Circular Migration and the Spaces of Cultural Assertion

Abstract

Harnessing primary and secondary evidence from India, our essay conceptualizes the cultural dynamics of migration. In so doing, it demonstrates the incompleteness of standard marginalist and Marxist accounts of labor circulation. As a corrective, we examine the linkages between culture, politics, space, and labor mobility and offer a way to think about them by building on poststructural critiques of development and postcolonial theories of migrant subjectivity. The proverbial compression of space-time not only has made extralocal work more viable for members of proletarianized groups but, more importantly, has allowed them to transfer their experiences of new ways of being into local contexts through acts of consumption and labor deployment that can become elements of a Gramscian counterhegemonic praxis. We argue that the possibility of this sort of “body politics” compels not merely a critique of the modernization paradigm that has organized classical migration studies but, more profoundly, a reassessment of the way we understand modernity itself. We advocate an approach that provincializes the Eurowest and foregrounds the existence of pluritopic “regional modernities.”

Key Words:

• consumption
• India
• migration
• regional modernities
• work

Acknowledgments

Special thanks to John Paul Jones III for his advice on content and organization, and to the various anonymous reviewers for their helpful comments on the original article. We particularly acknowledge the detailed critiques of two referees, who compelled us to undertake difficult but entirely necessary revisions and, we hope, write a substantially stronger article. We are grateful to Ron Aminzade, Ben Crow, Michael Dove, Ben Rogaly, James Scott, and Mark Steinberg for their meticulous comments on earlier versions of this article, and we thank Paul Alexander, Tania Li, Mary Beth Mills, David Mosse, Pauline Peters, and Jeff Romm for helpful conversations along the way. Thanks also to Sula Sarkar for her cartographic assistance. Responsibility for remaining errors or inconsistencies is entirely ours. The various pieces of research on which this essay is based were supported by grants from the American Institute of Indian Studies, the Izaak Walton Killam Foundation of Canada, the Population Council, the Social Science Research Council, the Wenner-Gren Foundation for Anthropological Research, the Institute of Development Studies, the University of Sussex, U.K. and the Royalty Research Fund, University of Washington. Our heaviest intellectual and emotional debts are to friends and acquaintances in Gujarat, Tamil Nadu, and West Bengal who made our research possible. Our abiding thanks as well to V. Arivudai Nambi and Muniappan for their crucial support during field research in Tiruvannamalai District, Tamil Nadu, in 1999, 2000, and 2001.

Notes

1. The observations developed in this essay apply to seasonal or circular migrants and relay migrants, not to permanent out-migrants. Labor circulation encompasses transhumance, rural to urban, and intrarural migration, whether of a short term or semipermanent nature. For taxonomies of migration, see Chapman and Prothero (1985) and Rogaly (1996).

2. Our stance on “modernity” deserves clarification: we reject both the “transition” narrative that portrays the history of societies as a discontinuous switch from the “traditional” to the “modern” (hence, implicitly suggests a defining moment when “modernity” takes hold of rural society) and the related “modernization” narrative that portrays history in progressivist, evolutionary terms. Instead, we take the position of South Asian and Latin American scholars of subalternity (Chatterjee 1993; Coronil 1997; Dussel 1998; Mignolo 1999; Prakash 1999; Chakrabarty 2000) that “modernity” is a pluritopic phenomenon, not confined to Europe, and that the rationalizing effects we observe in societies and ascribe to a singular, European modernity are in fact the operation of an Eurocentric discourse that has collapsed geographically differentiated processes of modernity (representing the articulation of precolonial, colonial, and postcolonial polities) into a unitary, historicist account. Watts (1995) makes a similar point.

3. Harvey (1989), Anderson (1998), and Twitchell (1999) offer trenchant analyses of these pathologies, particularly commodity fetishism.

4. Although we bracket the issue of gender in this article, obviously our intention is neither to suggest that the migrating body is only male nor to say that the renegotiation of gender identities is politically less salient than the tussles over class, ethnic, caste, and tribal identities. Rather, our emphasis in this article on caste and tribal identifications reflects our research trajectory, as well as a perceived gap in the migration literature on these struggles.

