**2. Householding**

Although a tradition of worldly ascetism, in which renunciation "is the ethos characterizing the ideal householder" is known in other Indian contexts (Cohen [1978] 2002, pp. 236–37), it is most well known in Sikhism. Guru Nanak's egalitarian, humanitarian, and universalistic reforms intended to eliminate Hindu rites of passage and the caste-based authority of priests which unwittingly separated social and spiritual life. Sikhism does not formally separate practitioners from priests, and thus needs no caste or gender divisions, for all Sikhs are to be both social and spiritual. In comparison with 'other worldly' Hinduism, Sikhism outlines a way of attaining divine realization in this world. In a single telling sentence, Surinder Jodhka observes that "it was perhaps the emphasis on just such a 'this-worldliness' which ensured that the social and personal world of the Sikh Gurus and subsequently of all Sikhs could not be caste-free" (Jodhka 2016, p. 585). Householding is the particular social formation that accomplishes Guru Nanak's this-worldly reform, but also, lays the framework for our further consideration of gender and caste: in providing a basis for kinship, gender, and caste, householding is the very formation that enables the continuation of gender and caste distinctions. Here, I briefly outline the Hindu notion of a´¯sramadharma, the religiocultural ground from which Sikh householding emerges and diverges. In this regard, it is important to note that my understanding of householding has emerged from and is largely concerned with the context that it provides for Sikhism, rather than a specific concern with Hinduism.

In the concept of a´¯sramadharma (or stages and duties of life), Hindus are provided with an explicit model for the life course. The a´¯sramas outline four age-appropriate roles for Hindu males from the elite, twice-born castes (Brahmin, K´satriya or Vai´sya) across the four stages of their lives, so that they may fulfill the dharma or responsible conduct of each. The a´¯sramas are linked to a series of rites of passage or sa m. skaras which enable life transitions and initiate men into their expected social roles, ¯ although most are concentrated in childhood. The sa m. skaras depend on the ritual authority of priests, ¯ the elite category of which is foundational to both the conduct of ritual and the caste hierarchy of Hindu society. The a´¯sramadharma, as described in the Dharma´sastras such as the Manusm ¯ r.iti, explains that in the Hindu life course, one is first brahmacar¯ı or student (and celibate), then gr,hastha or householder (and married), and then vanaprastha (or retiree), essentially in retreat from society and the world. ¯ The fourth 'stage' or status, sa m. nyasa or ascetic renunciation, which was not part of the original schema ¯ (Doniger 2014, p. 28), is neither incumbent upon all Hindus (for many simply 'retire'), nor always the final stage (for one can enter from any other point in the life course), but is—in allowing sa m. nyas¯ ¯ıs to ge<sup>t</sup> closest to moksha—the culminating phase of the ideal Hindu life. The sa m. nyas¯ ¯ı renounces the world and lives an ascetic life beyond society, or more precisely, the social world as represented through the activities, responsibilities, and relationships of study, householding, and retirement. Becoming a sa m. nyas¯ ¯ı requires social death (Narayan 1989, p. 74) and thus erases one's caste and kinship ties. Given that women can also renounce (Hausner 2007), arguably, if not always demonstrably, renunciation renders gender equality. In the permanent liminality of the world beyond the social, the renunciant is liberated in this life (Narayan 1989, p. 75).

In terms of the political economy and organization of society, gr,hastha supports all of the other stages, and within the household family structure in particular, the brahmacar¯ı and vanaprastha phases. ¯ As well, householders financed ritual. Gr,hastha is thus essential to the system. But importantly, gr,hastha inverts the normative social order of the other stages of a´¯sramadharma: the householder is spiritually subordinate to the student, retiree, and renunciant, and yet, socially powerful via control of household resources (Cohen [1978] 2002, pp. 202–3, 277). And, in terms of the potential for karmic accretions, gr,hastha is the most onerous stage (Klostermaier 2007, p. 123), in no small part as it is embedded in caste, gender, kinship, and other social formations and bound up with the expected pursuits of artha (material wealth) and kama (pleasure). G ¯ r,hastha thus lends spiritual force and necessity to vanaprastha and perhaps even sa ¯ m. nyasa since its bonds and responsibilities, not to ¯ mention its pleasures and desires, obscure engagemen<sup>t</sup> with spiritual questions. For Hindus, spiritual

development is best pursued in retreat to the forest ashram in vanaprastha or otherwise entirely beyond ¯ the social and bodily in sam. nyasa. ¯

