Picture a Maratha chief in the seventeenth-century Deccan. He holds a fort, a few hundred horse, and a patch of revenue land that three larger powers all claim. A Persian-speaking secretary keeps his accounts. He serves, this season, a sultan in Bijapur or Golconda whose grandfather came from somewhere west, and he will serve a different one next season if the price is right. Ask him the question that a modern news anchor would ask — "are you Hindu, or are you Muslim?" — and you have asked the wrong question. Not a hard question. The wrong one. He is calculating land, rank, marriage, the loyalty of his cavalry, and the mood of the court. The categories that organize his day are caste, lineage, patron, and fort. "Hindu" and "Muslim" are real words he knows. They are not the walls of his world.1
This is the situation James Laine asks us to hold in the mind before we say a single sentence about Shivaji. The mistake everyone makes — communalists, neo-traditionalists, and even some careful historians — is to start with two sealed boxes labeled HINDU and MUSLIM and then sort the seventeenth century into them. Laine's move is to refuse the boxes at the outset. In the seventeenth century these were not two settled identities with hard edges. They were contested categories: argued over, stretched, shrunk, and deployed for advantage by different people who could not even agree on what the words covered.2 The scope of each — how much of life it claimed to organize — varied wildly from one mouth to the next.3
The chief at the court is the anchor because he is the test case. Any definition of "Hindu" or "Muslim" that cannot find room for him — a man who is nominally Hindu, fluent in a Persianate court culture, indifferent to theological boundaries, and loyal to whoever pays — is a definition that has smuggled in a theology. And a smuggled theology is exactly what Laine wants to catch before it poisons the whole account of Shivaji.4
This is a methodological demand, and it has a sharp edge. Laine states it as a rule for honest description. If we want to be "descriptive rather than covertly or unconsciously theological," then our definitions of Islam or Hinduism "should not exclude 'bad Hindus' and 'bad Muslims.'"5 That single sentence is the spine.
Here is why it bites. When you define a religion by its best practitioners — the saint, the scholar, the orthodox jurist, the sincere devotee — you have already decided that the lax, the worldly, the heterodox, and the indifferent are not "really" of that religion. That decision feels neutral. It is not. It is the move a theologian makes when defending the faith against its embarrassments. The historian who makes it has, without noticing, taken sides on a question of who counts as a true believer — a question the seventeenth century itself never settled.6
So the contested-category frame is a refusal to let the analyst define a religion from inside the religion's own apologetics. A real definition leaves room for the bad Hindu and the bad Muslim because they were there, in numbers, and they were Hindus and Muslims. It leaves room for the man who reduced his Islam to a caste position and the man who reduced his Hinduism to private mystic feeling. It leaves room for the subaltern who never thought in these terms at all, and for the elite who found the shared court culture more useful than any pure religious alliance.7 The category is a tent with disputed edges, not a fortress with a guarded gate.
This is also, crucially, a counter-text to the project of reading Shivaji as the founder of a Hindu nation. If "Hindu" in 1660 was a fluid sign without primordial meaning, then a claim to Hindu identity in 1660 was a move inside an argument, not the awakening of an eternal essence.8
The internal machinery has three gears, and they turn together.
The first gear is the bad-Hindu, bad-Muslim test, already named. It is a filter you run any definition through: does this definition quietly exclude the people who would embarrass it? If yes, the definition is doing theology in a lab coat.9
The second gear is scope. Laine insists there was wide variation in how much social, political, and economic life each religion claimed to govern.10 This is the gear most accounts strip out. On one end stand the brahmin scholars in Hindu courts who, facing Central Asian invasion, set out to define and preserve a whole way of life — a totalistic Hinduism covering ritual, law, diet, marriage, kingship, everything.11 On the other end stand the neo-Hindu intellectuals — and Laine pointedly notes that this move is modern, born of the grapple with secularism — who shrank "religion" down to mystic insight and sincere devotion, deliberately excising its roots in political culture, claiming the political roots were never religious at all.12 The same word, "Hindu," is being used by both, and they mean radically different amounts by it. The Muslim side shows the mirror image: some Muslims worked out a mystic union between Hindus and Muslims while leaving social and political questions outside "religion" — and in doing so, perhaps unwittingly, reduced their Islamic identity to that of a caste and tacitly endorsed a Hindu social system underneath.13 Scope is not a detail. It is the variable that makes a single label cover wildly different things.
The third gear is the people for whom the labels had little or no purchase. Two populations sit here. The subalterns, for whom these distinctions "would have seemed meaningless" — they lived their lives by other coordinates entirely.14 And the elites, who were "more interested in maintaining power and sharing in a cosmopolitan shared culture of the court than in forging alliances based on purely religious grounds."15 Top and bottom of the social ladder, the religious boundary thins out. It is thickest in the middle, among the scholars and clerics who had a professional stake in making the boundary firm.
