Methodological Impasses: Facing Interrogation and Silence While Gathering Data on Sexual Violence in India

Feminist standpoint theorists highlight how relations of power and inequality impact our knowledge of the social world (Smith 1974). The hierarchical positioning of different social groups creates a world in which the experiences and perspectives of certain people are acknowledged while others are silenced (Hartsock 1998; Hekman 1997). Moreover, a researcher’s personal background—her race, gender, class, and sexuality, among other factors—condition what she is able to learn and how (Collins 2000). Together, this literature underscores how the social world—and who we are within it—shapes knowledge production.


F
eminist standpoint theorists highlight how relations of power and inequality impact our knowledge of the social world (Smith 1974).The hierarchical positioning of different social groups creates a world in which the experiences and perspectives of certain people are acknowledged while others are silenced (Hartsock 1998;Hekman 1997).Moreover, a researcher's personal background-her race, gender, class, and sexuality, among other factors-condition what she is able to learn and how (Collins 2000).Together, this literature underscores how the social world-and who we are within it-shapes knowledge production.
Building on feminist standpoint theory, this article shows how relations of power and inequality within research sites can work alongside a researcher's position to create methodological impasses for scholars of sexual violence.Drawing on experiences of fieldwork in India, we describe two methodological impasses: interrogation and silence.Interrogation occurs when respondents question a researcher's credibility; silence emerges when respondents refuse to answer questions.In our research, it is important to note that these impasses emerged not with survivors but rather in conversations with politicians, rights activists, and law-enforcement personnel-people who were broadly involved in the field of sexual violence.
Both authors identify and are perceived as middle-class, upper-caste Hindu women.They were young and unmarried at the time of their fieldwork.They both used qualitative methods to study diverse actors in the fields of communal and gender-based violence.Malik worked in Uttar Pradesh (UP), one of India's poorest and most violence-prone states.She interviewed politicians who may have been involved in communal riots or had known about sexual violence around such incidents.Roychowdhury worked in West Bengal, a state that scores relatively better on human development indicators.She interviewed and observed law-enforcement personnel who processed complaints, as well as rights activists and caseworkers who helped women advance legal claims.
Despite these important differences in the authors' sites and respondent groups, both encountered problems gathering information on sexual violence.When Malik asked politicians about the relationship between communal riots and sexual violence, her respondents interrogated and tested her political commitments.When Roychowdhury asked activists and lawenforcement personnel about rape outside of marriage, respondents became uncomfortable, reprimanded her, and abruptly ended interviews.