5. We understand “labor valorization,” in Marx's ([1876] 1976, ch. 7) sense, as the process whereby the commodity, labor power, is purchased and consumed for its use-value within a capitalist production process in order to generate surplus value. However, in this article we understand acts of consumption and labor valorization not as merely about the acquisition of use- or the transfer of exchange-values, but just as crucially, for their symbolic value in communicating social distinction. The communicative aspect of commodities is the subject of a rich literature. We owe debts primarily to Veblen ([1899] 1994), Douglas and Isherwood ([1979] 1996), Bourdieu (1984), Appadurai (1986), Baudrillard (1988), McCracken (1988), Parry and Bloch (1989), and Miller (1995a, 1995b). Like several of these authors, we want to underscore the signaling and gate-keeping functions of commodities—the ways in which they serve to mark social boundaries and hierarchies.

6. Bourdieu (1977) contrasts “heterodoxy” (the universe of competing possibilities) to “doxa” (the universe of undisputed ideas) and “orthodoxy” (the universe of straightened or rationalized ideas). We employ the word “heterodoxy” in this precise sense to convey the point that “body politics” is intrinsically about the dominated challenging the institutionalized or internalized censorships of the dominant.

7. Arnold (1984) discusses how Gramsci's ideas can augment our understanding of peasant subalternity in India.

8. Aside from the work of Stuart Hall (see, for instance, Hall and Jefferson 1976) and the Birmingham Centre for Contemporary Cultural Studies, the work of Michel de Certeau (1984) also deserves mention here. Witness this account (1984, xii–xiii):

To a rationalized, expansionist and at the same time centralized, clamorous, and spectacular production [i.e., a dominating media] corresponds another production, called “consumption.” The latter is devious, it is dispersed, but it insinuates itself everywhere, silent and almost invisibly, because it does not manifest itself through its own products, but rather through its ways of using the products imposed by a dominant economic order.

We appreciate the spirit of subversiveness that pervades de Certeau's account, but we find it overly instrumental and optimistic. Our view is much closer to Gramsci's, who recognizes the contradictions and brittleness of both hegemony and agency.

9. Significantly, Frederick Cooper's work claims that the origins of “development” were not exclusively European—that, in fact, it is tied to how the “labor question” was resolved in Europe's African colonies (vide Cooper 1992).

10. The key variable in the migrant's decision to move is net expected earnings. In other words, the rational migrant weighs the industry to agriculture wage differential by the probability of finding urban employment and subtracts migration costs from this expected income.

11. In the popular Harris-Todaro (1970) model of migration, the equilibrium condition reads: (L_F/L_F+ L_I)w_F+ (L_I/L_F+ L_I)w_I=w_A, where L_F is total formal employment available in the urban sector, L_I is total informal employment available in the urban sector, w_F is the fixed or mandated wage rate in the formal sector, w_I is the wage rate in the informal sector, and w_A is the wage rate in agriculture. The term (L_F/L_F+L_I) is simply the probability that the migrant will find formal employment in the urban sector; similarly, the term (L_I/L_F+L_I) represents the probability that the migrant will find informal employment in the urban sector. Thus, the left-hand side of the equation gives the migrant's expected earning in the urban sector. In equilibrium—that is, a state where no person wishes to migrate from one sector to the other—expected urban sector wage must equal the prevailing agricultural wage.

12. The decennial census of India provides figures on the stock of rural migrants (urban residents of rural origin) in the survey year and can therefore be used to track decennial changes in migrant inflows to urban areas. Unfortunately, the census data allow no way of discerning permanent migrants from circular migrants. As mentioned in the text, most estimates of circular migration rely on intelligent conjectures from smaller regional surveys and ethnographic studies. Papola (1997) and Srivastava (1998) discuss limitations of the Indian census data for migration studies.

