qualitative researchers probably have been the scholars most persistently concerned with
the definition (as opposed to quantitative operationalization) of the field’s key concepts.
We see this, for example, by considering the range of articles by qualitative
methodologists analyzing particular concepts: corporatism (Collier 1995), democracy
(Schmitter and Karl 1991; Collier and Levitsky 1997), ideology (Gerring 1997),
institutionalization (Levitsky 1998), peasant (Kurtz 2000), political culture (Gerring
2003), revolution (Kotowski 1984), and structure (Sewell 1992). Moreover, through
typological analysis (Elman 2005; George and Bennett 2005), qualitative researchers
routinely develop conceptual distinctions that play a major role in driving theories of
comparative politics. These typologies include, to name only a few, Linz and Stephan’s
(1996) specification of types of regimes (e.g., democratic, authoritarian, totalitarian),
Esping-Anderson’s (1990) types of welfare states (e.g., Christian, liberal, social-
democratic), and Evan’s (1995) typology of states (e.g., predatory, developmental,
intermediary).
Methodological writings have uncovered some of the reasons why there is an
affinity between qualitative research and conceptual development. One answer is the fact
that qualitative researchers match background understandings of preconceived concepts
with fine-grained evidence from their cases. This process of matching, which often
proceeds through many rounds of iteration, stimulates new conceptual understandings.
For example, O’Donnell’s (1973; 1979; 1994) creation of various new concepts over the
years – bureaucratic-authoritarianism, lo popular, brown areas – has always occurred in
conjunction with the close qualitative examination of actual cases. To capture a changing
political reality in Latin America, O’Donnell has found it essential to invent new
9
Convention