The QBist Interpretation of Quantum Mechanics and the Future of Science Communication
- Philip Ball
- 2 days ago
- 10 min read
Philp Ball on an interpretation of quantum mechanics that takes an agent's actions and experiences as the central concerns of the theory
In June I participated in a small and rather unique meeting at Royal Holloway University just outside London, convened by physicist Chris Fuchs and mathematician Rüdiger Schack, the two principal proponents of the so-called QBist interpretation of quantum mechanics. The premise of the gathering—which included other QBists, some non-QBist quantum experts, a historian and philosopher of physics, and several science writers who have written on the topic (hence my presence)—was that Chris is tired of QBism being misrepresented in science communication and wanted to get us writers, at least, to do the job right. As I’d largely anticipated, the meeting turned out to have a somewhat different flavor.
It was a very unusual affair, in that it was a frank but friendly and fun exchange between the scientists and the science writers where both were on an equal footing. As Chris dissected some of the popular articles that had been written about QBism, we debated what the errors were, and sometimes whether they were errors at all, and had discussions about the realities of popular science publishing: writers don’t get to write or often even see titles, standfirsts, infographics, or cover words, are discouraged or sometimes even prohibited from showing copy to those they interview, and so on. We spoke about how the colloquial use of words is sometimes different from their technical use (such that, for example, it can cause difficulties if scientists start using familiar words in ways that don’t match their everyday use). We talked about the goals of science communication: is it to provide an impeccably accurate (in the eyes of the scientist) description of their work, or to convey a flavour of that to the general reader?
It was, in short, what felt like an immensely valuable exchange of a sort that too rarely happens. The very fact that Chris convened the meeting shows that he is someone who is keen genuinely to engage with science writers and does not take the view, occasionally evident in science, that scientists do the intellectual work and the job of the science writer is merely to transcribe it in accessible language. I remember realising, many years ago at a “science meets the media” gathering at the Royal Society, that this was exactly how one eminent British Nobel laureate saw things. They were of course grateful for the work that science writers did, the Eminence allowed, but wished sometimes that the writers would listen to them more carefully. I’m reminded of the scientist who recently said to me that I might benefit from becoming a postdoc in a lab to find out how science actually works, because it was apparently inconceivable to him that what I had written on his field came from an informed but different point of view than his rather than from ignorance that could be brought in line with more training. (Such comments stick in the mind precisely because they are so rare.)
What emerged from our discussions with Chris and colleagues was far richer. I think and hope it was useful to Chris as well as to we writers to come to see that the challenge in conveying QBism well is partly that it is so very deeply radical. If we start, as we generally must, from the position that it is one of many interpretations of “what quantum mechanics means”, it becomes very hard to give it its due. It represents an entirely new way of seeing the world, quantum or not, which uproots pretty much all of one’s often deeply held convictions. On one level, it amounts to “taking probability seriously”: to truly engaging with the long-standing and ultimately philosophical questions about what probabilities can and should mean. One might think this would be obligatory once the probabilistic nature of quantum mechanics is acknowledged, but most quantum physicists take for granted that probabilities are unproblematic: they just refer to the likelihood of seeing the various possible outcomes of experiments on quantum systems, right? Well, not quite, or not obviously.
Unpacked in this way by QBists, quantum mechanics’ probabilistic nature leads them to conclude that there are actually no physical laws constraining what can happen in the world. The future is genuinely open. So why do we apparently see physical laws in the regularities of nature? That discussion needs an article in itself, but of course QBists believe they have an answer.
This perspective also leads QBists to assert that the quantum states we assign to quantum objects are just expressions that quantify our beliefs about the object, which get updated in a Bayesian manner (the original meaning of the “B”, albeit no longer – it’s just “B”). This means that assignations of probability are informed by our prior beliefs and refined as new information becomes available. Perhaps most challengingly of all: because they are beliefs, they are necessarily personal. Outcomes of measurements/experiments are then experiences that the individual observer has, and are personal to them – they are not objective facts about “the world” that must be the same for all individuals.
