Abstract
Drawing on Durkheim’s notion of effervescence, this paper explains the link between the increase in comfort in showing public affection during Mardi Gras for gay couples compared to the decrease in comfort in showing public affection where they live. It presents three major findings: (1) gay couples are more likely to feel comfortable showing public affection during Mardi Gras than in places where they live, compared to heterosexual couples; (2) gay couples who come from smaller cities are more likely to feel comfortable showing public affection during Mardi Gras compared to gay couples who come from larger cities; (3) gay couples who experience ‘effervescence’ during Mardi Gras are more likely to show affection in public during this event compared to gay couples who did not experience it.

Introduction: Mardi Gras, Effervescence, and Public Affection
Recent interest in civil society has brought the model of carnival to the fore of theoretical debates. Tucker (1996), a leading proponent of this model, advocates turning to carnival as a means to cultivate the aesthetics of civil society. Tucker argues that carnivals transfigure local social spaces and refashion civil society. Similarly, Jacobs and Smith (1997) have recently introduced the concept of carnival in sociology to understand how to change the aesthetics of public space and civil society. They argue that carnivals generate what Durkheim referred to as collective effervescence. Effervescence is a collective state of exaltation that influences people to think and act differently outside the structure, regulation, and routine of everyday life (Durkheim 1995, p. 220). Collective effervescence promotes difference, suspends hierarchical rules, and cultivates solidarity for local groups.

Another proponent of the model of carnival is Smith (1999), who provides the most systematic effort up to date to link effervescence with public space and civil society. Smith argues that a Durkheimian spatial paradigm that concentrates on producing collective effervescence as a resource for social change could be used as springboard to alter the meanings of public space rather than simply reproduce dominant meanings of it. According to Smith, change occurs because collective effervescence is a powerful social force that modifies the meanings, symbols, and subjectivity of people who occupy the settings where effervescence occurs. In this view, collective effervescence functions as a social resource to change dominant views of public space and civil society.

Still, the link between carnival and collective effervescence as a means to cultivate the aesthetics of public space and civil society remains undertheorized by contemporary sociologists. To date, sociologists have not provided a theoretical or empirical analysis of how collective effervescence can alter the aesthetics of public space and civil society and, as a result, cultivate different norms that changes the behaviors and attitudes of people.

In this paper we attempt to explore how collective effervescence during the Mardi Gras carnival in New Orleans transfigures civil society by cultivating a space that produces different sets of norms that govern the display of public affection, such as holding hands, kissing, and hugging, in public space. Specifically, we address how effervescence during Mardi Gras counters the heterosexual identity of space and the informal norms of heterosexuality placed on the display of public affection by gays and lesbians.

Informal normative constraints on public affection occur when specific groups are organized and classified around a distribution of ‘acceptable’ conduct imputed as normal, such as affectionate displays of kissing in public. Public affections that depart from what is ‘normal’ in certain public spaces are judged, assessed, and seen as irregular or abnormal. Gay couples in particular are subject to such normative constraints (Duncan 1996; Myslik 1996; Gamson 1998). Display of public affection among gay couples is often deemed as ‘abnormal’ and thus goes unseen in many parts of civil society, especially rural regions. Gay public affection is regulated and monitored by informal norms of heterosexuality, which results in enhancing the public visibility of heterosexual affection. These norms enforce a ‘normal’ status and natural quality to heterosexual affection, namely, the ‘heterosexual couple’ who kiss and hold hands while safely walking the streets in public spaces. Public affection that departs from the informal norms of heterosexuality is assigned a ‘polluted’ status and remains invisible due to restrictions and lack of spaces of difference.

In the present study, we use survey data collected during Mardi Gras to demonstrate how collective effervescence during Mardi Gras disrupts informal norms of heterosexuality and temporarily relieves the restrictions that such norms place on showing affection in public spaces. Furthermore, our data suggests that these processes interact with the social structural characteristics of people’s everyday experiences. Our approach can be seen as attempt to build on prior theories of civil society that use carnival as a framework to enhance the aesthetics of civil society and public life.

