Imperial legacies and southern penal spaces: A study of hunting nomads in postcolonial India

Abstract

Southern penal spaces are marked by resemblances and affinities with colonial regimes of control, yet they also reflect quite distinctive postcolonial social and political dynamics found in the global south. Here, legacies of control, forms of exile, status reductions, hierarchical social stratifications and other like forms come together in robust modes of containment suitable for managing ‘marginal’ and ‘suspect’ populations. We draw on ethnographic empirical work with two hunting nomadic groups in India by two of the co-authors who are working with the Kheria Sabar community in Purulia district in West Bengal and Pardhi community in Mumbai. The latter were subject to notification under the notorious Criminal Tribes Act 1871, marking them out as ‘criminal tribes’ until their de-notification shortly after India's independence in 1947, yet the Kheria Sabars too feel its effects. We draw attention here to the continual negotiation and (re)fabrication of both state and citizen at the point of their everyday contact. Our notion of southern penal spaces contributes to penal theory by breaking from northern societies’ focus on institutional carcerality and capturing instead both the variety and the dispersal of penal and punitive practices found in postcolonial societies of the south.

Introduction

Recent years have witnessed new interest among criminologists to understand penal practices in countries of the global south. In an attempt to distinguish the special character of punishment there and to recognise the long and complex colonial inheritance shaping systems of law, governance and the social orders of the south, it has been suggested we should seek to understand and theorise their distinctively ‘postcolonial penalities’ (Brown, 2017). Other scholars have drawn renewed attention (cf. Cohen, 1988; Sumner, 1982) to north–south penal transfers (Stambøl, 2021) and in doing so have coined terms such as ‘penal humanitarianism’ (Bosworth, 2017; Lohne, 2020) or ‘penal aid’ (Brisson-Boivin and O’Connor, 2013) and spoken of the ‘“the fiasco of the prison” in a global context’ (Drake, 2018: 1). In this special issue on Legacies of Empire, we write as a team of three southern scholars and one northern, aiming to advance criminological work not on penal transfers but on legacies, inheritances and forms of penal power in the contemporary postcolony. Our area of focus is colonial and postcolonial India and our particular target is the colonial birth and postcolonial life of a diverse set of penal practices directed toward people who from 1871 until after Indian independence in 1947 were legislatively designated as ‘criminal tribes’: criminality, in other words, as birth right, as inheritance, as mark, or as has sometimes been observed, as stain.

The colonial criminal tribes policy itself has attracted increasing interest over the last three or four decades leading to a small specialist literature principally turning on questions of historiographic importance (see generally, Major, 1999; Piliavsky, 2015; Radhakrishna, 2001; Studies in History, 2020; Yang, 1985). Our interest is somewhat different. To describe it, we first briefly describe two studies that provide our point of departure and set the context for this work. We then lay out our intended contribution and the plan of the paper. Our effort to connect a period of colonial rule in India that began in 1765 with contemporary life in 2021 builds on two previous works on the criminal tribes policy by one of us. The first (Brown, 2014) traced an arc from the enactment in June 1772 by Governor General Warren Hastings of Article 35 (a Bengal regulation directed toward dacoits – violent gang robbers – and their families), to the passage almost a century later of the Criminal Tribes Act 1871, and then onward to the extension of that legislation onto an all-India footing in the early 20th century. The focus of that work was upon penal power as an important modality of colonial rule. What it illustrated was how the armature of control established over those roughly 150 years of colonial governance was at once determinedly penal in character, yet frequently not institutionalised in typical penal forms. As an archetypal example of expansive colonial penal power, the criminal tribes policy ran alongside the ordinary penal law, as reflected in the Indian Penal Code 1860, Criminal Procedure Code 1861 and other key legislation, plus the system of ordinary policing and prisons (see Indian Jails Committee, 1920). The notification of groups and individuals as being or belonging to ‘criminal tribes’ was a penal response to a social problem that included criminality, but that was principally marked out by social marginality, by mobility and by resistance within certain segments on the population in rural north India to incorporation into the new sedentarised civil society and economic structures being fashioned under British rule.

The present work builds out also from a second study, Brown (2017), which examined the immediate postcolonial moment in India when a new, independent nation faced an opportunity to throw out such egregious examples of colonial oppression as the Criminal Tribes Act. While Jawaharlal Nehru (1947), independent India's first prime minister, spoke on the eve of Independence of the country awakening ‘to life and freedom’, provincial governments that had inherited the control architecture of colonial power in fact proved far from happy to let it go. To review the need for such deeply illiberal control measures a Criminal Tribes Enquiry Committee, known informally as the Ananthsayanam Ayyangar Committee after its Chairman, was established and toured the country during 1949–1950 collecting evidence. Reflecting the fact that colonial legacies cast their shadows in thought and attitude as much as in institutional structures, the Enquiry Committee struggled with the idea that those notified as belonging to criminal tribes might be owed the same freedom and liberty as all other Indian citizens. Certainly, it recognised that the notion of criminality by birth was inconsistent ‘with modern conceptions of right and justice’ (Criminal Tribes Enquiry Committee, 1951: 90). It also noted an ‘almost unanimity of feeling in the country among all sections of the people that the Criminal Tribes Act should be repealed as it brands members of certain communities as criminal by birth’ (p. 90). Nevertheless, it argued that ‘there is equally a large demand for some kind of control and restriction over the habitual offenders, to whatever community they may belong’ (p. 90). Moreover, like colonial rulers of the past, the Enquiry Committee rejected the notion that those currently captured within the criminal tribes apparatus could be dealt with under the ordinary criminal law. It noted, for example, that surveillance of ordinary citizens could not be achieved in the way it envisaged being necessary and nor would ordinary criminal law empower governments ‘to order the segregation of the children of habitual offenders from their parents, where such action may be desirable’ (p. 99). As such, the Committee recommended the repeal of the Criminal Tribes Act, duly achieved on August 31, by Act XXIV of 1952, and its replacement by Habitual Offender legislation that would achieve much the same effect but with a new grammar of habituality over writing the repugnant notion of born criminality.

