Seleucus Gave Chandragupta the Lands Along the Indus for 500 Elephants. Four Greek and Roman Writers Recorded the Deal, and It Took Until 1793 to Say Who the King Was

On 28 February 1793, Sir William Jones, president of the Asiatic Society in Bengal, delivered the society’s tenth annual discourse. Near the end he turned to “a discovery, which accident threw in my way.” Greek writers had described an Indian king they called Sandracottus, who ruled from a city they called Palibothra on a river they called Erannoboas, and who made a treaty with Seleucus Nicator. No one had settled which Indian names they stood for. Jones had found, in a Sanskrit book he called nearly two thousand years old, that the river Hiranyabahu was another name for the Sona, which the Greeks had turned into Erannoboas.

From the river he went to the king. Chandragupta had risen “from a military adventurer” to be sovereign of upper Hindustan and had set his capital at Pataliputra, Jones said, and “was no other than that very Sandracottus, who concluded a treaty with Seleucus Nicator.”

J. W. McCrindle, who translated the fragments of Megasthenes in 1877, wrote that the match linked Greek literature to Sanskrit literature and so supplied “for the first time a date to early Indian history, which had not a single chronological landmark of its own.” His grandson Ashoka has a tuput piece of his own. This one stays with the founder and with the sources that tell his story, nearly all of them written by outsiders or by later hands.

Five hundred elephants, four accounts

Strabo describes lands along the Indus that Indians held although they once belonged to the Persians. In H. L. Jones’s Loeb translation, Alexander “took these away from the Arians and established settlements of his own, but Seleucus Nicator gave them to Sandrocottus, upon terms of intermarriage and of receiving in exchange five hundred elephants.”

Appian, in M. M. Austin’s translation on Livius.org, says Seleucus “crossed the Indus and made war on Sandracottus, king of the Indians about that river, and eventually arranged friendship and a marriage alliance with him.” An editorial note on that page dates the peace to 304 BCE and says Seleucus “gave up all lands east of the Hindu Kush and received many war elephants.”

Plutarch, in Bernadotte Perrin’s translation of the Life of Alexander, says Androcottus “made a present to Seleucus of five hundred elephants.” Justin, abridging the lost history of Pompeius Trogus in J. S. Watson’s translation, says Seleucus, “after making a league with him, and settling his affairs in the east, proceeded to join in the war against Antigonus.”

The four disagree about what kind of deal it was. Strabo has an exchange with a marriage tie, Appian a friendship and marriage alliance, Plutarch a gift and Justin a league. None preserves the text of a treaty, and nothing in them settles whether Strabo’s “exchange” or Plutarch’s “present” is closer to the terms.

The war Seleucus went to join ended at Ipsus in 301 BCE, by the chronology on Livius.org. In Perrin’s Life of Demetrius, Plutarch gives Antigonus seventy-five elephants and his opponents four hundred, and says the enemy’s elephants cut Demetrius off from his own infantry. He does not say where they came from.

A stripling who saw Alexander

In September 326 BCE, according to a note in Perrin’s translation, Alexander’s army reached the river Hyphasis, the modern Beas in the Punjab, and refused to go on. Plutarch says the soldiers had learned that the kings of the Ganderites and Praesii were waiting beyond the Ganges with “eighty thousand horsemen, two hundred thousand footmen, eight thousand chariots, and six thousand fighting elephants.” Diodorus Siculus, in C. Bradford Welles’s Loeb translation, gives the same king, whom he calls Xandrames, twenty thousand cavalry, two hundred thousand infantry, two thousand chariots and four thousand elephants. The two writers agree only on the infantry.

Diodorus says the army had spent “almost eight years among toils and dangers,” and that Alexander’s prepared speech failed: “the Macedonians did not accept it, and he gave up the undertaking.” Before that, he says, Alexander asked Porus, the Indian king he had defeated, whether the reports were true. Porus confirmed the rest, “but that the king of the Gandaridae was an utterly common and undistinguished character, and was supposed to be the son of a barber.”

Plutarch then adds a line about the young Chandragupta: “Androcottus, when he was a stripling, saw Alexander himself, and we are told that he often said in later times that Alexander narrowly missed making himself master of the country, since its king was hated and despised on account of his baseness and low birth.” The words “we are told” mark it as hearsay.

Justin’s account is hostile. In Watson’s translation, India after Alexander’s death had “shaken, as it were, the yoke of servitude from its neck, and put his governors to death,” and Sandrocottus was “the author of this liberation” who afterwards “turned their semblance of liberty into slavery.” Justin says the young man offended Alexander “by his boldness of speech” and escaped an order to kill him “by swiftness of foot.” A lion licked the sweat off him as he slept, and, prompted by that omen, he “drew together a band of robbers, and solicited the Indians to support his new sovereignty.” A wild elephant later took him on its back. The lion and the elephant are portents, and none of the other accounts used here repeats them.

