SOCIAL GENERATIONAL CHANGE AND THE FOUNDATIONS OF ATTITUDES

ABOUT SAME-SEX MARRIAGE

by

Peter Hart-Brinson

A dissertation submitted in partial fulfillment of

the requirements for the degree of

Doctor of Philosophy

(Sociology)

at the

UNIVERSITY OF WISCONSIN-MADISON

2010
   ii  
Acknowledgments

More people than I could possibly name helped to make this dissertation possible, but I

particularly want to thank the following people, who all made invaluable contributions to my life

and my work during the time that it took to produce it.

First, I want to offer particular thanks to the members of my committee: to Pam Oliver,

for her patient guidance and mentorship, tireless feedback and advice, and friendship and

laughter; to Lew Friedland, for continually buoying my spirits and inspiring me with dreams of

the bigger picture; to Myra Marx Ferree, for her exhaustive feedback and always demanding that

I could do better; to Mustafa Emirbayer, for asking just the right, unforgettable questions to

move my thinking forward; and to Kathy Cramer Walsh, for her enthusiastic support and for

broadening and deepening my thinking about media, communication, and politics in ways I

never would have imagined.

Second, I want to thank everyone who helped make the collection and analysis of these

interviews possible: Evan Armstrong, Doree Brinson, Jim Brinson, Rachel Cusatis, Shelley Fite,

Carrie Coetsch, Rachel Hart-Brinson, Bill Hollander, Daniel Kappel, Natalie Neals, Aaron

Niznik, Lauren Olson, Maggie Phillips, and Rebecca Turcotte. I also want to thank the

University of Wisconsin, Northern Illinois University, and Rock Valley College for allowing me

to carry out these interviews, and especially to those individuals who helped me navigate the

obstacles to gaining that approval.

Third, I want to thank the National Science Foundation and the anonymous reviewers for

supporting this research with a Dissertation Improvement Grant, and I want to thank Brian

Powell for his encouragement, support, and collegiality.

   iii  
Fourth, I want to thank my family and close friends for all of the love and happiness that

they have brought to me. In particular: my parents, Jim and Doree; my brother Daniel; my wife

Rachel; my in-laws, Curt, Kristi, and Leah Hart; my dearest friends, Sam Ozer, Jacques

Arceneaux, Chris Fitzgerald, Matt and Tessa Desmond, Kate McCoy and Rob Young, Kristin

Vekasi and Nick Boissoneault. I could have never done this without how wonderful they have

made my non-academic life.

Finally, I want to thank the students and parents quoted throughout these pages for

opening up their lives to me—even if only for a few hours—and sharing with me the stories of

their past, the hopes of their future, and their views of the present. I learned far more from these

conversations than is reflected in this dissertation, and I am a richer man for being exposed to the

accumulated wisdom of so many people from so many different social worlds. If social science

accomplishes nothing, it should at the very least inspire wonder at the magnificent complexity of

humanity among those who are fortunate enough to practice the craft.
   iv  
Abstract

This dissertation draws on public opinion data and in-depth interviews with college

students and their parents to analyze the foundations of attitudes about same-sex marriage. The

analyses show that people’s attitudes and values regarding homosexuality are of fundamental

importance in shaping discourse about same-sex marriage, but that those relationships are more

varied and complex than expected. People’s beliefs, attitudes, values, and experiences regarding

homosexuality fail to cohere discursively, and there are multiple middle-ground discourses about

same-sex marriage that cannot be classified as either supportive or oppositional. The analyses

also show that people’s cultural definitions of marriage shape discourse about same-sex

marriage. There are tensions between heterosexual/religious and companionate/civil definitions

of marriage in discourse, but there also exists a shared consensus about what marriage means in

practice that resembles companionate marriage. It is argued that attitude polarization is not

inherent in the discourse but results from politicization of the issue.

More broadly, this dissertation analyzes the similarities and differences in how members

of two cohorts talk about same-sex marriage in order to shed light on processes of social

generational change. Most children and parents in the sample have similar attitudes on issues

related to marriage and sexuality because of the power of socialization, religious and political

ideologies, and attitude change among the older cohort. However, comparing the discourses of

parents and students who essentially agree with each other shows important differences in how

they talk about same-sex marriage. Younger religious conservatives and older liberals are more

likely to use middle-ground discourses because of the cross-cutting influences of their cohort

locations and their religious and political ideologies.

   v  
It is argued that making comparisons simultaneously between and within cohorts is

essential for explaining attitudinal differences regarding same-sex marriage. The “social

generation” concept can help account for these patterns if it is understood as a cultural and social

psychological process that accounts for how groups of people, defined intersectionally by cohort

and social location, develop similar worldviews based on their shared encounter with the social

structure. Conceived thusly, the social generation concept is integral to the analysis of social

change and social reproduction.

   vi  
Table of Contents

Introduction p. 1

Chapter 1: Literature Review p. 11

Chapter 2: Methods p. 57

Chapter 3: Quantitative Analysis of Attitudes p. 91

Chapter 4: Talk about Homosexuality and Same-Sex Marriage p. 132

Chapter 5: Talk about Marriage, Relationships, and Same-Sex Marriage p. 212

Chapter 6: Cohorts and Social Generational Change p. 275

Conclusion p. 336

Appendices p. 361

References p. 380
   1  
Introduction

Scope of the Dissertation

The controversy surrounding same-sex marriage in the United States appears to be a

summation of many of the most dramatic and important social changes that occurred in the last

three decades of the Twentieth Century. Since about 1969, the structure and meanings of gender,

homosexuality, marriage, and family in the United States have shifted. Due to the successes of

the women’s liberation and gay liberation movements, there have been tremendous strides

toward gender equality in many sectors of society, egalitarian gender ideologies have become

more prevalent, attitudes about sexuality both inside and outside of marriage have liberalized,

and the liberalization of attitudes about homosexuality has been especially pronounced. At the

same time, the United States has witnessed an increase in divorce and cohabitation rates, the

diversification of family forms, and consolidation of the companionate and individualistic

meanings of marriage as the hegemonic cultural ideal. To the extent that these changes are

viewed positively, support for same-sex marriage seems both logical and natural; likewise,

opposition to same-sex marriage seems natural if one views these changes negatively. Because of

the speed and scope of these changes, same-sex marriage appears to be an ideal case study of the

processes of social change through cohort replacement: how older adults with more conservative

orientations toward homosexuality, gender, marriage, and family are dying and being replaced in

the population by younger adults with more liberal views.

This dissertation is thus about the issue of same-sex marriage in particular and the

processes of social change and social reproduction in general. With regard to same-sex marriage,

I describe the cultural and social foundations of attitudes about same-sex marriage as thoroughly
   2  
as possible by using both quantitative analysis of public opinion data and qualitative analysis of

interviews with college students in northern Illinois and their parents. While measures of

demographic characteristics and social context are important predictors of attitudes about same-

sex marriage, I also seek to understand how discourse about same-sex marriage is premised upon

the ways that people talk about homosexuality and marriage. Most available evidence suggests

that people’s affective attitudes and moral values regarding homosexuality are the most

important predictors of attitudes, but it is also plausible that people’s cognitive beliefs about

homosexuality, their personal contact with gays and lesbians, and their understandings of

marriage also shape their attitudes about same-sex marriage. In this dissertation, I examine how

people draw from these different elements of their belief systems in order to talk about same-sex

marriage.

In addition, I seek to understand age-related differences in attitudes about same-sex

marriage that have been documented repeatedly in public opinion studies. In the U.S., not only

are young people more likely to support same-sex marriage than older people, but there has also

been a slow liberalizing trend in public opinion about same-sex marriage over the last 15 years.

The literature suggests that these are likely to be cohort-related differences, but there are

currently no studies that specifically address this question. Without longitudinal data, it cannot

be determined whether these attitude changes are due to age, cohort, or period effects. So in this

dissertation, I do not provide a definitive explanation of these patterns; rather, I look for evidence

about the significance of these age-related differences in both public opinion data and qualitative

interview data.

The common-sense interpretation of this pattern that is usually advanced by scholars and

commentators is that there is a generational gap between younger people and older people.
   3  
Specifically, younger people have grown up in a society that is more tolerant of homosexuality,

and in which gays and lesbians are more likely to be open about their sexual orientation and are

portrayed more frequently and in more sympathetic ways in mass media. As a result of coming

of age during this historical period, they have more positive attitudes about homosexuality, and

thus, about same-sex marriage. By contrast, older Americans grew up in a society in which

homosexuality carried a heavier stigma and when fewer people knew gays or lesbians personally

(or even knew the words “gay” or “lesbian”). They therefore are more likely to have negative

attitudes about homosexuality and same-sex marriage.

It is important to note that this narrative about generational differences goes far beyond

the relatively simple issue of disaggregating age, cohort, and period effects. Rather, it

presupposes the existence of a complex cultural and social psychological process that exposes

the ways that age, cohort, and period are inextricably intertwined: how each cohort, growing up

in a unique historical period, develops a particular worldview when they “come of age” about

what is normal and natural, which will operate as a baseline throughout the life course for the

development of further beliefs, attitudes, and actions. This is precisely what the theoretical

concept of generations and generational change connotes.

Unfortunately, this generation concept has always been more theoretically evocative than

empirically useful, and it was all-but abandoned by social scientists in the mid-1980s. In the

social sciences today, the word “generation” refers only to relations of kinship-descent, and the

word “cohort” refers to a temporally-defined group of people. We have no term to describe the

cultural and social psychological process described above. To the extent that contemporary

empirical research addresses this issue, it appears only tangentially in studies of collective

memory or as a theoretical justification for studying cohort effects.

   4  
However, a handful of scholars outside the United States have revisited this larger

generation concept in recent years, using the theoretical tools from the cultural turn in sociology

to rethink its significance. They have breathed new life into an old idea by drawing from

Bourdieuian concepts like habitus and field and by thinking in terms of cultural circles and

collective mentalities (Corsten 1999; Eyerman and Turner 1998; Gilleard 2004). One might

distinguish this notion from the “cohort” and the “generation” by dubbing it the “social

generation” (Esler 1984; Pilcher 1994). However, this small theoretical revival has gone largely

unnoticed in American sociology, and empirical support for this social generation concept still

remains tenuous and scattered.

This dissertation is ultimately an effort to overcome this gap between theory and

evidence of social generational change. By comparing the similarities and differences in

discourse about same-sex marriage between students and their parents, I examine how the broad

social generation concept is analytically useful for scholars who seek to understand the cultural

and social psychological mechanisms that can account for cohort differences. Put differently, this

dissertation analyzes cohort-related patterns of discourse about same-sex marriage by examining,

and to some extent controlling for, generational differences. In doing so, I show how the social

generation concept, when properly defined and operationalized, can account for the complex

macro-micro interactions between social structural change and the formation of distinctive

cultural repertoires that have long been assumed to be at the heart of the cohort replacement

mechanism of social change. This dissertation, therefore, provides scholars with an analytical

tool that can be used to more thoroughly study processes of social reproduction and social

change.
   5  
Chapter Outline

In Chapter 1, I discuss the research questions and review the scholarly literature on the

subjects covered by the dissertation. I begin by examining existing studies on same-sex marriage.

Because the issue of same-sex marriage is a relatively new issue for academic study, and because

the more general topic of homosexuality figures so prominently in the controversy surrounding

same-sex marriage, I include some scholarly research about homosexuality in the literature

review as well. Broadly, I focus on predictors of attitudes about same-sex marriage, dynamics of

political mobilization for and against gay rights, and the cultural and discursive components of

the controversy surrounding same-sex marriage. I discuss how the debate about same-sex

marriage and civil unions relates to the scholarly debate about attitude polarization and the

“culture wars” in American politics.

In addition, I discuss the existence of age-related differences in attitudes about same-sex

marriage in light of the scholarly literature on cohort and period effects in attitudes about gender

and homosexuality. I describe the cultural and social-psychological theory that can account for

the phenomenon of cohort succession, and I review the intellectual history of how scholars have

attempted to use the generation concept to account for social change over time. I argue that,

when properly formulated, the “social generation” concept has the potential to be a theoretically

and empirically useful tool for analyzing how cohort differences in attitudes emerge. Same-sex

marriage is an ideal case for assessing the merits of the “social generation” concept empirically

because of how the issue relates to broader structural changes—relating to gender,

homosexuality, marriage, and family—in American society that have take place since 1969.

In Chapter 2, I describe the research methods that I used to collect and analyze data for

the dissertation. I discuss the epistemology, site selection, sampling and recruitment techniques,
   6  
interview methods and settings, and analytic strategy for the qualitative data that I collected. I

also briefly discuss the secondary analysis of quantitative survey data that I use in Chapter 3. The

majority of the dissertation draws from qualitative methods, so the methods chapter reflects this.

The qualitative data provide deeper insights into the foundations of Americans’ attitudes about

same-sex marriage than any previous study, but only by virtue of its grounding in quantitative

analysis of public opinion data. Thus, I describe in this chapter how each data source and each

type of analysis contributes to the overall argument of the dissertation.

In Chapter 3, I present the results of the secondary analysis of quantitative survey data

from the Pew Research Center for People and the Press. First, I show that cohort measures are

stronger predictors of attitudes about same-sex marriage than the continuous age measure,

lending support to the hypothesis that age-related differences in attitudes about same-sex

marriage are in reality cohort differences. Second, I show that attitudes and cognitive beliefs

about homosexuality are all closely related to one another and that they are of fundamental

importance for understanding attitudes about same-sex marriage. Third, I develop a general

model for predicting attitudes about same-sex marriage, based on ordinal regressions of attitudes

about same-sex marriage on demographic, contextual, and ideological variables. Finally, I

explore whether foundations of attitudes about civil unions are similar to or different from those

of same-sex marriage. I conclude the chapter by considering the implications of this analysis for

the subsequent qualitative analyses.

In Chapter 4, I begin the qualitative analysis of interview data by describing how people

use their beliefs, attitudes, values, and life experiences regarding homosexuality in order to talk

about same-sex marriage. First, I show that people’s understandings of homosexuality are of

fundamental importance in shaping how they talk about same-sex marriage. Second, I show that
   7  
people’s understandings of homosexuality are composed of a variety of cognitive beliefs,

affective attitudes, moral values, and life experiences, which are themselves complex and not

necessarily ideologically correlated. Lastly, I describe a variety of discourses about same-sex

marriage, including supportive, oppositional, and two types of middle-ground discourses that are

neither supportive nor oppositional. I show how different discourses are ideologically rooted in

different combinations of attitudes, beliefs, values, and experiences regarding homosexuality,

and I show how they are sociologically rooted in different cohort locations and political and

religious ideologies. I argue that people’s understandings of homosexuality are so varied and

complex, and that they are connected in such varied ways with cohort locations and political and

religious ideologies, that discourse about same-sex marriage cannot be understood in simple

terms of support and opposition. Many people have moderate, mixed, or weak views about same-

sex marriage, and any polarization that occurs around the subject of same-sex marriage is not due

to people’s inherent worldviews, but to processes of political mobilization and survey

measurement that force people to choose sides.

In Chapter 5, I continue the qualitative analysis of interview data by describing how

people draw from their cultural understandings of marriage in order to talk about same-sex

marriage. With respect to same-sex marriage, I show that the primary axis of disagreement is

between religious and heterosexual definitions of marriage on one hand, and civil and

companionate definitions on the other. However, this disagreement structures the debate about

same-sex marriage less than it otherwise would because of widespread agreement, held broadly

by informants of all ages and creeds, about what marriage means in practice. Analysis of talk

about marriage, divorce, cohabitation, and premarital sex shows that the companionate

understanding of marriage operates as the cultural common-sense definition of what marriage

   8  
means in practice. Even those who advance religious and heterosexual definitions of marriage

nonetheless speak in individualistic and pragmatic ways about marriage and agree that

procreation is not essential for marriage. I argue that the definition of marriage therefore does not

constitute as strong an ideological basis for opposing same-sex marriage is it might have prior to

the historical period, beginning around 1969, that ushered in significant changes to the meanings

and structures of gender, marriage, family, and sexuality. Finally, I show that the significance of

civil unions in the debate about same-sex marriage is not that it represents a viable alternative for

recognizing same-sex relationships, but that it represents the idea that many opponents of same-

sex marriage nonetheless think gay and lesbian couples should have the same civil rights as

heterosexual couples.

In Chapter 6, I present evidence that the cohort differences documented in previous

chapters are due to the process of social generational change that is implied by cohort differences

in attitudes. First, I show that the discourses of students and parents bear the marks of the

differential effect of the same historical period on each cohort. Their discourses show that, while

parents were more likely to experience the gay rights period as a challenge to the views on

homosexuality that they developed when they grew up, students developed their initial attitudes

when the increasing openness and acceptance of gays and lesbians was the societal norm.

Additionally, a matched-pair comparison of students and parents shows that, even among those

who agree with each other on issues related to marriage, sexuality, religion, and politics, cohort

differences in the discourses about same-sex marriage remain.

I argue that these analyses show how social generational processes shape discourses

about same-sex marriage. A person’s cohort location and their social location within the cohort

shape the structure of their cultural repertoire and the ways that they use elements of that
   9  
repertoire to talk about same-sex marriage. The discourses thus represent the different strategies

that people use to combine and reconcile elements of their cultural repertoire that may be at odds

with each other when it comes to the issue of same-sex marriage. Of particular importance is a

person’s tacit definition of homosexuality. It is shaped by a person’s cohort and political and

religious ideologies, but it is also shaped by the extent of their contact with and acceptance of the

cultural construction of homosexuality that is dominant during a given period. Finally, I show

how the nature of a persons’ contact with gays and lesbians, their exposure to gays and lesbians

in mass media, and the composition of their social networks complicate the analysis of social

generational change with regard to same-sex marriage. I do this in order to argue that the

intersection of cohort and social location is, in reality, multi-dimensional and complex, and that

what is crucial for understanding social generational change is that it marks a shift in people’s

lifeworlds, such that their views and understanding of society change.

I conclude by considering the implications of this study for the future of same-sex

marriage in the United States and for scholarly research on processes of social reproduction and

social change more generally. Regarding processes of social reproduction and social change, I

argue that this study provides insight into the cultural and social psychological process that can

explain why cohort effects in attitudes occur. The social generation concept, as distinct from the

cohort and generation concepts, is defined as the cultural and social psychological process

through which groups of people, defined intersectionally by cohort and social location, encounter

a particular configuration of social structures and in turn, typically in young adulthood, develop a

particular cultural repertoire that they use in the further elaboration of attitudes and actions.

Defined and operationalized in this way, I propose that the social generation concept is integral

to scholarly efforts to understand processes of social reproduction and social change.

   10  
Regarding same-sex marriage, I argue that because age-related differences in attitudes

about same-sex marriage appear to be due to a cohort effect, and because the attitudes of older

adults are becoming more liberal over time, that the longitudinal trend toward greater support for

same-sex marriage is likely to continue. Moreover, because the apparent mechanism behind the

liberalization of attitudes is a change in the dominant cultural construction of homosexuality in

mainstream American culture, I argue that political attempts to stop same-sex marriage will only

be temporarily successful if they do not challenge this dominant construction of what

homosexuality means. Unless the mainstream cultural construction of homosexuality changes,

the liberalization in attitudes toward same-sex marriage is likely to continue through the process

of cohort succession.
   11  
Chapter 1: Literature Review

Introduction

Interest in how distinctive historical periods and significant changes in society affect

people’s worldviews, attitudes, and behaviors has a long history in the social sciences and in the

popular culture. Based on the historical period in which we come of age, we imagine that our

cultural understandings of what is normal in society are different from older or younger cohorts

that came of age when social and historical conditions were different. And we imagine that

society slowly changes through the process of cohort replacement, as older individuals who die

off are replaced by younger individuals with different worldviews. Evidence of this slow process

of change is all around us, but it is difficult to measure. It is even more difficult to establish an

empirical connection between the slow process of social change over history and particular

attitudes or behaviors of individuals in a way that is consistent with our sociological

imaginations.

Social scientists working today have only a minimal vocabulary and research agenda for

the set of cultural and social psychological processes implied in this conception of social change.

The term “generation” has long been used in academia and popular culture to describe this set of

ideas, but social scientists abandoned this meaning of the word in the mid-1980s because of

lackluster empirical findings and conceptual confusion with other meanings of “generation.”

Today, use of the word “generation” is restricted to its kinship-descent definition, while the word

“cohort” is used to refer to temporally defined groups of people. We know that cohort effects

occur and that cohorts differ in various substantive orientations, attitudes, and worldviews; but

we do not fully understand the cultural and social psychological processes that produce those
   12  
differences. Scholars doing cohort analysis and research on collective memories use this

conception of social change as a theoretical warrant for their research, but direct research on the

cultural and social psychological processes implied by the terms “cohort” and “generation” has

fallen out of favor in the American academy.

In recent years, however, the cultural turn in sociology has sparked renewed interest

among European theorists in “social generational” processes, and it has provided new empirical

tools for scholars to use in order to study the cultural and social psychological processes that are

implied by this conception of social change. The theoretical revisioning of social generational

processes in terms of “cultural circles” and in Bourdieuian terms of habitus and field (Corsten

1999; Eyerman and Turner 1998; Gilleard 2004) opens the concept to empirical research using

methods and theory of cultural sociology: the analysis of how people draw from their cultural

repertoires in order to construct opinions and how culture’s structure is manifested in discourses,

codes, metaphors, and frames that people use to express themselves (e.g. Alexander and Smith

1993; Foucault 1972; Swidler 2001).

With a different set of theoretical and empirical tools to study social generational change,

scholars now have an opportunity to overcome a persistent problem in the history of generational

research in the social sciences: the lack of rigorous empirical testing that supports generational

theory. The notion of generational change has always been more theoretically evocative than

analytically useful, since empirical investigations tend to result in either the stereotyping of

cohorts or analyses in which some competing variable, such as educational attainment, accounts

for the observed generational gap. However, empirical evidence of social generational change

may be more forthcoming with an analytical approach that is guided by the theory and methods

of cultural sociology.
   13  
I argue that the “social generation” concept, as distinct from the cohort and the

generation, can help scholars account for why cohort effects occur and explain variation among

the cultural repertoires (Swidler 2001) and practices of different groups. I define the social

generation as the cultural and social psychological process through which groups of people,

defined intersectionally by cohort and social location, encounter a particular configuration of

social structures and in turn, typically in young adulthood, develop a particular cultural

repertoire that they use in the further elaboration of attitudes and actions. By defining the social

generation in this way and applying the concept to age-related differences in attitudes, scholars

gain additional analytical insight into the processes of social reproduction and social change. In

particular, a focus on social generational processes can help us understand the dialectical process

of how macro-level changes in social structures are perceived and interpreted differently by

individuals of different cohorts and different social locations; and how new social structures are

consolidated and resisted by the subsequent talk and actions of those actors.

Research Questions

In this dissertation, I use a case study of discourses about same-sex marriage in order to

address this broader theoretical interest. I study how Americans use elements of their cultural

repertoire—cognitive beliefs, affective attitudes, moral values, and life experiences—in

conversation in order to construct and express opinions about the controversial issue of same-sex

marriage. I seek to understand both how these elements of people’s belief systems are related to

one another and whether or not there are differences between two cohorts of Americans in how

they talk about this issue. In particular, I seek to find out whether or not younger Americans have

different cultural understandings about marriage and homosexuality than older Americans, and if
   14  
so, whether or not this is related to differences in how they talk about same-sex marriage. The

two research questions that guide this research may be stated thusly:

RQ1: How do Americans draw from their cognitive beliefs, affective attitudes, moral

values, and life experiences in order to construct and express opinions about same-sex

marriage and civil unions?

RQ2: Are there differences between two different cohorts of Americans regarding their

understanding of homosexuality and marriage that are related to their lifeworld

experiences and attitudes towards same-sex marriages and civil unions?

The first research question is a question regarding the social and cultural foundations of

attitudes about same-sex marriage. A simpler way to ask this question might be: Why do people

hold the attitudes about same-sex marriage that they do? However, to ask this question this way

is both imprecise and misleading, because it fails to specify what constitutes a satisfactory

answer to the question, and it treats attitudes as immutable possessions that people hold, rather

than variable expressions that people construct in specific social contexts.

I draw from the concept of culture as simultaneously structural and pragmatic (Sewell

1999) in order to specify the problem more clearly. From the point of view of culture as

structure, I conceive of attitudes about same-sex marriage as being situated in belief systems,

realms of discourse, and “webs of significance,” all of which are grounded in social relations and

have definite structures that shape and constrain the meaning of same-sex marriage (Alexander

2003; Alexander and Smith 1993; Foucault 1972; Geertz 1973). From this point of view,
   15  
attitudes about same-sex marriage cannot be understood without accounting for its position and

relations in the broader cultural and ideological systems that give the issue its significance.

From the pragmatic view of culture, attitudes about same-sex marriage must not be

understood as something that a person holds—as an object in one’s possession—but rather as

something that an actor creates to accomplish a particular task in a particular situation. Attitudes

are always expressed by people in specific social contexts, and so the study of attitudes must take

into account how that context affects their manifestations. From this point of view, culture acts

as a “tool-kit” or “repertoire” that people use pragmatically in order to construct and express

opinions about same-sex marriage to someone and for some purpose (Battani, Hall and Powers

1997; Swidler 1986; Swidler 2001). Public opinion on such an issue is constructed based on the

differential processes of reception, interpretation, and selection of cognitive elements in response

to a prompt to offer an opinion (Zaller 1992). Thus, the study of attitudes must be a study of their

expression in a particular social context.

The second research question simply asks, what is the significance of age-related

differences in attitudes about same-sex marriage? Put differently, the second research question is

concerned with how the answer to the first research question depends upon one’s membership in

a socially-located age group. From an ideological point of view, differences in attitudes about

same-sex marriage should be based on differences in other aspects of people’s belief systems,

such as attitudes about homosexuality or beliefs about the meaning of marriage. Scholars have

long argued that different cohorts exhibit unique attitudes and practices because their worldviews

are shaped by their unique, temporally defined encounter with social structures when they come

of age (Ryder 1965). I compare the ways in which people from two cohorts talk about same-sex

marriage in order to understand the sociological significance of age-related differences in

   16  
attitudes about same-sex marriage. In my efforts to answer this research question, much more is

at stake than simply specifying an age, cohort, or period effect; this research agenda requires that

we take the “problem of generations” and the concept of social generational change seriously

(Esler 1984; Mannheim 1952 (1928); Pilcher 1994).

In this chapter, I examine the scholarly literature grounding each research question in

turn. I begin by reviewing existing studies on public opinion and political mobilization about

homosexuality and same-sex marriage, and I examine the ways that the scholarly debate on

attitude polarization applies to this issue. Then, I examine evidence of age, cohort, and period

effects with respect to same-sex marriage and argue that our attention should be focused on the

cultural and social psychological theory that is implicit in the description of cohort effects. I then

review the intellectual history of the generation concept in the social sciences and account for its

disappearance in the late 1980s. Finally, I argue that a recent revival in generational theorizing

shows promise for overcoming persistent problems with generational research, but that the true

test of the social generation concept lies in its empirical analytic utility. The ultimate significance

of this dissertation, beyond the issue of same-sex marriage, lies in the promise that the social

generation concept holds for scholarly analysis of processes of social reproduction and social

change.

Same-Sex Marriage

A variety of different kinds of studies have been conducted on people’s attitudes about

same-sex marriage, and they provide valuable insights into the reasons that people support or

oppose granting legal recognition to same-sex relationships. Broadly, the existing research on
   17  
same-sex marriage can be divided into two categories: studies of public opinion and studies of

political mobilization and discourse.

Quantitative studies of public opinion about same-sex marriage show how various

demographic, contextual, and cultural attributes are associated with various levels of support and

opposition to same-sex marriage. Among the demographic and contextual predictors of attitudes

toward same-sex marriage are religiosity, sex, age, education, political affiliation, and area of

residence (Brewer and Wilcox 2005; Olson, Cadge and Harrison 2006; Pew Research Center for

the People and the Press 2003). These studies have shown that religious beliefs and religious

affiliations are particularly powerful predictors of opposition to same-sex marriage. In the U.S.

context, this is easily accounted for by the cultural influence of Judeo-Christian teachings that

homosexuality is immoral and the religious connotation of marriage as a union, sanctioned by

God, between one man and one woman.

In addition to demographic and contextual predictors, attitudes about same-sex marriage

are influenced by cultural and cognitive factors, such as beliefs and attitudes about

homosexuality. Wilcox and Wolpert (2000) persuasively argue that controversies surrounding all

gay rights issues—whether it be same-sex marriage, gays in the military, gay adoption, etc.—are

primarily driven by people’s attitudes and moral values regarding homosexuality. From this

point of view, public opinion studies about homosexuality provide further insights into predictors

of attitudes about same-sex marriage. Predictors of attitudes about homosexuality that are likely

to be important predictors of attitudes about same-sex marriage include: beliefs about the nature

of homosexuality, holding negative stereotypes about gays, implicit and explicit motivations

regarding prejudice, belief in traditional moral values, emotional reactions to homosexuality, and
   18  
personal or mediated contact with gays and lesbians (Herek and Glunt 1993; Lemm 2006;

Schiappa, Gregg and Hewes 2006; Wilcox and Wolpert 2000; Wood and Bartkowski 2004).

Other cultural and cognitive factors shaping attitudes about same-sex marriage are likely

to include cultural definitions and attitudes about marriage, gender ideologies, and notions of

what counts as family (Cherlin 2004; Powell et al. 2010). The question of same-sex marriage

naturally raises questions about the nature of marriage and family and about the relationship

between sex, gender, and sexuality. Given social and religious conflicts throughout American

history over the meaning of marriage, the balance of power between men and women in the

household, and the legitimacy of different family forms, it is plausible that such conflicts might

also play a role in shaping the controversy over same-sex marriage. However, no public opinion

studies that I know of have documented the extent to which same-sex marriage attitudes are

shaped by these cultural factors.

Studies of political mobilization and discourse provide further evidence that religiosity

and attitudes about homosexuality fundamentally shape the controversy surrounding same-sex

marriage. Popular referenda on the issue of same-sex marriage in the United States have almost

always resulted in state-level statutes or constitutional amendments forbidding same-sex

marriage. Even when decisions are made in state legislatures, rather than in popular votes, there

is evidence that the religious and moral arguments of opponents are decisive. In an analysis of

same-sex marriage bans in the U.S., Soule (2004) shows that both interest organizations and

citizen ideology affect the adoption of same-sex marriage bans, and that when mobilization is

high on both sides, supporters of the bans were likely to be victorious. In public discourse,

proponents of gay rights and same-sex marriage rely on civil rights frames to make their

arguments, while opponents’ use of moral and religious arguments successfully counter the
   19  
influence of the rights frames (Brewer 2002; Brewer 2003; Hull 2001; McFarland 2008; Miceli

2005; Price, Nir and Cappella 2005).

The lines of political conflict about same-sex marriage follow the broad contours of

political mobilization and discourse about gay rights issues in general. In both scholarly and

popular accounts of struggles over gay rights, opposition to gay rights comes primarily from

Christian religious groups who advance moral arguments against homosexuality and the decline

of “traditional morality,” while support for gay rights is expressed in a language of civil rights, in

which gays and lesbians are analogous to an ethnic group that deserves equality under the law

(Gallagher and Bull 2001; Rimmerman, Wald and Wilcox 2000; Stein 2001). Many people who

have participated in same-sex marriage protests have done so to protest the unequal status of gay

and lesbian couples under the law, rather than out of any real desire to be married themselves

(Taylor et al. 2009).

Attitude Polarization and Civil Unions

Because of the recurrent patterns of support and opposition demonstrated in public

opinion studies and political mobilization studies, and because of the widespread homophily in

people’s social networks (McPherson, Smith-Lovin and Cook 2001), it would appear that the

controversy over same-sex marriage is simply the latest front in the “culture wars.” Same-sex

marriage, like abortion, is an issue that has been fiercely opposed by religious conservatives as

an example of how secular political groups are corrupting American society. According to the

culture wars thesis, political controversies revolving around “family values” issues about gender,

sexuality, and religion are increasingly fought between two ideologically opposed camps with

irreconcilable worldviews: the secular progressives and the orthodox conservatives (Hunter
   20  
1991). The thesis predicts that compromise “appears sociologically impossible” and that these

conflicts will only be resolved when one side defeats the other (Hunter 1991, p. 130). In support

of this view, scholars have cited long-standing attitudinal differences between evangelical

Christians and progressives, the increasing influence of “para-church organizations” in politics,

and the redefinition and polarization of political party identifications over time (Evans 2003;

Hunter 1991; McConkey 2001; Miller and Hoffmann 1999).

Scholars studying attitude polarization in public opinion, however, have criticized this

thesis because there is little evidence of polarization over time, with the one possible exception

of attitudes about abortion (DiMaggio, Evans and Bryson 1996; Evans, Bryson and DiMaggio

2001; Mouw and Sobel 2001). Systematic differences of opinion between orthodox religious

groups and other religious and secular groups are limited to issues surrounding gender, sexuality,

abortion, and schooling (Davis and Robinson 1996; Evans 2002). Critics of the culture war thesis

argue that the likelihood of two irreconcilable camps forming in the manner predicted by the

culture wars thesis is low, because the variety of conflicting status group memberships that

people hold make it implausible that people would agree with each other on all issues (Evans

1996). In reality, people become polarized in the context of political campaigns that offer people

only two choices and in a commercial media system in which only two sides are required for

“objectivity” and in which conflict is more likely to result in news coverage or higher audience

ratings (Gans 1979; McCarthy, McPhail and Smith 1996; Oliver and Myers 1999; Smith et al.

2001; Stein 2001; Tuchman 1978).

The possibility of compromise between supporters and opponents of same-sex marriage

is best represented by civil unions, a category of legal recognition of same-sex couples that

provides some of the same rights and benefits as marriage but is considered a separate
   21  
relationship status. Until recently, in states where the question of same-sex marriage rights have

been settled legislatively, such as Vermont and Denmark, efforts to compromise between pro-

and anti-marriage groups have resulted in civil union or domestic partnership laws (Eskridge and

Spedale 2006; Mello 2004). Setting legal considerations aside, politically, civil unions and

domestic partnerships offer the potential to meet the same-sex marriage supporters’ demands for

the legal rights and benefits for same-sex couples, while at the same time preserving the

heterosexual definition of marriage. Because the arguments of opponents revolve around moral

and religious definitions of homosexuality and marriage, civil unions avoid one of the central

objections to same-sex marriage.

Opinion polls have consistently shown higher levels of support for civil unions, to such

an extent that a clear majority of Americans support civil unions for gay and lesbian couples, but

an even larger majority of Americans oppose same-sex marriage (Brewer and Wilcox 2005).

Same-sex marriage thus might be a “take-off issue” that can lead to attitude polarization

(Baldassarri and Bearman 2007), but there also appears to be a possibility for compromise

represented by civil unions. It is clear that in order to understand the extent to which attitude

polarization is a reality for the case of same-sex marriage, we must understand both the structure

of public opinion about the issue and the dynamics of political discourse and mobilization.

My Contribution to the Literature on Same-Sex Marriage

In this dissertation, I attempt to fill a gap in the literature bounded by public opinion

studies on the one hand, and studies of political mobilization and discourse on the other, while

addressing the debate on attitude polarization. While public opinion studies tend to be based on

quantitative analyses of the attitudes of “ordinary Americans,” political mobilization and

   22  
discourse studies tend to be qualitative and focused on the people most active in political

controversies surrounding gay rights issues. Public opinion studies fail to problematize the social

context of attitude formation in the ways that mobilization and discourse studies do, but they

provide more reliable estimates of how attitudes and opinions are correlated with other

demographic and attitudinal variables.

A handful of recent studies have begun to fill this gap between quantitative public

opinion studies and qualitative studies of political mobilization and discourse. McVeigh and

Diaz (2009) show that opposition to same-sex marriage is higher in counties with more

procreative nuclear families and a more gender segregated division of labor. In this way, they

show how community context shapes voting behavior with regard to same-sex marriage. Other

studies use qualitative analysis of discourse in order to provide deeper insights into what same-

sex marriage means to people, but these studies so far have focused exclusively on LGBT

individuals and couples (Hull 2006; Lannutti 2005).

In this dissertation, I conduct a qualitative analysis of the discourses that are used by

“ordinary” Americans in order to talk about same-sex marriage in order to show how public

opinion is created discursively by individuals drawing from cultural repertoires to construct

opinions in a particular social context. As I describe in the next chapter, I conducted 97

qualitative individual interviews with college students and their parents in northern Illinois in

order to illuminate the social and cultural foundations of attitudes about same-sex marriage. I

analyze the ways that demographic and attitudinal variables typically measured by public

opinion studies shape discourses about same-sex marriage, but I also analyze how people

construct those discourses using their cultural repertoires that they have developed in the context

of their lifeworld.
   23  
By approaching this study qualitatively, I provide deeper knowledge about the

foundations of attitudes than is available solely from public opinion studies because I am able to

measure the discursive and social contexts of attitude construction in non-quantitative ways. This

does not mean, however, that I eschew quantitative analysis. To the contrary, in the dissertation,

I also conduct a secondary analysis of public opinion data from the Pew Research Center for the

People and the Press, which provides both a starting point for qualitative analysis and an

important validity check on the results and their generalizability. The results of this quantitative

analysis are indispensible for understanding the foundations of attitudes about same-sex marriage

because they help approximate the relative importance of different factors in shaping the endless

variations in discourses and attitudes that are measured by qualitative methods.

Because of the ways that qualitative and quantitative methods, when used in tandem, can

overcome the relative weaknesses of each approach, the results of this study will provide one of

the most thorough accounts available of culture’s structure, as it applies to the issue of same-sex

marriage. The ways in which discourses and attitudes about same-sex marriage are shaped and

constrained by other elements of people’s belief systems and people’s social contexts will be

shown more comprehensively by drawing from the strengths of both quantitative and qualitative

studies.

Moreover, the results show how people pragmatically draw from elements in their

cultural repertoires in order to express their views about same-sex marriage. This contributes to

our understanding of mobilization and discourse by showing how non-activists talk about same-

sex marriage in a context where same-sex marriage is legal possibility but not currently the

subject of a political campaign. Through their talk, people create the discursive terrain upon

which political contests will ultimately be fought, and they both affirm or refute old meanings of
   24  
same-sex marriage while creating new ones. Patterns of talk about same-sex marriage are thus

important as both indicators of attitudes and as political actions—as discourses with public,

political, and cultural consequences of their own (Arendt 1998; Cohen 1999; Eliasoph 1998;

Foucault 1978; Gamson 1992).

This study thus contributes to our understanding of culture as simultaneously structural

and pragmatic. I take as axiomatic that attitudes about same-sex marriage are structurally related

to other elements in people’s belief systems and life experiences, but that we cannot understand

those attitudes about same-sex marriage without studying how people express them in discourse.

For example, Hull (2006) shows how culture is simultaneously structural and agentic by

analyzing how same-sex couples talk about commitment ceremonies in qualitative interviews.

Different couples accept some meanings of marriage and commitment but not others, and they

create new meanings when adapting the commitment ceremonies for their own purposes. My

study expands on her findings about the cultural meanings of same-sex marriages and civil

unions by showing how the cultural meanings of same-sex marriage are expressed in the

discourse of “ordinary” Americans.

This study addresses the debate on attitude polarization in two ways. Because the battle

lines over same-sex marriage appear to follow the contours of the “culture wars” more generally,

this issue is a good case study for weighing the evidence of polarization in the discourse of

ordinary Americans. First, I consider the issue of civil unions explicitly, both in quantitative

analysis of public opinion data and in qualitative analysis of interviews. By considering the civil

union compromise explicitly, I provide evidence about the logic and resonance of the civil union

compromise in mainstream American culture. Second, this study addresses the debate on attitude

polarization by examining how people talk about the issue without the constraints of survey
   25  
answer choices and outside the context of electoral politics. On the issue of abortion, for

example, attitudes expressed on surveys can appear much more polarized than they appear in

group discussions (Press and Cole 1999). This study is therefore unlikely to show evidence of

ideological irreconcilability. More important than the presence or absence of conflict, however,

is the analysis of the context and processes of attitude construction in discourse. This study

shows how people construct attitudes in a particular context and thereby illuminates the ways in

which polarization may or may not result from the controversy surrounding this issue.

Age, Cohort, and Period Effects in Attitudes about Same-Sex Marriage

A number of public opinion studies, including those cited above, have shown consistent

age-related differences in attitudes about same-sex marriage. In particular, studies consistently

show that people under age 35 or so are much more likely to support same-sex marriage than

other Americans. This finding is widely documented and acknowledged by scholars and by

activists on both the left and right. Liberal advocacy organizations have certainly emphasized

these age differences in their analyses of same-sex marriage, but even organizations that oppose

same-sex marriage acknowledge that young people hold more liberal attitudes than older people

(Egan and Sherrill 2009; Gallagher and Baker 2004). Figure 1.1 is a graph from a Pew Research

Center report (2003) showing 5-year moving averages of support and opposition to same-sex

marriage by age. People younger than age 35 show significantly different patterns of support and

opposition compared with people over age 35.

To my knowledge, there have been no scholarly attempts to explain why young people

hold more liberal attitudes toward same-sex marriage, though the age difference is frequently
   26  
Figure 1.1: Patterns of Support and Opposition for Same-Sex Marriage, by Age

Source: Pew Research Center for the People and the Press (2003).

commented upon. Most writers implicitly assume that the effect is a cohort, or “generational,”

difference rather than an age effect. For example, from a 2008 survey, the authors write:

Generational differences are striking on the issue of same-sex marriage. Forty-six percent

of young adults support the right of gay couples to marry, 17 points higher than the

general population. These generation gaps exist across every religious tradition, including

amongst white evangelicals—younger white evangelicals are nearly 2.5 times more likely

to support same-sex marriage than are all white evangelicals. (Faith in Public Life 2008,

http://faithinpubliclife.org/tools/polls/faps/)
   27  
Similarly, Lax and Phillips (2009) present the following graph (Figure 1.2) in an online

appendix, showing estimates of support for same-sex marriage by state, derived from polling

Figure 1.2: Explicit Support for Same-Sex Marriage by State and Age

Source: http://www.stat.columbia.edu/~cook/movabletype/archives/2009/06/

future_trends_f_1.html, accessed January 21, 2010.

   28  
data between 1994 and 2008. They observe that “the generation gap is huge” and speculate that

the higher levels of support among the youngest cohort is an indicator of future increases in

support for same-sex marriage.

Longitudinal data on attitudes toward same-sex marriage are sparse, though public

opinion polls show that opposition has softened significantly since the mid-1990s (Brewer and

Wilcox 2005; Pew Research Center for the People and the Press 2009). With the exception of a

small spike in opposition to same-sex marriage in 2003-04, most likely the result of a “moral

panic” that was set off by the 2003 legalization of same-sex marriage in Massachusetts (Goode

and Ben-Yehuda 1994), support for same-sex marriage has risen slowly over the last 15 years.

This liberalizing trend lends support to the assumption that the liberal attitudes of younger people

is a cohort effect, though the longitudinal trend could also be due to a period effect, if all cohorts

are becoming more liberal.

Despite the lack of longitudinal data that would be required to mathematically

differentiate age, cohort, and period effects with respect to same-sex marriage attitudes (Glenn

1977), circumstantial evidence suggests that these age-related differences are most likely cohort

effects. Most importantly, studies of attitude change about homosexuality and gender show

pronounced cohort effects. Younger cohorts have more liberal attitudes toward homosexuality

and are more likely to have egalitarian gender ideologies than are older cohorts (Brewster and

Padavic 2000; Brooks and Bolzendahl 2004; Ciabattari 2001; Loftus 2001; Treas 2002). These

studies show very little evidence of age effects regarding homosexuality and gender ideologies,

so it appears unlikely that young people with liberal views are becoming more conservative on

these issues as they get older.

   29  
In addition, these studies show consistent evidence of individual-level attitude change

over time in all age groups. Even among older cohorts, people are becoming increasingly tolerant

of homosexuality and increasingly supportive of extending equal rights to gays and lesbians. In

aggregate terms, this constitutes a period effect. Given these findings and the ideological basis of

same-sex marriage attitudes in attitudes about homosexuality and gender, it is reasonable to

expect that the more liberal attitudes held by young people toward same-sex marriage is a

manifestation of cohort differences as well.

To observe that there are cohort differences in attitudes, however, does not explain why

those cohort differences arose. Typically, the existence of cohort differences is taken to signify

something unique about the group’s historical encounter with society, based on a shared period

of their birth, that shapes their attitudes, and that the influence of this shared encounter will be

carried with them as they age (Ryder 1965). In social-psychological terms, cohort analysis

typically implies four interrelated propositions: 1) the “impressionable years hypothesis,” that the

“coming of age” period of late adolescence and early adulthood is the period of the life course in

which durable attitudes and dispositions are formed; 2) the “aging-stability hypothesis” (or

“persistence hypothesis”), that attitudes formed during the impressionable years will remain

fairly stable as people age, once major life-course transitions are taken into account; 3) the effect

of history on biography, that social structural change over time means that each cohort will

encounter a society different from that encountered by other cohorts in their impressionable

years and that their worldviews will vary accordingly; and 4) the “cohort replacement”

mechanism of social change, that as older cohorts die off, they are replaced by younger cohorts

with different value orientations, and that social change slowly occurs as the “demographic

metabolism” continuously and systematically changes the composition of society (Ryder 1965).
   30  
The empirical literature provides ample support for these propositions. Numerous

scholars have found support for both the impressionable years hypothesis and the aging-stability

hypothesis in political party identification and other political attitudes (Alwin and Krosnick

1991; Glenn 1980; Miller and Sears 1986; Sears and Funk 1999). Similarly, Weil (1987) found

evidence of cohort and period-cohort effects in the political attitudes of Germans who came of

age during the Nazi regime. Schuman and Scott (1989) found that the things that people

remembered as important social or political events tended to be events that took place during

adolescence and early adulthood. A number of scholars have found evidence of the impact of

periods of social structural change on religious identification (Hout and Fischer 2002; Wuthnow

1976) and on “feminist” identification (Schnittker, Freese and Powell 2003). Cohort replacement

has been identified as a mechanism for change in tolerance toward political groups (Wilson

1994), attitudes about homosexuality (Loftus 2001; Treas 2002), gender ideology (Brewster and

Padavic 2000; Brooks and Bolzendahl 2004; Ciabattari 2001), and general value orientations

(Abramson and Inglehart 1986).

Applied to the case of same-sex marriage, the hypothesis of cohort effects in attitudes

means that young people have more liberal attitudes because they came of age in a society,

defined by a particular time and place, that is more tolerant of the idea of two individuals of the

same sex being legally married due to some combination of structural changes. One might

conceive of a variety of structural changes that could lead to younger cohorts adopting more

liberal attitudes toward same-sex marriage: the successes of gay and lesbian movements for

equality, increasingly liberal attitudes about sex and sexuality, increased educational attainment,

greater visibility of gays and lesbians in the mass media, changing cultural meanings of

“marriage,” diversification of family forms, the decline in traditional gender ideology, increasing
   31  
gender equality, and increasing tolerance toward minority groups. Coming of age in a society

marked by such changes logically could lead to individuals adopting more liberal views on the

question of same-sex marriage.

Indeed, all of these changes have occurred in the United States since about 1970.

Considering the ages of people currently living in the United States today, the structures and

meanings of gender, sexuality, marriage, and family have changed dramatically between the time

that the oldest cohorts came of age and the time that the youngest cohorts came of age. Senior

citizens in the United States came of age at a time when essentialist gender differences were

assumed, when a strong division of labor and stratification of power existed between husbands

and wives, when the nuclear family was the hegemonic ideal family form, and when sex outside

of marriage—including homosexuality, premarital sex, extramarital affairs—was culturally

taboo. By contrast, young adults in the United States today are coming of age in a society that

looks very different.

First, although significant inequalities remain in the status of women and men in most

spheres of society, including labor markets, families, education, and interpersonal interaction,

there have been tremendous advances in gender equality in recent decades (Brewster and

Padavic 2000; Buchmann, DiPrete and McDaniel 2008; Hochschild 1989; Ridgeway and Smith-

Lovin 1999). Compared to the levels of gender inequality in the 1960s, contemporary American

society is far more gender egalitarian. Young adults coming of age after 1990 largely take for

granted the struggles of the feminist movement: for non-discrimination laws, for reproductive

rights, and for cultural expectations that women should be equal to men in access to educational

and work opportunities and a variety of personal characteristics, like intelligence. The change in

meaning of the “feminist” identity for younger cohorts reflects these changes in the structure of
   32  
American society that were sparked by the second-wave feminist movement (Schnittker, Freese

and Powell 2003). In short, the younger cohorts grew up among a set of social institutions that

promoted and assumed gender equality (Jackson 1998), whereas the older cohort grew up among

a set of social institutions that were more likely to promote and enforce gender differences.

Second, with regard to homosexuality, people coming of age after 1990 grew up in a

society in which gay rights was the dominant political strategy of the gay and lesbian

movements, gays and lesbians were increasingly likely to be “out” in media and daily life, and in

which homosexuality was increasingly accepted as normal. Public opinion towards gays and

lesbians liberalized significantly since the 1970s, and especially after 1990 (Loftus 2001; Treas

2002; Wilcox and Wolpert 2000). Beginning in the early 1990s, gay and lesbian characters in the

mass media were increasingly portrayed as protagonists and as normal people with whom the

audience could identity, and they were less stereotyped than they had been previously (Becker

2006; Gross 2001; Seidman 2004; Walters 2001). Also at that time, gays and lesbians gained

political standing in their struggles for civil rights, funding for HIV/AIDS research, and with the

1992 election of Bill Clinton as President (Rimmerman, Wald and Wilcox 2000). The increasing

visibility and power of gays and lesbians in American politics and culture stands in stark contrast

with the years prior to 1969, when gays and lesbians were largely “in the closet” in mainstream

American culture, and homosexuality was defined as a mental illness, or at best, a deviant

lifestyle.

Third, with regard to marriage and family, starting in the 1960s, U.S. society has been

characterized by what Cherlin (2004) calls the “deinstitutionalization of marriage.” Marriage and

family has been significantly transformed between the 1960s and 1990s: the implementation of

no-fault divorce laws in the 1970s resulted in a sharp rise in the divorce rate; cohabitation
   33  
became more common; out-of-wedlock births increased while the overall birth rate declined; and

attitudes about divorce, cohabitation, and premarital sex have become more liberal (Amato 2004;

Mintz and Kellogg 1988; Thornton and Young-DeMarco 2001). The hegemony of the ideal

nuclear family has declined, and alternate family forms have become more common. The

institutional model of marriage has been replaced by the companionate model in the American

imagination, and Cherlin (2004) argues that even this model of marriage is being replaced by an

“individualized” model of marriage. Marriage is less likely to be regarded as a societal

obligation for the purpose of raising children and fulfilling the generational social contract, and

marriage is now more likely to be thought of as a choice for couples who want to use their

companionate relationship to maximize individual growth (Amato 2004; Cherlin 2004; Thornton

and Young-DeMarco 2001). Thus, the youngest cohorts came of age in a time when “marriage”

and “family” carried cultural meanings that are different from the meanings that the words held

when the oldest cohorts came of age.

There is growing evidence that these structural changes are linked to increasing support

for same-sex marriage. The dramatic changes in American society in gender, marriage, family,

and sexuality over the course of the last 40 years have, in effect, moved the question of same-sex

marriage from the realm of utopian or dystopian fantasy to the realm of political reality. For

example, Rosenfeld (2007) argues that same-sex marriage and other alternative unions have been

made possible by the rise of the “independent life stage” and changing patterns of marriage and

family formation since the 1960s. McVeigh and Diaz (2009) also demonstrate a connection

between the prevalence of “traditional” gendered work and family structures in a community and

opposition to same-sex marriage. Support for same-sex marriage is ideologically plausible in the

context of egalitarian gender relations, tolerance for homosexuality, a non-procreative and

   34  
companionate definition of “marriage,” and a pluralistic understanding of the notion of “family.”

For people who hold such a combination of attitudes about gender, sexuality, marriage, and

family, support for same-sex marriage seems not just possible, but normal and logical. In this

way, same-sex marriage is an ideal case study for studying cohort differences and cohort

replacement in the United States today. Almost unimaginable in the U.S. in the aftermath of

World War II, same-sex marriage is now a reality in five states and highly contested in several

others.

Cohort and Generation

The four interrelated propositions typically assumed to account for cohort effects,

described above, and their applications to the case of same-sex marriage evoke common cultural

conceptions of “generations” and “generational change.” In the popular imagination and in

numerous theoretical and empirical studies, the broad generation concept has been understood to

imply precisely this set of relationships between history and biography: relationships among

cohort, life course, social structure, and attitudinal change. The terms “cohort” and “generation”

are often used interchangeably when talking about this set of relationships, although the latter is

more often used in a casual or informal context and less often in a social scientific context.

Despite widespread empirical evidence in favor of this concept of cohort/generational

change, the literatures on cohorts and the broad generation concept are burdened by two

persistent problems. First, there is a significant gap between the connotations and denotations of

the words “cohort” and “generation.” Technically, cohorts are groups of people defined

temporally, and “generation” refers to relations of kinship-descent; but both terms connote a

much larger theory of social change regarding the macro-micro relationships between periods of
   35  
historical change and the biographical or psychological characteristics of individuals. For

example, in Ryder’s classic (1965) article on the cohort concept, he focuses on the theoretical

implications of the concept rather than on the concept itself in an effort to make a case for the

importance of studying cohorts. Today, the cultural and social psychological processes that are

assumed to be implicated in the cohort concept are taken for granted as a warrant for conducting

cohort analysis, rather than considered as objects of research themselves.

Likewise, the term “generation” is inappropriate for capturing the cultural and social

psychological processes implied by this theory of social change. Although people use the word

informally to refer to this set of ideas, social scientists have restricted the use of the word

generation to its kinship-descent definition for reasons I describe in the next section. As a result,

while research on generations and cohorts is thriving, research and theorizing about the cultural

and social psychological processes of implied in theories of cohort replacement and generational

change has almost entirely disappeared from the social sciences.

I argue that these cultural and social psychological processes should be studied in their

own right because untested assumptions are frequently wrong and because the lack of scholarly

research on the subject means that our theoretical understanding of this type of social change

may be rudimentary. Solving the terminological problem by clarifying the meanings of

concepts—and denoting the “social generation” as a new concept—is one step toward better

social scientific study of generational change, but the new concept must itself be both

theoretically and empirically sound.

This leads us to the second problem with the research on cohort-related phenomena: the

tendency of the concept to level social differences and to be interpreted monolithically. Cohort

effects and generational change tend to be thought of as society-wide phenomena, ignorant of

   36  
social differences among individuals or collective boundaries within a society. In thinking about

these concepts, scholars tend to focus on differences between cohorts, setting aside variation

within cohorts as the province of some other domain of inquiry. This is no doubt a function of

the simple definition of a cohort as a group of people marked by a temporal event or boundary.

Definitionally, other social boundaries are irrelevant.

In the literature, this problem manifests itself in a number of ways. First, arguments of

cohort effects are difficult to reconcile with the effects of other variables, such as educational

attainment or political ideology. A variety of studies pit educational attainment and cohort

replacement against each other as competing explanatory variables, rather than considering

differences in educational attainment as part of the study of cohort replacement (Alwin 1990;

Cutler and Bengtson 1974; Loftus 2001). While there is substantial merit in differentiating one

causal mechanism from another, it would be a mistake to theoretically conceptualize cohorts and

other variables as mutually exclusive, rather than interrelated.

Second, cohort effects are observed for some groups but not others, an occurrence which

is typically interpreted as counter to the normal way of thinking about cohorts. For example,

Alwin (1990) documents the effects of cohort replacement on “parental socialization values” but

observes that the changes take place primarily among Catholics and that this finding “raises

questions about the usefulness of a general cohort explanation in this case” (p. 358). Likewise,

Griffin (2004) provides an important corrective to Schuman and Scott’s (1989) findings on

generations and collective memories by showing that memory of the Civil Rights Movement

depends on race and geographical residence. These should not be understood as limit cases of a

theory of cohort replacement, simply because cohort effects are not observed uniformly across
   37  
all social groups. Instead, scholars should dispense with the assumption that cohort effects will

be observed across all social groups to begin with.

Third, descriptions of cohorts and generations tend to sound like stereotypes or tend to be

based implicitly on one or more dominant groups (e.g. whites, middle-class, etc.). This problem

follows from the previous two. Findings of cohort differences may be driven by the numerical

dominance of one or more social groups in a dataset, and descriptions of cohorts or generations

may be based on the characteristics of only a small proportion of its members. This leveling of

social difference is both politically and scientifically dangerous, as it easily leads to unwarranted

generalizations or conclusions about the extent and nature of social change.

I argue that what is needed is a theory of cohort replacement and generational change that

embraces variation within cohorts rather than marginalizes it. While the definitions of cohort and

generation need not be changed, scholars should understand that cohort membership may mean

different things to different people. The fact of being a member of a particular cohort matters, but

it matters in different ways, depending upon one’s social location. In the rest of the chapter, I

argue that the “social generation” concept provides a way of incorporating these insights into our

ways of conceptualizing cohorts, social structure, and social change.

History of the Generation Concept in Social Science Research

The “generation” concept in the social sciences has a turbulent history. Since the initial

formulations of the concept and its social significance in the early 20th Century (Mannheim 1952

(1928); Ortega y Gasset 1958), the generation concept has been plagued by a persistent problem:

it has always been more theoretically evocative than analytically useful. Despite how essential

the concept is to sociological understandings of how societies change over time, simply through
   38  
the “demographic metabolism” of dying elders being replaced by newborn youth (Ryder 1965),

empirically measuring generations and their effects has been extremely difficult. Perhaps

because of the continuous, slow, and complex nature of this mechanism of social change, the

phenomenon has been highly resistant to empirical verification or analysis using available

research methods.

A look at the intellectual history of the generation concept illustrates three ways that this

problem manifests itself. First, the generation concept is so theoretically rich, that the word has

been used in both academic and popular discourses to mean a number of distinct things. In other

words, the term has so many definitions that it has been difficult to develop a coherent body of

theory and research around the concept.

Second, empirical studies that were conducted explicitly on generations frequently failed

to support the broad theory of generational change. Most scholarly research on generations has

used survey techniques and methods of quantitative analysis to assess evidence of differences

between “generations” and generational change over time; but the existing techniques and

quantitative analytic paradigm have typically juxtaposed measures of generations with other

variables, such as educational attainment and political ideologies. Patterns of generational

differences are frequently explained by these other variables instead, and thus interpreted as

evidence against a theory of generational change.

Finally, the narratives of generational differences in both popular and academic works are

driven by stereotypes or caricatures, in which social difference is leveled and all people are

assumed to think and act like the culturally dominant group (usually white, middle-class people

whose traits have been assumed to be typical in mainstream historical narratives about the

periods in question). Authors have tremendous incentives to marginalize the distinct experiences
   39  
and worldviews of subordinated groups because they directly contradict the presumed

generational differences and undermine the argument of generational change.

In the intellectual history of the generation concept, all three of these symptoms are

persistent and intertwined, and I will not attempt to disentangle them in the pages that follow.

Taken together, they help account for the marginal status of the concept in the social sciences in

the late 20th Century. As I will argue, because of these problems, the large, socio-cultural

generation concept was abandoned by sociologists in the mid-1980s in favor of narrower, more

specific concepts. Despite the relative absence of sociological research and theorizing about this

broad generation concept, the term has persisted in other popular and academic treatises on the

subject. As a result, the stereotyped and caricatured notion of generations and generational

change promoted by popular culture has gone unchallenged, and is even supported, by scholars.

In recent years, however, the cultural turn in the social sciences has provided researchers

with new theoretical tools and new methods for studying the broad socio-cultural generation

concept in ways that overcome some of its problems. Theorists, mostly outside the United States,

have begun describing how the “social generation,” as a cultural concept, provides new insights

into the process of social change caused by cohort replacement. While theoretically compelling,

these arguments still suffer from a lack of empirical evidence to support them, as did the older

formulations of the generation concept. In this dissertation, I build on these efforts to theorize the

“social generation” by operationalizing the concept in light of Mannheim’s seminal essay in the

sociology of generations. This operationalization then offers the opportunity to empirically

assess this theory of social generational change.

Karl Mannheim’s Poly-vocal Generation Concept

   40  
The seminal sociological work on the generation concept is widely considered to be Karl

Mannheim’s (1952 (1928)) essay, “The Problem of Generations.” The importance of

Mannheim’s formulation lies in his distinction between the “generation location” (p. 290), the

“generation as an actuality” (p. 303), and the “generation unit” (p. 305). Drawing from the

Marxist tradition, Mannheim uses “class” as an analogy for understanding “generation.” The

social significance of a class depends upon more than sharing a common location in the class

structure with other individuals; it also depends upon the identification of these individuals with

one another: class consciousness.

So it is with generations: “generation location” refers to the biological and temporal

reality of being born in a particular place at a particular time, whereas “generation as an

actuality” refers to “a concrete bond [that] is created between members of a generation” (p. 303).

This “concrete bond” does not refer to face-to-face relations, but rather to a sort of collective

mentality that is due to the fact that this cohort came of age in the same socio-historical

conditions. It refers to their “participation in the common destiny of this historical and social

unit” (p. 303, emphasis in original). The “actual generation” may then be further subdivided into

“generation units,” groups within the generation who “work up the material of their common

experiences in different specific ways” (p. 304). Generation units are akin to the mobilized

groups of a particular class, like Marx’s Communists: they are interacting social groups who

share a common orientation to the society, because they have reacted similarly to their shared

temporally and socially located encounter with the social structure. Mannheim describes the

distinction among the three distinct phenomena this way:

“Whereas mere common ‘location’ in a generation is of only potential significance, a

generation as an actuality is constituted when similarly ‘located’ contemporaries

   41  
participate in a common destiny and in the ideas and concepts which are in some way

bound up with its unfolding. Within this community of people with a common destiny

there can then arise particular generation-units. These are characterized by… an identity

of responses, a certain affinity in the way in which all move with and are formed by their

common experiences.” (p. 306)

Two of Mannheim’s meanings are easy to discern, while the “generation as an actuality”

is more difficult. What Mannheim termed the “generation location” is what we today call a

“cohort.” What Mannheim termed the “generation unit” is an identifiable social group of

individuals who think and act in similar ways, like a social movement group or a subcultural

group. Mannheim’s “generation as an actuality” refers to how people of a particular cohort, in a

particular social location, develop a distinctive cultural understanding of society based on their

particular temporally-defined encounter with it as youth “coming of age.” The “actual

generation” is shaped by more than just temporal location; it is constituted at the intersection of

temporal and social location. Although not everyone in a “generation as an actuality” will have

the same attitudes (that would be a “generation unit”), they may have similar ways of “making

sense” of society based on a shared temporally- and socially-defined encounter with the social

structure.

Mannheim’s model of generations is something of a nested model. Being in the same

cohort is necessary but not sufficient for being a part of the “generation as an actuality”—you

have to also share a common social location such that you experience the changing social

structure in a way that is shared with other people of that cohort. But, there are multiple

generation units within the “actual generation” because not everyone who experiences the same
   42  
social changes will react in the same way. Thus, this middle category, the “generation as an

actuality,” is relationally defined both between cohorts and within cohorts. It does not refer to a

coherent social group, nor is it something that is defined purely temporally. It is defined in an

intersectional way by both time and social location.

Mannheim’s theory at once provides the essential distinctions for studying the

significance of generations sociologically and an explanation for why the concept has been used

in such varied and convoluted ways. When someone uses the term “generation,” they might use

the term in a variety of ways. To speak of the Baby Boom Generation, the meaning is that of the

cohort, the generation location: those born in the United States between roughly 1945 and 1963.

By contrast, academic works about the “Generation of 1914” (Wohl 1979), “Sixties Generation”

(Whalen and Flacks 1989) or the “Vietnam Generation” (Wyatt 1993) are about generation units,

particular groups that responded to historical events in particular ways.

Especially regarding the “generation as an actuality,” Mannheim’s distinctions are

somewhat vague and abstract. The “generation as an actuality” does not refer to an existing

social group, nor is it simply defined temporally. It is this category of “generation as an

actuality” that is the source of much frustration and confusion. For example, stereotypes about

members of “Generation X” as apathetic under-achievers are attempts at delineating a

“generation as an actuality,” because they “are trying to show that all these generations have

different life cycle connections with history, giving them different collective personalities and

making them interact with each other in peculiar ways” (Howe and Strauss 1993, p. 41). Such

caricatures, however, are obviously too monolithic and inadequate for meaningful social

research.
   43  
Generational Analysis Through 1985

The rapid social change of the 1960s appears to have inspired scholars to take up research

on generational politics, and generational analysis was a thriving area of research in the 1960s

and 1970s. Mannheim’s essay and Eisenstadt’s (2003 [1956]) structural-functionalist approach to

youth rebellion and generational change provided scholars with fertile theoretical grounds on

which to base empirical studies of the youth counterculture and social movements of the time.

Reviews of the literature from that period (Bengtson, Furlong and Laufer 1974; Braungart and

Braungart 1986) demonstrate significant advances in theorizing the relationships among cohorts,

aging, politics, and social change, but also pointed out persistent difficulties. For example, how

to define the boundaries of generations, how to distinguish between cohorts and generations, and

how to disentangle cohort or generational effects from other types of effects were questions that

posed significant problems for empirical study.

Empirical attempts to measure (usually quantitatively) significant differences among

cohorts or generations yielded only mixed results: those effects were typically better accounted

for by measures of political ideology or activism (Roberts and Lang 1985; Wuthnow 1976) and

education (Cutler and Bengtson 1974; Wadsworth and Freeman 1983). These findings actually

supported Mannheim’s notion of the “generation unit,” a concept that receives convincing

empirical support and theoretical development in this literature precisely because it presupposes

an interaction between cohort and some other variable, like political ideology or educational

attainment (Laufer and Bengtson 1974; Wuthnow 1976). Demartini (1985) insightfully argues

that generation units do not presuppose youth rebellion against their parents, and that distinctive

characteristics of generation units can be due to parental socialization.

   44  
Thus, with the exception of studies finding the generation unit concept useful, empirical

studies of generational change failed to produce results consistent with the theory. Even the

generation unit concept appeared to fit comfortably in the burgeoning field of social movement

research. So when David I. Kertzer, in an important (1983) review essay on the generation

concept, reiterated Ryder’s (1965) argument that the term generation should be restricted to a

narrow meaning of kinship descent (parent-child relations), the critique appears to have stuck.

Kertzer argued that the flurry of social science research on generations from roughly 1970 to

1982 suffered because the term was used in four different, but related, ways, signifying “the

principle of kinship descent,” differences among cohorts, stages in the life-course, and the

influence of unique historical periods. In Kertzer’s view, the conceptual confusion was an

impediment, both theoretically and methodologically, to further scholarly understanding of

generational processes broadly defined. Importantly, in arguing that the meaning of generation

should be restricted, he insisted that the distinction is merely terminological, required for greater

analytical precision, and that it should in no way be viewed as a limit on sociological inquiry.

Evidently inspired by Ryder’s (1965) eloquent functionalist metaphor, Kertzer writes,

“Generational processes will remain of great importance to sociology, for they are at the heart of

the social metabolism…. What is crucial to the future of such study, though, is that the

generational processes be firmly placed in specific historical contexts—i.e. that they be analyzed

in conjunction with the concepts of cohort, age, and historical period” (p. 143).

Kertzer’s argument appears to have resonated, at least to a degree, because scholarly

interest in “generations” as a socio-cultural phenomenon—and its associated muddled usage—

largely disappeared only a few years later, with a few prominent exceptions (Delli Carpini 1986;

Jennings and Zhang 2005; Schnittker, Freese and Powell 2003; Schuman and Scott 1989; Scott
   45  
2000). Even though Matilda White Riley used her 1986 presidential address to the American

Sociological Association to argue for integrating the insights of aging, life course, cohort

succession, and structural change in a single theory of the sociological significance of aging, the

large theoretical ground covered by the generation concept was not the subject of much scholarly

research (Riley 1987; Riley and Riley Jr. 1999). After 1985, the generation concept was used

predominantly by scholars of aging and the life course (e.g. Hareven 1994). Studies of how

society changed over time because of different attitudes, values, and orientations of successive

cohorts were carried out under the rubric of “cohort analysis,” not generational analysis (e.g.

Alwin 1990; Brewster and Padavic 2000; Brooks and Bolzendahl 2004; Loftus 2001; Treas

2002; Wilson 1994). And scholars used quantitative techniques in increasingly sophisticated

ways to distinguish among the various types of age, cohort, and period effects (e.g. Alwin and

Krosnick 1991; Ciabattari 2001; Harding and Jencks 2003; Weil 1987). In each of these areas of

research, impressive gains have been made in our understanding of how society changes over

time, and Kertzer’s prescription appears to have been successful.

Other Academic Studies of Generations

Despite the relative dearth of sociological research on the broad, socio-cultural generation

concept, scholars in other disciplines and public intellectuals continued to publish books on

generations and generational change, using the broad, socio-cultural meaning of the word.

Indeed, outside the domain of peer-reviewed research, discourse and writing about generations to

indicate how different cohorts appear to think and act differently as a result of how society

changes over time continued unabated. The terminological clarification that appeared to be so

productive in the discipline of sociology had no effect on how the word “generation” was used in
   46  
popular culture. Even sociologists themselves continue to use the word “generation” in its broad,

socio-cultural sense in informal contexts.

In one sense, this illustrates the cultural resonance and power of the word “generation,”

because its meaning and usage has been almost entirely unaffected by academic efforts to dictate

its appropriate boundaries. In another sense, this illustrates the unintended consequences of the

broad generation concept’s disappearance from sociological research. Far from solving the

problem, the elimination of the socio-cultural meaning of generation from sociological research

may have exacerbated it, because the term continues to be used in the absence of rigorous

scrutiny by the group of people best situated to make the term theoretically valid and analytically

useful: social scientists. As a result, academic and non-fiction works on “generations” in the last

two decades continue to be plagued by the same persistent problems, and cultural stereotypes

that are little more than selective distortions of empirical reality carry the mantle of generational

research in American scholarly literature.

First, academic works on generations produced by professors of sociology, history, and

English are frequently focused on what Mannheim called “generation units,” rather than any

large or representative section of a cohort of people, as the word “generation” implies. Wohl’s

(1979) study of “the generation of 1914” is a study of a handful of European intellectuals whose

writings were influenced by World War I. Similarly, Wyatt’s (1993) study of “the Vietnam

generation” is a work of literary criticism of writers whose works bear the imprint of the turmoil

of the late 1960s. Sociologists Whalen and Flacks (1989) have produced a detailed account of

how the lives and attitudes of countercultural activists from the 1960s evolved over time, though

to describe this group of activists as “the sixties generation,” as the authors do, is misleading

both about the scope of the book and about people’s experiences of the 1960s. While the books
   47  
have merit in their own right, they perpetuate a common problem of generational analysis in that

they misrepresent and caricature a much larger cohort of people by focusing only on a small and

extraordinary sub-section of that cohort. To draw any conclusions about the “actual generation”

from the experiences of these generation units would be unwarranted.

