The Pseudoscience of Fundamental Physics
Among contemporary physicists, Stephen Hawking is one of the better known. I sometimes ask friends and colleagues why he has not yet been awarded the Nobel Prize? Most people are at loss to respond.
The answer is related to three pillars of science, erected in the early 17th century, which have been the hallmark of science ever since: (i) development of theoretical models, (ii) expression of these models through mathematics, and (iii) verification by observation.
But today the third pillar is under threat, the principle of verification by observation. Cosmologists Ellis and Silk urge their readers to “Defend the integrity of physics,” as their article in Nature in 2014 was entitled. They go against colleagues who argue that if a theory is what they call “sufficiently elegant and explanatory,” then there is no need for experimental testing. The controversial theories are in particular those of string theory for the elementary particles and forces, and inflation theory for the extremely rapid expansion of space in the moment after Big Bang.
One of the consequences of inflation theory is the multiverse hypothesis. If the infant universe initially inflated faster than light, it may have spun off into separate entities which would later be unable to communicate. The multiverses are therefore inherently unobservable. The multiverse hypothesis and other similarly controversial theories either make unverifiable predictions, or they make so many predictions that anything can be explained. Ellis and Silk called for a conference to be convened where both sides of the testability debate should be involved. I will come back to this conference later, as it actually took place near the end of 2015.
How does Stephen Hawking and the Nobel Prize fit into this picture? His black hole radiation theory may in principle be verifiable, but the predicted radiation from these regions of space with super strong gravitation is too weak to be observed directly. Scientists have therefore looked for laboratory models that resemble the conditions of black holes, and there are claims that such experiments may have demonstrated Hawking’s predictions. But so far this is not generally accepted, and fortunately, the Nobel committee is adamant in its insistence that a physics theory should be supported by solid observation or experiment.
Indeed, Hawking is one of the cosigners of an article in Scientific American earlier this year, where he defended inflation theory as a real field of science. It was written by some of the architects of the theory, Guth and Linde, and it came as a response to Ijjas, Steinhardt, and Loeb’s article "Cosmic Inflation Theory Faces Challenges". In the latter article, the authors cast doubt on whether inflation theory should be regarded as an empirical science at all. Steinhardt, which is one of the heavyweights in the field of inflation theory, has turned into one of its most ardent critics.
Verification by observation is the scientific method
In my rather pessimistic moments, I view the reduction of the importance of verification by observation as the beginning of a regression back to the Greek science, where theorizing could outweigh observation. The most well-known example is the insistence on circular planetary orbits in the geocentric model of antiquity. It is grounded in an elegant theory that says that heaven is the realm of perfect gods and that this requires a perfect orbit—and what is more perfect than a circle? This theory was held onto despite observations (of retrograde movement of the planets) that contradicted the model. Instead observations were matched by adding more circles, and by offsetting both the orbits and the earth from the center of the orbit. The idea of circular orbits was so powerful that even Copernicus’ heliocentric model from 1543 held onto it.
In contrast, the modern scientific method is somewhat akin to Montesqieu’s principle of separation of powers into a legislative, an executive, and a judiciary system in the governance of states. Montesqieu's principle of separation of power was formed by several influences, among them the British constitutional system—thus the system of checks and balances could be said to have its feeble roots in Magna Carta of 1215. For Montesqieu, the independence of the courts was considered to be the most important one. In the same way, experiment and observation are the final arbiters of a physical theory. I rather like the way Nobel laureate Richard Feynman articulated the importance of a system of checks and balances: “Science is a way of trying not to fool yourself. The first principle is that you must not fool yourself—and you are the easiest person to fool.”
One of the best examples of the significance of verification by observation comes from the theory of general relativity. In 1915 it explained a problem that physicists had struggled with for decades, that of a small perturbation of the orbit of Mercury. During the solar eclipse of 1878 many astronomers had been searching for Vulcan, a hypothetical planet between the Sun and Mercury which was one of the hottest contenders for explaining the anomaly. But somehow a likely orbit could not be computed and more importantly, Vulcan was never even observed. Then came Einstein’s theory, which solved the problem very elegantly. Another prediction was the doubling of the amount of gravitational deflection of light compared to Newton’s theory. This was verified during yet another eclipse, that of 1919 and this catapulted Einstein to world-wide fame. The detection of gravitational waves in 2016 was but one of many verifications of Einstein’s theory.
