Skip to navigation – Site map

HomeThematic Issues21The Politics of Lineage: Caste, K...

The Politics of Lineage: Caste, Kinship and Land Control in an Agrarian Frontier

Girija Joshi


This paper traces the shifting loci of community in rural Haryana (“the Delhi frontier”) in the early nineteenth century, focusing particularly upon caste and kinship. It suggests that categories later identified by bureaucrats and scholars as “caste” and “tribe” in this region were in fact simply broad ethnic labels that represented only very abstract communities. It will further demonstrate that in this mobile agrarian frontier, kinship categories (such as the clan) had historically crystallized around the control of land. By implication, kinship was a vehicle of resource consolidation, whose boundaries were determined at least in part by this pragmatic consideration. An appreciation of the political dimension of kinship brings complexity to our understanding of rural politics, highlighting that the “village community” was far from being a historically stable polity.

Top of page

Full text

  • 1 My thanks to Delphine Thivet, Joël Cabalion and the anonymous reviewers whose comments have helped (...)

1The study of rural India is as old as the study of Indian society itself.1 Over the past half-century, perhaps no concept within South Asian rural history has been more criticized and more frequently revised than that of the “eternal village.” This might be summarized as the view that Indian rural communities have historically been timeless, self-contained units, whose cohesion was preserved over the centuries by rigid (religious) traditions (Srinivas and Shah 1960:1375). Once the dominant trope of rural life in the subcontinent, some fifty years of scholarship have progressively chipped away at much of its foundations. The avenues of criticism chosen by scholars have varied. Some have highlighted the enmeshment of villages in the subcontinent in wider circuits of trade and exchange (for instance, Bayly [1983] 2012; Yang 1998). Others have foregrounded the diverse modes of subsistence and ways of life that constituted the rural in pre-modern South Asia (for instance, Ludden 1994; Bhattacharya 1996; Parasher-Sen 1998; Skaria 1998; Guha 1999; Mayaram 2003). Whether through explicit engagement or by implication, each of these contributions has shown that rather than corresponding to a single, stable type that might be conveniently designated “the Indian village,” rural communities in the subcontinent have historically encompassed an impressive diversity of social, economic and political structures.

  • 2 Although there is great variety in the ethnography of this period, some of which is more sensitive (...)

2A third avenue of investigation has been to take a critical look at the social and cultural institutions that were once presumed to have served as the gatekeepers of the village community, to explore just how stable these in fact were. This approach has reaped rich analytical gains, bringing into relief the untenability of what might be termed a “tradition-centric” approach to the sociology of the South Asian countryside. Whereas within the “eternal village” paradigm, the social, political and economic life of rural communities is organized in accordance with supposedly timeless principles, a firmer grounding in historical context problematizes the very notion of “tradition.” The case of caste is exemplary of this tendency. Thus, while scholarship from the nineteenth and early twentieth centuries was apt to treat caste as a coherent institution founded upon distinct and timeless principles,2 a vast body of scholarship has undermined this conclusion. Far from conforming to any single definition, the historical manifestations of caste appear to be as varied as rural South Asia itself.

3The diversity and mutability of caste notwithstanding, it has proved difficult to do away with concept entirely (Jodhka 2012), and there is an abundance of historically grounded theoretical scholarship on the subject. At the risk of very considerable simplification, one might divide this literature into three broad approaches. The first of these identifies caste with particular sets of values (for instance, Dumont [1970] 1980; Das [1977] 1990; Bayly 1999). Although the simple ideological tension between purity and pollution, which Louis Dumont ([1970] 1980:46–61) proposed was the driving principle of caste stratification, is no longer generally accepted as such, it remains common for caste to be defined in terms of ideas (Guha 2013:2). The history of caste is therefore the history of the growing influence of those who profit from its values. Perhaps the best-known example of this approach is Susan Bayly’s careful and nuanced classic study (1999), in which she treats the development of caste as an historical process, that is bound up with indigenous and colonial state formation.

4The second approach divides the development of caste in South Asia into two periods, before and after colonialism. The Raj is treated as a watershed, during which caste was effectively stripped of the largely political role it had previously played, and instead assigned a purely social and religious meaning. As a colonial “technology of power” (Cohn and Dirks 1988; Dirks 2001:17)—a way for the state to order, other and understand South Asian society—it was lent a false homogeneity, and in this new avatar became a recurrent theme in both the study and governance of the subcontinent. This discursive dominance in turn reshaped reality in its image, such that a simplified, neo-traditional caste subsequently became “inscribed in ritual, familial, communal, socioeconomic, political, and public theaters of quotidian life” (Dirks 2001:15). Clearly the aforementioned approaches differ in the importance they assign to colonial rule in the historical trajectory of caste. Nonetheless, what they share in common is their treatment of caste as a discourse, sustained by certain techniques of information gathering and organization (Peabody 2001; Dirks 2001).

5A third broad approach to the study of caste concentrates not primarily upon its discursive aspect, but rather upon the mechanism of stratification and its political and economic determinants. This approach is by no means new. As early as 1971, Richard Fox published a study demonstrating how conceptions of caste in pre-modern rural Uttar Pradesh varied with the political priorities of rural populations. David Gilmartin (1994) has shown that similarly pragmatic considerations shaped the boundaries of baradari (brotherhood) in Panjab (p. 8–9). More recently, Sumit Guha (2013) has advocated shifting our attention from the ritual expression of caste and its ideological underpinnings to the drawing of caste boundaries. The myriad systems of stratification in South Asia that have been assigned the label of “caste,” he suggests, were simply varying forms of ethnic segregation. In each instance, differentiation was driven by local and regional politics, rather than by any set of values, although the latter certainly served a rhetorical purpose. Viewed through this prism, castes are no more than “bounded ethnic groups” (Guha 2013:5) that have historically coalesced around common political interest.

6According to Guha, the advantages of making stratification and segmentation the focus of an enquiry into caste are at least twofold. To begin with, this liberates scholars from the need to explain the ideological diversity of practices described as “caste” (Guha 2013:1). Another corollary benefit is that it dismantles the fence of cultural peculiarity which has hitherto prevented the historical sociology of South Asia from drawing upon the same pool of methodologies that are used in the study of other world regions (Guha 2015:53). More pertinent to the overarching theme of this special issue, a focus upon boundaries and politics grounds the study of caste firmly in local context. Rather than searching for adherence to specific principles, it privileges the particular ecology and economy of the locality. Similarly, rather than having to square locally particular observations of caste with dominant principles, focusing upon the mechanism of stratification is sensitive to the ways in which local institutions like kinship and clan served as the axis of “caste” stratification.