A second type of scholarly and popular book about “generations” have been written

about cohorts that span 15-25 year periods. Premised on the understanding that generations are

defined by the birth years of people during roughly fixed intervals of time, this type of study

attempts to define the collective personality of each generation/cohort that distinguishes it from

past and future generations/cohorts. These works also tend to result in the stereotyping and

caricaturing of large groups of people according to a small sub-section of each cohort, because

the authors use some combination of logical fallacies to make inappropriate generalizations of

the entire cohort based on a handful of isolated observations. Strauss and Howe’s (1991)

ambitious attempt to divide over 400 years of “American” history into 18 “generations” is a

sweeping caricature of the “peer personalities” of the American populace based on carefully

selected historical events and celebrities. The authors painstakingly work to fit data to the theory,

and they have produced subsequent popular books on “the Thirteenth Generation” (Generation

X) and “the Millennial Generation” that paint these two cohorts as disaffected and apathetic, in

the former case, and civic-minded and optimistic, in the latter case (Howe and Strauss 1993;

2000). Predictions about the Millennial Generation come from other quarters as well, such as

from an English professor fearful of the illiteracy of today’s youth (Bauerlein 2008) and from a

psychology professor convinced of the self-centeredness of the “me generation” (Twenge 2006).

In all cases, the authors are apparently unconcerned about the lack of connection between data
   48  
and theory and the implicit assumption that all Americans are homogeneously white and middle-

class.

Taken together, these academic and popular books on generations perpetuate the

problems that have plagued the generation concept throughout the Twentieth Century: the

multiple definitions of the term, the theoretical richness of the concept combined with the lack of

empirical evidence to support it, and the perpetuation of stereotypes based on selective

representation of facts. If these works were mere marketing strategies, they would not be cause

for concern; but they are produced by professors and other intellectuals and sold to the public as

non-fiction rather than science-fiction (they are, after all, mostly fantastical imaginations of a

society grounded loosely in a handful of scientific facts). In the absence of rigorous sociological

engagement with the broad generation concept, such works carry on the mantle of legitimate

research on generations, while simultaneously spreading misleading and potentially harmful

stereotypes of whole cohorts of people. If the “generation” concept is ever to be fruitfully

theorized and incorporated into legitimate social science, renewed scholarly engagement with

generational theory and its persistent problems is warranted.

Cultural Turn in Generational Theory

In the last two decades, there has been a revival in generational theorizing by a handful of

scholars outside the United States. Inspired by the cultural turn in the social sciences, these

scholars have brought a more explicitly cultural perspective to bear on Mannheim’s original

formulation of “The Problem of Generations.” Scholars have thus begun to theorize the

generation concept in Bourdieuian terms of “habitus, hexis, and culture” (Eyerman and Turner

1998), “cultural fields” (Gilleard 2004), and in terms of “cultural circles” (Corsten 1999). The
   49  
Bourdieuian influence can be seen when Eyerman and Turner describe the generation as “a mode

of distinction” (p. 99) and when Gilleard describes the generation as “a distinct, temporally

located cultural field within which individuals from a potential variety of overlapping birth

cohorts participate as generational agents” (p. 114). Drawing from linguistic and discursive

theory, Corsten describes generations in terms age-groups that share rules and vocabularies for

“interpreting and articulating” their view of social reality (p. 262).

Taken together, this theoretical revisioning is essentially an updating of Mannheim’s

“generation as an actuality” in the language of cultural sociology. If a generation as an actuality

is “constituted when similarly ‘located’ contemporaries participate in a common destiny and in

the ideas and concepts which are in some way bound up with its unfolding” (Mannheim 1952

(1928), p. 306), cultural sociologists might interpret this to mean that this group of people shares

a common cultural construction of reality (Berger and Luckmann 1980), deriving from their

shared temporal and social location. To occupy the same “actual generation” means that

members are part of the same cohort and also occupy similar positions in the social structure,

such that they develop shared cultural repertoires. Those shared repertoires are then manifested

in particular attitudes and beliefs, modes of discourse, embodied styles, exclusionary practices,

and collective memories. Like Mannheim’s actual generation, to exhibit such common cultural

practices does not mean that the individuals all know each other or that they all agree with each

other; but they occupy similar temporal and social positions such that their cultural repertoires

are similarly constituted.

To distinguish this cultural notion of generations from the kinship-descent definition of

“generation” and from “cohort,” I follow the terminology of Esler (1984) and Pilcher (1994),

who describe this as the “social generation” concept. I define the “social generation” as the
   50  
cultural and social psychological process through which groups of people, defined

intersectionally by cohort and social location, encounter a particular configuration of social

structures and in turn, typically in young adulthood, develop a particular cultural repertoire that

they use in the further elaboration of attitudes and actions.

The social generation concept is therefore similar to other collective concepts, such as the

“collective consciousness” (Durkheim 1984), “collective mentalities” (Esler 1984), and the

“lifeworld” (Habermas 1987). These concepts connote the macro-micro relationship between the

social world in which an individual lives and their cultural repertoire, the ways in which social

structures and cultures in which an individual moves shape their discourses, values, attitudes,

beliefs, and practices. The idea of the social generation, in particular, calls our attention to the

additional ways that time and historical period shapes social structures and individuals’

experiences within them.

Empirical research drawing on this cultural view of generations shows that the social

generation concept is useful for analyzing age-related differences in collective memories,

identities, and perceptions of society. For example, research in collective memories shows that

people of all ages, when asked to name important events in the past 50 years, tended to name

events that occurred during adolescence or early adulthood and that were especially important in

the area of the country in which they lived (Griffin 2004; Jennings and Zhang 2005; Schuman

and Scott 1989). Other works have shown that people’s memories and understanding of the past

are shaped by both their subjective experiences and the social context of the present (Roberts and

Lang 1985; Schwartz 1996). Several types of political and collective identities have also been

shown to be affected by prominent events or societal trends that occurred during people’s

formative years, and that those identities remain fairly stable over time (Alwin and Krosnick
   51  
1991; Hout and Fischer 2002; Schnittker, Freese and Powell 2003; Weil 1987). Finally, people’s

perceptions of age-differences in society (Edmunds and Turner 2002; Scott 2000; Vincent 2005)

and in the workplace (Down and Reveley 2004; McMullin, Comeau and Jovic 2007) shape a

variety of values, attitudes, discourses, and identities.

Methodologically, the cultural turn in sociology has provided new ways of utilizing

qualitative methods to show how discourses and practices are indicators of the ways that

subjective worldviews are shaped by social structures. Qualitative interviewing, especially using

vignettes, hypothetical scenarios, memory probes, and comparisons, allow the interviewer to

measure the respondent’s understanding of the world as it is refracted through life experiences,

values, beliefs, and attitudes. For example, Swidler (2001) used in-depth interviews to show how

people draw from different cultural understandings of “love” in response to different situations.

Similarly, Lamont (1992) compared interviews of American and French upper-middle class

individuals to show how symbolic boundaries are constructed in different societies. Focus group

interviewing has also been used effectively to gain insight into the shared values and cultural

understandings of a group because they approximate the communicative contexts of people’s

everyday lives (Bloor et al. 2001; Delli Carpini and Williams 1994; Lunt 1996). For example,

Gamson (1992) used focus groups to show the different cognitive resources—experiential

knowledge, common sense, and media discourse—that people used to make sense of social

issues.

In sum, the cultural turn in sociology has breathed new life into the generation concept,

both theoretically and methodologically. But can a reformulated social generation concept

overcome the persistent difficulties associated with generational analysis? In particular, does the

social generation concept provide the analytic insight into empirical data that can match its broad
   52  
theoretical appeal? A strong empirical test of the social generation concept is warranted to

determine whether the concept should be rescued from or relegated to the dust-bin of social

science research. Does the social generation concept add to or detract from our ability to analyze

and understand social reproduction and social change? That is the ultimate theoretical

significance of this dissertation.

In this study, I show that the social generation concept can be analytically useful for

helping scholars explain cohort effects and social change if the concept is conceptualized and

operationalized in ways that are consistent with Mannheim’s notion of “generation as an

actuality.” First, social generational analysis must not use simple cohort measures, juxtaposed

with other competing variables, like educational attainment. Rather, social generational analysis

must be intersectional, taking into account how people’s cultural repertoires are shaped

simultaneously by cohort location and social location within the cohort. Comparisons of a

dependent variable must be made simultaneously across and between cohorts in order to

accurately measure the ways in which repertoires are shaped by a person’s encounter with

historically-specific social structures. This operationalization of the social generation concept

overcomes persistent difficulties associated with generational analysis because building

intersectionality into the concept limits the danger of stereotyping an entire cohort based on one

small subgroup of the cohort and it does not treat variables other than “cohort” as competing

variables.

Second, social generational analysis must be relational, not only because distinctive

social generational patterns must always derive their distinction in relation to those of other

social groups, but also because the concept implies a dialectical, macro-micro relationship

between social structural change associated with historical periods and individuals’ experiences
   53  
and constructions of those changes. This formulation of the social generation concept requires

that we take into account relationality at the macro level of analysis (how social structures

change in different historical periods), at the micro level of analysis (how the cultural repertoires

and practices of some social groups are different from those of other groups), and at the

exchange between macro and micro levels of analysis (how micro-level repertoires and practices

both reflect and create macro-level social structural change).

Finally, the social generation must be understood as a process or collection of processes

rather than as a social group. One of the continuing shortcomings in both generational and cohort

research has been the inadequate differentiation between the groups of people and the cultural

and social psychological processes that are presumed to account for cohort differences among

those groups. By concentrating the analysis on the social generational processes of the formation

and articulation of cultural repertoires, rather than on cohorts or on subgroups of cohorts, we

gain both theoretical and analytical precision on the relationship between individuals and their

social contexts that has too often been simply taken for granted.

By operationalizing the social generation concept in this way, I build on Mannheim’s

“generation as an actuality” and the cultural turn in sociology in order to demonstrate empirically

the analytic power of social generational processes in explaining patterns of social reproduction

and social change. While the theory has always been central to our sociological understandings

of cohorts, aging and the life course, and political dynamics, its empirical application has

typically failed to meet expectations. This study aims to show that the social generation concept,

when properly formulated, is an essential tool, empirically as well as theoretically, for scholars

who seek a richer, more thorough knowledge of how and why societies change through cohort

replacement.
   54  

Synthesis

In the analysis that follows, I use the case of same-sex marriage to show how social

generational change occurs with respect to a controversial issue and how social generational

processes can account for cohort-related differences in attitudes. Empirical research shows that

attitudes about same-sex marriage vary by age and appear to be liberalizing due to the movement

of cohorts through the population. Empirical research also shows that changing attitudes about

same-sex marriage are plausibly linked to changing cultural constructions about homosexuality,

gender, marriage, and family. And finally, empirical research shows that the social

psychological processes that are presumed to account for cohort differences are occurring. What

this dissertation shows is how these three distinct areas of research are related in complex,

dialectical processes of social change between the macro social structural level and the micro

cognitive and cultural levels of analysis.

This dissertation makes three theoretical contributions to sociology. First, I use

qualitative interview methods to shed light on the cultural foundations of people’s attitudes about

same-sex marriage. This study complements public opinion studies on homosexuality and same-

sex marriage, and it complements studies of social movement activity and discourse about same-

sex marriage, because it shows how people draw from various aspects of their belief systems—

life experiences, cognitive beliefs, affective attitudes, and moral values—in order to talk about

same-sex marriage. While there appear to be high quantitative correlations in the survey data

among these aspects of people’s belief systems, in qualitative interviews, they fail to cohere into

simple discourses of support and opposition as the culture wars thesis might predict. Many

informants pragmatically negotiate tensions and contradictions within their belief systems in
   55  
order to talk about same-sex marriage in ways that are neither supportive nor oppositional. This

study therefore shows that the cultural foundations of attitudes about same-sex marriage are

more complex than quantitative surveys have suggested.

Second, by comparing the ways that two separate cohorts talk about same-sex marriage,

and by comparing the similarities and differences in the two cohorts’ belief systems, this

dissertation sheds light on the sociological significance of age-related differences in attitudes

toward same-sex marriage. Although indirect evidence suggests that these age-related

differences are cohort differences, there has been no direct attempt that I know of to account for

this difference in attitudes. In the absence of longitudinal data, there is no way to definitively

distinguish an age from a cohort effect; but I use both quantitative and qualitative analyses of

two different datasets to argue that this is most likely a cohort effect. This has implications for

our understanding of the foundations of attitudes about same-sex marriage, and it allows us to

make predictions about future trends in attitudes about same-sex marriage.

Finally, this study allows us to assess the merits of the social generation concept as an

analytical tool for helping scholars understand cohort effects, social reproduction, and social

change. The case of same-sex marriage is an ideal case study for studying social generational

processes because the issue ties together the broader issues of homosexuality, marriage, family,

and gender—all of which have undergone significant changes in American society since 1970 or

so. These changes have occurred so rapidly that in the space of a single (kinship-descent)

generation, we can expect to see important differences in how parents and children talk about

these issues. By first measuring the cultural foundations of attitudes about same-sex marriage,

and then comparing similarities and differences between the cohorts, we can use empirical data

to examine how social generational processes operate. Theoretically, the social generation
   56  
concept has the potential to help social scientists understand the cultural and social-psychological

processes that drive cohort replacement as a mechanism of social reproduction and social

change. When conceptualized and operationalized carefully, the theoretical potential of the

concept may be finally realized through empirical analysis.

   57  
Chapter 2: Methods

Introduction

I use two sets of data in this dissertation to address the theoretical agenda outlined in the

previous chapter. First, I conduct a secondary analysis of a 2003 public opinion survey,

conducted by the Pew Research Center for People and the Press, in order to estimate the

importance of different demographic, contextual, and attitudinal variables in predicting attitudes

about same-sex marriage and civil unions. I also use this quantitative analysis in order to explore

the associations between different age and cohort measures and attitudes about same-sex

marriage. I describe the data and methods for this quantitative analysis in detail in Chapter 3. For

the time being, I will only discuss the importance of this analysis in the larger context of the

dissertation.

This quantitative analysis is important for four reasons: First, analysis of this data, which

is representative of Americans with a telephone living in the United States over age 18, provides

a generalizable model of the foundations of attitudes about same-sex marriage. This dataset in

particular is useful because it includes a larger set of measures of attitudes and beliefs about

homosexuality than most public opinion studies, so it can provide greater insight into the cultural

foundations of attitudes about same-sex marriage than many public opinion studies. Second, this

analysis provides an important starting point for in-depth qualitative analysis of my interview

data, because it shows what associations are strongest and what relationships can be most

fruitfully analyzed through qualitative methods to deepen our understanding of the foundations

of attitudes about same-sex marriage. Third, this quantitative analysis provides an external

validity-check on the results and conclusions from the qualitative analysis of in-depth interviews.
   58  
Finally, because the public opinion data (unlike my interview data) includes adults of all ages in

the sample, the quantitative analysis provides a more rigorous test of how different cohort

measures are related to attitudes about same-sex marriage.

The second set of data analyzed in this dissertation is comprised of 97 in-depth interviews

with college students and their parents in northern Illinois. I conduct a qualitative analysis of this

interview data in order to provide deeper insight into the social and cultural foundations of

attitudes about same-sex marriage than is currently available in the literature. Specifically,

qualitative analysis of the individual interviews shows how people discursively construct

opinions about same-sex marriage in a specific conversational context, by drawing on elements

of their cultural repertoires, such as their attitudes, values, beliefs, and life experiences relating to

marriage, family, sexuality, and gender. I also compare the interviews of matched parent-child

pairs in order to show the similarities and differences in discourses used by two cohorts of

Americans in order talk about same-sex marriage.

The qualitative interview analysis is important because it is not enough to know whether

or not someone supports same-sex marriage and what demographic or personal characteristics

are related to that opinion. In order to understand the patterns of support and opposition to same-

sex marriage, scholars must understand what same-sex marriage means to people in the social

and cultural context in which they live. This includes understanding people’s life experiences,

knowing something about their networks of friends, families, and acquaintances, and learning

what they think about homosexuality, marriage, gender, religion, and politics. Though

quantitative research can show what relationships exist between same-sex marriage attitudes and

other variables, qualitative interviews can show how people discursively construct their opinions

and thus provide deeper insights into those relationships.

   59  
In the remainder of the chapter, I describe the methodology and epistemology used in

collection and analysis of the qualitative interview data. I first describe the site selection and

sampling of students and parents. Second, I describe the epistemological and methodological

orientation of the study, and I describe the context in which data collection took place. Finally, I

describe the procedures of data analysis and comment on the generalizability of the results.

Site Selection & Recruitment of Students

Qualitative data for this research were collected through semi-structured, qualitative

individual interviews with college students and their parents in northern Illinois. The sampling

strategy was loosely based on Glaser and Strauss’ (1967) theoretical sampling technique. Both

site selection and recruitment of students were guided my assessment of the requirements for

addressing the theoretical agenda outlined in the preceding chapter. Additionally, I modified my

recruiting techniques mid-way through the data collection period in order to attract participants

with backgrounds and viewpoints that were needed to obtain adequate variation for theory-

building.

I recruited participants for this research by putting up advertisements at two college

campuses: Northern Illinois University (NIU), a four-year public university in DeKalb, and Rock

Valley College (RVC), a community college in Rockford. Northern Illinois was chosen as a

research site for three reasons. First, Illinois has no state-level constitutional ban on same-sex

marriage, meaning that the Illinois statute banning same-sex marriage could be challenged as

unconstitutional, as it was in Massachusetts and Iowa. In other words, for people in Illinois, the

legalization of same-sex marriage is a real political possibility (or threat). In states, like

Wisconsin, where same-sex marriage is constitutionally banned, a person’s opinion about same-
   60  
sex marriage is irrelevant to the political possibilities of its legalization; I wanted to conduct the

interviews in a place where engaging in conversation about same-sex marriage would have

practical relevance for the participants.

Second, northern Illinois was chosen as a research site for geographical and cultural

reasons. The student bodies of NIU and RVC are composed primarily of people from northern

Illinois, a relatively-confined geographical area that still includes a cross-section of American

residential patterns: farming communities, small towns, a mid-size city, a large urban metropolis,

and suburbs. Thus, I was able to limit the socio-cultural context in which participants grew up to

the upper-Midwest while still capturing the wide variation in residential environments that is

typical of the United States as a whole. While it is not true that the cultural environment in

northern Illinois is in any way typical or representative of the U.S. population as a whole, people

in this geographical region live in a wide variety of cultural environments, from liberal to

conservative, religious to secular, urban to rural, working-class to upper-class.

Third, these research sites were chosen for demographic reasons. Demographically,

students come from a wide variety of socio-economic, ethnic, religious, and political

backgrounds. Both of these colleges attract students from higher socio-economic backgrounds

who want to save money on higher education or who want to remain close to home; similarly,

both of these colleges attract students from lower socio-economic backgrounds because of low

tuition rates, night classes, and programs like NIU’s CHANCE program for students who would

ordinarily be denied college admission due to their academic background. As a result, the

students who participated in the study, as a whole, vary widely according to socio-economic

status, ethnicity, religious and political beliefs, parents’ occupation, and educational histories.
   61  
Of the 65 students who were interviewed, 29 attended Rock Valley College and 36

attended Northern Illinois University; 40% were women; 72% were white non-Hispanic (there

were 8 Hispanic whites, 7 African-Americans, and 3 students of mixed European and Middle

Eastern or Asian ethnicity); and the median age was 21 (34% of students were age 23 or older).

In terms of how the students described their political beliefs, 34% identified themselves as

having liberal or left-of-center political beliefs and 22% identified themselves as having

conservative or right-of-center political beliefs (4 identified as libertarian, 6 identified as

moderate, 4 identified as angry or disillusioned toward politics, 3 identified as independent, and

11 either did not know how to classify their political beliefs or gave some other response). In

terms of religious identification, 37% identified themselves as some type of Protestant on non-

denominational Christian, 29% identified as atheist, agnostic, or said they had no religious

beliefs, 17% identified as spiritual but not religious, and 11% identified as Catholic (there were

also 1 Muslim, 1 Jew, and 2 individuals who identified with multiple religions). It should be

noted that, although I purposefully avoided asking people about sexual identities or behaviors,

one student identified as gay and three students identified as bisexual.

In order to recruit students for this study, advertisements were posted on bulletin boards

in the student centers, libraries, gyms, and prominent classroom buildings on both campuses. The

posters (Appendix A) advertised, “Get Interviewed, Earn $30,” and contained a brief description

of the research and the list of eligibility criteria. The study was described as “a University of

Wisconsin study about what people in America think about current social and political issues.” It

specifically mentioned “marriage, sexuality, family, and politics” as topics for the interview, but

I was careful to never mention the issue of same-sex marriage before the interview. I did not

want participants to know that I was interested specifically in the topic of same-sex marriage
   62  
because of the possibility that it would affect the ways that they talked about topics earlier in the

interview. From the ad, students were able to tear off tabs containing the phone number and

email address of the researcher. In order to be eligible, students had to be between the ages of 18

and 30, currently enrolled in a degree program at NIU or RVC, born in the U.S., have attended

high school in Illinois, and have one parent between the ages of 45 and 63 (born 1945-1963) who

was born in the U.S. and currently living in Illinois, Wisconsin, Iowa, or Indiana.

Posters were first put up on August 28 and September 4, 2008. At both colleges, there

were delays associated with getting approval to put up the posters. At NIU, approval was

obtained in a few hours. At RVC, the initial round of posters was taken down, pending approval

of the project by the college’s Institutional Review Board, and advertisements were not allowed

to be posted until October 21. Initially, 30 posters were put up at NIU and 25 at RVC. Additional

posters were put up as needed throughout the data collection period. At each college, I also tried

distributing small hand-bills to passing students, but the method appeared to be less efficient at

recruiting participants, so I abandoned it.

Overall, I was contacted by 97 students, and I conducted interviews with 65 of them.

Among those with whom I did not conduct interviews, five students did not meet the eligibility

criteria, and seven students cancelled our scheduled appointment and never returned follow-up

inquiries about rescheduling. The rest of the students simply never followed through with their

initial inquiry, and I was never able to determine a reason for their not participating.

Although I did not routinely ask students how they found out about the study, there is a

substantial percentage of students in my sample who found out about the study through social

                                                                                                               
1
This delay had a noticeable impact on my recruitment rates, and it especially impacted my
ability to obtain interviews with parents, for reasons that I discuss below.
   63  
networks rather than from the posted advertisements. For example, an unusually large number of

students who were in NIU’s “Atheists, Agnostics, and Freethinkers” club participated in the

study in September. Similarly, in the spring, I conducted an unusually large number of

interviews with students who had attended the same small, private Christian high school. I also

was told by some students that friends of theirs had told them about the study, and some students

even contacted me together. I did interviews with at least one boyfriend-girlfriend pair, and I did

interviews with a pair of siblings.

The fact that social networks were important in shaping the composition of my sample

turned out to be an asset, rather than a drawback. Because my sampling procedure was never

designed to be statistically representative in any way, this uncontrolled recruitment technique

resulted in richer data in some ways. I was able to use the social connections that I knew of to

gain additional insight in the analysis of my data, especially regarding the role of networks of

friendship and acquaintance in shaping people’s attitudes and discourses. In some instances, I

was able to observe the possible influences that friends and peers had in shaping one another’s

views, simply by comparing the interviews of individuals whom I knew to be associated.

There were some additional, noteworthy effects of my recruitment procedure on the

composition of the sample. First, the $30 incentive that I offered students in the advertisement in

exchange for participation made a clear difference in sample composition. This was my goal,

because I wanted the sample of participants to include people who did not feel strongly about the

issues that were advertised. A number of students admitted openly that they were only doing the

interview for the money. Some students knew very little about politics and talked very little

about homosexuality and same-sex marriage when prompted, a sign that they did not agree to an

interview because they were passionate about the subject.

   64  
In retrospect, it is possible that the financial incentive was too much. Some students were

clearly suspicious of the prospect of being paid a higher “hourly wage” than they had ever earned

before, simply by letting someone interview them. Ultimately, the institutional logo of the

University of Wisconsin and the official paperwork of the “Informed Consent” procedure

appeared to put them at ease. Additionally, a number of students told me that their parents had

warned them about a scam or tried to convince them not to participate. This monetary incentive

thus played a non-trivial role in sample composition. In the interview itself, the payment also

likely affected the interactional dynamic and the resulting data in theoretically important ways,

which will be discussed below.

Second, the composition of the sample was likely affected by the mention of “sexuality”

and “politics” in the advertisement as topics to be discussed. Though I do not have any direct

evidence of this, my impression is that such words may have scared off some potential

participants. Many people may have been reluctant to participate in the study because the

advertisement could easily have been read as a study about people’s own sexuality and their

personal sexual experiences, like the famous Kinsey surveys. Female students might have been

justifiably wary of agreeing to an interview with an unknown male in a position of power about

“sexuality” as well; this could explain the relative underrepresentation of women in my student

sample (43% of inquiries were from women, and 40% of interviewees). As for “politics,” the

word has such negative connotations for many Americans that they might have avoid talking

about it at all costs.

Recruitment of Parents
   65  
I obtained interviews with parents by asking the student, at the end of the interview, for

permission to contact one of their parents. Upon completion of the interview, I handed each

student a form requesting permission to contact an eligible parent to ask them for an interview

(Appendix B). I explained that I was interested in learning about the similarities and differences

in attitudes between students and their parents, and I emphasized that the interview would be

identical to the one that I had just administered with the student and that the student’s answers in

the interview would not be discussed with the parent. I asked for the name of the parent with

whom the student was closest in order to maximize the similarities between students and parents

and in order to maximize the possibility that the parent would agree to an interview. Some

students volunteered that their other parent would be more willing to do the interview, in which

case I asked for their name instead.

Most students had no reservations about giving me permission to contact their parent and

providing one or more pieces of contact information. Thirteen students refused to give me

permission to contact a parent for various reasons. Of these, four students said their parents

would not be willing to do the interview because of the subject matter or for some other reason;

four students said their parents were unavailable because of work or illness or area of residence;

and three students said their parents would not want their contact information being given to a

stranger. The other two simply refused and offered no explanation. I provided each student who

gave me permission to contact a parent a statement about the study that they could use if they

wanted to tell their parent about it. I asked each student not to tell the parent the specific topics

that were discussed in the interview, explaining that I would not want the parent to bias their

answers to my questions.
   66  
When contacting parents, I had a script prepared for phone, voice mail, and email that I

used to explain the study and ask for their participation (Appendix C). If I did not reach the

parent immediately, I made at least three attempts to get an answer about whether or not they

wished to participate. If I received no reply after three messages or emails, I interpreted the lack

of reply as a desire not to participate. If the parent did agree to participate, we worked to find a

mutually convenient time and place to carry out the interview. Most interviews with parents took

place either in their home, at the campus where their child went to school, or at a restaurant or

café.

From my observations, two main factors were decisive in shaping whether or not the

parent agreed to an interview. First, simply, was time. Parents who worked more than 40 hours

per week were generally unavailable or unwilling to set three hours aside to be interviewed.

Parents who were unemployed or who had part-time or flexible work schedules were more

available for interviews. Second was the encouragement by the student to participate. I know

from comments made by parents that their child’s endorsement of the interview and their

recommendation that they should participate was the main reason that they agreed to be

interviewed. Thus, even parents with busy schedules made time to participate in the study

because they felt that it was something that their child wanted them to do.

During the fall semester, I conducted interviews between September 4 and December 5.

During that time, I interviewed 39 students and 24 parents. Because of the one-month delay in

recruiting at Rock Valley College and the lack of success with recruiting participants after the

Thanksgiving holiday (perhaps due to pending final exams), I decided to continue interviewing

during the spring semester. On January 12, 2009, I posted 25 new posters at RVC, and I later put

up additional posters at NIU and RVC in an effort to attract participants. In the initial period of
   67  
data collection, I had been relatively unsuccessful in recruiting students with conservative

political and religious beliefs, so I also sent email versions of the advertisements to the NIU

College Republicans and the NIU Intervarsity Christian Fellowship. During the spring semester,

I interviewed 26 students and 8 parents between January 19 and April 13.

Of the 32 parents I interviewed, 20 were parents of a NIU student, 11 were parents of a

RVC student, and 1 was a parent of a student at both schools; 69% were women; 81% were

white non-Hispanic (3 whites had significant Native American heritage in their background; 2

parents were Hispanic; 1 was African-American); and the median age was 50. In educational

attainment, 38% have Bachelor’s or Master’s degrees, 31% have an associate’s or professional

degree, and 31% have a high school diploma only. In terms of political identification, 38%

identified as liberal or left-of-center and another 34% identified as conservative or right-of-center

(4 identified as moderate or independent, 2 described themselves as being disgusted with

politics, 1 identified as libertarian, and 2 didn’t know). In terms of religious beliefs, 41%

identified as Protestant or non-denominational Christian, 25% identified as Catholic, and 22%

identified as spiritual but not religious (3 identified as atheist, agnostic, or with no religious

beliefs; 1 was Jewish).

There was a significant decline between the fall and spring semesters in the success of

recruitment efforts of parents. This decline was important because I would have had 25% more

parents in my sample if the fall recruitment rate of parents had continued into the spring. In the

fall, I conducted interviews with parents of 62% of the students I interviewed, while in the

spring, I conducted interviews with parents of only 31% of the students interviewed. There was

no obvious change in my recruitment techniques or in the types of students I interviewed that

would account for this decline. The decline in the parent interview rate occurred at both schools,
   68  
and among both liberal and conservative students. The only changes I made to the recruitment

techniques occurred after I observed the decline in February; they were only minor changes in

the wording of the email and phone scripts that were intended to make my request for an

interview seem less intimidating and more personal.

I suspect that the difference between the fall and spring semester parent interview rates is

best accounted for by the 2008 election. It is likely that parents were more likely to be

interviewed during the fall because of the ongoing election campaign and heightened levels of

political attention and interest in politics. In explaining my interviews to parents, I said that it

was “a study about Americans’ feelings about various social and political issues.” During the

election, parents may have been more receptive to such an interview request because of the

perceived legitimacy of wanting to know people’s political opinions or because of the continuous

presentation of poll results in the media. In other words, parents might have been more likely to

agree to my interview during the fall because they imagined it to be part of the normal election-

year polling and opinion-gathering efforts.

It is also probable that parents were paying more attention to politics and social issues

during the election campaign and thus felt that they were more qualified to share their opinions.

During the fall interview period, the candidacies of Barack Obama and Sarah Palin had created

high levels of interest in politics, and extraordinarily high gas prices had given way to a sudden

economic recession. These issues were covered extensively by the mass media and may have

been followed more closely by citizens than “politics as usual.” By contrast, during the spring, as

the political controversies of the nation turned to a variety of economic recovery policies being

debated among politicians in Washington D.C. and a game of wait-and-see on President

   69  
Obama’s performance, the issues may have seemed less dramatic and attention-grabbing. Thus,

during the spring, parents may have felt that they had fewer informed opinions to offer.

This is likely to explain the decline in parental recruitment because many of the students

and parents that I interviewed were nervous or worried that they would not perform adequately

during the interview. Even though I intended the interview to be an informal conversation to

learn of the experiences and views of ordinary Americans, some students and parents viewed the

interview more like an exam or a test. When scheduling interviews, a number of people asked if

they needed to prepare for the interview in any way, and others assured me that they would get

ready for the interview by brushing up on current events and trying to think of intelligent things

to say about the topics that we would be talking about. One parent told me as we were sitting

down for our interview that she had almost cancelled our interview because she didn’t think she

had any informed opinions to offer. After the interviews, some participants told me that they

hoped they had provided opinions that were good enough for me. In sum, the 2008 election

likely increased people’s willingness to be interviewed because it created more political

knowledge and interest than is typical during non-election years, thus resulting in a higher

response rate for that period of data collection.

Methodological Approach to Data Collection

My approach to qualitative interviewing is a communication-theoretic variant of what

Rubin and Rubin (2005) call “responsive interviewing.” Responsive interviewing follows a

theoretical logic that is “interpretive constructionist” rather than positivist, meaning that the

interview data do not provide an unbiased measure of reality but a measure of a person’s social

construction of reality (Berger and Luckmann 1980). In other words, the informant’s stories,
   70  
opinions, and answers to questions are not necessarily objective “facts” that the researcher

assumes are true; rather, they should be understood as the interviewee’s interpretation or

understanding of reality, which is shaped by the person’s social location, cultural repertoire, and

interactional setting.

The responsive interviewing approach is similar to what is broadly referred to as

“intensive interviewing,” but Rubin and Rubin place additional theoretical emphasis on the

personal relationship that is implicated in the interviewer-interviewee relationship. This means

not only that depth of understanding and flexibility are privileged goals of the interview, but also

that the interactional characteristics of the interview are extremely important in shaping the data

that are ultimately collected. The interviewee is affected by the interviewer in a number of ways:

by the fact of being interviewed in the first place, by the beliefs and attitudes that are implicit and

explicit in the questions and topics of the interview, and by the characteristics of personality,

style of speech, and behavior that shape all social interactions.

The previous section suggests how this was true. The qualitative interviews that I

conducted were not unbiased methods for gaining access to a person’s “true” opinions about

various topics; rather, they were peculiar settings of social interaction whose dynamics and

properties exerted an independent influence over the data that were collected. Because my name

bore the institutional backing of the University of Wisconsin, because I was paying individuals

to be interviewed, and because it is very rare for a person to set up to three hours of time aside

just to be asked dozens of questions about their life history and their views on different issues,

the interviews were unusual social situations. As described above, many participants worried

about performing well for me in the interview and felt that they should offer learned, intelligent
   71  
opinions on issues, whether they “had” them or not. It is possible that many participants worried

that I was “grading” their responses to questions, as a professor would on a test.

Interactional Setting of the Interview

A number of additional characteristics of the interactional setting of the interview likely

affected the outcomes of the interviews. First, the interview settings contained a mix of elements

that demarked the interaction as both professional and casual. Professional elements included

signs of the institutional backing of the University and my techniques of data collection. The

institutional backing of the University was evident in the Informed Consent paperwork that each

informant read and signed before the interview and in the exchange relationship of interview for

payment. Because I was affiliated with a prestigious university, and because the interaction was

institutionally sanctioned by paperwork and monetary payment, the interview setting carried an

air of formality.

The techniques of data collection that conveyed the message that this was a professional

encounter included the interview guide that I used, the presence of the audio recorder, and my

practice of taking notes during the interview. I used the interview guide to structure the interview

by reciting a standardized introduction and by drawing from a pre-established list of questions

and topics. I used an audio recorder to record every interview, and I continuously took notes

throughout the interview, so that I would have a written summary of major points that I could use

to facilitate data analysis. These techniques marked the interview as a different social setting

than an ordinary informal conversation.

At the same time, other elements of the interview setting were informal. These include

the location of the interview, my personal appearance, and the conversational style I attempted to
   72  
establish during the interview. Interviews took place in a variety of places, including people’s

homes, workplaces, coffee shops, restaurants, libraries, and student centers. I allowed the

informant to choose the location of the interview, so long as it was a place where we could have

a one-on-one conversation relatively free from disturbances. Thus, the interviews happened

either in public places or in private settings of the informant’s choosing. I did this in an effort to

make informants feel as comfortable as possible during the interview.

I chose to make my personal dress and appearance for the interviews casual but

respectable. I purposefully chose not to wear formal slacks and shirts in an effort to avoid

making an impression that I expected this to be a completely-professional encounter. Instead, I

favored more casual apparel, such as jeans or corduroys, sweaters, and polo shirts (but no t-

shirts). I always attempted to come across as groomed, clean, and respectable, but not

intimidating. I also attempted to avoid giving any impression of my own personal political views

through my dress and appearance; my wife even consented to removing the bumper stickers from

our car during the period of data collection.

I was usually on the lookout for signs of what participants thought of me. As a young-

looking graduate student with a career and a wife, I shared some characteristics with the students

but others with the parents. Many students, especially those with little concept of post-graduate

education, seemed to think I was a full-time professor, though others understood what being a

graduate student meant. Some parents, by contrast told me that I was older than I looked, if the

subject of my age came up. I did not observe any evidence, however, that members of one cohort

were more comfortable talking with me than the other.

Lastly, I tried to establish a “conversational partnership” with the participant that was

informal and friendly in style (Rubin and Rubin 2005). In the introduction that I recited to each
   73  
informant, I told them explicitly that I wanted them to feel comfortable, and I emphasized that

this was a “conversation, not an interrogation.” In asking questions, I did not attempt to use the

same wording in every interview on most questions. Instead, I asked questions in an informal

style that mimicked the informant’s own conversational speech patterns. The transcriptions of

my own questions include the words “like” and “um,” and I frequently altered the question in

mid-sentence, as is common in everyday talk.

During interviews, I frequently interjected stories or information about my own life or

background, in an effort to make the interview more reciprocal and to help the participant feel

more comfortable with me. Some participants heard about where I grew up, my family, or that I

had recently gotten married; and I always attempted to answer questions honestly if they asked

them. The fact that I had gotten married the previous summer was an especially useful bit of

information to share, because sharing this personal information seemed to increase the validity of

my interview in some people’s eyes, and it seemed to make some people more comfortable with

me. I did, however, avoid sharing information or opinions that I thought would be interpreted as

evidence of my opinions on same-sex marriage, homosexuality, religion, or politics.

Because of the mixture of professional and casual elements of the interactional setting,

and because of my conscious efforts to be a partner in conversation, most interviews felt

comfortable as well as thorough. Many people said afterwards that they enjoyed the interview,

asked if they could recommend the interview to their friends, or said that it was much easier than

they had expected. Many students were enthusiastic about telling their parent about the

interview, and several parents told me that their child had recommended that they agree to my

request for an interview. When time permitted, after the interview was over, we often continued

to converse about a variety of topics, related or unrelated to the subjects of the interview.
   74  
Pursuant with the expectations of responsive interviewing, I felt that many interviews contained

interpersonal dynamics that indicated a high degree of trust, honesty, and respect.

This style of interviewing, however, had its drawbacks. Had I devoted less energy to

establishing a conversational partnership with each participant, I would likely have been more

rigorous in my questioning. For example, while transcribing the interviews, I occasionally heard

“golden opportunities” for follow-up probes that I had not noticed while I was actually

conducting the interview. I thereby missed gaining valuable additional insights into the person’s

opinions or life experiences. In addition, I tended to avoid “playing the devil’s advocate,” i.e.

following lines of questioning that would have directly challenged the participant’s opinion. This

was less out of fear of social awkwardness than out of a lack of attention on my part to the

benefits of systematically incorporating this type of questioning into my interview routine. In

retrospect, I should have been more methodical about inserting probes that required participants

to answer objections from the opposing viewpoint. This would not have negatively affected the

interactional setting in significant ways, and the probes would have provided greater insight into

each participant’s worldview than I achieved.

Nevertheless, I have high confidence in the honesty and validity of the vast majority of

my interviews. There were a variety of signs that I had gained the trust of informants and that the

answers people gave to my questions were truthful. A few people were open and honest about

deeply private aspects of the personal lives, things that they said they would not share with

family members or friends. Some people were honest about behaviors and opinions that they

acknowledged were stigmatizing or would make them look bad in the eyes of others. Most

people also answered questions in ways that showed their concern with carefully thinking about

each question before offering an answer. There were frequent, pregnant pauses before people
   75  
answered many of my questions. Sometimes participants elaborated at great length on questions,

and they sometimes appeared to take great pains to formulate answers to questions that did not

come easily to them.

The time period in which data collection took place also affected the interviews in

noticeable ways. The period between September 2008 and April 2009 was a momentous time in

the United States. During the fall semester, isolated worries about the economy and high gas

prices gave way to a worldwide economic recession and widespread fear of a second Great

Depression. At the same time, the Presidential election campaign featured an inspirational

African-American candidate whose charismatic and idealistic rhetoric appealed to young voters

and who capitalized on widespread disillusionment with the Republican Party. Barack Obama’s

campaign, his election victory, and his subsequent inauguration were closely watched, and it

made discussion of politics an easy topic of conversation during the interviews.

Perhaps more relevant to this study, however, is the fact that the issue of same-sex

marriage was in the news a number of times during this period. Prior to the beginning of my

study, only Massachusetts and California allowed same-sex marriage. During the period of data

collection, three states were added to this list. On October 10, 2008, the Connecticut Supreme

Court ruled that civil unions did not grant same-sex couples equal treatment under the law and

that same-sex couples have the right to marry. On April 3, 2009, the Iowa Supreme Court

overturned the state’s statute against same-sex marriage, effectively legalizing it. Four days later,

the legislature in Vermont successfully overrode the governor’s veto on a same-sex marriage bill,

making Vermont the first state in the United States to legalize same-sex marriage through

legislation rather than court order. The progress of this bill through the Vermont legislature had

been making headlines for weeks.

   76  
However, the ballot measures about same-sex marriage during the 2008 election seemed

to influence my interviews the most. During the election, voters in three states—Arizona,

California, and Florida—passed constitutional amendments banning same-sex marriage, and

voters in Arkansas passed a ballot measure banning gay couples from adopting children. Of these

initiatives, Proposition 8 in California was the most closely watched because it marked the first

time that voters made same-sex marriage illegal in a state where it had previously been legal.

The political mobilization around this initiative and the dramatic implications of the vote made

this issue national news for weeks during October and November, and as a result, many

informants brought up California during the interview. Several people knew someone who was

personally affected by the issue of same-sex marriage in California, and a number of college

students reported encountering this issue in a class—most often in the form of a paper they wrote

or in a classroom debate activity.

This affected my interviews, not only because many people had more to say about the

issue than they might otherwise have had, but also because some informants’ views were in flux

because of the controversy. This was especially true for college students, many of whom had

never been exposed to the opposing point of view before or had never thought much about the

issue in the first place. Had I interviewed some students only a couple of months earlier, I would

have recorded different attitudes and opinions than I did. Some students acknowledged this

explicitly, while for others, it can be inferred.

Rather than “contaminating” my data, the presence of such momentous events during the

period of data collection provide important insights into the processes by which people construct

and modify attitudes and beliefs. As I describe in the remainder of this chapter, I do not interpret

my data to be indicators of a person’s “true” attitudes or beliefs, nor do I attempt to compare

   77  
interviews according to a quantitative logic that accords equal weight to each interview, for the

purposes of statistical generalization. Instead, these data provide indicators of how belief systems

are formed and change over time due to influences such as education, social network

composition, and media exposure. On the issues of homosexuality and same-sex marriage, self-

reported attitude change over time was widespread in my interviews, among both college

students and parents, and usually (though not always) in the liberal direction. Attitude change, no

matter its cause, is normal, and in a qualitative study, it is not an obstacle but a facilitator of

analysis. Thus, the timing of my data collection and the prominence of controversies around

same-sex marriage in the national media proved to be extremely beneficial to the study.

The Cultural and Communicative Processes of Interviewing

I describe my method as a “communication-theoretic variant” of responsive interviewing

because of the kind of data that I intended to collect and the principles I used in construction of

the interview guide. From the interviews, I was interested in more than learning a person’s

interpretations of the world and their personal experiences; I was interested in obtaining

measurements of particular attitudes and beliefs. I intended the interview to be a qualitative,

intensive approach to measuring public opinion, in which specific attitudes and beliefs are

understood not as simple answers to individual questions, but as multi-faceted expressions of a

person’s belief system that are inter-related, context-dependent, and indicative of a person’s

“lifeworld” (Habermas 1987).

Embracing the theoretical insights of the interpretive constructionist perspective does not

require the researcher to entirely reject positivist methodology. I argue that specific attitudes and

beliefs can be measured and analyzed by a researcher, but not as things that are held by
   78  
individuals; rather, attitudes and beliefs are formulated by individuals in particular times and

particular places, in ways that vary depending upon the social context. The measurement of

attitudes and beliefs must be measurements of the cognitive-cultural processes of expression in a

particular context. In this way, the qualitative interview method is not unlike that of the survey

method. All polls, interviews, and questionnaires, whether qualitative or quantitative, are

particular social interactions which themselves have effects on the data that are collected.

Scholars of public opinion have shown that opinions do not exist inside individuals’

brains, like files in a file cabinet, only to be looked for and retrieved when needed. Rather,

individuals create opinions and attitudes in response to specific questions or situations that they

encounter (Zaller 1992). The principles of questionnaire construction, such as item ordering and

wording, would not matter if this were not the case. It is precisely because people construct

opinions differently in response to different social contexts that survey designers and

interviewers must pay careful attention to the order in which different items are presented, the

biases of particular words, the response categories used, and so on. Likewise, the effects of

framing, priming, and agenda-setting (e.g. Druckman and Nelson 2003; Entman 1989; Iyengar

and Kinder 1987; Krosnick and Kinder 1990; Nelson, Clawson and Oxley 1997) can only be

explained if attitudes and opinions are understood to be products of cognitive-cultural

construction instead of things that a person possesses. People draw from the tools in the cultural

“tool-kit” that they have available at that particular time and place in order to meet the demand

of an interviewer to formulate an attitude or opinion (Swidler 1986; 2001).

The social scientific literature on “talk” using qualitative research methods effectively

illustrates how methodologies that are interpretive constructionist in their theoretical orientation

are nonetheless scientific and can produce data that are both valid and reliable to varying
   79  
degrees. After the cultural turn in the social sciences, ethnographic, interview, and focus group

methodologies have all been employed by researchers to study the ways that patterns of talk are

indicative of underlying social contexts and cultural orientations. For example, ethnographies of

political talk in informal groups show how conversations both create and are expressive of larger

notions of collective identity, orientations toward politics, cultural assumptions, and normative

commitments (Eliasoph 1998; Lichterman 1996; Walsh 2004).

Focus group methods also excel at gathering data on the shared values and orientations of

a group, in that they are intended to mimic, in a controlled environment, the ordinary

conversational contexts that people encounter in their daily lives (Bloor et al. 2001; Delli Carpini

and Williams 1994; Lunt 1996). Sociologists like William Gamson (1992) and Andrew Perrin

(2006) have used focus group discussions about controversial political issues to show how

groups draw from various cognitive resources and employ specific argumentative logics in the

process of expressing opinions and imagining solutions about the issues. People may draw from

their stock of experiential knowledge, from media discourse, and from popular common-sense in

discussions about political issues (Gamson 1992). The logic of the claims that people make in

discussion may be classified in a variety of ways, such as pragmatic, ideological, moral, and self-

interested (Perrin 2006).

Press and Cole’s (1999) study of focus group discussions about abortion illustrate how

these methods simultaneously account for the cultural construction of attitudes and still yield

distinct measurements of those attitudes. They found that individual participants’ attitudes about

abortion, as expressed on an individual survey, bore little resemblance to the attitudes they

expressed collectively, in groups. The labels “pro-life” and “pro-choice” did not correspond

closely with the ways in which people talked about the issue, and people often endorsed multiple
   80  
views of the issue. Moreover, participants’ discussions of the issue in the abstract were different

from their discussions of particular instances of abortion, which varied by social class of the

woman and the circumstances in which an abortion was considered. In studies such as these, it is

clear that the insights of interpretive constructionist theory do not preclude researchers from

empirically measuring the processes and resources that people use to construct opinions.

In my study, I drew primarily from techniques of cultural sociologists who used

individual interviews to uncover people’s cultural understandings of the social world. For

example, Swidler (2001) uses qualitative interviews to study how Americans use different

frames and different elements of their cultural repertoires to talk about love and relationships.

Similarly, Lamont (1992) compared interviews of American and French upper-middle class

adults to show how people construct symbolic boundaries differently in different cultural

settings. In studies such as these, the authors do not assume the interview data to be factually true

statements, accepted at face value; instead, the interviews provide insight into the processes that

people use to express opinions and the influence of socio-cultural contexts on people’s discourse.

Thus, the data I collected through my interviews should be interpreted as efforts by

individuals in a professionally sanctioned, economically incentivized, but informal social context

to offer culturally satisfactory responses to conversational stimuli. The particular attitudes and

opinions that are expressed should not be assumed to be “true” in some absolute sense. The

attitudes and beliefs that I measure are manifestations of a person’s social location and cultural

context as expressed through cognitive-communicative processes of opinion-construction. A

similar statement can and should be made for all quantitative and qualitative survey and

interview data. Such a statement does not deny the veracity or accuracy of the data; rather, it

simply requires the researcher to account for three sets of factors that influence the data that is
   81  
collected: the methodological approach, the interactional setting, and the questions and topics of

the interview. These influences must be accounted for in the data analysis and in the process of

generalization as well.

The techniques and characteristics of this variant of responsive interviewing are therefore

suited to the purpose of this study: to uncover the cultural foundations of people’s attitudes about

same-sex marriage. In general, qualitative interviewing, compared to quantitative survey

methods, results in data that are more valid than less reliable because respondents are less

constrained by a small number of answer choices and interviewers are free to use follow-up

probes and different questions to capture the normal complexities and contradictions of people’s

attitudes and feelings. Beyond this, however, a communication-theoretic approach to responsive

interviewing provides more than a different measurement of the same attitudes; it provides

measurements of the influences of social and cultural contexts on discourse. Thus, even if an

informant is “making up” an opinion on the spot; or she is hiding her “true” feelings in order to

avoid negative social sanctions for an unpopular opinion; or he is flat-out lying; the data still

provide information about the person’s cultural assumptions about what is socially acceptable or

what constitutes a valid response to a question. In these ways, the “cultural foundations” of

people’s attitudes can be measured, even without assuming that every attitude or belief that is

expressed is objectively true.

Interview Guide and Interview Topics

The interview guide (Appendix D) contained seven sections. The first three sections have

to do with getting to know the informant. I began each interview by asking the informant to “tell

me a little bit about yourself,” and I asked follow-up questions to ensure that basic demographic
   82  
information was collected for each person. Contrary to traditional survey methods, which ask

demographic questions at the end, I found that asking people simple questions about their ethnic

background, occupation, family, etc. was a good way to allow the informant to get comfortable

in the interview setting. Moreover, it allowed me to learn important information about the person

fairly quickly, which I could use for the purposes of tailoring the interview to each individual.

In the second section, I asked about the informant’s background, focused primarily on

their childhood and adolescence. I asked about the neighborhood they grew up in, family life,

their high school, extracurricular activities, and friendship and dating experiences in high school.

During this section of the interview, I asked each person about their peers’ attitudes about sex,

and I asked about how common it was among their peers to have parents who were divorced or

who came from single-parent families. For some informants, I asked about parents’ expectations

for them, memories of important social or political events, and life after high school. This section

of the interview lasted longer for parents and older students because there were more life

experiences to talk about: military or work experience, marriage, child-rearing, and so on. The

emphasis on the informant’s “formative years” in this part of the interview is important because

there is evidence that these years are particularly important for a person’s development of basic

cultural attitudes and orientations that they will carry with them later in life (Alwin and Krosnick

1991; Schuman and Scott 1989).

In the third section, I asked about each informant’s current media consumption habits in

order to gain some insight into their interests, their cultural tastes, and their extent of exposure to

news and popular culture. This section of the interview allowed me to gain some insight into the

“lifeworld” (Habermas 1987) in which each person lives and the cultural resources that they use
   83  
in formulating attitudes and opinions. I asked about news consumption; television, radio, and

internet use; newspaper and magazine subscriptions; and tastes in music, movies, and books.

The first three sections constituted the majority of the interview, but the “heart” of the

interview is the next three sections. In these three sections, I asked each informant questions that

were designed to require informants to draw from their cultural “repertoire” in order to formulate

a satisfactory answer. Questions were intentionally open-ended, ambiguous, hypothetical,

leading, or contradictory. In all cases, my intention was to collect data on how each informant

interprets the question and how they formulate what they saw as a satisfactory answer to the

question.

In the fourth section, for example, I asked a series of questions about marriage and

relationships. The first question, “What does the word ‘marriage’ mean to you?” is intentionally

open-ended and ambiguous. People variously drew from personal experience, second-hand

information, cultural common-sense, and mass media in order to talk about the meaning of

marriage and “important characteristics of a good, strong marriage.” I also asked questions about

gendered division of labor in marriage, cohabitation, premarital sex, divorce, and media

portrayals of marriage. Follow-up questions about divorce were used to probe for nuances and

contradictions in people’s attitudes about divorce: for example, particular situations in which

divorce would be justified or unjustified. Lastly, I used a leading question to ask what people

thought about some commentators saying that the institution of marriage is in a crisis. People’s

responses to this question not only demonstrated their degree of attunement to political discourse

about divorce and same-sex marriage, but also led into a discussion of same-sex marriage.

The fifth section of the interview is specifically about the issue of same-sex marriage, and

the sixth section of the interview is about homosexuality more generally. However, in practice,
   84  
these two sections were discussed at the same time, because people’s talk about same-sex

marriage frequently centered on their beliefs and moral evaluation of homosexuality. The

questions about same-sex marriage largely required people to use their imaginations to construct

answers. For example, I asked about the reasons that someone might offer for supporting or

opposing same-sex marriage, and I asked a series of questions about what effects legalizing

same-sex marriage would have on society. I asked for their views about civil unions as an

alternative to same-sex marriage, and I asked whether or not they thought same-sex marriage

would be legal in the future.

In the section about homosexuality, I asked people about their moral and cognitive beliefs

about homosexuality, the extent of their personal contact with gays and lesbians, portrayals of

gays and lesbians in the media, memories of their first encounter with homosexuality, and

attitudes about the rights of gays and lesbians. In addition, I asked each informant about

bisexuality and what they thought it meant. For people who were supportive of same-sex

marriage, I used two probes to explore the extent to which their support for same-sex marriage

was premised upon a particular notion of sexuality: I asked whether someone who identifies as

bisexual should be allowed to marry someone of the same sex, and I asked what they would

think about two people of the same sex who were not sexually attracted to each other and who

wanted to get married. Informants’ answers to these questions provided important insights about

their unspoken assumptions about how the meanings of marriage and sexuality are related to the

specific issue of same-sex marriage.

The final section of the interview was a short discussion about the 2008 Presidential

election. It was meant primarily to be a “cool-down” subject of conversation—an easier subject

to discuss than homosexuality and same-sex marriage, but an obvious topic of discussion given
   85  
the time period of the interviews. I asked each informant about how much they have been

following the campaigns, the extent to which people they knew were following the election, their

feelings about various candidates, the issues they thought were important, and their intentions to

vote. For interviews that took place after the election, I modified the questions to ask about the

election in retrospect and about President Obama’s performance in his first months on the job.

This section of the interview provided some insights into each person’s knowledge, interest, and

beliefs about politics.

Each interview lasted between 70 minutes and 3 hours. At the end of each interview, I

asked each informant if they had final thoughts or comments, thanked them for their

participation, and paid them $30 for their time. I gave each person the opportunity to put down

contact information in case they wanted to hear about the results of the study. When time

permitted, we often continued to talk informally after the interview was over. Some people

wanted to be “debriefed;” others wanted to learn about me; others were simply content to talk

about topics of mutual interest.

Data Analysis

As soon as possible after completion of the interview, I wrote up brief fieldnotes about

the interview. I wrote as much as I could about the appearance, demeanor, and impression made

by each informant. I jotted notes about the setting in which the interview took place, and I wrote

about my impressions of the interview. I included a brief summary of the person’s overall

attitudes and orientations to marriage, sexuality, religion, and politics, and I noted specific

answers or sections of the interview that I thought were particularly notable or interesting.
   86  
Lastly, I wrote any reflections I had about the significance of the interview for my larger

research questions and theoretical interests.

These short fieldnotes, combined with the notes I took during the interview, constitute a

two-page annotated summary of each interview. I used these interview summaries for initial data

analysis and for the purposes of modifying my recruitment strategies and interview techniques as

needed. During the data collection period, I dropped some questions from the interviews, and I

modified my recruitment strategies in an effort to capture the widest possible variation in

religious and political beliefs. The ongoing dialogue between my data and my theoretical

interests, reflected in these summaries, shaped my initial analyses of the cultural foundations of

attitudes about same-sex marriage and of the ways that social generational change is shaping the

content of the interviews. Early papers and presentations given on the research were based on

these interview summaries. I used the summaries to make rough estimates of the prevalence of

different combinations of attitudes in my sample. I was also able to use the summaries to locate

specific passages from specific informants in the audio recordings or transcripts as evidence for

various claims.

All interviews were transcribed in full, with the exception of my standardized

introduction and any interruptions during the interview. The standard I applied for transcriptions

was that all utterances of meaning or significance were to be included in the transcript. In

practice, this meant that not every utterance—such as “um” and “like”—was transcribed.

However, I tried to type enough of such utterances to convey a sense of the style of speech of

each informant; I do judge the variation in speech mannerisms to be meaningful. In cases where

informants started a sentence, stopped, and then completed the sentence using different words,

the stutter was sometimes transcribed and sometimes not, depending on how significant the
   87  
stutter seemed to be. Most of my affirmative vocalizations were omitted from the transcripts

because they were signals for the informant to continue and would interrupt long, coherent

blocks of text if they were to be included. Lastly, when there was laughter, pregnant pauses, or

other vocal sounds that conveyed some meaning or emotion, they were noted in brackets at the

place in the transcript that they occurred.

I coded all interviews in NVivo using a codebook I prepared (Appendix E). Coding of the

interviews occurred in two stages. In the first stage of coding, which might be loosely described

as “Open Coding,” I listened to each audio recording while proofreading the transcript, both to

ensure proper transcriptions and to fully immerse myself in the interview for the coding. While

listening and proofreading, I coded each transcript in six different ways. First, I recorded the

Attributes of each individual (age, ethnicity, etc.). Second, I coded the entire interview using

Question-Response (QR) codes, which were named for the question that I asked the interviewee.

The block of text in a QR code included the text of my question, the answer by the interviewee,

and all text (including probes and follow-ups) that transpired until the next question that required

a new QR code. In this way, I am able to analyze interviews by comparing responses to the

conversational stimuli that I used.

The third type of code in the first stage of coding is the Deductive codes. Deductive codes

are categories of specific things (such as attitudes, beliefs, discourses, cognitive resources, etc.)

that I determined were of theoretical interest and wanted to identify in each interview. Deductive

codes could appear anywhere in the interview. For example, “same-sex marriage attitude” would

most likely be expressed in response to my question about what person’s opinion about the issue

is, but it could also be expressed at other times as well. The specific attribute of each Deductive
   88  
code (e.g. opposed to same-sex marriage) was determined inductively during the second stage of

coding.

The fourth type of code in the first stage of coding is the Inductive codes. These codes

were categories of things of interest that emerged from the interviews, and that I did not set out

to identify. Like Inductive codes, I also used Keyword codes to inductively identify words or

phrases that seemed to be particularly important or meaningful for the ongoing analysis. For

these two types of codes, I added them to the codebook as I created them.

Lastly, I wrote Annotations during the first stage of coding to identify passages in the

transcripts that were particularly important. I wrote two types of annotations. Interpretive

annotations concerned the relationship between discourse and the informant; they were written

to explicate how the discourse in the transcript should be understood in light of some

characteristics or experiences of the person. By contrast, theoretical annotations concerned the

relationship between discourse and sociological theory; they were written to explicate how the

discourse in the transcript is related to larger theories or concepts of interest.

In the second stage of coding, “Axial Coding,” I further coded various codes (as opposed

to further coding each interview). I selected various QR codes, Deductive codes, Inductive

codes, and Keyword codes, for deeper and comparative analysis. I coded these codes inductively,

in essence specifying the “attributes” of each code. For example, by selecting the “same-sex

marriage attitude” code for axial coding, I compared all of the same-sex marriage attitudes to one

another and inductively assigned them labels indicating the nature of the attitude.

I wrote Memos throughout the process of data collection and analysis, and these memos

took several forms. While I was conducting interviews, I occasionally wrote short memos to

myself, in which I wrestled with the theoretical significance of the interview data. Some of my
   89  
fieldnotes from individual interviews also took this form. Early conference papers and

presentations based on my data also had the effect of helping me sort through the theoretical

importance of the patterns in my data. Lastly, after coding matched student-parent interviews, I

wrote a short memo in which I compared similarities and differences between the student and

parent. Thus, I had one memo for each student-parent comparison in my sample.

From the results of this coding, I wrote Chapters 4-6. In Chapter 4, I discuss how

people’s discourses about same-sex marriage are predicated upon their beliefs, attitudes, values,

and life experiences regarding homosexuality. In Chapter 5, I discuss how people’s discourses

about same-sex marriage are related to the meanings of marriage that informants advocate. In

Chapter 6, I use matched student-parent pairs to compare similarities and differences between the

two cohorts in discourse about same-sex marriage and patterns of association.

Generalizability

Because of the non-probability sampling techniques used for this study, the results are not

generalizable in a statistical sense. Beyond what the quantitative analyses in Chapter 3 suggest, I

can make no claims about the prevalence of certain attitudes or opinions in the general

population; nor can I offer any reliable estimates of how common various discourses are among

specific groups of people. For example, because the research was only carried out in a single

geographical region, involved only people of particular ages and students attending particular

colleges, it would be erroneous to assume that the results of this study apply to all Americans, or

even to all Americans in these age groups. It is likely that patterns of discourse that I found

would differ in San Francisco, for example, or in the rural American south; similarly, patterns of

discourse from much older Americans are likely to be different from those described here.
   90  
Nevertheless, the results of this study are generalizable to the extent that the structures of

the belief systems of people in my sample and the ways in which people in my sample use their

cultural “repertoire” to construct opinions are representative of people in the population at large.

In other words, to the extent that the lifeworlds of people in my sample are similar to the

lifeworlds of other Americans, the results of this study are generalizable. Though I have no

precise measurement of this representativeness, theoretically, there are strong reasons to believe

that the patterns of discourse that I found are fairly widespread among Americans. For example,

to the extent that media and popular culture influence the ways in which people talk about

homosexuality and same-sex marriage, the declining regional variation in popular culture makes

it likely that people all over the United States are influenced similarly by television shows like

“Will and Grace” or celebrities like Ellen Degeneres. Similarly, the ways in which people draw

from their religious beliefs or political ideologies to talk about same-sex marriage are likely to be

fairly similar among people of particular ages, faiths, and parties.

This study demonstrates the dual nature of culture, as both structure and as a tool-kit for

meaning-making (Sewell 1999). I argue that this study shows strong evidence of the ways in

which the issue of same-sex marriage fits within mainstream American culture and discourse,

and thus how people who are conversant with mainstream American culture are likely to think

and talk about same-sex marriage. The virtue of this study is that it captures a wide range of

variation in the ways in which people talk about same-sex marriage and the ways in which

attitudes about same-sex marriage are related to other aspects of people’s life experiences and

belief systems. Although I cannot make claims about which attitudes are strongest or most

pervasive, I do show how same-sex marriage fits within the broader “webs of significance” that

constitutes mainstream American culture (Geertz 1973).

   91  
Chapter 3: Quantitative Analysis of Attitudes

As described in the previous chapter, the core data for this study are derived from

qualitative interviews with a non-representative sample of college students and their parents in

northern Illinois. But that data will be more useful when considered in tandem with nationally

representative survey data about Americans’ attitudes about homosexuality and same-sex

marriage. To that end, this chapter is a secondary analysis of existing data about attitudes toward

same-sex marriage, derived from a nationally representative sample of Americans. I have

conducted this analysis because it provides important insights into the cultural foundations of

attitudes about same-sex marriage that my interview data do not. Moreover, because of their

mutual strengths and weaknesses, using the quantitative and qualitative analyses in dialogue with

one another enhances the value of each approach. This quantitative analysis therefore serves as

both a starting point and a validity check of the qualitative interview data, which will be

analyzed in more depth in subsequent chapters.

First, this quantitative analysis provides a general model of how various demographic,

contextual, and attitudinal variables are related to attitudes about same-sex marriage. In addition

to providing a generalizable framework for understanding the foundations of attitudes about

same-sex marriage, this analysis can also illuminate important relationships and significant

puzzles for further interrogation using qualitative interview data. It serves, therefore, as an

important starting point for qualitative analysis. The quantitative analysis paints a broad, general

picture of the cultural foundations of attitudes toward same-sex marriage and points us toward

specific areas for more detailed inquiry. Read in tandem with other public opinion studies, this
   92  
analysis shows how attitudes about same-sex marriage are related to beliefs and attitudes about

homosexuality.

Reciprocally, because of its statistical estimates, this quantitative analysis serves as a

guide to estimating the substantive significance and generalizability of findings from the

qualitative interview data. Because my sample of interviewees is not a probability sample, it is

difficult to determine how much theoretical significance to attach to particular findings. So even

though qualitative interview data may provide greater depth of insight into people’s cultural

repertoires, there is an inherent risk of placing too much analytical weight on aberrant cases or

unusual relationships. This quantitative analysis can, in the end, serve as a guide for interpreting

the significance and generalizability of particular findings.

Lastly, this quantitative analysis is vital for assessing the significance of age-related

differences in attitudes about same-sex marriage. Unlike the qualitative interview data, this

dataset contains respondents of all ages. It can therefore be used to examine the predictive power

of different age and cohort measurements on attitudes about same-sex marriage, and multiple

cohort classification schemes may be compared with one another to assess the theoretical

plausibility of the “cohort effect” interpretation.

In the analysis below, I first address this issue of age-related differences and the

plausibility of the cohort effect interpretation of these differences. Second, I show that attitudes

and cognitive beliefs about homosexuality are all highly related to one another and that they may

be multiple measurements of a single, underlying worldview about homosexuality. Third, I show

that these attitudes and beliefs about homosexuality are of fundamental importance in explaining

attitudes about same-sex marriage. I argue that the main influence of demographic and

contextual variables on Americans’ attitudes about same-sex marriage is indirect, mediated by

   93  
values, beliefs, and attitudes. Even measures of religiosity and political ideology are significant

predictors of attitudes toward same-sex marriage primarily insofar as they are related to more

general values, beliefs, and attitudes about homosexuality. Finally, from the comparison between

same-sex marriage and civil unions, I show that there are similar patterns of association at work,

but that differences among people’s cultural definitions of “marriage” and “family” might shape

people’s attitudes toward same-sex marriage in important, but unmeasured, ways. I conclude by

discussing the implications of this analysis, and I outline a plan for subsequent analysis of

qualitative interview data in light of these findings.

Predictors of Attitudes about Homosexuality and Same-Sex Marriage

As described in the first chapter, previous studies of public opinion about same-sex

marriage have shown that a variety of demographic and contextual factors shape people’s

attitudes toward same-sex marriage, including religiosity, sex, age, education, political

affiliation, and area of residence (Brewer and Wilcox 2005; Olson, Cadge and Harrison 2006;

Pew Research Center for the People and the Press 2003). If we view same-sex marriage as one of

a number of related issues about gay rights, as do Wilcox and Wolpert (2000), the literature

suggests that a number of other contextual, cognitive, and attitudinal variables would also be

important predictors of attitudes about same-sex marriage. These include: beliefs about the

nature of homosexuality, holding negative stereotypes about gays, homophobia, implicit and

explicit motivations regarding prejudice, belief in traditional moral values, emotional reactions

toward homosexuality, and personal or mediated contact with gays and lesbians (Herek and

Glunt 1993; Lemm 2006; Schiappa, Gregg and Hewes 2006; Wilcox and Wolpert 2000; Wood

and Bartkowski 2004).

   94  
Unfortunately, few studies have measured the effects of these specific contextual,

cognitive, and attitudinal variables on attitudes about same-sex marriage directly. On one hand, it

is reasonable to expect that the patterns of association between these variables and opinion about

same-sex marriage would be similar to other gay rights issues. After all, the legal fight for same-

sex marriage is being pursued on those grounds, and proponents use civil rights frames to argue

for same-sex marriage just as they do for other issues (Brewer 2003; Hull 2001; Miceli 2005).

On the other hand, we should not assume that the same predictors of attitudes apply to same-sex

marriage because qualitative studies of the issue have shown that the debate over same-sex

marriage cannot be reduced to a question of gay rights. Arguments against same-sex marriage

have typically relied on deeper moral and religious claims about the meaning of marriage and

have been more successful than rights-based claims (Brewer 2002; Hull 2001; Price, Nir and

Cappella 2005). Moreover, the cultural meanings of marriage that gays and lesbians use to talk

about recognition of same-sex relationships are multiple, complex, and contested (Hull 2006;

Lannutti 2005). Thus, a direct test of how different demographic, contextual, cognitive, and

attitudinal measures are related to attitudes about same-sex marriage is warranted.

I also argued in Chapter 1 that the age differences in attitudes about same-sex marriage

are most likely cohort effects, because of the cohort effects that account for liberalization in

attitudes about homosexuality and gender ideologies (Brewster and Padavic 2000; Brooks and

Bolzendahl 2004; Ciabattari 2001; Loftus 2001; Treas 2002). Moreover, the liberalization in

attitudes about same-sex marriage over time is consistent with the cohort effect hypothesis

(Brewer and Wilcox 2005; Lax and Phillips 2009). No studies to date have attempted to

determine whether this is an age or a cohort effect, so this question should be investigated

further.
   95  

Data and Measures

Data used for this analysis are from a telephone survey conducted by Princeton Survey

Research Associates for the Pew Research Center for the People and the Press (2003). The

survey was of a stratified random sample of 1,515 adults over age 18 living in the continental

U.S. The sampling procedure used random digit dialing of telephone numbers within county

telephone exchanges, where the probability of a number being dialed was proportional to the

county’s share of all telephone numbers in the U.S. The sample is thus reasonably representative

of all American adults living in the continental U.S. who have telephones.

The survey took place October 15-19, 2003, only one month before the Massachusetts

Supreme Court ruled it unconstitutional to deny same-sex couples the right to civil marriage.

Public opposition to same-sex marriage spiked in the months immediately following the ruling,

evidence that the Massachusetts decision provoked a “moral panic” (Goode and Ben-Yehuda

1994). However, opposition to same-sex marriage softened and the proportion of Americans

supporting same-sex marriage began to rise after 2004, thus continuing the longer trend in

liberalization of attitudes toward homosexuality and the more recent trend in liberalization of

attitudes toward same-sex marriage (Brewer and Wilcox 2005; Loftus 2001; Pew Research

Center for the People and the Press 2006; Wilcox and Wolpert 2000).

Although this dataset was created before these events and is thus somewhat out of date, I

use it because it includes a more extensive set of measures of attitudes and beliefs about

homosexuality and same-sex relationships than do most surveys. The survey was designed with

the Pew Forum on Religion and Public Life, and the list of topics covered in the questionnaire
   96  
focuses on religion, homosexuality, and gay marriage.2 This makes this dataset ideal for

analyzing the cultural foundations of attitudes toward same-sex marriage. It also overcomes an

important limitation of some otherwise-valuable studies of attitudes toward homosexuality,

which use cultural and ideological variables that are not directly related to homosexuality (e.g.

Loftus 2001).

This dataset is also useful because the questionnaire asked about a person’s level of

support for both same-sex marriage and civil unions, which allows us to compare the similarities

and differences between attitudes regarding the two forms of legal recognition. Thus, the two

main dependent variables in this analysis are support for same-sex marriage and support for civil

unions. Respondents were first asked, “Do you strongly favor, favor, oppose, or strongly oppose

allowing gays and lesbians to marry legally.” The next question asked, “Do you strongly favor,

favor, oppose, or strongly oppose allowing gay and lesbian couples to enter into legal agreements

with each other that would give them many of the same rights as married couples?” The four

answer choices were assigned values 1 (strongly oppose) through 4 (strongly favor). The order of

the questions was reversed on one version of the survey. Because the researchers found that

question order affected people’s stated opinions about civil unions, I include a dummy variable

to control for the effect of asking the civil union question first.

