Stephen Turner

Robert Merton's essays on theories of the middle range and his essays on functional explanation and the structural approach are among the most influential in the history of sociology. But their import is a puzzle. He explicitly allied himself with some of the most extreme scientistic formalists and contributed to and endorsed the Columbia model of theory construction. But Merton never responded to criticisms by Ernest Nagel of his arguments or acknowledged the rivalry between Lazarsfeld and Herbert Simon, rarely cited the philosophical and methodological literature, and responded to critics with ambiguous concessions, leaving the Mertonian legacy profoundly ambiguous. Key Words: Robert Merton • Paul Lazarsfeld • theory construction • middle range theory • causal modeling • Émile Durkheim.

The writing of history typically involves opinions that cannot be established by historical evidence. This 'involvement' takes two main forms: first, the intimation of evaluative opinions is often the point of historical narratives; and second, as Weber maintained, opinion plays a constitutive role-the identification of historical objects, of explanatory problems, and perhaps even the selection of solutions to these problems is governed by opinions or commitments that cannot be proven historically. The comments of both Bulmer and Camic, for example, presume the validity of opinions about what was and was not important in the history of American sociology. I reject these opinions. In the book, the rejection was intimated. Perhaps the most useful thing I can do in this reply is to identify and sharpen some of the differences in unprovable opinions between myself and the critics though I must say that they were charitable and mild critics indeed-and make explicit some of the things that I wished to intimate in the historical discussion but could not prove. There are two reasons for doing this: Most of the questions raised by each of the commentators can be more easily addressed if these hidden themes and differences of opinion are made more explicit. A more important reason is this: some of these conflicts might be turned into answerable historical problems.