7This article will use these insights to explore the nature of caste and kinship categories in rural Haryana in the early nineteenth century. For want of space, it will confine its attention to dominant landholding groups. The analysis consists of four sections. The first two of these introduce respectively the geographical context and the primary sources consulted. The third explores some common social categories used in the region. One of these was the term qaum (people, community)—a malleable, multivalent concept, that appears to have designated groups defined broadly and imperfectly by occupation and geography. Another ethnic category was gota (clan, lineage), which, despite the implication of biological kinship that it carried, was defined by political pragmatism rather than by blood alone. The fourth section will therefore turn to the political context within which the boundaries of lineages took form.

Geographical Context: The Delhi frontier

8The analysis presented in this piece pertains to the countryside west of the city of Delhi, a space enclosed by the Yamuna River on the east and the Sutlej River on the west. On the south and southwest it merges with the hills of Mewat and the Thar Desert, and to the north-east with the Siwalik mountains. This irregular strip of mostly flat and arid land today falls largely within the states of Haryana and Punjab. At the turn of the nineteenth century, as in the present, it carried no single name. It fell within the jurisdictions of a number of different states, including, from 1803, that of the East India Company. Nor was it possible to assign this area a single ethnic or linguistic character. A variety of languages and dialects were spoken here, brought by mobile populations that circulated between Central Asia, Sind, the Deccan and the Gangetic Plains. Indeed, there was even a common saying that both bani aur pani (language and water) changed every forty kilometers (Wilson 1884:120).

  • 3 In fact, “Hariana” was only one of a number of sub-regional divisions that were popularly used in t (...)

9The hybridity of its ecology, political and cultural traditions mean that there are few appellations for this region that fit well. I have chosen here not to refer to it as “Panjab” or “south-eastern Panjab,” both of which seem insufficiently precise. They also carry linguistic and ethnic connotations which obscure the cultural heterogeneity of this space. “Haryana” is an equally poor match, for although it existed as a territorial category in the nineteenth century, it referred specifically to the arid prairieland that stretches west of Rohtak, merging with the desert near Hisar (Fagan 1893:3; Wilson 1884:29).3 In the absence of a fully satisfactory name, I have chosen to follow Jos Gommans (1998) in referring to the region as the “Delhi frontier” (p. 17–18). For Gommans, this is a tract whose specific historical unity derives foremost from its position as a hybrid frontier, or as what Bernard Cohn ([1987] 2005) has referred to as a “shatter zone” (p. 109). Gommans identifies three ways in which this region was a frontier. First, it was a climatic frontier, located between the humid Gangetic Valley to the east, and the western fringes of the vast Arid Zone that stretches across much of Central and West Asia and reaches east and south into the Indian subcontinent. Second, and relatedly, it was a subsistence frontier, where cultivation had historically overlapped with mobile pastoralism. Finally, it was a political frontier, where the control of formal states had always been limited, power resting instead in the hands of a variety of warlords and chieftains.

10These three facets of the Delhi frontier’s “in-betweenness” in fact fed into each other. Its limited ecological suitability to sedentary cultivation as opposed to pastoralism meant that its rural populations led a mobile existence. As a consequence of its vulnerability to drought and the absence of any perennial rivers except at its extremities, the cultivated extent in this frontier fluctuated greatly from year to year. Demographic pressure in neighboring regions periodically led to a flux of cultivators taking parts of the savannah under the plough. However, these settlements were ephemeral and, in years in which the monsoons failed, they would be abandoned, to be repopulated in a more favorable season. The uncertainties of cultivation in turn meant that this region did not serve as the revenue base for states dependent upon agrarian revenues. Rather, the polities that tended to thrive in the Delhi frontier were frequently diffuse, nomadic lineages, who exploited this arid but rich country for its excellent pastures, and for its proximity to the wealthy raiding grounds of the Doab.

A Window on Rural Society

11In order to explore the ethnic landscape of the Delhi frontier in the early nineteenth century, I have in this article relied considerably upon two historical accounts dating to the 1820s, authored by James Skinner. Skinner was one of the many elite mercenary soldiers active in the subcontinent during the Mughal “twilight.” Born in 1778, his father was a Scottish ensign in the English East India Company’s army, and his mother the daughter of a Benares landlord. Before he, too, came to fight under the Company’s flag, Skinner had led Daulat Rao Shinde’s Maratha troops in battle, working alongside English, Scottish, Irish, French and Savoyard mercenaries (Fraser [1851] 2012a:43). Despite the important role he played in the establishment of Company rule in northern India (Alavi 1993), the British were sparing in the tributes and recognition they extended to him. At least in part, this appears to have been for fear that any apparent partiality to their own compatriots—to which Skinner numbered when it so suited the Company (Alavi 1993:448)—would open them up to renewed charges of corruption (Fraser [1851] 2012b:44)

  • 4 I am grateful to the anonymous reviewer who urged me to reflect upon Skinner’s intended audience.

12Short-changed by his employers and patrons, Skinner nonetheless established himself not only as a gifted military commander, but also as a patron of the arts in the countryside west of Delhi. Based at his estate in Hansi, he seems to have maintained a fully functioning manuscript workshop, as well as to have periodically commissioned different artists from the region for specific assignments (McBurney 2014:2). Part of his literary and artistic testament includes two beautifully illustrated and illuminated social and political histories of the region. A total of three copies of each text were produced, and Skinner gifted a pair of these mutually complementary histories to three separate British officers. One of these men was John Malcolm, whose earlier ethnographic accounts of Central India may possibly have inspired Skinner to embark upon his own literary enterprise (McBurney 2014:7). However, both Skinner’s choice of writing in Persian and the considerable expense of producing such high-quality manuscripts (McBurney 2014:2) seem equally to have been motivated by his acquisition of various Mughal titles in May 1830 (McBurney 2014:8).4

  • 5 “Shareh-e-haqiqat-e-baramad-ye-aqvam az har farq-e-hanud ahl-e-Hind.” Library of Congress, Lessing (...)

