The trigger: a purge at Dhar, reported from inside the court
The best contemporary witness to the revolt was hostile to it and stood beside the man it was aimed at. Ziauddin Barani, seventeen years at Muhammad bin Tughluq's court, locates the origin of the rising not in ambition but in a massacre: the Sultan's appointee at Dhar "one day got together about eighty of the foreign amirs and chiefs of the soldiery" and "lie had them all beheaded in front of the palace." The slaughter, Barani says, "caused those of Deogir, and Gujar&t, and every other place to unite and to break out into insurrection." When the news reached the Sultan, "he sent 'Aziz a robe of honour and a complimentary letter." The other garrisons could now read the policy. [7]
The mechanism at Daulatabad was a summons. Ordered to send fifteen hundred horsemen and their senior officers to Broach, they stopped at the first stage and "con- sidered that they had been summoned to Broach in order to be executed, and if they proceeded thither not one would return." They killed the escorts, seized the governor, brought the treasure down from the fortress and distributed it to the troops; then "The Mahratta country was apportioned among these foreign amirs". The sultanate begins not as a conquest but as pre-emptive self-preservation by an officer corps that happened to be sitting on an imperial treasury. [7]
Who the rebels were, and a word a translator changed
The English literature calls these men the foreign amirs, and that phrase is a nineteenth-century substitution rather than a translation. Dowson's footnote corrects it: the Persian word is a rank, the "Mughal title for a centurion or commander of a hundred", and of Briggs, who rendered it as foreign chiefs, he writes that "he is not justified in his alteration of the original word." [7] The correction changes what kind of body revolted: not an ethnic bloc but a grade on the army list, whose members happened largely to be immigrants and who also held office. As Sahapedia puts it, "The Sultan was particularly suspicious of the amiran-i-sadah (centurions) who held important positions in the provincial administration." [12]
Delhi had spent a generation making the Deccan governable by its own hands, and the rebels inherited the apparatus whole. The transfer of the capital to Daulatabad "also involved the migration of the state officials and the entire population that drew patronage from the imperial court", and the Sultan rebuilt the fort so thoroughly that Ibn Battuta in 1342 "remarks that the fort has no equal in Hindustan in terms of impregnability". [12] The new state stood on Delhi's own investment: a transplanted population, a rebuilt fortress, and a corps Delhi had recruited and then frightened.
It was also not an exclusively Muslim affair: at Deogir "ho people of tho country joined them". [7] James Grant Duff, writing in 1826 from Marathi as well as Persian material, pressed further than his evidence could carry - "there is strong presumption of their having contributed more to its success, than the Mussulman historian was aware of" - and describes a settlement in which the sultan "either bestowed hereditary lands and immunities on the leading men, or confirmed them in those they already enjoyed, as Deshmookhs and Deshpandyas." [8]
The first rebel king, and where sovereignty was assumed
Hasan was not the rebels' first sovereign, and the man he replaced has two irreconcilable names. Barani is flat: "Then they made Makh Afghan, brother of Malik Yak Afghan, one of the foreign amirs, their leader, and placed him on the throne." [7] Ferishta, two and a half centuries later, calls him Nasir-ood-Deen Ismael and has him abdicate before a council, "declar, that on account of his age he was unfit to rule in such times". [6] The familiar modern name fuses the two: "Ismail Mukh used the excuse of his old age and abdicated in favor of Zafar Khan." [15] That compound is in neither source, and the three accounts of his later death - executed after a conspiracy, killed in the battle, poisoned - cannot all be true. [1]
Where and when Hasan took the throne is contested in the same way. Barani, the only contemporary, says only that "Hasan K&ngu then proceeded to Deogir and assumed royal dignity", with no date. [7] Ferishta puts the coronation at Daulatabad on 24 Rabi II 748, which Briggs's margin converts to 12 August 1347, and has Gulbarga become the seat afterwards: "Kool- burga, which now became the royal residence, was named Ahsunabad". [6] The Encyclopaedia of Islam agrees: "Shortly after being crowned the first Bahmanī sulṭān in Dawlatābād's mosque, ʿAlāʾ al-Dīn moved the capital 320 kilometres southwest, to Gulbargā." [5] Iranica dissents on both place and conversion: "Ẓafar Khan retreated to the fort of Gulbarga, where, on 24 Rabīʿ II 748/3 August 1347, he was proclaimed king." [1] The usual formula - 3 August 1347 at Daulatabad - takes its date from one authority and its place from another.
