A father dies, and his three sons start fighting over who gets to keep the family's stronghold. A neighbour — older, trusted, a friend of the dead man — offers to step in and settle it fairly. The brothers let him in. He claps all three in irons, takes the fort for himself, and only later hands them small land-grants and lets them go, now sworn to him. That neighbour was Shivaji, and the fort was Purandar.1 No ladders, no siege, no blood on the walls. He took it by being invited to fix a problem he then turned into his own opportunity. This is capture-by-mediation — a method as far from storming a wall as a handshake is from a punch.
Purandar sits a little to the south-east of Pune, off to one side of the three forts clustered on Pune's south-west. Its position mattered for a plain reason: it helped ring-fence the chief village of Shahaji's jagir, closing the protective circle around the Bhosle family's home ground.2 Like Kondhana, it sat under the Adil Shahi of Bijapur. Unlike an empty hilltop, it had a living chain of command — and that chain had just snapped.
The commandant of Purandar was a Brahmin named Nilo Nilkanth Sarnaik, also called Nilkanth Rao. He was not a stranger to the Bhosles. He was "a friend and neighbour of Shahaji and his family."3 And he had just died. His death did what the death of a fort-holder so often did in this world: it threw the question of succession open. He left three sons, and the three of them fell to squabbling over who would inherit the right to hold the fort.4
A succession quarrel inside a fort is a soft spot in a hard wall. The brothers were divided against each other, the chain of command was broken, and the family already knew and trusted Shivaji's house. Every condition that makes mediation possible was sitting there at once: a dispute, divided parties, and a credible outsider with standing to be asked in. Shivaji read the soft spot and moved toward it — not with troops, but with an offer.
Here is the mechanism, step by cold step. Shivaji offered to mediate the brothers' dispute.5 That offer is the whole trick. To mediate, you must be let inside the dispute — granted standing, access, and a measure of trust by all parties at once. The brothers, fighting each other, each had reason to want a neutral and powerful neighbour to tip the scales their way. So they opened the door to the very man who would use the opening against all three.
Once inside as the trusted arbiter, Shivaji reversed the role. He threw all three brothers into prison and took over the fort himself.6 The peacemaker became the captor in a single motion. The strength of the move is that it costs almost nothing in force — the brothers had already lowered their guard by accepting him as referee. He didn't have to beat their defences; they handed them over when they asked for his help.
Then comes the part that separates this from simple treachery: the release. Soon after, the brothers were let go, and each was given a watan grant — a hereditary land-right — in the country beneath the fort. They accepted the grants and pledged their loyalty to Shivaji.7 This is the closing move that makes the method durable. A man you imprison and humiliate becomes an enemy; a man you imprison, then release with land and a stake in your success, becomes a sworn dependent. Shivaji converted three rival claimants into three loyal watandars sitting at the foot of the fort he had taken from them. The fort, the men, and their future loyalty all came over in one sequence.
This page hands the vault a clean, named third method of fort-capture, distinct from the two on offer at Kondhana. Kondhana gives the assault-versus-subterfuge fork. Purandar adds mediation — and mediation is its own thing. Subterfuge bends one or two men inside the wall through bribery or trickery; mediation gets you invited in through the front door by the very people who hold the fort, by exploiting a dispute among them. The repertoire is now three-pronged: take it by force, slip it loose through a hidden contact, or be asked in to judge a quarrel and keep the prize.
The page also feeds the vault a model of conversion-of-rivals: the defeated party is not destroyed but re-anchored as a loyal stakeholder through watan grants. That is statecraft, not just tactics — the same logic that turns conquered enemies into administrators. And it sets up the forward arc: this is the fort that, eighteen years later, gives its name to the Treaty of Purandar and the great Mughal siege of 1665, so the quiet 1647 mediation is the first chapter of a much louder later story.
Walk the whole thing through as it happened. Nilo Nilkanth — Nilkanth Rao — the Brahmin commandant of Purandar, friend and neighbour of the Bhosles, dies.8 His death is not a battle, not a siege, not a defeat. It is a domestic event. But for Shivaji, hunting forts to close the ring around his father's jagir, a friendly neighbour's death is a door swinging open.
The three sons begin to squabble over the inheritance.9 This is the second gift. A united garrison under a single clear heir would have been a hard target. A garrison split three ways by a succession fight is already half-defeated from the inside — its strength is turned against itself before any outsider lifts a weapon. Shivaji does not have to create the division; he only has to recognize it and time his entry to it.
