The Battle of Marathon (490 BCE)

The first correction the sources make to the popular picture is one of scale and aim. The expedition of 490 was not an attempt to conquer Greece. Its forces were too small for that, and, as John Hyland stresses, it "did not aim to incorporate Greece into a provincial administrative system with direct Persian governors."^[2] It was a punitive operation with a short list of targets. Naxos, which had beaten off a Persian siege in 499, was still unpunished; Eretria and Athens had joined in the burning of Sardis during the Ionian Revolt, and Athens had additionally, on some accounts, insulted Darius by mistreating his heralds. The orders Herodotus records are correspondingly narrow: "He sent them with orders to capture Athens and Eretria and bring the captives before him."^[3] In Persian eyes Athens's guilt ran deeper than Sardis: its own envoys had given earth and water in 507/6, when the young democracy sought Persian help against Sparta, and however loudly the assembly later disowned the act, the court could regard the city as a subject polity in rebellion rather than a foreign enemy.^[4]

The power to reach out across the sea and punish a distant offender was itself a standard measure of royal legitimacy, inherited from Assyria and articulated in Darius's own inscriptions, where the king enjoins his heir to "punish well" the follower of the Lie and boasts at Naqsh-e Rostam of the proportionality of his vengeance.^[2] Roel Konijnendijk, surveying how the empire legitimised its wars, notes that the right to punish those who had once entered into relations with Persia and then defied it "was entirely taken for granted, and no actual reason is ever offered for these campaigns"; rebellion against the god-given king was ground enough.^[5] Herodotus's picture of Darius consumed by a personal obsession with Athens, reminded by a servant at every dinner to "remember the Athenians", is caricature, but the ideological basis underneath it, punishment as a duty of kingship, is sound.^[2] The deflationary reading is not unanimous. Cawkwell, pressing Herodotus's report that Darius's heralds demanded earth and water of the mainland cities, Sparta included, before the fleet ever sailed, holds that the aim in 490 as in 480 "was not just to punish" but "to begin the incorporation of mainland Greece within the Empire."^[6]

Seen this way, the campaign was the last act of the suppression of the Ionian Revolt, carried across the water to its most distant accomplices. It also had a constructive side. By enrolling Ionian and Aeolian contingents in the fleet, and by conspicuous acts of piety toward Greek sanctuaries, the expedition was meant to reintegrate the recently rebellious Greeks of Asia into the imperial order and to display Persian benevolence to the islanders who submitted.

Command was given not to a member of the royal house at its centre but to two competent subordinates: Datis, called "the Mede", and Artaphernes the younger, son of the satrap of Sardis and so Darius's nephew.^[3] Datis is one of the rare figures where a name in Herodotus can be matched to a name in the empire's own documents. A travel-ration tablet from the Persepolis Fortification archive records rations issued to one "Datiya", carrying a sealed authorisation of the king and travelling by the express relay from Sardis toward Persepolis in the winter of 494:

7 marriš (of) beer Datiya received as rations. He carried a sealed document of the king. He went from Sardis (via) express (service) to the king (at) Persepolis. Month XI, 27th year. (At) Hidali.

Persepolis Fortification tablet PF-NN 1809, trans. Kuhrt 2007^[7]

The identification, first made by David Lewis, is now widely accepted, and it is precious: it shows the Marathon commander moving on royal business through the imperial communications network four years before the battle, and it anchors a Greek narrative name in the bureaucratic reality of the empire.^[7] The battlefield itself has yielded one further possible Persian-side trace, a Babylonian cylinder seal found on the plain, perhaps lost by a Babylonian serving with the expedition.^[8] Datis, a Mede in high command, recalls Cyrus's Median general Harpagus and the old association of Medes with the conquest of Ionia; his appointment linked the new expedition to that first subjugation of the Greeks of Asia.^[2] The presence in the expedition of Hippias, the aged exiled tyrant of Athens, points to a further aim behind the punishment: the installation at Athens of a ruler bound to Persian interests. Hippias chose the Marathon landing site and drew up the army, remembering the plain from his own father's day.^[9]