Conveying what this implies accurately is difficult: the ideas are subtle and nuanced. QBism does not, for example, imply that there is no physical reality but just what is in each of our minds – it is not solipsism. QBism rejects realism, but does not deny that there is a world with particular characteristics that shape our experiences. Even in this much, I now sense Chris leaning over my shoulder and declaring “No, that’s not quite right!” But the key point is that QBism, as one participant said, “is playing a different game to the rest of science.” Try capturing that in a sentence or an infographic that explains how it differs from Bohmian mechanics or the Many Worlds view.
What this meant in practice was that often Chris’s suggestions for how things could have been better phrased in the articles he dissected led the science writers to object that no editor would allow those suggestions as they stood because they raise too many questions and seem to be unrelated to the ways quantum issues are generally discussed at all. You can’t, for example, throw in a word like “experience” without justifying what the hell it is doing there among talk of wavefunctions and entanglement.
This is a rather common experience for science writers: that scientists will offer them a “better” way of explaining things that the scientists thinks is fine because it uses no jargon, only for the writer to have to explain that no reader is going to be able to decrypt it with the necessary reading between the lines. Personally, however, such situations fuel my intense frustration at what is sometimes an excessive phobia in science publishing, inherited from journalistic norms more widely (where it exists often for very good reasons), of showing copy to the folks whose work you’re writing about. Especially in a topic like QBism—but in truth this applies to pretty much any area of science—it is all too easy even for the most diligent and well-informed writer to choose slightly the wrong words. Such problems could almost always be fixed with a little to-and-fro between the writer and the scientist, not necessarily by just inserting what the scientist suggests instead, but by negotiating good compromises with which everyone is happy. This does sometimes happen, but it is hindered by the insistence of some journals that such discussion should not happen, or will be subsumed by fact-checking by the editorial staff. Such interactions, however, would need to be conducted on the understanding that in the end it’s the writer’s (or editor’s) judgement, so that for example they can if necessary put their foot down and say “I’m not going to use those words”. In my experience, scientists almost always understand and accept this.
All this was very valuable. But what emerged in the meeting too is that it is typically not at all black and white whether a particular choice of words is good or not. There were, for example, some sentences about QBism that irked Chris but not Rüdiger, and vice versa. Sometimes scientists may not appreciate that what seems to them to be a straightforward error of fact in what a science writer writes might not seem that way even to others within their field.
But as we discussed in the meeting, the problems go deeper than that. Jo Marchant pointed out that misrepresentations of QBism are not confined to science writers, but can be even more egregious from other scientists. (Chris agreed!) I have occasionally encountered some bad popular science books in recent years, but I can confidently say that the worst—making bold but unsupported claims and awful factual errors, for example—are almost always from tenured professors and “experts”. There are several possible reasons for this—for example, editors assume the experts are right, and the experts also assume they are right and don’t get feedback from colleagues. But perhaps one of the most important reasons is that they can get away with it when science writers can’t. If scientists present a grotesquely one-sided account of an issue, they are merely presenting their “bold theory”; if science writers do that, they haven’t done their homework. If science writers routinely write wrong and dumb things, they’ll get pulled apart; if scientists do it, they still have tenure (and their colleagues probably don’t read their books anyway). It’s not to deny that science writers—perhaps even this one!—screw up, sometimes badly, to acknowledge that this “An expert pronounces” issue is a big problem in science communication. (There are of course also plenty of scientists who write excellent popular books and articles.)
Of course, such misrepresentations on quantum mechanics are not just about QBism. I have been railing for years about how, for example, the Copenhagen interpretation of quantum mechanics (itself not a well defined or unified thing) is routinely misrepresented by quantum physicists who prefer other interpretations. It was an utter joy, then, to see physicist Matt Leifer set out very clearly in his presentation the fact that the classic textbook accounts of quantum mechanics by the likes of Paul Dirac and John von Neumann are not the Copenhagen view, even though these are very commonly conflated. Likewise, Matt pointed out that it is not a good criticism of the Many Worlds view to complain that it has been so profligate in positing all those extra worlds—because it does not posit them. (I have made a point of saying this myself, although that has not prevented some Many Worlders from implying that I said the opposite.) Matt’s taxonomy of quantum interpretations was one of the most useful exercises of this sort I have ever seen, and should be required reading for anyone interested in quantum foundations. (The fact that Chris and QBism advocate John DeBrota objected to it showed precisely its value, in my view, as it will force folks to confront what their assumptions really are.) The crux of the Many Worlds view, by the way, is that it ascribes ontological reality to the wavefunction. There’s no reason one cannot do this—and indeed if one does, many worlds are not easy to avoid—but that assumption is not required by the quantum formalism and was shared by none of its originators: not Einstein, Bohr, Heisenberg, Born, Pauli, Schrödinger etc. So when Many Worlders insist that they are simply “taking quantum mechanics at face value”, this is the ultimate reason why that claim is plain false.
