An accession, and the man who actually ran the state
The reign begins in dispute. Ferishta, writing within the sultan's lifetime, says Ibrahim Quli Qutb Shah left three sons and that the eldest, Muhammad Quli, took the throne at twelve. H. K. Sherwani, working from the dynasty's own Persian chronicle, rejects both halves: a third son with elder brothers living, raised up in what looks like a coup engineered while a rival's prospective father-in-law was away at the siege of Naldrug. Modern reference works follow Sherwani on birth order while dating the reign 5 June 1580 to 11 January 1612 [22]. Ferishta's arithmetic will not close: he has the sultan dying 'in the 49th year of his age, after reigning nearly thirty-four years' [7].
Whoever engineered the accession, the next four decades of administration had one continuous author. Mir Muhammad Mu'min Astarabadi, an Iranian emigre who had tutored a Safavid prince, 'was appointed Peshwa of the Qutb Shahi Kingdom as early as 1581' and 'held this position for forty years without a break' [25]. Iranica identifies him as the courtier who praised the new foundation as a new Isfahan [8], and his correspondence with Shah Abbas states an intention to propagate the doctrine of the Ahl-e-Bayt in the new city [18]. That direction was set from the first year, when a half-fulus of 989/1581 carried the standard of Imam Husain [4]; the first purely religious building in the new capital was not a mosque but the royal Ashurkhana, begun 1593 [8].
Why the capital moved: water, crowding, and a road to the sea
The founding of Hyderabad was an infrastructure decision before it was a romance. Iranica states the mechanism plainly: Golconda's population had swelled, and overcrowding with lack of water pushed the sultan and his advisors to conceive a city able to hold the court, the bureaucracy and the overflow from the hill fort - four miles east, on the south bank of the Musi, on the highway running to the port of Masulipatam [8]. Ferishta, much closer in time, gives the same reasons. The direction had been fixed already by the previous reign, whose bridge over the Musi was finished in 986/1578 after westward expansion failed on rocky ground [2].
The site sat astride the artery that made the kingdom rich: Golconda lay on the route from Masulipatam inland, and prospered on diamonds, textiles and printed cloth [10]. The cost is unusually well documented, because Ferishta had access to the king's private treasurer: irrigation raised land rents alone by four lakh huns a year, seventy-eight lakh huns went on public works across the reign, and sixty thousand huns were given away annually in alms [6]. Ferishta closes his notice with the resulting list - the Ilahy Mahal palace and gardens, the Mahomedy gardens, the Nubat-Ghat and Nuddy Mahal palaces, an alms-house [7]. This was a decade-long capital programme, not a gesture.
The city as a designed object, and the arch at its centre
What went up was a plan. Sherwani describes a gridiron laid as a giant double cross, the Golconda-Masulipatam road made to intersect a new north-south road at precisely the point where the Charminar stands, with a ceremonial forecourt eighty yards north behind four arches fifty feet high [2]. Iranica supplies the commercial fill: once the layout was complete the sultan ordered 14,000 shops as well as mosques, schools and caravansaries along the roads radiating from the two squares [8].
The Charminar is that plan made visible - 'Constructed of plaster and stone in the years 1590-91', a triumphal arch of sorts, its 36-foot archways on four 60-foot sides framing the crossroads of the city's major thoroughfares [9], the whole rotated some ten degrees off the cardinal points toward Mecca. Ferishta, the earliest Deccan chronicler to describe it, is functional: the king 'built an elegant musjid and the chahar minar in the middle of the city', quadrangular, four arcades, a dome over the centre, a fountain of excellent water beneath it, and small apartments in each minaret for teachers and students [6]. Gateway, madrasa, mosque and waterworks at once.
The dating is unresolved, and UNESCO states why without evasion: the Charminar carries no foundation inscription, and its date of 1591 rests on contemporary Persian historical texts rather than on the fabric [10]. The dynasty's own chronicle gives 1000 AH / 1592 for the foundation of the city in the same year the monument supposedly went up, and the objection that a building of this scale could not have been finished in a little over a year comes from the historian reporting that date [18]. Popular accounts push the start to 1589, at a cost of nine lakh rupees [23], on no chronicle at all.
