Long before I became a committed academic, long before I was a college professor teaching dance history, long before terminal degrees and professional titles, I chanced upon an exhibition of early dance photographs at the Rodin Museum in Paris. I bought the small catalogue, and from time to time I would page through the striking black and white images searching for dancing inspiration. I always paused at a certain one of Loïe Fuller. There she is, radiant in the sunlight of Rodin's garden, chest open, arms spread like great wings, running full force towards the camera. It is an image of a strong, mature woman, one who exudes a joyful, yet earthy energy. A copy of this photograph taken in 1900 by Eugène Druet currently hangs above my desk.

With a nod to the meanings embedded in historical study, Walter Benjamin once wrote: “To dwell means to leave traces” (1999, 9). Indeed, traces are the material artifacts that constitute the stuff of historical inquiry, the bits and pieces of a life that scholars follow, gather up, and survey. The word itself suggests the actual imprint of a figure who has passed, the footprint, mark or impression of a person or event. These kinds of traces are omnipresent in the case of Loie Fuller. Some traces are more visible than others, some more easily located. But all traces—once noticed—draw us into another reality.

The promise of critical liberation that postcolonial and transnational perspectives offer by urging us to think the complex imbrication of the global in the local remains an unfulfilled promise in South Asian dance scholarship. I will elaborate this point by describing the global thrust of Rukmini Devi's art and education movement, which could not be recuperated within the territorializing intellectual framework of Indian nationalism, and explain why she, in fact, manifests herself as a discursive failure in standard scholarly accounts of Bharatanatyam in the United States.

Part I: Bharatanatyam manifests itself as a world form today, quite like Ballet, albeit with a different genealogy. It is researched in western academic institutions in the United Kingdom, Europe, the United States, Canada, Australia, as well as in India, Sri Lanka, Singapore, and Malaysia. Practitioners and dance scholars organize international conferences focused on Bharatanatyam all over the world today, and the Dance Studies Program at Roehampton, supported by a grant from the Arts and Humanities Research Board (AHRB), has assumed a leadership role in providing a new, global profile for the dance.

Ballipadu is a small village in the West Godavari district in Andhra Pradesh, South India. One of the most prominent markers in the village landscape is the Madanagopālasvāmi temple (Fig. 1). Pemmaraju Konayamatya constructed the temple sometime in the late eighteenth century (exact dates unknown). The central image in the temple is that of Kṛṣṇa as Madanagopālasvāmi, The Cowherd Beautiful as the God of Desire (Fig. 2). From the time of its consecration, the Ballipadu Madanagopālasvāmi temple had supported over fifteen devadāsīs (women artists who served in temples and/or courts) by providing them with tax-free land, money, and a platform to present their art. In return, some of the women performed ritual tasks such as the waving of the fiveflames (pañca-hārati) and the pot lamp (kumbha-hārati), as ritual duties in the temple. In 1948, at the end of a long struggle with the discursive contours of social reform, five women—Manikyam, Anusuya, Varahalu, Seshachalam and Maithili—were expelled from the temple. These five women from Ballipadu subsequently moved to the small town of Duwa, where they maintain a small matrifocal home, adhering to their traditional patterns of kinship as devadāsīs. They do not ostensibly retain any markers of devadāsī identity: they no longer sing or dance in public, have any ritual duties in temples, receive tax-free land (mãnyãlu) from temples or feudal kingdoms (zamīndāri samasthānas), and bless homes during auspicious occasions. Yet, they insist on referring to themselves as devadāsīs, despite the extreme social stigma attached to this identity.

BDG: Seán, I want to start with something that you said yesterday about never having been able to do Bill's [Bill T. Jones's] movement the way he did it—or was it the way he wanted it done? Could you talk about that?

SC: The way he did it. Bill has a very inventive, deeply personal, and unique way of moving, perhaps because he didn't come up through the sort of modern dance training sought by many African-American dancers. People in college told Bill that he should go to New York to be “finished” by Alvin Ailey and he really did not have an interest in that. Bill studied dance with Percival Borde and contact improvisation with Lois Welk and was a track star in high school and college. He did a lot of musical theater in high school with an English teacher he loved very much. Bill was about dancing his own way.

Arnold Genthe's 1915 photograph of Anna Pavlova, taken as she leaps into the air, is perhaps the earliest photograph of free movement in dance (Fig. 1). Unlike many other early images, with long exposure times necessitating static poses or wires to hold up the dancers, this photograph depicts actual movement. This claim to authenticity and actuality is a powerful part of its appeal; looking at the image, viewers are sure that they are witnesses to a faithful reproduction of Pavlova dancing, that they are seeing the dance of the past. Considered in this manner, the photograph is an example of the revelatory power of the camera to show us what has been.

