Picture mud creeks beneath three freshly captured hill towns, and four hundred men hauling timber to the waterline to lay the keels of twenty small ships. That is what Shivaji set in motion after taking Kalyan, Bhiwandi, and the country near Panvel. He had no fleet, no shipwrights of his own worth the name, and a religious code that told high-caste Hindus the sea was unclean. He built anyway. The twenty boats were small — each could carry twenty men — but they were the first armed ships a Deccan inland power had ever ordered into its own creeks, and they were aimed at one enemy: the Siddis of Janjira, whose island fort the Marathas could not touch by land.1 This page is about the moment a land animal decided to learn to swim, and about how badly that frightened the men who already owned the water.
The Mughals and the Deccan sultanates treated the Arabian Sea as somebody else's problem. They were army-states — horses, forts, cavalry — and they let the coastline run on a permission system. The Portuguese, the English, the Dutch, the French, and to a smaller degree the African Siddis held the seafront, and inland kings simply bought a "passport" so their goods and war-horses could move through the ports into the interior.2 Nobody connected that trade corridor to defence. If you do not control the water your cargo crosses, you do not control whether it arrives.
Two things made this worse for a Hindu state. First, the habit of insularity ran deep. Second, an orthodox diktat held that crossing the sea was "unholy" — a man who did it had to undergo a purification ceremony, a rule that lasted into the early twentieth century.3 Shivaji ignored it. He looked at the untapped water and the long coastline and saw not pollution but a defence zone nobody had claimed. The shipbuilding in the creeks was the first hard evidence that he meant it.
The mechanism here is borrowed competence applied to a borrowed element. Shivaji could not build ocean-capable hulls with local hands alone, so he commissioned Portuguese officers — by Purandare's account the better shipbuilders of the region — and set locals under them to learn.4 At the head of the teams were two Portuguese siblings, Roe Leitao Viegas and Fernao Leitao Viegas.5 Around four hundred workers, by the Portuguese records, did the labour.6
The design tells you the purpose. Twenty small craft, each rated for twenty men, is not a fleet for open-water battle against European galleons. It is a creek-and-coast harassment tool, sized to slip up to Janjira and Danda-Rajpuri, choke the supply lines feeding the island fort the Marathas were eyeing, and break up Siddi raids along the shore.7 The point was not to win the sea — it was to deny it to one enemy at one chokepoint. Shivaji built exactly the boat the job required and no more. The quiet genius is in the scaling: a land power's first naval program is deliberately small, cheap, and surgical, aimed at a problem land power alone could not solve.
This page supplies the founding moment of Maratha sea power, and it does it in a way that complicates the heroic version. The first Maratha ships were built by Portuguese hands, which means Maratha Maritime Power and Naval Strategy begins with dependence, not autonomy — a useful corrective for any page that treats Shivaji's navy as spontaneous indigenous genius. It hands The Siddis of Janjira its origin pressure: the navy exists because Janjira could not be taken by land, so the unconquerable fort is the actual author of the Maratha fleet. And it gives The Early Shivaji–Aurangzeb Correspondence its mirror: where Shivaji probed Aurangzeb in letters, here the Portuguese probe Shivaji through harbour permits and secret orders. Both pages are studies in the same operating method — read your opponent through the small administrative gestures, not the grand declarations.
The richest moment in the episode is not the building of the ships but the panic it caused one Portuguese official. All three creeks sat in the subdistrict of Bassein, where most ports answered to Portugal. Shivaji wanted timber brought to the creeks from the sea, and he asked for "an exit to the sea" through the Portuguese ports.8 The chief Bassein official, Antonio de Melo de Castro, wrote to his superiors in Goa with a flat recommendation: do not let these ships out, on any account.
His reasons read like a threat assessment written by a man who already knows he has lost the initiative. There would be born this way, he wrote, "a pirate at home." If Shivaji's men seized a few ships, their armada would only grow. Relations were not close enough to rule out trouble. If Shivaji had aims beyond the Siddis, the boats could one day threaten the island of Shashti north of Bombay.9 Then comes the line that gives the whole game away: Shivaji had put seafarers on regular salaries, and they had become so bound to him that the Portuguese could no longer find "a single sailor" for their own fleet.10
Sit with that last detail. The most powerful naval empire on the coast could not crew its own ships because a man with no navy had outbid them for the labour. De Melo de Castro is not afraid of the twenty small boats. He is afraid of what the boats reveal — that Shivaji understood the sea as a system of men, money, and timber before he owned a single hull. The captain is watching a competitor assemble the inputs of sea power and correctly reading the ships as the least important part.