The Kulliyat: what 'first' actually means
The claim usually made - first poet of Urdu - is looser than the evidence supports. Sherwani states it narrowly: the sultan 'was the first author of a dkvan in the language not merely in Tilang, but in the whole of south India' [2]. The operative word is divan, an alphabetically ordered collected book on the Persian model; Sherwani is explicit that Dakhni verse predates his subject. Matthews frames the same fact from the literary side - he 'preferred the Persian-style ÷azal, and his dµv≥n is one of the first substantial collections of secular Urdu poetry' [11]. Public summaries compress both into 'First Urdu poet to have compiled and published a Divan' [21]. The precise title is sahib-e-diwan.
That status is a modern discovery with a date. Maulvi Abdul Haq 'pushed the history further back by publishing a paper in 1922 on a manuscript of Quli Qutb Shah's divan, composed entirely in Urdu, consisting of some 50,000 couplets' [16], displacing Vali Deccani, whom northern criticism had installed as the Adam of Urdu poetry. The find was then lost: 'This priceless manuscript, which bore the signature of Muhammad Quli's successor, was discovered in the Asifiyya (now State Central) Library of Hyderabad, from which it promptly disappeared before it could be edited' [12]. Zor's printed edition of 1940 rests on incomplete Salar Jung manuscripts instead - so the circulating book and the book Abdul Haq described are not the same object.
What survives is still extraordinary in range. Sherwani insists on authorship rather than patronage: the sultan was 'a poet of no mean merit in that language, and the fifty thousand odd lines, which he is said to have composed, are remark- able for the ever-changing vista of life they represent' [3] - lines, where the press says couplets. He wrote odes, dirges, quatrains, masnavis and ghazals, on the rains, gardens, fruit and vegetables, down to coriander and asafoetida. Casadei confirms the shape: mainly ghazals, but concerning festivals religious and secular, the monsoon, the building of Hyderabad and the king's own life [14] - which makes the Kulliyat a historical source, not only a literary one.
The woman's voice, and the argument about the language
One formal feature carries far beyond the Deccan. A substantial body of these ghazals is narrated by a woman speaking of her lover, frequently the king himself - a convention predating northern rekhti by roughly a century. Carla Petievich reports the specialist judgement directly - “Sayeda Ja‘far has called Muhammad Quli’s poetry ‘the first in which lines of r®xtµ appeared in Dakani.’” [13] - and derives the female narrator from the virahini of Indic lyric rather than from anything Persian. Matthews notes the same convention [11]. The popular restatement, that the practice 'was pioneered in the Deccan, by Muhammad Quli Qutb Shah' [24], flattens a century of transmission, but the priority itself is not seriously contested.
What the language proves is contested. On one reading it is a genuinely mixed idiom: Sherwani finds it interspersed with Persian, Arabic and Hindi throughout, so much so that Hindi literary historians claim the sultan as readily as Urdu ones; Rekhta makes him a poet drawing on Hindu as well as Persian culture toward a secular view of love and life [21]. Casadei argues the reverse from the vocabulary - 'The choice of vocabulary and images are conventional, and one finds an almost total absence of Dravidian loanwords' [15]. Matthews sits between them: the poems show the increasing influence of Persian, and the metres are impeccably Persian.
Some of that disagreement is factual. Casadei also holds that before 1580 'there is no surviving evidence of works having been composed in Dakhni' in Golconda [14] - which Sherwani contradicts by naming Mahmud, Ahmad, Mulla Khiyali and a recovered manuscript of Firoz's Paratnamah. A practical obstacle keeps the corpus marginal on both sides: the diction carries words drawn from Telugu, Gujarati and Marathi in localised spelling, so that couplets can appear to scan wrongly to a reader who does not know the register - an error reported to have misled scholars as well as students [16].
The divan as a physical book
There is an object at the centre of this, and it is not the printed edition. Laura Weinstein's study of the earliest illustrated Urdu divan is built on one codex - 'Diwan of Muhammad Quli Qutb Shah, 1590-1605, folio 3v, watercolour, gold and ink on paper. Salar Jang Museum, Hyderabad, Ms. 153' [26] - a divan rather than the full kulliyat, 138 folios carrying eight paintings. Her most useful comparison is not Indian: the closest analogue is the Safavid divan of Shah Isma'il, a ruler's poems in Turcoman Turkish dressed in Persian book-arts.
The Metropolitan Museum makes the same manuscript the exemplar of Golconda's mixed workshop: 'The close working relationship of the Persian and Indian artists can be seen in the unique painting style of such manuscripts as the Kulliyat, a collection of verses in Urdu penned by Muhammad Quli Qutb Shah (r. 1580-1612), a contemporary of Ibrahim 'Adil Shah and, like him, a poet, statesman, and important patron of the arts' [17]. The pairing with Bijapur's Ibrahim Adil Shah II is worth holding on to: two neighbouring sultans, one generation, both writing in a vernacular and commissioning Persian high culture to carry it.
