In looking for narratives of identity in fragmentary conversations (created through the ad hoc and messy interview process) I created tables and mind-maps to identify the main issues at stake, in my case—the different kinds of identities in each group, the interplay among these competing/complementary identities within the group, and the hierarchy of identities between the groups. This process was one of the most difficult as the ‘groups’ so defined and as they stand in Table 1, coloumn one, were part of my findings, rather than being predefined at the beginning of analysis. This process of trying to map out the identities was aided by an intersectional approach to the data. Analysing the interplay of power in the researcher-respondent relationship (see section 3.4) I was able to identify some key themes which would then emerge as locations of identity. Masculinity(ies) was one of the earliest themes that emerged from the detailed narrative analysis of the interviews. I started from an initial position of looking at ‘ethnicity’ as a stable category to base my analysis on. However, as my analysis deepened, it revealed the myriad manifestations of ‘interactive intersectionality’ (Feree, 2011) which problematized such uniform structures.
Henry’s (2007) essay with its detailed account of her own experience in the field, shows how the researcher's mantle is often ‘unsettling’ in the field. The difficulties she experienced as a young, female researcher pertained to the ways in which power-relations expressed themselves at the intersection of the researcher-respondent intersections. She coins the term ‘authority’ and ‘authenticity’ to look at how the researcher’s role can be legitimised or discredited through operations of intersectional power relations.
Henry’s (2007) work has helped me reflect on and make sense of my field experiences. The feeling of being ‘in-authenticated’ and subordinated was very unsettling for me in a few interactions. During a group interview with rickshawpullers at a roadside tea-stall, one of the by-standers, snatched the copy out of my hands and started reading the questions I had written in Bengali. He stood out as better-dressed and when I tried to ascertain how he was related to the group of rickshaw-pullers, he dismissed me with easy authority. I felt humiliated and anxious at the same time. The note book had a lot of personal reflections on the rickshaw-pullers I had encountered so far. His insistence of reading out my writing in a low voice made me feel exposed to others who I felt had trusted me enough to speak to me. He clearly commanded some respect among the rickshaw-pullers because many—themselves illiterate—were looking at him expectantly. I had to wait till he finished reading my notes, taking his time, in spite of my insistence that I had to go somewhere else.
During the course of the same group-interaction, a rickshaw-puller asked me what they could expect from talking to me. I had anticipated such questions and in introducing myself usually emphasized my need to talk to them in order to complete an assignment. I had hoped that my emphasis on being a student and framing the interview within my educational requirements would decrease expectations. However, the question made me uncomfortable. My halting answers, in which I said that I did not think they would gain anything material from it, seemed to embarrass my respondents as much it made me uncomfortable. They all made expressions of dissent telling me that surely this will help. One of the younger rickshaw-pullers even went so far as to say tried that surely if they supported me I could become a ‘neta’ (leader) who would then espouse their cause. Researchers like Abbott (2007), in trying to address the dilemma caused by such questions, accentuate their positionality (Ibid. 226) in relation to the respondents. I have tried to do them same in my research.
When I failed to access the restaurant-owners in the LNB, I took the help of my friend. As a man from Sweden he was much easily accepted in the role of the researcher with me in a subordinate position as his translator. His race, gender and location all were a better fit for the role of the authentic and authoritative researcher. This created issues of exploitation of trust. Similarly, the instances where I gathered data from informal conversations/interviews from rickshaw-pullers were also ethically problematic. As I reflected on these two instances of not having explicit consent, I realised that the idea of written consent itself was problematic. The hierarchy of written consent seemed to embody a positivist separation between researcher and respondents. The realities of the context were that many of my respondents were illiterate and/or deeply suspicious of any form of documentation (digital or manual) produced through hierarchical relations of power by researchers (the knowledge-producing NGO worker/academic researcher and the scrutinising government official) with opaque and suspicious motives. And in adopting the role of the less authoritative translator or a chatty passenger, I was after all embodying, if consciously, the fluidity and fragmentary nature of identities. While I was less comfortable in these roles than as a feminist, for instance, even in asserting that identity of affinity I have encountered difficulties.
Lastly, in order to protect the identities of my respondents, all names have been changed and particularisation of certain facts that could potentially be used to identify them has been avoided.