Now, almost 75 years later, what has become of those communities once notified as criminal? What we do know is that the Criminal Tribes Act's repeal in 1952 did not instantly lift the spectre of criminality from their shoulders. Instead, they came to be known next as denotified tribes, or DNTs. This is where our present study begins. In it, we seek to make two primary contributions to work on contemporary global penalities in general and imperial legacies in particular, one empirical, the other theoretical. Empirically we seek to document and render visible the complex web of contemporary controls and disabilities experienced by members of DNT communities. Theoretically, we aim to develop further and indeed flesh out the term ‘postcolonial penality’. We do this by harnessing our empirical work to advance an idea of ‘southern penal spaces’, a notion that we think captures something of the distinctiveness of penal power in this part of the global south. To that end, the remainder of this article is divided into four parts. First, we position ourselves theoretically, describing what we hope will be the unique contribution we make to studies of contemporary penalities. The aim here is not to develop fully our thinking but rather to set out the intellectual contours of our work so as to frame the empirical contribution that follows. Second, we describe very briefly and, within the constraints of space here, necessarily schematically, the criminal tribes designation, the wider contours of its control apparatus and, importantly, the merging of criminality and nomadism into the hybrid DNT category in the years following independence in 1947. Third, we then move to two case studies of contemporary life, experience and the webs of control and disability within which two hunting nomadic groups – the Kheria Sabar community in Purulia district in West Bengal and the Pardhi community in Mumbai, Maharashtra – find themselves. Illustrated in this section is the way penality here extends far beyond the formal machinery of justice and deep into what we describe as the penalisation of lives and life-ways: of ways of being in the world. Finally, we draw these data on the controls that horizontally and vertically grid the spaces of DNT communities’ lives together with the theoretical framing described in section one to advance our key theoretical contribution, the idea of southern penal spaces. We conclude with some reflections upon the complex interplay of colonial legacies and postcolonial modernities that mark out penal power in sites across the global south.

Advancing penal theory: Southern penal spaces

Let us begin by unpacking this term southern penal spaces. When we speak of the south we do not refer the American south, although we recognise that this region has long been seen as distinctive in terms of criminal problems and penal responses (Borg, 1997; Currie, 2017; Michelle Brown, 2021). We are instead concerned with the global south, a term not entirely self-explanatory in itself, but something Nour Dados and Raewyn Connell (2012: 13) describe as capturing a certain part of the globe that has experience and ‘history of colonialism, neo-imperialism, and differential economic and social change through which large inequalities in living standards, life expectancy, and access to resources are maintained’. Thus, while we might have followed Brown (2017) in speaking of ‘postcolonial penalities’, we have sought instead to extend and elaborate that idea by drawing on the south as both a spatial zone and as a repository of cultural, economic, legal and other histories. Reflecting the theme of imperial legacies that organises this special issue, many aspects of these histories will be of colonial origin, but many also will not and it is in the intricate interplay of old and new, endogenous and exogenous, robust and subtle, that we situate our work.

Turning to the term penal, we note recent developments energised by the cross-disciplinary engagement of criminologists with geographers and the emergence of a new field of ‘carceral geographies’ and their attention to ‘carceral spaces’ (see generally, Moran, 2015). Moran and Schliehe (2017: 3) observe that ‘The recent development of carceral geography is directly related to the ‘spatial turn’ in criminology, and to the spatialisation of carceral studies’. We will deal with the spatial element in a moment. For now, we note that while our work and theirs shares important themes, we are less convinced of ‘the carceral’ as a sufficiently broad notion to capture the subtle arrangements of penal power in southern places. Indeed, geographers and criminologists working in this field seem to have become shackled to the terminology of the carceral even as they increasingly strain at its logical limitations. These constantly draw studies of the carceral back to the prison, stretching its connection with an increasingly diverse set of practices (Moran et al., 2018). At least partly for this reason we prefer to speak the penal and of penal spaces. However, we also prefer this for the important reason that the prison itself does not lie at the centre of this punitive order. Criminalisation and penalisation remain key, but as often as not the prison is peripheral to the harm and control that we seek to isolate from its messy background and so to capture and make recognisable. This is partly what made colonial penality distinctive. On the subcontinent, it was not first or principally an institutionally based set of practices and supporting discourses centred on the prison, just as it was also not typically disciplinary in character (Brown, 2014). And nor, as we discover, is it that way in the postcolony.

Finally, we must speak of space, spaces and spatiality. Andrew Jefferson (2014: 49) has argued, persuasively we think, that ‘to understand the experience of confinement we must look not only at institutions or sites but also at practices and meanings, or more crucially at the relations between sites, practices, social relations and subjectivity’. We extend his thinking on institutional penality into wider penal spaces. We understand these as assemblages of social, institutional and subjective activity that capture, control, position and target certain groups or types of persons. Connecting with Danielle Moran’s (2015: 110) work on carceral space, we seek to identify the contours of ‘a punitive state which operates in places far beyond the prison through pervasive and pernicious policies which incarcerate and confine without actually imprisoning’. So it is to these kinds of penal spaces we attend in this paper and, specifically, to the way in which southern penal spaces emerge in the global south as genealogical descendants of colonial regimes yet grow under distinctly postcolonial conditions. Movement and mobilities will form an important part of this. Indeed, the notion of mobility encourages us to attend not only to horizontal movement through space – such as Pardhis, as we will see, as they circulate through the peripheral spaces (pavements, underbridges, roadside camps) of urban Mumbai – but also vertical movement, or the social mobility – or lack thereof – of DNT communities seeking to escape their precarious social positioning. We follow Jennifer Turner and Kimberley Peters (2017: 99) in valuing the concept of mobilities for keeping us ‘attuned to the messy, complex, contradictory, unmappable realities of how, where, why and by what means people move or are unable to move’. Mapping out the forms of control that grid the spaces of DNT communities’ lives horizontally and vertically is part of our objective here.