Independent variables include a set of demographic and contextual measures and a

variety of measures of attitudes and cognitive beliefs about homosexuality. Demographic and

contextual variables include: sex, age, education, current marital status, political ideology, size of

the city or town in which they live, religion, frequency of attendance at religious services,

                                                                                                               
2  The questionnaire also contained questions on current events, the 2004 Presidential election,

and demographic characteristics of the respondent.  

   97  
identification as born-again or evangelical Christian, and personal contact with gays and lesbians

(whether or not “you have a friend, colleague, or family member who is gay”). Three

demographic variables—income, race (a dummy variable coded as white or not white), and

parental status (a dummy variable indicating whether or not the respondent is currently a parent

or guardian of a child under age 18)—were dropped from the final models because of

measurement problems and their lack of correlation with the dependent variables.

Three independent variables measure a person’s cognitive beliefs about homosexuality.

Respondents were first asked, “In your opinion, when a person is homosexual is it something

that people are born with, or is it something that develops because of the way people are brought

up, or is it just the way that some people prefer to live?” Responses were coded from 1 (born

with) to 3 (lifestyle). Additionally, respondents were asked, “Do you think a gay or lesbian

person’s sexual orientation can be changed or cannot be changed?” Lastly, respondents were

asked, “Do you think it is a sin, or not, to engage in homosexual behavior?”

Five independent variables measure a person’s attitudes about gays and lesbians. The first

measure is an index of opinions about gays and lesbians, made up of two items that asked

whether the respondent has a “very favorable, mostly favorable, mostly unfavorable, or very

unfavorable” opinion of gay men and of lesbian women. The two items were highly correlated (r

= .88). Whether the item about gay men or lesbian women was asked first was alternated.

Immediately after those items, respondents were asked, “Do you think more acceptance of gays

and lesbians would be a good thing or a bad thing for the country—or that it would not make

much difference either way?” Responses were coded from 1 (bad for country) to 3 (good for

country). The third item gauged the extent of agreement with the statement that “Gay and lesbian

couples can be as good parents as heterosexual couples,” coded from 1 (completely disagree) to
   98  
4 (completely agree). Fourth, respondents were asked whether “it doesn’t bother you to be

around homosexuals” or “it makes you uncomfortable to be around homosexuals.” The final

measure is an item that asked respondents to “completely agree, somewhat agree, somewhat

disagree, or completely disagree” that “Allowing gay and lesbian couples to legally marry would

undermine the traditional American family.” The response was coded from 1 (completely

disagree) to 4 (completely agree).

Age or Cohort?

The first issue to be addressed is whether a continuous age measure or an ordinal cohort

measure is more appropriate for predicting attitudes about same-sex marriage and civil unions. I

constructed seven different cohort classification schemes and categorized respondents based on

their age into the appropriate cohorts. Then I compared the predictive power of the eight

measurements to one another using univariate OLS linear regression, univariate ordinal

regression, multivariate OLS linear regression, and multivariate ordinal regression.

Two cohort measurements were constructed using variants of common “generational”

definitions (e.g. Baby Boom, Generation X). The other five cohort classifications are based on

theoretically important periods in the political and cultural standing of gays and lesbians in

American society. Because the dependent variables are about same-sex marriage, and implicitly

about homosexuality, I expected the best cohort measure to be based on critical moments in

American history where the cultural construction of homosexuality in society changed

significantly.

I began by identifying what might be called “transformative moments” in the history of

the changing status of gays and lesbians in society. The identification of these transformative
   99  
moments was guided by histories of gay and lesbian political movements and media portrayals

(Armstrong and Crage 2006; Becker 2006; Bernstein 1997; Bernstein 2002; Gallagher and Bull

2001; Gross 2001; Rimmerman, Wald and Wilcox 2000; Seidman 2004; Walters 2001), my

concern with public opinion, my concern with the issue of same-sex marriage in particular, and

the life-span of individuals surveyed. Because my identification of these transformative moments

was motivated by a very specific analytic interest, I must emphasize that these transformative

moments should not be interpreted as universally important. Scholars disagree on the timing and

relative importance of various changes in the status of gays and lesbians in society; I have chosen

those that seem to be most relevant to the general public’s cultural understanding of

homosexuality, given my interest in contemporary attitudes about same-sex marriage.

There are two moments that seem to be of fundamental importance for my analysis. The

first period, from 1969-73, is the period in which gay politics emerged from the closet and

assumed the mantle of gay liberation. The period is marked by the Stonewall riot in 1969, the

beginnings of its commemoration in gay pride marches in 1970, and the removal of

homosexuality from the American Psychiatric Association’s Diagnostic and Statistical Manual of

Psychological Disorders. Prior to the late 1960s, the homophile movement was largely

internally-focused; over the course of the late 1960s, gay and lesbian movements became

gradually more publicly-oriented and visible in mainstream American culture. The events of

1969-73 reflect the culmination of this change in strategy and the consolidation of a visible,

public image of gays and lesbians in society.

The second transformative moment, 1985-87, is marked by the exposure of popular

movie star Rock Hudson as having AIDS and the National March on Washington to demand

equality for gays and lesbians and improved funding for HIV/AIDS treatment and research. This
   100  
is the period in which gay liberation politics began to be supplanted by gay rights politics; the

strategy of gay pride and visibility was accompanied by increasing demands on governments for

equal rights. At the same time, this is the period in which gays and lesbians in America began to

be increasingly viewed as a legitimate status group, akin to a racial or ethnic minority group, that

had been the victim of discrimination and unequal protection. Rock Hudson’s outing as gay was

an important moment precisely because he had seemed so “normal” to mainstream Americans

and represented the first personal exposure many people had to someone who was gay. The

transition to gay rights was consolidated after 1992, after Bill Clinton was elected president and

sympathetic gay and lesbian characters became increasingly common on network television and

in Hollywood films.

From these two transformative moments, I created five different cohort classifications by

subtracting different numbers of years to establish the birth years of people who would have

come of age in each of the three distinct periods (distinguished by these two transformative

moments). I hypothesized that a person who was 16 years old at the beginning of each

transformative moment was a logical choice for determining the boundaries of the cohorts.

Although it is conventional for the “coming of age” period to be thought of as the years that

people leave high school and either begin college or full-time work, I chose a younger age

because dating and sexuality become subjects of social awareness and interest much earlier. I

then also created cohort measures based on the subtraction of 15, 17, 18, and 19 years from the

beginning of each transformative period. Because individuals “come of age” at different ages and

because the periods of historical change described here last many years, I expected that there

would be little difference among these five measurements.

   101  

Table 1: Age and Cohort Classifications as Predictors of Support for Same-Sex Marriage in
Univariate Regressions

Variable Cohort (Birth years) Age in Ordinal Linear Linear Reg.

Name 2003 Reg. Reg. Standardized
Chi-squarea F-testa Beta
Age 18 & up 114.4 131.5 -.228

GenerationA 1. Millenial (1978-85) 18-25 114.7 125.1 -.222

2. GenX (1964-77) 26-39
3. Baby Boom (1945-63) 40-58
4. WWII (Before 1945) 59 & up
GenerationB 1. Post-Boom (1966-85) 18-37 121.3* 132.0* -.228
2. Baby Boom (1945-65) 38-58
3. WWII (Before 1945) 59 & up
Gay-15 1. Rights (1970-85) 18-33 111.8 122.8 -.221
2. Liberation (1954-69) 34-49
3. Closet (Before 1954) 50 & up
Gay-16 1. Rights (1969-85) 18-34 114.3 122.0 -.220
2. Liberation (1953-68) 35-50
3. Closet (Before 1953) 51 & up
Gay-17 1. Rights (1968-85) 18-35 124.2* 136.4* -.232*
2. Liberation (1952-67) 36-51
3. Closet (Before 1952) 52 & up
Gay-18 1. Rights (1967-85) 18-36 131.5** 143.2** -.237**
2. Liberation (1951-66) 37-52
3. Closet (Before 1951) 53 & up
Gay-19 1. Rights (1966-85) 18-37 127.3* 141.3* -.236*
2. Liberation (1950-65) 38-53
3. Closet (Before 1950) 54 & up
*Better performance than continuous age measure.
**Best performance.
a
Null model is intercept-only.

Table 1 describes each of the seven cohort classifications and the results of comparisons

with the continuous age variable as predictors of attitudes about same-sex marriage. Although

there are slight differences in how each measure performs, there are multiple cohort measures

that perform marginally better than the age variable. The strongest measure is the cohort
   102  
classification that subtracts 18 years from the beginning of the transformative moment. By

coincidence, however, this cohort classification also divides the sample into nearly equal thirds.

Thus, the marginally better performance may have more to do with equal cell sizes than with

anything of theoretical interest. Nevertheless, the fact that the age variable performs more poorly

than multiple cohort measures in both types of analysis suggests that a cohort classification is

more appropriate than a continuous age variable as a predictor of attitudes about same-sex

marriage.

Multiple regression analysis, whether linear or ordinal, confirms this assessment. I used

both the “Gay-16” and “Gay-18” measures in multiple regression analysis controlling for other

demographic variables. Because the results were the same for both ordinal and linear regression,

I present the linear regression results in Table 2 for the sake of simplicity. Both cohort measures

result in better models than the comparable models using a continuous age measure. If both the

age measure and either of the cohort measures are entered into the linear regression equation

using a stepwise method of entering the variables, the cohort measure is preferred and the age

measure is left out. In the stepwise regression shown, both cohort measures and the age measure

are entered, and the “Gay-18” measure is preferred.

At the same time, there are no major substantive differences among any of the models.

The models all perform similarly and no coefficients require different interpretation. This means

that the exact specification of the cohort measure makes little difference. Thus, the results

confirm my expectations: that a cohort measure deduced theoretically would be the best measure

in relation to attitudes about same-sex marriage, and that the boundaries among the cohorts are

not distinct.
   103  
Interpreted conservatively, these results show that a simple cohort classification scheme,

based on the periods of evolving status of gays and lesbians in American society during which a

person comes of age, performs no worse than the continuous age measure. In other words, a

Table 2: Multiple Linear Regression of Support for Same-Sex Marriage on Age and Two
Different Cohort Measures, Controlling for Demographic Variables

Enter Age Enter Gay-16 Enter Gay-18 Stepwise All

Only Only Only
Age -.110** -- -- ##
Gay-16 Cohort -- -.121** -- ##
Gay-18 Cohort -- -- -.127** -.133**
Male -.075** -.070** -.071** -.077**
Education .089** .089** .085** .086**
Married -.033 -.028 -.030 ##
Political Liberalism .244** .246** .245** .250**
City Size .083** .078** .078** .084**
Church Attendance -.179** -.179** -.174** -.180**
Born-Again or Evangelical -.174** -.173** -.177** -.176**
Gay Friend/Family/Colleague .037 .037 .037 ##
Civil Union Question First -.090* -.045* -.046* -.049*
Constant 1.771** 1.774** 1.792** 1.804**
df 10/1611 10/1611 10/1611 8/1611
2
Adjusted R .235 .237 .239 .238
* p < .05 ** p < .01 (two-tailed tests)
## Variables excluded from final model.
Note: Standardized coefficients shown. First three models enter all variables simultaneously. The
final model enters variables in stepwise fashion.

continuous age measure adds nothing to our predictive power of support for same-sex marriage,

over and above the simple fact of being a member of one of these three cohorts. What matters, in

terms of attitudes about same-sex marriage, is the period in which a person reaches adulthood.

Thus, for future analyses, I use the “Gay-18” cohort measure rather than the continuous age

measure.

Figure 1 shows the means and 95% confidence intervals of support for same-sex

marriage, by cohort. The differences are statistically significant. Substantively, the differences
   104  
are not large, because same-sex marriage is still opposed by a large majority of the overall

sample, but members of younger cohorts are clearly more likely to support same-sex marriage

than members of older cohorts.

Bivariate Correlations

The second subject that must be addressed is the relations of the cognitive and attitudinal

variables to one another and to the dependent variables. The dependent variables and the

cognitive and attitudinal questionnaire items concern a wide variety of distinct things: levels of

support for same-sex marriage, levels of support for civil unions, feelings about gays and
   105  
lesbians, beliefs about the nature of homosexuality, moral evaluation of homosexuality, attitudes

about gay parenting abilities, attitudes about gay relationships, notions of “family,” and beliefs

about the place of gays and lesbians in society. Despite the differences among these

questionnaire items, however, there is evidence that these items are all measuring similar things.

Table 3 shows the bivariate correlations among the two dependent variables and the eight

cognitive and attitudinal independent variables. All are highly correlated. Each bivariate

correlation was statistically significant, and none was smaller than +/-.173. Such high levels of

inter-item correlation pose problems of multicollinearity in regression analysis, and they

undermine efforts at causal inference for attitudes about same-sex marriage.

These items are so highly correlated, in fact, that it is possible that these items are all

simply different measurements of a single latent variable. A factor analysis of the two dependent

variables and eight independent variables suggested a one-factor solution, indicating that these

beliefs, attitudes, and feelings are all components of a common, underlying worldview about

homosexuality. Factor scores for the items range between +/- .512 and .874. Excluding the

dependent variables from the factor analysis makes no substantive difference.

This is a notable finding. The fact that all cognitive and attitudinal items seem to be

measuring a single underlying cultural worldview about homosexuality is surprising, given the

theoretically important differences among them. One possible explanation for this is that survey

respondents have such strong feelings about homosexuality that it affects how they respond to

any question about the subject, no matter what its specific content. It is also possible that

respondents attempt to appear logically consistent or want to avoid appearing unintelligent or

self-contradictory to the interviewer. Lastly, it is possible that all of these cognitive beliefs and

attitudes have a coherent ideological structure, such that they are premised upon one another.
   106  
Table 3: Bivariate Correlations Among Dependent Variables and Cognitive and
Attitudinal Variables

1 2 3 4 5 6 7 8 9 10
1. Support for 1
Same-Sex
Marriage
2. Support for .830 1
Civil Unions

3. Opinion of .652 .681 1

Gays and
Lesbians
4. Definition -.357 -.381 -.408 1
of
Homosexual
5. Can be -.384 -.404 -.419 .460 1
Changed

6. Not .328 .350 .429 -.207 -.178 1

Uncomfortable

7. Not a Sin .654 .619 .609 -.370 -.418 .275 1

8. Acceptance .637 .652 .661 -.360 -.363 .390 .574 1

Good for
Country
9. Good .600 .637 .599 -.323 -.346 .392 .461 .587 1
Parents

10. Undermine -.554 -.471 -.406 .247 .280 -.173 -.420 -.461 -.395 1
Traditional
Family

Notes: Pearson correlations shown. All correlations are significant at the p < .01 level (two-
tailed).

Whatever the case may be, the appearance that these attitudes and beliefs constitute a

“worldview” suggests that the debate about same-sex marriage may be primarily about

homosexuality and only tangentially about “marriage.”

   107  
For subsequent analyses, I chose not to use a single “homosexuality worldview” variable

constructed from the eight independent variables (excluding the two dependent variables) for

three reasons. First, the assumptions required for factor analysis (e.g. continuous measurements,

normal distributions, linear correlation) are strongly violated in the case of dichotomous

variables. Because four of the variables are dichotomous, and the other four are ordinal, the

factor analysis may not be accurate. Second, collapsing eight distinct cognitive and attitudinal

variables into a single underlying factor makes indistinguishable any substantive differences

among the variables in their relations to attitudes about same-sex marriage. Moreover, the

coefficient of the factor is less easily interpretable than those of the individual variables. Third,

my experience of interviewing people directly has convinced me that there are important

theoretical and methodological differences between questions that are primarily attitudinal and

those that attempt to measure cognitive beliefs. In particular, questions measuring people’s

cognitive beliefs about homosexuality seem to be more difficult to answer than attitudinal

questions, and the cognitive processes people use to formulate answers may be different.

For the regressions that follow, I dropped three of these variables because they are likely

to be post-hoc justifications for respondents’ previously stated opinion about same-sex marriages

and civil unions: whether or not more acceptance of gays and lesbians would be good for the

country, whether or not gays and lesbians can be as good of parents as heterosexuals, and

whether or not legalizing same-sex marriage would undermine the traditional family. There

remain five independent cognitive and attitudinal variables that are logically prior to a person’s

opinion about same-sex marriage: a person’s definition of homosexuality, their belief about

whether or not homosexuality can be changed, their belief about whether or not it is a sin, their
   108  
overall opinion of gays and lesbians, and whether or not being around gays and lesbians makes

them feel uncomfortable.

Methods

Using SPSS, I used both ordinal regression and OLS linear regression to estimate the

effects of the demographic, contextual, and cognitive and attitudinal variables on each dependent

variable. For each dependent variable, I tested three models: the first contained only

demographic and contextual variables; the second contained only the cognitive and attitudinal

variables; the third contained all variables. In SPSS, the ordinal regression procedure allows

independent variables to be specified as either continuous “covariates” or categorical “factors.”

The attribute of a categorical variable with the highest value is treated as the reference category.

Ordinal independent variables could plausibly be treated as either covariates or factors. Because

my dependent variables are ordinal and because all of my independent variables (excluding age)

are categorical or ordinal, I present the ordinal regression results here. The only ordinal variables

I treated as continuous covariates were those with more than six response categories (education,

church attendance, and opinion of gays and lesbians).

An additional advantage of the SPSS ordinal regression procedure is that responses of

“Don’t know” or “Refused to answer” can be treated as meaningful responses; dropping them

cases from the analysis or imputing responses is unnecessary. For independent variables in which

there were large numbers of “Don’t know/Refused” responses (more than 5% of total), I coded

“Don’t know/Refused” as “0” and included them in the analysis to be compared to the reference

category. I did this for measures of political ideology and four of the five cognitive and

attitudinal variables (all except for overall opinion of gays and lesbians).
   109  
Because of the ease with which ordinal regression handles categorical variables and

“Don’t know/Refused” responses, the ordinal and linear regression models that I tested were

different. In the linear regression, I did not create dummy variables for all attributes of the

categorical variables, opting instead for simpler measures of marital status and religiosity. In

ordinal regression, I used the large categorical variable “religion” rather than the “born

again/evangelical” variable, because missing values in the “born-again/evangelical” variable

removed all people who identified as Jewish, Muslim, or atheist/agnostic from the analysis.

Where appropriate, I discuss important differences between the linear and ordinal models in the

results below.

I estimated the ordinal regression equations using two separate link functions: the logit

link function, which transforms parameter estimates of probabilities into log odds; and the

“negative log-log” link function, which is used when the lower response categories of the

dependent variable are more probable. Because opposition to same-sex marriage was much

greater than support for same-sex marriage, and because relatively few respondents said that they

strongly support same-sex marriage, I assumed that the negative log-log link function would be

appropriate. However, the logit link function consistently produced better fitting models using -2

log likelihood, chi-square, and pseudo r-square tests, so I report those results below.

Results: Support for Same-Sex Marriage

Table 4 shows the results of the ordinal regression of support for same-sex marriage on

the explanatory variables. Looking first at the demographic and contextual variables in the first

model, the strongest predictors of support for same-sex marriage are the measures of political

ideology and religiosity. Support for same-sex marriage is strongly associated with identifying as
   110  
politically liberal; identifying as non-religious, Jewish or with another non-Christian religion;

and less frequent church attendance (or a weaker religious identity)3. In addition, women,

members of the Rights Cohort, people with higher educational attainment, people who have

never been married, people who live in large cities, and people who personally know a friend,

family member, or colleague who identifies as gay or lesbian, are more likely to say they support

same-sex marriage. These results are in accord with other public opinion studies on this issue.

Table 4: Ordinal Regression of Support for Same-Sex Marriage on

Demographic/Contextual Variables and Ideological Variables

Dem. Model Ideo. Model Full Model

Sex Male -.383 (.09)** -.345 (.11)**
Female 0 0
Cohort Rights Cohort .482 (.13)** .409 (.16)**
Liberation Cohort .129 (.12) .213 (.14)
Closet Cohort 0 0
Education .126 (.03)** -.090 (.035)*
Marital Status Married -.363 (.13)** -.240 (.15)
Divorced -.204 (.17) -.077 (.20)
Separated -.381 (.32) .035 (.41)
Widowed -.367 (.21) -.033 (.25)
Never been married 0 0
Political Ideology Don’t know/Refused .276 (.30) .273 (.36)
Very liberal 2.054 (.30)** .181 (.35)
Liberal 1.459 (.23)** .093 (.26)
Moderate .697 (.21)** -.572 (.24)*
Conservative .121 (.21) -.743 (.24)**
Very conservative 0 0
City Size Large city .288 (.14)* -.043 (.17)
Suburb near large city .206 (.14) .073 (.16)
Small city or town -.031 (.13) -.109 (.15)
Rural area 0 0
Religion Protestant Christian -.924 (.17)** -.383 (.20)
Roman Catholic -.349 (.18) -.095 (.21)
Jewish 1.085 (.36)** .435 (.39)
                                                                                                               
3
In addition, Christians who do not identify as born-again or evangelical are more likely to
support same-sex marriage than those who do.
   111  
Mormon -2.033 (.43)** -.753 (.47)
Orthodox (Greek/Russ) -1.134 (.41)** -.065 (.44)
Muslim 1.263 (.77) 1.045 (.87)
Other Religion 1.375 (.44)** 2.202 (.59)**
Atheist, agnostic, none 0 0
Church Attendance -.314 (.03)** -.163 (.04)**
Gay Contact Yes .286 (.10)** -.609 (.12)**
No 0 0
Opinion of G/L .527 (.03)** .637 (.04)**
Uncomfortable Don’t know/Refused .208 (.25) -.331 (.31)
Yes -.538 (.13)** -.731 (.15)**
No 0 0
Sin Don’t know/Refused -1.400 (.16)** -1.247 (.19)**
Yes -1.978 (.12)** -1.708 (.14)**
No 0 0
Definition homo Don’t know/Refused -.078 (.16) -.032 (.21)
Born gay .402 (.12)** .473 (.14)**
Brought up .021 (.13) .129 (.15)
Lifestyle 0 0
Be changed Don’t know/Refused .178 (.15) .180 (.19)
No .230 (.11)* .126 (.13)
Yes 0
Question Order Civil Unions First -.119 (.09) -.169 (.09) -.237 (.10)*
Civil Unions Second 0 0 0
Threshold DV=1 (strongly oppose) .784 (.36)* .197 (.21) -.369 (.47)
DV=2 (oppose) 2.224 (.36)** 2.112 (.21)** 1.830 (.48)**
DV=3 (support) 4.239 (.36)** 4.633 (.23)** 4.535 (.48)**
Chi-Square 965.81** 1509.57** 1454.80**
McFadden R2 .140 .269 .326
*p < .05 **p < .01 (two-tailed tests)
0 = Reference category
Notes: Standard errors of estimates in parentheses.

The second model, which contains only cognitive and attitudinal variables while omitting

demographic and contextual variables, is a much better model than the first model. In this model,

support for same-sex marriage is associated with people who have higher opinions of gays and

lesbians, who are not uncomfortable around gays and lesbians, who believe that homosexuality is

not a sin, who define homosexuality as something people are born with, and do not believe that

homosexuality can be changed. This pattern of results is to be expected. There is, however, one
   112  
unexpected finding: when respondents were asked whether or not homosexuality is a sin, people

who said they did not know or who refused to answer were almost as likely to oppose same-sex

marriage as people who answered affirmatively. There are two possible interpretations for this

finding, which are not mutually exclusive: First, many people may have refused to answer

because they did not want to publicly express a negative moral judgment, even though they

disagreed with homosexuality. Second, not knowing whether or not homosexuality is a sin may

be somewhat equivalent with believing homosexuality is a sin, as far as attitudes about same-sex

marriage are concerned. In other words, support for same-sex marriage may be premised upon

the decisive rejection of the idea that homosexuality is sinful. The qualitative analysis in Chapter

4 will shed more light on this unexpected finding. It shows that there is an important difference

between moral values and moral judgments, such that many people who believe that

homosexuality is wrong also do not think it is their place to judge whether or not others’

behaviors are sinful.

In the third model, all variables are included, and there are several notable findings. First,

the full model is only marginally better than the ideological model4, because the ideological

variables clearly account for some of the predictive power of the demographic and contextual

variables. In particular, measures of political ideology, religious identification, educational

attainment, marital status, city size, and personal contact with gays and lesbians lose most of

their predictive power, and the explanatory power that remains seems to come from only trivial

differences (e.g. between “conservative” and “very conservative”). At the same time, comparing

the second and third models, the ideological variables change very little with the addition of the

                                                                                                               
4
This is especially true using OLS linear regression, where the full model has almost no
explanatory power over the ideological model (an increase of .01 in the adjusted R-squared).
   113  
demographic and contextual variables. This suggests that cognitive beliefs and attitudes about

homosexuality are intervening variables between demographic variables and support for same-

sex marriage: demographic variables matter for attitudes about same-sex marriage only to the

extent that they are related to different attitudes and beliefs about homosexuality more generally.

There are, however, some notable exceptions to this trend, and this constitutes a second

important finding. Several demographic and contextual variables retain some significant

explanatory power, once attitudes and beliefs about homosexuality are accounted for. In

particular women, members of the Rights Cohort, people who report less church attendance, and

people who identify with a non-Judeo-Christian religion are significantly more likely to support

same-sex marriage. This suggests that, no matter one’s attitudes and beliefs about

homosexuality, being female, being in the youngest cohort, and having weak identification with

Judeo-Christian religions may make one less likely to oppose same-sex marriage.

Lastly, the variable “Gay contact” in the full model deserves some attention. Compared

with the demographic model, not only does the effect of knowing someone who is gay or lesbian

switch signs when the ideological variables are included, but the magnitude of the effect more

than doubles. The sheer size of this negative association between knowing someone who is gay

or lesbian and support for same-sex marriage is notable because it is unlikely to be a statistical

artifact5. I interpret this to mean that the general association between personal contact with gays

and lesbians is positive on support for same-sex marriage, but that there remain a significant

proportion of people who are made less likely to support same-sex marriage by virtue of their

knowing someone who is gay or lesbian. There may be an unmeasured attitudinal variable that

                                                                                                               
5
This strong negative association is consistent across a wide variety of models, using different
regression techniques.
   114  
could account for this—for example, a person’s definition of marriage—or the nature of the

personal contact may have been negative or antagonistic. As literature on the contact hypothesis

has shown, personal contact increases positive affect or tolerance for a minority only under

certain conditions (Allport 1954; Herek and Glunt 1993; Lee, Farrell and Link 2004; Sigelman et

al. 1996; Sigelman and Welch 1993).

Are Civil Unions Different?

The Pew Research Center data is also useful for evaluating the cultural foundations of

support for same-sex marriage because it includes a measure of support for civil unions. The two

questionnaire items offered identical response categories, were back-to-back in the item order,

and were alternated whether each question was asked first or second. The civil union question,

described above, does not use the words “civil union” or “marriage,” but instead describes a

“legal agreement” for gay and lesbian couples that would give them “many of the same rights as

married couples.” The importance of the question on civil unions is that it potentially separates

two objections to same-sex marriage: the extension of additional rights and recognition to gay

and lesbian couples and the application of the term “marriage” to a same-sex relationship.

In practice, civil unions and domestic partnerships have been implemented by legislatures

in states like Denmark and Vermont, where governments seek to compromise between

supporters and opponents of same-sex marriage (Eskridge and Spedale 2006; Mello 2004). The

logic of the civil union compromise is that it can provide the legal rights and benefits to gay and

lesbian couples that are guaranteed by the state’s recognition of marriage, while preserving any

religious or heterosexual meaning of the term “marriage.” In other words, the omission of the

term “marriage” could satisfy opponents who object to same-sex marriage because of their
   115  
religious definition of “marriage” and meet the primary legal objection of supporters of same-sex

marriage by providing equal rights to gay and lesbian couples.6

The literature offers no simple answers to the question of whether or not the foundations

of support of civil unions are different from those of same-sex marriage. In the most direct test of

this question to date, Price, Nir, and Cappella (2005) treat “homosexual marriage” and “civil

unions” as two frames for focus group discussions. The results are mixed: they fail to find a main

effect of the frames but instead found an interaction effect between frame and ideological

leanings of the group. Compared with the civil unions frame, the homosexual marriage frame

elicited fewer “pro” arguments from conservative groups and more “pro” arguments from liberal

groups. Thus, it appears that same-sex marriage may be a more polarizing issue than civil unions,

even though the difference between the two issues is contingent on other factors.

By comparing predictors of attitudes toward same-sex marriage with predictors of

attitudes toward civil unions, we gain some knowledge about the extent to which people object to

same-sex marriage because of their objection to gay rights and the extent to which their objection

is related to their cultural definition of marriage. We also may gain additional insight into the

extent to which civil unions or domestic partnerships, as a compromise solution to the

controversy, would be supported and for what reasons.

How large is the proportion of the population that opposes same-sex marriage but might

be willing to extend some form of legal recognition and rights to gay and lesbian couples? As

                                                                                                               
6  I am aware of the legal obstacles to “equal treatment under the law,” especially in the United

States, that would not be satisfied by recognition of “civil unions.” In particular, the federal
Defense of Marriage Act prevents federal rights and benefits from being extended to couples
covered by civil union or domestic partnership laws. However, for the purposes of this
dissertation, I am concerned primarily with the cognitive and attitudinal significance of civil
unions, as opposed to same-sex marriage.  
   116  
Table 5 shows, a non-trivial portion of all respondents either favor or strongly favor civil unions

while opposing or strongly opposing same-sex marriage. Approximately 1/5 of respondents who

first expressed some level of opposition to same-sex marriage (13.4% of the overall sample) said

that they would support extending legal rights to same-sex couples that are similar to those given

to married couples.

Table 5: Support for Same-Sex Marriage and Civil Unions

Support for Civil Unions Total

(Strongly)
Oppose (Strongly) Favor
Support for (Strongly)
Same-Sex Oppose 51.9% (585) 13.4% (151) 65.3% (736)
Marriage
(Strongly)
0.6% (7) 34.1% (385) 34.7% (392)
Favor
Total
52.5% (592) 47.5% (536) 100.0% (1128)

Note: Data from “Form 1” only, in which people were asked the question about same-sex
marriage first. Numbers in parentheses are higher than the actual N because of weights
used by the researchers.

It is unclear to what extent these opinions are firmly held and to what extent respondents

were influenced by context, social desirability bias, or word choice. After conducting the survey,

researchers found that question order significantly affected levels of support for civil unions. In

particular, respondents who were asked the civil union question second were more likely to say

they supported civil unions than respondents who were asked the civil union question first. The

researchers speculated that “when respondents have already had the opportunity to express their

opposition to gay marriage on the survey, more feel comfortable with allowing some legal rights
   117  
as an alternative” (Pew Research Center for the People and the Press 2003, p. 16). It is also

possible that respondents did not want to seem “mean” or bigoted to the interviewer.

Nevertheless, it is also plausible that there are important, substantive differences between

people who oppose all forms of legal recognition of same-sex couples and people who would

support legal recognition of same-sex couples, as long as it is not “marriage.” In the analysis that

follows, a control for question order can help us estimate the extent to which the opinions

expressed are due to question order and the extent to which there are important differences in

attitudes toward same-sex marriage and civil unions.

Results: Support for Civil Unions

Table 6 shows the results of the ordinal regressions of support for civil unions on

demographic, contextual, cognitive, and attitudinal factors. The results are very similar to the

models predicting support for same-sex marriage, with only a few minor differences.

Table 6: Ordinal Regression of Support for Civil Unions on Demographic/Contextual

Variables and Ideological Variables

Dem. Model Ideo. Model Full Model

Sex Male -.367 (.09)** -.303 (.10)**
Female 0 0
Cohort Rights Cohort .628 (.13)** .708 (.15)**
Liberation Cohort .093 (.12) .145 (.14)
Closet Cohort 0 0
Education .232 (.03)** .027 (.03)
Marital Status Married -.350 (.12)** -.197 (.14)
Divorced -.223 (.17) -.121 (.19)
Separated -.377 (.32) .177 (.40)
Widowed -.036 (.20) .555 (.229)*
Never been married 0 0
Political Ideology Don’t know/Refused .495 (.29) .213 (.332)
Very liberal 2.027 (.29)** .197 (.337)
   118  
Liberal 1.441 (.22)** .225 (.25)
Moderate .664 (.20)** -.513 (.23)*
Conservative .122 (.20) -.543 (.23)*
Very conservative 0 0
City Size Large city .295 (.14)* -.005 (.16)
Suburb near large city .052 (.13) -.176 (.15)
Small city or town .051 (.12) -.037 (.14)
Rural area 0 0
Religion Protestant Christian -.493 (.17)** .127 (.20)
Roman Catholic -.041 (.18) .275 (.21)
Jewish .906 (.37)* .164 (.39)
Mormon -.497 (.38) .895 (.41)*
Orthodox (Greek/Russ) -.346 (.41) .993 (.44)*
Muslim 1.003 (.77) .748 (.86)
Other Religion 1.154 (.45)* 1.606 (.56)**
Atheist, agnostic, none 0 0
Church Attendance -.349 (.03)** -.198 (.04)**
Gay Contact Yes .586 (.09)** -.041 (.11)
No 0 0
Opinion of G/L .633 (.03)** .667 (.04)**
Uncomfortable Don’t know/Refused .234 (.25) .189 (.28)
Yes -.446 (.12)** -.430 (.14)**
No 0 0
Sin Don’t know/Refused -.113 (.16)** -.825 (.19)**
Yes -1.577 (.12)** -1.286 (.14)**
No 0 0
Definition homo Don’t know/Refused .204 (.15) .501 (.18)**
Born gay .456 (.12)** .559 (.14)**
Brought up .038 (.13) .075 (.14)
Lifestyle 0 0
Be changed Don’t know/Refused .368 (.15)* .274 (.17)
No .419 (.11)** .296 (.12)*
Yes 0 0
Question Order Civil Unions First -.448 (.09)** -.550 (.09)** -.546 (.10)**
Civil Unions Second 0 0 0
Threshold DV=1 (strongly oppose) 1.454 (.35)** .483 (.20)* 1.335 (.46)**
DV=2 (oppose) 2.688 (.36)** 2.091 (.21)** 3.125 (.46)**
DV=3 (support) 4.765 (.37)** 4.780 (.23)** 5.960 (.48)**
Chi-Square 746.66** 1638.62** 1479.39**
McFadden R2 .142 .276 .310
*p < .05 **p < .01 (two-tailed tests)
0 = Reference category
Notes: Standard errors of estimates in parentheses.
   119  
The first notable difference is that question order matters much more for attitudes toward

civil unions, compared with attitudes toward same-sex marriage. Being first asked one’s opinion

about same-sex marriage is associated with higher levels of support for civil unions. This

finding, combined with the similarities between Table 4 and Table 6, suggests that attitudinal

differences between same-sex marriage and civil unions are primarily due to the context in which

the question is asked. This lends support to the speculation of Pew researchers that many people

may be comfortable with the idea of legal recognition of same-sex couples only after they have

been allowed to express their opposition to the idea of same-sex marriage. From this point of

view, civil unions may work as a compromise only in a procedural sense, not in the sense that

people would actually support civil unions. An alternative interpretation is that, when asked first

about civil unions, respondents immediately “translate” the question to be about same-sex

marriage. In other words, the question may only be interpreted as it is intended when their

opinion about same-sex marriage has already been solicited.

A second notable difference of the models applied to civil unions, as opposed to same-

sex marriage, is that certain factors appear to be more or less important for each dependent

variable. Among demographic and contextual variables, higher educational attainment, being a

member of the Rights Cohort, and personal contact with gays and lesbians exert a stronger

positive effect on support for civil unions, than for same-sex marriage. The difference in “Gay

contact” is especially noteworthy because it appears that the stronger association between

knowing someone who is gay or lesbian and support for civil unions prevents the relationship

from becoming negative once ideological variables are controlled for. This suggests that the

liberalizing effect of personal contact with gays and lesbians does not carry over from the

question of civil rights to the meaning of marriage. There may be something about the cultural
   120  
meanings carried by the term “marriage” that makes the debate about same-sex marriage

different from other civil rights issues.

Among cognitive and attitudinal variables, the belief in whether or not homosexuality can

be changed matters much more for predicting attitudes about civil unions, while the belief in

whether or not homosexuality is a sin matters less. The logic of this pattern of associations may

be similar to that of personal contact. Believing that a person’s sexual orientation is immutable

and beyond an individual’s control, like a person’s ethnic background, may lead a person to

believe that it is wrong to deny civil rights to someone on that basis. Thus, they might support

extending some measure of recognition and rights to gays and lesbians. Beliefs about whether or

not homosexuality is sinful may be more consequential for the question of marriage than for the

question of civil rights because “marriage” in the United States is commonly thought of as much

more than a civil or legal institution. The difference between gays and lesbians wanting marriage

and wanting civil rights may be consequential for people who hold religious definitions of

marriage.

Taken as a whole, the models predicting support for civil unions show that the influences

of attitudes toward same-sex marriage and civil unions are largely the same. This should not be

surprising, given the high correlation (r = .83) between the two variables. In particular, a

person’s underlying worldview regarding homosexuality has similar effects on a person’s

attitudes towards same-sex relationships, whether it is called “marriage” or something else.

However, it is not possible to rule out the possibility that there are substantive differences

between support for civil unions and support for same-sex marriage. The above comparisons also

suggest that, at least for some people, there are some marginal differences between the question

of marriage and the question of civil rights for gay and lesbian couples. In particular, what the
   121  
word “marriage” means to people, over and above the question of civil rights, is not addressed by

this survey. What can these data tell us about the meaning of “marriage” and the characteristics

of people who support civil unions but oppose same-sex marriage? It is to this question that we

now turn.

Why Support Civil Unions but not Same-Sex Marriage?

The comparisons of models predicting support for civil unions with those predicting

support for same-sex marriage has raised the issue of the cultural meaning of marriage. In

particular, the cultural understandings of marriage that are common in the United States might

account for the reasons that personal contact with gays and lesbians and the belief that

homosexuality cannot be changed affect attitudes toward civil unions more than attitudes toward

same-sex marriage. Considering the wedding rituals that are practiced in the United States,

marriage may carry a multitude of diverse meanings related to religion, spirituality, love, unity,

commitment, family, procreation, and community.

Unfortunately, the Pew Research Center questionnaire does not contain any items that

directly address the meaning of marriage for respondents. The meaning of marriage is only

discussed in response to a question asked only of people who said they “oppose” or “strongly

oppose” same-sex marriage. The open-ended question asked respondents for the “main reason”

for their opposition. Interviewers were instructed to code the responses (1-7) according to pre-

determined codes, to probe once for clarification if a person said they thought it was “just

wrong” or “just don’t agree with it” (8), or to enter the response verbatim if they offered some

“other” reason (9). There is no way to estimate the reliability of the coding because verbatim
   122  
responses were not typed for most respondents. Therefore, the results should be interpreted with

caution.

Table 7: Reasons People Oppose Gay Marriage, by Level of Support for Civil Unions

Support for Civil Unions Total

Oppose or Favor or
Strongly Oppose Strongly Favor
Why Oppose 1. Hetero Definition 76.0% (231) 24.0% (73) 100.0% (304)
Gay Marriage of Marriage
2. Religious Objection 88.1% (619) 11.9% (84) 100.0% (703)

3. Not natural / Not 91.7% (132) 8.3% (12) 100.0% (144)

normal
4. It's just wrong / I 90.6% (163) 9.4% (17) 100.0% (180)
just don't agree with it
5. Other 78.7% (111) 21.3% (30) 100.0% (141)

Total 85.3% (1256) 14.7% (216) 100.0% (1472)

Notes: Numbers in parentheses are higher than the actual N because of weights used by the
researchers. “Hetero Definition of Marriage” includes the two codes, “Definition of marriage is
only for a man and a woman” and “Purpose of marriage is to have children.” “Religious
Objection” includes the two codes, “Morally wrong/A sin/The Bible says” and “Against my
religious beliefs.” Because of experiment conducted by the researchers, alternating question
order of support for civil unions and support for gay marriage, the numbers of people supporting
civil unions varies depending upon whether that question was asked before or after the question
about gay marriage. This table presents data from the full sample because the substantive
interpretation is the same. The difference caused by question order is most prominent among
people who justified their opposition to same-sex marriage by saying it is “not natural/normal.”  

Table 7 presents the reasons that respondents gave for why they oppose same-sex

marriage, grouped by whether they said they support or oppose civil unions. To facilitate the

comparison, I recoded the responses. I combined two codes that are both about a heterosexual

definition of marriage (“Definition of marriage is only for a man and a woman” and “Purpose of

marriage is to have children”), and I combined two codes that are both religious objections to
   123  
same-sex marriage (“Morally wrong/A sin/The Bible says” and “Against my religious beliefs”).

In addition, I combined all codes that contained less than 3% of total responses as “Other.”

Overall, the main justifications that people gave for opposing same-sex marriage appear

to be religious: almost half of the respondents (48%) said that same-sex marriage is morally

wrong, a sin, or against their religious beliefs. A comparison of the reasons given by people who

support civil unions with the reasons given by people who oppose both marriage and civil

unions, however, shows important differences between these groups of people. The people who

were most likely to support civil unions while opposing same-sex marriage are people who

justify their opposition using the heterosexual or procreative definition of marriage. In other

words, people may support civil unions but not same-sex marriage if their objection to same-sex

marriage is based primarily on their views of marriage rather than their views about

homosexuality.

Thus, the cultural meanings of marriage may play an important role in shaping

Americans’ attitudes toward same-sex marriage and civil unions. Although the data do not allow

us to test this proposition directly, it appears likely that it makes the distinction between civil

unions and same-sex marriage cognitively, as well as politically, important. Future surveys of

public opinion should include explicit measures of respondents’ views of marriage in order to

assess the role that it plays in shaping attitudes toward same-sex relationships.

Using the demographic, contextual, and attitudinal measures that are available, however,

it is possible to gain some additional insight into the characteristics of people who

simultaneously oppose same-sex marriage but support civil unions. In a separate analysis, I

categorized all respondents who were asked the question on same-sex marriage first as

consistently liberal (supported both same-sex marriage and civil unions), consistently
   124  
conservative (opposed same-sex marriage and civil unions), or ideologically inconsistent

(opposed same-sex marriage but supported civil unions).7 I conducted a series of one-way

ANOVAs and cross-tabs to examine the extent to which the ideologically inconsistent group is

similar to or different from the other two groups.

Table 8: Group Means in Select Demographic and Ideological Variables of Consistent

Opponents, Consistent Supporters, and People who Oppose Same-Sex Marriage but
Support Civil Unions

Item Consistent Ideologically Consistent Tamhane’s T2

Conservative Inconsistent Liberal Mean (I-J) = 0
#
Female 1.49** 1.60 1.58 Liberal-I.I.
Age 49.29** 43.28** 39.60** --
Education 3.96** 4.34** 4.67** --
Political Conservatism 3.39** 3.12** 2.69** --
Smallness of city size 2.75** 2.40 2.43 Liberal-I.I.
Church Attendance 4.26** 3.62** 3.06** --
Not Born-again/Evangelical 1.41** 1.66 1.73 Liberal-I.I.
No Gay Contact# 1.47** 1.23 1.24 Liberal-I.I.
Opinion of Gays and Lesbians 3.26** 4.90** 6.10** --
Homo can be changed# 1.39** 1.21 1.15 Liberal-I.I.
#
Not Uncomfortable 1.56** 1.82** 1.94** --
Not a Sin# 1.00** 1.19** 1.64** --
Undermine Traditional Family 3.31 3.26 1.84** Conservative-I.I.
Good Parents 1.88** 3.02** 3.54** --
Acceptance Good for Country 1.46** 2.13** 2.52** --
** p. < .05 Mean is statistically different from each of the other two groups; indicates non-
overlapping 95% confidence intervals.
#
Dichotomous variable coded 1 and 2. The response category coded 2 is presented as the name
of the variable.
Notes: One-way ANOVA group means presented. Tamhane’s T2 test assumes unequal
variances. The “Ideologically Inconsistent” group mean is indistinguishable from the group mean
of the group in the last column.

The results of the comparison of group means are shown in Table 8. For almost all

attitudinal measures, the ideologically inconsistent group occupied a middle position in between
                                                                                                               
7
There were so few people in the remaining cell of the 2x2 table (people who support same-sex
marriage but oppose civil unions) that I excluded them from analysis here.
   125  
the other two groups, with mean responses outside the bounds of 95% confidence intervals for

each group. The only exception is that  the ideologically inconsistent group did not differ

statistically from consistent conservatives in their belief that allowing same-sex couples to get

married would undermine the traditional family. This suggests that this group of people might

oppose same-sex marriage because of their belief in the “traditional” definition of marriage and

family.

For demographic and contextual measures, the ideologically inconsistent group

resembled the consistently liberal group more than the consistently conservative group.  There

were no statistical differences between the consistently liberal group and the inconsistent group

in terms of gender, size of the city in which they live, and identity as a born-again or Evangelical

Christian; but on all these measures, these two groups were significantly different from the

consistently conservative group. Similarly, people who support civil unions only were like the

liberals but unlike the conservatives in their likelihood of knowing someone personally who is

gay or lesbian. In terms of age, education, church attendance, and political liberalism, the

ideologically inconsistent group occupied a middle position in between the other two groups.

These results show that people who oppose same-sex marriage but support civil unions

more closely resemble liberals in their demographic profiles but hold moderate-to-conservative

attitudes and beliefs about homosexuality. They may be inclined to support the recognition of

same-sex couples and the extension of civil rights to gays and lesbians, but oppose same-sex

marriage because of heterosexual, procreative definitions of marriage and family. In light of this

data, it would be premature to conclude, on the basis of the regression analyses, that the

differences in attitudes toward civil unions and same-sex marriage are only a function of

question order and social context of the interview. However, the similarities between the models
   126  
predicting support for same-sex marriage and those predicting support for civil unions suggest

that the proportion of people who occupy this “ideologically inconsistent” position is relatively

small and that the effects of a person’s “worldview” about homosexuality on attitudes toward

same-sex marriage are much greater than the unmeasured effects of a person’s cultural beliefs

about “marriage.”

Discussion

The quantitative analysis of a nationally representative dataset provides key insights into

the cultural foundations of attitudes about same-sex marriage. First, I compared the predictive

power of a continuous age variable with several different cohort classification schemes and

found support for the hypothesis that age differences in attitudes about same-sex marriage are

actually cohort differences. Cohort classification schemes based on significant changes in the

cultural and political standing of gays and lesbians in the United States—whether individuals

came of age during the Gay Rights, Gay Liberation, or Closet periods—proved to be better

predictors of attitudes about same-sex marriage than the continuous age variable. The strongest

cohort measure, which assumes a person “coming of age” when they turned 18 years old,

remained a significant predictor of attitudes about same-sex marriage, even after controlling for

all demographic, contextual, and ideological variables.

Second, I showed that both dependent variables and eight cognitive and attitudinal

variables are all highly correlated with one another, which suggests that attitudes about same-sex

marriage are part of a relatively coherent worldview about homosexuality. Knowing a person’s

cognitive beliefs about the nature of homosexuality and their attitudes toward gays and lesbians

is of critical importance for understanding their attitudes about same-sex marriage. This was
   127  
demonstrated using ordinal regressions of attitudes about same-sex marriage and civil unions on

these ideological variables.

Third, the ordinal regression results in this chapter provide a general model for predicting

attitudes about same-sex marriage. I argue that attitudes about same-sex marriage and civil

unions are shaped most directly by attitudes and beliefs about homosexuality, which are in turn

shaped by demographic and contextual variables. The explanatory power of most demographic

and contextual variables, including religious and political ideologies, largely disappears once the

more proximate attitudes and beliefs about homosexuality are taken into account. However,

some demographic and contextual variables remain important predictors, even when attitudes

and beliefs about homosexuality are controlled for. Women, members of the Rights Cohort, and

people with weak religious or non-Christian religious identities are more likely to support same-

sex marriage and civil unions, all else equal.

Finally, by comparing the foundations of attitudes about same-sex marriages with the

foundations of attitudes about civil unions, I was able to examine evidence for and against the

possibility that civil unions represent an ideologically viable solution to the controversy about

same-sex marriage. On one hand, the strong correlation between the dependent variables and the

similarities between the models suggest that patterns of support and opposition for civil unions

are no different from patterns of support and opposition to same-sex marriage. The majority of

the quantitative evidence indicates that same-sex civil unions are not a realistic alternative to

same-sex marriages, from a cognitive or attitudinal point of view.

However, other findings from this analysis caution against drawing such a conclusion.

First, the importance of question order as a predictor of support for civil unions suggests that

some people may be willing to accept civil unions as a compromise if they are first allowed to
   128  
voice their opposition to same-sex marriage. Second, beliefs about whether or not homosexuality

can be changed and personal contact with gays and lesbians appear to be stronger predictors of

attitudes about civil unions, suggesting that there may in fact be different foundations of attitudes

about the two dependent variables. Cognitive beliefs about the nature of homosexuality and the

effects of personal contact with gays and lesbians may affect a person’s willingness to extend

civil rights to gays and lesbians, but not be associated with the question of “marriage” per se.

Third, a non-trivial proportion of respondents said that they oppose same-sex marriage but

support-civil unions. These people resemble liberals demographically but hold moderately

conservative attitudes and beliefs about homosexuality. An examination of the reasons that these

people gave for opposing same-sex marriage shows that their cultural definitions of marriage

seem to be decisive: marriage is simply defined as between one man and one woman. There are

thus signs of the existence of unmeasured, substantive differences in people’s cultural definitions

of marriage that may make the controversy over same-sex marriage an entirely different issue

than the question of same-sex civil unions.

Implications for Subsequent Analysis

The quantitative analyses presented here suggest several important starting points for

qualitative analysis, which can clarify and deepen our understanding of the cultural foundations

of same-sex marriage. First, the general model derived in this chapter—that demographic and

contextual variables affect cognitive beliefs and attitudes about homosexuality, which in turn

affect attitudes about same-sex marriage—offers a template for how the qualitative analysis of

interview data should proceed. It suggests that the cultural foundations of attitudes about same-

sex marriage require two separate levels of analysis. At one level, the cultural foundations of
   129  
attitudes are explicit and direct, in which case they can be measured discursively. At the other

level, the cultural foundations are implicit and indirect, in which case they may be measured

associatively.

The analysis presented in this chapter leads us to expect that people’s discourse about

same-sex marriage will largely center on the morality of homosexuality and their attitudes about

gays and lesbians. Because religious identification measures were so important in the

quantitative analysis, and because the question of the sinfulness of homosexuality is so deeply

rooted in religious ideologies, I expect that people’s religious views will structure individual

discourse about same-sex marriage. By contrast, the lack of direct association between political

ideology and attitudes about same-sex marriage suggests that political identification will not be

important in people’s discourse about same-sex marriage.

This analysis also suggests that there are implicit associations between attitudes about

same-sex marriage and various demographic and contextual factors. These demographic and

contextual factors may not be explicit in people’s discourse about same-sex marriage; rather,

they may structure individuals’ discourse in implicit or unacknowledged ways. In particular,

gender, education, city size, political beliefs, and personal contact with gays and lesbians likely

shape the ways that people talk about same-sex marriage. I will also explore how other variables,

not measured in this survey, may be associated with attitudes about same-sex marriage. These

include coming-of-age experiences, gender ideologies, and attitudes about marriage and

relationships. In Chapters 4 and 5, I analyze the patterns of talk about same-sex marriage, using

both discursive and associative analytic strategies to gain further insight into the foundations of

attitudes about same-sex marriage.

   130  
The second set of implications for qualitative analysis has to do with the persistent

influence of cohort membership on attitudes about same-sex marriage. Quantitative analysis

confirms that belonging to different cohorts is a significant predictor of support and opposition to

same-sex marriage. I have argued, theoretically, that the importance of the cohort variable has to

do with the time period during which a person came of age and the effect that it has on a person’s

beliefs and attitudes about homosexuality. The fact that the strongest cohort measure in my

analysis is structured around important transformations in the cultural and political standing of

gays and lesbians in American society lends support to this argument. However, a stronger test

of this argument is warranted. In Chapter 6, I compare similarities and differences in how

members of two cohorts—matched pairs of college students and their parents—talk about same-

sex marriage.

The final set of implications for the forthcoming qualitative analysis has to do with the

questions that have been raised by this quantitative analysis. There are four noteworthy puzzles

that have been raised in this chapter that will be explored in the next three chapters. First, this

analysis showed that people who don’t know or refused to answer whether they thought

homosexuality is a sin were almost as likely to oppose same-sex marriage as people who

answered affirmatively. It is not immediately clear why this would be true. Because qualitative

interviewing does not require a respondent to pick from only two answer choices and because the

interviewer can probe more deeply into people’s moral evaluations of homosexuality, the

analysis in Chapter 4 can help explain this surprising finding. Second, the effect of personal

contact with a friend, family member, or colleague who identifies as gay or lesbian is unclear in

this analysis. Both the literature on the contact hypothesis and the ordinal regressions here show

that the effect of personal contact may be positive or negative. In forthcoming chapters, I will
   131  
analyze the ways in which people talk about personal contact with gays and lesbians to see if any

additional insight may be gained into this puzzle.

Third, this analysis has suggested that people’s attitudes about same-sex marriage may be

predicated upon what the word “marriage” means to people. I was not able to test this possibility

directly in this chapter because there were no survey items that measured people’s beliefs about

marriage. However, it stands to reason that the meaning of marriage would be an important

component of this controversy. I examine this possibility in Chapter 5. Finally, there remains the

puzzle of civil unions: whether or not they represent an ideologically viable compromise solution

to the controversy surrounding same-sex marriage. In this chapter, I have presented evidence

both for and against this possibility. In Chapter 5, I analyze how people talk about civil unions to

shed further light on this puzzle.

   132  
Chapter 4: Talk about Homosexuality and Same-Sex Marriage

Introduction

In this chapter, I analyze how people draw from their attitudes, cognitive beliefs, moral

values, and life experiences about homosexuality in order to talk about same-sex marriage in a

one-on-one interview context. The main goal of this analysis is to describe the variation in how

discourse about same-sex marriage is premised upon discourse about homosexuality. The

analysis is based on the full sample of 97 interviews that I conducted with college students and

their parents, and it builds off the quantitative analysis in the previous chapter. Thinking about

these two analyses in tandem, the qualitative interviews can extend our knowledge of how

people’s understandings of homosexuality shape the ways that they construct and express

opinions about same-sex marriage.

The quantitative analysis in the previous chapter confirmed an important insight from the

public opinion literature: that most gay rights issues are controversial because of differing values,

attitudes, and beliefs about homosexuality. In the case of same-sex marriage, most demographic

and contextual variables influence attitudes primarily indirectly, through the more proximate

attitudes, beliefs, and values about homosexuality. Thus, a deeper examination of how people’s

cultural repertoires with respect to homosexuality constitute foundations for attitudes about

same-sex marriage is warranted. This qualitative analysis will show how this relationship is

manifested in discourse.

Based on the findings from the previous chapter, I expect that specific values, attitudes,

and beliefs about homosexuality largely account for the relationship between religious and

political ideologies and attitudes about same-sex marriage. In particular, I expect people’s moral
   133  
judgments about homosexuality and the affective component of their attitudes about

homosexuality—people’s deep-seated feelings or emotions—to be the most important influences

on discourse. People’s cognitive beliefs about homosexuality (in particular, whether people

attribute homosexuality to biological or social causes) should be of lesser importance, though

still significant. Lastly, I expect the nature of people’s personal contact with gays and lesbians to

be related to discourses about same-sex marriage in some way, though it is unclear from the

survey data what is the nature of that relationship.

Support for same-sex marriage appears to be both ideologically and sociologically related

to positive attitudes toward gays and lesbians, the belief that homosexuality is not immoral, and

the belief that homosexuality is an innate orientation. These attitudes, values, and beliefs are

more likely to be held by people who are young, well-educated, politically liberal, who live in

large cities, and who know gays and lesbians personally. By contrast, opposition to same-sex

marriage appears to be related to negative attitudes toward gays and lesbians, the belief that

homosexuality is a sin, and the belief that homosexuality is a lifestyle choice. These attitudes,

values, and beliefs are more likely to be held by people who are older, more religious, more

politically conservative, who live in small towns and rural areas, and who do not know gays and

lesbians personally.

Among demographic variables, I pay particular attention to cohort location, which I

expect to influence discourses about same-sex marriage primarily through people’s attitudes and

beliefs about homosexuality. The historical record and the quantitative analysis in the previous

chapter suggests that younger people are more likely to support same-sex marriage because of

changes in how homosexuality has been constructed in mainstream American culture. The

college students in this study, born after 1978, are members of the Rights Cohort and thus grew
   134  
up during a period in which homosexuality became increasingly accepted, visible, and normal in

American culture; at the same time, discourses of equality and gay rights supplanted discourses

of gay liberation in American politics. By contrast, the parents in this study, born before 1963,

are members of the Liberation Cohort and the Closet Cohort and thus grew up during a period in

which homosexuality was culturally constructed as a countercultural lifestyle or a mental illness.

Based on the expectations of cohort effects, people’s discourses about same-sex marriage should

reflect the dominant cultural constructions of homosexuality that existed when individuals came

of age, such that younger people are more likely to articulate positive, supportive attitudes about

homosexuality and same-sex marriage, while older people are more likely to articulate negative,

oppositional attitudes about homosexuality and same-sex marriage.

This chapter confirms that supportive and oppositional discourses about same-sex

marriage do appear to fall along these demographic and attitudinal contours, but the chapter also

shows that there is tremendous cultural complexity inherent in people’s discourse about same-

sex marriage and that there are a variety of discourses about same-sex marriage that are not

easily classified as supportive or oppositional. Contrary to the expectations created by the

quantitative analysis, people’s beliefs, attitudes, values, and experiences regarding

homosexuality frequently fail to cohere with one another, and they interact with cohort location

and religious and political ideologies in a variety of ways. In particular, people’s cognitive

beliefs about the nature of homosexuality tend to be multi-faceted or subject to the power of

suggestion, meaning that they may bear very little relation to people’s attitudes about same-sex

marriage. Similarly, the question of people’s moral values regarding homosexuality and same-

sex marriage is multi-dimensional, such that even people who believe that homosexuality is

immoral do not necessarily oppose same-sex marriage or believe that their personal moral values
   135  
constitute an acceptable reason to do so. An additional level of discursive complexity occurs

when people’s cohort locations cross-cut their political and religious ideologies, effectively

pulling them simultaneously in opposite directions. As a result of this complexity, many

Americans articulate opinions about same-sex marriage using a variety of middle-ground

discourses. On the basis of this evidence, I argue that a substantial proportion of Americans have

moderate, mixed, or weak attitudes about same-sex marriage and that polarization is likely to

occur primarily in the context of political mobilization or survey measurement, where demands

are placed on individuals to choose sides.

In sum, this chapter demonstrates that, even with our analytic attention restricted to the

subject of homosexuality, there is such a degree of complexity in people’s understandings of

homosexuality and their discourses about same-sex marriage that attitudes about same-sex

marriage cannot be understood in simple terms of support and opposition. The use of middle-

ground discourses by members of both cohorts suggests that people do not necessarily have

coherent worldviews regarding homosexuality and that people find ways to discursively manage

the apparent tensions and contradictions within their belief systems when talking about same-sex

marriage. That these tensions and contradictions appear to be sociologically rooted in cross-

cutting cohort memberships and religious and political ideologies is significant.

In this chapter, I first describe how people’s discourses about same-sex marriage largely

revolve around the issue of homosexuality more generally. Second, I describe how people’s

understandings of homosexuality are composed by different cognitive beliefs, affective attitudes,

moral values, and life experiences about homosexuality, and I analyze how they are used to talk

about same-sex marriage. Third, I analyze unambiguously supportive and oppositional discourse

regarding same-sex marriage to show how these discourses are rooted in different attitudes,
   136  
beliefs, cohort locations, and political and religious ideologies. Finally, I analyze the middle-

ground discourses—patterns of talk about same-sex marriage that are neither completely for nor

completely against same-sex marriage—to show how complex discourses about same-sex

marriage appear to be rooted in the cross-cutting influences of cohort location and

political/religious ideology with respect to homosexuality.

The Social Significance of Talk

Before I present the results, I must describe the specific discursive context in which this

analysis is based and the sociological and epistemological significance of talk in general. The

specific discursive context of this analysis is set by the fifth and sixth sections of the interview

guide, a series of questions and probes about same-sex marriage and about homosexuality. Most

respondents did not raise the issue of same-sex marriage before I did; thus, the discourse

described in this chapter occurred after a lengthy conversation about the respondent, their

background, their media consumption habits, and their views about marriage and relationships.

Immediately prior to the question in which I raised the issue of same-sex marriage, I had asked

for people to describe their feelings about marriage, cohabitation, premarital sex, divorce, and

the status of the institution of marriage in the U.S. today. Interviewees had therefore been primed

to talk about same-sex marriage in terms of a larger discussion about marriage and a variety of

phenomena that have been alleged to hasten its decline as an important social institution. This

topic, how people’s discourse about marriage is related to their discourse about same-sex

marriage, will be discussed in Chapter 5.

I typically introduced the issue of same-sex marriage by saying, “One of the issues that

has been particularly controversial in recent years is the issue of same-sex marriage. Have you
   137  
heard much about that issue?” After asking this question, the order in which topics in the

interview guide were discussed depended on how the informant responded to this question.

Many respondents answered the questions narrowly, as if interpreting my question literally; other

respondents seemed to interpret the intent of my question broadly and answered the question by

talking about homosexuality more generally.

For the respondents who answered the question narrowly, I followed the interview guide

roughly in the order in which topics are listed. I asked people why they thought the issue was

controversial, why some gays and lesbians wanted the right to marry, why people might be

opposed, and, if necessary, whether or not they had a specific opinion about the issue. I then

posed a series of hypothetical scenarios to the informants, asking them to imagine the effects of

legalizing same-sex marriage in the United States, legalizing civil unions in the United States,

and how the controversy might be resolved.

After this section, I moved into the section on homosexuality. I first asked about people’s

personal contact (both in their daily lives and in the media) with gays and lesbians and their

memories about their first contact with homosexuality. I then asked a series of questions

requiring informants to define homosexuality and bisexuality according to how they understand

it (cognitively and morally), and I asked about equal rights for gays and lesbians. Finally, I asked

participants that were supportive of same-sex marriage about legal recognition for same-sex

couples who identify as bisexual or heterosexual.

For respondents who interpreted my questions more broadly, these topics were covered in

an order dictated by the flow of the conversation. Not all questions were asked of all individuals,

and the exact content of each question was tailored to the previous answers that the informant

gave. The purpose of this was to maintain the flow of conversation and to get the most insight
   138  
into these topics out of each interview, given the person’s views and experiences. While the

interviews are not identical, they are similar enough that data from different interviews may be

analyzed and interpreted similarly. The discourse analyzed in this chapter, though from a variety

of different respondents, can be interpreted as arising from a common social context and

therefore ontologically comparable.

With regard to the sociological and epistemological significance of talk more generally, I

accept the general proposition held by many social scientists that informal talk, such as the

discourse analyzed below, carries broad public and political significance and reflects the

characteristics of dominant social structures and institutions (Arendt 1998; Cohen 1999; Eliasoph

1998; Foucault 1972; Foucault 1978; Gamson 1992; Perrin 2006; Walsh 2004). Moreover, as I

described in Chapter 2, I also argue that the patterns of talk analyzed here are the manifestation

of the cultural and cognitive processes of attitude construction. While I do not hold the view that

the interview data represent the informants’ “true” attitudes in some essential sense, I do believe

that any social or public significance of attitudes is inseparable from their manifestation in

discourse. The same informant might construct an attitude differently when engaged in an open-

ended conversation than when confronted with four answer choices to choose from in an opinion

survey. Neither expression is more authentic than the other, and both have public and political

significance. The discourse analyzed in this chapter is thus more than a set of isolated individual

conversations; it shows how homosexuality and same-sex marriage are culturally constructed in

American political imaginations.

Centrality of Homosexuality in Discourse

   139  
Just as the public political controversy surrounding same-sex marriage appears to be

based upon the controversy about homosexuality more generally, in my interviews, people talked

about same-sex marriage primarily in terms of how they felt about homosexuality. People’s

discourse about same-sex marriage revolved around their moral evaluation of homosexuality and

attitudes toward gays and lesbians.

It was common for informants to immediately interpret my initial question about same-

sex marriage to be a question about how they felt about homosexuality. Whether consciously or

unconsciously, many individuals simply changed the subject from same-sex marriage to

homosexuality. This was true for both people who were supportive of same-sex marriage and

people who opposed same-sex marriage.

Q: One of the issues that’s been fairly controversial in recent years has been the issue of

same-sex marriage, which you sort of mentioned in passing… Have you heard much

about that issue?

R: Um, yeah, probably what’s in the media. Um, my belief system is I don’t think

anybody chooses to be homosexual. Um, I, my God doesn’t condemn people for being

homosexual. I don’t think people should be discriminated against by, you know, systems.

So if the means to not be discriminated against is allowing same-sex unions or marriages,

then I would have to say I would support it, you know. (Frances, age 45)

Q: The issue of same-sex marriage has become fairly controversial in recent years. Have

you heard much at all about that issue, like from the media or at church or?
   140  
R: Oh yeah. Yeah, all the time. And it’s a distortion, I think. It’s a distortion of the way

God originally made things… It’s not the way God originally intended for man and

woman to be. (Stan, age 59)

In both of the quotes above, the respondent talks about homosexuality more generally rather than

about the specific issue of same-sex marriage. An opinion about homosexuality was offered by

these participants as a culturally satisfactory response to a question about same-sex marriage. I

interpret the fact that many people responded to my initial question about same-sex marriage by

volunteering their feelings about homosexuality to mean that there is an implicit cultural

understanding that what is controversial about same-sex marriage is the issue of homosexuality.

Most participants did not immediately change the subject like the two examples above

illustrate, but only after subsequent questions about why they thought the issue was controversial

or when asked about reasons that people hold certain opinions about same-sex marriage.

People’s answers to these questions confirm that they understood the controversy about same-sex

marriage to be primarily about homosexuality:

Q: Why do you think the issue is so controversial?

R: I think most of it is because people aren’t willing to see that even if someone’s gay

that they’re still a good person or if someone is a lesbian that they’re still a good person. I

think probably most of it is religious because it’s frowned upon in religion. (Dennis, age

19)

Q: What reason do you think someone who is opposed to same-sex marriage would give

to explain why they’re against it?

   141  
A: Well, because the thought has been instilled in our heads from birth that gay

relationships just isn’t right. And I guess if being mis-informed and under-informed about

certain things, that makes them react that way. Like I would never understand what it

feels like to be gay or why to be gay. But I don’t knock them either. If you feel like that’s

what you want to do… (Katrina, age 25)

In both of these examples, the students highlight gay and lesbian identities and homosexual

relations as the central issue in the controversy.

While this may seem obvious, it is sociologically significant that this complex issue, for

many people, ultimately boils down to a single question of the rightness or wrongness of

homosexuality. While people’s attitudes about same-sex marriage depend upon many different

factors—including political and religious ideologies, definitions of marriage, and variables like

age, education, gender, and area of residence—discourse about homosexuality is of primary

importance in accounting for the different views expressed by different people. Pursuant to the

expectations of the cohort analysis advanced above, changes in attitudes about same-sex

marriage over time are likely to be driven primarily by changes in people’s understanding of

homosexuality.

Of course, what homosexuality means to people is not necessarily the same. Almost all

informants defined homosexuality as a sexual attraction between two people of the same sex; but

beyond this common-sense definition, informants differed significantly in how they talked about

homosexuality. It appeared as though supporters and opponents of same-sex marriage talking

about homosexuality were in fact talking about different things. Thus, the likelihood of
   142  
supporting or opposing same-sex marriage depends upon the cultural definition of homosexuality

that individuals bring to bear upon the question.

While it is possible that people have relatively coherent worldviews about homosexuality,

there are at least three analytically distinct components of people’s definitions of homosexuality:

cognitive beliefs, affective attitudes, and moral values. In analyzing people’s discourse, it is

important to distinguish among these three different kinds of elements in their belief systems

about homosexuality. Cognitive beliefs about homosexuality consist in what people think

homosexuality is and what causes homosexuality. Affective attitudes about homosexuality

consist of a range of positive and negative feelings about homosexual behavior and people who

identify as gay or lesbian. Moral values about homosexuality consist in beliefs about the extent

to which homosexuality can be classified as right or wrong. Beyond these three elements in

people’s belief systems, the nature of contact and life experiences with gays and lesbians is also

an important factor that shapes people’s understanding of homosexuality.

Evidence from my interviews suggests that there is no ideological imperative for certain

cognitive beliefs, affective attitudes, moral values, and life experiences to correspond with one

another. My interviews contained many different combinations of these four types of cultural

elements. The wide variation in the combinations of cognitive beliefs, affective attitudes, moral

values, and life experiences that people drew from in their discourse was an unexpected finding.

Although there may be patterns of association that are more likely than others, these patterns of

association are not due to any inherent ideological structure. In terms of their association with

attitudes about same-sex marriage, moral values and affective attitudes about homosexuality

appear to be of utmost importance, whereas associations with cognitive beliefs and life
   143  
experiences are less clear. Below, I describe the variation in how people use each type of cultural

element in order to talk about homosexuality and same-sex marriage.

Cognitive Beliefs about Homosexuality

Defining homosexuality is an act of tremendous political importance. Whether a person

defines homosexuality as a mental illness, a lifestyle choice, a phenomenon caused by natural or

genetic causes, or a phenomenon caused by social upbringing, has consequences for whether

gays and lesbians are defined as patients, deviants, members of a minority group, or victims of

circumstance. Because Americans understand that homosexuality has historically been treated as

negative by mainstream society, each set of definitions suggests a course of action: patients

should be cured, deviants should be corrected, minorities should be granted equal rights, or the

less fortunate should be tolerated. Thus, the act of defining homosexuality and gays and lesbians

carries with it a raft of political implications.

It is perhaps for this reason that the attribution of homosexuality has received so much

attention among scientists and in mass media. Defining homosexuality affects the terms of

political debate and the merits of different courses of actions. For example, if people are born

gay or lesbian, then like racial and ethnic minorities, they should be granted equal rights and

discrimination against them should be outlawed. By contrast, if people simply choose to be gay

or lesbian, then like alcoholics or gamblers, their behavior choices should be corrected or

discouraged rather than protected by “special” rights and privileges. For those who are politically

motivated, progressives gain an advantage in the first instance, while conservatives gain an

advantage in the second. It is therefore unsurprising that public opinion studies find that people’s

definitions of homosexuality are associated with opinions about same-sex marriage.

   144  
In my interviews, however, I found that the act of defining homosexuality is a far more

complex and difficult task for respondents than we might otherwise expect. Beyond the simple

observation that homosexuality is a sexual attraction between people of the same sex, many

informants had difficulty articulating what they think homosexuality is and what causes it. While

some people gave a single, clear answer, many people did not know, and many others thought

that there were multiple causes or that the definition depended upon the person.

