A short thread on graph theory and network science. Both have long histories, but I'll focus on two people: Frank Harary, the "godfather of modern (American) graph theory", and Duncan Watts, the "reinventor" of network sociology, which morphed into network science.
The idea that all kinds of bilateral relationships can be simplified to some kind of network structure is quite old (Euler's Königsberg bridge problem is canonical), but for the longest time it was considered "toy mathematics" at best.
Most academics should probably not publish more than 3-5 pieces in their lives. One of the few exceptions was Frank Harary at the University of Michigan, who was on a mission to show the world that everything can be explained with dots and lines: graphs.
"Why Michigan" you ask? In the 1950-60s, when Harary was conducting his pioneering research, Detroit and its automotive industry was the crown jewel of American industrial might, and its supply network reached into all corners of the Midwest.
The best book about this network is "My Years With General Motors" by Alfred Sloan (he of MIT Sloan School fame). The Detroit supply network reached schools like Carnegie Mellon, Case Western, the Big Ten schools with their proud engineering traditions.
The exigencies of Detroit also reshaped post-War economics, as the nonlinearities of production lines occupied the brightest minds of their generation. One Pittsburgh paint factory revolutionized finance, forecasting, enterprise computing, org theory, AI.
In a bit of a longer trajectory, this paint factory also spawned the saltwater-sweetwater divide in macro, along with a whole bunch of Riksbank prizes, mostly to the wrong guys.
The nexus of all of this was the University of Michigan in cushy Ann Arbor, and Frank Harary was the nexus of that nexus, along with people like John Holland, who pioneered genetic algorithms also as a tool to tackle a supply network problem.
Harary's prodigious work touched on combinatorics, computational complexity, routing, metaalgorithms, ... and pre-internet filter bubbles. At a time when graph theory as mostly considered a passing fad.
As a side note, one of my favorite books by Harary is the 1973 book on graphical enumeration with Ed Palmer, which not only spawned many investigations into combinatorics and network structure (my own included), but also gave us the quote for our time.
Academia is a small world, and it might have taken a circuitous route, but eventually graph theory reached sociology, more precisely Mark Granovetter, then at Johns Hopkins. And if you look up the references to his pioneering 1973 paper "The Strength of Weak Ties", you get... 
Granovetter's papers, including the 1985 follow-up "Economic Action and Social Structure" on embeddedness, made a big splash in sociology, but soon enough the field descended into a lengthy intellectual winter.
While network sociology became an "everything is interrated, yay" field brimming with conference speaker types, other fields like transportation science picked up the slack and used the newly available compute power to create urban network flow models.
Cue the internet. And with the early internet, early internet memes. One of the early internet memes was Six Degrees of Kevin Bacon, the everything-is-interrelated parlor game based on Stanley Milgram's small worlds research.
Duncan Watts, who was just about to move Granovetter's network sociology into the computer age, happily rode the meme to endowed chair and blue check status.
The Internet, built on the ARPAnet, made it possible to make everything interrelated, and exponential compute power and demand from Silicon Valley, graph theory morphed into network science: It was finally possible to talk about network structure.
Finally, who remembers the contributions of Frank Harary to the field? My hunch is the number is roughly...
While Watts et al revolutionized sociology, Keith Poole and Howard Rosenthal brought peer-to-peer interaction to political science and ideological polarization. People didn't pay attention to polarization back then, so they did it without media buzz.
@threadreaderapp unroll plz
