Bhakti and Its Public
Christian Lee Novetzke
In this essay I want to chart a space for understanding bhakti that moves between the
two dominant, and in some ways mutually contradictory, modes of describing bhakti
in modern scholarship.1 These two positions are the idea, on the one hand, that
bhakti forms a social movement and, on the other, that bhakti is an act of personal
devotion. In between these poles of the broadly social and the strictly personal, I
want to suggest that bhakti seeks to form publics of reception rather than communities that imply a single cohesive issue or idiom. I advance the thesis that all manifestations of bhakti are performances and, more to the point, public ones, that is, performances that are part of, or help form, publics of reception. While I would not argue that
bhakti never describes a private affair of devotion, the sense in which bhakti enters the
history of India is not through the private realm, but through the social world of caste,
labor, media (both written and non-written), and markets outside the home and the
heart of an individual.
To Çakara, around the ninth century CE, is attributed the argument for an individualized, monist vision of bhakti, which he is said to have expressed in the Çivånandalahir by his metaphor of the river, or self, joining the ocean, or Brahman. His broad
influence had much to do with philosophical-religious understandings of bhakti in
the direction of a personal pursuit. Yet the metaphor is telling—the river may be an
individual stream on its way to the ocean, but it is also one of the central venues of
collective, social Hindu religious practice (think of the Kumbha Melå or the Gagå in
Banaras) and also, universally, a key site for commerce, economics, travel, and urban
development. The river is many things: a boundary, a threat, a source of sustenance, a
channel of trade. The river is an apt metaphor for the public, as much as it is for bhakti
and religious expression itself (for more on bhakti and rivers, see Feldhaus 1995).
Indeed, despite Çakara’s usage, we also find in Sanskrit an impressive series of texts
that associate bhakti with public performance. In treatises on aesthetics, and especially
in texts attributed to Abhinavagupta in the early eleventh century, the nature of bhakti
as affect is debated. Bhakti in this context is beyond rasa, beyond the “flavor” of a
performance, but is one of those key “experiences,” or bhåvas, that a rasa might
explore; all roads, as it were, may lead to bhakti, and it cannot be limited to any
particular kind of affect.2 It is thus understood not only to be a shared experience,
InternationalInternational
Journal of Hindu
Studies
11, 3Studies
(2007):11,
253–70
Journal
of Hindu
3 (2007): 255–72
© 2007 Springer
© 2008 Springer
DOI 10.1007/s11407-008-9049-9
256
254 / Christian Lee Novetzke
like love or anger, but an expressive one—an expression of the self, perhaps, but an
outward expression of the self, a performance of emotion. The relationship between
rasa and bhåva, and in particular its affective display, its public performance, suggests
both publics that theorize and publics that consume bhakti as a key “experience” of
life.
Therefore, whether we are talking about songs sung to God, recorded in performance
and text, or bhakti practices, such as p¨jå, darçan, pilgrimage, or keeping vows, these
things all take place in the context of some audience; if no one else, one can be sure
that at least God is always watching. I have always thought of the story of VålmFki as
illustrative of this point. The brigand is redeemed by accident. As penance for his
crimes, he sits alone, meditating on death by repeating the word for it in Sanskritic
languages, “måra.” Within this rote anamnesis on death the two syllables are slowly
transposed so that VålmFki is meditating upon “Råma” unintentionally. His accidental
repetition is “heard” by God, and he is saved; according to legend, he goes on to
compose one of the great texts of bhakti in Indian literary history, the Råmåya~a. What
leads to VålmFki’s redemption is not intention, or even any trace of “devotion to a
personal deity,” much less participation in a social movement, but rather the reception
of his sound by a small but powerful audience, in this case God. This is the foundational frame narrative for an epic text that has itself produced innumerable publics
through millennia of retelling in India and worldwide.
Bhakti seems to need an audience. Norman Cutler (1984) draws for his reader a
graphic representation of what I would call the public nature of bhakti. He offers a
triangle of lines that link gods, poets, and audience. By invoking the idea of a “public”
I do not wish to isolate any one of these three nodes—though that of “audience” might
seem the most appropriate. Instead, I want to point to the tripartite structure itself, to
the ways bhakti relies on the flow of its sentiment and information about the communal identity of fellow listeners, all communicated visibly, mediated by an audience.
We can see this evoked in the modern medium of film and its attendant idioms. In all
the great bhakti films of Indian cinema—such as the Damle and Fattelal films about
Tukåråm (1936) or Jñåneçvar (1940)—we are the observers of a bhakti public. Here
bhakti always represents an activity performed before audiences, which is in turn
projected on a screen and performed for a viewing public. As viewers of these films,
we are watching Cutler’s triangle in its entire form: we see the sant, we hear his songs
to God, and we observe the audience in turn observing and subsequently remembering
him as a bhakta. In a sense, bhakti is this very mise en scène—the composition in its
entirety, including layers of reception.
This mise en scène is essential to centuries of hagiography as well. The miracles
associated with the sants are usually public affairs, set before audiences that at times
adore the sant and at times persecute him or her. One can think of several examples
in the history of the miraculous in sant literature. For instance, both KabFr and
Nåmdev share the story of being persecuted by a sultån who threatens the sant with
the wrath of a stampeding elephant.3 Their persecution is thwarted by divine intervention, and this is witnessed by the sultån’s subjects, as well as by orthodox relig-
Bhakti and Its Public / 257
255
ious leaders who have come to see the famous sant’s tortures. When figures like
Kab>r, M>råbå>, Nånak, or Nåmdev are faced with persecution at the hands of some
mighty temporal authority, they are delivered through their reliance on faith, and the
miracle that intercedes on their behalf takes place in public. Such public persecution
of figures by powerful political authorities is regularly depicted in hagiography and
art, and the mise en scène in these visual-literary moments always supplies an
audience witnessing the triumph of the bhakti of the sant over the power of the king
or sultån. The highly personal “memoir,” I would suggest, is never to be found in
hagiography, though much hagiographical material is transmitted via a first-person
narration attributed to the sants themselves. The story of a sant’s life is also a story
of a context for that life and is coded with the reception of that life in ever-changing
contexts. The audience is never absent, but imagined and assumed in any iteration of
bhakti, whether literary, iconographic, performative, or ritualized. Of course, even
the highly theorized conjectures on bhakti found in speculative Sanskrit texts have
their audiences, but in narratives depicting bhakti in action the audience is more
pronounced, more clearly situated within a public, both in the narrative itself and in
the narrative’s social context of reception.
