Paradoxes of Complementarity in Public Administration: Part 1 – Can There Be a Theory of Everything?

The views expressed are those of the author and do not necessarily reflect the views of ASPA as an organization.

By Erik Devereux
November 8, 2024

This is the first column in a series on paradoxes across theories of public administration. Actually, these paradoxes are the same across theories of all types of organizations but I will respect boundaries and focus on public administration. Despite the focus on theory, this series is directly relevant to practitioners. As the reader will see, the tension between theory and practice lies at the heart of these paradoxes. PA research always seeks to be relevant but can struggle when these paradoxes are immediately and vividly evident to practitioners.

I recently began teaching core courses in public administration for the first time in my career, having previously taught mostly policy analysis. In doing so, I had to embrace a literature that, quite frankly, I’d been running away from since beginning my doctoral studies in a previous century. It turns out that I was running in an ellipse, not a straight line, and, thus, crashed into my backside at full force. I can’t explain my prior reticence to delve into PA research other than to allude to a younger self who was schooled in somewhat superficial political science theories of policymaking. As with most social phenomena, once you dive below the surface PA exhibits complexities that took some life experience (25 years as a practitioner) to appreciate.

PA in the United States over the period from the late 1880s to the present has developed theories that occupy three distinct levels of analysis:

• The constitutional/institutional level that focuses on the separation of politics and administration, federalism and inter-branch and inter-agency relations in efforts to explain overall government behavior.
• The organizational level that focuses on the internal structures of individual agencies and how those structures might influence agency behavior.
• The personnel level that focuses on what shapes the behavior of people in public agencies as they go about their daily work.

Various versions of public sector reform exclusively focus on one of these levels. Reformers might propose moving an agency from the jurisdiction of one congressional committee to another, relocating the agency from one department to another or shifting the agency’s function from the federal government to the states. At the next level down, reformers might propose an internal reorganization of a department or a specific agency. At the personnel level, reformers might try to inject new incentives to improve individual performance.

If this is beginning to sound familiar, I am paralleling Bob Behn’s 1995 PAR article, “The Big Questions of Public Management.” According to Bob, those questions are about the micromanagement of one part of the government by another part (institutional level), measuring agency performance (organizational level), and motivating people in the agency (personnel level). Behn’s three Big Questions thus are aimed at three completely distinct levels of problems.

The paradoxes begin to pile up quickly when you try to get theories and observations at one level to fit with those at the other levels. Very often, they just don’t get along. In fact, they often contradict each other. These contradictions emerge in greatest clarity at the personnel level which often does not produce observations that corroborate theories at the institutional and organizational levels. Later in this series I will delve into some examples of this in detail.

I’m now going to stress out any stray physicists that happen across this article and use some physics as a metaphor for the situation in PA. No worries—I took a lot of physics at MIT and continue read about physics for fun and profit. I promise to get the basic details correct.

Complementarity is a term in physics that pertains to an object simultaneously having incommensurable properties depending on how you measure those properties. At the quantum scale, objects simultaneously are particles and waves. If you go looking for particles, you will see particles; go looking for waves and you will see waves. One frontier in physics is experimentation to find out how large an object can be and still exhibit this complementarity. Some very large molecules have demonstrated complementarity (and, in theory, so can a car or a house).

Another version of complementarity is represented in the image accompanying this article of individual atoms on the surface of a crystal. At the macro scale, the crystal (and all matter) has measurable properties but connecting those properties to the properties of the individual atoms in the material remains a challenge. Materials science is content working at one level, quantum physics at another, and the two disciplines try to connect as best they can. There remain many incommensurable findings that get filed under complementarity.

Complementarity in PA refers to the difficulties of reconciling the institutional, organizational and personnel levels of theory and observation. Rather than seeking a theory of everything, this series explores the possibilities in asserting that PA must accept complementarity for what it is—the reality that incommensurable approaches simply cannot be reconciled but must be accommodated together. Again, where this gets interesting is at the personnel level where we see if our theories of public administration fit with the beliefs and behaviors of public administrators.

---

Author: Erik Devereux is Teaching Associate Professor in the Department of Public Policy, Management, and Analytics at the University of Illinois-Chicago. He has a B.S. from the Massachusetts Institute of Technology (Political Science, 1985) and a Ph.D. from the University of Texas at Austin (Government, 1993). He is the author of Methods of Policy Analysis: Creating, Deploying, and Assessing Theories of Change (available for free here). Email: [email protected]. More content is availab

(No Ratings Yet)
Loading...