It’s a trope among physicists that “nobody understands quantum mechanics”. It’s often offered as a humorous introduction to the subject, to either students or a general audience.
Recently, a particular example has been making the rounds on Twitter. Here Condensed Matter theorist Ramamaturi Shankar introduces his class on Quantum Mechanics1.
Yes, sure, Quantum Mechanics is hard. Yes, sure, Quantum Mechanics is very unlike Classical Mechanics. And yes, sure, this invitation to a hard and weird subject is certainly welcoming! For all these reasons I love Shankar’s introduction to the subject.
But there’s one problem I can’t get over.
Nobody Understands Quantum Mechanics?
“I think I can safely say that nobody understands quantum mechanics.”
Richard P. Feynman, The Character of Physical Law (1965)
The troupe is often couched as an implicit quote from the sixth in a series of public lecture Feynman gave over sixty years ago2:
Continuing this trope - that nobody understands Quantum Mechanics - is a persistent but fond allusion to Feynman.
Feynman’s point was two-fold.
First, that there is no classical analog to frame your own understanding. Matter behaves like matter behaves in Quantum Mechanics.
“Now we know how the electrons and light behave. But what can I call it? If I say they behave like particles I give the wrong impression; also if I say they behave like waves. They behave in their own inimitable way, which technically could be called a quantum mechanical way.”
Richard P. Feynman, The Character of Physical Law (1965)
This is the same point that Strassler was trying to make about “wavicles” that we spoke about a couple of weeks ago:
Second, nature is inherently stochastic. This fact is a departure from how physics had been done historically. This was a departure from the philosophical understanding of how Science itself worked3.
“A philosopher once said ‘It is necessary for the very existence of science that the same conditions always produce the same results’. Well, they do not. You set up the circumstances, with the same conditions every time, and you cannot predict behind which hole you will see the electron. Yet science goes on in spite of it — although the same conditions do not always produce the same results.”
Richard P. Feynman, The Character of Physical Law (1965)
From this perspective, what Feynman meant about understanding meant specifically, definitive prediction. He’s not saying that nobody understands the mathematics. That nobody understands the physics. He’s simply pointing out that our understanding of of understanding needs to evolve.
“The difficulty really is psychological and exists in the perpetual torment that results from your saying to yourself, ‘But how can it be like that?’ which is a reflection of uncontrolled but utterly vain desire to see it in terms of something familiar…
Do not keep saying to yourself, if you can possibly avoid it, ‘But how can it be like that?’ because you will get ‘down the drain’, into a blind alley from which nobody has yet escaped.
What machinery is actually producing this thing? Nobody knows any machinery. Nobody can give you a deeper explanation of this phenomenon than I have given; that is, a description of it. They can give you a wider explanation, in the sense that they can do more examples to show how it is impossible to tell which hole the electron goes through and not at the same time destroy the interference pattern. They can give a wider class of experiments than just the two slit interference experiment. But that is just repeating the same thing to drive it in. It is not any deeper; it is only wider.”
Richard P. Feynman, The Character of Physical Law (1965)
Who Understands Quantum Mechanics?
Many have misunderstood Feynman’s remark as if to imply there is a problem. That there is some forbidden knowledge on how the universe works that we do not yet have access to. For these polemics, humans do not understand Quantum Mechanics as if by definition.
Such people are not serious.
Many people understand Quantum Mechanics. We can explain it. We can teach it. We regularly use it to repeatedly make and successfully test predictions. Quantum Electrodynamics affords the most precise agreement between theory and experiment in all of human history. Indeed, any physicist who passed their qualification exams in graduate school understands Quantum Mechanics quite well4.
To the dedicated polemic, such refutations are themselves refuted by shifting the goal post. Precision tests are never good enough - for example - if you can’t “explain” the collapse of the wavefunction5. It’s all or nothing with them, a false dilemma. The polemic will complain either about a lack of “full” understanding of the mathematics or continuously and vaguely appeal beyond the bleeding edge of Scientific understanding. There’s always some kind of “missing piece” that prevents our “full understanding”. That’s the shifting goal post.
Science has never been about fully understanding anything. That’s a religious idea.
In Classical Mechanics, we can in principle have full knowledge of what nature is doing. Quantum Mechanics differs Classical Mechanics insofar as you cannot have full knowledge, by construction. Quantum Mechanics is a probabilistic theory, it can only tell you likelihoods of physical phenomena being observed.
