57. Captivity in antiquity: were several thousand Athenian prisoners starved to death?
Exemplary violence, assimilation by marriage, incarceration in quarries, running around with Plutarch and Arrian and Xenophon and Thucydides
Some things are becoming clear: The historical sources discuss men, their capture, and their occasional returns from captivity, as well as the conditions that attend such a return, far more often than they specifically mention the return of female or child captives.
But women and children were, in numeric and in proportionate terms, far more likely to be taken captive: to make a broad assumption, if one in four members of the free population of the ancient world one was an adult man of military age -- that is, between 14 and 60 -- then three in four were not. Leaving aside the small portion of men too old or too injured to bear arms, then in a free population of 10,000 people we have something like 7,000 potential non-fighting captives, as many as a third to one-half of whom, given life expectancy and mortality rates in antiquity, may be under the age of eighteen. (For demographic replacement purposes, it is estimated that each fertile woman must give birth to, on average, somewhere between six and nine children to beat the childhood mortality rates. To be clear, in my understanding this does not count pregnancies, but live births.)
Men may be taken as prisoners in the aftermath of battle, on the field or surrounded in camp with little chance to break out, like the Roman general Gaius Hostilius Mancinus and his troops, but they are generally taken and kept alive while hostilities are still in progress, or when the war has dragged on such that a settlement is preferred to further fighting, or when their conqueror does not mean to separate them from their families in defeat, but to incorporate them into a new diplomatic settlement. Elite men are, it seems, both more likely to be kept alive and less likely to be dispatched into slavery. At least when these circumstances apply.
We may note that exemplary and retaliatory violence is often visited -- or sometimes promised to be visited, by some law resolved in a council or assembly -- on male prisoners of war, which can be seen from these passages of Xenophon in the aftermath of the battle of Aegospotami, a naval clash near the (very bitter) bitter end of the Peleponnesian War:
"Thereupon many charges began to be urged against the Athenians, not only touching the outrages they had already committed and what they had voted to do if they were victorious in the battle —namely, to cut off the right hand of every man taken alive,—but also the fact that after capturing two triremes, one a Corinthian and the other from Andros, they had thrown the crews overboard to a man. And it was Philocles, one of the Athenian generals, who had thus made away with these men.
[32] Many other stories were told, and it was finally resolved to put to death all of the prisoners who were Athenians, with the exception of Adeimantus, because he was the one man who in the Athenian Assembly had opposed the decree in regard to cutting off the hands of captives... As to Philocles, who threw overboard the Andrians and Corinthians, Lysander first asked him what he deserved to suffer for having begun outrageous practices towards Greeks, and then had his throat cut."
Women were more likely to be taken prisoner in a sack, like at Melos:
"[The Melians] yielded to the Athenians, so that the Athenians could do as they wished concerning them. And the Athenians slaughtered the Melian men that they had taken, the ones that had attained puberty, while the women and children they turned into slaves,* and they settled a colony in the town, afterwards sending out five hundred settlers." Thuc. 5.116
*andrapodised, a word which Kathy L. Gaca very cogently argues contains brutal violation and violence.
After the defeat of a people, the slaughter of many adult men, the enslavement and brutalisation of women and children, and the dispossession of the remaining survivors of their moveable property and sometimes also their land and crops, prisoner restoration is not a common occurrence. If the defeated city was part of an alliance or coalition, its allies were victorious, and those allies insisted (were able to insist) on some form of the return of enslaved prisoners at the end of hostilities, some number of (surviving, identifiable, not-sold-too-far-afield-to-recover) captives might return to freedom and their despoiled homes. The victims, and the allies of the victims, of a town whose people had been subject to ἀνδραποδίζειν, andrapodising, viewed it as an outrage and an injustice, something that cried out for revenge (a revenge which would likely be andrapodising in its turn). The victors generally considered it a natural perk of their success.
In the Hellenistic period there are a handful of examples of kings essentially changing their minds and deciding to (take a few steps to) restore a dispossessed and enslaved populace: the city of Stageira, birthplace of Aristotle, is one of the few mentioned.
Plutarch Life of Alexander 7.3: "The city of Stageira, that is, of which Aristotle was a native, and which [Philip] had himself destroyed, he peopled again, and restored to it those of its citizens who were in exile or slavery." This took place when Alexander was old enough to need a tutor, so perhaps after 350BCE.
But Stageira was again deserted, or so it seems, by the time of Strabo (see Geography 7.8.35; Strabo was born c.64/3BCE and lived until at least midway through the reign of the Emperor Tiberius, dying perhaps c.23/4CE): either their restoration had not been sufficient to allow the town to thrive, or they had subsequently been andrapodised again during the wars of Alexander's successors, or in the Roman wars in Greece.
