Skip to navigation – Site map

HomeThematic Issues18Site, Sight, Cite: Conceptualizin...

Site, Sight, Cite: Conceptualizing Wayside Shrines as Visual Culture

William Elison


This article proposes a threefold framework for analyzing wayside shrines as artifacts of visual culture. Site casts shrine construction as a problem of the organization of space: a sacred space (the shrine) converges with a public space (the street). Sight defines the shrine as a visual phenomenon: a display framing an image that projects an ideological address. Cite directs attention to the semiotic operations of the image, focusing on effects of attraction and repulsion. The argument is organized in two parts. The first part is descriptive: the author presents a range of exemplary street shrines, from rudimentary (kaccha) to solid and well finished (pakka), encountered in the course of ethnographic fieldwork in Mumbai. The second part takes a systematic approach to analyzing the visual relations street shrines engage in contemporary urban public space. The analysis hinges on a proposed re-theorization of visual worship, darshan, as a gesture of mutual recognition.

Top of page

Full text

1Truth, Divinity, Beauty: Satyam Shivam Sundaram is the official motto of Indian government television, Doordarshan. It is also the title of a 1977 film by one of the great creative spirits of Indian cinema, Raj Kapoor. Nowadays, the film is chiefly remembered for its emphasis on female sexuality, which titillated some viewers and scandalized others. But Satyam Shivam Sundaram is also notable as a highly self-conscious statement on the part of the director, whose lens (voyeuristically at points, to be sure) exposes and questions multiple codes of Indian visual culture. What is attractive? What is modest, and what indecent? What is natural, and what is modern? And most germane for this essay: What does God look like, and what does He see?

2The film opens with the consecration of a wayside shrine. The camera zooms in on a rock projecting from a field at the side of a road. Two hands appear from the edge of the frame with a water vessel. As the rock is anointed, another hand emerges to paint it with the three stripes symbolic of Shiva. The camera pulls back to show some rustic-looking men and women assembling around the newly sacred object; a devotional chant in Sanskrit fills the soundtrack. Some villagers add to the adornment, while others stand by reverently. Once the visual conversion of the rock is complete, a gleam of light off its surface transforms into animated letters spelling out the film’s title. The credits begin.

3The off-screen voice that comments on the scene is that of Kapoor himself.

Now, if someone were to ask you what that is, what would you say? It’s a god. But in actuality, what your eye is seeing—what would that be? It’s a rock. So what is it that turned it into a god? Your faith did. Your belief did. Your love did. That which is true, is God, is beauty.

Kapoor’s formula seems anodyne, but it launches the film’s investigation—a searching and layered one—of visual relations and the cultural modes that structure them, including darshan, or visual worship, and cinema itself. For the purposes of this essay about street shrines, however, let’s lose no time in interrogating this scenario.

4What effect might the rock’s recognition as an embodiment of divine presence have on its environs? Are we to understand this patch of turf on the roadside as a sacred space? Would any diegetic passersby on the road be assumed to be from the same cult community? And if not, would they all be assumed to be Hindu? (Indeed, are the film’s viewers all assumed to be Hindu—and if not, what might Kapoor’s characterization of his audience have been?) What if the road were to run, not through the rural, “traditional” locus of the diegesis, but through a diverse, modern city like Mumbai?

5Furthermore, how does the formal modification of the way the rock looks affect its recognition as sacred? The intervention shown in the film does not go beyond some rudimentary marking and decoration of the object’s surface. But what if (as is the typical pattern in real life) a structure were erected around and over the god to shelter it? Or what if the embellishment involved products of industrial technology, like tinsel or printed images? Would development of the site along lines like these enhance the effect of sacrality, or dilute it? And let’s also consider that perhaps the local peasants’ intervention wasn’t wholly artless or naïve to begin with. What if their marking of the rock had failed to cite so clearly the iconic linga of Shiva, as found in temples run by Brahmins across India?

6Finally, if what is at stake here, as Kapoor states, is faith (shraddha)—a matter of an autonomous individual’s conscience—are we to understand that it’s up to each person passing by on the road to make up their own mind whether the rock is a god or not?

7In this essay I will follow up on Kapoor’s central concern. If godliness is adjudicated in the eye of the beholder, how can that interior vision be made legible to others? To be more specific: If a sacred site on the street is sacred by virtue of a visually grounded epistemology, how do we study it as a phenomenon of visual culture? I will build my argument with reference to a set of urban wayside shrines I have encountered in the course of ethnographic research in Mumbai and advance to some analytic propositions.

  • 1 I am indebted to Wendy Doniger for conceptualizing the framework in this way in an informal discuss (...)

8Let’s get started by setting aside the concepts of truth, divinity, and beauty, at least as an organizing rubric for this inquiry. In place of Satyam Shivam Sundaram, I propose a different alliterative trio: site, sight, and cite.1 Site, as in space: What effect does the sacred display have on the space it occupies? Sight, as in vision: What effect does it have on a viewing subject? Cite, as in signification: What sort of semiotic operation enables it to produce these effects?

Kaccha to pakka

9This paper adapts an argument that is presented more fully in a monograph currently in press, The Neighborhood of Gods: The Sacred and the Visible at the Margins of Mumbai (Elison 2018). The book is a study of visual relations that produce different kinds of urban spaces. It is centered on ethnographic research I pursued at various Mumbai localities on practices to do specifically with the consecration of space, as enacted in the main by members of subaltern communities. Inquiries into the city’s wayside shrines were one component of this research. Readers interested in the ethnographic details supporting the claims I make here are invited to consider my longer project.

10The handful of sites I describe here have been selected for their display of typical characteristics. If plotted on a map, most of my examples will be seen to be located in relatively prosperous neighborhoods across the southern part of Mumbai. In the built environment of this downtown area, with its heritage in colonial city planning, the effect of the shrines’ rupture of an ostensibly rational, modern order is heightened, to illustrative effect. Yet, as the other articles in this special issue will indicate, these Mumbai sites are representative of entirely commonplace features of the South Asian landscape. I have arranged my examples in an order that illustrates a typology of progressively more elaborate forms. In other words—Hindi words—they represent a movement from what Indians call kaccha—unripe, crude, or sketchy—to pakka—ripe, permanent, and legible. As I will go on to discuss, pakka also indicates, in this context, a process of development reflecting the investment of labor and money.

