Diagnosis and Inconsistency in the Axiochus

The Socrates of the dialogue Axiochus seems to advance incompatible arguments in his attempt to cure Axiochus of his fear of death. Is this incompatibility a foreseen and accepted consequence of the author’s therapeutic strategy? This paper argues that it is rather an intended and functional inconsistency: it serves to stimulate critical thinking in order to anchor philosophical conviction more deeply in the reader’s soul. The paper musters support for this reading by drawing attention to the different levels of inconsistency in the dialogue; the multiple ways in which the text thematizes inconsistency; the importance of exercising judgement in the text; and the motivating concern of superficial persuasion.


Dialogue, Genre, and Aim
The Axiochus, transmitted as part of the Platonic corpus, dates from anywhere between the second half of the third century BC to the first AD, but most likely from the first half of the first century BC.1 Its plot is simple: Socrates gets called to Axiochus' deathbed to cure him of his fear of death. Socrates produces a 2 Joosse The International Journal of the Platonic Tradition 16 (2021) 1-18 number of arguments. At the end of the dialogue, Axiochus looks forward to dying. One of the major interpretive problems of this text is that it seems to feature a Socrates who uses a set of jointly inconsistent arguments. And why would a Platonic author do that? I argue that the inconsistency is intentional and functional. It is part of a larger theme of inconsistency in the text. The text not only contains multiple instances of inconsistency but focuses our attention on it. In this way, I argue, the author aims for his readers to reason for themselves about the status of the soul in order that they may move beyond superficial persuasion and firmly internalize the Platonic position.
Before we look at the three levels of inconsistency to be observed in the text we need to note two generic characteristics. First, this is a Socratic dialogue. It does its best to signal this via frequent allusions to Socratic dialogues and specifically Plato's Socratic dialogues (see Feddersen 1895, 22-29;Chevalier 1914, 67-71;Joyal 2005, 98-104;Männlein-Robert 2012, 15-16 and 31-37;Erler 2012). It is also a therapeutic text, and close to the so-called genre of the consolatio. This therapeutic nature fits its origin in a time when philosophy was primarily seen as a kind of therapy (see e.g. Hershbell 1981, 19-20;O'Keefe 2006).2

Diagnosing Inconsistency: Level A
Let us turn now to inconsistency as a theme of the text. It is important I think to recognize that we in fact have at least three levels of inconsistency, presented in the following table for ease of reference: 3 Diagnosis and Inconsistency in the Axiochus The International Journal of the Platonic Tradition 16 (2021) 1-18 a diagnosis of Axiochus' condition as being not merely one of fear but also one of inconsistency (364b6-c1): ἀνιαρῶς τε φέρει τὴν τελευτήν, καίτοι γε τὸν πρόσθεν χρόνον διαχλευάζων τοὺς μορμολυττομένους τὸν θάνατον καὶ πρᾴως ἐπιτωθάζων.3 Wretchedly he endures his end, even though in times past he simply scoffed at those who were scared of death, and gently poked fun at them.
Where are your former boasts, and those perpetual praises of manly virtues, and that indescribable courage of yours? For like a timid athlete, though seeming brave in school exercises, you have failed in the actual contests.