INTERROGATION: STUDYING SEXUAL VIOLENCE AND COMMUNAL RIOTS IN UTTAR PRADESH
Malik's fieldwork in UP was centered in Meerut, Muzaffarnagar, and Shamli districts.These districts have emerged as key sites of declining (i.e., Meerut) and escalating (i.e., Muzaffarnagar and Shamli) communal violence between Hindus and Muslims in recent years (Pai and Kumar 2018;Malik 2021).Notably, in August 2013, politicians from the Hindu nationalist Bharatiya Janata Party (BJP) organized a grand village council (mahapanchayat) meeting at which they elevated the issue of protecting Hindu women's (and the broader Hindu community's) honor (Berenschot 2014;Pai and Kumar 2018;Malik 2021).They explicitly used a case of alleged harassment of a local Hindu girl by her Muslim classmate to rally Hindu voters.The mahapanchayat subsequently precipitated massive riots in Muzaffarnagar and Shamli.
To study these riots as well as declining communal conflict in neighboring Meerut district, Malik conducted more than 50 in-depth qualitative interviews in Uttar Pradesh.She traveled to UP multiple times between early 2013 and late 2015.Many of her interviews with political elites involved sensitive conversations about when, how, and why political leaders rely on parties as instruments for violence.
In preparing for her fieldwork, Malik received vital guidance on making contact with political elites and sequencing her interview questions.She also was rightly advised that building trust with politicians, some of whom had long been suspected of carrying out violence, would be difficult.However, she never received any direct advice about how to avoid hostility or respond to interviewees who tried to police her behavior in gendered terms.Given her preparation, Malik thus entered the field with the goal of accessing a group of respondents who have rarely been interviewed by scholars of communal conflict in India.
At the time of her research, Malik was in her twenties and single.Her first name indicated that she was Hindu and her appearance signaled her upper-caste and middle-class status.She owned an iPhone and an audio recorder, and she often traveled to villages in Muzaffarnagar and Shamli in a taxi-all markers of her relative wealth and considerable privilege.However, Malik also ventured into predominantly male spaces during fieldwork; indeed, all of the political elites whom she interviewed in UP were men.
To avoid drawing too much attention from male respondents, Malik ensured that she dressed traditionally.Her intersectional identity and appearance both advantaged and disadvantaged her in distinct ways.First, subsequent conversations with male researchers who sought access to similar respondent groups revealed that Malik's gender likely helped her to secure interviews with political elites. 1 The fact that she was a woman made her less threatening to politicians, who granted her meetings at their office or in their home.Second and simultaneously, however, it was precisely because Malik was a woman-particularly a young unmarried woman-that her male respondents often interrogated her motives in explicitly gendered terms.
When she interviewed political elites, Malik typically began by soliciting information on the background, goals, and aspirations of various parties.To build rapport with her respondents, she also frequently nodded to communicate her agreement with espoused party objectives-including those with which she personally disagreed-and never explicitly shared her political commitments.However, even these strategies did not completely shield her from receiving hostile responses, which were most noticeable when she asked focused questions about particular forms of violence.The following discussion describes two key moments of pushback that Malik received from BJP-affiliated elites in UP.
The first encounter occurred during an interview with a BJP member of parliament in Meerut at which other party functionaries were also present.Having other individuals within earshot or even physically present during interviews was not uncommon during Malik's fieldwork.On this occasion, however, when Malik probed the decline of communal riots in Meerut with her respondent, a party functionary who was not her interviewee pointedly asked her if she had read the Manusmriti, an ancient Hindu text that supports the caste system (Nadkarni 2003) and patriarchal norms of being a "good woman" (Ghosh 2018).When Malik stated that she had not read the text, she was admonished: "Yes, that's why you are asking these questions."The message was clear: by asking an elected representative about communal riots, Malik -unbeknownst to her-had violated core social norms and exposed herself as a Hindu woman who did not truly understand the boundary between acceptable and unacceptable behavior.
Another important test of Malik's credibility emerged during a conversation in Muzaffarnagar with a BJP member of the UP legislature.Before this interview, neither Malik nor any of the journalists she had interviewed could confirm that the alleged harassment of the Hindu girl in Muzaffarnagar had occurred.She thus hoped that her upcoming interview would enable her to probe how a case of reported harassment had culminated in the worst communal riots in India in more than a decade.
Malik's initial questions to her respondent-about the party's goals in UP-generated a comfortable back-and-forth dialogue, and the interviewee emphasized the developmental agenda of the BJP.Later, as the focus turned toward violence, the respondent mentioned new and low-intensity forms of communal conflict, including cow vigilante attacks, even though he was not explicitly asked about such assaults.However, the interview took a sharp turn for the worse when Malik asked about the alleged harassment and the 2013 riots.When she probed the relationship between these events, the respondent posed a counterquestion in Hindi, and followed it up by declaring: Anyone, whether he is Hindu or Muslim…if someone molests his sister, would he tolerate it?Absolutely not!Think about it.You too are a woman. 2   In each situation, Malik responded in a similar manner.Her funding constraints prevented her from changing her research design in the field (Hunt 2022).She thus bowed her head, accepted the rebukes, and stated that she was a scholar who was trying to learn.These responses enabled her to continue the interviews and avoid the silence that Roychowdhury experienced during her fieldwork.

