Bohr meets Rovelli: a dispositionalist account of the quantum limits of knowledge

I begin by examining the question of the quantum limits of knowledge by briefly presenting the constraints of the theory that derive from its mathematical structure (in particular the no-go theorems formulated by von Neumann and Kochen and Specker). I then argue that these theorems reflect on a formal level those practical and experimental settings that are needed to come to know the properties of physical systems. In particular, I discuss some aspects of this relationist and contextualist conception of reality by comparing, in their apparent diversity, Bohr holistic and Rovelli relationist interpretation of the formalism, that deep down share a unifying metaphysics of dispositions and propensities. Both interpretations are based on the widely shared fact that quantum mechanics does not describe previously definite quantities. In the final part I show that, as a consequence of a relationist and perspectival approach to quantum mechanics, the quantum state of the universe regarded as an isolated system cannot be known in principle, so that the universe must be described from within by dividing it into two arbitrary parts. This is in fact the only way in which any two systems can exchange information by being physically correlated.

💡 Research Summary

The paper investigates the “quantum limits of knowledge” from historical, sociological, mathematical, and physical perspectives, ultimately arguing that these limits are best understood through a dispositionalist, relationalist metaphysics shared by Niels Bohr and Carlo Rovelli. The introduction contrasts two 19th‑century attitudes toward scientific knowledge: the pessimistic “ignoramus‑ignorabimus” stance of Emile Du Bois‑Reymond and the optimistic “we must know – we will know” slogan of Hilbert (and earlier optimism of Lord Kelvin). The author notes that such attitudes re‑emerged in the late 20th century, for instance in Stephen Hawking’s claim that a final theory of physics might soon be achieved.

Moving beyond sociological explanations, the core of the paper focuses on the internal logical structure of quantum mechanics. Two major no‑go theorems are examined: von Neumann’s proof of the impossibility of hidden‑variable extensions that reproduce all quantum predictions, and the Kochen‑Specker theorem (1967) which shows that non‑contextual value assignments are incompatible with the Hilbert‑space formalism for dimensions ≥ 3. The author explains von Neumann’s linearity assumption for hidden‑variable values, why it fails for non‑commuting observables, and how Kochen‑Specker formalises the contextuality of quantum properties. Both theorems undermine the classical principle of “value definiteness” – the idea that every physical quantity possesses a pre‑existing, measurement‑independent value.

The philosophical import of these results is then linked to Bohr’s holistic interpretation and Rovelli’s relational quantum mechanics. Bohr famously insisted on the inseparability of the atomic system and the measuring apparatus, claiming that the conditions defining a phenomenon are part of the phenomenon itself. Rovelli, in a different language, argues that all quantum properties are relational: they acquire meaning only through interactions with other systems (including observers). Despite their apparent differences, both positions share a dispositionalist metaphysics: quantum systems possess dispositions or propensities that are actualised only in specific measurement contexts. The paper draws on David Lewis’s distinction between intrinsic and extrinsic properties, showing that quantum observables behave like extrinsic, context‑dependent properties rather than intrinsic ones.

A crucial extension of this relational view is the claim that the quantum state of the entire universe, regarded as an isolated system, cannot be known in principle. Information exchange requires physical correlation between subsystems, which in turn demands a division of the universe into at least two parts and an interaction that defines a measurement context. Since any “outside” observer would itself be part of the universe, there can be no external, perspective‑free description of the universal wavefunction. Consequently, a complete, observer‑independent knowledge of the universe is ruled out by the same dispositionalist framework that limits knowledge of individual systems.

In the final section the author argues that the quantum limits of knowledge are not merely mathematical curiosities but reflect a deeper epistemic structure: the world is not composed of pre‑determined properties awaiting discovery, but of potentialities that become actualised only through relational interactions. This view reconciles the formal no‑go theorems with Bohr’s holistic emphasis and Rovelli’s relationalism, presenting a unified picture in which the “limits” are precisely the conditions under which quantum properties acquire meaning. The paper thus concludes that quantum mechanics imposes a principled, context‑dependent bound on what can be known, and that this bound is rooted in the dispositional, relational nature of physical reality itself.