Lin3 ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 2018 Lijuan Lin Galen on to kalon and to agathon in De moribus Lijuan Lin OST both in the original Greek and the complete Arabic translation, Galen’s De moribus (Περὶ ἠθῶν, Fī l-Aḫlāq, “Character Traits”) fortunately still survives in an Arabic epitome. As the title suggests, this work centers on the explanation of different ēthē, their causes, signs, and treat- ments.1 In the first modern research on this text, by Walzer, its aim has been appropriately identified as anti-Stoic.2 While the Stoic intellectual doctrine defines ēthos as belonging only to the rational part of the soul, Galen argues that the non-rational powers also play a prominent role in the development of human character and thus ēthos belongs in no way to the rational soul alone. The Galenic theory of ēthos is not only based on empirical observation of animals and human beings, but also relies on sources of Greek philosophy in the classical period3—most notably, the Platonic tripartition of the soul and related ethical and educational thought. 1 See Ḥunain’s description: in Arabic, G. Bergsträsser, Ḥunain ibn Isḥāq: Über die syrischen und arabischen Galen-Übersetzungen (Leipzig 1925) 49; in Ger- man translation 40. 2 R. Walzer, “New Light on Galen’s Moral Philosophy,” Greek into Arabic (Cambridge [Mass.] 1962) 142–163. 3 For the intellectual background of Galen’s psychological and ethical thought, especially his relationship with other philosophical schools like the Middle Platonists and the Stoics, see J. Mansfeld, “The Idea of the Will in Chrysippus, Posidonius, and Galen,” Boston Area Colloquium in Ancient Philoso- phy 7 (1991) 107–145; T. Tieleman, “Galen’s Psychology,” in J. Barnes and J. Jouanna (eds.), Galien et la philosophie (Vanoeuvres/Geneva 2003) 131–161; Ch. Gill, Naturalistic Psychology in Galen and Stoicism (Oxford 2010). L 78 GALEN ON TO KALON AND TO AGATHON ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 Owing in part to the little attention this work has received so far,4 there are still some difficulties waiting to be solved. One of them has to do with the striking prominence of the terms to kalon (al-ǧamīl) and to agathon (al-ḫayr). In the newest introduction to De moribus, Singer (125–128) contributes a separate part to “the beautiful and the good,” reminding readers of the need to pay special attention to this topic. He rightly points out that such usage in fact echoes Plato, and provides evidence for the possible relation between the two authors. Still, he holds that sometimes these terms have been used by Galen in a prob- lematic way, which suggests that “there may be something more troubling taking place” (127). He especially puts forward (147 n.40) two passages, which are in his view hard to make sense in terms of the Platonic-Galenic vocabulary, namely 33.3–6, and 34.13–14. In this article, I intend to tackle these difficulties, and explain how these two passages could make good sense in terms of Platonic vocabulary. And by identifying 4 For studies on Galen’s De moribus see further Walzer, “A Diatribe of Galen,” in Greek into Arabic 164–174; G. Strohmaier, “Die Ethik Galens und ihre Rezeption in der Welt des Islams,” in Galien et la philosophie 307–329; Gill, Naturalistic Psychology 268–275; P. N. Singer, Galen: Psychological Writings (Cambridge 2013: hereafter ‘Singer’) 109–134; D. H. Kaufman, “Galen on Reason and Appetite: A Study of the De moribus,” Apeiron 50 (2017) 367–392; P. Adamson, “Health in Arabic Ethical Works,” in Health (Oxford, forth- coming). There are two editions: P. Kraus, “Kitāb al-Akhlāq li-Jālīnus,” Bull. Fac. Arts Egypt. Univ. 5 (1939) 1–51; A. Badawī, Dirāsāt wa-nuṣūṣ fī l-falsafa wa- l- ͑ ulūm ͑ inda al-Arab (Beirut 1981). Part of the work has been translated into English by R. Walzer in his two articles. F. Rosenthal translated part of it into German: Das Fortleben der Antike im Islam (Zurich 1965) 120–133. The first full English translation is J. N. Mattock, “A Translation of the Arabic Epitome of Galen’s Book Peri êthôn,” in S. M. Stern et al. (eds.), Islamic Philosophy and the Classical Tradition: Essays Presented by his Friends and Pupils to Richard Walzer (Oxford 1972) 235–260, while the newest English translation by D. Davies is in Singer 135–172. A number of quotations from sources in Arabic and Hebrew can add some further information to our knowledge of this text: for a full list and translation see Singer 173–201. In this paper I quote the Arabic text with page and line numbers from the Kraus edition, and I rely on the translation by Davies with slight modifications. LIJUAN LIN 79 ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 this, we may also add to the list of the Platonic dialogues that Galen used (as Singer 133–134 suggests, they are Republic, Phaedrus, and Laws) a new one, namely the Gorgias. In what follows, I begin with a general introduction to Galen’s ēthos theory in De moribus, so as to put these difficulties in their own theoretical context. I then consider the difficulties involved in the Galenic use of to kalon and to agathon, which are mainly twofold: in the first place, his confusing preference for to kalon over to agathon as the goal of the rational soul; second, his inconsistent use of the two value-terms. I tackle the first diffi- culty in section 2, arguing that Galen’s prima facie preference for to kalon is due to his equivalent use of the two terms at two different levels of the soul, namely the highest rational level and the second-highest honor-related level. This again has to do with the intimate relationship between the rational soul and the spirited soul in De moribus. Lastly, I show in section 3 that Galen’s diverse uses of the two value-terms are not contrary to Greek usage, but quite the opposite: they are both Greek and Platonic. By clarifying those difficulties, I hope this article could also cast new light on related questions like the relationship of the three parts of the soul in De moribus, Galen’s theory of moral education, and the ways in which he inherited and further developed the Platonic theory. 1. to kalon and to agathon in context One may wonder why to kalon and to agathon could play a prominent role in a work dedicated to character traits. My in- vestigation therefore starts with a brief introduction to Galen’s ēthos theory in De moribus, which, as is well known, is based on a Platonic tripartite psychological theory.5 For Galen, the soul is not, as the Stoics held, identified with the rational capacity alone, but it is rather, as Plato believed, composed of three parts: the appetitive soul (an-nafs aš-šahwānīya), the spirited soul 5 On this topic see further M. Schiefsky, “Galen and the Tripartite Soul,” in R. Barney et al. (eds.), Plato and the Divided Self (Cambridge 2012) 331– 349. 