Bhrigupati Singh

Department of Anthropology & Sociology Ashoka University (India)

and Carney Institute Department of Psychiatry
Brown University (E-mail: [email protected])

Can a Neighborhood Fall Sick? Opioid Addiction,

Collective Violence and Currents of Death in
Contemporary India
Opioid abuse is an increasingly global phenomenon. Rather than assuming it to be a
uniformly global or neoliberal pathology, how might we better understand compar-
ative and locally specific dimensions of opioid addiction? Working with neighbor-
hoods as a unit of analysis, this article analyzes the striking differences between pat-
terns of addiction and violence in two proximate and seemingly similar urban poor
neighborhoods in Delhi, India. Rather than global or national etiologies, I suggest
that an attention to sharp ecological variation within epidemics challenges social
scientists to offer more fine-grained diagnostics. Using a combination of quantita-
tive and ethnographic methods, I show how heroin addiction and collective violence
might be understood as expressions of what Durkheim called “suicido–genetic cur-
rents.” I suggest the idea of varying currents as an alternative to the sociology of
neighborhood “effects” in understanding significant differences in patterns of self-
harm and injury across demographically similar localities. [opioid epidemic, India,
addiction, violence, neighborhoods and health]

Introduction: An Etiology of Ecological Variations

Beyond the immediacy of death tolls, how might we better understand opioid abuse
as an increasingly global phenomenon?1 Comparative diagnostics often begin with
national and cross-national metrics. In India, for instance, the prevalence of opioid
use is three times the global average (Ambekar et al. 2019: 3) (see Figure 1). India has
lower figures than the United States at present (see Figure 2), but this may change
in years to come as pharmaceutical companies more aggressively target emerging
economies, and obliging governments, such as the one currently in power in India,
gradually widen the range of available opioids.2
Besides national metrics of opioid consumption, what measures, maps, and read-
ings of symptoms might anthropologists offer? I use the term “symptom” here
as a form of etiology that may occur outside the realm of biomedicine, as Gilles
Deleuze describes it, for instance, in the naming and investigation of psychiatric

MEDICAL ANTHROPOLOGY QUARTERLY, Vol. 35, Issue 2, pp. 159–189, ISSN 0745-
5194, online ISSN 1548-1387. © 2021 by the American Anthropological Association. All rights
reserved. DOI: 10.1111/maq.12629

159
160 Medical Anthropology Quarterly

Figure 1. Opioid use in Indiaa [This figure appears in color in the online issue]

Figure 2. Opioid use comparison between India and the United States (All figures
in percentages)b [This figure appears in color in the online issue]

symptoms in literary texts (1997: xv). In what ways do anthropologists diagnose

symptoms? A potential analytic trap for “social” diagnostics is the assumption, of-
ten implicit, of psychic life as a mere epiphenomenon of economic and institutional
shifts. This occurs in different ways, for instance, when the highly variable trajecto-
ries of opioid substitution treatment-seekers are reduced to a normalizing technique
of neoliberal governmentality (Harris 2015). Even in an exquisitely experience-rich
near ethnography like Righteous Dopefiend (Bourgois and Schonberg 2009), the
 Can a Neighborhood Fall Sick? 161

etiology rests on a disappointingly standardized diagnostic: the “collapse of the in-

dustrial economy” (2009: 150) and the resulting emergence of a “lumpen” residue
(2009: 243).
Rather than uniformly global etiologies such as neoliberalism or post-Fordism,
consider instead the diagnostic problem posed by ecological variation. As Robert
Sampson, a leading voice in the contemporary sociology of neighborhoods, asks:
“Why are so many health-related outcomes concentrated ecologically?” (2003: 132).
In one of the first in-depth studies of neighborhood ecologies of opioid addiction,
Juan Gamella studied the sharp rise of heroin injectors in Spain: “In 1978, there
were only a few dozen in Spain; by 1982, there were already tens of thousands”
(1994: 131). Studying patterns of initiation and spread with pioneers, recruiters, and
friendship networks in Valdemina, a neighborhood in Madrid, Gamella asks how
heroin comes to be “massively adopted, particularly during some periods, and why
by some groups rather than others.” Ecological variation is not simply a question of
local context. Rather, what is at stake are sharp variations within the “same” context
such as adjacent, socio–economically similar neighborhoods. As Gamella further
points out, a significant obstacle to focusing on these forms of variation lies in the
term epidemic itself: “In providing a fiction of an explanation, the epidemiological
metaphor (epidemic) hinders the understanding of some forms of drug use, especially
those more tabooed and destructive forms (1994: 154).
How might we approach questions of ecological variation in relation to addiction
trajectories? I use term ecology here as derived from its root word oikos (dwelling
space) (Simpson 2013). Taking urban neighborhoods as a an oikos, and as a viably
emic unit of analysis,3 in this article I examine a stark variation in the form that
heroin addiction takes in two demographically similar and proximate “resettlement
colonies”4 in Delhi, Trilokpuri, and Sundarnagri. Further, I contend that addiction
trajectories are better understood not as isolated pathologies, but as one among
other forms of life and death within specific ecologies. As such, I consider a sec-
ond ecological variation between these two neighborhoods, their sharply differing
propensity for collective violence. Rather than being causally related, I take these
two phenomena, heroin addiction and collective violence, to be varying expressions
of what Durkheim in Suicide called “suicido-genetic currents” (1997 [1897]: 325).
Suicide, I argue, is a founding text in the study of ecological variation, as Durkheim
compares administrative districts in Europe to ask, “how a definite number of peo-
ple kill themselves in a particular way, repeatedly over time” (1997[1897]: 48), and
how these forms of death vary, sometimes quite sharply, between proximate and
seemingly similar areas.
The first part of this article sets out the empirical puzzle of the two neighbor-
hoods in Delhi. Conceptually, I suggest an anthropologically oriented idea of suicido-
genetic currents as distinct from neighborhood “effects” studies in urban sociology
and public health, as a way of approaching the study of ecological variation. In
order to sharpen this contrast, I examine possible explanations for the significant
variations between these two neighborhoods in Delhi, drawing on the kinds of data
and methods that would be used in neighboring disciplines such as sociology, pub-
lic health and psychiatry. We see how this data demonstrates primarily similarity
rather than variation. Instead, as a step towards an anthropological diagnostics,
162 Medical Anthropology Quarterly

I emphasize compositional differences of caste that result in seemingly minor but

consequential differences in local economies. I show how these compositional ele-
ments are differently activated by currents animating particular forms of life and
death, for heroin addicts, and in the expression of political violence in these neigh-
borhoods. I conclude by suggesting that this form of anthropological diagnostics, an
etiology of ecological variation and the idea of currents as a way of sensing the pulse
of variations across particular spatially-defined units of analysis, might allow for a
more fine-grained picture of psychic distress rather than monolithic explanations of
neoliberal alienation.

Figure 3. Map of Trilokpuri and Sundarnagric [This figure appears in color in the
online issue]

The Empirical Puzzle: A Tale of Two Neighborhoods

My starting point for this article was a seemingly local difference between Trilokpuri
and another resettlement colony, Sundarnagri, where the hosts for my research, the
All India Institute of Medical Sciences (AIIMS, Delhi)5 , run Opioid Substitution
Therapy (OST) clinics, begun in 2003 with funding from the Ministry of Health,
the National Aids Control Organization, and the United Nations Office of Drug
Control. A strangely compelling ethnographic imperative was posed to me, with
some urgency, by my hosts, psychiatry faculty and residents at AIIMS, as I focused
primarily on Trilokpuri in the first few months of my research. “You must also visit
our other clinic,” some said, “and tell us why it is so different from Trilokpuri!”
“Different in what way?” I asked. “It’s more dangerous,” my hosts replied intrigu-
ingly. This other, more dangerous clinic was in Sundarnagri, 12 km away (Figure 3),
settled in the same year as Trilokpuri, in 1975–76 (Figure 4).
 Can a Neighborhood Fall Sick? 163

Figure 4. Brief municipal comparison of Trilokpuri and Sundarnagrid

At first glance, the Sundarnagri clinic did have an edgier atmosphere. Patients oc-
casionally threatened the clinic staff verbally or with knives. The patients, all male,
were gaunter and had hollower eyes, of a kind I had not seen in Trilokpuri. There
were fewer accompanying mothers and wives. Rather than only a psychiatric con-
struction of risk, the difference between the two clinics could also be posed in more
emic terms, in the language and experiences of addicts themselves.
As most nashe-baaz (addicts) describe it, there are three ways to ingest heroin.
It can be smoked or chased (inhaled) by heating on a tin foil (panni) (Strang and
King 1997), costing roughly Rs. 250 a piece.6 In a downward hierarchy, in a dif-
ferent category from smokers and chasers, come the injectors with their “set,” sui
(needle and syringe) and sheeshi (small glass bottles with buprenorphine and avil,
an antihistamine), available from secretly known chemist shops for Rs. 50 or less,
one-fifth the price of a smoker or chaser’s habit. As most addicts put it: “A chaser’s
life is ten years long, an injector’s life is two years at most.” In addiction psychiatry,
the heightened vulnerabilities of injectors are well known, including the risk of con-
tracting HIV or Hepatitis B through shared needles, with mortality rates up to 10
times higher than the general population (Solomon 2009).
So we have two quite different modes of consumption, with smokers and chasers
on the one hand, and injectors on the other. Methodologically, based on interviews,
patient files, and days and nights that I spent in both neighborhoods with treatment-
seekers, addicts, residents, and through household surveys,7 the initially intuitive dif-
ference emerged more sharply: Trilokpuri had barely a single injector. In contrast, in
Sundarnagri, injector groups and the detritus of used injections were commonplace
in neighborhood spaces such as parks. The difference in clinic figures is quite stark
164 Medical Anthropology Quarterly

Figure 5. OST clinic patient profile comparisone

(see Figure 5), with only four injecting drug users (IDUs) and six HIV+ patients in
Trilokpuri out of a total of 3,138 patients. In Sundarnagri, out of 1,773 patients,
300 are IDUs and 137 are HIV+.
On a national scale too, IDUs in India show a much higher rate of HIV than, for
instance, commercial sex workers, often assumed to be the primary drivers for the
spread of HIV (see Figure 6).
As such, I arrived at the guiding question for this article: Can a neighborhood fall
sick in ways different even from a proximate locality? Injecting is one among other
forms of self-injury. Inhabiting daily life in these localities, it felt unavoidable to also
consider a second ecological variation, where the balance sheet of injuries between
these two neighborhoods undergoes a sharp reversal, namely, collective violence.

