Skip to content

Introduction to Topology

Topology is the branch of mathematics that studies properties of spaces that are preserved under continuous deformations. Two objects are considered topologically equivalent if one can be continuously deformed into the other without tearing or gluing.

The classic informal example: a coffee mug is topologically equivalent to a donut (torus). Both have exactly one hole, and one can be smoothly deformed into the other.

Formally, topology generalises the notions of closeness, continuity, and convergence from analysis and geometry, stripping away the rigid structure of distance and angles.

Topology arises by definition from several directions:

  • Analysis: The ε\varepsilon-δ\delta definition of continuity in Rn\mathbb{R}^n depends only on open sets, not on the specific metric. Topology abstracts this to general spaces.
  • Geometry: Many geometric properties (e.g., the number of holes in a surface) are invariant under continuous deformations.
  • Algebra: Topological spaces carry algebraic invariants (fundamental group, homology groups) that classify spaces up to topological equivalence.

Key figures include Euler (Königsberg bridges, 1736), Riemann (Riemann surfaces, 1850s), Cantor (set theory, 1870s—80s), Poincaré (algebraic topology, 1890s), Hausdorff (metric and topological spaces, 1914), and Brouwer (fixed point theorem, 1911).

A topological space is a set XX together with a collection T\mathcal{T} of subsets of XX (called open sets) satisfying:

  1. \varnothing and XX are in T\mathcal{T}.
  2. Arbitrary unions of elements of T\mathcal{T} are in T\mathcal{T}.
  3. Finite intersections of elements of T\mathcal{T} are in T\mathcal{T}.

The collection T\mathcal{T} is called a topology on XX. A set is closed if its complement is open.

Example 1.1 (Standard Topology on R\mathbb{R}). The standard topology on R\mathbb{R} has open sets as arbitrary unions of open intervals (a,b)(a, b). This gives the usual notion of continuity.

Example 1.2 (Discrete Topology). On any set XX, let every subset be open. This is the finest topology: it makes every function f:XYf : X \to Y continuous.

Example 1.3 (Indiscrete Topology). On any set XX, let only \varnothing and XX be open. This is the coarsest topology. Only constant functions are continuous from this space.

Example 1.4 (Cofinite Topology). On an infinite set XX, let UXU \subseteq X be open if U=U = \varnothing or XUX \setminus U is finite. This topology satisfies only the finite intersection property for open sets.

A basis B\mathcal{B} for a topology on XX is a collection of subsets of XX such that:

  1. For each xXx \in X, there exists BBB \in \mathcal{B} with xBx \in B.
  2. If xB1B2x \in B_1 \cap B_2 for B1,B2BB_1, B_2 \in \mathcal{B}, then there exists B3BB_3 \in \mathcal{B} with xB3B1B2x \in B_3 \subseteq B_1 \cap B_2.

The topology generated by B\mathcal{B} consists of all unions of elements of B\mathcal{B}.

Example 1.5. The collection of all open intervals (a,b)(a, b) in R\mathbb{R} is a basis for the standard topology. The collection of all open balls Br(x)B_r(x) in a metric space forms a basis.

A function f:XYf : X \to Y between topological spaces is continuous if for every open set VYV \subseteq Y, the preimage f1(V)Xf^{-1}(V) \subseteq X is open.

Proposition 1.1 (Equivalent Characterizations). The following are equivalent:

  1. ff is continuous.
  2. For every closed set CYC \subseteq Y, f1(C)Xf^{-1}(C) \subseteq X is closed.
  3. For every xXx \in X and every neighborhood VV of f(x)f(x), there exists a neighborhood UU of xx such that f(U)Vf(U) \subseteq V.

Proposition 1.2. The composition of continuous functions is continuous. The identity map idX:XX\mathrm{id}_X : X \to X is continuous.

Example 1.6. Any function from a discrete space to any topological space is continuous. A function from any topological space to an indiscrete space is continuous.

Definition. A homeomorphism is a bijection f:XYf : X \to Y such that both ff and f1f^{-1} are continuous. Homeomorphic spaces are considered topologically equivalent.

For a subset AXA \subseteq X of a topological space:

  • The closure A\overline{A} is the intersection of all closed sets containing AA.
  • The interior int(A)\mathrm{int}(A) is the union of all open sets contained in AA.
  • The boundary A=Aint(A)\partial A = \overline{A} \setminus \mathrm{int}(A).

Proposition 1.3 (Properties).