tr. Hershbell 1981, mod. The inconsistency is between a past moment and the present but also one between words and action.4 Axiochus himself shows that he recognizes the inconsistency, in his response to Socrates (365c1-7): Ἀληθῆ ταῦτα, ὦ Σώκρατες, καὶ ὀρθῶς μοι φαίνῃ λέγων, ἀλλ' οὐκ οἶδ' ὅπως παρ' αὐτὸ τὸ δεινὸν γενομένῳ οἱ μὲν καρτεροὶ καὶ περιττοὶ λόγοι ὑπεκπνέουσιν λεληθότως καὶ ἀτιμάζονται, ἀντίσχει δὲ δέος τι ποικίλως περιαμύττον τὸν νοῦν εἰ στερήσομαι τοῦδε τοῦ φωτὸς καὶ τῶν ἀγαθῶν, ἀιδὴς δὲ καὶ ἄπυστος ὁποίποτε κείσομαι σηπόμενος, εἰς εὐλὰς καὶ κνώδαλα μεταβάλλων. tr. Hershbell 1981 This is a central passage for our purposes. Axiochus describes his own situation in terms of an inconsistency. He concedes to Socrates that what he says is true, in two ways (this I think is what the near-duplication ἀληθῆ ταῦτα … καὶ ὀρθῶς μοι φαίνῃ λέγων applies to): Socrates is right to say, in the part which I have not cited here, that his behaviour is inappropriate for an adult Athenian but he is also right in his diagnosis that Axiochus' past boasts do not fit with his present fears. Axiochus' statement shows that he still sees (some of) the truth of these past arguments, but they do not have the same persuasive force. His choice of words almost makes it seem like there is a physical component to this clash, which his present fear clearly wins: the arguments blow away (ὑπεκπνέουσιν) and no longer have the same value (ἀτιμάζονται), while his present fear persists (ἀντίσχει).5 This analysis in almost physical terms suggests an inconsistency between different kinds of factors: cognitively, Axiochus sees the truth of the arguments against fearing death, but in some other, non-cognitive way, the force of fear shoves that aside into insignificance. The continuation shows that this is not the full picture: Axiochus provides propositional content to the fear in terms of expected deprivation and putrefaction, which shows that the inconsistency plays out in the cognitive domain itself as well. Nevertheless the emphasis is on the mind-overwhelming aspect of αὐτὸ τὸ δεινόν, as Axiochus puts it, that has now become real for him. The καρτεροὶ λόγοι of which Axiochus speaks are clearly not strong enough (see the end of section 6 for the importance of this theme for the overall design of the dialogue).

Diagnosing Inconsistency: Level B
The second level of inconsistency that we find in the text is the reasoning behind Axiochus' present fear. Let us call this 'level B' inconsistency. Socrates' reply to the passage just cited is (365d1-5): But, Axiochus, because of your thoughtlessness, you uncritically connect sensation with absence of sensation; and you are doing and saying things contrary to yourself, not realizing that at one and the same time you lament the absence of sensation and are pained at decay and loss of pleasures, just as if by dying you entered into another life.
tr. Hershbell 1981 Contrary to level A, this inconsistency-the simultaneous commitment to the notion that we experience deprivation of the goods of life and even bodily degradation and to the idea that death implies absence of sensation-is not obvious, at least not to Axiochus. We should note, in fact, that Axiochus himself does not assert the belief that death involves the absence of sensation. He speaks of being deprived of the living light and of good things, as we just saw. On the other hand, Axiochus does not protest Socrates' ascription of this belief to him either. Perhaps we are to understand that the αὐχήματα of which we saw Socrates speak and the καρτεροὶ καὶ περιττοὶ λόγοι that Axiochus mentions include beliefs of this kind. Or perhaps this is a background belief that we are to understand Axiochus as actually entertaining. One further aspect to notice is that the inconsistent belief set at this level is itself involved in Axiochus' level A inconsistency: at that level, his past arguments conflict with his present fear, which has a cognitive component in the idea that he will be pained at his own decay. At level B, Socrates points to the inconsistency in this idea itself. Whether Socrates imputes this level B inconsistency to Axiochus correctly or not, it fits Socrates' therapeutic role that he is the one diagnosing it. Unlike the Socrates of Platonic dialogues, this Socrates does not apply extended questioning to bring the interlocutor himself to admit to inconsistency. He is more like a doctor who looks at symptoms, draws his own conclusion and communicates it to his patient.
Two further features of this passage deserve comment. First, a very similar passage occurs later on in the text that starts with almost the same words: Συνάπτεις γάρ, ὦ Ἀξίοχε, ἀνεπιλογίστως, τῇ στερήσει τῶν ἀγαθῶν ἀντεισάγων κακῶν αἴσθησιν (369e3-4, continuing to 370a6). This second diagnosis of level B inconsistency comes after a passage in which Axiochus complains that Socrates' arguments are just fair words and do not touch the soul. This order of the text is significant, as we will see below (see the end of section 5). Second, 6 Joosse The International Journal of the Platonic Tradition 16 (2021) 1-18 365d1-5 is remarkably emphatic about Axiochus' inconsistency, especially in its repetitions (the basic message being delivered three times with συνάπτεις … αἴσθησιν, σεαυτῷ … λέγεις, ἐπιλογιζόμενος … ἡδέων; the repetitions of ἀν-; καί … καί in d3), rhyme (note -σία/τία) and vocabulary (συνάπτεις; ὐπεναντία; repeated (ἀν)επιλογι-6). Both features serve to highlight Axiochus' inconsistency as an inconsistency-a stratagem to which we will return as well (see section 6).

