Identity enactment as collective accomplishment: Religious identity enactment at home and at a festival

Much research addresses the proposition that identifying with a group shapes individuals’ behaviour. Typically, such research employs experimental or survey methods, measuring or manipulating social identiﬁcation and relating this to various outcome variables. Although shedding much light on the processes involved in the identity – behaviour relationship,suchresearchtendstooverlookthevariousconstraintsthatlimitindividuals’abilitiestoactinaccordancewiththeiridentities.UsinginterviewdatagatheredinnorthIndia,weexplorethefactorsaffectingtheenactmentofareligiousidentity.More speciﬁcally, using data gathered at a religious mass gathering, we compare and contrast participants’ reports of identity enactment when they are at the event and when they are in their home villages. These two contexts differ in terms of their social organization, especially the degree to which they are marked by the presence of a shared identity. Exploringparticipants’accountsofsuchdifferencesinsocialorganization,weconsiderthe social processes that constrain or facilitate identity enactment. In so doing, our analysis contributes to a richer analysis of the identity – behaviour relationship.

. This explains the variability in behaviour across situations through the way in which different social identities vary in their salience. Simultaneously, it explains the homogeneity of behaviour amongst members of a social group in any given situation as deriving from the fact that they are acting in terms of the same social identity and hence the same norms and values.
We argue that the cognitive impact of social identification (and the consequent inclination to enact the group stereotype) tells only part of the story of social identitybased behaviour. It is one thing to identify and endorse certain identity-related positions, and it is another to be able to act upon them: We live in a world of constraint. Such constraints may originate in the active opposition of others. Elsewhere, they may arise less through active opposition but through the ways in which others fail to act in ways that accord with our social identification. For example, women working in professional occupations can find their ability to enact their professional identity compromised by the ways in which male colleagues orient to them in terms of other identities (e.g., in terms of their being a mother, their sexuality: Rafaeli, Dutton, Harquail, & Mackie-Lewis, 1997).
Below, we argue much research overlooks such constraints because it is increasingly conducted in artificial worlds where the inclination to act in a given way is taken as equivalent to acting in that way. We then discuss research that addresses some of the constraints on group members' ability to enact their identity. However, this research focuses on the relatively exceptional circumstances in which group members face punishment in relation to their attempts at identity enactment. We then argue that if we are to appreciate the more mundane constraints on identity enactment (and what may allow us to overcome such constraints), we should explore the shifts in social relations that occur when people view themselves and others as sharing a social group membership and identity. We argue that such relational changes allow a co-ordination of behaviour that is otherwise impossible when people are acting in terms of diverse identifications. Finally, we report interview data concerning the relative ease/difficulty of enacting a particular social identity in two different social contexts. Specifically, we report data obtained from participants at a religious mass gathering characterized by a sense of shared identity in which they reflect on their experience of identity enactment at the event and how it differs from when they are at home.
Important as these studies are, they equate choices with outcomes. First, there is the assumption that people are free to choose their identities (e.g., by completing an identification scale). However, one cannot always choose an identity at will: As sociologists of nationhood point out (e.g., Bechofer & McCrone, 2014), it may be better to regard identity as a claim, the outcome of which depends upon its acceptance by others. This is not always forthcoming (e.g., Cheryan & Monin, 2005;Hopkins, Reicher, & van Rijswijk, 2015;Pehrson, Stevenson, Muldoon, & Reicher, 2014;Wakefield et al., 2011). Second, there is the assumption that people are free to express themselves without any consideration of whether they have the necessary material resources and whether their action would be impeded or supported by others. This tendency to treat identity-related wishes as realities is encouraged by the fact that ever fewer of our studies involve actual behaviour and still less involve interaction with others (Haslam & McGarty, 2001). This is being exacerbated by the 'replication crisis' (Sassenberg & Ditrich, 2019) and an increasing reliance on online studies for adequately powered studies such that some now ask, 'is psychology still a science of behaviour' (Doli nski, 2018).

Constraints on identity enactment
One research tradition where there is attention to such constraints (and how they may be overcome) can be found in ethnographically informed studies of protest crowds facing police opposition (Reicher, 1984a(Reicher, , 1996. Such opposition can stop crowd members translating their beliefs and values into action. Yet, these constraints can sometimes be overcome if crowd members co-ordinate their behaviour in ways that support each other (Drury & Reicher, 1999, 2009. Addressing this ability to coordinate, social identity researchers emphasize the importance of crowd members having a shared identity such that they have a common understanding of how they should respond to those who oppose them and a strong ethos of mutual support, which allows crowd members to overcome outgroup opposition (Drury & Reicher, 1999, 2009Reicher, 1996;Stott, Drury, & Reicher, 2017;Stott et al., 2018). Research also shows that it is the ability to overcome this opposition and translate identity-related values into action (what is often referred to as 'collective self-realization' or 'collective self-objectification': Drury & Reicher, 1999, 2009Reicher, 1996), which underlies crowd members' positive emotional experience.
Laboratory studies also address such dynamics. Social identity research on the effects of anonymity shows it can result in increased conformity to group norms because it reduces cues to inter-individual difference and so facilitates a shift from individual to social identity (Reicher, 1984b; see too Postmes & Spears, 1998). Yet, in addition to this cognitive effect, anonymity in groups facing a powerful outgroup can result in greater expression of behaviours that are normative for the ingroup, but which would normally attract outgroup punishment (Reicher & Levine, 1994). Moreover, research addressing the effects of anonymity to fellow ingroup members showed it undermined group members' ability to co-ordinate social support such that they are less likely to express behaviours that are normative for the ingroup but which would attract punishment from the outgroup (Reicher, Levine, & Gordijn, 1998).
Although such research addresses some of the constraints on identity enactment, it is limited to relatively exceptional circumstances where identity enactment is vulnerable to punishment from a powerful outgroup. There is much less addressing the factors affecting identity enactment in more mundane circumstances. In order to highlight these, it is important to widen our understandings of the psychological transformations associated with social identity salience.

