Empedocles and metempsychÙsis: The critique of Diogenes of Oenoanda

Empedocles and metempsychÙsis: The critique of Diogenes of Oenoanda Theories of transmigration require a distinction between body and soul, but there are at least two ways, broadly speaking, in which the relationship of body and soul to each other can be understood, and the viability or even crude plausibility of a theory of transmigration depends crucially on which of these views one holds. On one approach, the soul is causally connected to the body in such close interdependence that it cannot be securely identified when considered separately from the body. On the other, the soul has a much looser connection to the body; it will have causal interactions with the body, but these will not affect our ability to identify the soul when considered separately from the body. This is but one possible way to categorize body-soul relationships and its focus on continuous identifiability of the soul answers to one set of concerns which appears in the ancient tradition. It is clear from the outset which kind of relationship is more favourable to theories of reincarnation (though both sorts of relationship could in principle permit it). Another approach to categorization addresses the ontological nature of the body and the soul. The body is material, of course, but the soul may be either material or immaterial. The position one takes on various metaphysical questions will have some impact on the issue of how significant the nature of the soul is for its interactions with the body. At one extreme, to be sure, is the Stoic view that only a body can interact causally with a body, and hence that the soul must be bodily since it manifestly interacts causally with the body. And another extreme would be a theory, such as that hinted at in the Gorgias (523e), of souls with a completely developed individual moral standing, each recognizable as an individual subject to evaluation for its deeds, and yet ‘naked’ in the sense that they are regarded without attachments of either body or material possessions. These souls, the description of which is embedded in a myth of some sort, are to be thought of as having a more or less contingent relationship to the bodies which eventually

clothe them. Such a soul could readily be thought of as a mere visitor to the body or bodies it inhabits. Clearly, this looser relationship between body and soul is quite close to the Empedoclean notion of a bearer of identity “clothed in an alien robe of flesh” (fr. 126 DK) and just about as remote as possible from the Epicurean theory of body/soul relations. The contrast between these two kinds of theory (perhaps we can call them the ‘essential interaction’ model and the ‘contingent visitor’ model) comes out clearly in consideration of the attack on Empedoclean transmigration theory found in the inscription erected by Diogenes of Oenoanda hundreds of years after the death of the brilliant Sicilian. But before we look more closely at that, some background is needed. Lucretius, at the climax of book III of De rerum natura (line 830), concludes “nil igitur mors est ad nos”: “Death, then, is nothing to us” (tr. Smith 2001). This translation into Latin of the opening of Epicurus’ second Principal Doctrine often serves as a keystone of Epicureanism, and not unreasonably so. The announcement is, of course, a conclusion, marked by igitur, and the earlier parts of book III lay out the doctrines, arguments and claims which purport to support it. And they are numerous, depending ultimately on the demonstration that the soul is corporeal. The interdependence of those parts of our atomic structure which count as soul (both animus and anima) and those which count as body is fundamental. Body and soul share experiences, cause changes in each other, and rely completely on each other for their continued existence; the spatial separation of soul and body leads to the dispersal of the former and the decay of the latter. At line 670, though, a new line of argument is opened up, one which addresses the theories of pre-existence, reincarnation, and metempsychÙsis which had been advanced earlier in the ancient philosophical tradition. Names are not named here, but the targets of his attack almost certainly were meant to include Plato and Pythagoreans, and perhaps also Empedocles (who was already in Lucretius’ mind as a model for didactic verse and who had been criticized by name for his theory of elements in book I; see line 716). The arguments are, in summary, as follows: 1. We have no recollection of any existence before this embodied life. But we would have some recollection if the soul had pre-existed and entered the body at birth. And if the mind pre-exists and enters the body at birth but is so changed by the process that no memory of