11The first site is located not far from the Dadar train station, one of the busiest junctions on Mumbai’s commuter rail network. It is a large house belonging to a Parsi family. Mounted on the wall surrounding the property, at heights from just above the pavement to approximately eye level, is a series of tiles bearing religious imagery. Every tile retailer in the city can offer designs like these, printed to represent the major religious communities residing in Mumbai and, in the case of Hindu gods, local deities metonymic of specific parts of India. Eclecticism is typical of such displays. Here are tiles of the Hindu deities Ganesh, Durga, Krishna, and Hanuman; a mosque; the Zoroastrian sacred fire; and Leonardo’s The Last Supper and the Virgin Mary. Some of the images show signs of having engaged cultic attention. Among the tiles, a three-dimensional figurine of the Maharashtrian holy man Sai Baba has been installed in a small shrine set into the wall, which also accommodates incense and other ritual paraphernalia. And a tile of Jesus mounted close to the ground has been adorned with a string of flowers. As a display of two-dimensional graphics that elides the conventional bourgeois partition between public and private space, this wall—and others like it—shares formal characteristics with a host of other urban sites. Collectively these can be categorized as the most rudimentary or kaccha form of street shrine.

Figure 1

Figure 1
  • 2 Blueprints by Malika Khurana. Photographs by the author.

Tiles with images of Durga and Krishna on a private wall fronting the street. Dadar, Mumbai2

12There are plentiful down-market analogs. We might, for example, head from Dadar to the Grant Road railway station, where a group of labor migrants have erected makeshift homes parallel to the tracks (the railway authority tends to be more lenient in the matter of evictions than the municipality). Spilling onto the fence that cordons off the rails, devotional prints of Hindu gods have been posted facing the street. What for an even slightly more prosperous family would be contained on an internal wall or shelf—the low-rent version of a prayer room—has become for these squatters an exhibit that exerts its effects in public. Elsewhere in the neighborhood, the same dynamic can be seen at work, with the Christian settlers of a pavement-based dwelling having installed figurines of Jesus and Mary on one of the metal switch boxes the power company maintains on the streets (their stiff postures next to each other recall the iconography of the Hindu deities Ram and Sita, or the Maharashtrian Vitthal and Rakhumai). And not far away is Chowpatty Beach, where the saintly Sai Baba makes his second appearance in the form of a miniature, mass-produced statuette. This time, the effigy inhabits a window-like little shrine set within a tree trunk facing a paan vendor’s stall. As with the fencing at the station and the electricity infrastructure, the tree thus consecrated is, legally speaking, public property. But if there is an incongruity here, it is surely lost on the shopkeeper and his betel-chewing clientele.

Figure 2

Figure 2

Small shrine dedicated to Sai Baba of Shirdi embedded in tree trunk. Chowpatty, Mumbai

Figure 3

Figure 3

Bas-relief of Ganesh on Charni Road pedestrian walkway. Girgaum, Mumbai

13Let’s continue one more stop down the Western Railway, to Charni Road, where the station is connected to the Girgaum business district by a pedestrian overpass. About a decade ago, the crowds of commuters would have been greeted daily by a great, metallic murti of Lord Ganesh framed in the window of a building overlooking the crossing. It was an eye-catching display, with an artful bit of staging behind it. The brass-colored Ganesh was as tall as a man (and of course considerably broader), and viewed through the windowpane it had a solid, three-dimensional appearance. A visit to the building that housed the image, however, showed it was a bas-relief, both shallow and hollow, made of vac-form plastic.

14Reactions among the crowd must have been mixed. Alongside those who registered a momentary transport, there would have been others who would have read it as a generic cipher of beneficence and moved on. And many other people, it is safe to say, must have seen in the rotund figure no blessing or welcome but rather a highly politicized symbol, for since the nineteenth century Ganesh has been the emblem of a specifically Maharashtrian Hinduism with a history of militancy. My tour of the premises made it plain that the image was intended strictly for public consumption; any would-be worshipper who entered the room would have been hard put to fit between Ganesh’s surface and the glass. Far from having functioned as a temple or prayer room, the unit seems to have served a less elevated purpose. The caretaker told me that the tenants who installed the god had been members of a “social club.” The implication was that the space had been devoted to drinking and possibly gambling, in which case the image’s function as a sort of mask of public piety would have come into play.

15Girgaum and the adjacent neighborhood of Khetwadi can furnish many other examples of wall-mounted displays that dominate the facades of their host structures. In some cases the site’s physical configuration manifests a process of growth through the development of pakka elements. Thus on one Khetwadi block, a tree growing on the public sidewalk has been annexed to the wall behind it by local Hindus who have enclosed it in a sort of booth. A window in the tiled surface exposes part of the tree trunk, framing it as the base component of a Hanuman shrine. The tiled enclosure has succeeded an earlier iteration of the shrine, one demolished by municipal workers as part of a citywide campaign against “illegal religious structures.” A fragment of the older installation remains in the form of the bottom part of a statuette of Lord Krishna, still fixed to the concrete slab at the foot of the tree. It has taken the shrine’s patrons less than a month to resurrect their site.

Figure 4

Figure 4

Hanuman shrine. Khetwadi, Mumbai

16My old address in Mumbai was on Nepeansea Road, an area not to be identified with any particular religious or caste community. Its chief association in the minds of Mumbai residents is probably with old money. This tony reputation notwithstanding, there was actually a fair amount of social diversity in the neighborhood, in part because the tenancy status of many residents was not strictly legal, and in part because of the many people engaged in service work, both as domestic staff and as workers in local businesses (many of which also had informal tenancy arrangements). Standing on the sidewalk about a block down from my door there was a shrine to Sai Baba. It had been erected by a man who lives next to it behind a small, illegally constructed shop.

Figure 5

Figure 5

Sai Baba shrine. Nepeansea Road, Mumbai

17This installation is, of course, another “illegal religious structure,” which is the principal way such sites are designated in the English-language press and in official documents. In colloquial speech, on the other hand, when people in Mumbai refer to street shrines the term they generally use is a Hindi or Marathi one: mandir. The word is conventionally translated as “temple.” On initial consideration, the image invoked of a full-blown Hindu temple—complete with an enclosure, an inner sanctum, and a priestly staff—seems like a sharp contrast with the reality of my modest example, which is not much bigger than a mailbox. Yet I would argue that the temple scenario is very much part of the picture. It can be thought of as an aspirational model, the destination of the journey a thriving shrine makes on its trajectory from kaccha to pakka—given the right convergence of variables. I enumerate the following conditions: 1) popularity, 2) capital investment, 3) sufficient physical room to expand, and 4) a blind or indulgent eye on the part of officialdom. The last feature also suggests that the interests of local power brokers are at play. As I have implied above, the question of the authorities’ forbearance can prove a crucial one, for the state has its own standards for adjudicating what is and what is not a real temple.

  • 3 The pakka intervention to the Hanuman shrine in Khetwadi can be seen as a tactical response to this (...)

18A wayside shrine like the last two examples, from Khetwadi and Nepeansea Road, violates the law on two counts. As an edifice, its construction has not been authorized. Nor, as a social entity, is it likely to have been registered as the occupant of an address under municipal jurisdiction—whether as a charitable trust, as would be the case with a legal temple, or alternatively as a private business or domicile.3 In this connection it is vital, however, to keep in mind that the homes of over half of Mumbai’s human population of approximately 12 million—the city’s slum dwellers—are, likewise, inadequately documented.

