Imagine you are in a train moving smoothly at constant speed. You throw a ball straight up, and it comes straight back to your hand. From your point of view inside the train, the ball just went up and down. But someone standing on the platform outside sees the ball tracing a parabola in the air because the train is moving. Who is right? Both. Velocity is not something the ball has “by itself.” Velocity is always relative to something else: to you in the train or to the person on the platform. This is a very classical idea—Galileo already understood it in the 17th century, and Einstein made it the cornerstone of special relativity. Properties like velocity, kinetic energy, or even the shape of rapidly moving objects are relational: they only make sense when we specify “relative to what.”
Classical physics, however, still believes that many other properties are absolute and belong to things in themselves. A stone has a definite position and a definite mass no matter who looks at it. A thermometer has a definite temperature. The universe as a whole has a definite configuration that exists independently of any observer. We just happen not to know it completely, but in principle it is there.
Quantum mechanics, especially in the interpretation developed since the 1990s by Carlo Rovelli under the name Relational Quantum Mechanics (RQM), says: no. All physical properties—not just velocity—are relational in exactly the same way. There is no absolute fact of the matter about the position of an electron unless we specify relative to which other system we are asking the question. The moon is not simply “there” when nobody looks; its being there is always its being there relative to something else that interacts with it.
From Classical Relations to Quantum Relations
Rovelli starts from the simple observation we began with: classical physics already contains relational quantities (velocity, angular momentum, electric potential, etc.). What quantum mechanics teaches us is that the non-relational, “absolute” quantities we thought remained—position, momentum, spin component—are actually relational too. The famous wave function ψ, instead of describing the objective state of a system in itself, describes the information that one system has about another system.
When a photon hits a molecule in your retina and you say “I see the moon,” the quantum state of the moon-photon-you system has changed. The moon now has definite properties relative to you (a definite approximate position, brightness, etc.), but those properties are not written somewhere in the moon independently of the interaction. Another observer—a neutrino passing by that barely interacts with the moon—has almost no information, so for that neutrino the moon is still largely indefinite, smeared out in quantum superposition.
This is the central claim of RQM: any physical event is an interaction, and any property is the manifestation of one system affecting another. There is no privileged “God’s-eye view” that sees everything absolutely.
What Is a Thing?
If everything is relational, what exactly is a “thing”? In everyday language and in classical physics we imagine things as little lumps of stuff that possess properties on their own and then enter into relations. Rovelli inverts the order. Relations come first. A “thing” is just a temporary node in a web of interactions—a bundle of relations that remains approximately stable for some time from the point of view of other bundles.
Think of a chair. Classically, the chair is made of atoms that have definite positions whether you look or not. In RQM the atoms only acquire definite relative positions when some system (your eye, a photon, the floor, the air molecules) actually interacts with them. The chair is a consistent pattern of actual and possible interactions. It is real, but its reality is always relative to something else.
This leads to a strange but consistent conclusion: there are no things-in-themselves. There are only relations. The electron is not a little ball hiding somewhere “really”; the electron is the way it affects detectors, atoms, photographic plates, etc. Its being is its relating.
Can There Be Relations Without Things?
Philosophers have asked for centuries: if everything is relation, what do the relations relate? Must there not be some underlying substance, some substratum that stands in relation? Rovelli’s answer, echoing Spinoza and echoing certain strands of Buddhist philosophy, is no. Relations do not need an underlying “thing-stuff.” Relations can be fundamental.
In the mathematical formalism of quantum mechanics this is almost obvious. The basic object is not a state vector assigned to a single system in isolation; the basic objects are correlations between systems, described by the density matrix or by the amplitudes for joint outcomes. The Hilbert space is always a tensor product space of at least two systems. There is no consistent way to assign a pure state to a single isolated system if it has ever interacted with anything else—and in practice nothing is ever completely isolated.
So the ontology of RQM is an ontology of events-of-interaction, not of objects-in-isolation. A relation without prior things is not only possible; it is what the theory forces upon us.
Why the Totality, the “Whole Universe,” Cannot Be Conceived
In classical physics we dream of a final theory that describes the state of the entire universe from t = –∞ to t = +∞. Even if we never reach it, the idea makes sense: the universe is a big box containing everything, and its total state is the list of all positions and velocities inside the box.
RQM says this idea is incoherent. Because every property is relational, to describe “the universe” we would need to say how it is relative to something else. But there is nothing else outside the universe. The universe is not relative to anything; therefore the universe as a whole cannot have a quantum state in the same sense that a subsystem can.
Rovelli likes to quote the physicist John Bell, who said that the word “measurement” should be replaced by “interaction with an external system,” but when the system is the whole universe there is no external system left. The notion of a complete, observer-independent description collapses. There is no “view from nowhere.” Every actual description is from the point of view of some subsystem relative to which events acquire definiteness.
