Here I part ways with Rovelli

© Sidney Harris, The American Scientist, 1977. 

My endorsement, promotion and defense of physicist Carlo Rovelli's Relational Interpretation of quantum mechanics has been very overt and public for years, on Scientific American and other publications. I have also never hid my personal liking and admiration for Rovelli as a scholar and a person: I find him exceptionally thoughtful and open, a bit of a renaissance man, something we so thoroughly miss in a world that often takes its cues from immature nerds passing for intellectual wizards—incomplete human beings who have very narrow relationships with life and themselves, but who happen to excel in fashionable niches or be good at rhetoric.

None of that has changed. I still hold Rovelli in the highest regard and have profound respect for him and his output. But a consequence of this very respect is that I cannot overlook recent output of his with which I also profoundly disagree. The latter is what this post is about. I have made criticisms of Rovelli's latest commitment to certain philosophical ideas in recent interviews and discussions, so I think it is appropriate that I summarize those criticisms in one go-to place. Unlike those about the work of other people I have criticized in this blog, the assessment below—I insist—comes from a place of respect and admiration, not of scorn or patronization.

Rovelli and I are in full agreement when it comes to our view of the nature of physical reality: there is no absolute world of tables and chairs with defined mass, position, momentum, etc., out there, but instead an entirely relational world. The observable properties of all physical systems are entirely relative to the particular vantage point of the observation. Measurement doesn't merely reveal what the properties of a physical system already were immediately prior to the measurement, but brings those properties into existence. In summary, the physical world has no standalone reality. Both Rovelli and I concur that this is the inevitable conclusion from quantum theory and the overwhelming experimental confirmation of its predictions over the past 42 years or so. (Unless, of course, one believes in a de facto infinitude of real physical universes popping up into existence every de facto infinitesimal fraction of a moment, for which we have precisely zero empirical evidence; I believe both Rovelli and I dismiss this alternative as little more than silly fantasy.)

In his original Relational Quantum Mechanics paper of 1996, Rovelli defends the conclusions of quantum mechanics discussed above, but explicitly and deliberately refrains from exploring their philosophical implications. I, on the other hand, am on record—both the popular and academic records—deriving precisely those implications. In my view, if the physical world has no standalone reality and is entirely relational, then there necessarily is a deeper, by definition non-physical but absolute (in the sense of not being relative) layer of reality that grounds the physical world, and of which the physical world is but a measurement image akin to a set of dials. I've known for a while now that Rovelli isn't comfortable with this conclusion of mine, but neither did I expect or require him—as someone approaching the problem from an eminently scientific perspective—to agree with my philosophical exploration of the topic.

Recently, however, Rovelli seems to have gone all the way into philosophical territory. Am I bothered that a scientist is making an incursion into philosophy? Absolutely not! Some scientists do philosophy while believing that they are doing science; that kind of cluelessness is dangerous and reprehensible, but that's not Rovelli's case at all. Perhaps atypically amongst scientists, Rovelli has clarity regarding the difference between science and philosophy and displays great care and thoughtfulness in treading on the latter. So I think it is fantastic that he is daring to do so and wholeheartedly welcome his foray. At the same time, entering philosophy territory does—of course—expose him to hopefully healthy and constructive criticism. This is my intent with the present post.

What Rovelli seems to be now saying is that, although the physical world is constituted of no more than relationships, there is no underlying, non-physical world to ground those relationships. This is problematic for a number of reasons. For one, it immediately runs into infinite regress: if the things that are in relationship are themselves meta-relationships, then those meta-relationships must be constituted by meta-things engaging in relationship. But wait, those meta-things are themselves meta-meta-relationships... You see the point. It's turtles... err, relationships all the way down.

This is surely bad enough, but it isn't the worst part. The worst is this: to speak of pure relationships without non-relational entities to constitute and ground those relationships is literally meaningless, in a semantic sense; there is just no discernible meaning pointed to by the words in this claim, even though the claim itself can be articulated in language. The issue here is analogous to the Cheshire Cat's grin, which stays behind after the Cheshire Cat disappears: there is no meaning in this statement, even though Lewis Carroll was able to articulate it in language, to great effect.

Let me try to illustrate this with an example: movement is a prime instance of a relational phenomenon, one which Rovelli himself uses in his original 1996 paper. Movement, indeed, is always relative: if you are sitting inside a high-speed train, relative to you the train is not moving; but relative to someone standing on a platform, the train is moving at high speed. Movement is relational. With this example in mind, Rovelli essentially maintains that the entire physical world is like movement; it's not made of things with standalone reality, but of relationships. Up until this point I agree wholeheartedly with him, for the theoretical and experimental results simply prove this to be the case. However, Rovelli now proceeds to deny that there is anything that moves. So we end up with a world in which everything is movement but there is nothing that moves. Is this coherent? Does this even have any meaning, in a semantic sense, beyond the words themselves?

How does Rovelli justify this rather surprising proposition? He cites 3rd-century Indian mystic Nāgārjuna, interpreting the latter's writings to mean that there is no ultimate essence to reality except emptiness. So the world is made of movement, although there is nothing that moves, because ultimately the world is empty; it's made of nothing. This surely would sound great in a late-romantic poetry book, but is it reasonable when taken literally? Does it make any explicit sense? After all, when I look around I do see a lot going on. That I deny naive realism doesn't entail or imply that I deny the obvious existence of something.