Figure 6

Figure 6

Schematic of typical shrine

19This humble Sai Baba shrine belongs to a category in which the majority of religious displays that occupy public space in Mumbai can be placed. This type of installation conforms to certain structural and typological standards. First among the physical components is a foundation substantial enough to stake a claim to its site with credibility, or at any rate unbudgeability: here, it is plastered bricks over rebar. (Built into the bricks is a cash box, complete with a slot to accommodate offerings.) This base supports a platform on which is mounted the murti—the image or object that functions as the material focus of cultic attention—which is exhibited at a height that can court the gaze of a pedestrian public. Crowning all this is a superstructure to provide shelter, and also to cite visually the spires or domes of architectural idioms associated with institutionally established houses of worship. Within these constraints there is as much room for variation as the urban bricoleur’s palette of scrap materials, found artifacts, and manufactured components will allow.

Figure 7

Figure 7


Figure 8

Figure 8

Sai Baba shrine. Nepeansea Road, Mumbai

20By way of illustrating this point I need look no further than the other side of Nepeansea Road, where a counterpart to the brick-and-stone booth used to stand next to my bus stop. Fronting the entrance to a middle-class residential colony, it was a columnar stack of concrete slabs of varying heights and diameters that had been topped with a shikha, or metal pinnacle. An arch-shaped backdrop of plywood stated the structure’s function—Sai Baba Mandir—and identified its sponsoring body—the Sai Mitra Mandal, or Sai Friendship Association, an organization whose membership mainly consisted, oddly enough, of residents of a different building up the block. Among this mandir’s accoutrements was a bell for worshippers to strike while performing their devotions, a feature shared by many of the other shrines to be found throughout the neighborhood.

21This concludes my list of kaccha and pakka examples. But a few more words about Sai Baba are in order before I wrap up this expository section. Historically speaking, Sai Baba was a holy man who settled in the Maharashtrian village of Shirdi, where he lived and taught in the style of a Sufi fakir, a Muslim mystic, from the late nineteenth into the twentieth century. I have written elsewhere (Elison 2014, 2018 chap. 3) about the saint and his extensive association with the wayside shrines of Mumbai. Here, let me summarize two points. First: Of all the figures enshrined on the city’s streets, the popularity of the Hindu-cum-Muslim Sai Baba exceeds that of all contenders. His teachings extend an ecumenical welcome, and his icon can be understood, in like fashion, to be projecting an appeal to an urban public that encompasses diverse communities.

22But there is also a second reason for Sai Baba’s ubiquity, which has to do with the historical affinity between the Sai cult and industrial technologies of visual reproduction—photography and lithography. From the early twentieth century onward, a central emphasis of the saint’s teachings and miracle stories has been that adoration of his person enables access to divine power, and that his person is present in his figural representations. To reiterate: the picture of Sai Baba is as powerful as his bodily presence. All devotional images circulating in the present day are derived from a handful of historical photographs, and in fact the murti installed at the cult’s origin place in Shirdi is itself a copy of the photograph that had been enshrined there since the Baba’s bodily departure from this world in 1918. The copy delivers the power of the original: once devised to validate the multiplication of sacred sights, it follows that the dictum would come to apply to the multiplication of sacred sites as well.

  • 4 Lawrence Cohen (2017a, 2017b) theorizes the problem of duplication in connection with the Indian go (...)

23Let me set things up for my analysis by proposing that recourse to a formulation like the “polytheistic imagination” attributed to Hinduism is inadequate when it comes to a phenomenon like Mumbai’s illegal religious structures. There are several reasons for this. The one in which I am most invested is the problem of failure, which I will develop below: that is to say, many, if not most, of these sites fail to be recognized as sacred or potent by many, if not most, members of the viewing public. A basic point of dispute for my fieldwork interlocutors was whether a given mandir was “real” (asli) or “fake” (nakli). Take the concern with originals and copies, which is a way of connecting the problem of failure or lack of efficacy with the problem of form.4 If a given site fails to convey divine power, that is, perhaps that has something to do with the imperfect way it looks. And there is also a spatial dimension to the relation between duplicate forms of questionable validity and their authentic originals. The very demand for local stations of God close to hand, right here in the big, modern city—Mumbai’s India-wide notoriety as a godless place is routinely cited by its residents, especially the poorer ones—can reflect a sense of remoteness from the original sources of divine power. Another reason not to pin the analysis to the teachings of Hinduism: it will be noted that not all Mumbai’s wayside shrines are Hindu.

Visual relations

24It seems appropriate to launch this exploration of visual effects by recalling some conversations from the period of my fieldwork I had with the Mumbai-based scholar Vidya Kamat, who is also a visual artist.

25I was introduced to Kamat at a seminar when we were independently developing research interests in the city’s illegal religious structures. Our discussions left me with two questions I chewed over for years as my own argument sought its pakka form. In pursuing inquiries on Dalal Street, where Mumbai’s stock exchange is located, Kamat had become intrigued by the discrepancy between attendance at one Sai Baba mandir, which enjoyed a reputation among the brokers as a wish-granting site, and its less favored neighbors on adjacent blocks. If they are all dedicated to the same divine personality, she asked, why should one shrine project more cultic power than its neighbors?

26She also made an observation about the formal characteristics of the displays enshrined on the streets. There seemed to be a correlation between the class status of the worshiper and the mimetic precision that defined the image worshiped, whether the style in question was a kind of anthropomorphic naturalism, as in the case of most metropolitan Hindu deities; mimetic referentiality to an original murti seated elsewhere, as in the case of the black Krishna image metonymic of the pilgrimage center of Nathdwara; or both, as in the case of Sai Baba of Shirdi. The better off the client, the slicker the aesthetic. Why should that be so?

27Reviewing these questions from the other end of my research, it now seems to me that the first question concerns a contest between two sources of power that the Dalal Street shrine cites at the same time: on the one hand, the Sai Baba personality and, on the other, the particular space the shrine occupies. The second question proposes a relation between an image’s formal attributes and the social positionality of its public. My answer here would have to do with the kaccha-pakka dynamic, specifically with a trajectory of increasing visibility and structural stability that can be conceptualized under a Hindi-English neologism: pakka-fication.

28These are complex questions, and I locate them both at the nexus of Site, Sight, and Cite. This opens the sacred image’s operations to inquiry along three vectors: the relation of the sight to the point it marks in space; the sitedness in society of the public to which it appeals; and the fields of discourse, affect, and bodily habitus it cites as a sign or material extension. And animating all three lines of inquiry is a common point of concern: efficacy. Which of these factors, in combination, will enable the mandir to do its work? Is an efficacious temple the same thing as a real temple? How does a sacred image—or indeed any kind of ideological spectacle—succeed in its bid for the desiring gaze? I propose that answers about the image’s power of address may best be sought through a rethinking of the Hindu ritual of darshan.