13The first of Skinner’s histories, the Tashrih al-aqvam (An Account of the Peoples, c. 1825), provides a description of a variety of different ethnic groups, beginning with a short summary of the origins of each that is drawn from folklore and mythology. Skinner then describes the hereditary occupations of each qaum, their religion (dharam) and their rites and customs, particularly those pertaining to marriage. In addition, he pays attention to their nature (svabhava, khasiyat zati) and their karam (actions). Every account is further accompanied by a painting of a member of the concerned group that, so Skinner claims, realistically depicts its garb and craft. Although some of the groups described—such as Rajputs and Brahmanas—were found across much of South Asia, the Tashrih’s focus is broadly north Indian. It was intended, as Skinner mentions in the preface, as an account of “Hindi people of all distinctions resident in Hind.”5 Even within this circumscribed geographical context, the richest ethnographic (as opposed to mythological) portraits are those of the landholding and martial populations of the country between the Sutlej and Yamuna Rivers and northern Rajasthan. This is natural enough, given that it was here that Skinner established himself as a magnate and a man of note.

  • 6 For Skinner’s account of the Bhattis, see his Tazkirat al-umara, ff.254r-261r. British Library, Ori (...)

14This geographical bias is articulated even more clearly in the Tazkirat al-umara (Biography of the Nobility, c. 1830), the companion (and likewise richly illustrated) history that Skinner wrote for the Tashrih. The Tazkira provides an account of a broad cross-section of the gentry and nobility of the Delhi frontier. A considerable chunk of its narrative appears to be drawn from lore and royal genealogies, especially in the case of the more illustrious and established royal houses. Thus, for instance, Rajput lineages such as the Chauhans of Bikaner are treated by Skinner as the descendants of the Chandravansha (lunar race). By comparison, the accounts of minor chieftains are somewhat more sober. For instance, Skinner narrates the manner in which Nawab Khan Bahadur Khan, the Bhatti chieftain of Fatahabad, began his career as a professional soldier in service of the kingdom of Bikaner, subsequently betraying his patron and giving himself up to a life of plunder (gharatgari).6 However far back its origins, each lineage’s history is traced down to Skinner’s own times, often including mentions of recent battles. Each dynastic account is concluded with a description of the precise extent of its domains and the revenues earned from these, as well as the strength and composition of its army. Together, the Tashrih and the Tazkira provide a window into rural society in the Delhi frontier in the transition to colonial rule. They are used in this analysis primarily to reconstruct ethno-political fault lines in the early nineteenth century.

15If Skinner’s accounts indicate where ethnic tensions lay in the Delhi frontier, their focus does not extend all the way down to the grassroots and they contain little by way of material explanation for ethnic conflict. In this respect, the large body of official correspondence and reports left by the colonial state in the region is a fuller source. This material includes the annals of conflicts at the level of the village and thus provides a wealth of information about the political life of rural communities and their relationship with political actors outside the village walls.

Social Categories: From Qaum to Caste

16The Tashrih al-aqvam, as its name suggests, provides a good point of departure to anyone seeking insights into the ethnic landscape of the Delhi frontier in the early nineteenth century. A half-century before the heyday of Victorian ethnography, it sheds light on those groups that colonial bureaucrats were subsequently to identify as “castes” and “tribes,” categories that have proved extremely enduring. The communities of whom we find accounts in its pages, 118 in all, are nominally organized along the principles of religion (dharam) and baran (from the Sanskrit varna; Brahmanical caste). The primary differentiation made is between Hindu and Muslim peoples. Once its religion has been specified, each community is further assigned a varna affiliation, a task rendered somewhat awkward by the fact that the number of groups mentioned in the text far exceeds the fourfold Brahmanical stratification. Skinner explains this incongruence as the product of the intermixing of varnas (baran samkara) brought on by the depravity of the times. As we shall see, however, this is only one instance of how the Tashrih’s neat façade is consistently undermined in the text by the hybridity and fluidity of its foundational categories.

  • 7 Although the plural form of the word “qaum is aqvam, for simplicity’s sake, I have restricted myse (...)
  • 8 “Basiyari ke az qaum-e-rabari-ye-asli musalman shodeh-and. Musalman rabari gofteh mishovand. Dharam (...)

17In fact, contrary to its formal schema, the social groups identified as “communities” (aqvam) in the Tashrih do not conform to any single organizational principle.7 Rather, they are each constituted around at least one of three common traits: a shared occupation (peesheh) and more generally, actions; customs (towr, tariq), specifically related to religious observance and marriage; and finally, a common geography. Skinner, borrowing from his sources, then superimposes a single origin myth upon this foundation and seals the category with a few remarks about the temperament of those belonging to it. It is worth noting that while religion could influence marriage customs and worship, it did not necessarily undermine the category of qaum itself, as long as one of the other two strands—occupation and geographical distribution—were present. Of the camel herding Rabari community, for example, Skinner says: “Many of the original Rabaris became Muslim [and therefore] came to be called Muslim Rabaris. Their beliefs and customs are separate (but) their work is the same.”8 Given this diversity of principles, it is unsurprising that Skinner mostly uses the generic Persian qaum and the terms farq (sort, kind, division) and zat (tribe, kind) to refer to community, even while paying token deference to Brahmanical categories.

  • 9 Tashrih, ff.106v-109v.

18Not only do the communities that feature in the Tashrih appear heterogeneous, but the boundaries between different qaum are frequently hazy as well. This is largely because the traits that serve to bind each together in the text were far from unique, and were in fact common to a broad cross-section of society. This holds less true of the service and artisanal groups that appear in the Tashrih, for their respective crafts serve as clear identifiers. By contrast, social groups who earned their livelihood from the soil or by herding livestock are much less precisely defined and internally quite disparate. Skinner’s account of the Jat qaum illustrates this point well. This category is perhaps the broadest of any social group mentioned in the Tashrih, tenuously bound together by a common purpose alone. This purpose, writes Skinner, was to take the land they inherited from their fathers and increase it until they became rajas themselves. To achieve their goal, they engaged in both cultivation (kisan) and trade (dad-o-setad). As zamindars, they also took care of their cultivators (ra’iyats). Their occupational diversity was matched by their confessional pluralism. Jats, writes Skinner, variously revered Brahmanas, the cow, Vishnu, various Shaiva ascetic orders, Guru Nanak and the Prophet Muhammad.9

  • 10 “Hukka-o-nan-o-ab-e-ahir-o-jat ek-ast,” Tashrih, f.101v.
  • 11 Tashrih, ff.422r-424v.

