The rise: a bookkeeper displaces his own patron
Madanna and Akkanna entered history in the household of Sayyid Muzaffar, the noble who had put Abul Hasan Qutb Shah on the Golconda throne in 1672 and then governed in his name. What followed is dated precisely because Dutch East India Company factors were writing it down as it happened: on 7 January 1674 the sultan placed the Sayyid and his sons under house arrest, and the next day the state records, accounts and treasury were carted from his house into the palace. Madanna, the Sayyid's own accountant, became the sultan's principal representative from one day to the next; the first farman he signed is dated 14 October 1674 [2]. French observers reached for the word revolution.
Jadunath Sarkar, writing from the Persian chronicles, tells it as court intrigue rather than administration: Abul Hasan 'bought over Madanna, the Brahman factotum of Muzaffar', corrupted the Sayyid's captains through him, and transferred the wazir-ship [9]. Richards and Narayana Rao register the outcome without melodrama: two Telugu Brahmins rose 'by means of administrative skill and political acumen' to become chief minister and chief army commander to the last independent king of Golconda, exercising virtually complete control of the state apparatus [14].
Their king is remembered under a nickname that half-explains how a bookkeeper could take the government: Abul Hasan became Tana Shah, 'King of Taste', for his love of the arts and of pleasure [21]. By the 1680s the relation had inverted. Company records for 1683 list five occasions on which the sultan went to Madanna's house to ask for money. Havart's version, in Sarkar's rendering, is that the sultan 'left everything to his care and was satisfied with receiving 75,000 dollars for his personal use' [13].
What office Madanna actually held
Popular accounts make Madanna Peshwa or Mir Jumla, dignified as Surya Prakasa Rao; the documents do not support it. Gijs Kruijtzer, working from at least ten original farmans signed 'Madaw Pandit' or 'Madaw Bhanji', finds that Madanna's standing office throughout was majmu'adar shahi - the king's collector or bookkeeper - and that 'There is no evidence that his position was ever officially described as peshwa or that he held the title of Mir Jumla, as is often assumed' [2]. What he carried was an honorific of intimacy rather than rank, Confidante of the Secrets of the King of Kings; Akkanna was chief bookkeeper of the crown estate.
The independent check comes from an unlikely archive. Two Golconda miniatures of the brothers, painted in the later 1670s and captioned in Persian for a Dutch buyer, survive in the Witsen Album at the Rijksmuseum, whose catalogue describes the first as a 'Portret van Madanna Panit, belastingklerk (majmudar) van Abul Hasan Padshah' - tax clerk [15]. In the picture they had made of themselves for export, the men who ran Golconda were accountants. Such portraits are a by-product of their politics: Madanna and Shivaji, both prominent in the surviving European series, 'were instrumental to the VOC's ascendancy in the region vis-a-vis other European companies' [16].
The reform: tax farming turned into salaried collection
Neither brother was an accident: the word Brahmin had become almost a synonym for financial administration in the Deccan, and Golconda, running on revenue farming rather than the Mughal jagir, had produced a class of non-combatant specialists half a century before the men of the pen rose in Hindustan. Madanna was a product of that system and, once in power, began turning it about. The best-attested measure is recorded by Daniel Havart, who served the Company at Masulipatnam during the ministry. The hawaldar of Masulipatnam had taken the port in farm for some hundred and fifty thousand ducatons a year, which he had to produce whatever the income, 'failing which he would be incredibly pestered, beaten or even killed. In the year 1676 Mister Madanna has introduced another practice. Now the hawaldar is on a salary, he has twelve hundred ducatons a year, free cost, free servants, and pomp and circumstance, and then so much inspection that he would not be able to embezzle so much as a penny' [3]. Private gifts stayed with the officer; anything presented openly went to the treasury.
Whether this was good government is exactly where the historiography splits. Sarkar's verdict is flat: 'During the ten years of Madanna 's ministry the internal administration of the country was marked by the same disorder and tyranny as in the reign of Abdullah, and matters naturally declined from bad to worse' [10]. He cites the Madras factors of July 1676 complaining that nothing was thought of but peeling and squeezing the people. Kruijtzer, reading the Dutch archive instead, sees a politique driven mainly by a desire for peace and stability - while conceding that the brothers' relations with Hindu zamindars were poor, plausibly because of those same reforms. That archive is not neutral either: in July 1685 the Company's factors accused Akkanna of extorting goods worth 145,030 hons, a charge repeated by later historians until documents found in April 1686 showed the money had been lost by the Dutch factors themselves in private trade.
