There are over a dozen different interpretations of quantum mechanics, not including the view that we shouldn’t be trying to interpret the theory at all, the “shut up and calculate” school. We could put that more elegantly perhaps as where we cannot speak, there we should be silent.
In his recent book Einstein’s Unfinished Revolution, Lee Smolin frames the issue as primarily one between scientific realism and what he calls anti-realism, though non-realism might be better. He argues that generations of scientists have been taught that quantum theory is, from a philosophical perspective, anti-realist. The established view, the Copenhagen interpretation, developed under the influence of Niels Bohr, is self-consciously anti-realist. There are several different versions of anti-realism in this context. One version is operationalism, which approaches quantum mechanics as a set of procedures for interrogating atoms rather than as a model of a natural system. Another version proposes that the properties of a quantum system are created in the moment we interact with it through observation.
Subsequent competing interpretations put forward by Louis de Broglie, David Bohm and Hugh Everett among others are attempts to provide a realist interpretation. Smolin acknowledges that these interpretations have not been successful, but continues to believe in the project, and puts forward his view of the principles that would ground a successful realist interpretation.
We need to be careful here to stay the right side of the line that separates philosophy from science. Realism is a necessarily philosophical theory and one that is easily misunderstood. I think that Smolin in fact does misunderstand it. He argues that realism asks for two things: firstly, that the way the natural world functions is independent of the way the human mind works and, secondly, that it is nonetheless possible for the human mind to comprehend the natural world sufficiently to explain its past and predict its future. Erwin Schrödinger in his essay Mind and Matter called these the hypothesis of the real world and the principle of the understandability of nature.
The problem is that if you press the second requirement too far, as far as, for example, invoking Leibniz’ principle of sufficient reason, as Smolin does, then you compromise the first requirement. If the natural world is constrained to function in such a way that there is a rational explanation that we can understand conceptually for everything that might have turned out differently in nature, then the natural world ceases to be properly concept and mind independent.
The hypothesis of the real world is the view that nature and the conceptual structures of science are independent. If nature is concept independent, then the degree that any observer can successfully understand nature through the application of scientific models will be accidental and contingent rather than necessary. How could the relationship be a necessary one, if the way nature works is independent of the conceptual model that the observer is deploying?
The hypothesis of the real world, or the idea that there is a real world out there, is, ironically, a non-realist proposition, and scientific realism is actually a version of idealism, the philosophical theory that, in some way, the rational ordering of the world is not an accident. The anthropic principle sometimes invoked in the arguments for string theory is another version of the same idealist impulse.
The debate between realism and non-realism is a philosophical question so the tests that can be applied are different from the tests that can be applied to scientific hypotheses. The way I would approach it is to look at the costs involved. What I mean by this is that realism comes with additional costs. If we argue that the rational order that we discover in the world is in some way necessary, we should then ask some further questions. What order does nature have to have, and why would it have more order than is necessary? What mechanism is in place to ensure that nature doesn’t get out of alignment with what can be conceptualised? For nature to be understandable in a way that is not contingent and accidental, either nature or concept or both must be constrained by some mechanism to remain in alignment.
Non-realism is a simpler hypothesis than realism. Realism requires some large claims about the relationship between concepts and nature to be true. However, if concepts can apply to nature, what more is added by wanting the relationship to be a necessary one. The general applicability of mathematical and conceptual models is a function of their levels of abstraction and there is therefore no need to posit any other mechanism to link concept and world. The link can be accidental rather than necessary. We don’t need to posit any kind of necessary relationship between concept and nature to explain the success of this process.
As an example, if we think about the use of trigonometry to map the surface of the earth, we don’t need to assume that there is anything inherently triangular in the way the earth’s surface is configured. Trigonometry allows us to overlay any terrain with a pattern of virtual triangles and draw inferences about directions and distances.
I think what Lee Smolin is looking for in an interpretation of physics that is not only internally consistent but also externally consistent; that is, it meets some criteria of continuity with the interpretation of the observable world and that both interpretations are built on the same set of philosophical assumptions. My view is that the hypothesis of the real world coupled with the view that conceptual and mathematical models can be meaningfully applied to nature is sufficient to meet this objective.
Smolin, L. (2019). Einstein’s Unfinished Revolution. Allen Lane.
Schrödinger, E. (1958). Mind and Matter. Cambridge: Cambridge University Press.