Sikhism explicitly rejects the Hindu stages of life framework, along with the sa m. skaras and the ¯ associated ritual authority and social status of Brahmin priests. Instead, it emphasizes permanent householding as the means of the ideal religious life. Yet some of the aspects of a´¯sramadharma will be familiar to Sikhs, who also use a temporal life course model derived from the Manusmr.iti in the idea of rat de char pehar or the four watches or quarters of the night. The r ¯ at de char pehar frames the stages ¯ of the Sikh life course around the womb, childhood, marriage and household life, and advancing age and preparation for death, which is mapped onto the amrit vela, or ambrosial period just before ¯ dawn, although there is no retreat to a forest ashram as in vanaprastha nor indeed the possibility of ¯ renunciation or sa m. nyasa at this later stage of life (or any). This may pose problems for individual Sikhs ¯ as they age and live the final stages of the life course (such as those with few social ties or desirous of a more solitary spiritual engagement), although the fourfold framework does address the later years of the life course, which are ideally to be lived with sons and grandsons in joint, extended, and patrilineal households which accommodate, and view as valuable, a heightened spiritual engagemen<sup>t</sup> among the elderly. Nevertheless, may¯ a or attachment is apparent throughout the stages ( ¯ Nayar and Sandhu 2007, p. 108). Even for Hindus, the idea of renunciation is not simply and easily contrasted with immersion in social life (e.g., Madan 1991). Indeed, "techniques of detachment" are a means of "dealing with the world and the intensity of a ffections and attachments that extended living in the world entails" (Lamb 2000, p. 141).

The a´¯sramadharma being a model for the heteropatriarchal male life course means that the reproductive capacities of women are an essential basis of the social system and thus are harnessed and constrained. At the same time, women are e ffectively beyond the system as marriage is the only life cycle rite of any consequence to a woman's life, in that it links her to the husband whose householding she will serve and whose retreat—in the classical sense, to the forest hermitage or ashram in vanaprastha—she will follow, should he so choose. Should the husband opt to renounce ¯ and become sa m. nyas¯ ¯ı, his wife e ffectively becomes a widow. Good Hindu wives are kanyadan¯ (virginal) upon marriage (Fruzzetti 1982; Malhotra 2002) and pativrata, (devotees of their husbands) ¯ thereafter (Leslie 1989; Malhotra 2002): servants, protectors, and conduits of their husbands' dharma (Nabar 1995), not to mention his honour (izzat), although this is a form of maya or illusion (Nayar 2010, p. 269). Hindu texts describe the duties of the pativrata, who was to be "like a slave when at work, ¯ a courtesan when making love, ... a mother when serving food, and a counsellor when her husband was ... distress[ed]" (Nabar 1995, p. 149). The idealized tradition of sati, the self-immolation of the wife upon her husband's death, reflected that she was a 'good woman', chaste, virtuous, devoted, and committed to the protection of her husband by making sacrifices on his behalf (Harlan 1994). Clearly, Hinduism sits within, if not emerges from, an explicitly heteropatriarchal and patrilineal milieu, and women—despite their importance to bhakti and other domestic rituals (e.g., Hancock 2000)—are socially positioned vis-à-vis their kin relations to men (fathers, husbands, and sons) at all stages of their lives, and ideally, at least in the religio-social sense, 'protected' by those men, often in ways which are culturally elaborated so as to subvert true protection, such as the gender paradigm of honour and shame and via traditions such as dowry. In addition to gender biases, the a´¯sramadharma system articulates significant inequalities around caste and asserts these di fferences on both ritual practice and everyday life. Indeed, the fact that the system is often termed var n. a´¯sramadharma emphasizes that caste is the overarching social and ritual construct of Hindu life (Uberoi 1996, p. 14). Of course, this social system constrains more than women and lower caste men, for it rigidly scripts a masculine, even hypermasculine, and exclusively heteronormative framework for gender, as well as ascribing fixed caste roles at birth with which individuals may or may not identify and which moreover assert, by various violent means, casteist and racialized social hierarchies.

Amid this context, Sikhism proposed a 'conscious sociological model' (cf. Leaf 1972) for an intentional and utopian community of householders. In contrast to Hinduism, Sikhism "holds

