[Epistemic status: highly likely]

Physics has been proclaimed to be dead a few times. Is this time for real?

I think there will certainly not be novelty, say for a thousand years. This thing cannot keep going on so that we are always going to discover more and more new laws. If we do, it will become boring that there are so many levels one underneath the other. It seems to me that what can happen in the future is either that all the laws become known... or it may happen that the experiments get harder and harder to make, more and moree expensive, so you get 99.9% of phenomena, but there is always some phenomenon which has just been discovered, which is very hard to measure... it always gets slower and slower and more and more uninteresting.... It has to end one way or the other... We are very lucky to live in an age in which we are still making discoveries. It is like the discovery of America – you only discover it once.

Richard Feynman - The Character of Physical Law (1965)

The more important fundamental laws and facts of physical science have all been discovered, and these are now so firmly established that the possibility of their ever being supplanted in consequence of new dis - coveries is exceedingly remote. Nevertheless, it has been found that there are apparent exceptions to most of these laws, and this is particularly true when the observations are pushed to a limit, i. e., whenever the circumstances of experiment are such that extreme cases can be examined. Such examination almost surely leads, not to the overthrow of the law, but to the dis- covery of other facts and laws whose action produces the apparent exceptions ... Our future discoveries must be looked for in the sixth place of decimals.

Albert Michelson, address at the opening of the Ryerson Physical Laboratory, University of Chicago, (1894)

Let us first define what we mean by useful physics. By useful physics I mean physics that can be applied to ends other than the knowledge of itself. For example, knowing about the Higgs boson is nice in terms of expanding our knowledge of how the world works. But it has not resulted in the improvement in the welfare of anyone, except those who are marginally more satisfied with the state of our knowledge, and these are a few (I count myself here!).

Progress does not necessarily depend on useful physics: The steam engine was invented and was working before science could explain why it worked. The why (thermodynamics) came after the what. But useful physics necessarily constraints progress: what is physically unfeasible is impossible.

The claim being discussed is not whether technological progress is possible without new fundamental discoveries. This is not true. Progress was there even before science was a thing, and tremendous progress also occurred after the birth of science. Even after the discoveries of electromagnetism, and the coming of modern physics, older discoveries in classical mechanics still enabled improvements, to the point where even today improvements in merely mechanical (or thermal) devices is still possible.

Nor is the claim being discussed that great technological progress is necessarily impossible. The list of General Purpose Technologies (those that are thought to have been great productivity enhancers) lists many that were not the direct result of new physical discoveries, like the printing press, or the concept of mass production.

Nor is the claim being discussed that physics won't make progress in the future. It will. There is a pile of unsolved problems in physics, the most blatant of them being the unification of General Relativity and Quantum Mechanics, and how consciousness arises. New particles and fields will be posited and hopefully confirmed. But to count, they would have to be useful discoveries.

Nor is the claim being discussed that science won't progress at all without new physics. It will. Complex systems that are governed by known laws at the fundamental level sometimes give rise to high level laws that help understand and control that system. For example, understanding how the human body works in detail would enable much better medicine and treatments.

Nor is the claim being discussed whether a Theory of Everything will ever be found in physics. I tend to think that it will. But regardless of whether it is found or not, the claim being discussed remains: A ToE can be an improvement over a pile of inconsistent hacks (QM+GR) in terms of scientific achievement and perhaps predictive power in the 100th decimal and at the same time not be useful.

Nor is even what is under discussion whether I am a pessimist or an optimist. I am by mood an optimist. When I hear about a new scientific discovery my reaction is "Nice!" and not "So what, how does that help me?"

Or whether useless science is worth pursuing. I for one think good things in life are useless: they are ends in themselves, not mere means. I value knowledge intrinsically, and thus I value useless science. But I admit others may not share this value. Most people, I reckon, won't be so thrilled by obscure tweaks to scientific theories and will only value science for its tangible results, and perhaps some cool display in a science museum. I am also not claiming that there is a clear line between useful and useless science, more on this later.

