Analogical reasoning and policymaking: Where and when is it used? DAVID PATRICK HOUGHTON Department of Government, University of Essex,Wivenhoe Park, Colchester, Essex CO4 3SQ, U.K. Abstract. A steadily growing literature has emerged in recent years on the role of analogical reasoning in policymaking contexts. However, there has so far been little attempt to anwer the question of where and when analogical reasoning tends to be used. Using Yaacov Vertzberger's `situational-motivational nexus' framework as starting point, the article examines the decision of the Kennedy administration to wait it out during the substantial stock market crash of 1962. Little evidence is uncovered that the Kennedy and his advisers relied on analogical reasoning to reach this decision, a ¢nding which is surprising given the number of situational and motivational induce- ments present in the case. The article concludes that a high degree of perceived risk and uncertainty ^ noticeably absent from the stock market case ^ is the key situational inducement to analogizing, but suggests that the case tells us something important about the prevalence of rule- as opposed to case-based reasoning. Introduction Recent years have seen an explosion of interest in the phenomenon of `policy learning' among scholars of both domestic (Heclo, 1974; Neustadt and May, 1986; Sabatier, 1988; Hall, 1993, Rose, 1991, 1993; Bennett, 1991; Robertson, 1991; Majone, 1991; Hoberg, 1991; May, 1992; Bennett and Howlett, 1992; Sabatier and Jenkins Smith, 1993; Leeuw et al., 1994) and foreign policy (Ravenal, 1978; Etheredge, 1985; Breslauer and Tetlock, 1991; Levy, 1994; Reiter, 1994), re£ecting a growing desire to determine how and to what extent individuals, organizations, administrations and sometimes entiere nation-states `learn' from past policy failures and successes. Not everyone has expressed satisfaction with this new literature, however; various de¢ciencies have been highlighted early on. One particular problem has been the profusion of single case studies and/or general discussions of policy learning without any sustained or coordinated attempt to assess where and when such learning e¡ects occur. As Paul Pierson has perceptively noted in a recent essay, almost all of the new learning perspectives have `su¡ered from an emphasis on illustrating processes rather than establishing the frequency of these processes. Because arguments about policy learning have been developed largely through single case studies ... the question of this phenomenon's scope has hardly been asked, much less answered. Yet surely this is a critical concern' (Pierson, 1993: p. 616). He points out with a fair degree of accuracy that `to date, what we have are plausible accounts of isolated cases,' and that we need to ask `more precise questions' about the role played by learning in various contexts and under various con- ditions (ibid: p. 627). Policy Sciences 31: 151^176, 1998. ß 1998 Kluwer Academic Publishers. Printed in the Netherlands.