a definite and uniform position against the path of renunciation as a valid means to liberation" (Nayar and Sandhu 2007, p. 4). Ang 71 of the Guru Granth Sahib describes householding as one among several forms of (Hindu) practice, including renunciation, ye<sup>t</sup> within this framework, it is ine ffective. To become so, householding must be conjoined with renunciation. The renunciant householder is exalted in several passages from the Guru Granth Sahib: "Blessed is the Gurmukh, householder and renunciate. The Gurmukh knows the Lord's Value" (131); "Those who are attuned to the Naam, the Name of the Lord, remain detached forever. Even as householders, they lovingly attune themselves to the True Lord" (230); "Immersed in family life, the Lord's humble servant ever remains detached; he reflects upon the essence of spiritual wisdom" (599). Kamala Elizabeth Nayar and Jaswinder Singh Sandhu argue that Guru Nanak 'denounced' the traditional householder framework even as he employed the ascetic terminology of the Nath yogis, a Saivite sect, to describe his vision ´ of 'living in this world' as a new kind of householder (Nayar and Sandhu 2007, p. 6). Sikhs, by name, were to actively learn throughout life (rather than in a distinct first stage), living simultaneously amid worldliness and asceticism, and pursuing spiritual development in the midst of family and community life. (Even Sikh ascetics, to the extent that they exist among subgroups like the Nihangs, marry and participate in household a ffairs). Enlightenment was not to be mediated by priests as in Brahmanical and later forms of Hinduism nor sought outside of family life and society as among sa m. nyas¯ ¯ıs, Buddhists, and Jains, but rather through disciplined engagemen<sup>t</sup> in the life of the world, and equally, the contemplative realization of its illusory nature. For Guru Nanak, detachment was best and truly practiced amidst the daily distractions and attachments of life, so that the ideal Sikh is what Nayar and Sandhu have called a 'socially involved renunciate' (Nayar and Sandhu 2007), or learning to become one, in the spirit of the meaning of Sikh as learner. J. P. S Uberoi on the other hand describes Sikh practice as 'the renunciation of renunciation', according to the following formulation: "Whereas (the Hindus) had sought to achieve emancipation and deliverance through individual renunciation and what amounted to social death, ... the new Sikh community was called to a ffirm the normal social order as itself the battleground of freedom. ... the Sikh initiation rites makes the positive theme of investiture prevail wholly over the negative theme of divestiture, and taking certain widely established customs of Hindu renunciation, emphatically inverts them" (Uberoi 1996, p. 11). Kesh or unshorn hair is an important example of this, as is the refusal of priestly (and caste-based) authority. The worldly engagemen<sup>t</sup> of householding thus also becomes privileged means to the realization of the divine for Sikhs. Thus, as Max Weber (Weber [1930] 2001) has so productively demonstrated of the Protestants, Sikhism too espouses the challenge of renunciation amid worldly realities.

For Guru Nanak, householding was both a means of personal spiritual realization and the moral means to social reform. The Guru Granth Sahib exhorts Sikhs, as gurmukhs, to "look upon all with the single eye of equality" for "in each and every heart, the Divine Light is contained" (598). At Kartarpur, radical praxes of equality in which everyone contributed to farming and everyone shared equally in its fruits placed the householder at the centre of a new society. Nanak wanted "the disciples who had gathered around him to continue to live di fferently" from the Hindus and Muslims around them (Madan 1986, p. 261). Life at Kartarpur was "a fellowship of men and women engaged in the ordinary occupations of life" (Nesbitt 2005, p. 23), but here in this consciously Sikh community, several key Sikh institutions—seva (voluntary social service), langar (community kitchen), and sanga<sup>t</sup> (congregation)—were developed and helped to establish "the values of equality, fellowship and humility, and in a ffirming a new and dynamic sense of 'family'" (23) in which there was neither marked retreat nor renunciation. Moreover, in these praxes, "the neat horizontal divisions and vertical hierarchies of society were broken down" as "Guru Nanak brought to life a 'Sikh' consciousness" (Singh 2011, pp. 11–12).

In laying out this new socioreligious ideal and the means to attain it, Sikhism fulfills several of Anthony Wallace's (1956b) elements of the revitalization movement: it intends to construct a more satisfying culture through utopian reform of beliefs, ideals, and practices, and it is led by not one but ten charismatic leaders, and now the eternal guru as word. As well, according to Wallace's (1956a) principle

of mazeway resynthesis, the development of Sikhism is both socially and individually therapeutic in its efforts to establish reform. The Sikh householder, whether amritdhari or not, adopts a set of values that support the establishment of a just society and is ideally engaged in a broad range of institutionalized social practices (such as vand chhakna, seva, langar, etc.) that enhance its equitable development. Sikh householding as a mutual embrace of quotidien and spiritual life is also echoed in the idea of sarbat da bhala or the wellbeing of all, and the later formulation of m¯ır¯ı-p¯ır¯ı, the refusal to distinguish between sacred and secular realms, and its figuration of the sant-sipah¯ı or saint-soldier. The inseparability of the historical and worldly domains of life from its sacred dimensions assert Sikhism's lived ethics of social justice. In his reformulation of householding and social life, Guru Nanak urged Sikhs to pursue the detachments of spiritual ends in the midst of family, community, society, and sociality. And as they rejected the separation between worldly and ascetic life, the Guru's Sikhs rejected the binary and bounded distinctions between Hindus and Muslims, Brahmins and Sudras, women and men. Had Sikhs continued to live collectively in what some have called 'communes'—as at Kartarpur—immersed in the worldmaking potential of the utopia, it may well be that the genderless, casteless society might have been realized. However, the continuation of normative Punjabi kinship practices upon which everyday Sikh householding is based has ensured the continuation of caste and gender inequalities for femininities and masculinities, caste identities, and group formation are grounded in the household.

