Department of Political Science

ERIN CASSESE

Department of Political Science
N735 Social and Behavioral Sciences Bldg.
Stony Brook University
Stony Brook, New York 11794-4392
ecassese@ic.sunysb.edu

DISSERTATION PROSPECTUS

CULTURE WARS AS IDENTITY POLITICS

Over the past two decades, religion has emerged as a significant cleavage in American political life. Religious adherence is a strong predictor of partisanship, vote choice, and political participation, and predicts a host of social policy preferences – such as reproductive rights and gay rights – in addition to intolerance towards a variety of social groups.  Differences in political preferences across religious groups have become so pronounced that religious voters are often perceived as a cohesive electoral bloc.  As a result, many argue religious differences form the basis of “culture wars” or a “values divide” in the United States (Hunter, 1991; White, 2002; Frank, 2004).  In these culture wars, debate over issues of morality and social policy has displaced the traditional economic disagreements driving policy conflict.  In addition, ideological labels have been redefined in moral terms, due to the growing salience of social policy issues and the close connection between religiosity and political conservatism (Miller and Hoffman, 1999). 
            Fiorina (2006) contests this notion of a nation embroiled in cultural conflict and focused exclusively on moral policy issues.  He argues the electorate is largely moderate and deeply ambivalent, even on flashpoint issues such as reproductive rights.  For Fiorina (2006), polarization on cultural issues is confined to elites and party activists.  In this project, I argue to the contrary.  While the citizenry is not polarized as a whole, there is strong evidence for polarization among significant subsets of the mass public.  Using data from the National Election Studies, I demonstrate the culture wars play out primarily among citizens who identify strongly with orthodox and progressive social groups.  These group members demonstrate high levels of consistency between values, political attitudes and behaviors - particularly on social policy issues.  Cultural issues are particularly salient for these individuals and weigh heavily on their electoral choices.  In this respect, these citizens can be thought of as issue publics (Converse, 1964).  In addition, I present data from a mail survey (N=537) and a web survey (N=2251) designed to address the limitations inherent in using the NES to study identity politics.   Specifically, I examine how identity salience, identity strength, and identity threat contribute to the culture wars phenomenon.
           To better understand the group basis of the culture wars, and the importance of collective identity on public opinion more generally, I draw on social identity theory.  Social identity theory (Tajfel and Turner, 1986) was originally developed to understand the psychological determinants of discrimination, but has been extended to explain a variety of phenomenon occurring both within groups and between them and has been applied to a wide range of group and intergroup contexts (see for example, Brown, 1995).  Social identity theory maintains that the self is composed of two distinct parts, the individual identity and the collective or social identity. The social identity is defined as: “that part of an individual’s self concept which derives from his knowledge of his membership in a social group (or groups) together with the value and emotional significance attached to that membership” (Tajfel, 1978).  Within groups, social identity is typically related to favorable evaluations of one’s group, adherence to shared values, conformity to group norms, and contributions to common pool resources.  In intergroup contexts, ingroup favoritism is fairly common.  There is evidence that collective identities also produce a host of negative attitudes and emotions towards oppositional groups – such as prejudice, anger, aggression, and intolerance (for a detailed review see Brown, 2000).  Chapters 1 and 2 provide a detailed overview of this literature.        
        The social identity literature suggests two factors moderate the expression of intergroup attitudes, emotions, and behaviors.  Attention to these moderators is critical for understanding the impact of identity on the culture wars at the mass level.  The first factor is individual differences in the strength of group identification.  Work on social identity theory has demonstrated the implications of group membership are most pronounced among individuals who possess a strong subjective sense of attachment to the group or have internalized their group membership (Luhtanen and Crocker, 1992; Ethier and Deaux, 1994; Kinket and Verkuyten, 1997).   In the political science literature, scholars typically ascribe respondents with a single group identity based upon their responses to survey items.  For instance, one might use the common religious denomination item to categorize respondents as Catholic, Protestant, or some other religious category.  For this method of establishing identity, strength of identification is assumed to be relatively homogeneous across group members.  This method also implicitly assumes a specific identity is primary or salient to respondents, and thus factors into their political orientations.  However, as the social identity theory literature indicates, identities vary in salience and strength across individuals with important consequences for attitudes and behavior (Huddy, 2001).  While one might indicate they are Roman Catholic, this person may place more importance on his or her identity as a feminist.  For this individual, the feminist identity could be more closely linked to political attitudes and behaviors.  Based on this work, I argue that the emotional and attitudinal processes characteristic of the culture wars will be most pronounced among individuals who primarily identify with orthodox and progress social groups and who possess a strong subjective sense of identification with these groups. 