Figure 6. HIV/AIDS in Indiaf

 Can a Neighborhood Fall Sick? 165

Figure 7. News media comparison of reported incidents of riots in Trilokpuri and

Sundarnagrig [This figure appears in color in the online issue]

Trilokpuri, the “healthier” neighborhood from the perspective of heroin addiction,

is known in Indian news media as the “riot laboratory” of Delhi, with the highest
death tolls in the city during the 1984 anti-Sikh riots, a major Hindu–Muslim riot
in 2014, and periodic outbreaks of violence since (Sharma 2018). In contrast, there
has never been a riot in Sundarnagri, as news archives8 (see Figure 7), interviews,
and surveys confirmed.
Like Trilokpuri, Sundarnagri is a mixed Hindu–Muslim neighborhood, accord-
ing to census data and our household surveys (see Figure 8) that show a number of
demographic similarities, including religion, place of birth, home ownership, posses-
sion of state documents, and levels of crime reported in neighborhood police stations
(see Figure 9), which I closely tracked over the course of 2015.
The striking ecological difference that interests me here is between two different
forms of self-harm. Trilokpuri does not inject itself, but it produces periodic riots.
No such riot has ever occurred in Sundarnagri. This is not to say that interreligious
animosity or the potential for violence does not exist in Sundarnagri. Since 2015,
I closely followed and interviewed key players in three incidents of Hindu–Muslim
disputes (one of which is discussed below) in Sundarnagri that stopped short of ex-
plicit violence each time, such that Sundarnagri residents could still say “ours is not
a danga-grast (riot-prone) area like Trilokpuri.” How might we find our conceptual
bearings for what is being compared here?

Suicido–Genetic Currents versus Neighborhood Effects

In what ways have adjacent disciplines engaged with this question of neighborhood
comparison? As Robert Sampson describes it in The Great American City (2013),
there is a century-long tradition of neighborhood effects studies in sociology. To an-
thropological eyes, in this genre of analysis, a forbiddingly large sample often leads
to disappointingly reductive comparative diagnostics. As an overview of neighbor-
hood studies of mental health puts it: “Better neighborhoods are associated with
better mental health” (Truong and Ma 2006: 151). How is “better” measured? Is
166 Medical Anthropology Quarterly

Figure 8. Demographic comparisons and possession of state documents among

residents of Trilokpuri and Sundarnagrih

it simply a function of socioeconomic advantage? How would the idea of better

be conceived when comparing two demographically similar urban poor neighbor-
hoods? As Sampson describes it, there have been a range of sociological theories
in response to this question of variation across economically and demographically
similar neighborhoods, drawing on social disorganization theory, ideas of social
capital, degrees of state “desertification,” spatial segregation, stigma, and compar-
isons of collective efficacy or “felt cohesion among community residents” (Sampson
2013: 367).
In comparing Trilokpuri and Sundarnagri through household surveys, health and
morbidity surveys, vignette-based and psychiatric instrument-based surveys, on the
face of it, most comparators were surprisingly similar. This was the case for a number
of indicators, including the presence of medical facilities, schools, public amenities,
shops, the number of below-poverty-line residents, possession of state documents,
and homeownership. We also undertook vignette-based surveys to study levels of
social cohesion, akin to the methods described by Sampson, where a description
of possible incidents and ongoing exchange relations are used to study feelings
of whether or not “people in this neighborhood can be trusted” (2013: 156) (see
Figure 10).
 Can a Neighborhood Fall Sick? 167

Figure 9. First Information Reports (FIRs) filed at neighborhood police station

from January to December 2015i

Psychiatric comparisons such as the Perceived Stress Scale and the WHO-Quality
of Life instrument also yielded surprisingly similar results across both neighbor-
hoods (see Figure 11).
How else might we conceptualize the study of ecological variation? The difference
here is not between quantitative and qualitative methods. Anthropological compar-
isons of ecological variation may also be quantifiable. Consider a striking instance of
an anthropological approach to ecological variation in Veena Das’s Life and Words
(2007) in the chapter, “The Force of the Local,” set in the resettlement colony of Sul-
tanpuri in Delhi, in the aftermath of the 1984 anti-Sikh riots. To document forms
of state complicity and as part of relief efforts, Das and her colleagues surveyed 523
Sikh households in 13 blocks of Sultanpuri (2007: 143). In creating block-wise tal-
lies of the loss of property and persons killed per block, Das encountered a striking
ecological variation. Within Sultanpuri, in Block A4, which had 145 Sikh house-
holds, 79 persons were killed. In contrast, in the neighboring two blocks, which had
a total of 180 and 40 Sikh households respectively, zero persons were killed (2007:
148). Das asks: “If the riots were simply an expression of blind rage […], how did
the crowds act to differentiate the kinds of persons against whom violence was un-
leashed? […] How do we account for the particularity of the choices that seem to
have been made?” (2007: 148). Das demonstrates how a national event folds into
everyday relations, in this case, neighborhood inter-caste rivalries between Chamars
and Siglikar Sikhs. Rather than being a spontaneous expression of social tensions,
168 Medical Anthropology Quarterly

Figure 10. Quotidian network comparisons among residents of Trilokpuri and

Sundarnagrij

Das emphasizes the ways in which acts of violence are differentiated and “autho-
rized” through local connections with representatives of the state.
In relation to ecological variations, what may be said to vary here? In his fore-
word to Life and Words, philosopher Stanley Cavell (2007: xvi) suggests that these
forms of violence (as a mode, however distorted, of collective action) might be read
not so much as forms of life in the Wittgensteinian sense of the term, but as “forms
of death” (2007: xvi). How might we study ecological variations in forms of death?
My focus here is on mundane and yet nationally and globally significant forms of
death such as heroin addiction and forms of collective violence that remain below the
threshold of a riot, for instance in what news media periodically report in Trilokpuri
as “skirmishes.” A form of death is not the same as morbidity or a death toll. Rather,
it is an expression, a current of something like the opposite of social vitality. Anthro-
pologically, how might we conceptualize the fluctuating currents that injure a social
body? What kinds of anti-vital currents are these?
 Can a Neighborhood Fall Sick? 169

Figure 11. Psychiatric instrument–based comparisons among residents of

Trilokpuri and Sundarnagrik

Here, I turn to Durkheim as a founding theorist not only of social structures, but
also of currents, a concept that recurrently appears at the margins of Durkheim’s
texts and is thus often ignored, although it may also be received as a key to rereading
Durkheim as a thinker of energies in ways resonant with but distinct from Bergson’s
“elan vital.” For instance, the commonly taught interpretation of The Elementary
Forms of Religious Life (Durkheim 2001) is of religion as society “worshipping it-
self,” resulting in the creation of collectivities with shared moral norms. A less taught
theme in the Elementary Forms is of religion as an engagement with a vital animat-
ing principle, “an anonymous and impersonal force … none possesses it entirely
and all share in it” (2001: 140). And further: “When we say that these principles
are forces, we are not using the word in a metaphorical way: they behave like real
forces. … If an individual comes into contact with them without taking precautions,
he receives a shock that can be compared to an electric charge” (2001: 142).
In continuation with these energetic preoccupations, Suicide marks Durkheim’s
move from a kind of collective Eros to Thanatos, a death drive in a non-Freudian
sense. Suicide is not just a comparison of death tolls. Rather, Durkheim is interested
in how social formations are “predisposed to contribute a definite quota of voluntary
deaths” (1997 [1897]: 51). As Durkheim puts it, these “predispositions” is the main
subject of Suicide. These predispositions are activated by “collective forces of a def-
inite amount of energy, impelling men (sic) to self-destruction” (1997 [1897]: 299).
Durkheim calls these forces “suicido–genetic currents” (1997 [1897]: 325). Such
currents vary in intensity and form (1997 [1897]: 363). Currents are not determin-
istic. They may encounter varying “degrees of resistance” (1997 [1897]: 323). This
flux notwithstanding, certain forms of death repeat with relative regularity. Such
regularities might be understood, not as a form of “imitation,” as Durkheim argues
contra Tarde,9 or as a quirk of “individual psychology” (p. 311), but in relation to
forces “in their common environment inclining them in the same direction” (p. 305).
170 Medical Anthropology Quarterly

As such, I mark two openings for empirical investigation: compositional elements of

a milieu (“their common environment”) and the ways in which these elements are
activated by specific forms and intensities of suicido–genetic currents, as these are
conducted or resisted. Let us take this formulation a step closer to Trilokpuri and
Sundarnagri by examining a compositional difference, after which we will consider
how these differences are variably activated by suicido–genetic currents in relation
to heroin addiction and collective violence.