previous actions survives, then this still counts as the destruction of the previous soul and the fresh creation of the current soul. So the present soul is dependent on the present body; hence it cannot exist before or after its present incarnation. 2. A pre-existent soul (on the contingent visitor model) would have enough ontological independence that we would not expect it to grow and change in interdependence with the body, as our soul manifestly does. Rather, we would expect it to remain isolated within the body as it maintained its psychological integrity in the new context. Indeed, the simple fact of interdependence during this life means that any untangling of soul and body is bound to entail the destruction of the soul. So the soul is not pre-existent, is created at birth and consequently must perish at death. 3. The spontaneous generation of maggots in dead flesh shows that some portions of soul remain behind in the body at death – so an escaping soul cannot be complete, and if it is incomplete it is no longer the same soul as it was when in the body. 4. The suggestion that such entities as maggots are enlivened by immigrant souls at birth is absurd. So is the notion that pre-existent souls might construct their own bodies to inhabit. 5. The correlation of psychological characteristics with the bodies of different species confirms the essential interdependence of mind and body. A transmigrating immortal soul or mind would either change its temperament between bodies (which is equivalent to destruction, says Lucretius; at the very least it would disrupt the continuity of identity needed for responsibility) or it would bring its psychology with it to the new body so that metempsychÙsis would lead to massive psychological mis-matches among the species. Even within one species (if human souls only migrate to human bodies) the lack of continuity of psychological traits shows that the same soul does not move from one body to another. 6. There is no reasonable account of how independently pre-existing souls could be allocated to new bodies. 7. Entities have natural places to exist in. Souls are naturally fit for their bodies. No change is possible in such natural fitness; but if it were, the least implausible change would be to a different part of the same body rather than to a completely distinct body. 8. The mortal and the immortal cannot be united as body and soul are. So soul too must be mortal.

9. An immortal object must be either a single atom (not a compound) or void – according to the principles of Epicurean physics. But no soul is either of these on Epicurean theory. These arguments are of varying character and value, some focussing on pre-existence, some on survival, and some on the process of transmigration. Many put particular emphasis on identity conditions for the soul, including those that rely on psychological continuities (e. g., 1., 2., 5.). Some (such as 8. and 9.) are virtually question-begging. Still, they are an appropriate basis for the conclusion that death is nothing to us, since they purport to show that, far from surviving to inhabit a different body or kind of body in a next life (something with which we would have many good reasons to be concerned) our soul and mind cannot be said to survive the death of the body at all. Lucretius’ reliance on the claim that significant qualitative change is tantamount to destruction of the previous entity may well be thought to employ an excessively demanding criterion for identity, but even this more vulnerable aspect of his argumentative strategy could be defended – though Lucretius does not do so here. Surely any Epicurean, including Epicurus, would have good reasons to argue against reincarnation theory in any of its forms; even theories which merely posit the survival of the soul would have to be rejected if they left room for any form of punishment or indeed any form of postmortem unpleasantness at all. And so Empedocles and his theory of transmigration must surely have been of particular interest to Epicureans. Yet in our surviving evidence he is remarkably seldom named as the target of such attacks – we have seen that Lucretius doesn’t name him in this connection. Silence on this point is surely the result of nothing more than the accident of our sources, especially since Empedocles is well known as a target of some Epicureans’ polemic. It is possible that some such attack on transmigration is a part of what Cicero alludes to in De natura deorum 1.93 when he refers to Epicurus, Metrodorus and Hermarchus speaking against Pythagoras, Plato and Empedocles. With regard to this passage, Francesca Longo Auricchio1 contends that the context (which does in fact suggest a more narrowly theological polemic) shows that at least part of the polemic focussed at least on anthropomorphic theology,2 and there are certainly other things for Epicureans to at1 2 Longo Auricchio 1988, 137. Compare Plutarch, De def. 420d = fr. 50 Longo Auricchio.

tack than reincarnation theories; D.L. 10.11 suggests that Pythagorean ideas about friendship could be a target, and Lucretius book I shows that Empedocles’ theory of matter was also worth refutation. Some of the polemic may have been quite diffuse, little more than verbal abuse (cf. Plutarch, Contra Epicuri beatitudinem 1086ef = fr. 237 Usener). Still, theories of transmigration are the most obvious thing shared among the three thinkers mentioned in Cicero and the emphasis here is on the audacity and indeed hypocrisy shown by Epicureans in criticizing more respectable philosophers (hence the faux outrage over Leontion’s criticism of Theophrastus at De natura deorum 1.93). So it is possible, at least, that Cicero is indicating awareness of attacks on reincarnation as proposed by Empedocles among others at the hands of the three earliest leaders of the Epicurean school. An important but obscure part of the anti-Empedoclean polemic undertaken by Epicureans is found in the treatise by Hermarchus entitled Against Empedocles, which Diogenes Laertius (10.25) tells us consisted of 22 books; of this lengthy work we have 8 fragments, according to Longo Auricchio (frs. 27 – 34 in her edition of 1988), including a substantial amount of paraphrase or quotation recovered from Porphyry’s De abstinentia (fr. 34). It is easy to see how the theory of social relations and justice sketched in the extract from Porphyry could be used to displace arguments for justice and respect for animals based on kinship with them (through transmigration of souls into animals), but in what we possess of Against Empedocles there is no recoverable critique of Empedocles’ theory of transmigration. Unless, that is, the fragment of the inscription erected by Diogenes of Oenoanda which is my particular interest here (fr. 42 Smith) is in fact derived from the work of Hermarchus. Longo Auricchio includes a portion of it as a testimonium, her fr. 52, but does so with considerable hesitation (174), despite the long tradition of rushing to judgement on this point, conveniently summarized in her commentary at 174 – 177. M.F. Smith3 is even more restrained. He regards Longo Auricchio’s caution as “a wise decision” and makes it clear that in his view there could be any number of sources for Diogenes’ attack on Empedocles’ doctrines. And given how little we know about Epicurean texts throughout the school’s history, that is certainly the case. Nevertheless, Diogenes’ inscription is of great interest to students of Empedocles, since it contains the only example (known to me at any Smith 1993, 493.