The origin question: four answers, none of them safe
Where Hasan came from is the central transmission problem, and the sources are irreconcilable: "ontemporary and later sources provide conflicting accounts of his birthplace, ancestry, and social background." [14] Isami, the only contemporary, says Ghazni. Iranica calls him a Khorasani adventurer claiming descent from the Sasanian Bahram Gur, while floating H. K. Sherwani's philological alternative: "Others have read the names Gangū or Kānkūy as Kākūya, suggesting his link to the Kākūya of Isfahan." [2] Iranica's Gulbarga entry keeps the hedge visible - "a rebel Tughluqid commander, perhaps a descendant of the Kakuyids of Isfahan". [3] Ferishta gives a fourth answer in six words: "The King himself was by birth an Afghan."
The famous story - that Hasan served a Brahman astrologer named Gangu, ploughed up a buried hoard, and took the dynastic name from his master - is Ferishta's, and Ferishta disowns it as he tells it: "I shall merely state that which is most generally believed in the Deccan." The reward in the tale is the "com- mand of one hundred horse" - the rank of amir-i sada itself, the legend explaining precisely the office from which the real revolt was made. The same passage demolishes the grander pedigree: Ferishta had seen a Persian genealogy in the Ahmadnagar royal library, but "it was probably only framed, after his accession to the throne, by flatterers and poets." Briggs then footnotes his own author, noting that "bramins, not only in his time but long anterior to the Mahomedan conquest of the Deccan, were the only persons who kept the public accounts" - the legend explains an administrative fact that predates it. [6]
Recent scholarship dates the legend rather than adjudicating it. Dusan Deak establishes the decisive negative: "none of the pre-17 th century Muslim historians who mention the king as Hasan Gangu relate the story of him being a servant of a Brahman. Only later historians do so." It surfaces around 1600, in Ferishta and in Rafiuddin Shirazi - whose version is a different story, with Gangu-Pandit not a master but a stranger who reads royal signs in a cobra shading the sleeping boy. Sumit Guha, cited by Deak, reads the Marathi retellings as "a legitimating myth of the politically subordinate, economically deprived, and yet socially elite Marathi Brahman scribes", the class that actually ran Deccan revenue; "he phonetic similarity of the words Bahman and Brahman" then fed a twentieth-century politics of dividing the Indian past in two. [4]
Against all of it stands the silence of the one man present. Isami "became the earliest panegyrist at the Bahmani court" [17], and his Futuh us-Salatin calls his patron Hasan and Zafar Khan, never Gangu. Deak's verdict: "Clearly, the panegyrist, apart from linking Hasan to Bahman, obscured the details of the origins of the sultan." [4] A last caution: the line often quoted as Barani's, that Hasan was "born in very humble circumstances" and spent thirty years as a field labourer, appears nowhere in Dowson's translation of Barani. [16]
What the state said about itself, in silver
One class of evidence is not transmission at all. A tanka of 10.29 g struck at Gulbarga under the mint-name Ahsanabad in AH 756 (1355) reads "al-sulṭān al-a‘zam ‘alā' al-dunyā wa'l-dīn abū'l-muzaffar bahman shāh al-sulṭān" and, on the reverse, "sikandar al-thānī yamīn al-khilāfa nāṣir amīr al-mū'minīn" - a second Alexander, right hand of the Caliph. [9] The claim made in metal is caliphal and Alexandrine: not Brahmanical, not explicitly Sasanian, and carrying no Gangu - an absence from the coins and inscriptions that has itself been used against the legend. Ferishta records the act at the accession: "The Khootba was read, and coins struck, in his name". [6]
Behind the propaganda sat an inherited fiscal machine. Wagoner and Tandon show that "he Bahmanis of the Deccan produced a limited copper coinage from the very outset of the state's founding in AH 748/1347 CE, but it was clearly secondary to the silver tankas and their fractions upon which the monetary system was based." The system adapted Delhi's: near-pure gold, a silver tanka of 11.02 g with fractions down to a twelfth, a copper coin of 3.67 g for the bazaar - all aniconic, bearing only the sultan's titles in calligraphy. The inversion toward heavy copper came only with the reform of the 1420s. [10]
Gulbarga: a capital, a borrowed Persia, and four provinces
Gulbarga held the capital for about seventy-seven years, "until 1424 when the ninth sultan, Ahmad Shah Wali (1422-36), moved his capital to his newly reconstructed town of Bidar." [13] What survives of the founder's city is thinner than its reputation: the citadel "contains only two surviving contemporary or near-contemporary buildings: a rectangular donjon or keep, perhaps of pre-Bahmanid date, and the Jāmeʿ Mosque", itself "nique among Indian mosques in being entirely covered" and possibly not designed as a congregational mosque at all. [3] Its date is unsettled - "he precise dating of the mosque and the reasoning behind its revolutionary plan are uncertain." [5] What is protected today is "the Gulbarga Fort and Great Mosque in the Fort and the Haft Gumbad complex with seven tombs". [11]
