Cultural Trauma

The concept of individual trauma and the related concept of post-traumatic stress have been used to describe the reactions of those who were victims in the 2002 Gujarati riots (Bushna, 2002; Swami, 2002; Human Rights Watch, 2002; Sen, 2005). However, there is much disagreement about whether such traumatic symptom presentation constitutes Posttraumatic Stress Disorder (PTSD) in a cross-cultural context (Marsella, Friedman, Gerrity, & Scurfield 1996). There is also the issue of utilizing culturally sensitive assessment measures (Wilson, 2007), despite there being at least one study using qualitative methods to validate PTSD criteria in India (Mehta, Vankar, & Patel, 2005).

There are also significant limitations to understanding such events solely through the lens of the individual, particularly when the individual is part of a larger religious in-group or out-group. Culture is a mediating factor in understanding the behavioral and symptomological manifestations and prognosis of reactions to trauma (Green, 2003). A World Health Organization report highlights cultural risk factors—associated with politics, society, and community—that play an important role in predicting collective violence (Bond, 2007). Bond (2007) believed that what has been neglected in understanding such violence is how culture is “a moderating role in exacerbating its members’ degree of perceived distress” (p. 39). Perceived distress is not only related to the intrapsychic and biological realms of the individual, but can be affected by many systems that include, but are not limited to, family, religious, social, and political communities (Danieli, 2007; Drozdek, 2007; Hoshmand, 2007; Fairbank et al., 2003). The degree to which integration of these systems does not occur following violence may leave an individual susceptible to future experiences of trauma (Danieli, 2007). Hoshmand (2007) cautions that assumptions should not be made that all communities have the same stress levels before an identified traumatic event, especially for those groups that have been historically oppressed, nor should assumptions be made that all groups will have the same capacity to respond to the same treatment. Culture influences which negative experiences are shared with others, and which ones an individual keeps to him or herself (Drozdek, 2007).

Wilson (2007) argues that there cannot be an experience of psychological trauma with an exclusion of or indifference to cultural history. A traumatic event can lead to an extended experience of stress and emotional dysregulation, particularly if three criteria are met:

Thus, it may be useful to conceptualize trauma in a cultural context, as opposed to one based on an individual’s symptoms. Cultural trauma is defined as “members of a collectivity feel[ing] that they have been subjected to a horrendous event that leaves indelible marks upon their group consciousness, marking their memories forever and changing their future identity in fundamental and irrevocable ways” (Alexander, Eyerman, Giesen, Smelser & Sztompka, 2004, p. 1). A cultural trauma becomes even more significant when understanding that it poses a threat to individuals’ personal identities in a given society (Alexander et al., 2004).

Shifts in Religious Identity for Hindus and Muslims Residing in India

According to Kakar (1996), a person who is “communal” in an Indian context – and therefore has a shared religious identity - can be defined as:

Kakar picks up on the historical tension that has existed between Hindus and Muslims in the Indian context. A broader social-psychological-theoretical formulation, of which Kakar’s communalism is an example, is Tajfel’s Social Identity Theory, which posits that an in-group’s distinctive character is defined and maintained principally in relation to relevant out-groups (Brown & Zagefka, 2005; Tajfel & Turner, 1986).

There is also a more recent literature describing shifts in religious identity and in-group/out-group positioning for Hindus and Muslims in the communal context. A new Hindu identity based on nationalism, Hindutva, has emerged in recent years (Stavenhagen, 1996). In stark contrast to the traditional practices of Hindus, the new movement ignores practices such as rigid caste divisions (Ashmore et al., 2001; Stavenhagen, 1996). Muslims, as a minority within India (154 million Muslims compared to 930 million Hindus), are invariably affected by the new Hindu identity. One possible result of this new movement may be that urban Muslims earn lower wages than any other religious group in the country, and Urdu (spoken primarily by Muslims) has become a displaced language that has further alienated Muslims from the job and public sector (Puniyani, 1998). In reaction to Hindu communalism, a broad Muslim identity has become increasingly endorsed (Puniyani, 1998). These shifts have laid the groundwork for religious conflict and violence and provide the backdrop for how individuals may process violence within a cultural framework.