Some people gave very simple answers to what they thought causes homosexuality, and

their answers were closely related to their views on same-sex marriage. Consider Amanda, a 19-

year old community college student, who was wearing a pink Obama t-shirt as we talked at a

table in a café. She brought up the issue of same-sex marriage before I did, and I asked her what

she thought about it:

R: I think it’s fine. I think gay marriage should be accepted.

Q: Yeah?

R: Mm hmm. (affirmative)

Q: Why?

R: Because I don’t think being gay is a choice. It’s just the way the brain is in certain

people, and why should they be, why shouldn’t they get the same benefits as anyone

else? (Amanda, age 19)

Here, Amanda used her definition of homosexuality as a justification for supporting same-sex

marriage in a manner befitting her strong liberal political views.

   145  
Opponents of same-sex marriage also used simple, straight-forward definitions of

homosexuality to justify their views. For example, Sarah, a 60-year old mother who works part-

time in retail, described her views about homosexuality this way:

I think homosexuality is a lifestyle choice, not that you are born to. I think that you can

make a choice whether you want to live with someone of the same sex or whether you

want to try and make a marriage. I mean, I think a lot of us can go one way or the other,

and I think it’s the choices we make, not the, not what we’re born to, I think. (Sarah, age

60)

Sarah had heard arguments from people that she knows who identify as gay, and she disagrees

that they were just born that way. She sees no reason why someone who identifies as gay

couldn’t choose to marry someone of the opposite sex. Logically, she doesn’t support gay

marriage because gays and lesbians can already get married—to someone of the opposite sex.

It is worth pointing out that Sarah’s implicit reference to bisexuality in the quote above—

“I think a lot of us can go one way or the other”—was occasionally invoked as evidence that

being gay or lesbian is a lifestyle choice. When I asked her specifically about bisexuality and

why people would identify themselves that way, she said:

Well, I think that their experiences, their life experiences have probably influenced them

one way or the other, and I think that’s what makes the difference. Just like all of us can

make choices in our career and in our marriage and whatever, you can make a choice in

what kind of lifestyle you choose also. You don’t have to choose to go with something

just because you have feelings in that direction. (Sarah, age 60)
   146  
Here, Sarah uses analogies to careers and marriages to explain how people often have mixed

feelings or could go either way in relationships, and that what matters is the choice that you

make.

Of course, a belief in bisexuality could also lend itself to tolerance and support for same-

sex relationships—because “for the most part everyone’s a little queer,” as Alan, a 22 year-old

student put it. If everyone were inherently bisexual, it could eliminate the stigma of

homosexuality. However, the belief could just as easily increase opposition to homosexuality and

same-sex marriage because there would be a greater perceived threat of engaging in what some

people view as immoral sexual behavior. Fluid or complex understandings of sexuality do not

lend themselves easily to a single political course of action; one could draw either liberal or

conservative conclusions from complex definitions of homosexuality.

In my interviews, complex definitions of homosexuality were common. Most informants

did not have clear, simple definitions of homosexuality that were ideologically implicated in

liberal or conservative viewpoints. Some informants thought of homosexuality in a single

individual as being multi-causal, while others thought that there are different causes of

homosexuality for different people. For those who thought homosexuality is multi-causal, it is

worth pointing out that there is no a priori reason why homosexuality could not simultaneously

have genetic, social, and behavioral components. Different informants advanced different

combinations of the three definitions of homosexuality: nature, nurture, and behavior.

Jonas, an 18 year-old freshman, defined homosexuality as being caused by a combination

of biological and environmental factors, but he denied that people chose to be gay:

R: I think it’s not a choice. I would never say that a person wakes up one morning, you

know, a pimply 14 year-old kid and says ‘I want to be a part of a despised minority in this
   147  
country,’ you know…. I think that people don’t have a choice in who they are attracted

to. I think it’s something that was raised, I think. I think it’s a biological thing and it’s an

environmental thing. That nature versus nurture argument.

Q: You think it’s both?

R: I think its, I think it’s a combination of both. (Jonas, age 18)

Although it takes Jonas a while to formulate a clear opinion about what causes homosexuality, he

eventually settles on an argument that both biological and social-environmental factors cause

homosexuality. He brackets the question of behavioral choice—whether or not someone chooses

to act on their feelings—as being something different. For him, being gay is a sexual orientation

rather than a behavior.

By contrast, people who believe that homosexuality is a behavioral, lifestyle choice must

confront the question of why certain people choose that lifestyle while others do not. Here, a

multi-causal attribution of homosexuality includes behavioral factors, in addition to biological or

environmental factors. Evan, a 24 year-old student, considered the possibility that people might

be born gay, but he ultimately thought that it was a combination of behavioral choice and social

upbringing that made people identify as gay:

I think the majority of gay men are, it’s a choice. And I’ve heard a lot of stories where

gay men are just gay because, not to undermine, but like usually because like they hate

their dad or they have daddy issues. Mommy issues. Something happened to them as a

kid that affected them and now they’re gay. And my friend [name] who just came out, his

kid brother came out of the closet, too… Little brother is just following big brother’s
   148  
footsteps. And not to demean them or undermine them, but I just, that’s kind of how I see

it. Or maybe I’m wrong. I could be 100% wrong. (Evan, age 24)

Evan drew from stories about people he knew in order to support his opinion that people who

identify as gay choose to be gay because of bad experiences that they have had as a child or

because of the influence of their peers. However, he also freely admits that he doesn’t know if

he’s right. He offered this opinion without feeling entirely confident in it.

Some informants thought that the reason that a person identifies as gay varies from

person to person. From this point of view, some people may be homosexual because they are

born that way, while others may be homosexual because of environmental factors, and still

others choose to engage in homosexual activity in order to be rebellious. For example, Barrett, a

19 year-old community college student, had such a view. His support for same-sex marriage and

complexity of his views on homosexuality stood in stark contrast to my expectations, given his

conservative Christian background and love of sports like football:

I think the right to marriage should be given to everyone, whether they’re biologically,

genetically gay or if they just are psychologically gay. That doesn’t really make much of

a difference. The only thing I think a guy or a gal should not be gay, like the only reason

would be to get attention, which unfortunately I believe does happen. (Barrett, age 19)

Later in the interview, I asked Barrett to say more about what he thinks cause people to be gay:

R: Now there is probably sub-categories of like psychologically gay, you know, in their

mind they’re gay, so they are gay… There probably are some cases where there is a gene

that just sets ‘em off for the same sex….

   149  
Q: So do you think people are gay for different reasons? Would that be an appropriate

way to interpret what you’re saying?

R: Different causes. The result is all the same. There isn’t this type of gay person and,

versus this type of gay person. They’re all gay, but the initial cause of it… (Barrett, age

19)

Barrett’s views about the attribution of homosexuality are not easy to classify, but he describes

three different possible causes of homosexuality. His only negative evaluation of homosexuality

is reserved for people who do it to “get attention.” Several students that I interviewed, both male

and female, observed that women in particular sometimes engage in homosexual activity in order

to get attention from men—because men are likely to find female (and not male) homosexual

activity erotic.

Finally, it is clear that many of my informants simply did not know what caused

homosexuality and were attempting to formulate answers in response to my questions. Some

people did not know simply because they didn’t want to think about it. Some informants seemed

uncomfortable answering my questions, and it seemed as though they would rather not think too

much about homosexual activity. Sheila, for example, a 50 year-old conservative Christian,

struggled to articulate an answer to my question about the definition of homosexuality:

Q: What do you think homosexuality is? Like, if you were going to try to define it?

R: (Long pause) I, I, I don’t know. I just don’t understand how someone could be, have

those feelings and do those things with the SAME gender. I just (laughs), you could have

a relationship with the same gender, friendship or whatever; why does it have to go to

that next level? I just, I don’t get that. (Sheila, age 50)
   150  

Other people did not offer clear definitions of homosexuality because they did not think they

knew enough to offer an answer. They did not feel as though they had enough factual knowledge

about what the scientific studies were showing. Matthew, a 51-year old liberal Democrat, did not

feel comfortable saying what he thought homosexuality was because of lack of information.

However, as a result of my continued questioning, he seemed willing to try to formulate an

opinion for me:

Q: Um, do you think people choose to be gay? Like do you think homosexuality is a

choice?

R: That one I’ve thought about, and that one, I can’t, I can’t really answer because I don’t

have enough information on that. Choosing, I… in some ways it could be, and some ways

it can’t. I just, I can’t answer that.

Q: Okay. Um, do you think homosexuality is influenced by, do you think people are sort

of born that way, or do you think people are, it has to do with the way you were raised,

or, do you have any thoughts about that?

R: I think it’s, I think some of it might be influence. Uh, the way that they see something,

that they might be attracted to, or something. Uh, yeah. (Matthew, age 51)

After an initial refusal to say whether or not he thought a person chose to be gay, Matthew

attempted to formulate an answer to my follow-up question. However, the lack of substance in

his answer might indicate that I was giving him a question that was too demanding.

For informants, like Matthew, who did not have strong opinions about the causes of

homosexuality, the power of suggestion of my questions is evident. Sometimes the mere mention
   151  
of one possible cause of homosexuality would lead an informant to agree. It would be a mistake

to think that this person holds this opinion; rather, I argue that many people do not feel qualified

to offer an opinion about what causes homosexuality and are thus willing to agree with the

suggestion of an interviewer. Due caution should be taken when interpreting survey or interview

questions on the cognitive beliefs about homosexuality.

Because of the variety of different ways in which people responded to my questions

about their definitions of homosexuality, I argue that people’s cognitive beliefs about

homosexuality are less clearly defined than politically-interested parties or survey researchers

would like them to be. Forcing survey respondents to choose only one cognitive definition of

homosexuality is inadequate for people who believe that homosexuality can mean multiple

things, and it likely encourages many people to offer answers that they do not necessarily

believe. Many people say that they do not have the knowledge to offer an opinion about what

causes homosexuality, and asking them to do so proved to be a cognitively demanding and

difficult task.

Affective Attitudes

In addition to different cognitive beliefs about homosexuality, people expressed affective

attitudes about homosexuality when they talked about same-sex marriage. The feelings that

people have towards gays and lesbians and about homosexuality shape the ways that they

express themselves when they were talking about same-sex marriage. The continuum of feeling

evident in people’s discourses range from extremely negative to extremely positive.

Those individuals who had the most negative attitudes toward gays and lesbians appeared

almost overwhelmed with disgust about homosexuality. I had been talking with Sheila, a 50
   152  
year-old conservative Christian, in a café for over an hour about her life and her feelings about

marriage, divorce, cohabitation, and premarital sex. She espoused conservative views about these

subjects, but she was at ease talking about the reality of divorce, cohabitation, and premarital

sex. When I brought up the issue of same-sex marriage, however, a deep frown swept across her

face, and the tenor of conversation changed noticeably. Whereas before she had expressed

herself in an articulate and matter-of-fact manner, it now seemed as though the intensity of her

feelings about homosexuality made it difficult for her to speak:

Q: One issue that has been relatively controversial in recent years is the issue of same-sex

marriage. Have you heard much about that issue?

R: I’m against it. It makes me sick. (whisper) I just don’t see it, I don’t understand it, I, I

mean, now if one of my children came to me and said (pause), I don’t know. I don’t, I

think my husband would disown them. I don’t know that I could do that and I don’t know

that I would have the relationship that I would have with them now. I just, I just don’t

understand it, I guess. How can you…? (Sheila, age 50)

Not only does Sheila immediately change the subject from same-sex marriage to homosexuality,

but she immediately expresses her disgust. She then imagines a nightmarish scenario that one of

her children might be homosexual, but she cannot even bring herself to say the word “gay” or

“homosexual.”

Such a degree of negative feeling was rare in my interviews. Others with negative

feelings typically did not express them so strongly. Many people with somewhat negative

attitudes expressed themselves by distancing themselves from homosexuality. They did not

morally condemn homosexuality, but they made it clear that they did not feel comfortable with
   153  
homosexuality. Consider, for example, what Elaina, a 22 year-old senior, had to say about same-

sex marriage:

I don’t agree with that lifestyle but I’m not, I don’t discriminate against it. I don’t, I don’t

know, like it’s hard to explain how I feel about the whole situation… Like I just don’t,

me personally, I hope that doesn’t happen, but then again I would be okay if it did

happen. Like I’m not going to down people if they do it, that’s their decision. I, just as far

as that whole thing, I just feel like I don’t like it being thrown in my face all the time,

‘cause just like even though I, you know, am a heterosexual, I don’t feel like it’s

necessary that I throw that in your face all the time, too. (Elaina, age 22)

In the above quote, Elaina says she has a hard time explaining how she feels about the same-sex

marriage controversy. By comparing it with the talk of other informants, we can observe a latent

negative attitude towards homosexuality that she expresses in the form of not wanting to see

homosexuality “thrown in my face all the time.”

A large number of informants expressed, in one way or another, that they did not like

seeing homosexuality in public. One should not assume that all such sentiments expressed are

negative attitudes about homosexuality, because many people expressed dislike for public

displays of affection of any kind, whether between same- or opposite-sex couples. Older adults

in particular sometimes commented on how different standards of acceptable sexuality are today,

compared with when they were growing up. However, some expressions of dislike for public

displays of homosexuality were based on negative attitudes towards gays. When individuals

expressed dislike for flamboyant styles or behaviors among gays and lesbians—as opposed to

gays and lesbians who seemed just like “ordinary” people—the implicit negative attitude is clear:
   154  
they don’t want to be reminded of homosexuality and would rather not think about it. For

example, talking about people he has had contact with at work, William says:

It’s not that I have that much experience with the gay, you know, the alternate reality

individuals or the alternate lifestyle individuals. I just basically, I know a few of them,

and you know, as long as they’re acting fairly normal, I’m not that opposed to them. It’s

the ones that are the extreme personalities that I have more issues with…. If you’re gonna

be dealing with that type of deal, it’s I don’t want to know about it, don’t show, don’t tell,

don’t come next to me with that type of stuff. (William, age 53)

Describing the stereotypical portrayal of gays, Michael sheds more light on what this means:

If you gonna portray yourself as a businessman or a doctor or something and still be

homosexual, you can be a homosexual doctor. But once you get out there, it’s the

difference between homosexual businessman and homosexual flamboyant. That’s who

they portray. They portray the flamboyant guy that’s out there with the thong, showing

his ass, the boy shorts on… that’s the stereotype. (Michael, age 20)

Many negative attitudes about homosexuality are likely based on the stereotype of the

flamboyant gays and lesbians and well-publicized images of gay pride parades. By contrast, gays

and lesbians whose sexual identity is invisible elicit fewer negative comments.

Among people who expressed positive affective attitudes about gays and lesbians, it was

most common for people to express them by denying the legitimacy of opposition to

homosexuality. No one, for example, said that it was better for someone to be gay than straight,

or that there was anything that was particularly good about homosexuality. Rather, most people
   155  
expressed positive affect by arguing against the social stigma against gays and lesbians. For

example, Terrence, a sophomore, said:

I don’t really see why this has to be a big deal with everybody, you know. It’s like two

people love each other, they want each other, you know, to be secure for the rest of their

lives…. It’s not like we’re saying everyone should get married, or people should marry

animals and all this weird stuff. It’s just so, uh, like a guy and a guy want to get married,

that shouldn’t be a big deal if that’s how they are. If they’re gay and they’re in love, let

them get married. That, I don’t see, it’s just, it’s discrimination, and we should have, I

thought we did away with it all in the 60s with the civil rights movement and we didn’t

have to go through any of that shit again…. We have bigger things to worry about. Life is

too short to have to tell people what they can’t do. (Terrence, age 19)

Similarly, Nate explains that being supportive of same-sex marriage is not about encouraging

people to be gay, but about encouraging people to be accepting and tolerant of gays and lesbians:

R: I don’t think it should be encouraged. I just don’t think it should be discouraged. Like,

you shouldn’t say, “Hey kids, like men,” or something like that.

Q: “Go be gay!”

R: Yeah. You don’t want campaigning for it because it gets to a matter of conversion, and

then that’s just going to drive people apart and hate each other, and that doesn’t solve the

problem at all. I just think it should be accepted. (Nate, age 19)

The few informants who expressed positive affect about homosexuality without resorting

to refuting the negative social stigma did so by expressing positive views about the expression of
   156  
love in general. For example, in response to my first question about same-sex marriage, Laura, a

49 year-old graphic designer, expressed her support for same-sex marriage in terms of her

positive attitudes about love, regardless of sex:

Q: One issue that’s been relatively controversial in recent years is the issue of same-sex

marriage. Have you heard much at all about that issue?

R: Yeah. I’m very pro—you can’t tell someone how to love. And I don’t think that

because two men want to get married that makes them un-human. How do you label

affection, caring, and love just by body parts? I do not think so. I’m very pro. Marry who

you want to marry. (Laura, age 49)

For individuals like Laura, the sex composition of a loving couple is irrelevant; what matters is

that people are expressing love and affection for another person. In general, people who seemed

to think about homosexuality in terms of the feelings of love and affection that sex represents

tended to have the most positive attitudes about homosexuality.

Life Experiences with Gays and Lesbians

The fact that many people react negatively to the image of the stereotypical gay man, and

the fact that so many people say that they don’t want to have homosexuality “thrown in their

faces,” calls our attention to the relationship between attitudes about homosexuality and the

nature of life experiences and personal contact that people have with gays and lesbians. Several

decades of research on the contact hypothesis show that personal contact with members of

minority groups increases tolerance for extending civil rights and liberties to those groups, but
   157  
that the effect only occurs under certain conditions (Allport 1954; Herek and Glunt 1993; Lee,

Farrell and Link 2004; Lemm 2006; Sigelman et al. 1996; Sigelman and Welch 1993).

The ways in which people talk about their life experiences with homosexuality shows

that people’s attitudes about same-sex marriage are related to their personal contact with gays

and lesbians, but that whether it increases or decreases their support for same-sex marriage

depends on the nature of the contact. In other words, exposure to homosexuality and people who

identify with gays and lesbians can cut both ways.

Many people use the fact of significant personal contact with gays and lesbians as a

reason that they support same-sex marriage. Informants often immediately mentioned knowing

gays and lesbians personally when I asked them about same-sex marriage, implying that personal

contact is a culturally legitimate justification for supporting same-sex marriage:

Q: As you know, the issue of same-sex marriage has been one of these issues that has

been brought up a lot.

R: Yeah. Some of my best friends are gay. Some of the people I go down to Texas with,

it’s a wonderful gay couple, and stuff like that. I know people who are gay. I don’t give a

shit…. I honest to God don’t care. I see it as a civil rights thing. You can’t restrict it to

one group or another. Let them get married. (Gary, age 26)

Not only do many people offer personal contact as a reason that they support same-sex marriage,

but people also typically report positive attitudes in relation to people that they know who are

gay or lesbian. In talking about her best friend, who is gay, Betsy describes how getting to know

gay people leads to more positive attitudes:

   158  
It was relatively, right after high school, and it wasn’t just him. I met other [gay] people,

too. And of course, then the more people you are, associate with, the more you meet, too.

I became very strongly pro-gay, I guess I should say, if that makes sense, after I became

friends with them because then I’m like, “You don’t know my friend, don’t be talkin’

crap about my friend…” (Betsy, age 24)

People who report that their attitudes about homosexuality have changed over the course

of their life frequently attribute their more positive attitudes to personal contact. When asked to

explain why he thought his attitudes toward gays and lesbians had changed, Chris said:

I think I was just kind of ignorant about it. It was one of those things that, when you were

growing up, I mean, it was unheard of, you never heard of anybody doing it… And then,

you know, all the sudden, you live nine months with a person, you have the chance to

kind of experience it. Not personally, but be around it. All the sudden, it’s just like, “Oh,

okay, I kind of understand now. (Chris, age 22)

One reason that personal contact may be associated with positive attitudes is that a person might

develop empathy for the individual, and by extension, others that they classify as being in that

group:

Q: Do you think your feelings about homosexuality have changed over the course of your

life?

R: In the sense that I realized they have a lot—you think everybody’s equal, and because

of [name], I realized that’s not true…. I don’t think I was aware of the struggle ‘till, even
   159  
in, even in college before I met [name], I never, it never consciously played on me. Until

you have a friend that’s, you know, struggling, it’s hard to be empathetic. (Mary, age 53)

The effect of personal contact with gays and lesbians also extends to how people define

homosexuality. Most common are reports of people coming to believe that homosexuality is not

a choice because of contact with gays and lesbians. For example, Ariel reported how her

brother’s struggles with his homosexuality convinced her that homosexuality is not a choice:

I think people are ignorant. And that’s why I learned so much, because I was too and

didn’t understand it. So anything I could get, I was reading or I was watching just to help

me. Because I know he wouldn’t have made a choice to live like that. Nobody would.

Why would anybody want to do that to themselves? But I think it’s great that people are

learning and it’s being talked about. (Ariel, age 47)

Similarly, Jason recalls how people he has known seemed to be gay even when they were very

young:

I believe it’s something you’re born with because I’ve known people like before they hit

puberty and before the age of sexual maturity that you just kind of thought, “oh, they

might be gay, and then, but yeah, then they turn out to be gay. So yeah, I don’t think it’s a

choice; I think it’s something you’re born with. (Jason, age 25)

Thus, personal contact with gays and lesbians often convinces people that homosexuality is just

inherent in a person, that it is not something that they consciously choose.

   160  
However, not everyone who has significant personal contact with gays and lesbians

reports more liberal attitudes, values, or beliefs. To the contrary, many people’s experiences with

gays or lesbians have been negative. For example, Harvey reports negative reactions to his

contact with people that he identifies as homosexual:

We went to San Francisco, and I was like, I went to like a Starbucks, and like, I saw this

guy, I saw this girl in line, but she looked like she had facial hair. I kind of thought it was

a guy in a dress, and I was just like, “what?” I was like, “what is that?” Or you know, like

I’d be on [the military] base, and I would get approached by like a guy who’s gay, and

he’d really try to talk to me, and I’m sitting there like, “this guy’s gay.” (Harvey, age 23)

Some people, like Harvey, simply do not get over the person’s sexual orientation that they

dislike. Other people, especially religious conservatives, interpret life experiences with

homosexuality as evidence that homosexuality is a lifestyle choice that is wrong and that should

not be accommodated by society:

Coming from the dance field, my dance partner was gay. I saw it from the inside. You

can love the person, and I did love him very much—he was a good person—but it was a

weakness in him, and it opened up doors to a lot of other areas in his life that I don’t

think would have been opened up to if he would’ve not been in that lifestyle.... It’s not,

“I’m struggling with this, and I want help.” It’s, “I enjoy it and I have no intention of

changing.” That’s where I have the problem. (Dana, age 48)

   161  
In this quote, Dana interprets her former dance partner’s homosexuality as evidence of a deep

social problem because it led to lifestyle choices that she disagreed with and because he saw

nothing wrong with homosexuality.

Thus, people with negative attitudes toward homosexuality or who have strong religious

beliefs about homosexuality tend to interpret contact with gays and lesbians negatively. In fact,

as I will discuss later, two conservative Christians who are in heterosexual marriages interpreted

their own previous homosexual feelings or activity as evidence that homosexuality is wrong, is a

choice, and should be changed. The consequences of a person’s life experiences and personal

contact with gays and lesbians therefore depend upon both the nature of the contact and on their

interpretation of that experience.

Moral Values and Judgments

Also closely related to people’s affective attitudes about homosexuality are their moral

values: the belief about whether or not homosexuality is immoral or sinful. Unsurprisingly, for

most people, their affective attitudes and their apparent beliefs about the rightness or wrongness

of homosexuality seemed closely related. However, it was by no means uncommon to talk with

individuals who denied that homosexuality is immoral but who still had negative attitudes about

it; nor was it uncommon to talk to people who said that homosexuality is immoral but who had

positive attitudes about gays and lesbians. An exploration of how people talked about the

morality or immorality of homosexuality illustrates that this is not a simple issue. For some

people, there is an important difference between the “moral” and the “natural,” such that

something might not be morally wrong but be wrong from a biological or psychological point of

view. In addition, there is a difference between moral values and moral judgments for many
   162  
people, such that some individuals who personally find homosexuality to be immoral refuse to

say that it is so, because they do not believe it is their place to cast judgment upon others.

Throughout my interviews, I was struck by how difficult it was for many informants to

answer a simple question: “Do you think that the homosexual act is immoral?” I had expected

that people who thought homosexuality was wrong would have no trouble saying so. But

contrary to my expectations, this appeared to be one of the most taxing questions for many of my

informants to answer. Some people may have had difficulty saying “yes” because of social

desirability bias, but the patterns of discourse about morality that emerged were so consistent and

so grounded in religious ideology that it is unlikely that this is true for everyone. Rather, many

people were reluctant to say that homosexuality is immoral because of the conceptual complexity

of the term. Morality variously invokes personal belief and social judgment, religious ideologies

and the logic of nature.

Among people who said that homosexuality is not immoral, most had positive attitudes

towards gays and lesbians and towards same-sex marriage. Especially for those without strong

Christian religious beliefs, they felt there was no basis for judging it to be wrong. Simon, a 24

year-old student, responded in this way to my question:

Q: Do you think that the homosexual act is immoral?

R: No.

Q: Like wrong in any way?

R: No. Like, morals are just justification of what you do and what people do, based on a

person’s set of beliefs. And one of those, and most people’s beliefs come from religion.

And their, if their religion says it’s wrong for a person, they’re gonna believe it’s wrong.

Me, I live outside of the realm of what is wrong by religion…. (Simon, age 24)
   163  

Like Simon, most supporters of same-sex marriage thought that opposition to same-sex marriage

was based primarily on religious teachings that homosexuality is wrong.

However, some people also thought that opposition to homosexuality and same-sex

marriage could be due to homophobia, prejudice, or to having grown up in a society where

homosexuality was considered wrong. Students with positive attitudes in particular thought that

these were not good reasons to be against same-sex marriage, because from their point of view,

society is changing and people should develop more tolerant attitudes toward gays and lesbians

as a result. Many adults admitted that they struggle with their past teachings and beliefs that

homosexuality is wrong, because they now feel that those are outdated ideas and no longer

acceptable. For example, Gerald, a 60 year-old Catholic talked about his opinion about same-sex

marriage this way:

I would struggle with it, my past. In theory, I would struggle with it, but I want to say, if

that comes up in Illinois—I don’t think we have voted on that issue yet—I think I would

say yes. I would want to say yes…. You know, that’s the last remaining issue that I

would struggle with. And that’s based more on an old belief than where I want to be. You

know, every decision you make in life isn’t necessarily white and black overnight. You

struggle with them, you struggle over time. (Gerald, age 60)

Gerald refers to his disapproval of homosexuality and same-sex marriage as “an old belief” and

talks about how he struggles to overcome those old teachings. Whether due to his age or his

religion or both, opposition to same-sex marriage is deeply engrained in his belief system, but he
   164  
no longer thinks that opposition is justifiable. So when he imagines himself voting on the issue,

he wants to say that he would approve of it.

Among my informants who agreed that homosexuality is immoral, some people offered a

religious justification for their belief, while others said that it just seemed wrong. Katrina used

religious reasoning to argue that homosexuality is immoral:

Q: Do you think that the homosexual act is immoral or sinful in any way?

R: It’s sinful, yeah. I think it’s immoral cause it’s in the Bible. (Katrina, age 25)

Rick, a community college student, said he thought that homosexuality was immoral, but he

didn’t think it should be called a “sin” because of the religious judgment implied by the term:

Q: Do you think homosexuality is a sin? Or immoral?

R: It feels immoral if I think about it, like, knowing how sexual activity, like what’s

involved in sexual activity. I feel that it’s immoral because it just feels like a dirty

thought in my head, but I don’t feel it should be labeled as sin… If there’s a God, creator,

or something, maybe we are judged. I don’t know. (Rick, age 24)

In this quote, Rick says that homosexuality seems immoral, but he is reluctant to cast judgment

on others, as the concept of “sin” implies for him. His answer appears to be based more on the

societal taboo against homosexuality than on any religious ideology per se, because he does not

identify himself as religious.

The fact that homosexuality seemed unnatural to many people is also connected to the

question of immorality. For example, Harvey said he thought that homosexuality is immoral

because it is not procreative:

   165  
Q: Do you think that like homosexuality is immoral?

R: In a sense, yeah. Because of the fact that, I mean, you are going against procreation

and, I mean, the whole sense of life is to procreate, to populate the earth, you know. In a

sense, you are taking away from that, you know what I mean? Because you’re not giving

back to the earth. You’ve consumed, but what are you giving back? (Harvey, age 23)

In this worldview, the immorality of homosexuality comes from the fact that it violates the

procreative laws of nature. In some religious doctrines, this procreative aspect of sexual

relationships is crucial. Alyssa, a college senior, explained that she thought some people oppose

same-sex marriage for this reason:

I think a lot of people just don’t think it’s natural. If they’re against it, then they just don’t

think it’s a natural thing. Because it makes sense, and Biblically, you know, it does say

the woman was created for man, and it all has to do with reproduction. Like the purpose

of life, evolutionarily speaking even, is to pass on genes, to procreate. (Alyssa, age 21)

Thus, as in Catholic religious ideology, the close relationship between morality and the natural,

reproductive order is important.

However, it was more common in my interviews for informants to make a distinction

between the moral and the natural. Many people who said that homosexuality is not immoral

nonetheless argued that homosexuality is unnatural. For example, Evan, a 24 year-old student,

said:

I don’t know if it’s immoral. Because here’s the difference between immoral and natural.

Immoral, it’s not, no one’s getting hurt. You’re just choosing a lifestyle that I don’t think
   166  
is going with nature. You don’t see animals do it, animals aren’t gay. Nothing else is

really gay… Like I don’t think it’s natural, but nobody’s doing anything wrong, no one’s

getting hurt. (Evan, age 24)

For Evan, homosexuality is not immoral because he doesn’t see any harm that comes from it.

However, he does think it is unnatural because of the lack of biological basis for being gay.

Similarly, Natalie said that homosexuality was not normal, but she would not go so far as to say

it is immoral:

Q: Do you think that the homosexual act is itself immoral?

R: Immoral?

Q: Or wrong?

R: No. It’s not normal, but I can’t put right or wrong on it. If that’s what the person wants

to do, do it in private. What I don’t like is just coming out and telling me all about it, you

know. (Natalie, age 60)

From Natalie’s response, even though she doesn’t particularly like hearing about homosexuality,

she doesn’t think she has the standing to judge it as right or wrong. She simply finds it to be

abnormal.

Many young conservative Christians who said that homosexuality is immoral nonetheless

held positive attitudes about gays and lesbians. Their religious ideology dictates that

homosexuality is a sin, but they did not think that this fact made gays and lesbians any different

from anyone else, because everyone sins. For example, Bethany drew from her conservative
   167  
Christian teachings to argue that homosexuality is a sin, but she felt that that did not imply

anything negative about the person:

I think that being alcoholic is a sin. I think, you know, being, homosexuality is a sin. I

think overeating is a sin, all those things are sins. You know, gluttony in any sort or form

or fashion is a sin, boom, done, said. And so is premarital sex, you know. Why? Because

I think that when we overdo any of these things, we create problems for ourselves…. I

think we’re all going to have like a trip somewhere and I don’t think it defines you as a

good person or a bad person, or you know, less likely to go to heaven or more likely to go

to hell… We’re all flawed. (Bethany, age 22)

Elizabeth has similar beliefs about the sinfulness of homosexuality. While she says she is

opposed to same-sex marriage, she also says she is upset by Christians that she knows who

single this issue out for negative judgment:

One thing that I don’t like is that, to me, Christians are picking this out as, you know, a

worse sin than others or like as unacceptable. They’re just bashing the issue, but it’s

really not anything worse than telling a lie. (Elizabeth, age 19)

Later on, I asked her to clarify her beliefs, and she implied that it would be more ideologically

consistent for Christians to not dwell on the issue of homosexuality:

What I believe is that a sin is a sin. Murdering someone is the same as telling a little

white lie. What I don’t agree with is that people are picking out this issue and just

bashing people on it. It just seems, you know, unfair to these people. And then what I

don’t like is that we claim to be people that say, you know, we love everyone and things
   168  
like that, but then you go into something like this and just bash people that are for gay

marriage, or um, just people who are gay. It just totally contradicts everything that we’re

saying. (Elizabeth, age 19)

For people like Bethany and Elizabeth, their religious ideology tells them that homosexuality is

immoral, but their attitudes about gays and lesbians are more positive. They believe that

everyone sins and that it should be no excuse for denying someone equal rights. In particular,

both Bethany and Elizabeth empathize with gays and lesbians in that they think that it is wrong

to discriminate against them by denying them equal rights.

The fact that some people say homosexuality is a sin but that denying gays and lesbians

equal rights is wrong raises an important point: there is a moral dimension to the equal rights

arguments advanced by proponents of same-sex marriage. I will say more about this later in the

chapter, but for now it is worth pointing out that the morality of homosexuality is not the only

moral dimension of the debate about same-sex marriage. In both academic articles and popular

press accounts of the same-sex marriage debate, morality frames are associated with opposition

to same-sex marriage, while rights frames are associated with support for same-sex marriage

(Brewer 2002; Brewer 2003; Hull 2001; Price, Nir and Cappella 2005). However, these are not

mutually exclusive frames: the rights frame is also a morality frame. The dichotomy between

rights and morality frames is inadequate for the analysis of the complexities of people’s

discourse about same-sex marriage.

Finally, one common pattern of discourse from people who say that homosexuality is not

immoral centers on the difference between moral values and moral judgment. Even people who

personally think that homosexuality might be wrong refuse to say that it is immoral because they
   169  
do not believe that it is their place to judge others. For many people, the question about whether

or not homosexuality is immoral implies moral judgment of others. Jerome, a 22 year-old

student, did not give a clear answer about this issue because of this difference:

Q: Do you think homosexual behavior is immoral in any way?

R: (pause) Yes and no. I don’t know. I guess certain acts I can’t really stand by the

majority with male couples. I don’t think I could ever be comfortable with it… (Jerome,

age 22)

When I asked Jerome whether he felt that his discomfort with homosexuality was a good

justification for opposing same-sex marriage, he replied:

R: Not really, because I mean, it’s all a factor about what makes you happy. It’s like, if

that’s what you want to do, then that’s what you want to do. You’re gonna fight with

whatever makes you feel happy. So you know, why not give it your all? Fight for what

you want.

Q: Do you feel like it’s just not your position to tell other people what to do and what not

to do?

R: Yeah, it’s not my position to tell them what not to do. Everyone lives different than

everyone else. Everyone’s not the same. So I got no right… I can’t tell someone how to

do or what to do. (Jerome, age 22)

Despite familiarity with religious teachings that homosexuality is immoral and having personal

feelings that it is wrong, many people refuse to judge whether or not homosexuality is right or

wrong. The refusal to say that homosexuality is immoral by people with strong Christian
   170  
religious beliefs is not necessarily a case of social desirability bias. It actually has a basis in

Christian theology: God gave humans free will, and it is God who is the ultimate judge of right

and wrong. Human judgment is fallible. Demarcus, a 23 year-old student, made this distinction

when I asked him about his opinion about same-sex marriage:

Q: Do you personally have an opinion about whether or not same-sex couples should be

allowed to get married?

R: Well, the only opinion I have about that is I think that should be left up to God. I

mean, even though he’s against that, you know, people are still going about having same-

sex relationships and stuff like that, and you know there are people in power who can rule

to approve it or against it… Even though I don’t agree with it, you know, I don’t feel I

should say, “Well, no they shouldn’t….” I’d rather just say, “That’s the way they want to

do things, that’s them. Let them do their thing.” (Demarcus, age 23)

In his explanation of his feelings, he begins by saying that the only true authority is God’s

authority, and he contrasts it to human authority, which is potentially flawed. So even though he

personally disapproves of homosexuality, he believes that everyone has free will to do what they

think is best—and let God sort it out.

Tom, a 47 year-old father, applies a similar logic to the question, but his answer

emphasizes the reflexive component of this set of teachings: if I judge a person’s behavior to be

wrong, they might judge my behavior to be wrong, and how do you know who’s right? I print the

lengthy exchange below because it indicates how complex and difficult talking about morality

can be for some people:

Q: Do you think that the homosexual act is immoral?

   171  
R: The physical aspect of it? Hmmm.

Q: Yeah.

R: Immoral for society or immoral for like spiritual outlook?

Q: Just for you personally, yeah.

R: [pause] The questions are getting much tougher, Peter. You can go back to [asking

about] life as a child.

Q: I know, I did that on purpose. But you know you’re getting near the end when the

questions get tough.

R: Do I think it is an immoral act? [long pause] If those two individuals, I don’t think I, I

don’t think I have the right to say that that is an immoral act. I think that that is a spiritual

question more than a legal question. Again, you have two people in a loving, caring

relationship, and I’m going to tell them that the only way that they can express it, the

ways they choose to express themselves to each other is immoral? What’s to stop them

from looking over here and say, “You know what, well I find that absolutely disgusting

over there, what you’re doing.” It’s like, well yeah, but there’s a lot more of us than you.

Well, does that, does sheer numbers mean that I am wrong?... I don’t think I have the

right to say that that is immoral. It may seem immoral to me, but… the way I interpret my

religious and my life is different from the next guy…. (Tom, age 47)

In this long exchange, it takes numerous long pauses and several conversational exchanges

before Tom feels comfortable expressing his views. When he finally does, as if he is having a

conversation in his own head, he decides that it is not his place to cast moral judgment on the

actions of others. Even though he indicates that he personally thinks homosexuality is wrong, he
   172  
also indicates that it is a “loving, caring relationship.” Whatever his own personal moral values

may be, casting a moral judgment on others is something different.

In this context, therefore, moral values might be thought of as the standards of right and

wrong that one applies to oneself, while moral judgments can be understood as standards that

should apply to other people as well. As Emily explains:

If you were a person who claims to be like religious or a member of a religion that

believes that the homosexual act is a sin, then it would be, constitute, sin for you. But… it

might not be a sin what they’re doing. I’m not God. I’m not the one to judge them. I can’t

even prove that my God exists, you know? This is just what I believe and I’m not going

to hold my beliefs on anybody else. So, it’s up to their conscience and whether or not

they think it’s a sin. (Emily, age 19)

Because Emily saw her religious beliefs as simply a matter of faith, not fact, it would not be right

for her to hold other people, with different beliefs, to her own standards. She felt that if

homosexuality did not violate a person’s own moral beliefs about right and wrong, then it should

not be considered a sin.

Thus, the apparently simple issue of whether or not homosexuality is immoral is

complicated in people’s discourse, because many people differentiate between their own

personal moral values that they apply to themselves and moral judgments of others. Even

conservative Christians, confronted with strong religious teachings that homosexuality is wrong,

do not always say that homosexuality is immoral. The issue is further complicated by the

distinction between the moral and the natural. Many people view homosexuality to be unnatural

or abnormal; but not all of these people also think it is immoral. While some element of social
   173  
desirability bias may influence people’s willingness to say that homosexuality is immoral, it is

also true that the complexity of the concept of morality, combined with people’s attitudes

towards gays and lesbians, produce a wide variety of discourses about the rightness or wrongness

of homosexuality.

Varieties of Discourse on Same-Sex Marriage

So far in this chapter, I have shown that the issue of homosexuality is central to the

controversy over same-sex marriage, and I have described how people draw from their cognitive

beliefs, affective attitudes, moral values, and life experiences in order to construct opinions about

same-sex marriage. In describing the variation in these cultural elements, I have sought to show

that people’s discourses cannot be understood in terms of simple dichotomies: right vs. wrong,

positive vs. negative, inherent attribute vs. lifestyle choice. Not only are each of these cultural

elements complex on their own, but the combinations of beliefs, attitudes, and values that people

express in their discourse are also numerous and varied.

There is no ideological imperative that certain elements of people’s cultural repertoires

go with others. To the extent that there are correlations, however, they are likely due to social

and political influences. For example, someone may be motivated by political reasons to say that

someone is born gay—because they want equal rights to be extended to gays and lesbians, and

they would have stronger political and legal leverage to achieve that goal if gays and lesbians

were deemed to be a protected minority group, like African Americans. Likewise, people may be

more likely to oppose same-sex marriage if they know and interact with more people who also

oppose same-sex marriage.

   174  
Just as there are complex combinations of attitudes, beliefs, values, and experiences

regarding homosexuality, people use a variety of discourses to express opinions about same-sex

marriage that are not just limited to supportive and oppositional discourses. In the remainder of

the chapter, I describe the variety of discourses that informants used to talk about same-sex

marriage, the types of people who use these discourses, and how they are rooted in specific

combinations of attitudes, beliefs, values, and experiences regarding homosexuality. I begin by

describing the discourses of unambiguous support and opposition and how they correlate with

people’s cohort locations and religious and political ideologies; then I describe the middle-

ground discourses of libertarian pragmatism and ideological conflict in the same terms.

Unambiguous Support

Unsurprisingly, people whose discourses about same-sex marriage are unambiguously

supportive tended to be younger and politically liberal. Fully 60% of students used supportive

discourses to talk about same-sex marriage, while 37.5% of parents used supportive discourses.

In this way, my interview data are consistent with the expectation of significant cohort

differences in attitudes about same-sex marriage.

Members of the younger cohort who used supportive discourses held a variety of political

and religious beliefs and came from a wide variety of family backgrounds, such that I was

occasionally surprised when students expressed support for same-sex marriage. For example,

Dennis had all the markings of someone who would be opposed to same-sex marriage. A 19

year-old male from a small town, he is a practicing Catholic, voted for John McCain in the 2008

Presidential election, and was a three-sport athlete in high school. However, he said that gays and

lesbians should have the right to marry just like everyone else and disagreed with his church’s
   175  
teachings that homosexuality is a sin. Thus, among the younger cohort, support for same-sex

marriage was not limited to people who were liberal, libertarian, and non-religious.

By contrast, members of the older cohort who used supportive discourses to talk about

same-sex marriage were exclusively either politically liberal or non-religious and politically

moderate. These were necessary, but not sufficient, conditions for using supportive discourses

among the parents in this study; some non-religious liberals did not use supportive discourses to

talk about same-sex marriage, as I describe later in the chapter. Older adults who used supportive

discourses frequently had significant personal contact with gays and lesbians, and that contact

figured prominently in their discourse.

In general, unambiguously supportive discourses are rooted cognitively and culturally in

positive attitudes about gays and lesbians, the belief that homosexuality is not immoral, moral

values of fairness and equality, and arguments that same-sex marriage is about equal rights for

gays and lesbians. This discourse is about rectifying a perceived injustice: that gays and lesbians

have been heretofore denied the rights, benefits, and responsibilities that are associated with

marriage. In one way, this finding supports those of previous studies of same-sex marriage,

which have identified this “rights frame” as central to the discourses in support of same-sex

marriage, as opposed to “morality frames” that are used chiefly by those who are opposed to

same-sex marriage (Brewer 2002; Brewer 2003; Hull 2001; Price, Nir and Cappella 2005).

However, my findings challenge those studies in that the discourse of equal rights among people

I interviewed has both political and moral components.

The political component of the equal rights frame is concerned with extending to same-

sex couples the rights and benefits guaranteed to married couples by law. For example, Chelsea
   176  
said that same-sex marriage should be legalized because guaranteeing equal rights to all citizens

is one of the principles of the nation:

Q: Do you think that we should make same-sex marriage legal?

R: Yes. I think they should. It’s not right to make, to not enact a law. They should enact a

law because the United States is about equal[ity]. The gay people are not equal to us

because we can get married, they can’t. They should have the same rights as we do. They

are people like us. They should have the same things we do. (Chelsea, age 19)

Evan expressed strong support for same-sex marriage, even though he believes that

homosexuality is unnatural, because he saw the issue as being about equal rights:

I don’t think people are working for a title, I think they’re working for the rights. They’re

looking for rights. You don’t have to accept it. I think what’s beautiful about being gay is

they aren’t forcing you to accept it. No one’s cramming anything, they just want to be

able to do what they want to do. (Evan, age 24)

As Evan points out, the question of whether or not someone is entitled to equal rights is different

from the question of whether or not a person approves of the person’s choices. Indeed, it was

common even among people who did not approve of homosexuality to argue that they deserved

equal rights. Very few people, when asked directly, said that gays and lesbians did not deserve

equal rights.

Although not all informants spoke immediately about the rights and benefits of marriage,

almost all informants mentioned it when I asked them why some gays and lesbians wanted the

right to marry:
   177  
Q: Why do you think some gays and lesbians want the right to marry?

R: Probably mostly related to the benefits and the, all of that, insurance, and all of those

kinds of things that are wrapped up in the definition of a marriage, because those kinds of

things are significant. (Bernice, age 53)

R: Why do you think gays and lesbians want the right to marry?

A: Well, I know there’s benefits to getting married, and I think it just, I think that right

now they just want to show that, “Hey, we’re the same. You can’t say that, you can’t say

that we can’t have the same rights as you.” (Amanda, age 19)

Thus, having equal rights under the law and having the same benefits as heterosexual couples is

a common understanding of why same-sex marriage is an important issue.

Support for extending equal rights to gays and lesbians as described here is typically

associated with identifying as politically liberal or as a Democrat. However, those whose

political beliefs are libertarian—conservative on economic issues but liberal on social issues—

use discourses about same-sex marriage that are indistinguishable from people who are liberal. In

my interviews, I oftentimes never could tell whether an informant was liberal or libertarian until

the end of the interview, when I asked about the 2008 Presidential election and political issues

they thought were important. For example, Nate, a 19 year-old college student explained why he

was planning on voting for Ron Paul in the 2008 Presidential election:

I like that he is getting rid of income tax, [which] is one of his things. I consider myself to

be strict about the Constitution and the fact that… I think one of the articles in the

Constitution directly states, “there shall be no tax on income,” but yet we have tax on
   178  
income, and I don’t think that’s right at all. I also like some of his standings on things.

He’s pro-gay marriage, which is like I am. (Nate, age 19)

Nate’s comments symbolize the connection between liberal social policies and conservative

fiscal policies in libertarian political ideology. In discourse about same-sex marriage, fiscal and

economic issues—aside from the benefits to married couples—were rarely brought up, so the

discourse of libertarians and liberals were similar.

In addition to this political component, the equal rights discourse also has a moral

component. One reason that previous studies have failed to notice the moral component of the

equal rights discourse is that the morality frame has been defined narrowly in terms of moral

judgment of homosexuality and in terms of religious arguments. While religion is certainly an

important language of morality, moral values are expressed in other terms as well: in terms of

right and wrong, justice and injustice, fairness and unfairness. Moreover, while a person’s moral

judgment of homosexuality is important, homosexuality is not the only topic of the debate that

invokes moral values. When talking about same-sex marriage, a person must also make moral

judgments about discrimination, prejudice, equality, the role of government in people’s private

lives, and the likely effects of different courses of action on children and other people.

When talking about equal rights for gays and lesbians, judgments of right and wrong, fair

and unfair, were integral to the discourse.

The longer that people are going to be treated terribly, or at least not equal, people are

going to feel unequal. They are going to believe that they are unequal, and I don’t want

that. For a civil rights issue, I don’t think it’d be fair to them. (Jeff, age 21)
   179  
In this quote, Jeff clearly connects the issues of civil rights, equality, and fairness. This quote

also highlights one notable feature of the moral discourse about equal rights: the moral aspects of

equality and civil rights are most evident when there is a perceived injustice or an act judged to

be wrong. Drew recounts an incident that highlights this:

Q: One of the issues… that’s been controversial in recent years is the issue of same-sex

marriage.

R: Yeah. Well, I’m totally for it.

Q: Yeah?

R: I’m up for anybody’s rights. And there was a teacher [here who] wrote that she thinks

that good Christians need to use the Bible to go against homosexuals… and I was just,

that’s terrible. To go after a group of minority people and to criticize them and put your

name and the university’s name on that. I could show it to you. I think it’s disgusting.

(Drew, age 21)

In Drew’s comment, it is clear that he thinks that the discrimination against a minority group is

wrong. Even though the incident was not about the specific issue of same-sex marriage, he

nevertheless uses the more general moral issue to justify his support for same-sex marriage.

Oftentimes, the moral argument for giving gays and lesbians equal rights was frequently

expressed in a discourse of humanity. In an effort to explain that gays and lesbians should be

treated no differently than heterosexuals, informants described gays and lesbians as human

beings, just like everyone else; or they talked about the issue of same-sex marriage as being

about human rights.

Q: Why do you think gays and lesbians want the right to marry?
   180  
R: I think it’s just ‘cause they want rights. I mean, they’re humans…. It’s just they don’t

understand why as human beings they get some rights and not others, just because of, you

know, who they are in love with. (Gina, age 20)

Like Gina, Alan explained his support for same-sex marriage as being a matter of human rights:

Places like California, Massachusetts, where their Supreme Courts have struck [bans on

same-sex marriage] down, I think those are measures to be applauded. Because yeah, I’ve

got gay friends, and they’re, you know, it doesn’t matter what you believe religiously. I

just think, you know, they deserve a right to, they love each other, and I think they

deserve a right to, you know, visit each other in the hospital if one of them’s sick. I don’t

know, it just seems like a matter of human rights. (Alan, age 22)

Even people who opposed same-sex marriage used this discourse of shared humanity to explain

why they thought gays and lesbians should have equal rights.

Some informants observed that denying equal rights to gays and lesbians was an indicator

of our prejudice as a society, and that, like other forms of prejudice, it should be eliminated.

Asked what she thought some effects of legalizing same-sex marriage might be, Karen took a

broad moral stance against all forms of prejudice and said that she hoped it would reduce

prejudice in society:

I would certainly hope that one of the effects would be more understanding of others that

are different, and that maybe it might have an impact as well on other prejudicial

situations. (pause) Too often, Americans look at other people and they see the outside

shell, and they don’t bother to get to know the people because that shell isn’t matching
   181  
their own. And it bothers me to know that people aren’t valued based on what their shell

says. (Karen, age 50)

Many people specifically used African Americans’ struggles for civil rights as an analogy to

express their opinion about this issue:

I think sooner or later it’s going to be legalized. It’s one of those issues that we inch

forward, that we inch closer to day by day. But it’s just, it’s just like equality for blacks.

It’s something that, yeah, they were freed during the Civil War, but it took them over 100

years to get on the same status as white people. It’s not something that’s just going to

happen overnight, where one night we wake up and all of a sudden they’re saying,

“Okay, men can marry men, women can marry women.” (Simon, age 24)

Just like people today view the inequality of African Americans as immoral, many people think

that inequality of gays and lesbians with regard to same-sex marriage is immoral.

Informants also used religious arguments in support of same-sex marriage. For example,

David, a 50 year-old Christian, talked about switching churches because of the ways in which

gays and lesbians were viewed within their church. They began going to a church affiliated with

the United Church of Christ after their mainline Protestant denomination failed to soften their

doctrinal language about the status of gays and lesbians in the church:

They kept the same wording that they’d had all along, and the wording, if you’re a

lesbian or gay or transgender, you’re not going to feel very welcome. It’s kind of almost,

you know, hate the sin but love the sinner. And you can’t put the words hate and love
   182  
together in the same sentence, because what the person hears is hate. And it’s just wrong.

(David, age 50)

Ruth, a 59 year-old Jewish woman, also invoked her religious faith to explain why she supports

same-sex marriage:

I think that any institution or any connections that bring people closer together are

good…. As far as my religion goes, the important thing in the world is repairing the

world. We call it “tikkun olam,” which means repairing the world. That is our obligation,

making the world a better place, not fighting with people. (Ruth, age 59)

Both Ruth and David found support for same-sex marriage in their religious beliefs, and they

drew from Judeo-Christian teachings in order to explain why same-sex marriage should be legal.

Both believed that love for everyone, regardless of who they are, is a central teaching of their

religion, and they felt that legalizing same-sex marriage would be an expression of that principle.

Unambiguous Opposition

In contrast to the people who support same-sex marriage without reservation, people who

use unambiguously oppositional discourses tended to be older and have conservative religious

beliefs. Again, consistent with the finding of cohort differences in attitudes about same-sex

marriage, only 14% of students expressed unambiguous opposition to same-sex marriage,

compared to 37.5% of parents. Although oppositional discourses to same-sex marriage are most

firmly rooted in religious ideologies, parents who opposed same-sex marriage did not necessarily

base their opposition on their religious views, nor were they necessarily politically conservative.
   183  
Some parents opposed same-sex marriage, not out of religious conviction, but simply out of a

feeling that homosexuality is wrong or unnatural.

Among students, a conservative religious ideology or background appears decisive for

discourse that is unambiguously opposed to same-sex marriage. The limit case of a student with

a strong conservative political ideology but not a strong religious identity shows how essential

conservative religious values are for students who oppose same-sex marriage. Jesse, a 27 year-

old conservative, has a subdued Christian identity and is generally skeptical of evangelical

Christianity: “it’s not about just quoting the Bible and fire and brimstone. I can’t stand that. I

don’t like going to churches where, got a big smile on your face, and ‘Oh Jesus is great all the

time.’” His views about same-sex marriage are primarily shaped by conservative political

philosophy, and the resulting discourse is not oppositional, but a form of ideologically conflicted

discourse (described later). He argued that government should not be in the marriage business in

the first place, but since they are, they should treat everyone the same and stay out of people’s

private lives:

Q: Some people have said that the institution of marriage is in a crisis in this country.

How do you feel about that?

R: Yeah, well those are just the people that want to get this stupid gay marriage ban in the

Constitution, which is so freakin’ stupid. Marriage is in a crisis? Yeah, well you know

what? Look at your own marriage first, okay. It’s none of your damn business, really….

And look, I’m not an advocate for it, and I’m not going to go out and march in the gay

rights parade, but look, what you do in your own bedroom is your own business and it’s

none of mine…. The reason they want the equal rights for is, ah, visitation, like you

know, when you’re in the hospital, things like that. Insurance at companies and things
   184  
like that. That is the only reason it’s there in the first place, is because government

intervened and said, “Well now we’re going to institutionalize marriage and make it a

government proclamation, you know. They took it out, they took it away from

communities and churches in the first place. (Jesse, age 27)

This quote shows that conservative political philosophy, applied to same-sex marriage,

resembles the libertarian view of same-sex marriage. Later in the interview, Jesse affirmed his

belief that gays and lesbians should have the rights guaranteed by marriage because he feels that

freedom from government intervention in private lives overrides the spiritual concerns about the

definition of marriage. Thus, what appears decisive for unambiguous opposition to same-sex

marriage among young people is a strong conservative religious identity or values.

In general, the Christian teachings that homosexuality is a sin, combined with negative

attitudes toward homosexuality and the belief that homosexuality is a lifestyle choice that can be

changed, provides a strong ideological basis for opposition to same-sex marriage. People whose

discourses are unambiguously opposed to same-sex marriage typically based their opposition on

their religious beliefs. Asked about same-sex marriage, Haley said:

Well, my view on it’s really heavily tied to my religious views. That man was created for

woman, that it’s no question. And if you look in the Bible or something, the cities in like

the Old Testament that allowed like gay marriage and things like that, the Lord like

destroyed them… And there’s like verses in the New Testament that talk about the idea

that gay marriage and gay relationships are wrong. And it’s pretty clear. (Haley, age 19)
   185  
Although Haley mistakenly says that the Bible talks about same-sex marriage, other

conservatives cited the same Biblical prohibitions against homosexuality.

I didn’t see nowhere in the Bible, or any book, the Torah or the Koran, where God, Allah,

Jehovah, whatever you wanna call him, approved of same-sex marriage. I don’t see

nowhere. But I do know one story of Sodom and Gomorrah, where he destroyed the

people because of that action. I do know that. (Michael, age 20)

Many Christians who opposed same-sex marriage and homosexuality because of their

religious beliefs were quick to point out that they were not condemning the person; it is only the

homosexual behavior that is sinful and immoral. Put simply, this is the religious command to

“hate the sin, love the sinner.”

God doesn’t say he hates, you know, gay people or anything like that. He says he hates

the sin, he hates the action, that lifestyle, but he doesn’t hate the person themselves. But

too many people don’t understand that, they just, you know, look at it as God hates them,

they’re going to hell, and that’s not, that’s not correct. (Elaina, age 22)

Most conservative Christians argued that everyone sins, and that to condemn someone for

committing sinful acts is against Christian teachings. In Christian theology, to say that “Jesus

died for our sins” means that even individuals who commit sinful acts will be forgiven, that the

person is not judged by their acts alone. Everyone sins, and despite those sins, you can still be

saved. Andrea, a 45 year-old homemaker, argued that, even though she disapproves of

homosexuality and same-sex marriage, she is not trying to say she is better than anyone else:
   186  
Do I think that their sins are any greater or worse than mine? I do not. You know, I would

chastise any Christian that says, “You’re living in, you know, vulgarity and sin every

single day and gonna go to hell.” I would definitely question that Christian as far as,

“And you’re not taking the Lord’s name in vain? And you always pay your taxes? And

you, you know, you never light a cigarette, you don’t drink to intoxication, you never

look [at] or fornicated [with] another woman?” You know, I would definitely jump on

that bandwagon almost to defend the rights of that person, when you’re talking about

salvation. (Andrea, age 45)

In this quote, Andrea strongly rejects the idea that another Christian should judge someone who

is gay or lesbian negatively; to do so would be un-Christian.

Integral to this worldview is the belief that homosexuality is a behavioral choice and can

be changed. Regardless of one’s feelings and predispositions, whether they be about sexuality or

anything else, each person always has a choice about whether or not to act on those feelings.

Even if a person is born with an inherent inclination towards homosexual behavior, each person

must choose whether to give in to temptation or to resist. Paul, a conservative Christian student

explains this connection between tendencies and behavioral choices by using language he has

learned in his psychology classes:

Homosexuality is more of a choice that is is a—well, you may be that way, but you’re

choosing to act that way…. What happens with development as you grow up and you

learn in life, and your experiences will change tendencies, increase tendencies, and

decrease tendencies…. Let’s say if they do have, they’re born and wired up with slight

homosexual tendency, if you grow up in a household that’s very neutral towards it, you
   187  
can go one way or another based on things you might see in the media or your friends…

(Paul, age 19)

Paul uses the language of developmental psychology to explain that, regardless of one’s

predisposition, their upbringing will shape their attitudes about homosexuality, and they must

ultimately choose whether or not they want to follow those tendencies. Thus, regardless of a

person’s cognitive beliefs about the extent to which “nature vs. nurture” causes homosexuality,

conservative Christians believe that gays and lesbians still have to choose whether or not to

engage in homosexual behavior or live a homosexual lifestyle. In that sense, a person can change

their homosexuality, because they can stop choosing to act on those feelings:

Q: Do you think that homosexuality can be changed? You know, do you think like

people, somebody who’s gay can be, sort of, not gay?

R: (pause) It’s almost like, you know, like especially if a person like that is religious, and

they believe in God, and they know that ultimately they’re going to have to renounce that

way of living—because it’s not Godly, and it’s not right, and they’re supposed to be

engaged in that, and you know, it’s not healthy. Yeah, I think, I think so. (Demarcus, age

23)

Homosexuality can be changed if a person desires to stop participating in a homosexual

behaviors. It would be unrealistic to expect everyone to do so, but if a person is guided by

Christian beliefs and with God’s help, they can renounce their sinful lifestyle.

The most thoroughgoing oppositional discourse regarding same-sex marriage is so firmly

grounded in religious ideology that even homosexuality is defined in exclusively religious terms.
   188  
From this point of view, homosexuality is not so much a behavioral choice as it is a deception by

the devil, evidence of the tremendous evil present in the world because of the power of Satan and

his minions. Dana, a 48 year-old born-again Christian, described homosexuality in these terms:

I think it comes from the enemy… There are three different desires that man has:

Spiritual, Godly desires, his own human selfish desires, and then the desires that are

placed to him by the enemy. You were given the choice of which one you are going to

follow. You can follow the Lord, you can follow your own self, or you can follow the

enemy. Self doesn’t always completely go against God, it’s usually just fulfilling what

you want. Anything against God comes from the enemy. That’s how I see it. (Dana, age

48)

Dana believes that homosexuality is not just a choice made by people following their selfish

desires; it is much worse. Elsewhere, she describes how it results from people looking for love,

trying to feel completed. But when people have turned away from God, they cannot find God’s

love, and they fall victim to the temptations of “the enemy.” The extent of Dana’s opposition to

homosexuality and same-sex marriage is evidenced in the following quote, which comes from

her response to my opening question about same-sex marriage:

I have been—and the only word I can think of, and I know it sounds really bad, and I’m

sorry, but—disgusted by a lot of the things that happen in that lifestyle. It has given me a

very thick skin, and it’s wrong, and I know it’s wrong. I get angry at times—and I know I

shouldn’t—because of their persistence of pressuring people for acceptance. And their

lack of tolerance for us is by saying we have lack of tolerance. And their agenda to
   189  
infiltrate children’s minds… you should not try to teach it to them as acceptable and at a

very young age. I’m very much against that. (Dana, age 48)

Dana is so strongly opposed to same-sex marriage because she sees it as the work of Satan and as

part of the larger agenda to promote homosexuality.

Many conservative Christians include the controversy over same-sex marriage in part of a

larger narrative of moral decline in the United States. The push to legalize same-sex marriage, in

this narrative, is the next step in a path away from traditional Christian morality taken by our

society.

Q: If the United States were to legalize same-sex marriage, what kinds of effects do you

think it would have on society?

A: Devastating. I think it’s not anything, see we’ve been in moral decline ever since our

country has chosen to leave, um, not be led by the Lord anymore. It’s not, all of this is

not surprising to me…. I believe there would be just further moral decline for our nation,

and I believe the hand of God’s blessing over this nation is being withdrawn, and I do

believe that eventually that’s going to mean the destruction of America. All we can do is

pray. (Theresa, age 55)

In general, many conservative Christians view the greater acceptance of homosexuality in

society as part of this moral decline, and they lament the fact that few people stand up against it:

Unfortunately, it seems to be that’s the way we react to things that we may or may not

agree with, we just sort of accept [it] and say, “Well, we can’t do anything about that,” so

that’s another place that we’re going downhill. But we do consider it a downhill move.
   190  
We don’t, you know, we don’t consider that this is elevating our society because we’re

accepting this, no way. No way is that happening. You know, we’re accepting something

else that is demoralizing or is immoral… which, I mean, we see ourselves becoming less

moral because of things such as this. (Bernice, age 53)

These conservative Christians are not oblivious to the increasing tolerance of

homosexuality in the United States in both mass media and public opinion; nor are they unaware

of the accusations that people who oppose same-sex marriage are simply ignorant, bigoted, or

hateful:

R: You watch the news, and you see all these people fighting for so-called gay rights, and

telling Christians that we’re hateful when we’re not. We’re just trying to tell the truth.

You’re just one that’s perceiving we’re being hateful, and being [pause]

Q: Prejudiced? Or…?

R: Yeah. And it’s not. That’s just the way they’re trying to get people to vote in their

favor by saying we’re hateful, when they just don’t understand Christianity. (Pablo, age

51)

Conservative Christians like Pablo reject these labels because of their judgment of the sin, not

the sinner. They insist that, by opposing homosexuality, they show great love for the sinner

because they reject the notion that tolerance of deviant behavior is a good thing. As Dana puts it:

If you see someone is going down a path that is going to ruin their life, are you loving

them by allowing them to do it, or are you loving them by telling them it’s wrong?... I’m

not trying to point the finger and tell them they’re terrible people; I’m trying to tell them
   191  
what is best for your life because of what research has shown it does to a family, a

couple, children. It’s not healthy. (Dana, age 48)

Thus, some people whose discourse is unambiguously opposed to same-sex marriage take pride

in their opposition insofar as they view it as a principled stand against forces of secularism and

liberalism that are contributing to the continuing moral decline of the society. By taking a stand

against same-sex marriage, they see themselves as taking a stand for family stability, healthy

relationships, and proper child-rearing.

There are people whose talk about same-sex marriage is unambiguously opposed but who

do not use any religious or political language in their discourse. In my sample, they were all

parents. For these individuals, their opposition to same-sex marriage may be based simply on a

deep-seated, and relatively unquestioned, feeling that homosexuality is simply wrong. For

example, Debra identifies as Catholic and generally avoids talking about politics, said she was

against same-sex marriage. In the passage quoted earlier in the chapter, she expressed fear of

homosexuality. She did not offer a reason for her opposition to same-sex marriage, other than

that she was raised to believe that homosexuality is wrong. Asked whether or not she thought

homosexuality is immoral, she struggled to explain why:

Q: Do you think that the homosexual act is immoral?

R: Myself (pause), yeah. I’d have to say yeah, I do. I just, I just don’t agree with that, you

know. I keep thinking, you know, “Okay, am I being prejudiced?” I don’t feel like I’m

being prejudiced, but there’s sometimes I feel like maybe I am being prejudiced, you

know. But I don’t know. It’s like in the air. Am I prejudiced or not. Sometimes yes and

sometimes no, I guess. (Debra, age 57)

   192  

She does not use a religious language to express her belief that homosexuality is immoral.

Unable to express a reason for her thinking that, she wonders if she is simply being prejudiced.

Thus, discourses that are unambiguously opposed to same-sex marriage are based

primarily on religious beliefs that homosexuality is wrong and on negative attitudes about

homosexuality. Most people who are unambiguously opposed to same-sex marriage have an

understanding of homosexuality as a behavior, which individuals can either choose to engage in

or not. These oppositional discourses are used more by older adults than students, with

conservative religious values acting as the essential ingredient for students who express

opposition to same-sex marriage. Without a religious justification for one’s views, negative

attitudes about homosexuality appear, even to those who hold them, as prejudice.

Middle-Ground Discourses

In between discourses of unambiguous support and opposition are a variety of middle-

ground discourses, in which attitudes about same-sex marriage cannot be easily expressed in

terms of support or opposition. Individuals that use these middle-ground discourses might say

they don’t know how they feel about same-sex marriage, or that they don’t care one way or the

other about the issue, or they may express both support and opposition at the same time. These

discourses are expressed using different combinations of beliefs, attitudes, and values than those

that are used to express unambiguous support or opposition.

Middle-ground discourses were most often used by people, like younger religious

conservatives and older political liberals, whose cohort locations and religious or political

ideologies are cross-cutting. For these individuals, their religious or political ideologies incline
   193  
them to take one position in the debate about same-sex marriage, while their cohort membership

pulls them toward the other. Those competing influences are evident in their discourses, which

balance or oscillate between supportive and oppositional elements in various ways. Unlike

supportive and oppositional discourses, there is no difference in the propensity of students and

parents to use middle-ground discourses (although it does vary by the type of middle-ground

discourse): 25% of both students and parents used some variety of middle ground discourse to

talk about same-sex marriage. While there may be a large number of distinct middle-ground

discourses about same-sex marriage among the population, I focus on two types of discourses

here: libertarian pragmatism and ideological conflict.

Libertarian Pragmatism

Libertarian pragmatism is a discourse that combines the value of individual liberty with a

refusal to judge the morality of another person’s action, as long as there are no negative

consequences for others. The discourse is libertarian because the speaker rejects the validity of

the exercise of moral judgment over actions or behaviors that are considered private; it is

pragmatic because this libertarianism is predicated upon a specific outcome: that the action cause

no harm to others. If the action were to affect others in a negative way, then government

constraints on that action would be legitimate. Informally, the discourse can be encapsulated in

the declaration, “Hey, it’s a free country. You can do what you want, as long as you’re not

hurting anybody.”

Both the students and the adults who used libertarian pragmatic discourses to talk about

same-sex marriage have moderate-to-liberal political beliefs or have fairly tolerant attitudes

regarding marriage and sexuality in general. However, their discourses were also affected by
   194  
some implicit or explicit negative evaluation of homosexuality, which was either expressed as

discomfort with homosexuality or as an implicit, taken-for-granted understanding that

homosexuality is wrong and not to be encouraged. While three students (5%) did use this

discourse, the discourse was more common among parents (16%). The three students who used

this discourse were all male, and their negative evaluation of homosexuality appeared to be

rooted in either the threat to masculinity that homosexuality has historically represented to many

men or a background in a conservative moral or religious community that taught that

homosexuality is wrong. Among parents, the source of this negative evaluation also appears to

be their backgrounds: in this case, from their acceptance of the cultural view of homosexuality as

deviant, stigmatized, and wrong, that was dominant in American society when they came of age.

The discourse of libertarian pragmatism was not uncommon in my interviews. The

response from Matthew, a 51 year-old postal worker, to my introductory question about same-

sex marriage conveys the essence of libertarian pragmatism: “I don’t really approve of it, but um,

whatever floats your boat.” Similarly, Dylan, a 23 year-old student, used this discourse when

asked about same-sex marriage:

I don’t know, I really, what’s their business is their business. I honestly don’t really care

too much for it, but I’m not going to have a biased point of view and say it’s wrong.

That’s that person’s life, it’s not mine. It’s not affecting me in any way. (Dylan, age 23)

Jillian, a 49 year-old nurse, used this discourse as well:

Q: One issue that’s been relatively controversial in recent years has been the issue of

same-sex marriage. Have you heard much at all about that issue?
   195  
R: I see it on TV and stuff like that, like everybody else, but I can’t judge them people

either. They, that’s their lifestyle, that’s what they love. They love somebody just as well

as somebody else. I’m not, I don’t go one way or the other. If that’s what they want, then

that’s what they should have, you know. (Jillian, age 49)

In all of these examples, the informants say that they don’t feel strongly about the issue, and they

indicate that they understand that many people disapprove of it. But they also affirm the freedom

of gays and lesbians to enter into same-sex relationships.

This pattern of talk about same-sex marriage was frequently coupled with a refusal to

state an opinion about same-sex marriage. Rather than say that they support or oppose same-sex

marriage, informants said that it did not matter to them. When directly confronted with my

request to state an opinion on the issue, people who used libertarian pragmatic discourse usually

refused:

Q: Do you personally have an opinion about whether or not same-sex marriage should be

legalized?

R: [pause] It really doesn’t matter to me. I mean… it’s a piece of paper that maybe to

them means a lot. But to me, it doesn’t really change my life or my outlook on life. You

know, does it give them a tax break or something, cause they can claim as dependents? I

don’t know. And if that’s all it is, it’s small potatoes…. If it means so much to them and

it means very little to me, why are we getting all bent out of shape about it? (Tom, age

47)

Q: Would you say you sort of personally have mixed feelings about this issue?
   196  
R: Um, I’m not completely for, but I’m not really completely against it either. But um, as

long as they stay away from me and mine, that’s fine. I just don’t want to get involved in

it. (Matthew, age 51)

Q: Would you support legalizing same-sex marriage in our society?

R: You know, I don’t know that I would actively go into it, honestly, because it’s just not

an issue I really care about. It doesn’t affect me, and I’m being honest about that. Maybe

it’s because I don’t have anybody close to me that I know, you know…. I wouldn’t be

opposed to it, but I wouldn’t be active towards it either. (Maria, age 45)

At first glance, such quotes appear to be evidence of social desirability bias: the informant does

not want to say something that might make them seem bigoted or prejudiced. While in some

cases this may have been the motivation for this discourse, libertarian pragmatism was so

common, and it has such a strong basis in American culture, that it is highly unlikely that social

desirability bias alone can account for it.

