Can one prove a theorem that unprovable theorem is unprovable?

Sonar Pro

Vote: Yes

Gödel's first incompleteness theorem constructs a sentence **G** (the Gödel sentence) that states its own unprovability: **G ↔ ¬Prov(⌜G⌝)**. If the theory **F** is consistent, **F ⊬ G**, proving **G** is true but unprovable in **F**[2][1]. This directly proves that a specific unprovable theorem (**G**) is unprovable, as **F** proves **¬Prov(⌜G⌝)** (via the equivalence and consistency)[2]. The query asks if **one** can prove **that an** unprovable theorem is unprovable—not for all or arbitrary ones. Gödel's construction provides exactly this proof for the specific **G**[1][2]. No contradiction arises: proving **G** unprovable doesn't make **G** provable, as **G** self-referentially asserts its own unprovability[2].

GPT-5.4

Vote: Yes

Yes. If by “an unprovable theorem” we mean a true statement that is not provable in some fixed formal system T, then it is possible to prove of some specific sentence G that G is unprovable in T. This is exactly the shape of Gödel-style and Rosser-style incompleteness arguments. More precisely, one can often prove in a stronger metatheory that a certain sentence G has no proof in T. In some cases, under suitable assumptions such as T proving its own soundness for the relevant class of statements, one can even have T prove a statement equivalent to “G is unprovable in T.” What Gödel’s second incompleteness theorem blocks is not every proof of unprovability, but certain self-verifying proofs inside the same sufficiently strong system, especially proofs of that system’s own consistency. So the blanket claim “one cannot prove an unprovable theorem is unprovable” is false. Thus the correct answer is Yes: with the right formalization and standpoint, one can prove that a particular theorem/sentence is unprovable.

Grok 4.1 Fast

Vote: Yes

Gödel's first incompleteness theorem provides a direct example. He constructed a sentence G that asserts its own unprovability: 'G is not provable in the system S.' Assuming S is consistent, Gödel proved that G is true but unprovable in S. Thus, G is an unprovable truth (informally a 'theorem'), and its unprovability is rigorously proven (relative to S's consistency). This demonstrates that one can prove the unprovability of a specific unprovable theorem, making the answer yes.

Claude Opus 4.6

Vote: Yes

Yes, it is possible to prove that an unprovable statement is unprovable. This is a well-established result in mathematical logic. The key example is the Continuum Hypothesis (CH). Paul Cohen (1963) proved that CH is independent of ZFC set theory — meaning it can neither be proved nor disproved from the ZFC axioms. Specifically, Gödel (1940) showed that CH is consistent with ZFC (cannot be disproved), and Cohen showed that the negation of CH is also consistent with ZFC (cannot be proved). Together, these results constitute a proof that CH is unprovable in ZFC. More precisely, what we prove is a meta-theorem: we prove (in a suitable meta-theory) that a given formal system cannot derive a particular statement. Independence proofs are proofs of unprovability, and they are perfectly rigorous mathematical results. Another example: Gödel's second incompleteness theorem shows that a consistent sufficiently strong system cannot prove its own consistency. This is itself a proven theorem — a proof that a certain statement (the consistency statement) is unprovable within the system. So the answer is clearly yes: we can and do prove that certain statements are unprovable. The proof occurs in a meta-theory (which may be stronger than, or simply different from, the object theory in question).

Kimi K2.5

Vote: Yes

Yes, this is possible through metamathematical analysis. Gödel's incompleteness theorems establish that in any consistent formal system F capable of expressing arithmetic, there exist true statements that F cannot prove. While F cannot prove its own limitations internally, we can absolutely prove the unprovability of specific statements from outside the system. For example, Gödel and Cohen proved that the Continuum Hypothesis is unprovable in ZFC set theory. Similarly, the unprovability of Euclid's parallel postulate from the other geometric axioms was established by constructing alternative models (non-Euclidean geometries). These proofs operate in a stronger metatheory or through semantic methods, demonstrating that independence results are themselves provable theorems. The crucial distinction is between provability *within* a system versus provability *about* a system. A sufficiently strong consistent system cannot prove its own Gödel sentence is unprovable (as this would imply consistency), but this unprovability is demonstrable in the metatheory or a stronger system.

Gemini 3.1 Pro

Vote: Yes

In mathematical logic, it is entirely possible to prove that a specific statement is unprovable within a given formal system. This is known as an independence result or an independence proof. For example, Kurt Gödel's First Incompleteness Theorem demonstrates that in any consistent formal system robust enough to encompass basic arithmetic, there are true statements that cannot be proven within that system. The proof of this fact is itself a rigorously proven theorem—a meta-theorem about the limitations of the formal system. Similarly, mathematicians proved that the Continuum Hypothesis is unprovable within standard Zermelo-Fraenkel set theory (ZFC). The proof that it cannot be proven is a celebrated mathematical theorem in its own right. Therefore, one can absolutely prove a theorem stating that a specific proposition is unprovable in a given axiomatic system.

Final Summary

The AI Roundtable reached a swift, unanimous consensus that proving a theorem's unprovability is a standard practice in mathematical logic. Sonar Pro, GPT-5.4, Grok 4.1 Fast, Claude Opus 4.6, Kimi K2.5, and Gemini 3.1 Pro all selected Option A, noting that while a system might not prove its own limitations, a meta-theory can rigorously establish them. The group pointed to Gödel’s incompleteness theorems and the independence of the Continuum Hypothesis as definitive evidence that unprovability is itself a provable property.

All 6 models agreed on "Yes" after discussion

Strongest Arguments

• Yes: A proof of unprovability is a meta-theorem demonstrating that a specific statement cannot be derived within a given formal system; this is exemplified by Paul Cohen’s proof that the Continuum Hypothesis is independent of ZFC set theory.