Etymology & Roots

An axiom is a statement adopted without proof as a starting point within a mathematical or logical system. Other statements may then be derived from the axioms using accepted rules of inference.

$$ \boxed{ \text{axioms} + \text{rules of inference} \longrightarrow \text{theorems} } $$

In modern mathematics, calling a statement an axiom does not necessarily mean that it is regarded as an obvious, universal, or metaphysically certain truth. It means that the statement has been selected as part of the assumptions defining the mathematical framework being studied.

Thus the role of an axiom is structural:

$$ \boxed{ \text{axiom} = \text{unproved starting assumption of a formal theory} }. $$

Different theories may use different collections of axioms, and changing an axiom can produce a different mathematical system.

Operational Nuance

Axioms as Starting Assumptions

A mathematical theory begins by specifying certain primitive objects, terminology, axioms, and rules of reasoning.

Schematically:

$$ \boxed{ \text{language} + \text{axioms} + \text{inference rules} = \text{formal theory} }. $$

Axioms Are Not Proved within the Same System

An axiom is not normally established by proving it from earlier results inside the theory. Instead, it is available for use at the beginning of derivations.

If a statement can be derived from the existing axioms, it functions as a theorem rather than as an additional necessary axiom.

Theorem

A theorem is a statement proved from axioms, definitions, earlier theorems, and accepted inference rules.

$$ \boxed{ A_1,A_2,\ldots,A_n \vdash T } $$

can be read as:

"The theorem $T$ is derivable from the axioms $A_1,\ldots,A_n$."

Axiom versus Theorem

$$ \boxed{ \begin{array}{c|c} \text{Axiom} & \text{Theorem}\\ \hline \text{assumed within the theory} & \text{proved within the theory}\\ \text{starting premise} & \text{derived conclusion} \end{array} } $$

Axiom versus Definition

A definition establishes the meaning of a term or symbol. An axiom asserts a relationship or condition that the theory assumes.

For example, defining an even integer as an integer of the form:

$$ n=2k, \qquad k\in\mathbb Z, $$

is a definition, not an axiom.

Axiom versus Postulate

Axiom and postulate have often been used with overlapping meanings.

A traditional distinction sometimes treats:

Modern usage is less rigid, and many authors use the words nearly interchangeably. Euclidean geometry, for example, is traditionally described using postulates, while modern formal treatments may call such assumptions axioms.

Example: Equality

A formal treatment of equality may include principles such as reflexivity:

$$ \boxed{ a=a }. $$

Depending on the logical framework, such properties may be built into the logic or introduced axiomatically.

Euclidean Geometry

Euclidean geometry is built from foundational assumptions governing points, lines, distances, angles, and parallelism.

The famous parallel condition can be represented in modern form by statements equivalent to the assertion that through a point outside a given line there is exactly one line parallel to the given line.

Changing an Axiom Can Change the Theory

Altering the parallel assumption leads to non-Euclidean geometries.

This is an important conceptual lesson:

$$ \boxed{ \text{different axioms} \longrightarrow \text{different mathematical theories} }. $$

The resulting theories may each be internally meaningful if their axioms are interpreted coherently.

Peano-Style Arithmetic

Arithmetic can be axiomatized by describing properties of natural numbers and a successor operation.

A simplified example of the intended structure is:

$$ 0,\ S(0),\ S(S(0)),\ldots $$

together with axioms governing the behavior of $0$, successors, and induction.

Mathematical Induction

In axiomatic arithmetic, induction is not merely a classroom technique. It is tied to foundational principles describing the natural numbers.

A typical induction principle has the form:

$$ \boxed{ P(0) \land \forall n\,[P(n)\rightarrow P(n+1)] \Rightarrow \forall n\,P(n) }. $$

Set-Theoretic Axioms

Modern mathematics is often formalized within an axiomatic set theory. Such systems specify principles governing the existence and construction of sets.

These axioms regulate operations and constructions such as:

Axiom Schema

An axiom schema is a rule describing an entire family of axioms rather than a single sentence.

Each admissible substitution or formula generates a particular axiom instance.

$$ \boxed{ \text{axiom schema} = \text{pattern generating many axioms} }. $$

Formal Language

Axioms in a formal system are expressed in a specified formal language. The language determines which symbols, variables, predicates, and formation rules are permitted.

Model of an Axiomatic System

A model is an interpretation in which the axioms of a theory are satisfied.

If a structure $\mathcal M$ satisfies an axiom $A$, one writes:

$$ \boxed{ \mathcal M\models A }. $$

Semantic Consequence

If every model satisfying a collection of axioms $\Gamma$ also satisfies a statement $\varphi$, one writes:

$$ \boxed{ \Gamma\models\varphi }. $$

This means $\varphi$ is a semantic consequence of $\Gamma$.

Syntactic Derivability

If $\varphi$ can be formally proved from $\Gamma$ using the system's inference rules, one writes:

$$ \boxed{ \Gamma\vdash\varphi }. $$

The symbols $\models$ and $\vdash$ therefore describe different notions: truth in all models versus formal derivability.

Consistency

A theory is consistent if it does not permit derivation of a contradiction.

Informally:

$$ \boxed{ \text{not both } P \text{ and } \neg P \text{ are derivable} }. $$

Consistency is a central requirement when evaluating an axiom system.

Why Contradictions Matter

In ordinary classical logic, if a contradiction is derivable, the theory becomes unusable for distinguishing mathematical consequences because arbitrary statements can then be derived through the principle of explosion.

This makes consistency fundamental to formal mathematics.

Independence of an Axiom

An axiom is independent of the remaining axioms if it cannot be derived from them and its negation cannot be derived from them, under the relevant consistency assumptions.

