Introduction to the symposium ‘Conceptual analysis in political science and beyond’

Reviewing literature and analyzing concepts are two research ‘tasks’ conducted by social scientists which are grossly undervalued. This kind of work is too often narrowly conceived as a prelude to serious research, not as a product of research in and of itself. Things are beginning to change for literature reviews under the impulse of what can be termed the evidence-based movement and in particular the leadership of organizations such as the Cochrane Collaboration, the Campbell Collaboration and the Evidence for Policy and Practice Information and Co-ordinating Centre (EPPI-Centre) (Davies, Nutley & Smith, 1999; Hansen & Rieper, 2009; Oakley, 2002; Pawson, 2002).

The perspective in the case of conceptual analysis, however, does not appear to be so bright. Admittedly, some extremely valuable work on the issue has been published recently by comparative politics scholars such as David Collier and John Gerring (2009), Gary Goertz (2006) and Andreas Schedler (2011). In addition, organizations such as the Committee on Concepts and Methods (C&M) of the International Political Science Association (IPSA) and the Research Committee on Conceptual and Terminological Analysis of the International Sociological Association (ISA) promote scholarly work on concepts. Yet, as a general rule, concepts and their analysis have received scant attention from scholars as objects worth studying and teaching in their own right, especially in comparison to data collection and analysis techniques (Gerring, 1999; Goertz, 2006; Schedler, 2011). Furthermore, most of the work on concepts is confined to political science and sociology, to the subfield of comparative politics in particular. Concepts and conceptual analysis seem to suffer somewhat from ‘benign neglect’. While all social scientists can probably recognize a concept when they see one, they do not always know how to assess its value or improve its structure, let alone develop new conceptual alternatives. When scholars venture into those tasks, they generally proceed informally, that is, without following clear rules or principles, thereby raising serious concerns in terms of construct validity.

As Giovanni Sartori (2009[1984]) wrote, more than 30 years ago, concepts are the basic units of thought and are, therefore, fundamental to any disciplined endeavour aimed at ‘discovering’ the nature of ‘reality’. Indeed, in political science as in other social sciences, academic tasks such as theorization, empirical research, the diffusion of scholarly work and teaching would simply be impossible in the absence of concepts. Since concepts are the ‘building blocks’ of theories and research, this benign neglect poses obvious problems for the validity and utility of our scientific work. Indeed, to put it bluntly, the social science ‘building’ incurs serious risk of collapse if we don’t first examine the quality of the material we use. This is admittedly a worst-case scenario, but in any event, the benign neglect of concepts stifles scientific progress.

The time has come for social scientists to pay more attention to the concepts they use. This symposium, entitled ‘Conceptual Analysis in Political Science and Beyond’, brings concepts and their analysis to the center stage of scientific work. The symposium’s objectives are manifold: to explain the nature of concepts, to outline some elements of conceptual analysis, to present a few systematic applications to important political concepts and, last but not least, to reflect on the importance of concepts and their use in the social sciences in general and in political science in particular. The ‘and Beyond’ part of the title is a deliberate inclusion. Indeed, whereas the perspective, literature and concepts of this symposium are grounded mainly in political science, social scientists in general could clearly benefit from the work presented here. My hope is that, regardless of their discipline, the readers of this symposium will become more conscious and self-reflective in the way they use concepts. I am especially pleased that three of the articles for this symposium are written in French, since this could certainly contribute to a more fertile dialogue and exchange of ideas between different traditions of conceptual analysis.

The symposium contains seven engaging articles. ¹ The first two tackle the issue of concepts and their analysis from a theoretical, methodological and sociological perspective. Steve Jacob and I outline the tridimensional nature of concepts (meaning, term and empirical referent) and offer a few guidelines taken from a ‘classical’ approach to analyze them. Good concepts are clear and precise, and their definition is parsimonious in the sense that each attribute should be necessary and all attributes should be collectively sufficient. Using a sample of 24 social-science methodology books written in French, our article also sets out to test the hypothesis according to which concepts suffer from benign neglect. We conclude that, while most books offer an explicit treatment of concepts and their analysis, few of them do so systematically.

Gary Goertz and James Mahoney argue in their article that qualitative and quantitative scholars display a different ontological and epistemological orientation toward concepts and measurement. Ontologically, qualitative scholars use a semantic approach to concept formation in which basic attributes and indicators are derived deductively, whereas quantitative scholars form concepts inductively from a set of indicators that are causally related to them. Epistemologically, qualitative researchers have more confidence in their description and coding of extreme cases (i.e. those that are close to ideal-types), whereas quantitative researchers are said to assume that extreme cases are more subject to measurement error. Their conclusion is that the long-standing differences between scholars from each orientation have hindered fruitful communication between them.

