🔬 Scientist
Pure mathematics — no exam syllabus
Set Theory & Logic — The Foundations of Mathematics
Set theory is the language in which all of mathematics can be expressed. Every mathematical object — number, function, space, group — can be defined as a set. Logic provides the rules for reasoning about these sets. Together, they answer the deepest questions: What can mathematics prove? What is the nature of infinity? Are there true statements that can never be proved? The answers are as beautiful as they are unsettling.
"Mathematics is the only science where we can prove things — but Gödel showed that even in mathematics, there are things we can never prove."
— inspired by Gödel's incompleteness theorems
Part 1
🧱 ZFC Axioms — The Rules of the Game
Building Mathematics from Nothing
The Axiom of Extensionality — Two Sets Are Equal If They Have the Same Elements
Set theory begins with a single primitive notion: a set is a collection of objects. The Zermelo-Fraenkel axioms with Choice (ZFC) provide the foundational rules. There are 8-10 axioms (depending on how you count): Extensionality (sets are determined by their elements), Pairing (given a,b, {a,b} exists), Union (∪ of a set of sets exists), Power Set (all subsets exist), Infinity (an infinite set exists), Separation (subsets of a set defined by a property exist), Replacement (the image of a set under a function exists), Foundation (no infinite descending chains), and Choice (a product of nonempty sets is nonempty). From these, all of mathematics can be derived: natural numbers as von Neumann ordinals (0=∅, 1={∅}, 2={∅,{∅}}), functions as sets of ordered pairs, real numbers as Dedekind cuts, and so on. Every theorem ever proved can be translated into ZFC — at least in principle.
ZFC is the common language of mathematics. Everything you've ever proved can be expressed in set theory — but at a cost: the foundation is never provably consistent.
The Axiom of Choice
The Axiom of Choice (AC) states: for any set of nonempty sets, there's a function choosing one element from each. It seems obvious for finite collections, but for infinite collections it has startling consequences: the Banach-Tarski paradox, the existence of a well-ordering of ℝ, the existence of non-measurable sets, and the fact that every vector space has a basis. Most mathematicians accept AC because it's so useful — without it, we lose foundational results in analysis, algebra, and topology. But the debate continues: is AC "obviously true" or does it create too many paradoxes?
Part 2
∞ Cardinals and Ordinals — Sizes of Infinity
Cantor's Paradise
Some Infinities Are Bigger Than Others
Cantor showed that infinity comes in many sizes. Two sets have the same cardinality if there's a bijection between them. ℕ (natural numbers) has cardinality ℵ₀ (aleph-null). ℝ (real numbers) has cardinality 2^{ℵ₀} — strictly larger than ℵ₀, as shown by Cantor's diagonal argument. The continuum hypothesis (CH) asks: is there a set whose size is strictly between ℵ₀ and 2^{ℵ₀}? Gödel proved CH is consistent with ZFC (1940); Cohen proved its negation is consistent (1963). CH is independent of ZFC — it can be neither proved nor disproved. Ordinal numbers extend the natural numbers into the transfinite: ω (the first infinite ordinal), ω+1, ω·2, ω², ω^ω, ε₀, and far beyond. Ordinals capture the idea of "position" in an infinite sequence; cardinals capture "size." The aleph hierarchy: ℵ₀, ℵ₁, ℵ₂, ... enumerates all infinite cardinalities. The continuum hypothesis says 2^{ℵ₀} = ℵ₁.
Cantor showed us that infinity is not one thing — it's a ladder of ever-larger infinities, stretching beyond imagination.
Cantor's Diagonal Argument
Suppose ℝ were countable. List all real numbers r₁, r₂, r₃, ... between 0 and 1. Construct a new number whose nth decimal digit is different from the nth digit of rₙ. This number differs from every rₙ — contradiction. Therefore ℝ is uncountable. This simple argument is the most important in set theory, and it generalizes: the power set of any set is strictly larger than the original set (Cantor's theorem). There is no largest infinity — the sequence of cardinalities ℵ₀, 2^{ℵ₀}, 2^{2^{ℵ₀}}, ... never ends.
Cantor's Theorem
|P(X)| > |X| for any set X
There is no set of all sets, no largest infinity, and no way to escape the endless ascent of larger and larger cardinalities.
In 1900, Hilbert listed 23 unsolved problems. The first was: prove the continuum hypothesis — that there's no set whose size is between ℕ and ℝ.
Gödel (1940) showed CH can't be disproved from ZFC (it's consistent). He built the constructible universe L, a model of ZFC where CH holds.
Cohen (1963) showed CH can't be proved from ZFC. He invented forcing — a technique to build models where CH fails.
Together, this proves CH is independent of ZFC. The question of whether CH is "true" is a question about what we mean by truth in mathematics — a philosophical question, not a mathematical one.