The last Nanda king and the dates

The Mahavamsa, a Pali chronicle from Sri Lanka attributed to Mahanama in the fifth century CE, tells it in a few lines. In Wilhelm Geiger’s translation of 1912, “the brahman Canakka” anointed “a glorious youth, known by the name Candagutta, as king over all Jambudipa, born of a noble clan, the Moriyas, when, filled with bitter hate, he had slain the ninth (Nanda) Dhanananda.” Jambudipa is the chronicle’s name for the Indian land. “Twenty-four years he reigned,” it continues, “and his son Bindusara reigned twenty-eight.”

Radhakumud Mookerji, in his 1943 study Chandragupta Maurya and His Times, reports that F. W. Thomas read the Agrammes of the Roman historian Curtius, and the Xandrames of Diodorus, as the last Nanda king, Dhana Nanda. A low birth for the Nandas appears in the Brahminical and Jain traditions too, Mookerji says, though Buddhist tradition does not give them one.

Mookerji put Chandragupta’s accession at 322 BCE. He reasoned that the generals who divided Alexander’s empire at Triparadisus in 321 left India east of the Indus out of the partition, which to him meant Chandragupta had already made himself ruler of the Punjab and Magadha. The EBSCO reference entry on Kautilya gives his reign as about 321 to about 298 BCE. The end date follows from the 24 years, a figure Geiger notes the Sri Lankan and Indian traditions share.

The detailed stories of Chanakya outwitting the Nandas come from later literature. The best known is the Sanskrit play Mudrarakshasa by Vishakhadatta, whom the EBSCO entry places in the fourth century CE or later. Richard Davis, reviewing Patrick Olivelle’s translation of the Arthashastra, calls him a Gupta-period playwright. That puts the play more than six centuries after Chandragupta.

The ambassador at Pataliputra

Seleucus sent an ambassador to Chandragupta’s court. In McCrindle’s translation, Arrian describes Megasthenes as a man who “lived with Siburtios the satrap of Arachosia, and who, as he himself tells us, often visited Sandrakottos the king of the Indians.”

Megasthenes’s book on India, the Indica, is lost. McCrindle says it was “so often abridged and quoted by the ancient writers” that its contents are fairly well known. Strabo, quoting it in McCrindle’s translation, describes Palibothra as “a city eighty stadia in length and fifteen in breadth,” “girded with a wooden wall, pierced with loopholes for the discharge of arrows,” with a ditch in front “for defence and for receiving the sewage of the city.” McCrindle’s footnote takes the stadium at about 202 yards, which makes the city roughly nine miles long and under two wide. Arrian, also quoting him, puts the ditch at six hundred feet wide and thirty cubits deep, and says the wall “was crowned with 570 towers and had four-and-sixty gates.”

The same fragment of Strabo describes how the state ran, as Megasthenes saw it. Officers measured the land, inspected the sluices that let water from the main canals into their branches, collected taxes, and built roads, setting up “at every ten stadia” a pillar “to show the by-roads and distances.” The city itself was governed by six bodies of five men each. The second attended to “the entertainment of foreigners,” assigning them lodgings, escorting them out of the country, and, if one died, sending his property to his relatives. The third recorded births and deaths. The sixth collected a tenth of the price of goods sold, and “fraud in the payment of this tax is punished with death.” The army had a board of the same shape, six divisions of five, one each for the fleet, the supply carts, the infantry, the cavalry, the chariots and the elephants.

Where the ambassador can be checked

The wooden wall is one place where the text meets the ground. McCrindle printed a report from a friend who inspected a dig in the Sheikh Mithia Ghari district of Patna in the cold season of 1876. Workmen digging a tank, twelve or fifteen feet below the swampy surface, found part of a long brick wall and, near it and running parallel, a line of palisades made of strong timber. McCrindle took the find as the wooden wall Megasthenes described, and as proof that the ancient city lay deep under modern Patna.

Other parts of the Indica cannot be tested that way. The fragments McCrindle collected include a section on gold-digging ants, and a division of the Indian people into seven classes. McCrindle quotes a note by the historian J. T. Wheeler that Herodotus had divided the Egyptians into seven castes, “and Megasthenes may therefore have taken it for granted that there were seven castes in India.”

Six hundred thousand men and a governor at Girnar

Plutarch says Androcottus “with an army of six hundred thousand men overran and subdued all India.” Pliny the Elder’s list of Indian peoples, in McCrindle’s collection, says the king of the Prasii “has in his pay a standing army of 600,000 foot-soldiers, 30,000 cavalry, and 9000 elephants.” That is three times the 200,000 infantry that Diodorus and Plutarch gave the Nanda king. Both figures are reports from outside India. We found nothing from India that confirms either, and we cannot tell whether the two writers are independent of each other.

The evidence from inside India is an inscription. F. Kielhorn published the Junagadh rock inscription of the Mahakshatrapa, or great satrap, Rudradaman in Epigraphia Indica and dated it to about 151 or 152 CE. It records the repair of the Sudarsana lake at Girinagara, the old Junagadh, after a storm and floods breached its dam and “all the water escaped” through a breach four hundred and twenty cubits long and seventy-five deep. Its one sentence on the lake’s earlier history, in Kielhorn’s translation, says it was “ordered to be made by the Vaisya Pushyagupta, the provincial governor of the Maurya king Chandragupta.” Kielhorn adds that the lake was improved under Ashoka by a Yavana (Greek) king, Tushaspha, who governed the district for him. The text is some four and a half centuries younger than the reign it describes, and it puts a Mauryan governor in Gujarat.