Compositional Differences in Trilokpuri and Sundarnagri

Alongside administrative spatial categories, how might we understand the composi-
tion of neighborhoods? My earlier work in rural India (Singh 2015) made me look
at Indian cities differently, better understood, like villages, primarily as caste-based
clusters in poor and elite neighborhoods. The seemingly anonymous slum dwellers
who were relocated to “resettlement” colonies in Delhi10 belong primarily to the
lowest castes of the Hindu and Muslim hierarchy. That said, rather than a unified
category of Dalits/Scheduled castes, consider instead the significant differences in the
political history of the two numerically dominant groups among north Indian Dalits,
the Balmikis (the “sweeper” caste), predominantly centrist Congress voters until the
early 2000s, with a discernible shift to the Hindu right-wing BJP over the last decade
(Jaoul 2011; Lee 2015), as distinct from the Chamars (the “tanner” caste, associated
with animal hides and carcasses, but also engaged in various forms of agricultural
and artisanal labor), political innovators among north Indian Dalits as founders of
the lower caste Bahujan Samaj Party (BSP), a major electoral force facing a decline
in recent years.
As such, instead of a more generalized category of Dalit, if we consider the specific
distribution of Hindu castes in these neighborhoods, in Sundarnagri we find a mix of
Koli-weavers, Jatav-Chamars, and other north Indian lower castes. (see Figure 12).
In contrast, the predominant group numerically in Trilokpuri are the Balmikis, ar-
guably the lowest-ranked social group in the Hindu caste hierarchy, whom Gandhi
called “Harijans,” earlier called “Bhangis,” a term that is now illegal for its pejora-
tive connotations.11 In affirmative action processes mandated by the early postcolo-
nial state, nearly every city sweeper employed by the Municipal Corporations of In-
dian cities, are Hindu Balmikis by caste.12 Ambiguous as this might sound given the
enduring hierarchies of labor, to be a state-employed sanitation worker is a salaried
profession with a middle-range monthly income, which can be life sustaining in the
landscape of the urban poor, around Rs. 35–40,000 per month for a pakka (perma-
nent) employee, and about half of that for a temporary employee.
In contrast to the salaried Balmikis, Muslim families in both neighborhoods
are far more precariously situated economically, although also working within the
waste, recycling, and meat economies as kabadiwallahs, butchers, and daily wage
laborers.13 Trilokpuri is divided into blocks numbered from 1 to 36. Some blocks
in Trilokpuri are mixed by caste and religion, while some are known quite defini-
tively as Balmiki (Block 8, Block 28) or as Muslim (Block15, Block 27) blocks. In
contrast, the 11 blocks of Sundarnagri are more mixed among Hindu castes and
religions, without the predominance of one social group, with the exception of Block
O, which is known primarily as a Qureshi-Muslim block.
 Can a Neighborhood Fall Sick? 171

Figure 12. Caste and religious demographicsl

A related compositional or experiential difference, harder to capture even in in-

terviews, is a qualitative feel of Trilokpuri as a neighborhood defined by a kind of
upwardly mobile Hindu Balmiki identity, as a caste of salaried employees linked to
the Municipal Corporation of Delhi, in ways that creates a form of economic sta-
bility and political presence, evidenced, for instance, by BJP-allied Balmiki leaders
dominating municipal ward and legislative assembly elections in the last decade. Hi-
erarchies of labor are also open to renegotiation, often informally. For example, a
Balmiki MCD employee may create an informal labor arrangement with a poorer
migrant (typically called “Bangladeshis,” although they may or may not have mi-
grated from Bangladesh) for more socially devalued tasks such as manual scavenging
or sewage work. Other ecological hierarchies are harder to renegotiate. Balmikis are
also the only caste who breed pigs and consume pig meat. That practice remains cen-
tral to Balmiki identity and neighborhoods, and is a source of stigma in the discourse
about Balmikis among neighboring Hindu castes and Muslims.
As distinct from Trilokpuri’s more sharply contrasting (salaried) Balmiki (in-
formal economy) Muslim identity, qualitatively Sundarnagri is more of a mixed,
daily wage-dependent, upper- and lower-caste Hindu and Muslim neighborhood,
distributed across a range of professions, some of which overlap with Trilokpuri,
such as shops, thelas (vegetable vendors), electronics and car spare parts markets,
and construction labor. This compositional difference has economic consequences.
It is possible to argue that Trilokpuri is a marginally better-off neighborhood. This
marginal contrast appears fleetingly in household surveys, for instance, in differences
between the two neighborhoods in the average monthly expenditure on electricity,
rents, private health care, and differences in income across similar professions (see
Figure 13).
172 Medical Anthropology Quarterly

Figure 13. Domains of higher income expenditure and earnings in Trilokpuri as

compared to Sundernagrim

Rather than a neighborhood “effect” that would allow us to determine uni-

formly better or worse outcomes for health and social cohesion instead as a more
fine-grained anthropological diagnostic, I consider how these compositional differ-
ences are differentially activated, in the contrast between the prevalence of IDUs and
expressions of collective violence, respectively considered not as an epidemic or a
culture of violence, but as suicido–genetic currents. A crucial feature of a current is
how it is conducted and the “degree of resistance” it encounters (Durkheim 1997
[1897]: 323). I now turn to each of these currents respectively, and the forms of re-
sistance they encounter, as we come closer to the pulse of these two neighborhoods.

The Afterlife of IDU Deaths

Consider a cause-and-effect-based answer for the presence of IDUs in Sundarna-
gri and their absence in Trilokpuri: Given that injecting is on average one-fifth the
price of smoking and chasing, the marginally better-off residents of Trilokpuri could
perhaps afford the more expensive habit as distinct from Sundarnagri, where the
 Can a Neighborhood Fall Sick? 173

cheaper, quicker, more intense rush of injecting took root within a more unstable
local economy. Such an answer, though, would beg further questions. Did potential
Balmiki addicts in Trilokpuri not even try the novel intensities of injecting? Why
were they not drawn further in, given that addiction trajectories do not necessarily
follow economic rationality? Further, what prevented the Balmikis’ neighbors, the
economically poorer Muslims in Trilokpuri, from creating an injector subculture
and finding their own spaces in the neighborhood?
Economic factors were not inconsequential, but the answer I offer begins in a
more contingent set of local events, as the chaska (craze) for injections first spread in
both neighborhoods, beginning in 2007, as most addicts describe it,14 starting from
Seemapuri, further north and then spreading to adjacent neighborhoods. Among a
group of injectors, there is often an expert skilled at finding veins and key to initiating
new IDUs. In Trilokpuri, as most ex-injectors describe it, botched injections in the
first few months of 2007 led to a series of sudden deaths. These deaths, recounted
with horror, initiated a strong degree of resistance to the current of heroin injecting
in Trilokpuri. For instance, Suresh Balmiki,15 in his mid-40s is one of the oldest
patients in the Trilokpuri clinic. He lives across the street from the clinic in Block
2 of Trilokpuri. His mother worked as a sweeper in an adjacent clinic. Suresh had
a key to her clinic and would use it as a “hotspot”16 with friends after hours. For
many years, he was second in command to one of the best-known gang leaders of
Trilokpuri. As such, he initiated many young men in Trilokpuri into smack (heroin).
I met Suresh every few months over the last five years and visited his mother and
sister, as well as the pride of their house, Pawan, a pet pig, weighing over a 100 kg,
or so it was said. Knowing my interest in the comparison between Trilokpuri and
Sundarnagri, Suresh introduced me to many other ex-injectors. He often returned
to an incident that had been a turning point in his own addiction trajectory:

One of my closest friends, Guddu died in front of me, thrrr thrr thrrr … with
blood spurting out of his veins. After that I decided it’s better to spend the
250 rupees for a panni, rather than these dangerous injections. We even
started beating up any injectors we saw, saying Hat! Gandi addat phela raha
hai. Karna hai tau panni pe kar. (Shoo! Stop spreading this dirty habit! If
you have to do it, then use a foil). We’d make them murgas (bending over
like chickens, a humiliating form of punishment).