The offer to mediate is the masterstroke, and it works precisely because Shivaji's family is known and trusted by the dead man's house.10 Trust is the lever. A stranger offering to settle the dispute would be suspected; a friend and neighbour offering the same thing is welcomed. The relationship Nilkanth Rao had with Shahaji becomes the crowbar that opens his own sons' fort. Then the reversal: imprison all three, take the fort.11 The men who invited the referee in find the referee has taken the field.
But the genius is in the ending, not the seizure. Shivaji releases the brothers and hands each a watan grant beneath the fort, and they pledge loyalty to him.12 Compare the two ways this could have gone. Kill or exile the brothers, and you hold a fort surrounded by aggrieved local families with blood-claims against you. Release them with hereditary land and a stake in your rise, and you hold a fort surrounded by men whose prosperity now depends on your success. He didn't just take Purandar. He converted its losing claimants into the foundation of his hold on it. That is why the episode reads less like a trick and more like a blueprint for absorbing a rival faction whole.
You hear that a fort-holder has died — a man your family knew, whose sons are now at each other's throats over the inheritance. Don't reach for a weapon. Reach for the relationship. You send word: let me help settle this fairly, as a friend of your father's. Feel how different this is from a siege. You are not battering a closed gate; you are waiting to be invited through an open one, and the invitation comes because the people inside are divided and each hopes you'll favour them.
Now you are inside as the arbiter, trusted by all three because trusted by none completely. This is the moment the method turns. You hold the standing they gave you and you use it — you secure the brothers, you take the fort. The hardest discipline is what comes next, because the reflex after a seizure is to crush the losers. Resist it. Open your hand. Give each man land beneath the fort, a hereditary stake, a reason to want you to win. Watch three rivals become three sworn dependents. You have not only taken the stronghold; you have bought the loyalty of the very families who might otherwise have spent a generation trying to take it back.
The precondition you must be honest about: this only works when there is a real dispute to mediate and you have real standing to be asked in. Manufacture neither, and you cannot force the door this way. Mediation-capture is a method that waits for the soft spot — a death, a divided house, an existing bond of trust — and then moves fast and ends generous.
The method fails first when there is no genuine quarrel and no standing. Try to mediate a dispute that doesn't exist, or push in where you are not trusted, and the front door stays shut. The whole approach depends on being invited; force the invitation and you are back to subterfuge or assault.
It fails second when the captor forgets the release. Imprison the rivals and stop there — keep them in irons, strip them bare, drive them out — and you have made permanent enemies camped around your new fort. The watan grants are not a soft afterthought; they are the load-bearing wall of the method. Skip them and the conversion never happens.
It fails third through reputational rot. Mediation-capture spends trust to buy a fort. Do it once and end generous, and you have absorbed a faction. Do it crudely, or betray the released men later, and word travels — the next divided family will not let you near their quarrel, and the front-door method is closed to you for good.
The sole modern source is Vaibhav Purandare's Shivaji: India's Great Warrior King (Juggernaut, 2022), a narrative biography for a general readership. [POPULAR SOURCE] The Purandar episode is told as a single clean sequence — death of Nilkanth Rao, succession squabble, offer to mediate, imprisonment, capture, release with watan grants and pledged loyalty — and Purandare does not flag a competing chronicle version here the way he does for Kondhana.13
That tidiness is itself worth flagging. [SINGLE SOURCE] Unlike the Kondhana capture, where Purandare openly weighs Sabhasad against the Jedhe Shakavali, the mediation account at Purandar is given without an internal contradiction surfaced. The reader should treat the smoothness of the narrative as a function of the popular-source register, not as proof that no alternative tradition exists. The mediation method is also the kind of episode that lends itself to retrospective tidying into a "clever Shivaji" anecdote, so the clean shape may flatten a messier reality.
Open questions: Did the watan grants come as a single planned act of statecraft, or were they a later accommodation reframed by the chroniclers as part of one smooth design? And does the contrast with Kondhana — one fort with two warring chronicle versions, one with a single seamless one — tell us more about the events or about how each was remembered?