Indian Diaspora in the United States and Collective Identity

Immigration to the United States opened up to professional and educated Indians in 1965 with the passing of the Immigration Act of 1965 (Durvasula & Mylvaganam, 1994; Quddus, 2005; Prashad, 2000). Acculturation may have been even more difficult for these immigrants as members of a collectivist culture of origin (Wong & Wong, 2006; Durvasula & Mylvaganam, 1994). After a literature review, Bhugra (2005) concluded that migrants who move from a collectivistic culture to an individualistic one, where there is little social support, experience increased stress. Changes due to modernization have led to geographic displacement for the purpose of seeking financial opportunities as well as associated feelings of rootlessness resulting from loss of friends and family and cultural values (Kakar, 1996; Ashmore, Jussim, & Wilder, 2001). Humiliation is not an uncommon feeling when non-Western citizens view the inequity in power and financial structures in comparison to their host country (Wong & Wong, 2006). Feelings of isolation for immigrants can also stem from perceived hostility from other Americans and the West in general (Wong & Wong, 2006).

To counter feelings of marginalization, Indians have formed student organizations at universities, which have grown in number and have become more specific to language, locality, and religion (Quddus, 2005). Spirituality grew in importance as Indian parents worried that their children would adopt the morals of the new (American) society (Prashad, 2000; Quddus, 2005). Another reason for Indians, particularly Hindus, to gravitate towards spirituality is the notion that they are somehow inherently advanced in this domain (Sen, 2005). This is typically a view held by Western individuals outside of Indian society, perpetuated in the era of colonialism (Sen, 2005).

The Vishnu Hindu Parishad of America (VHPA), along with affiliated organizations such as the Hindu Student Council, have put together youth camps that have been accused of portraying distorted views of Indian history and culture (Quddus, 2005; Prashad, 2000). These organizations have held world forums that have used scare tactics by advancing the notion that Indian Muslims threaten the physical safety and identity of Hindus (Puniyani, 2003). Many temples are run by the VHPA and serve to function as monetary supports and to propagate the idea of India as a Hindu nation (Quddus, 2005; Prashad, 2000). These factors, including funding from abroad, distortions of history, identification with a traditionally allocentric culture, and marginalization may color perceptions and experiences of the 2002 Gujarati riots for recent Indian immigrants.

Purpose

Despite the increasing literature suggesting that changing identities based on religion may underlie the riots, there is a dearth of psychological research in this area. While there is theoretical literature on how cultural traumas develop, it remains unclear how such trauma affects individuals psychologically, specifically in an Indian context. Furthermore, while the concepts of cultural trauma and collective identity are interrelated, the nature of the relationship is unclear. To our knowledge, there are no studies that have looked at how religious identities are negotiated and re-negotiated in the Indian Diaspora in the United States following religious riots. The purpose of this qualitative study was to better understand how such identity is negotiated, and its interplay with collective trauma, as well as in- and out-group affiliations and prejudices. Identifying cultural trauma in the Indian Diaspora may lead to the development of more appropriate mental health interventions and a better understanding of the psychological sequelae of exposure to similar events.

Hypothesis

Given the religious affiliations of those who perished in the Godhra train bombing (Hindu) and its aftermath (predominately Muslim), our tentative hypothesis is that Gujarati Hindus experienced the Godhra train bombing as a cultural trauma, while Gujarati Muslims instead experienced only the pogrom that followed as a cultural trauma. Our hypothesis is in line with Volkan’s (2001) concept of “chosen trauma” where distinct groups based on collective identity designate which events are traumatic and which ones are not. Chosen traumas are negative experiences that are the basis of group-level humiliation and unprocessed grief (Volkan, 2001). A trauma can help to strengthen or deteriorate a community or identity, usually some combination of the two (Alexander, 2004). Therefore, another distinction we think may become apparent is that the Godhra train bombing helped to unify and strengthen the religious and nationalistic identities of Gujarati Hindus, while for Muslims we would venture to say that it may have strengthened their religious identity, and yet weakened their national identity as Indians.

Participants

Inclusion criteria for participants were as follows: interviewees had to be individuals who were at least 18 years of age and identified themselves as either Gujarati Muslims or Hindus who were in the state of Gujarat, India during the riots (February 27-April 30, 2002). All participants currently resided in the U.S. The sample size was a total of 17 participants, 10 Hindus and 7 Muslims. Of the 10 Hindu participants, eight were male and two were female. All of the Muslim participants were male.

Demographics of the Hindu Participants

The Hindu participants averaged 30.6 [SD = 9.9] years in age. The participants averaged 1.16 [SD = 1.41] years residency in the United States, with half having permanent residency in the United States. With respect to highest education completed, 20% were high school graduates, 10% had some college credits, 40% were college graduates and 30% had graduate credits. Regarding employment, 40% of the Hindu participants were students, 30% were employed and 30% were unemployed at the time of the interview. Ten percent of the participants did not attend religious services, 20% attended one to ten times a year, 40% attended on a monthly basis and 30% attended weekly. Half of the participants read religious scripture one to ten times a year, 10% read on a monthly basis, 20% read weekly while the remaining 20% read daily. Seventy percent of the participants attest to being members of a religious or political group.