Suresh could afford this switch back to chasing because he is a municipal safai
karamchari (sanitation worker) in charge of three lanes in Block 2 of Trilokpuri.
Some who are affected worse than Suresh, such as his cousin Ramesh, one of the
few HIV+ patients in Trilokpuri, gave up smack altogether, supported by his mother,
an MCD sweeper.
What about Muslim smack addicts in Trilokpuri? While the religion, caste, and
even the family of an addict are almost always known in Trilokpuri, addict groups
and friendships can cross and maintain such boundaries. Hotspots are shared by
Hindu and Muslim addicts. The emergence of a form of resistance to the current of
heroin injecting in Trilokpuri, among economically worse-off Muslim addicts was
not simply a form of imitation of their Hindu neighbors. Nor was it necessarily the
threat of being beaten up, since this could have been resolved by finding separate
174 Medical Anthropology Quarterly

spaces. Rather, the resistance seemed to emerge from a kind of afterlife of particular
sharply remembered deaths and the impact that certain forms of death may leave on
proximate others. As a neighborhood, Trilokpuri is in no way a stranger to violent
deaths. Yet, there was something about injector deaths that was narrated as a kind
of desperate, undignified fatality.
For instance, among Muslim ex-addicts in Trilokpuri, the person I am closest to
is popularly known as Ashfaq Kebab-wallah, since he runs a buffalo-meat kebab
shop in Block 27, the Muslim garh (heart) of Trilokpuri. Between 2000 and 2005,
Ashfaq and two friends of his ran a taskari (peddling) business, selling smack
at a teashop outside Block 15, which stopped after they were arrested. For the
last three decades, the sale of smack in Trilokpuri has been a monopoly held by
two Hindu saansi (still popularly referred to by their colonial classification as a
“criminal” caste/tribe) families in Block 31, who pay a large “monthly” to the
local police. It is said that the saansi monopoly is protected by the police, based
on the understanding that their presence limits the further proliferation of smack
dealers. As such, Ashfaq’s foray into taskari had crossed an unspoken line within
neighborhood economies. Ashfaq’s brother Sultani died of an injection overdose in
January 2008, as did his close friend Irfan.
Ashfaq’s father had been among the first eight plot-holders in Trilokpuri in 1975.
Ashfaq and his four brothers were among the leading badmashes (miscreants/gang
leaders) of Trilokpuri and had been involved in a variety of violent clashes over the
last few decades. Nonetheless nearly every member of Ashfaq’s family—his father,
his mother, and his brother—had described Sultani and Irfan’s deaths to me, with
details that, as I heard them, constituted a kind of mourning and an afterlife that
exerted pressure on the living.
Here are some of the details I heard many times over, from Ashfaq and others:
Sultani’s screams the night before with tootan (“breaking,” the Hindi–Urdu term
for smack withdrawal); the 100 rupees that Ashfaq’s father gave Sultani, unable
to bear the screams, how this happened two days before Eid; all that was cooked
for Eid that year; the two days for which Sultani disappeared; how he died in a
garbage dump; how Ashfaq overheard news of his death from smack addicts in a
public bathroom; how they rushed to the nearby Guru Teg Bahadur hospital to
claim Sultani’s body; how his body was about to be cremated (in the Hindu style)
rather than buried when they reached the morgue. A few months later, Ashfaq gave
up smack, although now 10 years later, he still collects his biweekly Buprenorphine
dose from the Trilokpuri clinic.
Sundarnagri, too, saw a number of such deaths. As I tracked treatment dropouts,
comparing 100 patient files in both neighborhoods, I found that there had been a
much higher heroin-related death toll in Sundarnagri than in Trilokpuri (see Fig-
ure 14).
The deaths in Sundarnagri were not unmarked, but living companions continued
along a different trajectory of currents without the emergence of a form of resis-
tance. For example: Among injectors in the Sundarnagri clinic, I became fond of
Imtiaz, 20 years old, said to be a talented pickpocket, adept at stealing cellphones,
usually working with two friends, Waseem and Johar, both of whom died during
my fieldwork. Certain routine interactions in the Sundarnagri clinic would simply
not have been possible in Trilokpuri. For instance, Johar often went to the clinic
 Can a Neighborhood Fall Sick? 175

Figure 14. File-based comparison of patient records for HIV status and mortality
in OST clinics in Trilokpuri and Sundarnagrin

terrace to inject himself right after consuming his daily dose of Methadone. Some
of his peers grimly murmured that he had begun to inject himself on his ling (penis).
“Yeh last stage hota hai (this is the last stage) when there are no other veins left.”
Kamlesh-ji, the clinic’s needle-exchange outreach worker, in her early 60s, whom
many patients affectionately called “Mummy,” still treated Johar warmly. “Apni
chain tau upar kar le” (at least zip up),” she would yell, as he returned from the
terrace after injecting himself. It came as no surprise, one evening in October 2016,
when we heard from Kamleshji that Johar’s mother had called late at night to tell
her that he died of an overdose.
As we spoke after his death, Imtiaz praised Johar: “He wouldn’t steal wallets, only
mobiles (phones). If it was a small mobile, he would throw it back down and tell the
person they’d dropped it. He was a real artist.” Contrary to possible assumptions of
familial abandonment, Imitiaz and Johar both had relatively supportive families. In
2015, Imitiaz’s parents took a loan for Rs. 80,000 and bought him an e-Rickshaw,
but this was stolen, and Imtiaz relapsed. Some months after Johar, Waseem also died.
I met Imtiaz at the clinic, two days after Waseem’s death. Having known Imitiaz for
a year, I asked if I might record our conversation, punctuated though it was by his
tears. Here is a brief excerpt:

Imtiaz: Waseem died day before yesterday. He was my closest friend. We

were both in the Janta Flats school together. His marriage had been
arranged, just a few days ago. He died at the Tahirpur ditch where I sit with
chacha (an older injector, a leading initiator in the neighborhood). “Injection
laga reh gaya.” (“The injection remained stuck.”) That day he had “earned”
a big mobile, a Samsung. So he went to celebrate with Zulfikar (another
initiator friend). He had just started. A brother of his died similarly, learning
to do smack. It is Zulfikar’s fault too. He should have known that Waseem
wouldn’t be able to take it. He should have given him a “five point” (half a
syringe), but he filled it fully, tau usko overdose ho gaya (he overdosed).

Imtiaz himself had a near death experience six months ago with chacha. Imtiaz’s
account of the event was quite chilling for me:
176 Medical Anthropology Quarterly

Imtiaz: Chacha just left me there and went home. He didn’t even tell my
family. He thought “akad gaya thand mei.” (“I had died/stiffened in the
cold.”) A rickshawallah from my neighborhood came to the ditch to pee,
naseeb tha mera (“it was my fate”), seeing me, he informed my family. The
gypsy-vale (a police car) took me to GTB (a local hospital). My mother saw
my slippers at the ditch and rushed to the hospital. She was prepared for a
postmortem. They said I wasn’t breathing. My mother arrived and hearing
them say that, she started wailing. They gave me an injection. My mother
went on pressing me. Then upar vale ki dua se saans aa gayi (through a
divine blessing, I began breathing again). All the doctors were amazed. One
leg of mine had become huge. The SHO (police officer) asked me—who left
you there? But I didn’t give them chacha’s name.

I met Imtiaz’s mother a few times as well, since she would occasionally accom-
pany him to the clinic. Imtiaz’s family belongs to the Mansoori (weaver–artisan)
caste among Muslims, making rajai (comforters) in the winter and running a sug-
arcane juice stand in the summer. Two days after Waseem’s death, in mourning, or
so he said, Imitaz rejoined chacha at the ditch, having avoided him for six months
after his near-death abandonment.
Is there a way to compare mourning and the afterlife of particular forms of death?
I hesitate to call Imitiaz’s tears for Waseem “death without weeping,” as Nancy
Schepper-Hughes famously put it. Maybe there is no way to compare Imtiaz’s tears
with Ashfaq and his family’s grief for Sultani, or what Ashfaq might have become
had he lived in Sundarnagri rather than Trilokpuri. Instead, we might say that in
Trilokpuri a form of resistance developed to this current, in a way that it did not in
Sundarnagri, where this form of life and death took root and grew, leading to the
sharply different comparative figures between these two neighborhoods. These local
differences can also be nationally significant, as we see with the figures for HIV in
India. Let us now look at a different form of life and death to understand the quite
similar potentiality but very different actuality of collective violence across these
two mixed Hindu–Muslim neighborhoods.

