Rapsodia Ibero-Indiana: Transoceanic creolization and the mando of Goa

Abstract The mando is a secular song-and-dance genre of Goa whose archival attestations began in the 1860s. It is still danced today, in staged rather than social settings. Its lyrics are in Konkani, their musical accompaniment combine European and local instruments, and its dancing follows the principles of the nineteenth-century European group dances known as quadrilles, which proliferated in extra-European settings to yield various creolized forms. Using theories of creolization, archival and field research in Goa, and an understanding of quadrille dancing as a social and memorial act, this article presents the mando as a peninsular, Indic, creolized quadrille. It thus offers the first systematic examination of the mando as a nineteenth-century social dance created through processes of creolization that linked the cultural worlds of the Indian and Atlantic Oceans—a manifestation of what early twentieth-century Goan composer Carlos Eugénio Ferreira called a ‘rapsodia Ibero-Indiana’ (‘Ibero-Indian rhapsody’). I investigate the mando's kinetic, performative, musical, and linguistic aspects, its emergence from a creolization of mentalités that commenced with the advent of Christianity in Goa, its relationship to other dances in Goa and across the Indian and Atlantic Ocean worlds, as well as the memory of inter-imperial cultural encounters it performs. I thereby argue for a new understanding of Goa through the processes of transoceanic creolization and their reverberation in the postcolonial present. While demonstrating the heuristic benefit of theories of creolization to the study of peninsular Indic culture, I bring those theories to peninsular India to develop further their standard applications.


Introduction
In , the journal O Ultramar reported from the wedding of the daughter of Senhor Antonio Maria Xavier Rodrigues of Margao, South Goa, noting that, at the ball that followed the ecclesiastical proceedings and that lasted until six in the morning, about  couples at any given time were dancing to the music of a live orchestra. 1 The dances described therein were those prevalent worldwide wherever Europeans had settled and intermingled with other peoples in the course of creating their empires. But, within the gamut of European-derived social dances, from the 'old' contradances to the newer, 'frenetic polkas' and 'vertiginous waltzes', an item stands out by dint of its specificity to Goa: the 'classic Indian mando'. What kind of a dance is this, and what can it tell us about Goa and its relationship to India on the one hand and Europe on the other? 1 Antonio Bruto da Costa, ed., O Ultramar,  May , n.p.
A N A N Y A J A H A N A R A K A B I R from the 'Dakhan'/'Deccan'/'South' and associated with the Hindu kolvont) joins developments around the mando's performance to reveal how postcolonial cultural negotiations fragment and reassemble the worldview encapsulated in Rapsodia Ibero-Indiana. In contemporary Goa, the mando is still sung socially, in homes and at parties. But its dancing has moved out of the mansions of the Catholic elite to the state-sponsored Mando Festival, which has taken place annually since  in prestigious venues of


Nehruvian vintage. The festival showcases musical teams whose ensembles of violins, guitars, and the Goan percussion instrument, the gumott, accompany singing in European tonal harmony; some teams also compete in the dance section. Male participants dress in tuxedoes and tails; their female counterparts in sarongs, blouses, and stoles. Complemented by fans and flourished handkerchiefs, these outfits compound the difficulty in fitting the This article presents the mando as a peninsular Indic quadrille-a creolized music-dance form that attests to a web of people, cultures, and commodities connecting the Atlantic and Indian Oceans through the long timeline of the Portuguese empire and its relationships with other global and regional powers. By analysing the mando through an interdisciplinary, inter-imperial, interoceanic approach, I respond to the 'plea' that concludes Sanjay Subrahmanyam's seminal argument for connected histories: 'that we not only compare from within our boxes, but spend some time and effort to transcend them, not by comparison alone, but by seeking out the at times fragile threads that connected the globe, even as the globe came to be defined as such.' 7 The mando generates a history that connects Goa Portuguesa with other parts of the world touched by Portuguese expansionism. Subrahmanyam's 'fragile threads' are here spun out by creolization as an embodied sociocultural process that stands in complex relation to the graphic record. Accordingly, my analysis draws on nineteenth-century descriptions of the dance, early twentieth-century lyrics, their instrumental accompaniment and notation, the dance steps then and now, outfits worn by mando dancers today, and the contexts for postcolonial performances of mando and descriptions thereof. Infusing the 'archive' with the sounds, beats, and moves of the 'repertoire', 8 I shed light on a creolizing process that, by involving the Indian Ocean world, challenges 'the epistemological hegemony of the Atlantic model'. 9 In turn, through the mando, I bring creolization as a theory of cultural change to the study of a part of the world to which it not often applied: peninsular India.
I contextualize the mando within longue durée multiscalar interactions across the Portuguese empire as well as within the local world of the Deccan peninsula. These interactions, I argue, triggered certain Groesbeck, '"Classical music", "folk music", and the brahmanical temple in Kerala, India', Asian Music, vol. , no. , , pp. -. creolizing processes that resulted in the mando's emergence by the nineteenth century, even as its development during the twentieth century and beyond has been triggered by postcolonial decreolizing impulses. 10 I track these creolizing and decreolizing vectors on three analytical levels: the mando sung, the mando danced, and the mando remembered. Accordingly, I first demonstrate that the mando sung to instrumental accompaniment reveals the dialectical relationship between Konkani and Portuguese lyric worlds, and between European and Indic musicological principles. Next, I analyse how the mando danced re-enacts certain 'foundational scenarios', or repeated performances of the compromises, betrayals, collaboration, and accommodation accompanying colonial encounter. 11 Finally, I examine how the mando remembered-in song, dance, and description-activates memories of creolization's violent intimacies and the ritualistic erasure of colonial society's inherent complicities. I explain the interdependence of these semiotic levels by showing how the mando benefits from the ludic, mimetic, and structural features of the wider genre: the creolized quadrille. 12 In the mando's case, these features point to its mythopoetic relationship to dekhnis (such as 'Hanv saiba') and the fast-paced dulpods that close mando performances. Drawing on the creolized quadrille suite as a semiotic system, I argue for the mando, dekhni, and dulpod as collectively dramatizing creolization in Portuguese India through a set of scenarios, personages, and emotions that, inspired by Ferreira, I call Rapsodia Ibero-Indiana. I conclude that these embodied, performed histories allow the mando to function as a negotiating tool for postcolonial Goans trying to reconcile their incorporation within India's federal structure since  with the affective pull of a post-imperial, Portuguese-speaking world.
Some methodological clarifications: I am a literary historian working on creolized dance forms as embodied memory practices. From the traumas and violence of slavery and colonialism arose the cultural matrix of the 10 On postcolonial de-creolization in a transoceanic frame, see Ananya Jahanara Kabir, 'Elmina as postcolonial space: transoceanic creolization and the fabric of memory', Interventions, doi: https://doi.org/./X... 11 Taylor, Archive and repertoire, p. ; Teotónio de Souza, Medieval Goa: a socio-economic history (Saligao: Goa , ), p. .


Black Atlantic, 13 which included dance and music genres that now enjoy global popularity. My interest in investigating their histories alongside histories of creolized social dances of the Western Indian Ocean led me to the mando, whose memorialization occurs within the kinetic and sonic remnants of the transoceanic Portuguese empire, and the frameworks of subject formation that shape Goans through their interpellation within postcolonial India since . While my philological training attracted me to the textural density of mando lyrics, an interest in embodied methodologies necessitated fieldwork conducted at staged and social settings for mando performances in contemporary Goa, especially its Mando Festival. 14 I articulate Goa's position 'between empires', 15 oceans, and temporalities, through an embodied philological approach that dredges up non-narrative histories of cultural contact in pursuit of the ephemerality of performance and the elusiveness of affect. Furthermore, I acknowledge the mando's memorializing, creolizing topography by including, within the heading for each of the article's nine sections, a phrase from a mando lyric that incarnates lexically and affectively its multilingual world.