Governing the social margins: The criminal tribes policy and DNT precarity

The Criminal Tribes Act 1871 emerged out of a series of experiments undertaken in northern India from the mid 1850s to deal with what were perceived as the triple problems of mobility, invisibility and what was variously termed hereditary or professional criminality. The most formalised was the Kot system in the Punjab established under Book Circular, No. 18 of 1856, later struck down in 1867 by the Punjab Chief Court as unlawful. ¹ As a broader feature of political economy in India, the ‘problem’ of mobility was not unique to matters of law and disorder but it did create vexing problems for policing, which the Kots, settlements designed to hold in place ‘wandering’ criminal groups, had attempted to address. Such movement also rendered criminals less visible, for in the great flux of circulations in India they easily blended into the background, complicating further still efforts to prevent crime. Finally, there was the question of why certain Indian groups committed crime as they did. Colonial administrators first drew upon the same loose grammar of hereditary and professional conduct used ‘at home’ in Britain to describe subcultures with typical activities and supposed profession. In India, however, such ideas of social habituation found a timely fit with emerging notions of how the Hindu caste system actually worked (see Dirks, 2001). With caste increasingly understood as profession, the uniform application of caste categories to Indian communities made it a major unit of analysis in studying the characteristics of a supposedly rigid and timeless Indian social order.

In this emerging ‘ethnographic state’ with its clear line linking caste and profession, W. Nembhard, the Commissioner of East Berar, voiced a fairly unremarkable view when in May 1870 he wrote in support of the Criminal Tribes Bill:

We all know that traders go by castes in India; a family of carpenters now will be a family of carpenters a century or five centuries hence, if they last so long, so will grain dealers, blacksmiths, leather makers and every other known trade. […] If only we keep this in mind when we speak of ‘professional criminals’ we shall then realize what the term really does mean. ²

These words were later seized upon by James Fitzjames Stephen, Home member to the Government of India, as evidence of the need for some special measures, beyond those available under the ordinary criminal law, to counter this apparent menace posed by this newly recognised Indian type, the professional criminal tribe. Seven important features of the Criminal Tribes Act 1871 may be noticed. First, it constituted a piece of extraordinary legislation, sitting atop the ordinary criminal law. As late as 1918 the Secretary of State for India would continue to describe it as a piece of ‘emergency legislation’. ³ Second, it was an all-India Act, but was limited for a long time to just three local government areas across northern India: Oude, the North Western Provinces, and the Punjab. Third, it asked that such local governments provide evidence of ‘any tribe, gang or class of persons … addicted to the systematic commission of non-bailable offences’ (s.2), satisfaction of which would allow them to be notified as a ‘criminal tribe’. Fourth, once so notified, a range of restrictions could be imposed, including settlement of a nomadic group or restriction in their own village of a settled group. A system of roll calls and passports, the grounds on which the Punjab Kot system had been deemed unlawful, were now regularised in law. Further, after 1897, amendments would allow the removal of children from parents and the confinement of notified tribes in reformatory or industrial settlements. Fifth, any tribe, gang or class once notified found themselves beyond the reach of the very field of colonial law that had captured them. Section 6 of the Act instructed that ‘No Court of Justice shall question the validity of any such notification’. Sixth, the Act provided for sentence enhancements to any convictions made under the ordinary criminal law: a second conviction for a scheduled offence would attract a mandatory 7-years’ imprisonment; a third offence, transportation for life. And finally, but no less importantly, the whole architecture of the Act rested on a system of village surveillance. Since colonial officers could not be everywhere at once and, moreover, reflecting their apprehensions about the invisibility of criminal tribesmen against the background of their social milieu, village lambadars (head men) were saddled with the responsibility for keeping track of those notified under the Act and reporting absences to the police.

There is not space here to describe the operation of the Act, save to note that it differed in important ways geographically. Monograph-length treatments of the north Indian (Brown, 2014) and central and south Indian (Radhakrishna, 2001) experiences illustrate not only distinctive early- and later-years differences (respectively) in the nature of the legislation itself, but also what seems to have been a broader tendency in central and southern India, where the Act was only applied much later on, for it to be used against nomadic and forest-dwelling communities displaced from place and livelihood during the last decades of colonial occupation (Bokil and Raghavan, 2016).

How then, in the postcolonial life of measures directed toward now ‘denotified’ criminal tribes, did criminality become so strongly associated with nomadic and forest-dwelling communities? We offer two suggestions that we hope will help further frame and situate our work. First, from around the same time as the criminal tribes policy was taking shape, the Government of India created first a Forest Department in 1864 and then a hastily drafted Indian Forest Act of 1865, both of which aimed to deal with the unregulated environmental destruction brought on by spiralling needs for hardwood for both military (e.g. naval) and civil (e.g. railroad) applications. Madhav Gadgil and Ramachandra Guha (1993) provide one of the most detailed accounts we have of how the colonial state now ushered in the same dispossession of customary usage rights that had marked the great ‘enclosure movement’ in Britain in the early 18th century (see Thompson, 1975). Across India, peasants’ forest commons and ancient rights of hunt, forage and harvest were pushed aside with new visions of forest as property and of ‘productive use’ of forest assets advantageous to colonial power. Through various iterations leading to the Indian Forest Act 1927, colonial authorities claimed increasingly wide state jurisdiction over forests and forest produce, bringing them into conflict with forest-dwelling communities and other groups such as nomads and other tribal groups traditionally dependent on forests.