13. The word “dalit” applies to those downtrodden groups who, within the scheme of orthodox Hinduism, were considered ritually impure and therefore untouchable. Unlike the euphemism “harijan” (people of god) coined by Mahatma Gandhi, the term “dalit” has an explicitly political, anticaste connotation. Zelliot (1992), Omvedt (1995), Fernandes 1996, Mendelsohn and Vicziany (1998), Deliége (1999), Michael (1999), and Moon (2001) ably document the rise of dalit social consciousness and dalit movements in India. A note of clarification: we employ the term “dalit” for members of historically subordinated low-caste groups because this is increasingly how they choose to represent themselves (and when they do not, we substitute the proper term of reference—for instance, “harijan”). By contrast, the term “adivasi” (original inhabitant) does not carry the same positive or political connotations as the term “dalit.” Tribal groups do not use it self-referentially (instead, they use their proper group name—for example, “bhil”). And upper-caste groups often employ the term “adivasi” derogatorily. Hence, we prefer to use the term “tribal” in place of “adivasi.”

14. For our purpose, “space,” in any historical context, can be circumscribed by material social processes that involve the traffic of goods or ideas between sets of people; by contrast, “place” refers to the forms of consciousness with which individuals and groups apprehend and transform particular material spaces, taking for granted that these forms of consciousness are dialectically related to social power relations. This distinction builds on Heidegger (1977,; 332ff.) and the evocative discussion of place that is developed in Basso (1996,; 105–49). Creed and Ching (1997,; 7) offer a related definition of place as “a grounded metaphor.”

15. We thank two of the anonymous referees for urging us to foreground the feminist literature on “body politics.”

16. Parry (1999) goes on to provide a vivid account of a tension between young and old that crosses caste lines and is the result of a generalized disenchantment with agrarian life. He (117) notes that “[T]he fact is that agricultural work is now regarded with deep distaste—especially by the young…even unemployed youngsters resolutely refuse to so much as supervise the work of day laborers in the fields, let alone work in them themselves…the young see agriculture as emblematic of the rustic world.”

17. Whereas the Patels ultimately chose to emulate Vanias, the Kolis adopted the Kshatriya model—particularly after their political mobilization in the 1950s and 1960s by the Gujarat Kshatriya Sabha (on the role of caste associations in the prosecution of Rajput Kshatriya identity, see Kothari and Maru 1970; Shah 1975; Lobo 1995). Like Patels before them, the Kolis retained the services of Barots to build Rajput genealogies for them (Shah and Shroff 1959).

18. Again, it is difficult to sift fact from fiction in this contest over identities. If anything, the struggle over rank exposes that an essentialist account of caste, one that sees it as fixed in time and space, is irretrievably flawed.

19. Breman (1985,; 1996) does an especially fine job of this in the Gujarat context. Karlekar (1995), Menon (1995), Saradamoni (1995), and Teerink (1995) document the difficult work environment of women migrants in various other parts of India, based on detailed ethnographic fieldwork. Pryer (1992) discusses the Bangladesh case.

20. At the time of our field research in 1993–1994, the Lodhas were paid five kilos of rice (dhan) and two rupees for tiffin at the end of the workday. They could make the cash equivalent of that wage and sometimes slightly more performing coolie work in the nearby market town of Jhargram, if such work was in fact available. When they took a head-load of fuel-wood into the bazaar, it usually fetched fifty to sixty rupees (three times the daily earning from farm or coolie work). But the preparation of such a head-load usually involved two days of work by a Lodha man and half a day of walking to the location of sale by a Lodha woman.

21. The Kangsabati canal project failed in part due to the enactment of the Forest Conservation Act in 1980 by the government of India and the subsequent refusal of the Forest Department to allow conversion of forest land into a watercourse. For details, see Sivaramakrishnan (1998).

22. We are grateful to the authors for granting us permission to cite from their preliminary record of findings.

23. The name “Vankar” is a synthesis of the Gujarati noun van (meaning “unginned cotton”) and the verb kar (“to do”); hence, a Vankar is someone who transforms cotton—a weaver.

24. The figures on educational attainment are drawn from Gidwani (1996,; 8, table 1.1). Remaining information is distilled from Gidwani's field notes on changing agrarian relations in the village of Shamli in the Matar subdistrict of the Kheda District, Gujarat, recorded between 8 May 1994 and 31 August 1995. Village and respondent names here and below have been altered to preserve confidentiality.