Zaitu tempu zalo ('a long time has passed'): India, Goa, and Portuguese creolization Between the s and s, the mando was extensively showcased by Goan cultural brokers for local and wider Indian audiences. 16 Their 13 Paul Gilroy, The Black Atlantic: modernity and double consciousness (London: Verso, ). 14 The Mando Festival has been taking place annually in Goa since , under the aegis of various cultural organizations that, over the years, have been supported by government and civil-society sources. My fieldwork was conducted during its 'Golden Jubilee' year (), which included four days of mando performances at the Rabindra Bhavan, Margao, and the Kala Academy, Panjim. The festival is recorded in an annual souvenir, whose contents and paratexts (including evolving visualization of 'Goa' through the cover art) provide rich material for tracking the cultural politics of mando over half a century of Goa's political embedding within India. Apart from the festival, I witnessed informal performances of mando at the homes of Goan musicians, which featured the mando sung (not danced). 15 Rochelle Pinto, Between empires: print and politics in Goa (Delhi: Oxford University Press, ). 16 In numerous opinion pieces on the mando, including those published within Mando Festival souvenirs from  onwards, and the December  issue of the art magazine Marg dedicated to 'Goan art', with articles on the mando by Lucio Rodrigues, José writings present the mando as a quintessentially Goan product shaped by Goa's long exposure to European cultural influences as well as the tenacity of non-European, local elements, especially the Konkani language: a process succinctly captured by the quote within my section epigraph. 17 This nativist approach illuminates a particular phase in Goa's reconfiguration of identity within postcolonial India and I will analyse it as such later. But it does not shed much light on the mando's antecedents and innovations, or its similarities with (and divergences from) other dance-music genres across the Portuguese-speaking world. I attribute these gaps to the fact that 'creole' and 'creolization' are terms hardly ever applied to any aspect of India's history or culture. It is beyond the scope of this article to elaborate a case for 'creole Indias', although we will return to and reassess this concept in the conclusion. 18 Here, I provide a prolegomenon to this task by placing Goa within the space-time of Portuguese creolization, to explicate the mando as a creolized cultural product arising from this history. As an Indic contribution to the phenomenon of creolization, the mando reconfigures creolization as a historical process connecting the Indian and Atlantic Ocean worlds through transoceanic encounters between people of European, African, and Asian heritages. In turn, it imposes on considerations of 'Indian culture' two consequences of these encounters: Goa as a nodal point in the movement of elites across the Portuguese empire 19 and the movement of African and African-influenced cultural practices across the same space.
Pereira, Lourdino Barreto, and Antsher Lobo; these developments will be discussed later in this article. 17 The section epigraph is from the mando 'Sokanny'm furhem uttunum' ('arising early in the morning') by Ligório Costa  The British only established themselves as a ruling rather than a mercantile power in India after the Battle of Plassey in -two and a half centuries after the Portuguese conquest of Goa; with Indian independence in , the British were also the first European power to relinquish an imperial possession under the pressure of anti-colonialism. In contrast, although Brazil achieved independence from Portugal in , Portugal's African colonies did so only in the s, while Goa was seized from Portuguese control by India in . Right from the fifteenth century down through the twentieth, then, Portuguese cultural influence was disseminated through mercantile, religious, and juridical channels in Asia, Africa, and the Americas. This transoceanic longue durée of empire led to a distinctive history and culture of creolization. 24 If creolization indicates the creation of new demographies and cultures through the interaction (voluntary and forced) of peoples brought together in a compacted space, 25 then Cape Verde, where the Portuguese settled farmers and slaves in an inhospitable, uninhabited archipelago from the late fifteenth century onwards, became the world's first creolized society. 26 Portuguese colonialism played a key role in creating and sustaining the political, economic, and psychosocial conditions for the creolization process. 27 Consequently, a pervasive 'racial and cultural ambiguity and hybridity' became its characteristic. 28 Did Portuguese Goa's geographic contiguity to British India imbricate it within 'the early demise of Creole India' that British rule purportedly Answers lie in the 'dense and long temporality' of 'the vast, multi-secular contact zone' of Portuguese colonialism, 30 which nourished the circulation (and recreolization) of tangible products and intangible practices resulting from creolization. These spiralling flows blurred the distinction between 'metropole' and 'colonies' and resultant binaries of racialized culture. 31 Between Portugal, Cape Verde, Brazil, Angola, São Tomé and Príncipe, and Guinea Bissau, people, plants, recipes, furniture, fabric, instruments, melodies, rhythms, and modes of singing and dancing moved to and fro. 32 The study of these circum-Atlantic material and embodied circulations has deployed and deepened current understanding of creolization. 33 Research is increasingly focusing on parallel flows within the Indian Ocean: of textiles, relics, prestige objects, theological and political ideas, debt, soundscapes, and rhythms. 34 However, we await a systematic engagement with theories or evidence of creolization derived from the Indian Ocean space. 35 Such engagement could galvanize a comprehensive transoceanic history of slavery, colonialism, and modernity by turning attention to the links and divergences between Indian Ocean and Black Atlantic networks. Here, Goa plays an important role. Its music-and-dance culture, leading up to the mando, offers exciting, but underused evidence of how cultural contact, transmission, and production across the Indian Ocean world were impacted by the contact with the Atlantic world that the Portuguese empire and concomitant inter-imperial exchanges facilitated. In the two subsequent sections, accordingly, we shall see how elite cultural forms, including sacred and secular music, became the conduits of transoceanic creolization that deposited in Goa traces of European and African expressive culture. Sedimented in the mando, these traces can be recovered by reading against the grain the proscriptions, prohibitions, and prevarications that characterize the musical scores, conciliar notices, and lexicographic attestations that constitute its archive.
Flautach' toqui ('playing the flute …'): the transoceanic creolization of mentalités Benedict Anderson famously argued that 'print-capitalism' forged 'imagined communities' across supra-regional spaces in modernity. 36 In the case of the Portuguese empire, aural resources would appear to have performed this function of defining and consolidating a collective identity vastly dispersed across space and time. The Portuguese language as a source of this shared aurality is enshrined in the concept of lusofonia, now sometimes discredited for its imperial genealogy.


Recognizing the problematic nature of this formulation, but also the importance of sound as a connective force, ethnomusicologist Susana Sardo replaces it with 'lusosonia'. 38 While usefully overlaying the spoken with the sung word, her neologism nevertheless retains the imperialist overtones of lusofonia's first element. Sardo's exegesis of 'lusosonia' as the 'sound-trails of Portuguese culture' also perpetuates an insidious binary between 'Portuguese' and 'non-Portuguese'. In continuing to privilege aurality, furthermore, it leaves us none the wiser about the kinetic dimension that enfolds sound within dance, gesture, and performance; from another angle, it implies a potentially misleading contrast between the colonizer's language, disseminated through writing, and acoustic expression, transmitted through sound. In fact, it is a complex relationship between sound, text, word, and melody that makes possible this prima facie 'lusosonic' world. Its 'sound-trails' lead us to the creolization of mentalités along a labile interface between expressive culture's sonic, kinetic, and graphic dimensions. 39 The mando is a product of this deep structural transformation on epistemic and embodied levels.