The relational transformation of group psychology
The wider point in all the above is that although subjective self-definition in terms of a distinct social category is necessary for the behavioural enactment of identity-related values, it is not always sufficient. Or to put it in terms of self-categorization theory's formulation of the identity-behaviour relationship, there is more to identity enactment than individuals' processes of self-stereotyping. Addressing this, recent theorizing makes the point that a shared social identity in collective settings does not just lead to a cognitive transformation but also to a relational transformation (Reicher, 2017) such that group members come to see each other as 'fellows' rather than as 'other' (Neville & Reicher, 2011). This mutual recognition is manifested in greater respect, trust, and co-operation (Haslam, Reicher, & Levine, 2012;Hopkins, et al., 2019) and contributes to group members' sense of social support and well-being (Jetten, Haslam, & Haslam, 2012).
As already hinted, the relevance of such a relational transformation for overcoming constraints on identity enactment is particularly apparent in studies of crowds faced by opposition from others where the co-ordination of group members' behaviour is readily observable. In this paper, we investigate whether and how the cognitive and relational transformations described above are relevant to more common and mundane aspects of identity enactment. To be more explicit, we explore the complexities involved in enacting a very different identity to that of a protest crowd facing the authorities' opposition. The identity in question is a religious identity.
Our data derive from interviews conducted with religiously observant Hindus at a mass gathering pilgrimage event in India. Throughout, participants routinely reported on the pleasures of participation and proceeded to explain what was so distinctive about their mass gathering experience. In so doing, they referred to a range of social and relational factors concerning the informal social organization in their home villages and at the event, and how these shaped their ability to fulfil the demands of their religious identity. Before explaining our analytic approach to these data, we first describe this mass gathering.

A Hindu mass gathering
Hinduism is difficult to define. There are diverse traditions of belief and practice (Doniger, 2009). Moreover, Western (and colonial) assumptions about the nature of religion (using Judaeo-Christian belief as a normative framework) add confusion (King, 1999). Hinduism involves beliefs and practices that go beyond what Western observers typically consider 'religious' as opposed to 'secular'. For example, the concept of dharma refers to the customs and social activities that support the cosmic order and defines a whole 'way of life' (Hacker, 2006). Moreover, the conceptual contrast between 'religion' (which tends to focus on scripture and institutions) and spirituality (which tends to refer to a sense of interconnectedness and transcendent meaning), which is familiar in the West, is less relevant in Hinduism. In the latter, there is less institutional structure, a greater focus on everyday ritual and social practice (dharma), and an aspiration to transcend the concerns of the material world to engage with higher truths. Indeed, the Hindu concept of mayajaal conveys the idea that the everyday world is an illusion of limited importance that one should seek to transcend through one's religious practice. For example, the act of puja (loosely, 'prayer') is understood as entailing the opportunity to see the divine in one's environment (e.g., in an object) and have the experience of the divinity seeing the worshipper (this interaction being known as darshan).
Everyday demands routinely compromise engagement with such prescribed activities and spiritual concerns. One response is to attempt to free oneself from such mundane constraints through adopting the ascetic existence of the renunciant (e.g., naga sadhus who often eschew clothing and live on donated resources). Another is to undertake a pilgrimage (Buzinde et al., 2014). Hindu pilgrimage often centres on crossing points on rivers (tirthas) where it is believed one can catch sight of the deities. A key feature of such religious pilgrimages is the renunciation of everyday comfort and the pursuit of spiritual goals. Indeed, texts such as the Mahabharata and the Puranas talk of 'tirthas of the heart', which as Coleman and Elsner observe (1995: 149) implies that 'the pilgrim should not only bathe in the waters of earthly tirthas, but also in the inner virtues of truth, charity, patience and self-control'. Related to such injunctions is the pleasure of associating with others who listen to and speak spiritual truths (such assemblies being called satsang or satsangati).
One of the most significant Hindu mass gatherings is the annual month-long Magh Mela at Prayag (Allahabad, North India), which attracts millions to the confluence of the Ganga and the Yamuna rivers (Buzinde et al., 2014;Ruback, Pandey, & Kohli, 2008). Some work as market traders or as workers removing rubbish, cleaning latrines, etc. Many hundreds of thousands visit for a few days around particularly auspicious dates. Others (known as Kalpwasis), mostly rural and elderly (predominantly in their 60s and 70s), commit to undertaking a distinctive religious pilgrimage, which involves living for a full month on the floodplain and repeating this annually for 12 successive years. The norms associated with being a Kalpwasi prescribe a series of religious activities such as pre-dawn bathing in the Ganga, performing pujas, and listening to religious discourses (e.g., commentaries on the lives of the gods given by revered gurus). They eat one simple vegetarian meal a day (avoiding so-called tamasic ingredientse.g., onions and garlicwhich are believed to dull the senses and detract from one's clarity of thought). Through such activities, Kalpwasis seek to remove impure thoughts (e.g., jealousy), recognize others' divinity, live in harmony with others, and catch a glimpse of the gods (darshan) (Buzinde et al., 2014).
Living on the Ganga floodplain in January and February is demanding. The sanitation arrangements are basic (Balsari et al., 2016) and the tented encampments simple. The pre-dawn bathing presents many challenges: The trek to the bathing ghats can be long, the water is very cold, and entering the Ganga is complicated by the depth of the waters and the crowd (Pandey et al., 2014). Interview data show Kalpwasis differentiate themselves from others at the event (e.g., those who only come to the Mela for a few days), see themselves as a distinct group, and report a sense of mutual recognition as fellow Kalpwasis (Hopkins, et al., 2019). Complementing such findings, survey data show that many Kalpwasis report a strong sense of shared identity with other Kalpwasis and that variation in the degree to which they did so is consequential Khan et al., 2016). Specifically, these data show a relationship between levels of shared identity and strongly positive experiences of the event, which was mediated via participants' feeling that they (1) experienced relational intimacy with others and (2) were able to enact their Hindu identity (collective self-realization) . Moreover, longitudinal survey data concerning participants' (1) level of social identification as a Hindu and (2) their performance of religious practices, showed increases in both these variables to be a function of participants' level of shared identity (mediated by their ability to enact their Hindu identity; Khan, et al., 2016).
However, quite how a sense of shared identity impacts the ability to enact religious identity at the Mela requires further exploration. We address this using interview data gathered with Kalpwasis in the Mela. Specifically, we consider the contrasts participants drew between the informal social organization of life at home and in the Mela and how these shaped their ability to enact religious practices and pursue spiritual concerns. Such contrasts foreground the everyday constraints upon religious identity enactment and how these may be transcended in the Mela.