However, Genthe's photograph is not a powerful image simply because it is, authentically, of a dancer in motion. It might have mechanically frozen its subject in time, but the photograph communicates movement beyond the moment it depicts—beyond, in a sense, what it reveals photographically to what it evokes in the mind of the viewer. Viewers are able to see movement in details indicative of motion: the flowing fabric of the costume, Pavlova's bodily posture with raised and powerfully muscled thigh, the elevated arm gestures, and the sharply bent and thrusting toes. Additionally, the degree of blur in the photograph provides an indistinctness that is suggestive of something in motion; oddly, the partial obscurity of the picture prompts viewers to imagine more than they can see. All of these elements are evocative indications of movement; they are neither documentary nor part of what can be called photographic revelation, but are instead representational.

As Bharatanatyam dancers across the world talk about what they do through listservs, websites, and performance publicity—in academia's world dance courses and amidst international cultural diplomats—I keep re-encountering the Orientalist representation of a “pan Indian transhistorical” devadāsī. Her history is a linear deterioration of aesthetic quality and personal agency, from temple to courts and from courts to streets and to (deserved) abandonment from where the dancer and the dance must be rescued (see Hanna 1993; Banerji 1983).

The lifestyle of the sadir dancers of the early twentieth century was extensively researched by Amrit Srinivasan (1979–81) and documented in her ethnographic dissertation at Cambridge University 1984 and in subsequent articles.2 The devadāsī was selected for her talent. She was highly trained in dance, texts, and music, and she performed temple rituals. Her freedom from household responsibilities (grhastya) was made possible by the largesse of a patron, and bhakti theology legitimized “both the housewife and god's wife as parallel life-possibilities” both for women and for those men who could afford to support both kinds of liaisons. Temples frequently reimbursed devadāsī-s with bourses and land donations.

Dancers today persist in maintaining that dancers had no “technique” since they danced only for God, that they knew nothing of music or theory, performed in a vulgar manner, and that contemporary dancers are much more beautiful, intelligent, and better trained.

For the past few years, critics, fans, and even Bill T. Jones himself have been talking about the artist's move from explicitly political, identity-based works to an investigation of aesthetics and pure movement. They talk about the more conventional makeup of Jones's current ensemble—the fact that Lawrence Goldhuber and Alexandra Beller are no longer in the company, dancers described in the New York Times as “imperfect” because “chubbier than the norm” (Dunning 2002). They discuss the fact that Jones rarely uses text these days and is no longer confrontational. He dances to Beethover and performs at Lincoln Center with the Chamber Music Society. In a 1997 mterview with Richard Covington, Jones explains this shift in his work by stating, “It's not quite as sexy to talk about. What was being said in those earlier works was as important as how it was being danced. Here, I'm trying to think about how it's danced first, trusting that the political, social, all those things are in our bodies literally, and in the eyes of the beholder” (Covington 1997).

The Nāṭyaśāstra is a Sanskrit work on drama, dance, and music of uncertain date, which has acquired the status of Ur-text of Indian classical dance. It was in the twentieth century, through the modernization process of bharatanatyam, that the Nāṭyaśāstra and its related texts came to be regarded as setting out the grammar of Indian dance classicism (Coorlawala 1994; Meduri 1996; O'Shea 2001).

Chapter 4 of the Nāṭyaśāstra deals specifically with dance, here distinct from acting, and describes the 108 dance units—Karaṇas—of the tāṇḍava dance of Śiva and their combinations, dance phrases named aṇgahāras (NS 1988–89, 4, w. 1–245). This particular chapter of the Nāṭyaśāstra has attracted attention ever since Naidu, Naidu, and Pantulu published its translation into English, Tānḍavalakṇaṣam, in 1936. Since then there have been other translations of the Nāṭyaśāstra, among which the two volumes by Manmohan Ghosh, in 1951 and 1956 respectively, and different interpretations of the Karaṇas of chapter 4. Readings have drawn on the iconography of the Karaṇas as found at the temples at Thanjavur, Kumbakonam, and Chidambaram in southern India. The karana reliefs on the walls of Kumbakonam and Chidambaram have been identified by inscriptions found on each karaṇa slab, that give the Sanskrit name of the karaṇa as recorded in the Nāṭyaśāstra.

In the first few moments of Parallel Passions, the program devised for Mavin Khoo Dance and toured extensively around Britain in the fall of 2003, the halflight atmosphere creates difficulties in distinguishing among dancers (Fig. 1). Bodies, genders, technical backgrounds, and ethnicity become subsumed for a few short moments to a greater purpose that is unconcerned with the trappings of identity, and the viewer begins to wonder optimistically how this will evolve. The effect is not sustained, however, and another evening of worthy crosscultural experiments in dance forms unfolds. Being momentarily surprised in this way helped to develop some interesting ideas. What remains in memory are those extraordinary moments where the project of Modern Britain seemed suddenly to have shape and be given expression. The eclectic mix of the dancers and their heritages, cultural and professional, was abandoned as a preoccupation. This allowed, albeit momentarily, a protean practice to peep through the layers of cultural ritual. The embrace of ballet and bharatanatyam cited in the promotional literature became momentarily irrelevant. But then, as if recalling the essence of the project and its obligation to teach lessons about sympathy and tolerance through the arts, a touring show financed by six different funding sources remembered itself and resorted to comparing and contrasting, later abandoning seriousness altogether for a light-hearted jazzbased club number. During a ponderous second half the dancers looked to be willing a connection that would not come, and then it was over.