Shivaji, and the policy of paying rather than fighting
The brothers' foreign policy rested on one insight: Golconda could not survive a Mughal war and should therefore purchase everyone else. Sarkar records that Madanna 'had already made a subsidiary alliance with Shiva, promising him an annual tribute of one lak.h of hun for the defence of the realm' [11]. When Shivaji came to Hyderabad early in 1677 with sixty thousand men, a secret compact gave him three thousand hun a day, auxiliaries and a siege train for the Karnatak campaign against a share of the conquests. By the time Golconda fell, power was wielded by two Brahmin brothers, through whom the Qutb Shah had established a friendship even with Shivaji [26].
The alliance did not repay: Shivaji kept every fort he took, and Sarkar judges that the diplomatic system Madanna had built fell to the ground. After that the policy narrowed to cash. When a Mughal army approached Hyderabad in 1683, the Dutch observed no preparations beyond provisioning a dilapidated castle: 'the daft Brahmins do not know of any military affairs, they let all bounce on the purse and know how to deflect all wars with money and blindfold the rulers, as can be seen from the substantial amounts of money that have been sent to Aurangzeb over a short period' [7]. Kruijtzer totals the remittances of March to July 1683 at 855,000 rupees, over and above a tribute risen to some five million.
Twice the ministers restrained their own king. In 1683 the sultan schemed with the Persian envoy to fund the fugitive Mughal prince Akbar, concealing it from Madanna; that crisis produced Akkanna's private interview with a Dutch factor, in which he argued that Brahmin government served the king better than the incomers' did, 'ours being fullheartedly devoted to the welfare of the country, while we are not people who have or seek other countries, but that of the Moors is only to the end of becoming rich and then to leave for those places which they consider to be either their fatherland or holy' [1]. A year later, when Abul Hasan insisted on staging an elephant fight - an imperial prerogative - he mocked Madanna's objection before the court as that of a good Mughal ambassador.
The war of 1685 and the killing
In July 1685 Aurangzeb, holding a letter proving Golconda's aid to Bijapur, sent his son Shah Alam south. In Sarkar's narrative, east of Malkhed the Mughal vanguard was barred by a Golconda army under Mir Muhammad Ibrahim, Shaikh Minhaj and Rustam Rao - whom Sarkar calls Madanna's nephew and Kruijtzer, naming him Yenganna, calls his cousin. The line broke by defection rather than battle: Ibrahim was seduced by the Mughals, and Shaikh Minhaj told the sultan that since Ibrahim was Madanna's protege the desertion must have had the minister's consent. Shah Alam entered Hyderabad and plundered the palaces; his terms were 'the ceding of some disputed territories to the Mughals; grant of a large lump sum; and the dismissal of Madanna and Akkanna, the Brahmin ministers of the kingdom' [20]. The sultan agreed and then did nothing.
The killing is reported in two irreconcilable settings. Havart has the brothers leaving their town houses for the fortress, where they were killed by Sidi Makta - an Abyssinian whom Madanna had removed from office - and other slaves at the order of 'an old woman'; Khafi Khan has them struck down on the way home from the darbar by one Jamshid, the murder commissioned by Saruma and Jani Sahiba, aged widows of the previous sultan who wielded supreme power in the palace [5]. The bodies were dragged through the streets; 'The head of Madanna was severed from the body and sent to Aurangzib, while that of Akkanna was trampled under the feet of an elephant' [13]. The Dowager Sultana sent the heads on 'as the best of peace-offerings' [12], and looting of Brahmin houses spread fifty kilometres down the Masulipatnam road.
One report accuses the dead men of the treason for which they died: a Company peon said that Madanna, Akkanna and Muhammad Ibrahim had written to the Mughal commanders offering 'that they would present them with the keys of the kingdom, on the condition that they would leave to them the administration from there to the borders of Delhi' [6]. The territory claimed corresponds to the brothers' native Warangal country, which lends the report some weight, and Martin and Bhimsen repeat the charge. Against it stands the Dutch factory's own report, detailed on the murder and silent on treason.