The Persian pedigree Ferishta called a forgery was nevertheless built in stone. The early Bahmani tombs carry "a curious device of a pair of outspread wings cradling the crescent moon", likened to the Sasanian royal emblem; court protocol used a black parasol, and the turquoise throne stood in for the Sasanian standard. Iranica draws the inference this archive follows: "The various rumors about Ḥasan Gāngu's origins may have been partly responsible for the Bahmanid sultans' insistence on publicizing their claim to Sasanian lineage by every means at their disposal." [1] The other legitimating current was Sufi: the Encyclopaedia of Islam holds that Bahmani sovereignty hinged on the ruler's relationship with saints seen as prophesying kingship, while dating the best-known such story as " late-tenth/sixteenth-century tradition". [5] Ferishta shows the founder investing in it, distributing gold and silver at his accession "in the name of Nizam-ood-Deen Oulia, who had predicted his good fortune." [6]
The state he left was dynastic and fragile. Ferishta has him divide the kingdom in his last illness into four provinces, two of which went to the family of one minister who was also his son's father-in-law. He died "eleven years, two months, and seven days, after his accession to the throne", explaining his success as won "by kindnesses to my friends, generosity to my enemies, and by courte- sy and liberality to all mankind." [6] How nearly it came apart is visible in Barani: within a decade the founder's own son-in-law wrote to Delhi from Daulatabad that "a dispute had arisen between him and tho son of Hasan Kdngu, he therefore solicited the Sultdn to come himself and assumo the seat of royalty." The invitation to reconquer the Deccan came from inside the family; Muhammad bin Tughluq went to Thatta instead, and died there. [7]
Dates in this essay
Accession date: Ferishta (in Briggs) gives Friday 24 Rabi II 748, converted in Briggs's margin to 12 August 1347; Encyclopaedia Iranica (Shokoohy, 'Hasan Gangu') gives the same Hijri day as 3 August 1347. The two CE dates are rival conversions of one Hijri date, and the archive states both rather than choosing.
Place of accession: Barani (contemporary) and Ferishta both place the assumption of sovereignty at Daulatabad/Deogir; Encyclopaedia of Islam THREE agrees and gives Daulatabad as the first capital, 748-51/1347-50, with the move to Gulbarga following. Encyclopaedia Iranica places the proclamation at Gulbarga itself. The common modern formula '3 August 1347 at Daulatabad' takes the date from Iranica and the place from Ferishta/EI3.
Death of the founder: Ferishta (Briggs) gives 1 Rabi I 759, marginally converted as 10 February 1358, in his sixty-seventh year; Iranica's 'Hasan Gangu' entry gives '1 Rabiʿ I 759/11 February 1357', an internally inconsistent conversion; Iranica's 'Bahmanid Dynasty' entry ends the reign in Rabi I 759/February 1358.
Move of the capital to Bidar: Shokoohy (Muqarnas) says 1424; Iranica's Gulbarga entry says around 827/1424; Wikipedia's infobox says 1425. The archive follows the 1424 reading of the Hijri year.
End of the dynasty: Iranica's 'Bahmanid Dynasty' entry and Encyclopaedia of Islam THREE both run it to 934/1528 (Kalimullah); Iranica's Gulbarga entry gives 748-934/1347-1527; other accounts treat 1518, when Golconda became independent, as the effective end. By the archive's convention - dating a state to the settlement that extinguished it - the terminal date is the deposition of the last Bahmani sultan, but the rival spans 1527, 1528 and the effective end in 1518 should all be visible to the reader.
Date of the Gulbarga Jami Masjid: Encyclopaedia of Islam THREE holds that 'the precise dating of the mosque and the reasoning behind its revolutionary plan are uncertain', and sets out both readings - a stone inscription dated 769/1367-8 naming the patron Muhammad I and the architect Rafi b. Shams b. Mansur al-Qazvini, against an argument that the building dates to after 809/1407 under Firuz Shah and was conceived as an audience hall. Encyclopaedia Iranica's Gulbarga entry names the same patron and architect but gives no date. UNESCO's tentative-list dossier simply asserts 1367.
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
Was the accession at Daulatabad or at Gulbarga, and does the nine-day gap between the two published conversions of 24 Rabi II 748 reflect a calendrical dispute or simply an uncorrected error in one of them?
Who was the first rebel sultan? Barani's contemporary 'Makh Afghan, brother of Malik Yak Afghan' and Ferishta's 'Nasir-ood-Deen Ismael' cannot both be right, and the modern compound 'Ismail Mukh' appears in neither source.
