A theorem giving necessary and sufficient conditions for a contravariant homotopy functor on suitable homotopy categories (classically pointed CW complexes) to be representable by a space or spectrum, i.e., naturally isomorphic to homotopy classes of maps into a classifying object.

Principle

If a contravariant functor satisfies homotopy invariance, the wedge (coproduct) axiom (sending coproducts to products), and a Mayer–Vietoris or exactness condition with respect to homotopy colimits, then there exists an object X representing the functor so that the functor is naturally isomorphic to [–, X] or to maps into a spectrum; the organizing idea is that good homotopy-theoretic behaviour characterizes representability.

Demonstration

One uses Brown representability to construct spectra representing generalized cohomology theories: a cohomology functor satisfying the Eilenberg–Steenrod-like axioms (except dimension) extends to a representable functor, producing the classifying spectrum whose homotopy classes of maps compute the cohomology groups.

Misapplication

Attempting to apply Brown representability to functors that fail the wedge axiom, are not homotopy invariant, or are defined on categories lacking necessary smallness or closure properties (for instance wildly large categories without CW-type objects) yields nonrepresentable examples and misuse.

Consequence

When applicable it produces classifying spaces or spectra, allowing one to realize abstract cohomology theories as representable objects, to apply Yoneda-style arguments in homotopy theory, and to transfer homotopical problems into mapping-space questions.

Reversal

Nonrepresentability is the contrast: functors that violate the axioms exhibit genuinely nonclassifiable behaviour and cannot be encoded by a single universal object; this highlights limits of classification by spaces or spectra.

Boundary

Hypotheses include working over the appropriate homotopy category (often pointed CW complexes), assuming smallness or set-theoretic conditions, and verifying the wedge and exactness axioms; the theorem does not apply verbatim to arbitrary model categories or to functors lacking these axioms without suitable replacements.

Semantic Tension

Tension exists between representability in homotopy theory and purely algebraic Yoneda representability: Brown's theorem replaces algebraic exactness axioms by homotopy-theoretic coherence conditions, so one must reconcile categorical representability notions with homotopical axioms.

Synthesis

Brown representability identifies when a homotopy functor is realized by maps into a single object: demanding homotopy invariance, coproduct-to-product behaviour, and homotopical exactness guarantees a representing space or spectrum, thereby turning abstract functors into concrete classifying objects and enabling geometric methods for cohomological classification.