How much credence should be given to these rumors is unclear. Those who stood by Alexander in his final days would go on to control his image and to vie for his power; they manipulated the published records of his death (and indeed of his entire reign) to suit their own purposes. They were even capable of circulating false accounts to implicate rivals. (The stories in the so-called Liber de Morteor Book About the Death, a lurid narrative purporting to reveal the poisoning conspiracy, seem to have arisen in this way.) In the years following 323, the question of who killed Alexander, or whether he died of natural causes, would be spun one way and another for political advantage, making the truth very hard to recover.

One document has particularly bedeviled modern interpreters. The Ephemerides, or Royal Journals, is now lost but was drawn on by both Plutarch and Arrian in their accounts of Alexander’s last days. The Journalsdepicted Alexander as gradually succumbing to a long fever, not at all the quick, violent demise of a poisoning victim, and makes no mention of a suspicious detail reported by other sources, that Alexander cried out from a stabbing pain in his back after drinking a huge goblet of wine. The document is thought to have been composed by Eumenes, Alexander’s Greek scribe, a witness to the events it records and therefore, in some eyes, a sound authority. But it also might have been forged, or Eumenes himself might have tampered with it to cover up a plot. Further complicating the question are the differences, some of them significant, between the summaries Arrian and Plutarch give of the Journals.Clearly one of the two writers was looking at an altered copy—or both were.

The disputes among Alexander’s contemporaries over the cause of his death make it hard to accept anyevidence on its face. This is a hall-of-mirrors world where the more convincing an account seems, the more it might be suspected to be the work of clever assassins concealing their crimes. But historical research has to begin somewhere; if nothing can be trusted, nothing can be known. The events described here are based on Arrian’s summary of the Royal Journals, but with an awareness that none of our sources has an absolute claim on truth.

After the meeting on June 3, Alexander passed the night racked by fever but managed the next day to conduct the morning’s sacrifice and meet with his senior staff. On June 4 his condition was worse, but the next day he met again with the high command and continued to plan for the Arabian expedition (apparently postponed from its original launch date). So far, Alexander refused to admit that his condition might endanger the enterprise. In the past he had regarded campaigning as a kind of restorative. After the death of his closest friend, Hephaestion, had sent him into a prolonged depression, Alexander finally roused his spirits by charging off into the mountains of Media (now northern Iran), through deep snowdrifts, to attack a stiff-necked people called the Cossaeans, “using warfare to console his grief, as though going off on a hunt—a hunt of men,” as Plutarch says.

In the private quarters of the palace, meanwhile, at least one woman helped look after the ailing Alexander, if we can place any credence in the sources that mention her. Alexander’s wife Rhoxane, or Rauxsnaka (Little Star) as she was known in her native tongue, was much younger than Alexander, perhaps in her late teens, and as far removed from him in culture as Pocahontas from Captain John Smith. She came from a place (probably modern Uzbekistan) in the rugged, mountainous region the Greeks called Bactria. Alexander’s army had slogged through two tough years of guerrilla warfare there, and Rhoxane’s father, Oxyartes, had been one of its most determined enemies. After securing his surrender, Alexander had made him an ally, cementing the tie by marrying his daughter.

Rhoxane had become pregnant within a year of her marriage to Alexander, but the child either was stillborn or died in infancy. In June 323 she was in the third trimester of her second pregnancy. What passed between her and her dying husband during his illness is almost totally unknown, except that the Liber de Morteand the Alexander Romance, two works that contain much unreliable material, do record a bizarre and moving story involving the couple. According to their account, no doubt fictionalized but perhaps based on some real incident, Rhoxane entered the king’s sickroom one night to find his bed empty. Seeing a secret passageway standing open, she crept out of the palace in pursuit of her husband, catching up to him as he crawled feebly toward the Euphrates. There the two embraced, and Rhoxane, weeping, convinced her husband to give up what she realized was his plan to drown himself. “You have robbed me of immortality,” Alexander lamented as he obediently returned to the palace. He had been trying to make his body disappear, so that his followers might suppose he had really been a god.

Two other women besides Rhoxane must have monitored Alexander’s condition anxiously, for like Rhoxane they were totally reliant on him for their status, even their safety. These were Stateira and Parysatis, daughters of the last two Persian kings, who had become Alexander’s second and third wives about a year earlier. It is not clear whether the princesses were with their husband in Babylon, or had remained at Susa, one of the Persian royal capitals, where Alexander had kept them since 331 and where he had wed them in 324. Even at Susa, though, they must have known of Alexander’s illness within a day or two of its onset. News traveled quickly between the two cities, carried by the fleet Persian postal system and by fire signals.

Alexander’s marriages to the two Persian princesses were part of his effort to fuse his army’s leadership with the elites of Asia, to create a hybrid ruling class for his tricontinental empire. He had staged a mass wedding ceremony at Susa and had matched scores of his Companions with brides from the noble families of Persis and Bactria, carefully calibrating each bride with the favor he wished to bestow on the groom. He gave the greatest prize, the sister of his own bride Stateira, to Hephaestion, so that his children and Hephaestion’s would be first cousins. He singled others out as well for the high honor of inclusion in his extended family. Nearchus, Eumenes, and Ptolemy were each married to a relative of Barsine, a former mistress of Alexander’s and the mother of his only living child, a son named Heracles. Craterus, another of his top generals, was wed to a cousin of Stateira’s, a girl named Amastris. Like the other royal women, she had been captured in 331 and maintained in state ever since, learning high classical Greek from the tutors Alexander had appointed.

But Craterus was not happy with his new bride. He was older than other members of the king’s inner circle, in his late forties rather than his thirties, and his traditionalism put him at odds with Alexander’s Euro-Asian fusion. Though Craterus revered his king and was fiercely loyal, he felt entitled to tell him, on several occasions, that he had gone too far in embracing Persian ways. Alexander resented such interventions, especially since they made Craterus a hero to the rank-and-file soldiers, former peasants and farmers who, like Craterus, still regarded the Persians as vanquished enemies, not partners in rule. But Alexander needed Craterus’ talents too much to punish his dissent. By giving him highborn Amastris as bride, Alexander was perhaps making a last effort to enlist Craterus in a project he deeply mistrusted.

The mass wedding was held in the royal palace at Susa. A row of nearly a hundred richly wrought couches was laid out in a great hall, and the Companions reclined, holding cups of wine. A toast was drunk by all, and then, in a carefully choreographed movement, the Asian brides entered and each went to sit beside her groom. Alexander took his two brides by the hand and kissed them on the lips, the Persian custom for contracting a marriage; as though on cue, the rest of the Companions did likewise. A great feast was held, and each groom escorted his bride to a waiting bedchamber in the palace complex. How Alexander stage-managed his own double-wedding night has not been revealed by our sources.