Second, we may look to the dispersal of control that began during colonial times and has only expanded and hardened in the postcolonial era. This notion of dispersed measures of control is key to understanding the architecture of penal control we are concerned with here. Studies of the criminal tribes policy have concentrated primarily on the Criminal Tribes Act itself and its operation and demise. Yet as we saw above, the enclosure of forests and the restructuring of longstanding political economies, forms of circulation and mobility, and the shift from subsistence life to waged labouring were key features of the colonial reorganisation of Indian society. That reorganisation did not stop at Independence. Newly denotified criminal tribes were haphazardly re-scheduled alongside formerly ‘untouchable’ caste groups or nomadic and migratory tribes as constitutionally Scheduled Castes (SC), Scheduled Tribes (ST) or Other Backward Classes (OBC) or, in some states such as Maharashtra and Tamil Nadu, locally as Vimukta Jati (liberated or freed caste) Nomadic Tribes (DNT/VJNT) (see Idate Commission, 2017). Yet not all seemingly eligible groups were so classified, and as we will see, attempting to join such lists and the positive-discrimination social-uplift measures they hold the key to is far from straightforward.

The present project: Ethical considerations, case studies, histories

Fieldwork for this study was undertaken by two of the authors with two hunting nomadic groups in different parts of India: one, an ethnographic engagement with the Pardhi community in urban Mumbai, Maharashtra, and the second, a field action project with the Kheria Sabar community in rural Purulia, West Bengal. Another of us (Raghavan, 2013: 265) has observed that the ‘field-theory linkage’ that the latter projects develop are a key element of both professional education and university-led social advocacy in the country and have been highlighted by India's University Grants Commission as a best-practice model (Raghavan, 2013: 287). Data collected under both projects complied with ethical standards and best practice guidelines established by the two universities – Tata Institute for Social Sciences (TISS), and Jawaharlal Nehru University – under whose auspices the researchers worked. However, there is more to say of ethical matters than just that. Engagements such as these are typified by marked hierarchical power relationships that require acute attention to the personal ethics of practices of enquiry while, at the same time, that enquiry may be bound up within a broader relationship that is not purely extractive of research data. With respect to the field action project of TISS with the Kheria Sabar community, it may be noted that these are demonstration projects to pioneer initiatives in the field by working with marginalised sections of the community and identifying services that can be built into policy through long-term intervention. They highlight the role that higher educational institutions can play in engaging with marginalised communities from a constructivist viewpoint that strives for a phenomenological understanding of social phenomena (Dave et al., 2012). By their very nature, the ethical aspects of intervention are woven into these projects as they are aimed not to extract data from participants but to work with them over a period of time in attempting jointly to improve their lives at field and policy levels. This was further reflected in the present case with the co-production with the community of suitable research questions to draw out aspects of their lives, experiences and world.

The personal ethics of practice in relationships such as these requires sensitivity not only to what participants may wish to put on or off the record. In fieldwork relationships characteristic of long-term ethnographic enquiry, as was the case of the Pardhis of Mumbai, questions of boundaries inevitably also arise. Yet these are far sharper and more ethically weighty when working with highly vulnerable and deeply marginalised communities and individuals. In particular, this necessitates an appreciation of the extent to which individuals such as homeless pavement-dwellers expose their private as well as public lives to the outside world in ways over which they have little or no control. Layered atop this are abject experiences of police brutality, destruction of their makeshift shelters and offences again their personal dignity. To observe is thus to enter realms of privacy that one might not ever to wish pass, and so ethical practice demands a highest respect and constant vigilance to protecting the dignity of both individuals and the community itself.

Historically, certain sub-groups of the Pardhis were notified as criminal tribes in the Bombay Presidency and elsewhere. The Kheria Sabars, by contrast, consider themselves to be a former criminal tribe, yet there is no evidence of that ever having been so. Gupta (2011) suggests that since Kheria Sabars are often grouped together with Lodhas, who she claims were notified as a criminal tribe, the self-perception and stigma of criminality has crossed to the Kheria Sabars. Yet Gupta's claim regarding the Lodhas is only partially correct. For, as the Criminal Tribes Enquiry Committee (1951) noted, only two tribes were notified in West Bengal: the Bediyas and Karwal Nats. Members of any other tribe convicted under the Criminal Procedure Code 1861’s preventive provisions might, however, be notified individually as members of a criminal gang, their tribe duly recorded, and the Act's apparatus used to manage them. This illustrates how the Act's application to a few individuals within a tribe (in this case data indicate probably no more than a handful of Lodhas) created far wider ripples of impact. In fact, though, this is far from unusual and reflects the complex postcolonial intermingling of nomadic and denotified status, of memory and forgetting, lore and law. Varun Sharma (2020) found the same when he came to study the Pardhis of Chhattisgarh, a day's drive south-west of Purulia. The community there had welcomed their being listed as a denotified tribe by the Idate Commission (2017), though they had no memory themselves of ever having been notified and the archives of British administration show no evidence of it either. Yet, as Sharma notes, the historical fact of never having been notified as a criminal tribe has in no way hindered what he terms ‘the arrival of certain colonial logics in the forested pockets of Chhattisgarh, particularly the “police point of view” and the conceptualisation of “criminal classes”’ (p.102). What this reflects, he argues, is the capacity all these years later for the ‘colonial programme to enter and anchor itself in the region, giving the Pardhis a “history” other than their own’ (p. 102).

Fieldwork in Purulia, West Bengal, was undertaken between February 2019 and November 2020. Data reported in this article flow from an on-going household survey (n = 2182), focus group discussions using participatory methods conducted in three villages (involving around 20 participants in each village) and interviews with key community leaders and tribal activists. In Mumbai, fieldwork was conducted with Pardhis between June 2018 and July 2019, with further occasional visits through December 2019. The study used everyday observation, interactions with the community and non-community people, group discussions, a family survey (n = 621) and semi-structured interviews with 30 Pardhi participants at three sites named here: Musafir Nagar in Malad, Raj Mahal Pool in Chembur and Sadak Bazar in Colaba. ⁴ In addition, field notes of institutional visits to police stations, beggars’ homes and the children's homes are used to elaborate the findings. The tracing of Pardhi dwellings in Mumbai was done with the help of Pardhis at the above three sites, Koshish, a field action project of TISS and Ghar Bachao Ghar Banao Andolan, a civil society organisation, which assisted in a survey of 2105 families and helping form an estimate of the Pardhi population at 10,471.