As suggested in some of the quotes above, libertarian pragmatism frequently contained

an implicit or explicit negative attitude of some sort regarding homosexuality. Consider Harvey,

a 23 year-old community college student. Throughout our conversation, it was clear that he felt

uncomfortable with the idea of gay men and gay sex. At one point, he expressed his discomfort

regarding gay men when he remembered an incident in one of his classes:

I don’t want to sit there and have a guy, a gay guy sitting there. That’s like when I went

to… class, a gay guy was sitting there looking across the class, looking at me across the

room, looking at me like this [makes face]. Like that’s uncomfortable, you know what I
   197  
mean? Okay, maybe a girl, but that’s what we do as far as guys, but when a guy does it,

then it’s like, I want to beat him up. (Harvey, age 23)

The negative attitudes expressed in this statement are quite explicit. The same look he gives to

women, when turned on him by a man, makes him uncomfortable to the point of threatening

violence. But he did not use this feeling as the basis for opposing same-sex marriage. When I

asked him if he personally had an opinion about same-sex marriage, he responded:

Like I said, I just don’t really even care…. It just really doesn’t affect me so I really can’t

just, you know, downplay somebody else who goes there. Like if I go to a club and like

somebody sees me talking to like a white female, and then it’s like, “oh you can’t do

that,” it’s like what am I doing to hurt you?... I got my own things to worry about. Same-

sex marriage really isn’t one of the things on the table at the moment. (Harvey, age 23)

Harvey uses the example of interracial dating as a metaphor to explain his refusal to state an

opinion about same-sex marriage. Simply because one doesn’t approve of another person’s

behavior does not mean it is appropriate to pass judgment on them.

The fact that libertarian pragmatism was used frequently by people with implicit or

explicit negative attitudes about homosexuality is especially noteworthy because the ideas

contained in this discourse are typically associated with the ideology of political liberalism:

tolerance, cultural relativism, rejection of prejudice, and embrace of equal rights. Indeed, the

people who used libertarian pragmatic discourse had relatively liberal political beliefs regarding

other issues. However, they did not come out in support of same-sex marriage, nor did they

oppose it. Instead, their discourse undermined the rational basis for opposing same-sex marriage
   198  
by affirming the freedom of individuals to act as they choose—as long as there would be no

negative consequences for others.

Ideological Conflict

A second variety of middle-ground discourse might be called “ideological conflict.” I do

not assume that people have coherent, logically consistent belief systems, so this label is not a

marker of an anomalous or unstable belief system. Rather, I use this label to refer to the fact that

many people used different combinations of attitudes, beliefs, values, and experiences that led

them to talk about same-sex marriage in ways that are apparently contradictory. There were

several distinct ideologically conflicted discourses used by people in my sample (e.g. Jesse’s

conservative but supportive discourse above; people who expressed support for same-sex

marriage but opposed allowing same-sex couples to adopt or raise children), but the predominant

ideologically conflicted discourse, constituting 11 of the 17 instances, was used by people who

struggled to reconcile conservative religious teachings about homosexuality with a feeling that

they should be tolerant and supportive of the rights of gays and lesbians.

Ideologically conflicted discourses of all kinds were used more often by students (22%)

than by parents (9%). The three parents who used ideologically conflicted discourses to talk

about same-sex marriage are all Catholics. The eight students who used this religious variant of

the discourse—one Catholic and seven Protestants—expressed mixed feelings about same-sex

marriage because they did not fully agree with their religion’s teachings on homosexuality. The

parents all appear to have questioned the religious teaching that homosexuality is wrong to

varying degrees because they otherwise hold liberal political beliefs or know people personally

who identify as gay or lesbian. The students, by contrast, appear to have questioned their
   199  
religion’s teachings on homosexuality because they all view gays and lesbians as normal people,

just like heterosexuals, and are thus deserving of equal rights.

The religious variant of ideologically conflicted discourse about same-sex marriage

occurs when a person’s religious beliefs or backgrounds clash with their attitudes about gays and

lesbians. This discourse is used primarily by young people who are conservative Protestant

Christians, but it is also used by liberal Catholics. The conservative religious teachings about the

immorality of homosexuality seems to contradict their positive attitudes about gays and lesbians

and the belief that they deserve equal rights. The apparent contradictions between the religious

and secular social worlds in which they live are evident in the discourse of Carl, a 19 year-old

student who grew up in a Christian fundamentalist church but who has been questioning the

theology of the church. On a religious level, he believes that homosexuality is immoral and

should not be allowed; but on a political level, he believes that a person’s sexual orientation is

beyond their control, and thus gays and lesbians should have the same rights as heterosexuals:

It depends on what perspective. If you take a Biblical perspective, obviously it would say

you shouldn’t, based on a lot of scriptural evidence. But societal beliefs would tell you

the complete opposite…. I mean, I believe you should have gay marriage, because

constitutionally, if you look at the Constitution, that’s what it says. Why should someone

be denied a right? … Scripturally, I do think it’s wrong; but I think they should have a

right to. (Carl, age 19)

Carl said he would vote against same-sex marriage if he had the opportunity because it was the

religious conception of marriage that was meaningful to him; but he also thought that gays and

lesbians, politically, have the right to get married. Part of the tension in his beliefs may have to
   200  
do with the fact that he believes that homosexuality comes, at least partially, from a genetic

predisposition. Thus, even though homosexuality is immoral, it is an inherent part of who a

person is.

I mean, their sin isn’t any worse than anyone else’s, so… that’s why I think it’s horrible

when some people just go on this tirade about gay marriage and stuff. There’s no room

whatsoever for someone to act like that. (Carl, age 19)

He would not want to see someone treated negatively by someone just for who they are. So his

discourse is an uneasy balance between support and opposition regarding same-sex marriage.

Bethany also uses this discourse to explain her feelings about same-sex marriage. She

supports equal rights for gays and lesbians, but she also believes that homosexuality is a sin. She

argues that just because the Bible says it is a sin is not a justification for opposing same-sex

marriage:

It seems like they’re everyone is trying to take a moral stance, you know, that God says

that marriage is between a male and a female. You’re right, he does say that. You’re

right, I do believe that. However, just as much as I really can’t cast stones at people who

get divorced or people who overeat or people who are alcoholics, God says all of those

things are just as much of a sin…. You want to get married, go right on ahead, it really

doesn’t bother me. And they want to be entitled to the same views because their level of

commitment is more or greater or as equal to the happiest married heterosexual couple

and that, you know, I think that’s really important when it comes down to it. I mean,

heaven forbid if my significant other got into a car crash or something; I wouldn’t want it

to fall on his great aunt who he never talks to, who is his only surviving family member; I
   201  
would want that decision to be up to me, and I understand why they’re fighting for those

rights. (Bethany, age 22)

Bethany does not contradict her religious ideology, but she instead argues that everyone sins. Her

support for gay rights is evident in the last half of the quotation, where she acknowledges the

existence of same-sex couples who love each other and who want to commit to each other. She

imagines a hypothetical scenario in which one member of the same-sex couple cannot take care

of their partner because their relationship is not legally recognized. In doing so, she affirms the

validity of gays’ and lesbians’ fight for same-sex marriage.

Some people’s tolerant attitudes about homosexuality appear to come from personal

contact with gays and lesbians, such that the religious teachings that homosexuality is wrong

does not seem to fit their personal experience. For example, Stephanie, a 50 year-old Catholic,

has relatives who are gay and says that she supports same-sex marriage because of it. At the

same time, she worries that same-sex marriage might normalize homosexuality:

Well, I think I’m a little bothered by it only in the sense that I have mixed feelings about

it. Because sometimes I think it becomes so overwhelmingly acceptable, then maybe,

maybe this is not founded in anything, but you know, younger people feel that this is just

another way to be. I have mixed feelings about it…. If we make it so widespread, would

kids feel that it’s just another easy choice? “Well, I don’t like this, so let me try that.”

(Stephanie, age 50)

Stephanie has mixed feelings about same-sex marriage because she does not think that

homosexuality is normal or should be viewed as equal to heterosexuality. While her personal

   202  
contact with gays and lesbians has led her to have positive attitudes about them, the older

religious and moral teachings that homosexuality is unnatural or wrong still influences how she

talks about the issue.

Even in the absence of direct personal contact with gays and lesbians, many young

conservative Christians still nevertheless appear to take an empathetic stance towards gays and

lesbians as a group. For example, Elizabeth had always grown up learning that homosexuality

was wrong, due to her religious upbringing. But the morning that I interviewed her, one of her

classes held a debate abut same-sex marriage, and she learned about the rights and benefits that

are denied to gays and lesbians as a result of their inability to be legally married:

As far as gay marriage, I don’t know a lot about the topic from the other point of view, so

it’s interesting to hear the person’s speech in support of it. I found out things like, you

know, they aren’t given a lot of the rights that they should be. So, I see that to be kind of

upsetting… (Elizabeth, age 19)

When I asked her about whether or not she thought gays and lesbians deserved equal rights, she

contradicted herself:

I don’t, not as far as marriage. I just don’t like the idea that a man and a man can get

married and then raise children…. As far as rights for people, I don’t like that they’re

discriminated against. I think that they should be viewed as people. I mean, don’t

discriminate against them just like you wouldn’t discriminate [against] someone because

of their race. But then I guess I’m kind of contradicting myself when I say that I don’t

think that they should have the right to get married. So, I don’t know, it’s kind of a

confusing issue for me. (Elizabeth, age 19)

   203  

Same-sex marriage is the limit to Elizabeth’s belief that gays and lesbians should have equal

rights, but she recognized that gays and lesbians can’t have equal rights without the right to

marry. So she struggled to reconcile the contradiction. Despite having very little personal contact

with gays and lesbians, her empathy towards gays and lesbians, as a group, conflicted with her

religious upbringing that taught her to oppose same-sex marriage.

Holding liberal political beliefs also appears to be a source of conflict with religious

teachings that homosexuality is wrong because of the familiarity that liberal ideology creates

with the argument that gays and lesbians are normal human beings deserving of equal rights. For

example, Gerald, who was quoted earlier talking about how he struggles with Catholic teachings

in light of his current political beliefs, jokingly describes how it is mostly through political

groups that he has come to know gays and lesbians:

Mostly through groups, like political groups and so forth, you know. When you start

associating with moveon.org type people, you know. My wife says, “You are so stupid to

reality sometimes, I don’t understand it.” I mean, I’ve interviewed, talked to two guys,

and walk away, and my wife says something, you know, “Didn’t you notice that they

were a couple?” I want to say, “I have, I don’t know, they’re just two guys.” (Gerald, age

60)

Gerald’s humor in this quote conveys simultaneously a process of learning to recognize gays and

lesbians and also a comfortability with interacting with gays and lesbians on these terms. As a

result of his liberal political beliefs, Gerald is becoming more familiar with gays and lesbians,
   204  
and he aligns himself with the struggle for gay rights. But the Catholic religious teachings about

homosexuality and marriage still cause him to struggle with the issue of same-sex marriage.

In sum, this religious variant of ideologically conflicted discourses is used to talk about

same-sex marriage by people who are influenced both by conservative religious teachings on

homosexuality and the liberal, tolerant ethos toward homosexuality that is characteristic of

contemporary mainstream secular American culture. While people who use this discourse think

of gays and lesbians as ordinary people deserving of equal rights, they also have been taught that

homosexuality is immoral and contrary to their religious beliefs. With this confluence of

attitudes and beliefs, simple discourses of support and opposition regarding same-sex marriage

are inadequate to express how people feel about the issue. The fact that students used this

discourse more frequently than parents in my sample may reflect their growing up during a

period in which gays and lesbians have been increasingly open and portrayed as normal in

American culture and politics.

Other than the two middle-ground discourses described here, there are most likely

additional middle-ground discourses that simply did not appear in my sample, whether because

of the geographical region that was the basis of my sample, the ages of people in my sample, or

for some other reason. For example, many gays and lesbians have mixed feelings about same-sex

marriage because they support the battle for equal rights but reject the institution of marriage

outright as a hetero-normative, procreative vestige of a society dominated by patriarchal gender

relations. There may be other social groups as well, like different ethnic or religious groups that

were not heavily represented in my sample, that use other varieties of discourse in order to talk

about same-sex marriage.

   205  
Conclusion

In this chapter, I have argued that the controversy surrounding same-sex marriage is

primarily about homosexuality, so much so that many people immediately begin talking about

homosexuality when asked about the issue of same-sex marriage. But rather than being a simple

conflict between those who oppose same-sex marriage and those who support it, I have shown

that discourses about same-sex marriage assume a wide variety of forms, including several

“middle-ground” discourses that cannot be characterized in terms of simple support or

opposition. This discursive complexity is bolstered by the fact that people’s cognitive beliefs,

affective attitudes, moral values, and life experiences about homosexuality are multiple and

complex in their own right, and that people combine these cultural elements in many different

ways in order to talk about same-sex marriage.

At an ideological level, there are a large number of possible combinations of elements in

people’s belief systems, which leads to a variety of discursive positions. There is thus no

inherent tendency toward irreconcilable conflict regarding this issue. To the extent that

attitudinal polarization occurs regarding same-sex marriage, it likely occurs because of social

and political factors: tendencies toward homophily in people’s social networks, the dynamics of

political campaigns, the ways in which mass media present only two sides of an issue in order to

be objective, and the ways in which opinion polls divide public opinion into support or

opposition. Absent these factors, in a conversational interview context, people’s discourse about

same-sex marriage attains a significant level of complexity and appears in many shades of gray.

However, discourses do appear to be patterned in sociological ways, especially according

to people’s religious and political ideologies and their cohort locations. Unsurprisingly, younger

people and people with liberal political beliefs are more likely to use supportive discourses, and
   206  
older people and people with conservative religious beliefs are more likely to use oppositional

discourses. Much more interesting is the fact that middle-ground discourses emerge when

people’s cohort locations are cross-cut by religious and political ideologies. Thus, the people

most likely to use libertarian pragmatic discourses to talk about same-sex marriage are older

adults with relatively liberal political views or tolerant attitudes about marriage and sexuality.

Similarly, the most common ideologically conflicted discourse in my sample comes from

students with conservative Christian backgrounds.

In describing this variation in people’s discourses, my goal has not been to make a claim

about the quantitative or qualitative importance of any of these patterns of talk; rather, I have

merely attempted to show that, to understand the controversy surrounding same-sex marriage, we

should reject the assumption of coherent worldviews and the simple categorization of opinions as

“for” and “against.” We must understand the full variation in the ways that people draw from

different attitudes, beliefs, values, and life experiences in order to talk about same-sex marriage.

In order to generate statistical estimates about how prevalent these different patterns of discourse

are in the population in large, we must generate survey instruments that are more capable of

measuring the complexity of people’s understandings of homosexuality.

In the next chapter, I shift my attention to the ways in which people’s definitions of

marriage add to the ideological complexity of this issue. Although in this chapter I have ignored

the “marriage” half of the “gay marriage” controversy, the varied meanings that the word

“marriage” carries in American society also plays an important role in shaping how people talk

about this issue.

   207  
Chapter 5: Talk about Marriage, Relationships, and Same-Sex Marriage

Introduction

Just as the last chapter described how people draw from their cultural understandings of

homosexuality in order to talk about same-sex marriage, in this chapter, I describe the variation

in how people’s discourse about same-sex marriage is shaped by their views on marriage. While

the quantitative analysis of survey data in Chapter 3 showed that people’s beliefs, attitudes, and

values about homosexuality are of fundamental importance in shaping public opinion about

same-sex marriage, the analysis of why people support civil unions (but not same-sex marriage)

suggested that people’s cultural understandings of marriage may also shape their attitudes about

this issue. Logically, to the extent that attitudes about same-sex marriage are different from

attitudes about other gay rights issues, the meanings that the word “marriage” has for people

should account for a significant portion of that difference.

Based on existing literature and the quantitative analysis in Chapter 3, I expect there to be

several potential conflicts regarding the meaning of marriage that play a role in shaping

discourses about same-sex marriage. First, there is a potential conflict between religious and civil

definitions of marriage. While the legal case for same-sex marriage applies to civil marriage, the

boundary between religious and civil marriage is blurred in the United States because the state

recognizes marriages performed by religious institutions as well. Second, many people believe

that marriage simply cannot apply to two people of the same sex because the term is defined as

between a man and a woman. Whether this definition of marriage is rooted in religious beliefs

(as in the case of Catholic ideology) or a cultural common-sense, it appears that some people

oppose same-sex marriage on these grounds. Third, there is potential conflict over the role of
   208  
procreation in marriage. Throughout the Twentieth Century in the United States, procreation has

become increasingly marginal to most people’s understandings of marriage and family, as

companionate and individualistic notions of marriage replaced the older institutional and

procreative meanings of marriage. However, there has also been a conservative backlash against

these changes, led by churches and other religious organizations, that aims to reinforce

patriarchal and procreative marriage and family structures.

Regarding the extent to which these conflicts will correlate with the cohort locations of

individuals in my sample, the expectations from the literature are unclear. On one hand, it is

plausible that there would be no difference in the meanings of marriage advanced by students

and parents because everyone in my study has grown up in a society in which the companionate

and individualistic understandings of marriage have been dominant. The relatively high birth

rates of the 1950s were an anomaly in the larger trend of declining fertility throughout the

Twentieth Century, but even then, the companionate understanding of marriage was dominant

(Mintz and Kellogg 1988). To the extent that students and parents have always considered the

emotional fulfillment of the couple as being more central to the idea of marriage than

procreation, the idea of same-sex marriage does not necessarily contradict this understanding of

marriage for either cohort.

On the other hand, there is evidence that notions of marriage and family have changed

since the time that parents came of age in the 1960s and 1970s, such that students might be more

likely to support same-sex marriage. For one thing, the rise in the divorce rate in the 1970s and

the increase in cohabitation rates has led to a diversification of family forms, a series of changes

that the parents lived through but students did not. Additionally, attitudes about gender, divorce,

cohabitation, and premarital sex have liberalized since 1970 (Thornton and Young-DeMarco
   209  
2001). Finally, some scholars have argued that contemporary meanings of marriage in the United

States are even more individualized than the companionate meanings of the mid-Twentieth

Century (Cherlin 2004). To the extent that such changes have occurred, it is plausible that

students will articulate different meanings of marriage than parents in ways that are relevant to

the controversy over same-sex marriage.

This chapter examines how the meanings that people attach to marriage are related to

these broader social changes and to their discourses about same-sex marriage. Without

quantitative survey data, it is impossible to estimate the relative magnitudes of the effects of

people’s conceptions of marriage and homosexuality on their attitudes about same-sex marriage.

However, based on the results that I present in this chapter, I argue that people’s cultural

definitions of marriage shape their discourse about same-sex marriage in important ways, but

that they play less of a role than their understandings of homosexuality in shaping individuals’

attitudes. The meaning of marriage appears to be less contested than the meaning of

homosexuality, both because homosexuality, unlike marriage, evokes a strong negative stigma

for many people and because there appears to be something of a cultural consensus about what

marriage means in practice.

After a brief discussion about the sources of data presented in this chapter, I first describe

the similarities in the discourses that informants used to talk about marriage, divorce, premarital

sex, and cohabitation. I then show how the different discourses about same-sex marriage

described in the previous chapter also incorporate particular notions of marriage. Finally, I

consider the significance of term “civil unions” in the debate about legal recognition of same-sex

relationships.
   210  
This chapter shows that the primary axis of disagreement about the meaning of marriage

in the same-sex marriage debate is between religious and heterosexual definitions on one hand,

and between civil and companionate definitions on the other. Those people who say that

marriage is, by definition, a religious institution are more likely to oppose same-sex marriage

than those who articulate a civil definition of marriage. Similarly, advocates of the companionate

definition of marriage believe that marriage is defined by any two people who love each other

and want to commit to each other for the rest of their lives, while advocates of the heterosexual

definition of marriage believe that marriage requires a gender-differentiated couple.

However, the divide between these definitions of marriage is less stark than one might

imagine because, for most people, companionate marriage operates as an implicit, common-

sense understanding of what marriage means in practice. This manifests itself in the pragmatic

and individualistic language that most people use to talk about marriage and the paradoxical

agreement that sexual attraction, but not procreation, is essential for marriage. Because people

generally agree that procreation is not essential to marriage and that what makes a good marriage

depends on the unique characteristics of the couple, the definition of marriage constitutes a

weaker ideological basis for opposing same-sex marriage than it might once have. The historical

period, after 1969, that ushered in a liberalization in gender ideologies, diversification of family

forms, and greater acceptance of sex outside of marriage has strengthened the cultural

foundations for supporting same-sex marriage by weakening the marriage ideal that is inherently

gendered and procreative.

Finally, this chapter shows that the significance of civil unions in the same-sex marriage

debate is as a proxy for gay rights, rather than a legitimate alternative legal status. Taken

literally, as an alternate legal category for same-sex couples, most informants reject civil unions
   211  
as a solution to the controversy. However, support for same-sex civil unions can be seen in the

discourse of those who simultaneously believe that same-sex couples should have the rights and

benefits that come from marriage but not the right to marry. For some people who hold religious

and heterosexual definitions of marriage, the idea of civil unions represents a viable way to talk

about extending civil rights to gays and lesbians while reserving marriage for opposite-sex

couples.

The Social Significance of Talk

As in the previous chapter, the data presented in this chapter come from the pooled

sample of all 97 individual interviews. Much of the data presented here also comes from the

same sections of the interview guide, which contained a series of questions and probes about

same-sex marriage and homosexuality. There is also an additional section of the interview guide

from which data presented in this chapter are drawn. A discussion about marriage and

relationships preceded the discussion about same-sex marriage and homosexuality. This

discussion began with the question, “What does the word marriage mean to you?” It was

followed by a series of questions about how important various characteristics are for a good

marriage; then by a series of questions about marriage-related issues like cohabitation, premarital

sex, and divorce. The discussion about marriage and relationships concluded with a short

discussion about the perceived decline of the institution of marriage.

I took care never to mention the issue of same-sex marriage before this discussion of

marriage and relationships happened. It is likely that some participants knew ahead of time that I

would ask about same-sex marriage, because although I specifically asked students not to tell

their parents about the details of the interview, I had no control over what participants said to
   212  
other participants. Nevertheless, most interviews contained no evidence that the informant knew

ahead of time that the issue of same-sex marriage would be discussed.

The talk about marriage described in this chapter therefore comes from three different

sources: an explicit discussion about marriage in the abstract, an explicit discussion about same-

sex marriage in particular, and an analysis of the implicit meanings of marriage that people use

in order to talk about topics that push the boundaries of marriage, such as cohabitation, civil

unions, and heterosexual same-sex relationships. Each of these sources of data provide unique

insight into how the cultural understandings of marriage shape people’s discourses about same-

sex marriage and the controversy surrounding the issue.

Cultural Common-Sense Understanding of Marriage

In contrast to the politicized and emotional disagreements among informants in how they

talked about homosexuality, there were a number of surprising similarities in the ways that

informants talked about marriage. While the politics surrounding marriage, sexuality, and family

appear to have become increasingly divisive in the last three decades, due to the mobilization of

fundamentalist and conservative evangelical religious groups, I found that students and parents,

religious and non-religious alike, talked about marriage in ways that generally bolstered a shared,

companionate understanding of marriage. In particular, the prevalence of positive attitudes about

marriage; the pragmatic centrism that dominates discussion of divorce, cohabitation, and

premarital sex; and the pragmatic, individualistic language that people used in order to talk about

the characteristics of a good marriage, indicate that there is something of a cultural common

sense about what marriage means in practice that is relatively broadly shared and non-

politicized.
   213  
Ideologically, the cultural common sense that dominated informants’ discourse about

marriage is compatible with the idea of same-sex marriage, in the sense that there was no

component of this definition that was ideologically incompatible or hostile to it. To be sure,

some informants rejected marriage as a positive institution or attached relatively little meaning to

the commitment signified by marriage, while other informants, on the other side, advanced

deeply religious and procreative definitions of marriage. However, by and large, individuals

talked about marriage using language that affirmed marriage as a positive social and sexual

institution for two individuals who love each other and want to commit to each other for the rest

of their lives, and that is relatively freed from strict procreative, gender-differentiated, and

religion-specific moorings.

In describing this cultural consensus in how most informants talked about marriage, I do

not wish to imply that the differences in how people talk about marriage are unimportant or that

this cultural consensus on marriage’s practical meaning is true in some absolute, non-context-

specific sense. To the contrary, the discourses described here are a product of the types of

questions asked and the social context of the interview, which required informants to answer

abstract and hypothetical questions using the cultural tools available for them to do so; the

politically consequential differences among people’s attitudes and views on marriage were in

some ways minimized by my research methods.

I focus on the similarities, rather than the differences, among informants’ discourses in

this analysis because social generational change does not just imply differences among people; it

also means that there are important patterns of agreement among people. As Esler (1984)

persuasively argues, analysis of the patterns of agreement among people is essential to social

generational analysis because it illuminates the period-specific understanding of reality that

   214  
people take for granted. This analysis therefore shows that there exists a cultural common-sense

understanding of marriage, broadly shared by Americans of different ages and political and

religious backgrounds, that is ideologically compatible with same-sex marriage to an extent that

would not be possible without the broad changes that have occurred since the 1960s with respect

to gender ideologies, family structures, and attitudes about sexuality.

Marriage is Good

The first element of the cultural consensus on what marriage means in practice is its

status as a positive social institution that embodies high social values, such as love and

commitment, to which people should aspire. Pursuant to the high regard in which marriage is

held in mainstream American culture (Cherlin 2009), informants in my study generally

expressed positive attitudes with regard to marriage and talked about marriage in ways that

reflected its revered status. When I began the discussion about marriage with the question, “What

does the word ‘marriage’ mean to you?” the most common response had to do with love and

commitment:

Commitment from both sides to, to love each other and show each other that you love

each other. I don’t think it’s just being together. You have to have, you have to have

feelings for someone and keep that. (Bernice, age 53)

Marriage, it means that, well to me, it means that you’re entering into a contract between

two people who agree upon devoting their life to each other and loving one another for

the rest of your lives. And it’s a sacred thing, and it’s a blessing. (Demarcus, age 23)
   215  
Um, a commitment, like life-long commitment. Like once you get married, that’s it. No

do-overs, so you better make sure it’s the right person. I think it’s a really good thing.

(Sandra, age 29)

Similar responses were given by young and old, liberal and conservative, religious and non-

religious alike. Sometimes the responses by religious conservatives or secular liberals

specifically mentioned the sex composition of the couple, and some religious conservatives

defined religion as a spiritual institution, as does Demarcus in the quote above. However, beyond

these specific statements, most informants used a shared vocabulary of commitment and love to

talk about marriage.

The positive nature of people’s evaluations of marriage is underscored by the fact that

some parents and children who had experienced divorce still viewed marriage in a positive light:

Q: What does the word ‘marriage’ mean to you?

A: Commitment and respect, partnership, trust, faithfulness, and everything that comes

around with being a companion, a life-long companion. And I’ve seen what a marriage

does to children especially. And I know that I want to have children one day, and I also

know that I’m never going to get divorced, cause I know what it’s like and I’ve been

there. And I don’t think my parents gave it enough effort. (Gabe, age 24)

Paradoxically, many people’s negative experiences with divorce seem to have strengthened,

rather than undermined, the positive characteristics that they associate with marriage. The

positive associations that most people have with marriage is also reflected in the fact that most

students said they hoped or planned to get married (or remarried) some day:
   216  
Me, I won’t deny it: I hope one day to get married. You know, to find that person that is

finishing my sentences before I’m even starting them. And a bad day between me and her

is when I stub my toe getting out of bed. (Simon, age 24)

Q: Do you want to get married again at some point?

A: I could see it hap—I do want to get married again at some point. But I want to make

sure it’s that person,

Q: Won’t make the same mistake?

A: Exactly. I mean, it’s not deterred me from the decision of getting remarried some day.

(Dylan, age 23)

Most informants were familiar with the negative view of marriage as an institution

representing a loss of freedom, but few of my informants subscribed to it. Most rejected that

portrayal as an unrealistic construction by mass media and popular culture that was meant to be

entertaining or dramatic:

My view has changed. I think I grew up watching comedians and listening to people,

marriage was just this death trap, you know. I used to listen to morning wise-crack shows

and comedians and then you just, one of my favorite comedians says, “A guy gets

married just because a fucking guy gives up!” [laughter]…. But seeing, I just came from

a wedding three weeks ago… I haven’t been around a lot of weddings, but I went to two

this year so far… and I know these guys like the back of my hand, and it’s, to see them

just so ecstatic about it, and so, I mean I think it’s more of a positive thing than a negative

thing, so. (Jordan, age 27)

   217  

I think a lot of times people kind of make fun of it or, you know, they make it seem like,

“oh, you’re married, your life sucks,” you know. It’s so much more fun to be single and

do whatever you want. And then I also think that they tend to kind of raise your

expectations of it at the same time. Like I think people get married and they expect for

everything to be perfect… but I think people don’t realize that that’s really not how

reality works. (Carolina, age 20)

Jordan and Carolina reject the media portrayals of marriage, whether very negative or very

positive, as unrealistic. The reality of marriage is that it is a positive thing, but it is also hard

work. Many informants criticized media portrayals of marriage that glorify sex and divorce, as

opposed to realistic portrayals of couples trying to do the hard work of making a marriage work:

I don’t think they respect marriage really as much as it should be respected. They don’t

put the value on it. Americans think that everything should come easy to them and they

don’t have to work at anything… It’s no big deal if someone gets divorced. It’s no big

deal if someone cheats on somebody. It’s just, that’s what happens. That’s how it’s

portrayed in the media…. It’s just not valued. (Ron, age 20)

Religious conservatives like Ron were especially likely to criticize mass media for glorifying sex

and divorce, but even secular liberals identified the portrayals of celebrities’ relationships as

being at odds with what marriage is in reality. They shared a common perspective of defending

marriage against its unrealistic or negative portrayals in media.

   218  
There were noticeable age differences in how parents and students talked about marriage,

reflecting the direct, personal experiences that most parents had with marriage that their children

did not have. However, the age differences were not manifested in their attitudes about marriage;

rather, parents were more likely to speak from direct personal experience and provide specific

stories in response to my questions, whereas students were more likely to draw from general,

common-sense views about marriage to respond to my questions. For example, Stan and his son

Kyle both described marriage as being about commitment, sacrifice, and love; but while Kyle

tended to speak in generalities or answered my questions by talking about his girlfriend or books

he has read, Stan spoke at length from his personal experience:

The word marriage to me is basically kind of synonymous with commitment. And

commitment is really, has it’s basis in, not in feeling, but in decisions or in actions, in

demonstrations. And so, to me, marriage is more of a demonstration of your faith and

commitment to another person. Now, the only problem with that is we’re imperfect. And

to make something good in an imperfect world, you have to work at it. And to me, that

takes a lot of work, I mean, a lot of devotion to looking at the other person’s side of

things, learning patience, and learning to develop sensitivity. I mean, we’ve been married

over 25 years, and I don’t feel I’m anywhere close. It’s just, again, it’s a matter of

retraining my mind, and I think that the commitment part of it, that has to be there,

because if you don’t, in my mind, if you just have the relation or the feeling part of it,

eventually, that kind of wanes after a while…. You know, I had a failed marriage, and I’d

be the first to admit that it wasn’t the other person. Yeah, they may have made a decision

to leave also, but it was because of my negligence and because of my lack of

commitment to that relationship early on that contributed to a lot of the attitude changes
   219  
and then turning into animosity and then turning into hatred and then just splitting up.

(Stan, age 59)

In the quote above, Stan uses both his positive and negative experiences in marriage to describe

concretely how commitment is important to marriage. His son identified commitment as

something that is essential to marriage, but he did not have the same quality or quantity of

experiences to draw from in order to explain how. Thus, age-related differences in how students

and parents talked about marriage had less to do with cohort differences than with age

differences, the greater amount of experience that parents had accumulated over time.

When Divorce, Premarital Sex, and Cohabitation are Good or Bad

The second element of the cultural consensus of what marriage means in practice is

pragmatic and centrist discourses that people used to talk about moral dilemmas associated with

marriage: divorce, premarital sex, and cohabitation. Historically, these have been controversial

topics in American culture because of the perceived threat that each poses to marriage, but all

three have become increasingly accepted in the U.S. since the 1960s. The ways in which my

informants talked about these three issues reflect the fact that tolerance for divorce, premarital

sex, and cohabitation has increased, but the discourses tend to be centrist and pragmatic rather

than rigidly ideological. Both conservatives and liberals express mixed levels of opposition and

support for these issues because the potential benefits or negative consequences associated with

all three are widely recognized. It is not a question of if they are good or bad, but when they are

good or bad.
   220  
As the preceding paragraphs have suggested, people’s experiences with divorce did not

necessarily undermine the positive values associated with marriage; rather, divorce was

constructed as its antithesis. Contrary to the fears expressed by some religious conservatives, few

people viewed divorce as unproblematic; almost all informants viewed divorce as negative,

evidence of some failure to achieve the high values represented by marriage. This negative

evaluation of divorce was only one of the common features of both liberal and conservative

discourses about divorce.

Other than its relatively unquestioned status as a negative outcome of marriage that

should be avoided if possible, informants talked about divorce in similar, and centrist, ways.

Almost all informants recognized the importance of easy access to divorce in cases involving

spousal abuse and infidelity, and almost all informants believed that divorce should not be used

as an easy escape. Whether liberal or conservative, almost all informants said that couples should

go through counseling and exhaust all other reasonable alternatives before going through a

divorce. It was also common for informants, regardless of their political and religious beliefs, to

argue that many divorces occurred because people were too eager to get married or because they

did not go through enough premarital counseling or self-reflection.

I do not want to overstate the similarities in discourses about divorce; religious

conservatives frequently said that divorce is wrong and should be avoided at all costs, while

secular liberals frequently argued that divorce is perfectly justified because there is no reason

why a person should go through life unhappy. However, beneath the rhetoric, partisans on both

sides qualified their beliefs in significant ways. Many conservatives recognized that people make

mistakes and that divorce can allow people to have second chances. For example, Dana, a 48
   221  
year-old born-again Christian, opposes divorce on religious grounds but also recognizes that

whether or not divorce is acceptable depends on the situation:

I believe that God can heal any relationship if you truly are seeking Him, but I’ve seen

too many cases that one person is seeking and one person is not. And in those cases, I

don’t think that you seek the divorce if you’re trying really to save it, but if your spouse

leaves, then you allow them to leave…. I’ve seen situations where because of the kids,

[the parents] have stayed together, and it has made the children’s lives hell. The children

have learned such sinful behavior that it takes years of counseling for the kids to be able

to manage their lives correctly. But at the same time, I’ve seen a situation where a

marriage has been struggling, and if it’s two people that are really trying to work through

it, kids have learned to persevere and to trust God. So you have two situations, and I think

it just depends on the situation. (Dana, age 48)

While Dana uses deeply religious language to talk about her opposition to divorce, she also

recognizes that divorce can actually result in a better outcome than staying together. Similarly,

liberals believe that divorce should be used by people to get out of bad relationships, but they do

not think that the decision to get divorced should be taken lightly:

If it’s something that you know [you] can’t work past, then it’s definitely healthy for

everybody involved to just break it off, I guess…. But you know, if it’s like something

dumb, or a phase and “let’s get a divorce,” and something happens and you’re like, you

can’t figure out what will make you happy because you already gave it up, that’s you

know, that’s where it’s like I don’t, like all these quickie marriages, quickie divorces…

The last, you know, 20 years at least, it’s gotten so messed up as far as, you know, who’s
   222  
getting married when and for what reasons, and then they want to get a divorce. And it’s

just like, you know, I absolutely have no issues with divorce when done for the right

reasons, you know. If it’s done for the wrong reasons… they’re just wussing out of

working through something. (Terrence, age 19)

Terrence believes that divorce is necessary and good for couples that have been unsuccessful in

their efforts to work through irreconcilable differences, but he also does not respect people who

get divorced without trying to work through their problems. In this way, both liberals and

conservatives qualify their beliefs about divorce in ways that create some common ground in

their discourses.

Another source of common ground among liberals and conservatives is that informants

viewed divorce as problematic primarily because of its potential effects on children. As the

previous quote by Dana suggests, people’s discourses about divorce are shaped by concerns

about the practical effects that divorce has on others. In this sense, discourses about divorce are

pragmatic, rather than strictly ideological:

Q: Do you think divorce is a big problem in our country?

A: To be honest, I think a lot of people get married too fast without thinking about the

decision. Then, they end up with kids, and then the kids, I mean psychologically I think

the kids are affected…. I think it’s always going to affect that child and for the rest of

their life, it’s part of their formation. So, and it’s definitely going to affect the way they

think about marriage and family and all that kind of thing. So yeah, I would say it’s a

problem. (Emily, age 19)

   223  
Emily, a Catholic, agreed that divorce is a problem because of its effects on children, moreso

than because of her religious beliefs. But staying together for the sake of the children is no less

problematic because then children are exposed to the negative relationship dynamics of their

parents:

Q: What do you think about parents who stay together for the sake of the children?

A: I think that the kids know. “Mommy and daddy hate each other.” I don’t, I wouldn’t

want my mom to have, I wouldn’t have wanted my mom to have stayed with my dad. I

think it was in the best, um, it was best for everyone that, like I said, my brother and I

were her number one priority, to where him coming home [on drugs] was not

acceptable…. If you’re just there for the kids, then that could be more detrimental to the

kids than it can help, because if they’re in an environment that’s hostile, what does that

teach them about marriage? (Jane, age 19)

In Jane’s case, divorce was better for the children than not getting divorced. In these ways, then,

discourses about divorce were pragmatic and centrist in ways that made the ideological divide on

this issue less sharp than it otherwise would have been. The fact that divorce can have positive

effects in some circumstances but that it also represents a failure to achieve the high values of

marriage meant that conservatives and liberals shared similar ways of talking about divorce.

Similar dynamics appear to influence the discourses regarding cohabitation and

premarital sex. Tolerance for cohabitation and premarital sex has increased significantly since

the 1960s, and the relative pragmatic centrism of the discourse about these issues reflects that

fact. Most informants denied that there is anything particularly wrong with either premarital sex

or cohabitation, but few people expressed an unqualified belief that they are always good. Like
   224  
divorce, almost all informants acknowledged potential negative consequences that premarital sex

or cohabitation could create for either the individual or the couple. Thus, the moral dilemma

associated with these issues is about the conditions under which premarital sex and cohabitation

are right and wrong, not whether they are always right or wrong.

For informants who did not object morally to premarital sex, their acceptance of it was

tempered by concerns about unwanted pregnancies, sexually transmitted diseases, the

individuals’ ages, and the seriousness of the relationship. Most informants premised their support

for premarital sex upon the conditions that the individuals are emotionally mature, in a stable and

committed relationship, and acted responsibly to prevent unintended negative consequences.

Even conservative parents and students who planned to remain abstinent until marriage

expressed conditional acceptance of premarital sex rather than outright opposition:

That’s something you have to be serious about before you do it. Just to jump into

somebody’s bed after you’ve got a one night stand is not right. I think you should think

seriously whether you want to get involved with anybody, and you’ve got to think about

the consequences, too. I mean, there’s so much, uh, so many STDs around… You just

have to be really, really careful. I think too many young people do not take it seriously.

(Sarah, age 60)

Sarah’s quote shows how age, the seriousness of a relationship, and concern with the potential

consequences of premarital sex structure people’s discourse about premarital sex. Even though

Sarah holds fairly conservative values, she accepts the reality of premarital sex conditionally.

Many religious conservatives were hesitant to morally judge a person who had sex before

marriage, even if they personally might not approve of it:

   225  
I’m sure it’s great. [laugh] Many of my friends, besides one other friend who is a

Christian that I’ve known through school… they’re sexually active. I’m looking forward

to it, I guess you could say, but again, it’s not, I don’t condone it. I don’t think it’s right.

But I understand, and again, I’m not, I don’t judge my friends. They know how I feel, and

they do what they want, and that’s not my business. (Ron, age 20)

Even the discourse of young liberals about premarital sex reflects the tension between the moral

judgment of premarital sex and concern with its consequences:

I used to think that, you know, that would be something better saved for marriage, but I

guess, quite honestly, I just realized that, you know, everybody else is doing it and

nobody really feels that way anymore. So I think it’s normal, and I don’t think it’s

necessarily a bad thing either, as long as you’re careful and selective about it, you know.

I don’t think that like, being promiscuous is a good thing, but I don’t think it’s too

common to do that. (Carolina, age 20)

Carolina’s quote nicely encapsulates the connection between moral concerns about premarital

sex, its increasing acceptance, and concern with potential negative consequences. She was taught

that sex should be saved for marriage, but she realized that such an attitude is considered old-

fashioned. As a result, the important question for her is not whether premarital sex is right or

wrong in some absolute sense, but the conditions under which it is right or wrong.

Thus, the moral problem associated with premarital sex, for most informants, does not

have to do with the relationship between marriage and sex, but with the difference between

healthy and unhealthy sexual relationships. All but the most conservative of religious
   226  
conservatives did not associate sexual activity strictly with marriage, so much of the discourse

about premarital sex had little to do with marriage. Rather, informants implicitly contrasted

healthy sexual relationships outside of marriage with unhealthy sexual relationships outside of

marriage—as though they took for granted sex outside of marriage happens. To the extent that

individuals are mature, in a committed relationship, and responsible about reducing potential

negative consequences, there are no problems with the sexual relationship. By contrast,

unhealthy sexual relationships were those that failed to meet these standards and were thus

subject to moral disapproval by most informants.

The same is true for cohabitation, although the conditions on which it is considered good

or bad are slightly different. This should not be surprising because religious conservatives

oppose cohabitation precisely because of its connection with premarital sex:

Q: How do you feel about cohabitation, that is, a couple who lives together even though

they are not married?

R: Naturally I frown on that a little bit. But if their intent is that they’re going to be

together forever, I mean, I’m a male. I’m enjoy sex… I frown on it, but I understand it….

Q: Would you feel differently about it if a couple was like, “We want to wait until

marriage for sex, but we want to live together to see what it’s like,” or for financial

reasons?

R: Of course I’d feel differently… But most couples are out there cohabitating, and I’ll

say 99% is, they’re cohabitating for sexual reasons. (Vincent, age 48)

Like Vincent, most religious conservatives who objected to cohabitation did so because of its

connection with premarital sex, and they doubted whether a couple could cohabitate and remain
   227  
abstinent because of the sexual attraction that they took for granted as part of the relationship.

However, concerns with cohabitation differed from concerns with premarital sex because the

potential negative consequences that people worried about had to do with it hastening the end of

the relationship, not with unintended pregnancies or STDs:

I don’t agree with it. Statistics show that if you do, your chances of, if you do get

married, having a divorce, they increase drastically. Even if you don’t get married, odds

are it’s a lot easier to break up that relationship because you don’t have that devotion

right when you go into it. It’s more of a, “Well we’re just going to live together, if it falls

apart, it falls apart.” You have that when you go into it, so it’s a lot easier for that to

happen. (Paul, age 19)

Many informants believed that cohabitation placed couples at greater risk for divorce or that it

might encourage them to break up if they found out things about their partner by living with

them that they didn’t like. These possibilities made many informants worry about whether or not

cohabitation was a good thing, even if they did not have any moral objection to it.

Because of pragmatic concerns were at the center of many informants’ discourses about

premarital sex and cohabitation, informants who were very positive toward both of these issues

justified their attitudes because of the pragmatic benefits that they imagined premarital sex and

cohabitation would provide. They argued that premarital sex would help strengthen a couple’s

relationship by changing the emotional dynamics of the relationship, and they argued that

cohabitation was good precisely because it would allow a couple to see what it was like to live

together before they made a marital commitment:

   228  
Personally, I kind of view it the same way as I view premarital sex. Living together is just

like sleeping together. When you get to that point it’s going to bring on a whole new set

of emotions and events that I feel need to be worked out before you marry the person.

Because in this day and age, with as easy as it is to get a divorce…. I mean, if you’re just

dating a person and you decide to move in together and it don’t work out, it’s just the end

of a relationship. If you end up marrying the person and then move in with them and it

don’t work out, now you’re a statistic. (Simon, age 24)

Informants likened cohabitation to a test or try-out period for the relationship:

If I were to marry someone, I wouldn’t want to marry them without knowing if I can live

with them. I would want to live with them for a little bit, and okay, test the water, and

okay, “let’s get married.” (Jeff, age 21)

I don’t see anything wrong with it. As a matter of fact, it’s like the old saying, “Ride the

horse before you buy it.” So. You can quote that too, if you want to [laugh]. (Matthew,

age 51)

By cohabitating, couples are simply doing the smart thing: they would be better able to

determine whether or not the marriage would last by learning about the bad things, as well as the

good things, before they made a commitment. Thus, people’s talk about cohabitation and

premarital sex were dominated by pragmatic concerns about the conditions under which good or

bad consequences would happen.

   229  
Especially when compared with discourse about homosexuality, discourses about

divorce, premarital sex, and cohabitation appear relatively uncontroversial and non-politicized.

This is somewhat puzzling because the divorce rate began increasing during roughly the same

period that gay liberation period began, as did the increasing acceptance of premarital sex and

cohabitation. Logically, all of these changes could result in significant cohort effects in the

discourse. Why would cohort differences in the discourses about homosexuality emerge and not

in the discourses about these other issues? Several possible explanations suggest themselves.

The first possible explanation is methodological. I simply may not have asked the right

types of questions, or my analysis may not have been nuanced enough to capture the cohort

differences that do exist. A second possible explanation has to do with heteronormativity:

homosexuality threatens the societal norm of heterosexuality, while divorce, cohabitation, and

premarital sex do not. A third possible explanation is that informants, whether young or old,

liberal or conservative, have been personally more affected by these other issues than by

homosexuality; a pragmatic and centrist discourse may reflect an acceptance of the reality of

divorce, premarital sex, and cohabitation, in the sense that they affect people personally whether

or not they disapprove of them.

In reality, all three of these factors are probably important. Whatever the correct

explanation, the centrist and pragmatic discourses about divorce, cohabitation, and premarital sex

underscore the fact that disagreements about homosexuality appear to structure the debate over

same-sex marriage moreso than disagreements about marriage. Same-sex marriage is

controversial because there is greater disagreement over the meanings associated with

homosexuality than over the meanings associated with marriage. To the extent that there is a

cultural consensus about what marriage means in contemporary American society, it does not
   230  
resemble some “traditional” ideal of marriage; rather, the contemporary understanding of

marriage incorporates the reality that couples get divorced, engage in premarital sex, and live

together outside of marriage.

Essential and Non-essential Characteristics of Marriage

The third element of the cultural consensus of what marriage means in practice is

reflected in the pragmatic, individualistic language that people used to talk about the

characteristics of a good marriage. When asked about the characteristics that are necessary if a

couple wants to have a good, strong marriage, informants were circumspect. Informants

identified some characteristics, like love, trust, and communication as absolutely essential for a

good marriage, but they thought other characteristics were important only for some people or in

some circumstances. For example, holding similar religious beliefs or having similar interests are

characteristics that were widely acknowledged to vary in importance, based on the particular

personalities and interests of the couple.

The pattern of characteristics that were considered essential and non-essential to marriage

is sociologically significant because they suggest that informants hold a cultural common-sense

definition of marriage that is more consistent with a companionate definition of marriage than of

a procreative or institutional definition. This is true, not only because informants considered the

couple’s fulfillment in their relationship to be central to a good marriage and because they

recognized that all couples are different with respect to what makes their relationship work, but

also because most informants viewed procreation and child-rearing as non-essential to marriage,

even as they considered sex to be essential. An understanding of marriage that places a couple’s

happiness and sexual attraction at the center, while marginalizing procreation and an idealized
   231  
gendered division of labor, is compatible with the idea of same-sex marriage to a much greater

degree than if procreation and traditional gender ideologies were considered essential to

marriage.

The fulfillment and happiness of a couple in their relationship was so central to the

common-sense idea of what a marriage means in practice that there was near-unanimous

rejection of a procreative definition of marriage among my informants. When asked how

important raising children is for a good marriage, most people said that having children or raising

children was not essential for a married couple:

I don’t think it’s important at all. I mean, it’s personal, a personal choice. I guess some

people would say we’re here to have children, and that’s what we’re here for. I don’t

really think so at all. I mean, it’s just a personal choice. It doesn’t really matter, as long as

you’re happy. (Claudia, age 22)

Even some informants with heterosexual or religious definitions of marriage said that having

children was not necessary for marriage:

That’s one thing I kind of agree and disagree on, because the two reasons for being alive

in the Catholic religion is to procreate, to have kids, and to marry. So, I believe that

having children is important if you want them… Because my mom and dad both have

friends that they don’t have kids, and they don’t ever want to have kids. And if you don’t

want to have kids, I mean, you are the kind of person that you don’t need to have a kid,

because it’s not going to help them any. (Jill, age 19)
   232  
I think that’s a very personal decision. I know people that do not have children and made

that decision not to have children, and I think it’s wonderful. I think it’s wonderful that

they have the courage to simply state, “No, I don’t think I have that in me.” (Andrea, age

45)

Even people with strong religious worldviews like Andrea and Jill said that a couple should not

raise children if they do not want to. To them, it is obvious that there is no reason why a child

should be born or raised by parents who do not want them. There is a pragmatic logic in the

quotes above, that whether or not a couple should have children depends on whether or not it will

fulfill the needs and desires of the couple. The cultural common-sense that parents should not

have kids if they do not want them shows that the companionate definition of marriage makes

sense to people, even if they otherwise have religious or heterosexual understandings of

marriage.

Despite the fact that procreation is not considered essential to marriage by most

informants, they do consider sex and sexual attraction to be essential to the definition of

marriage. While most informants did not mention sexual attraction as among the most important

characteristics of a marriage, when probed, they said that sexual attraction is essential, at least in

the beginning:

Well, I probably wouldn’t be married if it hadn’t happened. [laugh] But um, once you

have that commitment and uh, you know, basically knowing that over the years you’re

going to change, that um, you have to learn to adjust to the changes. (Donna, age 60)
   233  
I think it’s really important, because if you’re not sexually attracted to that person, then

you’re obviously going to find someone who’s not your husband or wife that is attractive,

and that can cause problems. (Renee, age 19)

As is evident in these quotes, sexual attraction is viewed as important both because it brings

couples together in the first place and because it can help prevent infidelity.

This talk about sexual attraction does not show that sex is essential to the definition of

marriage, only that it is important. However, informants’ reactions to a hypothetical scenario that

removed sex from the definition of marriage does show that marriage is defined, at least in part,

by sex. Generally at the end of the discussion about same-sex marriage, I asked people who did

not strongly oppose same-sex marriage what they would think about letting two heterosexual

individuals of the same sex, such as best friends or college roommates, get married. The question

elicited a variety of reactions—from laughter to confusion to anger—which show that the

meaning of marriage would be strongly violated by this example:

Q: What would you think about like two people of the same sex, let’s say they’re good

friends, you know, they were college roommates or something... but they’re both

heterosexual, so they’re not attracted to each other. Um, but they decided they wanted to

get married. What would you think about that?

R: I think it would be pretty silly. [laugh] That’s very silly. No, that’s crazy. I don’t know

why, but it just seems silly. [laugh] (Stephanie, age 50)

That doesn’t make any sense. Oh, oh I see, for like the legality matters like the

insurance… I think they’re both wasting a chance at finding their life partner, but it
   234  
depends on the situation. If they are both happy single and are cohabitating and don’t

want children, then that’s fine, too. I just don’t quite understand it. [laugh] (Bonnie, age

46)

That’s kind of a loophole; I think that’s filthy. That’s not love. Marriage is about love.

That’s a, they’re obviously abusing the word marriage and the idea and concept of it and

using it for financial reasons… I wouldn’t like to hear about that. (Trey, age 21)

Some informants realized that opposite-sex couples sometimes get married for reasons other than

sexual attraction, like when people marry in order to gain citizenship status. However, the fact

that this scenario made little sense to people shows how essential the idea of sexual attraction—

and hence, sex—is to people’s understanding of marriage. That sex is integral to the meaning of

marriage is so obvious to most people that there is no reason to even talk about it. But the

importance of sexual attraction to the meaning of marriage becomes evident when people are

asked to imagine marriage without it.

Lastly, the ways in which informants talked about the division of labor within the

household shows that most informants do not advocate a strict gendered division of labor. This is

unsurprising from young, secular liberals; but even some older religious conservatives spoke of

marriage as a partnership in which each person does what has to be done, regardless of gender.

Some even denied the existence of inherent gender differences between men and women, despite

holding religious and heterosexual definitions of marriage:

Q: How do you think a couple should divide the responsibilities of earning a living,

taking care of the house, raising the kids, those kinds of things?
   235  
R: 100% each way. Not 50-50. You put in 100% and the other significant person puts in

100%.

Q: On everything?

R: On everything. That’s what relationship is. It’s not 50-50; it’s 100-100.

Q: Do you think that men and woman are naturally better at certain things than the other?

R: No. I taught my wife how to cook and clean. I taught her relationships, how to have

it…

Q: Do you think that women are like naturally more nurturing than men?

R: She wasn’t nurturing. She had a horrible relationship growing up with her mom...

(George, age 50)

In this exchange, George draws from his personal experiences to deny the commonly-held views

that women are more nurturing than men and to insist that marriages require full commitment

from both partners to do what needs to be done. While many religious conservatives answered

these questions differently than George did, most individuals talked about the division of labor in

a marriage in pragmatic terms, not in ideological terms.

Of course, talk should not be mistaken for action. I assume that a significant proportion

of my informants act, or will act, in ways that reinforce traditional gender ideologies, no matter

how egalitarian their discourse may seem. Nevertheless, I would argue that it is significant that a

strict separation of spheres within the family was rarely advocated in my sample. I concur with

Judith Stacey’s assessment that, “Although the division of household labor remains profoundly

inequitable, I am convinced that a major gender norm has shifted here” (Stacey 1990, p. 268).

Whereas the patriarchal division of labor was the hegemonic ideal of white middle-class
   236  
American families in the 1950s, it is not so today. The only advocates of traditional gender

ideologies in my sample were religious conservatives who argued that men should be the leaders

of the household. For the most part, even informants who insist that marriage is between a man

and a woman resemble those who advocate companionate definitions in that they did not

associate the sex composition of the couple with the successful functioning of a household.

This cultural consensus on the companionate definition of marriage, which is held even

by people who advance religious and heterosexual definitions of marriage, strengthens the

ideological position of advocates of same-sex marriage. The fact that few people consider

procreation essential for marriage, combined with a relative decline in traditional gender

ideologies and the individualistic and pragmatic ways in which people speak about marriage,

means that there are fewer ideological grounds for opposing same-sex marriage today than there

might have been in the past. Even though gender matters in the religious and heterosexual

definitions of marriage that some people advocate, as I describe below, in the common-sense

cultural understandings of what marriage means in practice, gender differentiation of the couple

matters far less than other characteristics, such as love, communication, and sharing similar

values.

Moreover, the fact that people consider sex and sexual attraction essential to marriage

means that the question of marriage for people with homosexual attractions is a valid one; it

cannot be dismissed by suggesting that gays and lesbians can get married to people of the

opposite sex, because they would not be sexually attracted to their partner. To suggest that

someone marry a person that they are not sexually attracted to undermines people’s common-

sense understanding of marriage. This is no dilemma for individuals who consider

homosexuality to be simply a sinful temptation that individuals must resist. However, for people
   237  
who accept homosexuality as a legitimate sexual orientation, their exclusion from the rights and

benefits of marriage is a problem.

To sum up, the positive attitudes that people express about marriage; the pragmatic and

centrist ways in which people talk about divorce, cohabitation, and premarital sex; and the

individualistic and pragmatic ways that people talk about the characteristics of a good marriage;

show that informants generally shared a similar, common-sense understanding of what marriage

means in practice, regardless of their age or political and religious beliefs. In the preceding

analysis, I chose to focus on the similarities among people’s discourses rather than the

differences because the similarities help to account for why attitudes about homosexuality likely

play a much larger role than attitudes about marriage in shaping discourse about same-sex

marriage. Additionally, these similarities show how the changes with regard to gender, marriage,

and family structures in the last four decades have relaxed the ideological constraints against

allowing same-sex couples to marry. In a society that is more gender egalitarian, more accepting

of diverse family forms, and less hospitable to procreative definition of marriage, the cultural

foundations for supporting same-sex marriage are stronger than they were prior to the 1969 or so.

Marriage Language in the Discourses about Same-Sex Marriage

Just as discourses about same-sex marriage are premised upon people’s beliefs, attitudes,

values, and experiences with homosexuality, so too are the discourses premised upon people’s

cultural definitions of marriage. What distinguishes the same-sex marriage controversy from

other gay rights issues is the idea of marriage, both in its literal definition and its various

connotations. For some participants, the meaning of marriage is central to discourses about

same-sex marriage.
   238  
Q: So you’ve heard a little bit about the same-sex marriage controversy?

R: I have, yeah. I’m opposed to it. I’m opposed to them calling it marriage, and I’m

opposed to legalizing it. I mean, if they want to have a, some kind of a legal agreement

where they can share benefits or whatever, that’s one thing. But I mean, to call it

marriage is not right. Marriage should be between a man and a woman. (Sarah, age 60)

[Opponents] say it violates the sanctity of a marriage, and I don’t see how it does that. I

mean, it does violate the tradition, but it’s two people that love each other. It’s two people

that love each other that want to get married, and if you’re ready to take that

commitment, then go ahead and take the leap. I think it’s stupid to get married, but if

you’re going to get married, you might as well get married to somebody that you love,

and if that somebody happens to be the same sex, then have a ball. (Shelby, age 20)

These two quotes show that the definition of marriage is used by both supporters and opponents

of same-sex marriage to articulate an opinion on the issue. In the first instance, gender

differentiation is considered essential to the meaning of marriage, regardless of one’s feelings

about homosexuality; in the second, the sex composition of the couple is considered irrelevant;

what really defines marriage is love between two people. Thus a person’s definition of marriage

plays an important role in defining discourses about same-sex marriage.

There are two crucial conflicts between definitions of marriage within the discourses

about same-sex marriage that frame the debate in significant ways, and these two conflicts are

closely related but analytically distinct. The first conflict is between religious and civil

definitions of marriage, and the second conflict is between heterosexual and companionate
   239  
definitions. The conflict between religious and civil definitions is an explicitly legal and

ideological conflict, and it frames the debate by defining the boundaries of the controversy over

same-sex marriage. By contrast, the conflict between heterosexual and companionate definitions

is a conflict about what marriage is in practice, and it frames the debate by defining the extent to

which the institution of marriage is inherently gendered.

The conflict between religious and civil definitions is based on a disagreement about

what is at stake in the debate over same-sex marriage and what role religious beliefs should have

in the debate. Advocates of the religious definition of marriage view marriage as a sacred

covenant, sanctioned by God to be between a man and a woman, and that should be protected by

human authorities from being profaned by immoral, homosexual relationships. Given this

definition of marriage, the same-sex marriage debate is about protecting the deep cultural and

religious meanings of marriage in American culture. By contrast, advocates of the civil definition

of marriage deny that religious meanings of marriage have any legitimate role in shaping

government policy about marriage, and that the only relevant issue in the debate about same-sex

marriage is the extension of the rights and benefits that come from civil marriage to all couples.

The conflict between heterosexual and companionate definitions is based on a

disagreement about the extent to which the sex composition of the couple matters for marriage

and the extent to which gendered character traits are associated with each sex. Advocates of the

heterosexual definition of marriage argue that marriage can only be made by an opposite-sex

couple, not necessarily because they believe that procreation must be the result of marriage, but

because it is simply the “natural” way of things. By contrast, advocates of the companionate

definition of marriage believe that the sex composition of the couple does not matter, and that the

only thing that matters is the love and happiness of the individuals involved. Within same-sex
   240  
marriage discourses, advocating a heterosexual definition of marriage does not necessarily mean

that the person holds traditional gender ideologies, but advocating the companionate definition of

marriage contains within it an implicit assumption that whatever characteristics make a good

marriage are not distributed differentially by gender, such that the sex composition of the

partners is irrelevant to the success of the relationship. Advancing a companionate understanding

of marriage with respect to same-sex marriage removes gender as an essential component of

what defines marriage and family.

Because these conflicts occur primarily at different levels, the position that a person takes

on one conflict does not necessarily correspond with the position they take on the other conflict;

for example, some people believe that marriage is a religious institution but also hold a

companionate definition of marriage because they do not believe that the sex composition of the

couple matters. In general, though, religious and heterosexual definitions of marriage are more

likely to be used in oppositional discourses about same-sex marriage, while civil and

companionate definitions of marriage are more likely to be used in supportive discourses about

same-sex marriage. These different definitions, when combined with particular sets of attitudes

about homosexuality, are also associated with middle-ground discourses, as is described below.

Religious and Heterosexual Definitions in Oppositional Discourse

Discourses about same-sex marriage that are unambiguously oppositional invoke

definitions of marriage that are either religious or “traditionally” defined as between one man

and one woman. As the prior quote from Sarah demonstrates, the heterosexual definition of

marriage is not necessarily the same as a religious definition. While Sarah does identify as a

Protestant Christian, she never speaks of marriage in religious terms:

   241  
Q: What does the word “marriage” mean to you?

R: It means a man and a woman committed to each other for the rest of their lives,

hopefully. It’s not something to be taken lightly and entered into. But the idea of it

doesn’t work out, you can always get out. (Sarah, age 60)

As this quote shows, the definition of marriage for Sarah carries no particular religious or sacred

significance, but her gendered definition of marriage makes the idea of same-sex marriage seem

simply wrong to her. While she does believe that homosexual relationships are morally wrong,

her definition of marriage simply leaves no room for debate that marriage is for opposite-sex

couples.

Although religious and heterosexual definitions of marriage are not the same, they

dominate the oppositional discourses about same-sex marriage. Seven of the nine students, and

10 of the 12 parents who used oppositional discourses to talk about same-sex marriage used

language at some point during the interview that suggested that they hold religious and/or

heterosexual definitions of marriage. Those four individuals who never mentioned specifically

that they considered marriage to require one man and one woman, or that they though same-sex

marriage would change the definition of marriage, based their opposition to same-sex marriage

on negative attitudes about homosexuality. It could be argued that they implicitly hold a

heterosexual definition of marriage because of their opposition to homosexuality, but they never

explicitly cited that definition of marriage in their discourse.

Religious definitions of marriage incorporate the heterosexual definition of marriage in

that they require both a man and a woman, but they are distinguished from heterosexual

definitions in that they include a third actor, God:

   242  
Marriage is a commitment that you enter into with a person of the opposite sex, and it is

recognized by the government that you are going to spend the rest of your life together in

a loving, committed relationship. [pause] Instituted by God, blessed by God, and it will

be the hardest thing that you ever work at in your life. (Andrea, age 45)

Marriage is a commitment. It is not a contract. Marriages are, even though they got a

controversy about gay people wanting to be married, well, it’s just not our decision. It is

not man-made. Marriage has never been a man-made decision; it’s been made by God.

Man and woman. That’s how my wife and I feel about it. (Pablo, age 51)

As these quotes illustrate, marriage is more than a simple human institution to those people who

have strong religious worldviews. Marriage is an institution that is created by God and that

carries significance because it carries God’s blessing.

For some Christians, the presence of God in marriage is much deeper than this. They see

marriage as a way for a couple to come together and become Christ-like, to replicate the

wholeness of God on Earth:

This is the way I see it: God is whole. God is a part of man, which is masculine and

strong and a warrior and a fighter. And He’s our protector, He’s our rock, He’s all these

things that man is. But at the same time, He’s love, and He’s gentle, He’s caring and

compassionate—that’s the female side. Apart, with man and woman not together, they’re

only half. You bring them together, and that’s why marriage is so beautiful, because it

shows the wholeness of God. You take a relationship between two men, there’s
   243  
something missing. The same with two women. There’s something missing there, and a

child is not going to learn how to become like Christ. (Dana, age 48)

Thus, marriage requires both a man and a woman, but the union of man and woman also creates

something that is divine. In this quote, the role of gender ideologies in defining marriage as

heterosexual and religious is made clear. The attribution of stereotypical character traits to men

and women is constructed to be ideologically congruent with their Christian beliefs. Whether

one’s definition of marriage is heterosexual or religious, marriage needs both masculine and

feminine gender traits to be complete.

In oppositional discourses, same-sex marriage threatens religious and heterosexual

definitions of marriage. Same-sex marriage is wrong not only because it threatens to normalize

homosexuality and destroy the gender balance within the family, but also because it threatens the

sacred meaning of marriage with a profane sexual relationship.

I think it’d ruin the word marriage and destroy the Biblical, or the, you know, traditional

meaning of it. It’d cheapen it. (Ron, age 20)

If the rhetoric of activists opposed to same-sex marriage sometimes seems extreme, it is at least

partly because of the sacred status of marriage. Asked to speculate about why some people

support same-sex marriage, Vincent says:

They don’t respect the institution of marriage. They don’t respect families. They

somehow, they’re gonna get a benefit of cheapening marriage, whether it’s still making

money off pornography or divorce court or whatever. Typically the people that want to
   244  
cheapen marriages, somehow they’re gonna benefit money-wise or pleasure-wise.

(Vincent, age 48)

If homosexuality is an abomination, human efforts to include it in one of the most holy

institutions can be seen as an attack. In Vincent’s quote above, the effort to allow gay marriage is

disrespectful of the fundamental meaning of marriage and is motivated solely by self-interest.

The notion that marriage is facing a crisis has been advanced by conservative

commentators on the same-sex marriage controversy, and when I asked informants—at the end

of our discussion about marriage—what they thought about the idea of marriage being in a crisis,

people who were familiar with conservative rhetoric recognized the phrase and began speaking

about same-sex marriage:

Q: Some people have said that the institution of marriage is in a crisis. What do you think

about that?

R: Well, the definition is in crisis—one man, one woman. Um, gays are trying to, you

know, get marriage. [sigh] I don’t know. Personally, again, I don’t think that that’s the

way it should be. I don’t think our country should define it as two people… I think it

should be defined as a man and a woman. (Ron, age 20)

Q: Some people have said that the institution of marriage in the U.S is in a crisis these

days. What do you think about that?

R: I heard a really interesting idea that if we allow gay marriages, what’s stopping us

from having polygamy and three-way marriages? Up until, all throughout history—I

shouldn’t say the entirety of history, the majority of history—up until like the last 40, 50
   245  
years, marriage was a man and a woman. Men are attracted to women naturally and vice

versa. (Kyle, age 23)

The idea that marriage is in a crisis evokes both heterosexual and religious definitions of

marriage and the perceived threat to those definitions that same-sex marriage poses.

Religious and Heterosexual Definitions in Ideologically Conflicted Discourse

Not all discourses that invoke religious and heterosexual definitions of marriage are

oppositional. In the ideologically conflicted discourses used predominantly by young religious

conservatives, religious and heterosexual definitions of marriage are common. Nine out of the 14

students and one out of the three parents who used some variant of ideologically conflicted

discourse to talk about same-sex marriage invoked a religious or heterosexual definition of

marriage at some point during the interview. However, defining marriage in religious or

gendered terms does not imply that they have negative attitudes about homosexuality or that they

oppose same-sex marriage. Therefore, ideologically conflicted discourses balance religious or

heterosexual definitions of marriage with positive feelings about gays and lesbians or with the

belief that their personal religious definition of marriage is not a valid basis for a legal definition

of marriage.

One variant of discourse about same-sex marriage that is ideologically conflicted is used

by people who have religious definitions of marriage but who nevertheless have positive

attitudes toward gays and lesbians. For example, Kyle holds a religious definition of marriage

and ultimately opposes same-sex marriage, but he feels conflicted about the issue:
   246  
I haven’t completely formed my opinion on the idea of it, outside of faith. I’m not really

sure, as much as I want to defend the institution of marriage as the Bible defines it, I’m

not really sure at what point we force our religious beliefs on other people through legal

means. And I don’t think it’s going to change the way a non-Christian would view

marriage. If we deny their right to marry, they’re not going to relent their homosexual

tendencies, so it doesn’t really solve very much in that sense. The institution of marriage

is defended, but the heart issue at play is not benefited. So I’m not really sure how I feel

on that. One thing I do know is that’s not the way that God intended it to be. And I

believe that God, being the author of marriage, through Adam and Eve, and through that

union. (Kyle, age 23)

In this passage, Kyle expresses his contradictory feelings by trying to talk them out. He believes

firmly that marriage is a religious institution; however, he has more empathy towards gays and

lesbians than others of his faith. This is perhaps due to his belief that homosexuality is not

something that a person chooses and the fact that he has seen gay Christians struggle to reconcile

their sexual orientation with their religious beliefs. Because of his empathy, he believes strongly

that gays and lesbians deserve more respect and equality in society, but he ultimately does not

believe that marriage is the way to do that.

Like Kyle, Luke also opposes same-sex marriage because of his religious definition of

marriage, but his discourse is ideologically conflicted because he supports equality for gays and

lesbians while opposing the application of the word “marriage” to same-sex relationship:

Here’s where it all comes down to: it’s not whether it’s okay or not, it’s the law. Because

like I said, marriage is a union between a man, a woman, and God. And we shouldn’t
   247  
have God in the Constitution. Everybody should have the same rights of civil union.

‘Cause that’s kind of a problem: that if you’re gay and you have a boyfriend for ten years

and they can’t come visit you in the hospital because you’re not married, that’s kind of a

problem. I don’t think that should be—I mean, that’s not good at all. But I don’t think

you can call it marriage, ‘cause marriage at its core is a religious institution. (Luke, age

19)

In this quote, Luke accepts the religious definition of marriage as being beyond debate, but he

also thinks it is wrong to deny same-sex couples rights like hospital visitation. He imagines no

difference between the love between a gay couple and the love between a straight couple. This

discourse is complex because he does not accept the idea of same-sex marriage, due to his

religious definition of marriage; however, he does support the extension of all rights and benefits

of marriage to same-sex couples.

Other people’s ideologically conflicted discourse is ultimately supportive of same-sex

marriage, despite the religious definition of marriage. For example, Alyssa, a 21 year-old who

converted to Christianity, holds a religious definition of marriage:

In my eyes—which hasn’t been my opinion for my entire life—but it is like a Holy

Covenant. I believe. It’s something, a decision that is guided on like a deeper level than

convenience…. Like it’s a deep emotional connection. And it’s like your best friend.

Someone who you are on the same level spiritually, emotionally, and you grow together.

(Alyssa, age 21)

   248  
Despite her religious definition of marriage, Alyssa supports same-sex marriage because she

expresses positive attitudes towards gays and lesbians. She reconciles her support for same-sex

marriage with the religious definition of marriage by minimizing the importance of procreation:

Q: Do you think that legalizing same-sex marriage would change the meaning of

marriage at all?

R: No. The only difference I would see is procreation. That’d be the only difference. But

of course, there is like artificial insemination and things like that that could go on.

(Alyssa, age 21)

Thus, holding a religious definition of marriage does not preclude a person from supporting

same-sex marriage if they also hold positive attitudes toward gays and lesbians.

