Learn set theory: the best books in order
This curriculum takes an intermediate learner from intuitive, informal set theory through rigorous axiomatic foundations, then into the deep waters of infinite cardinals, ordinals, and independence results. Each stage builds the vocabulary and proof fluency needed for the next, so the books are sequenced to minimize gaps and maximize understanding in self-study.
Naive Foundations & Intuition
BeginnerBuild a confident, informal understanding of sets, functions, relations, cardinality, and the paradoxes that motivate axiomatic set theory — all without heavy formalism.
▸ Study plan for this stage
Pace: 6–8 weeks, ~25–30 pages/day (Halmos first, then Devlin). Halmos is denser and more foundational (~160 pages); Devlin is more exploratory (~300+ pages). Allocate 3–4 weeks to Halmos, then 3–4 weeks to Devlin.
- Sets as fundamental objects: membership (∈), extensionality, and the informal notion that a set is determined by its elements
- Basic set operations: union, intersection, complement, power set, and Cartesian product—and how to visualize and manipulate them
- Functions and relations: definition, domain, codomain, injectivity, surjectivity, bijectivity, and composition as core structural tools
- Cardinality and infinite sets: finite vs. infinite, countable vs. uncountable, Cantor's diagonal argument, and the surprising fact that some infinities are 'larger' than others
- Paradoxes and their resolution: Russell's paradox, the Barber paradox, and why naive set theory fails—motivating the need for axioms
- Equivalence relations and partitions: how to classify and organize mathematical objects using sets
- Ordered sets and partial orders: introduction to structure beyond just membership
- Informal axiomatic thinking: recognizing that set theory requires careful rules (ZFC) to avoid contradiction, without diving into formal proofs
- What does it mean for two sets to be equal, and why is extensionality the key principle?
- Explain the difference between a set, an element of a set, and a subset. Give examples where these distinctions matter.
- What is Russell's paradox, and why does it show that the naive definition 'a set is any collection' is problematic?
- Define a function carefully: what are its domain, codomain, and range, and what does it mean for a function to be injective, surjective, or bijective?
- What is cardinality, and how can two infinite sets have different cardinalities? Sketch Cantor's diagonal argument.
- What is an equivalence relation, and how does it partition a set? Give a concrete example.
- Why do we need axioms (like ZFC) for set theory, and what informal intuitions do they formalize?
- Work through Halmos's exercises on set operations (union, intersection, complement): compute A ∪ (B ∩ C) and verify distributivity laws by hand for concrete sets.
- Construct power sets P(S) for small finite sets (e.g., S = {1, 2, 3}) and verify that |P(S)| = 2^|S|.
- Define three different functions from ℕ to ℕ (e.g., f(n) = n+1, g(n) = 2n, h(n) = n²) and classify each as injective/surjective/bijective with justification.
- Prove (or verify by example) that composition of injective functions is injective, and composition of surjective functions is surjective.
- Enumerate the first few elements of ℕ × ℕ using Cantor's diagonal enumeration, and explain why ℕ × ℕ is countable.
- Define an equivalence relation on a concrete set (e.g., congruence mod 3 on ℤ) and list the equivalence classes; verify they partition the set.
- Attempt to construct 'the set of all sets that do not contain themselves' and identify where Russell's paradox arises; discuss why this is not a valid set.
- Read Devlin's discussion of infinite cardinals and compare |ℕ| with |ℝ| using Cantor's argument; write a one-page explanation of why ℝ is uncountable.
Next up: This stage establishes the intuitive and historical foundations—why set theory matters, what goes wrong naively, and what the core structures (sets, functions, cardinality) look like—preparing you to engage rigorously with axiomatic set theory (ZFC axioms, formal proofs, and advanced topics like ordinals and the axiom of choice) in the next stage.
The perfect starting point: a short, beautifully written introduction to sets from a working mathematician's perspective. It builds intuition for all the core concepts — unions, intersections, functions, ordinals — that every later book assumes.
Bridges naive and axiomatic thinking by introducing the ZFC axioms informally and accessibly. Reading it second reinforces Halmos's ideas while gently preparing the reader for rigorous axiomatics.
Axiomatic Set Theory — First Rigorous Pass
IntermediateMaster the Zermelo–Fraenkel axioms (with Choice), understand how all of mathematics is built inside ZFC, and become comfortable with formal set-theoretic proofs.
▸ Study plan for this stage
Pace: 8–10 weeks, ~40–50 pages/day (Enderton first 4–5 weeks, then Suppes for 4–5 weeks)
- The Zermelo–Fraenkel axioms (Extensionality, Foundation, Pairing, Union, Power Set, Infinity, Replacement, Separation, Choice) and their intuitive justification
- How to construct natural numbers, integers, rationals, and reals purely from sets within ZFC
- Ordinal and cardinal numbers as fundamental objects in set theory, and their arithmetic
- Transfinite induction and recursion as proof and definition techniques
- The cumulative hierarchy (V-hierarchy) and how it models all of ZFC
- Formal set-theoretic proofs: working with definitions, quantifiers, and logical structure in a rigorous axiomatic system
- The role of the Axiom of Choice and its equivalents (Zorn's Lemma, Well-Ordering Theorem)
- How mathematical structures (functions, relations, sequences, topologies) are encoded as sets
- State the nine ZFC axioms and explain the intuition behind each one. Why is Foundation necessary, and what does it prevent?