#### **3. The Intersection of Gender and Caste**

It goes without saying that householding depends on the ghar or household, which in India is an economic, social, and ritual unit with important domestic, familial, and moral meanings. Social relationships within and beyond the household depend primarily on kinship, a foundational aspect of human society. Collectively, kinship, family, and household map and pattern social relations and produce human societies via an assemblage of social, cultural and moral practices. In most societies, "relationships to ancestors and kin have been the key relationships in the social structure; they have been the pivots on which most interaction, most claims and obligations, most loyalties and sentiments, turned" (Fox 1967, p. 13). In short, kinship provides one's basic identity and orientation to the social world. Another important aspect of kinship is its power to allocate access to and control over economic resources. Feminist anthropologists have long drawn attention to the patriarchal nature of the social customs and norms around particular kinship roles that enable these processes, delimit women from economic agency, and render them dependents (e.g., Sacks [1979] 1982). Even in societies that are not overtly and binarily gendered, men tend to have public cultural roles, while women's natural reproductive capacities tend to confine them to the domestic realm (e.g., Ortner 1972; Rosaldo 1974). For instance, Doris Jakobsh (2003, pp. 7–18) elucidates that the history of Sikhism silences and negates women while at the same time holding them to unrealistic ideals, at times within a colonial project. Sikh historiography puts forward a few extraordinary women as emblematic illustrations of gender equality, resilencing and further negating ordinary women in the process. As if in illustration, Nikky Singh (2005, p. 102) writes forcefully of her belated realization that the five Ks, "intrinsically paradoxical and multivalent", were just as much feminine symbols as masculine.

Central to all of these concerns are arranged, endogamous marriages, which reproduce gender, caste, and power. Marriage and the reproduction of (male) heirs it facilitates are paramount concerns across South Asia (e.g., Dumont 1970; Mandelbaum 1970). This is no less the case for Sikhs, among whom the cultural imperative is that both women and men be married, ideally as arranged by their parents (Mooney 2011, p. 92). Within these marriages, women's social roles are still overwhelmingly circumscribed to those located in kinship and household (daughter, wife, mother, etc.). Wives reproduce culture, community, and the caste group through bearing and raising children, especially sons; embody the community through chaste moral practice; and represent essentialized, sacralised, and feminized domesticity and interior modes of identity. Anjali Bagwe (1995) has proposed the telling phrase 'of woman caste' to sugges<sup>t</sup> the separate and unified category of South Asian women subjects beyond the pale of all other social di fferences. And yet, matrimonial advertisements clearly articulate caste

group identification, not to mention racial (e.g., 'wheatish complexion') and gendered stereotypes (e.g., 'innocent, issueless divorcee'). Even as love-cum-arranged marriages increasingly become the norm in India (Mooney 2011, p. 98) and the diaspora (Pande 2016, pp. 389–90), the fact that transnational marriage remains an important migration strategy (Mooney 2006) means that arranged marriages retain their force.

Perhaps unsurprisingly, most Sikhs continue to maintain caste endogamy, and to apply gendered cultural codes to their marital and family relations and within their everyday lives. North Indian gender practices broadly reflect the inferior social position of women. The feminist literature on South Asia demonstrates that a litany of gender issues accumulates around kinship, marriage, and family, as well as bodies, reproductive capabilities, and selves. These include patrilineality, patrilocality, hypergamy, dowry, son preference, sex selection/female infanticide/foeticide, levirate marriage, seclusion (purdah or ghund), socioeconomic subordination and dependency, disempowerment of women from rights over property, domestic abuse, and the paradoxical location of family, lineage, and caste status in the bodies of female kin, as reflected in the honour-shame paradigm. Presenting a list such as this risks essentializing gender practices and implying that all women are constrained and even victimized by them; my intention, rather, is to summarize and name those aspects of gender subordination that I have witnessed among or discussed with Sikh women in both my personal and professional life for the past quarter century. These practices are not Sikh-specific, and some of them are quite uncommon; moreover, many Sikh women live beyond these constraints. Nevertheless, all are still found in some measure in the Sikh community at large.

As this list suggests—cautions against gender universalism (Mohanty 1984) aside—India is an archetypal patriarchy. Men dominate socioeconomic relations and occupy most positions of power and authority, cultural norms and values are defined in relation to manhood, masculinity, and male dominance and control, and men are focal in most cultural spaces, and certainly those in the public realm. To varying degrees, and depending in considerable part on the organization of everyday household and kinship relations, women lack agency and rights to property, are discouraged from attempts to claim authority or exercise power, are subject to male (and proxy female) scrutiny, hegemony and sovereignty, and are held to di fferent standards of conduct, while men compromise women's social, economic, political, and bodily autonomy in multiple ways. Women's submission to this gendered hierarchy is policed and negotiated, or misogynistically coerced, via various cultural means, and in instances when it is not secured, it may be forced through violent subjugations of non-consenting women's bodies. Patriarchy, indeed heteropatriarchy, also demands the submission of lower status men to its hierarchical structure, as seen in the everyday submissions of sons to fathers and younger brothers to older ones, but also men outside the heteronormative frame, as well as men of other groups, who are often subjugated via their female kin, as evidenced far too amply by intergroup rape, for instance, as evidenced at Partition (e.g., Butalia 2000; Menon and Bhasin 1998; Mooney 2008).