29So let’s begin the analytic portion of this essay by reviewing the descriptive portion. What do the diverse sites I have already described here have in common?

30While all but the most kaccha sites are correctly described as mandirs, taken together they complicate on several counts the conventional understanding of just what a temple is. There is the absence of specialists, Brahmin by caste or otherwise, to provide ritual mediation between the deity and its audience. There is the question of design. A common criticism aired by middle-class Mumbai residents is that these shrines are not “real temples” because they do not conform to the Brahminical dicta governing the proportions of structures as laid down in the vastu shastras. But it is highly doubtful whether any of the bricoleur-architects whose constructions crowd the streets look to Sanskrit scriptures for their blueprints. Finally, there is the elision of one of the organizing principles of modern rationalized space: the demarcation between public and private. At what point does a domestic prayer room become a public temple?

  • 5 Andy Rotman’s recent ethnographic work assesses the semiotics of branded merchandise in connection (...)

31What brings my examples together—their common denominator—can be identified as follows: it is the mobilization of religious imagery to mark urban public space. Whether housed in open-faced shelters, showcased behind glass, or simply mounted on the wall, the sacred figures displayed at these sites take advantage of sight lines organized by the built environment to address not only the cult community associated with their construction, but also a transient audience within the dense and constant foot traffic on the Mumbai streets. They appeal, in other words, to diverse publics. I would like to suggest a connection with the advertising graphics with which Mumbai is similarly saturated; I think there is real promise in theorizing visual representations of Hindu gods along the lines of corporate branding that at once targets and constructs diverse constituencies of consumers.5

32But there is a more trenchant case to be made. My stated purpose in this essay is to theorize wayside shrines and related displays as a problem of visual culture. The argument I pursue is aimed at generating some analytic lessons for anthropologists and scholars of religion. To those ends, let’s get started by breaking down the operations of these sites in visual terms. That is to say, let’s situate a typical display in relation to the things, places, and persons it signifies, mediates, and addresses.

33Relation Number One: The display concretizes a claim to a specific piece of urban real estate. Typically, the titleholder—the legitimate owner in legal terms—is a state agency like the municipality or railway authority. The shrine’s rival claim provokes two questions. On whose behalf is the claim being made, and on the basis of what sort of legitimating logic? I propose the following answers: the claim is made in the name of the divine presence seated at that spot as a territorial sovereign. The constituents of the cult community of this sovereign—the people who have submitted to his or her authority—are the legitimate inhabitants of the realm thus established. To refer to the Sanskrit terms used in classical Hinduism, this is sacred space in the sense of pitha, or “seat,” as opposed to tirtha, or “crossing.” But let it be noted that my local interlocutors (of whom few have been Brahmins, and not all of whom would have described themselves as Hindu) never actually used these words in conversation. Textual antecedents of more direct relevance are to be found in records of territorial disputes between villages in precolonial Maharashtra, which involved a process called padthal, adjudication by tribunals of local deities (Wagle [1995] 2000). And as I will go on to discuss, the analogies Mumbai residents reach for when they themselves conceptualize how the shrines transmit divine power tend to come from modern science and technology.

Figures 9 and 10

Figures 9 and 10

Relation No. 1; Relation No. 2

34The arrow pointing backwards here signals the shrine’s relation to its home cult community, Relation Number Two. These are the local people who have erected the shrine to front the street, to provide the community with a public face of sorts. This schematic works best as an illustration of a slum or “encroachment” scenario, where behind the display would be found the community’s dwellings, along with other neighborhood infrastructure that benefits from being shielded—connections to water and electricity, for example. Alternatively, the human agency behind the shrine might be traced to the operators or employees of a local business. The relatively kaccha or pakka condition of a given settlement or business will be reflected in the shrine itself.

35Relation Number Three can be thought of along structuralist lines: the meaning of the display on one block is defined in part by its relation to the displays on the blocks to either side. In other words, the divine image’s function and the quality of its appeal to its client community and to more casual passersby—members of the urban public—are perceived in relation to who its divine neighbors are.

Figures 11 and 12

Figures 11 and 12

Relations No. 3; Relation No. 4

36Dominating the breakdown of visual operations so far has been a dynamic we could summarize as settlement. The shrines mediate the relations of distinct collectivities to the physical space they claim in the city and the social environment they create alongside each other. But with all of this going on in social and spatial dimensions, it wouldn’t do to lose sight of Relation Number Four: the way the murti—the enshrined cult object—corresponds to the divine power it emplaces at a particular spot.

37To refer to the murti as an “image,” as I frequently have in this essay, probably suggests a relation of signification: the murti is a representation of something else—a personality located elsewhere (perhaps Up There). Now, when it comes to images and symbols of the divine, complicating this straightforward relation of mimetic representation is an imperative for the academic study of religion, and so let’s run through some alternative formulations. For starters, the murti may be a vessel that seats material substance, as opposed to signified meaning. This is a commonly expressed understanding that reflects norms of participation with the divine personality through ritual. My urban fieldwork has introduced me to a complementary formulation, one that well suits the tech-savvy inhabitants of India’s media capital. Rather than a vessel, a modern murti may better be thought of as a channel or receiving station, conceived along the lines of the power conduits that are so contingent and precious for Mumbai residents who live in shantytowns, or squat in pakka neighborhoods under precarious conditions. That is to say that a slum-based temple, like a bootleg electrical connection, is a local port that delivers power from its source—a source that is by definition remote.

  • 6 For guardian or sentinel deities in rural contexts, see Fuller 1988; Mines 2005; Raheja 1988.

38Alternatively, to move from an emic to an etic formulation, productive insights might emerge from theorizing the murti as a mirror—a concept to which I will shortly return. But to get back to the original, semiotic level, I affirm the murti may indeed be read—but read critically, with a measure of suspicion, as a mask. In this connection I would cite many of the ubiquitous Sai Baba murtis, which, I would argue, are the ecumenical-looking public face of a cultic practice historically oriented around the maintenance of guardian deities at village boundaries.6

Figure 13

Figure 13

Relation No. 5

39And now let’s devote some space to interrogating the relation that has engaged most scholarly attention to this point: the one between the murti and the viewer. So long as this relation is to be analyzed in the sort of contexts I have described, I think I can safely move on from the iconographic model conventionally presumed in studies of the encounter. To the degree that I am interested in the semiotic relation between viewer-as-reader and image-as-text, my inquiry has been directed not at the symbolic meaning of details but rather the legibility of the form as a whole. The question of a display’s relative kaccha- or pakka-ness, then, extends from the solidity of its claim to physical space to the clarity of its religious appeal.