The Goa advisory council agreed with him. Its order of 19 July 1659, preserved in the Goan archives, names Shivaji as "a son of Shahaji, the rebel nobleman of the Adil Shahi court," notes that he has captured the country around Chaul and Bassein and grown "quite powerful," and instructs the Portuguese captain to keep the men-of-war from leaving the ports of Bhiwandi, Kalyan, and Panvel.11 An empire moved its bureaucracy to bottle up twenty creek boats. That is the measure of the alarm.
You are sitting where Antonio de Melo de Castro sat. A neighbour you do not fully control has taken three towns inland, and word comes that he is laying keels in the creeks. The instinct is to count the hulls — twenty, small, twenty men each — and file a calm report: not a threat. Resist that. Walk down to your own docks instead. Try to crew a ship. You cannot. Every able seaman on this stretch of coast is already drawing a salary from the man in the creeks.
Now you know something the hull-count would never tell you. He is not building a fleet; he is building a labour monopoly, and the boats are merely the visible edge of it. Read the inputs, not the output. Ask where his timber comes from, who pays his sailors, whether he is asking your harbours for an "exit to the sea" — because the request for the exit is the real intelligence. A man who needs your ports today is a man who intends to need them less tomorrow. Write the report that says: stop him before the hulls touch water, because by the time they float, the harder thing — the bound, salaried crew — already exists.
The way this goes wrong is the way every Deccan power before Shivaji got it wrong: you treat the sea as somebody else's floor. The diagnostic signs are easy to spot once you know them. You buy passports instead of building hulls. You let a foreign power hold the coast on the comfortable theory that you only need goods to move, not to be defended. You let a religious taboo — the sea is unclean, crossing it pollutes — quietly set your strategic boundary for you, so that the limit of your ambition is a purification ceremony.12 Each of these feels like prudence. Each is a slow surrender of the element your enemy's supplies cross.
The tell that a state has fallen into the land-mind trap is bewilderment after the fact — the same bewilderment Purandare flags when he asks why almost no one inland linked the trade corridor to defence and security.13 If you are surprised that the sea mattered, you have already lost it. Shivaji's whole move is the negative image of this failure: he saw the corridor and the coast as a defence zone first, and he started building before he had the skill, the crews, or the religious permission to do it cleanly.
The entire episode comes from Purandare's narrative biography, a [POPULAR SOURCE], and several of its sharpest details rest on a single documentary stream: Portuguese officials' own records and the Goan archives.14 [SINGLE SOURCE] The figure of "around 400" workers, the naming of the two Viegas brothers, and the "single sailor" complaint all trace to Portuguese reporting — which has its own motive to dramatize the Maratha threat and justify blocking the ships. A Portuguese captain arguing against letting a rival's navy sail is not a neutral witness to that rival's intentions.
The dating carries a small tension worth naming. Purandare places the shipbuilding "soon after capturing Kalyan, Bhiwandi and Prabalgad near Panvel in 1657," while the Goa council's blocking order is dated 19 July 1659.15 The two-year gap may reflect a slow build, a delayed Portuguese reaction, or loose chronology in the sources — the page treats the 1659 order as the firm dated anchor and the 1657 capture as the looser starting point. Open question: did any of these twenty boats ever actually clear the Portuguese ports, or did the Goa order strangle the Marathas' first fleet in its creeks before it ever reached Janjira?