80 GALEN ON TO KALON AND TO AGATHON ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 (an-nafs al-ġaḍabīya), and the rational soul (an-nafs an-nāṭiqa) (26.9–12).6 Plato’s tripartition of soul, especially his acknowl- edgement of the existence of the two non-rational souls, pro- vides support for Galen’s theory of ēthos, insofar as for Galen, ēthos should essentially be a non-rational condition of the soul.7 At the very beginning of De moribus, Galen provides his defini- tion of ēthos and stresses its non-rational character: “a charac- ter trait (al-ḫulq) is a state of the soul (ḥāl li-n-nafs) that induces someone to perform the actions of the soul without con- sideration (rawīya) or choice (iḫtiyār,8 25.4–5).” At another point in the treatise, he proceeds to talk of a twofold division of the soul, and identifies character traits with the non-rational state of the soul: “These states are divided into two categories: there are those that arise in the soul after [the exercise of] thought (al-fikr), consideration (ar-rawīya), and discrimination (at-tamyīz), and these are called ‘knowledge’ (ma ͑ rifa), ‘opinion’ (ẓann), or ‘judgement’ (ra ͗ y); and there are those that occur to the soul without [the exercise of] thought (ġayr fikr), which are called ‘character traits’ (aḫlāq)” (28.15–18). Galen then seeks to establish his view by appealing to empirical observations. As witnesses to the existence of the irreducibly non-rational elements in the soul, he refers to 6 It is to be noted that Galen furthermore equates the Platonic tripartition of the soul with the Aristotelian one: the appetitive soul is regarded as the same as the vegetative soul (an-nafs an-nabātiyya), the spirited soul as the animal soul (an-nafs al-ḥayawāniyya), and the rational soul as the cognitive soul (an-nafs al-mufakkira). 7 This is not to say that for Galen ēthos belongs only to the non-rational soul, since in the following argumentation he obviously also acknowledges the role of the rational part of the soul in the formation of character. In this regard I agree with Gill and Singer that Galen does not deny the role of rational soul in the formation of character, and his emphasis on the non- rational character of ēthos is to be explained only by his endeavor to reject the intellectual doctrine of the Stoics: Gill, Naturalistic Psychology 271–272; Singer 124–125. 8 Instead of the reading of Kraus (iḫtibār, precise knowledge) I follow here Badawī (iḫtiyār, choice). LIJUAN LIN 81 ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 people’s involuntary responses to certain stimuli, e.g. they laugh involuntarily when they hear something amusing, even if they do not want to do so.9 Furthermore, in order to show that character traits have to do with such non-rational elements, he also resorts to some other empirical facts, for example, that non-rational animals and pre-rational children already have different character traits.10 All these phenomena provide evi- dence in support of Galen’s view that character traits are inborn non-rational dispositions of the soul. However, there re- main some more complex questions which empirical evidence is unable to answer. These include: Why do people have differ- ent character traits at birth? Could the inborn character be changed? And if it could be changed, how and to what extent would it be changed, and in which direction should it be changed? In respect of such questions, the Platonic tripartite psychological theory is of special importance to Galen. This also explains why the whole summary is structured according to the Platonic tripartition of the soul, and much of the argu- mentation is also devoted to elaborating on the interrelation- ship of the three parts of the soul. Like Plato, Galen assigns various capacities and goals to the three souls: the appetitive soul is chiefly concerned with desires necessary for life and generation, and it inclines towards pleasure (al-laḏḏa); the spirited soul is the seat of anger, and it inclines towards conquest (al-ġalaba); the rational soul is where understanding (fahm) resides, and it inclines towards the pursuit of truth, wisdom, and the beautiful (al-ǧamīl, to kalon) (38.19–21, 9 See 25.2–4: “some people, when surprised by a terrible sound, are frightened and shocked, and that when they see or hear something amusing they laugh involuntarily (ġayr ͗ irāda); they often wish to refrain from this but they are unable to do so.” 10 For different character traits of animals see 25.11–26.1. For various character traits of still non-rational human beings like children see 29.10– 13, 29.18–30.1. Similar descriptions can also be found in Galen’s other works, see e.g. Plac.Hipp.Plat. V 337–338, 499–501 K.; Quod An.Mor. IV 768–769. 82 GALEN ON TO KALON AND TO AGATHON ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 26.9–12). The diverse roles of the three souls help to elucidate why people’s character traits differ: “The difference between the various categories of character traits is only because of a greater or lesser [degree] of inclination (al-kaṯra wa-l-qilla fī mayl) in each of the souls, according to the extent of its natural strength” (39.1–2). For example, greed as a character trait is based on a particularly strong appetitive soul, as it inclines by nature towards pleasure, and an innately bold person must have a very strong spirited soul, which inclines naturally towards conquest. Galen further suggests that as to a particular character trait, e.g. boldness (ǧasāra), we can also divide it into different types according to the interrelationship of different souls: steadiness (waqār) is only to be found in someone whose spirited soul is both strong and obedient to the rational soul, while rashness (tahawwur) resides in someone who has a strong and undisciplined spirited soul, which refuses to cooperate with the rational one (31.14–32.4, 32.14–15). The emergence of a character trait can thus always be clarified by a particular state of one soul or by the interrelationship of different souls.11 While emphasizing character as an inborn disposition of living beings, Galen does not ignore that education and train- ing also play a crucial role in the formation of character traits.12 11 Cf. Singer 124. 12 See e.g. Gill, Naturalistic Psychology 271–272: “Although Galen also highlights a process of development (signs of shame or shamelessness and of love of honour appear later than birth), he emphasizes that young children give early indications of whether they are inclined towards virtue or vice. This theme is subsequently linked with the tripartite psyche, in the idea that natural differences in character reflect variations in the relative strength of the parts of the psyche … The cumulative effect of these features is to suggest a rather different picture of the development of character from that offered by Plato in the Republic or Laws and Aristotle’s ethical works. There, the view is that the developed character of adult human beings is a product of the interplay between inborn nature, habituation in the family and com- munity, and rational decisions made by the person herself. This Platonic- Aristotelian view is reflected elsewhere in Galen’s writings as well as in other Imperial thinkers such as Plutarch.” LIJUAN LIN 83 ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 In the first book he mentions how character traits emerge from habituation (͑ āda): “A character trait comes about by means of continuous habituation (bi- ͑ āda dā ͗ im) to things that someone establishes in his soul and to things that he always does, every