Scattered Thoughts on Physics and Philosophy
Is there any other kind?
One of my bigger weaknesses as a blogger (other than the fact that every post comes out to be at least 1500 words) is that I’m not good at timeliness. I try to keep to writing only in the morning before everyone else comes in to work, so if I have an idea in the afternoon I mark it down for the next day. But then other events have a way of getting in the way of that— something urgent comes up at work so I can’t write at all that day, or some other topic jumps the queue— and as a result, I end up writing the actual topical blog post multiple days after I had the idea. Which might as well be putting it in a posthumous memoir when the inspiration was something from the deliberately ephemeral world of social media.
(Also, I have a tendency to lead off with self-indulgent prologues that have nothing to do with the meat of the post…)
Anyway, this past weekend I ended up in a somewhat lengthy tangle of threads on Bluesky relating to the role of philosophy in physics, kicked off by this quote-reply to Liam Bright:
Microblogging isn’t my best medium, and I was a bit distracted by various kid stuff; also, this quickly split into a couple of different sub-threads arguing slightly different things with different people. I thought “Oh, this is tomorrow’s blog post” and then it was the weekend, and then Kevin Drum died, and I had work, and then it was the Covid anniversary, and here we are almost a week later and I still haven’t written the post.
On the bright side, the relationship between philosophy and physics is a bit of an evergreen topic, so even a delayed write-up is probably worth something. But since this is already delayed, I’m going to do it as a series of brief(-ish, for me) comments about various aspects of the several different things that are going on here.
— The original Bluesky post is quoting something from Bright that is itself quoting an observation that Bohr and Einstein and Heisenberg were big readers of philosophy (the original cites Kirkegaard, Kant, and Plato, respectively, though you can find plenty of other philosophers of varying degrees on antiquity to connect them to). Many of the responses to this, like Bright’s, take this as (exaggerating a bit for effect) this explains why those guys were intellectual titans while modern physicists are plodding dullards. If only we had continued to make physics majors read Plato, we wouldn’t be stuck with the much-ballyhooed stagnation in fundamental physics.
My most surface-level response to this is “Why do you think physicists are unaware of the philosophical inclinations of these guys?” The main reason anyone outside of STEM is aware of the philosophical inclinations of Einstein and Bohr is that physicists bang on about this stuff endlessly (verging on tediously). It’s hard to find a description of quantum physics that doesn’t include an extended discussion of philosophical elements of the foundations of quantum theory. Popular treatments of the subject (my own included, though I’ve made an effort to step back from that a bit) positively revel in this stuff, particularly the decades-long back-and-forth between Einstein and Bohr about the status of quantum theory.
At the same time, though, we tend to gloss over the fact that this stuff was very much a sideshow while it was going on. The debates between Bohr and Einstein are widely celebrated after the fact, but as recalled by younger scientists who were there for the Solvay conferences and the Einstein, Podolsky, and Rosen (EPR) paper and all that fun stuff, the primary interest of young physicists at the time was that they suddenly had tools that let them make quantitative predictions about a staggering range of phenomena. Lots of those people went on to take a philosophical turn as they got older, which is a big part of why this stuff gets lionized. But a good deal of that philosophizing ended up running off in weird directions that we try to avoid talking about— Bohr had a whole phase of attempting to apply quantum ideas to human society, Schrodinger got into Eastern mysticism, Pauli was deep into Jung, etc. All of which leads pretty directly to this classic comic, and you can see the reason for the tension I refer to.
—In the specific case of Bohr and Einstein, I think it’s pretty telling that their celebrated philosophical debates get rolling in the late 1920’s— the Solvay conference of 1927 is a signature moment— but that none of those led to anything really substantive in their lifetimes. It’s really not clear (to me at least) that their big turn toward explicit philosophizing was a net win for them, and thus whether more of that is something younger physicists should aspire to. The EPR paper and Bohr’s response to it become hugely important, true, but only much later, after work by Bohm and Bell recast the problem in ways that could be directly investigated (about which more in a bit).
So, then, the question becomes whether their latent philosophical inclinations were in some sense responsible for their earlier successes. This slides a bit into the largely futile game of trying to put a dividing line between “philosophy” and “physics” (which, remember, started as “natural philosophy”) but I do think there’s pretty clearly a sense in which a concern with fundamental principles helped drive the discoveries that earned them intellectual immortality.
Einstein’s work on relativity can be cast as a kind of working-out of the logical consequences of taking the principle of relativity seriously, and he gets credited more than folks like Lorentz and Poincare (who had all the math of special relativity) in part because he carried the philosophical element farther, rejecting the idea of a privileged aether frame. General relativity is driven by extending those ideas further via the equivalence principle and (once Marcel Grossmann taught him the math) general covariance. His statistical model of photons works by taking the idea of quantized light and quantum atoms seriously, and working through the logical consequences of that.
Bohr’s career-making accomplishment was the old quantum theory, and stems from taking a couple of other ideas— Rutherford’s solar-system atom and the Planck relationship between energy and frequency—seriously and working out their logical consequences. His synthesizing of quantum ideas in the early days of the Copenhagen Interpretation also has a bit of this character, running with the idea of concretely observable predictions being the only truly essential element of a theory.
I think you can fairly characterize those developments as having some element of philosophy to them, in a good way.
— The question, then, is what distinguishes the early work that revolutionized physics from the later stuff that didn’t really go anywhere. Or, to put it a slightly different way, what did Bohm and Bell do that made something out of the mushy impasse that was the debate between Copenhagen and EPR?