4
Diagnosing Inconsistency: Level C Socrates' diagnosis leads into argumentative treatment almost immediately, which leads us to the third level of inconsistency, that of Socrates' own argumentation. Axiochus' treatment comes first in the form of a compressed Epicurean line of argument, combining the argument from insensibility, the argument from absence and the symmetry argument into a mere five lines (365d6-e2).7 A couple of lines later, however, Axiochus is offered a Platonic account of what happens after death, involving the soul's return to its home and the identification that we are immortal soul (365e3-366b1). Later in the text both kinds of argumentation recur. The tension between them is clear: the Epicurean arguments depend on the idea that we dissolve at the moment of death, whereas the Platonic arguments argue for persistence after death. This inconsistency, which we may call level C inconsistency, is not, it seems, diagnosed by anyone in the text. It is worth pointing to another level C inconsistency within the Platonic account itself. According to the initial statement of the Platonic account that I have just referred to, the soul is imprisoned in the body, will be released at death and will rejoin its kindred elements. This first account is prima facie consistent with a second, referred to in the text as οὐράνιος λόγος (372a11): cultural progress and our human capacity to understand the workings of the universe are so great that they could not have occurred if the human mind does not partake of divinity and hence is immortal (370b2-d6). A third Platonic argument, however, seems to be inconsistent with the earlier two (O'Keefe 2006, 392-3). It is a post-mortem myth of the kind we also find in Plato's Gorgias and Phaedo, 7 Diagnosis and Inconsistency in the Axiochus The International Journal of the Platonic Tradition 16 (2021) 1-18 in which souls are judged according to their quality of life and either suffer for their failures or enjoy pure pleasures afterwards (371a1-372a3). The problem here is that according to the first two arguments, human souls, and that would seem to include bad humans' souls, rejoin the heavenly spheres upon death as a matter of their nature, of their physical makeup, not as a matter of their embodied behaviour.
There is also another problem with the last Platonic argument. Socrates may seem happy and confident enough to promise Axiochus that he belongs to the latter group. But it is not obvious that he does (cf. O'Keefe 2006, 398-9). Nor does this prospect seem a very firm basis for driving out Axiochus' initial fear of death, now perhaps transformed into a fear of a bad post-mortem fate.
As I noted, Socrates' level C inconsistency is not commented upon by any of the speakers of the dialogue and remains in that sense undiagnosed. But it would certainly count as a disease of the soul for any Hellenistic philosopher and indeed for Plato's Socrates. Nor is it easy to dissociate Socrates himself from the inconsistency of his arguments (pace Benitez 2019, 19;Long 2019, 167). It is true that Socrates explicitly attributes large parts of his argumentation to a speech by Prodicus. This applies to the long description of the evils of life (366c1-369a2) as well as to the Epicurean argument of absence as formulated at 369b5-7. He attributes the final myth to Gobryas, ἀνὴρ μάγος (371a2). Socrates also claims that he is ζητητικός and disavows knowledge (366b5-8; cf. Beghini 2020, 255). However, attributions and disavowal can plausibly be taken (even by Axiochus) as a conversational ploy or a reminiscence of the many similar moves in Plato's (and Xenophon's) Socratic texts (by the reader). Note for instance that Socrates mixes elements from his own and Axiochus' experience into what purportedly is a report of Prodicus' speech (368d8-369a2). Moreover, Socrates himself claims that Prodicus' speech has made him long to die (ἐξ ἐκείνου θανατᾷ μου ἡ ψυχή, 366c8), although this could be read ironically. More importantly, however, even if Socrates himself does not subscribe to (a subset of) the arguments he puts forward, the text presents him as attempting to persuade someone to subscribe to them and thus as trying to get someone to hold inconsistent beliefs. Socrates' level C inconsistency is problematic, therefore.