Let me try to make this argument by expanding on the ideas of bhakti and public.
There are certain words in all languages that defy even heuristic definition, much less
an exact and lasting one, and the word bhakti is one of these enigmatic terms.4 It has
appeared in texts and practices for over two millennia and has attracted two centuries
of Western scholarly scrutiny, yet it remains an essentially ambiguous term (see
Prentiss 1999: 17–41; Sharma 1987). Still we can approximate the term’s contours
with recourse to etymology, the historicizing logic of philology. In this context the
word appears in a number of South Asian religions, but particularly in Hinduism,
where it signals a host of meanings that circulate around its Sanskritic verbal root,
bhaj, “to share, to apportion,” and hence comes most commonly to indicate love,
sharing, worship, and devotion. This verbal root has other associations too, however.
The most common include: to divide, distribute, and bestow; to obtain as one’s share,
to enjoy or possess; to resort to, engage in, assume (as a form), put on (garments); to
experience; to practice or cultivate; to choose; to serve and adore (Monier-Williams
1993: 743; Sharma 1987: 40–41). The noun bhakta refers to a person (or in some cases
a thing) in whom some qualities of bhakti inhere. Thus, a bhakta is someone who is
devoted, who serves, who is associated with a community, and who is faithful and
loyal.
A common scholarly convention interprets bhakti to mean “personal devotion” or a
sentiment of intimacy with a deity, but the term is also used in highly abstract contexts
where the “personal” is not present. In these cases, both in scholarship and within the
Indian public sphere, bhakti denotes a “movement” of social protest against caste,
class, religious, or gender inequities. Historically, no single social movement has
cohered around the term bhakti or its sentiments. Instead one finds innumerable
religious communities, practices, bodies of texts, and so on, that invoke bhakti as their
generative principle. Many scholars refer to a “bhakti movement” composed of a
256 / Christian Lee Novetzke
258
unified, if heterogenous, field of texts and practices produced and maintained in South
Asia over the last two millennia. These Indian spheres of discourse—both textual and
practical—use the word bhakti self-descriptively and reference one another, producing
intertextual consistencies among texts, making “publics” out of their own mutual
references. Sometimes this is purposeful, as, for example, when particular hagiographers create genealogical relationships among figures, as Anantadås (ca. the end of
sixteenth century) does for a collection of sants in northern India (see Callewaert and
Sharma 2000; Lorenzen 1991), or Rajab does for Dåd¨ Dayål (both ca. sixteenth
century) in Rajasthan (see Callewaert 1978), or MahIpati (eighteenth century) does for
the VårkarI sants in Maharashtra.5 And sometimes this is implied, as we see in the
similarities of stories that link KabIr and Nåmdev via a trope of persecution in the
collections I have just mentioned. This inter-referential practice helps produce the
effect of “movement” where it may not have a social basis.
Friedhelm Hardy in his book Viraha Bhakti (1983: 489–91) observes similar interreferential traits between South Indian Tamil devotional texts and the Bhågavata
Purå~a, what he calls an “opus universale,” a text composed in South India in the
tenth century CE and found throughout India by the fifteenth century. Here a single
text can serve as a kind of nodal archive for bhakti in its public reception over many
centuries. It produces a broad cohesion for concepts that circle around the idea of
bhakti both on its own in an elite sphere and more broadly, as it is performed and
interpreted. Likewise, we might consider the Bhagavad Gtå and the Råmåya~a,
two texts firmly within the discourse of bhakti in India that are articulated both in
Sanskrit and regional languages, criss-crossing the subcontinent in terms of historical
period and regional inflection. Such examples lead scholars to argue that a transregional movement with bhakti at its core is evident in the ways these core narratives
can travel.
Yet even in the case of texts such as these, which are so clearly cosmopolitan in
their reach, there are significant restrictions along sectarian and class lines. If one
thinks about Cutler’s triangle, the audiences targeted by these iterations of bhakti are
restricted, at least in their “original,” that is, Sanskritic forms. They represent an elite,
literate, Vai‚~ava sort of bhakti; they do not exhibit the entirely of what bhakti can
mean as a keyword in texts and practices known throughout India. Here is one of the
central quandaries occupying the current volume. Although scholarly work typically
associates bhakti with literary practices that signal the rise of devotional sentiment
(often also marking the first or second literary layers of regional literatures6), many
non-literary practices are also described as bhakti, including pilgrimage, daily worship,
the repetition of a deity’s name or names, and so on. In practical terms, then, bhakti
resists confinement to any particular action or utterance.
The history of the literary genealogy of bhakti is equally broad. Traces of bhakti
exist as early as the ¸g Veda, possibly composed toward the end of the second millennium BCE, where we find the variations of the verbal root bhaj appearing in the
context of entreaties to particular deities such as SarasvatI or, as in the GåyatrI mantra,
a call to the deity Savit® to protect a supplicant. But it is in a very early Buddhist text
Bhakti and Its Public / 259
257
that we find the first uses of the precise form bhakti as meaning “devotion” of some
kind. In the Theragåthå, probably composed in the fourth century BCE, bhakti appears
to indicate devotion to the way or “dharma”—another term that resists definition—of
the Buddha. Here, as so frequently in later usage, bhakti connects the personal and the
social, linking an individual to a shared social moral order (dharma). Around this
period, the word also appears in the A‚†ådhyåy" of På~ini indicating “devotion” to
something, but not necessarily a deity. A general scholarly consensus sees the first full
articulations of bhakti as the term is used today in the Bhagavad G"tå, which was
probably extrapolated from the Mahåbhårata and composed around the beginning of
the Common Era; in the Bhågavata Purå~a; and in the Bhaktis¨tras, probably composed around the tenth century CE, not long after the Bhågavata Purå~a. These are all
well-known and well-studied Sanskrit texts, yet regional languages also expressing
sentiments associated with bhakti began to appear in South India in the middle of the
first millennium CE. Tamil bhakti texts appear as early as the fifth century, and the
next twelve hundred years see a long efflorescence of regional literatures throughout
the subcontinent, within which the expression of bhakti appears a consistent and
prominent feature.
