The Fall of Babylon: reconciling Herodotus with the historical evidence

When Babylon fell to the Persian empire in 539 BC, it was probably the most populous city in the known world. It was famous for its thick double walls that were surrounded by a moat, and the city was supposed to be able to withstand a siege for several years. Aside from the city walls there was also an additional, second line of defense, that enclosed a much larger area around the city, more than 50 kilometers on each side. The northern border of this area was the famous "Median Wall" that Xenophon reports about in his Anabasis when the ten thousand Greek mercenaries passed through the region. Only a short section of this wall has been uncovered by archaeologists, but it's clear from the ancient sources that its western endpoint stretched to somewhere beyond Sippar, while its eastern endpoint was the city of Opis. The Babylonians also built a southern wall from Babylon to Kish, and from there towards the Tigris river, but it's not known where exactly the wall intersected the river in the south. The Euphrates and Tigris river acted as the western and eastern borders of this area, as these rivers were impassable for armies. 

The northern and southern walls were surrounded with moats, that were essentially canals connecting the Tigris and the Euphrates, and in times of crisis, they were flooded and turned into extensive swamps. With these, the entire core area of Babylonia was surrounded by water in all directions. This enclosed area included Sippar in the north-west corner, Opis in the north-east, Babylon in the south-west corner, Kish in the south, and the city of Kutha was also somewhere inside it. The entire defense system is described in the cuneiform inscriptions of Nebuchadnezzar II, but it's also extensively covered by Greek authors (Herodotus, Ktesias, Abydenus) whose descriptions had usually been dismissed on the grounds that the numbers they provide for its size seemed too fantastical for city walls, not realizing that their descriptions referred to the entire enclosed area of Babylonia and its walls and moats, and not just to the city walls of Babylon itself.

With this spectacular multi-layer defensive system, Babylonia should have been impenetrable, and yet the history of the Persian conquest proves otherwise: in the Babylonian Chronicle, Cyrus the Great defeats the Babylonian army in a decisive battle at Opis sometime between September 27 and October 9 of 539 BC. Then on October 10 he marches to Sippar and captures it without battle, and on October 12 his general Gobryas marches into Babylon and takes the city without a siege. There is no mention of breaching the walls of either the city of Babylon or of the enclosed Babylonian area, and the entire campaign is over in a few days.

Compared to this, Herodotus provides a completely different narrative about the Persian campaign: 1. Cyrus first spends a year diverting the Gyndes (Diyala) river by dividing it into 360 small channels, as he apparently needs to cross the river before he can march on Babylon. 2. Then the Babylonians come out of their city to meet the Persians but they are defeated in battle and retreat behind the walls of the city. 3. Cyrus then starts a siege and it looks like it could last for years. 4. After wasting a lot of time Cyrus decides to divert the Euphrates upriver of Babylon: the river is channeled into the lake/marsh that was part of the defensive system against the Medes, and as the riverbed bisecting the city dries up, the Persian soldiers can march through it unopposed, bypassing the walls. The city is then quickly taken in a surprise attack.

At first glance, Herodotus's account seems impossible to reconcile with the Babylonian cuneiform sources: in Herodotus, there is a long siege and canals dug to divert the Euphrates river, while in the Babylonian version the entire campaign is over in days, there is not enough time in the chronology for a siege of the city or for digging canals to divert the Euphrates from it. As a result, modern historians usually dismiss Herodotus' version of the events, and the general attitude is that "he made it all up". And as for diverting the Gyndes river before the battle, the explanations that modern academics came up with range from "it was part of a diversion maneuver" (makes no geographical sense), to "myths retaining a memory of how the Persians created new canals in the region for agriculture", and of course "Herodotus just made it up".

But neither Herodotus or the Babylonian version talk about the Persians neutralizing the second line of defense, yet this is a problem that the Persian army must have had to solve first, before they could march on the city of Babylon from Opis. The Persian army coming from the east couldn't have crossed the deep and wide Tigris river while the Babylonian army was waiting for them on the other side, and the Median Wall and its southern counterpart with their flooded moats would have been just as impenetrable. South of Opis and the Median Wall the Tigris merges with the Gyndes river and becomes navigable, and it's particularly hard to cross. So the Persians would be stuck, and they would have to start a siege *of the entire enclosed area* of Babylonia, a siege not just of the city of Babylon, but of the large 50x50 km region that Nebuchadnezzar II surrounded with walls and moats. And the only way to win this siege is by diverting one of the rivers that provide the water for its defenses (the Tigris or the Euphrates) before that river reaches the defense lines (so north of Sippar or Opis). But to divert the Tigris, one first has to divert the Gyndes that flows into it, otherwise the water redirected from the Tigris would just find its way back into its lower course through the Gyndes. 

This has to be the reason for Herodotus's story about Cyrus spending a year splitting the Gyndes river into countless small canals. Herodotus couldn't figure out how this action fit into the military campaign, so he provided a silly explanation for it, that Cyrus lost his favorite horse while trying to cross the Gyndes, which motivated him to make the crossing easier for everyone, and then ended up wasting the entire campaigning season on digging the canals for this. The real reason must have been that diverting the water of the Gyndes (and maybe diverting also part of the Tigris too after it, into the same canals that were dug for the Gyndes), would cause the water levels at Opis to drop significantly, and the Persian army could then march into the enclosed area through the dried up riverbed. This event must have preceded the Battle of Opis described in the Babylonian Chronicle, and probably the entire enclosed area had been under siege for over a year before the Persians finished working on the canals.

This solves the first issue with Herodotus' narrative, but it still leaves open the problem of the siege of the city of Babylon itself: in the cuneiform sources only 2 days pass between the fall of Sippar and the fall of Babylon, which is obviously not enough for a siege or for digging the canals to divert the Euphrates, so the only way to save Herodotus's narrative is to imagine that his description of the siege of the city of Babylon actually refers to the siege of the entire 50x50 kilometer enclosed area, and his story of diverting the Euphrates refers to diverting the Tigris instead. This is supported by the fact that in Herodotus' description, the Euphrates is redirected using the canals and lakes created as a defense against the Medes, but this is the northern side of the defensive system around the 50 x 50 km area that ends at Opis, and not the one around the city of Babylon. It seems probable that Herodotus was told the true story about how Babylon fell to the Persians, but he misunderstood it and thought that the story was about a siege of the city itself, since he couldn't imagine that there could be such a thing as a siege of the entire enclosed area around Babylon.