Small World Networks
Why Six Handshakes Connect You to a Stranger in Siberia
The most connected person you know is probably only two or three introductions away from someone who could change your life — and that's not luck, it's geometry.
The Idea
Most networks — social, biological, technological — seem chaotic when you're inside them. Billions of nodes, trillions of connections, no obvious order. Yet in 1998, mathematicians Duncan Watts and Steven Strogatz published a paper that revealed something quietly astonishing: many of these networks share a peculiar structural fingerprint. They called it the 'small world' property. The insight turns on two competing forces. Imagine a network where everyone only knows their immediate neighbours — highly clustered, but terrible for getting information anywhere fast. Now imagine the opposite: connections scattered entirely at random — fast travel between distant nodes, but no coherent local structure. Small world networks live in the tension between these extremes. They are mostly local and clustered, like a village, but laced through with a surprisingly small number of long-range shortcuts. Those shortcuts are doing extraordinary work. Adding even a tiny fraction of random long-distance links to an otherwise local network collapses the average path length — the number of steps between any two nodes — dramatically. The network stays tightly clustered in its neighbourhoods, but suddenly everyone is close to everyone else. This is why 'six degrees of separation' isn't just a cocktail party fact. It's a structural inevitability in any network that has this architecture. The world feels large. The graph is small.
In the World
In 2003, epidemiologist Martina Morris was trying to model how HIV spread through sexual contact networks in Uganda and the United States. The standard assumption — random mixing, like gas molecules bouncing off each other — produced models that didn't match reality. Infections spread faster and further than the maths predicted, then stalled in unexpected places. When Morris and her colleagues mapped the actual contact networks, the small world structure jumped out immediately. People weren't connecting randomly. They formed tight local clusters — couples, social circles, overlapping communities — exactly as the Watts-Strogatz model would predict. But within and between those clusters, a handful of individuals had a disproportionate number of connections reaching far outside their immediate group. These weren't random long-range shortcuts; they were specific people occupying specific structural positions in the network. Those few high-degree nodes — what network scientists call 'hubs' — were doing the majority of the long-distance transmission work. Remove them, and the network fragmented. Leave them in, and an infection that started in one tight cluster could skip across the continent in a handful of steps. The implication was sobering and clarifying at once: the geometry of the network, not just individual behaviour, was driving the epidemic. And it meant that targeted interventions — focused on hubs, on bridging nodes, on the shortcuts — could be orders of magnitude more effective than blanket approaches. The map of the disease was hiding inside the shape of the graph.
Why It Matters
Once you see small world structure, you start noticing the shortcuts everywhere — and, more importantly, the people who are them. In your professional life, the most valuable contacts are rarely the ones you know best. They're the people who sit at the bridge between your cluster and another one entirely: a former colleague now in a different industry, a friend who moves in very different social circles. Sociologist Mark Granovetter called this 'the strength of weak ties' — the loose acquaintances who carry genuinely new information precisely because they're connected elsewhere. The same logic applies to how ideas spread, how companies die or thrive, how misinformation travels. A message doesn't need to reach everyone directly; it just needs to find a few shortcuts. This also means fragility: small world networks are robust against random failures — knock out a random node and the path lengths barely change. But they are devastatingly vulnerable to targeted attacks on hubs. The internet, supply chains, even ecosystems all share this property. Knowing this changes how you read a room, build a team, or think about resilience. The question is never just 'who do I know?' — it's 'where do I sit in the graph?'
A Question to Ponder
Who in your life is a genuine shortcut — a connection that bridges two worlds that would otherwise never touch — and are you actually using that bridge?
Get a new one of these every morning.
Start learning with Thinkable