The mando's archival entry coincides with Indo-Portuguese patronage of dance-music genres that were being enjoyed worldwide by the last quarter of the nineteenth century. Thanks to Goa's long interpellation within the Portuguese empire, dance styles that waxed and waned in popularity and fashion, and their accompanying music, were continually absorbed into Indo-Portuguese society. This article's opening epigraph describes a nineteenth-century Margao wedding at which guests danced polkas, waltzes, and varsoviennes (a variant of mazurka), all of central European provenance, as well as 'contradances', deriving from an earlier internationalization of European courtly dance. 40 In Goa also 38 Susana Sardo, 'Lusosonia: post-colonial cartographies on sounds and memories', https://plataforma.com/formacao/lusosonia-post-colonial-cartographies-on-sounds-andmemories.htm (accessed  November ). 39 I marry here insights from Chaudenson, Creolization; and Jacques Le Goff, 'Mentalities: a history of ambiguities', in Jacques Le Goff and Pierre Nora (eds), Constructing the past: essays in historical methodology, with an introduction by Colin Lucas (Cambridge: Cambridge University Press, ), pp. -. 40 The social satirist 'Jip' (Francisco João da Costa) included a similar assortment of dances-waltzes, quadrilles, polkas, galopes, lanceiros, and mazurkas-in his description of an elite wedding, contained within his extremely popular account of Goan Christian life; he also describes mandos being danced and sung on the day after the wedding; see Jip, Jacob e Dulce: scenas da vida indiana, rd edition (Panjim: Typografia Sadananda, ), pp. -, -. On these dances, see Peter Manuel, Creolizing contradance in circulated dance-music genres specific to the Portuguese-speaking world -fados, modinhas, choros, and maxixes, or particularly popular within it, such as polkas and mazurkas. 41 This layered kinetic awareness signals more than a minuscule elite's superficial mimicry of 'Western dances'. By at least the end of the nineteenth century, sheet music for these circum-Atlantic genres was being printed in Goa, suggesting a parity of taste and demand linking it to Brazil, Cape Verde, Angola, Mozambique, and Portugal. 42 In the early twentieth century, the afore-mentioned Carlos Eugénio Ferreira rendered Goan themes in the tempo of the waltz, maxixe, and foxtrot, 43 confirming widespread Indo-Portuguese ability to read, compose, and dance to music transmitted through European notation. 44 'Lusosonia', then, is the tip of an iceberg. Through the channels of Portuguese colonialism circulated an analytical apprehension of musicality and its kinetic expression based on European aesthetic and compositional principles, which structured the relationship between melody and rhythm very differently from their Indic counterparts.
The heuristic of creolization interprets these diverse cultural principles in dynamic interaction rather than imprisoned in binaries. 45 Indeed, cultural transformation was in process from the moment 'culture' embarked in European-manned vessels for destinations 'ultramar' ('overseas'). The arrival of Vasco da Gama on the Malabar coast with trumpeters, organists, and chanters of mass amongst his crew mobilized ab initio 'instrumental diplomacy' in the mutually enhancing cause of Iberian Christianity and Portuguese expansionism. 46 The impact of these novel sounds unfolded through an ad hoc, unpredictable combination of curiosity, resistance, and collaboration from the locals, pragmatism and experimentation on the part of the arrivals, and two-way mimicry. 47 These responses constitute the uneven calculus of creolization in the longue durée. Because Jesuit priests insisted that mass was most efficaciously delivered to Goan neophytes in polyphony, 'by the s the practice of polyphony was being cultivated in Goan churches' and taught widely to Indian boys, while 'Indian instruments were being used along with the voices and organ'. 48 The pragmatic deployment of Indic resources to realize European musicality drives the creation of novelty through creolization. But, through a deeper transculturation, 'tonal harmony' itself becomes a creolizing force. 49 The pedagogy of polyphony necessitated, from the medieval European period onwards, an internalization of the ability to read and explicate musical scores. Centuries later in Goa, this musical literacy determined not only how mandos were performed, but the epistemic basis of their conceptualization and composition. Today, mandos continue to be 46  apprehended through musical scores that communicate its compositional reliance on tonal harmony and the European time signature of /.
'Playing the flute', as my section epigraph declares, 50 Goan musicians give breath to musical scores-a melding of Indic and Latin cultures that indexes a deep creolization of mentalités. From this milieu emerged the mando, a secular Indo-Portuguese product distilled from transoceanic history. European notation as a graphic technology of transmission enabled the same sacred music reverberate across the , miles separating the Se Cathedrals of Old Goa and Bahia. 51 It rapidly disseminated, across that immense space, verse forms associated with the lyric dimension of sacred music, particularly motets and villancicos, baroque forms well attested in Goa; indeed, the mando's lyrical structure follows the motet. 52 As notation's liturgical purpose became repurposed for the composition of secular genres, musical scores became the vehicle for the latter's circulation as leisure practices wherever Europeans were establishing plantations, port towns, and comptoirs as the building blocks of empire. However, there was an inevitable gap between European notation and circum-Atlantic creolized performance. 53 In retrieving embodied practice through graphic conventions, themselves nascent and evolving at the commencement of Iberian expansionism, an important role was played by the improvisation and creative adaptability that motors creolization in both sacred and secular expressive realms. Similar tendencies impacted the transmission of European dances through instruction manuals and lyrics with mnemonic instructions that tied dances to their corresponding music. 54 The most efficient vector for the transmission of dance-the body-was also the most porous to improvisation.
Disembarking dance teachers were eagerly awaited at colonial ports, only to have local kinetic codes re-inflect their movements. 55 From this dialectic between the codified, the notated, the improvised, and the embodied arises the creolized dance-music genres of the Atlantic and Indian Ocean worlds, including the mando-a distinct dance form, as well as the songs it was danced to.  59 In , the Fifth Church Council of Goa prohibited dancing to a local sung genre called the munda because such songs and their corresponding 'lascivious and dishonest' (lascivos e deshonestos) dances 'incited sensuality' (incite a sensualidade). 60 Is munda a precursor of mando? Sebastião Dalgado, the early twentieth-century lexicographer of Asian varieties of Portuguese, stated himself unsure of the etymological basis for this possibility. 61 But more telling is his ascription of an African origin to mando. Moreover, his munda and mando keep similar lexical company. The Fifth Church Council had proscribed munda alongside two other dances: sarabanda and cafrinho. 62 Three centuries later, Dalgado's Glossario Luso-Asiatico perpetuated this lexical environment through a series of cross-references that link mando, munda, cafrinho, batuque, and, via a citation of the council's triad of prohibited dances, back to the sarabanda. 63 The inter-referential circularity of these dance-music terms is the archive's testimony to transoceanic creolization operating through regional sub-circuits of cultural change within the Atlantic and Indian in Benin, connecting the Creole Atlantic', Atlantic Studies: Global Currents, vol. , no. , , pp. -. 59 Kabir, 'Creolization as balancing act'. 60 Joaquim Heliodoro da Cunha Rivara, Archivo portuguez oriental: Os concilios de Goa e o Synod de Diamper, st edition, Fasc. IV (Nova Goa: Imprensa nacional, ), p. : 'Como não ha cousa, que mais incite a sensualidade, que cantos, e baile e lascivos, e deshonestos, manda esta sagrada Synodo sob pena de excommunhão que nenhuma pessoa daquy por diante seja ousada a bailar ou cantar a sarabanda, nem as cantigas, que chamão mundã, ou cafrinho, nem os mande bailar, ou cantar' ('Since there is nothing that encourages sensuality more than songs and lascivious dishonest dances, this sacred Synod orders under penalty of excommunication, that nobody from now on dare to dance or sing sarabanda, nor the cantigas that are called munda or cafrinho, nor order to dance or sing them to others'). 61 Sebastião Rodolfo Dalgado, 'munda', Glossario Luso-Asiatico, vols I and II (Coimbra: Imprensa da Universidade, ), II, p. : 'Não me parece que o vocabulo seja outra forma de mandó, dança de origem Africana' ('it doesn't seem to me that this word may be another form of the mando, a dance of African origin'). 62 See ibid. 63 See Dalgado, 'munda', Glossario II, p. , for his citation of the Fifth Council's prohibition of munda, sarabanda and cafrinha; his entry for 'mando', ibid., pp. -, where, via a citation from F. N. Xavier, he equates batuque and mando, and his definition of 'batuque', Glossario I, p. , as 'dança especial entre os negros de Angola … na Africa Oriental tambêm há batuque. Na India, batuque é sinónimo do vernáculo gumate … batucar é tocar o gumate. Batucada é toque de gumate' ('a special dance among the Africans of Angola …. There is also batuque in Eastern Africa. In India, "batuque" is synonymous with the vernacular gumat. To "batuque" is to play the gumat. "Batucada" is percussion on the gumat'). See further notes , , and  below.