Demographics of the Muslim participants

The Muslim participants averaged 27.6 [SD = 9.0] years in age. The participants averaged 5.05 [SD = 7.63] years residency in the United States, with 29% having permanent residency in the United States. Fourteen percent of the participants did not complete high school, 14% graduated high school, 29% had some college credit, 14% graduated college, 29% attended graduate school. Regarding employment, 29% identified themselves as students, 57% unemployed and 14% employed. Fourteen percent of the participants attended religious service and read scripture one to ten times a year, 14% did so on a weekly basis and 72% daily. Lastly, none of the Muslim participants were affiliated with any religious or political groups.

Materials

The semi-structured interview protocol was developed using relevant topics found in the literature such as social identity (i.e., “Please describe what it means to be Indian”); role of government (i.e., “Do you think that local and state governments aided or deterred the riots? In what ways?”); acculturation (i.e., “What are some differences between Gujarati culture and American culture? Similarities?”); perceptions of the other religious group (i.e., “How do you see your community’s responsibility in the violence?, What about the other community,” etc.); proposed solutions (i.e., “What role do you think that Non-Resident Indians (NRIs) can play in preventing this from occurring in the future?”); and coping strategies (i.e., “How have you been able to put this experience behind you, did you talk to anyone about your experience before, how did your family deal with this experience,” etc.). The interview protocol was further refined after two pilot interviews with individuals that met the inclusion criteria. Due to the fact that the interviews were conducted in a semi-structured format, the interviewer attempted to ask follow-up questions using terms that the participant used over the course of the interview in an effort to reduce bias in gathering data between the two groups (Cope, 2004). The interviews ranged from 40 to 117 minutes (M = 67 minutes, SD = 19.7 minutes).

Procedure

Participants were a convenience sample, recruited using two primary methods. The first was the solicitation of potential participants outside of Indian stores in central New Jersey. All of the Hindus were recruited using this method. The Muslim sample was a snowball sample, where information about the study was sent from a posting on the internet to an activist and by contacting a leader in the Muslim Gujarati Association located in the United States; these two sources yielded seven participants. Individuals who were interested in participating gave us their contact information for further information regarding the study. They were told their participation would consist of the following procedure: receiving the consent form and having a chance to ask questions, filling out a demographic questionnaire and answering questions related to their experience of the 2002 Gujarati riots in an interview format that required audio taping. All ten of the Hindu participants and three of the Muslim interviewees lived in the New York/New Jersey area and were interviewed in person. Four of the Muslim participants lived in Illinois and were interviewed via phone.

Interviews were conducted either at the participants’ homes or at a meeting site that was mutually agreed upon by the participant and the interviewer. The sole interviewer was a first generation Hindu Indian-American male who was born and brought up in Central New Jersey and who spoke English and Gujarati (the interviews were conducted mostly in English but occasionally in Gujarati). Participants were offered $20 in compensation for participating upon signing the consent form.

Data Analysis

All interviews were recorded using a digital recorder and subsequently transcribed verbatim. Thematic content analysis was conducted using modified grounded theory approach (Strauss & Corbin, 1998; Coolican, 2004; Henwood & Pidgeon, 1992) with NVivo 7 software. Grounded theory is based on an inductive process where information is analyzed with few, if any, a priori hypotheses or theory; in fact, grounded theory aids in generating or developing theory (Strauss & Corbin, 1998).

Two stages of grounded theory, open coding and axial coding, were implemented. Open coding consists of reading the data line by line and highlighting discrete units of interest (Strauss & Corbin, 1998). Once a segment of the data is highlighted, it is given a short label that attempts to capture its essence. Parts of the interviews were then read by a coding group of qualitative researchers led by a senior expert in qualitative methodology and six graduate students. The group served as an audit trail (Lincoln & Guba, 1985), helping to verify the coding and reduce individual bias. The principle investigator, however, did the majority of the coding individually, using group meetings for verification and to increase validity.

The second step, axial coding, is accomplished by taking the individual codes and grouping them into larger conceptual categories, or themes (Strauss & Corbin, 1998). This process was accomplished by a process of linking open codes that were conceptually were related for each of the two religious groups. Once these steps were taken, the next step was to compare the themes and categories that developed between the two groups (Hindus and Muslims). A summary description of the major findings was created to highlight differences in how Hindus and Muslims perceived the 2002 Gujarati riots.