day, such as I shall describe later” (31.4–5). At the end of the same book he indicates that education will bring much benefit to one’s inborn character: “If it (the spirited part of the soul) is naturally bold (ǧasūra bi-ṭ-ṭab), education (al-adab) will acquire compliance and obedience for it; if it is lacking in boldness it will become better than it was (ḫayran mimmā kānat)” (34.15– 16).13 Like the Stoics, Galen also acknowledges the specific importance of education in the formation of human character; however, he disagrees with the Stoics insofar as they further suggest that character traits are formed by education and intellectual instruction alone. On the contrary, Galen keeps re- minding us that the extent to which someone can be educated depends greatly on the inborn nature of his different parts of soul, e.g. whether his spirited soul is amenable to education, or whether his rational soul is knowledgeable.14 Nevertheless, Galen seems quite pessimistic about the probability of the im- provement of a truly bad nature: “I think, [however], that someone who is, by nature, extremely cowardly and greedy will not, by means of education, become extremely brave and ab- stemious (28.13–14).”15 In a word, according to Galen, good character traits emerge from the perfect combination of inborn nature, habituation, 13 See further 47.16, where Galen describes habit as “a second nature (ṭabī ͑a ṯāniya).” 14 See 28.4–6: “Not every dog or horse is amenable to education … Sometimes the hunter and the horseman are unknowledgeable.” For a similar view see 34.17–19. 15 Besides, Galen also reminds us of other conditions necessary for an effective education, for example a child is in comparison with an old man more appropriate for education, just as a tree which has just begun to grow can be bent more easily than the same tree which is fully grown, see e.g. 31.5–9. 84 GALEN ON TO KALON AND TO AGATHON ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 and intellectual instruction. But what are good character traits —or to put it in another way, what are the goals of character development? In De moribus Galen writes at great length on this topic, to which he doubtless assigns crucial significance. It is also in this part that our central topic—the striking prominence of to kalon and to agathon—is brought into discussion. First of all, Galen argues that character education aims to bring different parts of the soul into a fine balance and agreement with one another, which again indicates a proper order of the three souls: “The aim is that the rational soul should employ the services of the spirited soul in subduing the appetitive soul, and that the spirited soul should be trained to obey the rational soul” (wa-l-maqṣūd istiḫdām an-nāṭiqa li-l-ġaḍabīya fī qam ͑ aš- šahwānīya wa-riyāḍa l-ġaḍabīya li-l-inqiyād li-n-nāṭiqa, 39.11–12, cf. 43.2 ff.). To achieve this, the spirited soul should be made compliant and the appetitive soul weak (28.11–12), so that both non-rational parts are subject to the rational one and serve its goal, i.e. the ultimate goal of the whole soul. Now to kalon (al- ǧamīl) is of special importance in this picture, as it has been identified as being this goal: “The rational soul, then, must love the beautiful (al-ǧamīl), hunger for truth, and be knowledgeable about the agreement and disagreement of things” (28.7–8). Since there is always a conflict between pleasure and to kalon in the soul, i.e. the conflict between what the appetitive soul desires and what the rational soul pursues, the goal of education is to make people learn how to live according to to kalon like the angels, and be free of various bestial pleasures as far as possible (34.3 ff., 38.1–3). Later in the fourth book, the goal of the rational soul is also described as to agathon (al-ḫayr): “we see that love of the good (ḥubb al-ḫayr) in people belongs to the rational soul” (48.16). On this argumentation, both to kalon (al-ǧamīl) and to agathon (al-ḫayr) would seem to be equally regarded as the goal of the rational soul and hence are co-extensively used in De moribus. However, a second look at the text will remind us that the case is much more complex. On the one hand, unlike Plato, Galen obviously prefers to kalon to to agathon as the ultimate goal of the rational soul, as in the whole De moribus we find only one pas- LIJUAN LIN 85 ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 sage in which to agathon is granted a role similar to that of to kalon; on the other hand, in some other passages the two terms are surprisingly not co-extensive any more, but on the contrary are set in opposition. But before we turn to a detailed explanation of the difficulties involved in the use of to kalon and to agathon, a brief comment is needed on the aim and the basic character of the whole summary. The aim of De moribus is twofold. Elucidated are not only the theoretical issues like the nature, the causes, and the signs of ēthos, but also practical issues like the treatments of ēthos and the final goal of an ethical life. The two aspects are certainly correlated: for Galen, an appropriate understanding of the nature of the three souls also teaches us how to live a virtuous and moral life. On the whole, the Galenic theory of ēthos is obviously under Plato’s influence, and this is not only due to Galen’s use of the Platonic tripartite theory of soul, but also because of his acceptance of other correlated Platonic educational and ethical ideas. It is well known that Plato paid special attention to the role of non-rational powers in the human soul. That is why he suggests that we should educate young people according to their different natures;16 that the first period of education should be mainly directed at the training of the non-rational souls like the thymoeides;17 and that 16 Cf. e.g. Phdr. 271B, 276E–277D; Resp. 485A–B. On this topic see further T. A. Szlezák, Platon und die Schriftlichkeit der Philosophie: Interpretationen zu den frühen und mittleren Dialogen (Berlin/New York 1985), and Platon Lesen (Stutt- gart/Bad Cannstatt 1993) 55–56; R. Geiger, Dialektische Tugenden (Paderborn 2006) 34–38; M. Erler, Platon, Grundriss der Geschichte der Philosophie: Die Phi- losophie der Antike II.2 (Basel 2007) 85–87. 17 D. L. Cairns, Αἰδώς: The Psychology and Ethics of Honour and Shame in Ancient Greek Literature (Oxford 1993) 387; J. Moss, “Shame, Pleasure, and the Divided Soul,” OSAPh 29 (2005) 137–170, at 166; G. R. Lear, “Plato on Learning to Love Beauty,” in G. Santas (ed.), The Blackwell Guide to Plato’s Republic (Oxford 2006) 104–124, at 106; I. Vasiliou, “From the Phaedo to the Republic: Plato’s Tripartite Soul and the Possibility of Non-philosophical Virtue,” in Plato and the Divided Self 9–32, at 29; J. Wilburn, “Moral Edu- cation and the Spirited Part of the Soul in Plato’s Laws,” OSAPh 45 (2013) 86 GALEN ON TO KALON AND TO AGATHON ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 true virtue comes from a combination of appropriate dis- position and reason (Leg. 653A–B). While setting knowledge of the good, i.e. what the rational soul desires, as the highest goal of education, Plato keeps reminding us of the existence and the importance of the non-rational elements of the soul, which could either impose limits on our character development, or, if well trained, give extra assistance to it. He is also well aware that to fulfill the noble goal of education at a later stage, we must prepare ourselves with appropriate training and habitu- ation of the non-rational souls in the early stage of education.18 All these Platonic thoughts, as we have seen, have been ac- cepted and well established in the Galenic theory of ēthos. 