Generated by Codex with GPT-5

From a Walk to a Theory

Jack Murtagh’s article begins with a puzzle that looks like a civic pastime, not the seed of a new field. In 18th-century Konigsberg, residents wondered whether someone could walk through the city and cross each of its seven bridges exactly once. The bridges connected two islands and two riverbanks, and every attempted route seemed to force a repeated crossing. The question was simple enough for nonmathematicians to ask, but it resisted the mathematical tools available at the time.

Leonhard Euler’s insight was to stop treating the map as a map. The precise lengths of the bridges, the shapes of the islands, the width of the river and the city’s orientation did not matter. What mattered was only connection: which landmasses were linked to which others, and how many links each one had. By stripping the problem down to points and lines, Euler turned a local walking problem into an abstract structure now called a graph.

That act of abstraction is the article’s central lesson. Good mathematics often advances by discarding detail, but not randomly. It keeps the features that control the question and throws away the rest. In the Konigsberg case, the essential feature was not distance or area but relationship. Once the bridges became edges and the landmasses became vertices, a puzzle about one city became a question about every possible network.

Why the Route Was Impossible

The solution rests on a compact counting argument. If a walker passes through a landmass in the middle of a route, every entry must be matched by an exit. That means the number of bridges connected to any middle landmass must be even. Only the start and end points are allowed to have odd numbers of bridges, because the start needs one more exit than entry and the end needs one more entry than exit.

From this reasoning comes a general rule. A graph has a route that uses every edge exactly once if all its vertices have even degree, or if exactly two vertices have odd degree. The Konigsberg bridge layout had four odd-degree landmasses, so the desired walk was impossible. The failure was not a matter of poor planning or missed cleverness. It was built into the structure of the network.

This kind of path is now called an Eulerian path. The article notes that a superficially similar problem, finding a route that visits every vertex exactly once, is far harder. That version leads to Hamiltonian paths and to questions that sit near the heart of computational complexity. The contrast is useful: two problems can look nearly identical in everyday language while belonging to very different mathematical worlds.

The Power of Naming the Pattern

Euler reportedly dismissed the bridge puzzle at first because it did not fit neatly into geometry, algebra or ordinary counting. That discomfort was productive. The puzzle demanded a language for connection itself, a way to reason about form without measuring lengths or angles. In solving it, Euler helped launch graph theory, which now underlies ideas in computer networks, social networks, biology, transportation, neuroscience and the web.

The article also links the bridge problem to topology, the branch of mathematics concerned with properties that survive stretching, bending and deformation. Topology asks what remains true when exact measurements are no longer the point. That perspective is already visible in Euler’s move from a detailed city map to a simplified network: the important information survives even after the picture is radically reshaped.

The lasting takeaway is that abstraction is not an escape from the real world. It is one of mathematics’ best ways of finding what is structurally real beneath local details. Konigsberg’s residents asked whether a particular walk could be done. Euler answered that question by inventing a framework that made countless other questions askable. A failed stroll through a Prussian city became a proof that the right simplification can open an entire discipline.