A Fort Taken Twice, in Two Different Books
Two separate nineteenth-century gazetteers preserve two accounts of how the Sidis first held Janjira, and neither is a clean origin story. The Imperial Gazetteer of India dates the seizure to 1489 and tells it as fraud: an Abyssinian in the service of the Nizam Shahi kings of Ahmadnagar, disguised as a merchant, "obtained permission from Ram Patel, the Koli captain, to land 300 boxes. Each of these boxes contained a soldier, and by this means the Abyssinians possessed themselves of Janjira island and the fort of Danda Rajpuri" [1]. The Bombay Gazetteer, compiled independently, gives a different lead account: Ahmad Shah, the founder of the Ahmadnagar dynasty, took Danda-Rajpuri around 1490 "after a long siege" and "gave the command to his Abyssinian slave Yakut" [3] — only to add, on the very next page, an alternate version in which "the Sidis got possession of the island by fraud," with Perim Khan and other Abyssinians shipping in armed men disguised as merchants carrying boxes of wine and silk from Surat [4]. Whichever account is closer to the event, Yakut's command is the fixed point both books share: a slave, not a conqueror in his own right, was set over the fort that would outlast every dynasty that tried to take it from him.
The record then goes dark for well over a century. "Local records seem to show that till 1618 the governors of Danda-Rájpuri were Moghal officers," the Bombay Gazetteer notes, before Sidi Sirul Khan was appointed governor that year [4] — the point at which this archive's traceable chain of Sidi rule actually begins.
An Elective Wazirate, 1618–1734
What followed for more than a century was not a hereditary court but a working elective office. Sirul Khan I gave way to Sidi Yakut in 1620, who gave way the following year to Sidi Ambar, known as Sanak or 'the Little' to distinguish him from the great Malik Ambar of Ahmadnagar [5]. When Janjira passed into Bijapur's orbit the rule was written down explicitly: "on the death of a Wazir, the first officer of the fleet, not the son of the late governor, was to succeed" [5]. Under that constitution Sidi Ambar was followed by Sidi Yusuf on his death in 1642, then — after an accession the record itself leaves unexplained, naming no installer — by Fateh Khan, and in 1670 by Sidi Sambal, chosen when two rival officers 'waived their claims in favour of' him [6]. Sidi Kasim received the command when Sambal was promoted to Mughal admiral, Sidi Khairiyat received the governorship from Kasim in 1677 when Kasim himself moved up to the Moghal admiralty, and in 1707 — on that same Kasim's death — 'the unanimous decision of the Sidis' elected Sidi Sirul Khan II, then commandant of the island fort of Padmadurg [7]. Ten transitions, 1618 to 1734: eight of them the record shows no violence at all — the office moving by waived claim or unanimous decision; one, in 1670, was decided by force, turned against the sitting Wazir's own decision to surrender the fort rather than by a rival claimant seeking his post; and the tenth, when Sidi Sirul Khan II died in 1734 and his son Sidi Abdullah inherited outright, passed by a son's birthright, and that inheritance is the hinge into the collapse the next sections trace. The next section takes the 1670 episode in full.
Three Officers and a Chain, 1655–1670
The system's sternest test came under Fateh Khan, Wazir from 1655, who held Janjira through the opening years of Shivaji's campaign against it. A modern account of that campaign records that by 1661 'relentless attacks by the Marathas' had worn Fateh Khan down to the point of resolving to surrender the fort [13] — and it was here that the elective office showed its teeth: three of his own Sidi officers, Qasim, Sambal and Khairiyat, 'did not approve of this surrender' and, having won the support of their compatriots, had Fateh Khan seized and put in chains [13]. The archive's own record of the sequel gives the mechanism rather than the drama: Kasim and Khairiyat, who were brothers, 'waived their claims in favour of Sidi Sambal, who was accordingly appointed governor' [6], and Sambal then transferred the Sidi fleet's loyalty to the Mughal emperor, receiving the title of admiral in return. The office survived its own Wazir's private decision to quit the fight; the fleet, not the Wazir, held the loyalty that mattered. It is the one moment in the whole elective period where force decided a succession — and even then the force deposed the Wazir who meant to surrender the fort rather than installing the men who used it: Kasim and Khairiyat waived their own claims in favour of Sambal, who took the post they might have kept for themselves.