Intuitively:

$$ \boxed{ \text{the remaining axioms do not decide it} }. $$

Redundant Axiom

If an alleged axiom can already be proved from the other axioms, it is logically redundant as a foundational assumption.

Removing it does not reduce the deductive strength of the system.

Independent Axiom Systems

A collection of axioms is often called independent when no individual axiom can be derived from the others.

Completeness

In one important sense, a theory is complete if for every sentence $\varphi$ expressible in its language, either:

$$ \varphi $$

or:

$$ \neg\varphi $$

is derivable.

Completeness in this sense should not be confused with other mathematical uses of the word complete.

Soundness

A deductive system is sound relative to its intended semantics if whatever can be derived is valid in the intended models.

$$ \boxed{ \Gamma\vdash\varphi \Rightarrow \Gamma\models\varphi }. $$

Axioms and Logical Rules

Axioms and inference rules play different roles.

An axiom supplies an accepted statement. An inference rule licenses passage from certain statements to another.

For example, modus ponens has the form:

$$ \boxed{ P, \qquad P\rightarrow Q \quad\Longrightarrow\quad Q }. $$

Logical Axioms versus Mathematical Axioms

Some formal systems distinguish between:

For example, assumptions specific to groups or sets are mathematical rather than purely logical.

Axioms for Algebraic Structures

Many familiar mathematical objects are characterized axiomatically.

A group, for instance, is a set with an operation satisfying requirements involving:

These properties function as defining axioms for the structure.

Associativity as an Axiom

For a group operation $\ast$:

$$ \boxed{ (a\ast b)\ast c = a\ast(b\ast c) } $$

for all elements $a,b,c$.

In group theory this is not proved from the other group properties; it is part of what it means for the operation to define a group.

Axiom Systems as Definitions of Structures

In abstract mathematics, axioms often characterize an entire class of objects. Rather than describing one particular structure, they specify the conditions that every structure of that type must satisfy.

$$ \boxed{ \text{axioms} \longrightarrow \text{class of admissible models} }. $$

Axioms Need Not Be Obvious

A common misconception is that an axiom must be self-evident.

Modern axiomatic mathematics does not require this. Axiom systems are often chosen because they define a useful structure, capture intended properties, or provide a productive foundation for deduction.

Axioms Need Not Be Unique

The same mathematical theory may sometimes be described by different but equivalent axiom systems.

One formulation may be shorter, more intuitive, more algebraic, or better suited for a particular proof technique.

Equivalent Axiom Systems

Two axiom systems may be regarded as equivalent when they generate the same relevant consequences or characterize the same intended structures.

Minimal Axiom Systems

Mathematicians may seek axiom systems with unnecessary assumptions removed. A minimal or independent presentation can clarify exactly which assumptions are responsible for which results.

Conditional Nature of Theorems

A theorem derived from axioms has the logical character:

$$ \boxed{ \text{if the axioms hold, then the theorem follows} }. $$

This conditional viewpoint is central to understanding modern formal mathematics.

Proof by Axiomatic Deduction

A proof may proceed through a chain such as:

$$ A_1,A_2 \Rightarrow L_1 \Rightarrow L_2 \Rightarrow T, $$

where the $A_i$ are axioms, the $L_i$ are intermediate lemmas, and $T$ is the theorem being established.

Primitive Terms

Some axiomatic systems begin with certain terms left undefined, called primitive terms.

Their behavior is characterized indirectly through the axioms.

In geometric axiomatizations, words such as point, line, and incidence may serve this role.

Why Primitive Terms Are Useful

Attempting to define every term using still earlier terms would create an endless regress. Formal systems therefore begin with some primitive vocabulary and constrain it through axioms.

Axioms and Models

An axiom system can have multiple models. What matters is that each model satisfies the required relationships.

The axioms characterize structure rather than necessarily naming one particular physical interpretation.

Mathematics versus Empirical Science

In pure mathematics, axioms function as assumptions defining a formal framework. In empirical sciences, starting principles may additionally be judged by how well their models describe observations.

This is an important difference between axiomatic deduction and empirical testing.

Common Misconception: "An Axiom Is Proven True"

If a statement has been proved from earlier assumptions, then relative to those assumptions it is functioning as a theorem or derived proposition.

Calling something an axiom instead means that it occupies a starting role in the selected formulation.

Common Misconception: "Every Axiom Is Universally True"

An axiom is relative to a theory.

A statement may be adopted in one system, rejected or replaced in another, or left undecided by a third.

$$ \boxed{ \text{axiom status is theory-dependent} }. $$

Common Misconception: "Axioms Cannot Be Questioned"

Mathematicians routinely study what happens when axioms are removed, weakened, strengthened, or replaced.

The axiomatic method makes those assumptions explicit precisely so their consequences can be investigated systematically.

Written vs. Spoken Syntax

Symbolic Notation (Written) Professional Articulation (Spoken)
$\Gamma\vdash\varphi$ "Phi is derivable from the assumptions Gamma."
$\Gamma\models\varphi$ "Gamma semantically entails phi," or "every model of Gamma satisfies phi."
$\mathcal M\models A$ "The model M satisfies axiom A."
$(a\ast b)\ast c=a\ast(b\ast c)$ "The operation star is associative."
$P,\ P\rightarrow Q\vdash Q$ "From P and P implies Q, infer Q."
$A_1,\ldots,A_n\vdash T$ "T is provable from axioms A one through A n."

In elementary mathematics, one often says simply, "Assume this axiom" or "By the axioms of the system." In formal logic, the relationship between axioms and conclusions may be expressed explicitly with symbols such as $\vdash$ and $\models$.

Defense Notation Advice: Avoid saying that an axiom is "proved true." Say instead that it is assumed, adopted, or postulated within the theory. Then distinguish statements derived from it as theorems, lemmas, corollaries, or other proved propositions.

Case Usage & Field Variations