The claim is just whether a new discovery will be made in physics that will enable new technologies to be designed. Examples of these exotic technologies include staples of sci-fi: faster than light travel, wormhole generators, warpdrives, new energy sources. But not only: improvements in existing technologies powered by new physics would also count. If a new computer can be designed on the basis of "Extended Quantum Mechanics" that can do things faster than quantum computers, it would count as one example of improvement in an existing technology. Similarly, if an antigravity device of some sort can be invented, we could have nice flying cars, even though we already have flying cars (albeit not as nice as the ones that antigrav would make possible).

The basic argument for the claim is that the way nature works is given. We have reason to believe the laws of physics are time and space symmetric and we have been progressively discovering more of what is out there. Each -fundamental- discovery has been succesively less useful. Just think of how much that surrounds you would be gone if you take out the discovery of the Higgs boson or quarks: Nothing. Now, take out general relativity: that would be the GPS in your phone. Take out quantum mechanics and you would lose lasers, and modern electronics. But that's still a fairly recognisable world. Take out now classical physics and see then what happens.

But not only that, now we can point the spotlight of our best theories at things and predict with awesome accuracy what is going to happen. As physicist Sean Carroll says in The Big Picture,

The logic behind our audacious claim [Physics is complete for everyday life purposes] is simple: 1. Everything we know says that quantum field theory is the correct framework for describing the physics underlying ev- eryday life. 2. The rules of quantum field theory imply that there can’t be any new particles, forces, or interactions that could be rele- vant to our everyday lives. We’ve found them all. [...]

The strength of effective field theory is what allows us to assert “This time is different” when we make our audacious claim that the laws of phys- ics underlying everyday life are completely known. When Newton and La- place contemplated the glory of classical mechanics, they may very well have considered the possibility that it would someday have to be superseded by more comprehensive theories.

And eventually it by special relativity, general relativity, and quan- tum mechanics. Newtonian theory is a good approximation in a certain domain of applicability, but ultimately it breaks down and we need a better description of reality.

What’s new is that Newton and Laplace, even if they had thought of their ideas as only accurate in a certain regime, had no way of knowing how far that regime extended. Newtonian gravity works very well for the Earth or Venus; it eventually starts breaking down when we consider the orbit of Mercury, whose tiny precession became some of the strongest evidence in favor of Einstein’s general relativity. But Newton would have had no idea how far his theory might be accurate.

With effective field theory, however, that’s exactly what we have. An effective field theory describes everything that happens to a certain set of fields, as long as the energies are lower than a certain cutoff, and distances are larger than a certain lower limit (as set by experiment). Once we have the parameters of the effective theory pinned down, we know what will happen to our fields in any experiment we can imagine within its domain of applicability, even if we haven’t done that experiment yet. It’s this special feature of quantum field theory that gives us the confi- dence to make such audacious claims about the scope of our knowledge.

One might say: do you even David Deutsch? Yes, I do David Deutsch. I've read The Beginning of Infinity (that's where the quotes at the beginning come from). But his arguments are hard to pin down.

In the book, Deutsch mentions also the case of John Horgan, who wrote The End of Science (in 1996). But his argument does not apply to my claim: I haven't said that physics is over, indeed I've conceded that there is work to do, new interpretations and theories to be discovered at the fundamental level, and many 'rules of thumb' at the level of biology and chemistry. Also, in defence of his more rosy view, Deutsch mentions the following discoveries since Horgan's book:

In cosmology, there has been revolutionary progress even in the few years since The End of Science was written – and also since I wrote The Fabric of Reality soon afterwards.

But the examples he goes on to mention are all examples of non-useful physics. One may not talk about an End of Science, but if we interpret Horgan's title to mean the End of useful fundamental Science, the book has held up perfectly since publication.

One may argue, returning to the point I made above, that useless physics today is useful physics tomorrow, just like maths. Indeed, this is true. For example, knowledge of a distant star may not be helpful at all right now, but in a future with spaceships, even if it takes a million years to reach it, that discovery will have allowed sending a spaceship there. One can nitpick with such examples, but the overall claim still stands. Cataloguing new stars is like cataloguing new species, each discovery is not a new fundamental breakthrough.