Oceans. Sarabanda (also zarabanda, sarabande) takes us to one such sub-circuit that constituted an early Iberian imprint on the New World. The histories of dance-music genres such as villancico, sarabanda, fandango, and rumba, enfold transcultural exchanges within the Andalusian contact zone that gained traction in the Spanish Americas to return, recharged with new rhythms and movements, to the Iberian Peninsula. 64 These developments are well documented, particularly in foundational ethnomusicological scholarship from Cuba. 65 A parallel sub-circuit radiates from cafrinho, which condenses the European impact on the 'pre-colonial traffic in goods, slaves, and ideas around the Indian Ocean'. 66 Cafrinho derives from the Arabic word for 'non-believer', nowadays usually rendered in English as 'Kafir'-a word that carries a charged 'history of relations [between] East Africa and the Indian Ocean, with Swahili-and Arabic-speaking traders, and Portuguese explorers'. 67 This precolonial, Western Indian Ocean label for 'non-Islamic Black people' took on the Portuguese-language diminutive suffix -inho/-inha, to name a new cultural product emerging with the Portuguese advent in the Indian Ocean world. Extant in several spellings in keeping with the fluidity of creole orthographies, 68 the term circulates through the Western and Eastern Indian Ocean to mark a distinct dance-music genre as manifestation of its creolized culture. It is a felicitous instance of linguistic creolization reflecting how European expansionism restructured the 'geography (and cosmology) of the connecting tissue of the Indian Ocean'. 69 This restructuring brings into dialogue creolization occurring on either side of the African continent. The Fifth Church Council's lexical clustering of Goan, Indian Ocean, and circum-Atlantic dances already signals slavery and colonialism's sedimented impress on this transoceanic creolized cultural matrix. To draw the Goan archive into this semantic world-system, a placeholder was needed for the concept of 'local dance being creolised through this matrix'. In , munda fulfilled this function. By , mando moved into its slot. Dalgado turns to the Fifth Church Council's proscription to link munda, cafrinho, and sarabanda. But the evolving repertoire enables him bring mando into the space earlier occupied by munda. Mando is now a synonym for cafrinho, whose cognates in 'Moluccan kafrinu' and 'Timorese kafrinia' allow Dalgado to define it as an 'Oriental' dance originally practised by Africans (cafres). 70 He also ascribes an African etymology to mando, which is defined as a dance-music genre of the Christians of Portuguese India. 71 This perception of a shared Africanity connecting the mando and cafrinha is strengthened by Dalgado's definition of the mando's percussive 68 Variations include: with 'k' rather than 'c', with double 'f', without elision of the vowel in the second syllable, and with the nasalized suffix variously rendered and in both masculine and feminine grammatical forms. 69 Baderoon, 'Provenance', p. . 70 Dalgado, 'cafrinho', Glossario II, p. : 'Parece que é o mesmo que mandó, praticado origináriamente pelos cafres. A dança devia estar muito generalizada no Oriente no século xix, visto que é conhecida nas Molucas com o nome de kafrinu, e em Timor, kafrinia' ('It seems that it is the same as mando, practised originally by cafres. The dance must have been very diffused through the East by the nineteenth century, seeing that it is known in the Moluccas by the name of kafrinu, and, in Timor, kafrinia'). 71 Dalgado, 'Mandó', Glossario II, p. : 'O termo é de origem Africana, mandoa, 'espécie de dança', na língua de Tete. Na India Portuguesa é o nome duma dança popular, entre os cristãos, ao toque de gumata ou batuque, e canto apropriado, que tambêm se denomina mandó' ('the term is of African origin, [from] mandoa, a type of dance in the Tete language. In Portuguese India, it is the name of a dance popular with the Christians, to percussion of the gumat or batuque, and appropriate song, which is also called mando').
instrument, the gumott, as batuque, the Portuguese-speaking world's generic term for African drums, the rhythms played on them, and their corresponding dances; the Glossario cites F. N. Xavier's  synonymous use of gumata, mando, and batuque. 72 A few decades later, A. B. Bragança Pereira's magisterial Etnografia da India Portuguesa not only asserts that the 'Portuguese introduced into its territories in India the mando', but declares it 'probably of African origin'. 73 Reiterating these transoceanic imaginaries, the Goan-born colonial administrator Fernando Leal even declared the mando 'a degenerate kind of lundum', the Afro-Portuguese precursor of creolized couple dances in the Southern Atlantic. 74 These equivalences between Goan and Creole Atlantic expressive practices are all made by nineteenth-and early twentieth-century colonial administrators. As declared by the mando lyric cited in the section heading above, from Goa to (East) Africa they went (and also came). 75 The intra-imperial movement of personnel between Mozambique and Goa consolidated precolonial mercantile comings and goings in the Western Indian Ocean. 76 The tentacular reach of Portuguese imperial bureaucracy set up a chain of relay and rumour across and beyond the eastern and western littorals of Africa, as reflected in Dalgado's yoking of mando to Afro-Atlantic batuque and an eastern African ethnic group. Dances (and the news of dances) being creolized through circum-Atlantic movement were being swirled into the Western Indian Ocean sub-circuit of cultural exchange. The lexical field occupied by munda and then mando demonstrates the Indian Ocean archive's covert memorialization of these transfusions of Africanity through ongoing creolizing processes. These archival traces of Africa lever on a transoceanic plane the circum-Atlantic world's obsession with 'mumbo jumbo': garbled vocalization of African percussive rhythms and chants, seen, for instance, in the chorus of 'Ese rigor a repente', Gaspar Fernandes's villancico from seventeenth-century Mexico: 'sarabanda tenge que tenge/sucusumba cucumbe' (Spanish, 'sarabanda has what it has: sucusumba cucumbe'). 77 Through the shadows cast by repertoire across 'the archive of slavery within and between the Atlantic and Indian Ocean networks' 78 and the transoceanic echo chamber of phantom rhythms, a common pool of 'lascivious' dances emerges. The memory of these dances generates an affective field around the mando within nineteenth-century accounts of elite Goan leisure practices, providing us with tools to interpret it as kinetic practice.  78 Arondekar, 'What more remains', p. .