Any would-be “perfect” knowledge of how the particles behave stands in direct contrast with observed reality. This point was formalized by John Stewart Bell in his now eponymous - and experimentally tested - theorem6.
The Measurement Problem
Measurement in Quantum Mechanics is an irreversible process that changes the physical state as it is being observed. In Quantum Mechanics, prior to measurement, a physical system can exist in multiple “classical” states simultaneously.
Because only one physical state can be observed at a time, the coherent combination of multiple “classical states” then collapses into single one upon direct measurement. Quantum Mechanics assigns a probability to each observable, classical state.
One profound insight from this idea is that a coherent combination of multiple classical states can constructively or destructively interfere with the likelihood of observing those states. Meaning that the final observations of some object moving in laboratory are different depending on how carefully it is monitored along the way7.
Experimental verification of Quantum Mechanics amounts to doing multiple experiments and verifying that statistical properties of the observations line up with those of the predictions.
Quantum Mechanics, in other words, is a probabilistic endeavor. And this really bothers some people. The definition of measurement is admittedly vague, and so there have been many attempts to explain how measurement works in Quantum Mechanics.
The Problem with the Measurement Problem
Most explanations of the measurement problem are mere interpretations, a distinction only relevant in Philosophy. A famous example of such an interpretation is Hugh Everett III’s Many Worlds hypothesis8, wherein multiple, coherent classical states correspond to various copies of reality allowed to exist consistently, and measurement effectively selects between them. It amounts to the statistical mechanics of observable copies of a physical system.
There are many such interpretations, but they all - at present anyway - predict the same observations. Occam’s Razor would then suggest we take the simplest, least informative9 one. This would be the Copenhagen interpretation.
At least, until we can come up with an experiment to run.
Is Quantum Mechanics Wrong?
If “nobody understands Quantum Mechanics”, then how can we know if it is right? Could it be wrong? Are we really living in multiple, parallel universes?
You can see how quickly these slippery questions can come at you.
Feynman’s remarks are meant to suggest that Quantum Mechanics is inherently probabilistic in nature. To understand Classical Mechanics was to know with certainty10 the trajectory of a physical object in time. Measurement - the act of knowing what state the physical object is in - irreversibly changes its reality.
Many who are dissatisfied with the Copenhagen interpretation of Quantum Mechanics want an explanation for why measurement has such an impact on reality.
Physicist David Deutsch famously has designs around a potential experiment that apparently could test the validity of Everett’s many worlds interpretation. This would be a welcome advancement if it could be consistently carried out. Although to date no serious and explicit experiment has been put forward to test it. But there are other ways to make progress.
Experiments can be done to test for alternative dynamics that could mimic - but also diverge from - the standard, Copenhagen interpretation. Sean Carroll - among others - has publicly advanced the idea that such experiments might be required to complete our understanding of gravitation11.
That said, serious proposals to explore interpretation-independent deviations from the Copenhagen interpretation have been put forward, notably by the late Steven Weinberg12 and some experimental limits have already been set.
Silly Answers to Silly Questions
Fantastical ideals about multiple realities and grand philosophical debates belong in the same domain of human knowledge as religious rites and cultural practices. Science progresses instead by posing and test hypotheses.
Do we live in multiple, parallel universes?
At the time of writing, there is no clear and coherent way to prove that this is true or false. Without an experiment, it is an ill-defined question13. Hence all we can say is that the Many Worlds Interpretation is a simply cute interpretation of the observable facts of nature.
Is Quantum Mechanics Wrong?
Quantum Mechanics successfully predicts phenomena. So it is not wrong. It is not wrong in the same sense that Classical Mechanics is not wrong. It’s a matter of experimental resolution. The theory of matter needed to be extended to include Quantum Mechanics.
If experiments are able to show deviations from the standard, Copenhagen predictions - like those Weinberg had discussed - then we have learned something new. Quantum Mechanics will be extended to included these new effects.
No doubt such evidence will only lead to more experiments, and more knowledge.
What about this « insert radical new idea here » ?
Cute. But it’s just an idea at this point. Radically new ideas require radically new evidence, and the inherent stochastic nature of Quantum Mechanics has already been experimentally established. Evidence for the nonlocality associated with entanglement in quantum measurement continues to pour in. Any new idea will need to be tested just like the old ones.
So What?