rate) of an explicit critique of Empedocles’ theory of transmigration (or metempsychÙsis – I mean to use the terms synonymously) and the bodysoul relationship from the Epicurean camp. Lucretius’ arguments may have been intended to catch Empedocles in their rather broad sweep, but that is not, of course, certain, given the generality of those arguments and the absence of any specific reference to him. The surviving relevant part of the inscription is not long, though we do have a pretty good sense of its place in the original context, which can be sketched briefly.4 Towards the end of fr. 34 Diogenes proposes to investigate how our life might become pleasant both in its stable states (katastÜmata) and in its actions. As to the former, the thing to note is that once disturbances are eliminated pleasure-producing elements replace them. These sources of upset are familiar: fear of the gods, fear of death, fear of pains, and excessive desires. Fear of the gods was dealt with first. Fr. 35 and fr. 36 seem to deal with this fear, though the condition of the inscription makes this a bit uncertain. Fr. 37, though, seems clearly to be about the fear of death. In its present condition the fragment opens with an account of the soul, its bodily nature, its interdependence with the body and vulnerability to its frailties. The stone breaks off with a re-assertion that the soul, by its presence or absence, is the ultimate determinant of life and death. Diogenes, quite sensibly, explains the nature of death before addressing the fear of it. In fr. 38 he continues to emphasize the relationship of soul to body. “One must learn that it too is a part … [of what?] For the soul cannot ever exist by itself, no matter how much nonsense Plato and the Stoics spout on the topic, nor can it move by itself – any more than the body can perceive once the soul has left it behind.” Fr. 39 is in very bad condition, and we can barely make out that Platonic views on immortality are being attacked. There is a clear mention of the eternal movement of the soul, followed by a few scrappy mentions of body/ soul relations. Then follows a vigorous apostrophe: “So, Plato, how will you get indestructibility? Or how will this [the soul] be indestructible in the generally understood sense?” He then attacks the Stoics, whose proneness to innovation is criticized. I assume this means their readiness to innovate on Plato, in whose camp they are otherwise regarded as being – echoes of an approach taken by Antiochus of Ascalon might be faintly heard here. For they decline to treat all souls as unqualifiedly indestructible. 4 I’ll be considering the inscription on the basis of Smith 1993 (the few changes in Smith 2003 do not have any effect on the sense that can be recovered).