Procedure
Interview process Interviews took place in the participants' encampments. Access was negotiated with the approval of the camp organizer and the interviewee. The interviews were scheduled so as avoid interrupting devotional activities and lasted 30-80 min. Questions focused on the experience of participation in the Mela: the conditions in which people lived; their daily routine; participation in religious activities; the degree to which participants experienced a shared identity with others; their experience of the crowds at various locations over the Mela site; the nature of their social relations with each other, etc. Throughout the interviews, participants compared and contrasted life in the Mela and in their home villages and spoke of the pleasures to be had in realizing the injunctions associated with being a Hindu. Sometimes, this was in response to interviewers' questions. Sometimes, it was spontaneous. No intrusive questions (e.g., relating to sexual abstinence) were asked.
The interviews were conducted in Hindi, were recorded, transcribed, and then translated into English (by the interviewers). Where appropriate, more literal translations were modified to comply with everyday English expression. Interviewees are identified by a number and information on their sex and age (e.g., P1M82 refers to Participant no. 1 who was male and 82 years old). This research was conducted according to the ethical guidelines of the British Psychological Society and was approved by the research ethics committee at the University of Dundee (UK) and the University of Allahabad (India). The data reported here are available at www.data-archive.ac.uk. The authors have no conflicts of interest to report in relation to this paper.

Analytic procedure
The interview transcripts were read and re-read. Sections relevant to the experience of participation in the Mela were subjected to particular attention and read through reference to the social identity approach to the psychological transformations found in crowds (Reicher, 2017). Initially, we focused on participants' reports of a sense of shared identity at the Mela and the pleasures that this brought (Hopkins et al., 2019). As the analysis progressed, we paid increasing attention to the contrast participants drew between their experience of life at home and in the Mela. For both contexts, we explored participants' reports of the informal social organization within which they were located and how these were relevant to their experience and behaviour. Such reports were compared against each other in an iterative process following the logic to Grounded Theory (Strauss & Corbin, 1998). Through doing so, we developed an analysis of participants' accounts of how the informal social organization found in the village and the Mela impacted their ability to pursue religious routines and spiritual concerns in these two contexts. Addressing such issues of social organization, we kept close to the explicit meaning of our participants' accounts (Boyatzis, 1998).
Undoubtedly, Hindu beliefs about the spirituality of a pilgrimage site (e.g., the presence of the gods, the holiness of the Ganga) are also relevant to our participants' experience of the Mela. However, we believe that our own analytic focus and categories (e.g., the concept of a shared social identity) help capture features of the social environment relevant to their Mela experience and how it differs from their everyday life.
Our analysis is illustrated through multiple examples (Elliott, Fischer, & Rennie, 1999) and negative or deviant cases (McPherson & Thorne, 2006). In the quoted extracts, excluded text is marked by square brackets [ ]. Text inside such brackets is to aid explanation. First, we consider religious identity enactment in the village. Second, we consider it in the Mela.