Interpreting the Inconsistency: The Therapeutic Interpretation
It is not a law of nature that texts cannot be inconsistent. It happens: an author may be confused, lack the necessary overview of his own work, lack the necessary time to revise his text. This may be the case with the Axiochus too. The principle of charitable interpretation, however, requires us to have more patience with the text and to try to find an explanation for Socrates' inconsistency that makes the text stronger.
The dominant interpretation today reminds us that this text is therapeutic. And, it is pointed out, the accumulation of arguments, even if that results in mutually contradictory arguments, is a common phenomenon in therapeutic texts, and consolations in particular.8 As O'Keefe writes, "Socrates is depicted as throwing arguments against a wall, one after another, to see which of them will stick" (2006,397). Moreover if we look at closely related texts, like the Plutarchan Consolatio ad Apollonium and Cicero's Tusculan Disputations book 1, we see that this reading works fine there. They too feature arguments of Epicurean origin side by side with Platonic ones and many others as well. Indeed, Cicero addresses this strategy explicitly in Tusculan Disputations book 3, when he lists different consolatory approaches and notes that some people prefer to combine them.9 According to this therapeutic reading, therefore, inconsistency is a foreseen and tolerated feature of consolations, here no less or more than elsewhere. Socrates' aim is to drive out fear, not to communicate a consistent view.10 The therapeutic reading faces three objections in particular. First, it would seem to accept a further inconsistency, an intertextual one. As noted at the beginning of this paper, the text alludes to many Platonic texts, both in terms of doctrine (e.g. the immortality of the soul) but particularly in terms of characterization: the Socrates that is evoked is recognizably the Platonic Socrates. This strategy creates expectations on the part of readers who know their Plato (Erler 2012, 101-102). Among these expectations is that the character named Socrates will value consistency and will be quick to detect inconsistency. To have this character argue an argumentatively inconsistent case produces tensions between the text and its intertexts.
Second, as Benitez 2019 has pointed out, therapeutic texts seem to be much less inconsistent than the therapeutic reading requires. Texts that also argue against the fear of death or seek to assuage grief at the passing away of a loved 8 D.S. Hutchinson's embrace of this approach in Cooper's collected volume of the Platonic corpus has done much for its becoming mainstream (1997: 1734-1735). Cf. Erler's concept of Argumentationshäufung as a therapeutic strategy (see for instance his 2018). 9 Tusc. 3.76: "Sunt etiam qui haec omnia genera consolandi colligant (alius enim alio modo movetur), ut fere nos in Consolatione omnia in consolationem unam coniecimus." 10 O'Keefe marks a contrast to the Sophist's concern (230b4-e4) with the purification of the soul: "Psychic therapy in the Axiochus aims at relief from pain, not the removal of the pollution of unjustified beliefs" (2006, 401). The Axiochus is markedly different from these and other therapeutic texts. It treats the alternatives on a conjunctive basis: the soul perishes at death and is immortal (cf. Furley 1986, 79).12 Third, the Axiochus presents its inconsistent arguments in passages that do nothing to soften the transition. Here are the two key moments of transition between the sets of Epicurean and Platonic arguments (365e2-6 and 370a7-b3): σὺ γὰρ οὐκ ἔσῃ περὶ ὃν ἔσται. πάντα τοιγαροῦν τὸν τοιόνδε φλύαρον ἀποσκέδασαι, τοῦτο ἐννοήσας, ὅτι τῆς συγκρίσεως ἅπαξ διαλυθείσης καὶ τῆς 11 Tusc. 1.82: "Video te alte spectare et uelle in caelum migrare. Spero fore ut contingat id nobis. Sed fac, ut isti uolunt, animos non remanere post mortem […] mali uero quid adfert ista sententia? Fac enim sic animum interire ut corpus: num igitur aliquis dolor aut omnino post mortem sensus in corpore est?" (Tr. Douglas 1985: "I see that you are setting your sights high and wanting to make the move to heaven. I hope that may be our lot. But assume, as those others claim, that souls do not survive after death. […] What evil does that view imply? For assume that the soul perishes in the same way as the body-well there isn't any pain or sensation at all in the body after death, is there?"). 12 An anonymous reader suggests that the inconsistencies in the Axiochus may be sufficiently explained by the structure of the Apology or the integration of different views in the early Academy. The point in the main text shows that this fails to account for the dialogue's specific features. It advances an inconsistent conjunction of arguments, not a consideration of alternative possibilities. On some possible relations between the Axiochus and Academic exegesis of Plato, see Beghini 2021.