The heteropatriarchy is infused with, oriented against, and expressed through a discursive moral code oriented to the gendered values of honour and shame, which are mapped onto a purity-pollution framework and script a feudal model for retributive justice. While these concepts refer broadly to social status, symbolic capital and public reputation, typically of the patriline and by extension the caste group, they also have explicitly gendered and sexualized dimensions. Veena Das (1976) proposes that honour is the cultural principle—and moral code—that is applied to the natural mechanisms, or biological and sexual aspects, of Punjabi kinship. Significantly, honour is associated with men: izzat is the male register, conveying reputation, dignity, and respect(-ability), and connoting influence, power, and authority, whereas sharam, the female register, is literally translated as shame, but more accurately refers to its prevention via modesty, humility, and sexual propriety, in other words, by maintaining purity as formulated and expected by men. When women are shameless, and even when they are not, they are called besharam (without shame), an epithet often used to regulate women's behaviour; while shameful behaviour is referred to as bezhti (without honour). (As an aside, it is interesting to consider that lower caste names are also used in dominating and perjorative ways). Jat Sikhs explain that 'a man's izzat is his women's sharam'. This epistemological containment of sharam, the concern of women, within izzat, the purview of men, evokes the containment of the domestic in the public (Rosaldo 1974), of the natural in the cultural (Ortner 1972), and of female lives and subjectivities in male ones. Women—mothers- and sisters-in-law being notorious here—themselves surveil and regulate other women's sharam so as to uphold (men's) izzat (Mooney 2010, 2011). At the same time, women are always guarded by male kin, with endogamous marriage transferring guardianship—along with body and progeny—from the paternal to the conjugal line; men always have authority over women (Chowdhry 2007, pp. 4–5). As part of this social contract, violations of male honour warrant retribution; this may take the form of "physical, mental, or emotional assault" (Virdi 2013, p. 111), and even, honour killing.

Although India has a diverse 'kinship map' (Karve 1993), Hindu kinship is "subtly equated" with and subsumes other forms of Indian kinship (Uberoi 1993, p. 39). In fact, there are many regional commonalities between Hindu and Sikh kinship in Punjab (and neighbouring states such as Haryana). The unreformed nature of Sikh kinship is illustrated in several ways in the course of Sikh history. It is often remarked that the gurus were Khatri and thus originated, married, and had children within a single caste community, and a twice-born one at that; as well, there are patrilineal connections among the later six gurus. Hew McLeod (1976, pp. 87–88) interpreted this in terms of the gurus' concern to reject caste as a vertical, but not a horizontal, principle of social organization. Meanwhile, Eleanor Nesbitt (2005, p. 21) speculates that Guru Nanak's name (and that of his older sister Nanaki) may reflect that they were born at their nanke (maternal parents' village). This tradition is rooted in ideas about pollution so as to prevent defilement of the patrilineal household into which the baby is born. Despite the Guru's defence of these natural and life-giving processes (Singh 2016, p. 612), the practice of giving birth at the nanke continues in India today. Village midwives (dais) were from those low caste communities now called Scheduled Caste or Dalit, who in addition to attending births were routinely called upon to perform female infanticide. This practice also continues, although it has been medicalized (Purewal 2010). Doris Jakobsh asserts that "a patriarchal value system was firmly established throughout the guru period, and by the end of the seventeenth century ... [it gave] religious, symbolic, and ritual sanctioning to a specific gender hierarchy" (Jakobsh 2003, p. 238). Anshu Malhotra (2010, p. 98) argues that seventeenth century Khatris (the caste of all of the Gurus) and Bedis in particular (the patriline of Guru Nanak) encouraged female infanticide so as to limit their daughter's exposure to marriage to men from lower castes such as the Jats who were at the time adopting Sikhism in large numbers and thus to maintain Bedi "ritual superiority" over them. The hypergynous ideal was thus of greater importance to the Bedis than female lives. This example conjures what Srinivas (1956, 1989) would term Sanskritization (and Sheel (1999) calls Brahmanisation), the concern of twentieth century Hindu castes to demonstrate their greater ritual purity so as to secure changes (or in this case maintain distinctions) in how the group is perceived and classified. Moreover, reading between the lines of Malhotra's account, we see concern for not only caste boundaries but also izzat, that profound attachment to masculine identity, ego, and lineage.