As I tried to say in the thread on Bluesky, I think there’s a level of concreteness to the early successes of Bohr and especially Einstein that’s missing in the later work. The theories that turned physics on its head have a kind of empiricism at their core— the first relativity paper discusses the need to spell out a procedure for synchronizing measurements that are separated in space, and the Bohr model of the atom takes pains to compare its predictions to experimental measurements of ionization energies and atomic spectra.
Maybe the very best example is Einstein’s work on photons. The first paper from 1905 is labeled (with uncharacteristic humility) a “heuristic model,” but stems directly from experimental measurements of the photoelectric effect. The 1916 paper on the statistics of radiation combines the 1905 “light quanta” with Bohr’s quantum atom and shows that you can use them to re-derive the Planck formula for thermal radiation. It also introduces stimulated emission and photon momentum, both of which could be verified.
By the late 1920’s, though, they’ve both shifted to dealing almost entirely in extremely abstract hypotheticals, and without that grounding empiricism the work sort of drifts off into areas of dubious relevance. The EPR paper sets up a scenario that we now recognize as fascinating and fruitful, but deploys it in a way that’s much more about abstract rhetoric than measurable reality. Bohr’s response is infamously muddled, for similar reasons— it’s become an argument about high-level abstract principles, and neither side seems all that interested in making a connection to empirical measurements.
What unsticks the debate is a return to that core of empiricism: Bohm re-casts the EPR argument in terms of angular momentum, which makes the whole thing a lot clearer and lends itself to explicit consideration of measurements in a way that the original EPR scenario doesn’t. Bell picks that up, and works out the consequences: actual quantities that you could think about measuring that would come out differently in the different pictures. Which inspires Clauser to start thinking of how to do the experiments, and then Aspect to do a better version, and as a result we get a big explosion in work on quantum foundations.
My feeling is that the big gap between EPR and Bell owes a lot from a shift in interests on the part of Bohr and Einstein from arguments grounded in empirical questions to wholly abstract philosophical questions. You can also see a bit of this in some of the self-mythologizing Einstein engages in during this same period, where he claims to have always been exclusively motivated by abstract principles and mathematical elegance. This includes things like claiming to be unaware of the negative result from the Michelson-Morley search for “aether drag” and other experimental results. I saw somewhere a comparison of his later autobiographical comments with statements that were (closer to) contemporaneous with his big results (in Pais, maybe?) that make the late stuff seem pretty clearly a back-projection of his later interests. I think that turn away from empiricism to abstract principle led him astray.
— There’s a kind of definitional question at this point about how to characterize the work in these different period. Is the early stuff physics and the later stuff philosophy? Is it all philosophy but the early stuff is done well and the later stuff badly?
That comes down to a question of what you consider the defining characteristics of each field. I would say that they overlap in a kind of shared commitment to the working-out of consequences of fundamental principles, but they diverge in that physics has a firm commitment to testing ideas against empirical reality, where philosophy is more accepting of appeals to intuition or even aesthetics. What pushes Einstein’s later work to the “philosophy” side of the Venn diagram, for me, is that it’s more likely to rely on abstract criteria than empirical ones. He rejects quantum theory because he sees it as not incorrect in its description of real phenomena, but as an “intolerable” affront to his intuition about how the world ought to work.
That seems to me to be a very different game than calculating the perihelion shift of Mercury or the diffusion rate of a particle in fluid or the frequency dependence of spontaneous emission and absorption of light by atoms. And, ultimately, I think it’s one that’s less likely to be productive.
(Applying this to the last forty-mumble years of particle theory is left as an exercise for the interested reader.)
—I’ll end this with something a little snide about philosophy, just to see if anyone actually reads all the way to the end of this, which is that I find it sort of telling that the philosophy cited— Kierkegaard, Kant, Plato— is all old. This is, admittedly, mostly a reflection of the fact that those are the philosophers one would read when Bohr and Einstein and Heisenberg were young, but it’s sort of interesting that even in the branches of social-media threads where people are all “Yeah! Physics would get unstuck in no time if physicists read more philosophy!” nobody’s stumping for philosophers from even the late 20th century. It’s all Quine and Hume and not Kuhn or Feyerabend1. And, you know, if the argument is that contemporary physics might be improved by reading more pre-1900 philosophy, that’s kind of a different proposition than the argument that contemporary philosophy-of-science is an essential prerequisite for progress…
I sometimes think that there’s actually something to that. For all that people with philosophy degrees of recent vintage roll their eyes at the naive positivism of physicists who dabble in philosophy, something like Popper’s notion of falsification at least has a constructive element. Asking yourself “How would I prove this wrong?” can be a productive way to come up with experimental ideas to pursue. I don’t know what the hell you would be inspired to do with, say, Kuhn’s social construction, other than switching careers to finance, where you can at least get flipping great wodges of cash.
But that’s a different topic altogether, so I’ll just stop here.
So, that’s at least more words than I could do on Bluesky; whether it’s any clearer or more convincing, who can say? If you’d like to see whether I come back to say more unclear (or snide) things about the topic, here’s a button:
And if you have thoughts in response to this, the comments will be open:
To be fair, I did once hear Lee Smolin talk at some length about how Feyerabend influenced his thinking about approaches to fundamental physics. I can’t say I followed his logic, and I bounced off Against Method in a big way because his writing is just chock full of things that my Usenet background taught me are “This is a a Crazy Person” signifiers…
>>> One of my bigger weaknesses as a blogger (other than the fact that every post comes out to be at least 1500 words) ...
A feature, not a bug, as far as I am concerned.
P.S. LOL @ fn1