Purandare is the only modern voice, so the productive tension here is between two of his own episodes side by side: the mediation capture of Purandar and the contested capture of Kondhana. On Kondhana, Purandare carries two chroniclers who disagree — Sabhasad's assault-and-thana versus the Jedhe Shakavali's diplomatic manoeuvre — and he refuses to choose. On Purandar, he carries a single, seamless account with no flagged rival version. The convergence is that both forts were taken in the same arc, both from the Adil Shahi, both to close the ring around the jagir, and both by methods short of open siege. The tension is in the texture of the telling: one capture arrives as a documented disagreement, the other as a polished anecdote. What the split reveals is a caution about the sources themselves. A capture that survives in two competing versions is, paradoxically, better evidenced — we can see the seams where chroniclers worked. A capture that survives in one frictionless version may be more shaped, more remembered-into-shape, than it looks. Purandare's even-handed restraint on Kondhana throws the smoothness of Purandar into relief, and the honest reading holds the mediation story as plausible and well-formed but less stress-tested by the surviving record than its noisier sibling.
Settling someone else's fight and slipping into the room as the trusted referee is the same maneuver whether the prize is a fort, a market, or a wounded psyche: the mediator's standing is itself the weapon.
Behavioral mechanics (persuasion through standing): Sun Tzu — Victory Without Fighting — Sun Tzu's highest skill is to subdue the enemy without battle, to take a thing whole rather than break it. Purandar is that ideal in flesh: the fort is taken intact, the garrison's strength turned by their own internal division rather than by force, the rivals absorbed rather than destroyed. The structural identity is "the cheapest conquest is the one where the target opens itself." The tension is the trust-cost — Sun Tzu's bloodless victory is usually framed as superior generalship, while Shivaji's bloodless victory spends a friendship (Nilkanth Rao's bond with the family) as ammunition. The insight neither domain gives alone: victory-without-fighting is not free. It is paid for in trust, and trust is a finite, non-renewable supply — every front-door capture quietly burns the relationship that opened the door.
History (statecraft and absorption): Arthashastra — Kingship and the Rajarshi Ideal — Kautilya's king is told to win over the conquered, to settle defeated parties with grants and a stake so they become props of the new order rather than perpetual rebels. The watan grants Shivaji hands the released brothers are exactly this conciliation-after-conquest. The parallel is "the conquest is not finished until the loser is converted into a stakeholder." The tension is the route in — the Rajarshi ideal imagines a king of conspicuous virtue, while Shivaji reached the throne of Purandar by betraying men who trusted him as a peacemaker. The insight: the manual's conciliatory ending and Shivaji's treacherous beginning sit inside one episode, exposing that durable statecraft can be built on a foundation of broken trust precisely because the generous ending overwrites the treacherous opening in memory. The grants don't just buy loyalty; they launder the method.
Cross-domain (the survival of humiliation): Psychological Resilience and the Survival of Humiliation — the three brothers are imprisoned by a man they invited in as a friend, then released to live as his dependents beneath the fort that was once theirs. Read from their side, this is an architecture of survivable humiliation: the watan grant gives them a way to live with the loss rather than be crushed by it. The structural parallel is "the powerful party can choose to make defeat survivable, and a survivable defeat produces loyalty where annihilation produces vengeance." The insight neither side generates alone: mercy here is not sentiment, it is engineering — the captor designs the loser's humiliation to be exactly bearable, because a wound that heals into dependence is worth more than a corpse. Too little and the brothers die enemies; too much, and they are humiliated past the point where a land-grant can buy them back. Shivaji found the dose. That precision is the part the psychology page and the history episode only see together: resilience is usually studied as something the wounded party summons from inside, but here the survivability of the wound is calibrated from the outside, by the very hand that inflicted it.
The Sharpest Implication
We tend to file "took the fort by force" as the strong move and "talked his way in" as the weak or sneaky one. Purandar flips that. The mediation capture is the more sophisticated and arguably the more ruthless act — Shivaji didn't beat the garrison, he got the garrison to open the door by pretending to help, then ended so generously that the men he betrayed swore loyalty to him. The uncomfortable thing it says about your own assumptions: the manipulations we find easiest to forgive are the ones that end in kindness. Shivaji's release-with-grants doesn't undo the betrayal — it makes the betrayal palatable, to the brothers and to history. If a generous ending can launder a treacherous beginning that cleanly, how much of what you read as "magnanimity" in any leader is actually the last move of a con?
Generative Questions