The Slaves Who Ruled, 1734–1792
The system that had run for eleven decades without bloodshed broke in a single year. Sidi Sirul Khan II died in 1734, and his eldest son Sidi Abdullah, briefly ruler, was murdered that same year — but the record splits on who did it. The main narrative blames Sirul's own 'youngest brother,' who then 'usurped the government to the prejudice of Sidi Reheman, an elder brother'; a separate telling, attributed to 'Janjira records,' instead names the killer as 'Sidi Sambal, a slave of Sidi Sirul,' without stating whom Sambal displaced [8]. Whichever account is closer to the event, the ruler who followed was Sambal: he held the throne for two years before three of his own fellow slaves, Sidi Sallam, Sidi Faraz and Sidi Syad, murdered him in turn and placed Sidi Reheman on the throne [8]. Reheman was removed from power in 1739 and his brother Sidi Hasan installed in his place [9]; on Hasan's death in 1745 one Syed Allana usurped the chiefship 'for a time,' before it was 'recovered in the following year by the rightful heir Sidi Ibrahim Khan' [9]. Ibrahim Khan I's reign ended in 1762 when he too was murdered by his own slave, who 'usurped the chiefship to the prejudice of Abdul Rahim, the nearest heir, and ruled as Sidi Yakut Sanni' [10] — Abdul Rahim only came into his inheritance a decade later, in 1772, when Yakut Sanni died shortly after a settlement that had finally promised him the succession. Abdul Rahim's own death in 1784 was followed at once by a fourth seizure: Sidi Johar, 'the commandant of Janjira,' took the chiefship 'to the exclusion of Abdul Rahim's eldest son Abdul Karim Khan' [10] — a disinherited son who instead went on to found the separate line of the Sachin Nawabs. Johar's own successor, Sidi Ibrahim Khan II, installed in 1789, was thrown into prison in 1792 by his own slave Sidi Jumrud Khan [10]. In this fifty-eight-year stretch the record shows ten transitions of power and five of them — half — were usurpations; three of the five are explicitly described as the ruling Sidi's own slave turning on his master, and two of the five ended in the ruler's murder rather than his mere imprisonment or exclusion.
British Arbitration Ends the Cycle, 1804–1879
The cycle closed as abruptly as it had opened. Jumrud Khan died in 1804 and Ibrahim Khan II was restored to a chiefship he held without further incident until his own death in 1826, when he was 'succeeded by his son Sidi Muhammad' [11] — the second time in twenty-five transitions that a son inherited on a father's natural death with no rival and no chains, and the only one of the two whose reign then went unmarked by murder: Sidi Sirul Khan II's own son Abdullah had inherited the same way in 1734, only to be murdered within the year (see 'The Slaves Who Ruled' above). Muhammad in turn abdicated voluntarily in 1848 'in favour of his son Sidi Ibrahim Khan' [11], a second peaceful handover in this closing period, though by gift rather than by death. Political authority was narrowing at the same time the succession was calming: the Imperial Gazetteer records that 'till 1868 the State enjoyed singular independence, there being no Political Agent,' but that year the 'maladministration of the chief, especially in matters of police and criminal justice,' brought internal administration under a British Political Agent, formalised by treaty in 1870 [2]. The last transition this archive traces, in 1879, shows that new arbitration at work directly: on Ibrahim Khan III's death, Sidi Ahmad Khan's own sardars first installed an elder but illegitimate half-brother 'in spite of the protest of the Assistant Agent'; the Bombay Government annulled the election, and 'Sidi Ahmad Khan was recognised as Nawab' [12] — the throne's first transition decided not by the fleet's own vote, nor by a slave's dagger, but by a colonial government's veto.