Purandare is the only modern voice on the page, so the real argument runs among the witnesses he leans on, and here that means the Portuguese colonial record set against Shivaji's own evident intentions. The Portuguese sources converge on one fact with the Maratha tradition: the ships were real, they were aimed at the Siddis, and they were built with Portuguese expertise. Both sides agree the boats existed and that Janjira was the target. The tension lives in motive and menace. Antonio de Melo de Castro reads Shivaji as a coming "pirate at home" whose stated anti-Siddi purpose is a cover for wider ambitions against Portuguese islands like Shashti.16 The Goa council, deliberating the same man, reaches a colder and more revealing conclusion — that it must "maintain amiable relations with him because he is powerful," while quietly arranging to help the Siddis in secret.17 The split between the alarmed field captain and the calculating council is itself the data: the men closest to Shivaji on the coast feared him most, while the men in Goa treated him as a force to be managed by duplicity. What the divergence reveals is that Shivaji's first naval program succeeded as deterrence long before it could succeed as a navy — it reorganized Portuguese policy around him before a single hull proved seaworthy. The panic, not the boats, was the achievement.
Sea power born of land necessity is, underneath, a story about competing for the same scarce resource and about reading an opponent before he acts — which is why this military-economic episode shakes hands with strategy theory and with the psychology of survival.
Sun Tzu — Field Intelligence and Signal Reading. Sun Tzu's core teaching is that the decisive intelligence is in the small signal read early, not the large event observed late. Antonio de Melo de Castro is a textbook Sun Tzu reader: he ignores the obvious metric (twenty small boats) and fixes on the buried one (he cannot crew his own ships because Shivaji has salaried every sailor on the coast). The structural parallel is exact — the meaningful signal was the labour market, not the shipyard. But there is a tension the military frame alone misses: Sun Tzu assumes the reader of signals can then act. De Melo de Castro read the signal correctly and could only recommend a port blockade — a defensive crouch. The insight neither domain produces alone is that reading an opponent perfectly and being unable to stop him are not opposites; an empire can diagnose its own decline in precise administrative prose and still issue an order that merely delays it. The Goa archive is a document of accurate, helpless foresight.
Arthashastra — Kingship and the Rajarshi Ideal. The Arthashastra treats the king as a manager of resources and dependencies — controlling timber, labour, and chokepoints is statecraft, not mere logistics. Shivaji's creek program is pure Arthashastra: he secures shipwrights, binds seamen with regular salaries, and sources timber through ports he is simultaneously pressuring for an "exit to the sea." The parallel is that both treat the inputs of power — men on wages, raw material, transit rights — as the real terrain of contest. The tension is religious: the Arthashastra's pragmatic king answers to dharma and custom, yet Shivaji here overrides a live orthodox taboo (the sea pollutes the crosser) in the name of state defence.18 The insight that neither the political-economy frame nor the religious frame produces alone: the founding act of Maratha sea power required a Hindu king to treat a sacred prohibition as a strategic cost to be paid, which means the navy was a quiet act of religious defiance before it was ever a military one. The taboo did not bend because belief changed; it bent because the chokepoint at Janjira mattered more.
Psychological Resilience and the Survival of Humiliation. Building twenty undersized boats to confront the Siddis, who held an island fort no land army could crack, is an act performed from a position of weakness — and the willingness to start a program you are not yet competent to run is a survival posture, not a triumphant one. The parallel to resilience under humiliation is the refusal to let a present incapacity define the ceiling of action; Shivaji begins the navy precisely because the Siddis have humiliated Maratha land power at Janjira. The insight the military record alone misses is psychological: the first fleet is an answer to a wound, and the smallness of the boats is the honest measure of how far the wound still had to be carried.
The Sharpest Implication. The most destabilizing thing this episode says is that the visible weapon is almost never the threat. Twenty small boats in a creek were harmless; the salaried crews, the timber routes, and the request for a harbour exit were the threat — and a watching empire that counted hulls would have filed a reassuring report while losing the coast. The reader who measures rivals by their finished capabilities is reading the wrong column. Measure them by what they are quietly assembling the inputs to do.
Generative Questions. (1) Why do land-based powers, again and again, treat the sea as someone else's floor until it is too late — is it geography, religion, or the simple fact that the people who run army-states cannot imagine the element they do not stand on? (2) Shivaji's navy began in dependence on Portuguese shipwrights and ended as a force the Portuguese feared; what is the general arc by which a borrowed competence becomes a sovereign one, and where does the original teacher get cut out?