All of these practices demonstrate the intersection of gender with caste. Indeed, kinship is the key site for the intersection of gender, class, race, ethnicity, and nation, as well as caste, and like all intersectional axes is a space of shifting and complex structures of power. Feminist scholars (e.g., Abu-Lughod 1991; Behar and Gordon 1996; Bridgman et al. 1999; Collins 1998; Kaplan and Grewal 1994; Kandiyoti 1991; Narayan 1993; Ortner and Whitehead 1981; Yuval-Davis and Anthias 1989) have variously described how gender is embedded in complex webs of relationships and di fferential power, drawing attention to forms and systems of social and cultural di fference—gender, race, class, language, religion, ethnicity, and so on—as mutually constituted, and to the ways in which they overlap, converge, and amplify each others' e ffects in producing identity, marginality, privilege, dominance, and discrimination. Social life is constructed and shaped by the borderzones in which frameworks, structures, systems, and experiences of identity, power, and oppression meet, make space for, refuse to accommodate, or subjugate di fference. The intersectionality approach recognizes that race (which we might read here as caste), class, and other forms of inequality are entangled with and compound those of gender (as well as sex and sexuality), etcetera, so as to produce a 'matrix of oppression' (Collins 2000). Associated with third wave feminism and critical race theory, intersectionality was proposed by Kimberlé Crenshaw (1991) in response to the white middle-class homogeneity of second wave feminism (although both are being challenged in the nascent fourth wave). The intersectional approach attends to multiple aspects of lived culture, multiple social locations, and multiple axes of both identity and oppression. In this, intersectionality in some ways eclipses earlier feminists' focus on patriarchy (Patil 2013), as it suggests that women's oppressions result from more than gender. Nevertheless, intersectionality critiques can be productively applied to understanding the complex but clear relationships between gender and caste.

To a considerable extent, ideas about what constitute equality and inequality are products of their times. Feminism "perennially ... [contends with] ... the important question of what it means to be a woman under different historical circumstances" (Brah and Phoenix 2004, p. 74). Before the advent of academic feminism, which has considerably addressed gender bias, writers were likely to read the Guru's egalitarianism primarily through the caste lens, so that gender inequality was a byline. This is in keeping with Padma Velaskar's argumen<sup>t</sup> that "caste and class studies have a strong masculinist sub-text" that both ignores women's voices and constitutes women as "gateways to caste" via their reproductive roles and a purity lens (Velaskar 2016, p. 391), suggesting that caste studies are ripe for feminist readings. The oppressions of gender find parallels and agglomerations in the oppressions of caste. Much like gender is arguably a key trait of South Asian society, Nicholas Dirks states that "caste has become a central symbol for India, indexing it as fundamentally different from other places as well as expressing its essence" (Dirks 2001, p. 3). Yet it cannot be taken as "a single term capable of expressing, organizing, and above all 'systematizing' India's diverse forms of social identity, community, and organization" (Dirks 2001, p. 5). Rather than the immemorial and immutable ideology of perfect social order (e.g., Dumont 1970) that orientalist texts might describe (Appadurai 1986), caste is viewed today as shifting and contextual (Raheja 1988), diverse and situated (Gorringe et al. 2017), and evoking the idea of "multiple hierarchies" (Gupta 2005, p. 424).

Thus, the term 'caste' is a vastly simplistic gloss on the complex and perhaps so wide-ranging as to be almost all-encompassing social phenomena it attempts to render into English. A colonial mistranslation of the Portuguese casta (for breed, race, or type), the English word caste—which is of course used in elite discourse in India—must render almost every nuanced aspect of Indian social relations. Little surprise that it has come to be so emblematic and essential, despite being a clumsy universalizing framework for the subtleties and intricacies of local socialities. In its Indian iterations, caste entails a far more particular—although still overlapping—range of social forms. These include varna, a ritual, social, and racialized ranking based in Brahminical Hindu ideals of hierarchy and purity; varna is 'caste writ large' (and sometimes rendered as 'class' by Indian scholars). Another aspect of caste is jati (Punjabi: z ¯ at), a hereditary category that traditionally denotes occupation and ascribed ¯ social role; this is 'subcaste'. Jati determines one's position within the jajm ¯ ani system of patron-client ¯ relations, the rural socioeconomic framework that organizes caste in relation to the—now diminishing (Srinivas 2003)—feudal or peasant economy. Still other aspects of the ascribed nature of caste are the kinship categories gotra (or clan, and thus surname), and biraderi (variously lineage, fraternal relations, and community). Ronald Inden observes that "the ideal, Brahmanical scheme of four varnas or classes is ever at odds, empirically and historically, with the multiplicity of jatis, castes and subcastes, ¯ and there are always discrepancies between caste rules and actual behaviour" (1986, p. 428). Here, we see the logic of McLeod's (1976) suggestion that the vertical (varna, jati) and horizontal (gotra, biraderi) ¯ principles of caste must be differentiated, as he argues they were by the gurus. Yet the conundrum remains, for the horizontal organization of kinship relations supports caste's vertical differentiation.