What the Chain Shows
Counted across all twenty-five transitions the register and the Bombay Gazetteer preserve for Janjira, the pattern falls into three distinct regimes rather than one continuous monarchy. From 1618 to 1734 — ten transitions over 116 years — the fleet's elective rule held with only one episode of force; that one, in 1670, deposed a Wazir who meant to surrender the fort rather than installing the officers who used it. From 1734 to 1792 — ten transitions in fifty-eight years — five were usurpations and three of those were a slave turning on his own master; violence, not election or inheritance, was the modal way the throne changed hands in this period. From 1804 onward, British political oversight closes the cycle: two peaceful hereditary transfers (1826, 1848) and one transfer settled by an external government's ruling rather than by the fort's own officers or families (1879). Ordinary inheritance from a father who had simply died happened twice in the whole span — Abdullah in 1734 and Muhammad in 1826 — but only Muhammad's succession went undisturbed by what followed; Abdullah was murdered within the year of taking the throne. What held the state together across two and a half centuries of Ahmadnagar, Bijapur, Mughal, Maratha and British contention was never a dynasty in the usual sense — it was an office, seized, elected, murdered for and finally arbitrated, but rarely, in the record this archive holds, simply passed down.
Dates in this essay
The founding of Sidi rule at Janjira has two independent tellings the archive holds separately: the Imperial Gazetteer dates the box-ruse seizure to 1489 and names Ram Patel as the Koli captain deceived [1]; the Bombay Gazetteer's own main account instead credits Ahmad Shah of Ahmadnagar with taking the fort by siege 'about 1490' and installing his slave Yakut, adding an undated fraud account (naming Perim Khan, not Ram Patel) directly after it as an alternate version [bg-p434-founding, bg-p435-box-ruse]. This essay keeps both tellings visible rather than collapsing them into one date.
This archive dates the mutiny that chained Fateh Khan and elected Sidi Sambal to 1670, following the Bombay Gazetteer's own sequencing of Bijapur Wazirs (Ambar d.1642, Yusuf d.1655, then Fateh Khan, then Sambal) [bg-p436-elective-wazirs, bg-p437-mutiny]. A tier-C secondary source, Wikipedia's article on Shivaji's invasions of Janjira, dates Fateh Khan's decision to surrender and the mutiny that stopped him to 1661 [13]; no primary source for that earlier year was opened during this research pass, so the archive keeps 1670 and records the discrepancy rather than silently adopting the uncited year.
The 1734 succession crisis is itself internally layered: the main narrative on Bombay Gazetteer p. 444 attributes Sidi Abdullah's murder to 'his youngest brother,' who usurped 'to the prejudice of Sidi Reheman, an elder brother,' while the same page's account of Janjira's own state records attributes the murder instead to 'Sidi Sambal, a slave of Sidi Sirul' without stating whom Sambal displaced [8] — two explanations on one page, both quoted above rather than resolved into one, and the essay body no longer merges the brother-account's 'to the prejudice of Reheman' clause into the slave-account's wording.
Where authorities disagree about when a state began or ended, the archive follows one rule — a polity is dated from the assumption of sovereign power in its own capital, and to the settlement that extinguished it — and records the alternatives rather than hiding them.
Open questions
Whether Fateh Khan's mutiny belongs to 1661 (Wikipedia, uncited) or 1670 (this archive's own reading of the Bombay Gazetteer's Wazir sequence) is unresolved; no primary source for the earlier date has been opened.
Whether Sidi Johar, who seized the chiefship in 1784, was himself a slave or freeborn officer is not stated on the page opened — he is called only 'the commandant of Janjira,' unlike Sambal, Yakut Sanni and Jumrud Khan, each explicitly named as a slave of the ruler he displaced.
No marriage, maternal, or building record for any Janjira ruler has yet entered this archive (the house records for this polity return empty MARRIAGES, MOTHERS and BUILT tables) — an evidentiary gap in what has been read so far, not a claim that none exist.
Whether the excluded heir Abdul Karim Khan (Balu Mia), who went on to found the separate Sachin Nawabs after being passed over at Janjira in 1784, kept any relationship with the Janjira line afterward is untouched by the pages read for this essay.
What became of Fateh Khan himself after being 'thrown into chains' in 1670 — whether he died in captivity, was released, or left Janjira — is not stated on the page opened.
Which of the two competing 1734 accounts (brother-usurper vs. slave-usurper Sambal) is more reliable is not resolved by the page opened; both are Bombay Gazetteer tellings of the same event and neither is corroborated by an independent source in this research pass.
Whether Syed Allana, who usurped the chiefship in 1745, held any prior position at Janjira is not stated on the page opened — he is introduced only by name, unlike the three usurpers explicitly identified as a ruling Sidi's slave.