BMCR 2005.05.30

Xénophon Mémorables. Tome 1. Introduction générale, Livre I. Collection Budé

, , , Mémorables. Collection des universités de France. Série grecque ; 399, 467-468. Paris: Les Belles lettres, 2000-. v. ; 20 cm.. ISBN 2251004823. €54.00.

In 1948, G.C. Field wrote that, “few scholars show themselves at their best when dealing with Xenophon” (140). While this is still true of some, a growing number of excellent publications have been changing that, and this book is one of them. This edition of the first book of Memorabilia is an essential reference work for anyone working on Xenophon, Socrates, or Plato. The text, produced by Michele Bandini, contains two apparatus critici, one containing variant readings, and a second indicating references to the text found in later authors. This second apparatus reminds us that whatever we may think of Xenophon, ancient readers and writers regarded him very highly and quoted him frequently. Michele Bandini is also responsible for the introduction to the history of the text. There are extensive notes on the text at the back of the book (consisting of 172 pages of small type), and shorter ones beneath the translation. These notes are mines of information about the literary, philological and philosophic issues raised by the text and include generous references to the literature on them.

In this review I will focus on the 250 page introduction in which Louis-André Dorion (D) takes a critical look at the treatment of Xenophon’s Socrates in modern scholarship, and offers some direction for the future. This introduction, which is essentially a monograph on Xenophon’s Socratica, will be appreciated by those who know this field, both for its thorough and painstaking documentation of the literature on Xenophon’s Socratica and for its original contributions. And it is even more valuable to those who have put off getting to know the “other” Socratic writer. They will get a fair, judicious and generously documented review of the important questions concerning Xenophon’s Socrates. I will summarize D’s main arguments, and take a closer look at his account of Xenophon’s attitude towards the elenchos.

The introduction is divided into six unequal sections. The first three sections, Réception des Memorabiles, Bilan des Critiques, and Réhabilitation de Xénophon, are devoted to a critical review of scholarship in Xenophon’s Socratic writings. In Réception, D takes us on the roller-coaster ride that Xenophon’s reputation has taken in recent centuries, from Brucker’s 1742-4 Historia critica philosophiae, which argued for the reliability of Xenophon’s Socratic writings, to L. Robin’s devastating 1910 critique, and on to the contemporary modest revival of interest. From this review emerge ten arguments against the accuracy and value of Xenophon’s Socratic writings, which D systematically discusses in Bilan des Critiques.

He dismisses the argument that Xenophon did not spend much time with Socrates, pointing out that he may have had up to ten years with him, and that some students can understand a teacher well after only a few months or less. He disposes of the charge that Xenophon was not a genuine disciple, a charge based largely on the fact that Plato never mentions him, by pointing out that Plato rarely mentions rival Socratic writers at all. As he argues, Plato only mentions them if their presence or absence from the trial and execution of Socrates was an issue. Xenophon, who was far off in Asia Minor at the time, needed no mention. D does not credit the argument that Xenophon’s works should be privileged on the grounds that he was an historian. He rejects the hypothesis that Xenophon took notes of his conversations with Socrates, and argues, rightly, that the lack of notes is no reason to dismiss his writings any more than we dismiss Plato’s because of a lack of notes.

D devotes special attention to Xenophon’s claim of autopsy. He is uncomfortable both with the naïve view that Xenophon was present everywhere he says he was and also with the overly skeptical view that insists he wasn’t, and that this lack of autopsy is a mark against him. He criticizes Vlastos for failing to note Sandbach’s distinction between Symposium (and D adds Economicus) where Xenophon’s claims for autopsy are highly problematic, and Memorabilia, where they are not (even if there are problems of anachronism). But he does not insist on the truth of these claims either; rather he treats them as literary devices which aim to lend authority to the account, and compares them (unfavorably) to the more elegant methods used by Plato to imply the accuracy of his dialogues. In the end, D points out that the question of autopsy is not crucial for our evaluation of Xenophon’s accuracy since his presence would not guarantee the accuracy of his accounts any more than his absence would guarantee their inaccuracy. And then, after disembarrassing us of these criteria of judgment, he proposes the reasonable view that Xenophon’s Socratic discussions are neither fiction nor history, but rather halfway houses with elements of each.