early blurring of the lines between dances of diverse European and African provenances. Given this 'pattern and flow of cultural influences in which it is often difficult to tell whether a European or non-European cultural influence is predominant', 79 I propose another category altogether: that of creolized dances. Through inter-and transoceanic circulation and the 'sticky webs of copy and contact', 80 they amalgamated, consolidated, and magnified kinetic signatures of various population groups. From West and Central African kinaesthetics came high-affect juxtaposition, body isolations, and polycentrism in response to polyrhythm, 81 and, from dances circulating within European courts came structures of dynamic interaction between men and women that generated pattern and symmetry through their movement in and out of couples. 82 By the nineteenth century, the partner hold would develop as the favoured heteronormative dance format emerging from central European urban culture. During the early phase of creolization on the plantation as a crucible for cultural change, however, features common to popular European and non-European dances, including geometric formations of circles and rows, predominated. During the seventeenth and eighteenth centuries, therefore, a spectrum of social dances emerged that exhibited elements from diverse sources drawn into a creolizing spiral. In the case of the mando, this creole mix includes Indic kinetic elements. These elements enable us to recalibrate through a transoceanic gauge cultural processes usually examined as pertinent to either the circum-Atlantic context or a Luso-Asian, Indian Ocean one, but rarely to both. I extract the mando's kinetic structure from performances I attended during the  Mando Festival in Goa. 83  83 See note  above. The video clips that this article links to were filmed by me at this festival. stage, comprising men and women, respectively. Reinforcing the frame are rows of male and female singers, and, to one side, the (usually male) musicians. When the music and singing start, men ceremoniously invite women from the opposite row, forming couples that dance simultaneously within the space demarcated by the frame (Figure ). Once face to face in the centre, they connect with each other through a mirroring technique (i.e. when the gentleman moves backwards, the lady moves forward, and vice versa; each circles the other in opposite directions). The mirroring is conducted on a diagonal axis with rotated torsos that, as with several Afro-Atlantic couple dances, break the body into two planes. On the lower plane, the feet move in small triplet steps, the mincing and shuffled effect exaggerated by the women's sarong-like skirt and stockinged, slipper-clad feet ( Figure ). On the upper plane, connection is generated through interlocked gazes and gender-specific accessories wielded by the hands: a handkerchief for the men and a fan for the women (Figures  and ). The dancing has a slow tempo, but it concludes with a faster segment, during which the handkerchief is rhythmically flourished. At that moment, the singing switches to the quicker-paced dulpods, and the dancing becomes more animated with lateral swings discernible in the women's hip movements. The geometry of the frame is maintained until the end ( Figure ).
Danced thus, the mando's structure coheres with creolized quadrilles and contradances popular from the eighteenth century onwards wherever Europeans lived, settled, or sojourned, sharing culture with non-Europeans. By the late eighteenth century, the dance known variously as 'contredanse', 'contradança', and 'contradanza' had spread across Europe and its colonies, cutting across social classes as well as linguistic and imperial boundaries; its progressive elaborations resulted in the quadrille. Both kinds of dances generated meaning and pleasure through the dynamic interface between the group, its constituent couples, and the musicians who render that interface audible. As a sign of their popularity, they were profusely creolized through the Atlantic and Indian Ocean worlds. 84 The mando's contemporary performances suggest it is a creolized contradance of the cotillion or longways variety, 85 where men and women are arranged in facing lines; this frame is a variant on the quadrille's idealized quadrilateral, while the 84 See Manuel, Creolizing contradance; Chaudenson, Creolization, pp. -; Kabir, 'Creolization as balancing act'. 85 Manuel, Creolizing contradance, p. .


male-female rows also recall those of Goan Cunbi dance. 86 The mando's mandatory pairing with the dulpod, and, as I will demonstrate later, its submerged connection to the dekhni, also reflects the quadrille's development into a multi-part music-dance form. Indeed, this semiotic interdependence between the mando, dulpod, and dekhni conforms to the creolized quadrille's adaptation of the multi-part structure to express a kinetic history of creolization itself. Hence I place the mando amongst other Indian Ocean quadrilles that creolized within a space of shared linguistic, musical, and kinetic material that Mahesh Radhakrishnan calls the 'bailasphere'. 87 Referencing the modern Sri Lankan and Mangalorean dances known as baila, derived in turn from baile (Portuguese, 'dance'), Radhakrishnan's bailasphere includes extant variations of the cafrinho, such as the kafriinha still danced by creole Burgher communities of Sri Lanka, 88 and the Goan mando. The dances of the bailasphere share an asymmetrical rhythm: an overlay of binary and ternary patterns with an accent on the fifth and first beats that, together with suspensions and syncopations, indicates an Indian Ocean rhythm substrate. 89 This rhythm's kinetic manifestations within the quadrille frame are deeply infused with peninsular Indic expressive culture. Heterosexual connection within the frame is managed 'remotely': never through interlaced arms or clasped hands through which men and women interchange places or turn around each other, but always through an interplay of gaze, glance, and gesture, 90 In some mando performances today, we see couples briefly holding hands, but this hold is never used actively to manage the connection between the dancers. Instead, the reliance on the gaze to manage this connection converges with the meaningfulness of the gaze (as darshan, nazar, and so on) including wrist movements whereby the lady manipulates her fan and the gentleman his handkerchief-accessories also ubiquitous in Iberian


bestowing on performers carte blanche to 'dress up' in order to recall another temporality. 99 The mando's agonistic structure as balancing act between opposing groups motivates a distinction between what male and female performers respectively wear. While the men wear formal European attire, the women's outfit, already described earlier in this article, is called pano-baju (Portuguese) or torhop-baz (Konkani). 100 Comprising a tubular skirt reaching the ankles, a long-sleeved tailored blouse, and a shoulder stole, its decorative elements include vegetal motifs embroidered in gold on the blouse and borders on the skirt running longitudinally down the middle and horizontally along the lower edge. Accessories include elaborately worked gold bangles, necklaces, and earrings; tortoise-shell head combs; filigreed fans in tortoise-shell, mother-of-pearl, or bamboo; and low-heeled, soft slippers. This ensemble recalls neither 'Western' nor 'Indian' traditional dress, but takes us instead to the sarong and blouse combinations of South-East Asia, as the Goan designer Wendell Rodricks demonstrated. 101 The embroidery, metalworking, and cutting techniques, as well as the materials used, attest to cultural transfer, exchange, and adaptation in the Indian Ocean world stretching from Japan to Persia, within which the Western Indian littoral was decisively positioned. 102 The names for these elements encode layers of taste-making-thus the woman's shoe, called a chinelo in Goa, recalls China, while combining elements from the Persian zapat (from Spanish zapato, 'shoe') and Mughal slippers; pano is Portuguese for wrapped cloth and is a word used widely throughout Africa, while baz is a Portuguese-influenced Konkani pronunciation of the Persian baju ('arm'), denoting the characteristic long sleeve, which nevertheless was shortened to three-quarter lengths to show off gold kanknas (Konkani, 'bangle', from Sanskrit kangana) 'similar in spirit to Deccan coastal gold ornaments'. While the male mando performer's outfit invites straightforward association with the Portuguese colonizer, 104 that of his female counterpart radiates multiple histories of transoceanic creolization enacted on the woman's body as a conduit of cultural change. By the beginning of the twentieth century, sarong-blouse outfits were firmly associated with women of mixed-race communities throughout the Eastern Indian Ocean. 105 Goan memory attributes the pano-baju's early twentieth-century popularity to its association with the mestiça wives that Portuguese administrators married in the Malaccas and brought to Goa. 106 Collective memory also associates the pano-baju with the heydays of the mando as a social dance practised amongst the elite denizens of Indo-Portuguese mansions, even though, in contemporary Goa, it is worn only during mando performances, including those staged for filmed documentation. 107 Thus, in Goa, this outfit is consistently recalled in conjunction with a danced memory of creolization. It materializes the structure of the dance as a courtship encounter between the colonizer and the colonized subject. This performance of creolization through a 'primal scene' staging 'the scandal of cultural miscegenation' is common across the circum-Atlantic world. 108 These 'foundation scenarios' assign the role of the colonizer to the man and that of the colonized to the woman, highlighting this role play through appropriate costume. 109 For instance, in the Angolan Rebita, also an open-format group dance like the mando, the men are dressed exactly like the male mando dancers, while the women wear 'local' dress. 110 104 Even while signalling the adoption of some version of 'Western' wear for the (post) colonized South Asian male in urban white-collar and aspirational contexts. 105 See Peter Lee, Sarong Kebaya: Peranakan fashion in an interconnected world, - (Singapore: Asian Civilisation Museum, ); Rodricks, Moda Goa, pp. -. 106 As attested in numerous conversations with Goans conducted in course of fieldwork. However, the precise South East Asian equivalent of this combination seems not to have crystallized until the late nineteenth century; see Patricia Ann Hardwick, '"Neither fish nor fowl": constructing Peranakan identity in colonial and post-colonial Singapore', Folklore Forum, vol. , no. , , pp. -. 107 See, for example, the reconstructions in Dances of Goa (dir. Nalini Elvino de Souza, ).