40The readiest way to gloss Relation Number Five, when a human subject recognizes a materially emplaced superhuman presence, is with a term associated primarily with Hinduism: darshan. This is typically translated as “visual worship.” At the level of colloquial usage, to speak in Hindi of “doing” or “taking” darshan is a shorthand way to denote the whole sequence of ritual interactions prescribed for a human subject before a sacred image. Visual recognition of the deity stands alone as a fulfilling act even if other exchanges mediated through light, music, scent, and material offerings such as food are not performed. Again, vision’s primacy as a channel of contact over communication through language, whether voiced or thought, is the norm observed in contemporary practice. (Parenthetically, I’ll add that vision is also the paradigm of perception in classical Hindu epistemology.) To have engaged in darshan, in short, is for the encounter between human and god to have “worked.”

41At once synecdochic and encompassing, darshan emerges as definitive of the experience of divine presence in Hinduism, central to the history of Indian visual culture, and undertheorized by scholars of both. John Cort (2012) gives an overview of the scholarly conversation, noting the continued salience of Diana Eck’s (1981) short monograph Darśan, with its emphasis on Brahminical norms and a rather totalizing concept of a Hindu “polytheistic imagination,” and following up with a call for reformulations. Cort’s own approach has been to turn to primary texts of a didactic character, a project in which he is joined by other scholars of Indic religions, including Richard Davis (1997:26–50) and Andy Rotman (2009). The route I have chosen follows two other approaches.

42One aspect of darshan that has important implications for my inquiry is that in granting the connection to the worshipper, the deity is perceived to gaze back. This is a visual exchange, then, between subject and master, and as such it invites analysis as a gesture of mutual recognition at the level of subjectification. In taking this view I look to a body of theoretical work on the gaze and its operations, influential in the field of film studies, that originates in the psychoanalytic theory of Jacques Lacan (1977). Darshan-effects have been theorized in the specific context of Indian cinema by Geeta Kapur (1987) and Ravi Vasudevan (2011). Complicating the psychoanalytic paradigm, however, I would also stress that the subject has been conditioned to enter this exchange as an affective, embodied experience; the somatic ground of the encounter comes to the fore in Christopher Pinney’s (1997, 2002, 2004) concept of corpothetics.

43Let me mention two more factors that destabilize the integrity of the viewing subject. The first is that, according to the logic of darshan that obtains among many Hindus, the exchange of glances is accomplished as an exchange of personal substance: as in other rituals, participation in darshan renders the organism porous. There is an established line of anthropological inquiry that theorizes this Hindu way of constructing the self through ritual and other transactional relations; the name most closely identified with the approach is McKim Marriott (1976), but a transactional model devised specifically around darshan is proposed by Lawrence Babb (1981), who extends the connection to film as well. What is more, the operations of the gaze—and, more broadly, of the faculties of sensory perception—are theorized along similar lines in Sanskrit texts, most notably in Tantric teachings that conceptualize the line of sight as an emission of light-energy from the eyes of the perceiving subject (see White 2009, especially 151–61; compare Gonda 1969).

44The second complicating factor is implied in the attribution by worshippers to their gods of control over their fortunes, and their consequent attempts to win favor through ritual transactions. To interact in this way with powerful nonhuman persons is to participate in a system of negotiating and redistributing agency. It follows that to localize the presence of these divine agents alongside oneself, and one’s fellow human dependents, is to inhabit a landscape contoured by flows of decentered and distributed agency (see Gell 1998:137–54).

45To ask, What kind of space is this? is then to ask, What kind of bodies inhabit it—and to what kind of regimes are they subject? Again, I submit that the recognition the murti compels in the viewing subject is not a simple matter of reading the image’s iconography. The sacred spectacle makes its appeal at a deeper level—affective, embodied, and performative. Local sovereignty is in contention, but this is sovereignty in the sense of a structural dispensation that has the power to organize personhood. It configures the subject in relation to the experience of space. And this relation is mediated at discrete and concrete sites throughout the city, through an ideological appeal that is delivered visually.

46Definitive among Hindus of daily ritual participation, darshan calls for rigorous theorization in contemporary and urban contexts. Its academic locus classicus has long been a Brahmin-run temple in a pilgrimage town. But recall that the vector in my graphic points forward, projecting an appeal into public space. The contemporary Indian scene—distinguished on the one hand by migration to larger, more cosmopolitan cities and on the other by the ascendancy of mass media—demands a reappraisal of darshan outside of the traditional temple scenario in which a client community of habitual worshippers can be taken for granted. Let’s review three points.

47First, deterritorialization. The conception of a local deity as a presence organically connected with a particular place over which it exerts sovereignty seems straightforward enough in the rural context. The definitive example is the village goddess (gaondevi) that personifies a discrete community in many rural Indian contexts, especially among Hindu peasants from the less prestigious castes as well as adivasis, or tribal people. An Islamic corollary is the case of a Sufi’s tomb (dargah) whose claim to space is anchored in the historical narrative of the saint’s life. But the movement of deities to the city along with their worshippers, and their re-emplacement alongside their constituents’ new homes, highlight the processual character of the connection between divine power and material space, and the human agency involved in maintaining it. The deterritorialization effect is particularly acute in slum neighborhoods, where the contingent and vulnerable character of housing for humans extends to the structures housing their gods (see Dianteill 2002; Headley 2004).

48Next, reproducibility. Modern Hindus, and other Indians whose devotional practice centers on darshan relations with a materially seated source of divine power, have been enthusiastic about adopting technologies of mass reproduction. But the attribution of authentic power to the cultic image or object is complicated in a milieu where a multitude of copies circulate (see Lambek 2002; Mallapragada 2010; Port 2011). The relevance of the ritual authorization of the murti—paradigmatically by a Brahmin priest—that would, in textbook Hinduism, be the determining factor recedes in an urban environment characterized by a multiplicity of sacred-looking images, on the one hand, and a diverse and transient client community, on the other.

49This brings us to the public. Both the spatial and social relations presumed in the temple-centered model of darshan are recast in the urban scenario of the street-facing prayer booth, at which communications and transactions with the personality enshrined within are open to whoever is moved to recognize it there. Such shrines may serve their client constituencies much in the same way as a “proper,” institutional temple with a priestly staff and a threshold to set off a sacred interior from a profane exterior. But their lack of state or Brahminical authorization, combined with the diverse and transient composition of the foot traffic to which they appeal, opens the question of each site’s efficacy to adjudication by the public eye.

50A telling indication of what is at stake in analyzing darshan as practiced nowadays on the Mumbai streets is given in the remark of a longtime collaborator of mine, a slum resident whose daily rounds in the hectic, media-saturated landscape are punctuated with restorative stops at local darshan stations in his neighborhood. “These days,” he told me, “the gods have become [television] channels.” Let it be noted that his attentions are not restricted to Hindu deities. Indeed, I suspect that the ease with which modern technologies inscribe, reproduce, and circulate typologically correct visual material has elevated mass-mediated darshan to hegemonic status as a mode of worship within metropolitan Hinduism and perhaps even encouraged a corresponding darshan-ization within other religious communities as well.