How far south the empire went is harder to say. Ashoka’s Rock Edict 2, in S. Dhammika’s translation, lists “the people beyond the borders, the Cholas, the Pandyas, the Satiyaputras, the Keralaputras, as far as Tamraparni and where the Greek king Antiochus rules.” In Rock Edict 13 Ashoka says conquest by dhamma has been won “even six hundred yojanas away, where the Greek king Antiochus rules,” and the editorial note identifies him as Antiochus II Theos of the Seleucid empire.

A textbook of statecraft with a disputed date

The Arthashastra is a manual of kingship and government attributed to Kautilya, whom later literature identifies with Chanakya. Davis writes that the work was widely cited until the early eighth century CE, then fell out of use from the ninth, and “resurfaced in about 1905,” when a single manuscript turned up in Tanjore and Rudrapatna Shamasastry published it from the Mysore Government Oriental Library.

In the modernised text of Shamasastry’s second edition (1923) on the Hanover College site, the king’s duty reads: “In the happiness of his subjects lies his happiness; in their welfare his welfare; whatever pleases himself he shall not consider as good, but whatever pleases his subjects he shall consider as good.” It warns that a king who “makes himself inaccessible to his people” will “engender confusion in business, and to cause thereby public disaffection.” It also grades punishments by caste: making a Brahman eat forbidden food earns “the highest amercement,” a tier of fine, and for a Shudra “a fine of 54 panas.”

Whether any of that describes Chandragupta’s government is the dispute. Mookerji, writing in 1943, quotes F. W. Thomas that the date of the work “clearly falls within or near the Maurya period,” a view he says Vincent Smith, Hermann Jacobi and K. P. Jayaswal had held, and he follows it. Davis says Kangle held to that position, but that the identification of Kautilya with Chanakya “has been substantially disproven by subsequent scholarship.” Olivelle, as Davis summarises him, finds three phases: an early expert text of the mid first century BCE, a Kautilya recension of 50 to 125 CE, and a Shastric recension of 175 to 300 CE, which is the version that survives. Mark McClish, in his own summary of his Cambridge book of 2019, argues “that the extant text is a 3rd century CE redaction of a treatise written a few centuries earlier.”

Olivelle goes on to propose, Davis says, that Chanakya was a real person and that the link between him and the author Kautilya was made only in Gupta times, as one of the ways the Guptas presented themselves as heirs of the Mauryas. The Guptas ruled about seven centuries after Chandragupta Maurya. Their Chandragupta II raised the iron pillar of Delhi, and the Gupta golden age has its own tuput piece. None of the rules above is attributed here to Chandragupta Maurya himself.

The Jain account of his end

The Digambara Jain tradition says Chandragupta gave up the throne to his son and became the disciple of the teacher Bhadrabahu. Mookerji sets out the sources. The literary versions are the Brihatkathakosha of Harishena, dated 931 CE, the Bhadrabahu-charita of Ratnanandi from about 1450, the Kannada Munivamsabhyudaya of about 1680 and the Kannada Rajavali-kathe. They agree that a famine in Bihar sent Bhadrabahu and the Jain community south, that Chandragupta followed him as his chief disciple to Shravanabelagola in Karnataka, and that he lived there as an ascetic for some years before dying of self-starvation in the Jain manner.

Mookerji adds the local evidence. A smaller hill at Shravanabelagola is called Chandragiri, and on it stand a cave named for Bhadrabahu and a temple called the Chandragupta basti, whose perforated stone screen carries sculpted scenes from the lives of Bhadrabahu and Chandragupta. The oldest inscription, of about 600 CE, links “the pair (yugma), Bhadrabahu along with Chandragupta Muni.”

The case for the tradition, as Mookerji gives it, is that Vincent Smith saw Jain tradition as the only explanation for a king who must have been under fifty disappearing from the record. Rice, who edited the Shravanabelagola inscriptions for the Archaeological Survey of Mysore in 1889, thought the first inscription recorded real events of about 297 BCE and that its Chandragupta was the Maurya emperor. His summary has Bhadrabahu foretelling a twelve-year famine, leading the community south, and dying on the smaller hill with a single disciple, and he describes Chandragupta as “stated to have been his disciple.” In that reading it is Bhadrabahu, not Chandragupta, who dies there. Chandragupta’s death on the hill comes from the later literary versions.

Others doubted it. Mookerji records that Fleet did not believe the tradition and read the Bhadrabahu of the first inscription as a later Bhadrabahu who became head of the community in 53 BCE, with a disciple called Guptigupta whose name had been confused with Chandragupta’s. Mookerji dates the oldest inscription to about 600 CE, some 900 years after the reign, and Rice himself allowed a window from about 250 BCE to about 650 CE for when it was cut. The Rajavali-kathe that Rice relied on for the life of Bhadrabahu was drawn up early in the nineteenth century, in Kannada on palm leaves, for a lady of the Mysore royal family named Deviramma.