108 Stuart Hall, 'Créolité and the processes of creolization', in Cohen and Toninato (eds), The creolization reader, p. , where he discusses the 'primal scene of the encounters between different worlds for which the Caribbean has historically provided the crucible'. 109 Taylor, Archive and repertoire, p. . 110 I am grateful to Federica Toldo for this information about Rebita. See Federica Toldo, '"Da geração" e "da simpatia": relacionalidade em prática em três danças da Ilha de Luanda (Angola)' (PhD diss., Paris , ).


Where the mando diverges from this and other examples of danced foundation scenarios is that its 'local woman' wears an outfit that asserts creolization through anterior and ongoing transoceanic encounters.
The sumptuous and lustrous materials (brocade, velvet, silk, tortoise-shell, mother-of-pearl, gold) of the pano-baju are metonymic of the elite status enjoyed by the community that the 'local woman' represents. 111 It consolidates the gravitas of the mando sung, which proclaims a message of cultural capital: of socio-economic gain rather than loss as a consequence of the creolization enacted by her courtship dance with the 'foreign man'. Its motet structure, rendered by voices in intervals of third and sixth parallels, 112 derives from the centuries-long creolization of mentalités catalysed by the Church that I argued for in an earlier section of this article. Mando lyrics deepen this acoustic alignment with institutional power through Konkani studded with elevated lexis drawn from two high-status languages. From Sanskrit derives a cosmological and planetary vocabulary-surya ('sun'), noketra ('stars'), tsondrima ('moon'), rendered in the prestigious Salcette accent on Konkani. From Latin, channelled through Portuguese, derives a vocabulary calquing liturgy with law-such as the lover's juramentu sagradu (from Portuguese juramento sagrado, 'sacred oath') in the section heading above. 113 The Portuguese phrase is seamlessly incorporated within the sentence through a conjugated Konkani verb (ditam) and the Indic vocative particle re that is integral to the cadence of the lyric's line. This deposition of a specialized Portuguese lexical field into the very syntax of Konkani is a ubiquitous feature of mando lyrics, as the section headings throughout this article demonstrate. The creolization processes that manifest in the mando on multiple levels are again in evidence-not through a linguistic creole, as might be expected, 114 but through a Konkani fundamentally transformed by the very encounter that the mando's performance of the status quo memorializes.
Polkist fulambai tuka kitea zai? ('why pursue polka dancers, flower-girl?'): desire's circuit The mando, like other creolized quadrilles, mobilizes the quadrille's kinetic structure to perform the collective memory of 'tabooed desire' and 'the scandal of cultural miscegenation'. 115 Its musical composition and linguistic style ratify the creolization of mentalités characterizing Goa under the Portuguese. However, its sartorial allusions to the rich stuffs of transoceanic trade and the classicizing Konkani of mando lyrics contrast sharply with the creolized quadrilles of the Caribbean and Western Indian Ocean islands, typically danced in clothes reminiscent of plantation-era sumptuary codes, to lyrics sung in various Creoles. While the insular creolized quadrille reflects the aspirations of the newly emancipated, then, its peninsular Indic counterpart transmits the retention of power by the Indo-Portuguese elite-mapping onto the difference between the geographical history of islands, which stimulates creolization by encouraging cultural rupture, and the continuing, complicated relationship with hinterlands that littoral and peninsular geographies afford. 116 These aligned differences crystallize in the mando, allowing us to refine a theory of creolization as the unpredictable generation of culture under conditions of duress. Divergences and overlaps between distinct sites of creolizing processes (for example, plantation versus enclave) and socio-economic axes of cultural transfer (for example, slavery versus mercantilism) can be calibrated through the mando's ludic and kinetic resources, 117 particularly its relationship with the music-dance genre dulpod, which furnishes the concluding section of a mando performance. Characterized as 'a song of joy', 118 the dulpod's quickened rhythm, rendered graphically as / rather than the mando's slower /, progressively distends the mando's stateliness. Its onset signals the performance's imminent descent into a world populated by an 115 Hall, 'Créolité', p. . 116 It would be illuminating, though beyond the scope of this article, to track the materialist histories corresponding to this theoretical abstraction by mapping it onto the intersecting inequalities of caste and conversion-as manifested, for instance, in the category of the 'Brahmin Christian'. 117 On these sites and axes, see Trouillot, 'Culture on the edges'; and Derek Bickerton, 'Creole languages', Scientific American, vol. , no. , , pp. -.


affectionately stereotyped cast, which completes the dramatis personae of Indo-Portuguese creolization. 119 The introduction to an influential anthology of dulpods provides an exhaustive list of these characters, including the advogad ('lawyer'), alfiad ('tailor'), beatinny ('devout spinster'), bikari ('beggar'), firngi ('white man, Portuguese'), forvoti ('sawyer'), harvi ('fisherman'), iscrivaum ('scrivener'), inglez ('Englishman'), kolvont ('temple dancer'), marinheir ('seaman'), maskany ('fisherwoman'), mistis ('mixed-race person'), padri ('priest'), and render ('toddy-tapper'), but somewhat mysteriously absent from the editors' lively litany are the two figures of my section heading above: polkist and fulambai, whose interaction forms the topic of at least one still popular dulpod, 'Ago fulambai' (Figure ). 120 Addressing a young woman as fulambai ('flower-girl'), it recounts her obsession with the motty-motte polkist (the 'expert polka dancer'). The blame is laid at Fulambai's door: 'Why are you after the expert polka dancers?' asks the first verse; 'expert polka dancers are in your heart and soul (kalliz any'm curassaum),' concludes the penultimate one. Yet it is clearly a mutual attraction: 'expert polka dancers are winking at you,' warns the second verse; 'expert polka dancers are always around you,' iterate the third and fourth. The chorus and the concluding verse thread these declarations of mutual attraction with the tropical flowers abolim ('orossandra') and mogra ('jasmine') that Fulambai accessorizes herself with and carries baskets of. 121 While these flowers, invoked through their Konkani names, bestow on Fulambai an intensely local aspect, 122 her dialogic relationship with the polkist establishes her as an aspirational figure within the dulpod's world. Whether expert or student, a polka dancer is a catch. 123 To attract him, Fulambai needs to be able to match his steps: 'without the polka,' one dulpod asserts, 'you cannot get stylish husbands.' 124 At the same time, Fulambai and her ilk infuse some necessary 'salt and feni' into the proceedings. 125 Whenever they enter the scene in dulpod lyrics, hips move and energetic contradanças and the rambunctious variety of quadrille called the Lanciers are mentioned. 126 The lyrics thus celebrate not only Fulambai's exposure to European social dances brought into the quotidian space by the polkist, but also her transformation of those dances into something altogether more zam zam ('vibrant'). 