To understand Quantum Mechanics in the present day is to use the Copenhagen interpretation, and be mindful of experiments that may detect deviations in the future. Ignorance is a feature of Quantum Mechanics, not a bug. It’s how Quantum Mechanics works. If you want to understand why it works this way, you’re asking questions outside the scope of Science. That is shifting the goal posts.
To say that “nobody understands” something like Quantum Mechanics can imply to a general audience that experts have no idea what they’re doing. Hence, there is no point in learning the subject, because the experts must be wrong. Even when well-intentioned, this negates the tremendous amount of work already done by individuals past and present.
It’s hard to express just how outrageous the many worlds proposal is14. Endorsing such a complicated reality without a shred of evidence amounts to little more than superstition15. It also trivializes the hard work required to experiment ourselves away from the Copenhagen interpretation.
Feynman’s quote was intentionally bombastic, as was his style. He leveraged his comment ironically to emphasize the philosophical break between classical and quantum physics. Without that context, his comment does not land the same in today16.
Progress in Science is made via negativa - that is, by way of negation. Attaining a positive truth - a full understanding - is not part of the Scientific Method. This is what makes statements like “Nobody understands Quantum Mechanics” at best, worthless and at worst, insidious. Such a statement aims at the religious or superstitious conception of complete knowledge.
Society stands the institutional knowledge of Science. If we lose that knowledge, we risk falling back into darkness.
Shankar wrote a big, red textbook on Quantum Mechanics that - in accordance with its size - had an outsized influence on me as an undergraduate. It helped push me on to the path of applying for graduate school. Regrettably, its one of the few textbooks that simply vanished from my library a decade ago.
The sixth lecture, “Probability and Uncertainty” starts in the video above at around 04:10:00. The quote itself comes at around 04:17:00. You can also find it in published transcription, The Character of Physical Law on page 132.
Laplace’s demon is a thought experiment which suggests that if we were to know with absolute certainty the trajectories of all particles through phase space, we would be able to predict the future with absolute certainly. For proponents of Human free will, that could have been a thorny intellectual problem. Fortunately, Nature persistently shuts down our academic hubris.
Philosopher Tim Maudlin here apparently disagrees, calling a physicist’s canonical understanding of Quantum Mechanics a “recipe” instead of a “theory”. To that, Feynman ends his 1965 lecture by saying “In fact it is necessary for the very existence of science that minds exist which do not allow that nature must satisfy some preconceived conditions, like those of our philosopher.” More on this in a future post.
This should be distinguished from a dissatisfaction with the measurement problem in Quantum Mechanics as it presently stands.
I say all this not to discount those still investigating the various remaining hidden variable types of theories which suggest that Bell’s Inequalities may be avoided by some as yet unknown kind of nonlocal determinism that can be experimentally verified. But based on our present knowledge and experience, most such explanations have been ruled out. But hopefully we have learned enough about nature as to avoid the hubris of Laplace.
Crucially, this is distinct from any associated momentum transfer or other physical interventions involved. If you’re not familiar with the “double slit experiment”, it’s a great toy model to get up to speed on these ideas. Feynman explains it clearly in a series of lectures. You can also check out this lecture we gave about an analogue - albeit conceptually simpler - set up.
This was put forward in his 1957 Doctoral Thesis “On the Foundations of Quantum Mechanics.”
Here we mean the one the requires the least amount of inductive knowledge or assumption.
In practice there is always the quantifiable uncertainty of measurement, which can enjoy a profound amplification in certain systems with nonlinear dynamics. The absolute certainty of Classical Mechanics should be thought of as Platonic construction, one that simply does not exist in Quantum Mechanics.
Roger Penrose also has some thoughts on this, suggestions a relationship between measurement and gravitation, the quantum theory of which has flummoxed physicists.
As a personal note, I was lucky enough to see Weinberg speak about these ideas. His speaking style was cogent to the point of comforting, and inspired many of us in the audience to finally consider this a discussion worth having.
Which makes the outlandishly false statements like those tucked in to Google’s Willow Announcement so frustrating. Hartmut Neven - the Lead of Google’s Quantum AI program and purported author of the document - should know better.
Although the implications associated with Quantum suicide, a Schrödinger’s Cat-like thought experiment, help demonstrate its a priori absurdity.
Which make’s it’s explicit appearance in Google’s Willow announcement all the more confusing.
Per Pew Research, public trust in scientists remains fairly high, although it is declining. For a more nuanced take with some interesting historical background, see Rowland, et. al.