Rethinking Representation Author(s): Jane Mansbridge Source: The American Political Science Review, Vol. 97, No. 4 (Nov., 2003), pp. 515-528 Published by: American Political Science Association Stable URL: http://www.jstor.org/stable/3593021 . Accessed: 16/08/2013 04:49 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. American Political Science Association is collaborating with JSTOR to digitize, preserve and extend access to The American Political Science Review. http://www.jstor.org This content downloaded from 88.119.17.198 on Fri, 16 Aug 2013 04:49:00 AM All use subject to JSTOR Terms and Conditions American Political Science Review Vol. 97, No. 4 November 2003 Rethinking Representation JANE MANSBRIDGE Harvard University long withthe traditional"promissory" form of representation,empirical political scientists have recently analyzed several new forms, called here "anticipatory,""gyroscopic," and "surrogate" representation. None of these more recently recognized forms meets the criteria for democratic accountability developed for promissory representation, yet each generates a set of normative criteria by which it can be judged. These criteria are systemic, in contrast to the dyadic criteria appropriate for promissory representation. They are deliberative rather than aggregative. They are plural rather than singular. ver the past two decades empirical political ing these more recently identified forms of represen- scientists have developed increasingly sophisti- tation are systemic, in contrast to the dyadic criteria cated descriptions of how American legislators appropriate for promissory representation. The crite- relate to their constituents. Yet although the empirical ria are almost all deliberative rather than aggregative. work has often been motivated by normative convic- And, in keeping with the conclusion that there is more tions that one way of relating is better than another, than one way to be represented legitimately in a democ- the normative theory of what constitutes "good" rep- racy, the criteria are plural rather than singular. resentation has not kept pace with current empirical The forms of representation identified here do not findings. This paper seeks to narrow the gap. map well onto the traditional dichotomy of "mandate" The traditional model of representation focused on and "trustee." Both mandate and trustee forms can the idea that during campaigns representatives made appear as versions of promissory representation (or, promises to constituents, which they then kept or failed alternatively, the trustee concept can figure as a subset to keep. I call this promissory representation. In addi- of gyroscopic representation), but the new concepts of tion, empirical work in the last 20 years has identified representation implied by recent empirical work do not at least three other forms of representation, which I have an obvious relation to the earlier dichotomy. call "anticipatory,""gyroscopic," and "surrogate" rep- In practice, representative behavior will often mix resentation. Anticipatory representation flows directly several of these forms. One cannot always tell by look- from the idea of retrospective voting: Representatives ing at a specific behavior what dynamics lie behind it. focus on what they think their constituents will approve Yet analyzing each form separately makes it possible at the next election, not on what they promised to do to identify the underlying power relation in each form, at the last election. In gyroscopic representation, the the role of deliberation in each, and the normative cri- representative looks within, as a basis for action, to teria appropriate to each. These normative criteria are conceptions of interest, "common sense," and princi- goals toward which to strive ("regulative ideals"), not ples derived in part from the representative's own back- standards that can be fully met. Conceiving of demo- ground. Surrogate representation occurs when legisla- cratic legitimacy as a spectrum and not a dichotomy, tors represent constituents outside their own districts. one might say that the closer a system of representation These are all legitimate forms of representation. comes to meeting the normative criteria for democratic None, however, meets the criteria for democratic ac- aggregation and deliberation, the more that system is countability developed for promissory representation. normatively legitimate. I argue that the appropriate normative criteria for judg- Addressing the norms appropriate to a system of representation assumes that representation is, and is intended to more than a de- Jane is Adams John School of normatively be, something Mansbridge Professor, E Kennedy fective substitute for direct Constituents Government, Harvard University, 79 JFK Street, Cambridge, MA democracy.1 02138 ([email protected]). choose representatives not only to think more care- This paper has evolved over time. Most recently I am grateful for fully than they about ends and means but also to nego- the suggestions of Douglas Arnold, David Brady, Martha Minow, tiate more perceptively and fight more skillfully than Mark Moore, Dennis Thompson, and participants in seminars at constituents have either the time or the inclination to Center for Advanced in the Behavioral Princeton Study Sciences, do. The difference between and direct University, the University of California Los Angeles, Stanford Uni- representation versity, and the University of Toronto. For excellent suggestions democracy creates a need for norms designed partic- on earlier versions I thank William Bianco, Carol Swain, Melissa ularly for democratic representation. Yet democratic Williams, Iris Marion Young, and participants in seminars at the Insti- comes in different with norms tute of representation forms, Governmental Studies at Berkeley, the Ohio State University, to each. Nuffield College Oxford, Indiana University, Princeton University, appropriate the University of California San Diego, Harvard University, and Northwestern University. I particularly thank Benjamin Page for his close reading and incisive comments at an early stage, and the insight- 1 Although deliberative forms of direct democracy can be effective ful reviewers for this journal. This paper, begun with support from methods of democratic governance in many circumstances, represen- the Institute for Policy Research at Northwestern University, was tative forms of democracy have their own uses, functioning not just completed while the author was a Fellow at the Center for Advanced as "transmission belts" for constituent opinion (Schwartz 1988; see Study in the Behavioral Sciences. I am grateful for financial support also Achen 1978, 476, Hibbings and Theiss-Morse 2002, Manin 1997, provided by National Science Foundation Grant SBR-9601236. and Pitkin [1967] 1972). 515 This content downloaded from 88.119.17.198 on Fri, 16 Aug 2013 04:49:00 AM All use subject to JSTOR Terms and Conditions Rethinking Representation November 2003 PROMISSORYREPRESENTATION with the voter as principal, statically conceived, trying to exercise power over the representative as agent.3 the traditional fol- Promissory representation, model, Promissory representation works normatively thro- lows the classic format. The principal-agent prob- ugh the explicit and implicit promises that the elected lem for the Bristol or is one of principal (in Ohio) representative makes to the electorate. It works pru- some control over the London or keeping agent (in dentially through the sanction the voter exercises at The in does not differ Washington). problem politics the next election (Time 3). That sanction is a reward from the of economic greatly problem keeping any or punishment for acting or failing to act according to to the desires of the Eco- agent responsive principal. the promise made at the previous election (Time 1). nomic and have focused on history theory recently Both normatively and prudentially, the electoral au- the of trade when there was no problem long-distance dit at Time 3 focuses on whether or not the promises infrastructure to enforce contractual ar- governmental at Time 1 were kept. George Bush thus angered his In the Mediterranean in the fourteenth rangements. supporters deeply by breaking an explicit campaign this situation necessitated either ties century, kinship promise ("Read my lips: No new taxes").4 or above-market rates to ensure that payment ships Promissory representation has the advantage that, at loaded with the value of thousands of workers surplus least in its more mandated versions, it reflects in a rel- returned with the received in trade actually goods (see atively unmediated manner the will (although not nec- Greif When control in a seabound or 1993). (as ship) essarily the considered will) of the citizenry. It comes information in relations with an is (as expert) asym- closer than any other model to an ideal in which the the for the is to make sure metric, problem principal simple imprint of the voter's will is transmitted through that the the the agent (the captain, lawyer, accountant) institutions