127 The local flower(-girl)'s transformative agency is enhanced by metonymic contiguity to Konkani, which, in absorbing the European dance terms syntactically, extrudes the creolizing process onto its surface: the word 'polka' always appears in dulpods already absorbed into Konkani syntax-either through the construction polkist or in conjunction with 'waltz' within the formulaic phrase valsam-polkam. 128 These European social dances function as signifiers of a particular world within dulpod lyrics, whose contours emerge through oppositional status with Fulambai and her tropical flowers. This opposition is strengthened through the lexical differences between the lyrics of dulpods and mandos. While the mando expresses courtship, love, and fidelity through its elevated lexis, the dulpod evokes daily life, often in a sexually suggestive tone. However, dulpod lyrics conjure up these opposed worlds precisely to suggest their capacity to seep into each other. Weddings, saints' days, market days, and carnivals 129 : this festive temporality punctuates the lyrics of several dulpods, offering its dramatis personae sites of encounter, through and in dance. Because the dulpods present lyric fragments, not concluded narratives, these encounters remain of the moment; nowhere do we hear of Fulambai and the polkist settling into a 'happily ever after'. These lyrics that memorialize dance as enabling encounter are matched by a dance in which desire becomes a non-teleological circuit: those playing the roles of Fulambai and the polkist advance and retreat towards each other; they circle each other. We note the same libidinal economy of couples interacting within a group that characterizes the mando. Within contemporary performances, dancers segue from mando to dulpod without changing their outfits; the singing style remains the same, as does the rhythm, although the tempo quickens and, as noted earlier, hips indeed move-though only as much as the narrow pano-baju permits. The dulpod thus displaces the mando's courtship between the European man and the elite local woman to a relatively Rabelaisian version, devolved onto the elite local man and non-elite local woman. The opposed social worlds of mando and dulpod lyrics reiterate these different moods of their corresponding dances. Yet, because always contiguous, these dances repeat kinaesthetically the performance of 'opposites attract'. This pattern of self-similarity and repetition with difference follows the fractal logic that dictates the creolized quadrille's multi-part structure or 'suite'. 130 In quadrille traditions and several couple dances that developed from them, suites function as repertoires that memorialize a spectrum of kinetic possibilities resulting from the creolization process. Hence the Tango 'tanda', the Antillean quadrille 'haute taille', or the Seychellois 'kamtole' all enfold within a set sequence, dances of different tempos, postures, affects, and lyrical emphases. 131 I suggest we consider the mando and the dulpod as comprising an attenuated mando suite in two parts. As constituent parts, the mando and the dulpod present respectively the local and transoceanic dimensions of the foundational scenario of creolization, while the suite as a whole performs these dimensions as complementary and interlocked.
Kolvontancho khellu ('the dancing girls play'): transoceanic dance, riverine crossings The mando suite's cultural intelligibility depends on what it includes as much as what is deemed to be outside it. It activates a particular memory of cultural encounter through its constituent parts, the mando and the dulpod, that together signify the dialectical relationship between creolization on transoceanic and local planes. But, in keeping with its underlying fractal logic of reduplicating self-similarity, the entire suite also exists in a parallel dialectic relationship with the dance-music genre dekhni. Structurally separated from the mando suite within contemporary performances but tied to it nevertheless through semiotic interdependence, the dekhni sheds light on how creolization is memorialized as having shaped not just the Indo-Portuguese elite-but also those elements within local society that, from the perspective of that elite, escaped creolization. The creolizing matrix here is understood as Christianity: conversion is the watershed that separated the new from the pre-existent. The figure within which this intimate otherness crystallized is the dancing girl of Hindu temples: the kolvont/bailadeira. Like Fulambai of the dulpods, she is associated with flowers: through her floral accessories and through the names she bears, which reference both the general 'flower' as well as specific local varieties. Dekhnis also name what she loves to eat: sweet and sour fruits, spicy vegetable preparations, and the addictive betel nut. 132 She dances not to court a would-be husband, but for enjoyment-her own and her audience'sand to placate her capricious gods. 133


the dekhni evokes the Hindu-temple culture that had been banished, symbolically and materially, to 'the other side of the river' in the course of the conversion of Goan elites to Catholicism. 134 The metonymy of the kolvont's desirability and the riverine cast of this symbolic topography is best captured by the dekhni 'Hanv saiba', contributing to its popularity long before and after its showcasing in the film Bobby. Its lyrics depict the kolvont and her friend pleading with a ferryman to take them to the river's farther bank, where the wedding of one 'Damulo' is taking place. In exchange for the crossing, they offer the ferryman their anklets, bracelets, necklaces, and nose rings; each verse focuses on the act of pointing to and taking off these items of jewellery from the corresponding parts of the body, bestowing the lyrics with excellent kinetic and mimetic potential. The ferryman's reiterated refusal is punctuated by a refrain that shifts the scene to Damulo's wedding canopy under which 'kolvonts play' (kolvontancho khellu). 135 The dekhni's interest, thus, is not so much in the success (or otherwise) of the kolvont's crossing, but in her ability to bargain for mobility through the accrued capital that she wears festooned on her body. 136 At the same time, the repeated reference to dancing girls enjoying themselves at Damulo's wedding emphasizes the kolvont's circulation outside the juridical and aspirational coupledom enacted in the mandos and dulpods. 137 Nevertheless, while her distance from the world of courtship and matrimony is emphasized by the fact that it is another's wedding she has to attend, her economic dependency on the institution of marriage is signalled by her willingness to invest her own capital to facilitate her ferry ride. The river, which both separates and connects, deepens the ambiguity of the kolvont's social position, but heightens her allure.
The riverine boundary can be read as the River Zuari, which, through the Portuguese New Conquests, demarcated Hindu from Catholic territories in Goa. But the kolvont's allure derives from an earlier 'flight 134 Ibid. On the conversion of Hindus through the Goan Inquisition, see Rowena Robinson, 'Some neglected aspects of the conversion of Goa: a socio-historical perspective', Sociological Bulletin, vol. , nos -, , pp. -; and Ananya Chakravarti, 'Mapping "Gabriel": space, identity and slavery in the late sixteenth-century Indian Ocean', Past & Present, vol. , no. , , pp. -. 135 Pereira and Martins, Sheaf, pp. -. The lyrics are widely disseminated, often in divergent versions. 136 Arondekar, 'What more remains', p. . 137 For some of these relationships, see ibid., p. .
