The traditional organization of universities honors specific substantive foci – e.g., biology, psychology, mathematics. The units bring with them a set of questions about different phenomena of interest. What is life and how do living organisms function? How does the human mind function and how does it affect behavior? What is the underlying logic of shape, quantity, and arrangement?
The disciplines and fields also, however, bring with them a set of methods of inquiry or approaches to scholarship. In many fields these are well-defined practices, which are prescribed by the discipline and sanctioned as legitimate ways to provide evidence for conclusions or arguments forwarded in the research.
Inside many of the social sciences, one finds a thriving mix of methods. It is common in such fields to have formal courses in “research methods,” to introduce the fledging student to the alternative approaches at knowledge acquisition. In some fields, the student would be exposed to collecting data from existing administrative or archival sources, to participant observation or ethnographic techniques, to forms of unstructured interviewing of persons, to randomized experiments with human subjects, to quantitative survey research, to statistical analysis of existing quantitative data. Fields that use multiple methods sometimes sort themselves into internal tribes, each of which touts the superiority of one method to discover truth and disparages the others.
As a provost, one is struck by the use of similar methods across disciplines studying very different phenomena. For example, behavioral economists use experimental laboratory methods with most of the features of psychologists’ methods in their laboratories. Organizational analysts sometimes use intensive observation and case study techniques that are common to anthropologists. Scholars in cultural studies in foreign language departments and English departments use techniques common to those in sociology departments. Some psychologists use the FMRI measurement in ways not dissimilar to those of neurologists. Some faculty in linguistics use research approaches similar to those in computer science. Statistical analysis of quantitative data is common in not only statistics but in political science, economics, sociology, psychology, public policy, business, etc.
This commonality of methods across fields is interesting at the university level, for three reasons. First, it produces a set of courses that cover similar content across different programs. For example, there are statistics courses spread throughout scores of departments. They differ in the mix of theory and application. They also tend to utilize data examples from the fields in which they are taught (e.g., in environmental studies, data on fish; in economics, data on businesses). Similar, the design of experiments or surveys can be taught in a variety of programs. Textual analysis is spread throughout many departments.
Second, it produces a set of faculty who share interests in advancing research methods but find themselves in different units. Good things can happen when such faculty get together. For example, at Georgetown there is a group of quantitative scientists from throughout the university, GQUADS, that convenes regularly to discuss new methods in statistics, computational science, and related issues. When methods-oriented faculty team-teach across fields, wonderful things can happen for students and faculty.
The third reason is more of a thought experiment – what would happen if a university organized itself about units that shared research methods and not in units that shared a set of substantive foci? If we organized that way, would we cumulate more insight into, for example, how economics and psychology might combine to explain human behavior? Would we end up in even more conflict between theory and application with our current disciplines? Would methodological development themselves advance at a greater pace; would we develop better measurement and observational tools faster?