19The Jat qaum that is portrayed by the Tashrih is not only a motley category, it also shares much in common with other landholding groups like the Ahirs, Gujars and Bhattis. Indeed, Skinner suggests that the Ahirs and the Gujars were simply a particular kind of Jat, whose distinctness as a category was rooted partly in their occupational specialization as cow- and goatherds, and partly in their geographical concentration on either bank of the Yamuna River. As such, writes Skinner, they were not recognized as a different qaum by the Jats, with whom they ate, smoked and drank without compunction.10 Meanwhile, “Bhatti” appears as another umbrella category, somewhat similar to “Jat,” although more specific in geographical focus. Skinner writes that the Bhattis were found mainly between Bhatner (now Hanumangarh) and Hisar, but half a century later, Denzil Ibbetson ([1883] 1916) observed that it was simply a common identification claimed by landholding communities from Jaisalmer to the banks of the Chenab River (p. 144). Like Jats, Bhattis too included a number of different segments, and practiced a mixture of cultivation, livestock breeding and plunder.11

20Skinner’s Tashrih provides one more indication that ethnic categories in pre-modern South Asia, which would later be recognized by the colonial state as distinct “castes,” were often simply regionally-specific appellations for those who engaged in a particular trade. In the Delhi frontier, the use of “Jat,” “Bhatti” and “Ahir” to refer to groups whose subsistence was broadly agro-pastoral, was akin to the use of the term “Rajput” to refer to soldiers (Kolff 1990:71–72). The salient differences between these groups, therefore, did not pertain to status or religion, but to geography. For the rest, their customs and social organization were extremely similar. For instance, each of these groups was said by Skinner to practice widow marriage (karao/karewa), a practice looked down upon by respectable (ashraf) people. Jats and Ahirs ate meat and drank alcohol, while amongst the Bhattis, little distance seems to have been maintained between men and women. Each group additionally partook of the broadly syncretic religious culture of Panjab, within which the emphasis lay upon personal devotion (bhakti) to the divine. Amongst the Jats, this bhakti had a Vaishnava, Shaiva and Sikh color. Amongst the Bhattis, devotion to Baba Farid Ganjshakkar was popular. Perhaps most notably, each of these communities was further subdivided into lineages or gotas, whose relationship with each other—as we shall see below—was often contentious.

  • 12 Constraints of space prevent a full treatment of the evolution of colonial ethnographic categories. (...)

21In light of this hybridity and overlap, it is little wonder that colonial ethnographers later in the nineteenth century struggled just as much as Skinner to place the rural populations of the Delhi frontier within a single sociological category. Amongst British administrators, the trans-Yamuna territories in general were regarded as exceptions to the rigid caste norms that were common further east in the Hindustani heartland. Although the term “caste” was still used in official reports on the region, it carried a specific connotation here, as a synonym for “tribe.” The latter, in turn, was used differently than in, for instance, central India, where communities identified as tribes were believed to be primitive peoples, as yet unexposed to the Brahmanical mainstream. By contrast, west of the Yamuna, many landholding groups identified as Rajputs or of Rajput origin and Brahmanical ideas about social stratification and distance were both known and observed to varying degrees. However, stratification seemed just as much to follow the lines of descent and kinship as it did those of occupation. The local apparition of caste, wrote Denzil Ibbetson, was thus of the “tribal type” (Caton 2004:44).12 Rather than seeking to resolve the tension between Brahmanical caste and local forms of stratification therefore, most colonial ethnographers simply used the term “caste” elastically, across a diversity of contexts.

Land control and lineage formation

22Their syncretic, egalitarian and “common” religious and social practices placed each of the landholding groups in the Tashrih on roughly the same social footing. A powerful rural presence, they appear in Skinner’s works as superior in status to menial service peoples, and inferior to those martial lineages who controlled enough land that they could abstract themselves from cultivation, living instead from their agrarian rents, protecting their ra’iyats and enhancing their territorial possessions through conquest. As both the Tashrih and the Tazkira suggest, however, in the late eighteenth and early nineteenth centuries, the gap between middling peasant-pastoralists and martial elite landlords was narrowing. This had led to the sprouting of multiple chiefdoms and royal houses from amidst the ranks of such “ordinary” qaum as the Jats and the Bhattis, including the Sikh Jat rajas of Patiala, Nabha, Jind and Kaithal, and the Bhatti nawabs of Rania and Fatahabad.

  • 13 Tashrih, f.119v.
  • 14 Tashrih, f.424r.

23In this transition from middling peasant-herdsman to local notable, the clan or lineage (gota; also referred to generally as qaum) was an important vehicle. As a military and political unit—a war band, no less—the clan established the material basis upon which charismatic opportunists from within its number could seek to establish themselves as warlords, and ultimately, as kings. Once this material foundation was in place, these arriviste rajas would emulate the customs and rituals of established elites, in an attempt to consolidate their status and thereby distance themselves from their humble origins. For instance, according to Skinner, the many Jat notables who dotted the Delhi frontier had stopped practicing widow remarriage (karao), taken to veiling their women and dressed in the elegant manner of respectable Brahmanas and Vaishyas.13 Karao had likewise been given up in the family of the Bhatti nawabs, who also claimed to be the only “true” Bhattis, related to the Bhatti Rajput lineage of Jaisalmer.14 Social status, however, followed, rather than preceded, effective political and military power. A powerful gota might not succeed in erasing its peasant provenance; however, its influence and clout could be sufficient to overshadow its social origins.

  • 15 Tazkira, f.155r-155v.

24To establish itself as a powerful rural presence, however, a clan was dependent to varying degrees upon larger actors and entities in the local political orbit, as well as upon the rural communities that constituted its manpower as a war band. As such, it exemplified the political continuum that linked the village with the state. In order to keep itself afloat and to consolidate its power, the clan was therefore obliged to balance the interests of its members with the interests of larger polities. This balance was extraordinarily delicate and in a political frontier such as the Delhi frontier, subject to frequent change. Were, for instance, a state to show signs of complacency or frailty, were its power to decrease or its attention be diverted, this would provide an opportunity for its client lineages to consolidate their position at their patron’s expense. This is precisely what Skinner indicates led to the mushrooming of local landlords and magnates in the Delhi frontier. As was mentioned previously, the Bhatti nawabs began as clients and subordinates of the Bikaner raj; however, when in the eighteenth century, the kingdom began to show signs of sluggishness (sosti), Khan Bahadur Khan seized the opportunity to break away from his patron and establish himself as an independent warlord. Similarly, the Raos of Rewari, who first served as agents of the Mughal state and were appointed the title of chaudhari (generic title for a notable), subsequently took advantage of their patron’s waning strength to establish themselves as autonomous chieftains.15

25Usurpation, however, also transformed the clan as a polity. As its star rose, the clan would itself become a patron, which might extend its support and protection to smaller political fry, in return for the payment of tribute and military manpower. Moreover, clan elites might see their growing influence as an opportunity for personal enrichment and seek to establish themselves as the leaders of their kinsmen, which could cause rifts within the clan, altering its boundaries. Once more, instances of such fission abound in the Tazkira. Not only had the Jat Sikh Khalsa ultimately splintered into twelve distinct streams (misl), but these in turn had further split into numerous sub-branches. Thus, there were four distinct states of the Phulkian misl in the Delhi frontier (Patiala, Nabha, Jind, Kaithal).