Nothing about Marathon has been more distorted than the size of the Persian force, and here the correction is sharpest. The later Greek tradition, and the modern imagination fed by it, pictures a vast Asiatic host. Herodotus himself, tellingly, gives no figure at all for the Persian army; the only number he attaches to the expedition is the fleet, "six hundred warships", and even that is suspect, being the identical round figure he assigns to two other Persian fleets.^[10] Modern estimates, built from the carrying capacity of the ships and the space of the battlefield, put the Persian combatants most plausibly between about 12,000 and 20,000, with the whole seaborne population, fighters, rowers and camp, perhaps 30,000 to 40,000 mouths to feed.^[2] Dandamaev sets the ceiling at fifteen thousand fighting men; reconstructions in the military-history tradition, following Lazenby, prefer eighteen to twenty-four thousand foot with a single thousand-strong cavalry regiment; the spread of modern estimates is wide, but all of them are small fractions of the hundred thousand that Cornelius Nepos later put ashore, for Dandamaev "a concoction from a much later period, meant as a tribute to the Athenian soldiers."^[8][11] The cavalry, though certainly present (the horse-transports recur throughout Herodotus's account), can only have numbered in the dozens or low hundreds, given the difficulty of shipping and feeding horses; it was nothing like the thousands of later legend.^[2]

This fits the wider demolition of Persian army sizes carried out by Sean Manning. The "millions" and the hundreds of thousands of the Greek sources belong, he argues, to a literary stereotype (the barbarian horde) and to the ancient habit of inflated enemy counts; realistic Achaemenid field armies, constrained like all pre-modern armies by transport, logistics and command, were comparable in scale to Hellenistic and Roman ones, that is, in the tens of thousands, not the hundreds.^[12] Manning also turns the same skepticism on the casualty figures. Herodotus reports the dead as "about six thousand four hundred" Persians against "a hundred and ninety and two" Athenians.^[13]

The Athenian figure, 192, is probably reliable, since Athens kept public casualty lists; the Persian figure is another matter. As Manning observes, 6,400 and 192 stand in a ratio of almost exactly 33⅓ to 1, and 6,400 is a suspiciously high, suspiciously round number, far larger than any loss Thucydides or Xenophon record for a battle between Greeks.^[14] Hyland notes that accurate counts of enemy dead depend on bureaucratic procedure on the losers' own side, which was absent here; the bodies lay in a marsh and were counted, if at all, by guesswork. A Persian toll in the hundreds, or at most the low thousands, is as much as the evidence will bear.^[2]

The voyage out was not a blundering detour but a methodical sweep, and it displayed the empire's characteristic pairing of reward and punishment. Setting out from Samos, the fleet took Naxos first (the island that had defied Persia in 499), burned its temples and city, and enslaved those it caught.^[15] It then moved from island to island through the Cyclades, taking hostages and levying contingents, punishing those who resisted, such as Carystus, and leaving alone those who submitted. At Delos, the sacred birthplace of Apollo and Artemis and a pan-Ionian cult centre, Datis did the opposite of sacking: he anchored offshore, recalled the fleeing islanders, and made an enormous offering: "He piled up three hundred talents of incense on the altar and burned it."^[16]

The scale of the gesture (three hundred talents of frankincense do not sail with a fleet by accident) shows it was planned, an act of calculated imperial patronage toward a shrine associated with the very Ionians the empire was reabsorbing.^[2] Briant reads the whole voyage as "a sort of propaganda tour, intended to show the islands that they had nothing to fear from the new master", and sees in it "the two complementary sides of Persian ideological strategy: patronage granted to the sanctuaries but pitiless repression in case of refusal to submit."^[1] Eretria, when the fleet reached Euboea, fell after a six-day assault when two of its leading men betrayed it; its temples were burned in reprisal for Sardis and its people taken captive, as Darius had ordered.^[17]

The army came ashore at Marathon, a stretch of coast on the far side of Attica from the city. The choice, on Hippias's advice, put the Persians on ground suited to cavalry and at a distance from Athens, perhaps to avoid pinning the enemy inside its walls, perhaps to buy time for sympathisers within the city to rise.^[2] The Athenians and their Plataean allies marched out and camped opposite; the Spartans, pleading a religious festival, set out too late and reached the field only after it was over. For reasons Herodotus does not explain, the two armies then sat facing each other for several days.

