Questions: Independence in ZFC and Limitations of Axiomatization

Question 1

A mathematician argues: 'Since CH is independent of ZFC, we can freely assume it either way — both assumptions are equally mathematically valid, so it's just a matter of taste.' A set theorist offers a more careful response. What is the most accurate objection?

• A. The mathematician is correct; independence means both options are mathematically interchangeable in all contexts
• B. Independence means ZFC cannot settle CH, but the true set-theoretic universe V may favor one answer; independence shows the limits of ZFC, not the limits of mathematical truth
• C. The mathematician is wrong because independence proves CH is false — ZFC + ¬CH has been shown more consistent than ZFC + CH
• D. The mathematician is wrong because independent statements cannot be added as axioms without creating contradiction

Answer: (B)

Independence means neither ZFC ⊢ CH nor ZFC ⊢ ¬CH — the axioms are silent. But 'ZFC cannot prove CH' is not the same as 'CH has no determinate truth value.' Many set theorists believe V (the true set-theoretic universe) satisfies CH or ¬CH; we simply cannot verify which from ZFC alone. Furthermore, independence results are not symmetric in all ways: large cardinal axioms and forcing axioms (like Martin's Maximum) tend to imply 2^ℵ₀ = ℵ₂, ruling out CH. The situation is analogous to the independence of the parallel postulate: both Euclidean and non-Euclidean geometries are consistent, but physical space has a specific geometry. Independence is a limitation of the formal system, not necessarily a statement about mathematical reality.

Question 2

Gödel's proof that ZFC + CH is consistent used which key strategy?

• A. Forcing: constructing a model extension M[G] where new reals are added to violate CH
• B. The constructible universe L — the smallest model of ZFC — where CH holds because every real is explicitly definable from ordinals
• C. Compactness: showing that no finite subset of ZFC axioms implies ¬CH, so by compactness ZFC + CH is consistent
• D. A diagonal argument showing that CH cannot even be stated in first-order set theory

Answer: (B)

Gödel constructed L, the 'constructible universe,' by iterating from the empty set and adding only sets that are explicitly definable (in first-order logic) from already-constructed sets. This produces the smallest possible model of ZFC — one where the reals are tightly controlled by their definitions from ordinals. In L, the cardinality structure is as compact as possible, and CH holds. The argument for consistency is: if ZFC is consistent (has any model), then L is a model of ZFC + CH, so adding CH to ZFC introduces no new contradiction. This is relative consistency — not a proof that CH is 'true,' but a proof that assuming it is safe.

Question 3

The independence of the continuum hypothesis from ZFC was established by combining Gödel's inner model technique with Cohen's forcing method.

• T. True
• F. False

Answer: True

Both directions are required. Gödel (1938–1940) proved the consistency of ZFC + CH using the constructible universe L: if ZFC is consistent, so is ZFC + CH. Cohen (1963) proved the consistency of ZFC + ¬CH using forcing: starting from any model of ZFC, he constructed a larger model where 2^ℵ₀ ≥ ℵ₂, so CH fails. Together, the two results establish that ZFC can prove neither CH nor ¬CH — the statement is genuinely independent. Either technique alone would only establish one direction of the independence result.

Question 4

Since CH is independent of ZFC, no mathematical proof — even one using large cardinal axioms or additional set-theoretic principles — can determine whether CH is true or false.

• T. True
• F. False

Answer: False

Independence from ZFC means ZFC alone cannot settle CH — but extensions of ZFC can and do. Many large cardinal axioms are consistent with both CH and ¬CH, but forcing axioms like Martin's Maximum (MM) imply 2^ℵ₀ = ℵ₂, directly settling CH in the negative. Other proposed axioms (like Woodin's Ultimate-L program) aim to settle CH affirmatively within a natural extension of ZFC. Independence is always relative to a specific axiom system; stronger systems can resolve questions that weaker ones cannot. The search for natural axioms that settle CH is an active area of set-theoretic research.

Question 5

Explain what it means for a statement to be 'independent of ZFC,' and why this is different from saying the statement is simply 'unknown' or 'unproven.'

Model answer: A statement φ is independent of ZFC if there is a proof (within ZFC itself) that neither ZFC ⊢ φ nor ZFC ⊢ ¬φ. This is established by exhibiting two models: one in which φ is true (Gödel's L for CH) and one in which φ is false (Cohen's forcing extension). 'Unknown' or 'unproven' suggests the proof exists but hasn't been found yet — a matter of mathematical effort. Independence is a proven result: no future ZFC proof, however clever or long, can settle CH, because we have already proved that both CH and ¬CH are consistent with ZFC. The statement is not waiting to be discovered; the formal system is provably unable to decide it.

The distinction is philosophically significant. Mathematical statements are typically either provably true, provably false, or genuinely open (we suspect they are true/false but lack proof). Independence is a fourth category: proven to be undecidable within the specified axiom system. Gödel's incompleteness theorems predicted this would occur for sufficiently strong axiom systems; CH was the first concrete mathematical question (not an artificial Gödel sentence) shown to have this property.