In a normal topological space X, for any two disjoint closed sets A and B there exists a continuous function f: X → [0,1] such that f equals 0 on A and 1 on B; this constructs a continuous separator between closed sets.

Principle

Normality (ability to separate closed sets by disjoint open neighborhoods) permits the construction of continuous maps that interpolate prescribed values on disjoint closed sets, realizing a basic partition-of-unity phenomenon.

Demonstration

In a metric space define f(x) = d(x,A)/(d(x,A)+d(x,B)) where d denotes distance; this continuous function takes value 0 on A and 1 on B and demonstrates Urysohn's lemma in the metric setting.

Misapplication

Attempting to construct such an f in a non-normal space can fail; spaces that are not normal (e.g., certain product topologies or pathological examples) may contain disjoint closed sets with no continuous separator.

Consequence

Urysohn's lemma is a stepping stone to the Tietze extension theorem and to partitions of unity: it provides concrete continuous functions used in embedding and extension arguments in topology and analysis.

Reversal

If no continuous function can separate two disjoint closed sets, the space cannot be normal; the failure of Urysohn's conclusion gives a certificate of non-normality.

Boundary

The lemma requires normality of the space (T1 + every two disjoint closed sets have disjoint neighborhoods); it does not hold in merely Hausdorff or in arbitrary topological spaces without additional separation axioms.

Semantic Tension

Urysohn's lemma is closely related to but weaker than extension theorems: it produces a separating function with fixed boundary values, whereas Tietze guarantees extension of arbitrary continuous functions from closed subsets to the whole normal space.

Synthesis

Urysohn's lemma translates the separation property of normal spaces into the existence of explicit continuous interpolants to [0,1], forming a practical tool to separate and extend functions in topological constructions.