Did Socrates die?

I have been reading Sextus Empiricus M 10 for a conference later this year. I have to talk about a particular chunk of it, the very last bit in fact, and I'm finding it fascinating. I spent a lot of yesterday wondering about this bit (10.346-7). (This also gives me a chance to try out posting some Greek):

εἰ ἀπέθανε Σωκράτης, ἤτοι ὅτε ἔζη ἀπέθανεν ἢ ὅτε ἐτελεύτα. καὶ ζῶν μὲν οὐκ ἀπέθανεν· ἔζη γὰρ δήπουθεν καὶ ζῶν οὐκ ἐτεθνήκει. οὔτε δ’ ὅτε ἀπέθανεν· δὶς γὰρ ἔσται τεθνηκώς. οὐκ ἄρα ἀπέθανε Σωκράτης.

If Socrates died, he died either while he was alive or when he was dead. And he did not die while alive, for he has living then and had not died while alive. Nor did he die when he was dead, for then he would be dead twice over. So Socrates did not die.

There's a lot we could say about this argument. Of course, Sextus’ intention is to cast doubt on the reasonable assumption that Socrates did die by raising difficulties for the assumption that it is possible for something, or someone, to pass away. (So this argument provides the counterpart to the concerns raised in 10.343 against the possibility that a baby might be born or come to be.) But it is also significant that Sextus has chosen to cast this argument not in general terms about ‘someone’ passing away, but about one person in particular: Socrates. The question whether Socrates died is not such a surprising one, at least perhaps not for a Platonist. A Platonist of a certain persuasion might, after all, argue that if, as is claimed in the Phaedo, the soul is immortal and if, as might again be a reasonable Platonist claim, each of us ought to think of himself principally as an immortal soul, then there is a reasonable sense in which Socrates – i.e. the immortal soul – does not die when the hemlock has been drained. It is in any case not implausible that some such debate might lie behind these remarks if we remember Sextus’ recent reference to an argument about whether a person (ἄνθρωπος) is a body, a soul, or some compound of the two. If Socrates is his soul and a soul is immortal then he does not die. If, on the other hand, Socrates is a compound of a body and soul and death – perhaps following the initial definition in the Phaedo (cf. M 9.196) – is the coming apart of body and soul, then Sextus' problem mighthave to be faced.

No doubt, there is something decidedly sophistic about the argument but it is not easy to say precisely what. A brief look back at the first appearance of this form of argument, in M 9.268–9 also reveals some important characteristics of its re-use here [1]. In M 10 Sextus does not bother with the possible added complication of wondering if it is in fact right and proper to talk of ‘a dead Socrates’. On another occasion, and with reference to something more like the M 9 version, he could have wondered if dying is not a change from one state (being alive) to another (being dead) but in fact an absolute passing away into not-being. And indeed, when summing up the overall picture from this argument and its successor, at 10.350, Sextus relies on the strong premise that to pass away is to exist no longer and argues that in that case neither can what is pass away (because it is) nor can what is not (because it is not). In the present argument, however, Sextus wants to rely only on the thought that living and being dead are exclusive and exhaustive states.

There is also something interesting going on with the tenses. In contrast with the earlier appearance of this sort of argument at M 9.268–9, the whole argument here in M 10 is cast in the past, beginning with the hypothesis: ‘If Socrates died (ἀπέθανε), he died so either when he was alive (ἔζη) or else when he was was dead (ἐτελεύτα: Sextus presumably varies the verb here just so as not to repeat the verb in the hypothesis). Both verbs are imperfect, contrasting with the aorist ἀπέθανεν, since he wants to offer the two possible states Socrates might have been in when he died. The dismissal of the two possibilities requires more intricate work with the tenses: Socrates did not die while alive (ζῶν: a present participle) for he was alive (ἔζη: imperfect) and he had not died while alive (ζῶν οὐκ ἐτεθνήκε: a present participle and a pluperfect). This last point is of course correct: it is no doubt true to say of some living person that he ‘has not died’ and Sextus has carefully crafted his claim to rely on this foundation. But it is not perhaps so obviously impossible to say of someone alive that ‘he is not dying’. There are two points here. First, Sextus has helped himself to something which supports his case without canvassing all the other possibilities. Although ‘When Socrates was alive he had not died’ is true, we might be less happy to accept as necessarily true the related but distinct ‘When Socrates was alive he was not dying’. Second, it is not legitimate to assume that it is possible to transform the tense of a proposition in this way and generate a true present tense claim from a true past tense claim. It is not legitimate to infer that ‘When Socrates is alive he is not dying’ must be true because ‘When Socrates was alive he had not died’ is true.

I think this is a good reason to see Diodorus Cronus' influence lying behind this argument, as well as the argument which follows for which he is explicitly credited (10.347). Diodorus has earlier been credited by Sextus with the observation that true past tensed propositions need have no simply related true present tensed proposition (M 10.97–100): that ‘Henry VIII had six wives’ is now true does not require us to believe that ‘Henry VIII has six wives’ was ever true. It is worth remembering, first, that this discussion follows a Diodoran paradox of motion that will be relevant to interpretation of the example of the wall which follows and, second, that Sextus himself takes issue with Diodorus and reasserts the necessity of a link between the truth values of past and present tensed forms of the same proposition. If that is the background to at least part of the Socrates argument then we might view it either as a crafty Diodoran example of the sort of paradox that might arise from being careless about transforming true past propositions into true present ones or else some attempted, and perhaps not entirely successful, response to his concerns. Either way, we would be right to think that Sextus’ engagement with Diodorus lies behind this brief argument no less than the more extended one to follow.

Another, more speculative thought then occurs. It's probably right to think that Diodorus was a strong influence on lots of Hellenistic thinkers, the Epicureans included. Certainly his paradoxes of motion seem to bear on problems in atomic physics. Now, if Diodorus is responsible for or inspired the worry about Socrates, it strikes me that he may also have set Epicurus thinking along the lines that would lead to one of his most famous (notorious?) arguments about the fear of death: death cannot be a harm to the living or the dead since ‘when death is we are not, and when it is we are not’ (see Letter to Menoeceus 125). This similarly trades on the absolute exclusivity of life and death: the living are alive and therefore cannot be affected by death; the dead are no longer and are therefore also immune.

[1] M 9.268–9: οἷον ὁ Σωκράτης ἤτοι ὢν θνῄσκει ἢ μὴ ὤν. δύο γὰρ οὗτοι χρόνοι, εἷς μὲν καθ’ ὃν ἔστι καὶ ζῇ, ἕτερος δὲ καθ’ ὃν οὐκ ἔστιν ἀλλ’ ἔφθαρται· διόπερ ἐξ ἀνάγκης ὀφείλει κατὰ τὸν ἕτερον τούτων θνῄσκειν. ὅτε μὲν οὖν ἔστι καὶ ζῇ, οὐ θνῄσκει· ζῇ γὰρ δήπουθεν· θανὼν δὲ πάλιν οὐ θνῄσκει, ἐπεὶ δὶς ἔσται θνῄσκων, ὅπερ ἄτοπον. οὐ τοίνυν θνῄσκει Σωκράτης.