This is not solipsism. The world is still out there, stubbornly real. But its reality is fragmented into a multitude of relative facts. My moon is definite relative to me; your moon is definite relative to you; the neutrino’s moon is almost completely indefinite. None of these perspectives is privileged.
Is This Physics or Philosophy? And Why Does It Feel Moral?
Many physicists dislike RQM precisely because it sounds “philosophical.” They prefer interpretations (Copenhagen, Many-Worlds, Bohmian mechanics) that try to keep an absolute, observer-independent description somewhere, even if it is hidden or inaccessible. Rovelli’s answer is that the discomfort comes from clinging to an old metaphysics of substance that quantum mechanics has simply outgrown.
But the issue runs deeper than physics. The habit of thinking that reality must ultimately consist of little things-in-themselves that then relate is the same habit that makes us think human beings are atomic individuals who then enter into social relations. If relations are primary, then a person is not a soul trapped inside a body that then decides to connect with others; a person is the web of relations—language, love, conflict, care—that constitute her from the beginning.
Rovelli himself often points out that accepting the relational nature of reality has ethical consequences. If there are no absolute facts, only facts relative to interactions, then the way we interact literally co-creates the properties of the world. Responsibility is not something added later; it is woven into the fabric.
Is Duality Ontologically Prior to Singularity?
In the end RQM suggests a very austere ontology: the world is made of interactions, and an interaction is irreducibly something that happens between at least two systems. Singularity—a lone thing existing by itself—is impossible. Every quantum event is a two-place relation (at minimum). The duality of observer-and-observed, system-and-environment, is not an artefact of our ignorance; it is the fundamental grain of reality.
This reverses the usual philosophical order. Western metaphysics since Parmenides has privileged “Being” as One, and treated relations as secondary. Even when relations were taken seriously (Leibniz’s monads that are “windowless” yet harmonized, or Whitehead’s process philosophy), the lone monad or the single actual occasion still came first logically. RQM says no: the isolated monad is the abstraction. The interaction is concrete and prior.
There is no description of a single system without reference to another. The tensor product comes before the individual factors; the relation comes before the relata. In the beginning was the relation.
Conclusion
Relational Quantum Mechanics, as Carlo Rovelli has articulated it over three decades, is not an exotic add-on to quantum theory. It is the consistent application of the same insight that made us understand velocity is relative: all properties revealed by quantum interactions are relative in exactly the same way. Things dissolve into bundles of relations; the universe as a whole cannot be grasped because there is no external standpoint; and reality turns out to be an immense web of encounters rather than a collection of solitary substances.
This is unsettling, because we are used to thinking of ourselves as separate spectators looking out at an objective stage. But it is also liberating. If reality is relational all the way down, then every interaction matters; every measurement, every glance, every conversation is a place where the world acquires definite properties. We do not just live in the universe. With every act we co-create the relative facts that are, as far as physics can tell, all the facts there are.
A brief bibliography for those who want to find out more:
Books
- Carlo Rovelli – Helgoland: Making Sense of the Quantum Revolution
- Italian edition: 2020; English edition: 2021
- Explains relational quantum mechanics for a general audience, under 200 pages, almost no equations.
- Carlo Rovelli – Covariant Loop Quantum Gravity (2014)
- Part III, Chapter 8: The relational strategy
- Shows how RQM ideas connect to quantum gravity.
Foundational Papers
- Carlo Rovelli – “Relational Quantum Mechanics”
- International Journal of Theoretical Physics 35, 1637–1678 (1996)
- The original paper introducing RQM.
- arXiv preprint
- Carlo Rovelli – “An Argument Against the Realistic Interpretation of the Wave Function”
- Foundations of Physics 46, 1454–1467 (2016)
- arXiv preprint: 1508.05533
Philosophical / Review Articles
- Federico Laudisa – Open Problems in Relational Quantum Mechanics
- arXiv: 1710.07556 (2017)
- Discusses open conceptual and foundational issues in RQM.
- arXiv link
- Bas C. van Fraassen – “Rovelli’s World”
- Foundations of Physics 40, 390–417 (2010)
- Philosophical discussion of RQM and its relation to relativity and quantum mechanics.
- PDF link
- Stanford Encyclopedia of Philosophy – “Relational Quantum Mechanics”
- Authors: Federico Laudisa & Carlo Rovelli (substantive revision, 2019)
- Excellent, accessible philosophical overview.
- SEP link
- Claudio Calosi & Timotheus Riedel – “Relational Quantum Mechanics at the Crossroads”
- Foundations of Physics, 2024
- Recent review of philosophical and foundational aspects of RQM.
- PhilPapers link