Of the battle itself the account is famously thin. Herodotus "forgets the cavalry entirely", as Hyland notes, mentioning it only in the Persians' astonishment that the Athenians would charge without horse or archers of their own; the absence of the Persian cavalry from the fighting is one of the enduring puzzles of the day, variously explained as a separate embarkation, a watering expedition, or simply a gap in a Greek source that did not understand what it was describing.^[2] The one ancient scrap bearing on it is a proverb, "cavalry apart", which the Byzantine Suda glosses with a tale of Ionians signalling to the Athenians "that the cavalry were away"; the notice is late and weak, as A. R. Burn observed, but it is one of the very few post-Herodotean statements about Marathon that offers a fact rather than patriotic embellishment, and the re-embarkation hypothesis built on it has run through the scholarship since the nineteenth century.^[18] How and Wells put the classic case for it: the horsemen, useless against a hill position but precious for a dash on Athens or a battle on the Phaleric plain, were already aboard for a move by sea, and their absence gave Miltiades his opening.^[19] Cawkwell reaches the same verdict by another road: only seven ships lost and a hundred and ninety-two Athenian dead are hard to square with cavalry engaged, and a battle joined when the Persians "were no longer able to use all their army" was "not the great affair that the Athenians represented it to have been."^[20] What Herodotus does give is the Athenian charge and its result:

They set forth at a run to attack the Barbarians... they were the first of all the Hellenes about whom we know who went to attack the enemy at a run, and they were the first also who endured to face the Median garments and the men who wore them, whereas up to this time the very name of the Medes was to the Hellenes a terror to hear.

Herodotus 6.112, trans. Macaulay^[21]

The course of the fighting, on the sober reading, was not the neat double envelopment of legend. In the centre, where the best Persian troops stood, the Persians and their Saka (Scythian) contingent broke the Athenian line and pursued it inland; on both wings the Athenians won and then turned in on the Persian centre, and the pressure caused a general flight to the ships.^[22] The heaviest Persian losses came in that final scramble at the beach and the marsh; there was no encircling manoeuvre. Herodotus's own words show the Persians fighting well and even winning where their picked men stood; the notion of a one-sided rout of a cowardly Asiatic mob by superior Greek spearmen is a later gloss, drawn, Hyland argues, from the very different battles of Xerxes's war a decade later and read back into the silence of the Marathon narrative.^[2]

What happened after the battle is the single clearest indication of what the campaign was, and was not. The Persians re-embarked, sailed around Cape Sunium hoping to reach an undefended Athens before its army could return, found the city forewarned and the army back in place, lay briefly at anchor off Phaleron, and then sailed home.

The Barbarians however came and lay with their ships in the sea which is off Phaleron... they anchored their ships, I say, off this place, and then proceeded to sail back to Asia.

Herodotus 6.116, trans. Macaulay^[23]