D brings out the weaknesses in the argument that Xenophon is unreliable because he plagiarized, pointing out that if he plagiarized from good sources, he may have got it right, and that it is vain to speculate about plagiarism from works which no longer exist. He argues similarly that Xenophon’s apologetic aims are no reason to discount the value of his testimony, and notes that Plato also has apologetic aims which he pursues more indirectly and more coherently than Xenophon. But it is precisely the incoherence of Xenophon’s apologetics, says D, which permits us to glimpse in his writings the outline of another Socrates whom both Plato and Xenophon have repressed. This is an important point, even if, as I believe, Xenophon’s apologetics deserve a little more consideration before we dismiss them as incoherent.

More serious is the charge that Xenophon placed his own mediocre ideas in the mouth of Socrates, just as Plato placed his own superlative ideas in it. The suspicions are based on two points: that many of Xenophon’s other characters say things similar to those said by Socrates, and that many of the things attributed to Socrates seem inappropriate to him, such as his discussions of horseback-riding, of military tactics, of farming techniques, and other such “Xenophontic” themes. Although acknowledging the possibility that these ideas may have had their origin in conversations with Socrates, D cannot shake the impression that there is some truth in this charge, and there may well be. But it is perhaps significant that the main argument he finds to support these suspicions is the evidence of Plato:

c’est à la fois la vie et l’oeuvre de Xénophon qui témoignent d’un intérêt soutenu pour ces thèmes, alors que, dans le cas de Socrate, si l’on fait abstraction du témoignage de Xénophon, on ne trouve pas, dans les autres écrits socratiques, le moindre indice du plus petit intérêt pour ces questions. (LXXIV)

The “other Socratic writings,” of course, are mostly the Platonic dialogues. And if Plato’s view of Socrates is idiosyncratic, it would be unfair to judge Xenophon by this standard. The subjects that Xenophon’s Socrates discusses were common topics for the ancient Greeks, and it is perfectly likely that any given person, Socrates for example, would have discussed them. Although the Platonic Socrates does not discuss these subjects at great length, he does say that a philosopher is in principle interested in everything (Republic 475b-c), and he does speak about common things like dogs, doctors, pilots and so on. In the Republic he offers practical observations on a variety of military and political subjects. Plato’s Socrates is quite possibly idiosyncratic, but even he gives evidence of concern for some common subjects.

Also serious is the charge that Xenophon was not enough of a philosopher to appreciate Socrates. But as D points out, in order for this objection to have any weight we would have already to know what kind of person Socrates was. Modern scholars use this argument not only to reject Xenophon’s account of Socrates as inaccurate, but also to denigrate Xenophon’s Socratic writings, even considered as fiction. Everyone will concede that Xenophon is not a philosopher of the Platonic or Aristotelian type: he did not investigate physics or metaphysics at all, and his discussions of ethical issues are far from systematic. But that did not prevent ancient writers from considering him a philosopher. Originality and systematic thought are not the only criteria, and using such criteria would force us to eliminate quite a few ancient philosophers from the rolls: D mentions Marcus Aurelius. But he does not tell us clearly in what sense Xenophon may be considered a philosopher.

In Rehabilitation de Xénophon, D discusses the conditions which make possible a renewed study of Xenophon’s Socratic writings. Chief among them is the abandonment of the “Socratic problem.” The obsession with the historical Socrates has impoverished the study of Socratic literature, especially Xenophon’s. This is all the more regrettable since the Socratic problem does not admit of any solution. If the Socratic logoi were indeed fictional, then it would be impossible, argues D, to use them to recover the historical Socrates. Of course it might be possible to recover something of the historical Socrates even from fictional writing, but D’s argument that we stand more in need of comparative work on the Socratic writers is a sound one, and may lead lead ultimately to a better appreciation of the historical Socrates as well. To adapt a remark of Xenophon’s — D not only exhorts us to pursue comparative studies of Socrates, he shows us how to do it. He does this in one of the most important sections of the introduction, Xénophon et l’elenchos, which I will discuss in some detail.

D compares the nature of the elenchos in Xenophon and Plato. He disputes the common opinion that Xenophon’s failure to record many elenctic disputes is due to his failure to comprehend Socrates, arguing convincingly that Xenophon’s portrait is designed precisely to offset the impression of Socrates that is created by accounts like Plato’s which make Socrates’ refutations his defining characteristic. The marginalizing of the elenchos is deliberate. But why? In Mem. 1.4.1 Xenophon offers some explanation:

If some people judge, deriving conclusions ( τεκμαιρόμενοι) from what some write and say about him, that Socrates was excellent at exhorting men to virtue, but unable to bring them to it, let them consider not only the refutations he inflicted ( ἤλεγχεν) for the sake of correction on those who thought they knew everything, but also the things he said in his daily conversations with his companions, and then judge whether he was able to improve his companions. (1.4.1)