25. Hence, Gooptu (1993,; 277–98) describes how the Adi Hindu movement in Uttar Pradesh in the early twentieth century, which gained momentum with urbanization and migration, transformed the labor practices of the untouchable Chamar caste—although she is careful to qualify that the rejection of degrading menial work by Chamars did not translate into a pointed rejection of the caste system itself. In short, subaltern agency can express itself in ways that are simultaneously unconventional and conservative.

26. Field notes, Gidwani (1996). Names of villagers and villages have been altered to preserve confidentiality. Gidwani used the term “harijan” rather than “dalit” because this is how his respondents referred to themselves.

27. It is important to clarify that we view aesthetics as an aspect of politics. In so doing, we abjure a common view (sometimes linked to the works of Adorno and Horkheimer) that interprets aesthetics as a form of escape from (the despair of) politics. We want to assert the materiality of aesthetics and hence its expression not just in consumption styles but also in the recasting of the working body. Having clarified this, we clearly believe that body politics exceeds aesthetics, because the domination of bodies cannot—and should not—be reduced to aesthetics alone. That would diminish the physical and psychological violence of domination.

28. Ferguson (1999,; 95) writes: “I use the term cultural style to refer to practices that signify differences between social categories. Cultural styles in this usage do not pick out total modes of behavior but rather poles of social signification, cross-cutting and cross-cut by other such poles.” He uses the term “style” to emphasize the accomplished performative nature of such practices. Ferguson cautions that we should not assume shared cultural commonalities can survive persistent stylistic differences that may arise in situational behavior. Equally, we must be careful not to presume that similar styles signify similar things for different people (Ferguson 1999,; 93–97).

29. This body politics of consumption echoes the notion of “cultural production” advanced by Lisa Lowe (1996). As she (158) observes, “[C]ultural forms of many kinds are important media in the formation of oppositional narratives and are crucial to the imagination and rearticulation of new forms of political subjectivity, collectivity, and practice.”

30. The struggle over attire and lower-caste consciousness of the uses of attire in making defiant gestures is not a new phenomenon in Tamil Nadu. Andre Beteille's classic study ([1966] 1996) of a Thanjavur village contains evidence of similar assertion by young men returned from breathing the fresh air of self-respect in cities where they studied or worked, even as casual labor. The interesting twist in the story is the way the struggle over attire in the 1960s was about violating proscriptions about covering parts of the body or affecting upper-caste styles of traditional dress. Thirty years later the contest is conducted in cultural styles evoking modern fashions and cosmopolitan attire.

31. Dressing in styles considered as “hip” or “chic” in the metropolitan centers visited by migrants in their work-related travels often proves to be the most easily available form of self-assertion when they return to social relations of dependency and subordination in their home villages. Osella and Osella (1999) provide comparative endorsement of this point in their study of the low-caste Izhavas of Kerala. Their study also notes how low-caste groups adopt differentiated, rather than uniform, consumption strategies in their efforts to assert new identities.

32. According to Geoghehan (1873,; 71) more than 20 percent of overseas migrants returned home, braving the most appalling traveling conditions, between 1842 and 1870. So this aspiration clearly cut across historical periods and the spatial seams of the migration experience.

33. Comins (1892) provides detailed information on wealth created and returned by emigrants to India. The scale of repatriation led to a fear in colonial governments of a drain of wealth and skills from the island colonies.

34. This description and the preceding paragraph draw on Sivaramakrishnan's field notes for June 1999. A study is presently being conducted in four villages, including one tribal village, and three multicaste villages with significant dalit populations in Tiruvannamalai District, Tamil Nadu, India. One of these villages has been specifically selected for its periurban location less than ten miles from the district town. In addition to open-ended interviews and periods of participant observation, a baseline survey of socioeconomic characteristics was conducted in all four villages during 1999 and 2000, covering 1,534 families in the multicaste villages and 56 families in the tribal village. Information on migration stories was collected in the course of these surveys.