As Briant emphasises, this decision to break off and depart, rather than to regroup, resupply and press the war, "confirms the fact that at this time Darius did not intend to establish long-term dominion over Greece. Datis was simply charged with making a raid, destroying sanctuaries and houses, taking prisoners, and setting sail for the coasts of Asia Minor."^[1] The expedition even continued its policy of cultic patronage on the way back, Datis pausing at Delos to restore a looted statue of Apollo, a sign that the failure at Marathon had not ended Persian courtship of Greek opinion.^[2] The Eretrian captives were carried to Susa and, far from being massacred, were settled by Darius at a place called Ardericca in Elam, where Herodotus says their descendants still lived in his own day, speaking their original tongue.^[24] The treatment of the prisoners is itself instructive: this was punishment and deportation on the Assyrian model, stopping well short of extermination, and the site chosen was no wasteland but a royal way-station, Herodotus's own word is stathmos, some 210 stades from Susa, perhaps identical with the Urdalika of the Assyrian annals.^[25] The deportation dossier assembled by Matarese adds two correctives to the tale of annihilation: the exiles' fate was famous enough among Greeks to serve, as Darius surely intended, as a warning, and Eretria itself lived on, manning ships at Artemision and Salamis and fielding hoplites at Plataea within a decade.^[25]

For Athens, Marathon was epochal, and became more so with time. The victory of the citizen-hoplites secured a young and fragile democracy, raised the city's standing against Sparta, and hardened into myth: the day the free citizens of Athens, standing almost alone, turned back the barbarian. Later Athenian art and oratory built the battle into the foundation stone of a whole ideology of Greek liberty against Asiatic despotism, and it is that Athenian reimagining, not the event of 490, that gives Marathon its outsized place in memory.^[26] The legend-making began at once: the runner Pheidippides's dash from the battlefield to Athens already belongs, as Rollinger notes, to "the rich production of legends" rather than to the event, and even the scale of what the Athenians defeated, an invading army or an exploring force, remains open to debate.^[27]

From the imperial side the reckoning was very different. Briant's summary is deliberately deflationary, and states the Persian-side view as sharply as it can be put:

In sum, from the Persian point of view, Marathon was nothing but a minor engagement that had no effect whatever on the Aegean strategy defined by Darius. This was surely the version found in Achaemenid propaganda.

Briant, From Cyrus to Alexander (2002), pp. 160–161^[1]

Darius kept the allegiance of the Ionian Greeks and the deference of most of the islands regardless of the setback; the Greeks of Asia, freshly and brutally reconquered, saw in Marathon no weakening of the empire that ruled them. On this reading the campaign's real objective, mastery of the Aegean islands, had already been achieved before the army ever landed in Attica, and the defeat "did nothing to change" it.^[1]

Recent scholarship has tempered the deflation without overturning it. Hyland cautions that to call Marathon a mere "minor engagement that had no effect whatsoever" is "somewhat reductive", given Athens's regional prominence and the resources the empire committed. The campaign had failed at two of its stated goals, the punishment of Athens and the restoration of Hippias, and that was a real and public loss of face; if it was meant to close the file on the Ionian Revolt, it left the file open and demanded further operations.^[2] Yet Hyland agrees that the failure was not catastrophic and was not read as such: the generals seem to have suffered no serious consequences, Datis's sons and Artaphernes both went on to high command under Xerxes, and the burning of Naxos and Eretria and the train of captives could be presented at court as substantial success.^[2] The likeliest imperial verdict lies between Briant's footnote and the Greek epic: an incomplete punitive campaign, a genuine but limited reverse at its far end, folded by its commanders into a favourable report and by Darius into unfinished business, which his son would take up on a wholly different scale.

That Herodotus is nearly all we have is underlined by what happens when other voices are heard. The court historian Ctesias, writing later still, gives a version so garbled that it has Datis killed at Marathon and his body refused to the Persians who asked for it, a claim flatly contradicted by Datis's continued activity afterward and by the survival of his family in royal favour.^[28] The variant is worth nothing as fact, but much as a warning: within a century the truth of the campaign had already dissolved into competing stories, and even in antiquity there was little independent purchase on it. Modern reconstructions of the battle's topography, the direction of the lines, the fate of the cavalry, and the true casualty count remain unsettled precisely because they all rest, in the end, on the same few pages of one Greek historian. Marathon is thus at once a battle and a lesson in method: an event of the empire's history that survives almost wholly in the memory of its enemies, and that must be recovered by reading that memory critically, weighing it against the empire's own logic, its documents, and the hard limits of the possible.