Making social science matter: why social inquiry fails and how it can succeed again

Abstract

ABSTRACT This book is a thoughtful antidote to the simple views that see social science as a science like any other-positivistic science. It begins with a well-grounded empirical case of the development and application of expert knowledge, then moves through consideration of context and values, the centrality of power, and a reconsideration of the Greek roots of modern knowledge. It concludes with some salient observations based on the author's own feedback and research practice. Making Science Matter: Why Inquiry Fails and How Can Succeed Again. Bent Flyvbjerg: Cambridge: Cambridge University Press, 205 Pages. INTRODUCTION is one of those phrases that, as an apprentice social scientist, I wished I had heard about 30 years ago. I was studying for a doctorate degree then, a process in which much musing was centered on some central issues in the philosophy of social science. In the texts that I read and the debates that I encountered there was a lot of about physics envy, that is. What was in short supply was much to make one feel confident as a social science researcher. Instead, rather like a character in a Beckett play, one was waiting for a character-the great deliverer of a truly social science-that never shows up. So I had to write a thesis regardless. Of course, there were various sightings that raised anticipations in different audiences, at different times: some barracked for Parsons, others for Marx, as a great unifier while others just got on with and didn't think too much about what it might mean. They just did it. How they did seemed to be based on an assumption that, while social science isn't physics, does have some formal similarities. It has hypotheses; contains propositions; and, maybe, some covering law-like explanations. (One colleague claims that the universal relationship between the size of organizations and their increasing need for centralization is such a relation.) If only Bent Flyvbjerg's book had been available then! BEYOND THE SCIENCE WARS The author begins with the hoax played by the physicist, Alan Sokal, one of the editors of the journal Text. Most people probably know about it. Sokal (1996) submitted an article that appeared to deconstruct physics and the editors accepted and published it. Thus was the latest salvo in the debate about the two cultures fired across the bows of the global social science community. The implications were clear: for bona fide-not to say macho-real scientists, social science is unreal and unnatural: simplistic in its assumptions, short on quality controls, and peopled by beings of lesser judgment, if not intelligence, than those to be found in the Natural Science Faculties. Flyvbjerg (page 3) is quite explicit about the implications of such warfare: Social science is locked in a fight that cannot hope to win, because has accepted terms that are self-defeating. What is to be done to change these terms and avoid defeat? Go back to Aristotle and start over again, says Flyvbjerg. Eschew Aristotle's path to knowledge that routes through the virtues of either techne or episteme and instead use Aristotle's account of a prudent and wise science-one founded on phonesis-and integrate with a Foucauldian conception of power. The route to Aristotle turns out to be quite contemporary-it works through the phenomenology of human learning in the light of the Dreyfus and Dreyfus (1986) model. This is the model that demonstrates that expert learning goes far beyond knowing and using the rules to accomplish an activity. Indeed, such explicit knowledge and referral is, in fact, counter to expertise that seems to rely on tacit knowledge and intuition embedded in context much more than does on explicit and disembedded knowledge. Physics envy is closely related to Cartesian anxiety-the fear of nihilism and relativism that lies outside the borders of a strict analytical and rational scientific tradition. …

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 140 citations exist upstream.