They Win Again?

To return to an issue I’ve discussed before: do the Melians have any hope of rescue, if they decide to resist the Athenians? According to the conventional Realist reading, they are simply deluded, grasping at straws (the Spartans will come, the gods will help us, you never know what might happen) rather than accept the reality of their position and the way the world works. Whether Thucydides intended us to believe this – whether here, if not elsewhere, he shares the Athenian respective – is less clear. Certainly the Spartans (let alone the gods) fail to turn up, and there’s no indication in the text that this was even a possibility; we could then assume that T takes this as a given, and wants us to reflect on (among other things) the capacity for the ‘weak’ to start pleading unicorns, or we could assume that he leaves the counterfactual possibility hanging, so we might reflect both on how far the Athenians got lucky (and so were confirmed in their irrational belief in their own omnipotence) and on the question of how much hope is enough to make the Melian gamble worthwhile.

Regardless of what Thucydides himself may have thought, there’s nothing to stop us exploring the implied counterfactual, and I’ve been persuaded by my colleague and collaborator that the Melian Dilemma game ought to include rescue by the Spartans as a possible outcome, even if it’s an unlikely one; after all, they had sent military reinforcements to Melos in the past, and not long after this event they sent Gyllipus and a small detachment to Syracuse, where he rallied them against the Athenians and helped defeat the attempted invasion. So, not unprecedented, not impossible – but exactly how likely was it? Assigning an estimated probability to the event isn’t normal historical procedure – but for the purposes of modifying the game, I do need at least an order of magnitude.

I asked this question on the Twitter and was roundly ignored, apart from one response (for which I’m very grateful):

50% they decide it is a good idea to help, 90% that the omens aren’t right and they can’t leave in time to help

— dylwah (@dylwah) January 23, 2019

These are certainly relevant factors, and there are others (we’ll come back to the issue of numbers). There clearly are reasons why the Spartans might see this as a wise move – as the Melians argue in the Dialogue, sense of obligation to old allies, concern that other allies might lose trust in them, wish to deny tactical advantage to Athenians – even if it’s by no means a sure thing. On the other hand, there must be a real chance that they might fail to act on such a decision – poor omens, yet another religious festival that leads them to delay until it’s too late, logistical problems – and that they are prevented from reaching Melos by Athens and its allies (easier to control access to Melos than to Syracuse?). Finally, even if the Spartans did send forces to Melos, there’s no guarantee of success, even if they sent substantial numbers rather than just a commander and a small detachment.

Can we attach numbers to these factors with any degree of plausibility? We could start with a basic 50% for each: so, chance that Spartan forces make it to Melos, 0.5 x 0.5 x 0.5 = 0.125; chance that Melos is saved, 0.5 x 0.5 x 0.5 x 0.5 = 0.0625. I’d be tempted to give the Spartans slightly more credit – okay, yes, Marathon – and, in the absence of a detailed study of the actual likelihood of them having a religious festival at the relevant time, reduce the chances of them failing to set out to 0.25; on the other hand, perhaps we need to add a choice between a large and small force (50:50?), with the latter having only a 0.25 chance of winning – but a better chance of getting to Melos in the first place?

So, chance of Melian escape version 2: 0.5 (Spartans decide to help) x 0.75 (Spartans actually set out) x 0.5 (despatch large force) x 0.5 (evades Athenians and reaches Melos) x 0.5 (victory!) PLUS 0.5 (Spartans decide to help) x 0.75 (Spartans actually set out) x 0.5 (despatch small force) x 0.75 (evades Athenians and reaches Melos) x 0.25 (victory!) = 0.046875 + 0.03515625 = 0.082.

This might of course be refined and modified further, if not indefinitely, but we’re still in the vicinity of under 10%, which is something I can work with for the game. What I’m not sure about is whether this approach has the slightest validity. Any thoughts, anyone?