Alongside this literary production, we can see a turn towards the physical representation of the public and devotional motivations of bhakti in the middle of the first
millennium CE. At the end of the Gupta Empire (320–647) and in the reign of the
Pallavas and the På~yas in South India (fourth to tenth centuries), we see a surge in
royally sponsored temple construction, the creation of “homes” for deities and loci for
public worship. These provide at least one early context for the practices associated
with bhakti, such as the process of making visual contact with a deity, or darçan, and
the offering of goods to a deity’s image, or p¨jå. These acts were replicated in homes,
and the transformations of the public economy of worship bear some significant relationship to the development of regional literatures that take up the ethos of bhakti. In
the midst of the cross-currents of regional literary developments and the opening of
public economies of devotion, we seem to find the rudiments of public performative
expressions of bhakti: the plays, dances, theatrics, and songs that have come to stand
for bhakti in modern scholarly discourse.
Unifying the myriad forms that bhakti has historically taken and continues to take is
the idea of a public, which I think of as a social unit created through shared cultural
phenomena and reinforced by demonstrations in public of these shared cultural
phenomena. Publics are not exclusive—indeed they can hardly be regulated at all.
The idea of a (or “the”) public has a long history in Western theoretical writing and
political life, ranging from Immanuel Kant’s (1995 [1784]) ideas about reason among a
reading public, John Dewey’s (1927) ideas of public political deliberation, and Jürgen
Habermas’s (1991 [1962]) influential theory of “the public sphere,” to the attribution
of all kinds of opinions to the “public” by politicians and the metrics of public opinion
employed by pollsters around the world.
The identification of publics in scholarship on South Asia is now a decades-long
project. The fields of political studies and historical anthropology have provided the
260
258 / Christian Lee Novetzke
most salient treatments of the functioning of publics. For example, the many publications by Paul Brass in political science (1974, 2003) have long engaged the formation of publics, as has more recent work, such as studies by Ashutosh Varshney
(2003) on civil society. In historical anthropology, the work of Bernard Cohn (1996)
pioneered a new genre of the study of publics and the colonial state. It has expanded
in the work of scholars such as Nicholas Dirks7 and Arjun Appadurai,8 as well as
more recent work by Sandria Freitag (1989), Anne Hardgrove (2004), and Douglas
Haynes (1991) (see also Yandell and Paul 2000). We have the excellent work of
Francesca Orsini (2002) on the Hindi public sphere in India, and Milind Wakankar
(2003) has shown how bhakti figures such as KabDr appear there prominently. In this
public sphere, bhakti can be integrated into ideas of what it is to be “Indian,” that is,
into a discourse of nationalism, of deç-bhakti, both a devotion to the nation and a
commitment to the public that comprises the nation.
When “publics” are introduced in the context of South Asia, it is usually as a
description of some formulation of the public sphere and/or of public culture. In
both cases the publics scrutinized tend to be set firmly within a designated modern
period (usually the nineteenth century to the present). Given the intimate association
between the rise of mass media, printing, and the nation that tends to characterize
scholarship on the public sphere, the association with modernity makes sense.
The case of “public culture” is somewhat different. The genealogy of this term
borrows something from the understanding of publics within political theory and the
notion of the public as a sphere of social activity opposed to the private. But the
major use of “public culture” appears to have begun with the second wave of cultural
studies initiatives of the 1980s. Its distinction from the notion of a “public” in the
political theoretical sense is its emphasis on a cultural studies or “culture as text”
approach. Used by cultural studies predominantly (and by anthropologists, in principle), public culture seems to me to have become a study of modern, urban cultural
forms in interstitial spaces: between the state and the individual, between political
and popular culture, between the global and the local, between the modern and the
traditional. Public culture is distinguished from the idea of the public sphere is at
least two ways. The idea of the public sphere is what I would call highly genealogical in that the parameters of its debate, and the dialectic of the debate’s continuity,
are located within the seminal work of Habermas. Following Habermas, the construction of the public sphere often relies upon at least the technologies of modernity
(the printing press, and so on) if not a good number of its key forms (the nation, civil
society, political life, and so on). Whatever the differences in these key terms in
social theory, both tend to emphasize the modern: for the public sphere, the core
features of modern social organization are a prerequisite, whereas for public culture,
the interstices of modern life appear to be the generative site of study.
Like Subaltern studies, the study of public culture entered mainstream academia
at the hands of South Asianist scholars. Though the term “public culture” predates
its usage in the influential work of Arjun Appadurai and Carol Breckenridge (see
Breckenridge 1995), through their stewardship of a series of influential publications
Bhakti and Its Public / 261
259
and journals,9 South Asia has remained a key site for the study of public culture in
general.10 However, as I have said, these studies tend to remain within modernity,
expressing the relationship between modern forms of the state (colonial or democratic) and publics, or within modern cultural formations like the public sphere. For
this reason, though these excellent studies make lucid use of ideas about the public
in modern contexts, they cannot provide a model for understanding bhakti in its
broadest, historically richest manifestations. They stop short at the threshold between
the pre-modern or non-modern and the modern.
In formulating a broader, usable outline of the idea of a public, we can turn to
Michael Warner, who provides a description of a public similar to the use I make of
the term here in the context of bhakti: “a public enables a reflexivity in the circulation
of texts among strangers who become, by virtue of their reflexively circulating
discourse, a social entity” (2005: 11–12). The circularity of this heuristic definition is
purposeful and recalls the same kind of circularity as that used by Clifford Geertz
(1973) to describe religion as a self-reinforced system of symbols, or a socially repetitive system, as in Peter Berger’s (1967) work, or as a principle of the “reflexive sociology” of the cultural field, as Pierre Bourdieu (1993) has described it. Importantly,
however, Warner (2005: 15) makes plain what at least Geertz implies: the object of the
belief, in this case of a public, is a fiction. It is not a physically demonstrable thing, like
a state, village, township, or other polity; nor does it exist in a carefully constructed
discourse, like a judiciary, a set of laws, or a dogma. A public relies as much on the
imagination of each individual as on a collective agreement as to its existence. People
must believe they are part of a public, and this gives it both its strength and its ephemeral quality.
