Axiomatic Probability

Probability is the mathematical language for measuring uncertainty. Before the 20th century, probability was described informally — people spoke of 'equally likely outcomes' or 'long-run frequencies' without a precise foundation. Axiomatic probability, developed by the Soviet mathematician Andrei Kolmogorov in 1933, changed everything by building probability on a small set of rigorous axioms, just as Euclid built geometry on postulates. The axiomatic approach defines probability as a function that assigns a number between 0 and 1 to every event in a sample space, subject to three rules called Kolmogorov's axioms. Any model — whether describing a coin flip, the weather, or radioactive decay — is valid as long as it satisfies these three axioms. This universality makes axiomatic probability the backbone of modern statistics, physics, finance, and computer science. For high school students, axiomatic probability might seem abstract at first, but it quickly becomes a powerful tool. Once you accept the three axioms, you can derive every other probability rule — the complement rule, addition rule, inclusion-exclusion principle, and more — purely through logical deduction, with no additional assumptions required.

Key takeaways
  • The Sample Space Ω
  • Events as Subsets
  • Axiom 1 — Non-negativity
  • Axiom 2 — Normalization
  • Axiom 3 — Countable Additivity
  • Deriving Key Rules from the Axioms

The Sample Space Ω

The sample space Ω (capital omega) is the set of all possible outcomes of a random experiment. For a single die roll, Ω = {1, 2, 3, 4, 5, 6}. For a coin flip, Ω = {H, T}. For measuring the lifetime of a lightbulb, Ω = [0, ∞). Every outcome must belong to exactly one element of Ω, and together they cover every possibility.

Events as Subsets

An event A is any subset of Ω. For the die, 'rolling an even number' is the event A = {2, 4, 6}. The empty set ∅ is the impossible event (nothing happens), and Ω itself is the certain event (something always happens). The collection of all valid events is called a σ-algebra (sigma-algebra), which is closed under complements and countable unions — this technical requirement ensures our probability function is well-defined.

Axiom 1 — Non-negativity

For any event A, its probability P(A) is a real number satisfying P(A) ≥ 0. Probability can never be negative. This mirrors our intuition: you cannot have a negative chance of something happening. Whether P(A) = 0 (impossible), P(A) = 0.5, or P(A) = 1 (certain), it must always be non-negative.

Axiom 2 — Normalization

The probability of the entire sample space is exactly 1: P(Ω) = 1. This means something in the sample space must happen — the total probability is always 100%. If you roll a die, one of {1,2,3,4,5,6} must come up, so the probability of the full set equals 1. This axiom lets us interpret probability as a proportion of certainty.

Axiom 3 — Countable Additivity

If A₁, A₂, A₃, … is a sequence of mutually exclusive events (meaning Aᵢ ∩ Aⱼ = ∅ for i ≠ j), then P(A₁ ∪ A₂ ∪ A₃ ∪ …) = P(A₁) + P(A₂) + P(A₃) + … This is the most powerful axiom. It says probabilities of non-overlapping events simply add up. For a finite case, if A and B cannot both occur, P(A ∪ B) = P(A) + P(B).

Deriving Key Rules from the Axioms

From just these three axioms, we can prove everything else. Complement rule: since A and Aᶜ are mutually exclusive and together form Ω, Axioms 2 and 3 give P(Aᶜ) = 1 − P(A). Impossible event: P(∅) = 0 (proved by noting Ω ∪ ∅ = Ω and applying Axiom 3). Addition rule for non-mutually-exclusive events: P(A ∪ B) = P(A) + P(B) − P(A ∩ B). Each of these follows by pure logic from the three axioms.

The Key Equation

P ⁣(⋃i=1∞Ai)=∑i=1∞P(Ai)when Ai∩Aj=∅ for all i≠jP\!\left(\bigcup_{i=1}^{\infty} A_i\right) = \sum_{i=1}^{\infty} P(A_i) \quad \text{when } A_i \cap A_j = \emptyset \text{ for all } i \neq j
  • P — Probability measure (dimensionless): A function mapping events to real numbers in [0, 1] satisfying all three axioms.
  • A_i — Event i (dimensionless): The i-th event, a subset of the sample space Ω.
  • \bigcup — Union (dimensionless): The set of outcomes belonging to at least one of the events A₁, A₂, A₃, …
  • A_i \cap A_j = \emptyset — Mutual exclusivity condition (dimensionless): No two events share any outcomes; they cannot both occur simultaneously.

Worked Example

Problem. A standard six-sided die is rolled once. Define events: A = {rolling a multiple of 3} = {3, 6}, B = {rolling a 5} = {5}. Using the axioms, find: (a) P(A), (b) P(B), (c) P(A ∪ B), (d) P(Aᶜ).

P(A)=13,  P(B)=16,  P(A∪B)=12,  P(Ac)=23P(A)=\tfrac{1}{3},\; P(B)=\tfrac{1}{6},\; P(A\cup B)=\tfrac{1}{2},\; P(A^c)=\tfrac{2}{3}

Answer. P(A) = 1/3, P(B) = 1/6, P(A ∪ B) = 1/2, P(Aᶜ) = 2/3

Figure
Diagram
Kolmogorov's Three Axioms — Visual SummaryAxiom 1: Non-negativityFor every event A: P(A) ≥ 0Probability is never negative.Axiom 2: NormalizationP(Ω) = 1The certain event has probability exactly 1.Axiom 3: Countable AdditivityIf A ∩ B = ∅, then P(A ∪ B) = P(A) + P(B)Probabilities of non-overlapping events add up.All other rules follow by logical deduction
Origins · 1 minAndrei Nikolaevich Kolmogorov and how this was discovered
AN
Andrei Nikolaevich Kolmogorov
1903–1987 · Soviet (Russian)

In 1933, Kolmogorov published 'Grundbegriffe der Wahrscheinlichkeitsrechnung' (Foundations of the Theory of Probability), in which he established the three axioms that now bear his name. This work unified the previously fragmented approaches to probability — classical, frequentist, and geometric — into a single rigorous framework built on measure theory. His axioms turned probability from an intuitive concept into a formal branch of mathematics. Kolmogorov made groundbreaking contributions to at least seven different areas of mathematics, including topology, turbulence theory, information theory, and the theory of algorithms. He was also passionate about education and founded a special boarding school in Moscow for mathematically gifted teenagers.

  • Sample Spaces and Events — The sample space Ω and the concept of events as subsets are the objects that axiomatic probability is defined on — you must understand these before the axioms make sense.
  • Conditional Probability — Conditional probability P(A|B) = P(A ∩ B)/P(B) is derived directly from Kolmogorov's axioms and is one of the most important theorems built on the axiomatic foundation.
  • Complement and Addition Rules — These everyday probability rules — P(Aᶜ) = 1 − P(A) and P(A ∪ B) = P(A) + P(B) − P(A ∩ B) — are theorems proved directly from the three Kolmogorov axioms.
  • Set Theory Basics — Events in probability are sets, and operations like union, intersection, and complement come directly from set theory — making set theory essential background for axiomatic probability.