Theoretical Framework: The Carnival Sphere
Durkheim provides the most influential theoretical link between carnival, effervescence, and public affection in The Elementary Forms of Religious Life. Durkheim (1995, p. 217) argues that effervescence serves as an exceptionally powerful experience that generates closeness and exaltation among people. Effervescence influences people who come together during carnivals to think and act differently outside mundane life (Durkheim 1995, p. 220; Smith 1999). The coming together of people to in this public arena cultivates emotional energy for people while participating in collective activities with each other (Durkheim, 1995; Collins 1992). Collective effervescence is an intense phenomenon that generates emotions to provide closeness and comfort among those who experience it (Durkheim, 1995, p. 218). The intensity of the moment facilitates the displaying of emotions and the shedding of normal ways of behaving. Collective effervescence often produces so much excitement that it moves people to transform the informal norms that guide behavior in public spaces. The development of informal norms in civil society based on emotion and affection during carnival transform public spaces into what we refer to as a ‘carnival sphere’.

We define a carnival sphere as a space where people meet in transitory or suspended moments to celebrate with each under conditions that emphasize difference. Carnival spheres are spaces where people unite to participate in emotional events such as laughter, holding hands, singing, hugging, dancing, talking, and kissing with the reduced likelihood of being subject to coercion or threats for doing so. According to Durkheim (1995, p. 218), this type of sphere generates collective effervescence to the point where people “feel a certain need to set themselves above and beyond ordinary morality. The sexes come together in violation of the rules governing sexual relations (Durkheim 1995, p. 218). Carnival spheres are “special places where everyday rules of life are seen as being held in abeyance. They are marked by ludic forms of behavior and emotional constellations which suspend the traditional moralities” (Smith 1999, p. 16).

When people experience effervescence they act differently, feel differently, and participate in behaviors and events that they normally cannot do (Durkheim 1995, p. 220). Most importantly, carnival spheres produce spaces where marginal groups can be a part of a place where they can articulate themselves, show affection, and be visible (Hetherington 1998, p. 72). In short, the collective effervescence that develops in carnival spaces can challenge the informal norms of heterosexuality when it creates public spaces where affectionate displays by gay people occur. These spaces “have some form of symbolic attachment for particular groups when they express themselves for the purposes of developing identification and solidarity with others of a similar mind” (Hetherington 1998, p. 72).

Elementary Forms provides a framework to examine how some gay couples who experience collective effervescence during Mardi Gras feel more comfortable participating in public affection during this festival compared to where they live. Starting with Durkheim's concept of collective effervescence, we argue that effervescence is not based on rules, regulations, or reason, but the development of emotions, feelings of awe, and solidarity through affectionate interaction rituals that transform informal norms of heterosexuality, public space, and civil society. Collective effervescence, then, provides a starting point to rethink civil society and the identity of space.

Informal Norms of Heterosexuality: Sexualized Spaces
Duncan has explored how everyday public space has a heterosexual identity. She argues that “it is assumed that sexuality is confined to private spaces. This is based on the naturalization of heterosexual norms” (Duncan 1996, p. 136). The result is that “naturalized heterosexuality makes sexuality in public spaces nearly invisible to straight populations” (Duncan 1996, p. 137). Informal norms of heterosexuality regulate public space, nonheterosexual public affection, and reinforce ‘appropriate’ conduct. Consequently, many gay couples rarely show public affection outside the context of the home. This is because public space is subject to various territorializing processes whereby local control fixes, claims, and monitors affection (Duncan 1996, p. 129). For example, public space is regulated by marginalizing groups or “keeping it free of passion or expressions of sexuality that are not naturalized, normalized or condoned” (Duncan 1996, p. 141). The aim of the regulation, normalization, and control of the spatial terrain of streets, as well as public affection, is to avoid any ‘disruptive’ or ‘lewd’ expressions by producing citizens who police themselves. The consequence is that informal norms of heterosexuality govern and restrain all forms of conduct that could be seen as resistance or out of place. Consequently, gay affection becomes excluded, marginalized, or concentrated in spatial locations outside of heterosexual space. In fact, it is often taken for granted that the identity of most public space is heterosexual. Myslik (1996) and Gamson (1998) illustrate how most heterosexual people claim that gays are ‘flaunting it’ when they make their affection or sexuality visible in public spaces. The assumption in this statement is that heterosexuals do not flaunt their sexuality outside the private sphere (Myslik 1996). However, “engagement announcements, wedding ceremonies, booking a double bed at a hotel, shopping together for a new mattress, casual references in conversation to a husband or wife, a brief peck on the cheek when greeting or leaving a spouse, photos on desks at work, and holding hands at the beach are all public announcements and affirmations of heterosexuality. The ‘normality’ of heterosexuality is so deeply ingrained in Western culture that it is not even seen” (Myslik 1996, p. 159). Once the taken for granted assumptions of ‘neutral’ space are exposed, the expression of heterosexualized public space can be seen.

