Continuum Hypothesis and Independence from ZFC

Core Idea

The Continuum Hypothesis (CH) asserts 2^ℵ₀ = ℵ₁—that there is no cardinal strictly between ℵ₀ and the continuum. Gödel proved CH is consistent with ZFC; Cohen proved its negation is also consistent. Thus CH is independent of ZFC: undecidable from the standard axioms alone.

Explainer

You already know from Cantor's diagonalization that the set of real numbers is strictly larger than the set of natural numbers — |ℝ| = 2^ℵ₀ > ℵ₀. And from cardinal arithmetic you know that power sets jump to strictly larger cardinalities. The natural question is: exactly how much larger is 2^ℵ₀? The Continuum Hypothesis is the claim that 2^ℵ₀ = ℵ₁ — the reals are precisely the *next* infinite cardinality after the naturals, with no cardinal in between. It is one of the most famous open questions in the history of mathematics, and its resolution revealed something unexpected: neither CH nor its negation is provable from the standard axioms.

Gödel established the first half in 1940 by constructing the constructible universe L — a minimal model of ZFC built by systematically defining only sets that are explicitly definable from earlier sets. In L, the axiom of choice holds, the generalized continuum hypothesis holds, and CH in particular holds. This means you cannot disprove CH from ZFC: if ZFC is consistent, then ZFC + CH is consistent, because L is a model of ZFC in which CH is true. Gödel's method is sometimes called an inner model argument — you build a carefully constrained universe inside any model of ZFC where the desired statement happens to be true.

Paul Cohen established the second half in 1963 using a radically new technique called forcing. Starting from a model of ZFC + CH, forcing constructs a carefully designed extension of the model — adding new "generic" sets — in which CH fails. The extended model satisfies all the ZFC axioms but contains enough real numbers to make 2^ℵ₀ > ℵ₁. Cohen's construction showed you can add arbitrarily many reals to a model without contradiction. CH is therefore undecidable: neither CH nor ¬CH can be derived from ZFC alone. Together, Gödel and Cohen showed that CH is independent of ZFC — it is a statement that ZFC is simply silent about.

What does independence mean philosophically? It does not mean CH is meaningless or random. It means ZFC does not determine the answer. Set theorists have since explored both CH and its negation as possible axioms, and explored stronger axioms (large cardinal axioms, Martin's Axiom, Woodin's Ultimate L program) that do decide CH one way or another. Some mathematicians hold a Platonist view — there is a real mathematical universe and CH is either truly true or truly false, even if ZFC cannot tell us which. Others take a pluralist view — there are many consistent universes of set theory, and CH is true in some and false in others. The independence of CH was a watershed moment that permanently changed mathematicians' understanding of what axioms do and do not determine.