Academic journal article Public Administration Quarterly

Coming to Terms with the Complementarity of Agent and Structure

Academic journal article Public Administration Quarterly

Coming to Terms with the Complementarity of Agent and Structure

Article excerpt


This article questions how to conceptualize the relation between different ontologies in the social sciences. Giddens' Structuration Theory is analyzed as an example of an unifying theory linking 'agent' and 'structure' ontologies. The concept of interaction is used to pinpoint Giddens' notion of the complementarity of these two. The resulting abstract unification does, however, not result in a unified social analysis. Rather it implies that we need to 'bracket' the one or the other approach in empirical observation. This raises the question: What is complementarity? In this article it is argued the notion of complementarity as developed in quantum theory (Q-complementarity) applies to the case at hand. Q-complementarity does not refer to the possible simultaneous attribution of characteristics, but establishes the validity of both attributes for describing reality depending on the circumstances of observation.


In the field of public administration several dichotomies are currently being debated. Notably, the dichotomy between politics and administration is a core topic. Many others may spring to mind, such as public and private, centralization and decentralization. These dichotomies all contribute substantially to our ideas about public administration and are therefore much debated (Svara, 1999, Rutgers, 2001). They indicate differences and similarities that assist in understanding and ordering reality, i.e. public administration. They provide the field with its ontology.

Although less conspicuous, there is also another important dichotomy, that of agent and structure. Students of public administration and public policy rarely, if ever refer, to this fundamental schism in understanding administrative reality, yet they all take a stance (cf. Denhardt, 1990). This dichotomy relates to allencompassing ideas on the nature of social reality, not just the delineation of public administration. We call it a formal dichotomy. In that sense it signifies ways of understanding or explaining public administration. If we look at public administration, do we look at agents, in the extreme case resulting in voluntarism, or do we look at structures, resulting in determinism? The two ideal types are regarded as competitive in the debates on the philosophy of the social sciences. There is even an abundance of phrases around to describe the dichotomy such as: individualistic versus holistic, or actor versus system. We will frame it here in terms of the schism between agent and structure as the starting point for conceptualizing social phenomena. The cleft between the two is deep because of the long historical strands (Burrel & Morgan, 1994; Astley & van de Ven, 1983; Fielding ed., 1988; Denhardt, 1990; Hollis, 1997; Barnes, 1995).

The argument in this article runs as follows. We will analyze one of the attempts to integrate or synthesize the two perspectives more closely in Giddens' Theory of Structuration.(1) Like Wamsley & Wolff (1996, 30) we regard the Theory of Structuration as "... valuable to public administration on both epistemological and ontological grounds because he rejects both positivism and subjectivism."

We will argue that Giddens' solution can be framed in terms of 'interactions'. By working out what interaction means, the complementariness of structure and agent in Giddens' theory can be highlighted as a non-linear and non-causal relation happening during the complex structuration processes over time. However, Giddens' theoretical and abstract 'unification' does not seem to apply to empirical research. Thus, Giddens presented us with the paradoxical situation that the complementarity of agency and structure at the ontological and epistemological level does not imply that (methodologically, i.e. on the observational level) social phenomena can, as such, be empirically studied simultaneously.

In order to conciliate or integrate the opposing ontological substantial dichotomies as Giddens did, it is increasingly popular to write about 'complementarity' (Rutgers, 2000). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.