As D shows, there are only two kinds of conversations highlighted here: refutations of know-it-alls about which others write and speak, thereby creating a misleading impression, and useful conversations with friends. Xenophon here proposes to concentrate on the latter, in order to correct the impression created by those who focused on the former. D comments:

Il est à remarquer que Xénophon ne conteste pas cette association entre l’elenchos et l’inaptitude à rendre vertueux; ce qu’il conteste c’est que l’elenchos épuisait à lui le discourse de Socrate. Si l’on n’a en vue que les elenchoi auxquels Socrate a soumis certains de ses interlocuteurs, le reproche d’impuissance à rendre virtueux est fondé. (CXXXVI)

Strictly speaking the conclusion does not follow. We cannot conclude that Xenophon accepts the charge simply because he does not refute it. But D does have other evidence that Xenophon shares the assumption that the elenchos does not instill virtue. He has shown earlier that Xenophon’s account of the elenchos is much closer to the simple non-philosophic meaning of the term, where to refute someone is simply to humiliate that person by exposing their flaws to public ridicule. If the elenchos is merely public humiliation, it is understandable that Xenophon would not defend it, but would seek to deflect attention to Socrates’ good deeds. Moreover, if one examines the elenchoi that do appear in Xenophon, none of them leads clearly to any education: Alcibiades’ interrogation of Pericles (1.2.40-46) does not; the interview with Aristippus (1.2) does not appear to do so; and although the examination of Euthydemus (4.2-3) is a necessary prerequisite to his eventual willingness to learn, he does not strictly speaking gain any virtue from the elenchos itself. This anecdotal evidence supports the position urged by Socrates’ critics. Xenophon seems to agree that the elenchos does not in itself promote virtue.

But the case is more complex than that. As D argues later, the elenchos does play an important educational role in Xenophon. There are places where Xenophon clearly highlights the positive effect of the elenchos, as in 4.8.11: “he was capable of testing others, of refuting them if they were in error and of exhorting them to virtue and gentlemanliness.” D remarks that in this passage Xenophon dissociates the elenchos from the acquisition of virtue (CXXV). But one could argue that by writing ἁμαρτάνοντας ἐλέγξαι καὶ προτρέψασθαι ἐπ’ ἀρετὴν, he actually means to link them. Even in 1.4.1 Xenophon accepts that Socrates was most able ( κράτιστον) at exhorting others to virtue by means of the elenchos, and he often praises Socrates for exorting others to virtue, as in the passage quoted above, as well as in 1.4.19, 1.6.14, and 2.1.1. Xenophon’s account of Socrates aims to supplement, not to replace ( μὴ μόνον … a)lla καὶ 1.4.1), other accounts in which the elenchos plays a greater role. All he denies is that the elenchos was the sum total of Socrates’ good deeds in the field of education in virtue.1 D’s later discussion of Euthydemus shows that although the elenchos does not instill virtue in Euthydemus, it is still a useful or necessary precursor to such instruction, since it assists in bringing about the recognition of ignorance, which is indispensable to learning. Thus Xenophon does ultimately defend the elenchos at least to this degree. D argues that this beneficial elenchos must be applied only to those who believe that they know everything, as indeed is the case with Euthydemus. And it is also true that in 1.4.1 Xenophon refers to those whom Socrates refuted as those who believed themselves to know everything. One might contest this conclusion, however, on the grounds that, as D shows, Socrates persists in reducing Euthydemus to an acknowledgement that he knows nothing at all. This suggests that the elenchos was also applicable to those who believed they knew anything at all.

The biggest challenge to D’s account of the elenchos, and to any coherent reading of Memorabilia, is the reported conversation between Alcibiades and Pericles. It is hard to know what this is doing here, since it seems to show that Socrates taught his students to engage in elenchoi of the very worst sort: this one does not even contain an exhortation to virtue, but seems to aim only at humiliating an elder leading citizen, who happens to have been Alcibiades’ guardian. The passage not only seems to show that Socrates allowed students to learn this kind of humiliating elenchos from him, but also that in so doing he contributed to the corruption of the youth. D argues that this is an example of the bad kind of elenchos. But if Xenophon wanted to illustrate that, why pick a student of Socrates? And if he did choose a student of Socrates, why not make it clear that he acted against Socrates’ wishes? D argues that “L’elenchos de I 2, 40-46 est probablement l’une de ces bêtises qu’Alcibiade commettait en l’absence de Socrate.” (CLXIX). But why does Xenophon leave us to guess? He does not say that Alcibiades’ conversation was conducted in Socrates’ absence, but introduces the conversation by saying that “while they [Alcibiades and Critias] were still together with Socrates” they sought to hold converse with politicians. And he concludes by saying that “as soon as they thought themselves superior to the politicians, they abandoned Socrates.” That makes it pretty clear that conversations such as these did occur under Socrates’ tutelage. Ultimately, D argues that the passage is “maladroit.” But this is not easy to accept either: the mistake in question is so glaring that no one, not even an arguably dull-witted Xenophon, could have made it. We would not object if a slip-up of this sort occurred somewhere else in Xenophon’s writings. But Xenophon has placed this scene right in the middle of a chapter devoted to the demonstration that Socrates did not corrupt the youth. Nothing compelled him to put it here or anywhere. Nothing compelled him to record it at all. Given its dramatic date (during Pericles’ lifetime) it is not a conversation that Xenophon could have witnessed. He may have heard a report about it from someone ( λέγεται), but it is also possible that he made it up himself. One has to assume that Xenophon had some reason for recording it, even though such a reason is difficult to grasp.