Afterlives: what survives, and what the sources will not settle
The murder is usually written as the end of Brahmin power at Golconda; the Company records show it was not. Shah Alam left within a month, the sultan resumed control, and the faction reassembled - in April 1686 Vessanna, Madanna's eldest brother, was invested with the dead man's dignities. The problem underneath is one of evidence: versions differ according to which of four bodies of material one relies on - the contemporaneous Dutch records, the Mackenzie manuscripts compiled over a century later, the Muslim court historians, and evolving Hindu folklore [19].
The material traces are scattered and indirect. Akkanna's costly Shiva temple at Maheshwaram was among the first buildings destroyed when Shah Alam entered the city - 'an event which confirms Richard Eaton's thesis that temple destructions by Muslim invaders generally concerned temples with a strong connection to ruling figures' [8]. India's UNESCO tentative-list dossier lists the 'Akkanna-Madanna Offices' and the Ramdas Jail inside the citadel at Golconda [17]. The Mahankali temple at Shah-Ali-Banda carries their names on the tradition that they worshipped there each morning before court [23]; the rock-cut caves at Vijayawada carry them anachronistically, being seventh-century or earlier [22].
Their most durable afterlife is devotional. Telangana's Endowments Department states that 'Kancharala Gopanna was the nephew of Akkanna and Madanna' and that on their recommendation he became tahsildar, going on to spend collected land revenue on building the great Rama temple at Bhadrachalam [18]; press retellings make them his cousins or brothers-in-law instead [24]. Sarkar reports the rest - six lakhs of government pagodas spent on shrines, arrest, torture, release through the intervention of Ram and Lakshman - as hagiography. In popular history the brothers have become a visionary ministry cut short by fanaticism, complete with a date the Dutch archive does not support [4]: one widely read account has them ambushed on 16 March 1686, dragged from their palanquins and beheaded [25].
Dates in this essay
Date of the murder: the standard date is March 1686. Sarkar's own narrative puts the killing 'early in March, 1686' and cites Khafi Khan, Ishwardas and Dilkusha for it; Kruijtzer traces Sarkar's dating to a remark about the severed heads of the brothers in the Ma'asir-i Alamgiri. The Wikipedia account of the siege (March 1686) and most popular retellings, including Live History India's 16 March 1686, follow him [wikipedia-siege-of-golconda] [livehistoryindia-madanna]. Gijs Kruijtzer rejects it, placing the killing 'in the last days of October 1685' on the strength of two detailed VOC reports, a November 1685 letter from East India Company personnel in Masulipatnam saying the brothers 'are certainly cut off', and the Mackenzie 'Annals of Condaved' date of Monday 2 Shukla Karttika, Krodhana - 29 October 1685 [kruijtzer-ch6-date-of-murder]. On Kruijtzer's reading the March 1686 entry records the arrival of the head at the imperial camp, not the death; it also makes the murder near-simultaneous with Muhammad Ibrahim's defection rather than a consequence of a long political crisis.
Dates of the ministry: Madanna's effective power begins with the arrest of Sayyid Muzaffar on 7 January 1674, though his first surviving signed farman is dated 14 October 1674; the ministry ends with the killing, on either dating. Sarkar and most later writers describe it as a decade of power. Akkanna's appointment as sar-lashkar dates from 1682.
Reign dates of Abul Hasan Qutb Shah: given as 1670-86 by Richards and Narayana Rao [richards-narayanarao-banditry] and as 1672-86 by The Hindu [thehindu-bhadrachalam-saint]; the accession is conventionally placed in 1672, and 1670 appears to reflect a different reckoning of the succession crisis.
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
Were the brothers Telugu Brahmins or Maharashtrian Brahmins? Richards and Narayana Rao, and Sanjay Subrahmanyam, call them flatly Telugu; Kruijtzer argues from Modi-script dating on the Telugu side of Madanna's bilingual farmans, and from Marathi records reported among descendants, that they may have belonged to a branch of the Maharashtrian Brahmins long settled in Telingana - a question he notes has been debated between Maharashtrian and Andhra historians.