Identity enactment in the village
Identities and social relations. Participants reported that in the village, various demands made it difficult to enact their religious identity and pursue their spiritual aspirations. When asked 'What was the reason you decided to do Kalpwas?' one woman (P27F70) replied: 'I had the motivation that "Oh Lord, please take away my impoverishment and give me your devotion"'. Asked to explain 'and what was your impoverishment?' she answered:

Extract 1
Impoverishment means that [if] I will stay at home, I will gossip, won't see satsangati, won't receive any knowledge, won't get darshan of the saints, won't be able to listen to discourses, won't get the darshan of mother Ganga. This was my wish.
In other words, her sense of impoverishment was spiritual in nature and bound up with the difficulty of enacting her religious identity at home, for example, the lack of opportunity to participate in discussions with those who listen to and speak spiritual truths (satsangati). As this interview progressed, the contrast been home and the Mela developed. Following an exchange in which the interviewee explained that there was sense of 'bonding' and 'love' at the Mela, the interaction continued:

Extract 2
Interviewer: So does it happen, that there is love among each other here [in the Mela], but these same people when they will go back and live in society again, then the love will decrease or perish? P27F70: Well, then this feeling inside will not remain Interviewer: It doesn't remain?
P27F70: Where will it remain? You have to clean away the cow dung, clean away the waste, boil the milk. The kids are going they haven't taken lunch. Throughout the material reported above, P27F60 draws a clear contrast between the opportunities at home and at the Mela to enact her religious identity. First, she describes her sense of spiritual impoverishment at home as motivating her pilgrimage. Second, she describes the ways in which her mundane household concerns and interactions (e.g., gossiping) limit her spiritual experience. Third, she describes how the Mela provides a context where she can be freed from such routine obligations and have the time and opportunity to enact religious injunction (e.g., listen to discourses, participate in discussions with those who speak spiritual truths).
Another constraint on identity enactment concerned the intrusions associated with social relations in the village. One (P37F62) commented that when engaged in the religious ritual of puja, her ability to fulfil her devotional aspirations was compromised by intrusive thoughts concerning everyday village interactions. Asked how the pilgrimage impacted her life, she contrasted her month at the Mela with the other eleven months of the year: Extract 3 P37F62: The daily routine, well I feel that the sadness, trouble and agitation etc., of these eleven months, all this is gone away in this one month. I forget everything, whatever has befallen me, whatever has befallen, it's all forgotten Interviewer: So this one month, you forget everything and P37F62: Here, there is a clear sense that everyday interactions (e.g., with neighbours, mothersin-law, sisters-in-law) compromise one's ability to focus on the spiritual. Indeed, even when formally engaged in practices and rituals associated with the religious identity (e.g., puja), 'your mind will wander' as you reflect on such interactions.
Other participants added that their attempts to pursue devotional activities and spiritual interests could easily be curtailed by obligations associated with particular social relationships. For example, after P9M66 described his Mela routine (bathing, listening to discourses, reading) he was asked with regard to 'the routine that you have, what do you like most?' He replied that at home his devotional activities were disturbed by his responsibilities as a parent but that at the Mela, he was freed from such duties. As he put it: Suppose that I am in my village, at home, praying, and my child falls from the roof, so will I pray or will I look after him? Here, it is secured from these things. If the head breaks, those who are there, they are looking after that. Here if I sit, whether I am trying to read Sundar Kand [a chapter in the Ramayana], or whether I want to read a chapter of the Gita, so there is no hurdle. So, how big a profit is this! Another (P25F60), explained that at the Mela, 'we should not have any feelings of malice' and that the 'the mind changes dear. This place is so good'. She continued that whilst one was able to experience a sense of love and harmony at the Mela, back home, concerns associated with farming changed everything. Illustrating this with changes in the behaviour of her brother-in-law, she explained: In these two extracts (4 and 5), our interviewees' contrasts between the experience of life at the Mela and at home focused on how routine village interactions associated with non-religious identities impacted their ability to enact their religious values and orient to spiritual concerns. In extract 4, the enactment of prayer is disturbed by the obligations of fatherhood, and in extract 5, the ability to experience spiritual feelings of love and harmony is compromised by the concerns of a farmer for his fields. In the next extract, the constraints on the pursuit of spiritual concerns were more general and entailed interactions based on selfishness. Specifically, the interviewee (P5M55) explained that at the Mela, there should be 'no selfishness that I will gain something from him, no selfishness. There is love where there is no selfishness'. However, they observed that in the village, interactions were shaped by people's selfishness and that he too acted selfishly with an eye to calculations of profit and loss:

Extract 6
Interviewer: OK grandfather, like you said, you have a friendly relation with people here, so when you will go back home so there also the behaviour is like this, meaning almost the same? P5M55: I also have selfishness, others also have selfishness. Again, the point is simple. In the village, participants were drawn into interactions that can best be characterized as instrumental in nature (involving calculations of 'loss' and 'gain') such that they failed to treat all with genuine regard for their spiritual worth. That is, in everyday interactions in the village, individuals reported being drawn into modes of interaction at variance with religious injunction.
Others as obstructing the enactment of one's religious identity. Thus far, we have considered how everyday village routines, social interactions, and competing demands could constrain participants' abilities to properly enact their religious identity and realize their spiritual concerns. We now consider how others' behaviour could impinge directly on their own ability to realize their devotional duties. Sometimes, this disruption was unintentional. For example, when P9M66 continued his response (reported in extract 4) to the question 'what do you like most?' about the routine of the Mela, he referred to the difficulty of concentrating on his prayers when others created a disturbance:

Extract 7
The environment here [the Mela] is spiritual and in the village, I am sitting for prayer, two people fight in the neighbourhood. There are arms flailing. Tell me, if I were in that situation, tan-tan, tan-tan [the sound of metal striking metal], could I concentrate?
In other examples, the disruption from others appeared more intentional. Thus, one participant (P37F62), who had described the opportunities for devotional practice in the Mela (see extract 3), was asked whether such opportunities were possible outside the Mela and explained not. More specifically (and referring to a devotional practice involving the ringing of bells), she explained: Extract 8 P37F62: No. In the village it cannot happen Interviewer: Why can't it happen?
P37F62: In villages it can't ever happen [ ] Will they be able to leave and sit in the village this way? Your daughters-in-law and all will kick you out of your home saying 'this old fellow has nothing to do, he just rings a bell the whole day'. People will make fun of you.
Thus far, we have discussed how religious identity enactment in the village was compromised and subject to social constraint. Sometimes, participants referred to the ways in which they omitted to pursue their religious routines and spiritual concerns, sometimes to the ways in which they acted in ways that actively compromised religious injunction, and sometimes to the ways in which their realization of devotional duty was impacted directly by the behaviour of others. We now turn to religious identity enactment in the Mela.

Identity enactment in the Mela
Participants routinely observed that the Mela provided a context in which they could more fully live in accordance with Hindu traditions and engage with spiritual concerns. One participant (P37F62), speaking to the female interviewer, put it as follows: 'For me, for one month, we see heaven. After that daughter, it's back to that same mayajaal'. Another (P20F60) commented: 'Staying at home, no one is able to do puja and all in this manner' and continued that when in the Mela they could have 'one hour with God in morning and in evening. Do puja. Talk to God. It feels that we have got peace of mind'.
In part, this experience is bound up with Hindu beliefs about the holiness of the Ganga. However, it is important to consider how the social organization of the Mela facilitated religious identity enactment.
Identities and social relations. Participants routinely described an absence of everyday obligations and social commitments when in the Mela. One (P36M66) explained that:

Extract 9
in villages everyone is busy making a living, gets trapped in mayajaal. There is a difference between there [the village] and here [the Mela]. No? Here we have only one task, but there we have many other tasks to do. So there certainly is a difference between the two. Another (P5M55), asked if 'the faith, the peace that you have come here for, you don't get anywhere else, you only get it here?' replied that it was possible on other pilgrimages but that the length of the Magh Mela (one month) allowed immersion in an environment characterized by the performance of religious rituals:

Extract 10
It is not like it cannot be found anywhere else, but here there are so many people, you don't have to go anywhere else. At home if we go somewhere [it is for] one day, two day and then we come back home. Here, at least for one month there is no other work. Here, bathing and prayers are the only things to be done. You don't look after business and you don't have anything to do with family. There is peace here! Moreover, when asked 'of your routine, which is the most favourite part, what would it be?' this interviewee explained:

Extract 11
Now, here the best is doing prayers. Here at least, we get time, with full devotion we have to perform prayers. We get time, we devote ourselves entirely to praying, we don't have to go anywhere, nobody is going to come, nobody is going to come and ask you to go with him, give responsibility of some work. So, we sit with full peace of mind and perform prayers.
Again, the point is that participants reported an absence of intrusive demands bound up with other relationships such that they could focus on their spiritual concerns ('So we sit with full peace of mind and perform prayers').
Another aspect of this focus was the absence of conflict and selfishness associated with everyday relationships in the village. Asked about their social relations with others, P12M65 explained that these were positive and that 'There is no clash of our selfishness and desire with others. So why any kind of clash? Only because of selfishness are there clashes. This is not present here'. When asked 'why is there no clash?' they referred to tensions in the village (e.g., over land ownership) and explained that in the Mela, people were engaged in practices that could be described as parmartha (actions of selfless service; Dalal, 2010):

Extract 12
Everyone has only one purpose, to bathe in the Ganga, praying to god. The purpose is not to fight. There is not the case of land partition: I will take this, I will cut this, I will do this. Such things are absent. So, there is no matter of selfishness. It is parmartha here. Everyone does it. Someone asks for alms, everyone gives, gives together. There is no controversy, no clash, there is no question regarding it! Needless to say, the contrast between this absence of selfishness and that reported to characterize life in the village (e.g., extract 6) is striking and can be understood as reflecting the lack of intrusion of other relationships and identities (e.g., 'farmer') associated with material concerns ('I will take this, I will cut this'). In turn, this lack of intrusion reflects the point that many of the interactions experienced at the Mela are shaped by Kalpwasis' shared identity.
With regard to the proactive commission of religious identity-related activity, participants also described social interactions at the Mela, which confirmed their religious practices and spiritual concerns. For example, referring to interactions with other Kalpwasis, P18F60 explained that whether they were from one's home village or not, 'everyone lives cooperatively, talks to each other. Crossing paths, they could just ask 'had your bath? And all this. People talk to each other'. In similar vein, another (P5M55), speaking of Kalpwasis' relations with each other, observed:

Extract 13
Here all are equal, have come to bathe in the Ganga, all are equal. Here people don't have bad feelings for others, no one is thinking good of me, nor am I thinking good of them. If someone meets with another person [we say] "Jai Ganga Maiyya!" [hail mother Ganga!], sit, have a talk, show some faith in relation to God. Just that and nothing else.
Such examples underline the point that interactions in the Mela are shaped by Kalpwasis' shared identity. Thus, in the extract above, the interviewee reports an absence of the usual inter-individual judgements experienced in other contexts. Moreover, this is reported to be mutual and reciprocated ('Here people don't have bad feelings for others, no one is thinking good of me, nor am I thinking good of them'). Second, and relatedly, there is a focus on (religious) identity-affirming topics and the exchange of identity-affirming greetings ('Jai Ganga Maiyya!'). Such conversational topics and identity-affirming greetings are socially important. They constitute forms of religious identity enactment. Moreover, they have an interactional dynamic in which others' behaviour (the issuing of a greeting) provides the context for one to enact one's own religious identity through reciprocating that greeting. Analytically, the wider point is that in the Mela, others' behaviour creates opportunities for one to enact one's religious identity and that this reciprocity is possible because there is a shared Kalpwasi identity (and thus an alignment of people's identities and identityrelated interests).
The significance of this alignment of identity for individuals' abilities to enact their religious identity and be spiritual is well-illustrated with the example of gossip. Gossip is understood to be problematic (see extract 1). Indeed, if one is to free oneself of jealousy and experience a sense of harmony in which the spiritual worth of all is recognized, gossip should be replaced with interactions in which one listens to and speaks spiritual truths. Thus, when asked what they liked the most about the Kalpwasi experience, P37F62 responded 'the best is satsang' and continued: Again, the point is that the pursuit of spiritual concerns is not simply a personal choice: If one hears gossip, one is inevitably drawn into everyday evaluation and judgement. That is, the ability to pursue one's devotional duties requires the co-operation of those in the immediate vicinity (no gossiping) with the corollary that religious identity enactment is something of a joint accomplishment. This returns us to the point that in the Mela, there is the potential for a shared identity and thus an alignment of individuals' identities and identity-related interests such that all act in ways that can support and scaffold each other's identity enactments.
Inevitably, participants reported disappointments concerning the Mela. Thus, P17M79 complained others did not always focus on their devotional practice and that this could detract from his own. Referring to attending sessions where they would sit and listen to speeches (or 'discourses') given by various gurus, he observed that it was inappropriate to talk at such meetings ('we cannot talk during discourses you know'), but that this injunction was not always followed by others. In response, the interviewer commented 'oh yes, you are not supposed to talk', prompting the interviewee to continue:

Extract 15
P17M79: If someone talks, the other people will forbid him saying "oh man, do what you have come for. Listen!" Interviewer: Yes, this is true P17M79: "And, if you want to talk about your home, then it is better you go away". As simple as that. And nobody speaks.
Here, three related points deserve attention. First, people attending the Mela felt able to censure their fellows if they behaved in an identity-discrepant manner. Second, such collective censure was possible because of a sense of shared identity such that people felt they could and should remind others of what constitutes 'correct' behaviour. Third, such censure contributes to creating an interactional context in which religious identity could be more fully enacted. Thus, although this negative case (McPherson & Thorne, 2006) makes the point that participants at the Mela did experience disturbance in pursuing spiritual concerns, it also implies a virtuous circle in which an intervention based on an understanding of Kalpwasi identity contributes to a social context in which identity enactment is facilitated rather than compromised. Again, this underscores the point that identity enactment is a social and collective accomplishment based on a sense of shared identity.
Others as supporting the enactment of religious identity. As noted in our introduction, many attending the Mela do so for non-spiritual purposes. Some attend in terms of workrelated identities (e.g., market traders), and some attend with a touristic rather than religious ethos (e.g., to shop for trinkets). Our Kalpwasi participants were alert to this diversity, and this was apparent as they spoke of their sense of shared identity with other Kalpwasis. For example, one was asked about the potential difficulties associated with a crowd where there can be 'pushing and shoving, people hit each other and move, sometimes they even make you fall, sometimes things are stolen, don't you have trouble with all these?' In response, P20F60 drew a contrast between the behaviour of the young men who came to watch the bathing crowds or to work as boatmen ferrying pilgrims on the rivers and Kalpwasis:

Extract 16
Young guys come. They run and move around. Some boatman are also like this who run and move around. The youngsters do this. They do not realize that some old man might fall over because of their [behaviour]. But the Kalpwasis who are there, they do not do this.
Another (P8M65), asked about the experience of the Mela crowds and how this compared to their experience of other crowds (e.g., at a bus station), responded: Such observations underscore the point that any sense of shared identity at the Mela is not all encompassing and that there is potential for a degree of identity disalignment (such that people act in terms of a range of identities and have divergent concerns and interests). Yet, if this is reminiscent of life in the village (where one can be surrounded by others acting in terms of different identities and pursuing non-spiritual concerns), there is one key difference: At the Mela, there is a body of other people (the community of Kalpwasis) for whom there is a sense of shared identity. This brings an alignment of interests (e.g., a common desire to pursue spiritual rather than material concerns) and shared understanding of appropriate behaviour.
Needless to say, in other contexts these same individuals' priorities would also be shaped by a range of disparate social identities (e.g., their diverse occupations). However, their shared identity in the Mela as Kalpwasis was such as to encourage mutual support. Sometimes, this support was manifested in mundane acts of solidarityas when people looked after others' clothing when they bathed. As P36M66 explained, 'one person looks after the belongings and the rest bathe. We have to look after the clothes also. If we don't look after the clothes, then perhaps someone just steals them while we are bathing'. Sometimes, it was more serious: P26F75 observed 'like for example we are in this place, somebody is in trouble and there is no one with him so the one who is there [available] here at least they should help'. Moreover, this concern for others was sometimes presented as an identity-related norm: Elaborating on how Kalpwasis were ready to help each other, P20F60 explained 'The meaning of Kalpwas is that the most important puja is cooperating with each other. Whatever is possible, help and cooperate with others. This is the main act'.
In turn, participants reported a sense of trust in other Kalpwasis, which made life in the Mela rather less daunting. One interviewee (P15M73) asked 'can you tell us how your relationships are?' answered: The mutual support and protection described above was important in several respects. Not only did it help in managing specific problems of direct relevance to one's ability to enact one's religious identity (e.g., through protecting one's clothes from theft whilst bathing), it also engendered a wider sense of confidence in one's ability to manage the environment and thus more fully engage with the opportunities to enact one's religious identity. Take, for example, P2M53 who responded to the question 'do you ever feel that there is some connection with everybody? That I am a Kalpwasi, they are also Kalpwasis?' replied 'Yes, there is a connection. If there was no connection, then we would not have the courage to come next year!' In similar vein, P20F60 spoke of the challenges involved in undertaking this pilgrimage and of the importance of others' identity-related behaviour in motivating her own. Discussing other pilgrimages, the interviewee explained that

Extract 19
P20F60: The people of the early days, like the kings and saints and all, they would go to mountains like the Himalaya and do penance. Where do we get such stamina? One sparrow will chirp and we will run from fear. Where do we get that courage? If no one bathes in Gangaji [the Ganga, venerated as a god] at midnight and we are asked to go for a dip, will we have the strength to go? First, we see [others] and only then do we go Interviewer: Ok, so you won't go?
P20F60: See, I won't tell a lie, only when we see the public do we go. If you do not have the courage, how will we go?