In Unité et Plan, D examines the various efforts that have been made to find some order in Memorabilia. He suggests that the idea of publishing a collection of independent dialogues on different subjects, rather than an extended treatment of one subject, as in Plato, was part of Xenophon’s originality. But he also finds a good deal of coherence here. He takes up Erbse’s approach of seeing the composition as a whole as an expansion of the first two chapters which deal more directly with the charges against Socrates. But he expands and develops the thesis, showing, with surprising success, that just about every part of it can be explained in this way. And he argues that the central aim of the work is to show Socrates’ utility in its manifold manifestations.

In Datation, D reviews the efforts of Delebecque and Kahn to offer a date for Memorabilia, and argues that these efforts are not convincing and not based on an adequate understanding of the text. Both authors want to divide Memorabilia into two parts, books 1-2 and 3-4. Kahn argues this on the basis of references to Plato, while Delebecque argues on the basis of a shift in the subject matter: the first part deals with family and household matters while the second deals with matters pertaining to the city. This, he argues, would be more appropriate after Xenophon’s hypothetical return to Athens late in life. But it is impossible to date the book on the basis of a biography whose outlines are extremely sketchy; and the alleged change in subject matter appears to be an illusion. As D shows, the division between household and political matters is not a division that Xenophon would have made, so there is no good reason to assume any gap between book two and book three. In fact, books two and three of Memorabilia are unified by the effort to describe what Socrates in Symposium calls his skill at “pimping.” The prominence of this skill, which consists in the ability to teach others how to present themselves before other people and before the city, shows just how closely related these two realms are.

D does not propose any date for Memorabilia, and does not even attempt to guess its relation to the Platonic dialogues, even though his argument that 1.4.1 refers to Plato could provide some basis for doing so. That passage assumes that the Platonic Socrates only makes use of the elenchos, and that he does not instill much virtue in his interlocutors. That suggests that Plato’s aporetic dialogues were the only ones known to Xenophon at the time of his writing. It may suggest that Memorabilia was written close to the transition between Plato’s aporetic period and his more dogmatic period — whenever these may have been. It may even suggest that Xenophon had a hand in inspiring that great change in Plato; but that is a kind of speculation in which D does not indulge.

As should be clear, this is a very thoughtful, well-researched and original study. Aside from the points I have mentioned above, I have only one general reservation about the portrait of Socrates that emerges here. D’s aim is to understand the conceptual world of Xenophon’s Socrates, not to describe his character or personality. That is an important aim and a fair one; but one may doubt that it was Xenophon’s main aim. Xenophon aimed to provide a portrait of Socrates’ character as well as his teachings, partly because the two are, in the case of Socrates, inseparable. In general the Socrates that emerges in D’s work is a serious man with a coherent, defensible and highly moral outlook on life. This is not necessarily the portrait that would emerge from a study of the Socrates of the Symposium. One could of course dismiss his behavior in theSymposium as private joking. But it might also be that Socrates’ sometimes wild antics, as well as his irony and even his sarcasm, need to be taken more into account in treating Socrates the philosopher.


1. D also argues that the prime beneficiary of the Socratic elenchos in Xenophon is not the victim but the audience (CXLVIII). But the passages he cites to show this (1.4.19; 1.6.14; 2.1.1) are at least ambiguous. The term συνόντες does not necessarily refer to those who were present at the conversation, but may simply refer to the victims of the refutation, considered as a class. Plato’s Socrates of course is notorious for highlighting the pleasure – – not necessarily the benefit — derived by youthful observers of his refutations (see Apology 23c).