We embrace the notion that public space has a sexual identity and can reproduce power relations, yet it is also possible to explore how collective effervescence transforms public space into an emotional atmosphere that enables marginal people to be visible. Bech (1999, p. 229) argues that the sociological study of sexuality and public space “should not ignore the emotional or sensual character of its subject matter.” Public space, then, is not only an emotional social institution that regulates intimate conduct, but also a site where relations of power can be altered to resist the informal norms of heterosexuality. For example, unlike gays and lesbians who live in large urban areas, those living in small nonurban cities have fewer spatial opportunities to show public affection compared to large urban cities (Castells 1983). Small, homogeneous cities with small gay and lesbian populations, especially those located in rural areas, subject gays and lesbians to the normative consensus of the dominant heterosexual groups.

One reason for this subjection is that urban spaces with high populations produce and sustain a mosaic of public behaviors, whereas nonurban spaces are less likely to produce these characteristics (Fischer 1975, p. 1323). In other words, the larger the population of a city, the greater the number of diverse persons who settle there. The result is that public spaces of difference develop in urban cities where people can participate in activities they may not be able to do in nonurban cities, such as display gay affection in public. Heterogeneous cities with large gay and lesbian populations weaken the informal norms of heterosexuality that exercise social control on public behavior. Because a larger population size encourages spatial difference and alternative lifestyles, the obstacle to “discovering friends and leading an unharassed, open life [decreases]” (Castells 1983, p. 145). Cities with large populations are more likely to comprise of people who are defined as living different lifestyles and can therefore mobilize and become visible in public space (Castells 1983).

Another result for marginalized groups who live in urban cities with large populations is that they can come out collectively and be safe together when showing public affection (Castells 1983). These behaviors are likely to “expand with the increasing capacity of gay people to defend themselves and to build up a series of autonomous institutions” (Castells 1983, p. 139). In fact, Castells (1983, p. 143-62) notes that in cities with large gay populations, like San Francisco, a comfortable environment to publicly live gay was developed around carnival spheres such as street celebrations, feasts, festivals, and carnivals.

“The gay community has done more through these increasingly diverse array of events than project its values or have fun: it has shown the city that streets are for people, that urban culture means gathering together to play in public places, and that music, politics, and games can intertwine in a revitalizing way, creating a new media for messages and establishing new networks of communication” (Castells 1983, p. 162).

Cities with a smaller population size are less likely to have safe-spaces where gays and lesbians can collectively organize carnivals to transform civil life into a cosmopolitan space. One reason for a lack of gay space in cities with small populations identified in this article is because heterosexual religious groups can intervene into gay couples’ lives attempt to ‘convert’ them.
One reason for unwanted religious intervention and lack of queer space is because heterosexual religious groups in spatially homogenous cities can effectively intervene into many spatial settings of the city and impose their moral beliefs on gays and lesbians, as well as the community, rather than encourage spatial difference. Some religious groups hold public events and actively intervene into the personal lives of gays. Their attempt is to persuade gays and lesbians to ‘repent’ and change their behavior.

Hypotheses for the Quantitative Analysis
People experiencing structural regulation in their everyday life are confronted with unwanted restrictions in their life. Among these are normative restrictions regarding their public behavior. We argue that the collective life during Mardi Gras temporary relieves many of these restrictions. This means that the indicators of everyday structural regulation should be positively related to reporting an increase in comfort in displaying public affection during Mardi Gras. In this line of thought, it may also be the case that people experiencing unwanted restrictions in their everyday life are more open to experiencing collective effervescence during Mardi Gras.