Caste references a complex organizational framework of 'traditional' social relations, both inter- and intra-group, with regard to daily social relations as expressed in commensality, marriage, inheritance, and other kinship practices (including jati endogamy and gotra and biraderi exogamy), as well as ¯

notions of ritual and racialized purity, ideas about rank, hierarchy, and status, and understandings of community. Originally manifest in the village context, today these aspects of caste have global reach. Subcaste, clan, and lineage are the most salient aspects of the caste system among Sikhs today, for these regulate everyday social obligations and alliances, determine the possibility of creating affinal relationships within arranged marriages, and express the importance of patriarchy to sociality, reiterating the ways in which caste is closely mapped onto gender. Since there is relatively little attention to varna among Sikhs, it is worth considering whether this might be read as evidence of Guru Nanak's successful eradication of caste, at least in the vertical register, even as jati, gotra and biraderi ¯ awareness continue. More critically, while caste performs numerous functions of social organization, including group formation and identity, it also expresses a racialized system of comparative power relations in which Dalits are subaltern. Since, like gender, caste is an ascribed social category embedded within everyday life, these marginalities are di fficult to avoid and indeed become normalized and hegemonic. Caste groups construct themselves such that "caste comes to be viewed, narrated, embodied, and performed by social actors simply as pre-existing 'natural' cultural di fference or identity" (Natrajan 2012, p. 5) rather than as highly structured and antagonistic power relations.

In his village fieldwork in the early 1970s, Paul Hershman (1981) observed that the "essentials of the Hindu caste system" remained prevalent in Sikh society, and despite the routine proposition that urbanization diminishes caste practices, I found considerable caste awareness in my fieldwork among urban Sikhs twenty-five years later; indeed, while I had planned to document the incidence of intercaste marriages for my doctoral dissertation, the Jat Sikhs I met insisted that they had never engaged in such practices. Rather than dying out, caste has adapted itself to contemporary sociopolitical regimes, arguably evolving from the basis of village social relations as apparent in "endogamy, heredity, and relative rank" (Reddy 2005, p. 548) to a forceful and entrenched political concept. There is certainly evidence that it has become the racialized and ethnicized basis of contemporary local, regional, national, and even transnational politics, and particularly so after the Mandal reforms (e.g., Gupta 2000, 2004, 2005; Jodhka 2012, 2014, 2015; Judge 2014; Srinivas 1959, 1987, 1996, 2003). Even in the diaspora, caste remains a deeply emotive and controversial practice, although some Sikhs are unwilling to admit its tenure (Takhar 2018). All of this suggests that caste still has considerable social force and impact.

These politicizations of caste, for the most part read in non-feminist terms, nonetheless reiterate its dependence on kinship and gender—including the honour and purity nexus—to maintain group identity (even as they may erase women). As well, the existence of dominant castes such as Jat Sikhs poses additional paradoxes for Sikh equality. According to Srinivas (1956, 1987), the dominant caste refers to those groups with low positions within the classical varna scheme, typically Shudras with traditional manual occupations in farming and as labourers, who are able to claim and assert social privileges and power owing to their demographic and socioeconomic preponderance—and as is readily apparent in the Jat case, their control of land as a key economic resource. Little surprise, then, that the incidence of landowning and independent cultivation among Punjab's scheduled castes is among the lowest in India (Jodhka 2002). A whole assemblage of everyday custom and popular culture coalesces around Jat dominance, privilege, and castecentrism in Punjabi village society (Mooney 2013b). At the same time, there is something of a narcissm of small di fferences or perhaps more accurately a status inferiority complex in the Jat claim to dominance, which emerges—highly aware of the Jats' original status in the varna formulation—from considerable and proliferating marginalities under post-coloniality and globalization. It is problematic to assert in this context (although I do elsewhere) that the particular losses and anxieties of Jats over the past three-quarters of a century—around Partition, the Green Revolution, 1984, migration, Hindutva, and so forth—are likely to amplify assertions of dominance, including over women and other castes, as these slights and traumas are read as assaults on the masculine Jat body, which, emasculated, retaliates in hypermasculine ways. As this point suggests, it is worth noting that the practices, performances, and meanings of caste patriarchy were solidified in the colonial encounter. For instance, hypermasculinity responds in part to British constructions of Indian men as feminized, and at the same time was actively channeled for some communities such

as Jat Sikhs via the martial race framework as part of the calculus of colonial rule (Mooney 2013a). Meanwhile, Anshu Malhotra has suggested that the very idea that Sikhism is an egalitarian religion may have been propagated by the British (2010), and, it is well worth acknowledging that the social virtues today attributed to the Gurus may be anachronistic. It is also provocative to consider the possibility that caste in diaspora may be evolving into an ethnic frame in response to the emergence of multiculturalism (as touched upon by Takhar 2018, p. 304). All of this points to the complexities of ye<sup>t</sup> other intersections—femininities and masculinities, tradition and modernity, colonizer and colonized, roots and routes, (and so on)—themselves complexly infracted through historical processes such as colonization, migration, the emergence of Hindu nationalism (and so forth). These are matters for future consideration. And ye<sup>t</sup> the point remains: if Sikhi had truly transformed Sikh society, would the very construct Jat Sikh exist?

