How Bengal got out of Delhi
Bengal was conquered before it was independent. "Bengal was gradually absorbed into the Delhi Sultanate during the 1200s. It began with Bakhtiyar's 1202–1204 conquest of Gauda during the reign of Muhammad of Ghor" [2]; the city of Gauda itself "was conquered by Bakhtiyar Khalji, a lieutenant of the Ghurid ruler Muhammad of Ghori, in 1203" [3]. "Iltutmish declared Bengal as a province of Delhi in 1225" [2].
It never held. The reason is stated plainly and it is geographic: "The Delhi Sultans attempted to govern Bengal through appointed governors, however, Delhi could not succeed given the considerable overland distance with Bengal. Ambitious governors rebelled and ruled as independent rulers until being suppressed militarily" [2]. The same distance that let the Deccan break away under the Bahmanis in 1347 was breaking Bengal away a generation earlier.
Delhi's last administrative attempt was a partition: "In 1325, the Delhi Sultan Ghiyath al-Din Tughluq reorganized the province into three administrative regions, with Sonargaon ruling eastern Bengal, Gauda ruling northern Bengal, and Satgaon ruling southern Bengal. Even this arrangement broke down. By 1338, the three administrative regions had separatist sultans" [2].
One of the three won. "Shamsuddin Ilyas Shah (or just Ilyas Shah) defeated Alauddin Ali Shah and secured control of Gauda. He then defeated Ikhtiyaruddin of Sonargaon. By 1352, Ilyas Shah emerged victorious among the Bengali triad" [2], and "unified the deltas of the Ganges, Brahmaputra, and Meghna Rivers into the Sultanate of Bengal" [2].
Beating Delhi twice
The new sultanate's independence was won in two sieges of the same fort, six years apart, and it lasted two centuries because of them.
Ilyas Shah overreached first — he "waged wars and raids against several city-states and kingdoms in the eastern subcontinent", conquered eastern Bengal and northern Bihar, "led the first Muslim army into Nepal, raided the Kathmandu Valley, and returned to Bengal with treasures", and "controlled an area stretching from Assam in the east to Varanasi in the west" [2]. Delhi answered: "In 1353, Ilyas Shah was defeated by Delhi Sultan Firuz Shah Tughluq in the Siege of Ekdala Fort during the Ekdala Wars. Bengal agreed to pay a tribute" [2].
His son settled it. "His son and successor, Sikandar Shah, defeated Delhi Sultan Firuz Shah Tughluq during the second Siege of Ekdala Fort in 1359. A peace treaty was signed between Delhi and Bengal, with the former recognizing the independence of the latter" [2]. The terms included a gesture worth recording: "Firuz Shah Tughluq gave a golden crown estimated to be worth 80,000 taka to Sikandar Shah. The peace treaty ensured Bengal's independence for two centuries" [2].
A defeated emperor crowning the man who beat him is the clearest statement of what 1359 meant, and everything that follows in this essay happens inside the space it bought.
The largest mosque in the subcontinent
Sikandar Shah spent his victory on a building, and the building says exactly what the victory meant.
The Adina Mosque at Pandua "was the largest structure of its kind in the Indian subcontinent and was built during the Bengal Sultanate as a royal mosque by Sikandar Shah, who is also buried inside" [4]. Commissioned 1373, completed 1375 — and the record draws the connection itself: "The early Bengal Sultanate harbored imperial ambitions after having defeated the Delhi Sultanate twice in 1353 and 1359 during the Ekdala Wars. The Adina Mosque was commissioned in 1373" [4].
The scale is imperial in the literal sense. It measured "172 by 97 metres (564 by 318 ft)", with a prayer hall five aisles deep and triple aisles on the other three sides, "260 pillars and 387 domed bays", and a courtyard façade of "92 arches surmounted by a parapet" [4]. Its model was Umayyad: "The vast architecture resembles the hypostyle of the Umayyad Mosque in Damascus, a style that was used during the introduction of Islam in new areas" [4].
And its most ambitious feature was Persian: "The entire western wall evokes the imperial style of pre-Islamic Sassanian Persia. The mosque's most prominent feature is its monumental ribbed barrel vault over the central nave, the first such huge vault built in the subcontinent... The mosque consciously imitated Persianate imperial grandeur" [4]. A Bengali sultan who had just beaten Delhi built himself a Damascus-plan mosque with a Sasanian vault — the argument is architectural and it is about rank.