Part 3
🔨 Independence — What Cannot Be Proved
Building New Realities
Forcing — Paul Cohen's Method for Creating Mathematical Universes
Paul Cohen invented forcing in 1963 to prove the independence of CH. The idea: start with a countable model M of ZFC (guaranteed by the Löwenheim-Skolem theorem). Add a "generic" set G that doesn't exist in M. The new model M[G] contains G and is still a model of ZFC. By choosing G carefully, we can control which statements are true in M[G]. Cohen added ℵ₂ many new reals to make CH fail. Forcing has since been used to prove independence of dozens of statements: the axiom of choice itself (from ZF), Souslin's hypothesis, the Whitehead problem, and many more. Today, set theorists study the forcing multiverse — the collection of all possible set-theoretic universes connected by forcing extensions. The question "what is true?" becomes "what is true in which universe?"
Set theory reveals that many mathematical questions have no single answer — they're true in some universes and false in others. Mathematics studies not one reality, but a multiverse of possibilities.
Large Cardinal Axioms
Just as the axioms of ZFC assert the existence of one infinity (ℕ), large cardinal axioms assert the existence of much larger infinities: inaccessible cardinals, Mahlo cardinals, measurable cardinals, Woodin cardinals, and supercompact cardinals. These cannot be proved from ZFC (if ZFC is consistent), but they're not arbitrary — they form a natural hierarchy, and they have consequences for ordinary mathematics (like projective determinacy). The study of large cardinals reveals the structure of the set-theoretic universe and suggests that the search for "stronger and stronger axioms of infinity" is the natural path of mathematical foundations.
Part 4
🧠 Gödel's Theorems — The Limits of Proof
The End of Certainty
Gödel's Incompleteness Theorems — Mathematics Can't Prove Its Own Consistency
In 1931, Kurt Gödel published a paper that shook mathematics to its core. His first incompleteness theorem: any consistent formal system strong enough to do arithmetic contains a statement that can neither be proved nor disproved within the system. His second incompleteness theorem: such a system cannot prove its own consistency. Gödel constructed a statement G that says "G is not provable." If G were provable, it would be false — contradiction. If G were not provable, it would be true — but unprovable. This is no trick — arithmetic really is incomplete. This means that ZFC (if consistent) cannot prove its own consistency. The search for a complete, consistent foundation for all mathematics is impossible. Mathematics will always have unsolved questions — not because we're not smart enough, but because the nature of proof has fundamental limits.
Gödel showed that mathematics is incomplete — there will always be true statements we can never prove. This is not a failure of mathematics; it's the deepest truth about its nature.
Gödel Numbering — Coding Statements as Numbers
Gödel's brilliance was encoding mathematical statements as numbers. Assign numbers to symbols (0, 1, +, ×, =, ∀, etc.). A formula is a sequence of symbols, its Gödel number is 2^{code1} × 3^{code2} × 5^{code3} × ... (prime exponent coding). A proof is a sequence of formulas, also encoded as a single number. Now the statement "formula with number p is provable" becomes an arithmetical statement about p. The self-referential sentence G says "the formula with Gödel number g is not provable," where g is the Gödel number of G itself. This self-reference is the heart of the proof — and it's completely legitimate within arithmetic.
Gödel's First Incompleteness Theorem
Any consistent formal system containing arithmetic is incomplete — there are true statements it cannot prove
This does not mean some truths are unknowable — it means they're unprovable in any given system. We can always move to a stronger system, but that system will have its own unprovable truths.
Truth vs Provability
Gödel's theorems establish that truth and provability are different concepts. In a given formal system, provable statements are a subset of true statements — but the converse fails. The Löb theorem says that if "if P is provable then P" is provable, then P is provable — a technical result about provability logic. The Halting problem (Turing, 1936) is the computability-theoretic version of incompleteness: there is no algorithm that determines whether a given program halts. Gödel's and Turing's results together define the limits of computation and proof — boundaries that no amount of cleverness can cross.
The classic Liar Paradox: "This sentence is false." If it's true, it's false. If it's false, it's true. This is a paradox because it involves self-reference and truth — but truth is not a formal concept within the language.
Gödel's insight: replace "true" with "provable." The sentence "This sentence is not provable" is not paradoxical — it's either true (and unprovable) or false (provable, which would be a contradiction if the system is consistent).
By using provability instead of truth, Gödel sidestepped the paradox and created a genuine mathematical theorem. The self-reference is legitimate because it's coded arithmetically — not a semantic trick but a syntactic construction within the formal system.
"The first incompleteness theorem says that mathematics is inexhaustible. No matter how many axioms you add, there will always be true statements that elude proof."
— Gregory Chaitin
🎯 Practice — Check Your Set Theory Intuition
These puzzles build set theory and logic intuition.