Currents of Collective Violence

As with the prevalence of IDUs, there could be a simple causal explanation for the
different levels of collective violence in these two neighborhoods, as an effect of
their distinct political formations. In modern Indian politics, castes are often linked
as a bloc to particular political parties. Sundarnagri remains competitively dispersed
among Hindu castes, each linked to a different political party, with the Kolis asso-
ciated with the Aam Aadmi Party,17 said to have begun in Sundarnagri where a
party leader and the current chief minister of Delhi, Arvind Kejriwal, co-founded
an NGO, Parivartan, with an assistant from the neighborhood, Santosh Koli. The
Jatavs/Chamars are allied primarily with the Bahujan Samaj party (BSP),18 while the
Hindu right-wing BJP in Sundarnagri retains its upper-caste base, characteristic of
the BJP nationwide until the early 2000s, with two Baniya shop-owners in conflict
with a Brahmin for the seat of ward councilor.
 Can a Neighborhood Fall Sick? 177

Figure 15. News media reportage of 2014 Trilokpuri riots (block 27/28)o
[This figure appears in color in the online issue]

In contrast, as I argue above, Trilokpuri is strongly defined by the increasingly

Hindu right-leaning Balmikis. The epicenter of the riot laboratory in Trilokpuri is
a marketplace situated between two blocks (see Figure 15): Block 28, an entirely
Hindu Balmiki block known for its pig-meat shops and Block 27, the Sunni Muslim
heart of Trilokpuri, dotted with Muslim-run buffalo meat shops.
This spatial proximity of the pig-breeders and buffalo-meat sellers is not neces-
sarily a recipe for violence. As anthropologists of collective violence in South Asia
have recurrently shown, latent tensions can only escalate with the support of state
mechanisms. Until even a decade ago, relations between Block 27 and 28 could have
been described as a form of “agonistic intimacy” (Singh 2011). At times, Muslims
of Block 27 still invoke agonistics in relation to the Balmikis, “humne unko kabhi
haavi nahi hone diya” (“we never let them subjugate us”). In recent years, though,
the affiliation of the Balmikis with the more brazenly violent Modi-led BJP, a former
agon has given way to a networked form of state-supported subjugation.
There can now be a line of continuity between the ex-addict Suresh Balimiki
above, to the formerly ill-reputed but now turned respectable MCD councilor Kamal
Beniwal (see Figure 16), a Balmiki leader of Block 28, and a step higher to the former
Member of the Legislative Assembly (MLA), Sunil Vaid (see Figure 17). Sunil Vaid
was a Balmiki resident of Block 8, who, in the widely reported 2014 Trilokpuri
violence, in a move that is perhaps an element of all major riots and pogroms, is said
to have told the police to stay out of Trilokpuri for three hours to give the Balmikis
178 Medical Anthropology Quarterly

Figure 16. Trilokpuri Balmiki political leaders (Kamal Beniwal)p [This figure
appears in color in the online issue]

Figure 17. Trilokpuri Balmiki political leaders (Sunil Vaid/Kiran Vaid)q [This
figure appears in color in the online issue]

an opportunity to “teach the Muslims a lesson.” Sunil Vaid died of a heart attack
some months later, onstage during a speech, in divine retribution for his crimes, as the
Sunnis of Block 27 insist. His wife, Kiran Vaid, has now emerged as a neighborhood
political leader. In step with the current regime, she is more brazenly anti-Muslim
than Sunil Vaid was.
The activation of these compositional differences into forms of violence in
Trilokpuri requires a closer examination of the movement of suicido–genetic cur-
rents, namely, the circuits through which a charge may travel and the forms of re-
sistance it might encounter. As the comparative table of police reports above shows,
both neighborhoods have very similar figures for petty crime. With this similarity in
mind, I offer examples of currents from each neighborhood specifically in relation
to Hindu–Muslim violence. In Trilokpuri it felt like every few days I would hear of a
potential or actual balva (skirmish). After staying there intermittently for two years,
I began to more acutely sense the circuits, for instance of rumor, through which
 Can a Neighborhood Fall Sick? 179

currents of violence travel. At times, my two main hosts in Block 27, Ashfaq (the
ex-addict) and Sufiji (a healer, in his early 70s) would voice misgivings characteris-
tic of more widespread Hindu–Muslim “regions of rumor” (Das 2007: 112): “Our
water supply is contaminated … by the Balmikis and the Jal (Water) Board (hinting
at Balmiki clout in municipal systems). Only Muslim households get this darkish
water.”
On July 3, 2017, I arrived in Trilokpuri at night and heard from Ashfaq and
others that a balva had been narrowly averted earlier that day. Sufiji described the
events: A married Balmiki man from Block 28 had begun living with a Muslim
woman. Enraged, the Balmiki man’s wife put the “jungli janvar’s” meat (Sufiji fol-
lows the Muslim etiquette of referring to pigs euphemistically, as the “wild animal”),
in front of the Muslim woman, saying: “If you are living with a Balmiki then eat
this!” An irate group of Muslims from Block 27 gathered as news of this incident
spread. Sufiji and other members of the Aman Ekta Committee (an interreligious
neighborhood committee, created through government/police initiative in the af-
termath of the 2014 Trilokpuri riots) convinced the enraged group to file a police
complaint rather than seeking direct retribution against the Balmiki family. The po-
lice responded by arresting the Balmiki man and wife, and the crowd quieted down,
“nahi tau balva ho jata” (“otherwise there would have been a riot”).
I spent that night in Sufiji’s one-room apartment in Block 27, which doubles as a
healing space. The next morning, I went to get a shave at the barbershop downstairs,
in the riot laboratory marketplace. The barber’s assistant, a young Muslim man, was
narrating the previous day’s events with added incendiary embellishments. This time,
the Balmiki woman threw pork in front of the Muslim woman during her morning
namaaz (prayers). The listeners, 10 in all, waiting for a haircut or a shave, seemed
unconvinced. Even as an outbreak of violence was averted, one could still feel the
circuits through which a charge is conducted.
In contrast, in Sundarnagri violence remained a more distant possibility, although
this might change as public expressions of anti-Muslim animosity become more per-
missible under the current Modi government. In June 2016, for the first time in the
neighborhood’s history since it was founded in 1975, Sundarnagri had a curfew. The
chain of events was as follows: A group of Muslim women in the mixed Hindu–
Muslim Block J approached a temple priest asking him to turn down the volume of
the temple loudspeaker and not to do the evening arti (prayers) at the same time as
the azan (the Muslim call for prayer), a classic Hindu–Muslim sonic conflict (Khan
2012). The priest responded sharply, and the women shouted back.
A few days later, the priest, Pandit Parvesh Sharma, helped produce an inflam-
matory YouTube “documentary” (see Figure 18) on “the threat to Hindus in Sun-
darnagri.” “I wanted to make this an international mamla (issue),” Sharma asserted,
when I interviewed him a few days after the curfew. “I called the Prime Minister’s
Office. And I called someone I know in Yogi Adityanath’s office (the rabidly right-
wing chief minister of the neighboring state of Uttar Pradesh). The next day Yogi-ji
sent 200 Hindus for the evening arti, shouting Har Har Mahadev.” The women
also called for help, and 50 or so Muslim men arrived from the Qureshi Block O,
although in his narrative the pandit pitched the number of Qureshis at 200, and
had them shouting “Allah ho Akbar.” The police, by all accounts, arrived almost
180 Medical Anthropology Quarterly

Figure 18. Media coverage of Pandit Parvesh Sharma in Sundarnagrir [This figure
appears in color in the online issue]

immediately, with the assistant commissioner of police there in person to announce

a curfew and install a police post at the temple for the next month.
As a long-time participant in local electoral politics, Kamleshji, the outreach
worker from the Sundarnagri clinic, knew most local notables and the neighborhood
police. One afternoon as the clinic closed for the day, I asked her if we could visit the
police post at the temple. We went, and Kamleshji introduced me to the two beat con-
stables, Hindu OBCs (Other Backward Castes/“middle” castes), in charge of protect-
ing the temple. As we chatted, the constables laughingly praised the priest Sharma
for his rousing speeches. We were joined by Mr. Arora, owner of a local milk supply
business, and a contender for BJP leadership in the neighborhood. I had interviewed
Mr. Arora a few weeks before. He had severely criticized his rivals, including the cur-
rent BJP councilor, a local shop owner, for their lack of acumen in forging alliances in
the neighborhood. Today Mr. Arora outdid himself in his ironic praise for “Panditji”
(Sharma), as the newfound rakshak (defender) of the Hindus of Sundarnagri. Sharma
responded to this ill-intentioned praise with nervous laughter, “Arey itna bhi mat
chadhaiye, Arora-sahib” (“don’t raise me so high, Arora-Sahib”). Sitting in this tem-
ple, I speculated: No Dalit had a place within these particular upper-caste neighbor-
hood machinations.
 Can a Neighborhood Fall Sick? 181

Kamleshji (herself, I realized, from the Sharma, Brahmin caste) joined this menac-
ing banter, making what felt like a feeble anti-Muslim joke: “They breed like chick-
ens, they bleat like goats.” Her listeners, including the policemen, guffawed. As we
walked back, I tested the waters, in terms of our political views. Gifted with sharp
social acumen, Kamleshji often subtly prompted our interlocutors in Sundarnagri,
including heroin dealers and police informants whom she was introducing me to,
into sharing information that they might be averse to offering an outsider. “Sir is
one of us,” she would say, referring to me. I thanked her for stoking the policemen,
to show me how partisan they are. She responded with surprise, “No sir, I wasn’t
joking! That is what I believe. I am very kattar (hardline).”
In subsequent conversations, the priest Sharma had some introspective moments:
“If it was a Congress government, like in the old days, they would have said:

Pandit mahaul kharab kar raha hai (he is ruining the atmosphere). Put him in
jail. I’ve been with the BJP since the 1980s, since the Jan Sangh time (the
electorally marginal precursor to the BJP). When we first started collecting
chanda (political/religious contributions) people weren’t willing to give even
50 paisa. We’ve done a lot of sangharsh (struggle). The police here are in
cahoots with the Qureshis (a Muslim caste) because of the car market (in
Sundarnagri, notorious for stolen car parts). The Qureshis give the police
huge monthly payments. That’s why they turn up so quickly to protect them.
I am working to get the car market shut down.