Its materials were local in a way that is recorded and should be stated without embellishment in either direction: "It was constructed from stones of dismantled Hindu temples and incorporates motifs inspired by pre-Islamic Hindu and Buddhist structures. Some exterior wall sections retain carvings, such as elephants and dancing figures, reflecting this reuse" [4]. The resulting ornament is genuinely hybrid — "The terracotta ornamentation of the Adina Mosque reflects a fusion of Islamic motifs with designs derived from Buddhist and Hindu sculptures of the Pāla–Sena era" [4]. The building is now "Abandoned; Preserved", a Monument of National Importance in Malda district, West Bengal [4].
Embassies, taka, and a trading nation
The Bengal Sultanate's self-presentation abroad was that of a maritime power, and the record supports it.
"The third Sultan, Ghiyasuddin Azam Shah, began expanding Bengal's influence abroad. He began to send embassies to Ming China, which continued as a tradition during the reigns of his successors" [2]. Later, Chinese fleets came the other way: "It is recorded in the Ming shi that a Chinese explorer, Cheng Ho, visited the city twice in 1421–22 and 1431–33" [6].
Its cities were mints. "The cities of the Bengal Sultanate are termed mint towns where the historical taka was minted" [2] — the currency Bangladesh still uses. Its network ran east and west at once: "The Bengal Sultanate was connected to states in Asia, Africa, the Indian Ocean, and Europe through maritime links and overland trade routes... Bengali ships and merchants traded across the region, including in Malacca, China, and the Maldives" [2].
Its vassals reached in every direction: "parts of Odisha in the southwest, parts of Bihar in the northwest, parts of Assam in the northeast, Arakan in the southeast, and Tripura in the east" [2]. And the verdict of visitors is quoted consistently: "Contemporary European and Chinese visitors described the Bengal Sultanate as a prosperous and thriving kingdom. Due to the abundance of goods in Bengal, the region was described as the 'richest country to trade with'" [2].
Persian was the language of state — "Persian was used as the primary official, diplomatic, court and commercial language" [2] — but the sultans did something the Bahmanids notably did not, and the archive's Dakhni essay makes the contrast worth drawing: they promoted the vernacular. The Bengal sultans "promoted the development of Bengali as a language and the writing of Islamic literature in Bengali, paving the way for the development of a distinct Bengali Muslim culture, while many intellectuals and scholars from throughout the Muslim world migrated to Bengal" [7]. In the Deccan the vernacular literature grew *outside* the court because the Bahmanids would not fund it; in Bengal the court funded it.
A throne that changed families five times
The sultanate "controlled large parts of eastern South Asia under its five unrelated dynasties" and was "a Sunni Muslim monarchy with Turco-Persian, Afghan, and Abyssinian elites" [2]. That last word is not decorative.
**The convert's son.** Jalaluddin Muhammad Shah was "born as Jadu", "Born a Hindu to his aristocratic father Raja Ganesha" [6]. The accession is a negotiated conversion: Raja Ganesha, "Facing an imminent threat of invasion at the behest of a powerful Muslim holy man named Nur Qutb Alam, he appealed to the saint to call off his threat. The saint agreed on the condition that Raja Ganesha's son Jadu would convert to Islam and rule in his place" [6].
Then it reversed, and reversed again. "Nur Qutb Alam died in 1416 AD and Raja Ganesha was emboldened to depose his son and accede to the throne himself as Danujamarddana Deva. Jalaluddin was reconverted to Hinduism by the Golden Cow ritual. After the death of his father he once again converted to Islam and started ruling his second phase" [6]. A sultan converted, deposed, ritually reconverted to Hinduism, and converted back — and the archive knows of no comparable sequence anywhere else in its corpus.
His own account of taking the throne the second time is quoted: "I believe in the truth of the religion of Muhammad, and I shall not deviate from this (belief). If you accept me and do not turn against my royal authority, then only I shall sit on this august throne; if not you shall raise my younger brother to the throne and excuse me" [6]. He ruled sixteen years, brought Arakan under Bengali suzerainty, dealt with the Timurids, Mamluk Egypt and Ming China, moved the capital from Pandua to Gauda, and "combined Bengali and Islamic architecture" [6].