2. to kalon and to agathon as co-extensive terms As Singer rightly observed, if we look closely at the use of to kalon and to agathon in De moribus, some ambiguity emerges. First of all, it is remarkable that Galen mostly identifies to kalon, but not to agathon, as the ultimate goal of the rational soul. This can be directly verified not only in the aforementioned 28.7–8, but also in many other passages, e.g. 38.19–20 in the second book: “Understanding [resides] only in the rational soul, and it is a capacity that perceives agreement and disagreement in all things. This soul inclines towards the beautiful (al-ǧamīl).” In some other passages Galen also indicates that to kalon belongs neither to the appetitive soul nor to the spirited one, but ex- clusively to the rational soul. In the third book he offers: “The beautiful and the ugly (wa-l-ǧamīl wa-l-qabīḥ) in the rational soul are like pain and pleasure in the vegetative soul” (42.4–5), and at the end of the first book: “A person ought to accustom his rational soul to the love of the beautiful (al-ǧamīl) and give it ___ 63–102, at 82. 18 Alcibiades’ tragic downfall is one of the best examples which demon- strate the importance of early moral education and especially the signifi- cance of appropriate training of the honor-loving spirited soul. For an excellent interpretation on this topic see J. Wilburn, “The Problem of Alci- biades: Plato on Moral Education and the Many,” OSAPh 49 (2015) 1–38. LIJUAN LIN 87 ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 authority over his spirited soul” (34.14–15). Moreover, we also find other passages in which Galen uses the rational soul’s aiming at the beautiful (or not) to explain the formation of a particular character trait like courage, rashness, or envy: “Courage, on the other hand, comes about from the rational soul’s aiming at the beautiful (al-ǧamīl), and of the spirited soul’s being readily obedient to it when there is in its nature the state in which it will combat terrible things. Rashness comes about in battle when the rational soul does not aim at the beautiful (al-ǧamīl) and the spirited soul is in a state in which it is not frightened by what frightens other people” (32.8–11); “When there are united in someone a spirited soul that is concerned with rule and a rational soul that does not love the beautiful (al- ǧamīl), his characteristic trait is envy” (50.13–14).19 In contrast to the frequent description of to kalon as the ultimate goal of the rational soul, there is only one passage in the whole of De moribus, namely the aforementioned 48.16 in the fourth book, in which to agathon (al-ḫayr) has replaced to kalon and is identified with the goal of both the rational and the spirited soul. Someone familiar with the Platonic description of the tri- partite soul might already realize that something confusing is happening here. According to what Plato himself presents in the Republic and other dialogues like Phaedrus and Laws,20 and according to a widely accepted view in scholarship,21 it is rather the spirited part of the soul which is primarily associated with 19 See also Quod An.Mor. IV 772 K. 20 Resp. 439E–440D, 548C, 571C; Phdr. 246B, 253E–254E; Leg. 679B, 792A; etc. 21 E.g. Cairns, Αἰδώς 383–384; G. R. Lear, Happy Lives and the Highest Good: An Essay on Aristotle’s Nicomachean Ethics (Princeton 2004) 139–144; Moss, OSAPh 29 (2005) 153–156; J. Wilberding, “Plato’s Two Forms of Second-Best Morality,” The Philosophical Review 118 (2009) 351–374, at 352; O. Renaut, “Challenging Platonic Erôs: The Role of Thumos and Philotimia in Love,” in E. Sanders et al. (eds.), Erôs in Ancient Greece (Oxford 2013) 95– 110; J. Wilburn, The Ally of Reason: Plato on the Spirited Part of the Soul (diss. Princeton 2011), and OSAPh 45 (2013) 63–102. 88 GALEN ON TO KALON AND TO AGATHON ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 feelings of admiration towards to kalon and disgust towards to aischron, whereas the rational part of the soul is associated with desires for to agathon and aversion to to kakon. In De moribus the reverse seems to be the case: the goal of the spirited soul turns into that of the rational one. Furthermore, this peculiar phe- nomenon is not confined to to kalon and to aischron, but also applies to other things which traditionally belong to the range of the spirited soul, e.g. honor-related emotions like sense of shame22 or honor-related values like nobility. All of them are now ascribed to the rational soul, as we can observe in the following passages: “The action of the rational [soul] is the contrary to the state of the appetitive [soul], for it prefers the knowledge (ma ͑ rifa) of things, especially those which it is more beautiful (aǧmal) to know; it is ashamed of (tastaḥā) the means of pleasure and conceals them, and it combats the appetitive capacity” (38.3–5). “The nobility (šaraf) of the soul comes from its knowledge and the most noble of things is complete knowl- edge” (44.1). Why does Galen prefer to kalon as the goal of the rational soul? Singer suggests that the reason ought to trace back to Plato: for there is some evidence in Plato that to kalon and to agathon can be used interchangeably, and both can represent the highest moral value, i.e. the ultimate goal of the rational soul. For example, the highest moral value is represented in the Republic by the form of the good, while in the Symposium the same role is undertaken by the form of the beautiful. Building on an interpretation offered by Hobbs,23 Singer (126) further suggests that the two terms “may ultimately be identical” in Plato, that is to say, “the two entities are (at the highest meta- physical level) identical, but may be different in our human perception of them; and in particular, the spirited soul can only 22 For shame as a moral emotion belonging to the spirited part of soul in Plato, see e.g. Resp. 571C, Phdr. 253C–254E. For literature on this topic cf. e.g. Cairns, Αἰδώς 383; T. Irwin, Plato’s Ethics (Oxford 1995) 213. 