Just a couple of brief points from Winner’s second chapter.
First, he argues very compellingly that we would be wrong to think that new technologies are adopted wholly, or even largely, for reasons of efficiency. For instance, he points out that in the 1880s, in Cyrus McCormick’s reaper manufacturing plant in Chicago, McCormick introduced a new machine to mold the casings of the reapers. Perhaps McCormick even claimed that the new machines would be more efficient, but they were not: they produced lower-quality casings at higher costs.
So why did McCormick bring the machines in? Because they could be manned by unskilled laborers, thus enabling McCormick to get rid of skilled laborers who had recently annoyed him by unionizing. Once he had broken the union, he dumped the inferior machines: they had done what he wanted them to do. But it had nothing to do with efficiency.
My second point leads to a question. Almost all of the examples Winner uses in this chapter come from the world of manufacturing: McCormick’s reaper factory, Engels’s thoughts about industrial organization in Manchester, various debates about nuclear power plants and their acceptable or unacceptable fuels. How do these ideas about technology and politics shift when the context is not manufacturing, but rather the production, management, and distribution of information? Do all the factors need to be thoroughly recalibrated, or can we get by with minor adjustments?
This is an excellent question, particularly since so many of the discussion about contemporary developments in information technology are premised on some notion of the essential characteristics of technologies or thinks (for example, the oft-repeated notion that "information wants to be free." Often these arguments are naively determinist: simply having access to networked communications will lead to xxx (where xxx = democracy, free market capitalism, secularization, whatever…)
The history that I turn to most in teaching on these issues is the history of the postal network in the United States, which is constructed, quite deliberately in such a way as to embody and reinforce representative democracy. Richard John's work is a good starting point, and David Henkins has a fabulous new book on the mid-19th century postal reform. There are similar discussions about the relationship between the architecture and "politics" of the telephone and telegraph networks. The weakest link is the Internet literature, which tends to fall into simplistic essentialism. One of the better histories here is Janet Abbate's book, which contains an extended discussion of the Cold War political context of the development of packet switching (and no, the arguments here is not that the ARPAnet was designed to survive a nuclear attack — it was not).
I'm very much interested in the question, too. Considering the date of the book, I observe that the U.S. has not yet shifted far to an information economy. The computing and communications revolutions of the 80s and 90s gave the information economy a serious boost, to say the least.
With respect to technology, the democratization of production been as one of the long-view effects of innovation and efficiency. But it has been modest in manufacturing and other materials-based processes, I think, but especially acute with anything in the digital realm. Many, many people now possess the tools to make films and photos (though the quality varies widely), else YouTube and Flicker would not exist. Similar effects are found in print and electronic publication. And although file-sharing isn't production per se, the ability of the user to duplicate and distribute media has resulted in content creators' inability to derive revenue from their own efforts.
So old expectations apply with respect to coffee beans and blue jeans: someone else makes them and we buy them. But with movies, music, news, books, etc., we are increasingly making our own and getting access to those made by others for free, often illegally.
I once read a great little book that was essentially a Marxist analysis of the screensaver. Alas, I have since lost track of the book, and don't have a copy handy, but if I remember correctly the basic argument was the screensavers, like many other computer technologies, provided the illusion of control and autonomy while reinforcing the structures of corporate surveillance and control. You might be able to choose your fonts, background colors, and screensaver (and email your friends, update Facebook, or post to your blog, all during work-time), but your corporate masters still own the means of production, appropriate your surplus value, deskill and routinize your labor, etc., etc. And with computers, they have increasingly fine-grained methods for monitoring your activities, enforcing rules and structuring your work, extracting your knowledge — and in the long term, replacing you altogether.
I am not endorsing the book or its argument, just noting its existence (there is also out there somewhere a similarly Marxist analysis of Dilbert: you might think by posting Dilbert on your cubicle you are engaged in an act of corporate resistance, but in the end its just false consciousness)
The point is that another way to think about the applicability of Winner to the information age might be to focus on the ways in which traditional social, economic, and political structures get reconstituted (and depoliticized) in information technologies. There is a little bit of literature on this: Paul Edward's The Closed World: Computers and the Politics of Discourse in Cold War America, Jon Agar's The Government Machine: A Revolutionary History of the Computer, Nathan Ensmenger's recent The Computer Boys Take Over: Computers, Programmers, and the Politics of Technical Expertise. The latter has a discussion of the ways in which particular programming languages are designed to accomplish social and organizational goals. COBOL, for example, was advertised as a language that "could be read by managers" (who often resented the power that the new breed of computer expert seemed to acquire within many organizations).
In any case, it seems to me that a productive way to think about how well Winner's analysis has aged over time is to think about the relationship between power and technology.