, but in doing so (a) I distinguish very clearly between what is tested or testable science and what is not, (b) I make strenuous efforts to consider what aspects of reality can be comprehended by a strict scientific approach, and what lie outside the limits of mathematically based efforts to encapsulate aspects of the nature of what exists.

Many key aspects of life (such as ethics: what is good and what is bad, and aesthetics: what is beautiful and what is ugly) lie outside the domain of scientific inquiry (science can tell you what kind of circumstances will lead to the extinction of polar bears, or indeed of humanity; it has nothing whatever to say about whether this would be good or bad, that is not a scientific question).

Attempts to explain values in terms of neuroscience or evolutionary theory in fact have nothing whatever to say about what is good or bad. That is a philosophical or religious question (scientists trying to explain ethics from these kinds of approaches always surreptitiously introduce some unexamined concept of what is a good life by the back door). And they cannot for example tell you, from a scientific basis, what should be done about Israel or Syria today. That effort would be a category mistake.

Horgan: Is your social activism--for example, you past efforts against apartheid--related in any way to your scientific work?

Not directly. It is based in concerns to do with human values and ethics that lie outside the scope of science per se.

Horgan: Are you concerned that today so much research—especially in the U.S.—is funded by the military?

This is not an issue that has been of specific concern to me – but it could become so, particularly as regards brain research.

Horgan: In some of your writings, you warn against excessive determinism in physics, and science. Could you summarize your concerns?

Many scientists are strong reductionists who believe that physics alone determines outcomes in the real world, This is demonstrably untrue – for example the computer on which I am writing this could not possibly have come into being through the agency of physics alone.

The issue is that these scientists are focusing on some strands in the web of causation that actually exist, and ignoring others that are demonstrably there – such as ideas in our minds, or algorithms embodied in computer programs. These demonstrably act in a top-down way to cause physical effects in the real world. All these processes and actual outcomes are contextually dependent, and this allows the effectiveness of processes such as adaptive selection that are the key to the emergence of genuine complexity.

As I stated above, mathematical equations only represent part of reality, and should not be confused with reality. A specific related issue: there is a group of people out there writing papers based on the idea that physics is a computational process. But a physical law is not an algorithm. So who chooses the computational strategy and the algorithms that realise a specific physical law? (Finite elements perhaps?) What language is it written in? (Does Nature use Java or C++? What machine code is used?) Where is the CPU? What is used for memory, and in what way are read and write commands executed? Additionally if it’s a computation, how does Nature avoid the halting problem? It’s all a very bad analogy that does not work.

Horgan: Einstein, in the following quote, seemed to doubt free will: "If the moon, in the act of completing its eternal way around the Earth, were gifted with self-consciousness, it would feel thoroughly convinced that it was traveling its way of its own accord…. So would a Being, endowed with higher insight and more perfect intelligence, watching man and his doings, smile about man’s illusion that he was acting according to his own free will." Do you believe in free will?

Ellis: Yes. Einstein is perpetuating the belief that all causation is bottom up. This simply is not the case, as I can demonstrate with many examples from sociology, neuroscience, physiology, epigenetics, engineering, and physics. Furthermore if Einstein did not have free will in some meaningful sense, then he could not have been responsible for the theory of relativity – it would have been a product of lower level processes but not of an intelligent mind choosing between possible options.

I find it very hard to believe this to be the case – indeed it does not seem to make any sense. Physicists should pay attention to Aristotle’s four forms of causation – if they have the free will to decide what they are doing. If they don’t, then why waste time talking to them? They are then not responsible for what they say.

Horgan: If you were the King of Physics, responsible for prioritizing and funding research, what would be your first decision?

Ellis: Condensed matter physics and quantum optics are where I’d concentrate first – it is where relatively modest investment is producing fabulous experiments and testable theories. The relation of physics to biology, medicine, and neuroscience is a fantastic area that should be strongly developed.

Yes of course I’d like to see astronomy and high-energy physics continued as far as possible, but the price has to be reasonable. Projects like Bicep2 give great returns, and gravitational wave astronomy is the last frontier and so must be explored, as also the many great observational cosmology projects under way at present. Megabucks for ever-greater colliders will need solid justification.

The views expressed are those of the author and are not necessarily those of Scientific American.

Share this Article:


You must sign in or register as a member to submit a comment.

Email this Article