23 A. Hobbs, Plato and the Hero: Courage, Manliness and the Impersonal Good (Cambridge 2000) 220 ff. LIJUAN LIN 89 ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 perceive things (rightly or wrongly) as either kalon or aischron; while the rational soul will understand the true identity of beautiful/noble as co-extensive with moral good, and not (as it can appear to the untutored soul) in conflict with it.” The first difficulty could thus be solved, as Singer indicates, by the alleged co-extensiveness of the two terms at the highest level, i.e. to kalon is identical with to agathon as the highest moral value, and they both can be realized only through the activities of the rational soul and its knowledge. This not only explains why Galen uses the two terms in an indistinguishable way, but also clarifies what Galen truly means when he identifies to kalon as the ultimate goal of the rational soul. To kalon in De moribus does not primarily refer to honor-related things in the normal sense, but instead it concerns knowledge as the truly honorable thing, as we can see from e.g. 43.8–11: “The soul’s beauty (ǧamāl) comes from knowledge and its ugliness from ignorance. As for the greater and lesser [degree] of its fineness (ḥusn), it accords with the greater or lesser [degree] of the nobility of the things that it knows and upon the depth or shallowness of its knowledge. The greater or lesser degree of its ugliness depends on the contrary of this.” Accordingly, the feeling of shame or nobility of the rational soul is also not to be understood at the level of honor-related things, but at the highest level: the rational soul is ashamed of not being eager for knowledge, and its nobility lies in nothing but its knowledge, as we have seen in 38.3–5 and 44.1.24 Galen’s preference for to kalon as the highest goal should thus no longer confuse us, as the sense of to kalon here basically has no essential difference from that of to agathon, and they both represent the knowledge and the highest moral 24 The location of shame, namely to which part of the soul (the rational or the spirited) it belongs, might have already been a controversial topic in the Platonic academy. In the horse-rider metaphor of Phaedrus (254B), Plato uses shame and kindred emotions to describe both the good horse (the spirited soul) and the rider (the rational soul). For shame as an emotion belonging to the rational soul in Aristotle see Top. 4.5, 126a6–12. 90 GALEN ON TO KALON AND TO AGATHON ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 value as the ultimate goal of the rational soul.25 There remains, however, a problem that the suggested in- terpretation is probably unable to solve, which concerns the possibility of an equivalence of the two terms at another level of the soul. Have to kalon and to agathon been equivalently used at the highest level of the soul alone? A careful examination of the 48.15–18 in the fourth book, where Galen identifies to agathon as the goal of both the rational and the spirited soul, shows us a different picture: As for love of the good (al-ḫayr), when we consider that God loves the good, and that the soul by means of which people imitate God is the rational soul, we see that love of the good in people belongs to the rational soul. When[, on the other hand,] we consider preference for doing good to others, and we see that beasts and infants may incline towards doing good to someone towards whom they incline, we see that love of the good belongs to the spirited soul. My own view is that it belongs to both the rational and the spirited souls. Here Galen obviously also talks about the love of to agathon in the spirited soul, and refers specially to things like doing good to people to whom we are akin—which is traditionally counted as honor-related things and to kalon in Plato. Galen’s ascribing to agathon to the spirited soul, which ought to aim at things it thinks kalon, suggests that the equivalence of the two terms is applied not only at the highest level, but also at the second highest level, namely the honor-related level. This in due course leads us to another point, which concerns the relationship between the spirited soul and the rational soul in De moribus. The equivalent use of the two terms, both at the highest level and at the second highest level of the soul, indi- cates the intimate relationship between the two souls. Although the two strive for different goals, their goals share remarkable similarities with each other, insofar as to kalon which the spirited soul strives for is in certain degree also agathos, and to agathon 25 Adamson, in Health, rightly notices that Galen describes the soul’s health in purely intellectualist terms. LIJUAN LIN 91 ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 which the rational soul is eager for is also in some way kalos. In this regard the relationship between the two souls is totally different from e.g. the relation between the spirited soul and the appetitive soul or that between the rational soul and the appetitive soul. As is well known, Plato defined the spirited soul as an assistant of the rational soul, in dialogues like the Republic, Phaedrus, and Laws. In De moribus Galen gives particular em- phasis to the same point: the relationship of the spirited soul to the rational soul is likened to that of the dog to the hunter, or the horse to the rider,26 and they both should fight together against a strong, greedy beast, i.e. the appetitive soul (42.9 ff.). In the struggle between the appetitive soul’s striving for pleasure and the rational soul’s striving for to kalon,27 the role of the spirited soul is to aid the rational one: “The Creator has helped man in this struggle by [giving him] the spirited soul, for 26 Cf. 27.20–21: “The relationship of the spirited soul to the rational soul is analogous to that of the dog to the hunter, or of the horse to the rider.” Doubtless this analogy has its origin in Plato, as the spirited soul is com- pared to a dog in the Republic (376A), and to a horse in the Phaedrus (352C– 256E). It is worth mentioning that the metaphor of the spirited soul as a dog was inherited in the Arabic tradition and widely used by Arab philosophers like al-Kindī, Miskawayh, al-Isfahānī and al-Ghazālī: see P. Adamson, Al- Kindī (Oxford 2007) 228; Y. Mohamed, “The Metaphor of the Dog in Arabic Literature,” Tydskrif Vir Letterkunde 45 (2008) 75–86. However, in the works of Islamic philosophers like Miskawayh and al-Isfahānī, the dog- hunter or horse-rider imagery gradually developed into a horse-dog-rider model. While the rational soul is still likened to the rider, the spirited soul is now his dog, and the appetitive soul his horse: see Mohamed 79–80. Still, Mohamed wrongly traced the dog/horse-rider model back to Galen, and ignored the fact that the animal imagery of the soul in the Arabic world is actually derived from Plato. 27 Cf. 38.1–5: “The desiderative capacity possesses no consideration or discrimination and does not know the beautiful actions or the angelic way of life since it devotes itself only to pleasure and avoids only what pains it. The action of the rational [soul] is the contrary to the state of the desiderative [soul], for it prefers the knowledge of things, especially those which it is more beautiful to know; it is ashamed of the means of pleasure and conceals them, and it combats the desiderative capacity.” 