We discuss above that everyday public life is guided by the norms of heterosexuality. Thus, gays are much more likely than are nongays to confront normative restrictions on expressing their sexual identity, for example, in regards to displaying affection toward their partners in public. If this is true, and that the collective life at Bourbon Street during Mardi Gras relieves these normative constraints, then the degree of increase in comfort in showing public affection during Mardi Gras should be greater among gays than nongays. Gays should report a greater release from normative constraints than nongays when participating in the collective life during Mardi Gras.

We argue that collective effervescence relieves some of the normative controls that people confront in their everyday life. On Bourbon Street during Mardi Gras, collective effervescence creates a space that relieves normative control and promotes difference. Taking part in this collective experience momentarily frees people from the normative constraints of everyday life. This line of thought implies particular empirical relationships. First, it implies that the statistical effect of having a gay identity on the increase in comfort in showing public affection during Mardi Gras should be contingent on the degree of effervescence experienced during Mardi Gras. Specifically, the statistical relationship between having a gay identity and reporting an increase in comfort in showing public affection should be stronger among persons experiencing higher levels of collective effervescence.

We have provided an argument suggesting that the intensity of normative heterosexuality varies directly with social indicators of the prevalence of normative constraints on sexuality, such as population size and the degree of religious intervention in hometown/city. With increased levels of structural regulation, gays should be confronted with a public space guided by more intense norms of heterosexuality. For these persons the collective life during Mardi Gras should lead to a greater release of unwanted regulation than among those reporting less structural regulation in their everyday life. Therefore, having a gay identity should positively correlate with increase in comfort in showing public affection during Mardi Gras; but this correlation should be stronger as people experience a greater structural regulation in their everyday life.

Data and Methods
The data used in this study were collected in New Orleans during Mardi Gras by interviewing and surveying a convenience sample of 102 (N=102) adults on Bourbon Street. We obtained a sample of fifty-one citizens who identified themselves as gay and fifty-one citizens who identified themselves as straight. The purpose of this type of quota sample was to compare differences between the former group with the latter. It was impractical to obtain either a random or representative sample of gays since it is impossible to know how many gays currently live in America, how many travel to Mardi Gras, or to know the size or universe for both (see Myslik 1996). Although the results cannot be generalized, they do apply to the sample in this study.

Measures
The dependent variable. The dependent variable indicates the change in comfort in participating in public affection from home to Mardi Gras. This variable is created by subtracting the degree of discomfort at home from the degree of discomfort experienced during Mardi Gras, thus providing an indicator of the increase in comfort in participating in public affection during Mardi Gras. The construct “comfort in showing public affection” refers to the degree to which a person feels comfortable holding hands, hugging, or kissing in public places. The construct is operationalized and measured using a Likert-scale, ranging from strongly agree to strongly disagree.

“Public affection” is defined as the degree to which a person feels comfortable holding hands, hugging, or kissing in public places. The following indicators were used to measure level of comfort in showing public affection: (1) I feel comfortable holding my partner’s hand in public places where I live, (2) I feel comfortable kissing my partner in public places where I live, (3) I feel comfortable hugging my partner in public places where I live, (4) I feel comfortable holding my partner’s hand in public places during Mardi Gras, (5) I feel comfortable kissing my partner in public places during Mardi Gras, and (6) I feel comfortable hugging my partner in public places during Mardi Gras. The answer scores were summed up in a scale (Cronbach’s Alpha is .95).
Also, respondents were asked how comfortable they felt doing each of these things in public during Mardi Gras. The answer scores to these questions were also summed in a scale (Cronbach’s Alpha is .90). The two scales, that is, comfort in showing public affection at home and at Mardi Gras, were then subtracted from each other, respectively. The product, which is the dependent variable in the analyses below, thus indicates the increase (or decrease) in comfort in showing public affection when moving from one social-spatial context (home) to the other (Mardi Gras).