What was their sectarian affiliation? Sarkar calls them Vaishnavas; Kruijtzer, reading the forehead marks in the Witsen Album and other miniatures, argues for Smarta, with Akkanna wearing Shaiva marks and Madanna a subtly Vaishnava one. Akkanna's Shiva temple at Maheshwaram and the Mahankali tradition at Shah-Ali-Banda pull in different directions.
Was Akkanna ever really a commander-in-chief? Richards and Narayana Rao and the Telangana Endowments Department both say so; Kruijtzer accepts a formal appointment as sar-lashkar in 1682 but strips it of military content, quoting Havart that Akkanna could handle a sword as iron can swim. A separate strand holds that he was never a general at all but a tarafdar for the Karnatak from 1675.
Was Kancherla Gopanna, the temple-builder of Bhadrachalam, their nephew, their cousin, or their brother-in-law? Government and encyclopaedic sources say nephew; The Hindu says cousin or brother-in-law; and the Telugu literary sources are contradictory enough that one reconciliation makes Madanna and Akkanna brothers-in-law rather than brothers, a reading Kruijtzer rejects.
Did the brothers actually offer the kingdom to the Mughals? The peon Dauwelsu's report, Martin and Bhimsen say yes; the Dutch factory's own detailed account of the murder and Khafi Khan are silent, and the Mackenzie tract alleges a forged letter. No intercepted letter itself has been produced.
Did Abul Hasan order, know of, or merely fail to prevent the killing? Kruijtzer counts the sources split evenly, and the archive takes no side.
S. Krishnaswami Aiyangar's article 'Abul Hasan Qutub Shah and his Ministers, Madanna and Akkanna', Journal of Indian History (August 1931), is the classic treatment argued with throughout the later literature; no copy of the text could be opened for this entry.
Daniel Havart's Op- en ondergang van Cormandel (Amsterdam, 1693), the single most important contemporary European account, was read here only through Sarkar's summary and Kruijtzer's quotations, never in the Dutch text or a scan.
K. V. Bhupala Rao, The Illustrious Prime Minister Madanna (Hyderabad, 1984), the fullest Telugu-side biography, could not be obtained in any form.
J. F. Richards, Mughal Administration in Golconda (Oxford, 1975), pp. 40-43, the standard account of the Brahmin appointments, was available only as a lending copy that could not be opened; the same applies to Sherwani, History of the Qutb Shahi Dynasty (Delhi, 1974), pp. 626-630 and 655-656, cited by Kruijtzer and others for temple building and for the disputed Golconda budget document.
The claim that Aurangzeb wrote in the Fatawa-i Alamgiri that the brothers deserved beheading circulates in popular accounts in Persian transliteration; no edition or scholarly citation of such a passage could be found, and it is not used here.
The economic programme attributed to Madanna in popular history - reopening of the Gollapalli, Kollur and Ravalkonda diamond mines, a fixed wage of fifteen hons for miners, a nine-year tax holiday for resettled villages - could not be traced to any source opened for this entry.
The claim that Madanna and the sultan granted the village of Kuchipudi to its dancers in 1678, and the Mackenzie-derived claim that the brothers financed a touring Yakshagana troupe, both await a primary or scholarly citation.
The archaeological reports cited elsewhere for the death narrative and for a jizya department - Z. Jung's Annual Report of the Archaeological Department of His Exalted Highness the Nizam's Dominions (1936), pp. 6-7, and A. H. Longhurst, Memoirs of the Archaeological Survey of India No. 17 Part 1, pp. 24-25 - were not opened.
What is the present physical condition and protection status of the Maheshwaram Shiva temple site, and are the 'Akkanna-Madanna Offices' inside Golconda's Bala Hisar securely identified as theirs by survey evidence rather than by tradition?
Akkanna's two or three trading ships, attested in Dutch records of 1683 and 1685, are the most concrete evidence of the brothers as portfolio capitalists in their own right; whether the vessels, routes or cargoes can be traced further is unresolved.
A press claim that Gandikota fort is also known as 'Akkanna Serai' was seen only in a search snippet and could not be verified against the article itself.