The case of Kheria Sabars in Purulia

We begin our two case studies with the case of the Kheria Sabars, a tribe who still draw upon their local forests for sustenance and income in part of the year. At the same time, however, their progressive squeezing out from these spaces since the 1960s has pushed them onto the fringes of village life and into new forms of migration, finding themselves participants in a ‘wage’ economy so precarious that parts of it sit outside the cash economy altogether. In the second case study, we will hear of the Pardhis of Mumbai, a group whose migrations have pushed even further out, into a kind of urban nomadism, and we will focus in that case on the more formal, intersecting controls, constraints and forms of penalisation that grid their lives. Here, in the forests and agricultural hinterlands of West Bengal, we focus instead on what Varun Sharma (2020) described as the easy slippage occurring from criminal tribes to criminal classes and the rehearsal of old colonial logics and implicit assumptions and biases about such groups. Further, the mutability of such biases becomes apparent as a nomadic hunting tribe, emasculated of its former social, economic and ecological niches, oscillates between the stigma of criminal caste and the reductions of caste untouchability. In the postcolonial era, as we will see, colonial legacies mix with contemporary failures of democratic governance.

Failure of social uplift: Kheria Sabars’ downward mobility

One of the first moves of the new postcolonial Indian state was to introduce constitutional recognition of scheduled castes and tribes, identifying them as special and important targets of governmental action for social uplift. Based on 2001 census figures, the Kheria Sabars form just a small fraction (1%) of West Bengal's scheduled tribes, who themselves represent just 5.5% of the state's population. Even among their peers, the Kheria Sabars are among the most marginalised. The same census, for example, ranks them last among scheduled tribes in terms of literacy, with 36% of males and just 16% of females aged above 7 literate (Census of India, 2001). The period since independence has also been one marked by livelihood destruction. Jaladhar Sabar, President of Kheria Sabar Kalyan Samiti, a Kheria Sabar organisation, recalls his grandparents being largely dependent on forest produce for sustenance. However, over the past 50 years conservation and forest management regimes, alongside the rapid depletion of forests themselves, have forced Kheria Sabars out of forests and onto the fringes of villages. Research for this case study estimated that some 114 Kheria Sabar hamlets have emerged in Purulia District in this way. The availability of forest lifeways being absent, the Kheria Sabars must either purchase land for tillage or enter the precarious wage labour economy. A household survey reveals just 578, or 26%, of families in the district possess any land. For the remainder, new forms of mobility beckon. In fact, 1180 families in the district report regular migratory searches for work, indicating that even for those holding land, supplemental wage labour is necessary. Focus groups in villages of Purulia reveal complex and unstable migrations, mixed with periods when collection and sale of forest products, such as wood, will still be possible. However, the stigma of criminal class limits migrating Kheria Sabars to only the most menial work. Earnings for a breadwinner are typically around INR 150–200 per day, or under 3 USD/2 GBP. Below this lies the sub-cash economy where Kheria Sabar labourers receive wages only in kind, such as in a share of harvested crop, instead of money payment.

Intersectional stigma: New biases multiply colonial continuities

As government management of diminishing forest resources pushes hunting nomadic groups like the Kheria Sabars progressively further from traditional lifeways, they are thrust not only into the precarious wage labour system just described, but also into the experience of both old and new forms of stigma. The pall of prior notification as a criminal tribe, even if slim evidence for that exists anywhere beyond local lore, colours their experience of life in contemporary India. Added to that, Kheria Sabars face discrimination based on various other social parameters reflecting their forest dwelling origins and the impact of their new contact with settled communities unaccustomed to their ways. Kharu Sabar, a key informant to the study who is also a Sabar tribal activist, describes how their community is infamous for its eating habits, which consist of a diet of rats, snakes and other wild forest animals. Such food habits provide grounds for a kind of untouchable status to be accorded to them, excluding them from social practices, such as festivals or the sharing of food and water with them and forcing them to reside beyond village peripheries. Focus group discussions with a roughly gender equal group (slightly more women) in one village indicated the extent of discrimination that colours and defines Kheria Sabars’ life. Asked how often different groups discriminate against them, participants responded ‘always’ or ‘often’ to government bureaucracy, religious leaders, and teachers (all 100%), affluent villagers (93%), police (90%), shopkeepers (80%), doctors (73%) and politicians (70%).

For Kheria Sabars, life as a DNT community in postcolonial India has thus produced anything but uplift. Their mobility has been constrained vertically as they migrate downward from subsistence forest life to penury on the fringes of rural villages, at the same time as it has forced them into wider migratory circulations in search of subsistence wage labour. Yet for all of this, they still retain something of their place, albeit constrained and possibly terminal, on the edges of the forests that were once their base of life as hunting nomads. In our next case, we move to Mumbai, a modern mega-city with an estimated 21 million inhabitants (UN 2021), including another such nomadic tribe, the Pardhis.

The case of Pardhis in Mumbai

The word ‘Pardhi’ originates from the Marathi word ‘Paradh’, which literally means prey, indicating the livelihood based on hunting. The description of Pardhis in colonial accounts terms them hunters and wanderers, pointing to their nomadic life patterns and distinguishing different types, such as Gaay Pardhis – who rode cows, Chittar Pardhis – engaged in hunting antelope and Phase Pardhis – noose hunters (Kennedy, 1908). Although we do not concentrate here on the complications of the postcolonial administrative identities provided to nomads once notified under the criminal tribes legislation, it is nevertheless crucial to note the impact of the label of ‘tribe’ itself. The ideal–typical construction of the category ‘tribe’ includes a common presumption against the possibility that tribes may dwell in the urban context (Srivastava, 2008). Thus, what is in fact a longstanding phenomenon of tribals migrating to and dwelling in cities remains a largely ignored and invisible one (Radhakrishna, 2007, Srivastava, 2008). Together with a range of other factors that will become clearer as we go along, this creates considerable difficulty in estimating population numbers. Agrawal and Sinha (2012) found estimates of the population of Mumbai Pardhis to vary from 5189 to 9600 but pointed to likely underrecording. Fieldwork that forms part of the present study identified 2105 Pardhi families who were traced in 35 settlements across Mumbai, constituting a population of 10,471.