Likewise, the public created when bhakti is invoked is ruled neither by dogma nor
coercion, but made cohesive by a kind of social agreement. Bhakti indicates a practice
of sharing, equal distribution, and mutual enjoyment, what Karen Prentiss calls
“participation” (1999: 24), an interaction that suggests the “embodiment” of bhakti as a
prerequisite for its practice. This is a crucial point when we are discussing systems of
memory that are often extra-textual and appeal to people who are not literate or who do
not engage with bhakti through literacy. Just as the public sphere requires literacy, the
publics of bhakti in South Asia require “embodiment,” the human as medium. This
very useful notion of “embodiment” does not simply exist as a trope of literature, but is
deeply engaged in the performance of the discourse of bhakti. By “discourse” I mean
the manifestations of bhakti not only in performance through song or literacy, but also
through all those actions and bodily displays that make up bhakti in the broadest sense,
such as those outlined above: pilgrimage, p¨jå, darçan, the wearing of signs on the
body, and so on. Embodiment, then, is not so much a technique of bhakti as its very
epicenter: bhakti needs bodies. In other words, bhakti needs the medium of the living
human or the remembered bhakta in hagiography, and the ways in which bodies are
objects of public display hardly need rehearsing here (for example, Berlant and Warner
1998; Foucault 1995; Zito and Barlow 1994). There is then almost a symbiotic equation between bhakti and performance. At this confluence, a public is necessarily
262
260 / Christian Lee Novetzke
created.
A bhakti public might have its precedent in what we might call the anthropological
etymology of bhakti—“anthropological” because it involves some of the old-fashioned
markers of community beloved of the anthropologist, like commensality. The sharing
of food constitutes a significant form of bhakti, both in its earliest articulations in
Sanskrit and throughout its literary and practical genealogy in India. We see it in the at
least partial breaking of the taboos of caste commensality during pilgrimage and in the
hagiographical and religious invocations of this act as a marker of the particular ethics
of bhakti.11 We also see this in the act of exchanging an item, often food, with a deity,
since the recipients of such food typically belong to a number of otherwise welldefined commensal groups. Prasåd, as it is called, is then often redistributed to friends
and family. In this way bhakti shows a parallel with older Sanskritic uses of the word
in the context of the preparation and distribution of food, which is one of the key
anthropological markers of community, alongside marriage (Monier-Williams 1993:
743). As early as the ¸g Veda, the verbal root bhaj and its associated nouns, such as
bhakti and bhakta, indicated the sharing, serving, and distributing of something,
whereas in the Manusm®ti and Mahåbhårata, the noun bhakta indicated prepared or
cooked food (Monier-Williams 1993: 743). På~ini uses the compound word bhaktakasa to indicate “a dish of food,” and many other compound words formed around
bhakta relate to food preparation and distribution (Monier-Williams 1993: 743). One
possible translation for bhakti, then, is “commensality,” the sharing of comestibles by a
community as a way of marking their kinship. In this way, bhakti as the exchange of
food becomes a metonym when applied to religious contexts. It takes on the meaning
of sharing food, as a symbol of other kinds of social circulation, with both deity and
fellow devotees. Here the comestible is “devotion” itself.
The word “public” often also marks places where the common good is situated.
Unlike the term “popular,” which makes a utilitarian appeal to a majority, or the word
“communal,” where the individual is subordinated to the whole, the idea of a public
implies a measure of resistance to homogeneous social entities that cause the erasure of
the individual. Note that I distinguish here between “a public” and “the public.” The
latter designation, “the public,” implies homogeneity, often enforced or primarily
influenced by the state. This is the public measured by pollsters and appealed to by
politicians. It is also the public invoked by scholars of phenomena such as “the public
sphere” and “public opinion.” This notion of the public is also usually associated
almost exclusively with modernity.12 In contrast, “a public” implies a much greater
flexibility of social organization. Warner (2005: Chapter 2) outlines a handful of
characteristics that help us identify “a public” when not qualified by its state-centered,
hegemonic sense as “the public.” Warner conditions a public with these principles:
publics (a) are self-organized; (b) exhibit a “relation among strangers”; (c) are both
personal and impersonal in their address to an audience; (d) require “mere attention”;
(e) construct a “social space…by the reflexive circulation of discourse”; (f) “act
historically,” which is to say they address the issues of their time—they historicize
themselves; and (g) enact a project of “poetic world-making,” which is to say that
Bhakti and Its Public / 263
261
publics give character to themselves and their participants, a character that is often then
embodied in signs of dress and bodily display that recall the poetics of a public.
There is much in Warner’s general rubrics that one can recognize when one thinks of
bhakti as a public. We certainly have a long tradition of the poiesis of bhakti, much of
which attempts to create “families” out of strangers, especially those made strangers by
differences in caste, class, language, and religion. The circulation of a shared discourse
that connects otherwise disparate bhakti traditions—such as the inter-relationship of
the hagiographies of KabDr and Nåmdev or the multiple emplotments of bhakti on the
imagined map of pre-modern India—is also readily apparent. And the shared social
space implied by these sorts of circulations is substantiated in pilgrimage and other
religious gatherings where strangers make common cause and identify their commonality by means of the language of bhakti. On another front—the sixth of Warner’s
principles—many studies have shown how bhakti texts and practices invest themselves
in their historical moments. The challenges faced by figures like Nåmdev, Nånak,
and KabDr at the hands of those who hold temporal authority immediately historicize
issues of persecution and faith that would have remained current concerns after the
purported lives of the sants. Indeed, the tendency to “act historically” is one of the key
“pathologies” of texts such as these. Scholars who address them must grapple with
long traditions of emendation and intercession upon the body of a bhakti text as it
passes through time, as it continues to locate itself in reference to a changing stream of
historical contexts (for more on this, see Callewaert and Lath 1989; Hawley 1984;
Novetzke 2008). Warner provides a useful, but not comprehensive, set of standards
that help us see the way bhakti can function as a public.
Bhakti traditions emphasize the social ethics that accompany the act of generating a
public. We have seen this through the notion of sharing food, places, and time in
religiously defined collectivities—performative moments, pilgrimage, the exchange of
prasåd, or the act of p¨jå. Sometimes the creation of public spaces, especially temples,
exemplifies this concern. Yet we also see the opposite tendency: to close temples to
other religions or caste communities, or to refuse to eat with a member of another
community or caste, even when engaged in a moment of bhakti—such as on a pilgrimage. The creation of shared publics is also always a creation of differences between
different publics.13 As Warner has suggested, publics represent a communication
among strangers, but this communication assumes a shared identity that can be differentiated from whatever is perceived about those who lack whatever is required to join
the quality of the public in question.14 One sees the bifurcation of large social groups
into communities designated Hindu and Muslim or Vai‚~ava and Çaiva. The latter pair
describes one of the key axes of difference that militates against the portrayal of bhakti
as a unified “movement.”15 A comprehensive history of the differences assumed or
overlooked within the so-called “bhakti movement” is a narrative awaiting an author.