35. Our skepticism on the possibility of exercising “freedom” within the framework of biopower is offset by an altogether different skepticism on the part of theorists such Wendy Brown (1995, 63–64) who fear that Foucault, in fact, yields to a naive “volunteerism” that arises from his tacit assumption about the “givenness and resilience of the desire for freedom.” She (64) attributes Foucault's optimism about “freedom” to his “rejection of psychoanalysis and his arrested reading of Nietzsche (his utter neglect of Nietzsche's diagnosis of the culture of modernity as the triumph of ‘slave morality’).” We find Brown's psychoanalytic critique curiously totalizing, unlike Gramsci's understanding of subaltern “consciousness” as a fragmentary, bizarre combination of conceptions and goals sutured together.

36. Foucault (1977, 1979,; 1980) asserted that he was offering an analytics of power rather than a theory of power. This is entirely consistent with his picture of power as a capillary force that suffuses bodies and the body society. Since power is everywhere (there is no “outside”) it is, then, logically impossible to give an account of power that is objective or a priori, as theories claim to be. By the same token, it is difficult to imagine a subjectivity that does not serve power—and if this is so, in what sense is it meaningful to speak of human agency? Silvey and Lawson (1999) contend that migrant narratives contained in life histories, folk songs, poems and other nonconventional sources of information offer us a way of recovering uncolonized subjectivities. But this fails to address the problem that Dreyfus and Rabinow (1983,; 203) concede in their largely sympathetic appraisal of Foucault's work: “The force of biopower lies in defining reality as well as producing it. This reality takes the world to be composed of subjects and objects and their totalizing normalization. Any solution that takes these terms for granted—even if it is to oppose them—will contribute to the hold of power.” Some of the methodological problems in recovering subaltern voices are discussed in Guha (1983), Spivak (1988), Bhabha (1994), and Pratt (1999). Foucault evidently began to move away from his pervasive emphasis on power and tackle the analytics of agency in his later writings (the three-volume History of Sexuality [beginning in 1979] is thought to represent the beginning of these endeavors). Regrettably, his task remained unfinished.

37. Butler's effort to combine theoretical insights from Jacques Derrida, Foucault, and Luce Irigaray to discuss the possibilities of politically transgressive agency is noteworthy in this regard. She introduces a temporal factor in Foucault's notion of discourse through the idea of “citationality.” This simply means that the terms of discourse must be reiteratively and performatively reproduced by those who are enacted as subjects by that discourse. This injects the possibility of slippage during reiterations and hence, for Butler, the possibility that the realm of the “abject”—the constitutive and excluded outside—which is ordinarily “uninhabitable” and “unlivable,” becomes imagined as an alternative to some normalizing “regulative ideal” (such as heterosexuality). But despite her rejection of voluntarism, even Butler seems to succumb to the need to smuggle in a moment of “free will” in order to argue the possibility of resistance. See Butler (1993, introduction; 1997, ch. 3). The work of Robert Sack (1997) offers provocative insights on the necessity of “free will.”

38. As Chakrabarty (2000, introduction and ch. 1,; 8) explains, historicism spawned a “not yet” or “waiting room version” of history on the part of colonial rulers—and even liberal intellectuals such as John Stuart Mill—who maintained that some “historical time of development and civilization (colonial rule and education, to be precise) had to elapse before they [Indians or Africans] could be considered prepared for [the task of self-rule or government].”

39. See Simon (1997), Moore (2000), Sivaramakrishnan and Agrawal (forthcoming), and Gidwani (forthcoming) for detailed critiques of postdevelopment.

40. Circular migration has a long history in the Indian subcontinent. Rarely, however, did it permit subaltern classes to repudiate traditional hierarchies and codes of conduct. But today, the spread of television, city editions of newspapers, retailing, transportation, and state-sponsored literacy programs has hugely expanded and accelerated the flow of materials and meanings and granted new potency to consumption as a source of counteridentity. Similarly, the political space in which subordinated groups like dalits and tribals can operate has also widened with the entrenchment of state affirmative-action programs and the rise of panregional dalit movements that recognize the influence dalits can exert in a democratic nation-state via their immense collective voting power. These sorts of structural changes in postcolonial India make subaltern “body politics” possible and effective.

41. For a discussion of the second democratic upsurge, see Yadav (2000). As he points out, this needs to be juxtaposed with the “first upsurge” in the 1960s, which witnessed the widening and deepening of political participation in the country through the channels of strong middle-of-the-road parties.