The independent variables. A dichotomous variable indicates a person’s sexual identity.
Respondents were asked the question “What is your sexual preference”. Those who indicated that they were “Gay”, “Lesbian”, “Bisexual”, or “Other (please specify)” were given the value one. Those who indicated that they were “straight” were given the value zero. Fifty-one individuals were thus given the value one, which is fifty percent of the sample. Females are coded zero and males are coded one.

We designed the questionnaire to determine whether the population of the city in which the couples lived influenced their level of comfort in showing public affection. To accomplish this we asked the respondents to identify the city and state they live in. We used the 1990 census to locate the population of the city where they live and coded the variable accordingly. Population in person’s home city was measured by the city and state in which the respondents live. After collecting the data, we recorded the population for each city by using the 1990 census data. A total of 74 cities were recorded. The mean age for the sample is 30.12 and the standard deviation is 8.83.

The construct collective effervescence is defined as a collective state of exaltation that assists people in thinking and acting differently outside the structure, regulation, and routine of everyday life (Durkheim 1995, p. 220). The following eight indicators were used to measure effervescence: “I feel a sense of belonging with the people at Mardi Gras”; “Mardi Gras is a place where I can freely express my emotions without having to feel embarrassed”; “During Mardi Gras, I don’t have to live up to the moral expectations set by other people”; “Compared to my everyday life, the environment at Mardi Gras allows me to be freer”; “During Mardi Gras, I don’t feel obligated to behave the way people usually expect me to behave”; “During Mardi Gras, I can be anyone I want to be”; “Mardi Gras is a good way to get away from my everyday routine”; and “Mardi Gras is a real stress reliever”. These indicators were based on four Likert-scale responses ranging from strongly agree to strongly disagree. The answer scores to the eight questions were summed in a scale. The scale has a relatively high reliability (Cronbach’s Alpha is .80).

Sense of belonging is defined as the degree to which a person feels a sense of belonging where he or she lives. The following indicator was used to measure sense of belonging: “I feel a sense of belonging where I live.”

Seven indicators were used to measure the independent variable “structural regulation:” “There are too many moral expectations to live up to in my life”; “There are too many annoying rules to follow in my life”; “I do not believe in most of the rules I follow in life”; “Most of the rules set by other people in my life are a burden”; “I follow many routines in life that I really do not want to follow”; “Other people place so many expectations on me that I cannot live up to them”; and “I have too many standards of behavior imposed on me in everyday life”. These indicators were based on four Likert-scale responses ranging from strongly agree to strongly disagree. The answer scores to the eight questions were summed in a scale. The scale has a high reliability (Cronbach’s Alpha is .86).

Two indicators were used to measure unwanted religious intervention in respondents’ home: “Too many religious people expect me to behave according to their expectations”; and “Religious groups try to influence my lifestyle against my will”. All indicators are Likert items based on strongly agree to strongly disagree (see Appendix A). The answer scores to the eight questions were summed in a scale. The Pearson’s correlation between the two items is .78.

Table 1. Descriptive statistics.

a Comfort in showing public affection during Mardi Gras minus comfort in showing public affection in hometown/city.

Results
To test the hypotheses, we use ordinary least square regression analyses. First, we regress the experience of collective effervescence during Mardi Gras on a set of predictors. Secondly, we regress the degree of increase in comfort in showing public affection during Mardi Gras on a set of predictors, in an attempt to see how the independent variables relate to the release of normative constraints experienced by some people during Mardi Gras. Finally, we explore the presence of statistical interactions between collective effervescence, gay identity, and the other independent variables, in order to assess the centrality of collective effervescence as the social mechanism that releases normative constraints during Mardi Gras.

In Table 2, effervescence during Mardi Gras is regressed on gay and lesbian identity and the indicators of structural regulation in everyday live, controlling for age and gender. As predicted, the indicators of structural regulation are positively related to experiencing effervescence during Mardi Gras. People are more likely to experience effervescence during Mardi Gras with increased level of structural regulation in their everyday lives. Thus, as people perceive religious groups to have more unwanted influence in their home town/city, the more likely they are to experience effervescence at Mardi Gras. Similarly, experiencing unwanted regulation in everyday life is significantly, positively related to experiencing effervescence at Mardi Gras. Age, gender, and sexual identity, on the other hand, are not significantly related to experiencing effervescence at Mardi Gras.