  • 7 The anthropologist and filmmaker Emmanuel Grimaud has recently completed a highly original interven (...)

51Now, to be sure, one of the things I learned in the course of my fieldwork is that focusing on what happens when a privileged connection is made with an image by a viewing subject is a limiting approach. Moreover, for all its promise of theoretical payoff, the methodological task it set for my research proved confounding.7 But more readily observed and, as I have already mentioned, of no less importance as an aspect of their operation are the conditions under which enshrined images fail to exert an attraction.

  • 8 As a psychoanalytic category, the abject owes its formulation to the post-Lacanian theorist Julia K (...)

52I suggest such negative effects fall into two categories: effects of abjection and effects of repulsion, or warding off. What I understand as abjection—the opposite of subjectification—involves selective blindness on the part of bourgeois subjects of the sort that consigns poor or “backward” people like slum-dwellers to a default position of invisibility.8 (A point of comparison from American cities is the occluded presence, in the gaze of middle-class restaurant patrons, of the busboys clearing their tables, who are generally migrants from Latin America.) The degree to which many of my interlocutors from Mumbai’s more privileged classes shun subaltern eyesores like shrines as legible features of their environment is pervasive enough to be understood as symptomatic of a broader subjective disposition.

53The second kind of negative effect I propose is that of warding off. If a display of religious imagery can inspire piety in some members of the public, inviting them to approach the site, it can also provoke others to avert their gaze, reminding them their own piety is directed elsewhere and repelling their advance. The clearest illustration I can offer is the first of the examples described in this essay: the wall decked with tiles of figures and symbols sacred to diverse traditions. Is everybody welcome here?

54Or is no one? The telltale clue is the positioning of the tiles, with most of them cemented low on the wall. In Such a Long Journey, Rohinton Mistry’s (1991) novel of middle-class life in Mumbai, a turning point comes with the realization of the hero—a Parsi, like the real-life owners of my exemplary wall—that if he decorates his home’s exterior with the images of gods he will be able to prevent passersby from urinating on it. What works as a compelling narrative device in Mistry’s fiction is indeed a quotidian fact of life in Mumbai. The display of tiles on another real-life wall, located in the suburb of Juhu, makes its purpose plain with an English-language caption: “For God’s sake, don’t piss here.”

  • 9 Patricia Spyer (2013) explores the complexities of boundary-marking in the Indonesian context of po (...)

55The owners of properties like these have been generous in their ecumenicalism, but also strategic; they seek comprehensive coverage against an incontinent (male) public, regardless of caste or creed. But it follows that this apotropaic effect can also be put to work selectively. For to the degree that religious images and symbols are perceived as metonymic of distinct communities, they are effective boundary markers in a crowded city where those very communities are contesting scarce urban space. By contrast with the abjection scenario, in which a viewing subject refuses to recognize a religious sight on the landscape, in this alternative scenario of negative address the subject does indeed recognize the gesture of address—and understands that its welcome is directed at others. Recall the brass-toned Ganesh at the Charni Road overpass. The operators of the gambling club who installed it did so, no doubt, in order to shield their project with a show of piety calculated to appeal to Maharashtrian Hindus, a dominant constituency within the ranks of the police (and other state agencies in a position to interfere). And yet, regardless of the intent or instrumentalism behind the display, the murti also presents non-Maharashtrians and non-Hindus with a warning signal, an admonition to tread lightly. Encountered at the end of a pedestrian crossing, this particular icon might be described as Checkpoint Charni: “You are now entering the Maharashtrian sector.”9

56Now, for the sake of the argument, let’s bracket the observed fact that no one engages in sustained worship with this particular murti and briefly consider Lord Ganesh as a territorial sovereign. For those whom a locally stationed icon interpellates, the deity thus manifested holds sway over a discrete tract of settled land, recognizing his subjects even as they recognize him, and looking over them with protection and blessings within the bounds of his domain. Such is the general outline of the site-sight-cite relations structuring the cults of village deities in rural Maharashtra and beyond. It is tempting to apply the same, “rural” logic to the use of the Ganesh icon by local Hindus making claims to parts of urban neighborhoods. The valorization of Ganesh among Maharashtrian Hindus (specifically, Kunbi-Marathas) as their particular champion invites comparison with the bond enjoyed by a village’s tutelary deity with its dominant caste. There is also a parallel to be noted between the boundary gods who keep watch from their stations at the village periphery and Ganesh’s role as the lord of the threshold, the Remover of Obstacles (he removes them, that is, for his petitioners, and by implication leaves them in place against outsiders).

  • 10 Much material on the Shiv Sena’s use of maha aaratis to incite violence against Muslims can be foun (...)

57Yet the vital factor to be noted in this connection is the political mobilization of the Ganesh icon in public space in Mumbai. The project begins in the 1890s with the nationalist leader B. G. Tilak’s promotion of the Ganesh festival, or Ganapati Puja, as a vehicle for generating solidarity among Maharashtrian Brahmins and Marathas (see Kaur 2005; Hansen 2001:29–31; Masselos 1991:50–52). It continues through to the present with the all-but-official sponsorship of the festival by the Shiv Sena, the Maharashtrian chauvinist party that has dominated politics in the city for the past four decades. And one thing that cannot be ignored about this alliance is how, as an appropriation of public space in the image of a militant Maharashtrian Hinduism, Ganapati Puja plays on memories of the bloody winter of 1992–93, when over a thousand people were killed in sectarian violence, the vast majority of them Muslims. That episode’s signal contribution to religious practice was the maha aaratis, or prayer rallies, the Shiv Sena staged on streets in mixed neighborhoods as an all-too-successful incitement to riot.10

58In light of this history, the idea voiced by several of my liberal, English-speaking interlocutors that the spread of the Hindu mandirs on the streets—not confined to Ganesh sites—keeps pace with the ascendancy throughout India of Hindu nationalist ideology suggests a correlation that calls for further investigation. Importantly, however, any inquiry along these lines would need to take into account the stress placed in the Maharashtrian context on bonds of blood and soil. According to the Shiv Sena and its ideological imitators, the Marathas are bhoomiputra, the “sons of the soil,” and this nativist construction, derived as it is from the self-image of an agriculturalist caste, complements the cultic focus on local tutelary deities. It should be noted that this Maratha nativism does not always mesh with the sort of Hindu nationalism that prevails on the all-India level, especially in its more Brahminical manifestations.

59Politics aside, another reason is often given to explain why illegal religious structures excite such offense in the eyes of Mumbai’s elites, and this one also hinges on their warding-off function. In much oral lore about Mumbai neighborhoods, wayside shrines are not merely cast as transgressive, but as the active nodes of an alternative, criminal geography of the city—outgrowths of the underworld, as it were. Stories about “what really goes on” at the ostensibly sacred site on the corner go beyond the efforts of small-time religious charlatans to exact profit from the credulous. In this characterization the cynically repurposed icon is at once a shield and mask, its worship providing the guise under which the shadiest members of a diverse urban public can come together to exchange illicit goods and services. Among the wayside activities the shrines are said to enable in this way are drug dealing, gambling, smuggling, and prostitution. Rumor even has it that certain sites are themselves the physical repositories of the wealth generated in such transactions.