Run all three gears and the picture is not a clean line between two communities. It is a field of varying densities — dense where someone is paid to police the edge, thin where life is lived by other rules. And then there is the fourth element, the one that turns description into strategy: Cynthia Talbot's rising chiefs.
The contested category is not only a thing that varies. It is a thing that gets picked up and swung. This is where Laine's frame shakes hands with the study of how identity becomes a weapon.
Talbot's work supplies the hinge. There were rising military chiefs, she shows, who used Hindu tropes of demonization — the Muslim as barbarian, beef-eater, defiler — but not, Laine stresses, "so much in order to defend 'Hinduism' or attack 'Islam' as to gain legitimacy, by taking upon themselves the classical role of the kshatriya."16 The demonization is instrumental. The chief who calls the enemy a demon is auditioning for the role of the dharma-protecting warrior-king, because that role licenses his power grab and dresses his new authority in an old and noble costume.17 While forging new regional-linguistic identities, these rajas wrote historical narratives linking themselves to a noble past — they manufactured, in Laine's phrase, "a sort of ethnicity."18
This is the handshake between the contested-category thesis and the mechanics of legitimacy. If "Hindu" were a sealed, primordial essence, the chief would be merely expressing his Hinduism. Because "Hindu" is a contested category with movable scope, the chief can operate it — inflate the religious-warrior frame when it pays, ignore it when it does not, and never once be a theologian. The same fluidity that makes the category hard to define is exactly what makes it usable as a tool of self-promotion. Description and strategy are two views of one fact: the edges are argued over because arguing over them wins things.
And this is the frame Laine will set Shivaji inside. Shivaji, and even more the men who later celebrated his heritage, claimed a Maharashtrian and a Hindu identity. But the claim was made "in the context of a debate (or perhaps just uncertainty) about the proper boundaries for those categories" — a context Laine calls "different from but analogous to similar debates today."19
Take Talbot's rising military chiefs as the working example, because they show every gear of the thesis turning at once.
The setting is the Deccan and south India in the centuries of Islamic political power, roughly 1200 to 1700, when local strongmen are clawing their way up through a landscape of shifting sultanates. A chief on the make needs two things he does not yet have: a story that says his rule is legitimate, and a class identity that outranks his actual origins. The classical Sanskrit tradition has a ready-made answer — the kshatriya, the warrior whose dharma is to protect the realm against threats to order. Claim that role and your power stops looking like banditry and starts looking like duty.20
So the chief reaches for Hindu tropes of demonization. He casts the rival power as the barbarian outsider, the defiler of temples, the eater of forbidden flesh. Crucially — and this is Talbot's and Laine's point — the goal is not to defend Hinduism. The goal is to become a kshatriya. The demonization is the costume fitting. By naming an enemy of dharma, the chief writes himself into the role of dharma's defender, and the role confers the legitimacy his birth and his sword cannot.21
Watch what this does to the two-box model. The chief is not a Hindu nationalist avant la lettre. He is an ambitious man using a religious idiom because that idiom happens to mint legitimacy in his market. The historical narratives he commissions — genealogies linking him to a noble past — are not the recovery of an ancient ethnic identity. They are the fabrication of one, in real time, for present advantage. Laine's word is exact: these rajas "created a sort of ethnicity."22 Ethnicity here is an output, not an input.
This is why the case study matters for Shivaji. The standard celebratory reading takes Shivaji's Hindu and Maharashtrian self-identification as the welling-up of a suppressed essence. Talbot's chiefs show a different and more documented mechanism: regional warlords routinely built identities out of religious tropes to secure power, and the religious content was a means, not the end.23 Put Shivaji in that company and his Hindu claim becomes legible as a move in a live game over boundaries — analogous, Laine says, to the boundary-debates of the present — rather than the first chapter of an inevitable national awakening.24
Here is the analyst at the desk, a seventeenth-century court chronicle open in front of her. The page describes a Maratha raja who has just routed a sultan's column, and the chronicler is calling the defeated commander a demon, an enemy of the gods.
She tries the first reading. Two sealed boxes. She files the raja under HINDU and the sultan under MUSLIM, and the demon-language under "ancient communal hatred surfacing." The story writes itself: two peoples who never really mixed, their buried tensions smoldering until they flare on this battlefield. Tidy. She can feel the pull of it — it matches the news on her phone.
Then she stops. She runs the bad-Hindu test. Where, in this account, are the Hindus serving the sultan? She knows they are there — the sultan's cavalry, his revenue officers, his Hindu generals. The two-box reading has quietly erased them. They do not fit, so they vanish. That erasure is the tell. A reading that has to disappear half the documented actors to stay clean is doing theology, not history.