To summarise, the act of knowledge-production itself is ethically problematic as feminists, as well as post-structuralist and post-colonial theorists have demonstrated. Therefore adopting a methodology which disables objective claims and objectification of the respondents is a necessary pre-requisite. As a feminist, ‘situated knowledge-production’ seemed the only answer to myriad ethical and political dilemmas. This chapter on methodology, following Feree’s (2011:63) exhortation to feminist researchers, is ‘a modest claim to limited, fallible but strategically useful framing [that opens the door for dialogues] allowing a reflexive approach…with which feminists can more broadly challenge the framework of inequality that enmesh us all’
Identities, Agency and Entitlements
This chapter analyses fixed and fluid identities through primary research data by looking at multiple sites of identity formation such as ethnicity and location, occupational class, and gender. It also looks at questions of power and agency through the lens of identities and concludes with reflections on entitlements, mobilised through the interplay of identities.
Problematising Forced Ethnic Identities
This chapter looks to question the concept of internally cohesive and externally bounded ethnic identities. Pieterse (2000) writes that the ability to name a group as an ethnic group points to relationships of power most often reserved by the state and other dominant groups. Chhachhi(1991) elaborates on her concept of ‘forced identities’ in the context of communal riots. A forced identity thrust upon an individual implies the burden of signifiers attached with that act of naming/labelling. For instance, Muslims in India carry the double burden of being ‘victims’ (such as in SCR’s recommendation for positive bias from the state in order to empower Muslims) as well as ‘aggressors’ (as in the depiction of Muslims by the Hindu right)8.
The following sections elaborate upon ‘forced’ ethnic identity in the construction of a unitary or cohesive Afghan/Bengali identity. These ethno-regional categories are reified and essentialised with the burden of imposed significations which are constantly contested in practice. The fieldwork was undertaken with the purpose of looking at two ethno-regionally different sections of Muslim migrant populations in Delhi in order to problematise the homogenous category of uniform ‘Muslim’ identity.
One of the first respondents, Mazar, who with his hour-long interview provided rich insights into the lives of rickshaw pulling migrants like him, was from Katihar—a district in the state of Bihar bordering the Malda district in West Bengal. He spoke in Hindi-laced Bengali. Halfway through the interview a fellow rickshaw-puller from Uttar Pradesh (UP) accosted him quite aggressively and said that obviously, being a Bihari, he could not have knowledge of Bengali. Mazar explained, in perfect, unaccented Hindi how he also knew Bengali, although he is a Bihari. His emphasis on the Bihari identity—the knowledge of Bengali not diluting that—was striking and definitely helped to diffuse the situation. Mobilising his linguistic skill as an embodied form of cultural capital, he was able to position himself as a ‘Bihari’ (to the aggressive UP man) and a Bengali (to me). The initial dilemmas about whether or not to use this interview—and others following it where the respondents were from Jharkhand or Bihar and not West Bengal—were soon resolved as it became clearer that such categorical purities were problematic.
None of the respondents self-identified as Bengali although they would display assent when I claimed affinity on the basis of that. Most often the respondents identified on the basis of their location, usually district: Malda (West Bengal), Farakka (Jharkhand/West Bengal), Katihar(Bihar)9 (see: Map1, Appendix I). However, within closer bonds of affinity, for instance among fellow rickshaw-pullers, the primary way of identifying collectively was through village-based kinship. Mitra et al. (2011) corroborate this finding by pointing out that fictive kinship based on villages of origin forms basis of social networks among migrants. A very common occurrence was introducing other rickshaw-pullers to me as, ‘He’s also from my village, didi10’.
So the category of ‘Bengali’ became broadened into ‘Bengali speakers’. In course of a group interview, the respondents unanimously agreed that discrimination was directed at the ‘Biharis’. When asked to elaborate, one respondent answered: People here do not know that there are rickshaw-pullers who are Bengalis. They only know that rickshaw-pullers are from Bihar. So they abuse us by calling us Bihari. Contrasting the response of Mazar to the rickshaw-puller from UP, with that of the respondent quoted above, it becomes clear how strategic the deployment of ethno-regional identities can be. In a country dominated by regional politics, the hierarchy of regional power is figured into identity. In the interaction between Mazar and the UP rickshaw-puller, being Bihari initially was synonymous with deception (deceiving the interviewer about Bengali identity). In the second case, being Bihari actually becomes a ‘forced’ identity that is synonymous with poverty and being powerless11, irrespective of actual geographical origins.