I don't claim that with 100% certainty, there will be no new useful fundamental physics. It's more like 90%, and that may even be optimistic.

Are there implications from raising one's confidence in the proposition "Useful fundamental science is over?". Yes: The more one believes in it, the more one should think that there is more value in doing other things. In the extreme case, if useful fundamental science is truly over, and taking social preferences for science as a mere means at face value (this is, assuming the basic funding-science-as-a-public-good model, with only a minority of weirdos like me values knowledge for the sake of knowledge), it would follow what research in theoretical physics, and possibly other similar fields should be defunded so that resources allocated to them can instead be put to a better use. Money spent in those fields is not much, but brainpower is not an unlimited resource. If the welfare-raising capability of a theoretical physicist is higher in mechanical engineering, then subsidising theoretical physics (or advanced maths) is socially destructive on the standard economic grounds. (And I say this as a fan of theoretical physics!)

This last paragraph is not "the" point of this blogpost, just one possible implication of the conjunction of some plausible views one may hold, designed for shock value so that the rest of the essay stays in your mind :).

To recap, the reasons for thinking that progress in fundamental science will continue are, along with the answers:

  • Similar claims have failed in the past
    • This time is different. We have mapped all stable elements out there, and measured with great accuracy how what surrounds us works, to a satisfyingly high degree.
    • Also, they didn't fail. Feynman's claim has still held. And so has Michelson's, charitably interpreted. Indeed most of useful physics is pre-modern physics, and QM and GR amount to tiny effects in comparison, which is why it took so long to see that they were there. Tiny effects that do matter, but tiny nonetheless.
    • As science progresses, the likelihood that "fundamental science is over" increases. The fact that Feynman made his claim in 1965 (Half a century ago!) and has not been proven wrong further raise the likelihood of the claim.
  • We can't predict future theories because if we could we would have them now
    • We can't know what they are, exactly but we can make educated guesses.
    • For example, we are fairly sure that future theories won't say that the moon is made of cheese, or that one can build a thermal engine more efficient that a (generalized) Carnot engine
  • The amount of knowledge we can extract about nature is infinite
    • Be that as it may, the claim is narrower: it is a claim about what is possible. I can imagine knowledge increasing greatly while at the same time the range of what is possible remaining constant.
  • But there will be progress within physics even if its borders (fundamental theory) does not improve
    • Yes, but I also said this
  • We won't have all the nice scifi stuff
    • No one said reality has to be cool, it just is how it is.
  • What about physical anomalies like the flyby anomaly
    • Some of these anomalies may not have day-to-day relevance if it is discovered why they happen
    • I said 90% what I say is true. For the other 10%, this could be true. The game is not fully over yet, but we also shouldn't be shocked if it indeed is.

As always comments and critiques are welcome.

Comments from WordPress

  • J J 2018-10-07T23:48:51Z

    That quote from Michelson is probably a misquote. His original quotation doesn't express as much certainty: "It is never safe to affirm that the future of physical science has no marvels in store which may be even more astonishing than those of the past; but it seems probable that most of the grand underlying principles have now been firmly established and that further advances are to be sought chiefly in the rigorous application of these principles to all the phenomena which come under our notice ... An eminent physicist has remarked that the future truths of physical science are to be looked for in the sixth place of decimals." (A. A. Michelson University of Chicago Quarterly Calendar 10 (August 1894): 15) Source: https://archive.org/stream/quarterlycal9495univ/quarterlycal9495univ_djvu.txt

  • Artir Artir 2018-10-08T00:11:36Z

    Oh! This indeed reveals a more cautious outlook. I quoted it verbatim from Deutsch's book and assumed it would be complete.

  • Psycicle Psycicle 2018-10-08T02:11:20Z

    This glosses over the difference between fundamental physics (in which case I agree), and non-fundamental physics, such as condensed matter physics and other areas (in which case, I anticipate that there's some physical effect that is not currently known that can be harnessed to be useful)