Consider, for example, the history of divisiveness in the literature and politics of South
India during the formative period of Tamil bhakti among nåyaŒårs, for example,
where Jain, Buddhist, Vai‚~ava, and Çaiva all stand at odds with one another. Or
consider the long antagonism in Marathi religious history between the VårkarDs and
264 / Christian Lee Novetzke
262
the Mahånubhåvas. Both revere K®‚~a in localized forms, and both devote ample
literary energy to mutual ridicule. Also consider the history of the Kumbha Melå for
examples of centuries of regularized bloodshed among Çaiva mendicants (see Pinch
1996, 2006).
Perhaps the most pernicious of these divisive uses of the bhakti public occurs when
“Bråhma~” and “non-Bråhma~” are made to stand for social opposites. As one can see
at various moments in the history and literary remnant of the VårkarE tradition in
Marathi, for example, a clear anti-Bråhma~ sentiment occurs in the works attributed
to figures like Nåmdev (fourteenth century) and Tukåråm (seventeenth century), and
this sits alongside periods of upper-caste discrimination, such as one sees in the many
centuries in which the Vi††hal temple in Pandhapur, under the management of a
particular Bråhma~ community, refused entry to the lowest castes. Yet even among
VårkarE sants of Bråhma~ical origin, the power of their voices in the creation of
publics runs counter to exclusivity and caste elitism. Jñåndev (thirteenth century),
though a Bråhma~ by birth, is remembered by some as having wrested that great
articulation of bhakti, the Bhagavad Gtå, from the confines of Sanskrit, rendering it
for the first time into a regional language meant for the consumption of the “lowly”
who otherwise would not have access to the text without Bråhma~ical mediation.16
Eknåth (sixteenth century), also a Bråhma~, is remembered to have been associated
with a ͨfE order of the Deccan, and his remembrance in Marathi public culture
features his engagement with the life of the non-elite, whose metaphors and performative cadences of labor appear in his speech. We can also see the formation of a new
bhakti public around ÇivåjE, exemplified in the violence and censorship that followed
the release of James Laine’s book in India in 2003 (for more on this, see Novetzke
2004). A new religion has been proposed by those who claim ÇivåjE as a champion of
the lower castes, a religion in which they refuse the participation of Bråhma~s and
draw deeply from a comingling of Vai‚~ava Marathi bhakti, Marå†hå militancy, and
anti-Bråhma~ical rhetoric.
Thus, publics in the context of bhakti are both created and opposed, they both unify
and divide. Initially this fact seems to disallow the possibility of a single social
movement configured around the idea of bhakti in India. Overall, the literature of
bhakti defines its publics as inclusive, contrasting them to associations that form along
lines of class and caste, which are their primary “others.” Yet exclusivity is also
expressed through the media of bhakti—in practice if not in theory. The idea of a
public allows us to engage both the inclusive and the exclusive assertions of the texts
and practices associated with bhakti and allows for social effects of both kinds.
Let me give an example of what I mean here from my work on Nåmdev. The need
for observers and participants is perhaps most apparent when Nåmdev enacts any
kind of performance, primarily krtan, which he does with regularity both in songs
attributed to him and in his received biography (see Novetzke 2003). Throughout
Nåmdev’s songs in Marathi, one finds the invocation of the public, or loka, a noun
almost always indicating the plural, the receptive public at large. The loka are often
the addressees of Nåmdev’s verses, and within songs he calls upon them to listen or
Bhakti and Its Public / 265
263
witness. This is not a theorized social entity, but rather a surrounding social force
with which Nåmdev’s songs and biography interact and take shape. His audience
becomes transformed into the loka, the public that receives his legacy long after he is
said to have died. Here we find the second meaning of loka in Marathi, indicating
“the world,” but a human world, not a physical one. This second meaning of loka
depends on the first to impart its sense of a human field of reception and interaction
and one unmarked by caste, class, or gender. What makes a loka in Nåmdev’s verses
is the human world alert to its own inward, referential gaze. We find the word loka
designating either the first person plural, as in “We people do this,” or the third
person plural: “That is the practice of those people.” The public as expressed through
this term creates both self and other. The loka form the audience of bhakti, both in
Nåmdev’s songs and in Marathi scholarship about Nåmdev; without this audience,
Nåmdev’s bhakti would remain silent, unrecalled, and lost to time. Bhakti without an
audience, without a loka, has no meaning.
A famous passage often used to evoke the seamless unity of bhakti over time and
place personifies bhakti as a woman. She grows old and withers as she moves northward, but she also blossoms anew as she follows that same course.17 The connotation
of gender here is particularly interesting: as the old cliché of anthropology suggests,
women preserve culture, and they do so within the “private” realm of public culture;
we might notice, without taking it too seriously, that bhakti is a feminine noun.
Female sex and sexuality have long been sites for the inscription of patriarchy, as is
well known, and this means both the homogenization of what it is to be female universally and the absolute heterogeneity of each woman, as marked in relation to some
specific patriarchal system. Something similar is perhaps implied in this gendered view
of the career of bhakti: a transmission over time and space that could be claimed
universally but marked with the minutiae of place and time in any given context. In
this sense, bhakti never achieves a hegemonic status in any place or time, as do categories like dharma, but rather permeates and inflects regional variations on the theme
of constructing publics of worship. Likewise in the nineteenth and twentieth centuries,
when bhakti enters the Indian public sphere (in the Habermasian sense), it comes to
stand for unity, for both “love of the nation” and, configured as “movement,” as a
social force that has unified a religiously striated culture. Yet it is used by the hegemonic discourse of “nation” and in the service of a nascent state, not as a historically
sound interpretation of a social movement on its own. As in the past, bhakti in the postcolonial period is often marshaled to form publics, an act that narrowly misses crossing
the boundary of social secularism, a mandate of the Indian constitutional state.
Similarly, in contemporary scholarship, we can see social forces shaping our
conceptions of bhakti. Is it a coincidence that in the context of colonialism and early
modernity, bhakti would be described as a “Protestant” movement? Or that so many
scholars who entered their professions as members of highly politicized university
environments of the 1960s and 1970s should find in bhakti a call to “social protest”?
