42 I refer here to a paper by Donald MacKenzie (“The Credit Crisis as a Problem in the Sociology of Knowledge”), who argues that a shift in cultures of evaluation within the ratings agencies, from older corporate collateralized debt obligations to the newer CDOs composed of mortgage-backed securities, accounted for a number of “slips” when it came to evaluation of the dangers posed by the latter. Of course, MacKenzie realized that the narrative of “technological error” seems on its face implausible (pp. 1830–32); but by focusing so intently upon the narrowly confined world of the low-level analysts and traders, he misses most of the action surveyed in this volume.

Philip Mirowski, Never Let a Serious Crisis …, loc. 2345