Table 2. Regressing effervescence during Mardi Gras on independent variables.

In Table 3, the dependent variable is the difference between reported discomfort in participating in public affection at home and the discomfort in participating in public affection at Mardi Gras. Thus, this variable indicates the increase in comfort in participating in public affection when moving from everyday life to Bourbon Street during Mardi Gras. In model I, the increase in comfort in participating in public affection is regressed on gay identity, indicators of structural regulation, and gender and age. The findings support our predictions. First, gay couples are significantly more likely than are different sex-couples to experience an increase in comfort in showing public affection during Mardi Gras. The relationship is moderately strong (Beta=.32). Thus, when gay persons enter Mardi Gras, they are more likely than are nongays to feel an increase in comfort in showing affection in public. Secondly, two of the indicators of the strength of heterosexual norms governing the public space of hometown/city are positively related to the increase in comfort in showing public affection during Mardi Gras. Thus, people who perceive greater religious domination in their hometown/city are significantly more likely to experience an increase in comfort in showing public affection during Mardi Gras (Beta=.17). Also, people from larger towns/cities experience, on the average, less increase in comfort in showing affection in public during Mardi Gras (Beta=-.16). Finally, people who experience a greater sense of belonging in their hometown/city are less likely to experience an increase in comfort in showing public affection during Mardi Gras (Beta=-.30). Reported experience of effervescence during Mardi Gras is added to the equation in Model II. The findings show that experiencing collective effervescence during Mardi Gras is positively related to an increase in comfort in showing public affection (Beta = .30). As people experience higher levels of collective effervescence during Mardi Gras, their level of comfort in showing public affection increases, on the average.

Table 3. Regressing the increase in comfort in showing public affection during Mardi Gras on independent variables.a

a The dependent variable is calculated: Comfort in showing public affection in context 2 (Mardi Gras) minus comfort in showing public affection in context 1 (at home).

In summary, the findings are consistent with our argument that Mardi Gras creates a space that relieves some of the regulations of normative heterosexuality and other normative constraints in everyday life. The findings in Table 3 are consistent with our argument that the collective life during Mardi Gras creates a space relieving some powerful normative restrictions of everyday life. First, normative restrictions in everyday life correlate positively with increased comfort in showing public affection during Mardi Gras. Second, gays experience a greater increase in comfort in showing public affection than do nongays. Finally, net of the indicators of structural regulation and sexual identity, experiencing collective effervescence during Mardi Gras is positively related to the increase in comfort in showing public affection during Mardi Gras.

If collective effervescence during Mardi Gras is the social process through which the release from the normative constraints occurs, two additional empirical findings should hold. First, having a gay identity should have differential effect on the release from the normative constraints of heterosexuality, depending on how strongly collective effervescence is felt during Mardi Gras. In comparison to nongays, the tendency for gays to experience an increase in being comfortable with displaying affection should be stronger as they report experiencing greater effervescence. Secondly, experiencing structural/normative constraints in everyday life should have differential effect on the release from these constraints, depending upon sexual identity. As normative regulation in everyday life increases—as indicated by population size, unwanted religious intervention, and lack of belonging in home community—having a gay identity should have a stronger positive effect on the increase in comfort in showing public affection during Mardi Gras.

To test these crucial hypotheses, five multiplicable interaction terms were created by multiplying gay identity by 1) collective effervescence, 2) unwanted religious intervention, 3) feeling a sense of belonging where one lives, 4) structural regulation, and 5) population size. Following Jaccard et al (1990), we centered the variables around its mean before creating the interaction terms.

The findings are presented in Table 4, where the increase in comfort in showing public affection is regressed on gay identity, the indicators of structural regulation, and the control variables. Moreover, in columns I through V the interaction terms are added to the model each at a time. The results show that, when added one at a time, all the interaction terms have significant effects on the dependent variable. First, the findings in column I show that the effect of having a gay identity on the increase in comfort in showing public affection is contingent on the level of effervescence reported during Mardi Gras. Thus, as effervescence is experienced more strongly during Mardi Gras, the positive relationship between having a gay identity and the increase in comfort in showing public affection during Mardi Gras increases considerably. This is consistent with our argument that experiencing effervescence during Mardi Gras reduces the normalization of heterosexual space.

