Episode Summary
Executive Summary: Sean Carroll and David Deutsch range from Everettian quantum mechanics to epistemology, constructor theory, and AI. Deutsch argues that human-like explanatory knowledge is a new physical force, that science progresses by good explanations rather than Bayesian credence updates, and that constructor theory may unify physics across scales. He is optimistic about AGI and suggests current quantum theory is likely incomplete.
Main Topics: Everettian quantum mechanics and Hugh Everett (Priority: 5/5): Deutsch recalls meeting Everett and explains how DeWitt and Everett’s technical understanding convinced him of the many-worlds interpretation. Monotony, progress, and the emergence of knowledge (Priority: 5/5): The conversation centers on Deutsch’s idea that major transitions in the universe—life, multicellularity, human explanation, enlightenment—appear after long periods of apparent stasis. Knowledge as a physical force (Priority: 5/5): Deutsch argues that explanatory knowledge has transformed Earth’s role in the universe, allowing humans to alter outcomes far beyond what ordinary physical processes would predict. Turing completeness and explanatory universality (Priority: 5/5): He distinguishes hardware universality (Turing completeness) from software universality (explanatory universality), claiming humans uniquely possess the latter. Constructor theory and the possible/impossible (Priority: 4/5): Deutsch describes constructor theory as a framework for expressing physics in terms of what transformations are possible or impossible, aiming to unify physics, biology, and thermodynamics. Critique of Bayesian epistemology (Priority: 5/5): He rejects the idea that knowledge is best modeled by probabilistic credences that increase with evidence, invoking the Popper-Miller theorem and the primacy of explanation. Limits of quantum theory and future revisions (Priority: 4/5): Deutsch says quantum theory and quantum field theory are likely incomplete and sketches alternative directions, including noncommuting qubits and modified field theories.
Key Arguments: Deutsch was persuaded toward Everettian quantum theory through direct conversation with Bryce DeWitt and later by Everett himself, who was technically current and not a naive outsider. Major scientific and biological advances often appear after very long periods of little visible change; this suggests open-ended explanatory progress rather than simple monotonic growth. Human explanatory knowledge has altered the universe in a physically testable way—for example, Earth may soon be able to deflect asteroids rather than attract them. Turing completeness is a settled property of brains/computers; Deutsch argues humans also have explanatory universality, meaning they can generate any explanation available to a universal constructor. Constructor theory should recast physics in terms of possible and impossible transformations, similar to how computation separates computable from noncomputable functions. Bayesian epistemology is, in his view, the wrong model for scientific knowledge because evidence can increase confidence in a theory while decreasing confidence in many of its consequences. What science really pursues is good explanation, not probability mass; explanations are rare, non-formalizable in a final way, and judged by explanatory power rather than likelihood. Quantum mechanics is likely not the final theory because it sits uneasily with general relativity and because quantum field theory has conceptual pathologies around commuting/noncommuting fields. AGI is plausible in principle but likely requires a philosophical breakthrough about creativity and explanation, not just more data or raw compute. Humans, aliens, and future AGI are all “people” in the relevant sense: entities capable of explanatory knowledge and therefore of transforming the world substantially.
Data Points: Hugh Everett seminar: late 1970s - Deutsch says he attended Everett’s talk at the University of Texas in the late 1970s. Seminar room smoking rule: waived for Everett - Wheeler asked that the no-smoking rule be waived because Everett was a chain smoker. Human/ape code difference: a few K of code - Deutsch says great apes and humans differ by only a few kilobytes of code, containing the bootstrap program. Turing paper date: 1936 - He references Turing’s founding work on computation. Quantum computer example: 2^N - He describes a quantum computation splitting into two to the N worlds/register states during the computation. Asteroid protection timeline: soon - He says Earth will soon be able to repel asteroids/comets rather than attract them.
Pivotal Quotes: "The explanation in this case is that there is explanatory knowledge on the Earth." — David Deutsch: On why Earth can alter asteroid trajectories and more generally transform physical outcomes. "The principle of optimism is that everything which is not forbidden by laws of physics is possible with enough knowledge." — David Deutsch: A core statement linking knowledge, possibility, and constructor theory. "We want to seek the best explanation, not increase our credence for true theories." — David Deutsch: His rejection of Bayesian epistemology as the model of scientific knowledge.
Implications: Deutsch frames science, AI, and physics as open-ended projects driven by explanation and constructive knowledge. If correct, future progress will come from new philosophical frameworks, not just more data, and could radically expand what humanity can build and control.
About Sean Carroll MindScape
Ever wanted to know how music affects your brain, what quantum mechanics really is, or how black holes work? Do you wonder why you get emotional each time you see a certain movie, or how on earth video games are designed? Then you’ve come to the right place. Each week, Sean Carroll will host conversations with some of the most interesting thinkers in the world. From neuroscientists and engineers to authors and television producers, Sean and his guests talk about the biggest ideas in science, ...