Although I think and work mostly under the value system of the Western Enlightenment—which takes objective, explicit, unambiguous, logically consistent, conceptually clear, empirically substantiated reasoning to be the reliable path to truth—I am known to admire Indian and Eastern philosophy in general as well. They embody a different avenue to knowledge: that of meditative introspection and self-inquiry, a subjective—as opposed to objective—path of exploration. Kierkegaard referred to the exponents of these two paths as 'geniuses' and 'apostles,' respectively, highlighting their differences.

Personally, I think both paths have their merits and are complementary. I myself have adopted both paths in different works. Although the majority of my output is based on objective reasoning and evidence, I've treaded the subjective path in e.g. my book, More Than Allegory. However, I don't think it is valid to mix and match these paths in the course of defending any particular point of view, because doing so is blatantly inconsistent; it's a way to indulge in confirmation bias. Allow me to elaborate.

Rovelli takes a purely objective path to the conclusion that the physical world is entirely relational. He uses explicit, conceptually clear logical reasoning and empirical evidence to do so. He goes where this reasoning and evidence take him, all the way until a point where the inevitable implication is something he doesn't seem to like: that there must be a deeper, non-physical and non-relational layer to reality, which grounds the relationships that constitute the physical world, giving semantic meaning to the very word 'relationship.' From that point on, Rovelli arbitrarily abandons all post-Enlightenment epistemic values and switches to a vague, ambiguous, hand-waving, second-hand appeal to the mystical insights of someone who is no longer around to clarify what he meant. Never mind that the result is a peculiar Frankenstein monster, neither objective nor subjective; that Rovelli managed to avoid a conclusion he doesn't like—he describes how relieved he was upon reading Nāgārjuna, because the latter freed him from the pressure of having to find out what the underlying essence of reality is—seems satisfactory to him.

It's far from satisfactory to me. The paths of the 'genius' and the 'apostle' are complementary in the sense that, when both are applied in an internally consistent manner and lead to the same conclusion, we get a particularly satisfying kind of reassurance that we are on to something. But switching between these two modes in the course of making a point is entirely akin to changing the rules of the game while it's being played: it's cheating. When Rovelli does this, he puts his subjective preferences ahead of an objective inquiry into nature, and abandons the post-Enlightenment epistemic values that he has been known to champion. We get Rovelli the mystic, the apostle, dressed in a lab coat. This is not okay, not only because it isn't honest—and by this I don't mean that Rovelli is being malicious or deliberately deceptive, just that he seems to be deceiving himself and inadvertently misleading his audience, which has come to expect level-headed objectivity from him—but also because it leads to a literally meaningless conclusion: that the world is made entirely of movement, although there supposedly is nothing that moves.

Not only is it internally inconsistent to mix and match objective and introspective modes, introspective insights are also well-known to be largely ineffable. Therefore, when put to words, they almost invariably fail to capture the salient nuances of the intended point. That's why whole schools of thought in the East (and some in the West) have entirely given up on trying to explain what reality is. Instead, their writings are what Peter Kingsley refers to as forms of 'Μῆτις' (Mêtis) or 'incantation': they are meant not to describe reality, but to trick you into seeing it for yourself; to make you 'trip over' your own conceptual narratives and finally see through them. In weaving these incantations, sages will freely and liberally use contradiction, cognitive dissonance, metaphor, sleight of hand, shocking absurdities pronounced with a solemn face, deliberate inconsistencies, lies and, sure enough, even true statements mixed in; only the desired effect counts (Nisargadatta Maharaj, the Eastern sage I admire the most, contradicts himself several times in each page of I Am That). And I believe this is all epistemically valid because it is entirely consistent with the stated goals. The problem only arises when one fishes out a particular statement from the mystical writings of someone else, interprets it literally—as if it had been written by an 18th-century European philosopher in the finest Apollonian tradition, as opposed to a 3rd-century Indian sage—and then uses it as an arbitrary bridge to change the course of what is otherwise meant as an objective argument. This just doesn't work and should be viewed with at least great suspicion.

Rovelli has been one of the greatest exponents of the post-Enlightenment epistemic values in the 21st century. I regret that he now seems to be so breezily departing from those very values, so as to acquiesce to his own subjective preferences about what nature should or should not be. Subjective, introspective paths of inquiry may even be the royal road to truth, but their value rests precisely in direct, personal insight. I would find it laudable if Rovelli decided to engage in self-inquiry and the whole arsenal of meditative techniques, in order to directly experience the nature of reality for himself; he might then find out that that 'emptiness' is mind at rest, a subject without objects, pregnant with the potential for every conceivable internal relationship. But fishing out statements from someone else's introspective insights is consistent neither with objective reasoning nor with the schools of direct knowing, for the words of the latter were never intended to be used in this manner (again, they were meant as 'incantations,' not descriptions). Instead, it's a disservice to both and dilutes the credibility of the otherwise priceless legacy Carlo Rovelli has been methodically building for decades.