Drawing on these instances of near-violence, I contend that the contrast is not

between a riot-prone Trilokpuri and amicable cohabitation in Sundarnagri. Instead,
I hope to have drawn out the ways in which compositional differences and political
alignments matter to the actualization of potential violence, and further, the spe-
cific circuits, resistances, and forms of Hindu–Muslim hostility in Sundarnagri and
Trilokpuri, which result in their sharply differing cumulative pictures of collective
violence so far.

Concluding Discussion: Varieties of Suicido–Genetic Currents

I began by asking how we might better understand the global nature of the opioid
crisis. The form of comparative ecological analysis I offer here is not merely local.
I hope to have offered a possible answer to Gamela’s concern above, on the dan-
gers of the term epidemic, and his question of why certain, more tabooed forms
of addiction take root in certain areas and not in others. This form of ecological
analysis can be continued across other geographies and vectors of significance. For
instance, recent opioid research in the United States has found variations between
seemingly similar demographic areas to be significant enough to ask whether this
is an entirely physician-induced iatrogenic epidemic (Wright et al. 2014), and if,
rather than a single opioid epidemic in the United States, the phenomenon would
be better understood as a series of three intertwined but distinct waves (Kiang et al.
2019), animated by different types of opioids and diverse geographical, temporal,
and socio–demographic patterns of mortality.19
182 Medical Anthropology Quarterly

Further, I asked what kinds of diagnostics anthropologists might offer for ques-
tions of ecological variation beyond global invocations of neoliberalism or pre-
dictable effects that a similarity of context might have on outcomes related to health
and well-being. Trilokpuri and Sundarnagri could both be classified as urban poor
neighborhoods. I enumerated a number of similarities between these neighborhoods,
including access to state services, levels of reported crime, vignette-based investiga-
tions of social efficacy, and psychiatric comparators of stress, somatoform disorders,
and quality-of-life indicators. Within this seeming similarity of context, rather than
uniform effects, I foreground the sharpness of ecological variations, depending on
the form of death under consideration. As such, either neighborhood could be seen
as better or worse, depending on the optic, if either heroin addiction or collective
violence or a generalized idea of neighborhood disorder had been considered in iso-
lation. The currents of addiction and violence are not causally related, but might be
understood as “divergent series” (Deleuze 1993: 110) within a milieu, with each se-
ries composed of specific events, networks, trajectories, and forms of life and death.20
Rather than a standardized explanation of social or psychic disorder, I have tried
to offer a more fine-grained form of ecological diagnostics. This involved a focus on
compositional elements, including differences in patterns of livelihood, related to
caste, and the continuing salience of caste hierarchies and differences in contempo-
rary India. While these compositional elements are ecological in a Batesonian sense,
as “recurrent patterns” (Bateson 1972: xxvi), the recurrences are not static forms
of repetition without differences. Even within stable inter-generational movements
such as the Balmiki’s status as the pig-breeding sweeper caste, we might read their
relative economic stability and relatively recent turn to the Hindu right as a crucial
part of the political ecologies described above.
Within these forms of stability and dynamism, Durkheim’s idea of suicido–genetic
currents offers a diagnostic tool to understand the entanglement of forms of as-
piration with forms of death, as with the political rise of the Balmikis leading to
an intensified expression of currents of hostility in their relations with neighbor-
ing Muslim communities. The concept of suicido–genetic currents also allows us to
diagnose the conditions under which Imtiaz, Waseem, and Johar’s heroin-addicted
peers in Trilokpuri, economically and demographically similar to them in so many
ways, are not injectors. These specific differences of life and death, I contend, are
best understood as a mode of conduction and resistance to specific suicido–genetic
currents, rather than a uniform epidemic. This is not to say that at a national level
India is not in the midst of a deepening opioid crisis, and a crisis of democratic life
and cohabitation, in ways described above. Within these crises, an attention to, at
times sharply varying “ecologies of experience”21 (Simpson 2013), might allow for
a deeper understanding of contemporary forms of psychic distress and the human
“predisposition to self-destruction,” as Durkheim presciently called it.

Notes
Acknowledgments. I wish to dedicate this article to the AIIMS OST team, led by
Dr. Ravindra Rao, Sister Bina, and their other colleagues and co-workers for the dif-
ficult work they do in the community clinics and vans in East Delhi. The research for
 Can a Neighborhood Fall Sick? 183

this paper was done prior to the Covid-19 pandemic. While the neighborhoods of
Trilokpuri and Sundernagri were deeply affected by the pandemic, I have chosen to
leave our argument as is, based on that earlier phase of research, rather than hastily
rewriting the experiences of my interlocutors entirely in a mode of pandemic-infused
crisis. In subsequent writings, I hope to engage more deeply with the experience of
the Covid-19 pandemic in these neighborhoods. The funding for this research was
provided by research grants from the Wenner-Gren Foundation and the American
Institute for Indian Studies. A writing fellowship at the Wissenschaftskolleg zu Berlin
(Institute for Advanced Study, Berlin) provided a lively and hospitable intellectual
home for the writing of this article. The article would not have been possible without
the sustained efforts of Ms. Anita Chopra in helping to build and analyze our project
data, and her sincerity as she continued to work with me after retiring as a research
scientist at AIIMS. My sincere thanks also to Naveeda Khan, who read a first draft
of this article and offered key suggestions for revisions. Professor Vincanne Adams
and the MAQ anonymous reviewers offered invaluable suggestions during the re-
view process, which significantly helped sharpen the article. This article was written
with the guidance and companionship of colleagues and interlocutors at the AIIMS
Psychiatry department in Delhi, including Dr. Ravindra Rao, Dr. Pratap Sharan, Dr.
Anju Dhawan, Dr. Rakesh Chadda, and Dr. Mamta Sood.

1. The British Medical Journal has begun an online discussion forum on

the global opioid crisis. https://blogs.bmj.com/case-reports/2019/06/25/the-global-
opioid-crisis/ (accessed November 24, 2019).
2. India’s Narcotic Drugs and Psychotropic Substances (NDPS) Act was
amended in 2014 to create a new class of “essential narcotics,” including fentanyl,
codeine, and hydrocodone, many of the stars of the American opioid epidemic,
gradually becoming more widely available in India. https://www.theguardian.com/
world/2019/aug/27/india-opioids-crisis-us-pain-narcotics (accessed November 24,
2019).
3. The epistemological problem here in defining the unit of analysis would be
what Appadurai had called the “production of locality” (1996). Scholars within the
tradition of neighborhood studies in sociology have long asked, “What is a neigh-
borhood?” (Sampson 2013: 136), if this is not to be taken simply as a census tract.
In this article, I take the neighborhoods of Trilokpuri and Sundarnagri to be viable
units of analysis based on the way in which my ethnographic interlocutors would
self-identify and locate themselves as residents of these neighborhoods in narrating
their lives and form of urban experience.
4. “Resettlement colony” is a government term for settlements where slum
dwellers were displaced from more gentrified parts of Delhi (Bhan 2016; Das and
Walton 2015; Tarlo 2003).
5. This article is based on five years of ethnographic fieldwork that I began in
2015–16, when I spent the year as a visiting faculty member at the Department of
Psychiatry, AIIMS (Delhi), with regular follow-up visits since.
6. One piece can produce 22 “lines.” A panni typically requires at least two lines.
7. In collaboration with Dr. Ravindra Rao (AIIMS, Psychiatry) I carried
out three phases of socio–demographic and psychiatric instrument-based surveys
184 Medical Anthropology Quarterly

between January and June 2016, in 100 randomly selected households respectively
in Trilokpuri and Sundarnagri.
8. The most common social science method for comparing variations in collec-
tive violence is through news archives. The largest such database, the Varshney-
Wilkinson dataset for Hindu–Muslim violence is, based on Times of India archives
from 1950 to 1995.
9. A debate I want to flag here, but maintain my distance from, is the rediscovery
of Gabriel Tarde in recent social theory (Candea 2010; Latour and Lepinay 2009),
as a supposedly more radical alternative to Durkheim. The reason I avoid this debate
is because it tends to present a caricatured picture of Durkheim as simply an institu-
tional powerbroker and pseudo-scientist of “social facts.” Tarde’s well-wishers are
perhaps still chagrined by Durkheim’s own caricature of Tarde’s work in Suicide. In
a very different rendering of this article, I could have taken Tarde’s ideas of “waves”
and “imitation” (Thomassen 2012) as the conceptual anchors for this analysis. I
will save my contribution to the Durkheim–Tarde debate for another occasion. For
now, I contend that the difference is much finer if we receive Durkheim not only as
a thinker of structures but also of currents. For reinterpretations of Durkheim as
a thinker of currents, see Olaveson (2001), Singh (2012), Mazzarella (2017), and
Coleman (2019).
10. According to municipal records, there are 55 resettlement colonies in Delhi,
with an average population of 150,000.
11. On the emergence of the term “Balmiki,” see Jaoul (2011).
12. On waste economies in urban India, see Gidwani and Maringanti (2016).
13. As social scientists (Gayer and Jafferlot 2012; Jamil 2017; Mehta 1997) and
the Government of India’s 2005 Sachar Committee Report have argued, a significant
proportion of the Muslim poor may be seen as Dalit by occupation, socioeconomic
status, and marriage-related practices. But since this social fact is denied theologi-
cally and governmentally, affirmative action programs are largely absent for Mus-
lims and available for Hindus, in what is perhaps one of the most systemic injustices
in postcolonial India, particularly sharply visible in resettlement colonies.
14. In India, the transformation of opium into heroin or “smack” is dated to
the mid-1980s (Mohan et al. 1985), when it first became available via Afghanistan,
according to the narratives of older addicts, and then began to be produced locally, in
Uttar Pradesh, Rajasthan, and Madhya Pradesh, the three states in India historically
where poppy can be legally grown for medicinal purposes. Part of the legal poppy
harvest is siphoned off for the production of heroin, which can be synthesized in one-
room cottage industry type production units. According to most national surveys by
addiction psychiatrists in India, the rise of injecting drug use began in the late 1990s
(Dorabjee and Samson 2000).
15. Patient names have been changed to protect their identity.
16. “Hotspot” (a meeting spot for addicts) is a term from AIDS–NGO discourse
that has made its way into addict slang in Delhi.
17. Grounded in civil society and anti-corruption initiatives, the Aam Aadmi
Party was founded in Delhi in 2012, see: https://aamaadmiparty.org/about/our-
history/ (accessed November 24, 2019).
 Can a Neighborhood Fall Sick? 185