Or that some members of the Subaltern Studies Collective, writing in the context of the
great post-modern moment and the rejuvenation of a Neo-Marxism in scholarship,
266
264 / Christian Lee Novetzke
have seen bhakti through the lens of old-school Marxism, where religion will always
be exploited by the elite to subjugate the powerless, but religion will therefore also
regularly form the content of the language that the subaltern throws back at the elite
(see Novetzke 2006)? Is it a coincidence that I should see as fruitful an engagement
with key words of contemporary cultural theory, such as “public culture” or “memory”
or “subaltern,” and see in these theoretical terms a convincing explanation for the force
of an ancient pan-Indian term such as bhakti?
Anachronisms and interpolations have long bedeviled the philological study of
religion, so much a part of the presence of bhakti in academic and popular discourse.
Some observers might feel that ascribing to bhakti the status of a modern social
movement or the character of a public-producing social phenomenon is just one
more shift away from the ur-text, from what bhakti “really is.” But my concern here
is not with origins but with process, not historical fact but discourse (communication
and power conjoined), not any material product (a single text or practice) but the
motivation of bhakti as a social act. Even when one takes this view, however, bhakti
emerges as being no more a social movement in India than love is a social movement
in America and Europe. Instead, bhakti serves as a subject of a complex series of
performances and mutual interactions that divides and unites, crosses all sorts of
borders and time periods, and remains fully intact only to the extent that its own finite,
largescale conception allows it to ride above particular social historical instantiations.
Bhakti is a locus for the creation of publics, not the formation of a single social or
literary movement. But publics can, in turn, portray bhakti in any way that serves the
constituency of a particular public: as a social movement, as a personal communication
with God, or as a Protestant revolution, a nationalist focal point, or a system of social
protest. The genealogy that constructs bhakti as a “social movement” is the genealogy
of one of the many publics produced by bhakti over millennia. It has its roots in
colonialism, Orientalist scholarship, and the Indian independence movement, on the
one hand, and the rise of sociology and structural analysis of cultures in EuroAmerican scholarship of the late nineteenth and twentieth centuries, on the other. But
the “bhakti movement” is not a fabrication of the colonial and post-colonial period.
The bhakti-as-woman metaphor cited above shows us that possibly as early as the
sixteenth century, bhakti was understood in a cosmopolitan, transregional way in India,
as a means of absorbing differences caused by regional variation.
The project of challenging the common scholarly understanding of bhakti as a
“movement” is important, not just for the way it can supply historically accurate
portrayals of India’s religious and literary past, but also for the way it illuminates the
motivations that lie behind our current scholarly conventions. Several scholars have
engaged the fallacy of a “bhakti movement” in print: Sharma (1987) is perhaps the
most forceful, albeit polemical, and Prentiss (2001) is the most nuanced. Both scholars break down the historical construction of bhakti as a simple social movement,
noticing how it actually serves to unite two broad, discursive projects that are substantially different. The first was to establish the similarity between the challenge
that regional bhakti traditions posed to cosmopolitan Sanskrit culture, on the one
Bhakti and Its Public / 267
265
hand, and the challenge of regional Protestant traditions to the Roman Catholic
Church, on the other. Colonial scholars first promoted this parallel, but it served the
purposes of Indian social and political activists as well. The second project noticed
by Sharma and Prentiss also sought connections between Hindu bhakti and the
Protestant Reformation, but was far more specifically an effort by Indian nationalists
of the late nineteenth and early twentieth centuries to use bhakti as a paradigm for a
social movement whose historical fulfillment was the national, transregional identity
of India, unified under the banner of “Saints of the Nation.” In both cases, the
Orientalist trace of scholarship so familiar from Edward Said and his contemporary
scholarly genealogy tracks the way an “academic” idea (one pairing missionary
concerns with the production of early colonial knowledge) becomes a bedrock for
an Indian “indigenous” idea shared by many of the “neo-Hindu” leaders. The irony
of the transmission of knowledge into shifting spheres of power is perfectly natural
in the colonial context, as many post-colonial scholars have endeavored to show.
The relationship of knowledge to power is a variable that can be manipulated.
Mountstuart Elphinstone seemed aware of this very specific threat to Empire when
in October 1819 he wrote a personal letter of concern to Captain Francis Irvine of
the Eleventh Regiment of the Bengal Native Infantry in Calcutta, concluding:
I have often considered the state of our Empire in India & it has always appeared
to me that…it would probably stand for a long time unless chance should raise up
some false prophet who should unite a plan for the reformation of the existing
religion with one for the deliverance of the country from foreigners (cited in
Ballhatchet 1957: 248–49).
One might hear in this the veiled threat of a figure like Rammohan Roy (1772–
1833), who, though a darling of the British Empire, provided the archetype for the
devout, yet thoroughly modern and British, Hindu subject. It was, after all, Roy who
first used that enigmatic term “Sanatåna Dharma” in a politico-religious context,
stripping it of its very particular meanings in texts such as the Mahåbhårata, the
Bhagavad Gtå, and the Manusm®ti (see Halbfass 1988). Elphinstone was likewise
forecasting generations of leaders, from SvåmE Dayånanda SarasvatE and SvåmE
Vivekånanda to Aurobindo Ghose and from Mohandas K. Gandhi to Vinayak
Damodar Savarkar and Keshav Baliram Hedgewar. They would all display the
characteristics of education and nativistic pride that Elphinstone attributed to his
portentous “false prophet.” In all these figures one sees the imbrication of scholarship and politics, and many drew from the body of modern scholarly literature that
had wedded bhakti to a social movement, meant to provide emancipation either from
perceived Bråhma~ical hegemony or from economic and political colonialism. The
project initiated here by Jack Hawley to interrogate claims made about bhakti is vital
because it recalls that scholarship is not simply “academic” but has some effect in
the world. Today the idea of a bhakti movement and of all the bhakti figures it
encompasses are used by actors across the political spectrum in India, Left to Right.