SILENCE: FAILING TO STUDY RAPE OUTSIDE OF MARRIAGE IN WEST BENGAL
Roychowdhury conducted 26 months of ethnography and interviews with survivors, activists, caseworkers, and lawenforcement personnel.She analyzed how survivors formulated grievances and why certain women were more successful However, Roychowdhury quickly encountered problems gathering information about rape.Respondents eagerly discussed domestic violence but became uneasy, silent, and hostile when asked about rape.This disparity in people's willingness to discuss different types of violence emerged among every group of respondents except self-identified feminists.Why did Roychowdhury have difficulty gathering data about rape cases?The difficulties were conditioned by the taboo nature of the topic and by her relationship to the topic as a middle-class, upper-caste Hindu woman.
In one of her first conversations with a police officer, Roychowdhury realized that respondents were comfortable when discussing domestic violence but not rape.For 30 minutes, she and the officer had an amicable conversation.The officer lamented that poor women had few options outside of abusive relationships, and he openly discussed the challenges that his station faced in processing cases.He promised to connect Roychowdhury with colleagues in other parts of the state.She believed that she had established rapport.
The rapport quickly dissipated, however, when she asked about rape.Until that point, the officer had looked Roychowdhury in the eyes and leaned forward in his chair while addressing her questions.Suddenly, he avoided eye contact, leaned away from her, and started shifting papers around on his desk."That's not an issue in this area.I don't handle those kinds of cases," he stated.He then looked at his watch and said that he had to take care of some urgent business.Realizing she had been dismissed, Roychowdhury tried to salvage the goodwill she had established earlier by thanking him profusely for his time and asking if she might contact his office again to obtain the names of his colleagues in other stations.The officer nodded curtly and remained noncommittal.Roychowdhury departed, fearing she had damaged her relationship with her respondent.She was right: he did not respond to her follow-up calls.
After several interviews that followed a similar pattern, Roychowdhury realized that her respondents were most comfortable discussing physical, emotional, and financial abuse within marriage.They were less comfortable discussing sexual violence within marriage but nevertheless were able to have a conversation about this issue through a coded term: rātēr kāj (i.e., night's work).This idea-that a wife must submit to a husband's sexual demands as part of the "work" of being a wife -allowed people to discuss nonconsensual sexual acts within marriage.Respondents had the most difficulty discussing rape outside of marriage.They wanted neither to acknowledge that unmarried women were raped nor to discuss the details of any specific case.
From the systematic differences in how respondents behaved when questioned about different forms of genderbased violence, Roychowdhury concluded that rape outside of marriage was the most taboo topic.Rape within marriage was less taboo because the boundaries of marriage allowed people to discuss sex-consensual or non-consensual-as a socially legible activity.Physical, emotional, and financial abuse within marriage was the least taboo topic, and almost everyone felt comfortable discussing these incidents because they did not involve sex.
The barrier that Roychowdhury faced while gathering data about rape outside of marriage, however, was not simply related to taboo topics.It also concerned how respondents perceived her and how their perceptions conditioned their willingness to discuss certain topics.The relational nature of the impasse she faced became clear during a conversation with Sila, a former NGO caseworker who conducted informal arbitrations with survivors. 3When Sila learned that Roychowdhury was researching the topic of rape, she looked askance and stated: Why are you getting involved with such things?A girl from a respectable family should not be asking about this.Leave that to others.You're not married; you shouldn't be discussing this.
At the time of her fieldwork, Roychowdhury was in her thirties and single.Her surname was Hindu and upper caste and her attire indicated that she was middle class.Sila's reprimand-that an unmarried woman from a "respectable family" does not discuss rape-indicated to Roychowdhury that her data-gathering problems were conditioned partially by her own social background.Her respondents were not simply uncomfortable with rape as an issue; they also were uncomfortable discussing the issue with her: a young, unmarried, upper-caste, middle-class woman.
Qualitative research is based on interactions between researchers and participants.As such, any discomfort or hostility is best interpreted not simply as an issue-specific matter but rather as an interaction-induced problem.Some of Roychowdhury's respondents may have felt more comfortable discussing rape with someone who was older, a married woman, a man, or a woman who did not have a Brahmin surname.Roychowdhury's social background placed her in a category of "virtuous" women, who should not know about sex-consensual or non-consensual-outside of marriage (Sunder Rajan 1993).By asking about this issue, she made herself suspect and potentially disrespectable and also made her interlocutors accomplices in the loss of her respectability.
The notable exception to the problems that Roychowdhury encountered-self-identified feminists-is important.It was only this group that did not ascribe to a classed, casterelated, and gendered framework of virtue.As a result, the feminists did not believe that rape was a taboo topic and did not find Roychowdhury's interest in the issue to be questionable.

CONCLUSION
Malik and Roychowdhury faced interrogation and silence from respondents when trying to gather data about sexual violence.It is theoretically possible that other factors may have led to these reactions.However, over the course of fieldwork, both authors concluded that the intersectional character of their identity combined with their research topics resulted in these impasses for two reasons.First, they noticed systematic 308 PS • April 2024 variation in respondents' willingness to talk to them about different forms of violence.Second, they faced direct admonishments that women of their social background should study a more "appropriate" topic.
In Malik's case, political elites repeatedly questioned her credentials.Even when they sometimes introduced information about other forms of violence, questions about sexual violence proved to be off limits.In Roychowdhury's case, lawenforcement personnel and activists who were not perpetrators of violence were comfortable discussing cases involving physical, emotional, and financial abuse.However, they became uncomfortable, upset, and silent when questioned about cases involving sexual violence outside of marriage.Both authors were told directly that women of their social background and age should not discuss sexual harassment and rape outside of marriage.
Scholars of sexual violence have long noted how the experience of violence generates feelings of "shame, intimacy, and fear" among survivors (Price 2012, 8).These feelings can make it difficult for survivors to speak openly not only to lawenforcement personnel but also to researchers (Schulz 2018;Tillman et al. 2010).This article describes how people who are involved in the field of violence-as activists, law enforcement, witnesses, and possible perpetrators-also have difficulty in discussing sexual violence.
How might researchers overcome or navigate these problems with data collection?First, researchers could use local code words to discuss sexual violence.Roychowdhury used this strategy to gather information on sexual violence within marriage, referring to "night's work."Second, researchers could work in teams alongside individuals who hold social identities distinct from their own.Those who are positioned differently in the field may be able to gather different types of data (Olukotun et al. 2021).