Adding to such as Ellis, Silk, and Steinhardt, science writer John Horgan is one of the most articulate critics of the new turn in the sciences. In The end of Science from 1996 (re-released in 2015), Horgan expresses concerns that science is pursued in a speculative, post-empirical mode which he calls "ironic science," or “pseudoscience.” It is a kind of science that does not converge on truth because it cannot be empirically verified. The demarcation between empirically verified science, that is "real science," and pseudoscience is blurred, and Horgan has continued to be critical ever since he wrote the book.
History of science as guideline
Listening to voices like these, I get worried that real science and pseudoscience are about to be confused. It is therefore good to hear what historian Helge Kragh says in order to get some perspective on the issues at stake. He was one of the participants at the conference “Why Trust a Theory?" in Munich (December 2015), called by Ellis and Silk.
In the paper "Fundamental Theories and Epistemic Shifts: Can History of Science Serve as a Guide?", Kragh discusses the demarcation problem between science and pseudoscience. He then makes some very interesting historical analogies with the vortex theory that particularly British scientists developed in the last half of the 19th century. This was a hydrodynamic theory of matter in the form of atomic particles swirling around in a cosmic fluid, usually identified as the ether. It was expected to (ultimately) explain gravity, and its proponents kept insisting optimistically that with some further development, that goal would be achieved. They were attracted to the theory just as much for its mathematical and aesthetic beauty and elegance as for its ability to explain phenomena. In that respect, it bears some resemblance to string theory.
The ideological and religious climate has changed over a period of more than 100 years. In the past God was thought to be found somewhere at the depth of the theories, now multiverse theories are used by some to explain away God and promote atheism. Whether this is a plausible argument is another matter. The fundamental question is “why there is something rather than nothing,” and the introduction of multiple universes rather than one universe does not really answer that question any better than if there is only a single universe.
The demarcation between science and philosophy
Having said that, even well-known physicists may have a rather unclear understanding of another demarcation, namely that between science and philosophy. Thomas Nagel said in his book Mind and Cosmos (2012):
Scientist are well aware of how much they don’t know, but this is a different kind of problem—not just acknowledging the limits of what is actually understood but of trying to recognize what can and cannot in principle be understood by certain existing methods.
Hawking is a typical example. In his book from 2010 he claimed: “Philosophy is dead. Philosophy has not kept up with modern developments in science, particularly physics. Scientists have become the bearers of the torch of discovery in our quest for knowledge.” But if philosophy is dead, why is so much of the book filled with a rather primitive version of it, without much regard for the great philosophical traditions that exist? When it comes to topics which touch the borders of philosophy, it is probably wiser to listen to Bob Dylan: “Don’t put my faith in nobody, not even a scientist.” In contrast to Hawking, Dylan is actually a Nobel laureate, for whatever that is worth.
With such unclear understanding of the boundary to philosophy among many scientists, one may therefore wonder if there are only purely scientific motivations for hypotheses such as the multiverse theory. I believe issues such as these will become clearer with time. If the theories do not deliver results of a purely scientific nature they will gradually disappear just like vortex theory once did.
Most scientists do normal science
To further support optimism on behalf of science, it should be said that most physicists deal with what can be called “normal science” and are not involved in developing fundamental theories. They are concerned with solving problems or puzzles within the existing paradigm without questioning it as John Horgan says in his book.
My own work in explaining mechanical properties of complex media with a view to medical imaging represents “normal science.” One example is a theory for wave propagation in fractals which is confirmed by our own experiments. Another is the study of properties of viscous media with theory from non-integer derivatives and verification by experiments of others. But there is always the danger that even theories of “normal science” (such as the one we use for non-integer or fractional derivatives), may lose touch with physics and become first and foremost an elegant and aesthetically pleasing mathematical theory.
I share Ellis and Silk’s concern that relaxation of the requirement for verification by observation opens the door for pseudoscientists to claim that their ideas are equally good as real science. Therefore, the principle of verification by observation cannot ever be abandoned. We must stick to the philosophy that Richard Feynman so pointedly expressed: “It doesn’t matter how beautiful your theory is, it doesn't matter how smart you are. If it doesn’t agree with experiment, it’s wrong.” Meanwhile we are waiting for the speculative theories either to deliver solid results or fade away.