What this covers

David Krakauer, president of the Santa Fe Institute, discusses what unifies complexity science by pointing to a single unifying concept: systems that internally encode the world they inhabit. The conversation with Sean Carroll moves through the distinction between complexity and mere self-organization—hurricanes are intricate but not complex, because they cannot store and use information about their environment. Krakauer argues that what separates complex adaptive systems from physics is agency: the ability to gather information, build an internal map, and navigate by it rather than simply responding to instantaneous gradients. The discussion traces this idea through examples ranging from bacterial chemotaxis to human decision-making, and backward through the history of control theory, natural selection, and how physicists have understood the difference between systems that follow equations and systems that represent the world.

The conversation covers substantial theoretical ground. It distinguishes adaptation from action principles in physics, explores emergence as a product of broken symmetry rather than new fundamental laws, and examines how Maxwell's Demon illuminates natural selection. Krakauer makes a case that life is fundamentally an inferential and representational process rather than a chemical one, discusses how complementary and competitive cognitive artifacts differ in opacity and internalizability, and explains how individuality itself is an emergent phenomenon detectable through information theory. The conversation also addresses the relationship between noise and innovation, the limits of description length as a bound on complexity, and why 21st-century problems like climate change require synthesis across disciplines rather than single-field approaches. Throughout, disagreements surface about whether existing frameworks adequately capture complexity or whether the field requires fundamentally new epistemology.

Sharpest takeaway

Krakauer argues that complexity science is unified by the study of 'teleonomic matter'—matter with purpose that internally encodes the world it lives in—which fundamentally distinguishes complex adaptive systems from merely self-organizing physical systems like hurricanes.

  • Complex systems are defined by internally encoding the world they live in and storing information adaptively, not by self-organization alone
  • Adaptation is fundamentally different from physics' action principles because it builds a memory/map rather than minimizing a functional
  • Emergence is the ability to write dynamically-sufficient effective theories enabled by broken symmetry, without requiring new fundamental laws

The argument · threads3 threads · 37 claims
0.57

Entropy-based criteria reveal subsystem structure in quantum mechanics and identify individuals in complex systems.

2 pointscentrality 3/5
  • The problem of carving the quantum world into subsystems (which physics normally takes as given) can be solved by an inverse procedure with criteria involving entropy—a quantity you minimize tells you where the subsystems are—directly paralleling the information-theoretic criteria for detecting individuals in complex systems.

    “Ashmeet Singh and I wrote a paper called Quantum Mereology... There is a set of criteria. It has to do with entropy. And there's a thing you minimize and that tells you where the subsystems are.”

  • An introduction describing the episode's topic and guest.

    “Hello everyone. Welcome to the Mindscape Podcast. I'm your host, Sean Carroll.”

0.56

Significant theoretical progress has occurred within established physics models despite no new foundational frameworks emerging.

1 pointcentrality 4/5
  • The claim that no major progress has occurred in fundamental theoretical physics since 1980 is adjacent to a true statement but not actually true: while no new model correctly mapping onto reality has emerged beyond the standard model, general relativity, and Big Bang theory, there have been enormous insights into those theories and speculative models—the black hole information problem alone has been a font of ideas like holography.

    “the statement that there's been no major progress in theoretical physics, fundamental theoretical physics since 1980 is adjacent to a true statement, but it is not actually a true statement.”

0.56

Agents' goal-directed states function as future boundary conditions, inverting the universe's low-entropy past boundary condition.

1 pointcentrality 4/5
  • Whereas physics has only a past boundary condition (the low-entropy Big Bang) and no future boundary condition, purposeful agents can be thought of as carrying a little mini future boundary condition with them—a state they want to reach—and it almost doesn't matter how they get there, raising the puzzle of how the universe's past low-entropy condition gets flipped into goal-states in agents.

    “purposeful agents can be thought of maybe... as carrying a little mini future boundary condition with them. There is a state they want to be at in the future, and it almost doesn't matter how they get there.”