  • 16 Haryana State Archives, Other Records, Tarikh-i-riyasat-i-kalsia, vol. II, f.4v.

26Similarly, an offshoot of the Karora Singh misl had, in the late eighteenth century, developed a small fiefdom at the village of Kalsian, south-west of Amritsar. This tiny raj subsequently split into two, and a second center was established nearby and christened Kalsian Khord (Little Kalsian), while the parent settlement was renamed Kalsian Kalan (Greater Kalsian). Subsequent tides of war and migration eventually carried part of the Kalsian family to the banks of the Yamuna River, where they established yet another offshoot of the Kalsian raj at Chhachrauli.16 The relationship between different branches of a clan were tempered by pragmatism. Amongst the Bhattis, for instance, Skinner writes that it was the sustained onslaught from the rajas of Bikaner and the Mughal state that led to the reconciliation of Khan Bahadur Khan and his son Zabiteh Khan, who had previously been fighting each other.

27While the examples so far have focused on lineages sufficiently large to receive mention in Skinner’s Tazkira, the same dynamic—of balancing the interests of smaller and larger political entities with the purpose of political aggrandizement—shaped every link in the chain that was the lineage. This dynamic operated, in other words, at the level of the village and pargana as well, molding the boundaries of rural communities at the very grassroots. In the Tazkira, Skinner provides us with a glimpse of some of these even smaller lineages and clans that still ruled as brotherhoods, from where a notable or magnate was yet to emerge, or whose elite were still too minor to be counted as nobility. These included, for instance, the Sangwan Jat gota, whose seat (sakunat) was between the towns of Jhajjar and Bhiwani, and consisted of 88 villages of which Charkhi and Jhojhu were the most important. Similarly, in Gohana near Rohtak, a block of 52 villages belonged to the Phogat Jat gota, and directly to the north, another 40 villages between Sanwar, Ranila and Bandh were held by another clan claiming to be affiliated to the Panwar Rajputs.

28The larger a clan territory (occasionally referred to in colonial literature as a khap, although I have not yet encountered this word in Skinner’s accounts), the more intricately it would be subdivided. Even clans that were no more than a single large settlement, were often subdivided into segments called pannahs or tholas of individual families. Within larger territories, such as those of the Phogats, Panwars and Sangwans mentioned above, it was not uncommon for clusters of villages—variously called thambas and tappas—to be demarcated (Ibbetson [1883] 1916:74). In principle, every gota, from the lineages of large states to village-level clans, developed from migration to a single parent site (variously, called malikan deh and tappadari mauzah), followed by natural increase. In practice, however, this was not the case. The weight of political considerations, the pressure of individual ambition and the resultant brittleness of the clan meant that its boundaries were not determined by kinship or blood ties alone.

  • 17 Muqaddam was in fact used for village headmen, but in many of the villages of the Delhi frontier, t (...)

29In principle, representatives of older settlements had a greater say over newcomers in the governance of the clan and its various segments, which included taking decisions regarding matters such as the distribution of spoils from plunder, the right to break up new land and the use of pasture. Moreover, new settlements would be compelled to pay older settlements a tribute, called a chaudhriat (the tribute due to the chaudhari or overlord). The same principle applied within the confines of a single village as well, where a distinction was made between the “original” residents (variously called maliks, pattidars, muqaddams and biswadars) and newcomers.17 The latter might have equal right to access of land, but a subordinate say in village affairs. When smaller clans were integrated within the domains of larger lineages or states, these divisions were frequently maintained, for they enabled the larger polity to govern its new possessions through the existing network of village and tappa elites. This arrangement was convenient, for established elites had an authority that state-appointees perhaps lacked; but, as the example of the Raos of Rewari mentioned above indicates, it could also prove to be dangerous for the state.

30In the Delhi frontier, every segment of a lineage had a corresponding expression in resources. For states, the most important of these resources were men and revenues. For smaller lineages, who served at once as their own army and their own revenue base, these resources included arable land, ponds and pasture; and every thola, deh (village) and tappa had its corresponding share of these. Quarrels over the partition of such resources could lead to fragmentation and protracted feuding. This friction also indicated that, while in theory at least, the component segments of a lineage were organized hierarchically, such that the village was subordinate to the village cluster (tappa), the village cluster to the clan elders or the raja (if there was one), and the elders or the raja in turn to the state, in reality, the center of power shifted between each of these different segments. This has already been demonstrated by the examples of the Bhatti nawabs and the Rewari rajas cited above. The same dynamic was, however, also replicated closer to the grassroots.

  • 18 British Library (BL), India Office Records (IOR), Board’s Collections (BC), F/4/212/4735, 15 June 1 (...)
  • 19 BL, IOR, BC, F/4/305/7014, 7 February 1809.

31The case of Abdul Samand Khan, a jagirdar (revenue-farmer) appointed by the East India Company to supervise the hinterland from Rohtak to Hisar, provides an excellent illustration of this point. When, on the force of the Company’s authority, Abdul Samand went to declare his rule in his newly acquired fiefdom, he was met with widespread hostility. As he recounted to the Company’s representative (“Resident”) at Delhi, Archibald Seton, the more intransigent villages belonging to martial Ranghars of an unspecified clan had met his declaration of overlordship with outright hostility. He had been told that “since the decline of the (Mughal) Empire no established authority had ever been acknowledged in their country, that every Governor who had come into it had been compelled to sit down like a petty Thanedar [policeman], that nearly forty Noblemen had at different periods arrived as Governors for the purpose of returning the province to subjection but that they had all been slain and that their graves could be pointed out.”18 Abdul Samand’s account was perhaps embellished to garner the Company’s support; but it is worth noting that he willingly relinquished his position as jagirdar and governor, because the intransigence of the rural communities under his supervision had worn down both his morale and his resources.19

  • 20 BL, IOR, BC, F/4/305/7015, 24 March 1809.