Secondly, as shown in columns II through V, the indicators of structural regulation interact positively with having a gay identity. That is, the positive effect of having a gay identity on the increase in comfort in showing public affection, increases by increased structural regulation in everyday life. Therefore, gays who experience structural regulation in everyday life, experience greater increase in comfort in showing public affection during Mardi Gras, compared to gays who experience less structural regulation. These effects are significant for all the three indicators of domination, that is, religious domination, structural regulation, and population size. Thus, with decreased population size in people’s home town/city, the relationship between sexual identity and the increase in comfort in showing public affection increases. Gays coming from a smaller population report greater increase in comfort than do gays from larger populations. The same finding applies to unwanted religious intervention and the measure of regulation. The effect of having a gay identity on the increase in comfort in showing public affection is greater as these indicators of unwanted intervention are more intense. Thirdly, gays who have a greater sense of belonging in their home town/city, experience less increase in comfort in showing public affection during Mardi Gras, compared to gays who have less sense of belonging in their everyday life. In other words, gays who have a weaker sense of belonging at home experience more increase in comfort in showing public affection during Mardi Gras than do gays who experience a strong sense of belonging.

These findings thus support our central argument that the collective social forces during Mardi Gras release the normative constraints of everyday life. Having a gay identity is positively related to such a release during Mardi Gras, but more so with increased reported effervescence during Mardi Gras and with increased structural regulation in everyday life.

Finally, in column VI all the interaction terms are added to the model simultaneously. Controlling for all the interaction terms, the interaction effects between having a gay identity, on the one hand, and effervescence, belonging at home, and population size, on the other, remain significant. Together the interaction effects add 10 percent to the explanatory power of the direct effect model. Thus, not only are the statistical interactions significant and consistent in the direction predicted, but they are quite strong. When all the interaction terms are controlled simultaneously three interactions emerge as significant: gay identity times effervescence, gay identity times population size and gay identity times belonging in hometown/city.

Table 4. Testing for statistical interactions: Regressing the increase in comfort in showing public affection during Mardi Gras on dependent variables and interaction terms.

a The dependent variable is calculated: Comfort in showing public affection during Mardi Gras minus comfort in showing public affection in hometown/city.

Conclusion
In this article we have illustrated that collective effervescence during Mardi Gras temporarily relieves many unwanted restrictions in life. In other words, people who experience unwanted regulations in their everyday life are more open to experiencing collective effervescence during Mardi Gras. We also discussed above that everyday space in public life is guided by the norms of heterosexuality. Thus, gay couples are much more likely than heterosexual couples to confront normative restrictions on showing public intimacy with their partner in public. Once these same people enter the collective life during Mardi Gras, however, these norms are relieved. As a result, the degree of increase in comfort in showing public affection during Mardi Gras was greater among gay couples than heterosexual couples.

The changing nature of the Mardi Gras carnival has given rise to a public arena where desire and affection take place in certain social spaces. An important social space identified in this paper is the carnival sphere. These intimate moments between people who are often invisible in the public sphere are part of an expanded and redefined cultural arena. Moreover, it is important to locate other carnival spheres that may provide public places for groups such as gays, lesbians, transgenders, and queers to assemble.

While many people may celebrate the changing and developing nature of the public sphere into a carnival sphere, these changes are far from liberatory. For instance, many sexually marginalized people are still concerned with the quality of public life. For this reason, they attempt to engender new public spaces, such as carnival spheres, to enhance their quality of life, namely, being able to walk in public places and show affection without getting attacked or assaulted. Though the carnival sphere is quickly developing alongside the public sphere, this does not imply that there will be a proliferation of public spaces for sexually marginalized groups to fully participate in public intimate life. For this to occur, several necessary elements identified in this article are needed.

The first element is collective effervescence. Collective effervescence relieves many normative regulations that sexual minorities confront in their everyday life. For instance, we have explained how collective effervescence cultivates a space that relieves normative control and promotes difference. Participating in this collective experience momentarily frees people from the normative constraints of everyday life. This line of thought implies several particular important conclusions.