The next four articles present insightful applications to concepts taken from the fields of political science and law. Maude Benoit conducts a thorough analysis and operationalization of the concept of meso-corporatism based on the three-level framework proposed by Goertz (2006). She identifies three secondary-level dimensions that are individually necessary and collectively sufficient to define meso-corporatism: (1) monopolistic or oligopolistic organization of interest representation; (2) presence of an interventionist state favouring collaborative management of a policy sector; (3) actual participation of the monopolistic/oligopolistic organization in the management or control of a policy sector. Once operationalized, this conceptualization is applied to the case of Quebec’s agricultural sector with which it displays a high level of congruence.

Louis Bélanger and Kim Fontaine-Skronski focus on legalization, a key concept for our understanding of contemporary international relations. Drawing on a deconstruction of the work of prominent theorists and a review of the recent empirical literature on legalization, their article proposes a compelling analysis of the structure of the concept, comprising three basic dimensions, namely: obligation, precision and delegation. They formulate three main conclusions. First, ‘if a better theoretical argument cannot be made linking precision to legalization ontologically’, this dimension should be dropped. If not, it should be limited to meaning ‘precision of obligations’ and be considered as the least important dimension. Second, because legalization theorists have failed to properly theorize the difference between type A delegation (dispute resolution) and type B (rule-making and implementation), only the second one should be kept as a constitutive dimension of legalization. Third, the dimension obligation should be given a quasi-necessary status in the construction of the concept of legalization. These conclusions lead the authors to propose a modified concept structure of international legalization and a typology that is simpler and more theoretically coherent.

Ryan Baird’s article makes a conceptual contribution to the literature on the determinants of economic outcomes for developing states. According to this body of literature, democracy provides superior institutions, such as the rule of law and an efficient bureaucracy, which, in turn, lead to economic development. Baird convincingly argues that, for many political scientists and economists, there is confusion between democracy and governance. There is an urgent need to better distinguish between the two concepts in order to understand economic development. He therefore proposes a new concept labelled governance infrastructure, which he defines as ‘the core domestic institutions that facilitate government competency and economic efficiency’, and which he clearly differentiates from democracy.

Julie McCann and Martin Thiboutot propose a systematic analysis of a concept that is relevant to the fields of law, political science and public administration, namely: public–private partnerships (PPP). They offer an overview of the process of developing normative legal knowledge, a process that goes through the ‘qualification’ of a concept and its interpretation. They argue that when a rule is based on a concept, the normative content of the concept is most important. Thus, the normative content in a legal approach is equivalent to the descriptive content in a social-science approach. The first is aimed at a legal decision, a judgment; the second at a descriptive statement, an empirical observation. The authors further argue that when the legal doctrine is insufficiently clear to define a given concept – which is the case of the PPP concept – it is necessary to draw from other disciplines. Therefore, they propose a research design aimed at characterizing the literature in various disciplines on the explicit definitions provided by the concept. In view of their preliminary results, they conclude that the PPP concept is empirically vague and needs a deeper multidisciplinary analysis.

In the conclusion to this symposium, Louis M Imbeau makes the observation that social scientists adopt various strategic approaches to conceptual analysis. He argues that these strategies depend on the ‘intelligibility scheme’ or explanatory logic that one chooses. In other words, conceptualization and explanation are intimately linked. His argument is based on the work of Jean-Michel Berthelot (1990), who has developed a six-fold typology of social explanation: causal, actantial, structural, hermeneutic, functionalist and dialectic. Imbeau develops his own typology based on two dimensions: (1) the joint/disjoint character of the observation of the explanandum and the explanans; and (2) the focus on actor/context. The contributions of this symposium which are applied in nature are then situated in the typology’s quadrants. He concludes that ‘intelligibility schemes’ matter, and makes a plea in favour of theoretical, methodological and explanatory pluralism.

Compared to ‘standard’ social-science practice, the contributions presented in this symposium are clearly atypical with respect to the level of self-consciousness and thoroughness they bring to concepts and their analysis. It is to be hoped that the current situation will evolve, and that the benign neglect of concepts by social scientists will come to an end. This expectation does not imply that every scientist should pursue conceptual work, however. Science is a collective enterprise and will (legitimately) remain so. My greatest hope is that social scientists will become increasingly conscious of the concepts they use.