#### **4. Householding and Equality, Revisited**

This paper has described gender and caste as mutually related aspects of Sikh society that have as ye<sup>t</sup> been elided from the Guru's egalitarian reforms. Guru Nanak proposed an idea of householding radically different – at least in its' emphases and goals - from the version evident in most contemporary Sikh households. As I have demonstrated, the Hindu construct of householding offers the possibility of social equality and spiritual liberation only to renunciant males from the upper twice-born castes, while Sikhism theoretically offers these possibilities to every Sikh regardless of caste or gender in every stage of the life course. Surinder Jodhka (2016) has noted that, on the whole, caste is different and more muted among Sikhs than Hindus. This would seem to imply that partial social reform of the caste hierarchy was indeed actualized as Sikhism became established. And yet, I have described how heteropatriarchal concerns with purity, honour, socioeconomic prowess, and group identity which link women's bodies and lives with masculine caste status and identity still circulate across both communities, producing inequalities of resources, status, privilege, influence, authority, and power.

Sangeeta Luthra has written that Sikh women's issues—like Sikh men's—are "existential threats" to the community demanding "parity between women and men" (Luthra 2018, p. 324). The problem of casteism that intersects gender subordination and oppression echoes this and also urgently demands resolution. I have located the prevailing inequalities of caste and gender in the ongoing kinship practices that Sikhism theologically and ideologically opposes. Given that gender and caste inequalities jointly underlie and emerge from the kinship system, the fundamental basis of human society, they perhaps are likely to persist for some time to come. Yet the signs are not all bleak. The intersectional approach does not privilege either gender or caste as the foundational basis of social discrimination for the premise of intersectionality theory is that forms of oppression augmen<sup>t</sup> each other rather than compete. Hence, since gender and caste are enmeshed phenomena, their eradication would seem to depend on mutually conjugated efforts. Opinderjit Takhar has recently described mobilizations around caste discrimination in Britain, noting that "mixed caste marriages are becoming more acceptable in the British Sikh community" (Takhar 2018, p. 304). Similarly, ongoing gender inequalities have recently materialized resistance in the form of challenges around gurdwara seva (Jakobsh and Walton-Roberts 2016; Singh 2011, p. 118). Meanwhile millennial Sikh activists employ new forms of "creative agency in translating Sikh tradition and identity through and across the contexts they inhabit" (Luthra 2018, p. 283) to engage not only Sikh issues but global ones (Mooney 2018). Luthra describes a "rapidly expanding Sikh American civil society" that enables "new social spaces, networks, and forms of expression for all Sikhs" that hope to establish "a Sikh praxis of equality for all" (Luthra 2017, p. 326). As well, as noted at the outset, there are other, more specific modes of Sikhi, unexplored here, that may already offer greater equality (or at least a greater tolerance of at least some forms of difference).

Sikhi's resistant and emancipatory egalitarianism rejects the very premise of social inequality based on caste and other distinctions such as gender and religion. At the same time, to state that Sikhism is egalitarian is perhaps effectively to claim that caste and gender inequalities do not exist, or at least are not meaningful, for there is no conceptual space for them, and they are silenced, which also

means they cannot be addressed. Thus, the demonstrated persistence of these inequalities potentially produces a cognitive dissonance reliant upon the separation of religion from culture. In opening this paper, I noted that Sikh youth today still grapple with the "shameful continuities" (Malhotra 2010) of gender and caste. Often, they attribute these di fficulties to their culture rather than their religion. In doing so, they are following a learned cultural script that locates unequal thinking and practice in an unevolved and even debased regional culture that they share with Hindus (and Muslims) rather than in the surely perfected and exalted religion that Guru Nanak initiated and they have learned to revere. At times, their regard for Sikhi is such that I too want to be convinced by their argumen<sup>t</sup> that culture rather than religion is to blame for these ongoing issues. It would be ideal to separate these two fields, but they are deeply intertwined, much as are gender and caste.

In this paper, I have explored the idea that the intersection of cultural and religious practices in the creation of a new Sikh society may, inadvertently and unintentionally, have left a religious space in the commitment to householding—which is, in its Hindu antecedents, the fundamental location of caste and gender—for the continuation of these unequal practices. Gurdwaras (despite being among the most private of public places) for the most part remain predominantly male spaces while caste-based gurdwaras are a detrimental diasporic norm. This being the case, contemporary programs aiming to reform gender (and caste) discrimination from within religious structures and institutions, while surely necessary and welcome forms of "consciousness raising" (Jakobsh 2016, p. 601), may ye<sup>t</sup> be inadequate to fully realize a genderless and casteless Sikh social order. At the same time, to seek to resolve inequities through rights-based legal channels within a secular context seems to divorce caste and gender from religion, negate the ethos of m¯ır¯ı-p¯ır¯ı, and potentially rupture the possibility that Sikhs might "live out the emancipatory praxis birthed by their guru" (Singh 2005, p. xi). Here, we return ye<sup>t</sup> again to the problematic intersection of religion and society: "Nanak's thought ... was marked by comprehensiveness and consistency: its theology entailed its sociology, or, ... its sociology is incomprehensible without reference to its theology" (Madan 1986, p. 260). While doctrine and practice remain at odds, or are described by Sikhs as being so, Guru Nanak's vision remains a deeply inspiring, but as ye<sup>t</sup> elusive, utopia.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The author declares no conflict of interest.