266 / Christian Lee Novetzke
268
It is present in the homes and personal lives of Hindus, Sikhs, and Jains, as in the
public lives of voters, activists, and sevaks. For all the good that the invocation of
bhakti in the Indian public sphere can do—as it assisted in the nationalist endeavor
to end the exploitation of colonialism—it can also marshal the “faithful” to commit
atrocities against others, as in Bombay in 1992–93 and in Gujarat in 2002. Here, as
we have seen earlier, bhakti is again the empty vessel into which both nectar and
poison can be poured. The power of bhakti as a concept for the formulation of
publics is precisely this dual ability: both to be exceedingly specific, targeting the
community, ethnicity, sect, nation, polity, or even individual; yet also to be exceptionally general, uniting and even co-opting through spreading outward from its
nebulous, never-defined core.
The question, “the bhakti movement—says who?” has its answer in these and other
discourses where we can see who indeed “says” there is a bhakti movement at a
particular time and for an identifiable purpose. People write, speak, perform, and
represent such a movement at specific points in time and space and thus make it real in
a social sense, but not in a transcendent one. At present, for example, a major lacuna in
the rich scholarship on bhakti is the presence of Islam, specifically of Sufism, a lacuna
that may be a product of the narrowing of historical vision that results from seeing
“Hinduism” as a religion and bhakti as a movement. Instead, one could talk of
localized, specific publics of bhakti that are as much “Hindu” as they are “Muslim”—
or indeed neither. Similarly, much work remains to be done on historical and contemporary (primarily political) uses of bhakti to set caste communities against each other
along the axes of “high” and “low” or “Bråhma~ical” and “Dalit” or “OBC.” Indeed,
rather than Hindu-Muslim antagonism, the primary locus of public political tension
across India for decades to come will be caste, and into this cauldron we can expect
expressions of bhakti as caste difference to be poured. When we talk of a “bhakti
movement” we are identifying—perhaps even representing—one of many publics that
have received this ancient word in a particular way, a public of which we, as scholars,
form an important part in the present. My point in this essay has not been to prove or
disprove this question of whether bhakti is a “movement,” but to attend to the “who” in
Hawley’s question. I have suggested that the multiple manifestations of bhakti in South
Asian history are predicated on asking the same questions: who says? And, just as
importantly, who listens?
Notes
1. My thanks to Jack Hawley, Sunila S. Kale, and Whitney Cox.
2. The orthodox theories of rasa debate but exclude bhakti as a rasa, but certain
Vai‚~ava theologians argue that bhakti is a rasa, particularly within the GauFya
Vai‚~ava tradition. In the Marathi songs attributed to Nåmdev, he often refers to the
rasa of bhakti, but this is of course in a very non-technical way.
3. In KabFr’s case the punishments also include being bound in chains and thrown
into a river and being set on fire!
Bhakti and Its Public / 269
267
4. In this section I draw on Novetzke (1999).
5. At the time of writing, no good study exists of MahJpati’s work, though Justin
Abbott’s translations are available in English.
6. Compare Pollock (1998), where he argues that royal and courtly discourse in
regional languages preceded religious discourse for the most part.
7. Dirks has explored the relationship between caste, colonialism, and public
displays of status in The Hollow Crown (1987) and more recently has studied the
reinscription of caste in Indian public culture through the hegemonic ordering of
social forms in the colonial period in Castes of Mind (2001).
8. Appadurai has explored publics or public culture in a number of venues, for
example, Worship and Conflict Under Colonial Rule (1981) and Modernity at Large
(1996).
9. In addition to their numerous independent publications, Appadurai and
Breckenridge founded the journal Public Culture in 1988.
10. What one might call a competing tradition of the study of public culture exists
in Australian cultural studies, inaugurated by the work of Horne (1986).
11. Caste commensality is as much maintained as transgressed in pilgrimage
settings, but I have yet to encounter a bhakti text that lauds caste commensality as a
virtue of bhakti. See Karve (1962).
12. This is almost universally true, even in influential studies of publics, especially
public culture and the public sphere, in South Asia, such as those conducted by
Appadurai and Breckenridge, Freitag, Hardgrove, Haynes, Orsini, and others (see
Yandell and Paul 2000). Warner, while not discussing South Asia, is equivocal on
the historical emergence of publics. Sometimes he attributes them to modernity, and
at other times he professes ignorance about the issue in non-Western, non-modern
contexts. In my reading, there is nothing about “a public” in Warner’s formulation
that fixes the phenomenon in modernity, nor aligns it necessarily with notions of
public culture shared by the authors cited above. By contrast, the public sphere, if
one follows Habermas, is by its definition a modern phenomenon, bounded by the
same conditions as the idea of the nation or of history, for example.
13. When this difference takes the form of a public in opposition to “the public,”
Warner (2005: Chapter 2) and others have described this as a “counterpublic,”
though I do not find this term nearly as helpful as his ideas about publics in general.
14. In America, for example, the “voting public” is also divided into the publics
that represent “conservative” and “liberal” voters. Likewise media are divided. The
conservative public reads The National Review, while the liberal public reads The
Nation; the conservative public watches Fox News; the liberal public, The News
Hour, and so on.
15. Many more examples can be found within very specific traditions. See, for
example, Carman’s (2000) essay on schisms among ÇrJ Vai‚~avas.
16. Here I am referring to the Bhåvårthad'pikå attributed to Jñåndev, who is
also know as “Jñåneçvar” and hence the text is in Marathi popularly called the
Jñåneçvar'.
268 / Christian Lee Novetzke
270
17. See the Bhågavata Måhåtmya (seventeenth century?), a text also attributed to
the Padma Purå~a.
References Cited
Appadurai, Arjun. 1981. Worship and Conflict Under Colonial Rule. New York:
Cambridge University Press.
Appadurai, Arjun. 1996. Modernity at Large. Minneapolis: University of Minnesota
Press.
Ballhatchet, Kenneth. 1957. Social Policy and Social Change in Western India,
1817–1830. New York: Oxford University Press.
Berger, Peter L. 1967. The Sacred Canopy. New York: Doubleday.
Berlant, Lauren and Michael Warner. 1998. “Sex in Public.” Critical Inquiry 24, 2:
547–66.
Bourdieu, Pierre. 1993. The Field of Cultural Production (ed. Randal Johnson). New
York: Columbia University Press.
Brass, Paul R. 1974. Language, Religion and Politics in North India. Cambridge:
Cambridge University Press.
Brass, Paul R. 2003. The Production of Hindu-Muslim Violence in Contemporary
India. Seattle: University of Washington Press.
Breckenridge, Carol A. 1995. Consuming Modernity: Public Culture in a South
Asian World. Minneapolis: University of Minnesota Press.