Urban mobilities: Place, circulation and risk

In the largest settlement of Pardhis in Mumbai – Musafir Nagar in Malad, where 494 families were based - many inhabitants reside in-place for only a limited time. Instead, they circulate through a variety of spaces, living with other community members elsewhere in the city, at homeless sites or on roadsides close to their current work. Such peripatetic practices seem to reflect nomadic characteristic that find new form in the context of Mumbai. Moreover, there is a conflict of opinion among the Pardhis regarding the virtue of their dwelling together in Musafir Nagar, since what might appear at first blush to be a settled home in fact functions equally well as a key interface with police. Here we quote from a discussion that took place at Musafir Nagar: Rajabhau:

⁵ It is good that we have our own people staying around. It provides us apnapan (sense of affinity/community).

Hambbir:

But there is too much of our population here which has made entire Musafir Nagar known as Pardhi basti. ⁶ Police are on regular patrolling in this area.

Tanna:

For police to catch hold of a Pardhi, all they have to do is ask the full name of the person here. If the surname of the person happens to be Pawar, Kale, or Chawan then they just pick up the person for enquiry or sometimes keep under custody, if they are found late night on the roads or some crime has taken place in the nearby by locality (Field Notes: Dec 2018).

Moving on, therefore, offers a measure of safety from police harassment. Yet even this has its limits. Kanta Tai, a Pardhi woman of about 60, reports that her family was also harassed by police when they stayed on the streets and at other homeless sites, pushing them into further nomadic circulations through the city where it was difficult for the police to keep a regular watch on them. The risks of police attention involve more than just harassment. In one instance reported to the researcher, the police picked up a Pardhi man in his mid-40s from the basti early one morning on suspicion in connection with a case of house-breaking and theft. The man was purportedly beaten to death in police custody, with the police later claiming it was a case of suicide. In another cases, police are reported to have invoked their own, brutal, summary justice in lieu of the more time-consuming process of arrest and charge.

Contested space

Part of the weakness of the position of Pardhi communities finds themselves in lies in the contested nature of the spaces they occupy: informal settlements, roadside camps, pavements, underbridges, beaches, and the like. Being an illegal slum, Musafir Nagar has been demolished by municipal authorities on multiple occasions. Yet resistance by slum dwellers to their homes’ destruction has been met by the municipality with cases filed against Pardhis under the Indian Penal Code Section 353, which criminalises action deterring a ‘public servant from discharge of his duty’, bringing with it the financial burdens of finding bail recognisance or paying fines. ⁷ To better understand the extent of resulting precarity experienced by Pardhi communities as they settle and circulate within Mumbai, 35 Pardhi dwelling sites were eventually traced. Of these Pardhi groups, 68% had adopted dwelling patterns that reflected complete homelessness: sleeping on footpaths, under bridges or outside railway stations. A further 14% dwelt in irregular slums, while 6% had huts in regularised slums and 3% each respectively in illegal hutments in regularised slums, partially regularised slums, transit camps, and apartments.

Citizenship: Legal identity and (in)visibility

As a denotified nomadic hunting tribe that has migrated from forest life to urban precarity, we are building here a picture of the way penal control grids their existence and how it does so in ways that defy many of citizenship's key protections. Among these, at the pinnacle one might say of citizenship, is to be recognised by the state to exist. In other words, to have legal personhood and identity. The capacity to claim any number of state benefits, even to participate in the most fundamental way – to be able to report a crime against oneself or family to the police – rests upon having legal identity. While the population census has been extensively critiqued as a tool of colonial control, it remains important today in India for fixing entitlements and a ‘nomadic’ community in the metropolis quickly finds that what defies census designers’ assumptions (such as ‘tribes’ in cities) will be rendered thereby invisible. The failure to count urban Pardhis is the first point in a cascading set of difficulties that undermine the community's capacity to claim elementary public goods and entitlements.

Local government efforts to mitigate the precarity or urban nomadic life also struggle to overcome problems of legal identity. For instance, Government of Maharashtra Resolution No. सविव्य - १००५/१६१२/प्र. क्र. १९६३/नापु -२८ of 12th September 2008 allots ration cards to the Pardhi community in recognition of their nomadic dwelling practices. Fieldwork at two of the main Pardhi living sites suggests, however, that Pardhis have frequently been unable to receive ration cards under this regulation and instead have had to rely on their abject homelessness to secure them. Two important findings emerge here. First, the invisiblisation of Pardhis in Mumbai is a consequence of them not possessing elementary documents such as a caste certificate that would recognise them for who they are. On the extreme social margins that Pardhis occupy, the evidentiary requirements of such a document often are insurmountable: a school leaving certificate: the semi-settled/nomadic life lived by the Pardhis in Mumbai rarely allows them to enrol children in school; a domicile certificate: yet most Pardhis either are homeless or live in irregular slums, as indicated above. Second, this leads to the imposition of a homeless identity onto, and into contest with, their nomadic culture and identity. Seen together, the state's failure not only to resolve but even to attend much to problems of DNT’ citizenship status squeezes and constrains their mobility in important ways. They are caught in an eddy, so to speak, circulating continuously through spaces on the precarious margins of society, yet held in place in terms of vertical mobility, unable to claim even the most basic entitlements of citizenship and social uplift.