Bhagmati: a transmission question, not a love story
The story attached to the founding - that the sultan named his city Bhagnagar after a courtesan called Bhagmati and renamed it Hyderabad after her death - is repeated everywhere, and its base is narrow and traceable. Ferishta says the king 'laid the found- ation of a new city, which was called Bhagnuggur (after his favourite mistress Bhagmutty); but after her death he ordered it to be called Hydrabad' [5]. Its earliest appearance is not a chronicle but a hostile petition attributed to Akbar's agent Faizi (999-1002/1591-94), whose whole notice of the sultan is a jibe. Ferishta is the only Deccan writer to carry it - 'The only contemporary chronicler who wrote in the Deccan and who' mentions Bhagmati and Bhagnagar is Ferishta' [1] - and Nihawandi and Khafi Khan both declare their dependence on him.
The counter-evidence is silence where one expects noise. The dynasty's own Tarikh Muhammad Qutb Shah (1017/1609) describes the new capital in detail and never mentions Bhagmati or Bhagnagar; the court histories of 1644 and 1681 are equally silent; the Kulliyat names seventeen mistresses with odes to each and Bhagmati is not among them, while the city is named three times, never as Bhagnagar. The mint record is hardest: 'While there are coins in the Plaidarabad Museum which were struck at the ‘Daru’s-Sultanat, Haidarabad’ in 1012/1603 there is not one com which was struck at ‘Bhagnagar’' [4]. Press accounts put the historiography bluntly - Sherwani 'had always refuted the existence of Bhagmati in his works' [19]; there is no historical evidence she existed [20].
The evidence will not bear the reverse claim either. An absent mint name does not disprove a popular one, and outsiders demonstrably called the place Baghnagar, the city of gardens - the etymology several historians prefer and seventeenth-century travellers support. Zor, who edited the Kulliyat, identified Bhagmati with a 'Haidar Mahal' named in the poems while conceding that the sultan never names a dancer. What can be stated is narrower than either camp's headline: the name in the mint, the chancery and the sultan's own verse is Haidarabad; the story enters the record from outside the Deccan, in hostile hands, within a few years of the founding, and reaches every later retelling through one chronicler.
Dates in this essay
Founding of Hyderabad: Iranica (Hutton) dates it 999/1590-91 and Sherwani says the plans were ready in 999/1591; the dynasty's own Tarikh-i Muhammad Qutb Shah, as reported by Salma Ahmed Farooqui, gives 1000 AH / 1592 for the laying of the foundation; the Hada'iqu's-Salatin (1092/1681) gives 1004/1595, which Sherwani reads as the completion of the building scheme rather than the start. This archive uses 1591 for the foundation of the new capital while recording the 1592 chronicle date.
Charminar: no foundation inscription survives, as UNESCO concedes; UNESCO gives 1591 (1000 AH), Iranica 'the years 1590-91', Sherwani completion in 1000/1592, and English Wikipedia an unsourced start of 1589. Farooqui raises the physical objection that a building of this scale and stucco could not plausibly have been finished in little over a year.
Reign span: this archive dates the reign 988/1580 to 1020/1611-12, from the accession at Golconda to the sultan's death. Sherwani places the accession at 21-4-988 / 5-6-1580; Iranica gives 988-1020/1580-1611; Wikipedia and Shaikh Musak Rajjak give 11 January 1612 for the death, Rekhta 11 January 1611, and Dawn simply 1611.
Death and age: Ferishta (Briggs) gives Saturday 17 Zi-qada 1020, glossed in the margin as 17 December 1611, and says the sultan died in his 49th year after reigning nearly thirty-four years - arithmetic inconsistent with his own accession date. Sherwani, following the Qutb Shahi chronicle, dates the birth to 14-9-973 / 4 April 1566; Wikipedia gives 4 April 1565.
Age and rank at accession: Ferishta calls him the eldest of three sons, succeeding at twelve; Sherwani, citing the Qutb Shahi chronicle, makes him the third son with two elder brothers living; Wikipedia says third son and age fifteen. The metrical Tawarikh-i-Qutbshahi sides with Ferishta.