A growing number of practitioners in the international community of choreographers and performers has begun to experiment with computer-assisted work linking dance and new technologies. This hardly comes as a surprise, since dance-on-film and videodance had already attracted considerable attention, at least since the 1980s. Earlier experiments, such as the astonishing films by Maya Deren, take us back to the 1940s, and today's motion capture-based animations find their historical roots in late nineteenth century motion studies in chronophotography and early cinema (Muybridge, Marey, Méliès). Furthermore, dancemakers, researchers, and teachers have used film or video as a vital means of documenting or analyzing existing choreographies. Some scholars and software programmers published tools (LabanWriter, LifeForms) that attracted attention in the field of dance notation and preservation as well as among choreographers (e.g., Merce Cunningham) who wanted to utilize the computer for the invention and visualization of new movement possibilities.

Robert Morris's dance Site (Fig. 1) premiered in February 1964 at the Surplus Dance Theater in New York City. Choreographed and performed by Morris, Site also featured the visual artist Carolee Schneemann and several sheets of four-by-eight foot plywood. Although it may seem odd to include these wooden panels among the performers, they assumed an active role in the choreography. Ironically, it was Schneemann who provided the background scenery. Nude and covered in white paint, she sat motionless throughout the performance, recreating the pose and persona of Edouard Manet's famous 1863 painting of Olympia while Morris manipulated the large wooden boards. In a graceful duet with inanimate partners, Morris spun the rectangular planes from a point on the ground, maneuvered them around his body, lifted them over his head, caressed their even form as he slowly moved his hand across one edge, and balanced the panels on his back as he moved across the stage. Not only did Morris never dance with Schneemann, he did not even seem to notice her.

In a career spanning over forty years, Robert Morris has produced theoretical articles, paintings, videos, installations, and environmental art in addition to his work in dance; nevertheless, the American artist remains best known for his Minimalist sculptures of the 1960s (Figs. 2 and 3). Like the works of his colleagues Donald Judd and Carl Andre, Morris's spare, geometrical objects of that period were three-dimensional and called attention to issues of site and artistic context. They also resisted past artistic conventions based in subjective methods of composition, expressivity, and metaphor. Morris, however, distinguished himself among this group of visual artists by the emphasis he placed on the viewer's bodily relationship with the art object, a distinction that derives directly from his unique involvement in avant-garde dance.

Today, Bollywood dance has become a term used by film professionals, amateur performers, and audiences to refer to dances choreographed to Hindi film songs. The question of what Bollywood dance is, however, remains debated in both movement and text. In classes, Bollywood dance movement varies in quality and style from song to song, instructor to instructor, and choreographer to choreographer. Costume choices, facial expressions, gestures (as interpretations of sung lyrics), “jhatkas and matkas” (percussive “emphasizing] … pelvic movement”) (Kabir 2001, 189), wrist whirls, and turns, all set to film music could be identified as essential ingredients of “Bollywood dance.” Yet, this definition only begins to skim the surface of what constitutes Indian film dance, as Hindi films make up approximately one-third out of the more than eight hundred films produced in India every year (Nandy 1995). Popular films produced by language-specific regional cinemas include song and dances that often retain a degree of regional specificity. While it is important to recognize the similarities shared by these dance traditions, it is also necessary to acknowledge the definitional slippage underlying the collo quial use of the term “Bollywood Dance.” For the sake of clarity, I limit the term “Bollywood dance” to a description of choreography inspired by Hindi films. This includes combinations taught in dance classes and performed on stages. I use “Hindi film dance” to refer to song and dance sequences contained in the films themselves.

This essay is the result of a long and tortured preoccupation stemming from my personal history of growing up in Kolkata, India, with a strong tradition of leftist cultural forums; the current escalation of religious fundamentalisms globally; and my questions about the signifying potential of performing bodies in this context. While I have tried to untangle and understand the complicated issues in this search for the secular dancing body in the context of Indian performance, I have been able to arrive only at a series of questions, which, in riotous recoiling, have constantly spun new questions and interjections. I offer my journey through these questions as considerations in thinking through one of the directions of contemporary Indian dance.