First and foremost, it implies that the effect of having a gay identity on the increase in comfort in showing public affection during Mardi Gras is contingent on the degree of effervescence experienced during Mardi Gras. Specifically, the statistical relationship between having a gay identity and experiencing an increase in comfort in showing public affection is stronger among persons experiencing the collective effervescence. In other words, not all gays experience an increase in comfort in showing public affection upon entering Bourbon Street during Mardi Gras, but those who do are likely to come from smaller cities where the identity of most space is heterosexual. Collective effervescence, then, disrupts heterosexual space and promotes difference, thereby accommodating plural groups rather than a few.

Moreover, we pointed out that the intensity of normative heterosexuality varies by indicators such as population size and the degree of unwanted religious intervention where people live. As the degree of structural regulation increases, gay couples encounter a public space guided by more intense and homogenous norms of heterosexuality where they live. The collective life at Bourbon Street during Mardi Gras, however, relieves unwanted regulation for gay couples more than heterosexual couples. In other words, gay couples report more structural regulation in their everyday life whereas heterosexual couples report less structural regulation. As a result, being gay correlates with an increase in comfort in showing public affection during Mardi Gras compared to heterosexual couples. This correlation, however, is stronger as people experience greater structural regulation in their everyday life.

And the final conclusion drawn is that carnivals matter---carnivals influence the concrete actions that people display on a daily basis. These “different behaviors," however, are not the result of "individual decisions." Instead, difference is highly influenced by social---not individual---facts such as: the degree of regulation, population size, having a sense of belonging, religious domination, experiencing collective effervescence, space, and sexual identity. These important variables all influence something as simple as the degree of comfort all people feel showing affection in public places, during both carnival and noncarnival times.

We believe that gay public affection enhances the aesthetics of public space and produces new sets of norms that change the dominant ones that govern behavior in public space. However, our model does not advocate a nostalgic ‘return’ to or ‘loss’ of a prior solidarity where society was once mythically ‘bonded’ and is now ‘fragmented’ into disorganized groups who bowl alone. Rather, our model asks, What is one way, among many, that civil society could look like? How can this one ‘look’ be accomplished in addition to numerous others? We believe that Durkheim’s notion of collective effervescence can be a political tool that enhances the cultivation and realization of our particular look. In particular, collective effervescence provides the public sphere with new strengths, vibrancy, and vitality.

Works Cited
Bech, Helmet. 1999. "Citysex: Representing Lust in Public." Theory, Culture, and
Society, 15, 3: 215-242.
Castells, Manuel. 1983. The City and the Grassroots: A Cross-Cultural Theory of Urban
Social Movements. Berkeley: University of California Press.
Collins, Randall. 1992. "Sociology of Emotions as the Solution to Rational Choice
Anomalies." Paper presented at the International Sociological Association.
Duncan, Nancy. 1996. Body/Space. New York, Routledge.
Durkheim, Emile. 1995. The Elementary Forms of Religious Life. Translated by Karen
Fields. The Free Press.
Fischer, Claude. 1975. "Toward a Subcultural Theory of Urbanism." American Journal of
Sociology, 80: 1319-1341.
Gamson, Joshua. 1998. Freaks Talk Back. Chicago: University of Chicago Press.
Hetherington, Kevin. 1998. Expressions of Identity. London, Sage.
Jaccard, James, Robert Turrisi and Choi K. Wan. 1990. Interaction Effects in Multiple
Regression. Newbury Park, CA: Sage.
Jacobs, Ronald. and Phillip Smith. 1997. "Romance, Irony, and Solidarity." Sociological
Theory 15, 1: 60-80.
Myslik, Wayne. 1996. "Negotiating the Heterosexual Street," in Body Space, ed. N.
Duncan. New York, Routledge.
Smith, Phillip. 1999. "The Elementary Forms of Place and Their Transformations: A
Durkheimian Model." Qualitative Sociology 22, 1: 13-36.
Tucker, Kenneth. 1996. "Harmony and Transgression: An Aesthetic Imagery and the Public Sphere in Habermas and Poststructuralism." Current Perspective in Social Theory, 16: 101-120.