- Construct the natural numbers from scratch using only set-theoretic operations. How does the Axiom of Infinity guarantee their existence?
- Define ordinal numbers and cardinal numbers. What is the difference between them, and how do you compare cardinalities?
- Prove by transfinite induction that every ordinal α satisfies a given property P(α). What makes this different from ordinary mathematical induction?
- State and prove the equivalence between the Axiom of Choice, Zorn's Lemma, and the Well-Ordering Theorem.
- Explain the cumulative hierarchy V_α and why it models ZFC. What does it mean to say 'all sets are in V'?
- Given a mathematical object (e.g., a function, a topological space, a group), show how to encode it as a set within ZFC.
- Work through Enderton's construction of ℕ, ℤ, ℚ, and ℝ step-by-step. For each, verify that the operations (addition, multiplication) are well-defined and satisfy their expected properties.
- Prove 5–10 basic theorems about ordinals (e.g., the class of ordinals is well-ordered, every well-ordered set is isomorphic to a unique ordinal, ordinal arithmetic is associative but not commutative).
- Prove Zorn's Lemma from the Axiom of Choice, and then use Zorn's Lemma to prove the Well-Ordering Theorem. Document each step carefully.
- For each ZFC axiom, write a short proof that a simple set-theoretic construction (e.g., {∅}, {∅, {∅}}, ℕ) satisfies the axiom's requirements.
- Solve 15–20 of the end-of-chapter problems from Enderton (Chapters 1–7), focusing on those involving formal proofs and axiom applications.
- Solve 10–15 problems from Suppes' chapters on ordinals, cardinals, and the Axiom of Choice, emphasizing rigorous justification.
- Write a 2–3 page formal proof of a non-trivial result (e.g., 'Every infinite cardinal is a limit ordinal' or 'The cardinality of ℝ is strictly greater than that of ℕ') using only ZFC axioms and previously proven lemmas.
- Create a reference sheet listing all nine axioms, their formal statements, and one concrete example of each. Update it as you learn more.
Next up: This stage equips you with the foundational rigor and axiomatic thinking needed to move into specialized topics—whether that's model theory (understanding different models of ZFC), large cardinals and independence results, or applications of set theory to topology and analysis.
The gold standard introductory axiomatic text: rigorous yet readable, with careful constructions of the natural numbers, integers, and reals inside ZFC. Its pacing is ideal for self-study after the naive stage.
Complements Enderton by emphasizing the logical and philosophical underpinnings of each axiom. Reading it second deepens understanding of *why* the axioms are chosen, not just what they say.
Cardinals, Ordinals & the Infinite
IntermediateDevelop a thorough command of transfinite arithmetic — ordinal and cardinal numbers, cofinality, the Axiom of Choice and its equivalents — and understand the landscape of infinite sets.
▸ Study plan for this stage
Pace: 8–10 weeks, ~40–50 pages/day (Hrbacek: 3–4 weeks; Kunen: 5–6 weeks)
- Ordinal numbers as order types and their arithmetic (addition, multiplication, exponentiation)
- Cardinal numbers, cardinal arithmetic, and the distinction between ordinals and cardinals
- Cofinality and regularity; the Cofinality Theorem and its implications for infinite cardinals
- The Axiom of Choice and its equivalents (Zorn's Lemma, Well-Ordering Theorem, Hausdorff Maximality Principle)
- Alephs and Beths; the Continuum Hypothesis and its independence from ZFC
- Cantor's Theorem and the hierarchy of infinite cardinals
- König's Theorem and the relationship between cofinality and cardinal exponentiation
- Cumulative hierarchy and the constructible universe as a foundation for understanding the landscape of sets
- What is the difference between an ordinal number and a cardinal number, and why is this distinction essential?
- State the Axiom of Choice and explain at least three of its equivalent formulations (e.g., Zorn's Lemma, Well-Ordering Theorem). Why are these equivalences non-trivial?
- Define cofinality and explain why ω₁ is regular while ω is not. What does regularity tell us about cardinal arithmetic?
- Prove or explain why there is no largest cardinal (Cantor's Theorem). What does this imply for the structure of the infinite?
- What is the Continuum Hypothesis, and why is it independent of ZFC? What does this independence reveal about set theory?
- Explain König's Theorem and its consequences for cardinal exponentiation. Why does cf(κ) < κ imply κ^cf(κ) > κ?
- Compute ordinal arithmetic: evaluate ω + 1, ω · 2, 2 · ω, ω^2, and ω^ω. Verify that ordinal addition is not commutative by showing ω + 1 ≠ 1 + ω.