Diogenes repeats the Stoic doctrine that fools’ souls are destroyed immediately when the soul/body compound is dissolved, whereas wise men’s souls last for a long time before they too are ultimately destroyed. There is another expostulation addressed to the readers: Do you see their manifest implausibility? They make this claim as though sages and non-sages didn’t possess the same kind of mortality, despite their intellectual differences. I wonder even more at their self-control: once the soul has the capability of existing without the body, even if we say it is for a split second, how … Exactly what sort of intellectual self-control is envisaged in this snide comment is not completely clear, but probably the idea is that once you allow the soul to survive death it takes heroic self-control to deny it true and complete immortality. Fragment 40, as reconstructed, brings in the Orphics and Pythagoras. “Let us not say that the soul hasn’t perished when it moves on [metabainousa], as the Orphics (and not just Pythagoras) seem to say in their madness.” Fr. 41 should probably be placed in this context too, since it connects Pythagoras and Empedocles (“not as the Empedocleans and Pythagoreans say, for …”). Fragment 42, unmistakably attacking Empedocles on this doctrine, follows. Three observations about Diogenes’ exposition here. First, it is in his usual vivid and exclamatory style. Second, Plato is not firmly associated with metempsychÙsis or transmigration, at least not in the bits we have. He is attacked for his views on the survival of the soul after death, indeed, for his views on immortality, while the Stoics are criticized for their intellectual sloppiness as well (needless innovations on Plato, inconsistent views about survivial of sages and fools). Of course, other themes of criticism could have been present in the full original text. Third, the ordering of the critiques is not chronological or school-based, since it seems to have gone from Plato to the Stoics, then back to Orphics and Pythagoreans and on to Empedocles. Most likely, then, the organizational principle was the doctrine under attack: survival, immortality, and transmigration were treated as distinct headings for the purposes of his polemic; indeed, we should imagine an ascending order of apparent absurdity as his organizing principle. Fragment 42 goes as follows (I reproduce and translate the text in Smith 1993, with an eye to the minor changes in Smith 2003 as well).5 5 Quite rightly, Smith 2003, 100 complains that I failed to update my translation in Inwood 2001 in light of Smith 1993.

Column I 1. – - – 9mpedojk/r 2. [peq· toºtym t]µm vi3. [kosov¸am pqos]´ka4. [bem 1j Puhacº]qou At this point, several lines are missing or too heavily damaged to be used. Column II 1. mor letaba¸meim [vg]2. s· t±r xuw±r 1j syl≤3. tym eQr sælata le4. t± tπ t± pq_ta diavha5. q/mai ja· 1p( %peiqom toO. 6. to ce¸meshai, ¶speq oq7. j 1qoOmtor aqt` timor7 8. “9m. p. edºjkeir, _ eQ l≥m 9. o. w. [m] d. º. m. amtai jah( 2au10. t. ±. r. [a].R x. uwa. · l´meim lg11. d. ≥. j. ·t. K. u. keir sºq. e.im 12. eQr f. []]ou. vºsim ja· toº13. tou w. ≤. qim. letav. ´qeim. 14. [a]q. t≤r, t¸ soi. dºmatai. B. Column III 1. let≤basir. _ 1m c±. [q t`] 2. letan. » wq. º. m\, dQ o[x tπ] 3. letaba¸meim aqta[ ?r ce¸]4. metai f]ou vºsim [di]5. ´wom, tπ. p÷m taq. a. wh. [¶]6. somtai. eQ d≥. lg[dal_r] 7. 5wousi l´me..im. %[meu] 8. sælator, t¸ l≤[kista] 9. 2aut` paq´wei[r pq≤]10. clata, l÷. kkom [d≥ 1]11. je¸mair, s. ºqy. [m aqt±r] 12. ja· letabib≤f. [ym eQr] 13. 6teqom 1n 2t´qou. [f`]14. om. ja· taOta po – - -

Column IV is so defective as to be unusable Column V 1. 2. 3. %]4. [leimom c]±. q "pk_r !5. vh≤qtour [Gm] t±r xu6. w±r jah( ´aut. ±r poi7. e?m ja· lµ eQr la8. jqπm 1mbake?m aqt±[r] 9. peq¸pkoum, Vma sou tπ 10. pam´swatom se. lmº11. teqom rp/qwe tπ xeOs. 12. la. _ C so· l´m, 9mpe13. dºjkeir, !pist¶sole[m] 14. t±r letab≤seir ta. [º]tar”] Column I Empedocles took his doctrine on these matters from Pythagoras6 … [about ten lines missing, that is, 2 12 times the amount of text preserved in this column] Column II … he says that the souls migrate from body to body after they are first destroyed and that this happens ad infinitum, as though someone won’t say to him, “Empedocles, so if souls can persist on their own and … not to drag them off into an animal nature and transfer them for this reason, what is the meaning of your Column III transmigration? For in the intervening time, in which they [the souls] experience transmigration, which punctuates animal nature, they will be utterly disrupted. But if they are in no way able to persist without a body, why do you bother yourself – or rather, bother them,7 dragging 6 7 Only the last three letters of the name are legible on the stone, but the restoration seems certain, esp. in light of fr. 129 DK and D.L. 8.56. See below for the significance of this apparently ironic suggestion of an unintended cruelty to the soul of such an animal.