18. The BSP is often described as a Dalit political party. However, for many Dal-
its, especially the Balmikis, the BSP is strongly associated with the Jatavs/Chamars
as the caste of its principal founders Kanshi Ram and Mayawati.
19. This research includes tracing neighborhood level variation (Visconti et al.
2015) and rural–urban and regional differences across neighboring states in North
American patterns of opioid abuse (Ruhm 2017; Unick and Ciccarone 2017).
20. In relation to medical sociology and public health, I find it helpful to dif-
ferentiate my argument here from the “social contagion” analyses associated with
Nicholas Christakis (Christakis and Fowler 2008; Fowler and Christakis 2008,
2010), that aim to show how “physical proximity” can enable particular psychic
states to spread (Fowler and Christakis 2008: 2338). The main difference would
rest in the kind of generalizability that network analysts like Fowler and Christakis
claim, for instance, in arguing that happiness can reliably be traced across “three
degrees of influence” (p. 2338). The disciplinary difference, as I see it, is not that
anthropology is content with a smaller scale or a lesser degree of generalizability,
but rather, what counts as a researchable question and possible insights within a
set of methods For instance, in Fowler and Christakis’s analysis of the spread of
happiness across three generations (published in the British Medical Journal) as he
qualifies: “Our data do not allow us to identify the actual causal mechanisms for the
spread of happiness” (p. 2338). In contrast, in this argument, I have attempted to
follow the causal chain of events that lead to specific patterns. As such, the anthro-
pological engagement with singularities and patterns might also yield rich insights at
different “scales” (without assuming that a larger sample is necessarily more com-
plex) in understanding ecological variation across demographically similar areas.
I thank the anonymous reviewer of MAQ for inviting me to make this important
clarification.
21. Recent scholarship in urban geography has also sought to explore more dis-
aggregated ecologies of psychic life beyond homogenizing experiential categories
like alienation and stress (Fitzgerald et al. 2016; Richaud and Amin 2019).

Figure Source
a. Ambekar, A., A. Agrawal, R. Rao, A. K. Mishra, S. K. Khandelwal, and R. K.
Chadda. 2019. Magnitude of Substance Use in India. New Delhi: Ministry of Social
Justice and Empowerment, Government of India (accessed February 26, 2019).
b. India:http://socialjustice.nic.in/writereaddata/UploadFile/Magnitude_Substan
ce_Use_India_REPORT.pdf (accessed February 26, 2019). USA: https://www.
samhsa.gov/data/sites/default/files/cbhsq- 2019).reports/NSDUHNationalFindings
Report2018/NSDUHNationalFindingsReport2018.pdf (accessed February 14,
2019).
c. Trilokpuri- https://www.google.com/maps/place/Trilokpuri,+New+Delhi,+
Delhi+110091/@28.6021062,77.2889023,14z/data=!3m1!4b1!4m5!3m4!
1s0x390ce4c222d9da97:0x3ea1ca718677356d!8m2!3d28.601243!4d77.
3031561 (accessed February 18, 2019).
186 Medical Anthropology Quarterly

Sundarnagri- https://www.google.com/maps/place/Sunder+Nagri,+Dilshad+
Garden,+Delhi/@28.6952768,77.3171691,16z/data=!3m1!4b1!4m5!3m4!
1s0x390cfbb9f9541be5:0xf0c2331c8b4642c2!8m2!3d28.6949907!4d77.
3223915 (accessed February 18, 2019).
d. Information compiled from https://www.electionsinindia.com/delhi/
trilokpuri-assembly-vidhan-sabha-constituency-elections and https://www.
electionsinindia.com/delhi/seema-puri-assembly-vidhan-sabha-constituency-
elections (accessed February 24, 2019).
e. Comparison chart for 2018 community clinic figures, prepared by Ms. Anita
Chopra, Scientist (Retd.) and Ms. Bina Rawat, Nurse, NDDTC, All India Institute
of Medical Sciences, New Delhi.
f. *http://naco.gov.in/sites/default/files/Fact%20Sheets_1.pdf (accessed Novem-
ber 11, 2019).
**http://www.naco.gov.in/sites/default/files/HIV%20DATA.pdf (accessed
November 11, 2019).
g. https://infoweb-newsbank-com. Access World News Research Collection for
Times of India Mumbai edition India [accessed by Ms. Anita Chopra via Brown
University Library Collections] (accessed September 1, 2019).
h. Information collected as part of study “A Mixed Method (Anthropological
and Psychiatric) Study to Examine and Compare Neighborhood Characteristics in
Determining High Risk Behavior and Treatment Outcomes in Relation to Opioid
Use in Two Resettlement Colonies in East Delhi.” Ethical approval reference no.
IEC/-151/04-03.2016; RP – 7/2016.
i. Delhi police records accessed as part of study “A Mixed Method (Anthropolog-
ical and Psychiatric) Study to Examine and Compare Neighborhood Characteristics
in Determining High Risk Behavior and Treatment Outcomes in Relation to Opi-
oid Use in Two Resettlement Colonies in East Delhi.” Ethical approval reference no.
IEC/-151/04-03.2016; RP – 7/2016.
j. Information collected from “Quotidian Networks” section in household survey
questionnaire as part of study “A Mixed Method (Anthropological and Psychiatric)
Study to Examine and Compare Neighborhood Characteristics in Determining High
Risk Behavior and Treatment Outcomes in Relation to Opioid Use in Two Resettle-
ment Colonies in East Delhi.” Ethical approval reference no. IEC/-151/04-03.2016;
RP – 7/2016. The Quotidian Networks Survey was developed by members of the
Institute for Socio-Economic Research in Development and Democracy (ISERDD,
Delhi).
k. Information collected from household survey using instruments -WHOQOL-
brief and Perceived stress scale as part of study “A Mixed Method (Anthropological
and Psychiatric) Study to Examine and Compare Neighborhood Characteristics in
Determining High Risk Behavior and Treatment Outcomes in Relation to Opioid
Use in Two Resettlement Colonies in East Delhi.” Ethical approval reference no.
IEC/-151/04-03.2016; RP – 7/2016.
l. Information collected as part of study “A Mixed Method (Anthropological
and Psychiatric) Study to Examine and Compare Neighborhood Characteristics in
Determining High Risk Behavior and Treatment Outcomes in Relation to Opioid
Use in Two Resettlement Colonies in East Delhi.” Ethical approval reference no.
IEC/-151/04-03.2016; RP – 7/2016.
 Can a Neighborhood Fall Sick? 187

m. Data from Household expenditure and Livelihoods Survey collected as part of

study “A Mixed Method (Anthropological and Psychiatric) Study to Examine and
Compare Neighborhood Characteristics in Determining High Risk Behavior and
Treatment Outcomes in Relation to Opioid Use in Two Resettlement Colonies in
East Delhi.” Ethical approval reference no. IEC/-151/04-03.2016; RP – 7/2016.
n. Based on Comparison Data of Patient Medical Record Files as Part of Study “A
Mixed Method (Anthropological and Psychiatric) Study to Examine and Compare
Neighborhood Characteristics in Determining High Risk Behavior and Treatment
Outcomes in Relation to Opioid Use in Two Resettlement Colonies in East Delhi.”
Ethical approval reference no. IEC/-151/04-03.2016; RP – 7/2016.
o. https://www.firstpost.com/india/trilokpuri-communal-riot-violence-leaves-
east-delhi-tense-35-hurt-section-144-imposed-1772819.html (accessed June 3,
2017).
Trilokpuri: The Riot laboratory. 2014. https://www.ndtv.com/video/shows/india-
matters/trilokpuri-the-riot-laboratory-343359. Published On: October 31, 2014
(accessed April 3, 2017).
p. Kamal Beniwal | Trilokpuri Nigam Parshad | ALL RIGHTS MAGAZINE. Pub-
lished February 25, 2016. https://youtu.be/ySi-g5L9oM (accessed April 3, 2017).
q. Ex-MLA accused of inciting riots dies of heart attack. https://www.
pressreader.com/india/hindustan-times-gurugram/20141218/page/2/textview (ac-
cessed April 3, 2017).
r. https://www.facebook.com/watch/?v=329070591102786 (accessed November
15, 2020).