60And yet, owing to norms of piety prevailing across social strata—and, indeed, to the official principle of “secularism” that stresses ecumenical respect for the diversity of religious practice in India—at the end of the day, the agents of the state find it hard to remove the structures. This is one reason why shrines connected to slum communities tend to get built in proximity to other, less valorized kinds of structures housing families or businesses. Again, the sponsorship of shrines by neighborhood power brokers who can claim important allies means that patronage of specific sites may, in practice, reach up to the higher tiers of officialdom. As case in point is a Hanuman temple that stands just off Lamington Road, not far from the downtown red-light district. The structure began as a shrine attached to the local Shiv Sena cell (shakha) but gradually expanded across legal boundaries to encompass an interior compound of its own; it now dominates the whole block.

61The exercise of strictly thisworldly clout seems inescapable here; the political edge to the gesture comes across all the sharper in view of the historically Muslim character of the neighborhood. And yet the site’s expansion is also a claim to public space at a deeper ideological level. The effect is a refiguration of the cityscape from which human agency recedes, an invitation to accede to the logic that the prime mover behind the site’s flourishing—as manifested in its visual and material increase—is the resident deity himself. And that is to say that the city’s guardian deities actually are extending protection over their human constituents in very concrete, material ways.

Concluding remarks

62Any ethnographic study of wayside shrines must investigate the human agency behind them, and its parochial and often instrumentalist motivations. But even blatantly bad-faith efforts at shrine construction take advantage of a presumption not many Mumbai residents are moved to contest: that the city holds some sites that are truly more sacred than others, where people can experience authentic darshan or other forms of interaction with divine power. It is, of course, this appeal to an extralegal yet valorized spatial order that distinguishes street shrines from other forms of subaltern encroachment on public space, such as slum housing or street vendors’ carts. The projection of this alternative domain of sovereignty and subjectivity onto the urban landscape is an aspect of the phenomenon that should be of particular interest to scholars of religion. And so here, let’s give the last word to the people who run the mandirs.

63Some shrine operators I spoke with struck me as more sincere than others. But regardless of the question of their personal bona fides, the thing that is characteristic about the way my interlocutors tell their stories—and the thing I choose to stress here—is the disavowal of their own agency in the shrines’ genesis. Shrines, in the prescribed telling, are the concrete manifestation of a divine agency that is heeded by human subjects. Immanent divinity can reveal itself at some places in the form of a symptom, a material clue like a swelling in the ground or a whorl in a tree that triggers recognition in the right person. This is the logic of the svayambhu, or “self-manifested,” icon, which anchors the origin stories of many of the famous sites of Brahminical Hinduism. At other sites, God—in one of His or Her myriad forms—may appear to the right person through the medium of a dream.

64The elect person, the visionary, is the one endowed with the insight to make the identification and the drive to make it visible to others. And in so doing, investing the blessings that gradually accrue through this practice of service (some of which take the form, in the project’s due course, of contributions from the patrons who collect around it), that human servant tends the site’s flowering. From flatness to depth, from obscurity into clarity, and—via assimilation within the visual codes prevailing in public—into legibility as an image of the sacred.

Top of page


Babb, Lawrence A. 1981. “Glancing: Visual Interactions in Hinduism.” Journal of Anthropological Research 37(4):387–401.

Cohen, Lawrence. 2017a. “Duplicate.” In “Keywords,” edited by M. Ashar, T. Brown, A. Doron, and C. Jeffrey, special section, South Asia: Journal of South Asian Studies 40(2):1–13.

Cohen, Lawrence. 2017b. “Duplicate, Leak, Deity.” Limn 6. Retrieved February 27, 2018 (

Cort, John E. 2012. “Situating Darśan: Seeing the Digambar Jina Icon in Eighteenth- and Nineteenth-Century North India.” International Journal of Hindu Studies 16(1):1–56.

Davis, Richard J. 1997. Lives of Indian Images. Princeton, NJ: Princeton University Press.

Dianteill, Erwan. 2002. “Deterritorialization and Reterritorialization of the Orisha Religion in Africa and the New World: Nigeria, Cuba, and the United States.” International Journal of Urban and Regional Research 26(1):121–38.

Elison, William. 2014. “Sai Baba of Bombay: A Saint, His Icon, and the Urban Geography of Darshan.” In “Indian Gurus.” Special issue, History of Religions 54(2):151–87.

Elison, William. 2018. The Neighborhood of Gods: The Sacred and the Visible at the Margins of Mumbai. Chicago: University of Chicago Press.

Fuller, C. J. 1988. “The Hindu Pantheon and the Legitimation of Hierarchy.” Man (New Series) 23(1):19–39.

Gell, Alfred. 1998. Art and Agency: An Anthropological Theory. Oxford: Oxford University Press.

Gonda, Jan. 1969. Eye and Gaze in the Veda. Verhandelingen der Koninglijke Nederlandse Akademie van Wetenschappen, Afd. Letterkunde. Amsterdam: North Holland Publishing Company.

Grimaud, Emmanuel. 2016. Ganesh Yourself. Film. Paris: Rouge International.

Hansen, Thomas Blom. 2001. Wages of Violence: Naming and Identity in Postcolonial Bombay. Princeton, NJ: Princeton University Press.

Headley, Stephen C. 2004. Durga’s Mosque: Cosmology, Conversion, and Community in Central Javanese Islam. Singapore: ISEAS.

Kapoor, Raj. 1977. Satyam Shivam Sundaram. Film. Mumbai: RK Studios.

Kapur, Geeta. 1987. “Mythic Material in Indian Cinema.” Journal of Arts and Ideas 14–15:79–108.

Kaur, Raminder. 2005. Performative Politics and the Cultures of Hinduism: Public Uses of Religion in Western India. London: Anthem.

Lacan, Jacques. 1977. Ecrits: A Selection. Translated by A. Sheridan. London: Tavistock.

Lambek. Michael. 2003. The Weight of the Past: Living with History in Mahajanga, Madagascar. New York: Palgrave Macmillan.

Mallapragada, Madhavi. 2010. “Desktop Deities: Hindu Temples, Online Cultures and the Politics of Remediation.” South Asian Popular Culture 8(2):109–21.

Marriott, McKim. 1976. “Hindu Transactions: Diversity without Dualism.” Pp. 109–42 in Transaction and Meaning: Directions in the Anthropology of Exchange and Symbolic Behavior, edited by B. Kapferer. Philadelphia: Institute for the Study of Human Issues.

