Reading :: Networks of Power

Originally posted: Mon, 04 Oct 2004 21:38:56
Networks of Power: Electrification in Western Society, 1880-1930
by Thomas P. Hughes
Spend any time at all in the sociotechnical studies literature (broadly speaking) and you'll run across the title of this book. Latour cites it, Law cites it, Bazerman draws liberally from it, and Pickering has a detailed critique. The book is a historian's examination of how Western society became electrified between 1880 and 1930, focusing on the US, England, and Germany. Since my current project (on telecommunications) has something to do with the construction of similar networks, I decided I had to read it. So I ordered a used copy from Amazon.
The book, frankly, is daunting because it is so immense. This thing is the size of a telephone directory -- 465 pages. It's taken a while to get through it, especially since I ended up reading other books in the meantime to avoid being bogged down. But the book is worth it -- even though I have to agree with Pickering's sharp critiques of it. Networks of Power provides an unusually thorough discussion of sociotechnical networks, supported by painstaking research and some strong theoretical work. Perhaps because he is a historian, Hughes often provides more than sufficient evidence to support his points, and sometimes I think the points get lost in the narrative, but I strongly suspect the difference is one of discipline rather than unfocused writing per se.
This book is rich enough that I'm not even going to try to summarize it. Instead, I want to draw out a couple of key points.
One key point is that sociotechnical systems -- which Hughes regards as cultural artifacts -- have what he calls unique styles. These styles are shaped by cultural factors such as "geographical, economic, organizational, legislative, contingent historical, and entrepreneurial conditions" (p.405). These factors shape the style of a sociotechnical system so profoundly that styles are easily recognized (p.443) and a system of one style simply can't work elsewhere if lifted and transplanted whole. Chapter 14 is devoted to this proposition and makes the case in great detail. The foundation of the case is laid much earlier, though, in Chapter 3: Hughes argues with great success that when one applies a sociotechnical system from one set of conditions to another (e.g., when Edison tried to reproduce his New York system in England and Germany), one must see the work as technology transfer rather than as simply applying the same solution elsewhere (p.47; see also Bowker's book on Schlumberger, Akrich's work with the gazogene, etc.). Hughes' real contribution here is in doing the hard work of demonstrating this point across the mentioned factors, providing us with several examples and with a set of factors to apply to other cases. I'll certainly be using them.
Another helpful point is what Hughes calls the reverse salient, a lag in development of part of the system (p.79). (Hughes prefers this term to "bottleneck" or "disequilibrium" because of problems with those two metaphors.) "The reverse salient usually appears as a result of accidents and confluences that persons presiding over or managing the system do not foresee, or, if they do foresee them, are unable to counter expiditiously" (p.80). One example included the problem of the high cost of transmission that was faced in the direct current system.
Now to the chief problem. Like Pickering, I think that Hughes makes a mistake in separating technology and politics. He argues that in Chicago, technology was preeminent; in London, politics were; and in Germany, both participated equally. The problem, of course, is that it's not especially illuminating to separate the two, as Hughes seems at the verge of admitting several times. As his analysis suggests over and over -- particularly in the chapter on style -- technology is political and politics are technological.
In all, this is a really useful (though overly long) study and I'll certainly be using it a lot.

Blogged with Flock