32This is, however, not to say that landlords or local notables had no recourse in the face of rebellion and dissension (fasad, fitna). Travelling through the qasbah (town) of Beri in 1808, Edward Gardner, an Assistant to the Resident of Delhi who the labor-starved Company had invested with a broad range of judicial and fiscal powers, was witness to one such conflict. The local zamindar of Beri, a notable named Faiz Talab Singh, had reportedly taken nine hostages from the people of the surrounding villages, whose revenues were in arrears.20 The conflict had become violent and Gardner had been forced to intervene, to restore the hostages to their families. Similarly, before he had conceded defeat and petitioned the Company to be relieved of his revenue farm, Abdul Samand Khan had attempted to enforce his mandate in the face of popular recalcitrance by force, an endeavor that had eventually bankrupted him. There is likewise plentiful evidence of conflicts between village and village in the colonial archive. In the largely pastoral economy of the Delhi frontier, where livestock—particularly bullocks—were a cherished possession, such conflicts often took the form of cattle theft. Indeed, cattle theft remained rife well into the twentieth century, serving as a way for rival villages and clans to score symbolic points against each other (Gilmartin 2003).

  • 21 BL, IOR, BC, F/4/305/7014, draft petition, 19 March 1809.
  • 22 BL, IOR, BC, F/4/237/5453, 18 July 1807.

33Against this background of frequent conflict, it is little wonder that rural settlements in the Delhi frontier were commonly fortified. Clans like Khan Bahadur Khan’s Bhattis used abandoned forts like those at Sirsa, Rania and Fatahabad as their strongholds, disappearing behind their walls after returning from a cattle raid or other plundering expeditions. However, it was common even for smaller villages to erect mud fortifications along their periphery. Watchtowers were also built outside the village walls and manned by men with matchlocks, to guard over wells and to make sure that cattle sent out to graze were not carried off (Fortescue 1911:112). In case a threat was spotted, guards would beat a large military drum called a tamak, to summon attention.21 The response to the alarm could be surprisingly rapid. One Company surveyor whose proximity to a village near Patiala caused the drums to be beaten reported that within a matter of minutes, “between 2 and 500 people appeared without the Town, stationing themselves under the protection of its walls.”22

  • 23 BL, IOR, BC, F/4/274/6109, 3 April 1808.

34Yet, if it was such apparent solidarity that led Charles Metcalfe to describe Indian rural communities as “little republics,” the village was, in fact, a far less cohesive polity than he conceded. Conflict was as common within the village walls as it was outside of them and the village itself was therefore prone to ruptures. One of the most common sources of friction pertained to the rights of individuals within the village to its land and pasture. This was a subject that the colonial state was at first unwilling to intervene in, since, as one Company servant put it, property rights were “blended with feelings of family and of ancestry.”23 Nonetheless, while conducting early revenue surveys, revenue officers found themselves repeatedly called upon to arbitrate between rival claims to the office of muqaddam (village headman). For the British, the office of headman was an important and dignified one, to be filled by a candidate of requisite influence and temperance. They were therefore surprised at the sheer number of persons who put themselves forward as prospective muqaddams (Edmonstone 1846:20).

35It was only gradually that the colonial state realized that there were as many muqaddams as there were tholas (segments) in a village, and that the role of the muqaddam was primarily to ensure that the families within his thola received their fair share of the village wealth, rather than simply to serve the welfare of the village as a community. The same transactional logic pervaded every link of the lineage chain. Indeed, kinship categories in general served as a kind of political coalition, to which biological kinship offered one, but not the only, path of entry. New members entered into the clan territories through conquest, migration, and compromise, while fragments of the clan would separate due to internal disagreements.

36The coalition-like nature of the clan is well illustrated in the account of the origins of the Dahiya Jat gota. By the 1870s, the territories of this clan extended from Karnal to Sonepat, and encompassed not only villages where Jats were in the majority, but also several that were predominantly Gujar and Chauhan Rajput. The roots of this multi-qaum lineage, according to Denzil Ibbetson, could be traced back to the days of Mughal dominance, when imperial governors had put their weight behind another Jat gota called Ghatwal, to counter the growing local influence of a recalcitrant Rajput gota called the Mandhars. Such was the success of the Ghatwals, however, that they posed a threat not only to the Mandhars, but also to other Jat clans in the area, such as the Dahiyas. In response to this threat, the Dahiyas began to consolidate their ranks, forging ties with several minor Jat, Gujar and Rajput lineages (Ibbetson [1883] 1916:82). Although the Mandhars had been politically eclipsed, the rivalry between the Ghatwal and Dahiya Jats persisted into the nineteenth century. Indeed, as one revenue officer noted, for members of both factions, the gota affiliation served as the primary identification, regardless of whether they were Hindus or Muslims, Jats, Rajputs or Gujars (Maconachie 1882:86).

37To stress the importance of pragmatic considerations in the drawing of clan boundaries is not to say that Brahmanical norms pertaining to varna and kinship were irrelevant to social and political life. On the contrary, as has been mentioned above, honoring these codes with overt gestures—whether by refraining from widow marriage or veiling one’s wife and daughters—signaled respectability and conferred high social status. Moreover, the language of family and varna was considered a necessary window-dressing to confer legitimacy upon relationships founded upon convenience. Ibbetson noted that in Karnal district, although it was not uncommon for non-kin members to be absorbed within the village as coparceners and fellow proprietors, “the fiction of common descent” was maintained. If such a newcomer were asked to explain how he had become a coparcener, he would respond “bhai karke basaya” (by making me a brother; Ibbetson [1883] 1916:75). In other words, “traditions” sanctified pragmatic associations, but did not determine their structure or boundaries.

Conclusion: Rural Polities and the Malleable Bonds of “Kinship”

38The preceding account has sought to demonstrate that in the Delhi frontier, rifts in rural communities from the village upward were common and that the source of these rifts had a clear material basis. The way in which communities disintegrated had consequences for the manner in which they were created as well, that is to say, not so much in accordance with “tradition” as with pragmatic interests. In a subsistence and political frontier, the absence of a single political center and the frequency of warfare meant these interests were regularly realigned and villages, as well as the clans or lineages that they were part of, accordingly reconstituted. The language of caste and kin would subsequently be graphed onto what were, effectively, coalitions. In this respect, then, the village as a polity was no different from the state, and kinship categories served as vehicles of political consolidation.