Webs of control: The criminalisation of life-ways and parenting

It is clear that while certain instruments of government key to the distribution of public goods invisibilise the Pardhis, others, focused on control and criminalisation, precisely locate and identify them. One example is the Maharashtra Prevention of Begging Act 1960. Section 2(1)(i) of the act, defining ‘begging’, equates certain traditional peripatetic nomadic occupations such as fortune telling (commonly practiced by the Vasudev and Nandiwale tribes in Maharashtra), street performing (engaged in also by communities like Madaris, Dombaris, Mang-Garudis, Nats, Makadwalas) with begging, and thus defines such labour as criminal. The section targets anyone who performs on the street entertaining people for money, which has long been a traditional livelihood of nomadic communities. A beggar may first be sentenced to not less than two but not more than three years in an institution for beggars. However, a second conviction requires a mandatory 10 years incarceration, part of which may be spent in a prison. The Act provides immense discretionary powers to the police to decide who is allegedly engaged in begging, who is likely to beg, and how they shall be charged or otherwise dealt with (payment of bribes being a common route out of this trouble). Pardhis, other traditional nomads and homeless migrants in the city all become targets of arrests and essentially arbitrary, long-term detention under this law. All of this is clearly penal, but the prison resides only on the fringes, as a last resort for those unable to be managed in beggar homes. The institutional arrangements to implement the Act, such as the establishment of these beggars’ homes and anti-beggary police squads are key to the control targeting of urban nomads. At Raj Mahal Bridge in Chembur, regular raids by the police lead to the demolition of Pardhi dwellings. Four such raids were conducted by police during the fieldwork period in which Pardhi dwellings and other belongings were burned, individuals picked up and ‘charged’ fines, or documents – so precious to Pardhis’ already tenuous legal identity – confiscated. The Maharashtra Juvenile Justice (Care and Protection of Children) Act, 2015 ends up targeting Pardhi families and dovetails neatly with the Maharashtra Prevention of Begging Act. Police targeting of Pardhis frequently involves the arrest of adult members under the beggary law and the removal and detention of their children in ‘care homes’ under the Juvenile Justice Act as parents are deemed unfit to retain care of their offspring. Instances of this pincer movement were observed during fieldwork.

Drawing these findings together, we can observe important impacts on Pardhi identity. Beginning with their non-recognition as a community in the census, their lack of legal documentation of individual and caste/tribal identity, their necessary recourse to homelessness over nomadic status to access public benefits, the further imposition of a beggar identity atop the homeless identity, as well as the constant threat not only of police harassment and loss of precious documents but of children being removed into custody. All of this comes together to erode and threaten erasure of who the Pardhis are, as a community, as a people. Here in the postcolonial life of a DNT tribe, we find strong continuities of experience at the hands of the state. To finish, then, we will reflect on the experiences of the Kheria Sabars in rural West Bengal and the Pardhis in urban Mumbai and what they can suggest to us about everyday postcolonial penalities and the way control operates through and across what we term southern penal spaces.

Contemporary nomadism and southern penal spaces

Imperial legacies cast long shadows over the lives of DNT tribal communities in contemporary India. Despite formal denotification of ‘criminal tribes’ in 1952, the new Indian state's more robust equation of nomadism and mobility with the latent spectre of criminality – brought together in the DNT category – has extended rather than abated the penalisation of lives and life-ways. It is worth dwelling here for a moment on the term penality, for as we look to develop the idea of southern penal spaces we may benefit from a lesson in its etymology. Penality, the Oxford English Dictionary tells us, derives from the Middle French term pénalité, meaning suffering, as well as punishment, and from post-classical Latin's penalitas and poenalitas, denoting also sorrow, both from the old Latin root poena, or pain. The idea of southern penal spaces captures this wider understanding of the penal as the imposition of punishment, hardship, suffering, sorrow and pain. Criminality – real or imagined, brought down through years as lore as well as law – remains at the root of all this. The Kheria Sabar and Pardhi communities have illustrated well how in contemporary India, the spectre of criminality colours such tribes’ standing in community and society, yet the penalisation of their lives in the postcolony is altogether more diffused than was achieved under the blunt instrument of the Criminal Tribes Act. Formal grids of control and criminalisation, reflected in anti-begging and juvenile justice laws, as well as others we have been unable to touch on here, such as around animal welfare and husbandry, intersect with a wider set of both formal and informal acts and omissions that position and hold DNTs in spaces marked by suffering, sorrow, pain and harm, as well as by punishment.

In this sense, the notions of postcolonial penality and southern penal space differ in important ways from northern, metropolitan visions and practices of punishment. These tend to hold the prison at their centre and they have encouraged geographers – who might prima facie be expected to take more expansive views of space – to narrow their vision of the carceral to sites of incarceration and then to a kind of bleed, or porosity, that allows the carceral to leak out into other domains, such as immigration regimes or forced labour. We agree with Sarah Armstrong and Andrew Jefferson (2017: 258) that this is deeply problematic, for as they note:

Accounts of prison's liminality and porosity ironically hold prison in place by staking out the territory between here and there, inside and outside without ultimately challenging the totalising and discrete qualities ascribed to the prison. In other words, talk about the way borders are transgressed instantiates rather than dissolves them.

For DNTs, the prison figures in their lives, but it is neither central nor completely peripheral. It is better understood as just one node in a networked or latticed space of control. It is one point in a series of flows that mark the lives of DNT communities as they strive for vertical mobility – ‘uplift’ – while navigating horizontally through a society and culture increasingly hostile to nomadic and forest-based life-ways. We do not suggest this architecture of control is a kind of diabolical scheme created by some malign intent. Rather, we imagine it more in terms of a distinction drawn by Gilles Deleuze (1992): ‘Enclosures’, he said, ‘are molds, distinct castings, but controls are a modulation, like a self-deforming cast that will continuously change’ (p. 4, original emphasis). The visible institution of prison we can imagine as an enclosure, a carceral space. For DNTs, against a background of Constitutional protections and reservations and a machinery of social reform and uplift, theirs is the invisibility of the self-deforming cast, modulating and adapting, drawing in as it does state and non-state, formal and informal means to capture and position citizen–subjects within grids of order and control that are distinctly penal in character. It is a spatially dispersed regime including formal punishment sanctions but imposing also myriad other forms of harm, suffering and also a certain kind of existential pain, dissolving nomadic identities and reflected in acts such as the Kheria Sabars’ appropriation of a putative former criminal status that is now actively adopted and recalled truly as theirs.