So she tries the second reading. The demon-language is not the eruption of an essence. It is a tool. This raja is auditioning for the kshatriya role; the chronicle is the audition tape. She asks who paid for the manuscript, what title the raja was claiming that season, what genealogy got commissioned alongside it. The demonization becomes evidence of a legitimacy strategy, not evidence of a primordial divide. The Hindus in the sultan's army stop being an embarrassment and become the proof: the boundary was porous, and the raja was building a harder one because hardening it served him.
| The two-bloc reading | The contested-category reading |
|---|---|
| Hindu and Muslim are sealed, eternal | Both are fluid signs with disputed edges |
| Demonization expresses ancient hatred | Demonization is a bid for kshatriya legitimacy |
| Cross-communal service is anomaly or treason | Cross-communal service is the normal middle ground |
| Religion sets the scope of identity | Scope varies by speaker — caste, court, lineage often outrank it |
| Shivaji awakens a suppressed nation | Shivaji moves inside a live boundary-debate |
She closes the laptop having written one harder, truer paragraph instead of one easy, false one.
You can catch the two-box error in the act. The signs:
The account excludes the inconvenient. Hindus in Muslim service, Muslims under Hindu patronage, the lax and the worldly — they go missing. When a reading has to delete documented people to stay coherent, it has failed the bad-Hindu test.25
The account holds scope fixed. It assumes "Hindu" meant the same amount to a totalizing brahmin scholar and to a neo-Hindu mystic and to a subaltern farmer. Flattening that variation is how a contested category gets smuggled back into a sealed box.26
The account reads demonization as sincerity. It takes "the enemy is a demon" as proof of religious devotion rather than asking what political role the speaker was buying with the insult. It misses Talbot's hinge entirely.27
The account erases the indifferent. The subaltern for whom the distinction was meaningless and the cosmopolitan elite who preferred shared court culture both disappear, because a two-box model has no shelf for people who didn't care about the boundary.28
The account is unconsciously theological. It defines the religion by its best self and calls the result history. The giveaway is that the definition would satisfy a defender of the faith — which means it is not description.29
Any one of these signs means the analyst has reached for the boxes before checking whether they were sealed. They were not.
The deepest tension Laine carries — and refuses to settle — is between two camps of serious historians.
On one side stand Sheldon Pollock, Louis Dumont, and Francis Robinson: the "different deep down" thesis. Pollock shows Hindu-court intellectuals, confronting the unassimilable Turkish invaders of the twelfth to fourteenth centuries, producing a flood of dharma-literature that defined a totalistic Hindu way of life against something genuinely other.30 Dumont grants that the two communities drew together at the level of social fact, yet insists their coexistence "produced no general ideological synthesis."31 Robinson reads the Persian-speaking Ashraf nobility as men who "had come from abroad to rule," cherishing international Perso-Islamic standards, and treats the eventual emergence of Pakistan as "the last striking expression of Perso-Islamic values in India."32 On this reading Hindus and Muslims might share a common culture on the surface, but "deep down" — by a presumed architecture of the soul — they held different visions, and their repressed tensions "smoldered" beneath.33
On the other side stands the anticommunalist scholarship: the "common world" thesis. Opposing modern communalism, many scholars argue that Hindu identity as opposed to Muslim identity is a modern phenomenon, a product of the British habit of taking religion as the primary marker of identity. This critique illuminates how Hindus and Muslims inhabited common social worlds in premodern times, and how cross-communal alliances were ordinary at every level of social life.34
Laine mediates without resolving. He grants the anticommunalists their common social worlds and their warning against British-made categories. He grants Pollock his dharma-literature and the reality of perceived otherness. But he refuses both endpoints. Against the smolder thesis, he denies that there were two sealed souls underneath. Against the happy-syncretism reading, he denies that it was all one common world with no real boundaries felt. His third position is the contested category: the boundary was real and felt and fluid, argued-over, variable in scope, and usable as a tool.35 He leaves the Pollock/Dumont/Robinson tension standing on purpose, as the live problem his frame is built to mediate rather than dissolve.
Open question: if scope varied this widely, by what evidence can we ever recover what "Hindu" meant to a specific seventeenth-century actor, rather than importing a modern default? Laine flags the danger; he does not fully solve the method.36
Laine is fighting a two-front war, and you feel him refusing the easy peace on either side.
