As scholarly interest in the concept of identity continues to grow,
social identities are proving to be crucially important for understanding
contemporary life. Despite—or perhaps because of—the sprawl of
different treatments of identity in the social sciences, the concept has
remained too analytically loose to be as useful a tool as the
literature's early promise had suggested. We propose to solve this
longstanding problem by developing the analytical rigor and methodological
imagination that will make identity a more useful variable for the social
sciences. This article offers more precision by defining collective
identity as a social category that varies along two
dimensions—content and contestation. Content describes the meaning
of a collective identity. The content of social identities may take the
form of four non-mutually-exclusive types: constitutive norms; social
purposes; relational comparisons with other social categories; and
cognitive models. Contestation refers to the degree of agreement within a
group over the content of the shared category. Our conceptualization thus
enables collective identities to be compared according to the agreement
and disagreement about their meanings by the members of the group. The
final section of the article looks at the methodology of identity
scholarship. Addressing the wide array of methodological options on
identity—including discourse analysis, surveys, and content
analysis, as well as promising newer methods like experiments, agent-based
modeling, and cognitive mapping—we hope to provide the kind of brush
clearing that will enable the field to move forward methodologically as
well.Rawi Abdelal is Associate Professor,
Harvard Business School (rabdelal@hbs.edu). Yoshiko M. Herrera is
Associate Professor, Government Department, Harvard University
(herrera@fas.harvard.edu). Alastair Iain Johnston is Professor, Government
Department, Harvard University (johnston@fas.harvard.edu). Rose McDermott
is Associate Professor, Political Science Department, University of
California at Santa Barbara (rmcdermott@polsci.ucsb.edu). Research for the
paper was made possible by the generous support of the Weatherhead
Initiative of the Weatherhead Center for International Affairs at Harvard
University. We are grateful to those who commented on earlier versions of
this paper: Peter Burke, Lars-Erik Cederman, Jeff Checkel,Michael Dawson, James Fearon, David Frank, Erin
Jenne, Michael Jones-Correa, Cynthia Kaplan, Peter Katzenstein, Herb
Kelman, Paul Kowert, David Laitin, Daniel Posner, Paul Sniderman, Werner
Sollors, Jeff Strabone, Philip Stone, Ronald Suny, Charles Tilly, Mary
Waters, and three anonymous reviewers. We would also like to thank
participants of the 2004 Identity as a Variable conference,
including Henry Brady, Kanchan Chandra, Jack Citrin, Neta Crawford,
Jennifer Hochschild, Jacques Hymans, Ted Hopf, Cynthia Kaplan, Ulrich
Krotz, Taeku Lee, Will Lowe, Jason Lyall, Kimberly Neuendorf, Roger
Petersen, Kevin Quinn, David Rousseau, Rogers Smith, Donald Sylvan, Kim
Williams, and Michael Young, for comments on this version.