Much has been written about corruption in recent times. Indeed, as was noted on page 1, thousands of articles are written on corruption and related terms every year. This was not always the case. Indeed, corruption analysis was often seen as peripheral to academic life. Back in 1957, Eric McKitrick, writing in the Political Science Quarterly, bemoaned the fact that the “investigation of corruption” did not seem to be something of “very intense interest to social scientists” (McKitrick 1957: 502). Michael Johnston, one of the political scientists who has done the most to put corruption analysis back on the map, agrees; he noted in 2006 that “American political science as an institutionalized discipline has remained steadfastly uninterested in corruption for generations” (Johnston 2006: 809). Ten years later, he further lamented the fact that corruption had “more or less dropped off the academic and international policy agendas for nearly a generation” (Johnston 2016: 13). This former indifference of political scientists towards the study of corruption was best described by John Peters and Susan Welch at the end of the 1970s, in their now much-cited piece for the American Political Science Review (APSR), “If political corruption is in the mainstream of American politics, why is it not in the mainstream of American politics research?” (Peters & Welch 1978: 974).

Peters and Welch posed an interesting question. Even post-Watergate, an event that caused corruption to lurch into the mainstream of American public consciousness, corruption analysis remained notable by its absence. The leading social science journals in America rarely discussed it, and when they did, corruption was only infrequently regarded as a phenomenon that needed to be explained; it was corruption’s impact on other things that generally grabbed people’s attention. Peters and Welch blamed this apparent neglect on the challenge of defining the concept, but the prevailing analytical and methodological assumptions of their discipline no doubt also played a role. When your aim is to use data systematically to test hypotheses and you want to use increasingly sophisticated statistical techniques to do so, then corruption is by no means a straightforward concept to get a handle on.