  • AAA \subseteq \overline{A}, and A=A\overline{A} = A iff AA is closed.
  • int(A)A\mathrm{int}(A) \subseteq A, and int(A)=A\mathrm{int}(A) = A iff AA is open.
  • A=A\overline{\overline{A}} = \overline{A}, int(int(A))=int(A)\mathrm{int}(\mathrm{int}(A)) = \mathrm{int}(A).

Example 1.7. In R\mathbb{R} with the standard topology:

  • (0,1)=[0,1]\overline{(0, 1)} = [0, 1], int([0,1])=(0,1)\mathrm{int}([0, 1]) = (0, 1).
  • Q=R\overline{\mathbb{Q}} = \mathbb{R} (the rationals are dense), int(Q)=\mathrm{int}(\mathbb{Q}) = \varnothing.
  • ((0,1])={0,1}\partial((0, 1]) = \{0, 1\}.

Subspace topology. If YXY \subseteq X, the subspace topology on YY is TY={UY:UTX}\mathcal{T}_Y = \{U \cap Y : U \in \mathcal{T}_X\}. A set is open in YY if it is the intersection of YY with an open set in XX.

Product topology. For two topological spaces XX and YY, the product topology on X×YX \times Y has basis {U×V:U open in X, V open in Y}\{U \times V : U \text{ open in } X,\ V \text{ open in } Y\}. Projection maps πX:X×YX\pi_X : X \times Y \to X and πY:X×YY\pi_Y : X \times Y \to Y are continuous.

Example 1.8. The product topology on R2=R×R\mathbb{R}^2 = \mathbb{R} \times \mathbb{R} coincides with the standard Euclidean topology. A basis is given by open rectangles (a,b)×(c,d)(a, b) \times (c, d).

A metric space (X,d)(X, d) induces a topology where UU is open if for every xUx \in U there exists ε>0\varepsilon > 0 such that Bε(x)UB_\varepsilon(x) \subseteq U. The open balls form a basis.

Proposition 1.4. Every metric space is Hausdorff: for any distinct x,yXx, y \in X, there exist disjoint open sets U,VU, V with xUx \in U, yVy \in V.

Example 1.9. The discrete metric (d(x,y)=1d(x, y) = 1 for xyx \neq y) induces the discrete topology. The usual metric on Rn\mathbb{R}^n induces the standard topology.

Problem 1. Show that the collection B={[a,b):a<b}\mathcal{B} = \{[a, b) : a < b\} forms a basis for a topology on R\mathbb{R} (the lower-limit topology, or Sorgenfrey line).

Solution. Every xRx \in \mathbb{R} belongs to [x,x+1)B[x, x+1) \in \mathcal{B}. If x[a,b)[c,d)x \in [a, b) \cap [c, d), then xx is in [max(a,c),min(b,d))[\max(a, c), \min(b, d)) which is in B\mathcal{B}. \blacksquare

Problem 2. Prove that f:XYf : X \to Y is continuous iff for every AXA \subseteq X, f(A)f(A)f(\overline{A}) \subseteq \overline{f(A)}.

Solution. Assume ff is continuous. Take yf(A)y \in f(\overline{A}), so y=f(x)y = f(x) with xAx \in \overline{A}. For any open VyV \ni y, f1(V)f^{-1}(V) is an open neighborhood of xx, so f1(V)Af^{-1}(V) \cap A \neq \varnothing. Thus Vf(A)V \cap f(A) \neq \varnothing, so yf(A)y \in \overline{f(A)}. Conversely, if the condition holds, let CYC \subseteq Y be closed. Then f(f1(C))f(f1(C))C=Cf(\overline{f^{-1}(C)}) \subseteq \overline{f(f^{-1}(C))} \subseteq \overline{C} = C, so f1(C)f1(C)\overline{f^{-1}(C)} \subseteq f^{-1}(C), hence f1(C)f^{-1}(C) is closed. \blacksquare

  1. Show that the standard topology on R\mathbb{R} is the same as the metric topology from d(x,y)=xyd(x, y) = |x-y|.
  2. Prove that continuous functions preserve compactness: if ff is continuous and KK is compact, then f(K)f(K) is compact.
  3. Show that every closed subset of a compact space is compact.
  4. Find the closure, interior, and boundary of QR\mathbb{Q} \subseteq \mathbb{R}.
  5. Prove that XX is Hausdorff iff the diagonal Δ={(x,x):xX}X×X\Delta = \{(x, x) : x \in X\} \subseteq X \times X is closed.