With the anticommunalist historians he half-agrees and then breaks. He stands with them in the demolition: yes, "Hindu" is a fluid sign without primordial meaning; yes, the British reified religion into a primary marker; yes, premodern Hindus and Muslims shared social worlds and crossed the line constantly.37 But where they want to conclude that the line therefore barely existed before 1800, Laine plants his feet. The line existed. It was felt. Pollock's dharma-literature is real evidence that someone, somewhere, was building a totalizing Hindu identity against a perceived other.38 To say the boundary is modern is to overshoot.
With Pollock, Dumont, and Robinson the move reverses. He grants them the felt boundary and then refuses their floor. The smolder thesis needs an "architecture of the soul" — two different visions held deep down, tensions repressed and waiting.39 Laine will not buy the deep-down. What looks like two souls smoldering is, on his account, a noisy surface argument over where the edges go, conducted by people with wildly different stakes and scopes — not a buried essence. The bad Hindu, the caste-Muslim, the indifferent subaltern, the cosmopolitan elite, and Talbot's opportunist chief all sit in the gap between "felt boundary" and "sealed soul."40
What the two refusals share, viewed from inside Laine's own experience of the sources, is a discomfort with any frame that lets the analyst stop arguing. The anticommunalist frame stops by declaring the boundary modern; the smolder frame stops by declaring it eternal. Laine keeps it open: contested then, contested now, "different from but analogous to similar debates today."41 He would rather hold a live tension than close a clean book — and he tells you so by leaving Pollock and the anticommunalists both standing, unreconciled, on the page.
Two connections do real work here, and neither is decorative.
The first runs to Religious Tolerance as Political Philosophy. The political-philosophy tradition asks how a state holds together populations of different faiths, and its classic answer is to privatize religion — push it out of the public, legal, political sphere and leave it in the conscience. Watch what Laine's neo-Hindu intellectuals are doing: shrinking "religion" down to mystic insight and sincere devotion, deliberately excising its roots in political culture.42 That is the privatization move performed from the inside of a tradition, by its own modern reformers, as an act of self-definition rather than state policy. And Laine's Muslims who left "social-political questions outside the realm of 'religion'" are doing the same thing from the other side — and, he notes, accidentally reducing Islam to a caste in the process.43 The handshake produces an insight neither field reaches alone: the privatization of religion is not only something liberal states impose from above to manage difference; it is also a scope-shrinking tactic that religious actors deploy on their own categories to make them governable, defensible, or modern. The contested-category frame shows tolerance-by-privatization as a thing done to a religion's own edges, not just to its place in the polity. The political philosopher treats scope as a settled background; Laine shows scope as the live battleground, and shows that shrinking it is itself a power move.
The second runs to Radical Reactionary Identity. The reactionary builds a hardened group identity by inventing a noble, threatened past and a defiling enemy at the gate — and the invention is recent even as it claims to be ancient. Talbot's rising chiefs are this exact engine caught at the moment of fabrication: they "created a sort of ethnicity" by commissioning genealogies that linked them to a noble past and by demonizing an outsider as the enemy of dharma.44 The handshake sharpens both sides. The reactionary-identity concept usually treats the manufactured past as an end — the point is the identity. Laine, through Talbot, shows the manufactured ethnicity as a means: the chief is not trying to be a Hindu nationalist, he is trying to be a kshatriya, and the invented ethnicity is the receipt for a power grab.45 The insight that neither domain produces alone: hardened reactionary identity and naked legitimacy-hunger are often the same act seen from two angles, and the religious sincerity is downstream of the strategic need. This reframes the modern boundary-debates Laine calls "analogous" — when a present-day movement demonizes the religious other and claims an ancient nation, the contested-category lens says: check who is auditioning for which role, and follow the legitimacy, not the theology.46
Both handshakes turn on the same hinge — scope and edges are not given, they are made and used — and each domain only sees half the hinge until the other is laid beside it. The tolerance theorist hands Laine a vocabulary for what the neo-Hindu reformer is doing to his own faith; Laine hands the tolerance theorist a warning that privatization is not always the state's gift but sometimes the believer's gambit. The reactionary-identity theorist hands Laine the engine that manufactures a noble past; Laine hands back the receipt that shows the engine was bought to buy power, not to honor ancestors. Put the three together and a single rule falls out that none states alone: when a religious boundary suddenly hardens or suddenly shrinks, look for the hand on the lever and the prize it is reaching for, because in a contested category the edge moves wherever it pays to move it.
Sharpest implication. If "Hindu" and "Muslim" were contested categories with movable scope in the seventeenth century, then every confident sentence that sorts that century into two communities is importing a settlement that did not exist — and usually importing it from today's news, not from the sources. The honest historian's first job is not to place actors in boxes but to ask what the boxes meant to them, and to whom, and at what scope. Shivaji's Hindu claim is not the dawn of a nation. It is a move in an argument that was live then and is live now.
Generative questions.