Precedent and the Laws

In the AI Story

The orthodox view in physics is Platonic: the laws of nature exist outside of time, as eternal mathematical truths that governed the universe's first moment and will govern its last. The universe obeys these laws the way a computer executes a program — the rules are fixed, the initial conditions are given, and everything that follows is determined by the combination. Smolin argues that this view is incoherent. If the laws exist outside of time, then the question of why these laws rather than others has no answer within physics — it requires an appeal to a Platonic realm that exists independently of the physical universe. This is metaphysics dressed as physics, and Smolin rejects it on the grounds that physics should be self-explanatory.

The principle of precedence offers an alternative. The laws of nature are habits — regularities that have emerged through repeated interactions and that continue to evolve as the universe itself evolves. The analogy to common law is structural, not merely illustrative. In common law, a dispute arises, a judge resolves it, and the resolution becomes a precedent governing subsequent similar disputes. Over centuries, the accumulation of precedents produces something that looks like a body of law — a set of principles that appears to have existed all along. But the appearance is retrospective. The principles were not there first. The cases were there first. The principles crystallized from the cases.

Smolin proposes that physical law works the same way. The first occurrence of a particular type of interaction — say, the formation of a particular kind of molecular bond — has no precedent. Its resolution is genuinely open, constrained only by the broadest features of the preexisting physics. Once the interaction has resolved in a particular way, that resolution establishes a precedent. Subsequent similar interactions are governed by the closest available precedent. Over cosmological time, the accumulation of precedents produces regularities so consistent that they appear to be timeless laws — the way an established legal principle appears to be a timeless truth when it is in fact the residue of centuries of cases.

The implication for the AI transition, as Segal develops it in Chapter 9, is that every institutional structure built now to govern AI is a precedent in exactly Smolin's sense. The EU AI Act is a case being decided for the first time. Its resolution — its specific provisions, its categories, its enforcement mechanisms — will propagate forward through every subsequent attempt to govern intelligent systems, shaping outcomes that its original drafters cannot foresee. The eight-hour day was a precedent established during the electrification transition. The research university was a precedent established during the printing transition. Both precedents shaped the governance of technologies their original designers could not have imagined. The precedents now being established around AI will do the same.

Origin

Smolin developed the principle of precedence in papers co-authored with Marina Cortês beginning around 2014, building on his earlier arguments for the reality of time and the evolution of laws. The principle draws on ideas from Charles Sanders Peirce (whose 1891 paper 'The Architecture of Theories' proposed that laws of nature are habits that have evolved) and from the legal philosophy of the common law tradition.

Key Ideas

Laws as habits. The regularities physicists call laws are not eternal truths but crystallized precedents established through past interactions.

Novel situations are open. Configurations with no precedent in cosmological history have genuinely undetermined outcomes, constrained only by broad structural features.

Resolutions propagate. Once a novel situation resolves in a particular way, that resolution becomes a precedent governing subsequent similar situations.

Precedents accumulate. Over time, the residue of past resolutions produces regularities so consistent they appear to be timeless laws.

Institutions are precedents. The governance structures established during technological transitions are precedents in exactly the same sense, shaping the resolution of situations their designers cannot anticipate.

Debates & Critiques

Critics argue that the principle of precedence cannot be formulated with the mathematical rigor that physics requires, and that the analogy to common law is illustrative rather than explanatory. Smolin and Cortês have developed mathematical frameworks for implementing the principle in specific contexts, including quantum gravity and cosmology. The debate remains genuinely open — which, Smolin would note, is itself an instance of the principle in operation.