Source
Deutsch & Marletto (2015), LIT-1663, §1, §2 and footnote 1; read in NOTE-1219.
What was actually shown
Principles constrain laws. The conservation of energy–momentum is the model. “It is a principle—a law of physics that constrains other laws rather than the behaviour of physical objects directly” (§1). Principles “do not make direct assertions about the outcomes of measurements” but are testable: “a principle P is refuted if some law violating P survives experimental tests, whereas all rival laws conforming to P are refuted” (§2). Footnote 1 gives the pattern: beta decay kept energy conservation by way of the neutrino, and parity invariance was refuted “in just that way”.
Constructors are realized and approximate. The constructor is the agent “whose defining characteristic is that it remains unchanged in its ability to cause the transformation again” (§2). “No perfect constructors exist in nature. Approximations to them, such as catalysts or robots, have non-zero error rates and also deteriorate with repeated use.”
Possibility needs no constructor. A task is possible “if the laws of nature impose no limit, short of perfection, on how accurately A could be performed, nor on how well things that are capable of approximately performing it could retain their ability to do so again” (§2). Subsidiary theories must measure the accuracy of approximate constructors “(including its own maintenance)”. “Constructor theory is the theory that the (other) laws of physics can be expressed without referring explicitly to constructors.”
Role, not kind. “Loops are excluded, because a substrate on a loop is a constructor” (§2). Whether a system is a constructor or a substrate depends on where it sits in a network of tasks.
What this does not say
- That principle I is true: that every law can be stated as possible and impossible tasks. That is the conjecture of the whole programme, and the paper does not argue it beyond its uses.
- That a law of possibility is a constraint “realized and maintained by something”, or that it is not. The paper treats possibilities as facts about the laws and says nothing about what realizes a law. Whether this makes it a case of a constraint realized by nothing is CLAIM-477’s question, not the paper’s.
- That a constructor is a constraint in Montévil and Mossio’s sense. The definitions are alike, but constructor theory has no time: “the notion of time does not appear” (§2). Its persistence is a limit over repetitions, not conservation at a time scale.
- Anything about loops beyond the one sentence. Closed networks of constructors are not treated.