of the deities' that was initiated during the Old Conquests. 138 Indeed, she enters the archive coterminous with the earliest appearances of 'lascivious dances' being creolized and censured in the Atlantic and Indian Ocean worlds. The same Fifth Church Council of  that prohibited the munda, sarabanda, and cafrinho, as we saw earlier, also banned groups of moças bailadeiras ('young dancing girls'). 139 Despite the Goan Inquisition that the Church Councils heralded, the bailadeira's role within Hindu rites continued, albeit within villages on the periphery of the Old Conquests to which the Hindu gods had fled. 140 By the time of the New Conquests, her symbolic freight increased through conflation with British obsession with nautch-girls and Indian nationalist discomfort with devadasis. 141 Meanwhile, by the early twentieth century, troops of such 'dancing girls' were crossing back and forth between these 'discrepant empires' centred in Goa and Bombay. 142 Their reinscription into the mythopoetic realm via the dekhni, a genre that Goans acknowledge as a product of elite Christian imaginings of the Hindu other, codifies in performance and lyric that other's symbolic persistence. The kolvont enables the dekhni to sublimate the alienation from a part of the autochthonous self that conversion psychically signifies. Dekhnis such as 'Hanv saiba', composed by Indo-Portuguese elites dwelling in the very districts that had become seats of Brahmin Christian culture following the Old Conquests, represent a ritual 'activation of memory' that propitiates perceptions of complicity with colonial violence, here crystallized in the Goan Inquisition. 143 Undoubtedly, versions of 'dekhni' as 'local song of the Deccan' existed ever since there were proximate 'non-local' cultures. But, by the end of the nineteenth century, Indo-Portuguese composers activated collective his social group. This group included Indo-Portuguese ladies who posed for photographs dressed up as their Hindu counterparts, draping the nine-yard sari between their legs in the Deccan manner, adorning themselves with gold ornaments that, like the altarpieces of their churches, marshalled Deccan craftsmanship into the service of Indo-Portuguese affective exigencies. 145 The dekhni performs creolization as mimicry that generates their collective Rapsodia Ibero-Indiana as a self-reflecting hall of mirrors. As song, dance, and dancing girl of the Deccan, the dekhni is an object of desire: for the composer, for those who danced it mimicking the dancing girls, for those who danced with them, and for the mando suite itself. Its auto-orientalizing oneiric realm projects onto the space of social dance: elite Christian imaginings of an intimately other(ed) local Hindu culture. The language and scenarios of dekhni lyrics verbalize this ludic seepage between genres as a mise-en-abyme of creolization that the mando suite manifests. Transoceanic and riverine circuits of desire intersect as the kolvont, wearing saris of seda (Portuguese, 'silk'), entices the desai (Konkani, 'village headman') to 'wiggle his hips' even as she 'bends' hers to the sounds of the 'cornet' (corneticha sadary kolvont/ox'm ox'm morhote). 146 Mannyka atam fel'cidade polleuchem ('I must now seek felicity'): a strategic forgetting Just as Portuguese words and a creolized habitus seeped into mando, dulpod, and dekhni lyrics, the boundaries between their corresponding dances and their affect-worlds were also permeable historically. In , António Lopes Mendes depicted a sari-wearing kolvont and seated musicians and singers, captioned dança do mandó em casa das bailadeiras (Portuguese, 'Mando dance in the house of dancing girls') and described as mandó rudimentar, à antiga ('a rudimentary mando of ancient times'). 147


the Mando and among the Dulpods' 148 and kinetically interpreted through the mando suite's creolizing structure. In , Carlos Eugénio Ferreira presented 'Hanv saiba' with an accompanying théorie de la danse par quatre couples, cavaliers et dames ('dance instructions for four couples comprising men and women'), in French ( Figure ). 149 This terminology and the choreography he provides are those of the quadrille; moreover, in line with creolization's constant innovations, he instructs the couples to incorporate the recently fashionable one-step into their footwork. Ferreira was not unique in thus interpreting the dekhni: it had been presented as a contradança de honra (Portuguese, 'contradance of honour') at an elite wedding in the s, under the direction of Lourenço Henrique Dias, leader of the Banda Nacional de Salcete. 150 Around the same time, the young Goan writer Floriano Barreto described mando dances descending into a wild denouement 'reminiscent of popular Hindu Goan music', starting in an 'elegant and delicate' manner, but moving to an 'unbelievable prestissimo … equal [ling] that of a gallop' during which 'all sing in a great chorus replying to the body of singers proper … and the gumott is played upon with impetuousness'; the result is a 'fear-arousing din, a vigorous orgy of drumming, which is soon mixed with sharp cries and piercing whistles'. 151 Contemporary mando performances seem quite distant from these boundary-breaching exertions. The dance today has a slow tempo and is decorous, as couples execute their dainty steps through postures of restraint. Even considering that these are staged performances that last for a pre-determined and short duration, this mando hardly accords with Barreto's frenetic dancers who, by the close of the evening, are 148  'This article, though written in the early years of this century, is the first authentic document we have of the mando in the days of its zenith. However, the author describes the dances as it was performed in Margao which being a town, was exposed to various mixed influences, with the result that the dance deviates in many points from the classical mando as danced in the classical centres Curtorim and Loutolim.' The significance of this rider will be returned to in the following section. mopping brows, sweating and panting, and reaching out for water. Hence, Susana Sardo considers Barreto's account of the mando evocative of the way in which the dulpod, rather than the mando, is danced. 152 Her


observation appears based on the accelerated section danced to dulpod lyrics that closes mando performances today, and on an assumed strict separation of these genres. However, Barreto's description is concordant with the concept of the mando suite that I have advanced, wherein mando names both the suite and one of its constituent dances. This polysemy, common in transoceanic creolized dances, derives from the fractal logic of creolization that generates segments within dance suites distinct in tempo and mood, yet prone to seeping into each other. 153 Barreto's contemporary, the satirist Jip (Francisco João da Costa), described dancers moving from o mando bom ('the good mando'), sung with voz dolente ('mournful voice') towards a dança louca ('the crazy dance'); the mando dolente reappears in Carlos Eugénio Ferreira's oeuvre ( Figure ). 154 These accounts illuminate how mando was danced at the turn of the century as an improvised social act. Its opposing dimensions, which, following da Costa, we may term dolente and louco, are reflected in Barreto's concluding summary of the mando as a diptych into whose panels condense this opposition. 155 The blurring of mando, dulpod, and dekhni through performance contrasts with their presentation as separate(d) genres within discourse. The archive registers successive attempts to sanitize and domesticate the creolized mando suite through such separation, conscripting the dulpod as mediator between the mando and the dekhni, and consigning the dekhni to the status of the noble mando's itinerant other: 'Like the monkeys on the trees on the other side of the river, the dekhni can never stay for long in one place.' 156 This comment, made in  by dekhni editors whose surnames align them with the same elite constituency that 'created' the dekhnis at the turn of the century, perpetuates the auto-orientalizing worldview of Rapsodia Ibero-Indiana. Ferreira's capture in writing and notation of 'Hanv saiba', and its presentation as the opening dekhni of the  edition on dekhnis, correspond to two political contexts for discursive interventions into the mando's performance: 'the birth of the republic in Portugal in , followed by a momentous and tumultuous decade when Goans hoped to govern themselves' 157 and the dramatic incorporation of Goa into postcolonial India in , which loomed as a possibility ever since India's independence from Britain in . Both moments triggered intense renegotiations of identity, status, and affiliation for the Indo-Portuguese elites whose creative and social world the mando represents. 158 The archive is sedimented by such interventions to mobilize repertoire into politically expedient versions of the self, even as repertoire performs and commemorates earlier processes of self-making in the face of expediency.
These processes enact a strategic forgetting of creolization's routes in search of 'felicity'-as signalled in the section epigraph. 159 The 'birth'


of the dekhni around the s signals readjustment of the semiotic system that I am conceptualizing as the mando suite. Mythopoetic resources that had congealed in foundation scenarios during the Old Conquests period were recycled through mimicry and play to bring forth a newly porous partition between Catholic 'self' and Hindu 'other'. Barreto's account of mando, dating from that period, was translated from Portuguese into English in , for a special issue of the Indian art journal Marg dedicated to Goan art and culture. 160 The system was shifting again in response to political currents: that year, visas became necessary to cross into Goa from elsewhere in India. 161 Tellingly, the Goan intellectuals contributing to this issue make Barreto's account available to a wider, English-speaking public while distancing themselves from its Dionysian emphases. Interventionist footnotes and translator's erasures vigorously repudiate the recognition, by Barreto and others of his generation, of a creolized Atlantic genealogy within the mando. 162 Instead, Goa's creolized culture is reinterpreted through Nehruvian discourses of syncretism, with an emphasis on its statelier elements that could resonate with an autochthonous antiquity seemingly inherited by other regions of India. Such reinterpretation, already foreshadowed in the  Marg issue, progressively repackaged the mando suite as an art dance after India's takeover of Goa in .
back to the last imperial power to exit a nascent national space, but as a more complex 'remembering-back' to an entangled inheritance of inter-imperial, mercantile, still dynamic transoceanic pasts.