It seems reasonable to assume that the destruction of a city, the death of (many of) its male citizens, and the enslavement of its non-fighting population (which includes the rape of women and girls and sometimes the mutilation of boys, and the continued enslavement and probable re-brutalisation of its population of already-enslaved people) is a situation where in the vast majority of cases, the enslaved have exceedingly little hope of a restoration of their freedom. In the case where freedom is restored to some number of the enslaved, with the idea of reconstituting their city as a living community, one must consider the ongoing consequences of such a set of traumatising experiences for the survivors.
Communities that surrendered under terms frequently did so to avoid this destructive violence and retain an existence as a community, albeit one now subject to the exactions of another power. Those exactions could include military service, hostages (though men on military service in the midst of another power could also be seen in light of hostages), tribute in money, moveable property, and/or people (slaves), cession of landholdings, and the marriage of appropriately-ranked women to individuals of high status among the new overlords (which can be seen through the lenses of hostageship, dispossession [of the subject men's control over the marriages and property of their female relatives, among other things], and forced assimilation: for this especially Alexander's weddings at Susa, Arrian, Anabasis 7.4.4 and following, though some of these royal women were captured in the aftermath of battle or sack, and only escaped enslavement by virtue of Alexander's aristocratic recognition of and respect for their own aristocratic status: Arrian describes Alexander's alleged generosity and graciousness to the captured female relatives of Darius after the Battle of Issus).
When Greek men were captured in the aftermath of battle, what happens to them frequently seems to depend on how high feelings are running, the balance of power in the wider war, and the personal inclinations of the victorious commander. Perhaps the best description of a military surrender that didn't end in a massacre or sale into enslavement is the one described by Thucydides (4.37 and following) in the wake of the battle of Sphacteria in 425 BCE. Four hundred and twenty hoplite soldiers fought on behalf of the Spartans until backed into a corner, at which point the Athenians asked them to give up their weapons and surrender. The surviving two hundred and ninety-two of them, of whom one hundred and twenty were Spartan citizens, chose to surrender rather than fight to the death. This, after the Spartans' reputation from Thermopylae onwards, was apparently quite a shock: they had been expected to fight on to the end.
The prisoners were brought back to Athens, and the Athenian assembly decided to keep the Spartans under guard in chains until they could come to terms. They also decided that if the Spartans invaded Attica in the meanwhile, they would kill the prisoners. (The prisoners were also hostages, in this sense, and we should consider any prisoners kept alive and not sold to likely also serve some form of hostage-purpose even if it is not stated in the sources.)
The Peace of Nicias that intervened between the two halves of the Peloponnesian War provided for the mutual return of captives (Thuc. 5.18): they would "give over" or "yield up" those captives of each principal (Athens and Sparta) and their allies that they "had". (Thucydides' text does not allow us to be sure, reading it, as to whether this included non-citizen-status captives, or captives sold abroad and beyond easy recovery.) The surrendered Spartans who returned home at least initially were restored to their citizen rights: I need to follow up Jean Ducat's study of "The Spartan 'Tremblers'" in Sparta and War (when I get access to it) to follow up what happened subsequently, and the scholarly discussion around why: it seems after a while they were deprived of their citizen rights.
The Spartan prisoners (and possibly their allied hoplites) from Sphacteria seem to have had the closest experience available in antiquity to that of a modern prisoner of war. It is not very close to what the Geneva convention prescribes. I do not mean to say that they were well-treated (though they were likely better-treated than the Athenians taken prisoner in Sicily), but while they were humiliated and under threat of death, they were not intentionally killed, they were apparently kept together, and they were not sold into slavery.
We should note that estimates for the size of the Spartan male citizen body during the Peloponnesian Wars vary (perhaps 6000? not more than 10,000) but that 100 men represents at least 1% of the whole number,
We can compare this experience with that of the Athenian disaster in Sicily in the latter part of the Peloponnesian War. This disaster compromised Athenian strength in men under arms very significantly.
The army of the Athenians and their allies ended up divided in two, and each section experienced defeat in detail. (Thuc. 7.82 and following.) Demosthenes, in command of one section, surrendered under terms: none of his troops were to be killed by violence, or in chains, or from being deprived of things necessary to life.
Nicias, in command of the other section, was told by the Spartans and the Syracusans that Demosthenes had surrendered. He tried to negotiate a settlement for his own troops to withdraw, offering to pay an indemnity and leave Athenians as hostages for the security of this payment. His attempt at terms was refused. He lacked supply, and his attempt at a fighting retreat turned into a rout and a slaughter. Nicias surrendered himself to the Spartan leader Gulippos, after which Gulippos gave the word to take live captives.