Another vexed issue is the idea that quantum superpositions place a system in “two states at once”. I have been waging war on this for years, sometimes to the bemusement of my colleagues, and so it was music to my ears to hear all the physicists present at the meeting condemn it as not merely incautious wording but as a fundamental misconception of quantum mechanics. Science writers (and, and perhaps especially, scientists writing popular books): please take note! There are short, accessible alternatives available for conveying superposition that do not do harm to the science.
These discussions reminded me of how scientists can sometimes be confident about the rightness of an idea or phrase while unaware that their colleagues are equally confident about it for wholly different reasons. I was reminded of this in some recent online discussion about “selfish genes”. There are various ways to interpret that metaphor. I can think of two quite different circumstances where I think the metaphor is apt, both referring to very specific aspects of genomics that make no claims about this somehow being a generic or default property of all genes. A shallow acceptance of the phrase can and does lead to some profound miscommunication and misunderstanding. What those and other debates have made apparent to me is that sometimes writers think and care about what words mean, and accordingly choose them carefully, much more than do scientists. In science, there’s sometimes a tendency to believe that, while technical terms need very precise definition, everyday words are crude tools that can be bandied around at liberty—or that can be appropriated for technical use like blank slates, free from any cultural baggage.
I suggested at the meeting that some if not many of the communication problems arise because there exist “mental attractors” to which we all (science writers, editors, scientists) get inevitably drawn unless we consciously resist. These seem to appear unbidden and are thereafter hard to escape. For example, in one article on QBism the editor had inserted, without the writer’s consent or knowledge, a pull quote along the lines that QBism asserts that the world is just a kind of illusion or hallucination. It says nothing of the sort, but this has become a popular cliché that makes (or is intended to make) readers say “oh, this blows my mind!”. It’s for this reason that scientists too need to be wary of popularizing such phrases. I was reminded too of how the existence of such attractors means that scientists are generally no better than general readers in constructing their understanding of the written word not from the words themselves but from a stock of preconceptions about what they think the writer means.
It's one thing being open to learning more, as a scientist, about how journalism works. But it’s to Chris’s immense credit that he has also been ready to take seriously what science writers say to him about the science. It seems that a years-long debate he has been having with science writer George Musser (who is far too smart to settle for vague explanations) has prompted Chris, Rüdiger and John to accept what they call the Musser Challenge and put together a paper outlining the QBist account of EPR correlations. (In QBism, entanglement is no big deal.) Such things reinforce my feeling that the best scientists, and indeed most of them, really don’t care about the credentials of people giving them feedback or engaging in scientific discussion, so long as it is useful to them.
I don’t know if this meeting delivered quite what Chris wanted, although I do hope and believe that it will make the science writers who were there less prone to infelicities when writing about QBism. But it convinced me that more interactions like this—good-faith exchanges in which both parties respect and want to learn from the other—would be very useful in all areas of science. I hope that can happen.
Philip Ball is a scientist, writer, and a former editor at the journal Nature. He has won numerous awards and has published more than twenty-five books, most recently How Life Works: A User’s Guide to the New Biology; The Book of Minds: How to Understand Ourselves and Other Beings, From Animals to Aliens; and The Modern Myths: Adventures in the Machinery of the Popular Imagination. He writes on science for many magazines and journals internationally and is the Marginalia Review of Books' Editor for Science and Senior Fellow of Scientific Integration and Natural Philosophy at Marginalia's Institute for the Meanings of Science. Follow @philipcball.bsky.social
