What this covers

William Egginton, author of The Rigor of Angels, argues that Immanuel Kant, Heisenberg, and Jorge Luis Borges each independently discovered a fundamental epistemological constraint: reality is never grasped directly but always through the mediating conditions of observation and thought. Egginton develops this comparison across three domains—philosophy, physics, and literature—showing how each thinker arrived at the Knower-Known Distinction from different starting points. Kant grounded it in the structure of human cognition, Heisenberg in the mathematics of quantum mechanics, and Borges in thought experiments that dramatize the impossibility of unmediated knowledge. The conversation with Sean Carroll explores how this shared insight applies across science, philosophy, and everyday reasoning.

The discussion ranges from Kant's argument that synthesis requires an a priori subject distinct from its objects, through Heisenberg's discovery that measurement itself introduces irreducible change, to Borges's fictional worlds (particularly Funes and the Library of Babel) where perfect perception or perfect memory collapse under their own weight. Egginton emphasizes that Heisenberg deliberately echoed Kant's language of forms of intuition, and that the Copenhagen interpretation physicists were engaged in epistemology as much as physics. The conversation also covers Grete Hermann's overlooked critique of von Neumann, the structure of Borges's fictional essays, and how Nietzsche's eternal return connects to modern infinite-universe arguments. Throughout, both speakers test whether the insight—that we must always operate within The Observation Filter—leads to Epistemic Humility or risks becoming metaphysical overreach, with Carroll pushing back on whether Kant's framework necessarily applies to physics.

Sharpest takeaway

Egginton argues that Kant, Heisenberg, and Borges each independently discovered the same fundamental limit on knowledge: we never perceive reality directly but only as mediated through the conditions of observation, and forgetting this filter leads us into error across science, philosophy, and life.

  • Kant showed that experience requires an a priori synthesizing subject distinct from what is known, so we cannot access reality unmediated
  • Heisenberg's uncertainty principle shows a baked-in minimal difference imported by the act of observation
  • Borges dramatized through fiction (Funes, the Library of Babel) that perfect perception or memory is incoherent, requiring a minimal distance between knower and known

The argument · threads4 threads · 32 claims
0.85

Our perception of reality is necessarily mediated and incomplete, requiring philosophical and scientific work to understand our relationship to fundamental nature.

3 pointscentrality 5/5
  • Kant did not know about quantum mechanics and was talking about human access to the world rather than the fundamental laws of physics, so it would be wrong to treat quantum mechanics as merely a version of classical-world problems; rather, quantum mechanics exploits in an unanticipated way the room left by the fact that our observations are necessarily incomplete and change the world.

    “the Immanuel Kant did not know about quantum mechanics. He was talking about a different thing than the fundamental laws of physics, it was our human access to the world... it takes advantage of the fact that our observations are necessarily incomplete and change the world.”

  • When we perceive reality we do not perceive it in a clear unadulterated form but only certain aspects, mediated rather than immediate, so understanding our relationship to the fundamental nature of reality requires work through philosophy, science, and even literature.

    “we don't perceive it in a clear, unadulterated form... We perceive something, and that's something has to do with reality but it's not unmediated, it's not immediate”

  • Rather than relying on a priori conditions of possibility to solve Hume's problem, one can hold that knowledge arises from a constant dialectic between our revisable models or hypotheses about structures in the world and the sensations we receive, requiring nothing a priori beyond a hope that there are discernible patterns.

    “we come up with hypotheses about structures out there in the world. We have some intuitions, we might revise them, but it's a constant dialectic between our models of the world and these sensations that we get. And I don't really need anything a priori other than a hope that there are patterns out there to be discerned.”

0.85

Infinite knowledge and complete representation are impossible; physicists, writers, and philosophers must embrace epistemic humility about precision and exhaustive explanation.

3 pointscentrality 5/5
  • Even if the Library of Babel contains a single book that correctly explains the library, it also contains a vastly larger number of books that purport to explain it but are incorrect — those with one error, one wrong word, and so on — so finding the true explanation is overwhelmed by near-identical falsehoods.

    “if there is a book that correctly explains the library in the library, there's also a huge number, vastly huger number of books that purport to explain the library and are incorrect.”

  • Physicists, fiction writers, and philosophers are three categories of people who often help themselves to infinite knowledge and precision in ways that are not quite warranted, so the lesson of epistemic humility is a useful corrective for them professionally as well as personally.

    “if we pick physicists, fiction writers and philosophers, these are three categories of people who often help themselves to infinite knowledge and precision in ways that are not quite warranted”

  • Just as Funes needs an entire day to reconstruct a day, the full card catalog of the Library of Babel would have to be the same size as the library itself — there is no compression possible of a complete representation of everything.

    “for the library the full card catalog would be the same as the library... There's no compression possible.”

0.51

Quantum mechanics advances empirical data-fitting without requiring the metaphysical rejection of physical understanding the 1920s founders mistakenly imposed.

1 pointcentrality 4/5
  • In the 1920s the founders of quantum mechanics — Heisenberg, Bohr, Pauli — were simultaneously fitting experimental data and carrying out a revolution in epistemology and metaphysics, but in declaring that understanding how the world really works was no longer the job of physics they went too far, and one can accept the data-fitting half of their program without the metaphysical half.

    “they were also carrying out a revolution in epistemology and metaphysics... And maybe they went too far. I think they went too far, honestly... And you could buy half of that program without the other half.”

0.24

Grete Hermann correctly identified a flaw in von Neumann's proof, leading through Einstein and Bohm to Bell and modern entanglement physics.

1 pointcentrality 2/5
  • Grete Hermann was the first to point out that John von Neumann's supposed proof of the impossibility of hidden variables did not work, but no one believed her; only Einstein independently appreciated the mistake and pointed it out to David Bohm, who then invented Bohmian mechanics, inspiring John Bell and leading eventually to the entanglement Nobel Prize.

    “She does make a brief appearance in the history of quantum mechanics as the first person to point out that John von Neumann's supposed proof about the impossibility of hidden variables didn't work.”