Discussion
Analyses of crowd protest reveal the importance of considering how identity enactment is a function of constraint and how such constraint may be overcome. Although a religious mass gathering has none of the drama of a protest crowd, our comparison of participants' understandings of how their mass gathering experience contrasts with their everyday experience also suggests that identity enactment is something of a social and collective accomplishment and that this is easier in the Mela than at home. With regard to life in their villages, interviewees reported acts of omission where they struggled to orient to spiritual concerns because of competing demands (extracts 1 and 2) or the distractions associated with everyday social relationships (even when engaging in religious ritual: extract 3). Moreover, they reported acts of commission where they were drawn into concerns associated with other identities, for example, parental (extract 4) or occupational (extract 5). They also reported instances where their attempts to pursue spiritual concerns were disturbed by otherssometimes unintentionally (extract 7) and sometimes intentionally (extract 8).
In contrast, social relations in the Mela were described as facilitating religious identity enactment. Participants reported a relative absence of demands in relation to work and family responsibilities and an absence of material concerns (extracts 9-12). They also reported engaging in socially shared activities and practices that were relevant to their identities as Hindus (e.g., extracts 13-16). So too, participants reported forms of support that provided them with the confidence necessary to undertake such a pilgrimage and the devotional duties it entails (extracts 17-19).
In exploring these contrasts, we have focused on the social organization to be found in the two contexts. Specifically, our analysis suggests that whereas in the village, there is a degree of identity disalignment (such that people act in terms of a range of identities and have divergent concerns and interests), in the Mela, there is potential for a degree of shared identity (as Kalpwasis) in which people's priorities and concerns are aligned. To the degree that this alignment obtains, one is less likely to be exposed to identityincongruent behaviour and demands, and more likely to be exposed to identity-congruent behaviour and demands. Moreover, a shared identity entails a sense of social connection with others and a concomitant sense of social support. In combination, the result is an increased ability for individuals to co-ordinate the enactment of identity-related behaviours. Our analysis does not imply religious identity enactment is impossible in the village: Many routine practices are organized according to Hindu traditions, homes have shrines, etc. Nor does it mean it is guaranteed in the Mela. Indeed, on this latter, survey data show Kalpwasis differ in the degree to which they report being able to realize their religious identity . Rather, our point is that identity enactment should not be taken for granted but depends on the local social organization that one experiences.
In making these claims, we should note our data concerning these different environments were gathered in just one site: the Mela. It is therefore possible that participants' reflections on life in the village were shaped by their Mela experiences and that if we had conducted interviews in the village, their accounts of religious and spiritual life in the village would be different. Moreover, our data comprise participants' reports of their experiences rather than data based on the independent observation of interactional practice. Set against this, our interviews have the merit of allowing our participants to make a direct comparison between the two contexts. Moreover, the psychological significance of a shared Kalpwasi identity is also apparent in survey data, which show that the experience of a shared identity is relevant to the positive emotional experience of this mass gathering  and to the impact of mass gathering participation on increasing participants' social identification as a Hindu and engagement in identityrelevant practices .
The direct implications of this work are threefold. First, we contribute to a longstanding social scientific interest in the pleasure of mass gathering participation. More than one hundred years ago, Durkheim argued that the pleasures of such gatherings reflected people's feelings that although in everyday life 'the mind is chiefly preoccupied with utilitarian and individualistic affairs' (Durkheim, 1912, p. 352, cited in Olaveson, 2001, participation allowed people to rise above such concerns and enact their shared ideas and values. We believe our analysis makes the point that this is not an individual achievement but rather a social and collective accomplishment made possible by a shared identity. Second, we believe that it is possible to generalize from our study of a Hindu mass gathering to other events. However, this generalization must be at the level of social psychological processes rather than specific behaviours. Indeed, it was a generalization at the level of social psychological processes that allowed us to find insights in studies of 'riots' (Drury & Reicher, 1999, 2009Reicher, 1996) relevant to a study on religious pilgrimage in a non-WEIRD (Henrich et al., 2010) context. As with our own study, any generalization from this pilgrimage to other mass gatherings (e.g., national commemoration events) would require careful attention to the particularities of the event and the identities involved. Certainly, it would be erroneous to assume that all such events facilitate identity enactment. Our approach implies that much depends on the degree to which a sense of shared identity is possible, and events may differ in the potential for various symbols and practices to establish or undermine a common identification (O'Donnell, et al., 2016;Stevenson & Abell, 2011). Indeed, pilgrimage events are frequently riven by tension as different factions contest the meaning of the event (Coleman, 2002;Eade & Sallnow, 1991;Sallnow, 1981) and subvert any sense of shared identity (and the associated cognitive and relational processes that follow).
Third, social psychological analyses of identity enactment need to take seriously the issue of constraint. There is more to identity enactment than self-stereotyping. Certainly, the cognitive transformation associated with identification is necessary for identityrelated action. However, identity enactment takes place in social contexts characterized by different levels of constraint and support. This is easily overlooked if our methods routinely create contexts in which identity-related action is easy (as when we manipulate or measure identification and observe effects on attitudinal expression or behavioural intention). In contrast, our interview data allow insight into social actors' own experiences of identity enactment and imply it is not simply an individual achievement but rather a collective accomplishment. Of particular importance are the cognitive and relational transformations associated with a shared identity. Where present, the potential for co-ordinated action in pursuit of identity-related values is greatly enhanced. Where absent, identity enactment is socially compromised.
With regard to the wider issues posed by generalizing from this research, two further observations are relevant. Throughout, we have emphasized the importance of attending to the social organization obtaining in any situation and how this impacts one's opportunities to enact one's social identity. This does not imply that individuals have no agency in overcoming the constraints they confront. For example, although professional women may find their male colleagues orienting to them in terms that compromise their opportunities to enact their professional identity, it is important to note that women routinely negotiate such constraints. Sometimes, this is through direct complaint, and sometimes, it is through seeking to limit the degree to which others interact with them on terms that disrupt their performance of their professional identity (e.g., through adopting 'professional' clothing: Rafaeli, Dutton, Harquail, & Lewis, 1997). Such agency is important to recognize. However, it has its limits (Hall, Schmader, Aday, & Croft, 2019) and much will depend on the way status and hierarchy reproduce constraint (hence the need for feminist analyses of organizational culture, e.g., Ledwith & Colgan, 1996).
A still wider implication of our analysis is to emphasize the need for a re-socialization of how we conceptualize the contextual determination of behaviour. A core tenet of social identity theorizing is that identities vary in their salience and that this variation is tied to context (Oakes, 1987;Oakes, Turner, & Haslam, 1991).
However, throughout such work there is an emphasis on perceptual processes relating to 'fit' and 'accessibility', which limits our understanding of what is involved in determining when and how individuals behave in terms of particular social identities. Individuals are not simply processors of perceptual information concerning others' behaviour (Huddy, 2002;Reicher & Hopkins, 1996, 2000. Rather, they are social actors located within informally organized social systems and thus are routinely positioned by others in ways that shape their abilities to enact their identities. Sometimes, the outcome will be to constrain the opportunities for the enactment of a particular identity. Sometimes, it will be to facilitate such enactment. We propose that a key determinant of such outcomes is the degree to which the social organization to be found in any situation entails a shared identity.