92 GALEN ON TO KALON AND TO AGATHON ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 his rational soul calls upon it for help to discipline the ap- petitive soul and restrain it from excessive movement” (38.8–9). In fact, it seems that in some places of De moribus Galen does not care much about distinguishing the role of the rational soul from that of the spirited one, but instead he is often content with roughly describing the two parts as a companionship, as the angelic part of the soul,28 which stands in opposition to the bestial part of the soul, i.e. the appetitive soul. Those who follow their angelic part of the soul will be godlike, just as those who follow their bestial part of the soul will be beasts, as Galen tells us at the end of the first book: “Someone who in his nature and his act makes [the attainment of] this pleasure his goal is like a pig, whereas someone whose nature and act love the beautiful follows the example of the angels. These [last], therefore, deserve to be called ‘godlike’, and those who pursue pleasures deserve to be called ‘beasts’ ” (34.8–10). In such a context, it might be hard to tell whether to kalon is the goal of the rational soul or that of the spirited soul; but rather, it has been used to represent roughly the goal of the two higher souls, which is, however, clearly distinguished from the pleasure of eating and breeding in the appetitive soul.29 In De moribus the spirited soul and the rational soul are so closely related that the borderline between the two is often not so easy to draw as that 28 The “angelic” part of soul seems strange and un-Galenic. It may originally have been the “godlike” part of soul, which has been replaced by the Christian translator Ḥunayn with “angelic,” see Singer 113. 29 Cf. 34.4–6: “Whoever chooses pleasure rather than the beautiful as his goal chooses to be like a pig rather than to be like an angel. For the angels do not eat or breed, since their substance remains in the same state.” In some passages from the third book, Galen also mentions that both the spirited and the rational souls should be made strong to fight against the beast: “The rational and spirited souls … accept education and benefit from training; these two ought to be trained so as to become strong, and the other neglected so as to become weak” (42.6–8); “He trains himself and his dog in everything that increases their combined strength” (42.15). In the ideal case, one would be freed from the other two souls, namely the de- siderative and the spirited souls, and would live like an angel, cf. 40.6–8. LIJUAN LIN 93 ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 between the appetitive soul and the rational one. In this light, the equivalence of to kalon and to agathon should also reflect Galen’s positive evaluation of the spirited soul as an assistant of the rational soul, which, again, has its background in Platonic thought. 3. to kalon and to agathon in opposition I now turn to the second difficulty concerning the use of to kalon and to agathon in De moribus, which has been described by Singer as “something more troubling.” Both passages are near the end of the first book, where Galen talks about the differ- ence between courage and cowardice: Because the things that someone naturally abhors and draws back from are of two types: one of them is that which he sees to be bad (šarr), and the other is that which he sees to be ugly (qabīḥ). There is greater courage (an-naǧda) in resisting what is ugly than there is in [resisting what is] bad, and cowardice (al- ġubn) is the reverse.30 The first is like one who prefers death in battle to defeat, and who will endure torture rather than tell a lie about a friend (33.3–6). Some of them choose what they think is beautiful (ǧamīl) rather than what they think is good (ḫayr); they are courageous and the opposite of these are cowards (34.13–14). In both passages, to kalon and to agathon, and correspondingly their antonyms, to aischron and to kakon, are surprisingly not co- extensive any more, but rather are set in opposition, as Singer (127) rightly observes on 33.3–6: Here ‘ugly’ and ‘evil/bad’ are set in opposition in the context of their avoidance; and, though the interpretation is far from un- problematic, the sense of the argument seems to be that it is morally preferable to avoid [doing] something ‘ugly’ rather than [suffering] something ‘bad’. But if this is right, ‘bad’ here is 30 Mattock’s translation: “There is greater courage in facing what is ugly than there is in facing what is evil, and there is less cowardice in refusing to face the former than there is in refusing to face the latter.” This translation makes no sense. On this topic see below. 94 GALEN ON TO KALON AND TO AGATHON ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 being used in a way which is not only not co-extensive with ‘ugly/shameful’, but in fact opposed to it; the sense here is presumably of ‘bad’ in the sense of ‘physically harmful’. Singer then points out (128) that similar problems also exist in 34.1, and concludes that the inconsistent use of these value- terms might originate from some distortion of Greek usage: Though sense can be made of the distinction offered in this pas- sage, then, it is difficult to see the terms for ‘beautiful’, ‘ugly’, ‘good’ and ‘bad’ here as consistently reflecting the Greek usage; it seems that at least some confusing distortion has taken place in the use of the value-terms. Against this interpretation, I shall demonstrate that the Galenic use of the two terms is not a distortion, but instead precisely corresponds to ordinary Greek usage, and especially Platonic vocabulary. First of all, it is worth mentioning that both passages occur in a context where Galen tries to explain character traits derived from the spirited soul: the trait of anger, and the corresponding virtue, courage. Broadly speaking, people equipped with a strong spirited soul are all bold by nature; however, difference arises as regards the relationship between the spirited soul and the rational one in them. As Galen tells us, true courage can only be found in a man whose spirited soul is both strong and obedient to the rational soul, which aims at to kalon: “Courage, on the other hand, comes about from the rational soul’s aiming at the beautiful (al-ǧamīl), and of the spirited soul’s being readily obedient to it when there is in its nature the state in which it will combat terrible things” (32.8–10). It is in this context that the distinction between to kalon and to agathon emerges. Galen (from 33 onward) then tries to use this distinction to elucidate the difference between true courage and cowardice: as to things that people naturally incline to, namely to kalon and to agathon, courageous persons always prefer the former to the latter, whereas cowardly persons are the reverse; just as with the things that people naturally draw back from, namely to aischron and to kakon, courageous persons usually prefer the second, whereas cowardly persons, on the contrary, prefer the first. In LIJUAN LIN 95 ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 fact, similar use of the two terms is not confined to these two places, but also emerges in some nearby passages (33.11–12 and 33.16–17): It is said that some beautiful acts (ba ͑ḍ al-af ͑āl al-ǧamīla) are what is aimed at by those who aim at certain good goals (ġāyāt al-ḫayr), such as the state of pleasure or security, when they prefer what leads to them, even if it is harmful… … for the nature of the ugly, both in the imagination and in reality, is not the same as the nature of the bad, and similarly the nature of the beautiful (al-ǧamīl) is not the same as the nature of the good (al-ḫayr). In all these