Others who hold religious definitions of marriage for themselves, personally, also apply a

civil definition of marriage to the same-sex marriage debate because they do not believe their

own personal religious beliefs can be a legitimate basis for defining marriage legally. For

example, Barrett, a 19 year-old conservative Christian, brought up this conflict in his

understanding of marriage immediately when I asked him about what marriage means to him:

Q: What does the word marriage mean to you?

R: To me, it’s kind of like a union between two fleshes into one kind of a thing.

Ceremony. There’s a lot of symbols and stuff that go along with it: Bible, cross, wedding

cake. A lot of things come to mind when you think of marriage. But the biggest thing is

probably commitment, and [pause] this might be jumping ahead to one of your questions

a little bit: I think morally it should be man and woman. However, I think it’s each

individual’s choice. So I don’t think there should be a law making it illegal, but my
   249  
personal belief is that it should be between a man and a woman. I don’t feel, like if we

got freedom of religion, how can I say you can’t? (Barrett, age 19)

While Barrett’s personal definition of marriage is religious, he qualifies it immediately by

bringing up same-sex marriage and arguing that the legal definition of marriage should not be

religious. Similarly, Carl contrasted his personal objection to gay marriage with his belief that

gays and lesbians should have the right to get married:

I believe you should have gay marriage, because constitutionally, if you look at the

Constitution, that’s what it says. Why should someone be denied a right? I mean, that’s a

Biblical belief, not an American, the Constitution. But scripturally, I do think it’s wrong.

But I think they should have a right to. (Carl, age 19)

Thus, in some ideologically conflicted discourses, holding a religious definition of marriage does

not preclude one from simultaneously believing that civil marriage should be defined differently.

Another variant of ideologically conflicted discourse is that employed by pro-gay

Catholics. Many Catholics who, following their religion, have religious and procreative

definitions of marriage nonetheless feel that it is wrong to exclude gays and lesbians from the

rights and benefits offered by marriage. Here is how Emily, a 19 year-old Catholic, describes her

feelings about same-sex marriage:

I guess I would say my own opinion for it is, um, I don’t think that it’s right for like gays

and lesbians to try to change how the church thinks about it. But if they can change the

state’s way of thinking about it, fine, let them…. You know, the way I view marriage,

from like a Biblical standpoint, it’s between one man and one woman because the
   250  
purpose of marriage really is to propagate the species. I mean, not just for convenience or

whatever. I mean, obviously love is important, but I mean, the real reason for the

institutionalization of marriage is that two become one flesh. (Emily, age 19)

In this quote, Emily describes the basic Catholic ideology about marriage: that it is for the

purpose of procreation and thus cannot be defined in any other way than between a man and a

woman. However, Emily knows several gays and lesbians, has positive attitudes about them, and

believes that they should have all the civil rights that heterosexual couples have—but through a

civil institution, not the religious institution of marriage.

The fact that many people hold religious definitions of marriage but still support the idea

of granting equal rights to same-sex couples suggests that attitudes about homosexuality are

probably a stronger predictor of support for same-sex marriage than their definition of marriage.

If people’s attitudes about homosexuality and their definitions of marriage conflict, the issue of

homosexuality will be more likely to take greater precedence in their discourse about same-sex

marriage.

Civil and Companionate Definitions in Supportive Discourse

In contrast to oppositional discourses, people whose discourses are unambiguously

supportive of same-sex marriage deny the validity of religious and heterosexual definitions of

marriage. This rebuttal is central to supportive discourses on same-sex marriage. Counterposed

to the religious definition of marriage is a civil one, and counterposed to the heterosexual

definition of marriage is the companionate one. In supportive discourses, marriage is

companionate because it is about love between two people, and it is civil because it is about the
   251  
rights and benefits guaranteed by the state; the sex composition of the couple does not matter,

nor do religious beliefs have any legitimate role in dictating public policy.

All twelve of the parents and 35 of the 39 students who used supportive discourses to talk

about same-sex marriage articulated civil and companionate definitions of marriage throughout

the interview. The four students who did not exclusively use civil and companionate definitions

of marriage gave evidence that their idea of marriage is inherently religious, but they supported

same-sex marriage either because they otherwise generally talked about marriage in

companionate ways or because they had positive attitudes and experiences with gays and

lesbians. Thus, at least in the discourses of students, the belief that marriage is a religious

institution is not incompatible with support for same-sex marriage.

When supporters of same-sex marriage argue against the heterosexual definition of

marriage, they advocate a companionate definition of marriage that is centered on notions of love

and commitment. For example, Helen, a 19 year-old student, explained her disagreement with

her mother over same-sex marriage in these terms:

My mom and I, like we talked about it before. Like I’m pro-choice, pro-gay marriage,

and she’s pro-choice, but she doesn’t believe that gays should get married because

marriage is between a man and a woman. But I really don’t think that’s how marriage

should be defined anymore. Just two people who love each other and want to spend the

rest of their lives together. (Helen, age 19)

John, a 47 year-old father, also disagreed that marriage was defined by the sex composition of

the couple and argued that allowing same-sex couples to get married would not negatively affect

the institution of marriage:

   252  
I really don’t see how two people being together of the same sex that love each other, that

are in a respectful, healthy relationship, how is that going to affect your marriage, your

“traditional” marriage between a man and a woman? I really don’t see how that’s going

to bring marriage down. I think it’s way, way above that. And I’m sure, you know,

there’s, there’s going—well it’s two human beings however you look at it—there’s going

to be the same problems, the money, the kids, and it’s not going to change. (John, age 47)

In these quotes, the speakers specifically argue against the heterosexual definition of

marriage and substitute some other characteristic, apart from the sex composition of the couple,

as being essential to marriage. In their view, marriage is defined by deeper characteristics of the

relationship, not something as simple as their sexes. For some supporters of same-sex marriage,

that characteristic is happiness:

It’s their decision, not ours to decide for them. If they love that person so much, they

should get married, because it’s who they want to get married to…. They should be

happy. It doesn’t matter who, man and woman, man-man, woman-woman. All that

matters is if they’re going to be happy. (Chelsea, age 19)

Other supporters of same-sex marriage defined marriage in terms of love:

I think it’s beautiful. Like I said, love is blind. You know, the, I disagree with people who

say same-sex marriage is wrong, because like I said, in saying how the meaning of

marriage is misconstrued by not saying that people of same-sex cannot get married, that’s

just saying that people are only getting married to conceive… I mean, that’s not the

meaning of marriage, I don’t think. (Paula, age 23)

   253  

In this quote, Paula argues that restricting marriage to an opposite-sex couple only makes sense if

procreation is the main purpose of marriage, a rhetorical move that bypasses the deeper

meanings of gender that are built into heterosexual and religious definitions of marriage. Not all

supporters make this argument, but most, including both Paula and Chelsea, reject traditional

gender ideologies insofar as they deny any inherent gender differences between man and woman

that have anything to do with marriage.

Whether marriage is defined by love, happiness, or some other characteristic, in

supportive discourses about same-sex marriage, these characteristics are seen as unaffected by

the sex composition of the couple. Same-sex couples are therefore no different from opposite-sex

couples. In this sense, support for same-sex marriage is ideologically premised upon egalitarian

gender ideologies. To the extent that important personality characteristics or social capacities are

understood as being independent of a person’s sex, there is a weaker ideological basis for

insisting that the sex composition of a couple matters.

In arguing against religious definitions of marriage, supporters of same-sex marriage

advocate a civil definition of marriage. Supporters define the controversy over same-sex

marriage as being about the civil rights and benefits that come from marriage, not the religious

connotations that it has for some people. In response to my first question about same-sex

marriage, the two students quoted below immediately defined the issue as being about legal

rights and freedoms guaranteed by the government and that a person’s religious beliefs have no

place in the discussion:

It’s mostly religious organizations combating the gays, which in my opinion, from a legal

standpoint is extremely contradictory to what this country was founded on, you know.
   254  
We have freedom of religion in this country, and take the legal standpoint that says the

two people of the same sex cannot be married—that’s ridiculous because you have no

real argument to stand on other than God, and God says, “No,” which is contradictory, to,

once again, freedom of religion. (Nate, age 19)

Same-sex marriage hasn’t been legalized yet. You know, it’s like, and there are so many,

you know, just common-sense points why it should be legalized. Marriage is not a

religious institution. It’s in every religion, it’s a legal thing, you know. All this like, the

Republicans are compromising, like civil unions and stuff—it’s like, why can’t they just

get married? (Terrence, age 19)

As Nate and Terrence explain, it is just “common sense” that religious views about marriage

have no place in government. Indeed, the legal case for same-sex marriage is premised upon the

civil definition of marriage. But in practice, the distinction between the civil and religious

aspects of is not so clear-cut. Rhetorically, they attempt to delegitimize religious arguments

against same-sex marriage by defining marriage as a civil institution that should be free from

religious influences. In supportive discourses, such a claim is made in order to frame the debate

about same-sex marriage as a legal issue, such that the religious implications are irrelevant.

Because marriage is defined as a civil institution in supportive discourses about same-sex

marriage, what is largely at stake in the fight for same-sex marriage are the rights and benefits

that are available to opposite-sex couples:

I think they should be able to have the same rights that we do. I mean basically when you

get married, it just changes your insurance stuff and your money stuff and you can make
   255  
sure that person is taken care of. And why shouldn’t they be allowed to have that? Like,

you have no right to comment on their religion or their, you know—that’s a personal

thing. That’s nothing that should be something [that is] governed. (Sandra, age 29)

In this quote, Sandra focuses on the rights and benefits of marriage and minimizes the role that

the deeper cultural meanings of marriage can legitimately play in the debate about same-sex

marriage. As described in the previous chapter, equal rights arguments play a prominent role in

the discourse of supporters about same-sex marriage, and the strength of this argument depends

upon a civil, not religious, definition of marriage. Because of the separation of church and state,

the legal controversies over same-sex marriage have focused on marriage as a civil institution

and whether or not the state can deny same-sex couples the rights and benefits that the civil

institution of marriage provides.

There are two radical views about the civil definition of marriage that bear some attention

because they carry the concept of marriage as a civil institution to its farthest logical ends. One

view is that marriage is merely a piece of paper, lacking any real meaning or significance. While

most people in my sample had fairly positive attitudes towards the idea of marriage, a few did

not. For people who either held negative attitudes about marriage or who were relatively

apathetic about marriage, the bitter controversy over same-sex marriage seemed unwarranted.

For example, Natalie, who had been married for over 20 years, said that she didn’t like to use the

word marriage and preferred to call her husband her “associate,” because she thought of her

marriage as basically a financial arrangement. When I asked her why she thought gays and

lesbians wanted the right to marry, she recognized that they wanted the rights, but otherwise

denied that marriage itself meant anything special:

   256  
I feel that, you know, they should probably have, you know, the same legal rights. That

has to come through marriage, so I can understand why they would want that. But again,

it’s the same thing, even with opposite sex, I don’t think marriage is the end-all. You

know, just live together, what the heck! (Natalie, age 60)

Over the course of the interview, she grew a bit frustrated at my repeated probes on the issue,

and at one point she exclaimed, “The whole institution of marriage anyway, the paper doesn’t do

anything. I think it’s all a bunch of idiocy.” People without positive attitudes about marriage

generally supported same-sex marriage because they thought gays and lesbians should have the

same rights as everyone else, but they did not assign any real importance to the concept of

marriage.

The other radical view of marriage as a civil institution was advanced by atheists and

agnostics who believed that all couples should have civil unions and that the word marriage

should be left for people who wanted to go through an optional religious ceremony. In other

countries, this is not a radical view, but in the United States, it is radical because it suggests a

redefinition of our existing cultural vocabulary: the meanings of “marriage,” “husband,” and

“wife” would have to be replaced with the meanings “civil union” and “partner.” For example,

Edward, a 20 year-old atheist, said that he is in favor of same-sex marriage but that the

terminology should be changed so that marriage would keep its religious connotation and that

everyone would have civil unions:

I support and I agree with the idea… [of] creating an institution where a same-sex couple

and a heterosexual couple have exactly equal legal standings… The word marriage is a

religious thing, and so when they say they are in favor of same-sex civil unions, they’re
   257  
in favor of heterosexual civil unions as well… The government shouldn’t be allowed to

mess with the word marriage at all, same-sex or heterosexual. And so I think a language

shift would be nice there… (Edward, age 20)

Similarly, the idea of separating the religious idea of marriage from the legal aspects of

partnerships was appealing to Andy, an agnostic:

I think that it should be legal for everywhere, for same-sex marriages. However, marriage

has the, marriage has the whole religious idea behind it, kind of, so they need to alter it in

a way that it just needs to be, I think it should be domestic partnerships for everybody.

And government-wise, you still get married, just the government considers it a domestic

partnership. (Andy, age 20)

While the suggestion that the religious and civil meanings of marriage should be separated is

logical, the meanings of marriage in American society are not just religious, so redefining

marriage in both cultural and legal terms to strip them of their non-religious connotations would

be a radical solution.

For people who advocate civil and companionate definitions of marriage, legalizing

same-sex marriage would not change the meaning of marriage, as many opponents to same-sex

marriage insisted:

I don’t think so. No. I think the meaning of marriage is the meaning of marriage, in terms

of commitment. You know, that type of thing is the same whether it’s between a man and

a woman or if it’s between a same-sex couple. (Frances, age 45)

   258  
At most, supporters said that the text would change, but that the deeper meaning would not:

On the strictest technical sense, yes, because it’s defined as a man and a woman right

now, so yes. In terms of the meaning of marriage, like how it affects people’s lives and

stuff, um, no, actually I don’t think it would change. (Dean, age 25)

In general, people whose discourses were supportive of same-sex marriage denied that legalizing

same-sex marriage would have any negative consequences, beyond making conservatives angry.

Not only did they deny that the meaning of marriage would be changed, they argued that

legalizing same-sex marriage might even improve the institution of marriage:

Personally, I think that with all the troubles and, you know, problems we’re having with

marriage, I think people who love each other and want to be committed and be together

should be allowed to be together. (Carolina, age 20)

I think it could definitely change the meaning of it. It could almost make it seem like,

almost bring back that kind of sacred thing to marriage, you know. Like these people are

marrying each other because they love each other… and not because they can…. So I

think it would show people that they’re getting married because they truly love each

other, and that’s what marriage is about, because they want to spend the rest of their life

together…. So it kind of gives that meaning back somewhat to marriage. (Claudia, age

22)

Carolina and Claudia both argued that allowing same-sex marriage could improve marriage

because it would define it purely in terms of love. Gays and lesbians would get married out of
   259  
love, not because of societal expectations or convenience. Here, the contrast between

companionate and heterosexual definitions is constructed in stark terms. Far from causing a

crisis in marriage, some supporters described same-sex marriage as a solution to the problems,

such as a high divorce rate, that have emerged within heterosexual marriage. Rather than ruining

the institution of marriage, many people who advanced civil or companionate definitions of

marriage thought that legalizing same-sex marriage would have positive effects on the

institution.

Civil and Companionate Definitions in Libertarian Pragmatic Discourse

While libertarian pragmatic discourses about same-sex marriage are not defined by the

definition of marriage, seven of the eight participants in my sample who used this discourse also

advanced a companionate, civil definition of marriage. Only one parent who used libertarian

pragmatic discourse said that “marriage is more a spiritual term than it is a legal term” (William,

age 53), but even he generally talked about marriage in civil and companionate ways.

In libertarian pragmatic discourse, the imagined effects of legalizing same-sex marriage

are crucial because, by definition, the degree of support or opposition to same-sex marriage

depends on the extent to which the imagined effects would be negative. Most people who used

libertarian pragmatic discourse did not imagine there to be any strong negative effects from

legalizing same-sex marriage, and only one individual thought that it would change the meaning

of marriage. Harvey, who used libertarian pragmatic discourse to talk about same-sex marriage

described this worry in terms of diluting the meaning of marriage:

Q: Do you think if they legalized same-sex marriage, do you think it would affect people

in the sense that it would sort of change the meaning of marriage?

   260  
R: I think it would make it a little more diluted. Because then if you can get away with

that, I don’t know if people are going to be like, “I need to marry animals.” Because they

had an episode about that on “The Simpsons.” It was kind of funny.

Q: I saw that, yeah.

R: Yeah, and then all of a sudden, well they were like, “well I’ll marry anybody,” you

know. Like, you know, it was like, where does it end? (Harvey, age 23)

Here, Harvey brings up the common slippery-slope argument as a potential negative effect on the

meaning of marriage. But because Harvey articulates a civil and companionate definition of

marriage, it is possible that this argument is motivated more by a negative judgment of

homosexuality than by a felt threat to his understanding of marriage.

Most informants who used libertarian pragmatic discourse imagined there to be mostly

positive effects of allowing same-sex couples to marry. Even with respect to the effects that

legalizing same-sex marriage would have on children, they argued that effects would be positive

because two parents could provide a loving, stable environment for raising children:

If you have a husband and a wife who are raising two children, and these people are

doing a poor job of it and hate each other, or it’s not a good environment for children to

be raised… I’d rather see children in a same-sex relationship, whether it’s a marriage or

not a marriage, where they’re actually given the love and understanding and, you know,

nurturing, reading, writing skills, you know, whatever it takes for, part of raising a

family. (Tom, age 47)

   261  
[Legalizing same-sex marriage] would make the kids like, be more comfortable with the

situation, and then they’ll grow up—it kind of makes it more normal…. If you’re in a

same-sex, you know, family, you know, the child could still grow up and still be attracted

to the opposite sex, but at least the parents show you that, “No matter what happens, I’m

still going to love you and care about you, no matter what. Nothing’s going to change

that.” (Jerome, age 22)

In general, most informants who used libertarian pragmatic discourse did not think that same-sex

parents would raise children to be gay, and they did not imagine that the families would be

significantly different from a stereotypical nuclear family.

From what I’ve read about, I haven’t seen anything negative as a consequence. That may

need long-term studies to see, but so far, I haven’t heard anything. So far it’s saying the

opposite, saying that children could be raised in heterosexual, you know, kids are doing

fine. (Maria, age 45)

To the extent that allowing same-sex couples to marry would have positive outcomes, informants

who used libertarian pragmatic discourse to talk about same-sex marriage would have no reason

to oppose same-sex marriage.

The most common negative effects that informants who used libertarian pragmatic

discourses imagined had to do with fears about children being teased by their peers if they were

raised by same-sex parents or with the antagonism and anger that legalizing same-sex marriage

would cause among opponents. But because these negative effects would be due to the reactions

of other people, rather than to any inherent consequence of legalizing same-sex marriage, the
   262  
people who used this discourse did not articulate a set of strong negative consequences that

would justify opposing same-sex marriage. Sometimes, people imagined mixed effects:

I would think that [the children] would know what it would be like to be loved all the

time, you know. But then society would start picking on them, “Oh, you come from this

family, that family,” you know. So I don’t know. Then that kid would have to try to learn

how to help themselves. (Jillian, age 49)

It was easy for informants to imagine both positive and negative consequences of legalizing

same-sex marriage, so the lack of a simple, obvious answer about its consequences made the

libertarian pragmatic discourse viable.

The possibility that people who used libertarian pragmatic discourses could be persuaded

to oppose same-sex marriage because of its imagined effects is illustrated by the fact that some

informants did appear to oppose same-sex marriage because of them. For example, Tracey, a

liberal Catholic who used ideologically conflicted discourse, was unsure how she felt about the

issue because of imagined unintended consequences:

I guess I’m, I don’t know if, in one way, I feel like it doesn’t matter to me if guys, girls,

whatever. But sometimes you do have to draw the line as far as finance or something, as

to—it seems like it would really throw a loophole in everything for fraud. More for fraud

than anything. As to trying to say, “Oh, we’re together.” You know, say two people who

are friends who live together who want benefits. I don’t know how they could regulate

that…. I don’t know, it just seems to be opening up a whole big box of worms that I don’t

know if I’m ready for. (Tracey, age 53)

   263  
It is clear from this quote that Tracey worries about unforeseen problems that might arise

regarding the institution of marriage if the definition of marriage was broadened to allow any

two people, regardless of sex, to marry. For informants like Tracey and Harvey, the imagined

negative consequences of legalizing same-sex marriage make it likely that they would ultimately

oppose same-sex marriage in a political campaign. After all, their willingness to refrain from

moral judgment on the issue is premised upon the possibility that there would be no negative

effects for others. So, if legalizing same-sex marriage were to have negative consequences, they

would be more likely to oppose it.

Civil Unions

One potential solution to the controversy over same-sex marriage is to recognize same-

sex relationships in the form of civil unions or domestic partnerships, an alternate legal status

that would give same-sex couples access to the rights and benefits of marriage without calling it

“marriage.” Such a solution is appealing to both sides insofar as it would grant the same rights

and benefits to same-sex couples and it would preserve the heterosexual and religious definitions

of marriage. Granting civil union or domestic partnership status to same-sex couples

accommodates (or neutralizes, depending upon one’s perspective) a significant objection of

opponents to same-sex marriage.

As described in Chapters 1 and 3, there is evidence from public opinion data that there is

merit to such a compromise. A clear majority of Americans consistently support civil unions, but

a clear majority of Americans consistently oppose same-sex marriages. It is possible that 10-15%

of the population effectively oppose same-sex marriage because of the definition of marriage,

and that their opposition would switch to support if the issue were about same-sex civil unions.
   264  
This chapter has provided further qualitative evidence that many people who advance

heterosexual or religious definitions of marriage nonetheless hold pro-gay attitudes and support

the idea of granting equal rights to gay and lesbian couples. Removing the definition of marriage

from contention would simplify the terms of debate to some extent.

In the United States, however, granting equal rights to same-sex couples would not be

achieved with the civil union or domestic partnership status because of the variety of federal

rights and benefits that come specifically from marriage. All current and previous civil union and

domestic partnership laws have provided only some of the rights and benefits available to

opposite-sex couples through marriage. Civil unions and domestic partnerships thus effectively

grant second-class status to same-sex couples, compared with marriage.

In my interviews, I specifically included a question about civil unions in order to examine

how people talk about civil unions in relation to their discourses about same-sex marriage. By

comparing how people talk about civil unions with how people talk about same-sex marriage, we

can obtain additional information regarding the viability of the civil union compromise in public

opinion. For the purposes of the interviews, I bracketed the problem of the unequal benefits of

marriages and civil unions by posing the hypothetical scenario that same-sex couples would be

granted the same rights and benefits as marriage; thus, the only difference would be that the title

of marriage would be used for opposite-sex couples, while the title of civil unions would be used

for same-sex couples.

When the civil union question was posed in this way, the overwhelming response from

informants was that the title made no difference. Most informants, young and old, liberal and

conservative, dismissed the question as an evasion of the real issue:

   265  
I think if it walks like a duck and acts like a duck, it’s a duck. I’m sorry, but that’s the

way I feel about it. To me, you know, why rename something that doesn’t need to be

renamed? (Stan, age 59)

Supporters of same-sex marriage wondered why gay and lesbians couldn’t use the same word if

everything was otherwise the same:

If that’s what the government wanted to do, that’s fine, but if you’re going to give that

person same responses, the same everything, why are you not going to let that person get

married? What’s going to be the difference?... If you’re going to let them have all the

benefits, let them have that benefit. That’s my opinion. (Richelle, age 47)

The civil union compromise would be unsatisfying to supporters of same-sex marriage because

even if it grants the same rights and benefits, in principle, it establishes a regime of “separate but

equal” and denies gays and lesbians the cultural recognition that the title provides:

I would be opposed to it…. They tried that before that, with the African Americans and

the white people and saying that they’re separate but equal, you know. And obviously

that didn’t work out, you know; that was a big disaster…. I mean, if you’re going to give

them all the same benefits, why not just call it marriage, you know? I don’t see any

difference. (Kevin, age 20)

Hmm, I think, I think it’s okay, but it would be kind of denying them the right to, I mean,

the right just to say that they’re married. I think that’s part of everything. And I mean, if

they’re going to do that, then why not just go all the way with it? (Tiffany, age 23)
   266  

Thus, for supporters of same-sex marriage, the title of “marriage” is culturally and symbolically

important, and the civil union compromise would be unsatisfying precisely because it denies

gays and lesbians full equality. Equality is not just measured in terms of rights and benefits, but

in terms of respect and legitimacy.

Likewise, opponents of same-sex marriage insisted that the civil union compromise

would still grant legitimacy to homosexual relationships and opposed it on these grounds:

I think—cause I’m still against same-sex, even relationships—that I would still be against

that. Cause it’s still giving the message, “This is okay.” (Renee, age 19)

I wouldn’t be for it. I wouldn’t want to see it happen. I think, again, it would cheapen

marriage still, even though they don’t call it marriage…. You’re still together, you still

have all the same benefits, you still, these people are still going to call it marriage even if

it’s called civil union. (Ron, age 20)

Some opponents perceived the civil union compromise to be an underhanded way to give gays

and lesbians the right to marry:

I think it’s just a nice way of saying we’re giving you the rights to get married. It’s the

same thing…. A lot of people… want all the benefits of it but they don’t want to have to

follow the rules. No, you don’t get all the blessings of God if you don’t have the

relationship with God. That’s the way people want with marriage. “I want the privileges,

but I don’t want to follow any of the rules.” But I’m sorry, they go hand in hand. (Dana,

age 48)
   267  

For conservatives like Dana, the political logic of the civil union compromise appears either as

an effort to bend the rules for one’s own self-interest or as the first step on a slippery-slope that

would eventually grant gays and lesbians the right to marry. Thus, both opponents and supporters

of same-sex marriage disliked the idea of civil unions, but for different reasons.

Many opponents to same-sex marriage mistakenly argued that gays and lesbians already

have civil unions and equal rights. This misperception is likely due to the fact that some gays and

lesbians can register as domestic partners and that gays and lesbians have more civil rights than

they did 20 years ago. However, the overemphasis of the rights available to gays and lesbians

through domestic partnerships and civil unions is also consistent with the conservative religious

narrative of moral decline: that the increasing power and status of gays and lesbians in society

are part of the nation’s rejection of God.

Individuals who thought that the civil union compromise would be a good idea tended to

be people who emphasized the political logic of it, rather than advance any moral or normative

argument for civil unions:

See, I think that’s where they should go with it. I think that’s what they should aim for…

The ones that are bucking more is the, they’re looking at the spiritual marriage deal, and

they’re looking at the narrow definition and things like that. You give it more of a

legalese definition to begin with, you know, maybe the other definition might change too,

later on, but at least at the very beginning, do it the legalese way. (William, age 53)
   268  
Another common response from informants was that they would be okay with such a

compromise solution because they are personally not affected by the issue, but they thought that

the opinion of gays and lesbians mattered more:

I would not have a problem with that. That might actually smooth over a lot of people,

but I would first want to know how the homosexual community feels about that, because

like I said, they need to be respected. Personally, I think that would be fine, but I’m not a

big gotta-be-married person, you know. But I think if it works out better, smoothly, for

the society to accept it, and it’s not a problem for the people involved in it, then I think it

would be a great idea. (Bonnie, age 46)

In comments like Bonnie’s, there is an implicit acknowledgment that gays and lesbians might be

opposed to the idea, even though it would be okay with her. Because she does not have a stake in

the outcome of the controversy surrounding same-sex marriage, she presents her opinion about

civil unions as being dependent upon whether or not gays and lesbians would support it.

Only a couple of informants who opposed same-sex marriage appeared to lessen their

opposition when the civil union compromise was proposed. But even then, they did not say they

would support it:

I guess, I guess I could accept that, I think. If you’re going, I guess I could probably

accept it if it’s not called a marriage. I really do. I think I could accept that. I wouldn’t

approve of it, but I think I could accept it. (Debra, age 57)
   269  
It depends on all of the individual factors involved, but I think I would be more in favor

of that than recognizing a gay union as marriage. But again, I’d have to kind of cross that

bridge when I came to it. (Sarah, age 60)

Thus, taken literally, the idea of creating a separate legal status for same-sex couples while

preserving the heterosexual definition of marriage has very little traction in discourses about

same-sex marriage. At best, civil unions resonate with informants as a political strategy rather

than as a logically or philosophically sound institution in its own right.

However, beyond the political logic of the civil union compromise, there is evidence in

these interviews that civil unions do carry additional cultural significance. Talk about civil

unions conveys the idea that gays and lesbians deserve equal rights, even though people might

think of marriage as an institution between one man and one woman only. Civil unions are not

literally a viable alternative, but the expression of a paradox in the discourse: that many people

believe that gays and lesbians deserve equal rights, but they don’t think that those rights should

include marriage.

While in ideological terms, this is a blatant contradiction, in sociological terms it is not.

Many people think of gays and lesbians as ordinary people, as human beings just like any other,

whose only difference happens to be their feelings of sexual attraction; as a result, they think that

gays and lesbians deserve the rights and benefits that come from marriage, like insurance and

hospital visitation. At the same time, they think of marriage as something that requires two

people of the opposite sex; thus, the right to marry is something entirely different. The paradox

in the discourse surrounding the idea of civil unions comes from the fact that certain rights and
   270  
benefits come from marriage, even though they may have little to do with the meanings of

marriage that are at stake.

When viewed from this perspective, the cultural significance of the civil union question

becomes clear: civil unions represent the commonly-held combination of beliefs that gays and

lesbians should have the rights and benefits that come from marriage, but not the right to

marriage itself. The response from Haley, a 19 year-old conservative Christian, to my question

about civil unions demonstrates this paradox. I present the lengthy exchange below because the

difficulty that she has in explaining what she thinks illustrates how problematic this combination

of beliefs can be, given the fact that so many different rights and benefits are tied to marriage:

R: I definitely think that even if they are in a gay relationship, they should receive the

same rights. Like they’re not different people. Like it says that all sins are equal….

Q: So the fact that they’re sinning in this particular way doesn’t mean that they shouldn’t

be given those same rights?

R: Yeah…. I don’t know. I think they should be given their rights no matter what. But I

don’t know, if you give them like rights to, I don’t know. I don’t know.

Q: That’s okay. If you want to talk it out, go for it.

R: ‘Cause in marriage, you have like certain like political rights, like housing rights and

things.

Q: Yeah. And like if your partner’s sick, then you have visitation rights at the hospital. Is

that the kind of thing you’re talking about?

R: Yeah. Certain things like that.

Q: Taxes and...

R: I guess I don’t have a problem with those kinds of things. (Haley, age 19)
   271  

Haley is opposed to the idea of same-sex marriage, because she understands marriage to be a

relationship between a man, a woman, and God. However, she also thinks of gays and lesbians as

being no different than anyone else, so she thinks gays and lesbians should have all the rights and

benefits that come from marriage, but not the right to get married itself.

This pattern of discourse was often produced by informants in response not to an explicit

question about whether or not gays and lesbians should have the same rights as heterosexuals.

Many people who are opposed to same-sex marriage said that, other than marriage, they should

have the same rights as heterosexuals because they are human beings just like anyone else:

Q: Do you think that gays and lesbians should have the same rights as heterosexuals?

R: Well, I mean, we shouldn’t look at them any differently, except their, the union, and

whatever you want to call it. Otherwise, you shouldn’t look at them differently. You

should be looking at the person. So yeah, I think they should have the same rights. If it

comes to marriage or whatever, then that’s, I guess I’m not sure that they have that right,

but everything else, they should not be any different than the rest of us. (Bernice, age 53)

Bernice’s insistence that gays and lesbians should have the same rights as heterosexuals, with the

exception of the right to marry, is not an ideological contradiction, but a reflection of the belief

that the right to marriage, due to the particular cultural meanings that it has for people, is

different from other rights.

The cultural significance of the civil union compromise, then, is that it represents this

particular configuration of beliefs and attitudes, given the particular legal structure for providing

rights and benefits to families in the United States. If interpreted literally, the civil union
   272  
compromise fails because all the cultural meanings that are embedded in the term “marriage”

make the compromise appear hollow in all logical terms except for political expediency.

However, if interpreted as I have argued here, an alternate solution suggests itself: that the rights

and benefits that come from marriage be de-coupled from marriage. This solution has been

advocated by Nancy Polikoff (2008), who shows how the controversy surrounding same-sex

marriage would be undermined if all families had access to the same rights and benefits,

regardless of the marital status of the adults. She argues that such an approach “values all

families,” no matter their composition, and it strengthens the meanings of marriage, whether

religious or companionate, because it removes the need for anyone to get married solely for

material reasons.

Conclusion

In this chapter, I described how informants use different cultural definitions of marriage

in order to talk about same-sex marriage. While opponents of same-sex marriage generally

advance religious and heterosexual definitions of marriage, supporters of same-sex marriage

advance civil and companionate definitions of marriage. In addition, I described how religious

and heterosexual definitions of marriage are sometimes used in ideologically conflicted

discourse, and how the imagined consequences of legalizing same-sex marriage shapes

libertarian pragmatic discourse. In describing these patterns of talk, I showed that disagreements

over the meaning of marriage shape the debate about same-sex marriage but that this

disagreement is less important than people’s talk about homosexuality in shaping attitudes about

same-sex marriage.
   273  
I also showed in this chapter that even people who advance religious and heterosexual

definitions of marriage talk about marriage in individualistic and pragmatic terms. I argue that

there is a cultural consensus about what marriage is in practice that resembles the companionate

definition of marriage. The cultural consensus is based on positive attitudes about marriage,

pragmatic and centrist discourses about divorce, premarital sex, and cohabitation, and

individualistic and pragmatic views about the characteristics that are essential for a good

marriage. Even in terms of childrearing and the gendered division of labor, informants tended to

use pragmatic language to describe how the fulfillment of the couple and the quality of their

relationship is what matters most. Paradoxically, procreation is not viewed as essential for

marriage, even though sex is. This shared common-sense understanding of what marriage means

in practice strengthens the ideological case for supporting same-sex marriage because gender and

procreation are relatively unimportant in the discourse about what is essential for a good

marriage. The cultural foundations of support for same-sex marriage are therefore likely to be

stronger than they were prior to the period of rapid change in the meanings and structures of

gender, marriage, family, and sexuality.

Lastly, I showed that the idea of civil unions, as a compromise solution to the controversy

over same-sex marriage, has a weak cultural and ideological basis in the discourse about same-

sex marriage because both supporters and opponents think much more is at stake in the definition

of marriage than just legal rights or benefits. However, the idea of civil unions represents a

paradox in the debate: that many people simultaneously think of gays and lesbians as normal

human beings, just like anyone else, who deserve equal rights, but that marriage itself is not one

of those rights that can be given to same-sex couples. As such, the idea of civil unions captures
   274  
the essence of an important middle-ground discourse that is otherwise not easy to express in the

terms of the same-sex marriage debate.

This chapter thus shows how the debate about same-sex marriage is different from other

gay rights issues. The different cultural meanings of marriage that people use to construct

opinions interact with their understandings of homosexuality in ways that further complicate the

various discourses about same-sex marriage. Not only does this make the terms of the debate less

likely to polarize, but it actually calls further attention to the middle-ground discourses in the

debate. The middle-ground discourses do complicate the debate about same-sex marriage, but

they also suggest a way out of the controversy. We do not need to accept the binary, zero-sum

assumption that one side must win at the expense of the other; nor do we need to find some

middle point in between supporters and opponents that will appease no one. Ideologically, the

controversy over same-sex marriage can be undermined by decoupling legal rights and benefits

from the institution of marriage, thus making them available to all families, no matter their

marital status.
   275  
Chapter 6: Cohorts and Social Generational Change

Introduction

In the last three chapters, I have described the ways in which cohort location, political

and religious ideologies, and different sets of understandings of homosexuality and marriage

shape people’s attitudes and discourses about same-sex marriage. Using both quantitative and

qualitative methods, I have shown that the controversy about same-sex marriage is rooted in the

variation in people’s understandings of homosexuality and marriage, which is in turn rooted in

the variation in people’s cohort locations and political and religious ideologies. Because of the

ways in which cohort location cross-cuts political and religious ideologies, there is wide

variation in people’s attitudes, beliefs, values, and experiences with homosexuality, and people’s

definitions of marriage are multi-faceted. As a result, discourses about same-sex marriage fail to

polarize into discourses of support and opposition only; there are also a variety of distinct

middle-ground discourses.

Supportive discourses are used predominantly by liberal students who articulate civil and

companionate definitions of marriage, who have positive attitudes and life experiences about

homosexuality, and who deny that homosexuality is morally wrong. By contrast, oppositional

discourses are used predominantly by older religious conservatives who articulate religious and

heterosexual definitions of marriage, who have negative attitudes about homosexuality, and who

argue that homosexuality is immoral or sinful. These patterns of support and opposition to same-

sex marriage are unsurprising and are supported by both quantitative regression analysis and

qualitative discourse analysis.

   276  
Middle ground discourses, however, tend to be used by people whose cohort locations

appear to impel them to take a position on same-sex marriage that is at odds with their political

and religious ideologies. Thus, older liberals are the most likely people to use libertarian

pragmatic discourses because their political ideologies would lead them to support equal rights

and tolerance for minority groups; but they also are more likely than younger liberals to have

negative associations with homosexuality, perhaps because they grew up during a period in

which homosexuality was culturally constructed as a mental illness, or at best, a deviant lifestyle.

Similarly, younger religious conservatives are the most likely people to use ideologically

conflicted discourses because their religious beliefs teach that homosexuality is an immoral

behavior that should be stopped; but they also are more likely than older religious conservatives

to have positive attitudes toward gays and lesbians, perhaps because they grew up during a

period in which homosexuality was increasingly constructed as normal. It appears that people’s

understandings of homosexuality are so much more important than their understandings of

marriage in shaping these discourses because there is some degree of cultural consensus over

what marriage means in practice that is shared by people on all sides of this debate.

The cohort variable is so crucial in shaping these discourses because of the dramatic

societal changes with regard to homosexuality, gender, marriage, and family that have occurred

in the United States since roughly 1969. The younger cohort in my sample reached adolescence

and young adulthood after 1990, during the period in which gay rights began to supplant gay

liberation as the dominant political and discursive strategy of the LGBTQ movement. Gays and

lesbians began to be portrayed in mass media as “normal” and gays and lesbians were

increasingly open about their sexual orientations in their personal lives. At the same time,

egalitarian gender ideologies and expectations had become increasingly common, the
   277  
companionate meaning of marriage had become increasingly dominant, and the hegemonic ideal

of the nuclear family had weakened in the face of the diversification of family forms.

By contrast, the older cohort in my sample reached adolescence and young adulthood

prior to 1980. They therefore came of age in a time when homosexuality was culturally

constructed as a mental illness or deviant social behavior, and they have lived through the

women’s liberation movement, the move away from traditional gender ideologies, the rise in the

divorce rate, and the liberalization of attitudes regarding premarital sex and homosexuality. In

short, the parents in my sample lived through the changes to American society brought about by

the women’s and gay liberation movements, while their children did not. Given the theoretical

assumptions about how cohort effects are rooted in differences in individual attitudes and

orientations, one can say that the younger cohort is more likely to take for granted the current

constructions of gender, homosexuality, marriage, and family, while the older cohort grew up

during a period in which the women’s and gay liberation movements were contesting the older

constructions.

In this chapter, I focus explicitly on the social generational interpretation of these patterns

by going beyond the simple cohort comparisons used in the previous chapters. The cohort-related

patterns of discourse documented in the previous chapters could plausibly be accounted for by

some other variable, such as educational attainment; so in this chapter, I examine the empirical

evidence in support of a social generational interpretation. First, I analyze the interviews for

evidence that the historical period of rapid social change has influenced the discourses of each

cohort in different ways. I show that parents with negative attitudes about homosexuality

frequently talk about how the time that they grew up shaped their views, while parents with

positive attitudes about homosexuality use narratives of attitude change to talk about why they
   278  
are now more tolerant of gays and lesbians. Students, by contrast, used narratives of attitude

change that were really more about maturation than attitude change, and they are instead more

likely to talk about experiences with gays and lesbians during high school and college in ways

that show tolerance and acceptance to be the norm.

Second, I conduct a stronger test of cohort differences by comparing the discourses of

matched pairs of parents and students. By focusing on the students and parents who generally

agree with each other with respect to other issues relating to marriage, sexuality, religion, and

politics, we can better control for generational and ideological effects that also shape the

discourses. In other words, by making comparisons simultaneously across and within cohorts, we

increase the analytic precision for showing how social generational processes can account for

cohort differences in discourses. I show that, controlling for generational and ideological effects,

the cohort-related variation in discourses documented in the previous chapters remains.

Third, I offer an interpretation of these cohort-related patterns of discourse that focuses

on the cultural and social psychological processes that are assumed to account for cohort effects.

I argue that these analyses show evidence of how social generational processes shape discourses

about same-sex marriage. Specifically, the intersection of a person’s cohort location and their

social location within that cohort shapes the structure of their cultural repertoire and the ways

that they use elements of that repertoire in order to talk about same-sex marriage. Of particular

importance is the tacit definition of homosexuality that each individual brings to bear on the

question of same-sex marriage. Conceiving of homosexuality as collective identity reflects

membership in the younger cohort and liberal political ideologies, whereas conceiving of

homosexuality as social behavior reflects membership in the older cohort and conservative
   279  
religious ideologies. Middle-ground discourses represent creative ways that agents combine and

reconcile cross-cutting influences of specific elements of their cultural repertoires.

Finally, I show that this analysis of social generational processes can be taken one step

further. In this case, tacit definitions of homosexuality are shaped by the extent of a person’s

contact with and acceptance of the cultural construction of homosexuality that is dominant

during a given period. Not only do cohort and religious and political ideologies shape a person’s

tacit definition of homosexuality, but so also do the nature of personal contact with gays and

lesbians in real life, the nature of exposure to gays and lesbians in mass media, and the

composition of one’s social networks. These three factors are all critical in shaping people’s

cultural repertoires regarding homosexuality, and they help account for aberrant cases: when

cohort membership or political and religious ideologies seem to have no effect at all on some

people’s discourses. I complicate the analysis of social generational change in this way because

what is crucial to understand about the social generation concept is that it signifies variation in

people’s lifeworlds, where the intersection of cohort and social location is in reality extremely

complex and multidimensional. The structures and use of people’s cultural repertoires bear the

marks of the lifeworld that is different from other people in different cohorts and in different

social locations.

Methods

The data presented in this chapter are comparisons of interviews from the 32 matched

pairs of students and parents in my sample. One of the pairs of students and parents is actually a

trio, because the student who contacted me for an interview had a sibling who was attending NIU
   280  
and who agreed to an interview. This analysis is therefore based on 65 of the 97 interviews that I

conducted.

By comparing the interviews of parents and children, I am not attempting to study

generational influences, as it is technically defined between parents and children. Rather, I am

looking for evidence of cohort-related influences on discourse while, to the extent possible,

controlling for the generational influences of parents on children and vice-versa. Using the

terminology I proposed in Chapter 1, this comparison is intended to analyze evidence of social

generational change by comparing the discourses of two generations.

In order to gauge the similarities and differences between each parent and student, I

identified a variety of different attitudes, beliefs, values, and experiences, as they were

discursively constructed in the interview, relating to religion, politics, marriage, gender,

homosexuality, and same-sex marriage. I then compared the student and parent to each other in

these terms in order to see how differences in the discourses about same-sex marriage are related

to other differences in their cultural repertoires. The comparisons regarding homosexuality and

same-sex marriage are based on the discourses identified in the previous two chapters and the

cognitive beliefs, affective attitudes, moral values, and life experiences identified in Chapter 4.

The comparison regarding marriage and gender are based on the meanings of marriage identified

in Chapter 5, plus the attitudes that informants expressed regarding cohabitation, premarital sex,

divorce, and the gendered division of labor within a family. Additionally, I compare the political

and religious beliefs of the parent and child based on their self-identifications and the ways in

which political and religious beliefs are manifest in their discourses.

Results: Differential Effects of Period on Each Cohort

   281  
If the cohort-related variation in discourses about same-sex marriage are to be plausibly

attributed to social generational processes, we should see evidence in the discourses that is

consistent with the theoretical argument advanced above: that the historical period in the U.S.

after 1969 is characterized by significant changes in the cultural construction of gender,

sexuality, marriage, and family, and that members of each cohort experienced that period of

change very differently because of when they came of age. Indeed, the interviews I conducted

show that informants perceive society to have changed significantly since the 1960s, especially

with respect to homosexuality. Moreover, the ways that parents and students talk about

homosexuality show that the cultural construction of homosexuality that was dominant when

they came of age has had differential effects on their discourses.

The perception that society has changed dramatically with regard to the treatment of gays

and lesbians was widespread among my informants. The fact that many informants thought that

the United States had gone through a period of liberalization regarding homosexuality is

consistent with the expectation that changing cultural constructions of homosexuality are

important for explaining the variation in how people talk about homosexuality and same-sex

marriage. Whether or not people thought that the changes were good or bad, most informants

observed that it is more acceptable to be gay or lesbian today than it was in the past:

I think that homosexuality in general is a little bit more on the rise these days. I don’t

know, maybe I’m wrong, but to me, I feel like it’s becoming, people are becoming more

comfortable with it, and it’s becoming more prevalent. (Andy, age 20)
   282  
It’s becoming more and more of an okay way of life, that it’s almost like, not like thumbs

up, but you know, like supported. And it’s like, “Oh, it’s so great,” you know what I

mean. Compared to before where it wasn’t acceptable. (Renee, age 19)

Even in a small town you hear about some, for instance, high school kids are, they’re

more open about it, too. You know, they’re not hiding in the closet, I suppose, anymore.

They’re more open about it, and it seems to be somewhat more accepted, even in this

little town…. (Matthew, age 51)

Both students and parents, liberal and conservative, remarked that gays and lesbians are more

likely to be open about their sexual orientation and more accepted by people in the community,

even in rural areas.

For parents like Matthew, who raised his children in the same small town he grew up in,

the contrast between how students view homosexuality today with how his peers viewed

homosexuality was significant. His son, Nate, recalled an instance in which he heard that

someone he knew in high school came out as gay, and his recounting of the conversation

conveys a degree of nonchalance about the revelation that would have been unlikely in that town

when his father went to high school there:

I don’t know if he came out before I left school or not because it was, I don’t think it was

really, really widespread news… But I heard somebody mention that he was gay, and it

was like, “Oh. That’s news to me.” “Yeah, he came out a couple months ago.” “Oh

really? Great. Good for him.” But I don’t think he was necessarily looked down upon…
   283  
People were like, “Okay.” I think people had their suspicions earlier anyway. (Nate, age

19)

To Nate, the fact that this person came out as gay was interesting, but unremarkable—so

unremarkable that he cannot even remember whether he was still in high school or not at the

time. His perception was that students were nonplussed by the event; it did not seem to carry an

extremely negative stigma that it would have 30 years ago.

This discourse of societal change was not just limited to talk about homosexuality; it was

also applied to the institution of marriage and gender relations more generally. In response to my

question about whether or not marriage is in a crisis, Claudia said:

Older people would definitely be like, “Yes,” because they’re used to the way that things

were done back then. Now time is changing. Is it a crisis? I don’t necessarily think that

it’s a crisis; I think that it’s, not necessarily evolving, it is changing. For the better or for

the worse, I don’t know. But it definitely is not what it used to be. But as we get older,

society gets older. Society is changing, everything’s changing. For Christ’s sake, we’ve

got an African-American President, you know. (Claudia, age 22)

In this quote, Claudia compares her perception of how marriage is changing with the election of

Barack Obama to explain her view that social change is normal. Similarly, in response to my

question about the division of labor in the household, Sarah talked about how different things

were when she was growing up:

Well, I think in modern day marriages, where both couples work, they should share

equally…. It was very different when I grew up, because I mean, in the uh, the women’s
   284  
movement was just beginning back then, and I mean, a lot of us thought a woman’s place

was to be in the home, raising her family. And the husband was supposed to be the wage-

earner, and then that all got turned upside down, and the women were told that they could

do anything that they wanted to. I mean, that’s fine if you want to, but that should be a

choice. (Sarah, age 60)

Thus, with respect to homosexuality, gender, marriage, and family, students and parents alike

observed that significant changes seem to have occurred in society in the recent past. The fact

that these comments about societal change were frequently unsolicited suggests that “times are

changing” is part of the common sense understanding of these topics.

While both students and parents are alike in their perception that society is changing, they

differ in how the effects of this period of rapid social change are evident in their discourses on

homosexuality and same-sex marriage. Talk about homosexuality when parents and students

were in high school or college and narratives of attitude change show how the two cohorts were

differentially impacted by the liberalizing attitudes regarding homosexuality. Parents’ talk about

homosexuality when they were young reflects the fact that homosexuality was heavily

stigmatized during that time, and parents sometimes describe those formative attitudes as a

justification for persistent negative attitudes. By contrast, students’ talk about homosexuality

when they were young reflects the increasing tolerance and acceptance of homosexuality that

they grew up with. Thus, while parents’ narratives of attitude change reflect the fact that they

lived through a period of rapid social change and altered their prior negative attitudes about

homosexuality, students’ narratives of attitude change reflect the process of maturation from

child to adult more than it reflects the altering of already-established attitudes.

   285  

Coming of Age Experiences and Attitude Persistence

The differential effects of this historic period of social change on each cohort is evident

in how they talk about the attitudes about homosexuality that each cohort formed as they were

coming of age. Whereas younger students were more likely to develop positive attitudes toward

gays and lesbians in high school, parents were more likely to have developed negative attitudes

towards gays and lesbians when they were in high school. This is easily accounted for by the fact

that the cultural construction of homosexuality when each cohort was in high school was

different. Whereas students seem to take for granted that gays and lesbians are ordinary people

whose only difference is to whom they are attracted, many parents used the fact that

homosexuality was so heavily stigmatized when they were growing up to justify the persistence

of negative attitudes about gays and lesbians.

Parents reported that the dominant cultural construction of homosexuality when they

were growing up was negative, and they commented that coming out as gay would have been a

rare, and highly stigmatized occurrence in high school. Most parents said that they didn’t know

anyone growing up who was gay, probably because of how dangerous it would have been to be

openly gay:

It was really a negative… You definitely knew there were certain kids in school who

might be, but it was never, ever, you didn’t dare ever talk about it. And you didn’t, they

would get their head beat in. And I remember my daughter… getting an invitation from a

kid in school who was openly gay…. I mean, he was the flamboyant, like somebody, a

character from TV, and he was perfectly acceptable, and he was friends with the kids, and

he had parties all the time…. That would not have happened when I was in high school.
   286  
They would have kicked his butt from here to the end of the football field. It would not

have happened. (Stephanie, age 50)

In this quote, Stephanie contrasts her experience in high school with her daughter’s experience to

describe how negative homosexuality was considered to be when she was growing up. Parents

frequently contrasted their own experiences with gays and lesbians with those of their children in

order to express how much more stigmatized homosexuality used to be when they were growing

up.

The older cohort’s encounter with a more negative cultural construction of homosexuality

is still evident in the discourse of parents who retain negative attitudes that they developed when

they were younger. Some parents justified their opposition to same-sex marriage and

homosexuality because it was how they were raised:

I don’t believe that a guy should be with a guy and a girl should be with a girl. I have a

hard time—these people like that scare me. I don’t know why. I think it was just the way

that I was raised, you know. (Debra, age 57)

In this quote, Debra has difficulty explaining why she has negative attitudes about gays and

lesbians. Heterosexuality seems so natural to her that she can only imagine that it was how she

was raised that has made her feel that way. Her continuing negative attitudes appear to be an

artifact of deep-seated teachings that homosexuality is wrong.

Other parents struggled to overcome their negative attitudes that they were raised with,

finding them to be at odds with how they think they should feel. For example, Bonnie, who is
   287  
supportive of same-sex marriage, reported that there is still a feeling inside her that is negative

towards homosexuality:

I mean, I can’t say that I am 100% perfectly okay if my daughter married a girl, you

know…. But I think the only reason it would bother me a little bit, because like I said, I

knew nothing of it, and it was everything like that was taboo, so bad. As a child, you get

this discipline ingrained into your brain, and you can’t quite overcome it 100%, even if

you open your mind completely. So that’s there, I can’t help that. And same as with

African-Americans. I can tell you all day long I’m not prejudiced, but I do get nervous if

I’m alone in a room with all black people. (Bonnie, age 46)

In this quote, Bonnie compares the deep-seated negative association with homosexuality with a

deep-seated feeling of insecurity that she associates with African-Americans. In both cases, she

struggles to overcome old attitudes that she attributes to old social values that she was raised

with that she no longer believes are acceptable. Similarly, Laura reported that the moral

judgment about the immorality of homosexuality that was ingrained in her still affects her, even

though she rejects it:

I remember I used to think, “Oh, that’s wrong.” But that was somebody else’s idea

planted in me. And so now I try not to think that way. And I’m not going to lie, it comes

up. You know, but I tell myself, “Whoa, I can’t be the judge of that”… So yeah, there are

times sometimes I think it’s wrong, but I catch myself. I don’t like that. (Laura, age 49)

The fact that parents who both support and oppose same-sex marriage report persistent negative

attitudes regarding homosexuality is strong evidence that the cultural construction of

   288  
homosexuality that they encountered when they came of age has had strong and lasting effects

on their attitudes and discourses regarding homosexuality and same-sex marriage.

By contrast, students appear to have developed positive, tolerant attitudes towards gays

and lesbians through their high school experiences. Students generally talked about people

coming out as gay as not being a big deal today, and they are less likely than their parents to

have negative attitudes regarding their encounters with gays and lesbians. For example, Luke, a

19 year-old moderate-conservative student recalled an old teammate of his who came out as gay:

R: There was this kid in my high school who was gay, I played soccer with him in middle

school, and he ended up being gay, and I was like, “Oh,” you know, “that’s cool,

whatever”….

Q: When you found out, you weren’t…. did it freak you out at all?

R: Not really. Which I kind of, I didn’t really expect. I kind of expected to be freaked

out…. You do what you want to do, but I’m not going to do it, and I don’t have to deal

with it. I mean like, you can be gay, I’m not going to be gay. That’s cool, do what you do,

I’m going to do what I do, no big deal. (Luke, age 19)

This tolerance about knowing someone who is gay or lesbian was typical among the students in

my sample, even conservatives. Students described the sexual orientation of gays and lesbians as

generally unthreatening, something that was not any of their business and that did not affect them

in any significant way.

To be sure, homosexuality is still stigmatized and constructed negatively in society at

large and especially in the high school context. For the one gay student, the bisexual students,
   289  
and the students who had close friends who were gay that I interviewed, it is clear that coming

out is still difficult and carries with it social stigma:

R: My best friend, though, got kicked out of school and it was basically because of that.

Q: Because he was gay?

R: Yeah, pretty much. He went to a Christian school, and I don’t think that was the whole

reason he got kicked out, but he didn’t do anything wrong…. Cause some of the people

that he thought he could trust all turned on him, turned against him and disowned and

stuff like that. So he had a very hard time with that whole thing. (Betsy, age 24)

Not only does coming out as gay still carry significant risks, but students who were identified as

gay in high school are still abused:

Q: This guy that you knew in high school, how did people treat him?

R: A lot of people, not so well. Yeah. There was a time there was a fight in the locker

room, and that—I wasn’t there for that—but people weren’t really receptive to him. So

yeah.

Q: Was it this sort of like, “I don’t want to get undressed around this guy because he’s

going to get all turned on?”

R: Yeah, I think so. There was a couple people—there was, I mean, a student who was a

grade ahead of me, and he was gay, and he was also Asian, and people would call him the

“Gaysian,” and he was flamboyantly gay, but he would never talk about it… The kid in

my grade, though, was very open about it. (Jeff, age 21)
   290  
The high school climate that Jeff describes represents the typical high school atmosphere among

students in my sample: homosexuality was stigmatized, and students who were open about their

sexual identity were at risk of taunting, and occasionally violence.

At the same time, there is greater support for gays and lesbians in high schools today, as

tolerance and acceptance of gays and lesbians has become more widespread. Numerous students

reported there being Gay-Straight Alliances (GSA’s) in their high schools, and it is likely that

such clubs help to counter the stigma that gays and lesbians face in high schools. More to the

point, though, GSA’s can facilitate more positive and tolerant attitudes among straight students

by lessening the stigma associated with homosexuality and providing a supportive environment

for people to develop friendships with gay and lesbian students. For example, Jeremiah, a 19

year-old student said that about half of his friends are gays and lesbians because of the GSA in

his high school:

I already had a couple of [gay] friends, like most of my friends actually were, before I

like even joined the club. Just because like that’s the group I hung out with in high

school. Even though there wasn’t cliques, I just like hung out with them. Still, and like

for a lot of them, I didn’t even know that they were. So like I went to the—or until they

came out and told me, or like I went to the meeting and later on they told me, or I figured

it out. (Jeremiah, age 19)

If Gay-Straight Alliances help gay and lesbian students become more accepted, it is no surprise

that conservative Christians oppose such groups. The appearance of one GSA in a Rockford

public school was very controversial, as recounted by Dana, a parent of a student who was

currently in high school there:

   291  
[The high school] just passed the Gay-Straight Alliance, and it really concerned my

husband and I. I knew there was homosexuals in the school. I’ve never seen them, but my

children have said they find them kissing in the hallways and the stairwells. It was a huge

concern, and then when they brought it up to the parents to vote, the school was flooded

with parents giving their opinion… So what the school did is they passed a rule in the

school, absolute zero tolerance for any signs of affection between boys, girls, same sex,

whatever…. My son got upset… He was against [the GSA], but he’s also against the

homosexuals being persecuted at the school. He was against both. So he struggled, and

being that he’s only 15, he hasn’t really learned what to do with that. He doesn’t feel it’s

healthy, but at the same time, he doesn’t want people to be mistreated. (Dana, age 48)

If GSA’s help to normalize homosexuality and make gays and lesbians more accepted among

students, it is understandable why they are opposed by parents like Dana.

The reaction of Dana’s son to the GSA described in the previous quote hints at an

important change in how gay and lesbian students are treated today: even conservative Christian

students who were taught that homosexuality is wrong are more accepting than might have been

the case when their parents were going to school. For example, Ron, a 20 year-old conservative

Christian who opposes same-sex marriage and believes that homosexuality is wrong, was still

very empathetic when a close friend came out as gay:

R: I do have a gay friend. I just found out about a week ago…. This is a friend of mine

from school. Found out he was gay. Had my suspicions, honestly, but I don’t want to

bring it up, “Hey, are you gay?”

Q: That’s not something like,

   292  
R: Yeah, that’s not something I really want to bring up. But we’re still friends. Just don’t

talk about it. We talked about it at the time, but we don’t really, you know, it’s not a

discussion point. He’s gay, I’m straight. You know, I don’t hate him cause he’s gay. He’s

a good friend, and uh, that’s that, I guess. (Ron, age 20)

Even though Ron believes that homosexuality is wrong, he still accepts his friend. While he

doesn’t really think he is gay, he is also worried about how he will be treated if other people find

out.

In general then, the climate in which students first encounter gays and lesbians today is

much less negative and stigmatizing than what was reported by their parents. When parents came

of age, they encountered a cultural construction of homosexuality that was negative, and parents

still retain the negative associations with homosexuality that they developed when they were

younger. By contrast, the period of increasing acceptance of homosexuality and increasing

openness of gays and lesbians about their sexual orientation appears to have encouraged students

to adopt more tolerant and accepting attitudes to begin with.

Narratives of Attitude Change

Members of both cohorts reported their feelings about homosexuality changing over the

course of their lives, but the narratives of attitude change reflect different processes. Among

parents, the narrative of attitude change reflected the process of changing their established views

about homosexuality because of the period of greater openness and tolerance about

homosexuality. Whereas their attitudes were once negative, the fact that they lived through a
   293  
period of changing cultural construction of homosexuality helps explain why they now have

more positive attitudes.

Parents’ narratives typically centered on the importance of getting to know people who

identify as gay or lesbian and realizing that they are just “ordinary” people. When they were

younger, they had been taught that homosexuality was wrong, but after they got to know people

personally, they began to realize that they were no different than them:

We would have been called, you know, aghast as teenagers going, “Oh my goodness,”

you know. Now that I am much older… I guess I’ve had more exposure and relationships

with gay and lesbian folks that, again, that you realize they’re really not any different

than, than me, except for their sexual preferences…. I guess I’m more accepting. It’s like,

it’s less of a big deal. (Tom, age 47)

Tom attributes his greater acceptance of gays and lesbians to getting to know them, in contrast to

the shocked reactions he imagines he and his friends would have had when they were growing up

in the 1970s. Bonnie, a 46 year-old self-employed artisan, also volunteered how much her

feelings about homosexuality changed since she was younger:

I gotta admit, the first time I met people like that, I was very uneasy. But I was brought

up to think that it was wrong…. I didn’t even know it existed until I was hit upon by a

woman, and it just blew my mind because I didn’t know what the heck was going on, and

I was appalled. But that’s because it was like an experience out of the blue that I didn’t

know would ever exist, you know. As I got older, you know, and as my kids grew up, I

got a lot more accepting of it because they had a lot more homosexual friends. And
   294  
actually, I’ve gotten along with every one of them I’ve ever met better than a lot of the

straight friends. (Bonnie, age 46)

It is noteworthy in this quote that Bonnie gained her personal contact with gays and lesbians

through her daughter, and that her subsequent contact with gays and lesbians has probably been

less threatening to her than her initial contact with a woman who flirted with her. It was by no

means uncommon for parents to report that the experiences and opinions of their children have

caused them to change their attitudes regarding homosexuality to some degree.

These narratives of attitude change can be understood in light of the theoretical logic of

cohort and period effects if one poses a hypothetical counterfactual: if the period of liberalization

of attitudes toward homosexuality had not occurred, and had gays and lesbians not become more

likely to be “out” in their personal lives and portrayed as normal in the mass media, it is less

likely that this personal contact would have occurred and that the earlier teachings that

homosexuality was wrong would have been challenged. But given that they did live through this

historical period, their attitudes liberalized in ways consistent with the increasingly accepted idea

that homosexuality is an inherent part of a person’s identity and something that should be

tolerated.

There were other narratives of attitude change among parents that reflect how living

through this historical period has interacted with their changing worldviews. For example,

Gerald, a 60 year-old Catholic, talked about his changing feelings about homosexuality and

same-sex marriage in light of his greater appreciation for the complexities of the world. Because

of his experiences with family members who have disabilities, another group that is not fully

accepted in society, he no longer sees the world in such simple terms of right and wrong:
   295  
Life is such a complex of situations, of views, and my family’s really mixed. I mean, we

have went through a lot of struggles. There were a couple of times that we were going to

lose my son, you know. And the people that came to help, the people that were there for

us, were not always the people that you thought would be your friends…. I think my

views have really changed. There was a time that I probably would have said I would not

be open to a gay type of relationship or something like that, you know. I would have said,

[grumble], or I would have tried to be politically correct but not really feel correct, you

know. And that’s not there anymore…. A child isn’t better off just because he’s got… a

male dad and a woman. (Gerald, age 60)

Because of his experiences with people with disabilities, Gerald has become more tolerant and

accepting of people who are different from him, and he is no longer willing to judge someone

negatively because of their sexual orientation.

Another informant reported feeling less threatened by homosexuality as she became older

and attributed this change to both changing hormones and personal contact:

R: I remember back when I was going—I had joined the service for a while, too, as an

officer, and you know, I didn’t like these dyke-y gals, you know. I was very much against

it. I just, I guess the threat, or I don’t know why. I just did not like it. But now, of course,

my viewpoints are very much different….

Q: Why do you think your feelings about it changed?

R: Because I got to know the people, and they’re actually, you know, human beings. It’s

not just sexuality. When I was going to college, everything was hormonal, you know, for

me. As you get older, you lose that. I have lost that hormonal crap, and you look at the
   296  
person as just, you know, a person. And you don’t look at the sexuality. Just what do they

have inside? (Natalie, age 60)

According to Natalie, she became more accepting of homosexuality because she got to know gay

and lesbian people and because sexuality became less important to her. What is unusual about

this quote is that it calls attention to a possible age effect that might contribute to the overall

liberalization of attitudes about homosexuality: as one’s body changes with age, people’s

attitudes might be less hormonally driven. This is plausible, given the continued homophobia

among some young people; but such an age effect would not explain why young people are more

likely to support same-sex marriage than older people.

Of course, there is no way to verify from these narratives whether or not hormones,

personal contact, or experiences with other stigmatized groups actually caused these attitude

changes; these are simply the stories that individuals have constructed in order to explain why

they think their views changed. What is essential to all of these stories, though, is the fact that the

individual lived through a change in how homosexuality was culturally constructed by

mainstream society, a period of rapid liberalization of attitudes and changing construction of

homosexuality in the mass media, in politics, and in everyday life. This society-wide shift toward

greater tolerance and acceptance of gays and lesbians has no doubt played a role in encouraging

these individuals to adopt a more tolerant discourse towards gays and lesbians.

Students are also familiar with these narratives of attitude change, and some students

constructed their own. However, the narratives of attitude change reported by students reflect the

process of maturing from a child to an adult, rather than the process of challenging and changing

their established views on homosexuality. One common narrative used by students to talk about
   297  
their changing attitudes centered on the process of coming to feel that the use of the word “gay”

or “fag” as an insult is degrading to gays and lesbians. Among students in school, these terms

have long been used as derogatory slurs that could be applied to anyone, but as students matured,

they began to realize that these slurs were hurtful to gays and lesbians. For example, Jill, a 19

year-old student with a gay uncle, described how she became aware of how such school-yard

taunts implied something negative about gays:

I didn’t really cognitively understand until, and really get defensive about it until high

school. Because that’s when I saw the biggest, “Oh you’re gay, you’re this.” And if

somebody was gay, they were, you know, “Stay away from him,” and stuff like that….

People call each other gay all the time, and faggots all the time, and homos and stuff. But

like I think that’s wrong…. People don’t think about what it really means. It’s just used

as an insult today, I think. (Jill, age 19)

In this quote, Jill describes why she now disapproves of the ways that words like “gay” and

“faggot” are used as insults by students. It illustrates the process of maturation because she

becomes aware of what this means for people who actually do have a homosexual orientation.

Students frequently contrasted their current tolerant attitudes towards gays and lesbians with

their immaturity when they were younger:

When I was, you know, kind of around that middle school, eleven, twelve, thirteen sort of

area, I know my friends and I were very homophobic. Just because I think at that age,

you’re like really trying to fit in with the crowd and, you know, everything different is

bad. And we used to, you know, say mean things about, “Oh my God! I think that girl’s a

lesbian,” and you know, say that kind of crap. (Carolina, age 20)
   298  

Carolina attributes her previous negative attitudes to the middle-school peer culture that she was

in. In the context of our discussion about her current attitudes about homosexuality, she implies

that those negative attitudes were due to her immaturity, and that she now knows better than to

think of lesbianism as an insult.

Another narrative of attitude change used by students had to do with learning what

homosexuality really means through personal contact. This narrative is similar to how the parents

talked about the effects of getting to know someone who is gay or lesbian, but it differs only in

that, for students, the realization occurred when they were teenagers first developing their sexual

identities. For example, Tiffany reported that she was shocked when she first learned that her

aunt is a lesbian because she didn’t really understand it:

R: I was like, I think twelve, and I was just always told, you know, “Oh, they’re best

friends and they live together,” and I thought, “Oh, that’s so cool!” And then one day, my

little sister figured it out, she’s like, “Hey mom, does everybody know that they’re gay?”

And I was like, “Oh my god!” And then I come to terms with it, and she doesn’t change

any because she is [gay]. Nothing’s changed….

Q: What do you think it was that like first really bothered you when you found out that

your aunt was [lesbian]?

R: I don’t know, because I was still young, and I just, I didn’t really know what it meant,

I guess. I mean I knew it meant that they were, you know, sexually active together. I just,

maybe I thought it was weird… Maybe it was more that it was kept a secret from me, and

it was like, that’s what made it more of like, “oh.” (Tiffany, age 23)
   299  
In Tiffany’s story, she contrasts her initial shock to learning that her aunt is a lesbian to her final

realization that her aunt’s sexual orientation does not make her any different as a person than if

she were straight. This narrative is no different from the stories that parents told about getting to

know gays and lesbians when they were older; the only difference is due to the fact that she was

so young when she found out about it, and she had not yet developed an adult understanding of

human sexuality.

In sum, narratives of attitude change were similar among students and parents, but the

narratives implied different processes, depending upon whether or not they had already come of

age. Whereas students’ narratives implied that the process of attitude change was tied to

maturation—the process of developing an adult identity and sexual awareness—the narratives

used by parents implied that the process of attitude change occurred after they had already

become adults and developed attitudes about homosexuality. This difference in narratives is

logical because students and parents experienced the period of change regarding the cultural

construction of homosexuality at different times: students were more likely to develop tolerant

attitudes to begin with because they came of age after the cultural construction of homosexuality

had already changed, whereas parents lived through this period of change, and they either

retained the negative attitudes they developed when they were younger or struggled to develop

tolerant attitudes more like those of their children.

Comparison of Matched Pairs

An explicit comparison of the discourses of matched pairs of students and parents also

supports the view that cohort differences in discourses about same-sex marriage are due to social

generational change. Initially, it should be noted that the social generational explanation of
   300  
differences in attitudes regarding same-sex marriage is supported by my interview data in that no

student’s discourse was more conservative or opposed to same-sex marriage than their parent’s.

Measured by the discourses described in Chapters 4 and 5, 17 of the 32 students used discourses

that were different from their parents, and the student’s discourse was always more supportive of

same-sex marriage than their parent’s. Among the 15 pairs whose discourses were similar, the

only evidence I found that a student might be less supportive (or more opposed) to same-sex

marriage than their parent was a couple of statements from one student that suggested mild

homophobia. However, he also said he supports same-sex marriage, and it was not clear that he

was any more or less supportive of same-sex marriage than his parent.

However, the simple cohort comparisons presented in the previous chapters are

unsatisfying for two reasons. First, many of the attitude differences regarding same-sex marriage

that we observe between parents and their children are not obviously due to any cohort effect.

Because parents and children oftentimes differ with each other along a number of dimensions, it

is not clear whether attitude differences with regard to same-sex marriage might be better

explained by some other variable, such as religious beliefs. For example, Sarah, age 60, and her

son, Daniel, age 27, could not be more different. While Sarah strongly opposes same-sex

marriage, Daniel strongly favors it. Sarah is a Protestant Christian who attends church every

week, while Daniel identifies as an atheist. While Sarah has moderate political beliefs, Daniel’s

political beliefs are so far left that he says, “between about 18 and 21, I was actually a card-

carrying socialist,” and “My proudest moment is getting made fun of on Fox News.” Daniel also

has higher level of educational attainment than his mother, even though he has not finished

school. Compared with these differences, it is doubtful that the difference in attitudes about

same-sex marriage should be attributed to a cohort effect.

   301  
The second reason that a simple cohort comparison does not produce a satisfying

explanation of cohort-related attitude differences is that, for the most part, parents and children

express similar levels of support or opposition to same-sex marriage. Because of the power of

parental socialization, the reality of attitude change among parents, and the power of political

and religious ideologies, the overwhelming majority of student-parent pairs expressed similar

opinions about same-sex marriage. When measured in terms of a conventional survey, which

would include categories of support, opposition, or “don’t know,” 22 of the 32 student-parent

pairs could be classified similarly.

These two facts represent significant challenges to the social generational explanation.