How did the abdicating rebel king die? Ferishta has him executed after a later conspiracy; Barani and Iranica have him killed at the battle of Daulatabad; R. C. Majumdar reportedly has him poisoned after being turned by a local chieftain named Narayana. The three accounts have not been reconciled against a primary text.
Where did Hasan actually come from - Ghazni (Isami), Khorasan (Iranica), Afghan birth (Ferishta), or the Kakuyid line of Isfahan (Sherwani's reading of Gangu/Kankuy as Kakuya)? No source read here settles it, and the coinage is silent on ancestry.
The line widely attributed to Barani, that Hasan was 'born in very humble circumstances' and spent 'the first thirty years of his life' as a field labourer, does not appear anywhere in Dowson's translation of the Tarikh-i Firoz Shahi in Elliot & Dowson vol. III, which was searched in full text. The wording appears to derive from J. D. B. Gribble, A History of the Deccan (1896), p. 15. The attribution to Barani should be treated as unverified and probably wrong until Gribble or Haig is opened directly.
J. D. B. Gribble, A History of the Deccan, vol. 1 (1896), and Wolseley Haig, Historic Landmarks of the Deccan (1907), pp. 29-31, are the load-bearing citations behind the modern 'Rebellion of Ismail Mukh' narrative. Both are on archive.org but the item identifiers tried during this research returned error pages; neither was opened.
S. A. Q. Husaini, Bahman Shah: The Founder of the Bahmani Kingdom (Calcutta, 1960), is the most extensive treatment of the origin question and is known here only through Deak's summary - including Husaini's dismissal of the Brahman story on the ground of Ferishta's Shi'a leanings, and his alternative derivation of 'Gangu' from Gangi, a village near Miraj. It does not appear to be online.
H. K. Sherwani, 'Gangu Bahmani', Journal of Indian History 10/1 (1941), pp. 65-99, and Sherwani, The Bahmanis of the Deccan (1953/1985) - the source of the Kakuyid hypothesis and of the account of the first Bahmani campaigns at Kodgir, Kalyani and Akkalkot - are known here only at second hand.
Tabataba's Burhan-i Maasir and Rafiuddin Shirazi's Tazkirat al-Muluk, in J. S. King's translations in Indian Antiquary 28 (1898-99), were not opened; the Shirazi cobra/Junaydi variant and Tabataba's 'illustrious descent from Kaiyumars' are therefore quoted here only through Deak.
Isami's Futuh us-Salatin itself (Agha Mahdi Husain edition, Agra 1938 / Asia Publishing House 1967; India Office MS 895) was not read; no verses on the accession were consulted directly, and everything said here about the only contemporary account of the founding is second-hand.
Only the abstract, key-takeaway summary and figure captions of Shokoohy, 'Sasanian Royal Emblems and Their Reemergence in the Fourteenth-Century Deccan', Muqarnas 11, pp. 65-78, were readable; the year is itself unsettled in the sources read here - the Academia.edu page dates it 1994, Encyclopaedia Iranica's bibliography 1994, while it is commonly cited as 1993; the argument itself is behind a paywall.
Does any coin or inscription of Bahman Shah anywhere carry the element 'Gangu'? Deak reports scholars citing its absence, but no numismatic or epigraphic corpus (Goron & Goenka; Epigraphia Indo-Moslemica) was consulted here to confirm the negative.
Marie H. Martin's 1980 Michigan dissertation, Bahmani Coinage and Deccani and North Indian Metrological and Monetary Considerations, 1200-1600, underlies the metrology cited here through Wagoner and Tandon; only its abstract surfaced in search, and the abstract page was not opened.
Simon Digby's account of Shaykh Sirajuddin Junaydi transferring his turban to Zafar Khan, cited by the Encyclopaedia of Islam, rests on a primary text not identified during this research.
Does the founder's own tomb west of the Gulbarga fort carry the winged Sasanian device? Iranica attributes the emblem generally to the early Bahmani tombs; the readable Shokoohy captions describe tombs only 'probably' belonging to the first sultans, and none is positively identified as Bahman Shah's.
The UNESCO tentative-list dossier used here for the composition and ASI protection of the Gulbarga group also twice asserts that Gol Gumbaz is the second-largest dome in the world, which is false, and asserts without date that the Gulbarga fort was built on the remains of a Warangal fort of a Raja Gulchand, which no scholarly source read here supports. Neither claim is relied on above, and the fort's pre-Bahmani history remains unresolved.
Was 'afaqi' a contemporary term for the immigrant faction? The Encyclopaedia of Islam holds that the contemporary word is gharibān and that āfāqī is largely a product of twentieth-century scholarship, following Fischel and Eaton; Iranica's Gulbarga entry uses āfāqī as though contemporary. The dispute is lexical, not factual, but the archive has not resolved which usage to standardise on.