Themes Identified in Interviews with Hindu Participants

There were 46 sub-themes that emerged during open coding of the responses by Hindu participants. Forty-three of these sub-themes were grouped into 11 themes. Only themes that occurred at least eight times are described below, ordered by the frequency in which they occurred.

Themes Identified in Interviews with Muslim Participants

There were 49 sub-themes that were used when using open coding when analyzing responses of Muslim participant. Forty-seven of these sub-themes were grouped into nine themes. Similar to the previous section, only themes that occurred at least eight times are described below, ordered by the frequency in which they occurred.

Limitations of the Study

The study was undertaken with the intention of generating theory and hypotheses about cultural trauma in an Indian context using grounded-theory analysis. Thus, the investigation is intended to present the groundwork for hypotheses that must be tested through future studies in order to refine and validate the conclusions drawn here. A major limitation of the study is the small sample size. Given that there were only two females in the Hindu sample and none within the Muslim sample, gender differences may not have been captured; therefore, the results are overall more reflective of males’ experience. Furthermore, the study may have been affected by selection bias, given the different strategies employed to recruit the participants across the two groups.

Another limitation involves the effect of the interviewer’s ethnic identity on the results. As is often the case in research, experimenter characteristics are ignored, being neither reported within the study, nor discussed for their possible influence on the participants and the study’s subsequent results (Kazdin, 2003; Anderson, 1993). The sole investigator’s religious affiliation (Hindu) was easily deduced by South Asians due to his first and last names and therefore, we believe it played a role in influencing participant responses. Anderson (1993) discussed the willingness of members from the majority group to participate in studies, due to a lower degree of fear that the researcher will exploit them. It is our opinion that even the Hindu participants that agreed to be interviewed would not have been able to speak at ease about the aggressiveness of Muslims and perceived injustices by Muslims upon their community without a perceived similarity with the interviewer. On the other hand, Muslim participants may have been able to discuss with him more openly their feelings of vulnerability without the perceived obstacle of an out-group interviewer.

Further studies with larger samples and more equal representation of women need to be conducted in order to validate the findings of the present study. To account for the question of acculturation, it may be useful to compare a non-Resident Indians sample to a sample still living in Gujarat. Some of the slum areas in Gujarat’s cities were the worst affected during the 2002 Gujarati violence (Human Rights Watch, 2002), and many participants in those areas are to a greater degree living with the direct consequences of the aftermath even today. As Human Rights Watch (2002) suggests, many segments of Indian society were involved and affected by the violence. Therefore, a stratified sample with respect to socio-economic status and geographical location might contribute to a more accurate understanding of the effects of the 2002 Gujarati riots. Furthermore, it would be important to collaborate with researchers (both Muslim and Hindu) in Gujarat who could interview participants, with the assumption that their fluency with local dialects spoken, and intimate knowledge of local cultural specificities, would lead to more trust by interviewees. In turn, the greater levels of trust that participants feel may lead them to be more forthcoming with their emotions and opinions concerning the 2002 Gujarati riots.

Reducing Communal Tension

One way to reduce intergroup bias is the framework of recategorization (Gaertner & Dovidio, 2005). Re-categorization occurs when the salience of in-group out-group designations is minimized by the creation of and adherence to a larger, more encompassing identity (Gaertner & Dovidio, 2005). Instead of attempting to classify Muslims as non-Indians, politicians of all religious backgrounds should seek to unite Indian citizens under a secular national banner. As Gaertner & Dovidio (2005) pointedly explain, “perceptions of group boundaries such as ‘us’ and ‘them’ can be transformed into a more inclusive ‘we’” (p. 77).

In asking Gujarati Muslims and Hindus what they felt could decrease animosity between the two groups, Shankar and Gerstein (2007) reported that responses included increasing social interaction between the two groups by way of mutual participation in community events and activities such as religious festivals may help to decrease feelings of tension. There is promising research that suggests that intergroup dialogues in which members of various groups are encouraged to speak directly about their feelings related to prejudice and stereotypes of the other groups, can effectively contribute to long-term behavioral and attitudinal change (Stephan & Stephan, 2005). The advantage of intergroup dialogues versus other interventions is that it focuses more on expressing affect, rather than didactic forms of multicultural education and diversity training (Stephan & Stephan, 2005). As seen in this study, there were stark differences in emotions expressed among Hindus and Muslims, and it may be useful for both sides to be aware of how the other religious group feels when misperceptions reinforce communal stereotypes and prejudices. This research may be a first step in highlighting such misperceptions at the prejudice, affect, and processing of trauma levels.