**The Abyssinians.** Between 1487 and 1493 the throne was held by four Habshi rulers — "the brief era of Siddi (referred to at the time as Habesha) rulers in Bengal" [5]. The mechanism is the one this archive documents in the Deccan: "Before the Ilyas Shahi dynasty, it was customary for Muslim rulers to purchase slaves and appoint them to royal palaces or important positions in the kingdom. Even during the reign of Jalaluddin Fateh Shah, some slaves were appointed as palace guards, who gradually increased their power among themselves. Jalaluddin tried to rein them in. But the Habshis rebelled under the leadership of Shahzada Barbak and killed the Sultan and seized the throne" [5].
The six years were violent even by the standards of this essay: Barbak killed the sultan and was killed within months by Malik Andil Khan, who reigned two years as Saifuddin Firuz Shah and is "credited by many as the main founder of Habshi rule"; his adopted infant son Mahmud Shah II was enthroned under a regent, and after a year "Sidi Badr killed him and Habash Khan and took over the rule" as Shamsuddin Muzaffar Shah, who "raised an army of thirty thousand soldiers; Among them were thousands of Afghans and five thousand Habshi", defeated the Kamata kingdom, and in 1494 was killed in a revolt led by his own wazir, Sayyed Hussain [5].
That wazir founded the fifth dynasty — the Hussain Shahi, under whom the sultanate reached its cultural height and whose sultan Alauddin Husain Shah is the ruler named on the 1505 inscription at Shah Jalal's shrine in Sylhet [10].
The Deccan connection is exact. The archive's essay on Malik Ambar sets out how Ethiopians entered the Indian Ocean slave trade and were "re-exported to the Deccan plateau to meet that region's insatiable demand for military labor", and how "Several former slaves rose to high ranks in the military and administration" [12]. Bengal ran the same system and produced the same result faster and further: where the Deccan's most famous Habshi governed for twenty-six years without ever claiming a throne, Bengal's took the throne four times in six years. Richard Eaton is cited in both places: "Habshis were initially pagans sold by Ethiopian Christians to Gujarati merchants for Indian textiles" [5].
Gaur, and a number the archive is correcting
The sultanate's capital moved — Pandua first, then Gauda, with Sonargaon as an eastern seat and a royal court under Ghiyasuddin Azam Shah [2] [9]. "For a period of 112 years, between 1453 and 1565, Gauda was the capital of the Bengal Sultanate" [3].
The city's size is where the archive has to correct itself. The Bengal Sultanate record states that "In 1500, the royal capital of Gaur was the fifth-most populous city in the world" [2], and this archive's own essay on Dhaka muslin repeated it. The city's own record puts a better-attributed figure directly against it: "Richard M. Eaton writes that the city had 40,000 population in 1515 AD. **According to other estimates**, in 1500, Gauda was the fifth-most populous city in the world, with a population of 200,000" [3].
Forty thousand and two hundred thousand cannot both be right about the same city fifteen years apart, and one of them carries a named historian whom this archive cites throughout its Deccan essays. The archive now prefers **Eaton's 40,000** and treats the fifth-largest-city claim as an unattributed estimate. The muslin essay's use of it is flagged; a claim that a delta capital was the fifth-largest city on earth needs better than "other estimates".
What is not in doubt is the building. "The sultans built a citadel, many mosques, a royal palace, canals, and bridges. Buildings featured glazed tiles", and "Most of the surviving structures in Gauda are from the period of the Bengal Sultanate" [3]. The Portuguese "left detailed accounts of the city" [3].
A second city survives better and is protected. Bagerhat — "formerly known as Khalifatabad, was founded by the Turkish general Ulugh Khan Jahan in the 15th century" — is a UNESCO World Heritage Site whose "infrastructure reveals considerable technical skill and an exceptional number of mosques and early Islamic monuments, many built of brick" [1]. It is the sultanate's best-preserved urban statement, and it was built by a general, not a sultan.
How it ended
"Its decline began with an interregnum by the Sur Empire, followed by Mughal conquest and disintegration into petty kingdoms" [2]. Gauda "was sacked by Sher Shah Suri", and when Humayun took the region he "renamed the city Jannatabad ('heavenly city')" [3].
Two things finished the capital that had nothing to do with armies: "An outbreak of the plague contributed to the city's downfall. The course of the Ganges was once located near the city, but a change in the river's course caused Gauda to lose its strategic importance" [3]. A delta capital is hostage to its river, and the river moved.