Linguistic particularities, in the form of dialect, vocabulary and pronunciation—as markers of difference, particularly of class,—was present in every interaction I had12. From Mazar giving me examples of his kind of Bengali (underscoring the difference with mine); to me misinterpreting words such as wife: ‘istiree’ in the dialect of one of my respondents from Malda and ‘stree’ in my tongue—the language used by my respondents and me were often different. Thus, even the category of ‘Bengali-speaking’ emerged as heterogenous.
The Bengali national consciousness evolved, primarily as a response to colonial discriminations, exclusively involving Hindu (upper castes) and (‘educated’) middle class (Sinha, 1999) and often actively alienating the marginalised groups like Muslim peasantry (Chowdhury, 1998: 161). And while it seemed very natural for me to deploy my ‘Bengaliness’ and through it, find Bengalis who were pulling rickshaws, such broad ethno-linguistic/regional categories had little meaning for the respondents.
Following Rapport and Dawson (1998), it can be claimed, then, that these circular/seasonal migrants do identify themselves in relation to a place of fixity, an idea of ‘home’. However, this home is not some abstract, ‘imagined community’ (Anderson, 1991) of Bengalis which cuts across other differences such as class, religion or region; but is the much more tangible physical places of origin. These places (villages) moreover, play an active role in their migration and livelihoods and continue to dictate social relations in the place of migration, as the following sections demonstrate.
‘Afghans’ in Delhi
As Table 1 demonstrates, the Afghans in Delhi are a myriad group of people. While the most obvious categories of internal difference might be the refugee/migrant status and specific ethno-religious categories, the reality is messier. In reports and articles Afghans have been polarised into dispossessed victims (Bose, 2004; The Hindu, 2007) or, more recently as successful entrepreneurs (Saini, 2012; Stillwagon, 2011a). In the field of entrepreneurial activities, Afghans have historically been associated in India with activities such as money-lending (Bose, 2004) and dry-fruits business (Stillwagon, 2011a). Moreover, a number of occupations such as translators, tourist guides (particularly related to the large-scale phenomenon of Afghan medical tourism in Delhi), restaurant-owners, real-estate agents, among others, have mushroomed around the increasing number of Afghans visiting and settling in Delhi (personal interviews with Anwar, Afzal and Asif).
Yet as Stillwagon’s (2011b) primary research in Bhogal shows, the categorical separation of refugee with economic migrants is not tenable:
[T]he restaurant manager: a tall, lean Afghan man wearing a fitted black dress shirt [says that h]e and his family migrated to India five years ago as refugees, because of high unemployment and child kidnapping at home.
Security (arising from conflict) and economic security emerge as contiguous issues through my primary research as well. Abid, the owner of a bakery in LNB, insists that he came to Delhi only for economic reasons. Yet, he is on a refugee card and not a visa. Asif, the co-owner of a restaurant in the LNB area is on a RSD waiting list. He left Afghanistan because of issues of security and came to Delhi on the invitation of a previous colleague who wanted a partner in his restaurant business. He has been in India for a few months and feels frustrated at the lack of opportunities to continue his study as a person without any stable legal status13.
On the other hand Tariq, the owner of a successful restaurant in Lajpat Nagar as well as with other business interests, is on a business visa and regularly goes back to Kabul where his family stays, ‘particularly during the hot months’. Anwar (from Khirki), currently unemployed and previously in a high profile job, insists, however, that there are only refugees who left Afghanistan fearing for their lives as most had well-paying jobs. This is clearly not the whole picture and one respondent, who shall remain unnamed, goes as far as to say that Afghans on medical and tourist visas are also setting up businesses in India, taking advantage of the legal loopholes. Thus, even internally the community has different perceptions about the reasons for leaving Afghanistan.
As far as internally recognised stable markers of identity goes, most respondents were reluctant to reveal their ethnic origins (like Tajik, Uzbek, Hazara, etc14) but less reluctant to identify by language (most seemed to be native Dari speakers with knowledge of Pashto and Hindi-Urdu, though they were equivocal about identifying as either Dari or Pashto speakers). Again, all readily identified as Muslim but were often reluctant to identify as Shia/Sunni.