Joosse
The International Journal of the Platonic Tradition 16 (2021) 1-18 ψυχῆς εἰς τὸν οἰκεῖον ἱδρυθείσης τόπον, τὸ ὑπολειφθὲν σῶμα, γεῶδες ὂν καὶ ἄλογον, οὐκ ἔστιν ὁ ἄνθρωπος. ἡμεῖς μὲν γάρ ἐσμεν ψυχή, ζῷον ἀθάνατον [.] For you, whom it would concern, will not exist. Away, then, with all this nonsense and realize this: that once the union of body and soul is dissolved and the soul has been established in its proper place, the corpse which remains, being earthly and irrational, is not the human person. For we are soul, an immortal living being.
tr. Hershbell 1981 νῦν δὲ περιτρέπεις σεαυτόν δειματούμενος στερήσεσθαι τῆς ψυχῆς, τῇ δὲ στερήσει περιτιθεῖς ψυχήν, καὶ ταρβεῖς μὲν τὸ μὴ αἰσθήσεσθαι, καταλήψεσθαι δὲ οἴει τὴν οὐκ ἐσομένην αἴσθησιν αἰσθήσει. πρὸς τῷ πολλοὺς καὶ καλοὺς εἶναι λόγους περὶ τῆς ἀθανασίας τῆς ψυχῆς· οὐ γὰρ ἂν θνητή γε φύσις εἰς τοσόνδ᾽ ὕψος διήρατο μεγεθουργίας, ὥστε [κτλ.] But now you upset yourself; fearing to be deprived of the soul, you confer on this deprivation a soul of its own. And you dread the absence of sensation, but you think that you will comprehend the future absence of sensation with sensation. In addition to the many and beautiful discourses on the immortality of the soul, a mortal nature would certainly not have arisen to such a height of great deeds that … [a catalogue follows of human achievements and feats of understanding].
tr. Hershbell 1981, mod. There is no attempt at smoothing the transitions, no buffer, no attempt to distract the reader from the very different message of the two sets of arguments. (To be clear, "Away, then, with all this nonsense" in 365e2-3, given the context, refers to Axiochus' description of his fear, not to Socrates' preceding Epicurean argument.) We find nothing so crass in other therapeutic texts. It is a marker of the harshness of these transitions that some scholars have seen them as fault lines of an ill-executed compilation of different draft sections of the text. To restore the text we should, they think, cut up the text and rearrange its parts, such that the Epicurean arguments come first and the Platonic ones last. Arranged properly, the two halves would be separated by Axiochus' expression of dissatisfaction with what Socrates has to offer (369d1-e2): Σὺ μὲν ἐκ τῆς ἐπιπολαζούσης τὰ νῦν λεσχηνείας τὰ σοφὰ ταῦτα προῄρηκας· […] τὰ δὲ παθήματα σοφισμάτων οὐκ ἀνέχεται, μόνοις δὲ ἀρκεῖται τοῖς δυναμένοις καθικέσθαι τῆς ψυχῆς. These fine sayings of yours are part of the current chatter of the times.
[…] Sufferings are not content with clever arguments, but are satisfied only with those things able to touch the soul.