While I had been trying to understand the secular traditions in Indian dance for a while, questions around it grew especially urgent with the growing power of Hindu fundamentalism in India, and in particular with the destruction of the Babri Masjid in Ayodhya on December 6, 1992, by an angry mob of “Hindus” on the grounds that the mosque had been built by Muslim ruler Babar in the sixteenth century, supposedly after destroying a Hindu temple dedicated to Rama which preexisted in that spot. This incident, followed by the Hindutva claim that a new temple dedicated to Rama be built on that spot, sparked offa horrifying spate of communal violence. That the Babri had been one of the last vestiges of the shaqri style of architecture in India was hardly raised. But the overwhelming of cultural icons by religious ideology as mobilized by fundamentalists was indubitable.

This essay is a descriptive analysis of a 2000 encounter with Mayan “mambo” dancing in a mountain community, an encounter that challenges assumptions about prevalent notions of exoticism, identity, and cultural authenticity. Traveling in Guatemala with a group of international scholars, I witnessed a public performance of the transnational mambo by costumed Guatemalans that was not mambo, not Mayan, and not social. Male performers, in celebration of Corpus Christi, dressed as Disney-style costume characters and executed routines to merengue music while nondancing participants watched the spectacle. This contradictory display of dancing encouraged me to reflect on the impact of popular social dance and to examine the complicated meanings communicated by performers who incorporate body-based art into indigenous social and economic paradigms. The performers' unique interpretation of mambo dance within the context of a public Corpus Christi festival underscored discrepancies between institutional perceptions of the mambo and the popular reuse of dance motifs. At the same time, the performance, which used clowning as a mechanism to engage the audience, inverted the solemnity of the religious feast day.

Onstage, a nineteenth-century German girl waits for her first encounter with the Middle East, which will come courtesy of Tchaikovsky and the many years of choreographic evolution that have preceded the current version of The Nutcracker in which she is performing. The girl's name is Clara (or sometimes Marie), and she has come to a fantasy land, where a lively Spanish dance has been given in her honor. Now, there is softer, slower music with a steady, insistent rhythm and a snaking melody carried by an English horn. Dancers glide onto the stage wearing gauzy harem pants and jeweled headdresses, their faces impassive, their gait deliberate and stately. Who are they? Clara's face seems to ask, and what will they do? Certainly it will be like nothing she has ever seen before because they are dressed like people from far away, a hot climate perhaps, where no one moves quickly and different customs prevail. There is a woman who walks like a princess, with a cool, internal gaze and limbs that stretch out imperially. A consort picks her up and swirls her arched figure around as if she were in need of a breeze. Then they stand side by side, pausing as if transfixed by a greater power, their hands drifting above them with palms facing the sky. When they disappear, Clara stares after them, wondering, no doubt, where they came from and what on earth that was all about.

Barnard College, Columbia University, hosted the 2004 Barbara Stoler Miller Conference in New York City, February 20–22, 2004. Called “Contesting Pasts, Performing Futures: Nationalism, Globalization, and the performing arts in Modern South Asia,” the conference recognized the growing scholarship in the field of South Asian performing arts. Since “performing arts” as a scholastic field of inquiry arrests little attention within the complex enormity of Asian Studies conferences, and diminishingly so, South Asian performing arts occupy just as little space within performing arts conferences (Association for Theater in Higher Education, Society for Ethnomusicology, American Anthropological Association, Society of Dance History Scholars, Congress on Research in Dance, and others), the 2004 Barbara Stoler Miller Conference certainly provided one of the few opportunities open to South Asian performance artists and scholars to discuss their most contemporary research works, to demonstrate and perform, and to get to know each other.

Barbara Stoler Miller, a Milbank Professor at Barnard, was a distinguished Sanskritist with strong interests in art history. She died of cancer at age fifty-two in April 1993. Among the corpus of the Sanskrit literature translations she published is Love Song of the Dark Lord: Jayadeva's Gitagovinda (Columbia 1977), a text that has been for centuries close to the hearts of many Indian dancers, and another edited by her, Theatre of Memory: The Plays of Kalidasa (Columbia 1984), presented a topic that has earned significant popularity in Euro-American academia.

Nearly twenty years ago, I took a graduate seminar on postcolonial theory with Edward Said in the Department of Comparative Literature at Yale University. It was a heady experience: competitive grad students and young faculty members vied for his imperious (I realize the irony of the word in reference to Said!) attentions. We read novels—potent, searing, difficult novels—alongside some of the theorists of political struggle who would indelibly alter my understanding of what cultural resistance could mean. Said's recent death, which of course coincided with a period of distressing shifts in political tides, has placed pressure on those of us engaged with cultural analysis to probe even more deeply the significance of anticolonial struggle in cultural forms. As a dance scholar, I have found myself reflecting on the ways in which postcolonial theory might inform my understanding of the power of choreography to affect political change—but also on the ways in which dance can inform our readings of postcolonial theory.