References Cited
Ambekar, A., A. Agrawal A., R. Rao, A. K. Mishra, S. K. Khandelwal, and R. K. Chadda.
2019. Magnitude of Substance Use in India. New Delhi: Ministry of Social Justice
and Empowerment, Government of India.
Appadurai, A. 1996. Modernity at Large. Minneapolis: University of Minnesota Press.
Bateson, G. 1972. Steps towards an Ecology of Mind. New York: Ballantine Books.
Bhan, G. 2016. In the Public’s Interest: Evictions, Citizenship, and Inequality in Contem-
porary Delhi. Athens: University of Georgia Press,
Bourgois, P., and J. Schonberg. 2009. Righteous Dopefiend. Berkeley: University of Califor-
nia Press.
Candea, M. 2010. The Social after Gabriel Tarde: Debates and Assessments. New York:
Routledge.
Cavell, S. 2007. Foreword. In Life and Words: Violence and the Descent into the Ordinary,
edited by V. Das, ix–xiv.Berkeley: University of California Press.
Christakis, N. A., and J. H. Fowler. 2008. The Collective Dynamics of Smoking in a Large
Social Network. New England Journal of Medicine 358: 2249–58.
Coleman, L. 2019. Widened Reason and Deepened Optimism: Electricity and Morality in
Durkheim’s Anthropology and Our Own. In Electrifying Anthropology: Exploring
Electrical Practices and Infrastructures, edited by S. Abram, B. Ross Winthereik, and
T. Yarrow, 43–63.New York: Bloomsbury.
Das, V. 2007. Life and Words: Violence and Descent into the Ordinary. Berkeley: University
of California Press.
Das, V. 2015. Affliction: Health, Disease, Poverty. New York: Fordham University Press.
188 Medical Anthropology Quarterly

Das, V., and M. Walton. 2015. Political Leadership and the Urban Poor. Current Anthro-
pology 56: S44–S54.
Deleuze, G. 1993. The Logic of Sense. Columbia University Press: New York.
Deleuze, G. 1997. Essays Critical and Clinical. Minneapolis: University of Minnesota Press.
Dorabjee, J., and L. Samson. 2000. A Multi-centre Rapid Assessment of Injecting Drug Use
in India. International Journal of Drug Policy 11: 99–112.
Duncan, D. T., and I. Kawachi, eds. 2018. Neighborhoods and Health. Oxford: Oxford
University Press.
Durkheim, E. 1997 [1897]. Translated by J. A. Spaulding and G. Simpson. Suicide: A Study
in Sociology. New York: Free Press.
Durkheim, E. 2001. The Elementary Forms of Religious Life, translated by C. Cosman.
New York: Oxford University Press.
Fitzgerald, D., N. Rose, and I. Singh. 2016. Living Well in the Neuropolis. The Sociological
Review Monographs 64: 221–37.
Fowler, J. H., and N. A. Christakis. 2008. Dynamic Spread of Happiness in a Large So-
cial Network: Longitudinal Analysis over 20 Years in the Framingham Heart Study.
British Medical Journal 337: 2338.
Fowler, J. H., and N. A. Christakis. 2010. Cooperative Behavior Cascades in Human Social
Networks. Proceedings of the National Academy of Sciences 107: 5334–38.
Gamella, J. F. 1994. The Spread of Intravenous Drug Use and AIDS in a Neighborhood in
Spain. Medical Anthropology Quarterly 8: 131–60.
Gayer, L., and C. Jaffrelot, eds. 2012. Muslims in Indian Cities: Trajectories of Marginal-
ization. London: Hurst and Company
Ghazala, J. 2017. Accumulation by Segregation: Muslim Localities in Delhi. Delhi: Oxford
University Press.
Gidwani, V., and A. Maringanti. 2016. The Waste-value Dialectic. Comparative Studies of
South Asia, Africa and the Middle East 36: 112–33.
Harris, S. 2015. To Be Free and Normal: Addiction, Governance, and the Therapeutics of
Buprenorphine. Medical Anthropology Quarterly 29: 512–30.
Jamil, G. 2017. Accumulation by Segregation: Muslim Localities in Delhi. New Delhi: Ox-
ford University Press India.
Jaoul, N. 2011. Casting the “Sweepers”: Local Politics of Sanskritisation, Caste and Labour.
In Cultural Entrenchment of Hindutva: Local Mediations and Forms of Convergence,
edited by D. Berti, N. Jaoul, and P. Kanungo, 273–306.New Delhi: Routledge.
Khan, N. 2012. Muslim Becoming: Aspiration and Skepticism in Pakistan. Durham: Duke
University Press.
Kiang, M. V., S. Basu, J. Chen, and M. J. Alexander. 2019. Assessment of Changes in the
Geographical Distribution of Opioid-related Mortality across the United States by
Opioid Type, 1999–2016. JAMA Network Open 2: e190040.
Latour, B., and V. A. Lepinay. 2009. The Science of Passionate Interest: An Introduction to
Gabriele Tarde’s Economic Anthropology. Chicago: Chicago University Press.
Lee, J. 2015. Recognition and Its Shadows: Dalits and the Politics of Religion in India. PhD
Dissertation, Columbia University, New York.
Mazzarella, W. 2017. The Mana of Mass Society. Chicago: University of Chicago Press.
Mehta, D. 1997. Work, Ritual, Biography: A Muslim Community in North India. Delhi:
Oxford University Press.
Mohan, D., S. S. Adityanjee, and S. Lal. 1985. Changing Trends in Heroin Abuse in India:
An Assessment Based on Treatment Records. Bulletin on Narcotics 37: 19–24.
Olaveson, T. 2001. Collective Effervescence and Communitas. Dialectical Anthropology
26: 89–124.
 Can a Neighborhood Fall Sick? 189

Richaud, L., and A. Amin. 2019. Mental Health, Subjectivity and the City: An Ethnography
of Migrant Stress in Shanghai. International Health 11: s7–s13
Ruhm, C. J. 2017. Geographic Variation in Opioid and Heroin Involved Drug Poisoning
Mortality Rates. American Journal of Preventive Medicine 53: 745–53.
Sampson, R. J. 2003. Neighborhood-level Context and Health: Lessons from Sociology.
In Neighbourhoods and Health, edited by I. Kawachi and L. F. Berkman, 132–
46.Oxford: Oxford University Press.
Sampson, R. J. 2013. The Great American City: Chicago and the Enduring Neighborhood
Effect. Chicago: University of Chicago Press.
Sharma, S. G. 2018. Delhi’s Trilokpuri Is Simmering again in Festive Season, This
Time over Inter-faith Love. https://swarajyamag.com/politics/delhis-trilokpuri-is-
simmering-again-in-festive-season-this-time-over-inter-faith-love (accessed June 3,
2018).
Simpson, P. 2013. Ecologies of Experience: Materiality, Sociality, and the Embodied Expe-
rience of (Street) Performing. Environment and Planning A: Economy and Space 45:
180–96.
Singh, B. 2011. Agonistic Intimacy and Moral Aspiration in Popular Hinduism: A Study in
the Political Theology of the Neighbor. American Ethnologist 38: 430–50.
Singh, B. 2012. The Headless Horseman of Central India: Sovereignty at Varying Thresholds
of Life. Cultural Anthropology 27: 383–407.
Singh, B. 2015. Poverty and the Quest for Life: Spiritual and Material Striving in Rural
India. Chicago: University of Chicago Press.
Solomon, S. 2009. Mortality among Injection Drug Users in Chennai, India (2005–2008).
AIDS 23: 997–1004.
Strang, J., and King, L. 1997. Heroin Is More than just Siamorphine. Addiction Research
5: R3-R7.
Tarlo, E. 2003. Unsettling Memories: Narratives of the Emergency in Delhi. Berkeley: Uni-
versity of California Press.
Thomassen, B. 2012. Émile Durkheim between Gabriel Tarde and Arnold van Gennep:
Founding Moments of Sociology and Anthropology. Social Anthropology 20: 231–
49.
Trilokpuri: The Riot Laboratory. 2014. https://www.ndtv.com/video/shows/india-matters/
trilokpuri-the-riot-laboratory-343359 (accessed April 3, 2017).
Truong, K. D., and S. A. Ma. 2006. A Systematic Review of Relations between Neighbor-
hoods and Mental Health. The Journal of Mental Health Policy and Economics 9:
137–54.
Unick, G. J., and D. Ciccarone. 2017. US Regional and Demographic Differences in Prescrip-
tion Opioid and Heroin-related Overdose Hospitalizations. International Journal on
Drug Policy 46: 112–19.
Visconti, A. J., G. M. Santos, N. P. Lemos, C. Burke, and P. O. Coffin. 2015. Opioid Overdose
Deaths in San Francisco: Prevalence, Distribution, and Disparities. Journal of Urban
Health: Bulletin of the New York Academy of Medicine 92: 758–72.
Wright, E. R., H. E. Kooreman, M. S. Greene, R. A. Chambers, A. Banerjee, and J. Wilson.
2014. The Iatrogenic Epidemic of Prescription Drug Abuse: County-level Determi-
nants of Opioid Availability and Abuse. Drug and Alcohol Dependence 138: 209–15.