tr. Hershbell 1981 The result would be a neat difference in status: the Epicurean arguments are rejected and the Platonic ones accepted. According to Immisch, who first proposed this approach (1896, 31-57), his restoration clearly shows that the author has polemical aims: the text is written against Epicurean therapeutics in favour of a Platonic version. If this is true, neither Socrates nor the text suffer from any level C inconsistency. In his recent edition and commentary of the text, Andrea Beghini (2020) revives this hypothesis of a serious mix-up of sections. The text as we have it, he argues, is the result of an editor's stitching together in an unfortunate way the separate sheets of papyrus that was the final version of the work of the author, who himself left the text unfinished and never got around to combining the component parts of his dialogue (2020, 51-67, esp. 57-59). However, this proposal has difficulty explaining why there are two very similar passages (as signaled above, section 3, the one beginning at 365d1, the other at 369e3-4) where we would expect only one. Immisch attempts to explain this duplication by positing two draft versions of this passage that have inadvertently both been included in the complete text (1896,(40)(41)(42)(43). Beghini too is open about this and other problems. He thinks the repetition can be explained as the author's having used similar wording in passages intended for different parts of the dialogue and located on different sheets, a similarity not noticed or corrected in the compilation (2020, 64-67). I do not wish to deny that such a cut and paste reconstruction is possible. However, Immisch and Beghini's proposals show that textual rearrangement offers no easy solution for the inconsistencies and harsh transitions of the text, and that we are forced into a number of ad hoc explanations. be explained as a parodic flourish" (32). He finds light-heartedness and satire in the text (31-32) and claims that "Socrates alleviates not fear, but seriousness" (32). In this proposal I think we should embrace the idea that the Axiochus is to some degree about the arguments it contains. The laughable elements Benitez detects, however, may well concern our 21st-century expectations more than ancient ones.13 The dialogue, I think, no less than the conversation it contains, seriously engages with a range of therapeutic arguments. Benitez himself recognizes that there is a serious point to what he sees as parody. As he sees it, the parody offers "not so much a cure for grief or fear, but an intellectual diversion from it" (32), consisting in "philosophical consideration" (33). The latter is right, I think, but we should specify: philosophical consideration of the very arguments proffered in the dialogue, not as a diversion but as a way of providing lasting cures of grief and fear. I suggest, then, that we view the text's level C inconsistency as functional, but not as parody. Rather, together with the level A and B inconsistencies in the text, it serves to thematize inconsistency as a psychic problem, and is presented as something to be diagnosed and overcome by Axiochus and the reader. As Cleinias and Axiochus were already able to diagnose his level A inconsistency and as Socrates quickly brought Axiochus' level B inconsistency to the surface, the reader is intended to diagnose the level C inconsistency. Like the other cases, the inconsistency in Socrates' argumentation is put there by the author on purpose, not as a tolerable side-effect of a therapeutic strategy, but as an instance of potential psychic disease that is meant to be diagnosed, and to be diagnosed by the reader. Making the inconsistency explicit in their own minds will force readers, who are certainly meant to engage seriously with these arguments, to take up a position themselves.
In the following I present three further considerations that support this reading, in addition to the text's stacked inconsistencies: its verbal highlighting of inconsistency; textual elements that point to the need for further reflection on the argumentation; and the fact that we can identify a clear motivation for this strategy that is also touched upon in the text itself, i.e. the attempt to go beyond superficial persuasion.
First, the text does not just happen to exhibit inconsistent argumentation. It positively draws our attention to it. It does so in three ways, which we have 13 For instance, Benitez asserts that "[t]here is no gravity or solemnity in the Axiochus"; and interprets Axiochus' sobbing and hand-wringing even though he has recovered from his bout of panic (365a3-5) as "affectations" (31). The argument about the evils of life is "more of a light-hearted whinge than a dirge", while the reference to Epicurean argumentation as φλυαρολογία is another indication that the text is a parody (32, omitting that this reference, a qualification of Socrates' words by Axiochus, covers Platonic material).

Diagnosis and Inconsistency in the Axiochus
The International Journal of the Platonic Tradition 16 (2021) 1-18 already briefly looked at. There is the very close proximity of the contradictory Epicurean and Platonic arguments. It is as if the reader needs to see this.