Callewaert, Winand M. 1978. The Sarvangi of the Dadupanthi Rajab. Lueven:
Orientalia Lovaniensia Analecta 4.
Callewaert, Winand M. and Mukund Lath. 1989. The Hindi Padåval9 of Nåmdev.
Leuven: Departement Orientalistiek, Katholieke Universiteit Leuven.
Callewaert, Winand M. and Swapna Sharma. 2000. The Hagiographies of Anantadås: The Bhakti Poets of North India. Richmond: Curzon Press.
Carman, John B. 2000. “Dissolving One Paradox and Discovering Another: Pillan’s
Interpretation of Nammalvar’s Poem.” In Keith E. Yandell and John. J. Paul, eds.,
Religion and Public Culture: Encounters and Identities in Modern South India,
149–61. Surrey: Curzon Press.
Cohn, Bernard S. 1996. Colonialism and Its Forms of Knowledge. Princeton:
Princeton University Press.
Cutler, Norman. 1984. “The Devotee’s Experiences of the Sacred Tamil Hymns.”
History of Religions 24, 2: 91–112.
Dirks, Nicholas B. 1987. The Hollow Crown: Ethnohistory of an Indian Kingdom.
Ann Arbor: University of Michigan Press.
Dirks, Nicholas B. 2001. Castes of Mind. Princeton: Princeton University Press.
Dewey, John. 1927. The Public and Its Problems. New York: H. Holt.
Feldhaus, Anne. 1995. Water and Womanhood: Religious Meanings of Rivers in
Maharashtra. New York: Oxford University Press.
Foucault, Michel. 1995 [1975]. Discipline and Punish: The Birth of the Prison. New
Bhakti and Its Public / 271
269
York: Vintage Books.
Freitag, Sandria B. 1989. Collective Action and Community: Public Arenas and the
Emergence of Communalism in North India. Berkeley: University of California
Press.
Geertz, Clifford. 1973. The Interpretation of Cultures. New York: Basic Books.
Habermas, Jürgen. 1991 [1962]. The Structural Transformation of the Public
Sphere: An Inquiry into a Category of Bourgeois Society (trans. Thomas Burger,
with Frederick Lawrence). Cambridge: Massachusetts Institute of Technology
Press.
Halbfass, Wilhelm. 1988 [1981]. India and Europe: An Essay in Philosophical
Understanding. New Delhi: Motilal Banarsidass.
Hardgrove, Anne H. 2004. Community and Public Culture: The Marwaris in
Calcutta. New York: Columbia University Press.
Hardy, Friedhelm. 1983. Viraha Bhakti: The Early History of K®‚~a Devotion in
South India. New York: Oxford University Press.
Haynes, Douglas. E. 1991. Rhetoric and Ritual in Colonial India: The Shaping of a
Public Culture in Surat City, 1852–1928. Berkeley: University of California Press.
Hawley, John Stratton. 1984. S¨r Dås: Poet, Singer, Saint. Seattle: University of
Washington Press.
Horne, Donald. 1986. The Public Culture: An Argument with the Future. London:
Pluto Press.
Karve, Irawati. 1962. “On the Road: A Maharashtrian Pilgrimage” (trans. D. D.
Karve and Franklin Southworth). Journal of Asian Studies 22, 1: 13–29.
Kant, Immanuel. 1995 [1784]. Foundations of the Metaphysics of Morals: And,
What is Enlightenment? (trans. Lewis White Beck). Upper Saddle River: Prentice
Hall.
Lorenzen, David N. 1991. Kabir Legends and Ananta-das’s Kabir Parachai: With a
Translation of the Kabir Parachai Prepared in Collaboration with Jagdish Kumar
and Uma Thukral and with an Edition of the Niranjani Panthi Recension of this
Work. Albany: State University of New York Press.
Monier-Williams, Monier. 1993 [1851]. A Dictionary: English and Sanskrit.
London: W. H. Allen.
Novetzke, Christian Lee. 1999. “Bhakti.” In Wendy Doniger, ed., Encyclopedia of
World Religions, 127–28. Springfield: Merriam-Webster.
Novetzke, Christian Lee. 2003. “Divining an Author: The Idea of Authorship in an
Indian Religious Tradition.” History of Religions 42, 3: 213–42.
Novetzke, Christian Lee. 2004. “The Laine Controversy and the Study of
Hinduism.” International Journal of Hindu Studies 8, 1–3: 183–201.
Novetzke, Christian Lee. 2006. “The Subaltern Numen: Making History in the Name
of God.” History of Religions 46, 2: 99–126.
Novetzke, Christian Lee. 2008. Religion and Public Memory. New York: Columbia
University Press.
Orsini, Francesca. 2002. The Hindi Public Sphere 1920–1940: Language and
272
270 / Christian Lee Novetzke
Literature in the Age of Nationalism. New York: Oxford University Press.
Pinch, William R. 1996. Peasants and Monks in British India. Berkeley: University
of California Press.
Pinch, William R. 2006. Warrior Ascetics and Indian Empires. Cambridge:
Cambridge University Press.
Pollock, Sheldon. 1998. “The Cosmopolitan Vernacular.” Journal of Asian Studies
57, 1: 6–37.
Prentiss, Karen. 1999. The Embodiment of Bhakti. New York: Oxford University
Press.
Sharma, Krishna. 1987. Bhakti and the Bhakti Movement: A New Perspective. New
Delhi: Munshriram Manoharlal.
Varshney, Ashutosh. 2003. Ethnic Conflict and Civic Life: Hindus and Muslims in
India. New Haven: Yale University Press.
Wakankar, Milind. 2003. “The Prehistory of the Popular: Caste and Canonicity in
Indian Modernity.” Ph.D. dissertation. New York: Columbia University Library.
Warner, Michael. 2005 [2002]. Publics and Counterpublics. New York: Zone
Books.
Yandell, Keith E. and John J. Paul, eds. 2000. Religion and Public Culture:
Encounters and Identities in Modern South India. Surrey: Curzon Press.
Zito, Angela and Tani E. Barlow. 1994. Body, Subject, & Power in China. Chicago:
University of Chicago Press.
CHRISTIAN LEE NOVETZKE is Assistant Professor in the South Asia Program
and the Comparative Religion Program of the Jackson School of International
Studies at the University of Washington, Seattle. <novetzke@u.washington.edu>