passages, the distinction between the two terms is obvious: while to kalon remains the highest goal of the rational soul, the term to agathon, as both 33.11 and 33.17 indicate, refers to a very different sense, namely the amoral physical security or pleasure, which brings direct benefit to its owner. In this picture, to kalon and to agathon not only have very different natures (33.15), but they are also quite differently ordered in the value system: to kalon is undoubtedly higher than to agathon. And that is why the one who chooses the latter rather than the former, or the one who chooses the former only for the sake of the latter, would be called coward and should be severely criticized (34.14, 33.10). Taking into account the role of this distinction in the whole argumentation, its consistent use and its frequent occurrence in 33 and 34, Singer’s conclusion that this distinction might result from a confusing distortion seems highly improbable. Still, it is worthwhile to ask why Galen uses to agathon in a different way here, and whether this use of the two terms corresponds to ordinary Greek usage, especially Platonic usage. As a matter of fact, if we look to the Platonic corpus and examine the evi- dence, a positive answer to this question emerges, as there to kalon and to agathon are not used in a unified and consistent way, but quite the opposite—they are notoriously used in a various 96 GALEN ON TO KALON AND TO AGATHON ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 way in different places.31 Sometimes the two terms seem to be treated as equivalent and represent equally the ultimate goal of the rational soul, i.e. the highest good at the moral level (Sym- posium, Republic); sometimes they are used separately to refer to the admirable goal of the spirited soul and the desirable goal of the rational soul (Republic, Laws); otherwise they are set, how- ever, in opposition. In the last case, to kalon concerns good in a moral sense, while to agathon is used to refer more generally to something good, good things (like physical health) which can bring direct pleasure and amoral benefit. The two confusing passages relate directly to this last usage. One of the best examples for this usage can be found in Plato’s Gorgias, especially in the exchange between Socrates and Polus, which centers on the relationship between to kalon and to agathon, to kakon and to aischron. While Polus takes the view that suffering injustice is worse (kakion) than doing injustice, and doing injustice is uglier/more shameful (aischion) than suffering injustice, Socrates holds that doing injustice is both uglier/ more shameful (aischron) and worse (kakion) than suffering in- justice.32 The two positions, however different they may seem, 31 As Barney has demonstrated in her thorough treatment of the Platonic use of the kalon and the good, the relation between the kalon and the agathon might be a puzzle also to Plato himself: R. Barney, “Notes on Plato on the Kalon and the Good,” CP 105 (2010) 363–377. On this topic see also Lijuan Lin, Die Helfer der Vernunft – Scham und verwandte Emotionen bei Platon (diss. Munich 2016). 32 Grg. 474CD: Σωκράτης: λέγε δή μοι, ἵν᾽ εἰδῇς, ὥσπερ ἂν εἰ ἀρχῆς σε ἠρώτων: πότερον δοκεῖ σοι, ὦ Πῶλε, κάκιον εἶναι, τὸ ἀδικεῖν ἢ τὸ ἀδικεῖσθαι; Πῶλος: τὸ ἀδικεῖσθαι ἔμοιγε. Σωκράτης: τί δὲ δή; αἴσχιον πότερον τὸ ἀδικεῖν ἢ τὸ ἀδικεῖσθαι; ἀποκρίνου. Πῶλος: τὸ ἀδικεῖν. Σωκράτης: οὐκοῦν καὶ κάκιον, εἴπερ αἴσχιον. Πῶλος: ἥκιστά γε. Σωκράτης: μανθάνω: οὐ ταὐτὸν ἡγῇ σύ, ὡς ἔοικας, καλόν τε καὶ ἀγαθὸν καὶ κακὸν καὶ αἰσχρόν. (Socrates:) “Well then, so that you may know, tell me, just as though I were asking you all over again, which of the two seems to you, Polus, to be the worse – doing wrong or suffering it?” (Polus:) “Suffering it, I say.” (Socrates:) “Now again, which is uglier – doing wrong or suffering it? Answer.” (Polus:) “Doing it.” (Socrates:) “And also worse, if uglier.” (Polus:) “Not at all.” (Socrates:) “I see: you hold, apparently, that beautiful and good LIJUAN LIN 97 ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 share nevertheless a common presupposition, namely that there is a clear distinction between aischros and kakos, kalos and agathos. Both Socrates and Polus agree, at least in this exchange, that kalos/aischros involves evaluation at the moral level, where- as agathos/kakos involves evaluation with respect to personal pleasure, benefit, and advantage in the amoral sense. Moreover, the exchange between Socrates and Polus pro- vides some further information for our understanding of 33.5: “There is greater courage in resisting what is aischron than there is in [resisting what is] kakon,” which Singer finds very puzzling (147 n.40). He especially finds it “difficult to see how a single verb (waqā, ‘guard against’, ‘resist’, ‘face’) can have the same function in relation to both ‘ugly’ (i.e. aischron) and ‘evil’ (i.e. kakon) here.” Singer therefore suggests the required sense of this sentence as follows: “there is greater courage in resisting (i.e. refraining from) something ugly (i.e. an ugly action) than in en- during something evil (i.e. something bad happening to one).” This suggestion, however, does not seem convincing. It is hard to comprehend why the one who resists something ugly should be more courageous than the one who endures something bad, since the two persons can be the same one, i.e. the one who endures something bad exactly because he resists doing some- thing ugly. I suggest instead that the single verb waqā (resist) can indeed serve both aischron and kakon, and thus the original sentence is right. That is to say, there is greater courage in re- sisting something ugly (and thus suffering something bad) than in resisting something bad (and thus doing something ugly). In fact, the same view is already indicated in the exchange be- tween Socrates and Polus in Gorgias, namely, if one must choose between resisting an ugly (aischros) action like immoral deeds (resisting doing injustice), and resisting something bad (kakos) like being physically tortured by others (resisting suffering in- justice), there is more courage in choosing the former rather ___ are not the same, nor bad and ugly.” (Polus:) “Just so.” (transl. W. R. M Lamb, slightly modified). 