We have also noted the near-duplication of 365d1-5 and 369e3-370a6, with the marked beginning Συνάπτεις γάρ, ὦ Ἀξίοχε, [παρὰ τὴν ἀνεπιστασίαν in 365d1] ἀνεπιλογίστως. Either this is indeed a major authorial or editorial oversight, or this passage bluntly insists on Axiochus' inconsistency and lack of progress. A further means of highlighting the theme of inconsistency is in Socrates' choice of words. For instance, in 370a7-b3 (cited above, section 5), Socrates accuses Axiochus of self-contradiction (περιτρέπεις σεαυτόν, translated as "you upset yourself" by Hershbell; alternatively: "you refute yourself") and in the next breath he contradicts himself: "you think that you will comprehend the future absence of sensation with sensation. In addition to the many and beautiful discourses on the immortality of the soul [etc.]". Similarly, we also saw how 365d1-5 (and 369e3-370a6) contains a number of terms that in their repetition, rhyme and rhythm force themselves onto the reader as markers of inconsistency. We should note at this point, however, that some of these emphatic markers are Epicurean terms of art (Immisch 1896, 25-26; Männlein-Robert 2012, 65 n. 23).14 It is worth pausing to consider their role. We are in Epicurean territory anyway in these passages, with terms like ἀναισθησία, but ἀνεπιλογίστως (365d2, 369e3) and οὐκ ἐπιλογιζόμενος (365d3) are terms with a particular function in Epicurean epistemology. As Schofield has convincingly shown, the term refers to a cognitive operation in which multiple phenomena are taken into account together and related to each other in a critical way; as he puts it more briefly: "epilogismos was a comparative form of appraisal" (1996,226).15 A fair number of Epicurean uses occur in the context of ethics and therapy, although I should note that the semantic spectrum of the word group is broader. In his On , Philodemus speaks of people who are informed of their disease but do not realize the serious consequences of it, either in general or ἐπιλογιστικῶς (10-11). They therefore act carelessly, until someone puts these consequences before their eyes, which makes them turn to their therapy. Diogenes of Oenoanda (fr. 44 Smith 1997) says that the superiority of psychic affections over bodily ones is δυσεπιλόγιστος for most people. Someone in bodily pain will rate this pain higher than psychic absence of pain, he observes, adding that what is present is more persuasive 14 The occurrence of ἐπιλογιῇ in 365b1 makes Beghini (2020, 237-238) think no special invocation of Epicurean vocabulary is at play, but in fact that occurrence too fits Epicurean usage and should I think be seen as preparing the accumulation here. 15 See Tsouna 2007, 55-57 for Philodemus' usage.

Joosse
The International Journal of the Platonic Tradition 16 (2021) 1-18 than what is absent.16 As these two cases illustrate, to restore the ability to set side by side present experiences and experiences that do not press themselves on someone as urgently seems to be the very goal of Epicurean therapy. In that sense Socrates' use of epilogismos terminology in this Epicurean context is not unexpected. Axiochus' present fear is stronger than his past arguments and so he needs epilogismos to overcome his level A no less than his level B inconsistency. Socrates' emphatic usage of these terms, however, goes beyond expectation. In having his Socrates use this terminology in such an obtrusive way, the author appeals to a key piece of Epicurean epistemology in order to flag the importance of comparative reasoning for Axiochus and for the reader. It allows the author explicitly to thematize the rational process of connecting different aspects, views and arguments, in the course of pointing out someone's failure to do so. Any comparative reasoning worth the name will diagnose Socrates' level C inconsistency; that is, I argue, exactly what the reader is expected to do. A further consideration that supports the idea that the reader is meant to diagnose Socrates' inconsistency is that the dialogue suggests that the conversation is not over when it ends (noted by Benitez 2019, 33). When Axiochus has pronounced his love of death, he announces his intention to review the conversation and invites Socrates back (372a13-15): νυνὶ δὲ ἠρέμα κατ' ἐμαυτὸν ἀναριθμήσομαι τὰ λεχθέντα. ἐκ μεσημβρίας δὲ παρέσῃ μοι, ὦ Σώκρατες.-Ποιήσω ὡς λέγεις.
And now quietly, by myself, I'll go over what has been said. But after midday, be with me, Socrates.-I will do as you ask.
tr. Hershbell 1981 16 Both passages are discussed by Schofield 1996, 228-229 (fr. 44 Smith 1997 is his 38). Clearly Socrates is committed to a position, but he also explicitly calls on Axiochus to exercise his own judgement about the things he has said. Narrowly conceived, the scope of this remark is Gobryas' myth. But in combination with the strong focus on language of judgement and inference earlier in the text this must be read as an appeal to readers to consider the arguments together, make up their own mind and search for the truth. This is not to deny that the reader is nudged into a Platonic direction, but the way towards it is through an explicit call for reflection.
