# What is truth? - Mathematician explains | Joel David Hamkins and Lex Fridman

Source: https://www.youtube.com/watch?v=Zv2mYE8adoc
Recap page: https://rapidrecap.app/video/Zv2mYE8adoc
Generated: 2026-01-04T22:01:08.789+00:00

---
## Quick Overview

The discussion between Joel David Hamkins and Lex Fridman centers on the fundamental distinction in logic between "truth" (semantic satisfaction within a model) and "provability" (syntactic derivation from axioms), a concept highlighted by Gödel's Incompleteness Theorems and Alfred Tarski's disquotational theory of truth, which separates the language being talked about from the language used to talk about it.

**Key Points:**
- Provability is defined as the syntactic derivation of a formula from axioms using inference rules, while Truth is the semantic satisfaction of a formula within a specific model.
- The discussion references Kurt Gödel's Incompleteness Theorems, which imply that any consistent formal system rich enough for basic arithmetic contains true statements that cannot be proven within the system, and cannot prove its own consistency.
- Alfred Tarski's disquotational theory of truth is introduced, stating that a sentence is true if and only if the content of the sentence is the case (e.g., "snow is white" is true if and only if snow is white), achieved by removing quotation marks from the assertion.
- The inability to algorithmically decide the provability of every true statement (the decision problem for arithmetic) is directly equivalent to the Halting Problem, a fundamental limitation in computability theory.
- Classical proof systems are characterized as being both 'sound' (proofs preserve truth) and 'complete' (all truths are provable), properties that Gödel's Incompleteness Theorems show cannot be simultaneously held by sufficiently complex formal systems.

![Screenshot at 00:14: A slide explicitly contrasts the logical concepts: Provability \(⊢\) is syntactic derivation, while Truth \(=\) is semantic satisfaction within a specific model, setting the stage for the entire discussion.](https://ss.rapidrecap.app/screens/Zv2mYE8adoc/00-00-14.jpg)

**Context:** This video features a deep dive into foundational concepts of mathematical logic between Lex Fridman and mathematician Joel David Hamkins. The core conversation revolves around clarifying the difference between 'truth' and 'provability' in formal systems, drawing heavily on the work of 20th-century logicians like Kurt Gödel and Alfred Tarski, and contrasting these concepts with the historical goals of mathematicians like David Hilbert.

## Detailed Analysis

The conversation meticulously unpacks the formal distinction between provability and truth in logic, concepts central to understanding the limits of formal mathematics established by Gödel's Incompleteness Theorems. Provability is framed as a syntactic process—deriving statements from axioms using fixed rules like Modus Ponens (MP)—whereas truth is a semantic concept, requiring satisfaction within a specific model or reality. Hamkins emphasizes that while Gödel's Incompleteness Theorems show that sufficiently strong axiomatic systems contain truths they cannot prove (and cannot prove their own consistency), Tarski’s disquotational theory of truth provides a formal way to define truth without reference to an external reality, by stating that a sentence like 'Snow is white' is true if and only if snow is white (i.e., removing the quotes). The discussion touches upon the decidability problem, noting that determining whether a given sequence of statements constitutes a proof is computable, unlike determining the truth of an arbitrary statement within the system. The inherent limitations shown by Gödel mean that no single formal system can capture all mathematical truths, leading to an ongoing tension between what is true and what is provable.

### Logic Foundations

- Provability vs. Truth: Provability is syntactic derivation from axioms using rules of inference
- Truth is semantic satisfaction within a specific model
- Gödel's Incompleteness Theorems imply true but unprovable statements exist in consistent systems rich enough for arithmetic.

### Key Logical Contributions

- Alfred Tarski established the disquotational theory of truth, separating the object language from the meta-language by removing quotation marks from assertions
- David Hilbert aimed for a complete and consistent formal theory of arithmetic, a goal undercut by Gödel's results.

### Proof Systems and Decidability

- Classical proof systems aim to be both sound (truth-preserving) and complete (all truths provable)
- Gödel's work shows that for systems complex enough to handle arithmetic, these two properties cannot fully coincide
- The decidability of proof checking is contrasted with the undecidability of determining truth for all statements, relating the latter to the Halting Problem.

![Screenshot at 00:14: Slide explicitly defining Provability \(syntactic derivation\) versus Truth \(semantic satisfaction\).](https://ss.rapidrecap.app/screens/Zv2mYE8adoc/00-00-14.jpg)
![Screenshot at 00:31: A historical photo showing logicians Kurt Gödel and Alfred Tarski, whose work is central to the discussion on truth and provability.](https://ss.rapidrecap.app/screens/Zv2mYE8adoc/00-00-31.jpg)
![Screenshot at 01:14: Slide illustrating Gödel's Incompleteness Theorems, stating that consistent formal systems rich enough for arithmetic contain unprovable true statements.](https://ss.rapidrecap.app/screens/Zv2mYE8adoc/00-01-14.jpg)
![Screenshot at 02:29: Illustration of Tarski's disquotational theory of truth using the example: '"snow is white" is true if and only if snow is white'.](https://ss.rapidrecap.app/screens/Zv2mYE8adoc/00-02-29.jpg)
![Screenshot at 05:45: Slide detailing the Modus Ponens \(MP\) logic rule, described as the central rule of inference in Hilbert systems.](https://ss.rapidrecap.app/screens/Zv2mYE8adoc/00-05-45.jpg)