98 GALEN ON TO KALON AND TO AGATHON ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 than the latter. For most people, as Polus argues, hold the view that doing injustice, although more shameful, brings more benefit and pleasure than suffering injustice, and thus they find doing injustice a better choice. This is, however, not what the courageous will choose. That is also why Galen, a few lines below, exemplifies the courageous with one “who will endure torture rather than tell a lie about a friend” (33.5).33 In Gorgias we find Socrates as such an example, since he explicitly prefers suffering injustice to doing injustice, if he must choose between the two behaviors.34 Given the similar discussion of to kalon and to agathon, to aischron and to kakon in both contexts, it appears that we have good reason to add Gorgias to the list of Platonic dia- logues which Galen might have used, and which form the intellectual background of De moribus. As we have said, in this context agathos is used not in the highest moral sense, but simply indicates “pleasant, advan- tageous, beneficial” in the amoral sense. This usage is by no means limited to Gorgias, but rather is quite widespread in the whole Platonic corpus. Similar passages are found in dialogues like Meno, Protagoras, or Republic, where agathos stands in parallel with ὠφέλιμος, χρήσιμος (beneficial), and can substitute for the latter or vice versa.35 At the very beginning of the second book of 33 Singer further raises the question that if here two kinds of behavior are intended, the one that resists what is ugly or shameful and the one that resists what is bad, then why does the following text (“The first is like one who prefers death in battle to defeat, and who will endure torture rather than tell a lie about a friend?”) provide only one example, namely the one of the former? I suppose here Singer wrongly traces “the first” back to the former behavior, while in fact it could only be related to “greater courage” as the former case. The example of the latter, namely “greater cowardice,” has been omitted either by Galen or by the Arabic translator on purpose, as doubtless is the case in reverse. 34 Grg. 469C: βουλοίμην μὲν ἂν ἔγωγε ούδέτερα: εἰ δ᾽ ἀναγκαῖον εἴη ἀδικεῖν ἢ ἀδικεῖσθαι, ἑλοίμην ἂν μᾶλλον ἀδικεῖσθαι ἢ ἀδικεῖν, (Socrates) “I should wish neither, for my own part; but if it were necessary either to do wrong or to suffer it, I should choose to suffer rather than do it.” 35 See e.g. Grg. 477A, Prt. 333D–334C, Men. 88D, Resp. 347CD, 379B, 505A. LIJUAN LIN 99 ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 the Republic, we find Glaucon trying to divide to agathon into three kinds (357A–D): (1) the kind of good which we welcome for its own sake, i.e. good things which bring us pleasure; (2) the good which we welcome both for its own sake and for the sake of what comes from it, i.e. good things which bring us both pleasure and benefit; (3) the good which we welcome for the sake of what comes from it, i.e. good things which bring us benefit. The equivalence of to agathon and benefit/pleasure is furthermore in accord with ordinary Greek usage, since the original meaning of agathos does not exclusively refer to the good in the moral sense, but concerns good things in a more general sense.36 For both persons and things, agathos only in later development is used to refer to moral good, whereas its original meaning concerns pleasant, beneficial, and morally neutral things like noble family, excellent ability, or useful instruments.37 In a context where to agathon represents beneficial things in the amoral sense, to kalon would often be set in opposition, and it represents moral excellence as a higher value, as we have observed both in Gorgias and in De moribus. However, it is also to be noted that to kalon in ordinary Greek usage, just like to agathon, is not limited to beauty in the moral sense, but can also refer to beauty of outward form or beauty in reference to use. This also explains why Socrates in Gorgias (474D–E) divides to kalon into the pleasant and the beneficial, and why Socrates’ morally infallible life style can be condemned as aischron by the ___ For a detailed discussion on this topic see P. Stemmer, “Der Grundriß der platonischen Ethik. Karlfried Gru ̈nder zum 60. Geburtstag,” Zeitschrift fur̈ philosophische Forschung 42 (1988) 529–569, especially 542 ff. 36 See e.g. B. Snell, Die Entdeckung des Geistes4 (Göttingen 1975) 170–177; E. Schwartz, Die Ethik der Griechen (Stuttgart 1951) 19–25; C. J. Classen, Sprachliche Deutung als Treibkraft Platonischen und Sokratischen Philosophierens (Munich 1959) 141–144. 37 See LSJ s.v.: “(i) of persons 1. well-born, 2. brave, 3. good, capable, in reference to ability, 4. good, in moral sense […]; (ii) of things 1. good, serviceable, 2. of outward circumstances, 3. morally good.” 100 GALEN ON TO KALON AND TO AGATHON ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 multitude in the Apology (28B). This use of to kalon is, however, not attested in our extant summary. For our present purpose, it is sufficient to note that the complicated relationship between the two essential Platonic terms to kalon and to agathon should chiefly be traced back to the complicated meanings the terms agathos and kalos originally have, insofar as semantically they both can mean very different things.38 As the Arabic epitome has shown us, the diverse uses of the two terms in classical Greek can also be detected in Galen’s De moribus. In this sense, the use of the two value-terms in this summary is not, as Singer thought, contrary to Greek usage, nor it has undergone “some confusing distortion”; but rather, it is Greek and Platonic. 4. Conclusion In this paper I have mainly concentrated on the relationship between to kalon and to agathon in the Arabic epitome of Galen’s De moribus. Through close examination of the related passages in Galen and Plato, I have demonstrated that the prima facie rather complicated relationship between the two terms in fact has its origin in the Platonic theory of soul and related ethical thought. Some confusion over this matter can thus be cleared up, if tackled in the context of certain Platonic dialogues like Gorgias. This research has thus added further evidence to the Platonic character of the Galenic works. I have also tried to call attention to some other aspects of De moribus which are related to Platonic thought. This concerns especially Plato’s tripartition of the soul, his emphasis on the role of the non-rational ele- ments in education, and his positive evaluation of the spirited soul as an assistant of the rational soul. Unlike some scholars, 38 On the relationship between agathos and kalos see further Schwartz, Die Ethik der Griechen 19–27; K. J. Dover, Greek Popular Morality in the Time of Plato and Aristotle (Oxford 1974) 71–72, and Plato: Symposium (Cambridge 1980) 136; T. Irwin, Plato’s Moral Theory (Oxford 1977) 49, 165; P. Woodruff, Plato, Hippias Major (Oxford 1982) 183–189; P. Stemmer, “Unrecht Tun ist schlechter als Unrecht Leiden. Zur Begründung moralischen Handelns im platonischen ‘Gorgias’,” ZPhF 39 (1985) 501–522. LIJUAN LIN 101 ————— Greek, Roman, and Byzantine Studies 58 (2018) 77–101 who are inclined to call into into question the role of the spirited soul in the Platonic theory of soul,39 Galen apparently takes Plato’s discussion of the spirited soul more seriously and duly appreciates its positive role in human life. As he repeatedly emphasizes, the hunter can only attain his goal if he trains his dog well.40 September, 2017 Center for Classical and Medieval Studies Department of History Peking University lijuanlin@pku.edu.cn 39 E.g. W. F. R. Hardie, A Study in Plato (Oxford 1936) 142; T. Penner, “Thought and Desire in Plato,” in G. Vlastos (ed.), Plato: A Collection of Critical Essays II (New York 1971) 113. 40 I would like to express my sincere gratitude and appreciation to Michael Erler, Peter Adamson, Andreas Lammer, Rik van Wijlick, Xi Wang, Boyun Liu, Wei Cheng, Ruobing Xian, and, last but not least, Kent Rigsby and the anonymous referees for their valuable feedback on earlier drafts.