Do paradoxes reveal deep truths about reality, or do they mostly expose the hidden ways our language and concepts misfire?
My links: https://linktr.ee/frictionphilosophy.
1. Guest
Michael Huemer is Professor of Philosophy at the University of Colorado, Boulder, and his work has focused on ethics, epistemology, paradoxes, politics, and a variety of other issues.
2. Interview Summary
The interview centers on Paradox Lost and uses it as a springboard to walk through several classic philosophical puzzles. The opening section focuses on the liar paradox, where Huemer argues that the apparent contradiction is a symptom of inconsistent “instructions” in ordinary language: in the liar case, the language seems to tell us to assign a proposition that would be true iff false, but there simply is no such proposition—so the sentence fails to express a proposition at all. He contrasts this with dialetheist approaches (e.g., Graham Priest), and defends a broadly classical picture on which genuine contradictions can’t be true once you understand how negation works.
From there, the conversation turns to vagueness and Huemer’s more radical “semantic nihilism” proposal: while classical logic and bivalence still apply to propositions, a huge proportion of everyday declarative sentences don’t manage to express a determinate proposition in the first place, so they aren’t straightforwardly true-or-false. To explain why such language can still be useful, he sketches an alternative way of thinking about meaning and mental content: our intentional states can be satisfied to degrees (not merely satisfied vs. unsatisfied), illustrated with simple desire-cases (cookie vs. brownie) and then extended—more controversially—to belief-like states.
A later stretch is devoted to decision/probability puzzles. On Newcomb’s problem, Huemer defends two-boxing by appealing to a dominance-style argument (if the opaque box has $0, both is better; if it has $1M, both is still better), while the interviewer pushes back about what you should do given the predictor story. They also discuss whether you should “defer” to a more informed version of yourself, and how Simpson’s paradox-style reversals complicate that intuition. The final portion moves into self-locating/probabilistic reasoning: in the “shooting room” case, Huemer’s diagnosis is that once you learn that a particular person (“Vic”) was called in, you gain evidence about how late in the process you’re conditioning on, which undercuts the simplistic 50% reasoning. The interview closes by zooming out to broader anthropic-style arguments (including a Bayes-y case for reincarnation given infinite time, under explicit assumptions) and the background metaphysical commitments those arguments quietly rely on.
3. Interview Chapters
00:00 - Introduction
01:04 - Liar paradox
03:47 - Propositions
10:15 - Logic
14:43 - Truth
16:36 - Other contradictions
18:02 - Non-existent objects
20:46 - Vagueness
25:59 - Semantic uses
29:14 - External factors
31:09 - Terence Horgan
34:43 - Newcomb’s problem
37:19 - Transparent box case
39:12 - Well-wishing friend
42:32 - Simpson’s paradox
47:08 - Deference
49:32 - Valuing evidence vs. reward
56:33 - Rationality
59:18 - Predictably rewarded
1:01:41 - Preparation
1:07:54 - Perfect predictor
1:10:29 - Free will
1:11:57 - Rationality
1:17:11 - Shooting room
1:22:39 - Variations
1:27:24 - Simulation
1:28:44 - Infinite expectations
1:29:44 - Possible in some other sense
1:35:32 - Immortality
1:38:27 - Names
1:46:10 - Old evidence
1:49:17 - Meaning of names
1:52:27 - Personal identity
1:54:33 - Moral relevance
1:56:38 - Moral realism and phenomenal conservatism
2:00:08 - Stance-dependent norms
2:02:48 - Ordinary English
2:05:51 - Value of philosophy
2:09:04 - Conclusion