            The second moderating factor is the presence of threat.  A growing body of literature suggests various forms of outgroup antipathy emerge consistently only under conditions of group threat (Brown, 2000; Flippen, Hornstein, Siegal and Weitzman, 1996; Postmes and Branscombe, 2002).  The culture wars rhetoric of elites and party activists contains numerous references to group-based threats. I expect orthodox and progressive group members will respond to cultural threats with anger, attitude polarization, and intolerance towards threatening groups.  These responses should be most pronounced among individuals who strongly identity with these groups.    
            In the first empirical chapter (Chapter 3), I investigate the extent to which the culture wars operate at the mass level by focusing primarily on individuals who identify with orthodox and progressive social groups.  Using data from the 1988-2004 NES and GSS, I develop measures of group identification and identity strength.  These measures are employed to evaluate the relationship between identity, attitude constraint, and consistency in political behavior.  In addition, I consider the relationship between identity, values, and political attitudes.  Using regression techniques, I demonstrate how identity moderates the relationship between values and attitudes.  These results suggest a stronger link between values and attitudes among respondents who identify with orthodox and progressive groups.  The analysis in this chapter supports the characterization of these group members as issue publics and points to the operation of the culture wars within a subset of the electorate.    
            Though the results from this chapter are suggestive, the data from these surveys allowed for a rather limited test of my core hypotheses.  To address questions raised about identity salience, gradations in identity strength, and dynamic responses to threat, I conducted a mail survey of “average” Americans (N=537) and a web survey of political sophisticates (N=2251).  Both were administered to an oversample of strong orthodox and progressive group identifiers to provide sufficient variance on key measures.  These surveys contained two measures of group identification and identity strength.  In Chapter 4, I evaluate the properties of these measures and consider the substantive implications of these results.  The first measure asks respondents to select all groups with which they identify from a checklist and indicate which group reflects their most important identity. Multidimensional scaling techniques are used to assess the relationship between these identities.  Preliminary results from this analysis suggest group identities are organized along two nearly orthogonal dimensions.  The first dimension is religiosity - anchored by Christian and Protestant identity at one end and atheist and agnostic identities on the other.  The second dimension reflects political ideology.  Interestingly, terms such as religious conservative, evangelical Christian, feminist, and secular humanist lie on the ideological dimension.  This finding suggests these identities are highly politicized for both the general population sample and the sophisticate sample.  This scaling technique was used to derive individual-level scores on these underlying dimensions, for use in subsequent chapters.  The second identification measure gauges gradations in identity strength.  I perform item-scale analyses on this measure to determine whether the scale has desirable properties and operates invariantly across the social groups of interest. 
            After establishing the quality of the identification measures and the relationship between the social identities of interest, I consider the effect of threat on the emotional and attitudinal processes characterizing the culture wars phenomenon in Chapter 5.   Both the mail and web survey contained a threat manipulation to allow for direct tests of the consequences of identity threat.  The content of the manipulation was varied to pose progressive or orthodox threat. The tone of the manipulation was varied as well, so that some of the respondents received a threat written in a personal tone (a personal vignette) and others received a threat written in a political tone.  Threats conveyed in a political tone reflect what we typically think of as culture wars rhetoric, while threat conveyed in a personal tone conveys a simple value violation.  If identity and identity threat are key components of the culture wars, one would expect responses to threat to be most pronounced in the political threat condition.  I hypothesize respondents will react to threat with anger, attitude polarization, and diminished tolerance for dissimilar groups.  Identity strength will moderate these responses to threat, such that they are most pronounced among individuals who identify strongly with key social groups.   
            While I have just begun data analysis for this project, the preliminary results look promising and are likely to shed new light on the micro-level dynamics of the culture wars. Fiorina’s (2006) recent work on the culture wars is rigorous but difficult to reconcile with anecdotal evidence from the American political life.  Recent controversies at both the elite and mass levels – such as bitter strife in Senate confirmation hearings for judicial nominees, the importance of referenda on gay marriage in many states for electoral outcomes in the 2004 elections, conflict at both levels regarding the outcome of the Terry Shaivo controversy and right to life (and death) issues more generally – suggest that issues where value priorities conflict produce anger, polarization, and intolerance.  By attending carefully to the nature and political consequences of collective identities, and drawing on research from social identity theory, we can develop a more nuanced understanding of the culture wars phenomenon and perhaps, more generally, obtain insight into the group basis of public opinion.

Works Cited

Frank, Thomas.  2004.  What’s the Matter with Kansas?  New York:  Henry Holt and Company. 

Miller, Alan S and John F. Hoffman. 1999.  “The Growing Divisiveness:  Culture Wars ora War of Worlds?”  Social Forces, Vol. 78, No. 2: 721-745.