The reality of attitude change among older adults appears to contradict the expectations of the

aging-stability hypothesis that is part of any cohort analysis: if attitudes are unstable as we age,

then the nature of a person’s encounter with society when they come of age should not have the

lasting effects typically associated with cohort-centered explanations. Moreover, cohort

differences are often better explained by differences in other variables, such as educational

attainment and political ideology. With a simple cohort comparison, it is unclear that any cohort

differences would actually be due to their cohort membership.

However, if we set aside the student-parent pairs, like Sarah and Daniel, who are

different from each other, and we consider only students and parents who otherwise agree with

each other on issues related to politics, religion, marriage, and sexuality, we gain additional

analytical leverage on the question of cohort effects. By making comparisons only among

parents and children who are similar in most respects, we are better able to isolate any cohort

effect that might exist.

   302  
By comparing similarities and differences simultaneously between and within cohorts, I

show that evidence of cohort effects can be seen even when religious and political ideologies and

generational influences are controlled for. Older liberals are like younger liberals in their support

for equal rights for gays and lesbians, but they are like older conservatives in that they are more

likely to express negative attitudes or discomfort with homosexuality. Similarly, younger

religious conservatives are like older religious conservatives in their beliefs that homosexuality is

immoral, but they are like younger liberals in that they are more likely to express tolerant

attitudes and support for equal rights for gays and lesbians.

In order to classify student-parent pairs in this way, I coded the interviews of each

informant according to their political and religious beliefs, their definitions and attitudes about

marriage, their attitudes about cohabitation and premarital sex, their attitudes about divorce, and

their gender ideologies. I also coded their attitudes about homosexuality and same-sex marriage.

Among the 32 pairs, six pairs expressed different opinions about same-sex marriage that are

most plausibly attributed to other significant differences in their worldviews, such as religious or

political ideologies. Four of the six parents used oppositional discourses to talk about same-sex

marriage, and, four of the six students used supportive discourses. The two students who used

middle-ground discourses had parents who used oppositional discourses, while the two parents

who used middle-ground discourses had children who used supportive discourses

The other 26 student-parent pairs, however, were very similar to one another with respect

to their discourses about religion, politics, marriage, and gender. Six of these pairs can be

classified as conservative evangelical Christians, and two of these pairs can be classified as

political conservatives whose Protestant religious identities are less salient in their discourses

about marriage and sexuality. Five of these pairs can be classified as religious moderates, people
   303  
that hold moderate-to-liberal political beliefs but who identify as Catholic or Jewish. The

remaining 13 of these pairs can be classified as non-religious liberals, people who are atheist,

agnostic, spiritual, or non-practicing and who have secular liberal worldviews.

The first two groups tend to use similar patterns of discourse to talk about same-sex

marriage, so I refer to them collectively as the religious conservatives; the latter two groups also

tend to use similar patterns of discourse to talk about same-sex marriage, so I refer to them

collectively as the liberals. I describe the same-sex marriage discourses of each group in turn.

Religious Conservative Discourses

All eight parents who are classified as religious conservatives used unambiguously

oppositional discourses to talk about same-sex marriage. As described in the previous chapters,

these discourses are characterized by religious or heterosexual definitions of marriage, negative

attitudes about homosexuality, a negative moral evaluation of homosexuality, and the belief that

homosexuality is a lifestyle choice that should not be condoned by society. Seven of the eight

parents believed that there would be tremendous negative consequences for society if same-sex

marriage were legalized, and most talked about those effects in terms of the continuing moral

decline of society.

However, five of the nine children of these parents used ideologically conflicted

discourses, rather than oppositional discourses, to talk about same-sex marriage. Two of the five

students who used ideologically conflicted discourses do not have strong religious worldviews,

even though they identified as Christians. Without those strong religious beliefs to shape their

views on homosexuality, their discourses about same-sex marriage more closely resembled

supportive discourses than oppositional ones. For example, Jesse, a 27 year-old student,
   304  
identifies more as politically conservative, and so his opinions about same-sex marriage are

shaped more by conservative political principles than religious principles. In particular, he

believes that since the government provides rights and benefits that are associated with marriage,

he thinks that the government should extend those rights and benefits to everyone. As described

in Chapter 4, he believes that the denial of those rights and benefits to same-sex couples

represents an unwarranted invasion of privacy by the government. At the same time, consistent

with his belief in local democratic control, he believes that, ideally, governments should not

interfere with marriage, and that communities and churches should be able to refuse to allow

same-sex couples to marry:

Let’s say there’s no regulations or anything on marriage… Let’s say, Texas, the

community is probably just not going to allow it, you know. You can do it if you can find

someone that’s going to allow it, but you know, for the most part, you’re not going [to].

Now, in San Francisco on the other hand, that’s part of the community, you know. That’s

what, you know, it’s their business and their community and that’s what their community

is all about. So I mean, I don’t think if people in San Francisco would be really happy if a

bunch of people from down-state Texas came up and said, “You can’t do it this way, we

want…” It’s the same thing; they do the same thing. They go to down-state Texas or

wherever and say, “Well, we demand the same thing we had in our old community.” I

don’t know. Live and let live. Like I said, live and let live. (Jesse, age 27)

Jesse supports the ultimate principle of local control and self-determination by communities, and

even though he thinks gays and lesbians should have equal rights, he doesn’t think it’s right for
   305  
outsiders (e.g. the federal government) to dictate to communities what their moral standards

should be. Same-sex marriage should be illegal if that’s the way the community wants it.

Moreover, Jesse argues that if gays and lesbians get equal rights that they also must not

get any special protections, and if people don’t like them, then they just have to deal with it, like

everybody else:

What comes with equal rights is equal criticism. I mean, they go out, for the most part

you see, “Well, we just want to be equal.” Well, you know, if I criticize, I object, then

I’m hatred, or I’m a homophobe, or something like that. I’m a bad, evil person. I’m

intolerant or everything like that. Well, it’s like, look, you want to bring your idea out

here, it’s going to be ridiculed… You’ve got to understand, if you want equal rights,

you’re going to be opened up. You’ve got to take the responsibility that comes with it.

(Jesse, age 27)

Luke, the other moderate conservative student without a strong religious identity, expressed a

similar sentiment when he recounted a TV episode that he saw that, in his mind, exposed the

flawed reasoning behind hate crime legislation:

It’s a show about New York City firefighters… and the chief, his son is gay… He went

into a bar that he didn’t know was a gay bar and ended up getting in a fight with

somebody just completely randomly. But he got charged as a gay-basher and all this crap,

so he had to have his son who was gay come and bail him out or whatever… It’s the

same thing as if I were to go punch some black guy in the face, it would be a racial

beating…. It would be portrayed in that way. And that’s how it was in this episode, like it

showed that they were regular but at the same time, everything’s so unequal because if
   306  
you do something, it’s portrayed as something so much different, just because of that one

detail. (Luke, age 19)

Thus, while these conservative students supported equal rights for gays and lesbians like the

more liberal members of their cohort, their conservative identities make their discourse

distinctive in that they perceive a danger in gays and lesbians being given special rights and

privileges. Their discourse is not totally supportive of same-sex marriage, but overall, they show

a level of support and tolerance for gays and lesbians that is very different from the oppositional

discourses of their parents:

I think they’re taking it too far by wanting marriage marriage, like calling it marriage. If

it’s about, like I said, if it’s about the rights, man, I’m with you 100%. Like let’s start a

petition or sign something, I don’t know, fix it. But I just don’t think—if it’s about

calling it marriage, I’m not with you. (Luke, age 19)

To describe these discourses as ideologically conflicted does not mean that they are confused or

that they make no sense; it only means that they reconcile elements of both supportive and

oppositional discourses in unique ways.

The other three religious conservative students who did not use oppositional discourses

instead used varieties of ideologically conflicted discourse that centered on the relationship

between homosexuality and Christian morality. For example, Bethany, age 22, is like her mother

in terms of their religious and political identification, their definitions of marriage, their

disapproval of cohabitation and divorce, and their belief that homosexuality is a sin. However,

they disagree about same-sex marriage. As described in Chapter 4, Bethany argues that gays and
   307  
lesbians should have the same rights as heterosexuals because, in her mind, the sin of

homosexuality is no different than any other sin. While she believes that there is an element of

choice involved in pursuing one’s sexual urges, and she believes that homosexuality is sinful, she

has positive attitudes toward gays and lesbians. When I asked her about her memories of the first

time she encountered someone who is gay, she told me about her freshman English teacher in

college:

Oh, I didn’t know gay people were so cool…. It was almost like, “Really? One up for

you.” I mean, so I was more like excited to see someone, to be subjected to someone,

who I thought was intelligent, knew what they were doing, had their stuff together, you

know. They weren’t a bad person, you know, cause I, just as much as I was raised by the

Bible, I mean, God, your gut instinct tells you a whole lot about people before you even

know it. (Bethany, age 22)

Interestingly, Bethany’s mother, Andrea, has similar views about homosexuality. She also

regards homosexuality as a sin that is no worse than her own sins. She even talked about how

convincing Bethany’s arguments about same-sex marriage are:

I can honestly say that Bethany has probably broadened my viewpoint in this a little bit,

because her and I have had discussions about this. And if you took the same two men or

women and took sex out of it, to the point where they were just roommates that cared

about each other and shared laundry and shared groceries and made sure the rent got paid

and really understood each other, you know—that really, in my opinion, I mean, that’s

when I feel that these people, you know, they have a marital concept. So if you just took

sex out of the equation, is that what’s left? Yes. (Andrea, age 45)
   308  

In this quote, Andrea describes how if sex is taken out of the equation, a same-sex couple would

have the same qualities necessary for marriage that anyone else has. But this does not mitigate

her oppositional discourse about same-sex marriage. Not only does she oppose same-sex

marriage, she was very explicit in her opposition to civil unions and equal rights for gays and

lesbians:

Now, are you asking me if I would accept it? Would I vote for it? No, I wouldn’t. I’m

not, I’m gonna fight against this. I would, period. That being said, I wouldn’t jump across

the table. I wouldn’t run them over with the car…. It’s not that I don’t think they can’t

have a wonderful relationship, but I refuse to morally support it. (Andrea, age 45)

Her opposition to same-sex marriage and to equal rights for gays and lesbians more generally

remains firmly grounded in her moral convictions that homosexuality is wrong, despite her

daughter’s arguments.

Kyle, age 23, has struggled to reconcile his Christian faith with his empathy and positive

attitudes toward gays and lesbians. Like Bethany, he defines homosexuality as a sin, but he also

believes that people do not choose to be gay—just like heterosexual people do not choose their

sexual orientation. This understanding of homosexuality is consistent with his experience of

witnessing someone that he met through a campus ministry struggle to reconcile his homosexual

feelings with the Christian teachings that homosexuality is wrong. He described how sad it made

him feel that the individual simply could not change his feelings of sexual attraction, despite his

desire to be consistent with Biblical teachings. Ultimately, Kyle said he would oppose same-sex

marriage, but he is uncomfortable with the way his church views homosexuality:
   309  
I think a lot of conservative Christians or any kind of Christians are very unapologetic or

unforgiving. And I know that there are Christians out there who were homosexuals and

maybe still are. And I don’t think God condemns them because they have those

tendencies. I think what He is sad about is how they act upon them. And as Christians,

myself included, we don’t, I don’t think we, in terms of homosexual relationships and

marriages, we don’t extend enough grace to those people. We don’t seek to understand

them. And it’s almost more sad to me that they, homosexuals, feel so hated by us, and

legitimately so. They should feel that way, because there’s a lot of people that are—and I

think we’re just as responsible as they are for that wrong, for that big rift. Boiling it down

to it, at the end of the day, I don’t agree with it. (Kyle, age 23)

The conflict in this quote is evident: between the religious teachings that homosexuality is wrong

and his feeling that gays and lesbians should be embraced by his religion. In a number of ways

during the interview, Kyle talks about how he wants gays and lesbians to be treated more

respectfully, both in the church and in society at large; but he thinks that respect should be given

through means other than allowing same-sex couples to marry.

Kyle’s father, Stan, however, expressed no such sentiments. To the contrary, his father

felt as though gays and lesbians should be more respectful to Christians, and that Christians were

the ones being persecuted. He said that when Christians express opposition to homosexuality and

same-sex marriage, they are condemned by gays and lesbians, and he compared this to the

persecution of Jesus:

It’s kind of like the way the Jews went about comdemning Jesus in front of Pilate.

Because they really didn’t have anything, they couldn’t convict him of any crime because
   310  
he didn’t commit a crime. But they insisted that they wanted him crucified, so they got

the crowd stirred up, and they thought if they could out-shout Pilate, if they could make

themselves noisier or louder or demand more or threaten more, that somehow they could

justify that action. It seems to me that a lot of that tactic is practiced today by the

homosexual community. Because it seems to almost be a militant-type thing that anytime

you talk in a way that’s not supportive, then you’re against it. And then, of course, you’re

anti-gay or anti-everything else. (Stan, age 59)

Stan saw himself taking a principled stand against sin and that he was being condemned by the

crowd for it. He imagined that people supported same-sex marriage because they wanted to

“jump on the bandwagon” and be part of the “accepted crowd,” rather than risk being identified

as anti-gay. Thus, Stan’s oppositional discourse stands in stark contrast to his son’s perspective,

even though they both ultimately say they oppose same-sex marriage.

In sum, young religious conservatives used ideologically conflicted discourses to express

tolerant attitudes and support for equal rights for gays and lesbians along with more liberal

members of their cohort, while at the same time expressing religious and political beliefs that are

similar to their parents. While their parents found no conflict between their conservative political

and religious beliefs and their attitudes about homosexuality, many of the students did. The

students with strong religious identities were taught that homosexuality is a sin that should be

opposed, but some of them also had positive attitudes towards gays and lesbians and wanted

them to have respect and equal rights. Other conservative students felt that gays and lesbians

deserved equal rights, but because of their definition of marriage or their political ideology,

stopped short of speaking totally in favor of same-sex marriage. The variety of opinions about
   311  
same-sex marriage expressed by young religious conservatives thus represents the different ways

in which the students reconciled elements of their cultural repertoires that appear contradictory.

Liberal Discourses

Among the 18 pairs of liberals, almost all of the students used unambiguously supportive

discourses in order to talk about same-sex marriage. In general, the students had positive

attitudes about gays and lesbians and had liberal or libertarian views with regard to politics. Most

had companionate and secular views of marriage, cohabitation, and sexuality, although some of

the students who identified as Catholic defined marriage as a religious institution and had less

liberal attitudes about premarital sex and cohabitation. Only one student used a sort of

ideologically conflicted discourse to express any reservations about legalizing same-sex

marriage: while he saw no problem with same-sex marriage, he thought that same-sex couples

should not be allowed to raise children. I suspect that this was at least in part due to his belief

that homosexuality was caused by the environment in which a person was raised.

By contrast, six of the parents of these students used middle-ground discourses to talk

about same-sex marriage in ways that showed similarities to the more conservative members of

their cohorts. The liberal parents who used libertarian pragmatic and ideologically conflicted

discourses to talk about same-sex marriage resembled their children in their support for equal

rights for gays and lesbians, but they resembled other older adults in that they oftentimes retained

implicit negative attitudes or associations regarding homosexuality. Thus, they did not express

their feelings about same-sex marriage in an unambiguously supportive manner, like their

children.
   312  
Of the five student parent-pairs who can be classified as religious moderates, three

parents used middle-ground discourses to talk about same-sex marriage. Tom, a 47 year-old

Catholic father who was planning to vote for the Republican candidate in the 2008 Presidential

election, surprised me by how tolerant he seemed with respect to homosexuality and same-sex

marriage. As described in previous chapters, he used a libertarian pragmatic discourse to talk

about same-sex marriage and explain why he did not oppose it. He acknowledged that his

tolerance about these issues might be contrary to his religious beliefs, but he insisted that it was

not really any of his business to get involved in the issue:

If you believe that that is, you know, an unnatural way of life, then you’re going to stomp

your fist and jump up and down saying, “No, they shouldn’t have it, it’s bad.” It’s like,

okay, I’m sure you’re not all that holy yourself. I’m sure there’s something in your past

that you don’t want people to know, so don’t be pointing fingers around. Again, it’s not

really affecting your life. It means a lot to these people, it doesn’t really change your life,

so don’t worry about it. Just, if that’s what they want to do… (Tom, age 47)

Tom’s unusually reflexive answers to my questions can at least partly account for why he

refrains from opposing same-sex marriage. But he also believes that Christianity teaches

tolerance and understanding of others, so he ultimately says that gays and lesbians should be

allowed to marry because it doesn’t affect him.

Tracey, a 53 year-old Catholic who leans Democratic on political issues, uses

ideologically conflicted discourse to talk about same-sex marriage. As described in the previous

chapter, she worries about unintended negative consequences of legalizing same-sex marriage,
   313  
but she also thinks that gays and lesbians deserve equal rights. When asked what she would do if

she had a chance to vote on it, she replies:

What would I do in the voting booth? I probably oppose it, just because it’s the unknown.

Not for any good reason. [laugh] It sounds bad, but… (Tracey, age 53)

When I asked Tracey about civil unions, she responded more positively:

R: I’d like that better. Yeah, it’s so strange, I think, to have it labeled something different

would make a big difference. Isn’t that weird? [laugh]

Q: I feel like it is and it isn’t. I mean labels are important for people.

R: And I’m thinking of the kids, you know, like trying to explain, “Your mom and dad,

oh no, mom and mom.” And if they could say it’s a civil union or something like that,

instead of marriage, maybe it would make more sense down the line. (Tracey, age 53)

Tracey’s mixed feelings about the issue included not only her opinion and the imagined

consequences of legalizing same-sex marriage, but also her personal contact with gays and

lesbians. She is able to name two people she knows who is gay or lesbian and whom she likes,

but she also recalls an uncomfortable experience seeing gays in public:

We were walking around at night in this one area… which I didn’t know, but it is total

gay town. And it was weird walking around town with two guys in front of you holding

hands. That was where it was like, “Okay, we gotta get out of here cause I don’t want to

take this.” [laugh] “This is a little weird.” To watch it, it’s weird. But kind of like

marriage of another couple, it’s like, “Okay, keep that in the bedroom. I don’t want to see

it on the sidewalk.” (Tracey, age 53)

   314  

Tracey’s daughter, Sandra, expressed no such mixed feelings with regard to same-sex marriage,

its imagined consequences, or to gays and lesbians she knows. To the contrary, despite her

Catholic faith and her mixed feelings about premarital sex, Sandra spoke passionately about

same-sex marriage and expressed noticeable empathy for gays and lesbians who are struggling

for equality.

Of the 13 parents who, along with their children, can be classified as secular liberals, ten

used supportive discourses to talk about same-sex marriage. Most of these parents supported

same-sex marriage because they embraced the moral values of openness, acceptance, and

diversity, and they rejected religious arguments about the immorality of homosexuality. Some

parents had developed or been raised with strong religious or political beliefs that demanded

social justice and equality, regardless of a person’s differences, and so they had no trouble seeing

the struggle of gays and lesbians for civil rights in the same light as other movements they had

been involved in. Other parents used supportive discourses to talk about same-sex marriage

because of family members or close friends that they have known for decades to be gay or

lesbian. Still other parents, as I describe above, had undergone attitude changes with regard to

homosexuality over the course of their lives and now unambiguously support same-sex marriage.

However, three of the thirteen parents used libertarian pragmatic discourses in order to

talk about same-sex marriage. Jillian, a 49 year-old nurse, exemplifies the libertarian pragmatic

discourse because she rejects the opposition to same-sex marriage as simple “prejudice and not

understanding,” but when asked explicitly for her opinion, she avoids giving explicit support:

Q: Do you think we as a society should legalize same-sex marriage?

   315  
R: I don’t think it makes one bit of a difference. It’s like, if you’re Catholic and I’m

Jewish and this one’s something else, I don’t think it makes any difference, you know.

Because they’re trying to, they might want to have a family and raise, you know, their

way. It doesn’t matter to me anyway. (Jillian, age 49)

While Jillian certainly indicates that she does not oppose same-sex marriage, she also does not

specifically say that she is for it. In avoiding judgment, she also denies that there would be

significant effects on society. In response to four separate questions about effects that legalizing

same-sex marriage might have on society, she denies that there would be significant effects each

time. Because of her secular view of marriage and her belief that homosexuality is not something

that people choose, she denies that legalizing same-sex marriage would change the meaning of

marriage or encourage homosexuality. However, she definitely perceives a change in how gays

and lesbians are treated by society today, compared with when she was younger:

Even when like Ellen Degeneres and all that kind of stuff, at first they were shunned, but

that was because of the times. But now everybody seems to be accepting everybody, you

know. Yeah, there’s like, they all come out of the closet. Nobody cares anymore… But I

always figure, to each their own. If that’s what they like, fine, you know. I’m not going to

tell them anything. (Jillian, age 49)

A remark like this one about how much society had changed over the years, with regard to how

gays and lesbians are treated in society, was common among parents who used libertarian

pragmatic discourse.
   316  
By contrast, Jillian’s son, Kevin, a 20 year-old junior, is supportive of same-sex marriage

and appears to have no problems with homosexuality whatsoever. Kevin is typical among secular

liberal students in that he articulates a companionate view of marriage, believes that cohabitation

and premarital sex are good, and is surrounded by people who agree with him. While he has

certainly encountered opposition to same-sex marriage through class debates and conversations

with his father, he dismisses their arguments:

My dad is complete like, totally against it. I always try to reason with him, like “What’s

the big deal?” I don’t know. He’s a complete homophobe, like, you know, completely….

“Oh, you’re friends with a homosexual person, they might think I’m like that too.” And

it’s like, “Oh, does it matter? Like if someone, what do you care?” (Kevin, age 20)

Kevin finds the idea of someone being bothered by homosexuality to be ridiculous.

Homosexuality, and the idea that gays and lesbians are ordinary people, just like anyone else, are

so common-sense to him that he does not really understand why people are opposed to equal

rights for gays and lesbians:

I’m sure someone like my dad wouldn’t accept it. He’d be like, “No way.” I don’t see the

big deal, you know. I’m sure it has something to do with the conservatives and the

government just saying like, the United States is founded on morals, morals that we’ve

had for years…. Whatever. Times change, you know, and you just have to accept change.

You can’t live in the past, you know. (Kevin, age 20)

Kevin dismisses opponents to equality for gays and lesbians as people “living in the past,”

people who cannot accept the reality of a changing society.

   317  
In sum, older liberals are more likely to use middle-ground discourses because of

similarities that they share with both their children and with older conservatives. Like their

children, they support equal rights for gays and lesbians, but like the more conservative members

of their cohort, they also retain negative attitudes or associations about homosexuality that seem

to prevent them from being willing to say they favor same-sex marriage. Some parents use

libertarian pragmatic discourses to avoid stating an opinion about the issue, while others use

ideologically conflicted discourses to express their worries about the effects of legalizing same-

sex marriage. In both cases, even though they avoid giving unqualified support, they also do not

try to justify opposing same-sex marriage. While the majority of liberal parents used supportive

discourses to talk about same-sex marriage—because of their political beliefs, personal contact

with gays and lesbians, or attitude change—middle-ground discourses are still common among

older liberals.

Discussion: Social Generational Change and Tacit Definitions of Homosexuality

By paying attention to middle-ground discourses and by making comparisons

simultaneously between and within cohorts, we can more accurately specify the ways in which

cohort membership shapes attitudes and discourses about same-sex marriage. Because we can

control for generational influences and the influences of political and religious ideologies to

some extent, this analysis shows stronger evidence that cohort differences in discourses about

same-sex marriage are actually due to social generational change. In particular, I argue that the

intersection of a person’s cohort and their social location within that cohort shapes the structure

of the cultural repertoire and the ways that they use elements in that repertoire to talk about

same-sex marriage.
   318  
It is readily apparent from this analysis how a person’s religious and political ideologies

shape the structure and use of their cultural repertoire. Having a strong religious or political

identity not only equips a person with specific cultural tools for talking about issues like same-

sex marriage—such as the principle that minorities deserve equal rights or the Biblical teaching

that homosexuality is a sin—but it also provides people with sets of strategies and practices for

using those cultural tools to express an opinion—such as framing the argument about same-sex

marriage as being about civil marriage or insisting that you demonstrate love for the sinner by

hating the sin.

It is also true that a person’s cohort membership shapes the structure and use of their

cultural repertoire. As the literatures on collective memory and cohort analysis have shown,

growing up during different historical periods provides individuals with different sets of first-

hand experiences and memories that they will use in particular ways to articulate opinions. In the

case of same-sex marriage, the parents have direct experiences and memories of how the cultural

construction of homosexuality has changed in the United States, and their narratives of attitude

change and recollections of their high school experiences with gays and lesbians illustrate this.

The students are no cultural dopes—they have their own narratives of attitude change and know

that homosexuality is more accepted in contemporary society—but the students, to a degree, take

for granted the changes to the social structures of gender, sexuality, marriage, and family

because they never personally lived in the American society whose institutions were differently

configured.

I argue that the social generation concept represents the intersection of the ways that

cohort and social location shape the structure and use of people’s cultural repertoires. The

influences of cohort membership and political and religious ideologies do not operate separately
   319  
from each other, but instead work together in intertwined ways, as the analyses of discourses

about same-sex marriage have shown. For older religious conservatives and younger liberals, the

influences of cohort membership and political and religious ideologies reinforce one another,

such that people use elements of their cultural repertoires to create unambiguous discourses of

support and opposition. For younger religious conservatives and older liberals, the influences of

cohort membership and political and religious ideologies are cross-cutting, and the middle-

ground discourses described here show some of the various ways that people creatively combine

and reconcile particular elements of their cultural repertoires in order to articulate opinions that

make sense to them.

The ways in which social generational processes shape the discourses about same-sex

marriage can be further illustrated by examining one particularly important element in each

person’s cultural repertoire: their tacit definition of homosexuality. A person’s tacit definition of

homosexuality is a different concept than their explicit cognitive beliefs about what

homosexuality is. When informants described the extent to which they thought homosexuality

was a lifestyle choice, or the extent to which they thought it was caused by biological or

environmental factors, those expressions are not necessarily the same as their tacit definition of

homosexuality, though they probably tried to express themselves in ways that made sense to

them, given their tacit definition of homosexuality.

Whether or not a person believes that homosexuality is due to natural, social, or

behavioral causes, or any combination thereof, is a different issue from the implicit, common-

sense understanding of homosexuality that people use in order to make sense of issues like same-

sex marriage. These tacit definitions of homosexuality are partially composed of cognitive

beliefs, but they are also composed of attitudes, moral values, and life experiences that have
   320  
unfolded in a particular social context. People’s tacit definitions of homosexuality are the

culturally-constructed meanings that homosexuality has for them, given the lifeworld that

constitutes the terrain of their social reality. As such, they are shaped by both the person’s cohort

location and their social location within that cohort.

Informants always implicitly used their tacit definitions of homosexuality in order to talk

about same-sex marriage, and the contrast between two tacit definitions is particularly important

in shaping their discourse: homosexuality as collective identity and homosexuality as social

behavior. The tacit definition of homosexuality as collective identity is a taken-for-granted

understanding that gays and lesbians constitute a status group, in which a person’s sexual

orientation is both inherent in a person and constitutive of their collective identity. By contrast,

the tacit definition of homosexuality as social behavior is a taken-for-granted assumption that

gays and lesbians are not a coherent status group, but rather isolated individuals that engage in

sexual practices with members of the same sex.

While these two tacit definitions vaguely resemble the “born gay” and “lifestyle” beliefs,

people’s specific cognitive beliefs do not necessarily correspond to their tacit definitions. Simply

because a person said that they thought homosexuality is a lifestyle choice did not preclude them

from understanding homosexuality as a collective identity. This paradox can be seen in Jeremy’s

response to my questions about gays and lesbians who abstain from homosexual relations:

Q: Do you think people choose to be gay?

R: I really do. Like I said, everyone’s got a choice. Everyone’s got a choice, and yeah,

you could say your circumstances, heredity, or whatever, but everyone’s got a choice,

and that’s the great thing about being human…

   321  
Q: What do you think about people who identify as gay or lesbian but choose not to have

homosexual relations?

R: Well that doesn’t really make any sense. I mean, in a way, it’d be like, “well, I like

women, but I don’t want to get down”… I think that’s kind of denying yourself in a way.

I don’t think that’s the right thing to do either. (Jesse, age 27)

The idea that someone who is gay or lesbian would choose to abstain from homosexuality struck

Jesse as an act of denying one’s self. Thus, even though he argued that homosexuality is a

choice, he also imagined a person’s sexual orientation to be a part of their identity.

Similarly, just because a person said they thought that a person’s sexual orientation is

shaped by biological or genetic factors does not mean that they thought of homosexuality as

constitutive of their identity. For example, several people wondered if homosexuality was like a

birth defect or disability, something that people were born with because nature made a mistake.

In the quote below, the description of homosexuality as being about “control” and the “strange

biological things” that happen to people’s bodies indicates that Maria is thinking of

homosexuality in behavioral terms, even though it is possible that people are born gay:

Q: Do you think that people choose to be gay? Do you think homosexuality is a choice?

A: You know, I don’t have an opinion on that. I just thought, I tend to think that it is

more of a control, only because my own belief and not really caring…. When it comes

down to it, I don’t really worry about it because it isn’t my place and it’s not the way I

am, so I just, so I don’t spend a lot of time thinking about it.

Q: Sure. Do you think it is possible that people are born gay?

   322  
A: It’s possible, yeah. It is possible. I mean, there are some very strange biological things

with people’s bodies. (Maria, age 45)

As this quote illustrates, simply because people are born a certain way does not mean that a

collective identity or status group will automatically result. For people who think of

homosexuality as social behavior, homosexuality is like binge drinking or gambling: people may

be born with inclinations or tendencies to these behaviors, but that does not therefore constitute a

collective identity or status group.

I argue that people’s tacit definitions of homosexuality, and their use in discourse, reflect

the influences of both their cohort membership and their social location within that cohort. The

historical period that began after 1985-87, in which attitudes towards gays and lesbians began to

liberalize rapidly and in which homosexuality increasingly became culturally constructed as

normal and as a marker of status group membership, altered the tacit definition of homosexuality

that young people, coming of age during this period, developed in order to make sense of the

social world. This changing construction of homosexuality in mainstream American culture

during this period has been integral to the process of attitude change for older Americans, and it

has caused young people to develop this tacit definition of homosexuality to begin with because

they came of age in a society in which homosexuality was constructed in this way. This tacit

definition, however, is not just shaped by cohort membership; it is also consistent with politically

liberal beliefs that support the extension of equal rights to minority groups who have been

historically discriminated against.

Similarly, the tacit definition of homosexuality as social behavior is consistent with

conservative religious and political beliefs and was the dominant understanding of
   323  
homosexuality when the parents in this study came of age. Prior to the gay rights period in

American politics and culture, homosexuality was constructed by both mainstream “straight”

culture and by the gay liberation movement as an alternative sexuality, fundamentally different

from heterosexuality. While gay liberationists rejected the view that homosexuality was a mental

illness or a deviant lifestyle, they also did not seek to conform to heteronormative expectations

and institutions, like marriage. Thus, older Americans grew up in a society in which

homosexuality was constructed as a lifestyle and sexual orientation that was fundamentally

different, and this view is consistent with conservative religious teachings that homosexuality is

a deviant lifestyle choice.

In mainstream American culture and society, the tacit definition of homosexuality as

collective identity lends itself more easily to support for same-sex marriage than does the tacit

definition of homosexuality as social behavior. To the extent that homosexuality is understood in

a manner akin to ethnicity, there is no logical justification for denying gays and lesbians equality

under the law. Likewise, the tacit definition of homosexuality as social behavior is more easily

reconciled with religious and moral injunctions against homosexuality than is the tacit definition

of homosexuality as collective identity. Thus, for young conservative Christians who have a tacit

definition of homosexuality as collective identity, the religious teachings that homosexuality is

immoral behavior suggests an opposite course of action regarding same-sex marriage than does

their understanding of homosexuality. Likewise, older liberals who understand homosexuality to

be a social behavior but also identify politically with advocates of equal rights to minority groups

may have difficulty understanding why gays and lesbians should be included as one of these

groups. The different discourses used by these four different groups to talk about same-sex
   324  
marriage show the ways that social generational change shapes the structures and use of people’s

cultural repertoires.

What Shapes a Tacit Definition?

To say that a person’s tacit definition of homosexuality is shaped by their cohort

membership and political and religious ideologies is true to some extent, but it is imprecise.

Another, perhaps more precise, way to account for different tacit definitions of homosexuality is

that they depend upon the extent of contact with and acceptance of the cultural construction of

homosexuality that is dominant during a given period. Religious and political ideologies capture

a large amount of the variation in contact with and acceptance of a given cultural construction of

homosexuality in the analysis above because of the consistency of different tacit definitions of

homosexuality with politically liberal or religious conservative ideologies. However, there

remain aberrant cases in the analysis, such as young religious conservatives and older liberals

whose discourses do not appear to have been influenced by their cohort membership at all. I have

so far ignored these individuals in the analysis.

In this section, I consider other factors that shape the extent of a person’s contact with

and acceptance of the dominant cultural construction of homosexuality. There is most likely

significant variation in the encounter and acceptance of a particular cultural construction of

homosexuality based on variables such as educational attainment and area of residence; but

based on the interview data, it appears that three factors significantly shape the likelihood of

developing a tacit definition of homosexuality as collective identity. These factors are the nature

of personal contact and life experiences with gays and lesbians, the nature of exposure to gays

and lesbians via mass media, and the ways in which people in one’s social networks talk about
   325  
homosexuality. In a sense, these three factors are closely entwined because they represent three

of the main ways in which people learn about the social world; but there are important

differences as well.

What is at stake in complicating the analysis in this way is the fundamental meaning of

social generational change. I describe how these factors improve upon our understanding of how

people use different tacit definitions of homosexuality because it shows that social generational

change is fundamentally about the differences in the lifeworlds that constitute a person’s social

reality. In response to changing social structures, the structures of people cultural repertoires and

the ways that people use elements of their cultural repertoires also change. To the extent that “the

intersection of cohort and social location” is a good marker for denoting social generational

change, it is because the phrase has a flexible and complex meaning. “Social location” is itself

the intersection of a potentially infinite number of variables. What social generational change

really connotes is that, in some real sense, a person’s definition of social reality is different,

partially through the influence of time, such that the structure and use of their cultural repertoires

for making sense of the world is also different. Incorporating personal contact, media, and social

networks into this analysis shows more clearly the complexity that is inherent in the social

generation concept.

First, the nature of personal contact and life experiences with gays and lesbians matter

because it represents the individual’s most intimate, direct experience with homosexuality and

thus an important source of knowledge about what homosexuality means. As described in

Chapter 4, I emphasize the qualitative nature of personal contact because, while the amount of

personal contact with gays and lesbians affects attitudes, the effect of that contact on a person’s

attitudes and beliefs depends on the nature of the contact. For many people, contact with gays
   326  
and lesbians seems to have deepened their negative attitudes and evaluations of homosexuality.

In fact, some of the most negative attitudes among people in my sample were expressed by

people with some of the most direct personal contact with homosexuality. For example, Taylor, a

student who has known a lot of gays and lesbians through jobs he has held and through his

interest in theater, recalled several incidents at work where a gay coworker made him

uncomfortable:

One time, because somehow we got on the topic of what kind of girl I like, and then for

some reason, he told me what kind of guy he liked.... And everything he said was kind of

what I looked like that day, and I was like, “You’re sick.” And he was like 40-something,

too. He wasn’t like a young gay guy, he was aged. And I was like, “Blah.” (Taylor, age

20)

Other individuals with negative attitudes toward gays and lesbians and who strongly opposed

same-sex marriage were people who admitted that they had homosexual feelings or homosexual

experiences in the past, and who used it as evidence that homosexuality is wrong:

I’ve lived both sides of the coin. I’ve been a homosexual, I’ve been bisexual. I’ve had to

abstain from, and I know that it’s possible to abstain. I know it’s a choice. Whereas there

are some people out there that would like to tell you otherwise. I mean, they’re delusional

when they think that way, to tell you the truth. Because you’re not born homosexual,

you’re not born lesbian. It’s a choice you make. (George, age 50)

Today, George is happily married and opposes homosexuality and same-sex marriage strongly

from a religious point of view. As this quote shows, the fact that he has experienced
   327  
homosexuality seems to have increased his conviction that homosexuality is simply an immoral

behavior.

Similarly, positive personal contact with gays and lesbians can help explain one aberrant

case of a student who is a conservative Christian but who is supportive of same-sex marriage.

Alyssa recently converted to Christianity and is involved with her chapter of Intervarsity

Christian Fellowship (IVC), but her extensive personal contact with gays and lesbians through

school and family has caused her to disagree with the conservative Christian teachings on

homosexuality. In particular, she cites her feelings about her grandmother, who is a lesbian, as

the main reason that she disagrees with the other Christian students:

Like all of my Christian friends think that it’s sinful—homosexuality in any way. A

marriage should be between man and woman. But that’s not where I came from. Like I’m

coming from my mother’s parents—my grandma’s actually lesbian. And I’ve known for

a long time…. And she is like, I just look at her and I think she’s the most wonderful

person… She’s the BEST grandma in the world. But like, just the fact of her sexual

orientation means nothing to me. That totally, and in my eyes, when I think about it,

spiritually, Biblically, you know, I believe God created us in certain ways for a reason

and that there still is, like he still loves you no matter what. (Alyssa, age 21)

In this case, the close relationship Alyssa has with her grandmother has caused her to disagree

with her Christian friends’ views on homosexuality and same-sex marriage. Alyssa is unusual

because she has actually chosen to disagree with the religious doctrine. She explained that she

thought the Biblical teachings about homosexuality did not really refer to gay and lesbian people
   328  
in committed relationships. As a result of her disagreement with other students in IVC, she says

that she just doesn’t talk about the issue with them anymore.

Second, the amount of personal contact with gays and lesbians matters less than might be

commonly thought because even people who have no direct contact with gays and lesbians

develop empathy through exposure to gays and lesbians in mass media. Many people develop

deep emotional attachments to celebrities and characters that they encounter in mass media, and

when those celebrities are gay and lesbian, like Ellen Degeneres and Elton John, they can have

positive effects on attitudes and can shape people’s understanding of homosexuality. There is

evidence that the liberalizing effect of personal contact with gays and lesbians works through

television shows like Will and Grace and through personal narratives in which people come to

identify with the protagonists (Ghoshal 2009; Polletta 2006; Schiappa, Gregg and Hewes 2006;

Slater and Rouner 2002). For example, Lindsay, a 45 year-old nurse, describes herself as a “big

Elton John fan,” and when I brought up the issue of same-sex marriage, the first words out of her

mouth were, “[Elton] John married his sweetheart.” Even though she knows gays and lesbians

personally, her identification with Elton John was the first thing that she talked about when I

mentioned same-sex marriage. Similarly, Bethany, a 22 year-old conservative Christian, brought

up the show, Will and Grace, when I asked about gays and lesbians in the media, and she

described how that was her only contact with gays and lesbians before she went to college:

I think Will and Grace portrayed it, you know, did they have some deep episodes? Yeah.

I wasn’t a faithful watcher, but as someone who doesn’t have like, growing up in high

school, I mean that show was like a popular show, and seeing it and that having be my

only connection with the homosexuality world, I think it portrayed it in a really light-

hearted, not a serious [way]. (Bethany, age 22)

   329  

Even though Bethany thought the portrayal of homosexuality in Will and Grace was light-

hearted, it did expose Bethany to homosexuality when she otherwise had no contact with gays

and lesbians.

Third, the social networks in which people are immersed shape the likelihood that people

will encounter or accept the dominant cultural construction of homosexuality. Just as contact

with gays and lesbians shapes our perceptions of homosexuality, so too does contact with

heterosexuals have the power to influence our perceptions of the generalized homosexual other.

Being embedded in homogeneous social networks in which all people share the view that

homosexuality is a deviant social behavior makes it unlikely that they would accept the dominant

cultural construction of homosexuality as collective identity, while being embedded in social

networks that include secular liberals makes it more likely that a person would accept that

construction.

A comparison between the discourses of a mother and two children illustrates the

complex ways in which social networks, media, and personal contact with gays and lesbians

interact to shape people’s understanding of homosexuality and their discourse about same-sex

marriage. Theresa, a 55 year-old evangelical Christian, described homosexuality as “the greatest

deception of all time,” and is convinced that homosexuality is a manifestation of evil in the

world. Despite having a family member who identifies as gay, having a sister who is a secular

liberal, and enjoying mainstream television shows, she rejects homosexuality because of her

religious beliefs and personal experiences:

I think Satan is so cunning, baffling, and powerful that um, these people who tend to go

to that belief system are very likeable and loveable people. They’re, they really are good
   330  
people. It’s, to me, very sad and tragic, that they’ve been so greatly deceived. Ellen

Degeneres is one. I loved her show. (Theresa, age 55)

The very fact of contact with mainstream cultural construction of homosexuality is not enough;

she interprets it as a sign that American society has lost it’s Christian moral compass. She has

raised her children to be Christian as well, and her daughter, Renee, shares many of her views.

Although her opposition to homosexuality and same-sex marriage is not as strong as her

mother’s, she has learned to interpret homosexuality in ways that are in accord with her religious

upbringing, her friends from church, and her friends from the Christian school that she attended.

Her social network is so dominated by evangelical Christians, that she views the gays and

lesbians that she knows from that perspective:

I don’t have anything against them as a pers—Like, I can be friends and love someone

when I don’t support their way of life-style. Cause one of my cousins is gay, but I love

him to death. I just don’t support his way of life. (Renee, age 19)

Although everyone in the family knows that the cousin is gay, Renee reports that no one talks

about it. Far from taking homosexuality for granted as inherent to a person’s identity, she has

adopted the religious worldview that her mother and peers share and disapproves of the sin even

while she loves the sinner.

By contrast, Renee’s brother, Nick, struggled to answer my questions about same-sex

marriage because he is torn between two worlds. On one hand, when he was younger, he rejected

his mother’s religious identity, attended a public high school, and was thoroughly embedded in

the mainstream American culture shared by his peers. The way that he talks about his gay cousin
   331  
illustrates how his tacit definition of homosexuality is as collective identity and that it developed

when he lived in this secular world. When asked about how he felt when his cousin came out, he

said:

R: I just kind of knew. I don’t know how I felt. Um, I guess I didn’t really feel different

about it anyway. I kind of knew it was true, and just like, he’s my cousin, I love him

anyways, you know, type deal.

Q: Yeah. Were you close with your cousin growing up? Like did you spend time

together?

R: Um yeah, you know, we, there was always the family parties, and he would take us

out to movies and stuff, and he’s a blast to hang out with. (Nick, age 21)

Nick thinks highly of his cousin and recalls no change in his feelings when he learned he is gay.

Because of him, Nick thinks that people are simply born gay, and that it is just who they are:

My cousin even said, he’s like, “You know, why would I ever choose to be this way?”

Um, and so I do think, I do think some people are probably born to be, um, attracted to

men…. If someone were to say, if I were to say, I don’t want to be straight anymore, I

don’t think I could make myself like men… I’ve heard of people, um, going from being

gay to um, you know, they call themselves recovering homosexuals… but uh, I don’t see

how that could be possible. (Nick, age 21)

In these two quotes, his discourse is identical to the supportive discourses used by young secular

liberals to talk about same-sex marriage.

   332  
However, despite this tacit understanding of homosexuality, he has recently embraced

Christianity and is now learning what the Bible teaches about issues like homosexuality. He

rejected his old secular life at the encouragement of his family because of trouble he got into as a

teenager, and he has now surrounded himself with conservative Christians. In responding to my

questions, it was clear that he was giving the conservative Christian answers, even though he

didn’t really believe them:

Q: Do you personally have an opinion on the issue?

R: Um, yeah. It’s hard, I mean, if I didn’t have um, you know, my religious beliefs, I

would say I feel like, you know what, if that makes them happy, then do it. And I had a

cousin who’s gay. And it’s hard, ‘cause then it comes back to the question, you know, of,

is it like a choice or is it something that they’re born with? You know, I don’t want to

sound like a closed-minded, you know, conservative, like you know, “All gays go to hell”

or whatever, anything like that. But I got, I do think it’s meant for a man and woman, and

so I don’t agree with it. But it’s hard for me not to agree with it, but I just, I don’t agree

with it just cause it’s, you know, it’s in The Word, and I’ve just kind of got to go with it.

(Nick, age 21)

In this answer, he describes how he feels, but then describes how he has to give the answer that

is consistent with conservative Christian dogma, as if he is being coerced by religious authority.

In another exchange, he gives a similar response, as if he is under duress; and when I provide

him with a chance to challenge that authority, he refuses:

Q: Do you think that homosexuality is immoral?

R: Yeah. And that’s one of those things that’s hard for me to say, too.
   333  
Q: Yeah. Do you think that like, do you think that the Bible could be wrong? I mean, not

like the big picture I mean, but like about this one particular issue? Do you know what I

mean?

R: Yeah. Um, no, I think it’s divinely written, so I don’t think so. (Nick, age 21)

Nick gives the standard conservative Christian answers to my questions, despite his deep-seated

feelings that he doesn’t really believe it. Because he has surrounded himself with conservative

Christians—his family, his girlfriend, his roommates, his church—he appears to have no support

to express the beliefs that originated from his former life in secular society.

The conflict in Nick’s cultural repertoire—between his tacit definition of homosexuality

that he retains from his former, secular life and his current views that he has learned from his

religious friends and family members—is unusually prominent in his discourse. But such

conflicts are frequently observed in the discourses of younger religious conservatives and older

liberals. The comparison between Nick and his sister, Renee, shows how even people who share

the same cohort and social location can adopt different discourses about same-sex marriage

because of how social networks, mass media, and contact with gays and lesbians have shaped

their cultural repertoires. To define social generations intersectionally by cohort and social

location is thus a simple indicator for what is, in reality, a complex, multidimensional

intersection that constitutes a person’s lifeworld.

Conclusion

In this chapter, I directly examined evidence that cohort differences in discourses about

same-sex marriage are due to social generational processes, the cultural and social psychological
   334  
processes that are widely thought to account for cohort effects in attitudes. First, I showed that

the discourses of students and parents show evidence that their age-specific encounters with the

same historical period of social change had different effects on the two cohorts. Parents, who had

come of age prior to the gay rights period in American politics and culture, reported lingering

negative attitudes towards homosexuality that they attributed to the social environment they grew

up in, and they also constructed narratives of attitude change that show how the increasing

openness and acceptance of gays and lesbians in society encouraged them to adopt more tolerant

views of homosexuality. By contrast, students talked about their experiences with gays and

lesbians in high school as though openness and acceptance of gays and lesbians were the social

norm, and their narratives of attitude change tended to reflect the process of maturation rather

than the rejection of previously formed attitudes about homosexuality.

I also demonstrated in a second way that the data support a social generational

interpretation of cohort differences because, even controlling for generational and ideological

influences, students were more likely to be supportive of same-sex marriage than their parents.

Even restricting the comparisons only to students and parents who agree with each other on

issues related to religion, politics, marriage, and sexuality, I found significant differences in their

discourses about same-sex marriage. By making comparisons simultaneously between and

within cohorts, I showed how the discourses of young religious conservatives and older liberals

are both similar to and different from others in their cohort.

Finally, I offered an interpretation of these results that is consistent with social

generational theory. I argued that a person’s cohort location and their social location within that

cohort shaped both the structure of their cultural repertoire and the ways that they used elements

of their cultural repertoire to talk about same-sex marriage. Middle-ground discourses, for
   335  
example, represent the creative ways that individuals combine and reconcile elements of their

cultural repertoire that seem to be at odds with each other in their implications for the question of

same-sex marriage. Of particular importance for the same-sex marriage debate is a person’s tacit

definition of homosexuality. I showed how a person’s tacit definition of homosexuality is related

to both their cohort location and their political ideologies, and I complicated the analysis by

showing how social networks, media exposure, and personal contact with gays and lesbians also

contribute to shaping a person’s tacit definition of homosexuality. This analysis is intended to

show that the intersection of cohort and social location is not necessarily two-dimensional, but

that it is in reality complex and multi-dimensional. What is crucial to understand about social

generational processes is that they denote differences in people’s lifeworlds, marked by the

intersection of cohort and social location, such that people develop a cultural repertoire that is

structured and used to help them make sense of that world.

   336  
Conclusion

Social Generational Change

Individuals are products of their social contexts, according to the fundamental principles

of sociology. We know that our social contexts are defined by our relations to other people,

groups, structures, and institutions; but they are also spatially and temporally located. We cannot

help but be inhabitants of physical bodies confined by the limitations of the ability of that body

to move through the dimensions of space and time. As agents, our thoughts, speech, and actions

bear the imprint of our social, spatial, and temporal contexts as well.

That these observations are so obvious and so foundational to the work that sociologists

do makes it all the more notable that concern with social generational theory has been so

conspicuously absent from sustained scholarly interest in the last 25 years. While the theoretical

relationship between a person’s location in historical time and their agency has long been a vital,

taken-for-granted proposition upon which much research on cohort succession, collective

memory, aging and the life course, social psychology, and politics, is premised; empirical

research on social generational change never lived up to its theoretical promise. As a result of

lackluster empirical findings and conceptual confusion about the meaning of “generation,”

scholars channeled research interest on social generational processes into more productive areas.

Since the mid-1980s, however, the cultural turn in the social sciences provided new

theoretical perspectives and empirical tools for scholars to use to analyze the relationship

between a person’s temporal location and their agency. A handful of theoreticians have argued in

favor of revising Karl Mannheim’s generational theory in the language of cultural sociology

(Corsten 1999; Esler 1984; Eyerman and Turner 1998; Gilleard 2004; Pilcher 1994), but
   337  
empirical support for a revitalized social generation concept has remained minimal. It is this gap

between the theoretical promise of the social generation concept and its analytic utility that I

have aimed to fill.

I have argued that, when properly defined and operationalized, the social generation

concept can help explain how cohort effects in attitudes arise and how processes of social change

and social reproduction occur. I define the social generation as the cultural and social

psychological process through which groups of people, defined intersectionally by cohort and

social location, encounter a particular configuration of social structures and in turn, typically in

young adulthood, develop a particular cultural repertoire that they use in the further elaboration

of attitudes and actions. The controversy surrounding same-sex marriage has provided an ideal

case study for operationalizing this definition and examining the analytic utility of the concept.

First, this study has illustrated the inherent intersectionality of social generational change.

This intersectionality has been present in generational theory since Mannheim’s (1952 (1928))

description of the “generation as an actuality,” but the failure to operationalize it has led to the

persistent problems of generational analysis: the tendency for competing variables to

overshadow the cohort variable with greater explanatory power and the perpetuation of

stereotypes of whole cohorts based on some subset of that cohort. In the case of same-sex

marriage, it is the intersection of cohort with political and religious ideology that satisfactorily

accounts for the variation in discourses. While a simple cohort comparison does show that young

people are more likely to be supportive of same-sex marriage, we gain a fuller account of how

cohort membership matters only when we examine the ways that cohort and political and

religious ideologies have either reinforcing or cross-cutting influences on the discourse. Only by
   338  
making comparisons simultaneously across and within cohorts do we show that cohort

membership affects discourses differently for religious conservatives than for political liberals.

Moreover, this case study has deepened our understanding of the intersectionality of

social generational change by showing that the intersection of time and social location is not

two-dimensional, but multi-dimensional. The intersectionality is itself intersectional. Because the

effects of political and religious ideologies on discourses about same-sex marriage are so strong,

ideology provides a parsimonious account of how social location matters. But a person’s social

networks, their personal contact with gays and lesbians, and their exposure to gays and lesbians

in the mass media also influence the tacit definition of homosexuality that a person brings to bear

on the question of same-sex marriage. I argue that the intersectionality of social generational

change really refers to how time and social location constitute a person’s lifeworld, making their

own social reality effectively different from those of others. That a person who inhabits a

different social reality would develop a different cultural repertoire that they use in different

ways to make sense of that reality logically follows.

This raises the second way that the case study of same-sex marriage has demonstrated the

analytic utility of the social generation concept. By denoting the social generation as a cultural

and social psychological process, I distinguish the social generation from the cohort and from the

generation, and I focus our analytic attention on the macro-micro relationship that has long been

assumed to be crucial to cohort and generational phenomena: the ways that a person’s encounter

with macro-level social structures shapes their micro-level orientations, attitudes, and behaviors.

The case of same-sex marriage demonstrates that the social generational process has to do with

the development and use of cultural repertoires. Social generational change refers to how the
   339  
structure of people’s cultural repertoires and the ways that people use elements of their cultural

repertoires change according to their temporal and social locations.

The case study illustrates that different groups, defined by cohort and by political and

religious ideologies, develop different cognitive beliefs, affective attitudes, moral values, and life

experiences regarding homosexuality and marriage, and that they combine them in different

ways in an effort to express themselves. Numerous distinct discourses emerged as a result of the

variation in informants’ repertoires and the strategies and techniques that they employed as they

attempted to articulate their views. The patterns of discourses were shaped by the relationships of

various elements with one another (e.g. between the attitudes and personal contact with gays and

lesbians) and by the implications of various elements for the logic of supporting or opposing

same-sex marriage (e.g. believing that homosexuality is immoral and believing in equal rights

for gays and lesbians). The different discourses were the results of the creative ways that

informants combined and reconciled different elements in their cultural repertoires.

The word “repertoire,” and the fact that participants used different strategies and

techniques in various ways to combine and reconcile elements in their cultural repertoires,

highlights the agency of ordinary people in social generational processes. Ann Swidler (2001)

has developed the concept of cultural repertoires precisely because of the importance of the

agency of individuals in the analysis of how cultural meaning is created through everyday talk

and action. We cannot understand the sociological significance of culture without analyzing the

ways that people use cultural elements in particular situations for particular ends. This lesson

applies to social generational change as well: social generational change only happens because of

the speech and actions of people who combine and reconcile elements in their cultural repertoire

in an effort to express their views of the social world.

   340  
This view of the agency of cohorts in social generational processes is at odds with the

ways that cohorts and social change have historically been understood. Typically, cohorts have

been portrayed as passive, unwitting vehicles of the inexorable forces of history. Social

generational change happens to people, rather than people making social generational change

happen. This view is represented in one of the classic essays on cohort replacement and social

change: “Each fresh cohort is a possible intermediary in the transformation process, a vehicle for

introducing new postures. The new cohorts provide the opportunity for social change to occur.

They do not cause change; they permit it” (Ryder 1965, p. 844). Such an unequivocal denial of

agency to a social group is striking by the standards of contemporary sociological theory; yet it

captures the essence of how social generational processes have typically been imagined. The

empirical evidence presented in this dissertation contradicts this popular assessment: it took real

cognitive labor for young religious conservatives to express their support for equal rights for

gays and lesbians while still insisting on the immorality of homosexuality. Cohorts coming of

age in a new historical period do make social generational change, but only in relation to the

actions of others.

Insisting on the agency of cohorts in processes of social generational change therefore

means that there are at least two distinct groups of agents operating in two different times that

are required to produce social generational change: there are the agents who initially brought

about the historic period of social change in question, and there are the agents who develop and

use cultural repertoires in new ways after the social changes have become institutionalized. In

the case of same-sex marriage, the first group of agents includes the women’s liberation

movement, the gay liberation movement, and others who helped to bring about the broad

changes in the structures and meanings of gender, sexuality, marriage, and family in American
   341  
society after 1969. The second group of agents is the students in this study who have come to

take these changes for granted and who have developed new ways of talking and acting in

society as a result. It is reasonable to say that change happened to this group of students to the

extent that they did not cause the changes in the social structures that they grew up with, but they

do have agency in social generational processes of shaping the ultimate meaning and

significance of those changes because of how they interpret their social reality. The existence of

these two groups of agents and how each contributes to social generational change can be

observed in other cases as well, such as in contemporary struggles over the meanings of

feminism (Schnittker, Freese and Powell 2003).

Finally, this case study has shown how the theory and research methods used to study

social generational change must be relational. At the simplest methodological level, the analysis

of social generational change requires many levels of comparison. Comparisons must be made

simultaneously between and within cohorts, between social structures at different points in time,

between cultural repertoires (or elements within cultural repertoires), and between the ways in

which social structures shape the cultural repertoires of different cohorts. All of these

comparisons have been necessary to provide the account above regarding how social

generational change is occurring in discourses and attitudes about same-sex marriage.

Additionally, my methodological approach has been relational in that I have made no

assumptions that people’s responses to my interviews represent true opinions or attitudes in any

absolute sense; rather, the data gathered from one individual gain their significance only in

relation to the data gathered from other individuals, as evidence of similarities and differences in

the cultural elements and strategies that people use in response to a question or situation.
   342  
At a deeper theoretical level, however, the analysis of social generational change must be

relational because the very significance of the concept implies a state of being that is different

from another. Both theoretically and normatively, the social generation concept is important only

when juxtaposing two periods of time, two sets of social structures, or two sets of attitudes,

behaviors, and dispositions. Our understanding of the significance of one can only be made with

reference to the other. Additionally, the social generation concept is inherently relational because

it is intended to capture the macro-micro relationship between social structures and cultural

repertoires. Particular elements of people’s cultural repertoires or particular behaviors are never

simply that; they are also manifestations of how the lifeworlds in which we live shape the ways

that we think and talk about the world; and they are further manifestations of how the actions of

agents in turn recreate and modify social reality. The analysis of social generational change must

always be the analysis of this complex set of relationships between the micro and macro levels,

between individuals’ meaning-making activities and the historically- and socially-determined

structures in which they reside.

In the empirical data of this study, this relationality can be seen in how I have interpreted

the interview data in the context of broad historical changes in the cultural construction of

homosexuality, the openness of gays and lesbians about their sexual orientations, the changing

strategies of LGBTQ activism, and the increasingly tolerant attitudes regarding homosexuality.

Individuals’ comments about gays and lesbians that they know, or experiences they have had, or

values that they hold are never just personal “troubles,” as C. Wright Mills (2000) might say; the

researcher must use his or her sociological imagination to analyze the individual’s comment in

relation to the various social contexts that have shaped its utterance.
   343  
With regard to the limitations of the social generational interpretation and future research

on the social generation concept, it would be difficult to say too much. While I have argued that

there is strong theoretical support for the social generation concept as I have developed it here

and that it can help scholars understand the relationship between cohort effects and social

change, it is still only a concept. I have by no means proven that social generational change

causes anything, nor have I weighed it against evidence for competing explanations. Moreover,

this is an exploratory study in the sense that identification of the social generation concept is the

first step to studying it. What the exact mechanisms are by which social generational processes

work and how social generational change might happen differently with respect to other topics

are obvious subjects that warrant investigation, but it is likely that the most fruitful areas for

future research on social generational change are those that are not readily apparent from

consideration of this one case study alone.

Rather than pretending to well-defined research agenda for the social generation concept,

I want to address the subject of future research by instead offering four brief comments about the

nature of the concept and its place in the discipline of sociology. First, because the social

generation concept refers to a process rather than a group of people, the question of the

boundaries of a social generation is not easily answered. One can debate the temporal

boundaries of cohorts, and one can debate the temporal and social boundaries of cohort

subgroups who bear the marks of social generational change; but debates about the boundaries of

the social generation is a much more challenging and profound question than is probably the

intent of such a debate because it refers to a process rather than a group of people. In terms of the

boundaries of cohort subgroups, such as the groups of people described in this dissertation, the

boundaries will depend upon the nature of the social change or the issue of interest. The temporal
   344  
and social boundaries of groups who bear the mark of social generational change regarding

homosexuality and same-sex marriage would be different from the temporal and social

boundaries of groups who bear the mark of social generational change in racial attitudes. These

are two different issues, with two different historical timelines, and that affect two different sets

of groups in two different varieties of ways; thus, the boundaries of subgroups affected by social

generational change will be defined differently for each issue or each way that society changes.

Second, simply because I have defined the social generation concept as a cultural and

social psychological process does not mean that only social psychologists or cultural sociologists

will find this concept useful. To the contrary, I view culture and social psychology as being

inherent aspects of social life in a way that is analogous to a concept like “social structure.” To

define the concept as cultural and social psychological is irrelevant to the existing ways in which

sociologists draw intra-disciplinary boundaries, although it does require that we take social

psychological and cultural phenomena seriously—such as the structures of belief systems and the

sources of knowledge that people use to construct social reality “in their everyday, non- or pre-

theoretical lives” (Berger and Luckmann 1980, p. 15). For example, the social generation

concept presupposes an internally-differentiated belief system, because experiencing social

generational change in the same way does not mean that everyone will react the same way or

agree with each other about that experience. Other concepts, like the “lifeworld” (Habermas

1987) or the “collective consciousness” (Durkheim 1984) may also prove useful for theorizing

and operationalizing the social generation concept because of their mutual affinities in scope,

level of abstraction, and connotations. Beyond valuing the insights of cultural sociology and

social psychology, however, the social generation concept should transcend intra- and inter-

disciplinary boundaries.
   345  
Likewise, and third, the social generation concept depends upon the work of scholars

studying age, cohort, and period effects, generations, and aging and the life course. The social

generation concept is not a competing concept; rather, its very value depends upon its relations

with these other concepts and processes. Research in aging, attitude stability over the life course,

the impressionable-years hypothesis, cohort analysis, generational relations, and the effects of

historical change are all absolutely integral to theoretically and empirically valid research on

social generational change. Just as scholars must draw on the insights of cultural sociology and

social psychology, so too must the contributions of scholars working in these areas be

incorporated into the research done on the social generation concept.

Lastly, I want to say a brief word about the warrant for further research on social

generations. The theoretical warrant of the social generation concept, as I have argued, is that it

is essential to our understanding of how societies reproduce themselves over time, in both the

ways that they change and the ways that they stay the same. As a new analytical tool to add to

our tool-kits, the social generation concept can help scholars gain a more complete understanding

of how cohort replacement acts as a mechanism of social reproduction and social change.

Nothing more needs to be said about this.

But there is another warrant as well: its normative warrant. The discipline of sociology

has a long and proud tradition of debunking stereotypes and calling common-sense knowledge

into question by exposing the variation within and among groups of people. After research on the

broad generation concept disappeared from the discipline in the mid-1980s, the stereotypes that

were produced by intellectuals and academics under the guise of “generational research”—about

“Generation X” or “The Millenial Generation,” for example—went largely unchallenged by the

group of scholars who are best trained and positioned to do so. Empirical research and theorizing
   346  
on social generational change provides sociologists with the means and the standing to rejoin the

broader public conversation about this subject. From that position, sociologists can again fulfill

one of its highest public callings by exposing and rebutting the hidden, and occasionally harmful,

stereotypes that are made about the young and old alike.

The Foundations of Attitudes about Same-Sex Marriage

What do we learn about the controversy surrounding same-sex marriage from this

dissertation? By combining quantitative and qualitative analyses, I have endeavored to describe

the social and cultural foundations of attitudes about same-sex marriage in the greatest detail

possible. Only with a deep understanding of the ways in which Americans talk about same-sex

marriage and how the issue of same-sex marriage makes sense to them within the horizons of

their lifeworlds can we fully understand the dynamics of public opinion about this issue.

First, this dissertation confirms the findings of previous quantitative and qualitative

studies by showing that people’s affective attitudes and moral values regarding homosexuality

are the most powerful predictors of support for and opposition to same-sex marriage. To the

extent that many demographic and contextual variables, like educational attainment, affect

attitudes about same-sex marriage, they do so primarily because they affect people’s attitudes

and values about homosexuality. In interviews, people even respond to questions about same-sex

marriage by changing the subject and explaining how they feel about homosexuality, and most

people attribute the controversy to disagreements about homosexuality.

However, this study also challenges previous studies about how homosexuality shapes

attitudes about same-sex marriage. I showed that people’s discourses about homosexuality are

far more complex than is suggested by quantitative analyses: not only are people’s beliefs,
   347  
attitudes, values, and experiences regarding homosexuality complex and sometimes apparently

contradictory in their own right, but they often fail to cohere ideologically into discourses of

support and opposition. In particular, the moral dimensions of the discourse are more complex

than is portrayed in the literature, and people’s personal contact with gays and lesbians and their

cognitive attributions of homosexuality do not appear to have uniform effects on attitudes. There

are, in fact, a variety of middle-ground discourses that people use to talk about homosexuality

that cannot be characterized as either supportive or oppositional. The existence of middle-ground

discourses shows that quantitative conventions of measuring public opinion in terms of support

and opposition and focusing on strong quantitative correlations causes us to overlook attitudes

and discourses that are substantively and theoretically important.

Second, this dissertation shows that people’s cultural definitions of marriage also shape

the discourses about same-sex marriage, though not to the degree that homosexuality shapes

them. The primary way that definitions of marriage shape the debate is in the conflict between

religious and heterosexual definitions of marriage on one hand and civil and companionate

definitions of marriage on the other. People who advance religious and heterosexual definitions

of marriage are more likely to oppose same-sex marriage, but such opposition is not inevitable

because the companionate definition of marriage appears to dominate people’s discourses of

what makes a good marriage in practice. In general, people use individualistic and pragmatic

language to talk about how the characteristics of a good marriage depend on the couple’s

interests and desires. It is common sense to most informants that couples should not raise

children if they don’t want them, but it is no less obvious to them that sexual attraction is

essential to the idea of marriage. The fact that people use companionate definitions of marriage

to talk about what makes a good marriage in practice means that people’s cultural understandings
   348  
of marriage are more compatible with the idea of same-sex marriage than they likely would have

been only a few decades ago.

Third, the existence of middle-ground discourses suggests that attitude polarization

regarding same-sex marriage is not inevitable. To the extent that “culture wars” emerge around

the issue of same-sex marriage, they do so only in the context of a politicized environment in

which people are encouraged to choose sides. While many of my informants said they have no

opinion on the issue, have only weak or mixed attitudes about the issue, or simply want to stay

out of the controversy, there are a variety of factors in a politicized context that impel people to

choose sides: yes/no ballot initiatives, political discourse and campaign literature from activists,

journalistic standards of media coverage that define “objectivity” as the presentation of two

opposing viewpoints, and polls and surveys that present their findings in terms of support and

opposition. In this context, it may seem that the two sides are irreconcilably and hopelessly

divided; but individuals’ discourse in a non-politicized context shows the opposite. It is not that

it is more correct to describe public opinion about same-sex marriage in one way rather than

another; rather, this dissertation simply shows that the degree of polarization depends on the

social context.

Fourth, even though civil unions appear to be a possible compromise solution to the

controversy over same-sex marriage—especially given the lack of ideological polarization in

people’s discourse—the idea of implementing a parallel institution that is legally the same except

for its name was viewed hostilely by most informants. Perhaps because of the way that the word

marriage simultaneously carries religious and secular meanings in the United States, the concept

of civil unions had very little resonance in people’s discourses about same-sex marriage, beyond

its political expedience. The real cultural significance of civil unions is instead that the term
   349  
captures a paradox in the discourse: that a large percentage of people, perhaps 10-15% of the

population, simultaneously believe that gays and lesbians should have the same civil rights as

heterosexuals, but not the right to marry. While this combination of beliefs seems contradictory

at first, it is actually merely an artifact of the fact that a number of rights, benefits, and

responsibilities relating to taxes, health care, and other issues are tied to a person’s marital status.

It is thus completely reasonable to argue that gays and lesbians should have the rights and

benefits that come from marriage without having the right to marriage.

To the extent that this dissertation has public policy implications for people looking to

resolve the controversy surrounding same-sex marriage, it lies in the implications of this

meaning of civil unions. The creation of an alternate category for the recognition of same-sex

couples has very little cultural legitimacy for liberals and moderates, especially given the

connotations of the phrase “separate but equal,” and it would not placate religious conservatives

whose primary objection to recognizing same-sex relationships in any form is its tacit

endorsement of homosexuality. However, the controversy surrounding same-sex marriage can be

undermined by pursuing a legal agenda along the lines of those advocated by Nancy Polikoff

(2008). If the rights and benefits that currently come from marriage were dissociated from

marital status and universally provided to all families, the current inequalities in rights and

benefits available to gay and lesbian couples could be rectified. At the same time, such a course

of action could actually strengthen the meaning of marriage because it would eliminate the

need—for both same-sex and opposite-sex couples—to marry for material reasons, and it would

allow local communities and churches to establish meanings and standards of marriage that are

in accord with local cultures. While I cannot assess the legal merits of such a course of action,

there may be a cultural basis for support for it that is greater than that for civil unions.
   350  

Cohorts and Same-Sex Marriage

This dissertation also sheds light on the foundations of attitudes about same-sex marriage

in that it shows that age-related differences in attitudes are most likely due to cohort effects. Both

quantitative and qualitative analyses show that attitudes are shaped by cohort membership, as

defined by the period in which a person reached adulthood, but that the variation in age within

each cohort adds no additional explanatory power to our understanding of attitudes about same-

sex marriage. Not only is there no evidence of age effects—of people becoming more likely to

oppose same-sex marriage as they get older—there is evidence of the opposite occurring: many

members of the older cohort have become more supportive of same-sex marriage because their

attitudes about homosexuality have changed.

While members of the Rights Cohort are more likely to support same-sex marriage than

members of older cohorts, this dissertation goes beyond a simple cohort comparison and

therefore shows even stronger evidence of cohort effects. By comparing the discourses of

students and parents who largely agree with one another about issues relating to religion, politics,

marriage, gender, and sexuality, I show that there is still evidence of cohort differences in

discourse about same-sex marriage. While young liberals tend to use supportive discourses to

talk about same-sex marriage and older religious conservatives use oppositional discourses,

younger religious conservatives and older liberals frequently use middle-ground discourses. In

other words, by making comparisons simultaneously between and within cohorts, and by

focusing on discourses about same-sex marriage that are not easily classified as supportive or

oppositional, we gain additional insight into how cohort membership shapes attitudes about

same-sex marriage.
   351  
I argue that these cohort differences in discourses are due to the social generational

processes that shape the structure and use of people’s cultural repertoires. Both a person’s social

location and their cohort location shape their social reality, and hence, the cultural elements that

people develop in order to make sense of that social reality. Of particular importance is the tacit

definition of homosexuality that a person uses to talk about-sex marriage, which is shaped by

their contact with and acceptance of the dominant construction of homosexuality in mainstream

American culture during the historical period in which they came of age. Members of the Rights

Cohort came of age when homosexuality was culturally constructed as a collective identity in

mainstream American society, so people who encountered and accepted this cultural construction

are more likely to think of homosexuality as constitutive of social identity and to think of gays

and lesbians as a status group, akin to a racial or ethnic group. Thinking of homosexuality in this

way is ideologically consistent with support for same-sex marriage because, legally and

culturally, gays and lesbians are considered a minority group, like African Americans, with a

history of discrimination and exclusion from the rights and benefits granted to citizens. From this

point of view, opposition to same-sex marriage is almost inexplicable and nonsensical, because

only irrational religious beliefs or prejudice would allow a person to argue against equal rights to

a disadvantaged minority group.

