The five features are (pp. 47-48):
- “the ability of technologies of transportation and communication to facilitate control over events from a single center or small number of centers.”
- “a tendency for new devices and techniques to increase the most efficient or effective size of organized human associations.”
- “the way in which the rational arrangement of socio-technical systems has tended to produce its own distinctive forms of hierarchical authority.”
- “the tendence of large, centralized, hierarchically arranged sociotechnical entities to crowd out and eliminate other varieties of human activity.”
- “the various ways that large sociotechnical organizations exercise power to control the social and political influences that ostensibly control them.” (e.g., regulatory capture)
While there are widely accepted criteria for placing regulatory limits on technology--Winner notes five (threats to health and safety, exhaustion of a vital resource, degrading environmental quality, threats to natural species and wilderness, and causing “social stresses and strains of an exaggerated kind,” pp. 50-51)--he suggests that these are insufficient. He cites a study by colleagues of electronic funds transfer (EFT) which suggested that it “would make possible a shift of power from smaller banks to larger national and international institutions” and create problems of data protection and individual privacy. But those problems don’t seem to fall under his five criteria, so he suggested, ironically, that “their research try to show that under conditions of heavy, continued exposure, EFT causes cancer in laboratory animals” (p. 51). Although I’d be surprised to find that EFT by itself had the effect Winner suggests, the recent global financial crisis as shown problems with allowing financial institutions to become “too big to fail” and motivated financial reform proposals (e.g., Sen. Dodd’s bill that would create new regulatory power over institutions with more than $50 billion in assets, including the ability to force such institutions into liquidation--“death panels” for large financial institutions).
In the 22 years since Winner’s book was published, most of his five features seem to continue to be relevant to developments such as the Internet. With respect to (2),(3), and (4) the Internet has greatly reduced the costs of organizing and allowed for social (non-market) production of goods. But the mechanisms which ease the creation of small, geographically dispersed groups have also facilitated the creation of larger groups, new kinds of hierarchical authority, and new kinds of centralization and monitoring (e.g., via applications used by hundreds of millions of people, provided by companies like Google, Facebook, and Twitter). It’s also allowed for new forms of influence by the same old powers-that-be, via techniques like astroturfing and faux amateur viral videos.
[A slightly different version of the above was written as a comment for my Human and Social Dimensions of Science and Technology core seminar. Thanks to Tim K. for his comments (though I declined to move the paragraph you suggested).]