Masselos, Jim. 1991. “Appropriating Urban Space: Social Constructs of Bombay in the Time of the Raj.” In “Aspects of ‘the Public’ in Colonial South Asia,” edited by S. B. Freitag. Special issue, South Asia: Journal of South Asian Studies 14(1):33–64.

McLain, Karline. 2016. The Afterlife of Sai Baba: Competing Visions of a Global Saint. Seattle: University of Washington Press.

Mines, Diane P. 2005. Fierce Gods: Inequality, Ritual, and the Politics of Dignity in a South Indian Village. Bloomington: Indiana University Press.

Mistry, Rohinton. 1991. Such a Long Journey. New York: Knopf.

Pinney, Christopher. 1997. Camera Indica: The Social Life of Indian Photographs. Chicago: University of Chicago Press.

Pinney, Christopher. 2002. “The Indian Work of Art in the Age of Mechanical Reproduction: Or, What Happens When Peasants ‘Get Hold’ of Images.” Pp. 355–68 in Media Worlds: Anthropology on New Terrain, edited by F. D. Ginsburg, L. Abu-Lughod, and B. Larkin. Berkeley: University of California Press.

Pinney, Christopher. 2004. “Photos of the Gods”: The Printed Image and Political Struggle in India. London: Reaktion.

Port, Mattijs P. J. van de. 2011. Ecstatic Encounters: Bahian Candomblé and the Quest for the Really Real. Amsterdam: Amsterdam University Press.

Raheja, Gloria Goodwin. 1988. The Poison in the Gift: Ritual, Prestation, and the Dominant Caste in a North Indian Village. Chicago: University of Chicago Press.

Rotman, Andy. 2009. Thus Have I Seen: Visualizing Faith in Early Indian Buddhism. New York: Oxford University Press.

Rotman, Andy. 2017. “Who Moved My Cheez: Brandism, Bazaarism, and the Future of Indian Markets.” Presented at the Annual Conference on South Asia, University of Wisconsin, Madison.

Sabrang. N.d. “Damning Verdict: Report of the Srikrishna Commission.” Retrieved June 22, 2013 (

Shinde, Kiran A., and Andrea Pinkney. 2013. “Shirdi in Transition: Guru Devotion, Urbanisation and Regional Pluralism in India.” South Asia: Journal of South Asian Studies 36(4):554–70.

Spyer, Patricia. 2013. “Images without Borders: Violence, Visuality, and Landscape in Postwar Ambon, Indonesia.” Pp. 101–26 in Images that Move, edited by P. Spyer and M. Steedly. Santa Fe, NM: School of Advanced Research Press.

Vasudevan, Ravi S. 2011. The Melodramatic Public: Film Form and Spectatorship in Indian Cinema. New York: Palgrave Macmillan.

Wagle, N. K. [1995] 2000. “On Relations amongst Bhūts, Gods, and Men: Aspects of Folk Religion and Law in Pre-British Maharashtra.” Pp. 181–220 in Folk Culture, Folk Religion and Oral Traditions as a Component in Maharashtrian Culture, edited by G. Sontheimer. Heidelberg University South Asian Studies. Delhi: Manohar.

White, David Gordon. 2009. Sinister Yogis. Chicago: University of Chicago Press.

Top of page


1 I am indebted to Wendy Doniger for conceptualizing the framework in this way in an informal discussion with me, many years ago now.

2 Blueprints by Malika Khurana. Photographs by the author.

3 The pakka intervention to the Hanuman shrine in Khetwadi can be seen as a tactical response to this second criterion. The improvement has been made over a formerly free-standing shrine on the public street as an extension of the wall behind it, which is private property.

4 Lawrence Cohen (2017a, 2017b) theorizes the problem of duplication in connection with the Indian government’s Aadhaar initiative: the introduction of a comprehensive new national I.D. that incorporates biometric data.

5 Andy Rotman’s recent ethnographic work assesses the semiotics of branded merchandise in connection with the piety and credit of Hindu merchants in the bazaars of Banaras. See Rotman 2017 for some preliminary findings.

6 For guardian or sentinel deities in rural contexts, see Fuller 1988; Mines 2005; Raheja 1988.

7 The anthropologist and filmmaker Emmanuel Grimaud has recently completed a highly original intervention in the ethnography of darshan. He led a team that designed an interactive robot in the form of a Ganesh murti and invited participation with it on the occasion of Mumbai’s 2014 Ganesh festival. Members of the public experimented with “playing God,” voicing their insights and questions through an audio prosthesis and projecting their own features onto the screen set in the center of the device’s stylized elephant head. The result is Ganesh Yourself (Grimaud 2016), an experimental documentary film that raises provocative questions about darshan, postcolonial subject positions, and ethnographic ethics.

8 As a psychoanalytic category, the abject owes its formulation to the post-Lacanian theorist Julia Kristeva (1982).

9 Patricia Spyer (2013) explores the complexities of boundary-marking in the Indonesian context of postwar Ambon.

10 Much material on the Shiv Sena’s use of maha aaratis to incite violence against Muslims can be found in the official government inquiry on the riots, “Damning Verdict: Report of the Srikrishna Commission” (Sabrang N.d.).

Top of page

List of illustrations

Title Figure 1
Caption Tiles with images of Durga and Krishna on a private wall fronting the street. Dadar, Mumbai2
File image/jpeg, 252k
Title Figure 2
Caption Small shrine dedicated to Sai Baba of Shirdi embedded in tree trunk. Chowpatty, Mumbai
File image/jpeg, 228k
Title Figure 3
Caption Bas-relief of Ganesh on Charni Road pedestrian walkway. Girgaum, Mumbai
File image/jpeg, 256k
Title Figure 4
Caption Hanuman shrine. Khetwadi, Mumbai
File image/jpeg, 764k
Title Figure 5
Caption Sai Baba shrine. Nepeansea Road, Mumbai
File image/png, 4.4M
Title Figure 6
Caption Schematic of typical shrine
File image/jpeg, 592k
Title Figure 7
Caption Axonometric
File image/jpeg, 188k
Title Figure 8
Caption Sai Baba shrine. Nepeansea Road, Mumbai
File image/png, 1.9M
Title Figures 9 and 10
Caption Relation No. 1; Relation No. 2
File image/png, 712k
Title Figures 11 and 12
Caption Relations No. 3; Relation No. 4
File image/png, 725k
Title Figure 13
Caption Relation No. 5
File image/jpeg, 496k
Top of page


Electronic reference

William Elison, Site, Sight, Cite: Conceptualizing Wayside Shrines as Visual CultureSouth Asia Multidisciplinary Academic Journal [Online], 18 | 2018, Online since 15 May 2018, connection on 23 May 2024. URL:; DOI:

Top of page

About the author

William Elison

University of California, Santa Barbara

Top of page



The text only may be used under licence CC BY-NC-ND 4.0. All other elements (illustrations, imported files) are “All rights reserved”, unless otherwise stated.

Top of page
Search OpenEdition Search

You will be redirected to OpenEdition Search