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
Neither Zor's 1940 Kulliyat-e Sultan Mohammad Quli Qutub Shah (Maktaba Ibrahimiya, Hyderabad) nor Sayyida Jafar's later edition (Taraqqi-e Urdu Bureau, New Delhi) could be opened in full during this research, so no Dakhni couplet here is quoted from a printed edition - every verse in the evidence base reaches us at second hand through Petievich, Matthews, or reference pages. A pass working directly from a printed Kulliyat would let the poems be quoted as poems rather than described.
How large is the Kulliyat, and in what units? Sherwani writes of 'fifty thousand odd lines'; Dawn and Rekhta say roughly fifty thousand couplets; Zor is cited elsewhere for 'nearly 50,000 verses'. Since the lost Asifiyya manuscript and the incomplete Salar Jung manuscripts that Zor edited are demonstrably different objects, the figures may not even be describing the same book.
What exact figure did Abdul Haq give in 1922 for the number of verses in the Asifiyya manuscript? The digits are missing from the extractable text of Matthews's PDF, and Dawn's '50,000 couplets' is a later restatement rather than a quotation of Abdul Haq's paper.
Where is the Asifiyya manuscript that Abdul Haq described, bearing the signature of Muhammad Quli's successor? It disappeared from the State Central Library before it could be edited, and no account of its fate was found.
Did the Asifiyya manuscript really contain Telugu verse alongside Urdu and Persian, as Matthews reports it was said to? Matthews himself doubts it, noting there is no evidence the king knew Telugu - which bears directly on the wider disagreement about Telugu patronage under this reign.
The verbatim texts of Faizi's petition to Akbar, Nizamu'd-din's Tabaqat-i Akbar Shahi and the Tarikh-i Muhammad Qutb Shah could not be opened; all three are quoted here only as Sherwani translates them in Appendix II. The Elliot and Dowson volume carrying Pritchard's translation of the Faizi letters was not opened either, so the earliest witness to the Bhagmati story has not been read in any edition.
Ferishta was read only in John Briggs's 1829 English translation, whose OCR is imperfect and whose marginal date glosses contain at least one clear error (AH 998 rendered 'A.D. 1598'). The Lucknow 1884 Persian edition that Sherwani cites was not opened, so no Ferishta date used here has been checked against the Persian.
No coin was examined directly. The numismatic case against 'Bhagnagar' rests entirely on Sherwani's citations of Hormuz Kaus (Spink's Numismatic Circular, May 1955) and Md. Abdul Wali Khan's catalogue of Qutb Shahi coins in the Andhra Pradesh Government Museum; neither catalogue was opened, and no museum accession record was consulted.
Laura Weinstein's chapter 'Variations on a Persian Theme' could be read only as an Academia.edu preview - abstract, figure captions and references. The full chapter in The Visual World of Muslim India (2014) was not accessible, so the account of Salar Jung Ms. 153 here rests on captions and on the Metropolitan Museum's survey rather than on her argument.
Is Casadei right that no Dakhni work composed in Golconda survives from before 1580? Sherwani names Mahmud, Ahmad, Mulla Khiyali and a recovered manuscript of Firoz's Paratnamah against exactly that claim, but none of those texts or their editions was opened here.
Marika Sardar's NYU dissertation 'Golkonda through time: A mirror of the evolving Deccan' (2007) would be the strongest single source on the Golconda-Hyderabad transition, but was not located in open access.
Sherwani's fuller History of the Qutb Shahi Dynasty (1974) and his article 'Town Planning and Architecture of Haidarabad under the Qutb Shahis' (Islamic Culture 50/2, 1976) are the standard references Iranica leans on; neither was opened, so the plan of the city here rests on the 1967 monograph and on Iranica's summary.
Ali Akbar Husain's Scent in the Islamic Garden: A Study of Deccani Urdu Literary Sources (Karachi, 2000) is the source behind Iranica's readings of the garden poems; it was not opened, so the garden material rests on paraphrase.
Serish Nanisetti's Golconda/Bagnagar/Hyderabad and Moupia Basu's Bhagmati: Why Hyderabad's Lost Queen Is the Soul of the City (Westland, 2025) are the two most recent interventions on the naming question, and both are known here only through press reports rather than from the books themselves.
The Archaeological Survey of India's own Charminar page timed out during this research, so the ASI's stated plague-eradication rationale for the monument is available only at second hand.
The English Wikipedia article on Muhammad Quli Qutb Shah currently contains a corrupted passage in its 'Patronage of art and literature' section; nothing from it was used here, and the article has been cited only for infobox dates and its plainly sourced sentences.