39While this limited foray into village politics cannot claim to yield novel theoretical insights, it does reinforce two known maxims for the broader study of caste and society in rural South Asia. The first of these is that although larger categories, such as Jat, Rajput and Gujar have served administrative and political purposes (and continue to do so), as categories of analysis, they belie as much as they reveal. The preceding analysis has demonstrated that caste formation is a continuous, ongoing process that occurs at multiple different levels at once (Guha 2013:51). An appreciation of this dynamism ensures that we do not, by default, keep returning to static “tradition” to explain socio-political life. The second point is that some of the attention that we bestow upon the language in which caste is expressed—in other words, the practices and gestures of inclusion and exclusion—might more fruitfully be invested in an exploration of the particular material context in which such behaviors occur. This in turn fosters attention to the way in which ecologically-specific forms of community have lived on under the modern guise of “caste.”

40I will conclude this piece with a reference to the present. A glance at the politics of caste and kin in present-day Haryana, especially the numerous honor killings that have made the news, might suggest that the contemporary weight of “tradition” is far greater than in the nineteenth century. Yet, practical considerations continue to inform how “tradition” is interpreted. Perhaps the best example of this comes from 2014, when the Satrol khap in Haryana announced that intra-khap marriages (hitherto considered incestuous) and inter-caste marriages were now acceptable to the community. The khap’s decision was rooted in pragmatism: Haryana’s skewed sex-ratio (approximately 877 women for every thousand men) means that such restrictions make many men bachelors for life. Moreover, as one of the leaders of the khap admitted, it was a way to enable men to find brides locally, women who would already be familiar with their husband’s family’s customs and would therefore adapt more easily. In this way, a first blow was struck to a marriage taboo, under the protective shroud of “guarding regional culture.”

Top of page


Archival Collections Consulted:

British Library, London, India Office Records, Board’s Collections

Haryana State Archives, Panchkula

Digitized Collections:

Oriental Manuscripts (British Library)

Lessing J. Rosenwald Collection, Manuscripts/Mixed Material (Library of Congress)

Secondary Literature and Printed Sources:

Alavi, Seema. 1993. “The Makings of Company Power: James Skinner in the Ceded and Conquered Provinces, 1802–1840.” Indian Economic and Social History Review 30(4):437–66.

Bayly, C. A. 2012 [1983]. Rulers, Townsmen and Bazaars: North Indian Society in the Age of British Expansion 1770-1870. 3rd ed. Oxford; New York: Oxford University Press.

Bayly, Susan. 1999. Caste, Society and Politics in India from the Eighteenth Century to the Modern Age. Cambridge: Cambridge University Press.

Bhattacharya, Neeladri. 1996. “Pastoralists in a Colonial World.” In Nature, Culture, Imperialism: Essays on the Environmental History of South Asia. Delhi: Oxford University Press.

Caton, Brian P. 2004. “Social Categories and Colonisation in Panjab, 1849–1920.” The Indian Economic & Social History Review 41(1):33–50.

Cohn, Bernard S. [1987] 2005. “Regions Subjective and Objective: Their Relation to the Study of Modern Indian History and Society”. Pp. 100-135 in An Anthropologist among the Historians and Other Essays. Delhi; New York: Oxford University Press.

Cohn, Bernard S. and Nicholas B. Dirks. 1988. ‘Beyond the Fringe: The Nation State, Colonialism, and the Technologies of Power’. Journal of Historical Sociology 1(2):224–229.

Das, Veena. [1977] 1990. Structure and Cognition: Aspects of Hindu Caste and Ritual. 2nd ed. Delhi: Oxford University Press.

Dirks, Nicholas B. 2001. Castes of Mind: Colonialism and the Making of Modern India. Princeton: Princeton University Press.

Dumont, Louis. [1970] 1980. Homo Hierarchicus: The Caste System and Its Implications. Chicago; London: University of Chicago Press.

Edmonstone, George F. 1846. ‘Report on the Settlement of the District of Pergunnah Paneeput’. Pp. 1–50 in Selected Reports on the Revision of Settlement Under Regulation IX of 1833 in the Delhie Territory, No.2. Agra.

Fagan, Patrick J. 1893. Gazetteer of the Punjab: Hisar District. Lahore: Civil and Military Gazette Press.

Fortescue, Thomas. 1911. ‘Report on the Revenue System of the Delhi Territory, 1820’. Pp. 69–130 in Records of the Dlehi Residency and Agency. Vol. I, Punjab Government Records. Lahore: Punjab Records Office.Fox, Richard G. 1971. Kin, Clan, Raja, and Rule: State-Hinterland Relations in Preindustrial India. Berkeley: University of California Press.

Fraser, James Baillie. [1851] 2012a. Military Memoir of Lieut.-Col. James Skinner, C.B. by James Baillie Fraser: For Many Years a Distinguished Officer Commanding a Corps of Irregular Cavalry in the Service of the H. E. I. C. Vol. I. Cambridge: Cambridge University Press.

Fraser, James Baillie. [1851] 2012b. Military Memoir of Lieut.-Col. James Skinner, C.B.: For Many Years a Distinguished Officer Commanding a Corps of Irregular Cavalry in the Service of the H. E. I. C. Vol. II. Cambridge: Cambridge University Press.

Gilmartin, David. 1994. “Biraderi and Bureaucracy: The Politics of Muslim Kinship Solidarity in 20th Century Punjab.” International Journal of Punjab Studies 1(1):1–29.

Gilmartin, David. 2003. “Cattle, Crime and Colonialism: Property as Negotiation in North India.” Indian Economic and Social History Review 40(1):33–56.

Gommans, Jos J. L. 1998. “The Silent Frontier of South Asia, c. A.D. 1100–1800.” Journal of World History 9(1):1–23.

Guha, Sumit. 1999. Environment and Ethnicity in India, 1200–1991. Cambridge: Cambridge University Press.

Guha, Sumit. 2013. Beyond Caste. Leiden: Brill.

Guha, Sumit. 2015. “States, Tribes, Castes: A Historical Re-Exploration in Comparative Perspective.” Economic & Political Weekly 50(46–47):21.