- Prove that the set of all countable ordinals (ω₁) is uncountable, and explain why this establishes a new cardinal above ℵ₀.
- Construct a well-ordering on ℚ (the rationals) using the Axiom of Choice, and verify that your construction satisfies the well-ordering property.
- Apply Zorn's Lemma to prove that every vector space has a basis. Identify the poset, the chain condition, and the maximal element.
- Work through Cantor's diagonal argument to show that 2^ℵ₀ > ℵ₀, and generalize it to prove Cantor's Theorem for arbitrary cardinals.
- Compute cardinal exponentiation: determine 2^ℵ₀, ℵ₀^ℵ₀, and (2^ℵ₀)^ℵ₀. Use König's Theorem to establish bounds on ℵ₁^ℵ₀.
Next up: This stage establishes the foundational machinery of transfinite arithmetic and the Axiom of Choice, preparing you to explore advanced topics such as large cardinals, forcing, and independence results that require fluency in ordinals, cardinals, and cofinality.
A rigorous yet approachable treatment of ordinals, cardinals, and transfinite induction that fills the gap between an introductory axiomatic course and research-level texts. Ideal as the first dedicated study of the infinite.
A landmark graduate text that systematically develops cardinals, ordinals, and combinatorics before introducing forcing. It is the essential bridge to advanced set theory and should be read after ordinal arithmetic is comfortable.
Advanced Topics — Independence & Forcing
ExpertUnderstand Gödel's constructible universe (L), Cohen's method of forcing, and the independence of the Continuum Hypothesis and the Axiom of Choice from ZFC — the deepest results in set theory.
▸ Study plan for this stage
Pace: 8–10 weeks, ~40–50 pages/day (Jech chapters 12–14, then Kanamori chapters 1–3). Allocate 5–6 weeks to Jech's forcing and constructibility material; 3–4 weeks to Kanamori's presentation of large cardinals and independence results.
- Gödel's constructible universe (L): definition via transfinite recursion, absoluteness, and the axiom of constructibility (V=L)
- Inner models and relative consistency: how L provides a model of ZFC where GCH and AC hold
- Forcing as a method: generic filters, dense sets, the forcing relation (⊩), and how to extend models of ZFC
- Independence of CH and AC: how forcing constructs models where CH fails (via Cohen's poset) and models where AC fails (via symmetric models)
- Large cardinals and their role in independence: inaccessible, measurable, and supercompact cardinals as calibrators of consistency strength
- The hierarchy of consistency strengths: ZFC < ZFC+V=L < ZFC+inaccessible < ZFC+measurable < ZFC+supercompact
- Absoluteness and reflection principles: which properties are absolute between V and L, and how this constrains what can be forced
- Canonical models and the universe of sets: how L and forcing extensions relate to the 'true' universe V
- What is the constructible hierarchy L_α, and why does L satisfy ZFC? How does the axiom of constructibility V=L differ from standard ZFC?
- Explain the forcing relation ⊩ and generic filters. Why must a generic filter be dense-open and meet all dense sets in the poset?
- How does Cohen's forcing with finite partial functions prove that CH is independent of ZFC? What does the generic set add to the model?
- What is a symmetric model, and how can it be used to show that the Axiom of Choice is independent of ZF (without C)?
- Define inaccessible, measurable, and supercompact cardinals. Why are large cardinals important for calibrating consistency strength?
- What does it mean for a formula to be absolute between V and L? Give examples of absolute and non-absolute properties.
- Work through Jech's construction of L_α for small ordinals (α < ω+5) by hand; verify that L_ω = V_ω and compute L_{ω+1}, L_{ω+2}.
- Prove that if φ is a Δ₀ formula (bounded quantifiers only), then φ is absolute between V and L. Then show that 'x is uncountable' is not absolute.
- Construct a simple forcing poset (e.g., finite partial functions from ω to 2) and identify its dense sets; verify that a generic filter meets all of them.
- Verify the forcing relation ⊩ for a few basic formulas (e.g., ⊩ ¬(0=1), ⊩ ∃x(x ∈ Ǧ)) in Cohen's poset; compute the forcing conditions explicitly.
- Outline the proof that V[G] (the forcing extension by a generic G) satisfies ZFC, focusing on the power set axiom and replacement.
- Construct a symmetric model of ZF where AC fails (e.g., using a countable support symmetric model over a model of ZFC+GCH); identify the set without a choice function.
Next up: Mastery of forcing and constructibility equips you to understand how large cardinals constrain the possible models of set theory and prepares you for exploring consistency hierarchies, inner model theory, and descriptive set theory in the context of large cardinal axioms.
The comprehensive reference for modern set theory: covers large cardinals, forcing, inner models, and descriptive set theory. After Kunen, this is the book that opens every frontier of the subject.
The definitive historical and mathematical account of large cardinal axioms. Reading it last places everything learned in a rich intellectual context and points toward current research directions.
Discussion
Keep reading
Paths that share books, cover the same subject, or open a related topic.