If I am correctly parsing the text of Thucydides at 7.85, the Spartan and Syracusan troops were taking occasional prisoners anyway, and after Gulippos's instructions, some (but not many) of these captives were given over to be held in common, or given over as public captives, but a great many of the captives were instead διακλαπέν, "kept back/kept alive by stealth," and Sicily was "glutted full" with them, "seeing as," Thucydides says, "[they were not taken] by an agreement like the men who were taken with Demosthenes."
Some escaped the battlefield, and some were made slaves and escaped later. Of those who did not escape and did not die, and (it's implied from that earlier διακλαπέν) were given over as captives-held-by-the-authorities, most were held in the quarries near Syracuse. (Thucydides is not completely clear, but the text suggests that this included those taken with Demosthenes and not just with Nicias.) Nicias, who gave his name to the earlier Peace, was killed, as was Demosthenes. Thucydides suggests that if Nicias had not been killed, he might have been tortured ("But some of the Syracusans who had been in correspondence with him were afraid, it was said, of his being put to the torture" 7.86.4). The conditions in the quarries were appalling, and some men -- the Athenians -- were kept here for eight months. Most of the Athenians' allies, however, were sold into slavery after seventy days, except those from Sicily or from Italy. Thucydides estimates seven thousand captives as a minimum: most modern scholarship considers this number high but not completely out of bounds, even as a total of men taken under arms. It is fairly reasonable if we consider that this number may also include any camp followers and enslaved people in whatever supply train remained with the armies of Demosthenes and Nicias, even if they are not specifically mentioned.
One thing to bear in mind about hoplite warfare is that its level of blunt force trauma is quite high. Another is that deaths are generally quite low among the victors, with most fatalities occurring among the defeated after they have turned to flee: morale breaks and deaths follow, rather than the reverse. (I simplify terribly.) So a large number of captives out of the whole is quite plausible if members of the broken army could not flee effectively.
From what Thucydides (7.87) tells us about the Athenian prisoners in Sicily, we can see that there is no particular consensus that one should offer one's not-yet-enslaved prisoners any standard of care: "they were kept in a hollow space exposed to the sun, and first the choking heat distressed them as they had no shelter; and then the nights, coming on the opposite way, autumnal and cold, forced them into sickly weakness by the change." They ate and slept and shat all close together, and the bodies of those who died from their wounds were not removed, or at least not very quickly, and during the eight months, each of them were only given "a measure of water and two measures of grain" (the measure is often translated as "a half-pint", so presumably this is a day's ration).
If these are the relatively high-value prisoners, kept together and just about alive in case they are needed for hostages, to get ransom money, or for exchange in order to close hostilities on better terms, that's not really a very good statement on treatment.
Diodorus Sicilus and Plutarch say that the Athenian prisoners were left to die in the quarries, but contemporary sources are silent, and Thucydides, who might be expected to mention this in his account of the disaster, says nothing of their ultimate fate. Kelly (1970) suggests that those who survived the eight months and whose friends and families, or at least ransom-minded wealthy fellow-citizens with an interest, could afford to ransom them were eventually brought home, with only the remainder condemned to enslavement and/or death. The Athenians took Syracusan prisoners later, and do not appear to have starved them to death, or otherwise killed them out of hand.
These Sicilian prisoners are mentioned in Xenophon: kept in quarries and escaping (409 BCE); a generation later, in 373 BCE, the Athenians are willing to ransom Syracusan captives via a fixed-price agreement: would they have, if several thousands of their own citizen-men (roughly estimating an adult citizen man population of between 25K and 50K at this time and a total Attic population of between 250K and 500K, demography in antiquity is difficult, these are rough figures, do not make me go into the weeds on Athenian demography, I don't like it there) had been starved to death within living memory? It seems like it would cry out for revenge in kind. So mistreatment followed by ransom seems more plausible: Plutarch's main interests in his biographies did not lie in the details of events, but the character of men he considered great leaders, while Diodorus condenses the whole episode.
I have gone through a lot of words and we are still mainly dealing with the Greeks. I'm aware that I'm perhaps sometimes repeating myself, and that my treatment has not been particularly chronological, nor particularly thorough in its details as yet: we're still in the read-all-the-things, throw-bits-at-walls-to-see-what-sticks phase of fucking around and finding out.
My kingdom for off-site access to my closest academic library’s resources, or a fortnight’s worth of childcare, or both. (I don’t have a kingdom. Kings are stupid.)