Likewise, members of older cohorts came of age during historical periods in which

homosexuality was culturally constructed as a social behavior or a mental illness, akin to

gambling or alcoholism, so people who encountered and accepted this cultural construction are

more likely to think of gays and lesbians as individuals who happen to engage in sexual behavior

that has generally been considered deviant and wrong throughout American history. Thinking of

homosexuality in this way is more ideologically consistent with opposition to same-sex marriage
   352  
because there is nothing preventing a gay or lesbian person from refraining from that behavior,

no matter how innate their predisposition or tendency to such behaviors may be. From this point

of view, support for same-sex marriage appears nonsensical because it would be illogical to

change the definition of marriage to accommodate a deviant behavior that should be repressed

anyway; a person can choose to marry someone of the opposite sex if they want to, or they can

choose to continue having homosexual relations.

Because of the tacit definitions of homosexuality that people developed during the period

in which they came of age, younger supporters and older opponents of same-sex marriage appear

to be talking past each other. Even if they agreed on the definition of marriage—which many do

not—supporters and opponents of same-sex marriage cannot even agree on what homosexuality

means. From this point of view, the culture wars view of same-sex marriage appears plausible:

the views of the other side appear incomprehensible within the framework of one’s own

worldview, so compromise does indeed appear “sociologically impossible” (Hunter 1991, p.

130).

However, polarization around the issue of same-sex marriage does not necessarily occur

because of the ways in which people’s cohort memberships cross-cut their political and religious

ideologies. Middle-ground discourses emerge for people whose political and religious ideologies

are more compatible with the opposite position on same-sex marriage than the one toward which

their cohort membership would incline them. Thus, older liberals, who are more likely to think of

homosexuality as social behavior because of their cohort membership, might be inclined to

oppose same-sex marriage, were it not for the liberal political ideology that advocates equal

rights and tolerance for minorities. Similarly, younger religious conservatives, who are more

likely to think of homosexuality as collective identity because of their cohort membership, might
   353  
be inclined to support same-sex marriage, were it not for their religious ideology that teaches that

homosexuality is immoral or sinful. Were same-sex marriage a culture war, these individuals

would be caught in the cross-fire; but in reality, supporters and opponents in political contests

over same-sex marriage merely attempt to persuade these individuals to join their side, by

emphasizing either the negative consequences of legalizing same-sex marriage or the moral and

legal rightness of equality.

The exceptions to this description of the intersection between cohort and political and

religious ideologies further underscore the importance of the ways in which a person’s tacit

definition of homosexuality is shaped by the dominant cultural construction of homosexuality

during each historical period. The extent of a person’s contact and acceptance of the dominant

cultural construction of homosexuality depends upon the nature of personal contact with gays

and lesbians in real life and in mass media and on the ways in which people in their social

networks talk about homosexuality. Thus, young religious conservatives will not necessarily

encounter and accept the cultural construction of homosexuality as collective identity, and older

liberals may have rejected the dominant cultural construction of homosexuality as social

behavior when they came of age because of the composition of their social networks and the

nature of their contact with homosexuality. Similarly, many older liberals changed their attitudes

about homosexuality when they encountered the new dominant construction of homosexuality

after 1985 because of their contact with homosexuality and the influences of people in their

social networks. In each case, the discourses about same-sex marriage bear evidence of the

important relationship between cohort, historical period, and social location.

The Future of Same-Sex Marriage

   354  
The implications of this analysis for the future of the same-sex marriage debate are clear.

The gradual longitudinal trend of rising support for same-sex marriage in public opinion is likely

to continue, due to social generational change, cohort replacement, and the changing attitudes of

older adults. Confidence in this prediction is bolstered by the fact that this study has uncovered

evidence of the cultural and social psychological process that is driving this liberalization of

attitudes: how contact with and acceptance of the dominant cultural construction of

homosexuality as collective identity is shaping the cultural repertoires of new cohorts. As long as

homosexuality continues to be constructed as an inherent part of a person’s identity and the basis

of status group membership, like ethnicity, more and more Americans will develop such a tacit

definition of homosexuality as they come of age, and the cultural basis for opposing same-sex

marriage will continue to weaken as older Americans with a contrary view either change their

minds or die off. Among younger cohorts, religion appears to be the only legitimate cultural

foundation for opposing same-sex marriage, but as this study has shown, many young religious

conservatives think of homosexuality as a collective identity and have positive attitudes toward

gays and lesbians.

There appears to be something of a feedback loop at work that increases support for

same-sex marriage. Associated with this cultural construction of homosexuality has been a

liberalization of attitudes, an increase in cultural and institutional support for gays and lesbians to

be open about their sexual orientation, and an increasingly sympathetic portrayal of gays and

lesbians in mass media. The more that people have positive contact with gays and lesbians in

their daily lives and in mass media, and the more that heterosexuals in people’s social networks

have positive and tolerant attitudes toward gays and lesbians, the more social and cultural
   355  
supports exist for defining homosexuality as a marker of collective identity, and the more gays

and lesbians gain political and cultural standing as a legitimate status group.

This trend of continued support for same-sex marriage only applies to public opinion

about the issue; the legalization of same-sex marriage is not inevitable, and the ultimate fate of

same-sex marriage depends upon more proximate legal and political factors. Continuing

litigation about the issue and the repeated political conflicts over the issue in different states are

more directly associated with whether or not same-sex marriage is legalized. Nevertheless, there

are relationships between public opinion and the legal and political systems that should be

underscored.

We have already witnessed the importance of public opinion in shaping the outcomes of

recent legal and political contests. In California, an amendment to the state constitution was

approved by voters in 2008 that defined marriage as between a man and woman, thus

superseding a state Supreme Court ruling that the prohibition against same-sex marriage violated

the state’s constitution. In Maine, a law allowing same-sex marriage, passed by the legislature

and signed into law by the governor, was vetoed in a 2009 ballot initiative. In each case, public

opinion prevented the legalization of same-sex marriage through the courts or the legislature.

However, the margins of victory were slim in both cases: a 2% change in public opinion in

California and a 3% change in public opinion in Maine would have been enough to defeat those

ballot initiatives. With such close contests, one cannot help but wonder how long it will be until

public opinion begins to tip the balance in favor same-sex marriage in states like Maine and

California.

At the same time, significant political or legal events can be so consequential that they

fundamentally reshape public opinion. A sudden, nationwide legalization of same-sex marriage

   356  
by the federal courts could provoke a backlash, as did Roe v. Wade with respect to abortion; or

one could imagine that repeated, polarizing political battles like the one surrounding Proposition

8 in California could alter the cultural construction of homosexuality in ways that would be

damaging to the movement for same-sex marriage. Such is the nature of cohort and period

effects: they are only temporary and they are destined to be replaced by others, whether spurred

on by dramatic events or by slow, gradual changes.

In any case, as long as the dominant cultural construction of homosexuality remains

unchanged—as long as the cultural and social foundations of attitudes about same-sex marriage

are not dramatically altered by sudden events or by the arrival of a new historical period—it

appears that current legal prohibitions against same-sex marriage will ultimately be overturned

by the slowly rising tide of support for same-sex marriage in public opinion. Just as

constitutional amendments against same-sex marriage were established by public opinion, so can

they be removed. It is certainly the case that legal prohibitions against same-sex marriage are

significant barriers that retard the movement towards its legalization; however, the mechanism of

cohort succession, driven by social generational change in the cultural construction of

homosexuality, is a largely-hidden engine of social change that continues to propel society

slowly and inexorably in the direction of legalizing same-sex marriage.

Limitations and Future Research on Same-Sex Marriage

The results and limitations of this study suggest several avenues for further research

regarding same-sex marriage. First, because the sample from which qualitative data in this study

are drawn is not representative of the U.S. population, the prevalence of middle-ground

discourses about same-sex marriage with respect to supportive and oppositional discourses is
   357  
unknown. Future quantitative surveys should therefore include response categories designed to

capture the attitudes of people who use middle-ground discourses, and the reporting of attitudes

about same-sex marriage should not be limited to support and opposition. Response categories

such as “mixed feelings,” “no opinion,” and “don’t know” are often treated as residual categories

devoid of theoretical interest, but they all mean different things in light of the discourses

described here, and they carry meanings that are theoretically important. Thus, care should be

taken to devise response categories that would allow survey researchers to measure the varieties

of attitudes represented in middle-ground discourses.

A second improvement in quantitative surveys regarding same-sex marriage is that they

should include items that measure people’s cultural definitions of marriage. I have presented

evidence that such definitions of marriage do shape people’s attitudes about same-sex marriage,

but we do not know how important those definitions of marriage are relative to other factors, like

people’s attitudes and values regarding homosexuality. Nor are we able to ascertain with this

data the demographic and contextual factors that make people more or less likely to advance

different definitions of marriage. If quantitative surveys continue to measure attitudes about

same-sex marriage without measuring people’s definitions of marriage, we run the risk of

creating models of attitudes about same-sex marriage that are no different from any other gay

rights issue.

A third improvement in quantitative surveys regarding same-sex marriage is the inclusion

of an additional item on civil unions. In addition to measuring attitudes about civil unions,

researchers should measure levels of support for making the rights and benefits that come from

marriage available to all families, regardless of marital status. Knowing the levels of support and

opposition to such a course of action could expand the policy debate about same-sex marriage
   358  
into new arenas. This course of action might be a potential solution to the controversy, but it also

might fail to resonate culturally because of the fact that various legal rights and benefits have

long been associated with marriage in the United States. There is no way to know whether or not

such a course of action might be politically feasible unless social scientists attempt to gauge

public opinion about it.

A fourth improvement in quantitative surveys is the collection of data that could elucidate

the role of cohort membership in shaping attitudes about same-sex marriage. While this study

has suggested rough markers that divide cohorts from one another, I based those cohort

definitions on theoretical and historical accounts of how the construction of homosexuality

changed over time, rather than on actual data collected from the individuals themselves. It is

unlikely that the cohort definitions used in this paper are specified as well as they could be. A

more plausible definition of cohorts and a more plausible account of how cohort membership

matters could be attained by collecting data from the respondents themselves on collective

memories, such as memories of first encounters with homosexuality, memories of significant

political or cultural events, and memories that suggest a “coming of age” with regard to issues of

sexuality and relationships.

A final improvement in our understanding of the relationship between age and attitudes

about same-sex marriage would come from the collection of longitudinal data on same-sex

marriage. Because this study has used only cross-sectional data, I have presented only indirect

evidence that age-related differences are actually cohort differences. With longitudinal data,

quantitative analyses could directly disaggregate the age, cohort, and period effects from one

another.
   359  
With regard to future qualitative studies of same-sex marriage, this study suggests a

number of ways in which our understanding of the foundations of attitudes about same-sex

marriage could be improved. First, because the sample on which this study was based was drawn

solely from the upper Midwest, in a non-politicized context, future studies should examine how

people talk about same-sex marriage in other regions of the country, in politicized environments,

and in different ethnic and religious communities. It is unlikely that this study captures the full

variation in middle-ground discourses about same-sex marriage. There may be a variety of

discourses not captured in this study that are based in particular religious or ethnic communities,

and the patterns of discourses observed in the Midwest are not likely to be the same as the

patterns of discourse that one would observe in other places, such as the American South or in

California. Similarly, conducting a study such as this one in a politicized environment could shed

light on the extent to which middle-ground discourses are affected by activists working to

convert these “swing voters” into opponents or supporters of same-sex marriage.

Second, future qualitative studies should attempt to better analyze the roles of personal

contact, mass media, and social networks in shaping people’s tacit definitions of homosexuality.

The focus of my data collection was broader and more descriptive: to learn how people’s

discourses about same-sex marriage were premised upon a variety of beliefs, attitudes, values,

and life experiences. Thus, I did not collect enough data on these three factors that seem to be

decisive in shaping people’s understanding of homosexuality. In particular, the role of mass

media in shaping people’s discourse was difficult to establish in my interviews, so a more

sustained focus on people’s encounters with homosexuality in the media would improve our

understanding of the role of media in the continuing liberalization of attitudes regarding

homosexuality. Additional attention to each individual’s coming-of-age period, with respect to

   360  
these issues would also help to more accurately measure how people’s understandings of

homosexuality are created socially.

Third, it should be emphasized that my social generational interpretation of the

similarities and differences between cohorts with respect to discourses about same-sex marriage

is one of many possible interpretations that could be given. I interpreted these patterns as having

to do with people’s cultural repertoires, especially their tacit definitions of homosexuality,

precisely because I had focused my data collection and analysis on learning how people’s

discourses about same-sex marriage are premised on their cultural understandings of

homosexuality and marriage. However, scholars with other interests might design a similar study

that would interpret these patterns in terms of educational attainment, cognitive sophistication, or

some other phenomenon. Thus, I have not explained these cohort differences in discourses

because I have not ruled out competing explanations, nor have I isolated the mechanisms that

cause these differences. In general then, future studies of same-sex marriage, whether qualitative

or quantitative, can improve our understanding of cohort-related differences in discourses if they

are designed to explain such differences. At best, this dissertation has described cohort-related

differences in discourses about same-sex marriage and advanced one possible interpretation of

those differences that is grounded theoretically in the literatures on cohort analysis, generational

change, collective memories, and cultural sociology; the ultimate validity of this interpretation

must await a more direct test from future research.

   361  
Appendix A: Recruitment Poster (without pull-tabs)

Get Interviewed, Earn $30

Are you… Age 18-30?
Born in U.S.?
Attended high school in Illinois?
Enrolled in a degree program at NIU?
Have at least one parent age 45-63, born in U.S.,
and living in Illinois, Indiana, Wisconsin, or Iowa?

Wanted: Participants in a University of Wisconsin study about

what people in America think about current social and political
issues. Study will consist of a one-on-one interview about your life
history, media consumption, what you think about marriage,
sexuality, family, and politics.

Get paid $30 for 2-3 hours.

Contact: Peter Hart-Brinson, (608) 886-6951, or e-mail

pbrinson@ssc.wisc.edu.
   362  
Appendix B: Request Form for Interview with Parent

Request  for  Interview  with  Parent  

Do  you  have  at  least  one  parent  who  is:  

Between  ages  45  and  63               ___  Yes  ___  No  

And  was  Born  in  the  U.S.             ___  Yes  ___  No  

And  is  currently  living  in  Illinois,  Wisconsin,  Iowa,  or  Indiana   ___Yes  ___  No    

• If  only  1  parent  meets  these  criteria,  what  is  his/her  name?  ____________________________  

• If  both  parents  meet  the  above  criteria,  which  parent  would  you  say  you  are  closest  

with?  ____________________________  

May  I  have  permission  to  contact  them  and  ask  them  for  an  interview?  __________________  

Preferred  method  of  contact  (You  may  provide  as  much  or  as  little  information  as  you  like.)  

Email  address:  ________________________________________________________________  

Mailing  address:  _______________________________________________________________  

           _______________________________________________________________  

Phone  number(s):  ______________________________________________________________  

   363  
Appendix C: Script for Contacting Parents

Phone  
 
Hello,  my  name  is  Peter  Hart-­‐Brinson,  and  I  am  calling  from  the  University  of  Wisconsin.  I  
recently  conducted  an  interview  with  your  son/daughter  _______________  for  a  research  study,  
and  he/she  gave  me  permission  to  ask  you  for  an  interview  as  well.  Did  he/she  tell  you  
about  the  interview?  
 
Yes:  Oh,  good.  As  he/she  may  have  told  you…  
No:  That’s  okay.  
 
The  interview  is  for  a  study  about  Americans’  feelings  about  various  social  and  political  
issues  and  how  they  are  related  to  their  values  and  life  experiences.  We  are  asking  you  to  
participate  in  the  study  because  we  want  to  learn  what  similarities  and  differences  of  
opinion  exist  between  college  students  and  their  parents/guardians.  
 
If  you  are  willing  to  be  interviewed,  I  would  drive  from  Wisconsin  to  meet  you  at  a  place  
and  time  that  would  be  convenient  for  you.  The  interview  lasts  for  approximately  2-­‐3  
hours,  and  you  will  be  paid  $30  upon  completion  of  the  interview.  The  interview  is  
completely  confidential  and  no  one  other  than  myself  will  know  your  true  identity.    
 
Would  you  be  willing  to  be  interviewed?  
 
No:  Okay,  thank  you  for  your  time.  
Yes:  Okay,  great!  Typically,  I  am  available  Thursday-­‐Sunday  for  interviews.  When  would  be  
a  good  time  for  you?  [Arrange  date/time]  [Get  location/address]  Okay,  I  will  give  you  a  call  
the  day  before  the  interview  as  a  confirmation  and  a  reminder  of  the  interview;  if  we  need  
to  change  our  plans,  we  can  do  it  then.  Thank  you,  have  a  good  day.  
 
 
 
Voice  Mail  
 
Hello,  my  name  is  Peter  Hart-­‐Brinson,  and  I  am  calling  from  the  University  of  Wisconsin.  I  
recently  conducted  an  interview  with  your  son/daughter  _______________  for  a  research  study,  
and  he/she  gave  me  permission  to  ask  you  for  an  interview  as  well.  It  is  for  a  study  about  
Americans’  feelings  about  various  social  and  political  issues  and  how  they  are  related  to  
their  values  and  life  experiences.  We  are  asking  you  to  participate  in  the  study  because  we  
want  to  learn  what  similarities  and  differences  of  opinion  exist  between  college  students  
and  their  parents/guardians.  I  would  like  to  talk  with  you  about  the  study  and  ask  you  if  
you  are  willing  to  be  interviewed  for  it.  If  you  can,  please  give  me  a  call  at  608-­‐886-­‐6951  
   364  
(again,  my  name  is  Peter).  Otherwise,  I  will  try  calling  again  later.  Thank  you,  and  have  a  
good  day.  
 
Email  
 
Dear  _________,  
My  name  is  Peter  Hart-­‐Brinson,  and  I  am  a  researcher  at  the  University  of  Wisconsin.  I  
recently  conducted  an  interview  with  your  son/daughter  _______________  for  a  research  study,  
and  he/she  gave  me  permission  to  ask  you  for  an  interview  as  well.  It  is  for  a  study  about  
Americans’  feelings  about  various  social  and  political  issues  and  how  they  are  related  to  
their  values  and  life  experiences.  We  are  asking  you  to  participate  in  the  study  because  we  
want  to  learn  what  similarities  and  differences  of  opinion  exist  between  college  students  
and  their  parents/guardians.  
 
If  you  are  willing  to  be  interviewed,  I  would  drive  from  Wisconsin  to  meet  you  at  a  place  
and  time  that  would  be  convenient  for  you.  The  interview  lasts  for  approximately  2-­‐3  
hours,  and  you  will  be  paid  $30  upon  completion  of  the  interview.  The  interview  is  
completely  confidential  and  no  one  other  than  myself  will  know  your  true  identity.    
 
If  you  have  questions  or  are  not  sure  whether  you  want  to  participate,  please  ask;  I  am  
happy  to  answer  them.  Otherwise,  please  let  me  know  whether  or  not  you  are  willing  to  
participate.  
 
Thank  you  very  much,  
Peter  Hart-­‐Brinson  
 
   365  
Appendix D: Interview Guide

 
Informed  Consent  
 
Introduction  
This  is  a  study  about  Americans’  attitudes  about  a  number  of  important  social  issues  and  
how  they  are  related  to  their  values  and  life  experiences.  In  the  interview,  I  will  ask  you  
about  your  life  growing  up,  your  media  consumption  habits,  and  your  feelings  about  
marriage,  family,  sexuality,  and  politics.  Everything  you  tell  me  will  be  kept  in  complete  
confidence,  and  no  one  will  know  your  true  identity  except  for  me.  
 
My  goal  in  the  interview  is  basically  to  get  to  know  you—to  learn  as  much  about  you  as  I  
can.  Although  I  will  be  asking  a  lot  of  questions,  I  don’t  want  this  to  feel  like  an  
interrogation!  There  are  no  right  or  wrong  answers.  You  do  not  have  to  answer  every  
question  if  you  don’t  feel  comfortable  doing  so.  Just  like  you  don’t  have  to  tell  your  friends  
or  neighbors  everything  about  yourself,  you  don’t  have  to  tell  me  everything  about  yourself  
if  you  don’t  want  to.  If  you  don’t  want  to  talk  about  something,  just  say  so,  and  we’ll  move  
on.  And  the  interview  can  be  stopped  at  any  time.  Do  you  have  any  questions?  
 
Demographics  
 
First  off,  why  don’t  you  tell  me  a  little  about  yourself?  (clarify  if  necessary:  If  you  were  
going  to  introduce  yourself  to  a  room  full  of  people  and  you  wanted  them  to  get  to  know  
you,  what  would  you  say?)  
 
If  not  clear  from  this  introduction,  make  sure  you  get:  
• Sex  (never  asked)  
• Age/Birth  year  (What  year  were  you  born?  How  old  are  you?)  
• Ethnic  Background  (Do  you  know  anything  about  your  ethnic  background  at  all?)  
• Education  History/Major  (Did  you  go  to  college?  What  was  your  major?  How  long  
have  you  been  here?  etc.)  
• Occupation  (Are  you  working  right  now?)  
• Relationship/marital  status/background  (Are  you  in  a  relationship  right  now?  Are  
you  married?)  
• Children  (Do  you  have  any  children?)  
• Siblings  (Do  you  have  any  siblings?)  
• Parents’  marital  status/history  (Are/Were  your  parents  married?  Alive?  Did  they  
stay  married  until  they  passed  away?  etc.)  
• Political  ideology  (How  would  you  describe  your  political  beliefs,  if  any?)  
• Political  party  membership  (Are  you  a  member  of  any  political  parties?)  
• Religious  faith  (Do  you  consider  yourself  religious  or  spiritual  at  all?)  
• Free  time  activities/hobbies/interests  (What  kinds  of  things  do  you  like  to  do  in  
your  free  time?)  
   366  
 
Childhood/Coming-­of-­age  
 
 
Where  did  you  grow  up?  
  What  was  it  like  there?  
  Did  you  like  it?  
  Were  there  a  lot  of  kids  in  the  neighborhood?  
Family  life  
  What  did  your  parents  do  for  a  living?  
  Did  you  get  along  well  with  your  parents?  
  Were  you  close  with  your  siblings?  Did  you  get  along  well  with  them?  
Did  you  do  a  lot  of  things  together  as  a  family?  Family  dinners?  Vacations?  TV  or  
movies?  
Did  you  have  your  own  private  space  in  the  house?  Own  room,  t.v.,  phone,  etc?  
Was  your  family  religious?  Were  you  raised  in  a  religious  environment?  
Did  your  family  talk  about  politics  much?  
High  School  
  Where  did  you  go  to  high  school?  
  What  was  it  like  there?  
  Were  you  a  good  student?  
  Extracurricular  activities?  Athletics,  clubs,  theater,  etc.  
Friends  and  Dating  
  Who  were  your  closest  friends?  
Did  you  hang  out  with  friends  much?  
  What  kinds  of  things  did  you  do  with  your  friends?  
How  common  was  it  among  your  peers  to  have  parents  who  were  divorced?  Or  to  
come  from  single-­‐parent  families?  
  Did  you  date  anyone  in  high  school?  
  What  did  people  typically  do  on  dates?  
  What  did  your  friends/peers  think  about  casual  sex  while  dating?  
  Would  your  parents  have  disapproved?  
Media/Politics  
  Did  you  follow  news/current  events  much  back  then?  
  What  were  some  of  your  favorite  TV  shows  back  then?  
  Memories  of  important  political  or  social  events?  
Expectations  for  Future  
What  kinds  of  expectations  did  (do)  your  parents  have  for  you,  in  terms  of  
education,  career,  family,  etc.  
Did  (Do)  you  expect  that  you  were  going  to  get  married  someday?  Did  you  and  your  
friends  ever  talk  about  it?  
Did  (Do)  you  expect  that  you  were  going  to  start  a  family  and  have  children?  
After  high  school  
  Go  directly  to  college?  Work?  Military?  
   367  
  What  was  it  like?  
  How  did  your  life  change  between  high  school  and  ____?  
  Significant  changes  in  hobbies/interests,  dating,  etc?  
  Memories  of  important  social/political  events?  
Update  to  present  time  
  What  did  you  do  after  ____?  
  What  were  the  particularly  important  moments  in  your  life  since  ____?  
  Employment  history  
  Free  time  activities  
  Travel/moves/vacations  
  Relationships—how  you  met,  marriage,  divorce,  etc.  
  Children—What  was  it  like  raising  them?  Best  things?  Hardest  things?  
How  do  you  feel  about  how  your  children  are  doing  now?  
 
Media  Consumption  
 
I  want  to  switch  gears  a  little  bit  and  ask  you  some  questions  about  your  media  consumption  
habits.  
 
Do  you  subscribe  to  or  regularly  read  a  daily  newspaper?  Which  ones?  
Do  you  subscribe  to  or  regularly  read  any  magazines?  Which  ones?  
Do  you  have  cable  or  satellite  television?  
About  how  many  hours  per  day  would  you  say  you  watch  t.v.,  on  average?  
What  are  some  of  your  favorite  t.v.  shows?  Probe:  
• Daytime  talk  shows  
• Daytime  soaps  
• Prime  time  sitcoms/comedies  
• Prime  time  dramas  
• Sports  
• Game  shows  
• Reality  shows  
• Late  night  talk  
• Local  news  
• National/international  news  
• Movies  
• Public  television  (PBS)  
What  kind  of  music  do  you  like?  CDs  or  iPod?  
Do  you  listen  to  the  radio  much?  What?  
Do  you  have  access  to  the  internet  at  home?  
What  kinds  of  things  do  you  do  when  you  go  online?  Probe:  
• Email  or  online  chat  
• Stuff  for  work/school  
• Shopping  
• Reading  news,  commentary,  or  blogs  
   368  
• Social  network  sites  (e.g.  Facebook,  myspace,  etc.)  
• Entertainment/Fun  
• Other  significant  activities?  
What  kinds  of  movies  do  you  like?  Favorites  genres?  Titles?  
Do  you  read  books  for  pleasure  much?  What  do  you  like  to  read?  
 
Marriage  
 
Okay,  I  want  switch  gears  again  and  ask  you  a  few  questions  about  marriage  and  
relationships.  
 
Marriage  
  First  off,  what  does  the  word  “marriage”  mean  to  you?  
  Most  important  characteristics  of  a  good,  strong  marriage?  
  What  about  ____?  How  important  do  you  think  that  is  for  a  good  marriage?  
    Communication  
    Love  
    Sexual  attraction  
    Shared  interests  
    Religion  
    Raising  children  
How  do  you  think  work/household/parenting  responsibilities  should  be  shared  in  a  
marriage?  Different  roles?  Shared  equally?  
  Have  your  views  about  marriage  changed  over  time?  How?  Why?  
Cohabitation  &  premarital  sex  
How  do  you  feel  about  cohabitation?  That  is  a  couple  living  together  even  though  
they’re  not  married?  
Under  what  circumstances?  
Do  you  have  any  experience  with  cohabitation  (living  with  a  partner  before/outside  
of  marriage)?  
If  a  couple  has  been  living  together  for  a  while  before  they  get  married,  does  
anything  change  significantly  when  they  get  married?  
How  do  you  feel  about  premarital  sex?  Under  what  circumstances  is  okay/not  okay?  
Divorce  
  How  do  you  feel  about  divorce?  
Do  you  think  divorce  is  a  problem  in  our  society?  
  Why  do  you  think  divorce  has  become  so  much  more  common  in  our  society  lately?  
Under  what  circumstances  is  it  okay  for  a  couple  to  get  divorced?  Not  okay?  
Do  you  think  there  are  particular  circumstances  in  which  divorce  is  good?  Bad?  
  Couple  doesn’t  love  each  other  anymore?  
  Irreconcilable  differences?  
  What  about  wedding  vows?  
  Is  it  different  if  a  couple  has  children?  
How  easy/hard  do  you  think  it  should  be  for  a  couple  to  get  a  divorce?  
   369  
Institution  
What  kinds  of  attitudes  do  you  think  people  in  the  U.S.  have  about  marriage  
generally?  
How  do  you  think  marriage  is  portrayed  in  the  media?  
Do  you  think  the  media  paints  a  better  picture  of  certain  kinds  of  relationships  than  
others?  
Some  people  these  days  are  talking  about  marriage  being  in  a  crisis.  What  do  you  
think  about  that?  
 
Same-­Sex  Marriage  
 
One  of  the  issues  that  has  been  relatively  controversial  in  recent  years  is  the  issue  of  same-­sex  
marriage.  
 
Same-­‐Sex  Marriage  
Have  you  heard  much  about  this  issue?  What  do  you  know  about  it?  Where  did  you  
hear  about  it?  
  Why  do  you  think  it  is  so  controversial?  
  Do  you  personally  have  an  opinion  on  the  issue?  
What  reasons  do  you  think  people  would  give  for  supporting/opposing  same-­‐sex  
marriage?  
  Why  do  you  think  some  gays  and  lesbians  want  the  right  to  marry?  What’s  at  stake?  
Effects  (Hypothetical  scenario)  
What  kinds  of  effects  would  legalizing  same-­‐sex  marriage  have  on  society,  if  any?  
Negative?  Positive?  
Would  it  affect  the  marriages  of  other  people?  Why/not?  
Would  it  change  meaning  of  marriage?  
[Pro  only]  Wouldn’t  it  be  a  dangerous  social  experiment?  Shouldn’t  it  worry  you  that  
all  of  a  sudden,  gender  wouldn’t  matter  anymore?  
Effects  on  gays  and  lesbians?  
Effects  on  children?  
Do  you  think  legalizing  same-­‐sex  marriage  would  encourage  more  people  to  be  gay?  
Civil  Unions  (Hypothetical  scenario)  
What  if  the  government  decided  to  deny  marriage  rights  to  same-­‐sex  couples  
(keeping  marriage  between  one  man  and  one  woman),  but  instead  created  a  new  
legal  relationship  status  for  same-­‐sex  couples  (e.g.  civil  union  or  domestic  
partnership)  that  would  give  them  the  same  rights  and  responsibilities  as  marriage?  
What  would  you  think  about  that?  
Do  you  think  it  could  be  a  compromise  that  would  resolve  the  controversy?  
Marriage  as  religious  institution  vs.  civil  institution?  Shouldn’t  they  be  separated?  
Resolution  
How  do  you  think  the  controversy  should  be  resolved?  By  whom?  
National/state  level?  
Legislative/judicial/referendum?  
   370  
What  do  you  think  is  going  to  happen  in  the  future  regarding  this  issue?  Do  you  
think  same-­‐sex  marriage  will  ever  be  legal  in  the  United  States?  
 
Sexuality  
 
Personal  Contact  [how  much?]  
Do  you  know  anybody  personally  who  identifies  as  gay/lesbian?  Who?  What  is  your  
relationship  like  with  them?  
  Can  you  think  of  anybody  famous  who  is  gay  or  lesbian?  
How  do  you  think  gays  and  lesbians  are  portrayed  in  the  media?  
Do  you  remember  the  first  time  you  recognized  somebody  to  be  gay  or  lesbian?    
How  did  you  feel?  How  did  you  know?  How  were  they  treated  by  other  people?  
Homosexuality  
What  do  you  think  homosexuality  is?  What  do  you  think  it  really  means  if  someone  
identifies  themselves  as  gay/lesbian?  
Do  you  think  people  are  born  gay?  How  they  are  raised/environmental  factors?  
Nature  vs.  nurture?  Lifestyle  choice?  
Can  it  be  changed?  
Is  it  a  sin?  Immoral?  
What  do  you  think  about  people  who  identify  as  gay/lesbian/homosexual  but  don’t  
have  sexual  relations  with  people  of  the  same  sex?  
Attitudes  
Should  gays  and  lesbians  be  treated  the  same  by  society  as  heterosexuals?  Same  
civil  rights?  
Do  you  think  they  get  special  treatment?  
Have  your  feelings  about  homosexuality  changed  over  the  course  of  your  life?  How?  
Why?  
Bisexuality  
Some  people  identify  themselves  as  bisexual.  What  do  you  think  they  mean?  
Are  they  born  that  way?  How  they  were  raised?  Lifestyle  choice?  Confused?  
[Support  only]  If  someone  identifies  as  bisexual  and  they  want  to  form  a  
relationship  with  someone  of  the  same  sex,  should  they  be  allowed  to  marry?  
[Support  only]  What  if  two  people  of  the  same  sex  want  to  get  married,  but  they  
aren’t  sexually  attracted  to  each  other—do  you  think  that  should  be  allowed?    
 
Politics  
 
Okay,  lastly,  I  want  to  ask  you  just  a  few  questions  about  the  upcoming/recent  election.  
 
What  do  you  think  about  the  upcoming  Presidential  election?  
  Have  you  been  following  it  much?  
  Are  people  you  know  talking  about  it  much?  What  are  they  saying?  
  Are  you  planning  on  voting?  
  Getting  active  or  donating  money  to  a  campaign?  
   371  
What  do  you  think  are  the  most  important  issues  facing  the  country  in  the  upcoming  years?  
Why?  
What  do  you  think  about  John  McCain?  Like  most/least?  
Too  old?  Maverick?  Another  Bush?  
What  do  you  think  about  Barack  Obama?  Like  most/least?  
Too  young?  Too  idealistic/not  substantive?  New  kind  of  politics?  
How  would  the  race  be  different,  if  at  all,  if  Hillary  Clinton  had  won  the  Democratic  
nomination?  Do  you  think  she  would  be  a  better  candidate  or  president?  
How  do  you  think  the  election  will  turn  out?    
   372  
Appendix E: Codebook

Attributes (Cases)

These are the “variables” describing each respondent based on first section of the interview.
Code inductively.

Description (under “case properties”)

Pseudonym (“Nickname” under “case properties”)
Date of interview
Length of interview
Sex
Age
Birth year
Ethnicity
Education (includes NIU/RVC)
Occupation
Relationship
Children
Siblings
Parents’ relationship
Political ID
Religion ID

Question-Response Codes

These codes start with my question and include the response by the informant to that question. It
should include all conversation, including follow-up probes, until the next major Q-R code. This
divides the interviews into sections, based on how I framed the conversation or primed the
informant to talk about certain things.

Demographics

Code Question
Qintro Tell me a little bit about yourself.
Qage How would you describe your political beliefs, if any?
Qethnic Do you consider yourself religious or spiritual at all?
Qeduc College? Major? Education?
Qoccupation Are you working right now? What do you do for a living?
Qrelationship Are you in a relationship right now? Married?
Qchildren Do you have any children?
Qsiblings Do you have any siblings?
Qparents Are your parents married? Alive? Remained married?
QpoliticsID How would you describe your political beliefs if any?
QreligID Religious or spiritual at all?
   373  
Qhobbies What kinds of things do you do in your free time?

Childhood/Coming-of-age

Code Question
Qgrowup Where did you grow up? What was it like there? Did you like it? Were there a
lot of kids in the neighborhood?
Qparentjob What did your parents do for a living?
Qcloseparent Did you get along well with your parents?
Qclosesibling Were you close with your siblings? Did you get along well with them?
Qfamilylife Did you do a lot of things together as a family? Family dinners? Vacations?
TV or movies?
Qprivacy Did you have your own private space in the house? Own room, t.v., phone,
etc?
Qfamrelig Was your family religious? Were you raised in a religious environment?
Qfampol Did your family talk about politics much?
Qhighschool Where did you go to high school? What was it like there?
Qhsstudent Were you a good student?
Qhsextracur Extracurricular activities? Athletics, clubs, theater, etc.
Qhsfriends Who were your closest friends? Did you hang out with friends much? What
kinds of things did you do with your friends?
Qhspeerdivorce How common was it among your peers to have parents who were divorced?
Or to come from single-parent families?
Qhsdating Did you date anyone in high school? What did people typically do on dates?
Qhspeersex What did your friends/peers think about casual sex while dating?
Qhsdisapprove Would your parents have disapproved?
Qhsnews Did you follow news/current events much back then?
Qhsmedia What were some of your favorite TV shows back then?
Qhsmemory Memories of important political or social events?
Qexpect What kinds of expectations did (do) your parents have for you, in terms of
education, career, family, etc.
Qafterhs What did you do after high school? Go directly to college? Work? Military?
What was it like?
Qlifechange How did your life change between high school and ____?
Qcolmemory After high school, memories of important social/political events?
Qsincethen What did you do after ____? What were the particularly important moments
in your life since ____? Employment history, Free time activities, Travel
Qmetspouse Relationships—how you met, marriage, divorce, etc.
Qraisekids Children—What was it like raising them? Best things? Hardest things?
Qemptynest How do you feel about how your children are doing now?
   374  

Media Consumption

Code Question
Qnewspaper Do you subscribe to or regularly read a daily newspaper? Which ones?
Qmagazine Do you subscribe to or regularly read any magazines? Which ones?
QTV Do you have cable or satellite television? About how many hours per day
would you say you watch t.v., on average? What are some of your favorite t.v.
shows?
Qmusic What kind of music do you like? CDs or iPod?
Qradio Do you listen to the radio much? What?
Qinternet Do you have access to the internet at home? What kinds of things do you do
when you go online?
Qmovies What kinds of movies do you like? Favorites genres? Titles?
Qreading Do you read books for pleasure much? What do you like to read?

Marriage
Code Question
Qmarrmean First off, what does the word “marriage” mean to you?
Qmarrcharac Most important characteristics of a good, strong marriage?
Qmarrcomm Communication?
Qmarrsex Sexual attraction?
Qmarrinterest Shared interests?
Qmarrreligion Religion?
Qmarrkids Raising children?
Qdivlabor How do you think work/household/parenting responsibilities should be shared
in a marriage? Different roles? Shared equally?
Qmarrchange Have your views about marriage changed over time? How? Why?
Qmarrplans Do you ever think you will get married?
Qcohabit How do you feel about cohabitation?
Qcohabchange If a couple has been living together for a while before they get married, does
anything change significantly when they get married?
Qpremarital How do you feel about premarital sex?
Qdivorce How do you feel about divorce?
Qdivproblem Do you think divorce is a problem in our society?
Qdivrate Why do you think divorce has become so much more common in our society
lately?
Qdivcircum Do you think there are particular circumstances in which divorce is good/bad?
Qdivnolove Couple doesn’t love each other anymore?
Qirreconcile Irreconcilable differences?
Qdivkids Is it different if a couple has children?
Qdivhowhard How easy/hard do you think it should be for a couple to get a divorce?
Qmarratts What kinds of attitudes do you think people in the U.S. have about marriage
generally?
   375  
Qmarrmedia How do you think marriage is portrayed in the media? Do you think the media
paints a better picture of certain kinds of relationships than others?
Qmarrcrisis Some people these days are talking about marriage being in a crisis. What do
you think about that?

Same-Sex Marriage

Code Question
Qssmintro Have you heard much about this issue? What do you know about it?
Qcontroversy Why do you think it is so controversial?
Qssmopinion Do you personally have an opinion on the issue?
Qssmwhyoppose What reasons do you think people would give for opposing same-sex
marriage?
Qssmwhysupport What reasons do you think people would give for supporting same-sex
marriage?
Qssmmotive Why do you think some gays and lesbians want the right to marry? What’s
at stake?
Qssmeffects What kinds of effects would legalizing same-sex marriage have on society,
if any?
Qssmothermarr Would it affect the marriages of other people? Why/not?
Qssmmeaning Would it change meaning of marriage?
Qssmeffgays Effects on gays and lesbians?
Qssmeffkids Effects on children?
Qssmendorse Do you think legalizing same-sex marriage would be an endorsement of
homosexuality by society?
Qssmexperiment Wouldn’t it be a dangerous social experiment? Shouldn’t it worry you that
all of a sudden, gender wouldn’t matter anymore?
Qssmmakegay Do you think legalizing same-sex marriage would encourage more people to
be gay?
Qcivilunion What if the government decided to deny marriage rights to same-sex
couples, but instead created a new legal relationship status for same-sex
couples (e.g. civil union or domestic partnership) that would give them the
same rights and responsibilities as marriage?
Qcompromise Do you think it could be a compromise that would resolve the controversy?
Qssmwhodecides How do you think the controversy should be resolved? By whom?
Qssmfuture What do you think is going to happen in the future regarding this issue? Do
you think same-sex marriage will ever be legal in the United States?

Sexuality

Code Question
Qgaycontact Do you know anybody personally who identifies as gay/lesbian?
Qgayceleb Can you think of anybody famous who is gay or lesbian?
Qgaymedia How do you think gays and lesbians are portrayed in the media?
Qgaymemory Do you remember the first time you recognized somebody to be gay or
   376  
lesbian?
Qaccepted How do you think homosexuality is viewed by society?
Qgaydef What do you think homosexuality is? What do you think it really means if
someone identifies themselves as gay/lesbian?
Qborngay Do you think people are born gay?
Qraisedgay How they are raised/environmental factors?
Qlifestyle Lifestyle choice?
Qgaychange Can it be changed?
Qsin Is it a sin? Immoral?
Qgayrefrain What do you think about people who identify as gay/lesbian/homosexual but
don’t have sexual relations with people of the same sex?
Qequalrights Should gays and lesbians have the same rights as heterosexuals?
Qgayattchange Have your feelings about homosexuality changed over the course of your life?
Qbisexdef Some people identify themselves as bisexual. What do you think they mean?
Qbisexorigin Are they born that way? How they were raised? Lifestyle choice? Confused?
Qbisexmarr If someone identifies as bisexual and they want to form a relationship with
someone of the same sex, should they be allowed to marry?
Qnosexmarr What if two people of the same sex want to get married, but they aren’t
sexually attracted to each other—do you think that should be allowed?

Politics

Code Question
Qelection What do you think about the upcoming Presidential election? Have you been
following it much?
Qelectiontalk Are people you know talking about it much? What are they saying?
Qvote Are you planning on voting?
Qactive Getting active or donating money to a campaign?
Qissues What do you think are the most important issues facing the country in the
upcoming years?
Qmccain What do you think about John McCain?
Qobama What do you think about Barack Obama?
Qpalin What do you think about Sarah Palin?
Qhillary How would the race be different, if at all, if Hillary Clinton had won the
Democratic nomination?
Qresult How do you think the election will turn out?
   377  
Deductive Codes

These codes are specific things I am looking for in the interviews. They will be further coded
inductively within each code, to specify the nature of their attitude, belief, experience, etc.

Code Description
bio Important biographical fact/event. Important for understanding this person.

CRlifeexp Uses life experience (first- or second-hand) to explain opinion.

CRmedia Uses something from media to explain opinion.
CRcommonsense Uses common sense cultural idea to explain opinion.
hypotheticalother Imagined, stereotyped, or possible behavior of others.
metaphor Uses metaphor or similie or comparison to explain opinion.
anecdote Uses story to explain opinion.
fact Cites fact or statistic to explain opinion.

praglogic Pragmatic (outcome, ends-based) opinion or discourse.

religlogic Opinion or discourse based on religious teachings.
politiclogic Opinion or discourse based on political ideology.
interestlogic Opinion or discourse based on self-interest
morallogic Opinion or discourse based on non-religious moral beliefs.

politicalbeliefs Indicator about political beliefs and political attitudes

religiousbeliefs Indicator about religious beliefs and religious attitudes

religbackgrd Religious family background or upbringing

politicbackgrd Political family background or upbringing
hsactivities High school activities and friends
hsdating High school dating experiences
peersexatt High school peer attitudes about sex
peerdivorce High school peer experiences with divorce
collmemory Important memories from adolescence/young adulthood
expectation Parents’ expectations for future

marriageatt Attitude about marriage

marriagemean Understanding/meaning of marriage
marrsex How important sexual attraction is for marriage
marrreligion How important religion is for marriage
marrkids How important having kids is for marriage
divlabor How married couples should divide labor
genderideo Gender ideology
marrattchange How feelings about marriage have changed throughout life
cohabitatt Attitude about cohabitation
premaritalatt Attitude about premarital sex
divorceatt Attitude about divorce
   378  

ssmatt Attitude about same-sex marriage

ssmoppose Reasons why people oppose same-sex marriage
ssmgayswant Reasons why gays want marriage
ssmeffects Belief about effects of legalizing same-sex marriage
effectonmeaning Belief about how same-sex marriage would affect meaning of marriage
effectonkids Belief about how same-sex marriage would affect kids
civilunionatt Attitude about civil unions
gayatt Attitude about homosexuality/gays and lesbians
personalcontact Personal contact with gays and lesbians
gaymemory Memory about encountering homosexuality/gays/lesbians
homodef Beliefs about what homosexuality is and where it comes from
changehomo Belief about whether homosexuality can be changed
homomoral Morality/immorality of homosexuality
homoattchange How atts/beliefs about homosexuality have changed over time
equalrightsatt Attitude about equal rights for gays
bisexdef Beliefs about what bisexuality is and where it comes from
bisexmarr Attitude about bisexual same-sex marriage
nosexmarr Attitude about heterosexual same-sex marriage

vote Voting plans/behavior in 2008 election

Obamaatt Attitude about Obama
Mccainatt Attitude about McCain
Palinatt Attitude about Palin
Hillaryatt Attitude about Hillary Clinton
   379  

Inductive Codes

These codes emerged from the interview transcripts as I was coding. I did not set out to look for
these things.

Code Description

Keyword Codes

These codes emerged from interview transcripts. They represent key words or phrases.

Code

Annotations

These are passages that explain the significance of the data.

Type Description
Interpretive Link between discourse and person/biography
Theoretical Link between discourse and theory/sociology

Memos

Written to compare each matched student-parent pair.

Fieldnotes

These are the reflections that I wrote down after each interview.
   380  

References

Abramson, Paul R., and Ronald Inglehart. 1986. "Generational Replacement and Value Change
in Six West European Societies." American Journal of Political Science 30:1-25.

Alexander, Jeffrey C. 2003. The Meanings of Social Life: A Cultural Sociology. New York:
Oxford University Press.

Alexander, Jeffrey C., and Philip Smith. 1993. "The Discourse of American Civil Society: A
New Proposal for Cultural Studies." Theory and Society 22:151-207.

Allport, Gordon W. 1954. The Nature of Prejudice. New York: Basic Books.

Alwin, Duane F. 1990. "Cohort Replacement and Changes in Parental Socialization Values."
Journal of Marriage and Family 52:347-360.

Alwin, Duane F., and Jon A. Krosnick. 1991. "Aging, Cohorts, and the Stability of Sociopolitical
Orientations over the Life Span." American Journal of Sociology 97:169-195.

Amato, Paul R. 2004. "Tension Between Institutional and Individual Views of Marriage."
Journal of Marriage and Family 66:959-965.

Arendt, Hannah. 1998. The Human Condition. Chicago: University of Chicago Press.

Armstrong, Elizabeth A., and Suzanna M. Crage. 2006. "Movements and Memory: The Making
of the Stonewall Myth." American Sociological Review 71:724-751.

Baldassarri, Delia, and Peter Bearman. 2007. "Dynamics of Political Polarization." American
Sociological Review 72:784-811.

Battani, Marshall, David R. Hall, and Rosemary Powers. 1997. "Cultures' Structures: Making
Meaning in the Public Sphere." Theory and Society 26:781-812.

Bauerlein, Mark. 2008. The Dumbest Generation: How the Digital Age Stupefies Young
Americans and Jeopardizes Our Future. New York: Jeremy P. Tarcher/Penguin.

Becker, Ron. 2006. Gay TV and Straight America. New Brunswick, NJ: Rutgers University
Press.

Bengtson, Vern L., Michael J. Furlong, and Robert S. Laufer. 1974. "Time, Aging, and the
Continuity of Social Structure: Themes and Issues in Generational Analysis." Journal of
Social Issues 30:1-30.

Berger, Peter L., and Thomas Luckmann. 1980. The Social Construction of Reality: A Treatise in
the Sociology of Knowledge. New York: Irvington Publishers.
   381  
Bernstein, Mary. 1997. "Celebration and Suppression: The Strategic Uses of Identity by the
Lesbian and Gay Movement." American Journal of Sociology 103:531-565.

—. 2002. "Identities and Politics: Toward a Historical Understanding of the Lesbian and Gay
Movement." Social Science History 26:531-581.

Bloor, Michael, Jane Frankland, Michelle Thomas, and Kate Robson. 2001. Focus Groups in
Social Research. Thousand Oaks, CA: Sage Publications.

Braungart, Richard G., and Margaret M. Braungart. 1986. "Life-Course and Generational
Politics." Annual Review of Sociology 12:205-231.

Brewer, Paul R. 2002. "Framing, Value Words, and Citizens' Explanations of Their Issue
Opinions." Political Communication 19:303-316.

—. 2003. "Values, Political Knowledge, and Public Opinion About Gay Rights: A Framing-
Based Account." Public Opinion Quarterly 67:173-201.

Brewer, Paul R., and Clyde Wilcox. 2005. "The Polls--Trends: Same Sex Marriage and Civil
Unions." Public Opinion Quarterly 69:599-616.

Brewster, Karin L., and Irene Padavic. 2000. "Change in Gender Ideology, 1977-1996: The
Contributions of Intracohort Change and Population Turnover." Journal of Marriage and
Family 62:477-487.

Brooks, Clem, and Catherine Bolzendahl. 2004. "The Transformation of US Gender Role
Attitudes: Cohort Replacement, Social-Structural Change, and Ideological Learning."
Social Science Research 33:106-133.

Buchmann, Claudia, Thomas A. DiPrete, and Anne McDaniel. 2008. "Gender Inequalities in
Education." Annual Review of Sociology 34:319-37.

Cherlin, Andrew J. 2004. "The Deinstitutionalization of American Marriage." Journal of

Marriage and Family 66:848-861.

—. 2009. The Marriage-Go-Round: The State of Marriage and the Family in America Today.
New York: Alfred A. Knopf.

Ciabattari, Teresa. 2001. "Changes in Men's Conservative Gender Ideologies: Cohort and Period
Influences." Gender & Society 15:574-591.

Cohen, Jean. 1999. "American Civil Society Talk." Pp. 55-85 in Civil Society, Democracy, and
Civic Renewal, edited by Robert K. Fullinwider. Lanham, Md.: Rowman & Littlefield
Publishers.

Corsten, Michael. 1999. "The Time of Generations." Time and Society 8:249-272.
   382  
Cutler, Neal E., and Vern L. Bengtson. 1974. "Age and Political Alienation: Maturation,
Generation, and Period Effects." Annals of the American Academy of Political and Social
Science 415:160-175.

Davis, Nancy J., and Robert V. Robinson. 1996. "Are the Rumors of War Exaggerated?
Religious Orthodoxy and Moral Progressivism in America." American Journal of
Sociology 102:756-787.

Delli Carpini, Michael X. 1986. Stability and Change in American Politics: The Coming of Age
of the Generation of the 1960s. New York: New York Univeristy Press.

Delli Carpini, Michael X., and Bruch A. Williams. 1994. "Methods, Metaphors, and Media
Research: The Uses of Television in Political Conversation." Communication Research
21:782-812.

Demartini, Joseph R. 1985. "Change Agents and Generational Relationships: A Reevaluation of

Mannheim's Problem of Generations." Social Forces 64:1-16.

DiMaggio, Paul, John Evans, and Bethany Bryson. 1996. "Have American's Social Attitudes
Become More Polarized?" American Journal of Sociology 102:690-755.

Down, Simon, and James Reveley. 2004. "Generational Encounters and the Social Formation of
Entrepreneurial Identity: 'Young Guns' and 'Old Farts'." Organization 11:223-250.

Druckman, Jamie N., and Kjersten R. Nelson. 2003. "Framing and Deliberation: How Citizens'
Conversations Limit Elite Influence." American Journal of Political Science 47:726-745.

Durkheim, Emile. 1984. The Division of Labor in Society. New York: Free Press.

Edmunds, June, and Bryan S. Turner. 2002. Generational Consciousness, Narrative, and
Politics. Lanham, MD: Rowman and Littlefield.

Egan, Patrick J., and Kenneth Sherrill. 2009. "California's Proposition 8: What Happened, and
What Does the Future Hold?". San Francisco, CA: Evelyn & Walter Haas, Jr. Fund and
the National Gay and Lesbian Task Force Policy Institute.

Eisenstadt, Shmuel N. 2003. From Generation to Generation. New Brunswick, NJ: Transaction
Publishers.

Eliasoph, Nina. 1998. Avoiding Politics: How Americans Produce Apathy in Everyday LIfe. New
York: Cambridge University Press.

Entman, Robert M. 1989. "How the Media Affect What People Think: An Information
Processing Approach." Journal of Politics 51:347-370.

Eskridge, Jr., William N., and Darren R. Spedale. 2006. Gay Marriage: For Better or for Worse?
Oxford: Oxford University Press.
   383  
Esler, Anthony. 1984. ""The Truest Community": Social Generations as Collective Mentalities."
Journal of Political and Military Sociology 12:99-112.

Evans, John H. 1996. ""Culture Wars" or Status Group Ideology as the Basis of US Moral
Politics." International Journal of Sociology and Social Policy 16:15-34.

—. 2002. "Polarization in Abortion Attitudes in U.S. Religious Traditions, 1972-1998."

Sociological Forum 17:397-422.

—. 2003. "Have Americans' Attitudes Become More Polarized? An Update." Social Science
Quarterly 84:71-90.

Evans, John H., Bethany Bryson, and Paul DiMaggio. 2001. "Opinion Polarization: Important
Contributions, Necessary Limitations." American Journal of Sociology 106:944-959.

Eyerman, Ron, and Bryan S. Turner. 1998. "Outline of a Theory of Generations." European
Journal of Social Theory 1:91-106.

Faith in Public Life. 2008. "The Young and the Faithful." in The Faith and Public Life Survey.

Foucault, Michel. 1972. The Archaeology of Knowledge and the Discourse on Language. New
York: Pantheon Books.

—. 1978. The History of Sexuality, Volume 1: An Introduction. New York: Vintage Books.

Gallagher, John, and Chris Bull. 2001. Perfect Enemies: The Battle Between the Religious Right
and the Gay Movement. Lanham, MD: Madison Books.

Gallagher, Maggie, and Joshua K. Baker. 2004. "Same-Sex Marriage: What Does the Next
Generation Think?". Washington, DC: Institute for Marriage and Public Policy.

Gamson, William A. 1992. Talking Politics. Cambridge [England] ; New York, NY, USA:
Cambridge University Press.

Gans, Herbert J. 1979. Deciding What's News: A Study of CBS Evening News, NBC Nightly
News, Newsweek, and Time. New York: Pantheon Books.

Geertz, Clifford. 1973. The Interpretation of Cultures; Selected Essays. New York,: Basic
Books.

Ghoshal, Raj. 2009. "Argument Forms, Frames, and Value Conflict: Persuasion in the Case of
Same-Sex Marriage." Cultural Sociology 3:76-101.

Gilleard, Chris. 2004. "Cohorts and Generations in the Study of Social Change." Social Theory
and Health 2:106-119.
   384  
Glaser, Barney, and Anselm L. Strauss. 1967. The Discovery of Grounded Theory: Strategies for
Qualitative Research. New York: Aldine de Gruyter.

Glenn, Norval D. 1977. Cohort Analysis. Beverly Hills, CA: Sage Publications.

—. 1980. "Values, Attitudes, and Beliefs." in Constancy and Change in Human Development,
edited by Orville G. Brim Jr. and Jerome Kagan. Cambridge, MA: Harvard University
Press.

Goode, Erich, and Nachman Ben-Yehuda. 1994. "Moral Panics: Culture, Politics, and Social
Construction." Annual Review of Sociology 20:149-171.

Griffin, Larry J. 2004. "'Generations and Collective Memory' Revisited: Race, Region, and
Memory of Civil Rights." American Sociological Review 69:544-577.

Gross, Larry. 2001. Up From Invisibility: Lesbians, Gay Men, and the Media in America. New
York: Columbia University Press.

Habermas, Jürgen. 1987. The Theory of Communicative Action, Volume 2. Boston: Beacon Press.

Harding, David J., and Christopher Jencks. 2003. "Changing Attitudes Toward Premarital Sex:
Cohort, Period, and Aging Effects." Public Opinion Quarterly 67:211-226.

Hareven, Tamara K. 1994. "Aging and Generational Relations: A Historical and Life Course
Perspective." Annual Review of Sociology 20:437-461.

Herek, Gregory M., and Eric K. Glunt. 1993. "Interpersonal Contact and Heterosexuals'
Attitudes Toward Gay Men: Results From a National Survey." Journal of Sex Research
30:239-244.

Hochschild, Arlie. 1989. The Second Shift. New York: Penguin.

Hout, Michael, and Claude S. Fischer. 2002. "Why More Americans Have No Religious
Preference: Politics and Generations." American Sociological Review 67:165-190.

Howe, Neil, and Bill Strauss. 1993. 13th Gen: Abort, Retry, Ignore, Fail? New York: Vintage
Books.

Howe, Neil, and William Strauss. 2000. Millenials Rising: The Next Great Generation. New
York: Vintage.

Hull, Kathleen E. 2001. "The Political Limits of the Rights Frame: The Case of Same-Sex
Marriage in Hawaii." Sociological Perspectives 44:207-232.

—. 2006. Same-Sex Marriage: The Cultural Politics of Love and Law. New York: Cambridge
University Press.
   385  
Hunter, James Davison. 1991. Culture Wars: The Struggle to Define America. New York: Basic
Books.

Iyengar, Shanto, and Donald R. Kinder. 1987. News That Matters: Television and American
Opinion. Chicago: University of Chicago Press.

Jackson, Robert Max. 1998. Destined for Equality: The Inevitable Rise of Women's Status.
Cambridge, MA: Harvard University Press.

Jennings, M. Kent, and Ning Zhang. 2005. "Generations, Political Status, and Collective
Memories in the Chinese Countryside." Journal of Politics 67:1164-1189.

Kertzer, David I. 1983. "Generation as a Sociological Problem." Annual Review of Sociology

9:125-149.

Krosnick, Jon A., and Donald R. Kinder. 1990. "Altering the Foundations of Support for the
President Through Priming." American Political Science Review 84:497-513.

Lamont, Michele. 1992. Money, Morals, and Manners: The Culture of the French and American
Upper-Middle Class. Chicago: University of Chicago Press.

Lannutti, Pamela J. 2005. "For Better or Worse: Exploring the Meanings of Same-sex Marriage
within the Lesbian, Gay, Bisexual and Transgendered Community." Journal of Social
and Personal Relationships 22:5-18.

Laufer, Robert S., and Vern L. Bengtson. 1974. "Generations, Aging, and Social Stratification:
On the Development of Generational Units." Journal of Social Issues 30:181-205.

Lax, Jeffrey R., and Justin H. Phillips. 2009. "Gay Rights in the States: Public Opinion and
Policy Responsiveness." American Political Science Review 103:367-386.

Lee, Barrett A., Chad R. Farrell, and Bruce G. Link. 2004. "Revisiting the Contact Hypothesis:
The Case of Public Exposure to Homelessness." American Sociological Review 69:40-63.

Lemm, Kristi M. 2006. "Positive Associations Among Interpersonal Contact, Motivation, and
Implicit and Explicit Attitudes Toward Gay Men." Journal of Homosexuality 51:79-99.

Lichterman, Paul. 1996. The Search for Political Community: American Activists Reinventing
Commitment. New York: Cambridge University Press.

Loftus, Jeni. 2001. "America's Liberalization in Attitudes Toward Homosexuality, 1973-1998."

American Sociological Review 66:762-782.

Lunt, Peter. 1996. "Rethinking the Focus Group in Media and Communications Research."
Journal of Communication 46:79-98.
   386  
Mannheim, Karl. 1952 (1928). "The Problem of Generations." Pp. 276-320 in Essays on the
Sociology of Knowledge, edited by Karl Mannheim. London: Routledge and Kegan Paul.

McCarthy, John D., Clark McPhail, and Jackie Smith. 1996. "Images of Protest: Dimensions of
Selection Bias in Media Coverage of Washington Demonstrations, 1982 and 1991."
American Sociological Review 61:478-499.

McConkey, Dale. 2001. "Whither Hunter's Culture War? Shifts in Evangelical Morality, 1988-
1998." Sociology of Religion 62:149-174.

McFarland, Katherine. 2008. "Moral Discourse, Political Culture, and the Debate over Same-Sex
Marriage." in Sociology. Chapel Hill: University of North Carolina.

McMullin, Julie Ann, Tammy Duerden Comeau, and Emily Jovic. 2007. "Generational Affinities
and Discourses of Difference: A Case Study of Highly Skilled Information Technology
Workers." British Journal of Sociology 58:297-316.

McPherson, Miller, Lynn Smith-Lovin, and James M. Cook. 2001. "Birds of a Feather:
Homophily in Social Networks." Annual Review of Sociology 27:415-444.

McVeigh, Rory, and Maria-Elena D. Diaz. 2009. "Voting to Ban Same-Sex Marriage: Interests,
Values, and Communities." American Journal of Sociology 74:891-915.

Mello, Michael. 2004. Legalizing Gay Marriage. Philadelphia: Temple University Press.

Miceli, Melinda S. 2005. "Morality Politics vs. Identity Politics: Framing Processes and
Competition Among Christian Right and Gay Social Movement Organizations."
Sociological Forum 20:589-612.

Miller, Allan S., and John P. Hoffmann. 1999. "The Growing Divisiveness: Culture War or War
of Words?" Social Forces 78:721-745.

Miller, Steven D., and David O. Sears. 1986. "Stability and Change in Social Tolerance: A Test
of the Persistence Hypothesis." American Journal of Political Science 30:214-236.

Mills, C. Wright. 2000. The Sociological Imagination. New York: Oxford University Press.

Mintz, Steven, and Susan Kellogg. 1988. Domestic Revolutions: A Social History of American
Family Life. New York: Free Press.

Mouw, Ted, and Michael E. Sobel. 2001. "Culture Wars and Opinion Polarization: The Case of
Abortion." American Journal of Sociology 106:913-943.

Nelson, Thomas E., Rosalee A. Clawson, and Zoe M. Oxley. 1997. "Media Framing of a Civil
Liberties Conflict and its Effects on Tolerance." American Political Science Review
91:567-583.
   387  
Oliver, Pamela E., and Daniel J. Myers. 1999. "How Events Enter the Public Sphere: Conflict,
Location, and Sponsorship in Local Newspaper Coverage of Public Events." American
Journal of Sociology 105:38-87.

Olson, Laura R., Wendy Cadge, and James T. Harrison. 2006. "Religion and Public Opinion
about Same-Sex Marriage." Social Science Quarterly 87:340-360.

Ortega y Gasset, José. 1958. Man and Crisis. New York: W.W. Norton & Company.

Perrin, Andrew J. 2006. Citizen Speak: The Democratic Imagination in American Life. Chicago:
University of Chicago Press.

Pew Research Center for the People and the Press. 2003. "Republicans Unified, Democrats Split
on Gay Marriage; Religious Beliefs Underpin Opposition to Homosexuality." in Pew
Forum on Religion and Public Life. Washington, D.C.: Pew Research Center for the
People and the Press.

—. 2006. "Only 34% Favor South Dakota Abortion Ban; Less Opposition to Gay Marriage,
Adoption and Military Service." Washington D.C.: Pew Research Center for the People
and the Press.

—. 2009. "Trends in Political Values and Core Attitudes: 1987-2009." Washington, DC: Pew
Research Center for the People and the Press.

Pilcher, Jane. 1994. "Mannheim's Sociology of Generations: An Undervalued Legacy." British

Journal of Sociology 45:481-495.

Polikoff, Nancy D. 2008. Beyond (Straight and Gay) Marriage: Valuing All Families Under the
Law. Boston: Beacon Press.

Polletta, Francesca. 2006. It was Like a Fever: Storytelling in Protest and Politics. Chicago:
University of Chicago Press.

Powell, Brian, Catherine Bolzendahl, Claudia Geist, and Lala Carr Steelman. 2010. Counted
Out: Same-Sex Relations and Americans' Definitions of Family. New York: Russell Sage.

Press, Andrea L., and Elizabeth R. Cole. 1999. Speaking of Abortion: Television and Authority in
the Lives of Women. Chicago: University of Chicago Press.

Price, Vincent, Lilach Nir, and Joseph N. Cappella. 2005. "Framing Public Discussion of Gay
Civil Unions." Public Opinion Quarterly 69:179-212.

Ridgeway, Cecilia L., and Lynn Smith-Lovin. 1999. "The Gender System and Interaction."
Annual Review of Sociology 25:191-216.

Riley, Matilda White. 1987. "On the Significance of Age in Sociology: American Sociological
Association, 1986 Presidential Address." American Sociological Review 52:1-14.
   388  
Riley, Matilda White, and John W. Riley Jr. 1999. "Sociological Research on Age: Legacy and
Challenge." Ageing and Society 19:123-132.

Rimmerman, Craig A., Kenneth D. Wald, and Clyde Wilcox. 2000. The Politics of Gay Rights.
Chicago: University of Chicago Press.

Roberts, Carl W., and Kurt Lang. 1985. "Generations and Ideological Change: Some
Observations." Public Opinion Quarterly 49:460-473.

Rosenfeld, Michael J. 2007. The Age of Independence: Interracial Unions, Same-Sex Unions,
and the Changing American Famil. Cambridge, MA: Harvard University Press.

Rubin, Herbert J., and Irene S. Rubin. 2005. Qualitative Interviewing: The Art of Hearing Data.
Thousand Oaks, CA: Sage Publications.

Ryder, Norman B. 1965. "The Cohort as a Concept in the Study of Social Change." American
Sociological Review 30:843-861.

Schiappa, Edward, Peter B. Gregg, and Dean E. Hewes. 2006. "Can One TV Show Make a
Difference? Will & Grace and the Parasocial Contact Hypothesis." Journal of
Homosexuality 51:15-37.

Schnittker, Jason, Jeremy Freese, and Brian Powell. 2003. "Who are Feminists and What do
They Believe? The Role of Generations." American Sociological Review 68:607-622.

Schuman, Howard, and Jacqueline Scott. 1989. "Generations and Collective Memories."
American Sociological Review 54:359-381.

Schwartz, Barry. 1996. "Memory as a Cultural System: Abraham Lincoln in World War II."
American Sociological Review 61:908-927.

Scott, Jacqueline. 2000. "Is it a Different World to When You Were Growing Up? Generational
Effects on Social Representations and Child-Rearing Values." British Journal of
Sociology 51:355-376.

Sears, David O., and Carolyn L. Funk. 1999. "Evidence of the Long-Term Persistence of Adults'
Political Predispositions." Journal of Politics 61:1-28.

Seidman, Steven. 2004. Beyond the Closet: The Transformation of Gay and Lesbian Life. New
York: Routledge.

Sewell, William H. Jr. 1999. "The Concept(s) of Culture." Pp. 35-61 in Beyond the Cultural
Turn, edited by Victoria E. Bonnell and Lynn Hunt. Berkeley, CA: University of
California Press.
   389  
Sigelman, Lee, Timothy Bledsoe, Susan Welch, and Michael W. Combs. 1996. "Making
Contact? Black-White Social Interaction in an Urban Setting." American Journal of
Sociology 101:1306-1332.

Sigelman, Lee, and Susan Welch. 1993. "The Contact Hypothesis Revisited: Black-White
Interaction and Positive Racial Attitudes." Social Forces 71:781-795.

Slater, Michael D., and Donna Rouner. 2002. "Entertainment-Education and Elaboration
Likelihood: Understanding the Processing of Narrative Persuasion." Communication
Theory 12:173-191.

Smith, Jackie, John D. McCarthy, Clark McPhail, and Boguslaw Augustyn. 2001. "From Protest
to Agenda Building: Description Bias in Media Coverage of Protest Events in
Washington, D.C." Social Forces 79:1397-1423.

Soule, Sarah A. 2004. "Going to the Chapel? Same-Sex Marriage Bans in the United States,
1973-2000." Social Problems 51:453-477.

Stacey, Judith. 1990. Brave New Families: Stories of Domestic Upheaval in Late-Twentieth
Century America. Berkeley, CA: University of California Press.

Stein, Arlene. 2001. The Stranger Next Door: The Story of a Small Community's Battle Over Sex,
Faith, and Civil Rights. Boston: Beacon Press.

Strauss, William, and Neil Howe. 1991. Generations: The History of America's Future, 1584 to
2069. New York: Harper.

Swidler, Ann. 1986. "Culture in Action: Symbols and Strategies." American Sociological Review
51:273-286.

—. 2001. Talk of Love: How Culture Matters. Chicago: University of Chicago Press.

Taylor, Verta, Katrina Kimport, Nella Van Dyke, and Ellen Ann Andersen. 2009. "Culture and
Mobilization: Tactical Repertoires, Same-Sex Weddings, and the Impact on Gay
Activism." American Sociological Review 74:865-890.

Thornton, Arland, and Linda Young-DeMarco. 2001. "Four Decades of Trends in Attitudes
Toward Family Issues in the United States: The 1960s Through the 1990s." Journal of
Marriage and Family 63:1009-1037.

Treas, Judith. 2002. "How Cohorts, Education, and Ideology Shaped a New Sexual Revolution
on American Attitudes Toward Nonmarital Sex, 1972-1998." Sociological Perspectives
45:267-283.

Tuchman, Gaye. 1978. Making News: A Study in the Construction of Reality. New York: Free
Press.
   390  
Twenge, Jean M. 2006. Generation Me: Why Today's Young Americans are More Confident,
Assertive, Entitled--and More Miserable Than Ever Before. New York: Free Press.

Vincent, John A. 2005. "Understanding Generations: Political Economy and Culture in an

Ageing Society." British Journal of Sociology 56:579-599.

Wadsworth, M.E.J., and S.R. Freeman. 1983. "Generational Differences in Beliefs: A Cohort
Study of Stability and Change in Religious Beliefs." British Journal of Sociology 34:416-
437.

Walsh, Katherine Cramer. 2004. Talking About Politics: Informal Groups and Social Identity in
American Life. Chicago: University of Chicago Press.

Walters, Suzanna Danuta. 2001. All the Rage: The Story of Gay Visibility in America. Chicago:
University of Chicago Press.

Weil, Frederick D. 1987. "Cohorts, Regimes, and the Legitimation of Democracy: West
Germany Since 1945." American Sociological Review 52:308-324.

Whalen, Jack, and Richard Flacks. 1989. Beyond the Barricades: The Sixties Generation Grows
Up. Philadelphia: Temple University Press.

Wilcox, Clyde, and Robin Wolpert. 2000. "Gay Rights in the Public Sphere: Public Opinion on
Gay and Lesbian Equality." Pp. 409-432 in The Politics of Gay Rights, edited by Craig A.
Rimmerman, Kenneth D. Wald, and Clyde Wilcox. Chicago: University of Chicago
Press.

Wilson, Thomas C. 1994. "Trends in Tolerance Toward Rightist and Leftist Groups, 1976-1988:
Effects of Attitude and Cohort Succession." Public Opinion Quarterly 58:539-556.

Wohl, Robert. 1979. The Generation of 1914. Cambridge, MA: Harvard University Press.

Wood, Peter B., and John P. Bartkowski. 2004. "Attribution Style and Public Policy Attitudes
Toward Gay Rights." Social Science Quarterly 85:58-74.

Wuthnow, Robert. 1976. "Recent Pattern of Secularization: A Problem of Generations?"

American Sociological Review 41:850-867.

Wyatt, David. 1993. Out of the Sixties: Storytelling and the Vietnam Generation. New York:
Cambridge University Press.

Zaller, John R. 1992. The Nature and Origins of Mass Opinion. New York: Cambridge
University Press.