Graph Theory
The Seven Bridges That Broke Mathematics Open
A city's walking-tour puzzle, dismissed as a mere curiosity in 1736, accidentally invented a field of mathematics that now keeps the internet alive.
The Idea
Graph theory begins with a deceptively simple question: can you cross all seven bridges of Königsberg exactly once and return to where you started? Leonhard Euler, arguably the most prolific mathematician in history, proved it was impossible — but more importantly, he showed *why*, and in doing so created an entirely new way of thinking about structure. The insight was to strip the problem down to its skeleton. The actual geography of Königsberg — the rivers, the landmasses, the architecture — was irrelevant. What mattered was only this: how many connections does each region have? Euler realised that for a complete circuit to be possible, every node in the network must have an even number of connections. Königsberg's bridges violated this rule at every point. Case closed, but the method was revelatory. What Euler had quietly invented was the concept of a *graph* — not a bar chart, but a mathematical object made of vertices (points) and edges (connections between them). No distances, no coordinates, no geometry in the traditional sense. Just relationships. This act of radical abstraction — caring only about what connects to what, not where anything sits in space — turned out to be one of the most powerful moves in the history of mathematics. It gave us a language precise enough to describe anything from a social network to a protein interaction to the routing logic of the internet.
In the World
In the 1990s, two PhD students at Stanford were trying to understand which academic papers were genuinely important — not just frequently cited, but cited by *other frequently-cited papers*. Larry Page and Sergey Brin treated the entire web as a graph: every webpage was a vertex, every hyperlink was a directed edge. Their insight, which became PageRank and then Google, was essentially Eulerian in spirit — forget the content of the pages, just study the structure of the connections. The same logic applies in biology. When researchers mapped the protein interaction network of a simple yeast cell, they found it shared a structural property with the internet, with power grids, and with the network of Hollywood actors connected by shared film credits: a small number of nodes with an enormous number of connections (hubs), and a vast majority of nodes with very few. This 'scale-free' pattern, invisible to the naked eye, only becomes legible when you represent the system as a graph. Epidemiology runs on graph theory too. During the early spread of COVID-19, contact-tracing models weren't primarily about biology — they were about network topology. Who connected to whom, and how densely? Superspreader events made sense not because of viral load alone, but because certain individuals occupied high-degree nodes in the social graph, broadcasting outward to dozens of loosely-connected clusters at once.
Why It Matters
Graph theory is one of those rare mathematical frameworks that becomes more useful the more different the problems it touches. The same conceptual toolkit — vertices, edges, degree, connectivity, paths — lets you ask structurally identical questions about wildly different systems. That's not a coincidence; it's the reason abstraction is so valuable. Knowing this changes how you look at complexity. When a system feels overwhelmingly complicated — a supply chain, a friendship group under stress, a failing organisation — it's often because you're focused on the content of each node rather than the shape of the connections. Graph theory trains you to ask different questions: Where are the hubs? What happens if this single connection breaks? Which nodes are bridges between otherwise separate clusters? You don't need to do the mathematics to benefit from this instinct. Thinking in graphs means noticing that influence in a network rarely travels evenly — it concentrates, it bottlenecks, it shortcuts. Most real-world failures of communication, coordination, or resilience have a network topology underneath them. Euler just gave us the language to see it.
A Question to Ponder
In the networks you're part of — professional, social, informational — where are the bridges: the single connections whose removal would split the whole structure in two?
Get a new one of these every morning.
Start learning with Thinkable