I chose to focus on restaurant owners from LNB because they represented a melange of these different positions and also because of their visibility (section 3.2). Visibility of the ‘Afghans’ is, however, problematic. As Anwar pointed out, Hindu and Sikh Afghans were mostly indistinguishable to the Indians because of language, physical appearance and cultural practices. I discovered this to be true on many occasions when I would be surprised to learn that a particular business was being run by Afghan Sikhs or when I would hear Sikh gentlemen speaking with each other in Dari. Thus, those who are visibly identifiable as ‘Afghans’ are usually, almost always, Muslims. Samira used the term ‘ethnic’ Afghan to refer to the Muslims which seems to imply that only Muslim Afghans have a real claim upon the term ‘Afghan’ and that Hindus and Sikhs are not authentic Afghans. This was elaborated by Afzal in his interview:
Afzal: Afghan Sikhs are basically Indian origin [if you] trace back and that’s why Indian government provide[s] them with visa easily. So they’re staying legally.
Me: So they are coming on business visas?
Afzal: No…as of now they have residential permit, RP. So they have no problem, they can work. Government can’t object [to] them, you know. Because they are Indian origin so till the time they can get their naturalisation they are treated as an Indian, half-Indian…
Thus, these interviews reveal that the ‘Afghan’ as a stable identifier is not a tenable category. While the respondents strategically lay claim to an Afghan identity for different reasons, it is internally differentiated.
Forced Identities and Discrimination
It interestingly emerged that while the forced identification as ‘Muslim’ created negative stigma associated with Bengali speakers (as ‘Bangladeshi infiltrators’15), it did not seem to correspond with the experiences of Afghans living in Delhi. Most of my respondents of Afghan origin confirmed that while they had to pay higher rents as compared to ‘locals’, they were not routinely harassed by the authorities as for instance is commonplace in Iran, for Afghan refugees (personal interview). Yet, many also agreed that the general perception about people from Afghanistan equated them with Talibans (personal interviews). Abid, one of my respondents from Mazhar-e-Sharif and the most cautious and tight-lipped of the respondents, tried to distance himself from the ‘disturbances’ in Afghanistan repeatedly during the two interviews. He said he did not know about the situation since nothing ‘like that’ happened in the cities but only in remote places. He even pointed to my collaborator saying:
Actually I don’t know much about the situation…These people [white foreigners] will know much more than me about those things.
In explaining why the ‘housing apartheid’ (Ashok and Ali, 2012) did not affect Muslims of Afghan origin in a non-Muslim residential neighbourhood like Lajpat Nagar, Afzal offered an explanation. According to him the neighbourhood, originally a refugee colony set up for refugees (Hindus and Sikhs) leaving Pakistan during the partition of 1947, was more sympathetic to the plight of refugees having to leave their country of origin. This seems not entirely probable—in Bhogal with its higher concentration of Muslim Afghans, the Delhi Police’s public service announcements in the busy market area consists of messages cautioning residents about the necessity of verifying the identities of people seeking employment or seeking to rent property so as to avoid terrorists16. All the respondents of Afghan origins mentioned the fact that they had to pay higher rents, which might also act as a mitigating factor for the landlords who benefit more. Overall, it does seem that the ‘Taliban-Afghan’ stereotype has not become hardened into official practise and mainstream discourse (through the efforts of the Hindu right) as in the case of the ‘Bangladeshi-infiltrators’. And thus the ‘official script’ (Ramachandran, 1999) seems to considerably affect and influence everyday social practices of discrimination. While people may conflate Afghans with Taliban and terrorists17, the official leniency towards Afghans has influenced actual practice.
Interestingly enough, the rickshaw-pullers, although very forthcoming on most subjects were very reluctant to discuss religious discrimination. When pressed, the respondents in the group interview identified the police as harassers. But, particularly, when I tried to ascertain harassment pertaining to being identified as Bengali-speaking Muslims, there was uncomfortable, but resolute silence in several interview situations18. As Poland and Pederson (1998:294) point out, silences in qualitative research can be ‘profoundly meaningful’. However, rather than trying to analyse the meaning of the silences, I would like to suggest that the silences on part of the respondents was one of the most important way of resisting the efforts of an outsider-researcher to probe the realms of the personal. While the silence might indicate the fear of being identified as Muslim to an obviously Hindu (-named) researcher, it was also an indication that resistance to someone from a superior class position cannot always take the form of a direct, verbal refusal to answer. And silence, therefore, is a strategy to assert certain boundaries that I was stepping over through my questions. Silencing, is usually seen in a negative light, and silences often taken to affirm subordination, fear and oppression. But whether or not it indicates fear, it clearly demonstrates agentic capacity in the respondents’ refusal to reveal.