Content area
Full Text
Theory and Formalization: Some Reflections on Experience*
I describe in this paper some of my efforts in developing formal theories of social processes. These include work on models of occupational mobility, on models to describe the emergence of expectations out of performance evaluations, and on the graph theory formulation of the Status Characteristics theory. Not all models have been equally significant in developing theory. However, the graph theory formulation has played a central role in the growth of the Expectation States program. It has been involved in the generalization of theories, the integration of theories, and in the construction of highly sensitive tests of theories that would be impossible without the inferential capacities of formalization.
Throughout my career, my primary research goal has been to develop abstract and general theories of different types of interpersonal processes and in particular developing theories that evolve, proliferate and grow. It is in this context that I have come to view formalization as being instrumental to achieving this goal.
There was very little in my graduate training in sociology at Harvard in the beginning of the 1950s that prepared me for work on formal theory. While there was much talk about the future prospects of formalization, and Bush and Moesteller were beginning their collaboration on learning models, few individuals at Harvard during that time were actually trying to develop formal theory in sociology.
DARTMOUTH AND MOBILITY MODELS
My first real exposure to the idea of formal theory and formalization came with my appointment at Dartmouth in 1954. 1 arrived at Dartmouth at the same time as John Kemeny. John Kemeny, an extremely talented individual, was a man with a mission. His mission was to develop new mathematics-mathematics based on the specific substantive needs and the specific substantive problems found in the social sciences. To realize that objective, he was in the process of assembling a group of bright young mathematicians including J. Laurie Snell, Gerald L. Thompson, and Robert Z. Norman to work with him. I was very impressed with this group of researchers and with the work they were doing, and I soon became involved with these activities.
In 1956, in collaboration with J. Laurie Snell, I worked on applying finite Markov Chain theory to the...