TO SOME, "POLITICAL SCIENCE" IS A CONTRADICTION IN TERMS. They regard politics as an art which defies systematic study and, hence, offers no basis for a "science." Others contend, as I do, that human behavior is subject to systematic study, explanation, and prediction-and this includes man's political behavior. While the attitude of the professional student of politics toward this issue may still reveal his attitude toward the computer as a useful or even "legitimate" tool in his research, the argument over the "behavioral approach" in political science is fast becoming irrelevant to computer applications in political research. Not only is the computer becoming a "conventional" research tool in patently humanistic studies like literature,1 music,2 and art,s but it is also winning favor as a useful aid to hard-nosed professional politicians-witness the conference held in Chicago last spring on data processing for Republican party workers.4 Exactly how have computers been used in political science? The version of this paper to be published in the FJCC Proceedings seeks to answer this question by reviewing actual computer applications in three methodological categories: data analysis, information processing, and simulation. In this abbreviated version, I will select for review only the use of computers in information processing, the application most likely to be of interest to readers of Computers and the Humanities. In contrast to "data analysis," where input to the computer is in numerical form, "information processing" utilizes the computer capabilities for accepting natural language text as input. This is an important feature of the computer for political scientists, whose material often resists easy quantification. Consequently, we find a considerable amount of research being done with the