Will Constructor Theory REWRITE Physics?

Constructor theory is gaining attention as a potential “rewrite” of physics by shifting the foundation from dynamical equations to a cleaner set of rules about what transformations are possible or impossible. Instead of treating physical laws as step-by-step mechanisms that evolve numbers over time, the framework treats reality as a network of “tasks” (input-to-output transformations) that are either allowed or forbidden for a given physical system. That change matters because it promises broad conclusions about nature without needing the full, detailed equations of motion—an advantage when the governing dynamics are unknown, such as in the search for quantum gravity.

In the mechanistic picture that dominates mainstream physics, a system is described by quantitative variables—temperature, pressure, position, velocity—and then equations predict how those variables change. Those laws are considered correct when predictions match reality, and the approach has been extraordinarily successful. Yet unification has stalled: quantum mechanics and general relativity can each be derived from deeper principles, but the missing link between them has resisted nearly a century of attempts to build a single dynamical “theory of everything.” Constructor theory, developed by David Deutsch and Chiara Marletto at Oxford University, aims to bypass that bottleneck by replacing “how things evolve” with “what can happen.”

The framework’s core objects are counterfactuals—meta-facts about whether a particular task is possible or impossible for a constructor, a system capable of performing physical transformations. Inspired by information theory and quantum computation, it borrows the idea of a universal constructor from John von Neumann: a general system that can carry out any computation or physical task. Deutsch’s version of the program leans on the claim that if a quantum computer can simulate any process in physics, then physics can be expressed in the language of quantum computation. In this view, conservation laws and other principles don’t just constrain trajectories; they rule out entire classes of output states.

A simple illustration uses Newton’s falling apple. In a constructor-theory framing, the question becomes which tasks the apple can perform given its initial state. Remaining hovering in place is forbidden by general relativity because free-falling bodies must follow spacetime geodesics. Turning into gold is forbidden by quantum mechanics and energy conservation. The only allowed task is the one consistent with the familiar prediction that the apple falls.

Constructor theory also targets problems where mechanistic reasoning becomes cumbersome. Perpetual motion machines of the first kind can be ruled out more generally by combining conservation of energy with the second law of thermodynamics, rather than analyzing every specific device’s torque and power balance. More ambitiously, Marletto uses the framework to propose an experiment that tests whether gravity is quantum. By treating entanglement as a task possible for quantum “superinformation media” but impossible for purely classical information media, she argues that if gravity can induce entanglement between spatially separated qubits, then gravity must have quantum properties. The selling point is that the argument doesn’t depend on a particular dynamical theory of quantum gravity.

The episode then pivots to audience questions about quantum tunneling, clarifying that tunneling is defined by motion across a barrier where the wavefunction decays exponentially, not merely by “being somewhere else” due to wavefunction spread. It also addresses how long tunneling takes, noting that crossing-time definitions since the 1980s suggest it’s not instantaneous and can be faster than light only in ways that don’t straightforwardly imply causality violations. The discussion ends with speculative reflections—simulation-like analogies and causality-as-statistics—highlighting how far people’s intuitions can drift from the underlying physics when they try to map quantum behavior onto everyday notions of motion and distance.