Ibbetson, Denzil. [1883] 1916. Report on the Revision of Settlement of the Panipat Tahsil & Karnal Parganah of the Karnal Allahabad: Pioneer Press.

Jodhka, Surinder. 2012. “The Problem.” Seminar (633). Retrieved 8 May, 2018 (

Kolff, Dirk H. A. 1990. Naukar, Rajput, and Sepoy: The Ethnohistory of the Military Labour Market in Hindustan, 1450–1850. Cambridge: Cambridge University Press.

Leonard, Zak. 2016. “Colonial Ethnography on India’s North-West Frontier, 1850–1910.” The Historical Journal 59(1):175–96.

Ludden, David. 1994. “History Outside Civilisation and the Mobility of South Asia.” South Asia: Journal of South Asian Studies 17(1):1–23.

Maconachie, R. 1882. “Delhi: Final Report on the Settlement of Land Revenue in the Delhi District, 1872–77.” Lahore: O Wood & R Maconachie.

Mayaram, Shail. 2003. Against History, against State: Counterperspectives from the Margins. New York: Columbia University Press.

McBurney, Nick G. 2014. The 1836 Tazkirat Al-Umara of James Skinner: A Catalogue. London: Bernard Quaritch Ltd.

Parasher-Sen, Aloka. 1998. “Of Tribes, Hunters and Barbarians: Forest Dwellers in the Mauryan Period.” Studies in History 14(2):173–91.

Peabody, Norbert. 2001. “Cents, Sense, Census: Human Inventories in Late Precolonial and Early Colonial India.” Comparative Studies in Society and History 43(4):819–50.

Skaria, Ajay. 1998. “Being Jangli: The Politics of Wildness.” Studies in History 14(2):193–215.

Srinivas, Mysore Narasimhachar and Arvind M. Shah. 1960. “The Myth of Self-sufficiency of the Indian Village.” Economic Weekly 12(37):1375–78.

Wilson, James. 1884. Sirsa: Final Report on the Revision of Settlement of the Sirsa District, 1879–83. Calcutta: Calcutta Central Press.

Yang, Anand A. 1998. Bazaar India: Markets, Society, and the Colonial State in Gangetic Bihar. Berkeley, CA: University of California Press.

Top of page


1 My thanks to Delphine Thivet, Joël Cabalion and the anonymous reviewers whose comments have helped me refine this article.

2 Although there is great variety in the ethnography of this period, some of which is more sensitive to local peculiarities and divergence from “classical” scriptural principles than has been conceded by generic critiques of colonial ethnography as a whole (for instance, Dirks 2001). A good example of this is the rich settlement literature of the nineteenth century, which by virtue of genre, as well as the “ethnographic intellectualism” (Leonard 2016:187) of colonial officers at the grassroots often contains detailed accounts of the diverse permutations and combinations of “caste” practices found within a particular district.

3 In fact, “Hariana” was only one of a number of sub-regional divisions that were popularly used in the Delhi frontier. Others included the “Bagar,” or the desert tract stretching south-west from Bhiwani and Fatahabad; the “Rohi” and “Budlada” or “Jangal” which, like Hariana, referred to specific stretches of prairie; and the “Nali,” which referred to the arid bed of the Ghaggar stream. Further, in the Delhi frontier (as across much of the Indo-Gangetic Plains), it was common to distinguish between the uplands (bangar, dhaia, utar) and lowlands (khadar, bet, hitar) cut by the shifting courses of rivers. Settlement reports from the late nineteenth century provide rich information on the ethno-ecological connotations of these different terms.

4 I am grateful to the anonymous reviewer who urged me to reflect upon Skinner’s intended audience.

5 “Shareh-e-haqiqat-e-baramad-ye-aqvam az har farq-e-hanud ahl-e-Hind.” Library of Congress, Lessing J. Rosenwald Collection, Manuscript/Mixed Material. James Skinner, James Watson (owner), Tashrih al-aqvam. 1825, f.16v.

6 For Skinner’s account of the Bhattis, see his Tazkirat al-umara, ff.254r-261r. British Library, Oriental Manuscripts, AD MS 27254.

7 Although the plural form of the word “qaum is aqvam, for simplicity’s sake, I have restricted myself to the use of qaum throughout the text, regardless of class.

8 “Basiyari ke az qaum-e-rabari-ye-asli musalman shodeh-and. Musalman rabari gofteh mishovand. Dharam o tariq-e-har ek joda-st. Karam yek-ast.” Tashrih, f.312v.

9 Tashrih, ff.106v-109v.

10 “Hukka-o-nan-o-ab-e-ahir-o-jat ek-ast,” Tashrih, f.101v.

11 Tashrih, ff.422r-424v.

12 Constraints of space prevent a full treatment of the evolution of colonial ethnographic categories. For a comprehensive overview, interested readers might consult Caton (2004).

13 Tashrih, f.119v.

14 Tashrih, f.424r.

15 Tazkira, f.155r-155v.

16 Haryana State Archives, Other Records, Tarikh-i-riyasat-i-kalsia, vol. II, f.4v.

17 Muqaddam was in fact used for village headmen, but in many of the villages of the Delhi frontier, this title was simply granted to every individual who owned land in the settlement.

18 British Library (BL), India Office Records (IOR), Board’s Collections (BC), F/4/212/4735, 15 June 1807.

19 BL, IOR, BC, F/4/305/7014, 7 February 1809.

20 BL, IOR, BC, F/4/305/7015, 24 March 1809.

21 BL, IOR, BC, F/4/305/7014, draft petition, 19 March 1809.

22 BL, IOR, BC, F/4/237/5453, 18 July 1807.

23 BL, IOR, BC, F/4/274/6109, 3 April 1808.

Top of page


Electronic reference

Girija Joshi, “The Politics of Lineage: Caste, Kinship and Land Control in an Agrarian Frontier”South Asia Multidisciplinary Academic Journal [Online], 21 | 2019, Online since 01 August 2019, connection on 23 June 2021. URL:; DOI:

Top of page

About the author

Girija Joshi

Institute for History, Leiden University

Top of page
  • Logo Centre d'études de l'Inde et de l'Asie du sud
  • Logo European Association for South Asian Studies
  • DOAJ - Directory of Open Access Journals
  • Journal supported by the Institut des Sciences Humaines et Sociales (CNRS)
    CNRS - Institut national des sciences humaines et sociales
  • OpenEdition Journals
Search OpenEdition Search

You will be redirected to OpenEdition Search