A final consideration is the motivation behind this strategy of inconsistency. If we do read the dialogue as calling for readers to offer their own judgement, we are not doing much out of the ordinary if one of Plato's dialogues were concerned. Platonic scholarship has gotten used to taking the dialogue form seriously; a common view of the very choice to write dialogues is that an author means to involve the reader's response (see the now classic exposition of Frede 1992). The case is no different for this pseudo-Platonic dialogue. And in this case there is a specific concern that motivates the attempt to activate the reader's judgement: the danger of superficial persuasion.
In the text itself, it is Axiochus, ironically, who wishes to receive nonsuperficial therapy. In 369d1-e2 (cited above, section 5) he qualifies Socrates' attempts at driving out his fear as part of "current superficial chitchat" and requests instead something that is able "to reach through to the soul" (my trans.). I call Axiochus' wish ironic because he himself seems to be persuaded only superficially at the end of the dialogue, having become swayed to a (Platonic?) desire to die in an implausibly short span of text. The reader should once more be on his guard, particularly when realizing that Axiochus has a history of being persuaded by arguments that do not hold out against incoming fear. As we saw in the texts that describe Axiochus' level A inconsistency (364b6-c1, 365a6-b1 and 365c1-7, cited above, section 2) the arguments Axiochus himself was once fond of hurling at those who fear death are "tough and abundant" (365c3), but that did not make them stick when fear showed up, "tearing up the mind in all kinds of ways" (c4-5, my trans.). As we noted above, Axiochus describes the force of fear in almost physical terms. In this self-diagnosis of his level A inconsistency, then, Axiochus shows the need for true therapy to consist in deep persuasion, or else emotion will tear up and blow away philosophical arguments. This motivates the dialogue's strategy of inconsistency, which stimulates the reader's own thinking and so helps anchor the truth of Platonism deeply into their soul. The Axiochus is not alone in its philosophico-literary context to be concerned with superficial persuasion. Compare the following exchange in the first book of Cicero's Tusculan Disputations, when the interlocutor expresses his wish to be persuaded of the view that the soul returns to its home after death. The main speaker ('Cicero') replies (Tusc. 1.24): Quid tibi ergo opera nostra opus est? Num eloquentia Platonem superare possumus? Euolue diligenter eius eum librum qui est de animo: amplius quod desideres nihil erit.-Feci mehercule et quidem saepius, sed nescio quo modo, dum lego, adsentior, cum posui librum et mecum ipse de inmortalitate animorum coepi cogitare, adsensio omnis illa elabitur.
Why do you need my help then? Surely I can't outdo Plato in eloquence. Read carefully that book of his on the soul: you will want nothing more.-I have done that, many times too. But somehow while I am reading, I am convinced, but when I put the book down and begin to think for myself about the immortality of souls, I lose all that conviction. tr. Douglas 1985 This time it is even Plato's work itself that is found wanting, a touch of emulation on Cicero's part with which I see no correspondence in the Axiochus.18 A difference between the two texts is also that the interlocutor of the Tusculans is not dying and sees his assent get away as a result of reflection rather than emotions. Nevertheless this passage points to a concern shared with the Axiochus, that reading alone may not cut it.19

Conclusion
Few philosophical documents from antiquity seem so clearly self-contradictory as the pseudo-Platonic Axiochus. If the case I have made here stands, its contradictions serve the dialogue's ultimate aim of producing deep persuasion in 18 There is a rich literature on Cicero's relation to Plato; Dörrie 1987, 483-543

Diagnosis and Inconsistency in the Axiochus
The International Journal of the Platonic Tradition 16 (2021) 1-18 its reader. Against a background of concern with superficial persuasion which it shares with other texts from the period, the Axiochus' strategy can be understood as an attempt to reach the souls of its readers by making them do the philosophical work themselves, in hopes that, when they do so, they work towards rather than away from Platonic views. The Axiochus was written in (or not much before) a period in which a less skeptical Platonism comes to the intellectual forefront again. Among the defining characteristics of this renewed Platonism is a concern with systematizing Plato's philosophy. It is part of the significance of the Axiochus that it shows how Plato's emphasis on the value of discovering truth for oneself continued to be recognized and to motivate philosophers' literary choices. The dialogue also stands as a fascinating document of one Platonist's attempt to confront the popularity of Epicurean therapeutics and to harness it to serve a Platonist agenda.