The Mughal province that replaced the sultanate was "established following the dissolution of the Bengal Sultanate, a major trading nation in the world, when the region was absorbed into the Mughal Empire" [8], with its capital eventually at Dhaka — "the capital of a proto-industrialized Mughal Bengal for 75 years (1608–39 and 1660–1704)" [13]. Aurangzeb would call Bengal "the paradise of Nations" [8], and the muslin trade the archive treats separately reached its height under that province, not under the sultanate.
The last word on what the sultanate was belongs to its own record, and it is a claim about rank rather than size: it "was also the largest and most prestigious authority among the independent medieval Muslim-ruled states in the history of Bengal", and "The Bengal Sultanate left a strong architectural legacy. Buildings from the period show foreign influences merged into a distinct Bengali style" [2].
Dates in this essay
The archive dates the Bengal Sultanate 1352–1576 — from Ilyas Shah's victory over the Bengali triad to the Mughal conquest. The region's Muslim rule is older (Bakhtiyar 1202–04, Delhi province 1225, separatist sultans by 1338) and the archive treats those as the Delhi-provincial phase, not the sultanate.
Gaur's population is contested by a factor of five. Richard Eaton gives 40,000 in 1515; unattributed "other estimates" give 200,000 in 1500 and the fifth-largest-city-in-the-world claim. The archive now prefers Eaton's figure, and the dhaka-muslin essay's repetition of the larger claim is flagged there for correction. A named historian beats an unattributed estimate.
The Adina Mosque is dated two ways within one source: "commissioned in 1373" and "Completed in 1375" in the summary, against "Built between 1364 and 1374 CE" in the section on its terracottas. The archive uses 1373–1375, which fits the stated political motive (the Ekdala victories of 1353 and 1359).
The Habshi dynasty is dated 1487–1493 in the record, but its last sultan is said to have been killed in a revolt in 1494. One of the two is wrong; the archive uses 1487–1493 for the dynasty and notes that the terminal date does not reconcile with the account of its end.
Jalaluddin Muhammad Shah's reign is recorded in two phases — 1415–1416 and 1420–1433 — with his father Raja Ganesha ruling between. The narrative says his second phase began after his father's death and elsewhere dates it 1418. The archive uses 1415–1416 and 1420–1433 as the record's own regnal table gives them.
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
**The forced-conversion claim needs proper handling and this essay does not settle it.** The account of Jalaluddin Muhammad Shah quotes James Wise, writing in the Journal of the Asiatic Society of Bengal in **1894**, that "the only condition he offered was either the Koran or death". That is a single colonial-era article, and it sits in the same source as a description of a peaceful sixteen-year reign with embassies to three empires. The archive reports its existence and provenance and asserts nothing: a nineteenth-century British claim of mass forced conversion is exactly the kind of statement its method requires be traced to its earliest attestation before use. It is also directly contradicted, as a general explanation, by the agrarian account of Bengal's Islamisation that the archive's Shah Jalal essay sets out.
No Bengali or Persian source was read for this essay, and none of the sultanate's own chronicles, inscriptions or coins. For a state defined by its mint towns, an essay that never examines a taka is thin.
The Adina Mosque's temple spolia is stated as fact by the source and not evidenced there — no excavation report, no inscription, no inventory of reused pieces is cited. The claim is architecturally commonplace and probably right, but the archive should establish which survey records it before treating the detail as settled.
The five dynasties are named and only two are described here. The Ilyas Shahi (150 years) and the Hussain Shahi (the cultural peak, and the dynasty of the 1505 Sylhet inscription) both deserve their own treatment; the Hussain Shahis in particular are credited with the sultanate's height and appear in this essay only as the family that succeeded the Habshis.
Ulugh Khan Jahan built a UNESCO World Heritage city and is one clause in this essay. Who he was, what his relationship to the sultans was, and why a general founded an urban complex of that scale at Bagerhat are unexamined.
Bengal's relationship with Arakan — brought under Bengali suzerainty by Jalaluddin, and later the destination of Bengali Muslim migration — is named twice and never described. It is the sultanate's most consequential external relationship for the modern history of the region.
This is the archive's third Bengal essay and Bengal still has no collection: the region node holds two directly-tagged essays against a threshold of three. That is the progress instrument working as designed, and the next cross-branch Bengal essay will light it up.