Through all of this, we have found spatial thinking both theoretically and analytically useful. However, we also recognise its lack of dynamic quality. Thus, it is possible to speak of movement within vertical or horizontal spaces, but that movement is not a theoretical process but rather an act of imagination we impose upon these actually existing spaces. The data themselves, however, reveal clear dynamic elements or processes and that fact itself requires some sort of response. Here, we are drawn to John Braithwaite and Bina D’Costa's work on cascades of violence, based on their fieldwork in Sri Lanka (2016) and South Asia more widely (2018; see also Braithwaite, 2020). ‘As in the cascading of water’, they argue, ‘violence and nonviolence can cascade down from commanding heights of power (as in waterfalls), up from powerless peripheries, and can undulate to spread horizontally (flowing from one space to another)’ (2016: 11). Suzanne Karstedt (2020: 173) summarises that the notion of cascades they invoke is something ‘encompassing all types of behaviours, contacts and structures that relate individual decisions, actions and actors across time and space, including contagion, networks, herd behaviour, or spill-over effects’.

In our own data, we find clear evidence of such dynamics. Recall the question of the Kheria Sabars’ former criminal status, a status that in fact never was. Here we observe a cascade effect through time and space, as colonial-era notification of a few serious offenders within the Lodha tribe taints the whole Lodha name with a mark of criminality, which in turn taints that of the Kheria Sabars with whom they are commonly associated, which then flows into the contemporary self-identification of Kheria Sabars with this former criminal status and their adoption of, to use Sharma’s (2020: 102) phrase, ‘a “history” other than their own’. In the case of the Pardhis, we see how lack of legal identity documents produces cascades both through time, as parents’ lack of documented status affects children's opportunities, and through space as lack of legal documentation precludes access to public goods and raises risks across a variety of spheres of interaction with the state, not least of all with the police. And in both cases, we find impacts on subjectivity and identity, related to both precarity and processes of deinviduation as nomadic status gives way to replacement identities reflecting the harms, hardships, suffering and sorrow of life on the margins. This is reflected again in the Pardhis’ cascading identities as nomad gives way to homeless identity and eventually to beggar personhood. All of these processes, configurations of penal power, sites, networks and vital state/non-state dynamics and spatial formations currently escape metropolitan penal theory.

Conclusion

In the opening pages of this article we asked the question ‘what [today] has become of those communities once notified as criminal?’. In light of what has been revealed here it seems worthwhile to close with some reflections upon possible futures. Three things stand out. First, nomadic life-ways are not static and nomadism in itself is neither inherently desirable nor virtuous: it is good when it ‘works’ and makes sense in the way that it once did. Imagining a future for the Kheria Sabar and Pardhi communities at the centre of this article is not necessarily about recuperating, reconstructing and making possible again the life-ways of old. It is about imagining and making possible ways of life that release these communities from the penal grip within which they are currently held and finding ways of life that ‘work’ and make sense to them today.

Second, for both communities, their current nomadism may at least in part be understood as circulations of despair, the solution to which will involve a certain measure of sedentarisation and development of pathways into meaningful and sustainable livelihoods. The routes to achieving that in each community are of course different, reflecting the remote location of one and urban position of the other. To give an example, Jaladhar Sabar, a Kheria Sabar community leader, recalls that in the early 2000s the United Nations Development Programme (UNDP) and Government of West Bengal initiated work on livelihoods, education and entitlements (including land entitlements) which played an important role in the socio-economic development of Kheria Sabar in 20 villages. The lessons of that remain valid today, but concentrated and large-scale efforts to bring uplift to this community have long ago ceased. At the same time, organisations like UNDP now understand that viable, sustainable solutions are not imposed but imagined in partnership with the communities whose futures they will become. Understanding the way the social spaces of the majority intersect with and in meaningful ways create the penal spaces of nomadic and denotified communities will be critical to thinking through such possible futures.

Finally, imagining a future for either of these communities will be impossible without solving perhaps the most fundamental degradation and penal infliction to which they are subject: the failure to recognise them for who they are. Thus, recognition – in the sense of being properly recognised as DNT communities, being recorded in the all-important exercises that provide access to benefits and being capable of participating in life and society in the way of ordinary rights-bearing citizens – is a lynchpin upon which all efforts to escape their entrapment in these penal spaces depends. No uplift, no vertical social movement, nor any meaningful cascades into better lives, better prospects for children, more secure places of living and sources of income seem possible without this. This would seem to require change (such as in Census categories) at the level of the behemoth that is the Government of India. Yet that is not the only route available. In 2017, for example, the Government of Maharashtra, recognised that nomadic and denotified communities frequently did not have the documents required to access welfare schemes. As a result, special rules were introduced making it possible for members of these communities to obtain the crucial caste certificate. ⁸ Having said that, its effect in practice has been minimal and such examples are dispiritingly rare. Every week, month and year that passes compounds what is now a very long history of these communities’ penalisation. In the arc of time between 1871 and 2021 – 150 years – the moments of Independence and denotification of criminal tribes now sit almost exactly midway. Theorising penal power in the contemporary postcolony and imagining futures beyond it are thus tasks not only of marking out imperial legacies but of mapping the terrains of reconfigured penal spaces and the cascading dynamics that inhibit or enable citizen-subjects’ movement toward freedom and dignity.