Schaffer’s Elucidating Social Science
Concepts: Notes of a Conceptualist in the Field
University of Notre Dame
[**Original document page numbers preserved in brackets for citation purposes.**]
[Start Page 47]
Gary Goertz is Professor at the University of Notre Dame’s Kroc Institute for International Peace Studies. He can be reached at email@example.com. He thanks Pamina Firchow and Jim Mahoney for comments
on an earlier draft of this essay.
I am quite sympathetic to many aspects of the anthropological and ethnographic approach defended by Fred Schaffer in his
Elucidating Social Science Concepts: An Interpretivist Guide.1 Much of my methods work is motivated and informed by what I call “methodological anthropology,” which I define as the examination of the practices of social scientists
and philosophers regarding concept formation and construction. My original interest in concepts started from the observation that social scientists and philosophers spend a lot of energy and effort defining, disputing, and thinking about concepts.
At the same time, concept methodology was completely absent from methods, statistics, and research design textbooks. These have chapters on measurement but nothing about concepts.
Schaffer’s chapter 2 has the subtitle “how people under- stand a concept.” He focuses on “everyday people” as subjects of his methodology; in contrast, I have focused on social scientists and philosophers as subjects. This is a major difference between
our interests. For example, in addition to understanding social science practice, I give advice to my subjects (social scientists) on how to do things better. Fred is not telling everyday people how to do concepts better.
At the same time, I completely agree with almost all the “lessons” he has for elucidating concepts and would apply these lessons to my target groups. For example, his postulate to “investigate ordinary use” is critical because people, including social
scientists, do all kinds of odd things with concepts. This analysis is critical in producing better social science concepts. “Compare the use of the same word in different language games” means, for example, that one should look at how political theorists
or philosophers work on a concept, say democracy, versus quantitative social scientists. “Examine opposites and negations” is absolutely essential to distinguishing between what I call the positive and negative poles. Terminology is critical and signals
all sorts of issues. For example, social scientists cannot agree on what to call “not democracy” and this has varied over time, with popular options like monarchy (19th century), dictatorship,2 authoritarian, totalitarian, etc. So I completely
endorse his recommendation to “follow the clouds of etymology.”
In short, much of Schaffer’s ethnographic advice works very well in understanding how social scientists develop and use concepts.
Much of chapter 3 resonated with me as well. The analysis of historical developments and genealogy is critical to understanding social science concepts. For example, one cannot understand the polity or Freedom House datasets without an understanding of
their history. Many things that seem odd or curious about these datasets arise from the fact that they were not meant to capture concepts of democracy at the beginning! They have evolved and been adapted over time, but still retain traces of their
origins. Freedom House was about the concept of liberty—social, economic, and political. It eventually morphed into a democracy dataset. The polity concept of anocracy, which is now used to refer to competitive-authoritarian regimes, originated in
the concept of anarchy.3
In short, much of chapters 2 and 3 is directly relevant to thinking about how social scientists and philosophers develop, debate, and use concepts and is good advice to all those interested in concept methodology.
Chapter 4 is about “elucidating power.” A good example of this practice is the literature on gender and politics because one of the first moves of a gender scholar is to deconstruct and analyze the gender bias of traditional concepts. For ex- ample, it
is fascinating to see how the World Bank conceptualizes “indigenous people,” a concept that is very politicized and that has large real-life implications for these peoples. The discussion in the literature is reminiscent of Foucault talking about
an institutionalization of “insanity-madness” (folie in the 18th century). To apply this practice more broadly, theories involving democracy and democratization would probably require some significant changes if women’s voting were included
in the major concepts and datasets, where women are quite notable by their absence.4 The same issue applies to minorities, e.g., African-Americans, in democracy concepts and datasets: for example, the USA in 1920 is coded a maximal democracy
Schaffer contrasts “positivist reconstruction” with “interpretivist elucidation.” What “positivism” means is a hotly
contested concept itself. I see Schaffer as engaged in positivistic empirical research. His long discussion about what “family” means in different cultures is meant to be an accurate representation of these differences.
So Schaffer and I agree on many points dealing with the semantics of concepts. We part ways on the role of concepts in describing the world and their use in explaining how the world works. His book stresses the value of understanding how people use concepts
and what they mean by various concepts. But concepts have an instrumental value as well. We can ask how well they describe the world and if they are useful in explaining the world.
The biggest difference between us lies in the role of explanation, causation, and causal hypotheses. I am interested in concepts because they are essential in describing the world, but also very much because they are core to explaining the world. This
of course makes me a “positivist,” but some interpretivists also want to explain the world.5 Causal explanation is the goal of my main target group, social scientists, and is my goal in my substantive work. I work from the philosophy that
high quality concepts are critical to high quality social science. Bad concepts, e.g., terrorism, lead to bad research. A very big chunk of my applied work over the years involves very serious conceptual analysis. Before I can explain international
peace I needed to think very hard about the concept of peace.6 To analyze how people, say Wolofs in Gambia, differ in their concept of democracy, is interesting to me if that somehow “matters.” Mattering is that it influences behavior or
is influenced by something. These are causal questions. I am interested in differences in meaning and concepts, but only those that somehow matter in causal explanations, hypotheses and theories.
The title of Schaffer’s book indicates that it is about “social science” concepts. To advance social science we need to know how interpretivist methodologies of concepts help or relate to causal explanations and hypotheses.
Foucault, Michel. 1972. Histoire de la folie à l’age classique. Paris: Gallimard.
Goertz, Gary, Paul F. Diehl, and Alexandru Balas. 2016. The Puzzle of Peace: the
Evolution of Peace in the International System. Oxford: Oxford University Press.
Gurr, Ted Robert. 1974. “Persistence and Change in Political Systems 1800-1971.” American
Political Science Review vol. 68, no. 4: 1482–1504.
Paxton, Pamela. 2000. “Women in the Measurement of Democracy: Problems of Operationalization.”
Studies in Comparative International
Development vol. 35, no. 3: 92–111.
Paxton, Pamela, Kenneth A. Bollen, Deborah M. Lee, HyoJoung Kim. 2003. “Problems of a Half-Century of Suffrage: New Data and a Comparative Analysis.” Studies
in Comparative International Development vol. 38, no. 1: 93–117.
Przeworski, Adam, Michael E. Alvarez, Jose Antonio Cheibub and Fernando Limongi. 2000. Democracy and
Development: Political Institutions and Well-Being in the World, 1950-1990. Cambridge: Cambridge University Press.
Schaffer, Frederic. 2016. Elucidating Social Science Concepts: An Interpretivist
Guide. London: Routledge.
Wedeen, Lisa. 2002. “Conceptualizing Culture: Possibilities for Political Science.” American
Political Science Review vol. 96, no. 4: 13–28.
1 Schaffer 2016.
2 Przeworski et al. 2000.
3 Gurr 1974.
4 See Paxton (2000) and Paxton et al. (2003) for nice discussions.
5 e.g., Wedeen 2002; Foucault 1972.
6 Goertz et al. 2016. There are many war-conflict datasets, we offer the first peace dataset.